{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.357348703170029,
  "eval_steps": 3000,
  "global_step": 87000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.7425968170166,
      "epoch": 0.0004803073967339097,
      "grad_norm": 5.0,
      "learning_rate": 2e-06,
      "loss": 10.7893,
      "mean_token_accuracy": 0.0001694567094091326,
      "num_tokens": 10855.0,
      "step": 5
    },
    {
      "entropy": 10.742561435699463,
      "epoch": 0.0009606147934678194,
      "grad_norm": 4.8125,
      "learning_rate": 4.5e-06,
      "loss": 10.8164,
      "mean_token_accuracy": 0.0,
      "num_tokens": 24110.0,
      "step": 10
    },
    {
      "entropy": 10.742608547210693,
      "epoch": 0.001440922190201729,
      "grad_norm": 4.875,
      "learning_rate": 7e-06,
      "loss": 10.7754,
      "mean_token_accuracy": 6.963788182474673e-05,
      "num_tokens": 35984.0,
      "step": 15
    },
    {
      "entropy": 10.742629623413086,
      "epoch": 0.0019212295869356388,
      "grad_norm": 4.6875,
      "learning_rate": 9.5e-06,
      "loss": 10.7236,
      "mean_token_accuracy": 7.215007208287716e-05,
      "num_tokens": 48152.0,
      "step": 20
    },
    {
      "entropy": 10.742585182189941,
      "epoch": 0.0024015369836695487,
      "grad_norm": 4.46875,
      "learning_rate": 1.2e-05,
      "loss": 10.6191,
      "mean_token_accuracy": 0.0010923755064141006,
      "num_tokens": 59810.0,
      "step": 25
    },
    {
      "entropy": 10.74241237640381,
      "epoch": 0.002881844380403458,
      "grad_norm": 3.65625,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.4993,
      "mean_token_accuracy": 0.00965826828032732,
      "num_tokens": 70852.0,
      "step": 30
    },
    {
      "entropy": 10.74171323776245,
      "epoch": 0.0033621517771373678,
      "grad_norm": 3.578125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.3736,
      "mean_token_accuracy": 0.025041871704161167,
      "num_tokens": 83378.0,
      "step": 35
    },
    {
      "entropy": 10.74036684036255,
      "epoch": 0.0038424591738712775,
      "grad_norm": 2.75,
      "learning_rate": 1.95e-05,
      "loss": 10.241,
      "mean_token_accuracy": 0.0312205046415329,
      "num_tokens": 95505.0,
      "step": 40
    },
    {
      "entropy": 10.738578128814698,
      "epoch": 0.004322766570605188,
      "grad_norm": 2.3125,
      "learning_rate": 2.2e-05,
      "loss": 10.1218,
      "mean_token_accuracy": 0.03706384804099798,
      "num_tokens": 106812.0,
      "step": 45
    },
    {
      "entropy": 10.737641048431396,
      "epoch": 0.004803073967339097,
      "grad_norm": 2.0,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.0587,
      "mean_token_accuracy": 0.03593158051371574,
      "num_tokens": 118572.0,
      "step": 50
    },
    {
      "entropy": 10.737699699401855,
      "epoch": 0.005283381364073006,
      "grad_norm": 2.0,
      "learning_rate": 2.7e-05,
      "loss": 9.9722,
      "mean_token_accuracy": 0.033468486927449706,
      "num_tokens": 130051.0,
      "step": 55
    },
    {
      "entropy": 10.737590789794922,
      "epoch": 0.005763688760806916,
      "grad_norm": 1.96875,
      "learning_rate": 2.95e-05,
      "loss": 9.9388,
      "mean_token_accuracy": 0.04036580752581358,
      "num_tokens": 141920.0,
      "step": 60
    },
    {
      "entropy": 10.736310005187988,
      "epoch": 0.006243996157540826,
      "grad_norm": 1.8046875,
      "learning_rate": 3.2e-05,
      "loss": 9.8341,
      "mean_token_accuracy": 0.04443504847586155,
      "num_tokens": 152706.0,
      "step": 65
    },
    {
      "entropy": 10.733973693847656,
      "epoch": 0.0067243035542747355,
      "grad_norm": 1.890625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.8099,
      "mean_token_accuracy": 0.04265851955860853,
      "num_tokens": 165253.0,
      "step": 70
    },
    {
      "entropy": 10.73133602142334,
      "epoch": 0.007204610951008645,
      "grad_norm": 1.875,
      "learning_rate": 3.7e-05,
      "loss": 9.7265,
      "mean_token_accuracy": 0.04272236041724682,
      "num_tokens": 176708.0,
      "step": 75
    },
    {
      "entropy": 10.72813663482666,
      "epoch": 0.007684918347742555,
      "grad_norm": 1.8828125,
      "learning_rate": 3.95e-05,
      "loss": 9.6935,
      "mean_token_accuracy": 0.04440820924937725,
      "num_tokens": 188240.0,
      "step": 80
    },
    {
      "entropy": 10.724117946624755,
      "epoch": 0.008165225744476465,
      "grad_norm": 1.890625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.5348,
      "mean_token_accuracy": 0.04667272008955479,
      "num_tokens": 198355.0,
      "step": 85
    },
    {
      "entropy": 10.719026947021485,
      "epoch": 0.008645533141210375,
      "grad_norm": 1.8046875,
      "learning_rate": 4.45e-05,
      "loss": 9.493,
      "mean_token_accuracy": 0.043789859488606454,
      "num_tokens": 209497.0,
      "step": 90
    },
    {
      "entropy": 10.712520503997803,
      "epoch": 0.009125840537944284,
      "grad_norm": 1.953125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.4335,
      "mean_token_accuracy": 0.04756799638271332,
      "num_tokens": 220859.0,
      "step": 95
    },
    {
      "entropy": 10.70284299850464,
      "epoch": 0.009606147934678195,
      "grad_norm": 1.8046875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.3827,
      "mean_token_accuracy": 0.05129999294877052,
      "num_tokens": 231550.0,
      "step": 100
    },
    {
      "entropy": 10.686277961730957,
      "epoch": 0.010086455331412104,
      "grad_norm": 1.78125,
      "learning_rate": 5.2e-05,
      "loss": 9.2483,
      "mean_token_accuracy": 0.05116281881928444,
      "num_tokens": 244210.0,
      "step": 105
    },
    {
      "entropy": 10.665186405181885,
      "epoch": 0.010566762728146013,
      "grad_norm": 1.90625,
      "learning_rate": 5.45e-05,
      "loss": 9.1189,
      "mean_token_accuracy": 0.058265451714396475,
      "num_tokens": 255745.0,
      "step": 110
    },
    {
      "entropy": 10.634323501586914,
      "epoch": 0.011047070124879923,
      "grad_norm": 1.71875,
      "learning_rate": 5.7e-05,
      "loss": 9.0189,
      "mean_token_accuracy": 0.05389787927269936,
      "num_tokens": 266180.0,
      "step": 115
    },
    {
      "entropy": 10.591359043121338,
      "epoch": 0.011527377521613832,
      "grad_norm": 1.7578125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.8917,
      "mean_token_accuracy": 0.057663530483841895,
      "num_tokens": 277736.0,
      "step": 120
    },
    {
      "entropy": 10.54497652053833,
      "epoch": 0.012007684918347743,
      "grad_norm": 1.8203125,
      "learning_rate": 6.2e-05,
      "loss": 8.7685,
      "mean_token_accuracy": 0.061545657366514205,
      "num_tokens": 289069.0,
      "step": 125
    },
    {
      "entropy": 10.48107271194458,
      "epoch": 0.012487992315081652,
      "grad_norm": 1.671875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.6457,
      "mean_token_accuracy": 0.06588748730719089,
      "num_tokens": 300240.0,
      "step": 130
    },
    {
      "entropy": 10.40806179046631,
      "epoch": 0.012968299711815562,
      "grad_norm": 1.6171875,
      "learning_rate": 6.7e-05,
      "loss": 8.558,
      "mean_token_accuracy": 0.06453255712985992,
      "num_tokens": 311698.0,
      "step": 135
    },
    {
      "entropy": 10.328728771209716,
      "epoch": 0.013448607108549471,
      "grad_norm": 1.484375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.4291,
      "mean_token_accuracy": 0.06200851388275623,
      "num_tokens": 322844.0,
      "step": 140
    },
    {
      "entropy": 10.22898988723755,
      "epoch": 0.013928914505283382,
      "grad_norm": 1.3984375,
      "learning_rate": 7.2e-05,
      "loss": 8.3111,
      "mean_token_accuracy": 0.06522167101502419,
      "num_tokens": 335092.0,
      "step": 145
    },
    {
      "entropy": 10.104230976104736,
      "epoch": 0.01440922190201729,
      "grad_norm": 1.3515625,
      "learning_rate": 7.45e-05,
      "loss": 8.1711,
      "mean_token_accuracy": 0.06845104768872261,
      "num_tokens": 347033.0,
      "step": 150
    },
    {
      "entropy": 9.963459777832032,
      "epoch": 0.014889529298751201,
      "grad_norm": 1.3125,
      "learning_rate": 7.7e-05,
      "loss": 8.0479,
      "mean_token_accuracy": 0.06833775304257869,
      "num_tokens": 358696.0,
      "step": 155
    },
    {
      "entropy": 9.770106410980224,
      "epoch": 0.01536983669548511,
      "grad_norm": 1.2890625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.9389,
      "mean_token_accuracy": 0.07227320410311222,
      "num_tokens": 369390.0,
      "step": 160
    },
    {
      "entropy": 9.595024299621581,
      "epoch": 0.01585014409221902,
      "grad_norm": 1.21875,
      "learning_rate": 8.2e-05,
      "loss": 7.832,
      "mean_token_accuracy": 0.07254205383360386,
      "num_tokens": 380540.0,
      "step": 165
    },
    {
      "entropy": 9.395240879058838,
      "epoch": 0.01633045148895293,
      "grad_norm": 1.171875,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.7561,
      "mean_token_accuracy": 0.07409532256424427,
      "num_tokens": 391243.0,
      "step": 170
    },
    {
      "entropy": 9.157040977478028,
      "epoch": 0.01681075888568684,
      "grad_norm": 1.015625,
      "learning_rate": 8.7e-05,
      "loss": 7.6871,
      "mean_token_accuracy": 0.07649487368762493,
      "num_tokens": 403336.0,
      "step": 175
    },
    {
      "entropy": 8.964170169830322,
      "epoch": 0.01729106628242075,
      "grad_norm": 1.2890625,
      "learning_rate": 8.95e-05,
      "loss": 7.5772,
      "mean_token_accuracy": 0.07687406204640865,
      "num_tokens": 413886.0,
      "step": 180
    },
    {
      "entropy": 8.755004787445069,
      "epoch": 0.01777137367915466,
      "grad_norm": 1.1875,
      "learning_rate": 9.2e-05,
      "loss": 7.5785,
      "mean_token_accuracy": 0.07747036479413509,
      "num_tokens": 425277.0,
      "step": 185
    },
    {
      "entropy": 8.616313934326172,
      "epoch": 0.01825168107588857,
      "grad_norm": 1.0546875,
      "learning_rate": 9.45e-05,
      "loss": 7.6259,
      "mean_token_accuracy": 0.079135612398386,
      "num_tokens": 436868.0,
      "step": 190
    },
    {
      "entropy": 8.522105503082276,
      "epoch": 0.018731988472622477,
      "grad_norm": 1.25,
      "learning_rate": 9.7e-05,
      "loss": 7.4359,
      "mean_token_accuracy": 0.08061296418309212,
      "num_tokens": 448349.0,
      "step": 195
    },
    {
      "entropy": 8.420132637023926,
      "epoch": 0.01921229586935639,
      "grad_norm": 1.1328125,
      "learning_rate": 9.95e-05,
      "loss": 7.4205,
      "mean_token_accuracy": 0.08393056243658066,
      "num_tokens": 459447.0,
      "step": 200
    },
    {
      "entropy": 8.309343433380127,
      "epoch": 0.0196926032660903,
      "grad_norm": 1.0,
      "learning_rate": 0.000102,
      "loss": 7.3675,
      "mean_token_accuracy": 0.08649685978889465,
      "num_tokens": 470734.0,
      "step": 205
    },
    {
      "entropy": 8.280020236968994,
      "epoch": 0.020172910662824207,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.4174,
      "mean_token_accuracy": 0.08485044017434121,
      "num_tokens": 482015.0,
      "step": 210
    },
    {
      "entropy": 8.232032871246338,
      "epoch": 0.020653218059558116,
      "grad_norm": 1.125,
      "learning_rate": 0.000107,
      "loss": 7.3525,
      "mean_token_accuracy": 0.08708564937114716,
      "num_tokens": 493339.0,
      "step": 215
    },
    {
      "entropy": 8.191392135620116,
      "epoch": 0.021133525456292025,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001095,
      "loss": 7.289,
      "mean_token_accuracy": 0.0901091642677784,
      "num_tokens": 504924.0,
      "step": 220
    },
    {
      "entropy": 8.096070671081543,
      "epoch": 0.021613832853025938,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000112,
      "loss": 7.2659,
      "mean_token_accuracy": 0.08543256968259812,
      "num_tokens": 516603.0,
      "step": 225
    },
    {
      "entropy": 8.060660076141357,
      "epoch": 0.022094140249759846,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001145,
      "loss": 7.1691,
      "mean_token_accuracy": 0.09287247136235237,
      "num_tokens": 528347.0,
      "step": 230
    },
    {
      "entropy": 8.039314794540406,
      "epoch": 0.022574447646493755,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.2454,
      "mean_token_accuracy": 0.09620365053415299,
      "num_tokens": 539328.0,
      "step": 235
    },
    {
      "entropy": 8.03833818435669,
      "epoch": 0.023054755043227664,
      "grad_norm": 1.171875,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.2157,
      "mean_token_accuracy": 0.10354483127593994,
      "num_tokens": 549297.0,
      "step": 240
    },
    {
      "entropy": 7.88909420967102,
      "epoch": 0.023535062439961577,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000122,
      "loss": 7.1749,
      "mean_token_accuracy": 0.09601260498166084,
      "num_tokens": 560306.0,
      "step": 245
    },
    {
      "entropy": 7.962506532669067,
      "epoch": 0.024015369836695485,
      "grad_norm": 1.9375,
      "learning_rate": 0.0001245,
      "loss": 7.2445,
      "mean_token_accuracy": 0.0955030769109726,
      "num_tokens": 571972.0,
      "step": 250
    },
    {
      "entropy": 7.930570125579834,
      "epoch": 0.024495677233429394,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000127,
      "loss": 7.1279,
      "mean_token_accuracy": 0.10435850247740745,
      "num_tokens": 582962.0,
      "step": 255
    },
    {
      "entropy": 7.853069543838501,
      "epoch": 0.024975984630163303,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0001295,
      "loss": 7.0975,
      "mean_token_accuracy": 0.0972387246787548,
      "num_tokens": 597193.0,
      "step": 260
    },
    {
      "entropy": 7.889965534210205,
      "epoch": 0.025456292026897216,
      "grad_norm": 1.046875,
      "learning_rate": 0.000132,
      "loss": 7.1716,
      "mean_token_accuracy": 0.10085091665387154,
      "num_tokens": 608982.0,
      "step": 265
    },
    {
      "entropy": 7.903727865219116,
      "epoch": 0.025936599423631124,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.206,
      "mean_token_accuracy": 0.09627965837717056,
      "num_tokens": 619953.0,
      "step": 270
    },
    {
      "entropy": 7.821546459197998,
      "epoch": 0.026416906820365033,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.1526,
      "mean_token_accuracy": 0.10060698464512825,
      "num_tokens": 631039.0,
      "step": 275
    },
    {
      "entropy": 7.850963020324707,
      "epoch": 0.026897214217098942,
      "grad_norm": 0.96875,
      "learning_rate": 0.0001395,
      "loss": 7.1375,
      "mean_token_accuracy": 0.10203886106610298,
      "num_tokens": 642656.0,
      "step": 280
    },
    {
      "entropy": 7.698967504501343,
      "epoch": 0.027377521613832854,
      "grad_norm": 1.078125,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.0134,
      "mean_token_accuracy": 0.11590334996581078,
      "num_tokens": 653748.0,
      "step": 285
    },
    {
      "entropy": 7.7186966896057125,
      "epoch": 0.027857829010566763,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001445,
      "loss": 6.9933,
      "mean_token_accuracy": 0.10406283661723137,
      "num_tokens": 665618.0,
      "step": 290
    },
    {
      "entropy": 7.69569730758667,
      "epoch": 0.028338136407300672,
      "grad_norm": 1.15625,
      "learning_rate": 0.000147,
      "loss": 7.0226,
      "mean_token_accuracy": 0.10578782483935356,
      "num_tokens": 677329.0,
      "step": 295
    },
    {
      "entropy": 7.657517051696777,
      "epoch": 0.02881844380403458,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001495,
      "loss": 6.9769,
      "mean_token_accuracy": 0.10939329639077186,
      "num_tokens": 688278.0,
      "step": 300
    },
    {
      "entropy": 7.689532470703125,
      "epoch": 0.029298751200768493,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000152,
      "loss": 7.0338,
      "mean_token_accuracy": 0.10749343782663345,
      "num_tokens": 700739.0,
      "step": 305
    },
    {
      "entropy": 7.570545291900634,
      "epoch": 0.029779058597502402,
      "grad_norm": 1.09375,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.9673,
      "mean_token_accuracy": 0.10927367061376572,
      "num_tokens": 712527.0,
      "step": 310
    },
    {
      "entropy": 7.609434270858765,
      "epoch": 0.03025936599423631,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000157,
      "loss": 6.9962,
      "mean_token_accuracy": 0.10492227971553802,
      "num_tokens": 724514.0,
      "step": 315
    },
    {
      "entropy": 7.632580757141113,
      "epoch": 0.03073967339097022,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001595,
      "loss": 6.9339,
      "mean_token_accuracy": 0.11308366954326629,
      "num_tokens": 735679.0,
      "step": 320
    },
    {
      "entropy": 7.555677700042724,
      "epoch": 0.03121998078770413,
      "grad_norm": 1.125,
      "learning_rate": 0.000162,
      "loss": 6.942,
      "mean_token_accuracy": 0.11490793004631997,
      "num_tokens": 747896.0,
      "step": 325
    },
    {
      "entropy": 7.581246328353882,
      "epoch": 0.03170028818443804,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.9856,
      "mean_token_accuracy": 0.10933408513665199,
      "num_tokens": 759081.0,
      "step": 330
    },
    {
      "entropy": 7.607112407684326,
      "epoch": 0.03218059558117195,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.9207,
      "mean_token_accuracy": 0.10591650679707527,
      "num_tokens": 770459.0,
      "step": 335
    },
    {
      "entropy": 7.448120260238648,
      "epoch": 0.03266090297790586,
      "grad_norm": 1.015625,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.8814,
      "mean_token_accuracy": 0.11157295480370522,
      "num_tokens": 783960.0,
      "step": 340
    },
    {
      "entropy": 7.518631458282471,
      "epoch": 0.03314121037463977,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.9692,
      "mean_token_accuracy": 0.1153689943253994,
      "num_tokens": 795425.0,
      "step": 345
    },
    {
      "entropy": 7.512917995452881,
      "epoch": 0.03362151777137368,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.927,
      "mean_token_accuracy": 0.10967304334044456,
      "num_tokens": 807536.0,
      "step": 350
    },
    {
      "entropy": 7.475287914276123,
      "epoch": 0.034101825168107586,
      "grad_norm": 1.078125,
      "learning_rate": 0.000177,
      "loss": 6.9761,
      "mean_token_accuracy": 0.10532209947705269,
      "num_tokens": 818801.0,
      "step": 355
    },
    {
      "entropy": 7.540158653259278,
      "epoch": 0.0345821325648415,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001795,
      "loss": 6.8451,
      "mean_token_accuracy": 0.11356438398361206,
      "num_tokens": 831497.0,
      "step": 360
    },
    {
      "entropy": 7.401552820205689,
      "epoch": 0.03506243996157541,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000182,
      "loss": 6.8086,
      "mean_token_accuracy": 0.11573421731591224,
      "num_tokens": 842491.0,
      "step": 365
    },
    {
      "entropy": 7.4459820747375485,
      "epoch": 0.03554274735830932,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001845,
      "loss": 6.8907,
      "mean_token_accuracy": 0.10651533678174019,
      "num_tokens": 854560.0,
      "step": 370
    },
    {
      "entropy": 7.452930879592896,
      "epoch": 0.03602305475504323,
      "grad_norm": 1.09375,
      "learning_rate": 0.000187,
      "loss": 6.8048,
      "mean_token_accuracy": 0.12114926502108574,
      "num_tokens": 866688.0,
      "step": 375
    },
    {
      "entropy": 7.454508543014526,
      "epoch": 0.03650336215177714,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001895,
      "loss": 6.9043,
      "mean_token_accuracy": 0.10833153203129768,
      "num_tokens": 879484.0,
      "step": 380
    },
    {
      "entropy": 7.407199048995972,
      "epoch": 0.036983669548511046,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000192,
      "loss": 6.8868,
      "mean_token_accuracy": 0.1142148919403553,
      "num_tokens": 890807.0,
      "step": 385
    },
    {
      "entropy": 7.467933511734008,
      "epoch": 0.037463976945244955,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001945,
      "loss": 6.824,
      "mean_token_accuracy": 0.11662854552268982,
      "num_tokens": 901759.0,
      "step": 390
    },
    {
      "entropy": 7.318777418136596,
      "epoch": 0.037944284341978864,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.7487,
      "mean_token_accuracy": 0.1253423847258091,
      "num_tokens": 912212.0,
      "step": 395
    },
    {
      "entropy": 7.409648275375366,
      "epoch": 0.03842459173871278,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.851,
      "mean_token_accuracy": 0.1168945498764515,
      "num_tokens": 923947.0,
      "step": 400
    },
    {
      "entropy": 7.329876947402954,
      "epoch": 0.03890489913544669,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000202,
      "loss": 6.7936,
      "mean_token_accuracy": 0.1201321728527546,
      "num_tokens": 935732.0,
      "step": 405
    },
    {
      "entropy": 7.254326105117798,
      "epoch": 0.0393852065321806,
      "grad_norm": 1.25,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.7414,
      "mean_token_accuracy": 0.1142658345401287,
      "num_tokens": 948261.0,
      "step": 410
    },
    {
      "entropy": 7.434876155853272,
      "epoch": 0.039865513928914506,
      "grad_norm": 1.203125,
      "learning_rate": 0.000207,
      "loss": 6.8486,
      "mean_token_accuracy": 0.12036997452378273,
      "num_tokens": 959574.0,
      "step": 415
    },
    {
      "entropy": 7.29233832359314,
      "epoch": 0.040345821325648415,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002095,
      "loss": 6.9081,
      "mean_token_accuracy": 0.11429130733013153,
      "num_tokens": 970329.0,
      "step": 420
    },
    {
      "entropy": 7.388989067077636,
      "epoch": 0.040826128722382324,
      "grad_norm": 1.15625,
      "learning_rate": 0.000212,
      "loss": 6.7648,
      "mean_token_accuracy": 0.12700393423438072,
      "num_tokens": 982037.0,
      "step": 425
    },
    {
      "entropy": 7.1995869159698485,
      "epoch": 0.04130643611911623,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002145,
      "loss": 6.7255,
      "mean_token_accuracy": 0.11979713216423989,
      "num_tokens": 994612.0,
      "step": 430
    },
    {
      "entropy": 7.330949687957764,
      "epoch": 0.04178674351585014,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.8381,
      "mean_token_accuracy": 0.12672944292426108,
      "num_tokens": 1005960.0,
      "step": 435
    },
    {
      "entropy": 7.2600654602050785,
      "epoch": 0.04226705091258405,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002195,
      "loss": 6.7001,
      "mean_token_accuracy": 0.1206528477370739,
      "num_tokens": 1017618.0,
      "step": 440
    },
    {
      "entropy": 7.283382511138916,
      "epoch": 0.042747358309317966,
      "grad_norm": 1.140625,
      "learning_rate": 0.000222,
      "loss": 6.7395,
      "mean_token_accuracy": 0.12196639478206635,
      "num_tokens": 1029307.0,
      "step": 445
    },
    {
      "entropy": 7.244419002532959,
      "epoch": 0.043227665706051875,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002245,
      "loss": 6.7559,
      "mean_token_accuracy": 0.11804269179701805,
      "num_tokens": 1042027.0,
      "step": 450
    },
    {
      "entropy": 7.249811124801636,
      "epoch": 0.043707973102785784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.7648,
      "mean_token_accuracy": 0.11966601833701133,
      "num_tokens": 1053125.0,
      "step": 455
    },
    {
      "entropy": 7.156477594375611,
      "epoch": 0.04418828049951969,
      "grad_norm": 1.25,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.7438,
      "mean_token_accuracy": 0.12282297015190125,
      "num_tokens": 1064908.0,
      "step": 460
    },
    {
      "entropy": 7.249273777008057,
      "epoch": 0.0446685878962536,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.6219,
      "mean_token_accuracy": 0.12831630930304527,
      "num_tokens": 1076328.0,
      "step": 465
    },
    {
      "entropy": 7.203511905670166,
      "epoch": 0.04514889529298751,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.8416,
      "mean_token_accuracy": 0.11294705122709274,
      "num_tokens": 1088469.0,
      "step": 470
    },
    {
      "entropy": 7.142552328109741,
      "epoch": 0.04562920268972142,
      "grad_norm": 1.171875,
      "learning_rate": 0.000237,
      "loss": 6.6489,
      "mean_token_accuracy": 0.12703610360622405,
      "num_tokens": 1099408.0,
      "step": 475
    },
    {
      "entropy": 7.071178340911866,
      "epoch": 0.04610951008645533,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002395,
      "loss": 6.5735,
      "mean_token_accuracy": 0.131545989215374,
      "num_tokens": 1111101.0,
      "step": 480
    },
    {
      "entropy": 7.069411039352417,
      "epoch": 0.046589817483189244,
      "grad_norm": 1.1875,
      "learning_rate": 0.000242,
      "loss": 6.5925,
      "mean_token_accuracy": 0.1311088353395462,
      "num_tokens": 1122877.0,
      "step": 485
    },
    {
      "entropy": 7.144361734390259,
      "epoch": 0.04707012487992315,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002445,
      "loss": 6.6995,
      "mean_token_accuracy": 0.11579547598958015,
      "num_tokens": 1133956.0,
      "step": 490
    },
    {
      "entropy": 7.092379188537597,
      "epoch": 0.04755043227665706,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000247,
      "loss": 6.5997,
      "mean_token_accuracy": 0.12872846126556398,
      "num_tokens": 1146101.0,
      "step": 495
    },
    {
      "entropy": 7.134016275405884,
      "epoch": 0.04803073967339097,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002495,
      "loss": 6.6909,
      "mean_token_accuracy": 0.1324249878525734,
      "num_tokens": 1157432.0,
      "step": 500
    },
    {
      "entropy": 7.112411451339722,
      "epoch": 0.04851104707012488,
      "grad_norm": 1.203125,
      "learning_rate": 0.000252,
      "loss": 6.6805,
      "mean_token_accuracy": 0.12597657218575478,
      "num_tokens": 1167601.0,
      "step": 505
    },
    {
      "entropy": 7.080900621414185,
      "epoch": 0.04899135446685879,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002545,
      "loss": 6.5746,
      "mean_token_accuracy": 0.1313086122274399,
      "num_tokens": 1178818.0,
      "step": 510
    },
    {
      "entropy": 7.224948310852051,
      "epoch": 0.0494716618635927,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000257,
      "loss": 6.8153,
      "mean_token_accuracy": 0.12766674906015396,
      "num_tokens": 1189977.0,
      "step": 515
    },
    {
      "entropy": 7.031567239761353,
      "epoch": 0.049951969260326606,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002595,
      "loss": 6.6461,
      "mean_token_accuracy": 0.13004279211163522,
      "num_tokens": 1201039.0,
      "step": 520
    },
    {
      "entropy": 7.180213594436646,
      "epoch": 0.05043227665706052,
      "grad_norm": 1.078125,
      "learning_rate": 0.000262,
      "loss": 6.6694,
      "mean_token_accuracy": 0.13014231398701667,
      "num_tokens": 1212573.0,
      "step": 525
    },
    {
      "entropy": 7.090188646316529,
      "epoch": 0.05091258405379443,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.6139,
      "mean_token_accuracy": 0.1312407836318016,
      "num_tokens": 1223382.0,
      "step": 530
    },
    {
      "entropy": 7.126926517486572,
      "epoch": 0.05139289145052834,
      "grad_norm": 0.90625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6812,
      "mean_token_accuracy": 0.13135352060198785,
      "num_tokens": 1236501.0,
      "step": 535
    },
    {
      "entropy": 7.070159816741944,
      "epoch": 0.05187319884726225,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.6779,
      "mean_token_accuracy": 0.12742857187986373,
      "num_tokens": 1246798.0,
      "step": 540
    },
    {
      "entropy": 7.063770055770874,
      "epoch": 0.05235350624399616,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.601,
      "mean_token_accuracy": 0.13103493079543113,
      "num_tokens": 1258182.0,
      "step": 545
    },
    {
      "entropy": 7.035881757736206,
      "epoch": 0.052833813640730067,
      "grad_norm": 1.125,
      "learning_rate": 0.0002745,
      "loss": 6.5479,
      "mean_token_accuracy": 0.12442576736211777,
      "num_tokens": 1270273.0,
      "step": 550
    },
    {
      "entropy": 6.987351989746093,
      "epoch": 0.053314121037463975,
      "grad_norm": 1.25,
      "learning_rate": 0.000277,
      "loss": 6.6115,
      "mean_token_accuracy": 0.1252695493400097,
      "num_tokens": 1281136.0,
      "step": 555
    },
    {
      "entropy": 7.070683813095092,
      "epoch": 0.053794428434197884,
      "grad_norm": 0.96875,
      "learning_rate": 0.0002795,
      "loss": 6.5936,
      "mean_token_accuracy": 0.12710341513156892,
      "num_tokens": 1294488.0,
      "step": 560
    },
    {
      "entropy": 6.9461894035339355,
      "epoch": 0.05427473583093179,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.5271,
      "mean_token_accuracy": 0.12947569712996482,
      "num_tokens": 1304113.0,
      "step": 565
    },
    {
      "entropy": 7.051584100723266,
      "epoch": 0.05475504322766571,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002845,
      "loss": 6.5709,
      "mean_token_accuracy": 0.13638901486992835,
      "num_tokens": 1315417.0,
      "step": 570
    },
    {
      "entropy": 6.921914052963257,
      "epoch": 0.05523535062439962,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000287,
      "loss": 6.533,
      "mean_token_accuracy": 0.13163114339113235,
      "num_tokens": 1328084.0,
      "step": 575
    },
    {
      "entropy": 7.019269561767578,
      "epoch": 0.05571565802113353,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002895,
      "loss": 6.687,
      "mean_token_accuracy": 0.13050052151083946,
      "num_tokens": 1338696.0,
      "step": 580
    },
    {
      "entropy": 7.040796899795533,
      "epoch": 0.056195965417867436,
      "grad_norm": 1.03125,
      "learning_rate": 0.000292,
      "loss": 6.6282,
      "mean_token_accuracy": 0.12728999853134154,
      "num_tokens": 1350240.0,
      "step": 585
    },
    {
      "entropy": 7.014576768875122,
      "epoch": 0.056676272814601344,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002945,
      "loss": 6.5687,
      "mean_token_accuracy": 0.1278102345764637,
      "num_tokens": 1361720.0,
      "step": 590
    },
    {
      "entropy": 6.951624011993408,
      "epoch": 0.05715658021133525,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000297,
      "loss": 6.5698,
      "mean_token_accuracy": 0.13706584498286248,
      "num_tokens": 1373286.0,
      "step": 595
    },
    {
      "entropy": 6.970668458938599,
      "epoch": 0.05763688760806916,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002995,
      "loss": 6.5226,
      "mean_token_accuracy": 0.12837623208761215,
      "num_tokens": 1384274.0,
      "step": 600
    },
    {
      "entropy": 6.928129005432129,
      "epoch": 0.05811719500480307,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000302,
      "loss": 6.5685,
      "mean_token_accuracy": 0.1299145996570587,
      "num_tokens": 1395355.0,
      "step": 605
    },
    {
      "entropy": 6.85921688079834,
      "epoch": 0.05859750240153699,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003045,
      "loss": 6.4921,
      "mean_token_accuracy": 0.13815943896770477,
      "num_tokens": 1406664.0,
      "step": 610
    },
    {
      "entropy": 7.0746745586395265,
      "epoch": 0.059077809798270896,
      "grad_norm": 1.125,
      "learning_rate": 0.000307,
      "loss": 6.6592,
      "mean_token_accuracy": 0.12862858772277833,
      "num_tokens": 1418450.0,
      "step": 615
    },
    {
      "entropy": 6.919266176223755,
      "epoch": 0.059558117195004805,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003095,
      "loss": 6.5555,
      "mean_token_accuracy": 0.13008297383785247,
      "num_tokens": 1430048.0,
      "step": 620
    },
    {
      "entropy": 6.971873044967651,
      "epoch": 0.060038424591738714,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000312,
      "loss": 6.5905,
      "mean_token_accuracy": 0.12921996414661407,
      "num_tokens": 1441820.0,
      "step": 625
    },
    {
      "entropy": 6.887925004959106,
      "epoch": 0.06051873198847262,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003145,
      "loss": 6.4336,
      "mean_token_accuracy": 0.1331394538283348,
      "num_tokens": 1453209.0,
      "step": 630
    },
    {
      "entropy": 6.868549394607544,
      "epoch": 0.06099903938520653,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000317,
      "loss": 6.4539,
      "mean_token_accuracy": 0.1323256403207779,
      "num_tokens": 1465423.0,
      "step": 635
    },
    {
      "entropy": 6.915077447891235,
      "epoch": 0.06147934678194044,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003195,
      "loss": 6.4443,
      "mean_token_accuracy": 0.1385464273393154,
      "num_tokens": 1476575.0,
      "step": 640
    },
    {
      "entropy": 6.748648643493652,
      "epoch": 0.06195965417867435,
      "grad_norm": 1.203125,
      "learning_rate": 0.000322,
      "loss": 6.4001,
      "mean_token_accuracy": 0.13120974749326705,
      "num_tokens": 1486932.0,
      "step": 645
    },
    {
      "entropy": 6.933510446548462,
      "epoch": 0.06243996157540826,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.4429,
      "mean_token_accuracy": 0.13331692218780516,
      "num_tokens": 1498494.0,
      "step": 650
    },
    {
      "entropy": 6.856443071365357,
      "epoch": 0.06292026897214217,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.4923,
      "mean_token_accuracy": 0.1305932119488716,
      "num_tokens": 1509937.0,
      "step": 655
    },
    {
      "entropy": 6.934806203842163,
      "epoch": 0.06340057636887608,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.5179,
      "mean_token_accuracy": 0.14017625749111176,
      "num_tokens": 1519823.0,
      "step": 660
    },
    {
      "entropy": 6.8656535148620605,
      "epoch": 0.06388088376560999,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.5463,
      "mean_token_accuracy": 0.1294463723897934,
      "num_tokens": 1529943.0,
      "step": 665
    },
    {
      "entropy": 6.814459609985351,
      "epoch": 0.0643611911623439,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.3731,
      "mean_token_accuracy": 0.14023564606904984,
      "num_tokens": 1540618.0,
      "step": 670
    },
    {
      "entropy": 6.802310371398926,
      "epoch": 0.06484149855907781,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000337,
      "loss": 6.4921,
      "mean_token_accuracy": 0.12996583580970764,
      "num_tokens": 1553208.0,
      "step": 675
    },
    {
      "entropy": 6.938099765777588,
      "epoch": 0.06532180595581172,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003395,
      "loss": 6.4714,
      "mean_token_accuracy": 0.14118467569351195,
      "num_tokens": 1563975.0,
      "step": 680
    },
    {
      "entropy": 6.912463045120239,
      "epoch": 0.06580211335254563,
      "grad_norm": 1.15625,
      "learning_rate": 0.000342,
      "loss": 6.5185,
      "mean_token_accuracy": 0.13118280842900276,
      "num_tokens": 1575998.0,
      "step": 685
    },
    {
      "entropy": 6.780094909667969,
      "epoch": 0.06628242074927954,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.4184,
      "mean_token_accuracy": 0.14230406507849694,
      "num_tokens": 1586041.0,
      "step": 690
    },
    {
      "entropy": 6.827849006652832,
      "epoch": 0.06676272814601344,
      "grad_norm": 1.1875,
      "learning_rate": 0.000347,
      "loss": 6.4684,
      "mean_token_accuracy": 0.13458749651908875,
      "num_tokens": 1597531.0,
      "step": 695
    },
    {
      "entropy": 6.795176601409912,
      "epoch": 0.06724303554274735,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003495,
      "loss": 6.4518,
      "mean_token_accuracy": 0.1343138784170151,
      "num_tokens": 1609255.0,
      "step": 700
    },
    {
      "entropy": 6.759337377548218,
      "epoch": 0.06772334293948126,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000352,
      "loss": 6.4421,
      "mean_token_accuracy": 0.13904302567243576,
      "num_tokens": 1621098.0,
      "step": 705
    },
    {
      "entropy": 6.774127531051636,
      "epoch": 0.06820365033621517,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003545,
      "loss": 6.3399,
      "mean_token_accuracy": 0.14766101837158202,
      "num_tokens": 1631941.0,
      "step": 710
    },
    {
      "entropy": 6.74300217628479,
      "epoch": 0.0686839577329491,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000357,
      "loss": 6.3446,
      "mean_token_accuracy": 0.1429346315562725,
      "num_tokens": 1643117.0,
      "step": 715
    },
    {
      "entropy": 6.714703035354614,
      "epoch": 0.069164265129683,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003595,
      "loss": 6.3839,
      "mean_token_accuracy": 0.13676466047763824,
      "num_tokens": 1653595.0,
      "step": 720
    },
    {
      "entropy": 6.830532836914062,
      "epoch": 0.06964457252641691,
      "grad_norm": 1.1875,
      "learning_rate": 0.000362,
      "loss": 6.4213,
      "mean_token_accuracy": 0.13829002231359483,
      "num_tokens": 1664495.0,
      "step": 725
    },
    {
      "entropy": 6.722404527664184,
      "epoch": 0.07012487992315082,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003645,
      "loss": 6.4066,
      "mean_token_accuracy": 0.13997879922389983,
      "num_tokens": 1674923.0,
      "step": 730
    },
    {
      "entropy": 6.769494438171387,
      "epoch": 0.07060518731988473,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000367,
      "loss": 6.2971,
      "mean_token_accuracy": 0.14468497931957244,
      "num_tokens": 1685904.0,
      "step": 735
    },
    {
      "entropy": 6.729282712936401,
      "epoch": 0.07108549471661864,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0003695,
      "loss": 6.5209,
      "mean_token_accuracy": 0.1288589984178543,
      "num_tokens": 1699133.0,
      "step": 740
    },
    {
      "entropy": 6.7413102149963375,
      "epoch": 0.07156580211335255,
      "grad_norm": 0.96875,
      "learning_rate": 0.000372,
      "loss": 6.3281,
      "mean_token_accuracy": 0.13741401731967925,
      "num_tokens": 1711559.0,
      "step": 745
    },
    {
      "entropy": 6.672541761398316,
      "epoch": 0.07204610951008646,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003745,
      "loss": 6.2503,
      "mean_token_accuracy": 0.1432784304022789,
      "num_tokens": 1722526.0,
      "step": 750
    },
    {
      "entropy": 6.6785728454589846,
      "epoch": 0.07252641690682037,
      "grad_norm": 1.171875,
      "learning_rate": 0.000377,
      "loss": 6.3132,
      "mean_token_accuracy": 0.13960239216685294,
      "num_tokens": 1734261.0,
      "step": 755
    },
    {
      "entropy": 6.65572600364685,
      "epoch": 0.07300672430355427,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003795,
      "loss": 6.3011,
      "mean_token_accuracy": 0.14777525439858435,
      "num_tokens": 1745151.0,
      "step": 760
    },
    {
      "entropy": 6.867705583572388,
      "epoch": 0.07348703170028818,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000382,
      "loss": 6.4582,
      "mean_token_accuracy": 0.14053600579500197,
      "num_tokens": 1755463.0,
      "step": 765
    },
    {
      "entropy": 6.743470144271851,
      "epoch": 0.07396733909702209,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003845,
      "loss": 6.4049,
      "mean_token_accuracy": 0.13823116198182106,
      "num_tokens": 1767717.0,
      "step": 770
    },
    {
      "entropy": 6.601284313201904,
      "epoch": 0.074447646493756,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.2542,
      "mean_token_accuracy": 0.14758941754698754,
      "num_tokens": 1779115.0,
      "step": 775
    },
    {
      "entropy": 6.523760461807251,
      "epoch": 0.07492795389048991,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.2018,
      "mean_token_accuracy": 0.14831501990556717,
      "num_tokens": 1789644.0,
      "step": 780
    },
    {
      "entropy": 6.752778768539429,
      "epoch": 0.07540826128722382,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.4227,
      "mean_token_accuracy": 0.14102208092808724,
      "num_tokens": 1800606.0,
      "step": 785
    },
    {
      "entropy": 6.677890968322754,
      "epoch": 0.07588856868395773,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.4269,
      "mean_token_accuracy": 0.13399276584386827,
      "num_tokens": 1812168.0,
      "step": 790
    },
    {
      "entropy": 6.844178104400635,
      "epoch": 0.07636887608069164,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.3673,
      "mean_token_accuracy": 0.13583303466439248,
      "num_tokens": 1823830.0,
      "step": 795
    },
    {
      "entropy": 6.592782688140869,
      "epoch": 0.07684918347742556,
      "grad_norm": 1.125,
      "learning_rate": 0.0003995,
      "loss": 6.3581,
      "mean_token_accuracy": 0.14064425751566886,
      "num_tokens": 1835611.0,
      "step": 800
    },
    {
      "entropy": 6.721078968048095,
      "epoch": 0.07732949087415947,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000402,
      "loss": 6.4246,
      "mean_token_accuracy": 0.13046615198254585,
      "num_tokens": 1847036.0,
      "step": 805
    },
    {
      "entropy": 6.640185070037842,
      "epoch": 0.07780979827089338,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004045,
      "loss": 6.3494,
      "mean_token_accuracy": 0.13754188790917396,
      "num_tokens": 1857476.0,
      "step": 810
    },
    {
      "entropy": 6.721815967559815,
      "epoch": 0.07829010566762729,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.373,
      "mean_token_accuracy": 0.1429060995578766,
      "num_tokens": 1869073.0,
      "step": 815
    },
    {
      "entropy": 6.464849758148193,
      "epoch": 0.0787704130643612,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004095,
      "loss": 6.2412,
      "mean_token_accuracy": 0.14811026602983474,
      "num_tokens": 1880439.0,
      "step": 820
    },
    {
      "entropy": 6.810800933837891,
      "epoch": 0.0792507204610951,
      "grad_norm": 1.125,
      "learning_rate": 0.000412,
      "loss": 6.4175,
      "mean_token_accuracy": 0.13784201666712761,
      "num_tokens": 1891600.0,
      "step": 825
    },
    {
      "entropy": 6.659660720825196,
      "epoch": 0.07973102785782901,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004145,
      "loss": 6.3795,
      "mean_token_accuracy": 0.14037746116518973,
      "num_tokens": 1903126.0,
      "step": 830
    },
    {
      "entropy": 6.764267301559448,
      "epoch": 0.08021133525456292,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000417,
      "loss": 6.3437,
      "mean_token_accuracy": 0.14527037143707275,
      "num_tokens": 1913913.0,
      "step": 835
    },
    {
      "entropy": 6.566430568695068,
      "epoch": 0.08069164265129683,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004195,
      "loss": 6.2667,
      "mean_token_accuracy": 0.14265700429677963,
      "num_tokens": 1924961.0,
      "step": 840
    },
    {
      "entropy": 6.74987416267395,
      "epoch": 0.08117195004803074,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000422,
      "loss": 6.3443,
      "mean_token_accuracy": 0.13290713876485824,
      "num_tokens": 1936773.0,
      "step": 845
    },
    {
      "entropy": 6.521253681182861,
      "epoch": 0.08165225744476465,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004245,
      "loss": 6.2669,
      "mean_token_accuracy": 0.14235577881336212,
      "num_tokens": 1948190.0,
      "step": 850
    },
    {
      "entropy": 6.609525537490844,
      "epoch": 0.08213256484149856,
      "grad_norm": 1.140625,
      "learning_rate": 0.000427,
      "loss": 6.2573,
      "mean_token_accuracy": 0.14473007172346114,
      "num_tokens": 1960038.0,
      "step": 855
    },
    {
      "entropy": 6.668189859390258,
      "epoch": 0.08261287223823247,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004295,
      "loss": 6.3464,
      "mean_token_accuracy": 0.14037344232201576,
      "num_tokens": 1970535.0,
      "step": 860
    },
    {
      "entropy": 6.6438507556915285,
      "epoch": 0.08309317963496637,
      "grad_norm": 1.1875,
      "learning_rate": 0.000432,
      "loss": 6.3179,
      "mean_token_accuracy": 0.1358397677540779,
      "num_tokens": 1981386.0,
      "step": 865
    },
    {
      "entropy": 6.555328369140625,
      "epoch": 0.08357348703170028,
      "grad_norm": 1.0,
      "learning_rate": 0.0004345,
      "loss": 6.2349,
      "mean_token_accuracy": 0.14654369950294494,
      "num_tokens": 1993196.0,
      "step": 870
    },
    {
      "entropy": 6.608275127410889,
      "epoch": 0.08405379442843419,
      "grad_norm": 1.046875,
      "learning_rate": 0.000437,
      "loss": 6.2625,
      "mean_token_accuracy": 0.15001987367868425,
      "num_tokens": 2004756.0,
      "step": 875
    },
    {
      "entropy": 6.563670921325683,
      "epoch": 0.0845341018251681,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004395,
      "loss": 6.2449,
      "mean_token_accuracy": 0.14094123244285583,
      "num_tokens": 2016020.0,
      "step": 880
    },
    {
      "entropy": 6.522834205627442,
      "epoch": 0.08501440922190202,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000442,
      "loss": 6.288,
      "mean_token_accuracy": 0.14180351421236992,
      "num_tokens": 2027747.0,
      "step": 885
    },
    {
      "entropy": 6.644138765335083,
      "epoch": 0.08549471661863593,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004445,
      "loss": 6.2648,
      "mean_token_accuracy": 0.1389330357313156,
      "num_tokens": 2038841.0,
      "step": 890
    },
    {
      "entropy": 6.5890850067138675,
      "epoch": 0.08597502401536984,
      "grad_norm": 1.125,
      "learning_rate": 0.000447,
      "loss": 6.3071,
      "mean_token_accuracy": 0.13798698037862778,
      "num_tokens": 2049905.0,
      "step": 895
    },
    {
      "entropy": 6.653052091598511,
      "epoch": 0.08645533141210375,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3869,
      "mean_token_accuracy": 0.13413772657513617,
      "num_tokens": 2062492.0,
      "step": 900
    },
    {
      "entropy": 6.5456493377685545,
      "epoch": 0.08693563880883766,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.3093,
      "mean_token_accuracy": 0.13950769752264022,
      "num_tokens": 2073840.0,
      "step": 905
    },
    {
      "entropy": 6.569193363189697,
      "epoch": 0.08741594620557157,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.2225,
      "mean_token_accuracy": 0.14058948308229446,
      "num_tokens": 2085720.0,
      "step": 910
    },
    {
      "entropy": 6.613348484039307,
      "epoch": 0.08789625360230548,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.3019,
      "mean_token_accuracy": 0.14243812635540962,
      "num_tokens": 2096649.0,
      "step": 915
    },
    {
      "entropy": 6.631076240539551,
      "epoch": 0.08837656099903939,
      "grad_norm": 0.90625,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.3033,
      "mean_token_accuracy": 0.14374962896108628,
      "num_tokens": 2109030.0,
      "step": 920
    },
    {
      "entropy": 6.521146059036255,
      "epoch": 0.0888568683957733,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000462,
      "loss": 6.2097,
      "mean_token_accuracy": 0.14402098059654236,
      "num_tokens": 2121384.0,
      "step": 925
    },
    {
      "entropy": 6.506090831756592,
      "epoch": 0.0893371757925072,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004645,
      "loss": 6.1669,
      "mean_token_accuracy": 0.15031797289848328,
      "num_tokens": 2131377.0,
      "step": 930
    },
    {
      "entropy": 6.554385089874268,
      "epoch": 0.08981748318924111,
      "grad_norm": 1.0625,
      "learning_rate": 0.000467,
      "loss": 6.3537,
      "mean_token_accuracy": 0.13973637223243712,
      "num_tokens": 2142364.0,
      "step": 935
    },
    {
      "entropy": 6.5829205989837645,
      "epoch": 0.09029779058597502,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004695,
      "loss": 6.2265,
      "mean_token_accuracy": 0.14725347831845284,
      "num_tokens": 2153040.0,
      "step": 940
    },
    {
      "entropy": 6.523045444488526,
      "epoch": 0.09077809798270893,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000472,
      "loss": 6.2703,
      "mean_token_accuracy": 0.14759753867983819,
      "num_tokens": 2165571.0,
      "step": 945
    },
    {
      "entropy": 6.537745571136474,
      "epoch": 0.09125840537944284,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004745,
      "loss": 6.2456,
      "mean_token_accuracy": 0.14311620220541954,
      "num_tokens": 2177241.0,
      "step": 950
    },
    {
      "entropy": 6.572008037567139,
      "epoch": 0.09173871277617675,
      "grad_norm": 1.078125,
      "learning_rate": 0.000477,
      "loss": 6.2812,
      "mean_token_accuracy": 0.13619020357728004,
      "num_tokens": 2187475.0,
      "step": 955
    },
    {
      "entropy": 6.5255202293396,
      "epoch": 0.09221902017291066,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004795,
      "loss": 6.3112,
      "mean_token_accuracy": 0.14339686632156373,
      "num_tokens": 2198185.0,
      "step": 960
    },
    {
      "entropy": 6.506760311126709,
      "epoch": 0.09269932756964457,
      "grad_norm": 1.15625,
      "learning_rate": 0.000482,
      "loss": 6.2432,
      "mean_token_accuracy": 0.145152947306633,
      "num_tokens": 2210404.0,
      "step": 965
    },
    {
      "entropy": 6.614426374435425,
      "epoch": 0.09317963496637849,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004845,
      "loss": 6.2227,
      "mean_token_accuracy": 0.1446111634373665,
      "num_tokens": 2222188.0,
      "step": 970
    },
    {
      "entropy": 6.4725706577301025,
      "epoch": 0.0936599423631124,
      "grad_norm": 1.0625,
      "learning_rate": 0.000487,
      "loss": 6.2734,
      "mean_token_accuracy": 0.14453680962324142,
      "num_tokens": 2233418.0,
      "step": 975
    },
    {
      "entropy": 6.50886721611023,
      "epoch": 0.0941402497598463,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004895,
      "loss": 6.2401,
      "mean_token_accuracy": 0.1501119926571846,
      "num_tokens": 2245053.0,
      "step": 980
    },
    {
      "entropy": 6.550162982940674,
      "epoch": 0.09462055715658022,
      "grad_norm": 1.0625,
      "learning_rate": 0.000492,
      "loss": 6.2607,
      "mean_token_accuracy": 0.14686937928199767,
      "num_tokens": 2256375.0,
      "step": 985
    },
    {
      "entropy": 6.408747339248658,
      "epoch": 0.09510086455331412,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004945,
      "loss": 6.1015,
      "mean_token_accuracy": 0.14864389300346376,
      "num_tokens": 2267074.0,
      "step": 990
    },
    {
      "entropy": 6.412355279922485,
      "epoch": 0.09558117195004803,
      "grad_norm": 1.125,
      "learning_rate": 0.000497,
      "loss": 6.1563,
      "mean_token_accuracy": 0.15030216574668884,
      "num_tokens": 2277168.0,
      "step": 995
    },
    {
      "entropy": 6.463335084915161,
      "epoch": 0.09606147934678194,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995,
      "loss": 6.1421,
      "mean_token_accuracy": 0.15055060535669326,
      "num_tokens": 2288178.0,
      "step": 1000
    },
    {
      "entropy": 6.375842523574829,
      "epoch": 0.09654178674351585,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999999983283737,
      "loss": 6.1518,
      "mean_token_accuracy": 0.14721509963274002,
      "num_tokens": 2299765.0,
      "step": 1005
    },
    {
      "entropy": 6.613855504989624,
      "epoch": 0.09702209414024976,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999999915373924,
      "loss": 6.2717,
      "mean_token_accuracy": 0.1355791248381138,
      "num_tokens": 2312047.0,
      "step": 1010
    },
    {
      "entropy": 6.423848724365234,
      "epoch": 0.09750240153698367,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999999795225793,
      "loss": 6.2421,
      "mean_token_accuracy": 0.13908444195985795,
      "num_tokens": 2324118.0,
      "step": 1015
    },
    {
      "entropy": 6.541734027862549,
      "epoch": 0.09798270893371758,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999999622839347,
      "loss": 6.2441,
      "mean_token_accuracy": 0.14666304737329483,
      "num_tokens": 2335171.0,
      "step": 1020
    },
    {
      "entropy": 6.506129121780395,
      "epoch": 0.09846301633045149,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999999398214593,
      "loss": 6.1528,
      "mean_token_accuracy": 0.14639462679624557,
      "num_tokens": 2346338.0,
      "step": 1025
    },
    {
      "entropy": 6.4336082458496096,
      "epoch": 0.0989433237271854,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999999121351532,
      "loss": 6.1809,
      "mean_token_accuracy": 0.1509660616517067,
      "num_tokens": 2357185.0,
      "step": 1030
    },
    {
      "entropy": 6.422454690933227,
      "epoch": 0.0994236311239193,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999998792250173,
      "loss": 6.1198,
      "mean_token_accuracy": 0.1525423288345337,
      "num_tokens": 2368494.0,
      "step": 1035
    },
    {
      "entropy": 6.528336238861084,
      "epoch": 0.09990393852065321,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999998410910524,
      "loss": 6.318,
      "mean_token_accuracy": 0.13721144571900368,
      "num_tokens": 2380800.0,
      "step": 1040
    },
    {
      "entropy": 6.4995053768157955,
      "epoch": 0.10038424591738712,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999997977332592,
      "loss": 6.2573,
      "mean_token_accuracy": 0.1406011924147606,
      "num_tokens": 2391753.0,
      "step": 1045
    },
    {
      "entropy": 6.355279207229614,
      "epoch": 0.10086455331412104,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999997491516389,
      "loss": 6.1348,
      "mean_token_accuracy": 0.14223827719688414,
      "num_tokens": 2403324.0,
      "step": 1050
    },
    {
      "entropy": 6.485898113250732,
      "epoch": 0.10134486071085495,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999996953461925,
      "loss": 6.2381,
      "mean_token_accuracy": 0.14303919151425362,
      "num_tokens": 2414873.0,
      "step": 1055
    },
    {
      "entropy": 6.298157644271851,
      "epoch": 0.10182516810758886,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999996363169212,
      "loss": 6.0291,
      "mean_token_accuracy": 0.15766309648752214,
      "num_tokens": 2425308.0,
      "step": 1060
    },
    {
      "entropy": 6.33884563446045,
      "epoch": 0.10230547550432277,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999995720638266,
      "loss": 6.0714,
      "mean_token_accuracy": 0.15607305616140366,
      "num_tokens": 2436835.0,
      "step": 1065
    },
    {
      "entropy": 6.4138219356536865,
      "epoch": 0.10278578290105668,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00049999950258691,
      "loss": 6.1795,
      "mean_token_accuracy": 0.14858196377754213,
      "num_tokens": 2446798.0,
      "step": 1070
    },
    {
      "entropy": 6.343221044540405,
      "epoch": 0.10326609029779059,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999994278861731,
      "loss": 6.0823,
      "mean_token_accuracy": 0.1503281533718109,
      "num_tokens": 2457308.0,
      "step": 1075
    },
    {
      "entropy": 6.410136365890503,
      "epoch": 0.1037463976945245,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999993479616175,
      "loss": 6.1524,
      "mean_token_accuracy": 0.14141261354088783,
      "num_tokens": 2468917.0,
      "step": 1080
    },
    {
      "entropy": 6.4494935989379885,
      "epoch": 0.1042267050912584,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999992628132451,
      "loss": 6.1666,
      "mean_token_accuracy": 0.14505101442337037,
      "num_tokens": 2481363.0,
      "step": 1085
    },
    {
      "entropy": 6.386939859390258,
      "epoch": 0.10470701248799232,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999991724410582,
      "loss": 6.1591,
      "mean_token_accuracy": 0.14499209821224213,
      "num_tokens": 2493082.0,
      "step": 1090
    },
    {
      "entropy": 6.401579475402832,
      "epoch": 0.10518731988472622,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999990768450583,
      "loss": 6.1193,
      "mean_token_accuracy": 0.14764227569103242,
      "num_tokens": 2503849.0,
      "step": 1095
    },
    {
      "entropy": 6.376060819625854,
      "epoch": 0.10566762728146013,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999989760252482,
      "loss": 6.1642,
      "mean_token_accuracy": 0.14896273389458656,
      "num_tokens": 2514528.0,
      "step": 1100
    },
    {
      "entropy": 6.361701393127442,
      "epoch": 0.10614793467819404,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004999988699816299,
      "loss": 6.1386,
      "mean_token_accuracy": 0.15386009961366653,
      "num_tokens": 2524971.0,
      "step": 1105
    },
    {
      "entropy": 6.2997602939605715,
      "epoch": 0.10662824207492795,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999987587142058,
      "loss": 6.0389,
      "mean_token_accuracy": 0.1510354369878769,
      "num_tokens": 2535674.0,
      "step": 1110
    },
    {
      "entropy": 6.495864677429199,
      "epoch": 0.10710854947166186,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999986422229789,
      "loss": 6.2935,
      "mean_token_accuracy": 0.13760789409279822,
      "num_tokens": 2547108.0,
      "step": 1115
    },
    {
      "entropy": 6.3444905281066895,
      "epoch": 0.10758885686839577,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999985205079514,
      "loss": 6.0935,
      "mean_token_accuracy": 0.14668779224157333,
      "num_tokens": 2559474.0,
      "step": 1120
    },
    {
      "entropy": 6.281940221786499,
      "epoch": 0.10806916426512968,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999983935691265,
      "loss": 5.9473,
      "mean_token_accuracy": 0.16225168406963347,
      "num_tokens": 2571264.0,
      "step": 1125
    },
    {
      "entropy": 6.2712925434112545,
      "epoch": 0.10854947166186359,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499998261406507,
      "loss": 6.1068,
      "mean_token_accuracy": 0.1521009087562561,
      "num_tokens": 2583731.0,
      "step": 1130
    },
    {
      "entropy": 6.588572835922241,
      "epoch": 0.10902977905859751,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999981240200958,
      "loss": 6.1624,
      "mean_token_accuracy": 0.1474340245127678,
      "num_tokens": 2595497.0,
      "step": 1135
    },
    {
      "entropy": 6.184768009185791,
      "epoch": 0.10951008645533142,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999979814098966,
      "loss": 6.1175,
      "mean_token_accuracy": 0.15334238335490227,
      "num_tokens": 2607358.0,
      "step": 1140
    },
    {
      "entropy": 6.4007182121276855,
      "epoch": 0.10999039385206533,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999978335759121,
      "loss": 6.0488,
      "mean_token_accuracy": 0.15680247992277146,
      "num_tokens": 2618936.0,
      "step": 1145
    },
    {
      "entropy": 6.322919797897339,
      "epoch": 0.11047070124879924,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999976805181461,
      "loss": 6.1873,
      "mean_token_accuracy": 0.1458278477191925,
      "num_tokens": 2631840.0,
      "step": 1150
    },
    {
      "entropy": 6.365493202209473,
      "epoch": 0.11095100864553314,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499997522236602,
      "loss": 6.138,
      "mean_token_accuracy": 0.15330254882574082,
      "num_tokens": 2642412.0,
      "step": 1155
    },
    {
      "entropy": 6.349469995498657,
      "epoch": 0.11143131604226705,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999973587312837,
      "loss": 6.1062,
      "mean_token_accuracy": 0.14750108644366264,
      "num_tokens": 2653890.0,
      "step": 1160
    },
    {
      "entropy": 6.458298873901367,
      "epoch": 0.11191162343900096,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999971900021947,
      "loss": 6.1649,
      "mean_token_accuracy": 0.1521790385246277,
      "num_tokens": 2664888.0,
      "step": 1165
    },
    {
      "entropy": 6.242096948623657,
      "epoch": 0.11239193083573487,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999970160493391,
      "loss": 6.0665,
      "mean_token_accuracy": 0.14765168353915215,
      "num_tokens": 2675550.0,
      "step": 1170
    },
    {
      "entropy": 6.323576927185059,
      "epoch": 0.11287223823246878,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999968368727209,
      "loss": 6.0731,
      "mean_token_accuracy": 0.15634231716394426,
      "num_tokens": 2688022.0,
      "step": 1175
    },
    {
      "entropy": 6.322928762435913,
      "epoch": 0.11335254562920269,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999966524723442,
      "loss": 6.0667,
      "mean_token_accuracy": 0.15129867494106292,
      "num_tokens": 2698737.0,
      "step": 1180
    },
    {
      "entropy": 6.321136522293091,
      "epoch": 0.1138328530259366,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999964628482135,
      "loss": 6.0445,
      "mean_token_accuracy": 0.15983350723981857,
      "num_tokens": 2709844.0,
      "step": 1185
    },
    {
      "entropy": 6.330933856964111,
      "epoch": 0.1143131604226705,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999962680003328,
      "loss": 6.0873,
      "mean_token_accuracy": 0.1584131069481373,
      "num_tokens": 2720273.0,
      "step": 1190
    },
    {
      "entropy": 6.310527086257935,
      "epoch": 0.11479346781940442,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499996067928707,
      "loss": 6.1154,
      "mean_token_accuracy": 0.14796678125858306,
      "num_tokens": 2731354.0,
      "step": 1195
    },
    {
      "entropy": 6.2873091220855715,
      "epoch": 0.11527377521613832,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999958626333406,
      "loss": 6.0851,
      "mean_token_accuracy": 0.15372786819934844,
      "num_tokens": 2742966.0,
      "step": 1200
    },
    {
      "entropy": 6.31016173362732,
      "epoch": 0.11575408261287223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999956521142383,
      "loss": 6.0253,
      "mean_token_accuracy": 0.1610312283039093,
      "num_tokens": 2755010.0,
      "step": 1205
    },
    {
      "entropy": 6.299917507171631,
      "epoch": 0.11623439000960614,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999954363714051,
      "loss": 6.0553,
      "mean_token_accuracy": 0.1533173903822899,
      "num_tokens": 2766176.0,
      "step": 1210
    },
    {
      "entropy": 6.396469974517823,
      "epoch": 0.11671469740634005,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999952154048459,
      "loss": 6.1947,
      "mean_token_accuracy": 0.14717967435717583,
      "num_tokens": 2777861.0,
      "step": 1215
    },
    {
      "entropy": 6.267566394805908,
      "epoch": 0.11719500480307397,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499994989214566,
      "loss": 6.0127,
      "mean_token_accuracy": 0.15340577512979509,
      "num_tokens": 2788725.0,
      "step": 1220
    },
    {
      "entropy": 6.235751724243164,
      "epoch": 0.11767531219980788,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999947578005705,
      "loss": 6.0311,
      "mean_token_accuracy": 0.15211454331874846,
      "num_tokens": 2801613.0,
      "step": 1225
    },
    {
      "entropy": 6.352841997146607,
      "epoch": 0.11815561959654179,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999945211628648,
      "loss": 6.1142,
      "mean_token_accuracy": 0.1481211692094803,
      "num_tokens": 2812474.0,
      "step": 1230
    },
    {
      "entropy": 6.22174391746521,
      "epoch": 0.1186359269932757,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999942793014544,
      "loss": 6.0198,
      "mean_token_accuracy": 0.1554572179913521,
      "num_tokens": 2823178.0,
      "step": 1235
    },
    {
      "entropy": 6.201157283782959,
      "epoch": 0.11911623439000961,
      "grad_norm": 0.859375,
      "learning_rate": 0.000499994032216345,
      "loss": 6.0082,
      "mean_token_accuracy": 0.15631193220615386,
      "num_tokens": 2836486.0,
      "step": 1240
    },
    {
      "entropy": 6.298185110092163,
      "epoch": 0.11959654178674352,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999937799075422,
      "loss": 6.0849,
      "mean_token_accuracy": 0.15345329195261,
      "num_tokens": 2847902.0,
      "step": 1245
    },
    {
      "entropy": 6.128469324111938,
      "epoch": 0.12007684918347743,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499993522375052,
      "loss": 5.9631,
      "mean_token_accuracy": 0.15789376199245453,
      "num_tokens": 2859991.0,
      "step": 1250
    },
    {
      "entropy": 6.42838487625122,
      "epoch": 0.12055715658021134,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999932596188802,
      "loss": 6.1567,
      "mean_token_accuracy": 0.14668479561805725,
      "num_tokens": 2870269.0,
      "step": 1255
    },
    {
      "entropy": 6.199470043182373,
      "epoch": 0.12103746397694524,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999929916390331,
      "loss": 6.0278,
      "mean_token_accuracy": 0.15147914439439775,
      "num_tokens": 2882191.0,
      "step": 1260
    },
    {
      "entropy": 6.238399505615234,
      "epoch": 0.12151777137367915,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999927184355169,
      "loss": 6.0384,
      "mean_token_accuracy": 0.15111806318163873,
      "num_tokens": 2892775.0,
      "step": 1265
    },
    {
      "entropy": 6.218603134155273,
      "epoch": 0.12199807877041306,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999924400083377,
      "loss": 6.022,
      "mean_token_accuracy": 0.15733990743756293,
      "num_tokens": 2904750.0,
      "step": 1270
    },
    {
      "entropy": 6.357618188858032,
      "epoch": 0.12247838616714697,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999921563575022,
      "loss": 6.0938,
      "mean_token_accuracy": 0.1490491099655628,
      "num_tokens": 2916150.0,
      "step": 1275
    },
    {
      "entropy": 6.287556076049805,
      "epoch": 0.12295869356388088,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999918674830169,
      "loss": 6.0557,
      "mean_token_accuracy": 0.15047229453921318,
      "num_tokens": 2928452.0,
      "step": 1280
    },
    {
      "entropy": 6.229531908035279,
      "epoch": 0.12343900096061479,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999915733848886,
      "loss": 6.052,
      "mean_token_accuracy": 0.14816432297229767,
      "num_tokens": 2940577.0,
      "step": 1285
    },
    {
      "entropy": 6.193968200683594,
      "epoch": 0.1239193083573487,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499991274063124,
      "loss": 6.0271,
      "mean_token_accuracy": 0.15466209426522254,
      "num_tokens": 2952302.0,
      "step": 1290
    },
    {
      "entropy": 6.348705244064331,
      "epoch": 0.12439961575408261,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999909695177301,
      "loss": 6.0836,
      "mean_token_accuracy": 0.1576863333582878,
      "num_tokens": 2964611.0,
      "step": 1295
    },
    {
      "entropy": 6.245805740356445,
      "epoch": 0.12487992315081652,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499990659748714,
      "loss": 6.0714,
      "mean_token_accuracy": 0.1507440909743309,
      "num_tokens": 2975668.0,
      "step": 1300
    },
    {
      "entropy": 6.332144260406494,
      "epoch": 0.12536023054755044,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999903447560828,
      "loss": 6.1263,
      "mean_token_accuracy": 0.15026951879262923,
      "num_tokens": 2987303.0,
      "step": 1305
    },
    {
      "entropy": 6.330797815322876,
      "epoch": 0.12584053794428435,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999900245398439,
      "loss": 6.0264,
      "mean_token_accuracy": 0.15367057919502258,
      "num_tokens": 3000400.0,
      "step": 1310
    },
    {
      "entropy": 6.232060718536377,
      "epoch": 0.12632084534101826,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999896991000047,
      "loss": 5.9647,
      "mean_token_accuracy": 0.1533641867339611,
      "num_tokens": 3012336.0,
      "step": 1315
    },
    {
      "entropy": 6.258577537536621,
      "epoch": 0.12680115273775217,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999893684365729,
      "loss": 6.0286,
      "mean_token_accuracy": 0.152130339294672,
      "num_tokens": 3023004.0,
      "step": 1320
    },
    {
      "entropy": 6.220389127731323,
      "epoch": 0.12728146013448607,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999890325495559,
      "loss": 6.0975,
      "mean_token_accuracy": 0.14989801868796349,
      "num_tokens": 3035147.0,
      "step": 1325
    },
    {
      "entropy": 6.253002405166626,
      "epoch": 0.12776176753121998,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999886914389617,
      "loss": 5.9242,
      "mean_token_accuracy": 0.15818189233541488,
      "num_tokens": 3045611.0,
      "step": 1330
    },
    {
      "entropy": 6.172885894775391,
      "epoch": 0.1282420749279539,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999883451047981,
      "loss": 5.9499,
      "mean_token_accuracy": 0.1549012377858162,
      "num_tokens": 3056420.0,
      "step": 1335
    },
    {
      "entropy": 6.169215822219849,
      "epoch": 0.1287223823246878,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999879935470733,
      "loss": 5.9238,
      "mean_token_accuracy": 0.15970174074172974,
      "num_tokens": 3068770.0,
      "step": 1340
    },
    {
      "entropy": 6.298935747146606,
      "epoch": 0.1292026897214217,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999876367657954,
      "loss": 6.062,
      "mean_token_accuracy": 0.14842666387557985,
      "num_tokens": 3080806.0,
      "step": 1345
    },
    {
      "entropy": 6.276106071472168,
      "epoch": 0.12968299711815562,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999872747609725,
      "loss": 6.0784,
      "mean_token_accuracy": 0.1495455838739872,
      "num_tokens": 3091769.0,
      "step": 1350
    },
    {
      "entropy": 6.219928741455078,
      "epoch": 0.13016330451488953,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999869075326132,
      "loss": 5.9957,
      "mean_token_accuracy": 0.1552440881729126,
      "num_tokens": 3103121.0,
      "step": 1355
    },
    {
      "entropy": 6.204897928237915,
      "epoch": 0.13064361191162344,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000499986535080726,
      "loss": 5.9743,
      "mean_token_accuracy": 0.16277382969856263,
      "num_tokens": 3115606.0,
      "step": 1360
    },
    {
      "entropy": 6.145621061325073,
      "epoch": 0.13112391930835735,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999861574053196,
      "loss": 5.8792,
      "mean_token_accuracy": 0.16227115243673323,
      "num_tokens": 3127961.0,
      "step": 1365
    },
    {
      "entropy": 6.100908708572388,
      "epoch": 0.13160422670509125,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999857745064027,
      "loss": 5.8918,
      "mean_token_accuracy": 0.16080396622419357,
      "num_tokens": 3138316.0,
      "step": 1370
    },
    {
      "entropy": 6.145740222930908,
      "epoch": 0.13208453410182516,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499985386383984,
      "loss": 5.8898,
      "mean_token_accuracy": 0.16010251343250276,
      "num_tokens": 3150818.0,
      "step": 1375
    },
    {
      "entropy": 6.221674299240112,
      "epoch": 0.13256484149855907,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999849930380729,
      "loss": 6.0332,
      "mean_token_accuracy": 0.15478831380605698,
      "num_tokens": 3162066.0,
      "step": 1380
    },
    {
      "entropy": 6.167523431777954,
      "epoch": 0.13304514889529298,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999845944686781,
      "loss": 5.9967,
      "mean_token_accuracy": 0.1528627470135689,
      "num_tokens": 3172209.0,
      "step": 1385
    },
    {
      "entropy": 6.064603567123413,
      "epoch": 0.1335254562920269,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999841906758093,
      "loss": 5.8208,
      "mean_token_accuracy": 0.16878736317157744,
      "num_tokens": 3183248.0,
      "step": 1390
    },
    {
      "entropy": 6.167459440231323,
      "epoch": 0.1340057636887608,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999837816594757,
      "loss": 5.9321,
      "mean_token_accuracy": 0.16111771613359452,
      "num_tokens": 3194748.0,
      "step": 1395
    },
    {
      "entropy": 6.154092168807983,
      "epoch": 0.1344860710854947,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999833674196865,
      "loss": 5.8874,
      "mean_token_accuracy": 0.1668296754360199,
      "num_tokens": 3205669.0,
      "step": 1400
    },
    {
      "entropy": 6.156223106384277,
      "epoch": 0.13496637848222862,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999829479564518,
      "loss": 5.989,
      "mean_token_accuracy": 0.15401702225208283,
      "num_tokens": 3216035.0,
      "step": 1405
    },
    {
      "entropy": 6.196358156204224,
      "epoch": 0.13544668587896252,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499982523269781,
      "loss": 5.9783,
      "mean_token_accuracy": 0.15676602721214294,
      "num_tokens": 3227192.0,
      "step": 1410
    },
    {
      "entropy": 6.147973108291626,
      "epoch": 0.13592699327569643,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999820933596842,
      "loss": 6.0023,
      "mean_token_accuracy": 0.15647790133953093,
      "num_tokens": 3240237.0,
      "step": 1415
    },
    {
      "entropy": 6.207319927215576,
      "epoch": 0.13640730067243034,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499981658226171,
      "loss": 5.8811,
      "mean_token_accuracy": 0.1610751822590828,
      "num_tokens": 3251963.0,
      "step": 1420
    },
    {
      "entropy": 6.076681470870971,
      "epoch": 0.13688760806916425,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499981217869252,
      "loss": 5.9913,
      "mean_token_accuracy": 0.1612226501107216,
      "num_tokens": 3263101.0,
      "step": 1425
    },
    {
      "entropy": 6.216303873062134,
      "epoch": 0.1373679154658982,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499980772288937,
      "loss": 5.8768,
      "mean_token_accuracy": 0.16368011087179185,
      "num_tokens": 3275100.0,
      "step": 1430
    },
    {
      "entropy": 6.141348648071289,
      "epoch": 0.1378482228626321,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999803214852367,
      "loss": 5.9678,
      "mean_token_accuracy": 0.15624959021806717,
      "num_tokens": 3287025.0,
      "step": 1435
    },
    {
      "entropy": 6.223503255844117,
      "epoch": 0.138328530259366,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999798654581613,
      "loss": 5.9712,
      "mean_token_accuracy": 0.1579916626214981,
      "num_tokens": 3299867.0,
      "step": 1440
    },
    {
      "entropy": 6.127595090866089,
      "epoch": 0.13880883765609991,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999794042077214,
      "loss": 5.9159,
      "mean_token_accuracy": 0.1553739383816719,
      "num_tokens": 3311183.0,
      "step": 1445
    },
    {
      "entropy": 6.166766166687012,
      "epoch": 0.13928914505283382,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999789377339279,
      "loss": 5.9827,
      "mean_token_accuracy": 0.15575652718544006,
      "num_tokens": 3322247.0,
      "step": 1450
    },
    {
      "entropy": 6.165034675598145,
      "epoch": 0.13976945244956773,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999784660367915,
      "loss": 5.892,
      "mean_token_accuracy": 0.1582799434661865,
      "num_tokens": 3333369.0,
      "step": 1455
    },
    {
      "entropy": 6.110639047622681,
      "epoch": 0.14024975984630164,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999779891163231,
      "loss": 5.9384,
      "mean_token_accuracy": 0.15631188750267028,
      "num_tokens": 3345876.0,
      "step": 1460
    },
    {
      "entropy": 6.116653394699097,
      "epoch": 0.14073006724303555,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999775069725339,
      "loss": 5.8225,
      "mean_token_accuracy": 0.1637310653924942,
      "num_tokens": 3357323.0,
      "step": 1465
    },
    {
      "entropy": 6.105227518081665,
      "epoch": 0.14121037463976946,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499977019605435,
      "loss": 5.9005,
      "mean_token_accuracy": 0.160066157579422,
      "num_tokens": 3367689.0,
      "step": 1470
    },
    {
      "entropy": 5.980920314788818,
      "epoch": 0.14169068203650337,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999765270150378,
      "loss": 5.8722,
      "mean_token_accuracy": 0.16156274527311326,
      "num_tokens": 3379472.0,
      "step": 1475
    },
    {
      "entropy": 6.227704000473023,
      "epoch": 0.14217098943323728,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999760292013536,
      "loss": 5.9185,
      "mean_token_accuracy": 0.15907305255532264,
      "num_tokens": 3390929.0,
      "step": 1480
    },
    {
      "entropy": 6.240896797180175,
      "epoch": 0.14265129682997119,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999755261643941,
      "loss": 5.9174,
      "mean_token_accuracy": 0.16511183232069016,
      "num_tokens": 3401242.0,
      "step": 1485
    },
    {
      "entropy": 6.003513145446777,
      "epoch": 0.1431316042267051,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999750179041709,
      "loss": 5.9087,
      "mean_token_accuracy": 0.15776129364967345,
      "num_tokens": 3411169.0,
      "step": 1490
    },
    {
      "entropy": 6.05569019317627,
      "epoch": 0.143611911623439,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999745044206959,
      "loss": 5.7394,
      "mean_token_accuracy": 0.1670708954334259,
      "num_tokens": 3423265.0,
      "step": 1495
    },
    {
      "entropy": 6.043114852905274,
      "epoch": 0.1440922190201729,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999739857139809,
      "loss": 5.8494,
      "mean_token_accuracy": 0.1674238234758377,
      "num_tokens": 3434793.0,
      "step": 1500
    },
    {
      "entropy": 6.026560688018799,
      "epoch": 0.14457252641690682,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499973461784038,
      "loss": 5.7887,
      "mean_token_accuracy": 0.1741930678486824,
      "num_tokens": 3445732.0,
      "step": 1505
    },
    {
      "entropy": 6.085477066040039,
      "epoch": 0.14505283381364073,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999729326308792,
      "loss": 5.9612,
      "mean_token_accuracy": 0.15714176446199418,
      "num_tokens": 3457090.0,
      "step": 1510
    },
    {
      "entropy": 6.235566902160644,
      "epoch": 0.14553314121037464,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499972398254517,
      "loss": 5.9443,
      "mean_token_accuracy": 0.15230439007282257,
      "num_tokens": 3468087.0,
      "step": 1515
    },
    {
      "entropy": 6.1251365661621096,
      "epoch": 0.14601344860710855,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499971858654964,
      "loss": 5.8869,
      "mean_token_accuracy": 0.1600414589047432,
      "num_tokens": 3478820.0,
      "step": 1520
    },
    {
      "entropy": 6.028464698791504,
      "epoch": 0.14649375600384246,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999713138322321,
      "loss": 5.9044,
      "mean_token_accuracy": 0.15837728679180146,
      "num_tokens": 3489878.0,
      "step": 1525
    },
    {
      "entropy": 6.158157634735107,
      "epoch": 0.14697406340057637,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999707637863346,
      "loss": 5.8931,
      "mean_token_accuracy": 0.16006086468696595,
      "num_tokens": 3500944.0,
      "step": 1530
    },
    {
      "entropy": 6.039820289611816,
      "epoch": 0.14745437079731027,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999702085172838,
      "loss": 5.8805,
      "mean_token_accuracy": 0.16458703875541686,
      "num_tokens": 3511383.0,
      "step": 1535
    },
    {
      "entropy": 6.116418266296387,
      "epoch": 0.14793467819404418,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999696480250929,
      "loss": 5.9682,
      "mean_token_accuracy": 0.1588382303714752,
      "num_tokens": 3523300.0,
      "step": 1540
    },
    {
      "entropy": 6.157420682907104,
      "epoch": 0.1484149855907781,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999690823097747,
      "loss": 5.8922,
      "mean_token_accuracy": 0.15603983476758004,
      "num_tokens": 3534371.0,
      "step": 1545
    },
    {
      "entropy": 6.085301923751831,
      "epoch": 0.148895292987512,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999685113713426,
      "loss": 5.8566,
      "mean_token_accuracy": 0.1630222499370575,
      "num_tokens": 3544847.0,
      "step": 1550
    },
    {
      "entropy": 6.091690015792847,
      "epoch": 0.1493756003842459,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999679352098096,
      "loss": 5.8394,
      "mean_token_accuracy": 0.1639106124639511,
      "num_tokens": 3555859.0,
      "step": 1555
    },
    {
      "entropy": 6.032490921020508,
      "epoch": 0.14985590778097982,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999673538251891,
      "loss": 5.8622,
      "mean_token_accuracy": 0.16305326968431472,
      "num_tokens": 3568283.0,
      "step": 1560
    },
    {
      "entropy": 6.074030351638794,
      "epoch": 0.15033621517771373,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999667672174947,
      "loss": 5.9212,
      "mean_token_accuracy": 0.15798406153917313,
      "num_tokens": 3581442.0,
      "step": 1565
    },
    {
      "entropy": 6.221883392333984,
      "epoch": 0.15081652257444764,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049996617538674,
      "loss": 5.9677,
      "mean_token_accuracy": 0.15445063039660453,
      "num_tokens": 3594055.0,
      "step": 1570
    },
    {
      "entropy": 6.099120283126831,
      "epoch": 0.15129682997118155,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999655783329386,
      "loss": 5.9341,
      "mean_token_accuracy": 0.1539287731051445,
      "num_tokens": 3605952.0,
      "step": 1575
    },
    {
      "entropy": 6.155495500564575,
      "epoch": 0.15177713736791545,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999649760561046,
      "loss": 5.958,
      "mean_token_accuracy": 0.1572701022028923,
      "num_tokens": 3618544.0,
      "step": 1580
    },
    {
      "entropy": 6.109302043914795,
      "epoch": 0.15225744476464936,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999643685562519,
      "loss": 5.8982,
      "mean_token_accuracy": 0.1646600529551506,
      "num_tokens": 3630445.0,
      "step": 1585
    },
    {
      "entropy": 6.1016254901885985,
      "epoch": 0.15273775216138327,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999637558333945,
      "loss": 5.9039,
      "mean_token_accuracy": 0.16291006058454513,
      "num_tokens": 3642516.0,
      "step": 1590
    },
    {
      "entropy": 6.074013328552246,
      "epoch": 0.15321805955811718,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999631378875467,
      "loss": 5.8209,
      "mean_token_accuracy": 0.16821238100528718,
      "num_tokens": 3654425.0,
      "step": 1595
    },
    {
      "entropy": 5.979668712615966,
      "epoch": 0.15369836695485112,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999625147187228,
      "loss": 5.8486,
      "mean_token_accuracy": 0.1578590378165245,
      "num_tokens": 3666521.0,
      "step": 1600
    },
    {
      "entropy": 6.134301042556762,
      "epoch": 0.15417867435158503,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999618863269373,
      "loss": 5.8947,
      "mean_token_accuracy": 0.15352863371372222,
      "num_tokens": 3679121.0,
      "step": 1605
    },
    {
      "entropy": 6.100239419937134,
      "epoch": 0.15465898174831894,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999612527122049,
      "loss": 5.894,
      "mean_token_accuracy": 0.15724605321884155,
      "num_tokens": 3691095.0,
      "step": 1610
    },
    {
      "entropy": 6.060277366638184,
      "epoch": 0.15513928914505284,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999606138745402,
      "loss": 5.8731,
      "mean_token_accuracy": 0.1618647649884224,
      "num_tokens": 3703426.0,
      "step": 1615
    },
    {
      "entropy": 6.1450238704681395,
      "epoch": 0.15561959654178675,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999599698139581,
      "loss": 5.9336,
      "mean_token_accuracy": 0.16260550543665886,
      "num_tokens": 3715429.0,
      "step": 1620
    },
    {
      "entropy": 6.146528911590576,
      "epoch": 0.15609990393852066,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999593205304734,
      "loss": 5.9163,
      "mean_token_accuracy": 0.15805355310440064,
      "num_tokens": 3726327.0,
      "step": 1625
    },
    {
      "entropy": 6.100754690170288,
      "epoch": 0.15658021133525457,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999586660241012,
      "loss": 5.8732,
      "mean_token_accuracy": 0.1542663834989071,
      "num_tokens": 3736818.0,
      "step": 1630
    },
    {
      "entropy": 6.088619899749756,
      "epoch": 0.15706051873198848,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999580062948569,
      "loss": 5.8789,
      "mean_token_accuracy": 0.16102785766124725,
      "num_tokens": 3747776.0,
      "step": 1635
    },
    {
      "entropy": 5.949653625488281,
      "epoch": 0.1575408261287224,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999573413427556,
      "loss": 5.7321,
      "mean_token_accuracy": 0.16540554612874986,
      "num_tokens": 3758990.0,
      "step": 1640
    },
    {
      "entropy": 6.039187049865722,
      "epoch": 0.1580211335254563,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999566711678128,
      "loss": 5.8119,
      "mean_token_accuracy": 0.16214931458234788,
      "num_tokens": 3769686.0,
      "step": 1645
    },
    {
      "entropy": 6.065354490280152,
      "epoch": 0.1585014409221902,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999559957700442,
      "loss": 5.8824,
      "mean_token_accuracy": 0.1635928988456726,
      "num_tokens": 3781815.0,
      "step": 1650
    },
    {
      "entropy": 6.073367595672607,
      "epoch": 0.15898174831892412,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999553151494653,
      "loss": 5.9146,
      "mean_token_accuracy": 0.16144528090953827,
      "num_tokens": 3793392.0,
      "step": 1655
    },
    {
      "entropy": 6.073708391189575,
      "epoch": 0.15946205571565802,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999546293060919,
      "loss": 5.8568,
      "mean_token_accuracy": 0.1598874419927597,
      "num_tokens": 3804974.0,
      "step": 1660
    },
    {
      "entropy": 5.9182981014251705,
      "epoch": 0.15994236311239193,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00049995393823994,
      "loss": 5.7125,
      "mean_token_accuracy": 0.17333305180072783,
      "num_tokens": 3817166.0,
      "step": 1665
    },
    {
      "entropy": 6.048591375350952,
      "epoch": 0.16042267050912584,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999532419510255,
      "loss": 5.8338,
      "mean_token_accuracy": 0.16252198666334153,
      "num_tokens": 3828151.0,
      "step": 1670
    },
    {
      "entropy": 6.027536058425904,
      "epoch": 0.16090297790585975,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499952540439365,
      "loss": 5.8397,
      "mean_token_accuracy": 0.16017384976148605,
      "num_tokens": 3839439.0,
      "step": 1675
    },
    {
      "entropy": 6.0826294898986815,
      "epoch": 0.16138328530259366,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999518337049743,
      "loss": 5.8916,
      "mean_token_accuracy": 0.1598974958062172,
      "num_tokens": 3851694.0,
      "step": 1680
    },
    {
      "entropy": 6.004607820510865,
      "epoch": 0.16186359269932757,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00049995112174787,
      "loss": 5.8737,
      "mean_token_accuracy": 0.16189273744821547,
      "num_tokens": 3863593.0,
      "step": 1685
    },
    {
      "entropy": 5.980313777923584,
      "epoch": 0.16234390009606148,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999504045680687,
      "loss": 5.7892,
      "mean_token_accuracy": 0.169136680662632,
      "num_tokens": 3874588.0,
      "step": 1690
    },
    {
      "entropy": 6.042408800125122,
      "epoch": 0.1628242074927954,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999496821655869,
      "loss": 5.8829,
      "mean_token_accuracy": 0.16141626685857774,
      "num_tokens": 3884662.0,
      "step": 1695
    },
    {
      "entropy": 6.158259296417237,
      "epoch": 0.1633045148895293,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999489545404414,
      "loss": 5.9778,
      "mean_token_accuracy": 0.15626356303691863,
      "num_tokens": 3896569.0,
      "step": 1700
    },
    {
      "entropy": 6.13565845489502,
      "epoch": 0.1637848222862632,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999482216926493,
      "loss": 5.8183,
      "mean_token_accuracy": 0.16687567681074142,
      "num_tokens": 3907691.0,
      "step": 1705
    },
    {
      "entropy": 6.031500291824341,
      "epoch": 0.1642651296829971,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999474836222273,
      "loss": 5.8433,
      "mean_token_accuracy": 0.1687850758433342,
      "num_tokens": 3918794.0,
      "step": 1710
    },
    {
      "entropy": 6.0354866027832035,
      "epoch": 0.16474543707973102,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999467403291928,
      "loss": 5.8302,
      "mean_token_accuracy": 0.17379950284957885,
      "num_tokens": 3929773.0,
      "step": 1715
    },
    {
      "entropy": 6.017301607131958,
      "epoch": 0.16522574447646493,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999459918135628,
      "loss": 5.8537,
      "mean_token_accuracy": 0.16206772923469542,
      "num_tokens": 3940264.0,
      "step": 1720
    },
    {
      "entropy": 5.961511898040771,
      "epoch": 0.16570605187319884,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499945238075355,
      "loss": 5.7453,
      "mean_token_accuracy": 0.16794898062944413,
      "num_tokens": 3951500.0,
      "step": 1725
    },
    {
      "entropy": 5.9861691951751705,
      "epoch": 0.16618635926993275,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999444791145865,
      "loss": 5.8345,
      "mean_token_accuracy": 0.16556786149740219,
      "num_tokens": 3963580.0,
      "step": 1730
    },
    {
      "entropy": 5.995738363265991,
      "epoch": 0.16666666666666666,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999437149312754,
      "loss": 5.7682,
      "mean_token_accuracy": 0.17584467083215713,
      "num_tokens": 3975994.0,
      "step": 1735
    },
    {
      "entropy": 6.007660579681397,
      "epoch": 0.16714697406340057,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499942945525439,
      "loss": 5.7844,
      "mean_token_accuracy": 0.1600303217768669,
      "num_tokens": 3987897.0,
      "step": 1740
    },
    {
      "entropy": 6.043190240859985,
      "epoch": 0.16762728146013448,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999421708970954,
      "loss": 5.924,
      "mean_token_accuracy": 0.15594586208462716,
      "num_tokens": 3999829.0,
      "step": 1745
    },
    {
      "entropy": 5.972902059555054,
      "epoch": 0.16810758885686838,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999413910462625,
      "loss": 5.7781,
      "mean_token_accuracy": 0.16753159314393998,
      "num_tokens": 4010882.0,
      "step": 1750
    },
    {
      "entropy": 5.9610387802124025,
      "epoch": 0.1685878962536023,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999406059729586,
      "loss": 5.756,
      "mean_token_accuracy": 0.1666981115937233,
      "num_tokens": 4021423.0,
      "step": 1755
    },
    {
      "entropy": 6.116128730773926,
      "epoch": 0.1690682036503362,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999398156772016,
      "loss": 5.894,
      "mean_token_accuracy": 0.15112203359603882,
      "num_tokens": 4033590.0,
      "step": 1760
    },
    {
      "entropy": 5.959614181518555,
      "epoch": 0.16954851104707014,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049993902015901,
      "loss": 5.7608,
      "mean_token_accuracy": 0.17025730162858962,
      "num_tokens": 4043978.0,
      "step": 1765
    },
    {
      "entropy": 6.023910427093506,
      "epoch": 0.17002881844380405,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999382194184023,
      "loss": 5.8956,
      "mean_token_accuracy": 0.1614847645163536,
      "num_tokens": 4054513.0,
      "step": 1770
    },
    {
      "entropy": 6.124105310440063,
      "epoch": 0.17050912584053796,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999374134553972,
      "loss": 5.8399,
      "mean_token_accuracy": 0.16457433849573136,
      "num_tokens": 4066019.0,
      "step": 1775
    },
    {
      "entropy": 5.968501949310303,
      "epoch": 0.17098943323727187,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999366022700131,
      "loss": 5.7905,
      "mean_token_accuracy": 0.16719537675380708,
      "num_tokens": 4077688.0,
      "step": 1780
    },
    {
      "entropy": 6.039582633972168,
      "epoch": 0.17146974063400577,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999357858622691,
      "loss": 5.8791,
      "mean_token_accuracy": 0.16837128698825837,
      "num_tokens": 4089803.0,
      "step": 1785
    },
    {
      "entropy": 6.028085994720459,
      "epoch": 0.17195004803073968,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999349642321842,
      "loss": 5.8319,
      "mean_token_accuracy": 0.16663650572299957,
      "num_tokens": 4101969.0,
      "step": 1790
    },
    {
      "entropy": 6.029037094116211,
      "epoch": 0.1724303554274736,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999341373797772,
      "loss": 5.808,
      "mean_token_accuracy": 0.1627863183617592,
      "num_tokens": 4113567.0,
      "step": 1795
    },
    {
      "entropy": 5.954306650161743,
      "epoch": 0.1729106628242075,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999333053050675,
      "loss": 5.7683,
      "mean_token_accuracy": 0.16414574086666106,
      "num_tokens": 4125191.0,
      "step": 1800
    },
    {
      "entropy": 6.0340282917022705,
      "epoch": 0.1733909702209414,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999324680080744,
      "loss": 5.8148,
      "mean_token_accuracy": 0.16474771052598952,
      "num_tokens": 4135050.0,
      "step": 1805
    },
    {
      "entropy": 5.960515642166138,
      "epoch": 0.17387127761767532,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999316254888172,
      "loss": 5.8762,
      "mean_token_accuracy": 0.16312166899442673,
      "num_tokens": 4146874.0,
      "step": 1810
    },
    {
      "entropy": 5.994801616668701,
      "epoch": 0.17435158501440923,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999307777473157,
      "loss": 5.7959,
      "mean_token_accuracy": 0.16597876250743865,
      "num_tokens": 4158118.0,
      "step": 1815
    },
    {
      "entropy": 6.025922155380249,
      "epoch": 0.17483189241114314,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999299247835893,
      "loss": 5.778,
      "mean_token_accuracy": 0.17315936088562012,
      "num_tokens": 4169035.0,
      "step": 1820
    },
    {
      "entropy": 5.942285585403442,
      "epoch": 0.17531219980787704,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499929066597658,
      "loss": 5.7655,
      "mean_token_accuracy": 0.17127336263656617,
      "num_tokens": 4180314.0,
      "step": 1825
    },
    {
      "entropy": 6.031781435012817,
      "epoch": 0.17579250720461095,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999282031895418,
      "loss": 5.8419,
      "mean_token_accuracy": 0.16611838191747666,
      "num_tokens": 4192238.0,
      "step": 1830
    },
    {
      "entropy": 5.9649334907531735,
      "epoch": 0.17627281460134486,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999273345592604,
      "loss": 5.7689,
      "mean_token_accuracy": 0.16682324707508087,
      "num_tokens": 4203346.0,
      "step": 1835
    },
    {
      "entropy": 5.9124401092529295,
      "epoch": 0.17675312199807877,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999264607068343,
      "loss": 5.8301,
      "mean_token_accuracy": 0.16939765363931655,
      "num_tokens": 4213763.0,
      "step": 1840
    },
    {
      "entropy": 5.962861585617065,
      "epoch": 0.17723342939481268,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999255816322837,
      "loss": 5.7927,
      "mean_token_accuracy": 0.16723354011774064,
      "num_tokens": 4225553.0,
      "step": 1845
    },
    {
      "entropy": 6.043817090988159,
      "epoch": 0.1777137367915466,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499924697335629,
      "loss": 5.6976,
      "mean_token_accuracy": 0.17336428612470628,
      "num_tokens": 4236058.0,
      "step": 1850
    },
    {
      "entropy": 5.887497758865356,
      "epoch": 0.1781940441882805,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999238078168906,
      "loss": 5.7904,
      "mean_token_accuracy": 0.17044568955898284,
      "num_tokens": 4248299.0,
      "step": 1855
    },
    {
      "entropy": 5.978550386428833,
      "epoch": 0.1786743515850144,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999229130760894,
      "loss": 5.7232,
      "mean_token_accuracy": 0.17197388857603074,
      "num_tokens": 4259704.0,
      "step": 1860
    },
    {
      "entropy": 5.893118095397949,
      "epoch": 0.17915465898174832,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499922013113246,
      "loss": 5.607,
      "mean_token_accuracy": 0.18167430013418198,
      "num_tokens": 4270480.0,
      "step": 1865
    },
    {
      "entropy": 5.973291683197021,
      "epoch": 0.17963496637848222,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999211079283814,
      "loss": 5.8578,
      "mean_token_accuracy": 0.16774737685918809,
      "num_tokens": 4282104.0,
      "step": 1870
    },
    {
      "entropy": 6.07144513130188,
      "epoch": 0.18011527377521613,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999201975215164,
      "loss": 5.8297,
      "mean_token_accuracy": 0.16095103695988655,
      "num_tokens": 4294251.0,
      "step": 1875
    },
    {
      "entropy": 5.909284830093384,
      "epoch": 0.18059558117195004,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999192818926725,
      "loss": 5.7078,
      "mean_token_accuracy": 0.1667791336774826,
      "num_tokens": 4305569.0,
      "step": 1880
    },
    {
      "entropy": 6.000306129455566,
      "epoch": 0.18107588856868395,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999183610418706,
      "loss": 5.8418,
      "mean_token_accuracy": 0.16189501956105232,
      "num_tokens": 4317845.0,
      "step": 1885
    },
    {
      "entropy": 5.922011661529541,
      "epoch": 0.18155619596541786,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999174349691322,
      "loss": 5.7119,
      "mean_token_accuracy": 0.1709301382303238,
      "num_tokens": 4329987.0,
      "step": 1890
    },
    {
      "entropy": 5.898713827133179,
      "epoch": 0.18203650336215177,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999165036744788,
      "loss": 5.7182,
      "mean_token_accuracy": 0.17281702160835266,
      "num_tokens": 4341628.0,
      "step": 1895
    },
    {
      "entropy": 5.981558704376221,
      "epoch": 0.18251681075888568,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999155671579322,
      "loss": 5.7952,
      "mean_token_accuracy": 0.16310359686613082,
      "num_tokens": 4352379.0,
      "step": 1900
    },
    {
      "entropy": 5.8705666065216064,
      "epoch": 0.1829971181556196,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499914625419514,
      "loss": 5.7502,
      "mean_token_accuracy": 0.16969154328107833,
      "num_tokens": 4364800.0,
      "step": 1905
    },
    {
      "entropy": 5.977327680587768,
      "epoch": 0.1834774255523535,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999136784592459,
      "loss": 5.7488,
      "mean_token_accuracy": 0.16629694402217865,
      "num_tokens": 4376048.0,
      "step": 1910
    },
    {
      "entropy": 5.91033673286438,
      "epoch": 0.1839577329490874,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999127262771502,
      "loss": 5.7375,
      "mean_token_accuracy": 0.16977874636650087,
      "num_tokens": 4388072.0,
      "step": 1915
    },
    {
      "entropy": 6.031687164306641,
      "epoch": 0.1844380403458213,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999117688732487,
      "loss": 5.8366,
      "mean_token_accuracy": 0.16268031895160676,
      "num_tokens": 4399843.0,
      "step": 1920
    },
    {
      "entropy": 5.8528807163238525,
      "epoch": 0.18491834774255522,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999108062475638,
      "loss": 5.6895,
      "mean_token_accuracy": 0.1747806504368782,
      "num_tokens": 4411373.0,
      "step": 1925
    },
    {
      "entropy": 5.9012157917022705,
      "epoch": 0.18539865513928913,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499909838400118,
      "loss": 5.6783,
      "mean_token_accuracy": 0.17514891028404236,
      "num_tokens": 4421857.0,
      "step": 1930
    },
    {
      "entropy": 5.992180585861206,
      "epoch": 0.18587896253602307,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999088653309334,
      "loss": 5.7655,
      "mean_token_accuracy": 0.17122769355773926,
      "num_tokens": 4432728.0,
      "step": 1935
    },
    {
      "entropy": 5.874881219863892,
      "epoch": 0.18635926993275698,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999078870400329,
      "loss": 5.7109,
      "mean_token_accuracy": 0.1737685203552246,
      "num_tokens": 4444683.0,
      "step": 1940
    },
    {
      "entropy": 6.048246431350708,
      "epoch": 0.18683957732949089,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999069035274391,
      "loss": 5.8239,
      "mean_token_accuracy": 0.16185275614261627,
      "num_tokens": 4456961.0,
      "step": 1945
    },
    {
      "entropy": 5.890452718734741,
      "epoch": 0.1873198847262248,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999059147931747,
      "loss": 5.669,
      "mean_token_accuracy": 0.1743285223841667,
      "num_tokens": 4468424.0,
      "step": 1950
    },
    {
      "entropy": 5.918470287322998,
      "epoch": 0.1878001921229587,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999049208372629,
      "loss": 5.87,
      "mean_token_accuracy": 0.15737151950597764,
      "num_tokens": 4479813.0,
      "step": 1955
    },
    {
      "entropy": 6.117772960662842,
      "epoch": 0.1882804995196926,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999039216597267,
      "loss": 5.8843,
      "mean_token_accuracy": 0.16440917253494264,
      "num_tokens": 4491172.0,
      "step": 1960
    },
    {
      "entropy": 5.863306999206543,
      "epoch": 0.18876080691642652,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999029172605892,
      "loss": 5.7603,
      "mean_token_accuracy": 0.17048608511686325,
      "num_tokens": 4503063.0,
      "step": 1965
    },
    {
      "entropy": 6.016164255142212,
      "epoch": 0.18924111431316043,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999019076398738,
      "loss": 5.8321,
      "mean_token_accuracy": 0.15428075492382048,
      "num_tokens": 4514188.0,
      "step": 1970
    },
    {
      "entropy": 5.956648588180542,
      "epoch": 0.18972142170989434,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499900892797604,
      "loss": 5.7373,
      "mean_token_accuracy": 0.17256650179624558,
      "num_tokens": 4525293.0,
      "step": 1975
    },
    {
      "entropy": 5.915363073348999,
      "epoch": 0.19020172910662825,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998998727338031,
      "loss": 5.8306,
      "mean_token_accuracy": 0.16908561289310456,
      "num_tokens": 4536589.0,
      "step": 1980
    },
    {
      "entropy": 5.890870523452759,
      "epoch": 0.19068203650336216,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998988474484952,
      "loss": 5.5742,
      "mean_token_accuracy": 0.19122034162282944,
      "num_tokens": 4547594.0,
      "step": 1985
    },
    {
      "entropy": 5.90146050453186,
      "epoch": 0.19116234390009607,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998978169417038,
      "loss": 5.7832,
      "mean_token_accuracy": 0.17002302259206772,
      "num_tokens": 4559850.0,
      "step": 1990
    },
    {
      "entropy": 5.936874294281006,
      "epoch": 0.19164265129682997,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998967812134529,
      "loss": 5.7186,
      "mean_token_accuracy": 0.16649213433265686,
      "num_tokens": 4570727.0,
      "step": 1995
    },
    {
      "entropy": 5.786225271224976,
      "epoch": 0.19212295869356388,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998957402637664,
      "loss": 5.6629,
      "mean_token_accuracy": 0.17317176163196563,
      "num_tokens": 4582248.0,
      "step": 2000
    },
    {
      "entropy": 5.951917409896851,
      "epoch": 0.1926032660902978,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998946940926687,
      "loss": 5.6879,
      "mean_token_accuracy": 0.17081644088029863,
      "num_tokens": 4592604.0,
      "step": 2005
    },
    {
      "entropy": 5.852265024185181,
      "epoch": 0.1930835734870317,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499893642700184,
      "loss": 5.7288,
      "mean_token_accuracy": 0.16796955615282058,
      "num_tokens": 4604398.0,
      "step": 2010
    },
    {
      "entropy": 5.932278966903686,
      "epoch": 0.1935638808837656,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998925860863368,
      "loss": 5.805,
      "mean_token_accuracy": 0.16799372732639312,
      "num_tokens": 4616434.0,
      "step": 2015
    },
    {
      "entropy": 5.984751844406128,
      "epoch": 0.19404418828049952,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998915242511516,
      "loss": 5.7565,
      "mean_token_accuracy": 0.17852777987718582,
      "num_tokens": 4627577.0,
      "step": 2020
    },
    {
      "entropy": 5.979996967315674,
      "epoch": 0.19452449567723343,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998904571946528,
      "loss": 5.8059,
      "mean_token_accuracy": 0.16536686867475509,
      "num_tokens": 4639698.0,
      "step": 2025
    },
    {
      "entropy": 5.992480897903443,
      "epoch": 0.19500480307396734,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998893849168655,
      "loss": 5.8193,
      "mean_token_accuracy": 0.16457681208848954,
      "num_tokens": 4650643.0,
      "step": 2030
    },
    {
      "entropy": 5.908215618133545,
      "epoch": 0.19548511047070125,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998883074178144,
      "loss": 5.7539,
      "mean_token_accuracy": 0.16797305792570114,
      "num_tokens": 4662897.0,
      "step": 2035
    },
    {
      "entropy": 5.982371234893799,
      "epoch": 0.19596541786743515,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998872246975247,
      "loss": 5.8354,
      "mean_token_accuracy": 0.17146392464637755,
      "num_tokens": 4673701.0,
      "step": 2040
    },
    {
      "entropy": 6.009561920166016,
      "epoch": 0.19644572526416906,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998861367560213,
      "loss": 5.7898,
      "mean_token_accuracy": 0.16622848212718963,
      "num_tokens": 4685873.0,
      "step": 2045
    },
    {
      "entropy": 5.924004793167114,
      "epoch": 0.19692603266090297,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998850435933296,
      "loss": 5.6709,
      "mean_token_accuracy": 0.1732612207531929,
      "num_tokens": 4697179.0,
      "step": 2050
    },
    {
      "entropy": 5.862003564834595,
      "epoch": 0.19740634005763688,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998839452094749,
      "loss": 5.7114,
      "mean_token_accuracy": 0.17544772624969482,
      "num_tokens": 4707752.0,
      "step": 2055
    },
    {
      "entropy": 5.796081399917602,
      "epoch": 0.1978866474543708,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998828416044829,
      "loss": 5.5796,
      "mean_token_accuracy": 0.17987633645534515,
      "num_tokens": 4718413.0,
      "step": 2060
    },
    {
      "entropy": 5.9167498588562015,
      "epoch": 0.1983669548511047,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499881732778379,
      "loss": 5.7688,
      "mean_token_accuracy": 0.16329588294029235,
      "num_tokens": 4730033.0,
      "step": 2065
    },
    {
      "entropy": 5.891122531890869,
      "epoch": 0.1988472622478386,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499880618731189,
      "loss": 5.6187,
      "mean_token_accuracy": 0.17821569591760636,
      "num_tokens": 4742084.0,
      "step": 2070
    },
    {
      "entropy": 5.9228325366973875,
      "epoch": 0.19932756964457252,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998794994629388,
      "loss": 5.8539,
      "mean_token_accuracy": 0.16390760689973832,
      "num_tokens": 4753885.0,
      "step": 2075
    },
    {
      "entropy": 5.88261046409607,
      "epoch": 0.19980787704130643,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998783749736545,
      "loss": 5.6934,
      "mean_token_accuracy": 0.1716991052031517,
      "num_tokens": 4765686.0,
      "step": 2080
    },
    {
      "entropy": 5.954660081863404,
      "epoch": 0.20028818443804033,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998772452633619,
      "loss": 5.7332,
      "mean_token_accuracy": 0.1718806430697441,
      "num_tokens": 4777157.0,
      "step": 2085
    },
    {
      "entropy": 5.845113325119018,
      "epoch": 0.20076849183477424,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998761103320876,
      "loss": 5.6655,
      "mean_token_accuracy": 0.17630862295627595,
      "num_tokens": 4788583.0,
      "step": 2090
    },
    {
      "entropy": 5.971330165863037,
      "epoch": 0.20124879923150815,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998749701798577,
      "loss": 5.7852,
      "mean_token_accuracy": 0.16313367635011672,
      "num_tokens": 4800749.0,
      "step": 2095
    },
    {
      "entropy": 5.862610292434693,
      "epoch": 0.2017291066282421,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998738248066986,
      "loss": 5.6986,
      "mean_token_accuracy": 0.1719326362013817,
      "num_tokens": 4812488.0,
      "step": 2100
    },
    {
      "entropy": 5.943162345886231,
      "epoch": 0.202209414024976,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998726742126372,
      "loss": 5.6826,
      "mean_token_accuracy": 0.17237975001335143,
      "num_tokens": 4823495.0,
      "step": 2105
    },
    {
      "entropy": 5.845049905776977,
      "epoch": 0.2026897214217099,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998715183976999,
      "loss": 5.7416,
      "mean_token_accuracy": 0.16694231554865838,
      "num_tokens": 4834450.0,
      "step": 2110
    },
    {
      "entropy": 5.872321510314942,
      "epoch": 0.20317002881844382,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998703573619137,
      "loss": 5.6479,
      "mean_token_accuracy": 0.17984378933906556,
      "num_tokens": 4846826.0,
      "step": 2115
    },
    {
      "entropy": 5.981400775909424,
      "epoch": 0.20365033621517772,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998691911053056,
      "loss": 5.838,
      "mean_token_accuracy": 0.15721772760152816,
      "num_tokens": 4859668.0,
      "step": 2120
    },
    {
      "entropy": 5.889471960067749,
      "epoch": 0.20413064361191163,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998680196279026,
      "loss": 5.7188,
      "mean_token_accuracy": 0.17116449922323226,
      "num_tokens": 4871727.0,
      "step": 2125
    },
    {
      "entropy": 5.965567874908447,
      "epoch": 0.20461095100864554,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998668429297319,
      "loss": 5.7581,
      "mean_token_accuracy": 0.17664661183953284,
      "num_tokens": 4882191.0,
      "step": 2130
    },
    {
      "entropy": 5.91290979385376,
      "epoch": 0.20509125840537945,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998656610108208,
      "loss": 5.7024,
      "mean_token_accuracy": 0.17135777622461318,
      "num_tokens": 4892416.0,
      "step": 2135
    },
    {
      "entropy": 5.7849208354949955,
      "epoch": 0.20557156580211336,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998644738711969,
      "loss": 5.6744,
      "mean_token_accuracy": 0.1663479372859001,
      "num_tokens": 4903572.0,
      "step": 2140
    },
    {
      "entropy": 6.012947654724121,
      "epoch": 0.20605187319884727,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004998632815108874,
      "loss": 5.766,
      "mean_token_accuracy": 0.16735016852617263,
      "num_tokens": 4915417.0,
      "step": 2145
    },
    {
      "entropy": 5.828745889663696,
      "epoch": 0.20653218059558118,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998620839299203,
      "loss": 5.6506,
      "mean_token_accuracy": 0.1731554314494133,
      "num_tokens": 4926943.0,
      "step": 2150
    },
    {
      "entropy": 5.879217290878296,
      "epoch": 0.2070124879923151,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998608811283233,
      "loss": 5.612,
      "mean_token_accuracy": 0.17777990251779557,
      "num_tokens": 4937724.0,
      "step": 2155
    },
    {
      "entropy": 5.948641300201416,
      "epoch": 0.207492795389049,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998596731061244,
      "loss": 5.7627,
      "mean_token_accuracy": 0.16541688293218612,
      "num_tokens": 4949970.0,
      "step": 2160
    },
    {
      "entropy": 5.916024541854858,
      "epoch": 0.2079731027857829,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998584598633516,
      "loss": 5.7691,
      "mean_token_accuracy": 0.1677638292312622,
      "num_tokens": 4961389.0,
      "step": 2165
    },
    {
      "entropy": 5.9447978973388675,
      "epoch": 0.2084534101825168,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998572414000329,
      "loss": 5.8134,
      "mean_token_accuracy": 0.16801437735557556,
      "num_tokens": 4973888.0,
      "step": 2170
    },
    {
      "entropy": 5.877821445465088,
      "epoch": 0.20893371757925072,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998560177161969,
      "loss": 5.7626,
      "mean_token_accuracy": 0.15845638066530227,
      "num_tokens": 4985423.0,
      "step": 2175
    },
    {
      "entropy": 5.847891712188721,
      "epoch": 0.20941402497598463,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998547888118718,
      "loss": 5.7327,
      "mean_token_accuracy": 0.1661007210612297,
      "num_tokens": 4997711.0,
      "step": 2180
    },
    {
      "entropy": 5.9377055168151855,
      "epoch": 0.20989433237271854,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998535546870862,
      "loss": 5.7449,
      "mean_token_accuracy": 0.17071173936128617,
      "num_tokens": 5009633.0,
      "step": 2185
    },
    {
      "entropy": 5.87849235534668,
      "epoch": 0.21037463976945245,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998523153418687,
      "loss": 5.6893,
      "mean_token_accuracy": 0.17723384648561477,
      "num_tokens": 5021523.0,
      "step": 2190
    },
    {
      "entropy": 5.969219064712524,
      "epoch": 0.21085494716618636,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998510707762481,
      "loss": 5.7792,
      "mean_token_accuracy": 0.1698889523744583,
      "num_tokens": 5033513.0,
      "step": 2195
    },
    {
      "entropy": 6.002408790588379,
      "epoch": 0.21133525456292027,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998498209902533,
      "loss": 5.7803,
      "mean_token_accuracy": 0.16734371781349183,
      "num_tokens": 5047055.0,
      "step": 2200
    },
    {
      "entropy": 5.902544927597046,
      "epoch": 0.21181556195965417,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998485659839134,
      "loss": 5.6634,
      "mean_token_accuracy": 0.171344156563282,
      "num_tokens": 5057613.0,
      "step": 2205
    },
    {
      "entropy": 5.888894319534302,
      "epoch": 0.21229586935638808,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998473057572575,
      "loss": 5.7793,
      "mean_token_accuracy": 0.16772594749927522,
      "num_tokens": 5068886.0,
      "step": 2210
    },
    {
      "entropy": 5.889117670059204,
      "epoch": 0.212776176753122,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998460403103146,
      "loss": 5.7516,
      "mean_token_accuracy": 0.1667829394340515,
      "num_tokens": 5079978.0,
      "step": 2215
    },
    {
      "entropy": 5.892506170272827,
      "epoch": 0.2132564841498559,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998447696431146,
      "loss": 5.7088,
      "mean_token_accuracy": 0.1731807380914688,
      "num_tokens": 5091021.0,
      "step": 2220
    },
    {
      "entropy": 5.812002897262573,
      "epoch": 0.2137367915465898,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998434937556865,
      "loss": 5.6145,
      "mean_token_accuracy": 0.1779699981212616,
      "num_tokens": 5101483.0,
      "step": 2225
    },
    {
      "entropy": 5.916818428039551,
      "epoch": 0.21421709894332372,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998422126480602,
      "loss": 5.7339,
      "mean_token_accuracy": 0.16566304564476014,
      "num_tokens": 5113116.0,
      "step": 2230
    },
    {
      "entropy": 5.918116569519043,
      "epoch": 0.21469740634005763,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998409263202653,
      "loss": 5.6905,
      "mean_token_accuracy": 0.17165909111499786,
      "num_tokens": 5124824.0,
      "step": 2235
    },
    {
      "entropy": 5.821660041809082,
      "epoch": 0.21517771373679154,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998396347723318,
      "loss": 5.6186,
      "mean_token_accuracy": 0.16778021901845933,
      "num_tokens": 5137567.0,
      "step": 2240
    },
    {
      "entropy": 5.834303140640259,
      "epoch": 0.21565802113352545,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998383380042895,
      "loss": 5.6873,
      "mean_token_accuracy": 0.1687231481075287,
      "num_tokens": 5149016.0,
      "step": 2245
    },
    {
      "entropy": 5.800395917892456,
      "epoch": 0.21613832853025935,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998370360161688,
      "loss": 5.5679,
      "mean_token_accuracy": 0.177195705473423,
      "num_tokens": 5160356.0,
      "step": 2250
    },
    {
      "entropy": 5.877714490890503,
      "epoch": 0.21661863592699326,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998357288079996,
      "loss": 5.7659,
      "mean_token_accuracy": 0.1638329356908798,
      "num_tokens": 5172100.0,
      "step": 2255
    },
    {
      "entropy": 5.947152042388916,
      "epoch": 0.21709894332372717,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998344163798125,
      "loss": 5.7387,
      "mean_token_accuracy": 0.1652805581688881,
      "num_tokens": 5183984.0,
      "step": 2260
    },
    {
      "entropy": 5.8969450950622555,
      "epoch": 0.21757925072046108,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004998330987316379,
      "loss": 5.7033,
      "mean_token_accuracy": 0.16846372336149215,
      "num_tokens": 5195853.0,
      "step": 2265
    },
    {
      "entropy": 5.786798810958862,
      "epoch": 0.21805955811719502,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998317758635062,
      "loss": 5.5582,
      "mean_token_accuracy": 0.17903037667274474,
      "num_tokens": 5206995.0,
      "step": 2270
    },
    {
      "entropy": 5.688874101638794,
      "epoch": 0.21853986551392893,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998304477754484,
      "loss": 5.5976,
      "mean_token_accuracy": 0.1773568794131279,
      "num_tokens": 5219291.0,
      "step": 2275
    },
    {
      "entropy": 5.856671667098999,
      "epoch": 0.21902017291066284,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998291144674952,
      "loss": 5.6779,
      "mean_token_accuracy": 0.1779756113886833,
      "num_tokens": 5230856.0,
      "step": 2280
    },
    {
      "entropy": 5.781990480422974,
      "epoch": 0.21950048030739674,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998277759396776,
      "loss": 5.5337,
      "mean_token_accuracy": 0.18482960611581803,
      "num_tokens": 5242871.0,
      "step": 2285
    },
    {
      "entropy": 5.7626872062683105,
      "epoch": 0.21998078770413065,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998264321920265,
      "loss": 5.6465,
      "mean_token_accuracy": 0.18032118529081345,
      "num_tokens": 5253835.0,
      "step": 2290
    },
    {
      "entropy": 5.782493829727173,
      "epoch": 0.22046109510086456,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004998250832245734,
      "loss": 5.6194,
      "mean_token_accuracy": 0.17594331204891206,
      "num_tokens": 5266195.0,
      "step": 2295
    },
    {
      "entropy": 5.801488208770752,
      "epoch": 0.22094140249759847,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998237290373494,
      "loss": 5.5986,
      "mean_token_accuracy": 0.17856477349996566,
      "num_tokens": 5277499.0,
      "step": 2300
    },
    {
      "entropy": 5.931110095977783,
      "epoch": 0.22142170989433238,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499822369630386,
      "loss": 5.706,
      "mean_token_accuracy": 0.16192054152488708,
      "num_tokens": 5288622.0,
      "step": 2305
    },
    {
      "entropy": 5.866655731201172,
      "epoch": 0.2219020172910663,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998210050037148,
      "loss": 5.7657,
      "mean_token_accuracy": 0.16639903783798218,
      "num_tokens": 5299664.0,
      "step": 2310
    },
    {
      "entropy": 5.874531269073486,
      "epoch": 0.2223823246878002,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998196351573674,
      "loss": 5.6538,
      "mean_token_accuracy": 0.17477090954780578,
      "num_tokens": 5311627.0,
      "step": 2315
    },
    {
      "entropy": 5.721544981002808,
      "epoch": 0.2228626320845341,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998182600913757,
      "loss": 5.555,
      "mean_token_accuracy": 0.18625836670398713,
      "num_tokens": 5323000.0,
      "step": 2320
    },
    {
      "entropy": 5.832249641418457,
      "epoch": 0.22334293948126802,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998168798057715,
      "loss": 5.6003,
      "mean_token_accuracy": 0.1812728613615036,
      "num_tokens": 5333811.0,
      "step": 2325
    },
    {
      "entropy": 5.826548004150391,
      "epoch": 0.22382324687800192,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499815494300587,
      "loss": 5.6142,
      "mean_token_accuracy": 0.17571379840373993,
      "num_tokens": 5344762.0,
      "step": 2330
    },
    {
      "entropy": 5.80838451385498,
      "epoch": 0.22430355427473583,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998141035758542,
      "loss": 5.602,
      "mean_token_accuracy": 0.18153773993253708,
      "num_tokens": 5356112.0,
      "step": 2335
    },
    {
      "entropy": 5.876251316070556,
      "epoch": 0.22478386167146974,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998127076316054,
      "loss": 5.7258,
      "mean_token_accuracy": 0.16987995207309722,
      "num_tokens": 5367339.0,
      "step": 2340
    },
    {
      "entropy": 5.799464511871338,
      "epoch": 0.22526416906820365,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998113064678734,
      "loss": 5.6526,
      "mean_token_accuracy": 0.183486644923687,
      "num_tokens": 5378627.0,
      "step": 2345
    },
    {
      "entropy": 5.8673011302948,
      "epoch": 0.22574447646493756,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998099000846901,
      "loss": 5.6825,
      "mean_token_accuracy": 0.1747091993689537,
      "num_tokens": 5390209.0,
      "step": 2350
    },
    {
      "entropy": 5.885054302215576,
      "epoch": 0.22622478386167147,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998084884820887,
      "loss": 5.6786,
      "mean_token_accuracy": 0.17270639091730117,
      "num_tokens": 5401578.0,
      "step": 2355
    },
    {
      "entropy": 5.7969471454620365,
      "epoch": 0.22670509125840538,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998070716601016,
      "loss": 5.6028,
      "mean_token_accuracy": 0.17538144290447236,
      "num_tokens": 5413831.0,
      "step": 2360
    },
    {
      "entropy": 5.863705444335937,
      "epoch": 0.2271853986551393,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998056496187618,
      "loss": 5.652,
      "mean_token_accuracy": 0.1691875770688057,
      "num_tokens": 5425430.0,
      "step": 2365
    },
    {
      "entropy": 5.6353754043579105,
      "epoch": 0.2276657060518732,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998042223581025,
      "loss": 5.5009,
      "mean_token_accuracy": 0.1855420872569084,
      "num_tokens": 5435353.0,
      "step": 2370
    },
    {
      "entropy": 5.862425804138184,
      "epoch": 0.2281460134486071,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998027898781565,
      "loss": 5.7041,
      "mean_token_accuracy": 0.17443186193704605,
      "num_tokens": 5446925.0,
      "step": 2375
    },
    {
      "entropy": 5.729289484024048,
      "epoch": 0.228626320845341,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998013521789574,
      "loss": 5.585,
      "mean_token_accuracy": 0.1821834087371826,
      "num_tokens": 5456613.0,
      "step": 2380
    },
    {
      "entropy": 5.850210475921631,
      "epoch": 0.22910662824207492,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997999092605384,
      "loss": 5.6193,
      "mean_token_accuracy": 0.17688928544521332,
      "num_tokens": 5467790.0,
      "step": 2385
    },
    {
      "entropy": 5.8598381042480465,
      "epoch": 0.22958693563880883,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499798461122933,
      "loss": 5.6118,
      "mean_token_accuracy": 0.1773025169968605,
      "num_tokens": 5479166.0,
      "step": 2390
    },
    {
      "entropy": 5.695592832565308,
      "epoch": 0.23006724303554274,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997970077661748,
      "loss": 5.5783,
      "mean_token_accuracy": 0.18406546413898467,
      "num_tokens": 5490186.0,
      "step": 2395
    },
    {
      "entropy": 5.785607433319091,
      "epoch": 0.23054755043227665,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997955491902977,
      "loss": 5.5746,
      "mean_token_accuracy": 0.17407838255167007,
      "num_tokens": 5500416.0,
      "step": 2400
    },
    {
      "entropy": 5.80128846168518,
      "epoch": 0.23102785782901056,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997940853953354,
      "loss": 5.6449,
      "mean_token_accuracy": 0.17734454870223998,
      "num_tokens": 5512189.0,
      "step": 2405
    },
    {
      "entropy": 5.753947496414185,
      "epoch": 0.23150816522574447,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499792616381322,
      "loss": 5.5149,
      "mean_token_accuracy": 0.1830824211239815,
      "num_tokens": 5523631.0,
      "step": 2410
    },
    {
      "entropy": 5.727394866943359,
      "epoch": 0.23198847262247838,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997911421482914,
      "loss": 5.5761,
      "mean_token_accuracy": 0.17973856627941132,
      "num_tokens": 5535637.0,
      "step": 2415
    },
    {
      "entropy": 5.7857983112335205,
      "epoch": 0.23246878001921228,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499789662696278,
      "loss": 5.5863,
      "mean_token_accuracy": 0.1800864651799202,
      "num_tokens": 5546470.0,
      "step": 2420
    },
    {
      "entropy": 5.888076305389404,
      "epoch": 0.2329490874159462,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997881780253162,
      "loss": 5.7413,
      "mean_token_accuracy": 0.16848194748163223,
      "num_tokens": 5558633.0,
      "step": 2425
    },
    {
      "entropy": 5.80600266456604,
      "epoch": 0.2334293948126801,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004997866881354403,
      "loss": 5.6487,
      "mean_token_accuracy": 0.17162267118692398,
      "num_tokens": 5570427.0,
      "step": 2430
    },
    {
      "entropy": 5.762806415557861,
      "epoch": 0.23390970220941404,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499785193026685,
      "loss": 5.6197,
      "mean_token_accuracy": 0.17541880607604982,
      "num_tokens": 5580991.0,
      "step": 2435
    },
    {
      "entropy": 5.884730529785156,
      "epoch": 0.23439000960614795,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997836926990851,
      "loss": 5.6853,
      "mean_token_accuracy": 0.17449010610580445,
      "num_tokens": 5592777.0,
      "step": 2440
    },
    {
      "entropy": 5.745317554473877,
      "epoch": 0.23487031700288186,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997821871526752,
      "loss": 5.5765,
      "mean_token_accuracy": 0.18271358013153077,
      "num_tokens": 5603326.0,
      "step": 2445
    },
    {
      "entropy": 5.73594331741333,
      "epoch": 0.23535062439961577,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997806763874905,
      "loss": 5.5693,
      "mean_token_accuracy": 0.18031217455863952,
      "num_tokens": 5614504.0,
      "step": 2450
    },
    {
      "entropy": 5.782830619812012,
      "epoch": 0.23583093179634967,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997791604035659,
      "loss": 5.6286,
      "mean_token_accuracy": 0.17803447991609572,
      "num_tokens": 5625150.0,
      "step": 2455
    },
    {
      "entropy": 5.804438877105713,
      "epoch": 0.23631123919308358,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997776392009366,
      "loss": 5.6445,
      "mean_token_accuracy": 0.17143210619688035,
      "num_tokens": 5636815.0,
      "step": 2460
    },
    {
      "entropy": 5.8043129444122314,
      "epoch": 0.2367915465898175,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997761127796381,
      "loss": 5.6297,
      "mean_token_accuracy": 0.17545064836740493,
      "num_tokens": 5648272.0,
      "step": 2465
    },
    {
      "entropy": 5.757105493545533,
      "epoch": 0.2372718539865514,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997745811397056,
      "loss": 5.5526,
      "mean_token_accuracy": 0.1801515832543373,
      "num_tokens": 5659227.0,
      "step": 2470
    },
    {
      "entropy": 5.795153522491455,
      "epoch": 0.2377521613832853,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997730442811748,
      "loss": 5.673,
      "mean_token_accuracy": 0.17654881477355958,
      "num_tokens": 5670411.0,
      "step": 2475
    },
    {
      "entropy": 5.748981523513794,
      "epoch": 0.23823246878001922,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997715022040814,
      "loss": 5.5096,
      "mean_token_accuracy": 0.18419251292943956,
      "num_tokens": 5681570.0,
      "step": 2480
    },
    {
      "entropy": 5.7172455310821535,
      "epoch": 0.23871277617675313,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499769954908461,
      "loss": 5.5052,
      "mean_token_accuracy": 0.18976044952869414,
      "num_tokens": 5693021.0,
      "step": 2485
    },
    {
      "entropy": 5.727937793731689,
      "epoch": 0.23919308357348704,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997684023943498,
      "loss": 5.5945,
      "mean_token_accuracy": 0.17496601939201356,
      "num_tokens": 5704043.0,
      "step": 2490
    },
    {
      "entropy": 5.732729482650757,
      "epoch": 0.23967339097022095,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997668446617837,
      "loss": 5.6705,
      "mean_token_accuracy": 0.16925858408212663,
      "num_tokens": 5715735.0,
      "step": 2495
    },
    {
      "entropy": 5.932486486434937,
      "epoch": 0.24015369836695485,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997652817107989,
      "loss": 5.6455,
      "mean_token_accuracy": 0.17272803783416749,
      "num_tokens": 5725778.0,
      "step": 2500
    },
    {
      "entropy": 5.75950403213501,
      "epoch": 0.24063400576368876,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997637135414315,
      "loss": 5.6535,
      "mean_token_accuracy": 0.17402508705854416,
      "num_tokens": 5737224.0,
      "step": 2505
    },
    {
      "entropy": 5.841374206542969,
      "epoch": 0.24111431316042267,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997621401537183,
      "loss": 5.6819,
      "mean_token_accuracy": 0.1714730903506279,
      "num_tokens": 5749226.0,
      "step": 2510
    },
    {
      "entropy": 5.859533452987671,
      "epoch": 0.24159462055715658,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997605615476955,
      "loss": 5.65,
      "mean_token_accuracy": 0.17602848410606384,
      "num_tokens": 5760282.0,
      "step": 2515
    },
    {
      "entropy": 5.705944013595581,
      "epoch": 0.2420749279538905,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997589777234,
      "loss": 5.5527,
      "mean_token_accuracy": 0.18297999501228332,
      "num_tokens": 5771756.0,
      "step": 2520
    },
    {
      "entropy": 5.783916234970093,
      "epoch": 0.2425552353506244,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997573886808684,
      "loss": 5.5929,
      "mean_token_accuracy": 0.17033566832542418,
      "num_tokens": 5783237.0,
      "step": 2525
    },
    {
      "entropy": 5.766665458679199,
      "epoch": 0.2430355427473583,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997557944201375,
      "loss": 5.6765,
      "mean_token_accuracy": 0.17114062905311583,
      "num_tokens": 5794825.0,
      "step": 2530
    },
    {
      "entropy": 5.83233232498169,
      "epoch": 0.24351585014409222,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997541949412445,
      "loss": 5.5704,
      "mean_token_accuracy": 0.18051684647798538,
      "num_tokens": 5805578.0,
      "step": 2535
    },
    {
      "entropy": 5.825575923919677,
      "epoch": 0.24399615754082613,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997525902442266,
      "loss": 5.6636,
      "mean_token_accuracy": 0.1699126750230789,
      "num_tokens": 5818201.0,
      "step": 2540
    },
    {
      "entropy": 5.761399126052856,
      "epoch": 0.24447646493756003,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997509803291207,
      "loss": 5.5885,
      "mean_token_accuracy": 0.17694975286722184,
      "num_tokens": 5830319.0,
      "step": 2545
    },
    {
      "entropy": 5.706338500976562,
      "epoch": 0.24495677233429394,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997493651959647,
      "loss": 5.5395,
      "mean_token_accuracy": 0.1815444216132164,
      "num_tokens": 5840638.0,
      "step": 2550
    },
    {
      "entropy": 5.751677846908569,
      "epoch": 0.24543707973102785,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997477448447955,
      "loss": 5.5617,
      "mean_token_accuracy": 0.17573375552892684,
      "num_tokens": 5852472.0,
      "step": 2555
    },
    {
      "entropy": 5.835761976242066,
      "epoch": 0.24591738712776176,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997461192756512,
      "loss": 5.6,
      "mean_token_accuracy": 0.17359656244516372,
      "num_tokens": 5863455.0,
      "step": 2560
    },
    {
      "entropy": 5.677509737014771,
      "epoch": 0.24639769452449567,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997444884885694,
      "loss": 5.5109,
      "mean_token_accuracy": 0.18119774758815765,
      "num_tokens": 5873141.0,
      "step": 2565
    },
    {
      "entropy": 5.6975038051605225,
      "epoch": 0.24687800192122958,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997428524835879,
      "loss": 5.6097,
      "mean_token_accuracy": 0.1763081893324852,
      "num_tokens": 5884363.0,
      "step": 2570
    },
    {
      "entropy": 5.83483624458313,
      "epoch": 0.2473583093179635,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997412112607446,
      "loss": 5.6652,
      "mean_token_accuracy": 0.17418509721755981,
      "num_tokens": 5895856.0,
      "step": 2575
    },
    {
      "entropy": 5.710783958435059,
      "epoch": 0.2478386167146974,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997395648200778,
      "loss": 5.5001,
      "mean_token_accuracy": 0.18241955786943437,
      "num_tokens": 5906657.0,
      "step": 2580
    },
    {
      "entropy": 5.73748025894165,
      "epoch": 0.2483189241114313,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997379131616257,
      "loss": 5.6229,
      "mean_token_accuracy": 0.17517986595630647,
      "num_tokens": 5919496.0,
      "step": 2585
    },
    {
      "entropy": 5.801119375228882,
      "epoch": 0.24879923150816521,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997362562854266,
      "loss": 5.6814,
      "mean_token_accuracy": 0.16362314075231552,
      "num_tokens": 5932593.0,
      "step": 2590
    },
    {
      "entropy": 5.805383825302124,
      "epoch": 0.24927953890489912,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997345941915187,
      "loss": 5.5946,
      "mean_token_accuracy": 0.17460722476243973,
      "num_tokens": 5944080.0,
      "step": 2595
    },
    {
      "entropy": 5.7451694965362545,
      "epoch": 0.24975984630163303,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997329268799412,
      "loss": 5.5769,
      "mean_token_accuracy": 0.18260987102985382,
      "num_tokens": 5955703.0,
      "step": 2600
    },
    {
      "entropy": 5.733469676971436,
      "epoch": 0.25024015369836694,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997312543507322,
      "loss": 5.6574,
      "mean_token_accuracy": 0.1740378275513649,
      "num_tokens": 5966979.0,
      "step": 2605
    },
    {
      "entropy": 5.814836883544922,
      "epoch": 0.2507204610951009,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997295766039309,
      "loss": 5.5405,
      "mean_token_accuracy": 0.17929624170064926,
      "num_tokens": 5978808.0,
      "step": 2610
    },
    {
      "entropy": 5.754059314727783,
      "epoch": 0.25120076849183476,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997278936395761,
      "loss": 5.7145,
      "mean_token_accuracy": 0.16822077333927155,
      "num_tokens": 5992145.0,
      "step": 2615
    },
    {
      "entropy": 5.775521326065063,
      "epoch": 0.2516810758885687,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997262054577071,
      "loss": 5.5776,
      "mean_token_accuracy": 0.1735659971833229,
      "num_tokens": 6003723.0,
      "step": 2620
    },
    {
      "entropy": 5.785316133499146,
      "epoch": 0.2521613832853026,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997245120583627,
      "loss": 5.6083,
      "mean_token_accuracy": 0.18019691854715347,
      "num_tokens": 6014064.0,
      "step": 2625
    },
    {
      "entropy": 5.703524494171143,
      "epoch": 0.2526416906820365,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997228134415825,
      "loss": 5.5219,
      "mean_token_accuracy": 0.18180668652057647,
      "num_tokens": 6025455.0,
      "step": 2630
    },
    {
      "entropy": 5.781266164779663,
      "epoch": 0.2531219980787704,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997211096074059,
      "loss": 5.6394,
      "mean_token_accuracy": 0.1656498432159424,
      "num_tokens": 6037347.0,
      "step": 2635
    },
    {
      "entropy": 5.690671968460083,
      "epoch": 0.25360230547550433,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997194005558722,
      "loss": 5.5216,
      "mean_token_accuracy": 0.1789523884654045,
      "num_tokens": 6049236.0,
      "step": 2640
    },
    {
      "entropy": 5.6680278301239015,
      "epoch": 0.2540826128722382,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004997176862870216,
      "loss": 5.538,
      "mean_token_accuracy": 0.17978094518184662,
      "num_tokens": 6060982.0,
      "step": 2645
    },
    {
      "entropy": 5.819158840179443,
      "epoch": 0.25456292026897215,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997159668008933,
      "loss": 5.5366,
      "mean_token_accuracy": 0.18156028538942337,
      "num_tokens": 6070925.0,
      "step": 2650
    },
    {
      "entropy": 5.680534505844117,
      "epoch": 0.25504322766570603,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997142420975277,
      "loss": 5.487,
      "mean_token_accuracy": 0.18826171904802322,
      "num_tokens": 6081279.0,
      "step": 2655
    },
    {
      "entropy": 5.710468053817749,
      "epoch": 0.25552353506243997,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997125121769647,
      "loss": 5.5967,
      "mean_token_accuracy": 0.17861386388540268,
      "num_tokens": 6091797.0,
      "step": 2660
    },
    {
      "entropy": 5.825000762939453,
      "epoch": 0.25600384245917385,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997107770392444,
      "loss": 5.6073,
      "mean_token_accuracy": 0.18347439020872117,
      "num_tokens": 6103435.0,
      "step": 2665
    },
    {
      "entropy": 5.79217495918274,
      "epoch": 0.2564841498559078,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499709036684407,
      "loss": 5.6709,
      "mean_token_accuracy": 0.17622565925121308,
      "num_tokens": 6114531.0,
      "step": 2670
    },
    {
      "entropy": 5.653947401046753,
      "epoch": 0.25696445725264166,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997072911124932,
      "loss": 5.5192,
      "mean_token_accuracy": 0.18308780789375306,
      "num_tokens": 6126110.0,
      "step": 2675
    },
    {
      "entropy": 5.768945932388306,
      "epoch": 0.2574447646493756,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997055403235432,
      "loss": 5.6082,
      "mean_token_accuracy": 0.17530772536993028,
      "num_tokens": 6137114.0,
      "step": 2680
    },
    {
      "entropy": 5.69290862083435,
      "epoch": 0.2579250720461095,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997037843175978,
      "loss": 5.5674,
      "mean_token_accuracy": 0.179079769551754,
      "num_tokens": 6148696.0,
      "step": 2685
    },
    {
      "entropy": 5.712290477752686,
      "epoch": 0.2584053794428434,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997020230946978,
      "loss": 5.5568,
      "mean_token_accuracy": 0.17987955063581468,
      "num_tokens": 6160235.0,
      "step": 2690
    },
    {
      "entropy": 5.794245862960816,
      "epoch": 0.25888568683957736,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997002566548841,
      "loss": 5.5384,
      "mean_token_accuracy": 0.17859422713518142,
      "num_tokens": 6172031.0,
      "step": 2695
    },
    {
      "entropy": 5.621943998336792,
      "epoch": 0.25936599423631124,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996984849981976,
      "loss": 5.415,
      "mean_token_accuracy": 0.18926973193883895,
      "num_tokens": 6183547.0,
      "step": 2700
    },
    {
      "entropy": 5.752551746368408,
      "epoch": 0.2598463016330452,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996967081246794,
      "loss": 5.6197,
      "mean_token_accuracy": 0.17076316028833388,
      "num_tokens": 6194768.0,
      "step": 2705
    },
    {
      "entropy": 5.741240787506103,
      "epoch": 0.26032660902977905,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996949260343711,
      "loss": 5.6212,
      "mean_token_accuracy": 0.17027782052755355,
      "num_tokens": 6206099.0,
      "step": 2710
    },
    {
      "entropy": 5.780304861068726,
      "epoch": 0.260806916426513,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996931387273137,
      "loss": 5.61,
      "mean_token_accuracy": 0.1808237463235855,
      "num_tokens": 6217530.0,
      "step": 2715
    },
    {
      "entropy": 5.834391117095947,
      "epoch": 0.2612872238232469,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004996913462035487,
      "loss": 5.6319,
      "mean_token_accuracy": 0.17484427392482757,
      "num_tokens": 6228564.0,
      "step": 2720
    },
    {
      "entropy": 5.670606851577759,
      "epoch": 0.2617675312199808,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499689548463118,
      "loss": 5.5138,
      "mean_token_accuracy": 0.17770184427499772,
      "num_tokens": 6239945.0,
      "step": 2725
    },
    {
      "entropy": 5.735776376724243,
      "epoch": 0.2622478386167147,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996877455060631,
      "loss": 5.622,
      "mean_token_accuracy": 0.17381711900234223,
      "num_tokens": 6251829.0,
      "step": 2730
    },
    {
      "entropy": 5.871343946456909,
      "epoch": 0.2627281460134486,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996859373324259,
      "loss": 5.7197,
      "mean_token_accuracy": 0.16585395336151124,
      "num_tokens": 6264823.0,
      "step": 2735
    },
    {
      "entropy": 5.686900472640991,
      "epoch": 0.2632084534101825,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996841239422485,
      "loss": 5.4025,
      "mean_token_accuracy": 0.18778327703475953,
      "num_tokens": 6276247.0,
      "step": 2740
    },
    {
      "entropy": 5.597873067855835,
      "epoch": 0.26368876080691644,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996823053355729,
      "loss": 5.5143,
      "mean_token_accuracy": 0.18251601308584214,
      "num_tokens": 6287593.0,
      "step": 2745
    },
    {
      "entropy": 5.837592267990113,
      "epoch": 0.2641690682036503,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996804815124413,
      "loss": 5.6832,
      "mean_token_accuracy": 0.1740195035934448,
      "num_tokens": 6299918.0,
      "step": 2750
    },
    {
      "entropy": 5.680815124511719,
      "epoch": 0.26464937560038426,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996786524728962,
      "loss": 5.5247,
      "mean_token_accuracy": 0.18393096774816514,
      "num_tokens": 6311147.0,
      "step": 2755
    },
    {
      "entropy": 5.613561296463013,
      "epoch": 0.26512968299711814,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996768182169797,
      "loss": 5.4431,
      "mean_token_accuracy": 0.18843217045068741,
      "num_tokens": 6323239.0,
      "step": 2760
    },
    {
      "entropy": 5.7647748470306395,
      "epoch": 0.2656099903938521,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996749787447349,
      "loss": 5.5256,
      "mean_token_accuracy": 0.1798029989004135,
      "num_tokens": 6334625.0,
      "step": 2765
    },
    {
      "entropy": 5.730550050735474,
      "epoch": 0.26609029779058596,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499673134056204,
      "loss": 5.5671,
      "mean_token_accuracy": 0.18138138502836226,
      "num_tokens": 6346068.0,
      "step": 2770
    },
    {
      "entropy": 5.688610410690307,
      "epoch": 0.2665706051873199,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996712841514303,
      "loss": 5.5462,
      "mean_token_accuracy": 0.18104476630687713,
      "num_tokens": 6357097.0,
      "step": 2775
    },
    {
      "entropy": 5.780763864517212,
      "epoch": 0.2670509125840538,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996694290304563,
      "loss": 5.6012,
      "mean_token_accuracy": 0.1663811519742012,
      "num_tokens": 6367481.0,
      "step": 2780
    },
    {
      "entropy": 5.648175573348999,
      "epoch": 0.2675312199807877,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996675686933255,
      "loss": 5.5164,
      "mean_token_accuracy": 0.1855094313621521,
      "num_tokens": 6378873.0,
      "step": 2785
    },
    {
      "entropy": 5.778855705261231,
      "epoch": 0.2680115273775216,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996657031400807,
      "loss": 5.5687,
      "mean_token_accuracy": 0.1777387112379074,
      "num_tokens": 6390651.0,
      "step": 2790
    },
    {
      "entropy": 5.656892156600952,
      "epoch": 0.26849183477425553,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996638323707655,
      "loss": 5.4324,
      "mean_token_accuracy": 0.18739653527736663,
      "num_tokens": 6401631.0,
      "step": 2795
    },
    {
      "entropy": 5.608145618438721,
      "epoch": 0.2689721421709894,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996619563854232,
      "loss": 5.5187,
      "mean_token_accuracy": 0.1855126902461052,
      "num_tokens": 6413875.0,
      "step": 2800
    },
    {
      "entropy": 5.782154989242554,
      "epoch": 0.26945244956772335,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996600751840974,
      "loss": 5.5557,
      "mean_token_accuracy": 0.176130411028862,
      "num_tokens": 6425764.0,
      "step": 2805
    },
    {
      "entropy": 5.670776462554931,
      "epoch": 0.26993275696445723,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996581887668317,
      "loss": 5.4646,
      "mean_token_accuracy": 0.18302996158599855,
      "num_tokens": 6437911.0,
      "step": 2810
    },
    {
      "entropy": 5.640679788589478,
      "epoch": 0.27041306436119117,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00049965629713367,
      "loss": 5.4898,
      "mean_token_accuracy": 0.18281519412994385,
      "num_tokens": 6449942.0,
      "step": 2815
    },
    {
      "entropy": 5.719023561477661,
      "epoch": 0.27089337175792505,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004996544002846561,
      "loss": 5.5995,
      "mean_token_accuracy": 0.17978287041187285,
      "num_tokens": 6461729.0,
      "step": 2820
    },
    {
      "entropy": 5.751323080062866,
      "epoch": 0.271373679154659,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996524982198343,
      "loss": 5.5719,
      "mean_token_accuracy": 0.18161486983299255,
      "num_tokens": 6472046.0,
      "step": 2825
    },
    {
      "entropy": 5.721522521972656,
      "epoch": 0.27185398655139287,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996505909392485,
      "loss": 5.5477,
      "mean_token_accuracy": 0.1771852806210518,
      "num_tokens": 6483308.0,
      "step": 2830
    },
    {
      "entropy": 5.551406669616699,
      "epoch": 0.2723342939481268,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996486784429429,
      "loss": 5.4219,
      "mean_token_accuracy": 0.1857747346162796,
      "num_tokens": 6495093.0,
      "step": 2835
    },
    {
      "entropy": 5.71174054145813,
      "epoch": 0.2728146013448607,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996467607309622,
      "loss": 5.5127,
      "mean_token_accuracy": 0.17964418977499008,
      "num_tokens": 6505933.0,
      "step": 2840
    },
    {
      "entropy": 5.72988600730896,
      "epoch": 0.2732949087415946,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996448378033507,
      "loss": 5.559,
      "mean_token_accuracy": 0.17828564196825028,
      "num_tokens": 6517280.0,
      "step": 2845
    },
    {
      "entropy": 5.678931474685669,
      "epoch": 0.2737752161383285,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996429096601532,
      "loss": 5.6158,
      "mean_token_accuracy": 0.17375251650810242,
      "num_tokens": 6528980.0,
      "step": 2850
    },
    {
      "entropy": 5.71303505897522,
      "epoch": 0.27425552353506244,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996409763014144,
      "loss": 5.6106,
      "mean_token_accuracy": 0.1804870694875717,
      "num_tokens": 6540670.0,
      "step": 2855
    },
    {
      "entropy": 5.702807521820068,
      "epoch": 0.2747358309317964,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996390377271791,
      "loss": 5.57,
      "mean_token_accuracy": 0.18366750180721284,
      "num_tokens": 6551302.0,
      "step": 2860
    },
    {
      "entropy": 5.705173969268799,
      "epoch": 0.27521613832853026,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996370939374924,
      "loss": 5.5216,
      "mean_token_accuracy": 0.17237444818019867,
      "num_tokens": 6563177.0,
      "step": 2865
    },
    {
      "entropy": 5.844460535049438,
      "epoch": 0.2756964457252642,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996351449323994,
      "loss": 5.6155,
      "mean_token_accuracy": 0.1813505157828331,
      "num_tokens": 6573323.0,
      "step": 2870
    },
    {
      "entropy": 5.725948524475098,
      "epoch": 0.2761767531219981,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996331907119455,
      "loss": 5.5713,
      "mean_token_accuracy": 0.1717742770910263,
      "num_tokens": 6585382.0,
      "step": 2875
    },
    {
      "entropy": 5.586256790161133,
      "epoch": 0.276657060518732,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996312312761758,
      "loss": 5.4521,
      "mean_token_accuracy": 0.1891452044248581,
      "num_tokens": 6596629.0,
      "step": 2880
    },
    {
      "entropy": 5.739269733428955,
      "epoch": 0.2771373679154659,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499629266625136,
      "loss": 5.5628,
      "mean_token_accuracy": 0.18007983416318893,
      "num_tokens": 6608408.0,
      "step": 2885
    },
    {
      "entropy": 5.759958505630493,
      "epoch": 0.27761767531219983,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996272967588715,
      "loss": 5.5799,
      "mean_token_accuracy": 0.17467228472232818,
      "num_tokens": 6619375.0,
      "step": 2890
    },
    {
      "entropy": 5.750079774856568,
      "epoch": 0.2780979827089337,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996253216774283,
      "loss": 5.6357,
      "mean_token_accuracy": 0.17795114517211913,
      "num_tokens": 6631317.0,
      "step": 2895
    },
    {
      "entropy": 5.730680131912232,
      "epoch": 0.27857829010566765,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996233413808521,
      "loss": 5.484,
      "mean_token_accuracy": 0.18032360523939134,
      "num_tokens": 6642009.0,
      "step": 2900
    },
    {
      "entropy": 5.708155965805053,
      "epoch": 0.27905859750240153,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996213558691889,
      "loss": 5.6328,
      "mean_token_accuracy": 0.1716104194521904,
      "num_tokens": 6654713.0,
      "step": 2905
    },
    {
      "entropy": 5.699761247634887,
      "epoch": 0.27953890489913547,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996193651424848,
      "loss": 5.5966,
      "mean_token_accuracy": 0.1802009731531143,
      "num_tokens": 6667157.0,
      "step": 2910
    },
    {
      "entropy": 5.7038122653961185,
      "epoch": 0.28001921229586935,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000499617369200786,
      "loss": 5.5383,
      "mean_token_accuracy": 0.18896565586328506,
      "num_tokens": 6679573.0,
      "step": 2915
    },
    {
      "entropy": 5.7405517578125,
      "epoch": 0.2804995196926033,
      "grad_norm": 0.875,
      "learning_rate": 0.0004996153680441389,
      "loss": 5.6213,
      "mean_token_accuracy": 0.17383338063955306,
      "num_tokens": 6691768.0,
      "step": 2920
    },
    {
      "entropy": 5.770690536499023,
      "epoch": 0.28097982708933716,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00049961336167259,
      "loss": 5.5717,
      "mean_token_accuracy": 0.174806410074234,
      "num_tokens": 6701964.0,
      "step": 2925
    },
    {
      "entropy": 5.705366468429565,
      "epoch": 0.2814601344860711,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996113500861857,
      "loss": 5.5588,
      "mean_token_accuracy": 0.18040292710065842,
      "num_tokens": 6713506.0,
      "step": 2930
    },
    {
      "entropy": 5.7421009063720705,
      "epoch": 0.281940441882805,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996093332849729,
      "loss": 5.5832,
      "mean_token_accuracy": 0.17432304471731186,
      "num_tokens": 6724616.0,
      "step": 2935
    },
    {
      "entropy": 5.721134281158447,
      "epoch": 0.2824207492795389,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996073112689983,
      "loss": 5.5541,
      "mean_token_accuracy": 0.17643650472164155,
      "num_tokens": 6735054.0,
      "step": 2940
    },
    {
      "entropy": 5.75977578163147,
      "epoch": 0.2829010566762728,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004996052840383088,
      "loss": 5.6062,
      "mean_token_accuracy": 0.17887143716216086,
      "num_tokens": 6746756.0,
      "step": 2945
    },
    {
      "entropy": 5.688607835769654,
      "epoch": 0.28338136407300674,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004996032515929516,
      "loss": 5.4822,
      "mean_token_accuracy": 0.1823216959834099,
      "num_tokens": 6759566.0,
      "step": 2950
    },
    {
      "entropy": 5.691004610061645,
      "epoch": 0.2838616714697406,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996012139329738,
      "loss": 5.4995,
      "mean_token_accuracy": 0.17962852418422698,
      "num_tokens": 6771375.0,
      "step": 2955
    },
    {
      "entropy": 5.770864868164063,
      "epoch": 0.28434197886647455,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995991710584228,
      "loss": 5.6127,
      "mean_token_accuracy": 0.17344165444374085,
      "num_tokens": 6783252.0,
      "step": 2960
    },
    {
      "entropy": 5.664705038070679,
      "epoch": 0.28482228626320844,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995971229693459,
      "loss": 5.5973,
      "mean_token_accuracy": 0.17210681587457657,
      "num_tokens": 6795525.0,
      "step": 2965
    },
    {
      "entropy": 5.780670309066773,
      "epoch": 0.28530259365994237,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004995950696657909,
      "loss": 5.5285,
      "mean_token_accuracy": 0.18434083163738252,
      "num_tokens": 6807212.0,
      "step": 2970
    },
    {
      "entropy": 5.617880916595459,
      "epoch": 0.28578290105667625,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995930111478051,
      "loss": 5.4638,
      "mean_token_accuracy": 0.18253830969333648,
      "num_tokens": 6819367.0,
      "step": 2975
    },
    {
      "entropy": 5.67643985748291,
      "epoch": 0.2862632084534102,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995909474154365,
      "loss": 5.5384,
      "mean_token_accuracy": 0.18423326760530473,
      "num_tokens": 6830405.0,
      "step": 2980
    },
    {
      "entropy": 5.620176124572754,
      "epoch": 0.28674351585014407,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004995888784687331,
      "loss": 5.5139,
      "mean_token_accuracy": 0.17934491336345673,
      "num_tokens": 6841479.0,
      "step": 2985
    },
    {
      "entropy": 5.667870092391968,
      "epoch": 0.287223823246878,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995868043077428,
      "loss": 5.5504,
      "mean_token_accuracy": 0.17985389232635499,
      "num_tokens": 6851585.0,
      "step": 2990
    },
    {
      "entropy": 5.701903438568115,
      "epoch": 0.2877041306436119,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995847249325137,
      "loss": 5.5311,
      "mean_token_accuracy": 0.1836439073085785,
      "num_tokens": 6863176.0,
      "step": 2995
    },
    {
      "entropy": 5.728053426742553,
      "epoch": 0.2881844380403458,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995826403430942,
      "loss": 5.586,
      "mean_token_accuracy": 0.1827200844883919,
      "num_tokens": 6874021.0,
      "step": 3000
    },
    {
      "epoch": 0.2881844380403458,
      "eval_entropy": 5.467494524030904,
      "eval_loss": 5.550236225128174,
      "eval_mean_token_accuracy": 0.187719893863362,
      "eval_num_tokens": 6874021.0,
      "eval_runtime": 20.6705,
      "eval_samples_per_second": 1587.532,
      "eval_steps_per_second": 198.448,
      "step": 3000
    },
    {
      "entropy": 5.704174375534057,
      "epoch": 0.2886647454370797,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995805505395328,
      "loss": 5.5426,
      "mean_token_accuracy": 0.18029418736696243,
      "num_tokens": 6884999.0,
      "step": 3005
    },
    {
      "entropy": 5.699745559692383,
      "epoch": 0.28914505283381364,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004995784555218778,
      "loss": 5.523,
      "mean_token_accuracy": 0.1831611156463623,
      "num_tokens": 6897021.0,
      "step": 3010
    },
    {
      "entropy": 5.637732887268067,
      "epoch": 0.2896253602305475,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995763552901779,
      "loss": 5.5302,
      "mean_token_accuracy": 0.18181773722171785,
      "num_tokens": 6908320.0,
      "step": 3015
    },
    {
      "entropy": 5.748047924041748,
      "epoch": 0.29010566762728146,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995742498444818,
      "loss": 5.5131,
      "mean_token_accuracy": 0.18566226214170456,
      "num_tokens": 6919957.0,
      "step": 3020
    },
    {
      "entropy": 5.710698127746582,
      "epoch": 0.2905859750240154,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004995721391848387,
      "loss": 5.4732,
      "mean_token_accuracy": 0.17768509238958358,
      "num_tokens": 6930531.0,
      "step": 3025
    },
    {
      "entropy": 5.712391090393067,
      "epoch": 0.2910662824207493,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995700233112972,
      "loss": 5.6125,
      "mean_token_accuracy": 0.1782066062092781,
      "num_tokens": 6942556.0,
      "step": 3030
    },
    {
      "entropy": 5.708733320236206,
      "epoch": 0.2915465898174832,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004995679022239066,
      "loss": 5.5547,
      "mean_token_accuracy": 0.1831905037164688,
      "num_tokens": 6954410.0,
      "step": 3035
    },
    {
      "entropy": 5.706331968307495,
      "epoch": 0.2920268972142171,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995657759227162,
      "loss": 5.5624,
      "mean_token_accuracy": 0.17520011961460114,
      "num_tokens": 6964970.0,
      "step": 3040
    },
    {
      "entropy": 5.6419037818908695,
      "epoch": 0.29250720461095103,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995636444077751,
      "loss": 5.4598,
      "mean_token_accuracy": 0.18813293129205705,
      "num_tokens": 6976016.0,
      "step": 3045
    },
    {
      "entropy": 5.64836859703064,
      "epoch": 0.2929875120076849,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995615076791333,
      "loss": 5.4822,
      "mean_token_accuracy": 0.18480593264102935,
      "num_tokens": 6987199.0,
      "step": 3050
    },
    {
      "entropy": 5.643376255035401,
      "epoch": 0.29346781940441885,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995593657368399,
      "loss": 5.5014,
      "mean_token_accuracy": 0.18682510256767274,
      "num_tokens": 6999174.0,
      "step": 3055
    },
    {
      "entropy": 5.666988182067871,
      "epoch": 0.29394812680115273,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499557218580945,
      "loss": 5.5725,
      "mean_token_accuracy": 0.1785376712679863,
      "num_tokens": 7012148.0,
      "step": 3060
    },
    {
      "entropy": 5.61878514289856,
      "epoch": 0.29442843419788667,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995550662114981,
      "loss": 5.4427,
      "mean_token_accuracy": 0.18653325736522675,
      "num_tokens": 7023238.0,
      "step": 3065
    },
    {
      "entropy": 5.654002618789673,
      "epoch": 0.29490874159462055,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995529086285495,
      "loss": 5.5267,
      "mean_token_accuracy": 0.1808794245123863,
      "num_tokens": 7034944.0,
      "step": 3070
    },
    {
      "entropy": 5.678547668457031,
      "epoch": 0.2953890489913545,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499550745832149,
      "loss": 5.392,
      "mean_token_accuracy": 0.19032707810401917,
      "num_tokens": 7046880.0,
      "step": 3075
    },
    {
      "entropy": 5.637848234176635,
      "epoch": 0.29586935638808837,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995485778223471,
      "loss": 5.4831,
      "mean_token_accuracy": 0.17996930032968522,
      "num_tokens": 7057678.0,
      "step": 3080
    },
    {
      "entropy": 5.570685338973999,
      "epoch": 0.2963496637848223,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995464045991939,
      "loss": 5.4519,
      "mean_token_accuracy": 0.18621571809053422,
      "num_tokens": 7068336.0,
      "step": 3085
    },
    {
      "entropy": 5.675237321853638,
      "epoch": 0.2968299711815562,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00049954422616274,
      "loss": 5.5212,
      "mean_token_accuracy": 0.17852572798728944,
      "num_tokens": 7080341.0,
      "step": 3090
    },
    {
      "entropy": 5.806090068817139,
      "epoch": 0.2973102785782901,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995420425130359,
      "loss": 5.6666,
      "mean_token_accuracy": 0.17285532206296922,
      "num_tokens": 7090618.0,
      "step": 3095
    },
    {
      "entropy": 5.632469844818115,
      "epoch": 0.297790585975024,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995398536501324,
      "loss": 5.4237,
      "mean_token_accuracy": 0.18980091959238052,
      "num_tokens": 7101843.0,
      "step": 3100
    },
    {
      "entropy": 5.602583503723144,
      "epoch": 0.29827089337175794,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995376595740801,
      "loss": 5.4945,
      "mean_token_accuracy": 0.18240708112716675,
      "num_tokens": 7112014.0,
      "step": 3105
    },
    {
      "entropy": 5.6829828262329105,
      "epoch": 0.2987512007684918,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995354602849302,
      "loss": 5.57,
      "mean_token_accuracy": 0.17665938287973404,
      "num_tokens": 7123860.0,
      "step": 3110
    },
    {
      "entropy": 5.68782525062561,
      "epoch": 0.29923150816522576,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995332557827337,
      "loss": 5.5429,
      "mean_token_accuracy": 0.1798730105161667,
      "num_tokens": 7135901.0,
      "step": 3115
    },
    {
      "entropy": 5.678207349777222,
      "epoch": 0.29971181556195964,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995310460675416,
      "loss": 5.4903,
      "mean_token_accuracy": 0.19226625710725784,
      "num_tokens": 7148743.0,
      "step": 3120
    },
    {
      "entropy": 5.718117713928223,
      "epoch": 0.3001921229586936,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995288311394053,
      "loss": 5.55,
      "mean_token_accuracy": 0.18411389291286467,
      "num_tokens": 7160731.0,
      "step": 3125
    },
    {
      "entropy": 5.695175981521606,
      "epoch": 0.30067243035542746,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004995266109983764,
      "loss": 5.5379,
      "mean_token_accuracy": 0.18051327168941497,
      "num_tokens": 7172861.0,
      "step": 3130
    },
    {
      "entropy": 5.663706064224243,
      "epoch": 0.3011527377521614,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995243856445062,
      "loss": 5.4866,
      "mean_token_accuracy": 0.1794595777988434,
      "num_tokens": 7183954.0,
      "step": 3135
    },
    {
      "entropy": 5.54837441444397,
      "epoch": 0.3016330451488953,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995221550778466,
      "loss": 5.455,
      "mean_token_accuracy": 0.18403159081935883,
      "num_tokens": 7195466.0,
      "step": 3140
    },
    {
      "entropy": 5.650097274780274,
      "epoch": 0.3021133525456292,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995199192984491,
      "loss": 5.445,
      "mean_token_accuracy": 0.18828442096710205,
      "num_tokens": 7207173.0,
      "step": 3145
    },
    {
      "entropy": 5.701932287216186,
      "epoch": 0.3025936599423631,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995176783063657,
      "loss": 5.5847,
      "mean_token_accuracy": 0.1841903731226921,
      "num_tokens": 7220095.0,
      "step": 3150
    },
    {
      "entropy": 5.633075094223022,
      "epoch": 0.30307396733909703,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995154321016487,
      "loss": 5.5079,
      "mean_token_accuracy": 0.1826324224472046,
      "num_tokens": 7230664.0,
      "step": 3155
    },
    {
      "entropy": 5.709785985946655,
      "epoch": 0.3035542747358309,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995131806843499,
      "loss": 5.4593,
      "mean_token_accuracy": 0.18663873821496962,
      "num_tokens": 7241278.0,
      "step": 3160
    },
    {
      "entropy": 5.560795783996582,
      "epoch": 0.30403458213256485,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995109240545218,
      "loss": 5.6204,
      "mean_token_accuracy": 0.1759345531463623,
      "num_tokens": 7252999.0,
      "step": 3165
    },
    {
      "entropy": 5.699498128890991,
      "epoch": 0.3045148895292987,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995086622122167,
      "loss": 5.4472,
      "mean_token_accuracy": 0.1813385471701622,
      "num_tokens": 7263949.0,
      "step": 3170
    },
    {
      "entropy": 5.637405109405518,
      "epoch": 0.30499519692603266,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995063951574871,
      "loss": 5.4896,
      "mean_token_accuracy": 0.18765443116426467,
      "num_tokens": 7275467.0,
      "step": 3175
    },
    {
      "entropy": 5.65950870513916,
      "epoch": 0.30547550432276654,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995041228903856,
      "loss": 5.4633,
      "mean_token_accuracy": 0.19078702926635743,
      "num_tokens": 7285534.0,
      "step": 3180
    },
    {
      "entropy": 5.745648288726807,
      "epoch": 0.3059558117195005,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499501845410965,
      "loss": 5.5733,
      "mean_token_accuracy": 0.1831236869096756,
      "num_tokens": 7297252.0,
      "step": 3185
    },
    {
      "entropy": 5.611358976364135,
      "epoch": 0.30643611911623436,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004994995627192781,
      "loss": 5.4468,
      "mean_token_accuracy": 0.18962527066469193,
      "num_tokens": 7308492.0,
      "step": 3190
    },
    {
      "entropy": 5.604902219772339,
      "epoch": 0.3069164265129683,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994972748153781,
      "loss": 5.4833,
      "mean_token_accuracy": 0.18148828595876693,
      "num_tokens": 7319703.0,
      "step": 3195
    },
    {
      "entropy": 5.685352182388305,
      "epoch": 0.30739673390970224,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499494981699318,
      "loss": 5.4493,
      "mean_token_accuracy": 0.19088191539049149,
      "num_tokens": 7331022.0,
      "step": 3200
    },
    {
      "entropy": 5.615092754364014,
      "epoch": 0.3078770413064361,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499492683371151,
      "loss": 5.4995,
      "mean_token_accuracy": 0.182320274412632,
      "num_tokens": 7342977.0,
      "step": 3205
    },
    {
      "entropy": 5.697661113739014,
      "epoch": 0.30835734870317005,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994903798309306,
      "loss": 5.4744,
      "mean_token_accuracy": 0.18243792206048964,
      "num_tokens": 7353227.0,
      "step": 3210
    },
    {
      "entropy": 5.6787608623504635,
      "epoch": 0.30883765609990393,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994880710787102,
      "loss": 5.5267,
      "mean_token_accuracy": 0.16992310285568238,
      "num_tokens": 7364165.0,
      "step": 3215
    },
    {
      "entropy": 5.5938324451446535,
      "epoch": 0.30931796349663787,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994857571145432,
      "loss": 5.476,
      "mean_token_accuracy": 0.1885462373495102,
      "num_tokens": 7374800.0,
      "step": 3220
    },
    {
      "entropy": 5.550031423568726,
      "epoch": 0.30979827089337175,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994834379384837,
      "loss": 5.4209,
      "mean_token_accuracy": 0.1900305286049843,
      "num_tokens": 7386360.0,
      "step": 3225
    },
    {
      "entropy": 5.643242835998535,
      "epoch": 0.3102785782901057,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994811135505851,
      "loss": 5.4517,
      "mean_token_accuracy": 0.1852445274591446,
      "num_tokens": 7397066.0,
      "step": 3230
    },
    {
      "entropy": 5.69812912940979,
      "epoch": 0.31075888568683957,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994787839509018,
      "loss": 5.5215,
      "mean_token_accuracy": 0.1741228461265564,
      "num_tokens": 7408349.0,
      "step": 3235
    },
    {
      "entropy": 5.671992778778076,
      "epoch": 0.3112391930835735,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994764491394876,
      "loss": 5.5781,
      "mean_token_accuracy": 0.17537114769220352,
      "num_tokens": 7420343.0,
      "step": 3240
    },
    {
      "entropy": 5.708970642089843,
      "epoch": 0.3117195004803074,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994741091163969,
      "loss": 5.4766,
      "mean_token_accuracy": 0.18881572633981705,
      "num_tokens": 7431683.0,
      "step": 3245
    },
    {
      "entropy": 5.49902081489563,
      "epoch": 0.3121998078770413,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499471763881684,
      "loss": 5.3839,
      "mean_token_accuracy": 0.18738958537578582,
      "num_tokens": 7443327.0,
      "step": 3250
    },
    {
      "entropy": 5.648282718658447,
      "epoch": 0.3126801152737752,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004994694134354031,
      "loss": 5.479,
      "mean_token_accuracy": 0.18875627666711808,
      "num_tokens": 7454002.0,
      "step": 3255
    },
    {
      "entropy": 5.579559707641602,
      "epoch": 0.31316042267050914,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499467057777609,
      "loss": 5.4831,
      "mean_token_accuracy": 0.1835149794816971,
      "num_tokens": 7464074.0,
      "step": 3260
    },
    {
      "entropy": 5.561008071899414,
      "epoch": 0.313640730067243,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994646969083565,
      "loss": 5.4059,
      "mean_token_accuracy": 0.19485885351896287,
      "num_tokens": 7475543.0,
      "step": 3265
    },
    {
      "entropy": 5.693277263641358,
      "epoch": 0.31412103746397696,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994623308277002,
      "loss": 5.471,
      "mean_token_accuracy": 0.18558466881513597,
      "num_tokens": 7486818.0,
      "step": 3270
    },
    {
      "entropy": 5.665642356872558,
      "epoch": 0.31460134486071084,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000499459959535695,
      "loss": 5.51,
      "mean_token_accuracy": 0.18450151830911637,
      "num_tokens": 7499046.0,
      "step": 3275
    },
    {
      "entropy": 5.634830045700073,
      "epoch": 0.3150816522574448,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994575830323962,
      "loss": 5.4406,
      "mean_token_accuracy": 0.17926069498062133,
      "num_tokens": 7509853.0,
      "step": 3280
    },
    {
      "entropy": 5.514278316497803,
      "epoch": 0.31556195965417866,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994552013178586,
      "loss": 5.3046,
      "mean_token_accuracy": 0.19069029539823532,
      "num_tokens": 7521091.0,
      "step": 3285
    },
    {
      "entropy": 5.596073818206787,
      "epoch": 0.3160422670509126,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499452814392138,
      "loss": 5.4375,
      "mean_token_accuracy": 0.19329715222120286,
      "num_tokens": 7531317.0,
      "step": 3290
    },
    {
      "entropy": 5.64233808517456,
      "epoch": 0.3165225744476465,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994504222552894,
      "loss": 5.5567,
      "mean_token_accuracy": 0.18084498941898347,
      "num_tokens": 7542822.0,
      "step": 3295
    },
    {
      "entropy": 5.722558689117432,
      "epoch": 0.3170028818443804,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994480249073684,
      "loss": 5.5067,
      "mean_token_accuracy": 0.18076840192079544,
      "num_tokens": 7552434.0,
      "step": 3300
    },
    {
      "entropy": 5.598572635650635,
      "epoch": 0.3174831892411143,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994456223484308,
      "loss": 5.3801,
      "mean_token_accuracy": 0.1927143394947052,
      "num_tokens": 7563895.0,
      "step": 3305
    },
    {
      "entropy": 5.454520845413208,
      "epoch": 0.31796349663784823,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994432145785323,
      "loss": 5.3961,
      "mean_token_accuracy": 0.18793661743402482,
      "num_tokens": 7575391.0,
      "step": 3310
    },
    {
      "entropy": 5.677673196792602,
      "epoch": 0.3184438040345821,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994408015977288,
      "loss": 5.5531,
      "mean_token_accuracy": 0.18976033180952073,
      "num_tokens": 7587119.0,
      "step": 3315
    },
    {
      "entropy": 5.694156169891357,
      "epoch": 0.31892411143131605,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994383834060764,
      "loss": 5.5264,
      "mean_token_accuracy": 0.17661510556936263,
      "num_tokens": 7598615.0,
      "step": 3320
    },
    {
      "entropy": 5.696703481674194,
      "epoch": 0.31940441882804993,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994359600036311,
      "loss": 5.5015,
      "mean_token_accuracy": 0.19055715948343277,
      "num_tokens": 7610159.0,
      "step": 3325
    },
    {
      "entropy": 5.683110094070434,
      "epoch": 0.31988472622478387,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994335313904493,
      "loss": 5.4601,
      "mean_token_accuracy": 0.18581361472606658,
      "num_tokens": 7620922.0,
      "step": 3330
    },
    {
      "entropy": 5.582138109207153,
      "epoch": 0.32036503362151775,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004994310975665873,
      "loss": 5.3831,
      "mean_token_accuracy": 0.19046609699726105,
      "num_tokens": 7632343.0,
      "step": 3335
    },
    {
      "entropy": 5.732189750671386,
      "epoch": 0.3208453410182517,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004994286585321017,
      "loss": 5.5902,
      "mean_token_accuracy": 0.17252802550792695,
      "num_tokens": 7644748.0,
      "step": 3340
    },
    {
      "entropy": 5.710432291030884,
      "epoch": 0.32132564841498557,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499426214287049,
      "loss": 5.5229,
      "mean_token_accuracy": 0.19114425033330917,
      "num_tokens": 7655449.0,
      "step": 3345
    },
    {
      "entropy": 5.64078426361084,
      "epoch": 0.3218059558117195,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994237648314862,
      "loss": 5.5035,
      "mean_token_accuracy": 0.18543385416269303,
      "num_tokens": 7665623.0,
      "step": 3350
    },
    {
      "entropy": 5.554352760314941,
      "epoch": 0.3222862632084534,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994213101654697,
      "loss": 5.3906,
      "mean_token_accuracy": 0.19078583717346193,
      "num_tokens": 7676860.0,
      "step": 3355
    },
    {
      "entropy": 5.686653995513916,
      "epoch": 0.3227665706051873,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499418850289057,
      "loss": 5.5767,
      "mean_token_accuracy": 0.18165411055088043,
      "num_tokens": 7687778.0,
      "step": 3360
    },
    {
      "entropy": 5.667432641983032,
      "epoch": 0.32324687800192126,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994163852023048,
      "loss": 5.4775,
      "mean_token_accuracy": 0.18488942384719848,
      "num_tokens": 7699154.0,
      "step": 3365
    },
    {
      "entropy": 5.591152572631836,
      "epoch": 0.32372718539865514,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994139149052706,
      "loss": 5.496,
      "mean_token_accuracy": 0.18603985756635666,
      "num_tokens": 7711010.0,
      "step": 3370
    },
    {
      "entropy": 5.668866777420044,
      "epoch": 0.3242074927953891,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994114393980117,
      "loss": 5.5134,
      "mean_token_accuracy": 0.18332684338092803,
      "num_tokens": 7721969.0,
      "step": 3375
    },
    {
      "entropy": 5.679357814788818,
      "epoch": 0.32468780019212296,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994089586805856,
      "loss": 5.4463,
      "mean_token_accuracy": 0.1889062002301216,
      "num_tokens": 7733762.0,
      "step": 3380
    },
    {
      "entropy": 5.640239667892456,
      "epoch": 0.3251681075888569,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994064727530496,
      "loss": 5.456,
      "mean_token_accuracy": 0.1815125659108162,
      "num_tokens": 7744614.0,
      "step": 3385
    },
    {
      "entropy": 5.564548492431641,
      "epoch": 0.3256484149855908,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004994039816154618,
      "loss": 5.4026,
      "mean_token_accuracy": 0.1951647639274597,
      "num_tokens": 7755799.0,
      "step": 3390
    },
    {
      "entropy": 5.5753570079803465,
      "epoch": 0.3261287223823247,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00049940148526788,
      "loss": 5.4578,
      "mean_token_accuracy": 0.1877640664577484,
      "num_tokens": 7768140.0,
      "step": 3395
    },
    {
      "entropy": 5.5929913997650145,
      "epoch": 0.3266090297790586,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993989837103618,
      "loss": 5.4482,
      "mean_token_accuracy": 0.1846913829445839,
      "num_tokens": 7778494.0,
      "step": 3400
    },
    {
      "entropy": 5.708591794967651,
      "epoch": 0.3270893371757925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993964769429657,
      "loss": 5.5226,
      "mean_token_accuracy": 0.19035352319478988,
      "num_tokens": 7789234.0,
      "step": 3405
    },
    {
      "entropy": 5.599575138092041,
      "epoch": 0.3275696445725264,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004993939649657498,
      "loss": 5.4999,
      "mean_token_accuracy": 0.1874484598636627,
      "num_tokens": 7800602.0,
      "step": 3410
    },
    {
      "entropy": 5.562217473983765,
      "epoch": 0.32804995196926034,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004993914477787721,
      "loss": 5.3376,
      "mean_token_accuracy": 0.19393931627273558,
      "num_tokens": 7812803.0,
      "step": 3415
    },
    {
      "entropy": 5.748968839645386,
      "epoch": 0.3285302593659942,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993889253820915,
      "loss": 5.6342,
      "mean_token_accuracy": 0.1738002836704254,
      "num_tokens": 7825432.0,
      "step": 3420
    },
    {
      "entropy": 5.6201917171478275,
      "epoch": 0.32901056676272816,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993863977757663,
      "loss": 5.4553,
      "mean_token_accuracy": 0.18626101016998292,
      "num_tokens": 7837258.0,
      "step": 3425
    },
    {
      "entropy": 5.545711612701416,
      "epoch": 0.32949087415946204,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993838649598552,
      "loss": 5.3317,
      "mean_token_accuracy": 0.1965160608291626,
      "num_tokens": 7847573.0,
      "step": 3430
    },
    {
      "entropy": 5.670653295516968,
      "epoch": 0.329971181556196,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993813269344171,
      "loss": 5.4727,
      "mean_token_accuracy": 0.18039093613624574,
      "num_tokens": 7857957.0,
      "step": 3435
    },
    {
      "entropy": 5.603359937667847,
      "epoch": 0.33045148895292986,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993787836995108,
      "loss": 5.3905,
      "mean_token_accuracy": 0.19513991624116897,
      "num_tokens": 7867996.0,
      "step": 3440
    },
    {
      "entropy": 5.527387523651123,
      "epoch": 0.3309317963496638,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993762352551954,
      "loss": 5.4387,
      "mean_token_accuracy": 0.1835164800286293,
      "num_tokens": 7879245.0,
      "step": 3445
    },
    {
      "entropy": 5.659226322174073,
      "epoch": 0.3314121037463977,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993736816015301,
      "loss": 5.5261,
      "mean_token_accuracy": 0.17599146217107772,
      "num_tokens": 7891186.0,
      "step": 3450
    },
    {
      "entropy": 5.6871460437774655,
      "epoch": 0.3318924111431316,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993711227385742,
      "loss": 5.5404,
      "mean_token_accuracy": 0.19238765239715577,
      "num_tokens": 7902231.0,
      "step": 3455
    },
    {
      "entropy": 5.626007223129273,
      "epoch": 0.3323727185398655,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993685586663871,
      "loss": 5.5157,
      "mean_token_accuracy": 0.1862488716840744,
      "num_tokens": 7913364.0,
      "step": 3460
    },
    {
      "entropy": 5.866927194595337,
      "epoch": 0.33285302593659943,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993659893850281,
      "loss": 5.6859,
      "mean_token_accuracy": 0.17142274528741835,
      "num_tokens": 7925217.0,
      "step": 3465
    },
    {
      "entropy": 5.592819738388061,
      "epoch": 0.3333333333333333,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004993634148945573,
      "loss": 5.4461,
      "mean_token_accuracy": 0.17677210867404938,
      "num_tokens": 7937636.0,
      "step": 3470
    },
    {
      "entropy": 5.596171140670776,
      "epoch": 0.33381364073006725,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993608351950341,
      "loss": 5.4622,
      "mean_token_accuracy": 0.18027846217155458,
      "num_tokens": 7948958.0,
      "step": 3475
    },
    {
      "entropy": 5.671500778198242,
      "epoch": 0.33429394812680113,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004993582502865185,
      "loss": 5.4924,
      "mean_token_accuracy": 0.18664607107639314,
      "num_tokens": 7960013.0,
      "step": 3480
    },
    {
      "entropy": 5.620531797409058,
      "epoch": 0.33477425552353507,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993556601690706,
      "loss": 5.5105,
      "mean_token_accuracy": 0.1812283620238304,
      "num_tokens": 7971041.0,
      "step": 3485
    },
    {
      "entropy": 5.660557413101197,
      "epoch": 0.33525456292026895,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993530648427505,
      "loss": 5.5333,
      "mean_token_accuracy": 0.18094693422317504,
      "num_tokens": 7982752.0,
      "step": 3490
    },
    {
      "entropy": 5.649030065536499,
      "epoch": 0.3357348703170029,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993504643076184,
      "loss": 5.4105,
      "mean_token_accuracy": 0.18630249798297882,
      "num_tokens": 7993681.0,
      "step": 3495
    },
    {
      "entropy": 5.596584844589233,
      "epoch": 0.33621517771373677,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993478585637347,
      "loss": 5.4389,
      "mean_token_accuracy": 0.18762964010238647,
      "num_tokens": 8004727.0,
      "step": 3500
    },
    {
      "entropy": 5.62527117729187,
      "epoch": 0.3366954851104707,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004993452476111599,
      "loss": 5.4575,
      "mean_token_accuracy": 0.19001320451498033,
      "num_tokens": 8015423.0,
      "step": 3505
    },
    {
      "entropy": 5.621340608596801,
      "epoch": 0.3371757925072046,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004993426314499546,
      "loss": 5.4243,
      "mean_token_accuracy": 0.19290892481803895,
      "num_tokens": 8027911.0,
      "step": 3510
    },
    {
      "entropy": 5.620342636108399,
      "epoch": 0.3376560999039385,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004993400100801796,
      "loss": 5.4526,
      "mean_token_accuracy": 0.18564309179782867,
      "num_tokens": 8038831.0,
      "step": 3515
    },
    {
      "entropy": 5.51328182220459,
      "epoch": 0.3381364073006724,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993373835018956,
      "loss": 5.3275,
      "mean_token_accuracy": 0.19699581563472748,
      "num_tokens": 8049906.0,
      "step": 3520
    },
    {
      "entropy": 5.513483619689941,
      "epoch": 0.33861671469740634,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004993347517151638,
      "loss": 5.4294,
      "mean_token_accuracy": 0.18792010992765426,
      "num_tokens": 8061158.0,
      "step": 3525
    },
    {
      "entropy": 5.646618938446045,
      "epoch": 0.3390970220941403,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993321147200452,
      "loss": 5.3849,
      "mean_token_accuracy": 0.18583539128303528,
      "num_tokens": 8071958.0,
      "step": 3530
    },
    {
      "entropy": 5.588662624359131,
      "epoch": 0.33957732949087416,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499329472516601,
      "loss": 5.3926,
      "mean_token_accuracy": 0.18485624343156815,
      "num_tokens": 8084068.0,
      "step": 3535
    },
    {
      "entropy": 5.4956859111785885,
      "epoch": 0.3400576368876081,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993268251048925,
      "loss": 5.3184,
      "mean_token_accuracy": 0.198596853017807,
      "num_tokens": 8096132.0,
      "step": 3540
    },
    {
      "entropy": 5.58132586479187,
      "epoch": 0.340537944284342,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993241724849814,
      "loss": 5.4759,
      "mean_token_accuracy": 0.1903732657432556,
      "num_tokens": 8107327.0,
      "step": 3545
    },
    {
      "entropy": 5.608106994628907,
      "epoch": 0.3410182516810759,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499321514656929,
      "loss": 5.4608,
      "mean_token_accuracy": 0.18965245932340621,
      "num_tokens": 8118584.0,
      "step": 3550
    },
    {
      "entropy": 5.643239259719849,
      "epoch": 0.3414985590778098,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993188516207972,
      "loss": 5.4951,
      "mean_token_accuracy": 0.18048142790794372,
      "num_tokens": 8130081.0,
      "step": 3555
    },
    {
      "entropy": 5.573701095581055,
      "epoch": 0.34197886647454373,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993161833766478,
      "loss": 5.4541,
      "mean_token_accuracy": 0.18741901069879532,
      "num_tokens": 8141463.0,
      "step": 3560
    },
    {
      "entropy": 5.632593965530395,
      "epoch": 0.3424591738712776,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993135099245426,
      "loss": 5.4959,
      "mean_token_accuracy": 0.18373015522956848,
      "num_tokens": 8153863.0,
      "step": 3565
    },
    {
      "entropy": 5.53346095085144,
      "epoch": 0.34293948126801155,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993108312645438,
      "loss": 5.4089,
      "mean_token_accuracy": 0.18515754789113997,
      "num_tokens": 8165695.0,
      "step": 3570
    },
    {
      "entropy": 5.614653158187866,
      "epoch": 0.34341978866474543,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993081473967135,
      "loss": 5.4701,
      "mean_token_accuracy": 0.1853488028049469,
      "num_tokens": 8176456.0,
      "step": 3575
    },
    {
      "entropy": 5.65216178894043,
      "epoch": 0.34390009606147937,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993054583211143,
      "loss": 5.4656,
      "mean_token_accuracy": 0.18880873173475266,
      "num_tokens": 8189050.0,
      "step": 3580
    },
    {
      "entropy": 5.548917770385742,
      "epoch": 0.34438040345821325,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004993027640378081,
      "loss": 5.3775,
      "mean_token_accuracy": 0.18787043541669846,
      "num_tokens": 8200011.0,
      "step": 3585
    },
    {
      "entropy": 5.564371919631958,
      "epoch": 0.3448607108549472,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499300064546858,
      "loss": 5.3903,
      "mean_token_accuracy": 0.19150739163160324,
      "num_tokens": 8211770.0,
      "step": 3590
    },
    {
      "entropy": 5.599589443206787,
      "epoch": 0.34534101825168106,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992973598483264,
      "loss": 5.4214,
      "mean_token_accuracy": 0.19162243753671646,
      "num_tokens": 8223582.0,
      "step": 3595
    },
    {
      "entropy": 5.68660888671875,
      "epoch": 0.345821325648415,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499294649942276,
      "loss": 5.5506,
      "mean_token_accuracy": 0.18258577585220337,
      "num_tokens": 8234336.0,
      "step": 3600
    },
    {
      "entropy": 5.620280456542969,
      "epoch": 0.3463016330451489,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992919348287699,
      "loss": 5.4521,
      "mean_token_accuracy": 0.18839440494775772,
      "num_tokens": 8244605.0,
      "step": 3605
    },
    {
      "entropy": 5.633523941040039,
      "epoch": 0.3467819404418828,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004992892145078711,
      "loss": 5.5118,
      "mean_token_accuracy": 0.18184645473957062,
      "num_tokens": 8255876.0,
      "step": 3610
    },
    {
      "entropy": 5.564020156860352,
      "epoch": 0.3472622478386167,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992864889796427,
      "loss": 5.3966,
      "mean_token_accuracy": 0.18950104862451553,
      "num_tokens": 8266602.0,
      "step": 3615
    },
    {
      "entropy": 5.597898054122925,
      "epoch": 0.34774255523535064,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992837582441481,
      "loss": 5.3843,
      "mean_token_accuracy": 0.19015932381153106,
      "num_tokens": 8279804.0,
      "step": 3620
    },
    {
      "entropy": 5.6535993099212645,
      "epoch": 0.3482228626320845,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004992810223014506,
      "loss": 5.4885,
      "mean_token_accuracy": 0.18701547533273696,
      "num_tokens": 8291020.0,
      "step": 3625
    },
    {
      "entropy": 5.622987365722656,
      "epoch": 0.34870317002881845,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992782811516137,
      "loss": 5.4299,
      "mean_token_accuracy": 0.19102343469858168,
      "num_tokens": 8302192.0,
      "step": 3630
    },
    {
      "entropy": 5.582995176315308,
      "epoch": 0.34918347742555234,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004992755347947011,
      "loss": 5.4053,
      "mean_token_accuracy": 0.1902191162109375,
      "num_tokens": 8313649.0,
      "step": 3635
    },
    {
      "entropy": 5.5383587837219235,
      "epoch": 0.34966378482228627,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992727832307766,
      "loss": 5.3974,
      "mean_token_accuracy": 0.18865602910518647,
      "num_tokens": 8324694.0,
      "step": 3640
    },
    {
      "entropy": 5.71724762916565,
      "epoch": 0.35014409221902015,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992700264599039,
      "loss": 5.5476,
      "mean_token_accuracy": 0.18377173095941543,
      "num_tokens": 8336517.0,
      "step": 3645
    },
    {
      "entropy": 5.599456405639648,
      "epoch": 0.3506243996157541,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004992672644821473,
      "loss": 5.5135,
      "mean_token_accuracy": 0.18044322580099106,
      "num_tokens": 8349001.0,
      "step": 3650
    },
    {
      "entropy": 5.642662906646729,
      "epoch": 0.35110470701248797,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992644972975707,
      "loss": 5.4246,
      "mean_token_accuracy": 0.19142407476902007,
      "num_tokens": 8361230.0,
      "step": 3655
    },
    {
      "entropy": 5.532097005844117,
      "epoch": 0.3515850144092219,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004992617249062383,
      "loss": 5.3433,
      "mean_token_accuracy": 0.19701765477657318,
      "num_tokens": 8372159.0,
      "step": 3660
    },
    {
      "entropy": 5.58645806312561,
      "epoch": 0.3520653218059558,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004992589473082147,
      "loss": 5.4707,
      "mean_token_accuracy": 0.18735242784023284,
      "num_tokens": 8383228.0,
      "step": 3665
    },
    {
      "entropy": 5.640217876434326,
      "epoch": 0.3525456292026897,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992561645035641,
      "loss": 5.4179,
      "mean_token_accuracy": 0.18576795011758804,
      "num_tokens": 8394582.0,
      "step": 3670
    },
    {
      "entropy": 5.58097825050354,
      "epoch": 0.3530259365994236,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992533764923515,
      "loss": 5.4246,
      "mean_token_accuracy": 0.187009534239769,
      "num_tokens": 8406784.0,
      "step": 3675
    },
    {
      "entropy": 5.567069339752197,
      "epoch": 0.35350624399615754,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004992505832746412,
      "loss": 5.4055,
      "mean_token_accuracy": 0.19539184421300887,
      "num_tokens": 8418405.0,
      "step": 3680
    },
    {
      "entropy": 5.581828355789185,
      "epoch": 0.3539865513928914,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004992477848504983,
      "loss": 5.3563,
      "mean_token_accuracy": 0.19186690151691438,
      "num_tokens": 8430432.0,
      "step": 3685
    },
    {
      "entropy": 5.608822822570801,
      "epoch": 0.35446685878962536,
      "grad_norm": 0.875,
      "learning_rate": 0.0004992449812199877,
      "loss": 5.5098,
      "mean_token_accuracy": 0.18246612846851348,
      "num_tokens": 8442423.0,
      "step": 3690
    },
    {
      "entropy": 5.603720855712891,
      "epoch": 0.3549471661863593,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004992421723831745,
      "loss": 5.512,
      "mean_token_accuracy": 0.1835420086979866,
      "num_tokens": 8454951.0,
      "step": 3695
    },
    {
      "entropy": 5.596175956726074,
      "epoch": 0.3554274735830932,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992393583401239,
      "loss": 5.368,
      "mean_token_accuracy": 0.19326992928981782,
      "num_tokens": 8467758.0,
      "step": 3700
    },
    {
      "entropy": 5.587682628631592,
      "epoch": 0.3559077809798271,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992365390909014,
      "loss": 5.4549,
      "mean_token_accuracy": 0.18845424950122833,
      "num_tokens": 8479728.0,
      "step": 3705
    },
    {
      "entropy": 5.616272878646851,
      "epoch": 0.356388088376561,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004992337146355721,
      "loss": 5.504,
      "mean_token_accuracy": 0.18585399091243743,
      "num_tokens": 8492099.0,
      "step": 3710
    },
    {
      "entropy": 5.6486982822418215,
      "epoch": 0.35686839577329493,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004992308849742019,
      "loss": 5.4578,
      "mean_token_accuracy": 0.17614103853702545,
      "num_tokens": 8504657.0,
      "step": 3715
    },
    {
      "entropy": 5.613508033752441,
      "epoch": 0.3573487031700288,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992280501068563,
      "loss": 5.4172,
      "mean_token_accuracy": 0.19100760966539382,
      "num_tokens": 8514728.0,
      "step": 3720
    },
    {
      "entropy": 5.631121015548706,
      "epoch": 0.35782901056676275,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992252100336012,
      "loss": 5.5379,
      "mean_token_accuracy": 0.1856612503528595,
      "num_tokens": 8525588.0,
      "step": 3725
    },
    {
      "entropy": 5.616485548019409,
      "epoch": 0.35830931796349663,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992223647545027,
      "loss": 5.4989,
      "mean_token_accuracy": 0.18350607752799988,
      "num_tokens": 8537468.0,
      "step": 3730
    },
    {
      "entropy": 5.631916236877442,
      "epoch": 0.35878962536023057,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004992195142696266,
      "loss": 5.4096,
      "mean_token_accuracy": 0.1893299102783203,
      "num_tokens": 8548598.0,
      "step": 3735
    },
    {
      "entropy": 5.418419742584229,
      "epoch": 0.35926993275696445,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992166585790391,
      "loss": 5.2948,
      "mean_token_accuracy": 0.20219719260931016,
      "num_tokens": 8560301.0,
      "step": 3740
    },
    {
      "entropy": 5.572213268280029,
      "epoch": 0.3597502401536984,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992137976828067,
      "loss": 5.3931,
      "mean_token_accuracy": 0.19558291882276535,
      "num_tokens": 8571186.0,
      "step": 3745
    },
    {
      "entropy": 5.578411054611206,
      "epoch": 0.36023054755043227,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004992109315809955,
      "loss": 5.3799,
      "mean_token_accuracy": 0.1940329760313034,
      "num_tokens": 8580725.0,
      "step": 3750
    },
    {
      "entropy": 5.555987167358398,
      "epoch": 0.3607108549471662,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004992080602736725,
      "loss": 5.5245,
      "mean_token_accuracy": 0.180860997736454,
      "num_tokens": 8594598.0,
      "step": 3755
    },
    {
      "entropy": 5.739662742614746,
      "epoch": 0.3611911623439001,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004992051837609039,
      "loss": 5.5226,
      "mean_token_accuracy": 0.18189345002174379,
      "num_tokens": 8606733.0,
      "step": 3760
    },
    {
      "entropy": 5.661942768096924,
      "epoch": 0.361671469740634,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992023020427568,
      "loss": 5.4456,
      "mean_token_accuracy": 0.19026065915822982,
      "num_tokens": 8618863.0,
      "step": 3765
    },
    {
      "entropy": 5.462978744506836,
      "epoch": 0.3621517771373679,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991994151192979,
      "loss": 5.291,
      "mean_token_accuracy": 0.19249616265296937,
      "num_tokens": 8629270.0,
      "step": 3770
    },
    {
      "entropy": 5.561847591400147,
      "epoch": 0.36263208453410184,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991965229905943,
      "loss": 5.5043,
      "mean_token_accuracy": 0.1881070166826248,
      "num_tokens": 8641363.0,
      "step": 3775
    },
    {
      "entropy": 5.688853645324707,
      "epoch": 0.3631123919308357,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991936256567133,
      "loss": 5.4613,
      "mean_token_accuracy": 0.18743191063404083,
      "num_tokens": 8653233.0,
      "step": 3780
    },
    {
      "entropy": 5.547352027893067,
      "epoch": 0.36359269932756966,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000499190723117722,
      "loss": 5.4458,
      "mean_token_accuracy": 0.1837211713194847,
      "num_tokens": 8665192.0,
      "step": 3785
    },
    {
      "entropy": 5.648187112808228,
      "epoch": 0.36407300672430354,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991878153736877,
      "loss": 5.5293,
      "mean_token_accuracy": 0.17802536785602568,
      "num_tokens": 8677669.0,
      "step": 3790
    },
    {
      "entropy": 5.492604207992554,
      "epoch": 0.3645533141210375,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991849024246781,
      "loss": 5.329,
      "mean_token_accuracy": 0.1957615062594414,
      "num_tokens": 8688002.0,
      "step": 3795
    },
    {
      "entropy": 5.566503572463989,
      "epoch": 0.36503362151777136,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991819842707608,
      "loss": 5.3763,
      "mean_token_accuracy": 0.19245349913835524,
      "num_tokens": 8698396.0,
      "step": 3800
    },
    {
      "entropy": 5.637855195999146,
      "epoch": 0.3655139289145053,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991790609120035,
      "loss": 5.3994,
      "mean_token_accuracy": 0.18937206119298935,
      "num_tokens": 8711135.0,
      "step": 3805
    },
    {
      "entropy": 5.548817920684814,
      "epoch": 0.3659942363112392,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499176132348474,
      "loss": 5.4421,
      "mean_token_accuracy": 0.19292772859334945,
      "num_tokens": 8723707.0,
      "step": 3810
    },
    {
      "entropy": 5.611570119857788,
      "epoch": 0.3664745437079731,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991731985802405,
      "loss": 5.3797,
      "mean_token_accuracy": 0.194192536175251,
      "num_tokens": 8734193.0,
      "step": 3815
    },
    {
      "entropy": 5.5339254379272464,
      "epoch": 0.366954851104707,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991702596073708,
      "loss": 5.4636,
      "mean_token_accuracy": 0.18580521792173385,
      "num_tokens": 8745619.0,
      "step": 3820
    },
    {
      "entropy": 5.510940504074097,
      "epoch": 0.36743515850144093,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004991673154299335,
      "loss": 5.368,
      "mean_token_accuracy": 0.1902906373143196,
      "num_tokens": 8757331.0,
      "step": 3825
    },
    {
      "entropy": 5.5497801303863525,
      "epoch": 0.3679154658981748,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004991643660479967,
      "loss": 5.3793,
      "mean_token_accuracy": 0.19248363822698594,
      "num_tokens": 8768840.0,
      "step": 3830
    },
    {
      "entropy": 5.608413505554199,
      "epoch": 0.36839577329490875,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991614114616289,
      "loss": 5.3387,
      "mean_token_accuracy": 0.19261340647935868,
      "num_tokens": 8781214.0,
      "step": 3835
    },
    {
      "entropy": 5.616617870330811,
      "epoch": 0.3688760806916426,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004991584516708988,
      "loss": 5.4095,
      "mean_token_accuracy": 0.18974703997373582,
      "num_tokens": 8791645.0,
      "step": 3840
    },
    {
      "entropy": 5.630726337432861,
      "epoch": 0.36935638808837656,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991554866758751,
      "loss": 5.5918,
      "mean_token_accuracy": 0.18070310354232788,
      "num_tokens": 8803286.0,
      "step": 3845
    },
    {
      "entropy": 5.639148473739624,
      "epoch": 0.36983669548511044,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991525164766265,
      "loss": 5.3792,
      "mean_token_accuracy": 0.18947552740573884,
      "num_tokens": 8814207.0,
      "step": 3850
    },
    {
      "entropy": 5.567518186569214,
      "epoch": 0.3703170028818444,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004991495410732222,
      "loss": 5.4329,
      "mean_token_accuracy": 0.18093401342630386,
      "num_tokens": 8825540.0,
      "step": 3855
    },
    {
      "entropy": 5.613472604751587,
      "epoch": 0.37079731027857826,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991465604657311,
      "loss": 5.5639,
      "mean_token_accuracy": 0.1754475399851799,
      "num_tokens": 8838182.0,
      "step": 3860
    },
    {
      "entropy": 5.599302816390991,
      "epoch": 0.3712776176753122,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991435746542224,
      "loss": 5.4306,
      "mean_token_accuracy": 0.19523921459913254,
      "num_tokens": 8850211.0,
      "step": 3865
    },
    {
      "entropy": 5.560464334487915,
      "epoch": 0.37175792507204614,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004991405836387655,
      "loss": 5.4749,
      "mean_token_accuracy": 0.1841433361172676,
      "num_tokens": 8862804.0,
      "step": 3870
    },
    {
      "entropy": 5.578405046463013,
      "epoch": 0.37223823246878,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991375874194298,
      "loss": 5.4249,
      "mean_token_accuracy": 0.18286560624837875,
      "num_tokens": 8874112.0,
      "step": 3875
    },
    {
      "entropy": 5.550269412994385,
      "epoch": 0.37271853986551395,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499134585996285,
      "loss": 5.4388,
      "mean_token_accuracy": 0.19159672558307647,
      "num_tokens": 8885114.0,
      "step": 3880
    },
    {
      "entropy": 5.586460971832276,
      "epoch": 0.37319884726224783,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991315793694004,
      "loss": 5.3307,
      "mean_token_accuracy": 0.192276206612587,
      "num_tokens": 8895555.0,
      "step": 3885
    },
    {
      "entropy": 5.524693441390991,
      "epoch": 0.37367915465898177,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991285675388463,
      "loss": 5.3293,
      "mean_token_accuracy": 0.20285780131816863,
      "num_tokens": 8906073.0,
      "step": 3890
    },
    {
      "entropy": 5.544021129608154,
      "epoch": 0.37415946205571565,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991255505046922,
      "loss": 5.4864,
      "mean_token_accuracy": 0.18590356409549713,
      "num_tokens": 8916587.0,
      "step": 3895
    },
    {
      "entropy": 5.61430516242981,
      "epoch": 0.3746397694524496,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991225282670083,
      "loss": 5.3813,
      "mean_token_accuracy": 0.18587377518415452,
      "num_tokens": 8927923.0,
      "step": 3900
    },
    {
      "entropy": 5.538893175125122,
      "epoch": 0.37512007684918347,
      "grad_norm": 1.0,
      "learning_rate": 0.000499119500825865,
      "loss": 5.4267,
      "mean_token_accuracy": 0.19215198904275893,
      "num_tokens": 8939939.0,
      "step": 3905
    },
    {
      "entropy": 5.523986291885376,
      "epoch": 0.3756003842459174,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991164681813323,
      "loss": 5.4006,
      "mean_token_accuracy": 0.19619202464818955,
      "num_tokens": 8951748.0,
      "step": 3910
    },
    {
      "entropy": 5.533074951171875,
      "epoch": 0.3760806916426513,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991134303334807,
      "loss": 5.3196,
      "mean_token_accuracy": 0.19790138006210328,
      "num_tokens": 8962922.0,
      "step": 3915
    },
    {
      "entropy": 5.467417430877686,
      "epoch": 0.3765609990393852,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004991103872823807,
      "loss": 5.3028,
      "mean_token_accuracy": 0.19722643196582795,
      "num_tokens": 8974013.0,
      "step": 3920
    },
    {
      "entropy": 5.515323543548584,
      "epoch": 0.3770413064361191,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499107339028103,
      "loss": 5.3855,
      "mean_token_accuracy": 0.1921772316098213,
      "num_tokens": 8986032.0,
      "step": 3925
    },
    {
      "entropy": 5.610967826843262,
      "epoch": 0.37752161383285304,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004991042855707184,
      "loss": 5.3714,
      "mean_token_accuracy": 0.1876501262187958,
      "num_tokens": 8996889.0,
      "step": 3930
    },
    {
      "entropy": 5.524027967453003,
      "epoch": 0.3780019212295869,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991012269102977,
      "loss": 5.3662,
      "mean_token_accuracy": 0.18921342343091965,
      "num_tokens": 9007594.0,
      "step": 3935
    },
    {
      "entropy": 5.449644565582275,
      "epoch": 0.37848222862632086,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990981630469119,
      "loss": 5.3452,
      "mean_token_accuracy": 0.19195079207420349,
      "num_tokens": 9018097.0,
      "step": 3940
    },
    {
      "entropy": 5.576449728012085,
      "epoch": 0.37896253602305474,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990950939806323,
      "loss": 5.4713,
      "mean_token_accuracy": 0.18259283453226088,
      "num_tokens": 9029554.0,
      "step": 3945
    },
    {
      "entropy": 5.5739020824432375,
      "epoch": 0.3794428434197887,
      "grad_norm": 0.9375,
      "learning_rate": 0.00049909201971153,
      "loss": 5.34,
      "mean_token_accuracy": 0.1885339617729187,
      "num_tokens": 9042518.0,
      "step": 3950
    },
    {
      "entropy": 5.542482852935791,
      "epoch": 0.37992315081652256,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004990889402396763,
      "loss": 5.392,
      "mean_token_accuracy": 0.19385820478200913,
      "num_tokens": 9054524.0,
      "step": 3955
    },
    {
      "entropy": 5.561400604248047,
      "epoch": 0.3804034582132565,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004990858555651431,
      "loss": 5.3564,
      "mean_token_accuracy": 0.1895818293094635,
      "num_tokens": 9065375.0,
      "step": 3960
    },
    {
      "entropy": 5.515637397766113,
      "epoch": 0.3808837656099904,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990827656880015,
      "loss": 5.3131,
      "mean_token_accuracy": 0.1950634926557541,
      "num_tokens": 9076338.0,
      "step": 3965
    },
    {
      "entropy": 5.506933546066284,
      "epoch": 0.3813640730067243,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004990796706083235,
      "loss": 5.3694,
      "mean_token_accuracy": 0.188057541847229,
      "num_tokens": 9088407.0,
      "step": 3970
    },
    {
      "entropy": 5.518592023849488,
      "epoch": 0.3818443804034582,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004990765703261809,
      "loss": 5.3255,
      "mean_token_accuracy": 0.19157813787460326,
      "num_tokens": 9099833.0,
      "step": 3975
    },
    {
      "entropy": 5.4132482528686525,
      "epoch": 0.38232468780019213,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004990734648416458,
      "loss": 5.2993,
      "mean_token_accuracy": 0.19470689445734024,
      "num_tokens": 9111126.0,
      "step": 3980
    },
    {
      "entropy": 5.551502323150634,
      "epoch": 0.382804995196926,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990703541547901,
      "loss": 5.3853,
      "mean_token_accuracy": 0.19405279755592347,
      "num_tokens": 9121979.0,
      "step": 3985
    },
    {
      "entropy": 5.61278281211853,
      "epoch": 0.38328530259365995,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990672382656863,
      "loss": 5.4075,
      "mean_token_accuracy": 0.1920866474509239,
      "num_tokens": 9132929.0,
      "step": 3990
    },
    {
      "entropy": 5.5851963520050045,
      "epoch": 0.38376560999039383,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004990641171744064,
      "loss": 5.3742,
      "mean_token_accuracy": 0.19273888766765596,
      "num_tokens": 9143903.0,
      "step": 3995
    },
    {
      "entropy": 5.577231502532959,
      "epoch": 0.38424591738712777,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004990609908810231,
      "loss": 5.4469,
      "mean_token_accuracy": 0.18918763995170593,
      "num_tokens": 9154416.0,
      "step": 4000
    },
    {
      "entropy": 5.5459951877594,
      "epoch": 0.38472622478386165,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004990578593856089,
      "loss": 5.4271,
      "mean_token_accuracy": 0.18706282377243041,
      "num_tokens": 9165613.0,
      "step": 4005
    },
    {
      "entropy": 5.517359685897827,
      "epoch": 0.3852065321805956,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990547226882366,
      "loss": 5.3865,
      "mean_token_accuracy": 0.19604209959506988,
      "num_tokens": 9177884.0,
      "step": 4010
    },
    {
      "entropy": 5.608504676818848,
      "epoch": 0.38568683957732947,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004990515807889788,
      "loss": 5.5114,
      "mean_token_accuracy": 0.18097308576107024,
      "num_tokens": 9190041.0,
      "step": 4015
    },
    {
      "entropy": 5.668904542922974,
      "epoch": 0.3861671469740634,
      "grad_norm": 1.0,
      "learning_rate": 0.0004990484336879087,
      "loss": 5.4049,
      "mean_token_accuracy": 0.19387364238500596,
      "num_tokens": 9202390.0,
      "step": 4020
    },
    {
      "entropy": 5.502331256866455,
      "epoch": 0.3866474543707973,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990452813850992,
      "loss": 5.4057,
      "mean_token_accuracy": 0.18983091711997985,
      "num_tokens": 9213437.0,
      "step": 4025
    },
    {
      "entropy": 5.569727373123169,
      "epoch": 0.3871277617675312,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004990421238806236,
      "loss": 5.4686,
      "mean_token_accuracy": 0.1795946553349495,
      "num_tokens": 9226310.0,
      "step": 4030
    },
    {
      "entropy": 5.586350727081299,
      "epoch": 0.38760806916426516,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990389611745551,
      "loss": 5.3893,
      "mean_token_accuracy": 0.19350333213806153,
      "num_tokens": 9236271.0,
      "step": 4035
    },
    {
      "entropy": 5.618629360198975,
      "epoch": 0.38808837656099904,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004990357932669672,
      "loss": 5.4615,
      "mean_token_accuracy": 0.1930390328168869,
      "num_tokens": 9247755.0,
      "step": 4040
    },
    {
      "entropy": 5.52394847869873,
      "epoch": 0.388568683957733,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990326201579335,
      "loss": 5.309,
      "mean_token_accuracy": 0.2017490491271019,
      "num_tokens": 9259821.0,
      "step": 4045
    },
    {
      "entropy": 5.552427816390991,
      "epoch": 0.38904899135446686,
      "grad_norm": 1.125,
      "learning_rate": 0.0004990294418475274,
      "loss": 5.4188,
      "mean_token_accuracy": 0.1905425786972046,
      "num_tokens": 9270663.0,
      "step": 4050
    },
    {
      "entropy": 5.550879764556885,
      "epoch": 0.3895292987512008,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004990262583358231,
      "loss": 5.4288,
      "mean_token_accuracy": 0.1851213574409485,
      "num_tokens": 9282588.0,
      "step": 4055
    },
    {
      "entropy": 5.5647125244140625,
      "epoch": 0.3900096061479347,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004990230696228943,
      "loss": 5.4029,
      "mean_token_accuracy": 0.18009695112705232,
      "num_tokens": 9293368.0,
      "step": 4060
    },
    {
      "entropy": 5.598783493041992,
      "epoch": 0.3904899135446686,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004990198757088149,
      "loss": 5.4702,
      "mean_token_accuracy": 0.18485325574874878,
      "num_tokens": 9305962.0,
      "step": 4065
    },
    {
      "entropy": 5.559340906143189,
      "epoch": 0.3909702209414025,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004990166765936593,
      "loss": 5.3495,
      "mean_token_accuracy": 0.1973833203315735,
      "num_tokens": 9317955.0,
      "step": 4070
    },
    {
      "entropy": 5.53150486946106,
      "epoch": 0.3914505283381364,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990134722775016,
      "loss": 5.3658,
      "mean_token_accuracy": 0.19113391786813735,
      "num_tokens": 9329491.0,
      "step": 4075
    },
    {
      "entropy": 5.537538051605225,
      "epoch": 0.3919308357348703,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004990102627604162,
      "loss": 5.483,
      "mean_token_accuracy": 0.19353909343481063,
      "num_tokens": 9341612.0,
      "step": 4080
    },
    {
      "entropy": 5.59714822769165,
      "epoch": 0.39241114313160425,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990070480424778,
      "loss": 5.4206,
      "mean_token_accuracy": 0.18844473958015442,
      "num_tokens": 9352302.0,
      "step": 4085
    },
    {
      "entropy": 5.568612432479858,
      "epoch": 0.3928914505283381,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004990038281237608,
      "loss": 5.3716,
      "mean_token_accuracy": 0.1887407511472702,
      "num_tokens": 9363303.0,
      "step": 4090
    },
    {
      "entropy": 5.498561525344849,
      "epoch": 0.39337175792507206,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990006030043401,
      "loss": 5.3233,
      "mean_token_accuracy": 0.18864162415266036,
      "num_tokens": 9375878.0,
      "step": 4095
    },
    {
      "entropy": 5.547961282730102,
      "epoch": 0.39385206532180594,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989973726842906,
      "loss": 5.3781,
      "mean_token_accuracy": 0.18958245515823363,
      "num_tokens": 9388342.0,
      "step": 4100
    },
    {
      "entropy": 5.48880181312561,
      "epoch": 0.3943323727185399,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989941371636872,
      "loss": 5.279,
      "mean_token_accuracy": 0.19420120269060134,
      "num_tokens": 9399047.0,
      "step": 4105
    },
    {
      "entropy": 5.567055034637451,
      "epoch": 0.39481268011527376,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004989908964426051,
      "loss": 5.3882,
      "mean_token_accuracy": 0.19254230260848998,
      "num_tokens": 9410172.0,
      "step": 4110
    },
    {
      "entropy": 5.607863855361939,
      "epoch": 0.3952929875120077,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004989876505211194,
      "loss": 5.5177,
      "mean_token_accuracy": 0.18579558283090591,
      "num_tokens": 9422287.0,
      "step": 4115
    },
    {
      "entropy": 5.619776344299316,
      "epoch": 0.3957732949087416,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004989843993993056,
      "loss": 5.394,
      "mean_token_accuracy": 0.19312429577112197,
      "num_tokens": 9433709.0,
      "step": 4120
    },
    {
      "entropy": 5.447456026077271,
      "epoch": 0.3962536023054755,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989811430772392,
      "loss": 5.2821,
      "mean_token_accuracy": 0.19651771783828736,
      "num_tokens": 9445138.0,
      "step": 4125
    },
    {
      "entropy": 5.522433710098267,
      "epoch": 0.3967339097022094,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989778815549957,
      "loss": 5.4104,
      "mean_token_accuracy": 0.19089603275060654,
      "num_tokens": 9455263.0,
      "step": 4130
    },
    {
      "entropy": 5.561570167541504,
      "epoch": 0.39721421709894333,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989746148326508,
      "loss": 5.3728,
      "mean_token_accuracy": 0.18802272528409958,
      "num_tokens": 9465491.0,
      "step": 4135
    },
    {
      "entropy": 5.456939458847046,
      "epoch": 0.3976945244956772,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004989713429102805,
      "loss": 5.3368,
      "mean_token_accuracy": 0.18885795921087264,
      "num_tokens": 9477601.0,
      "step": 4140
    },
    {
      "entropy": 5.559215688705445,
      "epoch": 0.39817483189241115,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004989680657879607,
      "loss": 5.4053,
      "mean_token_accuracy": 0.1909678354859352,
      "num_tokens": 9489385.0,
      "step": 4145
    },
    {
      "entropy": 5.4989770412445065,
      "epoch": 0.39865513928914503,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004989647834657675,
      "loss": 5.2718,
      "mean_token_accuracy": 0.20021265596151352,
      "num_tokens": 9501131.0,
      "step": 4150
    },
    {
      "entropy": 5.475941276550293,
      "epoch": 0.39913544668587897,
      "grad_norm": 0.921875,
      "learning_rate": 0.000498961495943777,
      "loss": 5.428,
      "mean_token_accuracy": 0.19261130541563035,
      "num_tokens": 9513094.0,
      "step": 4155
    },
    {
      "entropy": 5.646708154678345,
      "epoch": 0.39961575408261285,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004989582032220656,
      "loss": 5.5307,
      "mean_token_accuracy": 0.18555604368448259,
      "num_tokens": 9524538.0,
      "step": 4160
    },
    {
      "entropy": 5.593398761749268,
      "epoch": 0.4000960614793468,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989549053007096,
      "loss": 5.3436,
      "mean_token_accuracy": 0.19916174411773682,
      "num_tokens": 9535284.0,
      "step": 4165
    },
    {
      "entropy": 5.515568971633911,
      "epoch": 0.40057636887608067,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004989516021797858,
      "loss": 5.4242,
      "mean_token_accuracy": 0.1882516548037529,
      "num_tokens": 9546472.0,
      "step": 4170
    },
    {
      "entropy": 5.554485511779785,
      "epoch": 0.4010566762728146,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000498948293859371,
      "loss": 5.4202,
      "mean_token_accuracy": 0.18679715245962142,
      "num_tokens": 9558358.0,
      "step": 4175
    },
    {
      "entropy": 5.641470861434937,
      "epoch": 0.4015369836695485,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004989449803395415,
      "loss": 5.4463,
      "mean_token_accuracy": 0.19269262552261351,
      "num_tokens": 9570653.0,
      "step": 4180
    },
    {
      "entropy": 5.546628665924072,
      "epoch": 0.4020172910662824,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004989416616203747,
      "loss": 5.3869,
      "mean_token_accuracy": 0.1908806025981903,
      "num_tokens": 9582150.0,
      "step": 4185
    },
    {
      "entropy": 5.560229682922364,
      "epoch": 0.4024975984630163,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004989383377019476,
      "loss": 5.3314,
      "mean_token_accuracy": 0.19352059811353683,
      "num_tokens": 9592462.0,
      "step": 4190
    },
    {
      "entropy": 5.506600999832154,
      "epoch": 0.40297790585975024,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004989350085843371,
      "loss": 5.3262,
      "mean_token_accuracy": 0.19509673863649368,
      "num_tokens": 9604027.0,
      "step": 4195
    },
    {
      "entropy": 5.445934009552002,
      "epoch": 0.4034582132564842,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004989316742676207,
      "loss": 5.3209,
      "mean_token_accuracy": 0.19528971016407012,
      "num_tokens": 9616325.0,
      "step": 4200
    },
    {
      "entropy": 5.4403135776519775,
      "epoch": 0.40393852065321806,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004989283347518757,
      "loss": 5.2888,
      "mean_token_accuracy": 0.19427725523710251,
      "num_tokens": 9628133.0,
      "step": 4205
    },
    {
      "entropy": 5.6427380561828615,
      "epoch": 0.404418828049952,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989249900371797,
      "loss": 5.4941,
      "mean_token_accuracy": 0.1834379181265831,
      "num_tokens": 9639686.0,
      "step": 4210
    },
    {
      "entropy": 5.576334476470947,
      "epoch": 0.4048991354466859,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989216401236103,
      "loss": 5.3872,
      "mean_token_accuracy": 0.19191829711198807,
      "num_tokens": 9650222.0,
      "step": 4215
    },
    {
      "entropy": 5.450829315185547,
      "epoch": 0.4053794428434198,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989182850112455,
      "loss": 5.3003,
      "mean_token_accuracy": 0.2054572194814682,
      "num_tokens": 9661792.0,
      "step": 4220
    },
    {
      "entropy": 5.533667945861817,
      "epoch": 0.4058597502401537,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004989149247001629,
      "loss": 5.4054,
      "mean_token_accuracy": 0.18864203095436097,
      "num_tokens": 9673000.0,
      "step": 4225
    },
    {
      "entropy": 5.57403678894043,
      "epoch": 0.40634005763688763,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989115591904407,
      "loss": 5.3594,
      "mean_token_accuracy": 0.19470216035842897,
      "num_tokens": 9685253.0,
      "step": 4230
    },
    {
      "entropy": 5.4957459449768065,
      "epoch": 0.4068203650336215,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004989081884821569,
      "loss": 5.3394,
      "mean_token_accuracy": 0.18996310979127884,
      "num_tokens": 9697245.0,
      "step": 4235
    },
    {
      "entropy": 5.490689039230347,
      "epoch": 0.40730067243035545,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004989048125753899,
      "loss": 5.3659,
      "mean_token_accuracy": 0.19019463509321213,
      "num_tokens": 9710095.0,
      "step": 4240
    },
    {
      "entropy": 5.472829914093017,
      "epoch": 0.40778097982708933,
      "grad_norm": 0.984375,
      "learning_rate": 0.000498901431470218,
      "loss": 5.2533,
      "mean_token_accuracy": 0.20080252438783647,
      "num_tokens": 9721488.0,
      "step": 4245
    },
    {
      "entropy": 5.3933182716369625,
      "epoch": 0.40826128722382327,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004988980451667198,
      "loss": 5.2043,
      "mean_token_accuracy": 0.1991969019174576,
      "num_tokens": 9733280.0,
      "step": 4250
    },
    {
      "entropy": 5.4942357540130615,
      "epoch": 0.40874159462055715,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004988946536649737,
      "loss": 5.3359,
      "mean_token_accuracy": 0.18908474445343018,
      "num_tokens": 9744514.0,
      "step": 4255
    },
    {
      "entropy": 5.509702730178833,
      "epoch": 0.4092219020172911,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988912569650585,
      "loss": 5.3416,
      "mean_token_accuracy": 0.19426414370536804,
      "num_tokens": 9754931.0,
      "step": 4260
    },
    {
      "entropy": 5.42921404838562,
      "epoch": 0.40970220941402496,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004988878550670533,
      "loss": 5.3049,
      "mean_token_accuracy": 0.19834061712026596,
      "num_tokens": 9765635.0,
      "step": 4265
    },
    {
      "entropy": 5.563821077346802,
      "epoch": 0.4101825168107589,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988844479710369,
      "loss": 5.4092,
      "mean_token_accuracy": 0.18279904276132583,
      "num_tokens": 9777512.0,
      "step": 4270
    },
    {
      "entropy": 5.6497407913208,
      "epoch": 0.4106628242074928,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004988810356770884,
      "loss": 5.4388,
      "mean_token_accuracy": 0.17943356335163116,
      "num_tokens": 9790128.0,
      "step": 4275
    },
    {
      "entropy": 5.516012334823609,
      "epoch": 0.4111431316042267,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000498877618185287,
      "loss": 5.3604,
      "mean_token_accuracy": 0.19427798241376876,
      "num_tokens": 9802549.0,
      "step": 4280
    },
    {
      "entropy": 5.486606025695801,
      "epoch": 0.4116234390009606,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988741954957121,
      "loss": 5.3152,
      "mean_token_accuracy": 0.19245170205831527,
      "num_tokens": 9813736.0,
      "step": 4285
    },
    {
      "entropy": 5.4289391040802,
      "epoch": 0.41210374639769454,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988707676084432,
      "loss": 5.2994,
      "mean_token_accuracy": 0.20583047419786454,
      "num_tokens": 9823785.0,
      "step": 4290
    },
    {
      "entropy": 5.480359363555908,
      "epoch": 0.4125840537944284,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988673345235597,
      "loss": 5.2796,
      "mean_token_accuracy": 0.1970909520983696,
      "num_tokens": 9834910.0,
      "step": 4295
    },
    {
      "entropy": 5.525894498825073,
      "epoch": 0.41306436119116235,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004988638962411416,
      "loss": 5.3082,
      "mean_token_accuracy": 0.19386228919029236,
      "num_tokens": 9845593.0,
      "step": 4300
    },
    {
      "entropy": 5.436275625228882,
      "epoch": 0.41354466858789624,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004988604527612685,
      "loss": 5.1992,
      "mean_token_accuracy": 0.20476569086313248,
      "num_tokens": 9856763.0,
      "step": 4305
    },
    {
      "entropy": 5.573514127731324,
      "epoch": 0.4140249759846302,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004988570040840205,
      "loss": 5.4485,
      "mean_token_accuracy": 0.18456612378358841,
      "num_tokens": 9869528.0,
      "step": 4310
    },
    {
      "entropy": 5.480060815811157,
      "epoch": 0.41450528338136405,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988535502094774,
      "loss": 5.3357,
      "mean_token_accuracy": 0.19747838228940964,
      "num_tokens": 9881170.0,
      "step": 4315
    },
    {
      "entropy": 5.5313794136047365,
      "epoch": 0.414985590778098,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004988500911377198,
      "loss": 5.4322,
      "mean_token_accuracy": 0.1888141453266144,
      "num_tokens": 9893119.0,
      "step": 4320
    },
    {
      "entropy": 5.4053281307220455,
      "epoch": 0.41546589817483187,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004988466268688276,
      "loss": 5.2654,
      "mean_token_accuracy": 0.20344866067171097,
      "num_tokens": 9905339.0,
      "step": 4325
    },
    {
      "entropy": 5.571810483932495,
      "epoch": 0.4159462055715658,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004988431574028814,
      "loss": 5.3586,
      "mean_token_accuracy": 0.19604763239622117,
      "num_tokens": 9917500.0,
      "step": 4330
    },
    {
      "entropy": 5.524722290039063,
      "epoch": 0.4164265129682997,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988396827399618,
      "loss": 5.4356,
      "mean_token_accuracy": 0.18719332963228225,
      "num_tokens": 9929667.0,
      "step": 4335
    },
    {
      "entropy": 5.520600318908691,
      "epoch": 0.4169068203650336,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004988362028801495,
      "loss": 5.3503,
      "mean_token_accuracy": 0.19482496231794358,
      "num_tokens": 9941102.0,
      "step": 4340
    },
    {
      "entropy": 5.45862774848938,
      "epoch": 0.4173871277617675,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004988327178235253,
      "loss": 5.2373,
      "mean_token_accuracy": 0.2036656454205513,
      "num_tokens": 9951986.0,
      "step": 4345
    },
    {
      "entropy": 5.434820699691772,
      "epoch": 0.41786743515850144,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004988292275701699,
      "loss": 5.2608,
      "mean_token_accuracy": 0.19801096320152284,
      "num_tokens": 9964486.0,
      "step": 4350
    },
    {
      "entropy": 5.481801605224609,
      "epoch": 0.4183477425552353,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004988257321201646,
      "loss": 5.3643,
      "mean_token_accuracy": 0.19375134259462357,
      "num_tokens": 9975909.0,
      "step": 4355
    },
    {
      "entropy": 5.512979364395141,
      "epoch": 0.41882804995196926,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004988222314735902,
      "loss": 5.3665,
      "mean_token_accuracy": 0.19729440063238143,
      "num_tokens": 9986951.0,
      "step": 4360
    },
    {
      "entropy": 5.606730222702026,
      "epoch": 0.41930835734870314,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988187256305284,
      "loss": 5.4677,
      "mean_token_accuracy": 0.1853739634156227,
      "num_tokens": 9999234.0,
      "step": 4365
    },
    {
      "entropy": 5.498926258087158,
      "epoch": 0.4197886647454371,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004988152145910603,
      "loss": 5.3337,
      "mean_token_accuracy": 0.20228377282619475,
      "num_tokens": 10010178.0,
      "step": 4370
    },
    {
      "entropy": 5.419414520263672,
      "epoch": 0.420268972142171,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004988116983552675,
      "loss": 5.2865,
      "mean_token_accuracy": 0.194709937274456,
      "num_tokens": 10021183.0,
      "step": 4375
    },
    {
      "entropy": 5.632723474502564,
      "epoch": 0.4207492795389049,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004988081769232317,
      "loss": 5.5847,
      "mean_token_accuracy": 0.1804965242743492,
      "num_tokens": 10033686.0,
      "step": 4380
    },
    {
      "entropy": 5.490194416046142,
      "epoch": 0.42122958693563883,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004988046502950346,
      "loss": 5.2806,
      "mean_token_accuracy": 0.19382549822330475,
      "num_tokens": 10045923.0,
      "step": 4385
    },
    {
      "entropy": 5.338625478744507,
      "epoch": 0.4217098943323727,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000498801118470758,
      "loss": 5.1978,
      "mean_token_accuracy": 0.20119253993034364,
      "num_tokens": 10057196.0,
      "step": 4390
    },
    {
      "entropy": 5.4828390121459964,
      "epoch": 0.42219020172910665,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498797581450484,
      "loss": 5.3776,
      "mean_token_accuracy": 0.19160341322422028,
      "num_tokens": 10069655.0,
      "step": 4395
    },
    {
      "entropy": 5.533946561813354,
      "epoch": 0.42267050912584053,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004987940392342948,
      "loss": 5.2625,
      "mean_token_accuracy": 0.19662789553403853,
      "num_tokens": 10080876.0,
      "step": 4400
    },
    {
      "entropy": 5.507131290435791,
      "epoch": 0.42315081652257447,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004987904918222726,
      "loss": 5.3612,
      "mean_token_accuracy": 0.19040922522544862,
      "num_tokens": 10091986.0,
      "step": 4405
    },
    {
      "entropy": 5.542103242874146,
      "epoch": 0.42363112391930835,
      "grad_norm": 1.0,
      "learning_rate": 0.0004987869392144996,
      "loss": 5.4415,
      "mean_token_accuracy": 0.19244588911533356,
      "num_tokens": 10104027.0,
      "step": 4410
    },
    {
      "entropy": 5.522120523452759,
      "epoch": 0.4241114313160423,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004987833814110584,
      "loss": 5.3062,
      "mean_token_accuracy": 0.19760800004005433,
      "num_tokens": 10114665.0,
      "step": 4415
    },
    {
      "entropy": 5.473827123641968,
      "epoch": 0.42459173871277617,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004987798184120316,
      "loss": 5.3303,
      "mean_token_accuracy": 0.19874691665172578,
      "num_tokens": 10126032.0,
      "step": 4420
    },
    {
      "entropy": 5.56437931060791,
      "epoch": 0.4250720461095101,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004987762502175018,
      "loss": 5.3934,
      "mean_token_accuracy": 0.18887989372015,
      "num_tokens": 10137256.0,
      "step": 4425
    },
    {
      "entropy": 5.421947860717774,
      "epoch": 0.425552353506244,
      "grad_norm": 0.953125,
      "learning_rate": 0.000498772676827552,
      "loss": 5.2537,
      "mean_token_accuracy": 0.19916727691888808,
      "num_tokens": 10149445.0,
      "step": 4430
    },
    {
      "entropy": 5.51547532081604,
      "epoch": 0.4260326609029779,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004987690982422652,
      "loss": 5.3958,
      "mean_token_accuracy": 0.1895904004573822,
      "num_tokens": 10161607.0,
      "step": 4435
    },
    {
      "entropy": 5.549444389343262,
      "epoch": 0.4265129682997118,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004987655144617243,
      "loss": 5.4129,
      "mean_token_accuracy": 0.19335524141788482,
      "num_tokens": 10173184.0,
      "step": 4440
    },
    {
      "entropy": 5.520701122283936,
      "epoch": 0.42699327569644574,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004987619254860126,
      "loss": 5.2833,
      "mean_token_accuracy": 0.20056545436382295,
      "num_tokens": 10184617.0,
      "step": 4445
    },
    {
      "entropy": 5.515553379058838,
      "epoch": 0.4274735830931796,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004987583313152134,
      "loss": 5.3041,
      "mean_token_accuracy": 0.19518428593873977,
      "num_tokens": 10195608.0,
      "step": 4450
    },
    {
      "entropy": 5.488814115524292,
      "epoch": 0.42795389048991356,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004987547319494104,
      "loss": 5.3998,
      "mean_token_accuracy": 0.19026617109775543,
      "num_tokens": 10206763.0,
      "step": 4455
    },
    {
      "entropy": 5.541830825805664,
      "epoch": 0.42843419788664744,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004987511273886867,
      "loss": 5.3406,
      "mean_token_accuracy": 0.19727638810873033,
      "num_tokens": 10218714.0,
      "step": 4460
    },
    {
      "entropy": 5.502747392654419,
      "epoch": 0.4289145052833814,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004987475176331263,
      "loss": 5.3599,
      "mean_token_accuracy": 0.18761258721351623,
      "num_tokens": 10229902.0,
      "step": 4465
    },
    {
      "entropy": 5.464094400405884,
      "epoch": 0.42939481268011526,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004987439026828129,
      "loss": 5.2389,
      "mean_token_accuracy": 0.1978443130850792,
      "num_tokens": 10241578.0,
      "step": 4470
    },
    {
      "entropy": 5.39894585609436,
      "epoch": 0.4298751200768492,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004987402825378305,
      "loss": 5.206,
      "mean_token_accuracy": 0.2003094583749771,
      "num_tokens": 10252109.0,
      "step": 4475
    },
    {
      "entropy": 5.5236156463623045,
      "epoch": 0.4303554274735831,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004987366571982631,
      "loss": 5.3827,
      "mean_token_accuracy": 0.1925773411989212,
      "num_tokens": 10263357.0,
      "step": 4480
    },
    {
      "entropy": 5.555547142028809,
      "epoch": 0.430835734870317,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004987330266641948,
      "loss": 5.3822,
      "mean_token_accuracy": 0.18810010105371475,
      "num_tokens": 10275536.0,
      "step": 4485
    },
    {
      "entropy": 5.5469659805297855,
      "epoch": 0.4313160422670509,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004987293909357101,
      "loss": 5.3566,
      "mean_token_accuracy": 0.19789600372314453,
      "num_tokens": 10286901.0,
      "step": 4490
    },
    {
      "entropy": 5.410230827331543,
      "epoch": 0.43179634966378483,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004987257500128933,
      "loss": 5.2626,
      "mean_token_accuracy": 0.1930641770362854,
      "num_tokens": 10298961.0,
      "step": 4495
    },
    {
      "entropy": 5.526727294921875,
      "epoch": 0.4322766570605187,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004987221038958288,
      "loss": 5.3875,
      "mean_token_accuracy": 0.19502759724855423,
      "num_tokens": 10310911.0,
      "step": 4500
    },
    {
      "entropy": 5.566359043121338,
      "epoch": 0.43275696445725265,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004987184525846015,
      "loss": 5.3842,
      "mean_token_accuracy": 0.18737707734107972,
      "num_tokens": 10322267.0,
      "step": 4505
    },
    {
      "entropy": 5.477842903137207,
      "epoch": 0.4332372718539865,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987147960792958,
      "loss": 5.4036,
      "mean_token_accuracy": 0.19320885390043258,
      "num_tokens": 10335111.0,
      "step": 4510
    },
    {
      "entropy": 5.582201528549194,
      "epoch": 0.43371757925072046,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004987111343799971,
      "loss": 5.3399,
      "mean_token_accuracy": 0.19570576399564743,
      "num_tokens": 10345672.0,
      "step": 4515
    },
    {
      "entropy": 5.567843389511109,
      "epoch": 0.43419788664745435,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00049870746748679,
      "loss": 5.3295,
      "mean_token_accuracy": 0.19766930043697356,
      "num_tokens": 10357369.0,
      "step": 4520
    },
    {
      "entropy": 5.494852113723755,
      "epoch": 0.4346781940441883,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004987037953997598,
      "loss": 5.3546,
      "mean_token_accuracy": 0.19237277954816817,
      "num_tokens": 10368842.0,
      "step": 4525
    },
    {
      "entropy": 5.473773574829101,
      "epoch": 0.43515850144092216,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004987001181189918,
      "loss": 5.3001,
      "mean_token_accuracy": 0.19634409248828888,
      "num_tokens": 10380096.0,
      "step": 4530
    },
    {
      "entropy": 5.398287105560303,
      "epoch": 0.4356388088376561,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004986964356445713,
      "loss": 5.3269,
      "mean_token_accuracy": 0.1937148690223694,
      "num_tokens": 10391996.0,
      "step": 4535
    },
    {
      "entropy": 5.496431159973144,
      "epoch": 0.43611911623439004,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986927479765837,
      "loss": 5.2735,
      "mean_token_accuracy": 0.1934656947851181,
      "num_tokens": 10403607.0,
      "step": 4540
    },
    {
      "entropy": 5.511260318756103,
      "epoch": 0.4365994236311239,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004986890551151148,
      "loss": 5.2977,
      "mean_token_accuracy": 0.1925230547785759,
      "num_tokens": 10413580.0,
      "step": 4545
    },
    {
      "entropy": 5.47296986579895,
      "epoch": 0.43707973102785785,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004986853570602503,
      "loss": 5.3258,
      "mean_token_accuracy": 0.19467985332012178,
      "num_tokens": 10426456.0,
      "step": 4550
    },
    {
      "entropy": 5.578516864776612,
      "epoch": 0.43756003842459174,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004986816538120758,
      "loss": 5.3594,
      "mean_token_accuracy": 0.18571384996175766,
      "num_tokens": 10438869.0,
      "step": 4555
    },
    {
      "entropy": 5.476364612579346,
      "epoch": 0.43804034582132567,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986779453706778,
      "loss": 5.3735,
      "mean_token_accuracy": 0.18391507416963576,
      "num_tokens": 10450672.0,
      "step": 4560
    },
    {
      "entropy": 5.489961862564087,
      "epoch": 0.43852065321805955,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986742317361419,
      "loss": 5.2794,
      "mean_token_accuracy": 0.20194608271121978,
      "num_tokens": 10461890.0,
      "step": 4565
    },
    {
      "entropy": 5.555417490005493,
      "epoch": 0.4390009606147935,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986705129085546,
      "loss": 5.4116,
      "mean_token_accuracy": 0.18422490060329438,
      "num_tokens": 10473866.0,
      "step": 4570
    },
    {
      "entropy": 5.551934766769409,
      "epoch": 0.43948126801152737,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004986667888880021,
      "loss": 5.3176,
      "mean_token_accuracy": 0.19357246905565262,
      "num_tokens": 10484889.0,
      "step": 4575
    },
    {
      "entropy": 5.465532112121582,
      "epoch": 0.4399615754082613,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004986630596745709,
      "loss": 5.355,
      "mean_token_accuracy": 0.19461666494607927,
      "num_tokens": 10496108.0,
      "step": 4580
    },
    {
      "entropy": 5.503142023086548,
      "epoch": 0.4404418828049952,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004986593252683477,
      "loss": 5.3168,
      "mean_token_accuracy": 0.19193545132875442,
      "num_tokens": 10505472.0,
      "step": 4585
    },
    {
      "entropy": 5.371403980255127,
      "epoch": 0.4409221902017291,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004986555856694191,
      "loss": 5.2186,
      "mean_token_accuracy": 0.2004853069782257,
      "num_tokens": 10516954.0,
      "step": 4590
    },
    {
      "entropy": 5.551608276367188,
      "epoch": 0.441402497598463,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004986518408778718,
      "loss": 5.3399,
      "mean_token_accuracy": 0.19154966324567796,
      "num_tokens": 10528166.0,
      "step": 4595
    },
    {
      "entropy": 5.461075639724731,
      "epoch": 0.44188280499519694,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986480908937929,
      "loss": 5.2563,
      "mean_token_accuracy": 0.19935419112443925,
      "num_tokens": 10538112.0,
      "step": 4600
    },
    {
      "entropy": 5.530212450027466,
      "epoch": 0.4423631123919308,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986443357172695,
      "loss": 5.3944,
      "mean_token_accuracy": 0.19479911625385285,
      "num_tokens": 10549888.0,
      "step": 4605
    },
    {
      "entropy": 5.653611516952514,
      "epoch": 0.44284341978866476,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004986405753483887,
      "loss": 5.4656,
      "mean_token_accuracy": 0.1801862746477127,
      "num_tokens": 10561710.0,
      "step": 4610
    },
    {
      "entropy": 5.439924049377441,
      "epoch": 0.44332372718539864,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004986368097872377,
      "loss": 5.3356,
      "mean_token_accuracy": 0.1889050930738449,
      "num_tokens": 10574564.0,
      "step": 4615
    },
    {
      "entropy": 5.5105362892150875,
      "epoch": 0.4438040345821326,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004986330390339042,
      "loss": 5.3037,
      "mean_token_accuracy": 0.19357553124427795,
      "num_tokens": 10586639.0,
      "step": 4620
    },
    {
      "entropy": 5.502204561233521,
      "epoch": 0.44428434197886646,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004986292630884755,
      "loss": 5.3128,
      "mean_token_accuracy": 0.1955375999212265,
      "num_tokens": 10598730.0,
      "step": 4625
    },
    {
      "entropy": 5.417471694946289,
      "epoch": 0.4447646493756004,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004986254819510393,
      "loss": 5.2261,
      "mean_token_accuracy": 0.20689018219709396,
      "num_tokens": 10610352.0,
      "step": 4630
    },
    {
      "entropy": 5.460666942596435,
      "epoch": 0.4452449567723343,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004986216956216835,
      "loss": 5.3129,
      "mean_token_accuracy": 0.1962559252977371,
      "num_tokens": 10621951.0,
      "step": 4635
    },
    {
      "entropy": 5.489536237716675,
      "epoch": 0.4457252641690682,
      "grad_norm": 1.046875,
      "learning_rate": 0.000498617904100496,
      "loss": 5.2536,
      "mean_token_accuracy": 0.1979408711194992,
      "num_tokens": 10633207.0,
      "step": 4640
    },
    {
      "entropy": 5.551165294647217,
      "epoch": 0.4462055715658021,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004986141073875646,
      "loss": 5.366,
      "mean_token_accuracy": 0.186821448802948,
      "num_tokens": 10645853.0,
      "step": 4645
    },
    {
      "entropy": 5.432143926620483,
      "epoch": 0.44668587896253603,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004986103054829779,
      "loss": 5.27,
      "mean_token_accuracy": 0.19958215057849885,
      "num_tokens": 10656892.0,
      "step": 4650
    },
    {
      "entropy": 5.428584432601928,
      "epoch": 0.4471661863592699,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004986064983868237,
      "loss": 5.2548,
      "mean_token_accuracy": 0.19150467962026596,
      "num_tokens": 10670110.0,
      "step": 4655
    },
    {
      "entropy": 5.511739206314087,
      "epoch": 0.44764649375600385,
      "grad_norm": 1.0,
      "learning_rate": 0.0004986026860991906,
      "loss": 5.3878,
      "mean_token_accuracy": 0.19259512722492217,
      "num_tokens": 10681255.0,
      "step": 4660
    },
    {
      "entropy": 5.514628124237061,
      "epoch": 0.44812680115273773,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004985988686201672,
      "loss": 5.4364,
      "mean_token_accuracy": 0.19111472219228745,
      "num_tokens": 10692631.0,
      "step": 4665
    },
    {
      "entropy": 5.506016063690185,
      "epoch": 0.44860710854947167,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004985950459498419,
      "loss": 5.2841,
      "mean_token_accuracy": 0.19589586406946183,
      "num_tokens": 10704880.0,
      "step": 4670
    },
    {
      "entropy": 5.485089635848999,
      "epoch": 0.44908741594620555,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004985912180883037,
      "loss": 5.2646,
      "mean_token_accuracy": 0.20430310666561127,
      "num_tokens": 10715561.0,
      "step": 4675
    },
    {
      "entropy": 5.413974857330322,
      "epoch": 0.4495677233429395,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004985873850356411,
      "loss": 5.2773,
      "mean_token_accuracy": 0.19555141776800156,
      "num_tokens": 10727232.0,
      "step": 4680
    },
    {
      "entropy": 5.413888311386108,
      "epoch": 0.45004803073967337,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004985835467919436,
      "loss": 5.2991,
      "mean_token_accuracy": 0.20086562782526016,
      "num_tokens": 10739404.0,
      "step": 4685
    },
    {
      "entropy": 5.5062575340271,
      "epoch": 0.4505283381364073,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004985797033572999,
      "loss": 5.322,
      "mean_token_accuracy": 0.19117717891931535,
      "num_tokens": 10751948.0,
      "step": 4690
    },
    {
      "entropy": 5.458573246002198,
      "epoch": 0.4510086455331412,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004985758547317994,
      "loss": 5.296,
      "mean_token_accuracy": 0.19499971568584443,
      "num_tokens": 10764611.0,
      "step": 4695
    },
    {
      "entropy": 5.518351459503174,
      "epoch": 0.4514889529298751,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004985720009155315,
      "loss": 5.3248,
      "mean_token_accuracy": 0.19280190467834474,
      "num_tokens": 10775954.0,
      "step": 4700
    },
    {
      "entropy": 5.475862789154053,
      "epoch": 0.45196926032660906,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004985681419085856,
      "loss": 5.3332,
      "mean_token_accuracy": 0.18717591017484664,
      "num_tokens": 10788723.0,
      "step": 4705
    },
    {
      "entropy": 5.48907470703125,
      "epoch": 0.45244956772334294,
      "grad_norm": 1.0,
      "learning_rate": 0.0004985642777110513,
      "loss": 5.3185,
      "mean_token_accuracy": 0.19800368696451187,
      "num_tokens": 10799879.0,
      "step": 4710
    },
    {
      "entropy": 5.421358299255371,
      "epoch": 0.4529298751200769,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004985604083230183,
      "loss": 5.2791,
      "mean_token_accuracy": 0.19696285873651503,
      "num_tokens": 10811838.0,
      "step": 4715
    },
    {
      "entropy": 5.494543981552124,
      "epoch": 0.45341018251681076,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985565337445765,
      "loss": 5.2937,
      "mean_token_accuracy": 0.19829083234071732,
      "num_tokens": 10822910.0,
      "step": 4720
    },
    {
      "entropy": 5.533810424804687,
      "epoch": 0.4538904899135447,
      "grad_norm": 0.875,
      "learning_rate": 0.0004985526539758158,
      "loss": 5.3543,
      "mean_token_accuracy": 0.19275055080652237,
      "num_tokens": 10835344.0,
      "step": 4725
    },
    {
      "entropy": 5.468215227127075,
      "epoch": 0.4543707973102786,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004985487690168263,
      "loss": 5.3316,
      "mean_token_accuracy": 0.19680845141410827,
      "num_tokens": 10846043.0,
      "step": 4730
    },
    {
      "entropy": 5.478121185302735,
      "epoch": 0.4548511047070125,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000498544878867698,
      "loss": 5.2595,
      "mean_token_accuracy": 0.20072862654924392,
      "num_tokens": 10857783.0,
      "step": 4735
    },
    {
      "entropy": 5.474717378616333,
      "epoch": 0.4553314121037464,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004985409835285215,
      "loss": 5.3229,
      "mean_token_accuracy": 0.19704768657684327,
      "num_tokens": 10870527.0,
      "step": 4740
    },
    {
      "entropy": 5.464604425430298,
      "epoch": 0.45581171950048033,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004985370829993873,
      "loss": 5.304,
      "mean_token_accuracy": 0.20295322835445403,
      "num_tokens": 10882285.0,
      "step": 4745
    },
    {
      "entropy": 5.4915252208709715,
      "epoch": 0.4562920268972142,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004985331772803857,
      "loss": 5.3263,
      "mean_token_accuracy": 0.1975204363465309,
      "num_tokens": 10895319.0,
      "step": 4750
    },
    {
      "entropy": 5.524920082092285,
      "epoch": 0.45677233429394815,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004985292663716074,
      "loss": 5.3107,
      "mean_token_accuracy": 0.1974620059132576,
      "num_tokens": 10906253.0,
      "step": 4755
    },
    {
      "entropy": 5.351269483566284,
      "epoch": 0.457252641690682,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004985253502731435,
      "loss": 5.2019,
      "mean_token_accuracy": 0.2055060312151909,
      "num_tokens": 10918197.0,
      "step": 4760
    },
    {
      "entropy": 5.476984643936158,
      "epoch": 0.45773294908741596,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004985214289850845,
      "loss": 5.4009,
      "mean_token_accuracy": 0.19004748165607452,
      "num_tokens": 10930771.0,
      "step": 4765
    },
    {
      "entropy": 5.489283180236816,
      "epoch": 0.45821325648414984,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004985175025075217,
      "loss": 5.2985,
      "mean_token_accuracy": 0.19462554901838303,
      "num_tokens": 10942759.0,
      "step": 4770
    },
    {
      "entropy": 5.6803840637207035,
      "epoch": 0.4586935638808838,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004985135708405462,
      "loss": 5.4936,
      "mean_token_accuracy": 0.18637332171201706,
      "num_tokens": 10953557.0,
      "step": 4775
    },
    {
      "entropy": 5.436570262908935,
      "epoch": 0.45917387127761766,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004985096339842493,
      "loss": 5.2622,
      "mean_token_accuracy": 0.20460412800312042,
      "num_tokens": 10963142.0,
      "step": 4780
    },
    {
      "entropy": 5.396737003326416,
      "epoch": 0.4596541786743516,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985056919387224,
      "loss": 5.2322,
      "mean_token_accuracy": 0.202386274933815,
      "num_tokens": 10974321.0,
      "step": 4785
    },
    {
      "entropy": 5.550132369995117,
      "epoch": 0.4601344860710855,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004985017447040569,
      "loss": 5.43,
      "mean_token_accuracy": 0.19050837010145188,
      "num_tokens": 10985524.0,
      "step": 4790
    },
    {
      "entropy": 5.525373315811157,
      "epoch": 0.4606147934678194,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004984977922803447,
      "loss": 5.3236,
      "mean_token_accuracy": 0.19777408093214036,
      "num_tokens": 10997606.0,
      "step": 4795
    },
    {
      "entropy": 5.440893602371216,
      "epoch": 0.4610951008645533,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004984938346676772,
      "loss": 5.3185,
      "mean_token_accuracy": 0.1897636130452156,
      "num_tokens": 11010692.0,
      "step": 4800
    },
    {
      "entropy": 5.451299047470092,
      "epoch": 0.46157540826128723,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004984898718661468,
      "loss": 5.2502,
      "mean_token_accuracy": 0.20564369410276412,
      "num_tokens": 11022517.0,
      "step": 4805
    },
    {
      "entropy": 5.449737644195556,
      "epoch": 0.4620557156580211,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004984859038758451,
      "loss": 5.252,
      "mean_token_accuracy": 0.20158778578042985,
      "num_tokens": 11033141.0,
      "step": 4810
    },
    {
      "entropy": 5.4242347240447994,
      "epoch": 0.46253602305475505,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004984819306968642,
      "loss": 5.2538,
      "mean_token_accuracy": 0.19843853563070296,
      "num_tokens": 11044619.0,
      "step": 4815
    },
    {
      "entropy": 5.555520963668823,
      "epoch": 0.46301633045148893,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004984779523292966,
      "loss": 5.2999,
      "mean_token_accuracy": 0.19956816583871842,
      "num_tokens": 11055934.0,
      "step": 4820
    },
    {
      "entropy": 5.463156366348267,
      "epoch": 0.46349663784822287,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004984739687732345,
      "loss": 5.1856,
      "mean_token_accuracy": 0.2006239727139473,
      "num_tokens": 11066203.0,
      "step": 4825
    },
    {
      "entropy": 5.269961309432984,
      "epoch": 0.46397694524495675,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004984699800287705,
      "loss": 5.1521,
      "mean_token_accuracy": 0.2056365504860878,
      "num_tokens": 11079664.0,
      "step": 4830
    },
    {
      "entropy": 5.376195335388184,
      "epoch": 0.4644572526416907,
      "grad_norm": 0.921875,
      "learning_rate": 0.000498465986095997,
      "loss": 5.1928,
      "mean_token_accuracy": 0.2033955082297325,
      "num_tokens": 11091186.0,
      "step": 4835
    },
    {
      "entropy": 5.497678804397583,
      "epoch": 0.46493756003842457,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004984619869750069,
      "loss": 5.3304,
      "mean_token_accuracy": 0.18890146911144257,
      "num_tokens": 11102710.0,
      "step": 4840
    },
    {
      "entropy": 5.408234453201294,
      "epoch": 0.4654178674351585,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000498457982665893,
      "loss": 5.2202,
      "mean_token_accuracy": 0.2038560301065445,
      "num_tokens": 11114746.0,
      "step": 4845
    },
    {
      "entropy": 5.460979413986206,
      "epoch": 0.4658981748318924,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004984539731687483,
      "loss": 5.2911,
      "mean_token_accuracy": 0.195130418241024,
      "num_tokens": 11126572.0,
      "step": 4850
    },
    {
      "entropy": 5.4305055141448975,
      "epoch": 0.4663784822286263,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004984499584836659,
      "loss": 5.1793,
      "mean_token_accuracy": 0.20321067571640014,
      "num_tokens": 11137830.0,
      "step": 4855
    },
    {
      "entropy": 5.377715015411377,
      "epoch": 0.4668587896253602,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000498445938610739,
      "loss": 5.2293,
      "mean_token_accuracy": 0.2016708493232727,
      "num_tokens": 11148860.0,
      "step": 4860
    },
    {
      "entropy": 5.479203367233277,
      "epoch": 0.46733909702209414,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004984419135500608,
      "loss": 5.3466,
      "mean_token_accuracy": 0.18884660452604293,
      "num_tokens": 11161311.0,
      "step": 4865
    },
    {
      "entropy": 5.443951034545899,
      "epoch": 0.4678194044188281,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004984378833017249,
      "loss": 5.2379,
      "mean_token_accuracy": 0.19662537276744843,
      "num_tokens": 11173124.0,
      "step": 4870
    },
    {
      "entropy": 5.412505626678467,
      "epoch": 0.46829971181556196,
      "grad_norm": 1.0,
      "learning_rate": 0.0004984338478658248,
      "loss": 5.3144,
      "mean_token_accuracy": 0.19596934467554092,
      "num_tokens": 11184879.0,
      "step": 4875
    },
    {
      "entropy": 5.530852127075195,
      "epoch": 0.4687800192122959,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004984298072424542,
      "loss": 5.3221,
      "mean_token_accuracy": 0.19306136071681976,
      "num_tokens": 11196243.0,
      "step": 4880
    },
    {
      "entropy": 5.437544488906861,
      "epoch": 0.4692603266090298,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498425761431707,
      "loss": 5.1889,
      "mean_token_accuracy": 0.20993078351020814,
      "num_tokens": 11207485.0,
      "step": 4885
    },
    {
      "entropy": 5.378513956069947,
      "epoch": 0.4697406340057637,
      "grad_norm": 0.9375,
      "learning_rate": 0.000498421710433677,
      "loss": 5.2263,
      "mean_token_accuracy": 0.19690736532211303,
      "num_tokens": 11219891.0,
      "step": 4890
    },
    {
      "entropy": 5.515623712539673,
      "epoch": 0.4702209414024976,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004984176542484584,
      "loss": 5.321,
      "mean_token_accuracy": 0.19684761762619019,
      "num_tokens": 11231329.0,
      "step": 4895
    },
    {
      "entropy": 5.420218229293823,
      "epoch": 0.47070124879923153,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004984135928761452,
      "loss": 5.2182,
      "mean_token_accuracy": 0.20384916067123413,
      "num_tokens": 11241367.0,
      "step": 4900
    },
    {
      "entropy": 5.4436530590057375,
      "epoch": 0.4711815561959654,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004984095263168317,
      "loss": 5.3041,
      "mean_token_accuracy": 0.19205039143562316,
      "num_tokens": 11254532.0,
      "step": 4905
    },
    {
      "entropy": 5.55198187828064,
      "epoch": 0.47166186359269935,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004984054545706124,
      "loss": 5.3604,
      "mean_token_accuracy": 0.18903273940086365,
      "num_tokens": 11265223.0,
      "step": 4910
    },
    {
      "entropy": 5.435238647460937,
      "epoch": 0.47214217098943323,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000498401377637582,
      "loss": 5.3017,
      "mean_token_accuracy": 0.19942667335271835,
      "num_tokens": 11278228.0,
      "step": 4915
    },
    {
      "entropy": 5.531127023696899,
      "epoch": 0.47262247838616717,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000498397295517835,
      "loss": 5.4152,
      "mean_token_accuracy": 0.1892762526869774,
      "num_tokens": 11289654.0,
      "step": 4920
    },
    {
      "entropy": 5.441771411895752,
      "epoch": 0.47310278578290105,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004983932082114659,
      "loss": 5.194,
      "mean_token_accuracy": 0.20304778963327408,
      "num_tokens": 11301911.0,
      "step": 4925
    },
    {
      "entropy": 5.547307014465332,
      "epoch": 0.473583093179635,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983891157185699,
      "loss": 5.3721,
      "mean_token_accuracy": 0.19476744085550307,
      "num_tokens": 11312945.0,
      "step": 4930
    },
    {
      "entropy": 5.594564008712768,
      "epoch": 0.47406340057636887,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004983850180392421,
      "loss": 5.422,
      "mean_token_accuracy": 0.186136656999588,
      "num_tokens": 11324126.0,
      "step": 4935
    },
    {
      "entropy": 5.489442443847656,
      "epoch": 0.4745437079731028,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004983809151735775,
      "loss": 5.345,
      "mean_token_accuracy": 0.19110574275255204,
      "num_tokens": 11336395.0,
      "step": 4940
    },
    {
      "entropy": 5.506710433959961,
      "epoch": 0.4750240153698367,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004983768071216713,
      "loss": 5.2532,
      "mean_token_accuracy": 0.1983468234539032,
      "num_tokens": 11347387.0,
      "step": 4945
    },
    {
      "entropy": 5.399397993087769,
      "epoch": 0.4755043227665706,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983726938836189,
      "loss": 5.2472,
      "mean_token_accuracy": 0.20188201814889908,
      "num_tokens": 11358467.0,
      "step": 4950
    },
    {
      "entropy": 5.4775426387786865,
      "epoch": 0.4759846301633045,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004983685754595159,
      "loss": 5.4131,
      "mean_token_accuracy": 0.18820505440235138,
      "num_tokens": 11370322.0,
      "step": 4955
    },
    {
      "entropy": 5.378913688659668,
      "epoch": 0.47646493756003844,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004983644518494578,
      "loss": 5.2103,
      "mean_token_accuracy": 0.20417440831661224,
      "num_tokens": 11381719.0,
      "step": 4960
    },
    {
      "entropy": 5.415813398361206,
      "epoch": 0.4769452449567723,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983603230535403,
      "loss": 5.2357,
      "mean_token_accuracy": 0.19860472083091735,
      "num_tokens": 11393561.0,
      "step": 4965
    },
    {
      "entropy": 5.506166219711304,
      "epoch": 0.47742555235350626,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004983561890718594,
      "loss": 5.3325,
      "mean_token_accuracy": 0.19491951316595077,
      "num_tokens": 11405411.0,
      "step": 4970
    },
    {
      "entropy": 5.578386402130127,
      "epoch": 0.47790585975024014,
      "grad_norm": 1.09375,
      "learning_rate": 0.000498352049904511,
      "loss": 5.4209,
      "mean_token_accuracy": 0.1887726828455925,
      "num_tokens": 11417419.0,
      "step": 4975
    },
    {
      "entropy": 5.438750505447388,
      "epoch": 0.4783861671469741,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983479055515914,
      "loss": 5.2286,
      "mean_token_accuracy": 0.19747813940048217,
      "num_tokens": 11428145.0,
      "step": 4980
    },
    {
      "entropy": 5.37105541229248,
      "epoch": 0.47886647454370795,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004983437560131964,
      "loss": 5.1843,
      "mean_token_accuracy": 0.20008998066186906,
      "num_tokens": 11439224.0,
      "step": 4985
    },
    {
      "entropy": 5.454404306411743,
      "epoch": 0.4793467819404419,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004983396012894228,
      "loss": 5.292,
      "mean_token_accuracy": 0.1926464319229126,
      "num_tokens": 11451731.0,
      "step": 4990
    },
    {
      "entropy": 5.503137922286987,
      "epoch": 0.47982708933717577,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004983354413803666,
      "loss": 5.2531,
      "mean_token_accuracy": 0.20034031867980956,
      "num_tokens": 11463058.0,
      "step": 4995
    },
    {
      "entropy": 5.568387126922607,
      "epoch": 0.4803073967339097,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004983312762861248,
      "loss": 5.3678,
      "mean_token_accuracy": 0.19228482842445374,
      "num_tokens": 11472618.0,
      "step": 5000
    },
    {
      "entropy": 5.427599239349365,
      "epoch": 0.4807877041306436,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004983271060067939,
      "loss": 5.2494,
      "mean_token_accuracy": 0.19887056648731233,
      "num_tokens": 11483114.0,
      "step": 5005
    },
    {
      "entropy": 5.37850604057312,
      "epoch": 0.4812680115273775,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004983229305424707,
      "loss": 5.2093,
      "mean_token_accuracy": 0.20153291821479796,
      "num_tokens": 11494281.0,
      "step": 5010
    },
    {
      "entropy": 5.467651128768921,
      "epoch": 0.4817483189241114,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004983187498932522,
      "loss": 5.3153,
      "mean_token_accuracy": 0.19636951982975007,
      "num_tokens": 11505962.0,
      "step": 5015
    },
    {
      "entropy": 5.497737312316895,
      "epoch": 0.48222862632084534,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004983145640592354,
      "loss": 5.3634,
      "mean_token_accuracy": 0.19193184673786162,
      "num_tokens": 11517558.0,
      "step": 5020
    },
    {
      "entropy": 5.4847935199737545,
      "epoch": 0.4827089337175792,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004983103730405176,
      "loss": 5.3696,
      "mean_token_accuracy": 0.19255296885967255,
      "num_tokens": 11529184.0,
      "step": 5025
    },
    {
      "entropy": 5.436256265640258,
      "epoch": 0.48318924111431316,
      "grad_norm": 0.953125,
      "learning_rate": 0.000498306176837196,
      "loss": 5.2808,
      "mean_token_accuracy": 0.19038698375225066,
      "num_tokens": 11540727.0,
      "step": 5030
    },
    {
      "entropy": 5.436978578567505,
      "epoch": 0.48366954851104704,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004983019754493681,
      "loss": 5.2094,
      "mean_token_accuracy": 0.20015370547771455,
      "num_tokens": 11551510.0,
      "step": 5035
    },
    {
      "entropy": 5.481416273117065,
      "epoch": 0.484149855907781,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004982977688771314,
      "loss": 5.3544,
      "mean_token_accuracy": 0.19754620641469955,
      "num_tokens": 11563203.0,
      "step": 5040
    },
    {
      "entropy": 5.4130464553833,
      "epoch": 0.4846301633045149,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004982935571205835,
      "loss": 5.2229,
      "mean_token_accuracy": 0.20334694683551788,
      "num_tokens": 11576013.0,
      "step": 5045
    },
    {
      "entropy": 5.343645143508911,
      "epoch": 0.4851104707012488,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004982893401798223,
      "loss": 5.1661,
      "mean_token_accuracy": 0.21171488612890244,
      "num_tokens": 11587535.0,
      "step": 5050
    },
    {
      "entropy": 5.405810928344726,
      "epoch": 0.48559077809798273,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004982851180549456,
      "loss": 5.2146,
      "mean_token_accuracy": 0.20077610909938812,
      "num_tokens": 11598487.0,
      "step": 5055
    },
    {
      "entropy": 5.349521541595459,
      "epoch": 0.4860710854947166,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004982808907460515,
      "loss": 5.1012,
      "mean_token_accuracy": 0.22083092629909515,
      "num_tokens": 11609457.0,
      "step": 5060
    },
    {
      "entropy": 5.382025146484375,
      "epoch": 0.48655139289145055,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004982766582532382,
      "loss": 5.1731,
      "mean_token_accuracy": 0.20699179470539092,
      "num_tokens": 11620251.0,
      "step": 5065
    },
    {
      "entropy": 5.375451374053955,
      "epoch": 0.48703170028818443,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004982724205766038,
      "loss": 5.1636,
      "mean_token_accuracy": 0.20996559262275696,
      "num_tokens": 11630956.0,
      "step": 5070
    },
    {
      "entropy": 5.420528221130371,
      "epoch": 0.48751200768491837,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004982681777162468,
      "loss": 5.2211,
      "mean_token_accuracy": 0.20217099487781526,
      "num_tokens": 11642560.0,
      "step": 5075
    },
    {
      "entropy": 5.33175368309021,
      "epoch": 0.48799231508165225,
      "grad_norm": 1.0,
      "learning_rate": 0.0004982639296722657,
      "loss": 5.1842,
      "mean_token_accuracy": 0.20277092307806016,
      "num_tokens": 11654050.0,
      "step": 5080
    },
    {
      "entropy": 5.378251647949218,
      "epoch": 0.4884726224783862,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004982596764447591,
      "loss": 5.3346,
      "mean_token_accuracy": 0.20321715325117112,
      "num_tokens": 11664947.0,
      "step": 5085
    },
    {
      "entropy": 5.449995803833008,
      "epoch": 0.48895292987512007,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004982554180338258,
      "loss": 5.2456,
      "mean_token_accuracy": 0.19882605373859405,
      "num_tokens": 11676927.0,
      "step": 5090
    },
    {
      "entropy": 5.577300882339477,
      "epoch": 0.489433237271854,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004982511544395646,
      "loss": 5.3554,
      "mean_token_accuracy": 0.18948118835687638,
      "num_tokens": 11688573.0,
      "step": 5095
    },
    {
      "entropy": 5.363681936264038,
      "epoch": 0.4899135446685879,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004982468856620745,
      "loss": 5.2582,
      "mean_token_accuracy": 0.19672578871250151,
      "num_tokens": 11698704.0,
      "step": 5100
    },
    {
      "entropy": 5.34480619430542,
      "epoch": 0.4903938520653218,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004982426117014545,
      "loss": 5.1847,
      "mean_token_accuracy": 0.20228682458400726,
      "num_tokens": 11709466.0,
      "step": 5105
    },
    {
      "entropy": 5.421897935867309,
      "epoch": 0.4908741594620557,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004982383325578041,
      "loss": 5.2921,
      "mean_token_accuracy": 0.19451844096183776,
      "num_tokens": 11721120.0,
      "step": 5110
    },
    {
      "entropy": 5.504036855697632,
      "epoch": 0.49135446685878964,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004982340482312226,
      "loss": 5.2679,
      "mean_token_accuracy": 0.1939774975180626,
      "num_tokens": 11732120.0,
      "step": 5115
    },
    {
      "entropy": 5.4026013851165775,
      "epoch": 0.4918347742555235,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004982297587218092,
      "loss": 5.1478,
      "mean_token_accuracy": 0.20662265568971633,
      "num_tokens": 11743501.0,
      "step": 5120
    },
    {
      "entropy": 5.384142446517944,
      "epoch": 0.49231508165225746,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004982254640296637,
      "loss": 5.2361,
      "mean_token_accuracy": 0.20638473331928253,
      "num_tokens": 11755051.0,
      "step": 5125
    },
    {
      "entropy": 5.443932723999024,
      "epoch": 0.49279538904899134,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004982211641548857,
      "loss": 5.3835,
      "mean_token_accuracy": 0.1939889207482338,
      "num_tokens": 11767663.0,
      "step": 5130
    },
    {
      "entropy": 5.499396657943725,
      "epoch": 0.4932756964457253,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004982168590975752,
      "loss": 5.2535,
      "mean_token_accuracy": 0.20122796595096587,
      "num_tokens": 11778828.0,
      "step": 5135
    },
    {
      "entropy": 5.513611745834351,
      "epoch": 0.49375600384245916,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004982125488578321,
      "loss": 5.4264,
      "mean_token_accuracy": 0.19013044238090515,
      "num_tokens": 11790654.0,
      "step": 5140
    },
    {
      "entropy": 5.448870611190796,
      "epoch": 0.4942363112391931,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982082334357563,
      "loss": 5.2255,
      "mean_token_accuracy": 0.1969570130109787,
      "num_tokens": 11801489.0,
      "step": 5145
    },
    {
      "entropy": 5.391196393966675,
      "epoch": 0.494716618635927,
      "grad_norm": 1.0,
      "learning_rate": 0.0004982039128314481,
      "loss": 5.2229,
      "mean_token_accuracy": 0.19950801879167557,
      "num_tokens": 11813818.0,
      "step": 5150
    },
    {
      "entropy": 5.391459226608276,
      "epoch": 0.4951969260326609,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004981995870450079,
      "loss": 5.2281,
      "mean_token_accuracy": 0.1980131134390831,
      "num_tokens": 11824814.0,
      "step": 5155
    },
    {
      "entropy": 5.479894018173217,
      "epoch": 0.4956772334293948,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004981952560765361,
      "loss": 5.2744,
      "mean_token_accuracy": 0.19537818133831025,
      "num_tokens": 11836252.0,
      "step": 5160
    },
    {
      "entropy": 5.3897405624389645,
      "epoch": 0.49615754082612873,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004981909199261331,
      "loss": 5.1831,
      "mean_token_accuracy": 0.1983597010374069,
      "num_tokens": 11847715.0,
      "step": 5165
    },
    {
      "entropy": 5.345670604705811,
      "epoch": 0.4966378482228626,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004981865785938998,
      "loss": 5.1961,
      "mean_token_accuracy": 0.2015163406729698,
      "num_tokens": 11860309.0,
      "step": 5170
    },
    {
      "entropy": 5.3873132228851315,
      "epoch": 0.49711815561959655,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004981822320799367,
      "loss": 5.2035,
      "mean_token_accuracy": 0.19946749955415727,
      "num_tokens": 11872569.0,
      "step": 5175
    },
    {
      "entropy": 5.385909605026245,
      "epoch": 0.49759846301633043,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004981778803843449,
      "loss": 5.1969,
      "mean_token_accuracy": 0.2004099726676941,
      "num_tokens": 11884778.0,
      "step": 5180
    },
    {
      "entropy": 5.452798223495483,
      "epoch": 0.49807877041306436,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004981735235072256,
      "loss": 5.277,
      "mean_token_accuracy": 0.2006050869822502,
      "num_tokens": 11897324.0,
      "step": 5185
    },
    {
      "entropy": 5.475766944885254,
      "epoch": 0.49855907780979825,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004981691614486796,
      "loss": 5.3007,
      "mean_token_accuracy": 0.19719384908676146,
      "num_tokens": 11909145.0,
      "step": 5190
    },
    {
      "entropy": 5.360778284072876,
      "epoch": 0.4990393852065322,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004981647942088084,
      "loss": 5.1975,
      "mean_token_accuracy": 0.2016976147890091,
      "num_tokens": 11921021.0,
      "step": 5195
    },
    {
      "entropy": 5.543325519561767,
      "epoch": 0.49951969260326606,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004981604217877135,
      "loss": 5.3512,
      "mean_token_accuracy": 0.19207874685525894,
      "num_tokens": 11932565.0,
      "step": 5200
    },
    {
      "entropy": 5.409695863723755,
      "epoch": 0.5,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000498156044185496,
      "loss": 5.2685,
      "mean_token_accuracy": 0.19969164580106735,
      "num_tokens": 11943225.0,
      "step": 5205
    },
    {
      "entropy": 5.400139284133911,
      "epoch": 0.5004803073967339,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004981516614022579,
      "loss": 5.2701,
      "mean_token_accuracy": 0.20002519637346267,
      "num_tokens": 11954821.0,
      "step": 5210
    },
    {
      "entropy": 5.445433759689331,
      "epoch": 0.5009606147934679,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004981472734381008,
      "loss": 5.2092,
      "mean_token_accuracy": 0.20269747227430343,
      "num_tokens": 11966090.0,
      "step": 5215
    },
    {
      "entropy": 5.445778322219849,
      "epoch": 0.5014409221902018,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004981428802931267,
      "loss": 5.2383,
      "mean_token_accuracy": 0.20094207674264908,
      "num_tokens": 11977410.0,
      "step": 5220
    },
    {
      "entropy": 5.433689832687378,
      "epoch": 0.5019212295869356,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004981384819674375,
      "loss": 5.2401,
      "mean_token_accuracy": 0.19651618897914885,
      "num_tokens": 11989119.0,
      "step": 5225
    },
    {
      "entropy": 5.504806232452393,
      "epoch": 0.5024015369836695,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004981340784611354,
      "loss": 5.3355,
      "mean_token_accuracy": 0.19821322858333587,
      "num_tokens": 12000165.0,
      "step": 5230
    },
    {
      "entropy": 5.426196527481079,
      "epoch": 0.5028818443804035,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004981296697743224,
      "loss": 5.2837,
      "mean_token_accuracy": 0.1921847268939018,
      "num_tokens": 12012118.0,
      "step": 5235
    },
    {
      "entropy": 5.454052591323853,
      "epoch": 0.5033621517771374,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004981252559071012,
      "loss": 5.3446,
      "mean_token_accuracy": 0.19141038358211518,
      "num_tokens": 12023432.0,
      "step": 5240
    },
    {
      "entropy": 5.45819616317749,
      "epoch": 0.5038424591738713,
      "grad_norm": 1.0,
      "learning_rate": 0.0004981208368595739,
      "loss": 5.2359,
      "mean_token_accuracy": 0.20807536840438842,
      "num_tokens": 12034323.0,
      "step": 5245
    },
    {
      "entropy": 5.384243202209473,
      "epoch": 0.5043227665706052,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004981164126318435,
      "loss": 5.2354,
      "mean_token_accuracy": 0.19932233542203903,
      "num_tokens": 12045532.0,
      "step": 5250
    },
    {
      "entropy": 5.437687397003174,
      "epoch": 0.5048030739673391,
      "grad_norm": 0.875,
      "learning_rate": 0.0004981119832240124,
      "loss": 5.2598,
      "mean_token_accuracy": 0.20076098144054413,
      "num_tokens": 12057346.0,
      "step": 5255
    },
    {
      "entropy": 5.396479225158691,
      "epoch": 0.505283381364073,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981075486361837,
      "loss": 5.2005,
      "mean_token_accuracy": 0.2075444668531418,
      "num_tokens": 12068670.0,
      "step": 5260
    },
    {
      "entropy": 5.467753219604492,
      "epoch": 0.5057636887608069,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004981031088684601,
      "loss": 5.3338,
      "mean_token_accuracy": 0.19293148219585418,
      "num_tokens": 12079664.0,
      "step": 5265
    },
    {
      "entropy": 5.470578527450561,
      "epoch": 0.5062439961575408,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004980986639209448,
      "loss": 5.2611,
      "mean_token_accuracy": 0.20865996778011323,
      "num_tokens": 12089984.0,
      "step": 5270
    },
    {
      "entropy": 5.336924886703491,
      "epoch": 0.5067243035542748,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000498094213793741,
      "loss": 5.2147,
      "mean_token_accuracy": 0.20369258522987366,
      "num_tokens": 12101182.0,
      "step": 5275
    },
    {
      "entropy": 5.461707401275635,
      "epoch": 0.5072046109510087,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498089758486952,
      "loss": 5.2908,
      "mean_token_accuracy": 0.19222524017095566,
      "num_tokens": 12112002.0,
      "step": 5280
    },
    {
      "entropy": 5.52132477760315,
      "epoch": 0.5076849183477425,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004980852980006812,
      "loss": 5.3713,
      "mean_token_accuracy": 0.18709225356578826,
      "num_tokens": 12124194.0,
      "step": 5285
    },
    {
      "entropy": 5.383759450912476,
      "epoch": 0.5081652257444764,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004980808323350323,
      "loss": 5.2764,
      "mean_token_accuracy": 0.2057688355445862,
      "num_tokens": 12133966.0,
      "step": 5290
    },
    {
      "entropy": 5.434381294250488,
      "epoch": 0.5086455331412104,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004980763614901089,
      "loss": 5.2242,
      "mean_token_accuracy": 0.20607974529266357,
      "num_tokens": 12145643.0,
      "step": 5295
    },
    {
      "entropy": 5.415269994735718,
      "epoch": 0.5091258405379443,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004980718854660146,
      "loss": 5.2754,
      "mean_token_accuracy": 0.19458574503660203,
      "num_tokens": 12156804.0,
      "step": 5300
    },
    {
      "entropy": 5.415568923950195,
      "epoch": 0.5096061479346782,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004980674042628537,
      "loss": 5.236,
      "mean_token_accuracy": 0.1962417483329773,
      "num_tokens": 12168700.0,
      "step": 5305
    },
    {
      "entropy": 5.462696838378906,
      "epoch": 0.5100864553314121,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00049806291788073,
      "loss": 5.2451,
      "mean_token_accuracy": 0.19924858808517457,
      "num_tokens": 12181050.0,
      "step": 5310
    },
    {
      "entropy": 5.497040414810181,
      "epoch": 0.510566762728146,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004980584263197477,
      "loss": 5.3312,
      "mean_token_accuracy": 0.1936553090810776,
      "num_tokens": 12192001.0,
      "step": 5315
    },
    {
      "entropy": 5.576595830917358,
      "epoch": 0.5110470701248799,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004980539295800111,
      "loss": 5.4345,
      "mean_token_accuracy": 0.19061971008777617,
      "num_tokens": 12202436.0,
      "step": 5320
    },
    {
      "entropy": 5.417716455459595,
      "epoch": 0.5115273775216138,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004980494276616246,
      "loss": 5.2237,
      "mean_token_accuracy": 0.2038380205631256,
      "num_tokens": 12214454.0,
      "step": 5325
    },
    {
      "entropy": 5.4226940155029295,
      "epoch": 0.5120076849183477,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004980449205646926,
      "loss": 5.2415,
      "mean_token_accuracy": 0.20273517370223998,
      "num_tokens": 12225924.0,
      "step": 5330
    },
    {
      "entropy": 5.481082773208618,
      "epoch": 0.5124879923150817,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049804040828932,
      "loss": 5.2504,
      "mean_token_accuracy": 0.2039707839488983,
      "num_tokens": 12236456.0,
      "step": 5335
    },
    {
      "entropy": 5.45840368270874,
      "epoch": 0.5129682997118156,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004980358908356113,
      "loss": 5.2707,
      "mean_token_accuracy": 0.2003647953271866,
      "num_tokens": 12247719.0,
      "step": 5340
    },
    {
      "entropy": 5.392190551757812,
      "epoch": 0.5134486071085494,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004980313682036717,
      "loss": 5.3122,
      "mean_token_accuracy": 0.2013251528143883,
      "num_tokens": 12259141.0,
      "step": 5345
    },
    {
      "entropy": 5.524492168426514,
      "epoch": 0.5139289145052833,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004980268403936058,
      "loss": 5.365,
      "mean_token_accuracy": 0.1915614753961563,
      "num_tokens": 12269748.0,
      "step": 5350
    },
    {
      "entropy": 5.469030904769897,
      "epoch": 0.5144092219020173,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004980223074055189,
      "loss": 5.3143,
      "mean_token_accuracy": 0.199795925617218,
      "num_tokens": 12281456.0,
      "step": 5355
    },
    {
      "entropy": 5.447213125228882,
      "epoch": 0.5148895292987512,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004980177692395164,
      "loss": 5.3019,
      "mean_token_accuracy": 0.19139748066663742,
      "num_tokens": 12293763.0,
      "step": 5360
    },
    {
      "entropy": 5.445377683639526,
      "epoch": 0.5153698366954851,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004980132258957035,
      "loss": 5.2248,
      "mean_token_accuracy": 0.20402918457984925,
      "num_tokens": 12305398.0,
      "step": 5365
    },
    {
      "entropy": 5.37006196975708,
      "epoch": 0.515850144092219,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004980086773741856,
      "loss": 5.2008,
      "mean_token_accuracy": 0.19856112748384475,
      "num_tokens": 12316582.0,
      "step": 5370
    },
    {
      "entropy": 5.4883135795593265,
      "epoch": 0.516330451488953,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004980041236750685,
      "loss": 5.3332,
      "mean_token_accuracy": 0.19816807955503463,
      "num_tokens": 12328463.0,
      "step": 5375
    },
    {
      "entropy": 5.471945953369141,
      "epoch": 0.5168107588856868,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004979995647984577,
      "loss": 5.3031,
      "mean_token_accuracy": 0.2046574831008911,
      "num_tokens": 12341040.0,
      "step": 5380
    },
    {
      "entropy": 5.377148008346557,
      "epoch": 0.5172910662824207,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004979950007444593,
      "loss": 5.2058,
      "mean_token_accuracy": 0.20244609713554382,
      "num_tokens": 12353024.0,
      "step": 5385
    },
    {
      "entropy": 5.388622713088989,
      "epoch": 0.5177713736791547,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004979904315131792,
      "loss": 5.2118,
      "mean_token_accuracy": 0.20672389566898347,
      "num_tokens": 12366100.0,
      "step": 5390
    },
    {
      "entropy": 5.371060800552368,
      "epoch": 0.5182516810758886,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004979858571047233,
      "loss": 5.2067,
      "mean_token_accuracy": 0.20542769879102707,
      "num_tokens": 12377829.0,
      "step": 5395
    },
    {
      "entropy": 5.531446743011474,
      "epoch": 0.5187319884726225,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004979812775191979,
      "loss": 5.3243,
      "mean_token_accuracy": 0.19638566672801971,
      "num_tokens": 12390830.0,
      "step": 5400
    },
    {
      "entropy": 5.379356050491333,
      "epoch": 0.5192122958693564,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004979766927567094,
      "loss": 5.183,
      "mean_token_accuracy": 0.20557043850421905,
      "num_tokens": 12401642.0,
      "step": 5405
    },
    {
      "entropy": 5.416276645660401,
      "epoch": 0.5196926032660903,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004979721028173643,
      "loss": 5.2809,
      "mean_token_accuracy": 0.19905493408441544,
      "num_tokens": 12411653.0,
      "step": 5410
    },
    {
      "entropy": 5.433461380004883,
      "epoch": 0.5201729106628242,
      "grad_norm": 1.03125,
      "learning_rate": 0.000497967507701269,
      "loss": 5.1845,
      "mean_token_accuracy": 0.2078799471259117,
      "num_tokens": 12422891.0,
      "step": 5415
    },
    {
      "entropy": 5.475499200820923,
      "epoch": 0.5206532180595581,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004979629074085303,
      "loss": 5.2669,
      "mean_token_accuracy": 0.19606062769889832,
      "num_tokens": 12434190.0,
      "step": 5420
    },
    {
      "entropy": 5.457970666885376,
      "epoch": 0.521133525456292,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004979583019392548,
      "loss": 5.3354,
      "mean_token_accuracy": 0.19612389653921128,
      "num_tokens": 12445796.0,
      "step": 5425
    },
    {
      "entropy": 5.535390472412109,
      "epoch": 0.521613832853026,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004979536912935497,
      "loss": 5.3988,
      "mean_token_accuracy": 0.1937570571899414,
      "num_tokens": 12456212.0,
      "step": 5430
    },
    {
      "entropy": 5.380803632736206,
      "epoch": 0.5220941402497599,
      "grad_norm": 1.0,
      "learning_rate": 0.000497949075471522,
      "loss": 5.1219,
      "mean_token_accuracy": 0.2022882655262947,
      "num_tokens": 12467871.0,
      "step": 5435
    },
    {
      "entropy": 5.396414804458618,
      "epoch": 0.5225744476464937,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004979444544732786,
      "loss": 5.221,
      "mean_token_accuracy": 0.1953339844942093,
      "num_tokens": 12478626.0,
      "step": 5440
    },
    {
      "entropy": 5.4133953094482425,
      "epoch": 0.5230547550432276,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000497939828298927,
      "loss": 5.304,
      "mean_token_accuracy": 0.19671797156333923,
      "num_tokens": 12491487.0,
      "step": 5445
    },
    {
      "entropy": 5.487530469894409,
      "epoch": 0.5235350624399616,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004979351969485747,
      "loss": 5.2534,
      "mean_token_accuracy": 0.19640773683786392,
      "num_tokens": 12503240.0,
      "step": 5450
    },
    {
      "entropy": 5.477804231643677,
      "epoch": 0.5240153698366955,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004979305604223291,
      "loss": 5.1975,
      "mean_token_accuracy": 0.19627538919448853,
      "num_tokens": 12513860.0,
      "step": 5455
    },
    {
      "entropy": 5.403817749023437,
      "epoch": 0.5244956772334294,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004979259187202978,
      "loss": 5.2836,
      "mean_token_accuracy": 0.19977164268493652,
      "num_tokens": 12525884.0,
      "step": 5460
    },
    {
      "entropy": 5.434985589981079,
      "epoch": 0.5249759846301633,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004979212718425887,
      "loss": 5.2083,
      "mean_token_accuracy": 0.19778544455766678,
      "num_tokens": 12536709.0,
      "step": 5465
    },
    {
      "entropy": 5.4217602729797365,
      "epoch": 0.5254562920268973,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004979166197893096,
      "loss": 5.1909,
      "mean_token_accuracy": 0.2063028946518898,
      "num_tokens": 12549727.0,
      "step": 5470
    },
    {
      "entropy": 5.387630319595337,
      "epoch": 0.5259365994236311,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004979119625605683,
      "loss": 5.2883,
      "mean_token_accuracy": 0.19354690313339235,
      "num_tokens": 12562053.0,
      "step": 5475
    },
    {
      "entropy": 5.365799236297607,
      "epoch": 0.526416906820365,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004979073001564734,
      "loss": 5.1571,
      "mean_token_accuracy": 0.20725233256816863,
      "num_tokens": 12574096.0,
      "step": 5480
    },
    {
      "entropy": 5.444792222976685,
      "epoch": 0.5268972142170989,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004979026325771328,
      "loss": 5.3429,
      "mean_token_accuracy": 0.19553550630807875,
      "num_tokens": 12585416.0,
      "step": 5485
    },
    {
      "entropy": 5.421392726898193,
      "epoch": 0.5273775216138329,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004978979598226549,
      "loss": 5.1898,
      "mean_token_accuracy": 0.20420124977827073,
      "num_tokens": 12596861.0,
      "step": 5490
    },
    {
      "entropy": 5.3878334045410154,
      "epoch": 0.5278578290105668,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004978932818931483,
      "loss": 5.2117,
      "mean_token_accuracy": 0.20604118704795837,
      "num_tokens": 12607761.0,
      "step": 5495
    },
    {
      "entropy": 5.473101997375489,
      "epoch": 0.5283381364073007,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004978885987887216,
      "loss": 5.3143,
      "mean_token_accuracy": 0.19876412749290467,
      "num_tokens": 12619889.0,
      "step": 5500
    },
    {
      "entropy": 5.52455768585205,
      "epoch": 0.5288184438040345,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004978839105094833,
      "loss": 5.2833,
      "mean_token_accuracy": 0.2017087683081627,
      "num_tokens": 12630604.0,
      "step": 5505
    },
    {
      "entropy": 5.3455263614654545,
      "epoch": 0.5292987512007685,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004978792170555426,
      "loss": 5.1887,
      "mean_token_accuracy": 0.2012774810194969,
      "num_tokens": 12641172.0,
      "step": 5510
    },
    {
      "entropy": 5.3482582569122314,
      "epoch": 0.5297790585975024,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004978745184270083,
      "loss": 5.107,
      "mean_token_accuracy": 0.21653910875320434,
      "num_tokens": 12651731.0,
      "step": 5515
    },
    {
      "entropy": 5.3865655899047855,
      "epoch": 0.5302593659942363,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004978698146239893,
      "loss": 5.2281,
      "mean_token_accuracy": 0.20066481977701187,
      "num_tokens": 12663050.0,
      "step": 5520
    },
    {
      "entropy": 5.507231616973877,
      "epoch": 0.5307396733909702,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004978651056465952,
      "loss": 5.3351,
      "mean_token_accuracy": 0.19464969784021377,
      "num_tokens": 12674732.0,
      "step": 5525
    },
    {
      "entropy": 5.335496616363526,
      "epoch": 0.5312199807877042,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000497860391494935,
      "loss": 5.1646,
      "mean_token_accuracy": 0.2027897372841835,
      "num_tokens": 12685981.0,
      "step": 5530
    },
    {
      "entropy": 5.45561728477478,
      "epoch": 0.531700288184438,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004978556721691183,
      "loss": 5.2877,
      "mean_token_accuracy": 0.19806865006685256,
      "num_tokens": 12697139.0,
      "step": 5535
    },
    {
      "entropy": 5.408766794204712,
      "epoch": 0.5321805955811719,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004978509476692547,
      "loss": 5.2436,
      "mean_token_accuracy": 0.1951587051153183,
      "num_tokens": 12708268.0,
      "step": 5540
    },
    {
      "entropy": 5.4117907047271725,
      "epoch": 0.5326609029779059,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004978462179954538,
      "loss": 5.2159,
      "mean_token_accuracy": 0.20486142337322236,
      "num_tokens": 12720715.0,
      "step": 5545
    },
    {
      "entropy": 5.401797723770142,
      "epoch": 0.5331412103746398,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004978414831478253,
      "loss": 5.1928,
      "mean_token_accuracy": 0.20377576798200608,
      "num_tokens": 12732409.0,
      "step": 5550
    },
    {
      "entropy": 5.3635334968566895,
      "epoch": 0.5336215177713737,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004978367431264794,
      "loss": 5.3332,
      "mean_token_accuracy": 0.19053958356380463,
      "num_tokens": 12745174.0,
      "step": 5555
    },
    {
      "entropy": 5.448608493804931,
      "epoch": 0.5341018251681076,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004978319979315261,
      "loss": 5.2754,
      "mean_token_accuracy": 0.20215798765420914,
      "num_tokens": 12756116.0,
      "step": 5560
    },
    {
      "entropy": 5.492583560943603,
      "epoch": 0.5345821325648416,
      "grad_norm": 1.0,
      "learning_rate": 0.0004978272475630752,
      "loss": 5.2044,
      "mean_token_accuracy": 0.20121425837278367,
      "num_tokens": 12768183.0,
      "step": 5565
    },
    {
      "entropy": 5.337130117416382,
      "epoch": 0.5350624399615754,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004978224920212374,
      "loss": 5.2125,
      "mean_token_accuracy": 0.20143257826566696,
      "num_tokens": 12778537.0,
      "step": 5570
    },
    {
      "entropy": 5.33579421043396,
      "epoch": 0.5355427473583093,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004978177313061232,
      "loss": 5.2536,
      "mean_token_accuracy": 0.19911010861396788,
      "num_tokens": 12789691.0,
      "step": 5575
    },
    {
      "entropy": 5.470923042297363,
      "epoch": 0.5360230547550432,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004978129654178426,
      "loss": 5.2871,
      "mean_token_accuracy": 0.19690274745225905,
      "num_tokens": 12801438.0,
      "step": 5580
    },
    {
      "entropy": 5.426489925384521,
      "epoch": 0.5365033621517772,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004978081943565067,
      "loss": 5.238,
      "mean_token_accuracy": 0.1983571782708168,
      "num_tokens": 12812425.0,
      "step": 5585
    },
    {
      "entropy": 5.3816330909729,
      "epoch": 0.5369836695485111,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004978034181222261,
      "loss": 5.2057,
      "mean_token_accuracy": 0.19511340707540512,
      "num_tokens": 12824735.0,
      "step": 5590
    },
    {
      "entropy": 5.472543573379516,
      "epoch": 0.537463976945245,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004977986367151119,
      "loss": 5.2914,
      "mean_token_accuracy": 0.20105631202459334,
      "num_tokens": 12835454.0,
      "step": 5595
    },
    {
      "entropy": 5.470243644714356,
      "epoch": 0.5379442843419788,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004977938501352747,
      "loss": 5.3567,
      "mean_token_accuracy": 0.19113609939813614,
      "num_tokens": 12847086.0,
      "step": 5600
    },
    {
      "entropy": 5.43897271156311,
      "epoch": 0.5384245917387128,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004977890583828259,
      "loss": 5.2946,
      "mean_token_accuracy": 0.1976897582411766,
      "num_tokens": 12857713.0,
      "step": 5605
    },
    {
      "entropy": 5.3829254627227785,
      "epoch": 0.5389048991354467,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004977842614578768,
      "loss": 5.2637,
      "mean_token_accuracy": 0.19672221690416336,
      "num_tokens": 12869967.0,
      "step": 5610
    },
    {
      "entropy": 5.49492883682251,
      "epoch": 0.5393852065321806,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004977794593605386,
      "loss": 5.2954,
      "mean_token_accuracy": 0.19768787920475006,
      "num_tokens": 12881230.0,
      "step": 5615
    },
    {
      "entropy": 5.434672880172729,
      "epoch": 0.5398655139289145,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000497774652090923,
      "loss": 5.2561,
      "mean_token_accuracy": 0.19740487784147262,
      "num_tokens": 12892376.0,
      "step": 5620
    },
    {
      "entropy": 5.423199844360352,
      "epoch": 0.5403458213256485,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004977698396491414,
      "loss": 5.2622,
      "mean_token_accuracy": 0.2015289470553398,
      "num_tokens": 12903709.0,
      "step": 5625
    },
    {
      "entropy": 5.338968229293823,
      "epoch": 0.5408261287223823,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004977650220353055,
      "loss": 5.096,
      "mean_token_accuracy": 0.20367807447910308,
      "num_tokens": 12914958.0,
      "step": 5630
    },
    {
      "entropy": 5.394136428833008,
      "epoch": 0.5413064361191162,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004977601992495274,
      "loss": 5.2204,
      "mean_token_accuracy": 0.20167420655488968,
      "num_tokens": 12927418.0,
      "step": 5635
    },
    {
      "entropy": 5.442663192749023,
      "epoch": 0.5417867435158501,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004977553712919189,
      "loss": 5.2625,
      "mean_token_accuracy": 0.20043358504772185,
      "num_tokens": 12939874.0,
      "step": 5640
    },
    {
      "entropy": 5.459798288345337,
      "epoch": 0.5422670509125841,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004977505381625921,
      "loss": 5.2809,
      "mean_token_accuracy": 0.1963661327958107,
      "num_tokens": 12951113.0,
      "step": 5645
    },
    {
      "entropy": 5.350559663772583,
      "epoch": 0.542747358309318,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004977456998616593,
      "loss": 5.1751,
      "mean_token_accuracy": 0.20273139774799348,
      "num_tokens": 12961940.0,
      "step": 5650
    },
    {
      "entropy": 5.300940227508545,
      "epoch": 0.5432276657060519,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004977408563892327,
      "loss": 5.1719,
      "mean_token_accuracy": 0.20215410143136978,
      "num_tokens": 12973938.0,
      "step": 5655
    },
    {
      "entropy": 5.438532543182373,
      "epoch": 0.5437079731027857,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004977360077454249,
      "loss": 5.1819,
      "mean_token_accuracy": 0.20405146181583406,
      "num_tokens": 12985400.0,
      "step": 5660
    },
    {
      "entropy": 5.416611433029175,
      "epoch": 0.5441882804995197,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004977311539303483,
      "loss": 5.247,
      "mean_token_accuracy": 0.20964492708444596,
      "num_tokens": 12996402.0,
      "step": 5665
    },
    {
      "entropy": 5.33284559249878,
      "epoch": 0.5446685878962536,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004977262949441158,
      "loss": 5.1338,
      "mean_token_accuracy": 0.207969768345356,
      "num_tokens": 13006991.0,
      "step": 5670
    },
    {
      "entropy": 5.431616878509521,
      "epoch": 0.5451488952929875,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004977214307868399,
      "loss": 5.2149,
      "mean_token_accuracy": 0.20395167768001557,
      "num_tokens": 13016969.0,
      "step": 5675
    },
    {
      "entropy": 5.395513486862183,
      "epoch": 0.5456292026897214,
      "grad_norm": 1.046875,
      "learning_rate": 0.000497716561458634,
      "loss": 5.1664,
      "mean_token_accuracy": 0.20513697415590287,
      "num_tokens": 13027759.0,
      "step": 5680
    },
    {
      "entropy": 5.496779155731201,
      "epoch": 0.5461095100864554,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004977116869596107,
      "loss": 5.3761,
      "mean_token_accuracy": 0.1954101249575615,
      "num_tokens": 13039881.0,
      "step": 5685
    },
    {
      "entropy": 5.441585397720337,
      "epoch": 0.5465898174831892,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004977068072898834,
      "loss": 5.2172,
      "mean_token_accuracy": 0.2028578534722328,
      "num_tokens": 13051443.0,
      "step": 5690
    },
    {
      "entropy": 5.487551069259643,
      "epoch": 0.5470701248799231,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004977019224495652,
      "loss": 5.3149,
      "mean_token_accuracy": 0.1925104007124901,
      "num_tokens": 13063474.0,
      "step": 5695
    },
    {
      "entropy": 5.36223611831665,
      "epoch": 0.547550432276657,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004976970324387698,
      "loss": 5.1564,
      "mean_token_accuracy": 0.21396491527557374,
      "num_tokens": 13074365.0,
      "step": 5700
    },
    {
      "entropy": 5.3667303085327145,
      "epoch": 0.548030739673391,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004976921372576104,
      "loss": 5.2345,
      "mean_token_accuracy": 0.20565143525600432,
      "num_tokens": 13087354.0,
      "step": 5705
    },
    {
      "entropy": 5.39310245513916,
      "epoch": 0.5485110470701249,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004976872369062011,
      "loss": 5.204,
      "mean_token_accuracy": 0.20052028000354766,
      "num_tokens": 13099306.0,
      "step": 5710
    },
    {
      "entropy": 5.456270217895508,
      "epoch": 0.5489913544668588,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004976823313846552,
      "loss": 5.3527,
      "mean_token_accuracy": 0.1909172460436821,
      "num_tokens": 13111259.0,
      "step": 5715
    },
    {
      "entropy": 5.505323982238769,
      "epoch": 0.5494716618635928,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004976774206930869,
      "loss": 5.2629,
      "mean_token_accuracy": 0.19202751219272612,
      "num_tokens": 13123589.0,
      "step": 5720
    },
    {
      "entropy": 5.3730621337890625,
      "epoch": 0.5499519692603266,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004976725048316101,
      "loss": 5.2531,
      "mean_token_accuracy": 0.20131368041038514,
      "num_tokens": 13136485.0,
      "step": 5725
    },
    {
      "entropy": 5.442380285263061,
      "epoch": 0.5504322766570605,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004976675838003388,
      "loss": 5.2377,
      "mean_token_accuracy": 0.19371712058782578,
      "num_tokens": 13148067.0,
      "step": 5730
    },
    {
      "entropy": 5.425840091705322,
      "epoch": 0.5509125840537944,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004976626575993877,
      "loss": 5.2184,
      "mean_token_accuracy": 0.19743252247571946,
      "num_tokens": 13159813.0,
      "step": 5735
    },
    {
      "entropy": 5.4400359153747555,
      "epoch": 0.5513928914505284,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976577262288705,
      "loss": 5.2761,
      "mean_token_accuracy": 0.19976773113012314,
      "num_tokens": 13170828.0,
      "step": 5740
    },
    {
      "entropy": 5.354959201812744,
      "epoch": 0.5518731988472623,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004976527896889023,
      "loss": 5.1102,
      "mean_token_accuracy": 0.2092746376991272,
      "num_tokens": 13181883.0,
      "step": 5745
    },
    {
      "entropy": 5.348735427856445,
      "epoch": 0.5523535062439962,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004976478479795974,
      "loss": 5.1869,
      "mean_token_accuracy": 0.20560824871063232,
      "num_tokens": 13193530.0,
      "step": 5750
    },
    {
      "entropy": 5.479697370529175,
      "epoch": 0.55283381364073,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004976429011010706,
      "loss": 5.409,
      "mean_token_accuracy": 0.1900714859366417,
      "num_tokens": 13205822.0,
      "step": 5755
    },
    {
      "entropy": 5.410556364059448,
      "epoch": 0.553314121037464,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004976379490534366,
      "loss": 5.1295,
      "mean_token_accuracy": 0.20924041867256166,
      "num_tokens": 13216698.0,
      "step": 5760
    },
    {
      "entropy": 5.373366641998291,
      "epoch": 0.5537944284341979,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004976329918368107,
      "loss": 5.2313,
      "mean_token_accuracy": 0.19905991554260255,
      "num_tokens": 13228389.0,
      "step": 5765
    },
    {
      "entropy": 5.449615812301635,
      "epoch": 0.5542747358309318,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004976280294513079,
      "loss": 5.2777,
      "mean_token_accuracy": 0.18989023566246033,
      "num_tokens": 13239628.0,
      "step": 5770
    },
    {
      "entropy": 5.44436993598938,
      "epoch": 0.5547550432276657,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004976230618970431,
      "loss": 5.2859,
      "mean_token_accuracy": 0.2014923021197319,
      "num_tokens": 13251149.0,
      "step": 5775
    },
    {
      "entropy": 5.518587255477906,
      "epoch": 0.5552353506243997,
      "grad_norm": 0.859375,
      "learning_rate": 0.000497618089174132,
      "loss": 5.3325,
      "mean_token_accuracy": 0.1971034973859787,
      "num_tokens": 13264846.0,
      "step": 5780
    },
    {
      "entropy": 5.339368152618408,
      "epoch": 0.5557156580211335,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004976131112826898,
      "loss": 5.1132,
      "mean_token_accuracy": 0.21116390824317932,
      "num_tokens": 13275409.0,
      "step": 5785
    },
    {
      "entropy": 5.2891651630401615,
      "epoch": 0.5561959654178674,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004976081282228323,
      "loss": 5.0913,
      "mean_token_accuracy": 0.21006590127944946,
      "num_tokens": 13287173.0,
      "step": 5790
    },
    {
      "entropy": 5.436860179901123,
      "epoch": 0.5566762728146013,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000497603139994675,
      "loss": 5.1804,
      "mean_token_accuracy": 0.20149831473827362,
      "num_tokens": 13298225.0,
      "step": 5795
    },
    {
      "entropy": 5.383152008056641,
      "epoch": 0.5571565802113353,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004975981465983338,
      "loss": 5.183,
      "mean_token_accuracy": 0.20298803150653838,
      "num_tokens": 13309685.0,
      "step": 5800
    },
    {
      "entropy": 5.355784368515015,
      "epoch": 0.5576368876080692,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975931480339246,
      "loss": 5.1935,
      "mean_token_accuracy": 0.20758654028177262,
      "num_tokens": 13320837.0,
      "step": 5805
    },
    {
      "entropy": 5.400020551681519,
      "epoch": 0.5581171950048031,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004975881443015635,
      "loss": 5.2981,
      "mean_token_accuracy": 0.1994555786252022,
      "num_tokens": 13333512.0,
      "step": 5810
    },
    {
      "entropy": 5.499358034133911,
      "epoch": 0.5585975024015369,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004975831354013667,
      "loss": 5.3286,
      "mean_token_accuracy": 0.19726525396108627,
      "num_tokens": 13345189.0,
      "step": 5815
    },
    {
      "entropy": 5.400446796417237,
      "epoch": 0.5590778097982709,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004975781213334503,
      "loss": 5.1814,
      "mean_token_accuracy": 0.20379290580749512,
      "num_tokens": 13356123.0,
      "step": 5820
    },
    {
      "entropy": 5.37984790802002,
      "epoch": 0.5595581171950048,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004975731020979309,
      "loss": 5.2375,
      "mean_token_accuracy": 0.19814090579748153,
      "num_tokens": 13366902.0,
      "step": 5825
    },
    {
      "entropy": 5.458560657501221,
      "epoch": 0.5600384245917387,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004975680776949249,
      "loss": 5.2688,
      "mean_token_accuracy": 0.1988992065191269,
      "num_tokens": 13377567.0,
      "step": 5830
    },
    {
      "entropy": 5.446119594573974,
      "epoch": 0.5605187319884726,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004975630481245492,
      "loss": 5.2135,
      "mean_token_accuracy": 0.21275554597377777,
      "num_tokens": 13387297.0,
      "step": 5835
    },
    {
      "entropy": 5.448615026473999,
      "epoch": 0.5609990393852066,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004975580133869202,
      "loss": 5.2517,
      "mean_token_accuracy": 0.203067584335804,
      "num_tokens": 13397723.0,
      "step": 5840
    },
    {
      "entropy": 5.4316037654876705,
      "epoch": 0.5614793467819404,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004975529734821552,
      "loss": 5.302,
      "mean_token_accuracy": 0.19728813618421553,
      "num_tokens": 13409875.0,
      "step": 5845
    },
    {
      "entropy": 5.390988826751709,
      "epoch": 0.5619596541786743,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004975479284103708,
      "loss": 5.2099,
      "mean_token_accuracy": 0.20556127578020095,
      "num_tokens": 13421338.0,
      "step": 5850
    },
    {
      "entropy": 5.444341087341309,
      "epoch": 0.5624399615754082,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004975428781716845,
      "loss": 5.2321,
      "mean_token_accuracy": 0.20571468025445938,
      "num_tokens": 13431373.0,
      "step": 5855
    },
    {
      "entropy": 5.428695297241211,
      "epoch": 0.5629202689721422,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004975378227662134,
      "loss": 5.267,
      "mean_token_accuracy": 0.20486005395650864,
      "num_tokens": 13443158.0,
      "step": 5860
    },
    {
      "entropy": 5.476019096374512,
      "epoch": 0.5634005763688761,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004975327621940746,
      "loss": 5.3266,
      "mean_token_accuracy": 0.19317104965448378,
      "num_tokens": 13454559.0,
      "step": 5865
    },
    {
      "entropy": 5.550100421905517,
      "epoch": 0.56388088376561,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004975276964553861,
      "loss": 5.387,
      "mean_token_accuracy": 0.19499229043722152,
      "num_tokens": 13466934.0,
      "step": 5870
    },
    {
      "entropy": 5.421635150909424,
      "epoch": 0.5643611911623438,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004975226255502651,
      "loss": 5.1407,
      "mean_token_accuracy": 0.21336206346750258,
      "num_tokens": 13477770.0,
      "step": 5875
    },
    {
      "entropy": 5.469849634170532,
      "epoch": 0.5648414985590778,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004975175494788297,
      "loss": 5.3372,
      "mean_token_accuracy": 0.194351489841938,
      "num_tokens": 13490093.0,
      "step": 5880
    },
    {
      "entropy": 5.486195516586304,
      "epoch": 0.5653218059558117,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004975124682411974,
      "loss": 5.2018,
      "mean_token_accuracy": 0.2025473579764366,
      "num_tokens": 13500663.0,
      "step": 5885
    },
    {
      "entropy": 5.460525321960449,
      "epoch": 0.5658021133525456,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004975073818374863,
      "loss": 5.2964,
      "mean_token_accuracy": 0.19814062863588333,
      "num_tokens": 13512369.0,
      "step": 5890
    },
    {
      "entropy": 5.4164266109466555,
      "epoch": 0.5662824207492796,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004975022902678145,
      "loss": 5.181,
      "mean_token_accuracy": 0.20385072082281114,
      "num_tokens": 13523181.0,
      "step": 5895
    },
    {
      "entropy": 5.385796976089478,
      "epoch": 0.5667627281460135,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004974971935323003,
      "loss": 5.1334,
      "mean_token_accuracy": 0.20066388696432114,
      "num_tokens": 13534113.0,
      "step": 5900
    },
    {
      "entropy": 5.329702281951905,
      "epoch": 0.5672430355427474,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004974920916310619,
      "loss": 5.1819,
      "mean_token_accuracy": 0.20250354558229447,
      "num_tokens": 13545037.0,
      "step": 5905
    },
    {
      "entropy": 5.408524751663208,
      "epoch": 0.5677233429394812,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004974869845642178,
      "loss": 5.2304,
      "mean_token_accuracy": 0.20249072313308716,
      "num_tokens": 13555541.0,
      "step": 5910
    },
    {
      "entropy": 5.388197135925293,
      "epoch": 0.5682036503362152,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004974818723318866,
      "loss": 5.2069,
      "mean_token_accuracy": 0.2064737856388092,
      "num_tokens": 13566951.0,
      "step": 5915
    },
    {
      "entropy": 5.446853971481323,
      "epoch": 0.5686839577329491,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004974767549341868,
      "loss": 5.2924,
      "mean_token_accuracy": 0.19983698725700377,
      "num_tokens": 13578492.0,
      "step": 5920
    },
    {
      "entropy": 5.443757009506226,
      "epoch": 0.569164265129683,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974716323712376,
      "loss": 5.1818,
      "mean_token_accuracy": 0.20955992639064788,
      "num_tokens": 13589183.0,
      "step": 5925
    },
    {
      "entropy": 5.421390628814697,
      "epoch": 0.5696445725264169,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004974665046431576,
      "loss": 5.2909,
      "mean_token_accuracy": 0.19809137284755707,
      "num_tokens": 13600588.0,
      "step": 5930
    },
    {
      "entropy": 5.376427507400512,
      "epoch": 0.5701248799231509,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004974613717500659,
      "loss": 5.192,
      "mean_token_accuracy": 0.20923979729413986,
      "num_tokens": 13612107.0,
      "step": 5935
    },
    {
      "entropy": 5.5154797554016115,
      "epoch": 0.5706051873198847,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004974562336920818,
      "loss": 5.359,
      "mean_token_accuracy": 0.19478144943714143,
      "num_tokens": 13623973.0,
      "step": 5940
    },
    {
      "entropy": 5.39094648361206,
      "epoch": 0.5710854947166186,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004974510904693245,
      "loss": 5.1651,
      "mean_token_accuracy": 0.20794936567544936,
      "num_tokens": 13634994.0,
      "step": 5945
    },
    {
      "entropy": 5.455427503585815,
      "epoch": 0.5715658021133525,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004974459420819134,
      "loss": 5.3206,
      "mean_token_accuracy": 0.19929085075855255,
      "num_tokens": 13646361.0,
      "step": 5950
    },
    {
      "entropy": 5.420032691955567,
      "epoch": 0.5720461095100865,
      "grad_norm": 1.0,
      "learning_rate": 0.000497440788529968,
      "loss": 5.2128,
      "mean_token_accuracy": 0.20158523619174956,
      "num_tokens": 13656975.0,
      "step": 5955
    },
    {
      "entropy": 5.401870012283325,
      "epoch": 0.5725264169068204,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004974356298136081,
      "loss": 5.252,
      "mean_token_accuracy": 0.20242744088172912,
      "num_tokens": 13668434.0,
      "step": 5960
    },
    {
      "entropy": 5.414079713821411,
      "epoch": 0.5730067243035543,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974304659329533,
      "loss": 5.2351,
      "mean_token_accuracy": 0.2013985440135002,
      "num_tokens": 13679266.0,
      "step": 5965
    },
    {
      "entropy": 5.358348417282104,
      "epoch": 0.5734870317002881,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004974252968881236,
      "loss": 5.2507,
      "mean_token_accuracy": 0.19810654670000077,
      "num_tokens": 13690921.0,
      "step": 5970
    },
    {
      "entropy": 5.451554870605468,
      "epoch": 0.5739673390970221,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000497420122679239,
      "loss": 5.1919,
      "mean_token_accuracy": 0.2040702223777771,
      "num_tokens": 13702329.0,
      "step": 5975
    },
    {
      "entropy": 5.353938722610474,
      "epoch": 0.574447646493756,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004974149433064196,
      "loss": 5.1619,
      "mean_token_accuracy": 0.20967793166637422,
      "num_tokens": 13713356.0,
      "step": 5980
    },
    {
      "entropy": 5.297184705734253,
      "epoch": 0.5749279538904899,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004974097587697856,
      "loss": 5.1537,
      "mean_token_accuracy": 0.19970266073942183,
      "num_tokens": 13724718.0,
      "step": 5985
    },
    {
      "entropy": 5.3087554454803465,
      "epoch": 0.5754082612872238,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004974045690694575,
      "loss": 5.178,
      "mean_token_accuracy": 0.20804403126239776,
      "num_tokens": 13736113.0,
      "step": 5990
    },
    {
      "entropy": 5.485042858123779,
      "epoch": 0.5758885686839578,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004973993742055557,
      "loss": 5.2045,
      "mean_token_accuracy": 0.20523133128881454,
      "num_tokens": 13748322.0,
      "step": 5995
    },
    {
      "entropy": 5.351669263839722,
      "epoch": 0.5763688760806917,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004973941741782007,
      "loss": 5.2036,
      "mean_token_accuracy": 0.19881592094898223,
      "num_tokens": 13759433.0,
      "step": 6000
    },
    {
      "epoch": 0.5763688760806917,
      "eval_entropy": 5.173574446934365,
      "eval_loss": 5.231561183929443,
      "eval_mean_token_accuracy": 0.20952744792778133,
      "eval_num_tokens": 13759433.0,
      "eval_runtime": 20.3933,
      "eval_samples_per_second": 1609.104,
      "eval_steps_per_second": 201.144,
      "step": 6000
    },
    {
      "entropy": 5.317676401138305,
      "epoch": 0.5768491834774255,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004973889689875135,
      "loss": 5.1242,
      "mean_token_accuracy": 0.20577383935451507,
      "num_tokens": 13770181.0,
      "step": 6005
    },
    {
      "entropy": 5.34284462928772,
      "epoch": 0.5773294908741594,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004973837586336147,
      "loss": 5.1693,
      "mean_token_accuracy": 0.2017221912741661,
      "num_tokens": 13781792.0,
      "step": 6010
    },
    {
      "entropy": 5.397495365142822,
      "epoch": 0.5778097982708934,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004973785431166254,
      "loss": 5.223,
      "mean_token_accuracy": 0.19736747443675995,
      "num_tokens": 13792101.0,
      "step": 6015
    },
    {
      "entropy": 5.408102750778198,
      "epoch": 0.5782901056676273,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004973733224366666,
      "loss": 5.2252,
      "mean_token_accuracy": 0.20430949330329895,
      "num_tokens": 13803640.0,
      "step": 6020
    },
    {
      "entropy": 5.417121553421021,
      "epoch": 0.5787704130643612,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004973680965938597,
      "loss": 5.149,
      "mean_token_accuracy": 0.20893548280000687,
      "num_tokens": 13815017.0,
      "step": 6025
    },
    {
      "entropy": 5.47672209739685,
      "epoch": 0.579250720461095,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004973628655883258,
      "loss": 5.3584,
      "mean_token_accuracy": 0.19625595062971116,
      "num_tokens": 13826119.0,
      "step": 6030
    },
    {
      "entropy": 5.360985326766968,
      "epoch": 0.579731027857829,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004973576294201865,
      "loss": 5.1834,
      "mean_token_accuracy": 0.20206461399793624,
      "num_tokens": 13837869.0,
      "step": 6035
    },
    {
      "entropy": 5.567944145202636,
      "epoch": 0.5802113352545629,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004973523880895633,
      "loss": 5.3422,
      "mean_token_accuracy": 0.1880539357662201,
      "num_tokens": 13849333.0,
      "step": 6040
    },
    {
      "entropy": 5.459596252441406,
      "epoch": 0.5806916426512968,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004973471415965779,
      "loss": 5.1904,
      "mean_token_accuracy": 0.20356349796056747,
      "num_tokens": 13860648.0,
      "step": 6045
    },
    {
      "entropy": 5.490988540649414,
      "epoch": 0.5811719500480308,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497341889941352,
      "loss": 5.3429,
      "mean_token_accuracy": 0.1951858326792717,
      "num_tokens": 13872979.0,
      "step": 6050
    },
    {
      "entropy": 5.409925270080566,
      "epoch": 0.5816522574447647,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004973366331240078,
      "loss": 5.2426,
      "mean_token_accuracy": 0.19907674342393875,
      "num_tokens": 13884363.0,
      "step": 6055
    },
    {
      "entropy": 5.414319229125977,
      "epoch": 0.5821325648414986,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000497331371144667,
      "loss": 5.2063,
      "mean_token_accuracy": 0.1978744849562645,
      "num_tokens": 13895424.0,
      "step": 6060
    },
    {
      "entropy": 5.446793508529663,
      "epoch": 0.5826128722382324,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004973261040034521,
      "loss": 5.3301,
      "mean_token_accuracy": 0.1934633269906044,
      "num_tokens": 13907319.0,
      "step": 6065
    },
    {
      "entropy": 5.335920190811157,
      "epoch": 0.5830931796349664,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004973208317004852,
      "loss": 5.168,
      "mean_token_accuracy": 0.20753708779811858,
      "num_tokens": 13920013.0,
      "step": 6070
    },
    {
      "entropy": 5.3865642070770265,
      "epoch": 0.5835734870317003,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004973155542358889,
      "loss": 5.138,
      "mean_token_accuracy": 0.20259324014186858,
      "num_tokens": 13932033.0,
      "step": 6075
    },
    {
      "entropy": 5.333774566650391,
      "epoch": 0.5840537944284342,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004973102716097853,
      "loss": 5.1623,
      "mean_token_accuracy": 0.2091463476419449,
      "num_tokens": 13943324.0,
      "step": 6080
    },
    {
      "entropy": 5.408417177200318,
      "epoch": 0.5845341018251681,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004973049838222973,
      "loss": 5.2317,
      "mean_token_accuracy": 0.21122615337371825,
      "num_tokens": 13954291.0,
      "step": 6085
    },
    {
      "entropy": 5.465994453430175,
      "epoch": 0.5850144092219021,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004972996908735479,
      "loss": 5.3392,
      "mean_token_accuracy": 0.19264212399721145,
      "num_tokens": 13966264.0,
      "step": 6090
    },
    {
      "entropy": 5.490592527389526,
      "epoch": 0.585494716618636,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004972943927636597,
      "loss": 5.2102,
      "mean_token_accuracy": 0.19654978811740875,
      "num_tokens": 13977785.0,
      "step": 6095
    },
    {
      "entropy": 5.3439209938049315,
      "epoch": 0.5859750240153698,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004972890894927558,
      "loss": 5.2046,
      "mean_token_accuracy": 0.20348718762397766,
      "num_tokens": 13989704.0,
      "step": 6100
    },
    {
      "entropy": 5.414724111557007,
      "epoch": 0.5864553314121037,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004972837810609592,
      "loss": 5.1788,
      "mean_token_accuracy": 0.2061559408903122,
      "num_tokens": 14000565.0,
      "step": 6105
    },
    {
      "entropy": 5.300512218475342,
      "epoch": 0.5869356388088377,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004972784674683933,
      "loss": 5.1034,
      "mean_token_accuracy": 0.21265693455934526,
      "num_tokens": 14011521.0,
      "step": 6110
    },
    {
      "entropy": 5.342484712600708,
      "epoch": 0.5874159462055716,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004972731487151815,
      "loss": 5.1484,
      "mean_token_accuracy": 0.20898157060146333,
      "num_tokens": 14022966.0,
      "step": 6115
    },
    {
      "entropy": 5.496384477615356,
      "epoch": 0.5878962536023055,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004972678248014471,
      "loss": 5.2818,
      "mean_token_accuracy": 0.20139852464199065,
      "num_tokens": 14034905.0,
      "step": 6120
    },
    {
      "entropy": 5.296892976760864,
      "epoch": 0.5883765609990393,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004972624957273139,
      "loss": 5.1224,
      "mean_token_accuracy": 0.20901353806257247,
      "num_tokens": 14045816.0,
      "step": 6125
    },
    {
      "entropy": 5.331165027618408,
      "epoch": 0.5888568683957733,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004972571614929055,
      "loss": 5.1156,
      "mean_token_accuracy": 0.20691063702106477,
      "num_tokens": 14057316.0,
      "step": 6130
    },
    {
      "entropy": 5.4494264125823975,
      "epoch": 0.5893371757925072,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004972518220983457,
      "loss": 5.2315,
      "mean_token_accuracy": 0.194955912232399,
      "num_tokens": 14067542.0,
      "step": 6135
    },
    {
      "entropy": 5.300818586349488,
      "epoch": 0.5898174831892411,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004972464775437586,
      "loss": 5.1082,
      "mean_token_accuracy": 0.2038309842348099,
      "num_tokens": 14079467.0,
      "step": 6140
    },
    {
      "entropy": 5.322385835647583,
      "epoch": 0.590297790585975,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004972411278292683,
      "loss": 5.0573,
      "mean_token_accuracy": 0.20967037081718445,
      "num_tokens": 14090695.0,
      "step": 6145
    },
    {
      "entropy": 5.365872621536255,
      "epoch": 0.590778097982709,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004972357729549988,
      "loss": 5.1054,
      "mean_token_accuracy": 0.21425690948963166,
      "num_tokens": 14101380.0,
      "step": 6150
    },
    {
      "entropy": 5.316104984283447,
      "epoch": 0.5912584053794429,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004972304129210746,
      "loss": 5.158,
      "mean_token_accuracy": 0.20969926416873932,
      "num_tokens": 14112265.0,
      "step": 6155
    },
    {
      "entropy": 5.376876020431519,
      "epoch": 0.5917387127761767,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004972250477276202,
      "loss": 5.1411,
      "mean_token_accuracy": 0.20484964549541473,
      "num_tokens": 14124294.0,
      "step": 6160
    },
    {
      "entropy": 5.383788347244263,
      "epoch": 0.5922190201729106,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004972196773747599,
      "loss": 5.1806,
      "mean_token_accuracy": 0.2072070226073265,
      "num_tokens": 14135230.0,
      "step": 6165
    },
    {
      "entropy": 5.435587596893311,
      "epoch": 0.5926993275696446,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004972143018626186,
      "loss": 5.2706,
      "mean_token_accuracy": 0.199320986866951,
      "num_tokens": 14147576.0,
      "step": 6170
    },
    {
      "entropy": 5.357582330703735,
      "epoch": 0.5931796349663785,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004972089211913211,
      "loss": 5.1463,
      "mean_token_accuracy": 0.20706107169389726,
      "num_tokens": 14158834.0,
      "step": 6175
    },
    {
      "entropy": 5.406814289093018,
      "epoch": 0.5936599423631124,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004972035353609923,
      "loss": 5.284,
      "mean_token_accuracy": 0.1946965351700783,
      "num_tokens": 14170694.0,
      "step": 6180
    },
    {
      "entropy": 5.382691860198975,
      "epoch": 0.5941402497598463,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004971981443717572,
      "loss": 5.1948,
      "mean_token_accuracy": 0.20373794585466384,
      "num_tokens": 14183184.0,
      "step": 6185
    },
    {
      "entropy": 5.499075698852539,
      "epoch": 0.5946205571565802,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004971927482237409,
      "loss": 5.2727,
      "mean_token_accuracy": 0.19923094511032105,
      "num_tokens": 14194761.0,
      "step": 6190
    },
    {
      "entropy": 5.346008777618408,
      "epoch": 0.5951008645533141,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004971873469170689,
      "loss": 5.0874,
      "mean_token_accuracy": 0.2069292798638344,
      "num_tokens": 14205820.0,
      "step": 6195
    },
    {
      "entropy": 5.285473346710205,
      "epoch": 0.595581171950048,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004971819404518664,
      "loss": 5.1194,
      "mean_token_accuracy": 0.2080278217792511,
      "num_tokens": 14217826.0,
      "step": 6200
    },
    {
      "entropy": 5.355868816375732,
      "epoch": 0.5960614793467819,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000497176528828259,
      "loss": 5.0693,
      "mean_token_accuracy": 0.2167549431324005,
      "num_tokens": 14228632.0,
      "step": 6205
    },
    {
      "entropy": 5.407436847686768,
      "epoch": 0.5965417867435159,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004971711120463722,
      "loss": 5.3142,
      "mean_token_accuracy": 0.19621515572071074,
      "num_tokens": 14240231.0,
      "step": 6210
    },
    {
      "entropy": 5.441270589828491,
      "epoch": 0.5970220941402498,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000497165690106332,
      "loss": 5.1491,
      "mean_token_accuracy": 0.20620052814483641,
      "num_tokens": 14251718.0,
      "step": 6215
    },
    {
      "entropy": 5.402524423599243,
      "epoch": 0.5975024015369836,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004971602630082642,
      "loss": 5.1877,
      "mean_token_accuracy": 0.21023498028516768,
      "num_tokens": 14263129.0,
      "step": 6220
    },
    {
      "entropy": 5.316794395446777,
      "epoch": 0.5979827089337176,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004971548307522947,
      "loss": 5.1042,
      "mean_token_accuracy": 0.20603148490190507,
      "num_tokens": 14274465.0,
      "step": 6225
    },
    {
      "entropy": 5.442112064361572,
      "epoch": 0.5984630163304515,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004971493933385498,
      "loss": 5.2854,
      "mean_token_accuracy": 0.20513668954372405,
      "num_tokens": 14286190.0,
      "step": 6230
    },
    {
      "entropy": 5.433134078979492,
      "epoch": 0.5989433237271854,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004971439507671556,
      "loss": 5.2285,
      "mean_token_accuracy": 0.2017318844795227,
      "num_tokens": 14297638.0,
      "step": 6235
    },
    {
      "entropy": 5.346106767654419,
      "epoch": 0.5994236311239193,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004971385030382384,
      "loss": 5.1661,
      "mean_token_accuracy": 0.20881230533123016,
      "num_tokens": 14309773.0,
      "step": 6240
    },
    {
      "entropy": 5.369094181060791,
      "epoch": 0.5999039385206533,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004971330501519248,
      "loss": 5.1494,
      "mean_token_accuracy": 0.2099878966808319,
      "num_tokens": 14320543.0,
      "step": 6245
    },
    {
      "entropy": 5.409315919876098,
      "epoch": 0.6003842459173871,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004971275921083414,
      "loss": 5.2246,
      "mean_token_accuracy": 0.19912684857845306,
      "num_tokens": 14332076.0,
      "step": 6250
    },
    {
      "entropy": 5.338934898376465,
      "epoch": 0.600864553314121,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000497122128907615,
      "loss": 5.1418,
      "mean_token_accuracy": 0.20467256158590316,
      "num_tokens": 14343768.0,
      "step": 6255
    },
    {
      "entropy": 5.336265516281128,
      "epoch": 0.6013448607108549,
      "grad_norm": 0.875,
      "learning_rate": 0.0004971166605498722,
      "loss": 5.1778,
      "mean_token_accuracy": 0.20175106227397918,
      "num_tokens": 14357051.0,
      "step": 6260
    },
    {
      "entropy": 5.380846071243286,
      "epoch": 0.6018251681075889,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004971111870352401,
      "loss": 5.2108,
      "mean_token_accuracy": 0.19954999536275864,
      "num_tokens": 14368800.0,
      "step": 6265
    },
    {
      "entropy": 5.387369632720947,
      "epoch": 0.6023054755043228,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004971057083638458,
      "loss": 5.1885,
      "mean_token_accuracy": 0.21062051504850388,
      "num_tokens": 14379617.0,
      "step": 6270
    },
    {
      "entropy": 5.346527814865112,
      "epoch": 0.6027857829010567,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004971002245358166,
      "loss": 5.1559,
      "mean_token_accuracy": 0.21145128458738327,
      "num_tokens": 14391454.0,
      "step": 6275
    },
    {
      "entropy": 5.3886960506439205,
      "epoch": 0.6032660902977905,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004970947355512795,
      "loss": 5.1591,
      "mean_token_accuracy": 0.2130463242530823,
      "num_tokens": 14402379.0,
      "step": 6280
    },
    {
      "entropy": 5.3505042552947994,
      "epoch": 0.6037463976945245,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004970892414103622,
      "loss": 5.1614,
      "mean_token_accuracy": 0.20644438415765762,
      "num_tokens": 14415040.0,
      "step": 6285
    },
    {
      "entropy": 5.376429271697998,
      "epoch": 0.6042267050912584,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004970837421131921,
      "loss": 5.2567,
      "mean_token_accuracy": 0.19653793871402742,
      "num_tokens": 14426677.0,
      "step": 6290
    },
    {
      "entropy": 5.396064376831054,
      "epoch": 0.6047070124879923,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004970782376598972,
      "loss": 5.1587,
      "mean_token_accuracy": 0.20533006340265275,
      "num_tokens": 14436676.0,
      "step": 6295
    },
    {
      "entropy": 5.392739105224609,
      "epoch": 0.6051873198847262,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004970727280506048,
      "loss": 5.2039,
      "mean_token_accuracy": 0.19733990132808685,
      "num_tokens": 14448294.0,
      "step": 6300
    },
    {
      "entropy": 5.315736722946167,
      "epoch": 0.6056676272814602,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004970672132854431,
      "loss": 5.1622,
      "mean_token_accuracy": 0.20562706589698793,
      "num_tokens": 14460642.0,
      "step": 6305
    },
    {
      "entropy": 5.338218450546265,
      "epoch": 0.6061479346781941,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004970616933645403,
      "loss": 5.0822,
      "mean_token_accuracy": 0.21485713571310044,
      "num_tokens": 14471370.0,
      "step": 6310
    },
    {
      "entropy": 5.4621439456939695,
      "epoch": 0.6066282420749279,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004970561682880242,
      "loss": 5.2419,
      "mean_token_accuracy": 0.20577811002731322,
      "num_tokens": 14482358.0,
      "step": 6315
    },
    {
      "entropy": 5.4311422348022464,
      "epoch": 0.6071085494716618,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004970506380560233,
      "loss": 5.2623,
      "mean_token_accuracy": 0.20287067741155623,
      "num_tokens": 14494413.0,
      "step": 6320
    },
    {
      "entropy": 5.398257827758789,
      "epoch": 0.6075888568683958,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004970451026686659,
      "loss": 5.1725,
      "mean_token_accuracy": 0.21675312221050264,
      "num_tokens": 14506370.0,
      "step": 6325
    },
    {
      "entropy": 5.395339870452881,
      "epoch": 0.6080691642651297,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004970395621260806,
      "loss": 5.1651,
      "mean_token_accuracy": 0.2110365942120552,
      "num_tokens": 14517198.0,
      "step": 6330
    },
    {
      "entropy": 5.46751856803894,
      "epoch": 0.6085494716618636,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000497034016428396,
      "loss": 5.2693,
      "mean_token_accuracy": 0.19731477200984954,
      "num_tokens": 14529434.0,
      "step": 6335
    },
    {
      "entropy": 5.328270530700683,
      "epoch": 0.6090297790585975,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004970284655757409,
      "loss": 5.1596,
      "mean_token_accuracy": 0.2077682361006737,
      "num_tokens": 14540857.0,
      "step": 6340
    },
    {
      "entropy": 5.332006740570068,
      "epoch": 0.6095100864553314,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004970229095682439,
      "loss": 5.1729,
      "mean_token_accuracy": 0.20433960258960723,
      "num_tokens": 14552594.0,
      "step": 6345
    },
    {
      "entropy": 5.420663452148437,
      "epoch": 0.6099903938520653,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004970173484060344,
      "loss": 5.2244,
      "mean_token_accuracy": 0.20081303864717484,
      "num_tokens": 14563599.0,
      "step": 6350
    },
    {
      "entropy": 5.411350440979004,
      "epoch": 0.6104707012487992,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004970117820892414,
      "loss": 5.1663,
      "mean_token_accuracy": 0.20423975884914397,
      "num_tokens": 14575905.0,
      "step": 6355
    },
    {
      "entropy": 5.417771673202514,
      "epoch": 0.6109510086455331,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004970062106179939,
      "loss": 5.1986,
      "mean_token_accuracy": 0.2102581739425659,
      "num_tokens": 14587800.0,
      "step": 6360
    },
    {
      "entropy": 5.382213973999024,
      "epoch": 0.6114313160422671,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004970006339924214,
      "loss": 5.1492,
      "mean_token_accuracy": 0.2105979710817337,
      "num_tokens": 14600654.0,
      "step": 6365
    },
    {
      "entropy": 5.389059114456177,
      "epoch": 0.611911623439001,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004969950522126534,
      "loss": 5.1494,
      "mean_token_accuracy": 0.2078918397426605,
      "num_tokens": 14611985.0,
      "step": 6370
    },
    {
      "entropy": 5.326954650878906,
      "epoch": 0.6123919308357348,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004969894652788196,
      "loss": 5.1482,
      "mean_token_accuracy": 0.20942535400390624,
      "num_tokens": 14625004.0,
      "step": 6375
    },
    {
      "entropy": 5.39363808631897,
      "epoch": 0.6128722382324687,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004969838731910494,
      "loss": 5.1972,
      "mean_token_accuracy": 0.20212942659854888,
      "num_tokens": 14635381.0,
      "step": 6380
    },
    {
      "entropy": 5.327234983444214,
      "epoch": 0.6133525456292027,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004969782759494729,
      "loss": 5.1525,
      "mean_token_accuracy": 0.211282180249691,
      "num_tokens": 14646582.0,
      "step": 6385
    },
    {
      "entropy": 5.334175634384155,
      "epoch": 0.6138328530259366,
      "grad_norm": 0.9375,
      "learning_rate": 0.00049697267355422,
      "loss": 5.2168,
      "mean_token_accuracy": 0.20744510143995284,
      "num_tokens": 14658024.0,
      "step": 6390
    },
    {
      "entropy": 5.537168884277344,
      "epoch": 0.6143131604226705,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004969670660054208,
      "loss": 5.2515,
      "mean_token_accuracy": 0.1993764966726303,
      "num_tokens": 14669933.0,
      "step": 6395
    },
    {
      "entropy": 5.426605176925659,
      "epoch": 0.6147934678194045,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004969614533032054,
      "loss": 5.1954,
      "mean_token_accuracy": 0.2035425618290901,
      "num_tokens": 14681331.0,
      "step": 6400
    },
    {
      "entropy": 5.25716814994812,
      "epoch": 0.6152737752161384,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004969558354477041,
      "loss": 5.0418,
      "mean_token_accuracy": 0.21936360746622086,
      "num_tokens": 14691361.0,
      "step": 6405
    },
    {
      "entropy": 5.337737512588501,
      "epoch": 0.6157540826128722,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004969502124390474,
      "loss": 5.1451,
      "mean_token_accuracy": 0.2076416000723839,
      "num_tokens": 14701791.0,
      "step": 6410
    },
    {
      "entropy": 5.4165130138397215,
      "epoch": 0.6162343900096061,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004969445842773658,
      "loss": 5.2218,
      "mean_token_accuracy": 0.20131406635046006,
      "num_tokens": 14713672.0,
      "step": 6415
    },
    {
      "entropy": 5.3496606826782225,
      "epoch": 0.6167146974063401,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00049693895096279,
      "loss": 5.0661,
      "mean_token_accuracy": 0.2134489744901657,
      "num_tokens": 14724789.0,
      "step": 6420
    },
    {
      "entropy": 5.311793231964112,
      "epoch": 0.617195004803074,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004969333124954508,
      "loss": 5.1384,
      "mean_token_accuracy": 0.20452271401882172,
      "num_tokens": 14737212.0,
      "step": 6425
    },
    {
      "entropy": 5.303396415710449,
      "epoch": 0.6176753121998079,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004969276688754791,
      "loss": 5.1577,
      "mean_token_accuracy": 0.20288337469100953,
      "num_tokens": 14748387.0,
      "step": 6430
    },
    {
      "entropy": 5.390523481369018,
      "epoch": 0.6181556195965417,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004969220201030059,
      "loss": 5.1743,
      "mean_token_accuracy": 0.20002119094133378,
      "num_tokens": 14758477.0,
      "step": 6435
    },
    {
      "entropy": 5.424677038192749,
      "epoch": 0.6186359269932757,
      "grad_norm": 1.0,
      "learning_rate": 0.0004969163661781624,
      "loss": 5.2673,
      "mean_token_accuracy": 0.2052892789244652,
      "num_tokens": 14769650.0,
      "step": 6440
    },
    {
      "entropy": 5.289681673049927,
      "epoch": 0.6191162343900096,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004969107071010798,
      "loss": 5.0647,
      "mean_token_accuracy": 0.20910587459802626,
      "num_tokens": 14780988.0,
      "step": 6445
    },
    {
      "entropy": 5.361701822280883,
      "epoch": 0.6195965417867435,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004969050428718895,
      "loss": 5.1561,
      "mean_token_accuracy": 0.20649783462285995,
      "num_tokens": 14792458.0,
      "step": 6450
    },
    {
      "entropy": 5.406031084060669,
      "epoch": 0.6200768491834774,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000496899373490723,
      "loss": 5.22,
      "mean_token_accuracy": 0.20498424172401428,
      "num_tokens": 14804772.0,
      "step": 6455
    },
    {
      "entropy": 5.400968217849732,
      "epoch": 0.6205571565802114,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000496893698957712,
      "loss": 5.1514,
      "mean_token_accuracy": 0.2003758192062378,
      "num_tokens": 14816126.0,
      "step": 6460
    },
    {
      "entropy": 5.400296592712403,
      "epoch": 0.6210374639769453,
      "grad_norm": 1.0,
      "learning_rate": 0.0004968880192729882,
      "loss": 5.2769,
      "mean_token_accuracy": 0.1981361836194992,
      "num_tokens": 14829376.0,
      "step": 6465
    },
    {
      "entropy": 5.335179996490479,
      "epoch": 0.6215177713736791,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004968823344366835,
      "loss": 5.0786,
      "mean_token_accuracy": 0.21548613011837006,
      "num_tokens": 14841098.0,
      "step": 6470
    },
    {
      "entropy": 5.350776815414429,
      "epoch": 0.621998078770413,
      "grad_norm": 1.0,
      "learning_rate": 0.0004968766444489298,
      "loss": 5.1178,
      "mean_token_accuracy": 0.2144631862640381,
      "num_tokens": 14852690.0,
      "step": 6475
    },
    {
      "entropy": 5.3790772438049315,
      "epoch": 0.622478386167147,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004968709493098593,
      "loss": 5.0996,
      "mean_token_accuracy": 0.2103243201971054,
      "num_tokens": 14863327.0,
      "step": 6480
    },
    {
      "entropy": 5.40360050201416,
      "epoch": 0.6229586935638809,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004968652490196041,
      "loss": 5.2792,
      "mean_token_accuracy": 0.20150311589241027,
      "num_tokens": 14875213.0,
      "step": 6485
    },
    {
      "entropy": 5.375523424148559,
      "epoch": 0.6234390009606148,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004968595435782967,
      "loss": 5.1665,
      "mean_token_accuracy": 0.21407082676887512,
      "num_tokens": 14886129.0,
      "step": 6490
    },
    {
      "entropy": 5.438185405731201,
      "epoch": 0.6239193083573487,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004968538329860695,
      "loss": 5.3223,
      "mean_token_accuracy": 0.19463388323783876,
      "num_tokens": 14897217.0,
      "step": 6495
    },
    {
      "entropy": 5.393808460235595,
      "epoch": 0.6243996157540826,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004968481172430549,
      "loss": 5.2077,
      "mean_token_accuracy": 0.20633927583694459,
      "num_tokens": 14908438.0,
      "step": 6500
    },
    {
      "entropy": 5.309189128875732,
      "epoch": 0.6248799231508165,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000496842396349386,
      "loss": 5.108,
      "mean_token_accuracy": 0.21646459102630616,
      "num_tokens": 14920472.0,
      "step": 6505
    },
    {
      "entropy": 5.331929683685303,
      "epoch": 0.6253602305475504,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004968366703051952,
      "loss": 5.1589,
      "mean_token_accuracy": 0.20861062556505203,
      "num_tokens": 14932075.0,
      "step": 6510
    },
    {
      "entropy": 5.352813959121704,
      "epoch": 0.6258405379442843,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004968309391106157,
      "loss": 5.192,
      "mean_token_accuracy": 0.2066570281982422,
      "num_tokens": 14942971.0,
      "step": 6515
    },
    {
      "entropy": 5.33895583152771,
      "epoch": 0.6263208453410183,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004968252027657806,
      "loss": 5.0448,
      "mean_token_accuracy": 0.21391605734825134,
      "num_tokens": 14954288.0,
      "step": 6520
    },
    {
      "entropy": 5.299142169952392,
      "epoch": 0.6268011527377522,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004968194612708229,
      "loss": 5.1024,
      "mean_token_accuracy": 0.20947352200746536,
      "num_tokens": 14966017.0,
      "step": 6525
    },
    {
      "entropy": 5.404318571090698,
      "epoch": 0.627281460134486,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004968137146258759,
      "loss": 5.1899,
      "mean_token_accuracy": 0.20726121813058854,
      "num_tokens": 14978022.0,
      "step": 6530
    },
    {
      "entropy": 5.298225545883179,
      "epoch": 0.6277617675312199,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004968079628310732,
      "loss": 5.0961,
      "mean_token_accuracy": 0.213259819149971,
      "num_tokens": 14990370.0,
      "step": 6535
    },
    {
      "entropy": 5.273942232131958,
      "epoch": 0.6282420749279539,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004968022058865482,
      "loss": 5.1,
      "mean_token_accuracy": 0.21425193250179292,
      "num_tokens": 15001535.0,
      "step": 6540
    },
    {
      "entropy": 5.437747240066528,
      "epoch": 0.6287223823246878,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004967964437924346,
      "loss": 5.1834,
      "mean_token_accuracy": 0.210428886115551,
      "num_tokens": 15012560.0,
      "step": 6545
    },
    {
      "entropy": 5.273789119720459,
      "epoch": 0.6292026897214217,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004967906765488662,
      "loss": 5.0904,
      "mean_token_accuracy": 0.21402514725923538,
      "num_tokens": 15024594.0,
      "step": 6550
    },
    {
      "entropy": 5.3567702293396,
      "epoch": 0.6296829971181557,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004967849041559769,
      "loss": 5.1326,
      "mean_token_accuracy": 0.207964426279068,
      "num_tokens": 15034735.0,
      "step": 6555
    },
    {
      "entropy": 5.427696561813354,
      "epoch": 0.6301633045148896,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004967791266139006,
      "loss": 5.2278,
      "mean_token_accuracy": 0.2034089207649231,
      "num_tokens": 15046955.0,
      "step": 6560
    },
    {
      "entropy": 5.351802921295166,
      "epoch": 0.6306436119116234,
      "grad_norm": 1.0,
      "learning_rate": 0.0004967733439227716,
      "loss": 5.1081,
      "mean_token_accuracy": 0.2113296240568161,
      "num_tokens": 15058250.0,
      "step": 6565
    },
    {
      "entropy": 5.387771797180176,
      "epoch": 0.6311239193083573,
      "grad_norm": 0.921875,
      "learning_rate": 0.000496767556082724,
      "loss": 5.2238,
      "mean_token_accuracy": 0.20387782156467438,
      "num_tokens": 15069451.0,
      "step": 6570
    },
    {
      "entropy": 5.345231580734253,
      "epoch": 0.6316042267050913,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004967617630938924,
      "loss": 5.1825,
      "mean_token_accuracy": 0.21100624650716782,
      "num_tokens": 15080920.0,
      "step": 6575
    },
    {
      "entropy": 5.305248832702636,
      "epoch": 0.6320845341018252,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000496755964956411,
      "loss": 5.1219,
      "mean_token_accuracy": 0.20580895841121674,
      "num_tokens": 15091579.0,
      "step": 6580
    },
    {
      "entropy": 5.3537815570831295,
      "epoch": 0.6325648414985591,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004967501616704147,
      "loss": 5.225,
      "mean_token_accuracy": 0.20618314743041993,
      "num_tokens": 15103097.0,
      "step": 6585
    },
    {
      "entropy": 5.358901882171631,
      "epoch": 0.633045148895293,
      "grad_norm": 0.90625,
      "learning_rate": 0.000496744353236038,
      "loss": 5.1302,
      "mean_token_accuracy": 0.20893874168395996,
      "num_tokens": 15114646.0,
      "step": 6590
    },
    {
      "entropy": 5.435190343856812,
      "epoch": 0.633525456292027,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000496738539653416,
      "loss": 5.1712,
      "mean_token_accuracy": 0.2017595499753952,
      "num_tokens": 15126019.0,
      "step": 6595
    },
    {
      "entropy": 5.388354825973511,
      "epoch": 0.6340057636887608,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004967327209226835,
      "loss": 5.1644,
      "mean_token_accuracy": 0.20799960643053056,
      "num_tokens": 15136132.0,
      "step": 6600
    },
    {
      "entropy": 5.357957077026367,
      "epoch": 0.6344860710854947,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004967268970439755,
      "loss": 5.2352,
      "mean_token_accuracy": 0.20312527120113372,
      "num_tokens": 15149016.0,
      "step": 6605
    },
    {
      "entropy": 5.4254063129425045,
      "epoch": 0.6349663784822286,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004967210680174274,
      "loss": 5.2036,
      "mean_token_accuracy": 0.21304135918617248,
      "num_tokens": 15160302.0,
      "step": 6610
    },
    {
      "entropy": 5.364810657501221,
      "epoch": 0.6354466858789626,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004967152338431745,
      "loss": 5.1851,
      "mean_token_accuracy": 0.205369770526886,
      "num_tokens": 15171770.0,
      "step": 6615
    },
    {
      "entropy": 5.3285229206085205,
      "epoch": 0.6359269932756965,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004967093945213523,
      "loss": 5.0719,
      "mean_token_accuracy": 0.2132847711443901,
      "num_tokens": 15182283.0,
      "step": 6620
    },
    {
      "entropy": 5.329556131362915,
      "epoch": 0.6364073006724303,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004967035500520962,
      "loss": 5.1796,
      "mean_token_accuracy": 0.20531734675168992,
      "num_tokens": 15193917.0,
      "step": 6625
    },
    {
      "entropy": 5.485039520263672,
      "epoch": 0.6368876080691642,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004966977004355421,
      "loss": 5.2488,
      "mean_token_accuracy": 0.20010878443717955,
      "num_tokens": 15205854.0,
      "step": 6630
    },
    {
      "entropy": 5.307362651824951,
      "epoch": 0.6373679154658982,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004966918456718256,
      "loss": 5.116,
      "mean_token_accuracy": 0.2113931208848953,
      "num_tokens": 15217884.0,
      "step": 6635
    },
    {
      "entropy": 5.3968860626220705,
      "epoch": 0.6378482228626321,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004966859857610828,
      "loss": 5.2363,
      "mean_token_accuracy": 0.19845329076051713,
      "num_tokens": 15229493.0,
      "step": 6640
    },
    {
      "entropy": 5.36131911277771,
      "epoch": 0.638328530259366,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004966801207034497,
      "loss": 5.1907,
      "mean_token_accuracy": 0.20187900811433793,
      "num_tokens": 15241402.0,
      "step": 6645
    },
    {
      "entropy": 5.421045923233033,
      "epoch": 0.6388088376560999,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004966742504990624,
      "loss": 5.1971,
      "mean_token_accuracy": 0.2028847262263298,
      "num_tokens": 15252981.0,
      "step": 6650
    },
    {
      "entropy": 5.372963190078735,
      "epoch": 0.6392891450528339,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004966683751480572,
      "loss": 5.1243,
      "mean_token_accuracy": 0.21094227433204651,
      "num_tokens": 15264171.0,
      "step": 6655
    },
    {
      "entropy": 5.313787841796875,
      "epoch": 0.6397694524495677,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004966624946505706,
      "loss": 5.1267,
      "mean_token_accuracy": 0.20988318622112273,
      "num_tokens": 15275595.0,
      "step": 6660
    },
    {
      "entropy": 5.406278848648071,
      "epoch": 0.6402497598463016,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004966566090067391,
      "loss": 5.2067,
      "mean_token_accuracy": 0.20010184794664382,
      "num_tokens": 15286074.0,
      "step": 6665
    },
    {
      "entropy": 5.420538902282715,
      "epoch": 0.6407300672430355,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004966507182166993,
      "loss": 5.1726,
      "mean_token_accuracy": 0.20725206285715103,
      "num_tokens": 15296230.0,
      "step": 6670
    },
    {
      "entropy": 5.338460397720337,
      "epoch": 0.6412103746397695,
      "grad_norm": 0.890625,
      "learning_rate": 0.000496644822280588,
      "loss": 5.1458,
      "mean_token_accuracy": 0.2085949957370758,
      "num_tokens": 15307428.0,
      "step": 6675
    },
    {
      "entropy": 5.454903888702392,
      "epoch": 0.6416906820365034,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000496638921198542,
      "loss": 5.2716,
      "mean_token_accuracy": 0.19481648355722428,
      "num_tokens": 15318796.0,
      "step": 6680
    },
    {
      "entropy": 5.352904796600342,
      "epoch": 0.6421709894332372,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004966330149706984,
      "loss": 5.163,
      "mean_token_accuracy": 0.2108139529824257,
      "num_tokens": 15330914.0,
      "step": 6685
    },
    {
      "entropy": 5.312903547286988,
      "epoch": 0.6426512968299711,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004966271035971944,
      "loss": 5.1306,
      "mean_token_accuracy": 0.2060234397649765,
      "num_tokens": 15341756.0,
      "step": 6690
    },
    {
      "entropy": 5.429139471054077,
      "epoch": 0.6431316042267051,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004966211870781671,
      "loss": 5.2248,
      "mean_token_accuracy": 0.20073795765638353,
      "num_tokens": 15353217.0,
      "step": 6695
    },
    {
      "entropy": 5.409663105010987,
      "epoch": 0.643611911623439,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000496615265413754,
      "loss": 5.1183,
      "mean_token_accuracy": 0.2033244863152504,
      "num_tokens": 15363813.0,
      "step": 6700
    },
    {
      "entropy": 5.33024377822876,
      "epoch": 0.6440922190201729,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004966093386040923,
      "loss": 5.1356,
      "mean_token_accuracy": 0.20895603746175767,
      "num_tokens": 15375223.0,
      "step": 6705
    },
    {
      "entropy": 5.3554637908935545,
      "epoch": 0.6445725264169068,
      "grad_norm": 1.0,
      "learning_rate": 0.00049660340664932,
      "loss": 5.1214,
      "mean_token_accuracy": 0.20896363705396653,
      "num_tokens": 15385523.0,
      "step": 6710
    },
    {
      "entropy": 5.4246556758880615,
      "epoch": 0.6450528338136408,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004965974695495746,
      "loss": 5.1904,
      "mean_token_accuracy": 0.20112065821886063,
      "num_tokens": 15397262.0,
      "step": 6715
    },
    {
      "entropy": 5.298054838180542,
      "epoch": 0.6455331412103746,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004965915273049938,
      "loss": 5.1156,
      "mean_token_accuracy": 0.20933966040611268,
      "num_tokens": 15409043.0,
      "step": 6720
    },
    {
      "entropy": 5.3289378643035885,
      "epoch": 0.6460134486071085,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004965855799157158,
      "loss": 5.11,
      "mean_token_accuracy": 0.2075978308916092,
      "num_tokens": 15420859.0,
      "step": 6725
    },
    {
      "entropy": 5.38963098526001,
      "epoch": 0.6464937560038425,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004965796273818787,
      "loss": 5.2011,
      "mean_token_accuracy": 0.2000960275530815,
      "num_tokens": 15433650.0,
      "step": 6730
    },
    {
      "entropy": 5.307419919967652,
      "epoch": 0.6469740634005764,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004965736697036206,
      "loss": 5.0846,
      "mean_token_accuracy": 0.21013231873512267,
      "num_tokens": 15445158.0,
      "step": 6735
    },
    {
      "entropy": 5.410332536697387,
      "epoch": 0.6474543707973103,
      "grad_norm": 0.875,
      "learning_rate": 0.0004965677068810798,
      "loss": 5.347,
      "mean_token_accuracy": 0.19520421028137208,
      "num_tokens": 15456974.0,
      "step": 6740
    },
    {
      "entropy": 5.460703659057617,
      "epoch": 0.6479346781940442,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004965617389143947,
      "loss": 5.2038,
      "mean_token_accuracy": 0.20826771557331086,
      "num_tokens": 15467395.0,
      "step": 6745
    },
    {
      "entropy": 5.342369890213012,
      "epoch": 0.6484149855907781,
      "grad_norm": 1.4375,
      "learning_rate": 0.000496555765803704,
      "loss": 5.2001,
      "mean_token_accuracy": 0.20690712183713914,
      "num_tokens": 15479396.0,
      "step": 6750
    },
    {
      "entropy": 5.332846927642822,
      "epoch": 0.648895292987512,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004965497875491462,
      "loss": 5.1564,
      "mean_token_accuracy": 0.21249084174633026,
      "num_tokens": 15490992.0,
      "step": 6755
    },
    {
      "entropy": 5.493760395050049,
      "epoch": 0.6493756003842459,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004965438041508604,
      "loss": 5.2238,
      "mean_token_accuracy": 0.1961396560072899,
      "num_tokens": 15502413.0,
      "step": 6760
    },
    {
      "entropy": 5.415083122253418,
      "epoch": 0.6498559077809798,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004965378156089851,
      "loss": 5.1621,
      "mean_token_accuracy": 0.1977938786149025,
      "num_tokens": 15512588.0,
      "step": 6765
    },
    {
      "entropy": 5.344873046875,
      "epoch": 0.6503362151777138,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004965318219236597,
      "loss": 5.1392,
      "mean_token_accuracy": 0.20562929958105086,
      "num_tokens": 15523252.0,
      "step": 6770
    },
    {
      "entropy": 5.317601299285888,
      "epoch": 0.6508165225744477,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004965258230950232,
      "loss": 5.2299,
      "mean_token_accuracy": 0.2040314793586731,
      "num_tokens": 15534975.0,
      "step": 6775
    },
    {
      "entropy": 5.35293869972229,
      "epoch": 0.6512968299711815,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004965198191232148,
      "loss": 5.107,
      "mean_token_accuracy": 0.2115544080734253,
      "num_tokens": 15545709.0,
      "step": 6780
    },
    {
      "entropy": 5.400387811660766,
      "epoch": 0.6517771373679154,
      "grad_norm": 0.953125,
      "learning_rate": 0.000496513810008374,
      "loss": 5.1939,
      "mean_token_accuracy": 0.21018994152545928,
      "num_tokens": 15557270.0,
      "step": 6785
    },
    {
      "entropy": 5.407934236526489,
      "epoch": 0.6522574447646494,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004965077957506403,
      "loss": 5.1593,
      "mean_token_accuracy": 0.20628803521394729,
      "num_tokens": 15569156.0,
      "step": 6790
    },
    {
      "entropy": 5.340046644210815,
      "epoch": 0.6527377521613833,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004965017763501533,
      "loss": 5.1144,
      "mean_token_accuracy": 0.20741754472255708,
      "num_tokens": 15579270.0,
      "step": 6795
    },
    {
      "entropy": 5.257709932327271,
      "epoch": 0.6532180595581172,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004964957518070529,
      "loss": 5.0634,
      "mean_token_accuracy": 0.21847713738679886,
      "num_tokens": 15589912.0,
      "step": 6800
    },
    {
      "entropy": 5.367315435409546,
      "epoch": 0.6536983669548511,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004964897221214788,
      "loss": 5.1686,
      "mean_token_accuracy": 0.20783090144395827,
      "num_tokens": 15601088.0,
      "step": 6805
    },
    {
      "entropy": 5.432088708877563,
      "epoch": 0.654178674351585,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000496483687293571,
      "loss": 5.1712,
      "mean_token_accuracy": 0.20100946724414825,
      "num_tokens": 15612520.0,
      "step": 6810
    },
    {
      "entropy": 5.322639274597168,
      "epoch": 0.6546589817483189,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004964776473234696,
      "loss": 5.096,
      "mean_token_accuracy": 0.21446669548749925,
      "num_tokens": 15623991.0,
      "step": 6815
    },
    {
      "entropy": 5.256708145141602,
      "epoch": 0.6551392891450528,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000496471602211315,
      "loss": 5.0236,
      "mean_token_accuracy": 0.21122566908597945,
      "num_tokens": 15635389.0,
      "step": 6820
    },
    {
      "entropy": 5.375541400909424,
      "epoch": 0.6556195965417867,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004964655519572475,
      "loss": 5.1366,
      "mean_token_accuracy": 0.21136708408594132,
      "num_tokens": 15646427.0,
      "step": 6825
    },
    {
      "entropy": 5.289290475845337,
      "epoch": 0.6560999039385207,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004964594965614072,
      "loss": 5.1356,
      "mean_token_accuracy": 0.20655087977647782,
      "num_tokens": 15657518.0,
      "step": 6830
    },
    {
      "entropy": 5.3904680728912355,
      "epoch": 0.6565802113352546,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004964534360239353,
      "loss": 5.1551,
      "mean_token_accuracy": 0.19926086515188218,
      "num_tokens": 15669775.0,
      "step": 6835
    },
    {
      "entropy": 5.3419993877410885,
      "epoch": 0.6570605187319885,
      "grad_norm": 0.859375,
      "learning_rate": 0.000496447370344972,
      "loss": 5.1549,
      "mean_token_accuracy": 0.19673027694225312,
      "num_tokens": 15682566.0,
      "step": 6840
    },
    {
      "entropy": 5.310968637466431,
      "epoch": 0.6575408261287223,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004964412995246584,
      "loss": 5.1433,
      "mean_token_accuracy": 0.21229192465543748,
      "num_tokens": 15693168.0,
      "step": 6845
    },
    {
      "entropy": 5.342676162719727,
      "epoch": 0.6580211335254563,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004964352235631354,
      "loss": 5.1464,
      "mean_token_accuracy": 0.20767595022916793,
      "num_tokens": 15703879.0,
      "step": 6850
    },
    {
      "entropy": 5.394798994064331,
      "epoch": 0.6585014409221902,
      "grad_norm": 0.9375,
      "learning_rate": 0.000496429142460544,
      "loss": 5.1755,
      "mean_token_accuracy": 0.2011292904615402,
      "num_tokens": 15716015.0,
      "step": 6855
    },
    {
      "entropy": 5.4194292545318605,
      "epoch": 0.6589817483189241,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004964230562170254,
      "loss": 5.196,
      "mean_token_accuracy": 0.20597222745418547,
      "num_tokens": 15728254.0,
      "step": 6860
    },
    {
      "entropy": 5.352735090255737,
      "epoch": 0.659462055715658,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004964169648327209,
      "loss": 5.1521,
      "mean_token_accuracy": 0.20761849135160446,
      "num_tokens": 15738778.0,
      "step": 6865
    },
    {
      "entropy": 5.39863657951355,
      "epoch": 0.659942363112392,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000496410868307772,
      "loss": 5.1609,
      "mean_token_accuracy": 0.20410102009773254,
      "num_tokens": 15750305.0,
      "step": 6870
    },
    {
      "entropy": 5.380156660079956,
      "epoch": 0.6604226705091258,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004964047666423203,
      "loss": 5.1467,
      "mean_token_accuracy": 0.2071267694234848,
      "num_tokens": 15761303.0,
      "step": 6875
    },
    {
      "entropy": 5.327047538757324,
      "epoch": 0.6609029779058597,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004963986598365072,
      "loss": 5.1144,
      "mean_token_accuracy": 0.21683207303285598,
      "num_tokens": 15773095.0,
      "step": 6880
    },
    {
      "entropy": 5.38857135772705,
      "epoch": 0.6613832853025937,
      "grad_norm": 1.0,
      "learning_rate": 0.0004963925478904746,
      "loss": 5.1791,
      "mean_token_accuracy": 0.2092995211482048,
      "num_tokens": 15784405.0,
      "step": 6885
    },
    {
      "entropy": 5.2769053936004635,
      "epoch": 0.6618635926993276,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004963864308043645,
      "loss": 5.1071,
      "mean_token_accuracy": 0.2081150382757187,
      "num_tokens": 15795178.0,
      "step": 6890
    },
    {
      "entropy": 5.344074344635009,
      "epoch": 0.6623439000960615,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004963803085783189,
      "loss": 5.1303,
      "mean_token_accuracy": 0.20570641607046128,
      "num_tokens": 15806205.0,
      "step": 6895
    },
    {
      "entropy": 5.331649971008301,
      "epoch": 0.6628242074927954,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004963741812124799,
      "loss": 5.0506,
      "mean_token_accuracy": 0.21467693150043488,
      "num_tokens": 15817474.0,
      "step": 6900
    },
    {
      "entropy": 5.301028442382813,
      "epoch": 0.6633045148895294,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004963680487069898,
      "loss": 5.1485,
      "mean_token_accuracy": 0.20411243587732314,
      "num_tokens": 15829728.0,
      "step": 6905
    },
    {
      "entropy": 5.446570825576782,
      "epoch": 0.6637848222862632,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004963619110619908,
      "loss": 5.2387,
      "mean_token_accuracy": 0.20543691217899324,
      "num_tokens": 15840082.0,
      "step": 6910
    },
    {
      "entropy": 5.397112512588501,
      "epoch": 0.6642651296829971,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004963557682776256,
      "loss": 5.1318,
      "mean_token_accuracy": 0.20543725341558455,
      "num_tokens": 15851450.0,
      "step": 6915
    },
    {
      "entropy": 5.375704574584961,
      "epoch": 0.664745437079731,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004963496203540368,
      "loss": 5.2728,
      "mean_token_accuracy": 0.19991056323051454,
      "num_tokens": 15864168.0,
      "step": 6920
    },
    {
      "entropy": 5.405896139144898,
      "epoch": 0.665225744476465,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004963434672913671,
      "loss": 5.1945,
      "mean_token_accuracy": 0.20860163271427154,
      "num_tokens": 15875634.0,
      "step": 6925
    },
    {
      "entropy": 5.489336633682251,
      "epoch": 0.6657060518731989,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004963373090897592,
      "loss": 5.2406,
      "mean_token_accuracy": 0.19939687252044677,
      "num_tokens": 15888411.0,
      "step": 6930
    },
    {
      "entropy": 5.367930603027344,
      "epoch": 0.6661863592699327,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004963311457493563,
      "loss": 5.2067,
      "mean_token_accuracy": 0.20343247950077056,
      "num_tokens": 15901084.0,
      "step": 6935
    },
    {
      "entropy": 5.468456649780274,
      "epoch": 0.6666666666666666,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004963249772703015,
      "loss": 5.2239,
      "mean_token_accuracy": 0.20327554941177367,
      "num_tokens": 15912061.0,
      "step": 6940
    },
    {
      "entropy": 5.43146448135376,
      "epoch": 0.6671469740634006,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004963188036527378,
      "loss": 5.2862,
      "mean_token_accuracy": 0.20486033260822295,
      "num_tokens": 15925439.0,
      "step": 6945
    },
    {
      "entropy": 5.288822364807129,
      "epoch": 0.6676272814601345,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004963126248968087,
      "loss": 5.0241,
      "mean_token_accuracy": 0.2155329942703247,
      "num_tokens": 15937762.0,
      "step": 6950
    },
    {
      "entropy": 5.24741907119751,
      "epoch": 0.6681075888568684,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004963064410026577,
      "loss": 5.0452,
      "mean_token_accuracy": 0.21581644266843797,
      "num_tokens": 15948656.0,
      "step": 6955
    },
    {
      "entropy": 5.389021873474121,
      "epoch": 0.6685878962536023,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004963002519704281,
      "loss": 5.1301,
      "mean_token_accuracy": 0.20223325192928315,
      "num_tokens": 15960376.0,
      "step": 6960
    },
    {
      "entropy": 5.403183937072754,
      "epoch": 0.6690682036503363,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000496294057800264,
      "loss": 5.162,
      "mean_token_accuracy": 0.20470741689205169,
      "num_tokens": 15971913.0,
      "step": 6965
    },
    {
      "entropy": 5.385066366195678,
      "epoch": 0.6695485110470701,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004962878584923089,
      "loss": 5.2015,
      "mean_token_accuracy": 0.20076711922883989,
      "num_tokens": 15984775.0,
      "step": 6970
    },
    {
      "entropy": 5.373220205307007,
      "epoch": 0.670028818443804,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004962816540467068,
      "loss": 5.1285,
      "mean_token_accuracy": 0.20235749930143357,
      "num_tokens": 15996717.0,
      "step": 6975
    },
    {
      "entropy": 5.352170085906982,
      "epoch": 0.6705091258405379,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004962754444636017,
      "loss": 5.1253,
      "mean_token_accuracy": 0.21042097061872483,
      "num_tokens": 16007964.0,
      "step": 6980
    },
    {
      "entropy": 5.3342327117919925,
      "epoch": 0.6709894332372719,
      "grad_norm": 0.921875,
      "learning_rate": 0.000496269229743138,
      "loss": 5.1361,
      "mean_token_accuracy": 0.2070094719529152,
      "num_tokens": 16019185.0,
      "step": 6985
    },
    {
      "entropy": 5.382532596588135,
      "epoch": 0.6714697406340058,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004962630098854597,
      "loss": 5.1412,
      "mean_token_accuracy": 0.21081747412681578,
      "num_tokens": 16029983.0,
      "step": 6990
    },
    {
      "entropy": 5.362025403976441,
      "epoch": 0.6719500480307397,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004962567848907113,
      "loss": 5.0801,
      "mean_token_accuracy": 0.20711022466421128,
      "num_tokens": 16040574.0,
      "step": 6995
    },
    {
      "entropy": 5.39690465927124,
      "epoch": 0.6724303554274735,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004962505547590374,
      "loss": 5.2372,
      "mean_token_accuracy": 0.20249408334493638,
      "num_tokens": 16052037.0,
      "step": 7000
    },
    {
      "entropy": 5.408071565628052,
      "epoch": 0.6729106628242075,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004962443194905826,
      "loss": 5.2155,
      "mean_token_accuracy": 0.20055394023656845,
      "num_tokens": 16063413.0,
      "step": 7005
    },
    {
      "entropy": 5.406305170059204,
      "epoch": 0.6733909702209414,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004962380790854916,
      "loss": 5.1046,
      "mean_token_accuracy": 0.20647710859775542,
      "num_tokens": 16074373.0,
      "step": 7010
    },
    {
      "entropy": 5.407031011581421,
      "epoch": 0.6738712776176753,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004962318335439094,
      "loss": 5.2426,
      "mean_token_accuracy": 0.20001827776432038,
      "num_tokens": 16086575.0,
      "step": 7015
    },
    {
      "entropy": 5.355895090103149,
      "epoch": 0.6743515850144092,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004962255828659809,
      "loss": 5.0847,
      "mean_token_accuracy": 0.20987340211868286,
      "num_tokens": 16098529.0,
      "step": 7020
    },
    {
      "entropy": 5.3135683059692385,
      "epoch": 0.6748318924111432,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004962193270518513,
      "loss": 5.1494,
      "mean_token_accuracy": 0.2066950723528862,
      "num_tokens": 16110085.0,
      "step": 7025
    },
    {
      "entropy": 5.3063897609710695,
      "epoch": 0.675312199807877,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004962130661016659,
      "loss": 5.0173,
      "mean_token_accuracy": 0.2101002663373947,
      "num_tokens": 16120249.0,
      "step": 7030
    },
    {
      "entropy": 5.438611888885498,
      "epoch": 0.6757925072046109,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004962068000155699,
      "loss": 5.3008,
      "mean_token_accuracy": 0.1928768277168274,
      "num_tokens": 16132645.0,
      "step": 7035
    },
    {
      "entropy": 5.368090295791626,
      "epoch": 0.6762728146013448,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004962005287937088,
      "loss": 5.1745,
      "mean_token_accuracy": 0.20920299738645554,
      "num_tokens": 16143808.0,
      "step": 7040
    },
    {
      "entropy": 5.340124797821045,
      "epoch": 0.6767531219980788,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004961942524362283,
      "loss": 5.1821,
      "mean_token_accuracy": 0.21055777072906495,
      "num_tokens": 16154309.0,
      "step": 7045
    },
    {
      "entropy": 5.3663732528686525,
      "epoch": 0.6772334293948127,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004961879709432741,
      "loss": 5.129,
      "mean_token_accuracy": 0.20459160655736924,
      "num_tokens": 16164654.0,
      "step": 7050
    },
    {
      "entropy": 5.417678356170654,
      "epoch": 0.6777137367915466,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000496181684314992,
      "loss": 5.1882,
      "mean_token_accuracy": 0.20391224175691605,
      "num_tokens": 16177837.0,
      "step": 7055
    },
    {
      "entropy": 5.296876859664917,
      "epoch": 0.6781940441882806,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004961753925515279,
      "loss": 5.067,
      "mean_token_accuracy": 0.21334256529808043,
      "num_tokens": 16189073.0,
      "step": 7060
    },
    {
      "entropy": 5.4126255989074705,
      "epoch": 0.6786743515850144,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496169095653028,
      "loss": 5.2545,
      "mean_token_accuracy": 0.20514075458049774,
      "num_tokens": 16200371.0,
      "step": 7065
    },
    {
      "entropy": 5.371976518630982,
      "epoch": 0.6791546589817483,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004961627936196384,
      "loss": 5.0329,
      "mean_token_accuracy": 0.21760908663272857,
      "num_tokens": 16210526.0,
      "step": 7070
    },
    {
      "entropy": 5.340885639190674,
      "epoch": 0.6796349663784822,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004961564864515055,
      "loss": 5.152,
      "mean_token_accuracy": 0.20996493846178055,
      "num_tokens": 16221687.0,
      "step": 7075
    },
    {
      "entropy": 5.24901442527771,
      "epoch": 0.6801152737752162,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004961501741487757,
      "loss": 5.0353,
      "mean_token_accuracy": 0.21863269060850143,
      "num_tokens": 16233828.0,
      "step": 7080
    },
    {
      "entropy": 5.31521201133728,
      "epoch": 0.6805955811719501,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004961438567115955,
      "loss": 5.0253,
      "mean_token_accuracy": 0.2200038269162178,
      "num_tokens": 16243900.0,
      "step": 7085
    },
    {
      "entropy": 5.282388162612915,
      "epoch": 0.681075888568684,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004961375341401116,
      "loss": 5.1168,
      "mean_token_accuracy": 0.21111399531364441,
      "num_tokens": 16256347.0,
      "step": 7090
    },
    {
      "entropy": 5.329246616363525,
      "epoch": 0.6815561959654178,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004961312064344708,
      "loss": 5.0302,
      "mean_token_accuracy": 0.2127418264746666,
      "num_tokens": 16267743.0,
      "step": 7095
    },
    {
      "entropy": 5.278157663345337,
      "epoch": 0.6820365033621518,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049612487359482,
      "loss": 5.0589,
      "mean_token_accuracy": 0.21328497529029847,
      "num_tokens": 16278280.0,
      "step": 7100
    },
    {
      "entropy": 5.223898696899414,
      "epoch": 0.6825168107588857,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004961185356213062,
      "loss": 5.0603,
      "mean_token_accuracy": 0.21954626590013504,
      "num_tokens": 16288568.0,
      "step": 7105
    },
    {
      "entropy": 5.3232855796813965,
      "epoch": 0.6829971181556196,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004961121925140767,
      "loss": 5.1444,
      "mean_token_accuracy": 0.21324246376752853,
      "num_tokens": 16300730.0,
      "step": 7110
    },
    {
      "entropy": 5.363188552856445,
      "epoch": 0.6834774255523535,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004961058442732786,
      "loss": 5.1427,
      "mean_token_accuracy": 0.20338241755962372,
      "num_tokens": 16311789.0,
      "step": 7115
    },
    {
      "entropy": 5.308821392059326,
      "epoch": 0.6839577329490875,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004960994908990594,
      "loss": 5.1065,
      "mean_token_accuracy": 0.20517774373292924,
      "num_tokens": 16324439.0,
      "step": 7120
    },
    {
      "entropy": 5.404965448379516,
      "epoch": 0.6844380403458213,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004960931323915665,
      "loss": 5.2544,
      "mean_token_accuracy": 0.21529368460178375,
      "num_tokens": 16335344.0,
      "step": 7125
    },
    {
      "entropy": 5.350891351699829,
      "epoch": 0.6849183477425552,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004960867687509475,
      "loss": 5.0887,
      "mean_token_accuracy": 0.21074139922857285,
      "num_tokens": 16349018.0,
      "step": 7130
    },
    {
      "entropy": 5.351697254180908,
      "epoch": 0.6853986551392891,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004960803999773504,
      "loss": 5.1675,
      "mean_token_accuracy": 0.20665929168462754,
      "num_tokens": 16360137.0,
      "step": 7135
    },
    {
      "entropy": 5.396191358566284,
      "epoch": 0.6858789625360231,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004960740260709228,
      "loss": 5.1727,
      "mean_token_accuracy": 0.20857713222503663,
      "num_tokens": 16372277.0,
      "step": 7140
    },
    {
      "entropy": 5.390155935287476,
      "epoch": 0.686359269932757,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004960676470318128,
      "loss": 5.1691,
      "mean_token_accuracy": 0.20712728649377823,
      "num_tokens": 16383440.0,
      "step": 7145
    },
    {
      "entropy": 5.3785594463348385,
      "epoch": 0.6868395773294909,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004960612628601683,
      "loss": 5.1547,
      "mean_token_accuracy": 0.20781178623437882,
      "num_tokens": 16394330.0,
      "step": 7150
    },
    {
      "entropy": 5.285098600387573,
      "epoch": 0.6873198847262247,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004960548735561379,
      "loss": 5.0762,
      "mean_token_accuracy": 0.21686847060918807,
      "num_tokens": 16405734.0,
      "step": 7155
    },
    {
      "entropy": 5.334474325180054,
      "epoch": 0.6878001921229587,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004960484791198697,
      "loss": 5.092,
      "mean_token_accuracy": 0.2046303242444992,
      "num_tokens": 16416025.0,
      "step": 7160
    },
    {
      "entropy": 5.311826276779175,
      "epoch": 0.6882804995196926,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004960420795515121,
      "loss": 5.1006,
      "mean_token_accuracy": 0.20956576615571976,
      "num_tokens": 16427416.0,
      "step": 7165
    },
    {
      "entropy": 5.333488845825196,
      "epoch": 0.6887608069164265,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004960356748512138,
      "loss": 5.0915,
      "mean_token_accuracy": 0.21433757096529008,
      "num_tokens": 16438073.0,
      "step": 7170
    },
    {
      "entropy": 5.313464593887329,
      "epoch": 0.6892411143131604,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004960292650191236,
      "loss": 5.0614,
      "mean_token_accuracy": 0.2105144217610359,
      "num_tokens": 16449994.0,
      "step": 7175
    },
    {
      "entropy": 5.303148984909058,
      "epoch": 0.6897214217098944,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004960228500553899,
      "loss": 5.0544,
      "mean_token_accuracy": 0.21942657977342606,
      "num_tokens": 16460355.0,
      "step": 7180
    },
    {
      "entropy": 5.391108226776123,
      "epoch": 0.6902017291066282,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004960164299601623,
      "loss": 5.135,
      "mean_token_accuracy": 0.20661765486001968,
      "num_tokens": 16472274.0,
      "step": 7185
    },
    {
      "entropy": 5.407241725921631,
      "epoch": 0.6906820365033621,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004960100047335892,
      "loss": 5.1563,
      "mean_token_accuracy": 0.20781127214431763,
      "num_tokens": 16482959.0,
      "step": 7190
    },
    {
      "entropy": 5.370867395401001,
      "epoch": 0.691162343900096,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004960035743758202,
      "loss": 5.1994,
      "mean_token_accuracy": 0.20494930148124696,
      "num_tokens": 16494533.0,
      "step": 7195
    },
    {
      "entropy": 5.2687249183654785,
      "epoch": 0.69164265129683,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004959971388870044,
      "loss": 5.0578,
      "mean_token_accuracy": 0.21412606686353683,
      "num_tokens": 16506124.0,
      "step": 7200
    },
    {
      "entropy": 5.32615475654602,
      "epoch": 0.6921229586935639,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004959906982672912,
      "loss": 5.0819,
      "mean_token_accuracy": 0.20936821401119232,
      "num_tokens": 16517867.0,
      "step": 7205
    },
    {
      "entropy": 5.248544454574585,
      "epoch": 0.6926032660902978,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004959842525168302,
      "loss": 5.0086,
      "mean_token_accuracy": 0.21626189798116685,
      "num_tokens": 16529075.0,
      "step": 7210
    },
    {
      "entropy": 5.336199426651001,
      "epoch": 0.6930835734870316,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004959778016357712,
      "loss": 5.0686,
      "mean_token_accuracy": 0.2134622648358345,
      "num_tokens": 16540326.0,
      "step": 7215
    },
    {
      "entropy": 5.347323656082153,
      "epoch": 0.6935638808837656,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004959713456242637,
      "loss": 5.1025,
      "mean_token_accuracy": 0.21234775632619857,
      "num_tokens": 16551570.0,
      "step": 7220
    },
    {
      "entropy": 5.4202698230743405,
      "epoch": 0.6940441882804995,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004959648844824576,
      "loss": 5.1504,
      "mean_token_accuracy": 0.2128515362739563,
      "num_tokens": 16562636.0,
      "step": 7225
    },
    {
      "entropy": 5.355170202255249,
      "epoch": 0.6945244956772334,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004959584182105032,
      "loss": 5.0377,
      "mean_token_accuracy": 0.21433595418930054,
      "num_tokens": 16573867.0,
      "step": 7230
    },
    {
      "entropy": 5.254151916503906,
      "epoch": 0.6950048030739674,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004959519468085504,
      "loss": 5.0509,
      "mean_token_accuracy": 0.21470070779323577,
      "num_tokens": 16584533.0,
      "step": 7235
    },
    {
      "entropy": 5.303434753417969,
      "epoch": 0.6954851104707013,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004959454702767494,
      "loss": 5.2485,
      "mean_token_accuracy": 0.21080477982759477,
      "num_tokens": 16596562.0,
      "step": 7240
    },
    {
      "entropy": 5.348945045471192,
      "epoch": 0.6959654178674352,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004959389886152507,
      "loss": 5.0921,
      "mean_token_accuracy": 0.21597741097211837,
      "num_tokens": 16607508.0,
      "step": 7245
    },
    {
      "entropy": 5.36681900024414,
      "epoch": 0.696445725264169,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004959325018242048,
      "loss": 5.1746,
      "mean_token_accuracy": 0.20786135047674179,
      "num_tokens": 16617737.0,
      "step": 7250
    },
    {
      "entropy": 5.389381217956543,
      "epoch": 0.696926032660903,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004959260099037622,
      "loss": 5.1742,
      "mean_token_accuracy": 0.20632247775793075,
      "num_tokens": 16628518.0,
      "step": 7255
    },
    {
      "entropy": 5.329285192489624,
      "epoch": 0.6974063400576369,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004959195128540737,
      "loss": 5.1874,
      "mean_token_accuracy": 0.2045375347137451,
      "num_tokens": 16639644.0,
      "step": 7260
    },
    {
      "entropy": 5.452426290512085,
      "epoch": 0.6978866474543708,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00049591301067529,
      "loss": 5.2328,
      "mean_token_accuracy": 0.19840320199728012,
      "num_tokens": 16650882.0,
      "step": 7265
    },
    {
      "entropy": 5.395485305786133,
      "epoch": 0.6983669548511047,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004959065033675623,
      "loss": 5.1382,
      "mean_token_accuracy": 0.20692969709634781,
      "num_tokens": 16662464.0,
      "step": 7270
    },
    {
      "entropy": 5.353216886520386,
      "epoch": 0.6988472622478387,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004958999909310414,
      "loss": 5.177,
      "mean_token_accuracy": 0.19958139210939407,
      "num_tokens": 16673718.0,
      "step": 7275
    },
    {
      "entropy": 5.310863065719604,
      "epoch": 0.6993275696445725,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004958934733658788,
      "loss": 5.1105,
      "mean_token_accuracy": 0.20269549638032913,
      "num_tokens": 16684957.0,
      "step": 7280
    },
    {
      "entropy": 5.392301034927368,
      "epoch": 0.6998078770413064,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004958869506722256,
      "loss": 5.1609,
      "mean_token_accuracy": 0.20818491131067277,
      "num_tokens": 16695782.0,
      "step": 7285
    },
    {
      "entropy": 5.30381064414978,
      "epoch": 0.7002881844380403,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004958804228502332,
      "loss": 5.0584,
      "mean_token_accuracy": 0.21515030711889266,
      "num_tokens": 16707448.0,
      "step": 7290
    },
    {
      "entropy": 5.25730562210083,
      "epoch": 0.7007684918347743,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004958738899000534,
      "loss": 5.0832,
      "mean_token_accuracy": 0.20368140041828156,
      "num_tokens": 16718074.0,
      "step": 7295
    },
    {
      "entropy": 5.325642538070679,
      "epoch": 0.7012487992315082,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004958673518218377,
      "loss": 5.1229,
      "mean_token_accuracy": 0.2102039560675621,
      "num_tokens": 16728656.0,
      "step": 7300
    },
    {
      "entropy": 5.346181964874267,
      "epoch": 0.7017291066282421,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004958608086157379,
      "loss": 5.1757,
      "mean_token_accuracy": 0.21052978038787842,
      "num_tokens": 16740703.0,
      "step": 7305
    },
    {
      "entropy": 5.34031572341919,
      "epoch": 0.7022094140249759,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000495854260281906,
      "loss": 5.0407,
      "mean_token_accuracy": 0.22064048647880555,
      "num_tokens": 16753320.0,
      "step": 7310
    },
    {
      "entropy": 5.238544130325318,
      "epoch": 0.7026897214217099,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004958477068204941,
      "loss": 5.1031,
      "mean_token_accuracy": 0.21317285746335984,
      "num_tokens": 16764889.0,
      "step": 7315
    },
    {
      "entropy": 5.404721593856811,
      "epoch": 0.7031700288184438,
      "grad_norm": 0.90625,
      "learning_rate": 0.000495841148231654,
      "loss": 5.1417,
      "mean_token_accuracy": 0.21132533699274064,
      "num_tokens": 16776301.0,
      "step": 7320
    },
    {
      "entropy": 5.378520059585571,
      "epoch": 0.7036503362151777,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004958345845155383,
      "loss": 5.1264,
      "mean_token_accuracy": 0.21723831295967103,
      "num_tokens": 16786935.0,
      "step": 7325
    },
    {
      "entropy": 5.313399696350098,
      "epoch": 0.7041306436119116,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004958280156722992,
      "loss": 5.1314,
      "mean_token_accuracy": 0.20934344679117203,
      "num_tokens": 16799335.0,
      "step": 7330
    },
    {
      "entropy": 5.243146371841431,
      "epoch": 0.7046109510086456,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004958214417020894,
      "loss": 5.0713,
      "mean_token_accuracy": 0.2091888189315796,
      "num_tokens": 16811376.0,
      "step": 7335
    },
    {
      "entropy": 5.403880453109741,
      "epoch": 0.7050912584053795,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004958148626050614,
      "loss": 5.1196,
      "mean_token_accuracy": 0.2032211810350418,
      "num_tokens": 16822859.0,
      "step": 7340
    },
    {
      "entropy": 5.329212093353272,
      "epoch": 0.7055715658021133,
      "grad_norm": 1.0,
      "learning_rate": 0.000495808278381368,
      "loss": 5.1134,
      "mean_token_accuracy": 0.2076030045747757,
      "num_tokens": 16833831.0,
      "step": 7345
    },
    {
      "entropy": 5.315768384933472,
      "epoch": 0.7060518731988472,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000495801689031162,
      "loss": 5.1383,
      "mean_token_accuracy": 0.20439638048410416,
      "num_tokens": 16846827.0,
      "step": 7350
    },
    {
      "entropy": 5.369126892089843,
      "epoch": 0.7065321805955812,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004957950945545965,
      "loss": 5.113,
      "mean_token_accuracy": 0.2099258854985237,
      "num_tokens": 16858166.0,
      "step": 7355
    },
    {
      "entropy": 5.3943541049957275,
      "epoch": 0.7070124879923151,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004957884949518246,
      "loss": 5.2375,
      "mean_token_accuracy": 0.1968243569135666,
      "num_tokens": 16870201.0,
      "step": 7360
    },
    {
      "entropy": 5.387752628326416,
      "epoch": 0.707492795389049,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004957818902229992,
      "loss": 5.109,
      "mean_token_accuracy": 0.2092154875397682,
      "num_tokens": 16880891.0,
      "step": 7365
    },
    {
      "entropy": 5.3780457973480225,
      "epoch": 0.7079731027857828,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004957752803682741,
      "loss": 5.1874,
      "mean_token_accuracy": 0.19942378848791123,
      "num_tokens": 16893498.0,
      "step": 7370
    },
    {
      "entropy": 5.40191125869751,
      "epoch": 0.7084534101825168,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004957686653878024,
      "loss": 5.1438,
      "mean_token_accuracy": 0.20957574099302292,
      "num_tokens": 16904959.0,
      "step": 7375
    },
    {
      "entropy": 5.427788352966308,
      "epoch": 0.7089337175792507,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000495762045281738,
      "loss": 5.1783,
      "mean_token_accuracy": 0.20638214647769929,
      "num_tokens": 16916547.0,
      "step": 7380
    },
    {
      "entropy": 5.316069078445435,
      "epoch": 0.7094140249759846,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004957554200502344,
      "loss": 5.1134,
      "mean_token_accuracy": 0.2140609458088875,
      "num_tokens": 16928580.0,
      "step": 7385
    },
    {
      "entropy": 5.35137710571289,
      "epoch": 0.7098943323727186,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004957487896934454,
      "loss": 5.15,
      "mean_token_accuracy": 0.20508789867162705,
      "num_tokens": 16941247.0,
      "step": 7390
    },
    {
      "entropy": 5.423005771636963,
      "epoch": 0.7103746397694525,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000495742154211525,
      "loss": 5.1648,
      "mean_token_accuracy": 0.2094101682305336,
      "num_tokens": 16951119.0,
      "step": 7395
    },
    {
      "entropy": 5.3181304931640625,
      "epoch": 0.7108549471661864,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004957355136046272,
      "loss": 5.0905,
      "mean_token_accuracy": 0.21093301326036454,
      "num_tokens": 16962608.0,
      "step": 7400
    },
    {
      "entropy": 5.309222221374512,
      "epoch": 0.7113352545629202,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004957288678729064,
      "loss": 5.1209,
      "mean_token_accuracy": 0.20930130630731583,
      "num_tokens": 16973291.0,
      "step": 7405
    },
    {
      "entropy": 5.401194095611572,
      "epoch": 0.7118155619596542,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004957222170165166,
      "loss": 5.136,
      "mean_token_accuracy": 0.20743726789951325,
      "num_tokens": 16984895.0,
      "step": 7410
    },
    {
      "entropy": 5.386478519439697,
      "epoch": 0.7122958693563881,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004957155610356124,
      "loss": 5.169,
      "mean_token_accuracy": 0.20269262194633483,
      "num_tokens": 16997109.0,
      "step": 7415
    },
    {
      "entropy": 5.32469916343689,
      "epoch": 0.712776176753122,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004957088999303481,
      "loss": 5.0971,
      "mean_token_accuracy": 0.2049662411212921,
      "num_tokens": 17008640.0,
      "step": 7420
    },
    {
      "entropy": 5.332112789154053,
      "epoch": 0.7132564841498559,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004957022337008787,
      "loss": 5.0653,
      "mean_token_accuracy": 0.21241938471794128,
      "num_tokens": 17018924.0,
      "step": 7425
    },
    {
      "entropy": 5.315089893341065,
      "epoch": 0.7137367915465899,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004956955623473587,
      "loss": 5.0955,
      "mean_token_accuracy": 0.21295803785324097,
      "num_tokens": 17029932.0,
      "step": 7430
    },
    {
      "entropy": 5.348790550231934,
      "epoch": 0.7142170989433237,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000495688885869943,
      "loss": 5.1957,
      "mean_token_accuracy": 0.20275085419416428,
      "num_tokens": 17042164.0,
      "step": 7435
    },
    {
      "entropy": 5.391345739364624,
      "epoch": 0.7146974063400576,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004956822042687868,
      "loss": 5.1731,
      "mean_token_accuracy": 0.1998043403029442,
      "num_tokens": 17054729.0,
      "step": 7440
    },
    {
      "entropy": 5.29535026550293,
      "epoch": 0.7151777137367915,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004956755175440451,
      "loss": 5.039,
      "mean_token_accuracy": 0.2244885817170143,
      "num_tokens": 17066537.0,
      "step": 7445
    },
    {
      "entropy": 5.314731502532959,
      "epoch": 0.7156580211335255,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004956688256958731,
      "loss": 5.0943,
      "mean_token_accuracy": 0.21349663585424422,
      "num_tokens": 17077787.0,
      "step": 7450
    },
    {
      "entropy": 5.293307971954346,
      "epoch": 0.7161383285302594,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004956621287244262,
      "loss": 5.1445,
      "mean_token_accuracy": 0.20553401112556458,
      "num_tokens": 17089555.0,
      "step": 7455
    },
    {
      "entropy": 5.377112102508545,
      "epoch": 0.7166186359269933,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004956554266298599,
      "loss": 5.0951,
      "mean_token_accuracy": 0.21415567249059678,
      "num_tokens": 17099911.0,
      "step": 7460
    },
    {
      "entropy": 5.296390771865845,
      "epoch": 0.7170989433237271,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004956487194123298,
      "loss": 5.0732,
      "mean_token_accuracy": 0.21108074188232423,
      "num_tokens": 17110880.0,
      "step": 7465
    },
    {
      "entropy": 5.301234579086303,
      "epoch": 0.7175792507204611,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004956420070719918,
      "loss": 5.0492,
      "mean_token_accuracy": 0.2123343512415886,
      "num_tokens": 17122272.0,
      "step": 7470
    },
    {
      "entropy": 5.330184984207153,
      "epoch": 0.718059558117195,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004956352896090014,
      "loss": 5.1828,
      "mean_token_accuracy": 0.2031962588429451,
      "num_tokens": 17133995.0,
      "step": 7475
    },
    {
      "entropy": 5.359682512283325,
      "epoch": 0.7185398655139289,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004956285670235147,
      "loss": 5.0621,
      "mean_token_accuracy": 0.2137000471353531,
      "num_tokens": 17145970.0,
      "step": 7480
    },
    {
      "entropy": 5.412320518493653,
      "epoch": 0.7190201729106628,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004956218393156879,
      "loss": 5.2512,
      "mean_token_accuracy": 0.19951253086328508,
      "num_tokens": 17157747.0,
      "step": 7485
    },
    {
      "entropy": 5.321125650405884,
      "epoch": 0.7195004803073968,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004956151064856772,
      "loss": 5.1911,
      "mean_token_accuracy": 0.21149196475744247,
      "num_tokens": 17168357.0,
      "step": 7490
    },
    {
      "entropy": 5.360573768615723,
      "epoch": 0.7199807877041307,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004956083685336386,
      "loss": 5.1275,
      "mean_token_accuracy": 0.21438027173280716,
      "num_tokens": 17179871.0,
      "step": 7495
    },
    {
      "entropy": 5.362541389465332,
      "epoch": 0.7204610951008645,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004956016254597289,
      "loss": 5.0693,
      "mean_token_accuracy": 0.21592396199703218,
      "num_tokens": 17190456.0,
      "step": 7500
    },
    {
      "entropy": 5.316376209259033,
      "epoch": 0.7209414024975984,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004955948772641044,
      "loss": 5.1698,
      "mean_token_accuracy": 0.2022445410490036,
      "num_tokens": 17201623.0,
      "step": 7505
    },
    {
      "entropy": 5.289179086685181,
      "epoch": 0.7214217098943324,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000495588123946922,
      "loss": 5.0701,
      "mean_token_accuracy": 0.2154449552297592,
      "num_tokens": 17212891.0,
      "step": 7510
    },
    {
      "entropy": 5.3309040546417235,
      "epoch": 0.7219020172910663,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004955813655083384,
      "loss": 5.1261,
      "mean_token_accuracy": 0.20933630019426347,
      "num_tokens": 17223983.0,
      "step": 7515
    },
    {
      "entropy": 5.364512252807617,
      "epoch": 0.7223823246878002,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004955746019485104,
      "loss": 5.154,
      "mean_token_accuracy": 0.21106259077787398,
      "num_tokens": 17233701.0,
      "step": 7520
    },
    {
      "entropy": 5.334096384048462,
      "epoch": 0.722862632084534,
      "grad_norm": 0.875,
      "learning_rate": 0.000495567833267595,
      "loss": 5.1542,
      "mean_token_accuracy": 0.20766069740056992,
      "num_tokens": 17245958.0,
      "step": 7525
    },
    {
      "entropy": 5.422809076309204,
      "epoch": 0.723342939481268,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004955610594657496,
      "loss": 5.1555,
      "mean_token_accuracy": 0.19982282370328902,
      "num_tokens": 17257687.0,
      "step": 7530
    },
    {
      "entropy": 5.387662744522094,
      "epoch": 0.7238232468780019,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004955542805431313,
      "loss": 5.2068,
      "mean_token_accuracy": 0.20973837822675706,
      "num_tokens": 17270247.0,
      "step": 7535
    },
    {
      "entropy": 5.3202332019805905,
      "epoch": 0.7243035542747358,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004955474964998976,
      "loss": 5.1216,
      "mean_token_accuracy": 0.21431013494729995,
      "num_tokens": 17281920.0,
      "step": 7540
    },
    {
      "entropy": 5.414140367507935,
      "epoch": 0.7247838616714697,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004955407073362058,
      "loss": 5.2871,
      "mean_token_accuracy": 0.19534719735383987,
      "num_tokens": 17293602.0,
      "step": 7545
    },
    {
      "entropy": 5.420685434341431,
      "epoch": 0.7252641690682037,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004955339130522136,
      "loss": 5.0799,
      "mean_token_accuracy": 0.2121052160859108,
      "num_tokens": 17304484.0,
      "step": 7550
    },
    {
      "entropy": 5.2246194839477536,
      "epoch": 0.7257444764649376,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000495527113648079,
      "loss": 5.0059,
      "mean_token_accuracy": 0.21230011135339738,
      "num_tokens": 17314953.0,
      "step": 7555
    },
    {
      "entropy": 5.308328247070312,
      "epoch": 0.7262247838616714,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004955203091239596,
      "loss": 5.1482,
      "mean_token_accuracy": 0.2041084423661232,
      "num_tokens": 17326813.0,
      "step": 7560
    },
    {
      "entropy": 5.433790779113769,
      "epoch": 0.7267050912584054,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004955134994800134,
      "loss": 5.2177,
      "mean_token_accuracy": 0.1998712569475174,
      "num_tokens": 17337968.0,
      "step": 7565
    },
    {
      "entropy": 5.352131271362305,
      "epoch": 0.7271853986551393,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004955066847163984,
      "loss": 5.1089,
      "mean_token_accuracy": 0.20739677548408508,
      "num_tokens": 17350406.0,
      "step": 7570
    },
    {
      "entropy": 5.270753717422485,
      "epoch": 0.7276657060518732,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004954998648332731,
      "loss": 5.0378,
      "mean_token_accuracy": 0.2157138392329216,
      "num_tokens": 17361888.0,
      "step": 7575
    },
    {
      "entropy": 5.341862487792969,
      "epoch": 0.7281460134486071,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004954930398307956,
      "loss": 5.1435,
      "mean_token_accuracy": 0.20940286815166473,
      "num_tokens": 17374047.0,
      "step": 7580
    },
    {
      "entropy": 5.3741772174835205,
      "epoch": 0.7286263208453411,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004954862097091245,
      "loss": 5.1396,
      "mean_token_accuracy": 0.20097160637378692,
      "num_tokens": 17386175.0,
      "step": 7585
    },
    {
      "entropy": 5.308564138412476,
      "epoch": 0.729106628242075,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004954793744684184,
      "loss": 5.143,
      "mean_token_accuracy": 0.21635485887527467,
      "num_tokens": 17398168.0,
      "step": 7590
    },
    {
      "entropy": 5.309036827087402,
      "epoch": 0.7295869356388088,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004954725341088358,
      "loss": 5.0228,
      "mean_token_accuracy": 0.21216989010572435,
      "num_tokens": 17408825.0,
      "step": 7595
    },
    {
      "entropy": 5.309152460098266,
      "epoch": 0.7300672430355427,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004954656886305356,
      "loss": 5.1221,
      "mean_token_accuracy": 0.21166825592517852,
      "num_tokens": 17419813.0,
      "step": 7600
    },
    {
      "entropy": 5.38578987121582,
      "epoch": 0.7305475504322767,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004954588380336768,
      "loss": 5.1714,
      "mean_token_accuracy": 0.20159507989883424,
      "num_tokens": 17431134.0,
      "step": 7605
    },
    {
      "entropy": 5.311600255966186,
      "epoch": 0.7310278578290106,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004954519823184184,
      "loss": 5.078,
      "mean_token_accuracy": 0.21454625129699706,
      "num_tokens": 17442614.0,
      "step": 7610
    },
    {
      "entropy": 5.34098448753357,
      "epoch": 0.7315081652257445,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004954451214849196,
      "loss": 5.057,
      "mean_token_accuracy": 0.21475362479686738,
      "num_tokens": 17454615.0,
      "step": 7615
    },
    {
      "entropy": 5.360263872146606,
      "epoch": 0.7319884726224783,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004954382555333397,
      "loss": 5.1168,
      "mean_token_accuracy": 0.21009002029895782,
      "num_tokens": 17467025.0,
      "step": 7620
    },
    {
      "entropy": 5.301818799972534,
      "epoch": 0.7324687800192123,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000495431384463838,
      "loss": 5.0963,
      "mean_token_accuracy": 0.21603968143463134,
      "num_tokens": 17477681.0,
      "step": 7625
    },
    {
      "entropy": 5.361507606506348,
      "epoch": 0.7329490874159462,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004954245082765741,
      "loss": 5.1253,
      "mean_token_accuracy": 0.21298671513795853,
      "num_tokens": 17489814.0,
      "step": 7630
    },
    {
      "entropy": 5.368234872817993,
      "epoch": 0.7334293948126801,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004954176269717077,
      "loss": 5.132,
      "mean_token_accuracy": 0.2104403242468834,
      "num_tokens": 17501701.0,
      "step": 7635
    },
    {
      "entropy": 5.3271644115448,
      "epoch": 0.733909702209414,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004954107405493984,
      "loss": 5.0356,
      "mean_token_accuracy": 0.2159121409058571,
      "num_tokens": 17513585.0,
      "step": 7640
    },
    {
      "entropy": 5.358423662185669,
      "epoch": 0.734390009606148,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004954038490098064,
      "loss": 5.1798,
      "mean_token_accuracy": 0.21037847995758058,
      "num_tokens": 17525557.0,
      "step": 7645
    },
    {
      "entropy": 5.293937301635742,
      "epoch": 0.7348703170028819,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004953969523530914,
      "loss": 5.04,
      "mean_token_accuracy": 0.20951283425092698,
      "num_tokens": 17538312.0,
      "step": 7650
    },
    {
      "entropy": 5.427788209915161,
      "epoch": 0.7353506243996157,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004953900505794136,
      "loss": 5.1923,
      "mean_token_accuracy": 0.20821893364191055,
      "num_tokens": 17550248.0,
      "step": 7655
    },
    {
      "entropy": 5.321779489517212,
      "epoch": 0.7358309317963496,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004953831436889332,
      "loss": 5.0401,
      "mean_token_accuracy": 0.21040415316820144,
      "num_tokens": 17560624.0,
      "step": 7660
    },
    {
      "entropy": 5.344732618331909,
      "epoch": 0.7363112391930836,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004953762316818106,
      "loss": 5.0942,
      "mean_token_accuracy": 0.20328582674264908,
      "num_tokens": 17572439.0,
      "step": 7665
    },
    {
      "entropy": 5.398155641555786,
      "epoch": 0.7367915465898175,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004953693145582064,
      "loss": 5.1457,
      "mean_token_accuracy": 0.20126384794712066,
      "num_tokens": 17583120.0,
      "step": 7670
    },
    {
      "entropy": 5.239375448226928,
      "epoch": 0.7372718539865514,
      "grad_norm": 0.96875,
      "learning_rate": 0.000495362392318281,
      "loss": 4.9714,
      "mean_token_accuracy": 0.21710849404335023,
      "num_tokens": 17593677.0,
      "step": 7675
    },
    {
      "entropy": 5.31401481628418,
      "epoch": 0.7377521613832853,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004953554649621951,
      "loss": 5.1353,
      "mean_token_accuracy": 0.2044165164232254,
      "num_tokens": 17605180.0,
      "step": 7680
    },
    {
      "entropy": 5.387196826934814,
      "epoch": 0.7382324687800192,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004953485324901098,
      "loss": 5.0941,
      "mean_token_accuracy": 0.2123894587159157,
      "num_tokens": 17617459.0,
      "step": 7685
    },
    {
      "entropy": 5.339674425125122,
      "epoch": 0.7387127761767531,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004953415949021857,
      "loss": 5.0877,
      "mean_token_accuracy": 0.2147862806916237,
      "num_tokens": 17628024.0,
      "step": 7690
    },
    {
      "entropy": 5.280680418014526,
      "epoch": 0.739193083573487,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004953346521985843,
      "loss": 5.1071,
      "mean_token_accuracy": 0.21122694313526152,
      "num_tokens": 17639833.0,
      "step": 7695
    },
    {
      "entropy": 5.383000087738037,
      "epoch": 0.7396733909702209,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004953277043794663,
      "loss": 5.1907,
      "mean_token_accuracy": 0.19969362765550613,
      "num_tokens": 17651057.0,
      "step": 7700
    },
    {
      "entropy": 5.367031621932983,
      "epoch": 0.7401536983669549,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004953207514449933,
      "loss": 5.0906,
      "mean_token_accuracy": 0.215635746717453,
      "num_tokens": 17662288.0,
      "step": 7705
    },
    {
      "entropy": 5.322967958450318,
      "epoch": 0.7406340057636888,
      "grad_norm": 0.875,
      "learning_rate": 0.0004953137933953267,
      "loss": 5.0931,
      "mean_token_accuracy": 0.2102230355143547,
      "num_tokens": 17673885.0,
      "step": 7710
    },
    {
      "entropy": 5.322059631347656,
      "epoch": 0.7411143131604226,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000495306830230628,
      "loss": 5.1252,
      "mean_token_accuracy": 0.20912605226039888,
      "num_tokens": 17685030.0,
      "step": 7715
    },
    {
      "entropy": 5.315001726150513,
      "epoch": 0.7415946205571565,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004952998619510589,
      "loss": 5.066,
      "mean_token_accuracy": 0.21422436237335205,
      "num_tokens": 17696624.0,
      "step": 7720
    },
    {
      "entropy": 5.3054591655731205,
      "epoch": 0.7420749279538905,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004952928885567811,
      "loss": 5.0518,
      "mean_token_accuracy": 0.20987910330295562,
      "num_tokens": 17707386.0,
      "step": 7725
    },
    {
      "entropy": 5.447640895843506,
      "epoch": 0.7425552353506244,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004952859100479566,
      "loss": 5.3013,
      "mean_token_accuracy": 0.20340465903282165,
      "num_tokens": 17718605.0,
      "step": 7730
    },
    {
      "entropy": 5.337006759643555,
      "epoch": 0.7430355427473583,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004952789264247474,
      "loss": 5.0881,
      "mean_token_accuracy": 0.2116430938243866,
      "num_tokens": 17730275.0,
      "step": 7735
    },
    {
      "entropy": 5.415854597091675,
      "epoch": 0.7435158501440923,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004952719376873156,
      "loss": 5.1833,
      "mean_token_accuracy": 0.204722361266613,
      "num_tokens": 17741390.0,
      "step": 7740
    },
    {
      "entropy": 5.36515965461731,
      "epoch": 0.7439961575408262,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004952649438358234,
      "loss": 5.0831,
      "mean_token_accuracy": 0.21425745487213135,
      "num_tokens": 17752354.0,
      "step": 7745
    },
    {
      "entropy": 5.360581254959106,
      "epoch": 0.74447646493756,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004952579448704334,
      "loss": 5.0705,
      "mean_token_accuracy": 0.21331725269556046,
      "num_tokens": 17762839.0,
      "step": 7750
    },
    {
      "entropy": 5.340327024459839,
      "epoch": 0.7449567723342939,
      "grad_norm": 0.90625,
      "learning_rate": 0.000495250940791308,
      "loss": 5.1718,
      "mean_token_accuracy": 0.2019715890288353,
      "num_tokens": 17775800.0,
      "step": 7755
    },
    {
      "entropy": 5.336835718154907,
      "epoch": 0.7454370797310279,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004952439315986096,
      "loss": 5.0913,
      "mean_token_accuracy": 0.21300239115953445,
      "num_tokens": 17787804.0,
      "step": 7760
    },
    {
      "entropy": 5.420383405685425,
      "epoch": 0.7459173871277618,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004952369172925012,
      "loss": 5.1926,
      "mean_token_accuracy": 0.21128985583782195,
      "num_tokens": 17800291.0,
      "step": 7765
    },
    {
      "entropy": 5.346131610870361,
      "epoch": 0.7463976945244957,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004952298978731454,
      "loss": 5.0623,
      "mean_token_accuracy": 0.2144858330488205,
      "num_tokens": 17811548.0,
      "step": 7770
    },
    {
      "entropy": 5.319376707077026,
      "epoch": 0.7468780019212296,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004952228733407055,
      "loss": 5.0082,
      "mean_token_accuracy": 0.21832434386014937,
      "num_tokens": 17822589.0,
      "step": 7775
    },
    {
      "entropy": 5.246201896667481,
      "epoch": 0.7473583093179635,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004952158436953444,
      "loss": 5.0327,
      "mean_token_accuracy": 0.2151762828230858,
      "num_tokens": 17834203.0,
      "step": 7780
    },
    {
      "entropy": 5.329451227188111,
      "epoch": 0.7478386167146974,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004952088089372252,
      "loss": 5.1195,
      "mean_token_accuracy": 0.2087213635444641,
      "num_tokens": 17846238.0,
      "step": 7785
    },
    {
      "entropy": 5.339178371429443,
      "epoch": 0.7483189241114313,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004952017690665114,
      "loss": 5.0359,
      "mean_token_accuracy": 0.2133630856871605,
      "num_tokens": 17857640.0,
      "step": 7790
    },
    {
      "entropy": 5.300387716293335,
      "epoch": 0.7487992315081652,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004951947240833664,
      "loss": 5.0483,
      "mean_token_accuracy": 0.212492036819458,
      "num_tokens": 17868755.0,
      "step": 7795
    },
    {
      "entropy": 5.285845375061035,
      "epoch": 0.7492795389048992,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004951876739879537,
      "loss": 5.1171,
      "mean_token_accuracy": 0.20655798763036728,
      "num_tokens": 17881078.0,
      "step": 7800
    },
    {
      "entropy": 5.276634550094604,
      "epoch": 0.7497598463016331,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004951806187804371,
      "loss": 5.0482,
      "mean_token_accuracy": 0.21204878985881806,
      "num_tokens": 17893888.0,
      "step": 7805
    },
    {
      "entropy": 5.35845980644226,
      "epoch": 0.7502401536983669,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004951735584609804,
      "loss": 5.1258,
      "mean_token_accuracy": 0.205987349152565,
      "num_tokens": 17904443.0,
      "step": 7810
    },
    {
      "entropy": 5.3006621360778805,
      "epoch": 0.7507204610951008,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004951664930297474,
      "loss": 5.1187,
      "mean_token_accuracy": 0.2139347091317177,
      "num_tokens": 17918090.0,
      "step": 7815
    },
    {
      "entropy": 5.254575872421265,
      "epoch": 0.7512007684918348,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000495159422486902,
      "loss": 5.0504,
      "mean_token_accuracy": 0.21954036206007005,
      "num_tokens": 17929233.0,
      "step": 7820
    },
    {
      "entropy": 5.316081428527832,
      "epoch": 0.7516810758885687,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004951523468326088,
      "loss": 5.0974,
      "mean_token_accuracy": 0.2085426151752472,
      "num_tokens": 17940580.0,
      "step": 7825
    },
    {
      "entropy": 5.377710056304932,
      "epoch": 0.7521613832853026,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004951452660670317,
      "loss": 5.1798,
      "mean_token_accuracy": 0.20402376651763915,
      "num_tokens": 17953993.0,
      "step": 7830
    },
    {
      "entropy": 5.318881797790527,
      "epoch": 0.7526416906820365,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004951381801903352,
      "loss": 5.0197,
      "mean_token_accuracy": 0.21420904248952866,
      "num_tokens": 17966033.0,
      "step": 7835
    },
    {
      "entropy": 5.1989460468292235,
      "epoch": 0.7531219980787704,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004951310892026839,
      "loss": 4.9955,
      "mean_token_accuracy": 0.22273213565349578,
      "num_tokens": 17977943.0,
      "step": 7840
    },
    {
      "entropy": 5.31523871421814,
      "epoch": 0.7536023054755043,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004951239931042424,
      "loss": 5.0643,
      "mean_token_accuracy": 0.2181277170777321,
      "num_tokens": 17990135.0,
      "step": 7845
    },
    {
      "entropy": 5.287261819839477,
      "epoch": 0.7540826128722382,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004951168918951753,
      "loss": 5.1015,
      "mean_token_accuracy": 0.21650127023458482,
      "num_tokens": 18002126.0,
      "step": 7850
    },
    {
      "entropy": 5.2205640316009525,
      "epoch": 0.7545629202689721,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004951097855756476,
      "loss": 4.9926,
      "mean_token_accuracy": 0.21751293540000916,
      "num_tokens": 18013147.0,
      "step": 7855
    },
    {
      "entropy": 5.362655401229858,
      "epoch": 0.7550432276657061,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004951026741458243,
      "loss": 5.107,
      "mean_token_accuracy": 0.21097714751958846,
      "num_tokens": 18025146.0,
      "step": 7860
    },
    {
      "entropy": 5.297344970703125,
      "epoch": 0.75552353506244,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004950955576058705,
      "loss": 5.0441,
      "mean_token_accuracy": 0.20839287489652633,
      "num_tokens": 18036412.0,
      "step": 7865
    },
    {
      "entropy": 5.333431053161621,
      "epoch": 0.7560038424591738,
      "grad_norm": 1.0,
      "learning_rate": 0.0004950884359559513,
      "loss": 5.1184,
      "mean_token_accuracy": 0.21789794266223908,
      "num_tokens": 18048041.0,
      "step": 7870
    },
    {
      "entropy": 5.348143005371094,
      "epoch": 0.7564841498559077,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004950813091962324,
      "loss": 5.1664,
      "mean_token_accuracy": 0.20152033120393753,
      "num_tokens": 18060163.0,
      "step": 7875
    },
    {
      "entropy": 5.341336822509765,
      "epoch": 0.7569644572526417,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004950741773268788,
      "loss": 5.1109,
      "mean_token_accuracy": 0.2069680780172348,
      "num_tokens": 18071628.0,
      "step": 7880
    },
    {
      "entropy": 5.24476900100708,
      "epoch": 0.7574447646493756,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004950670403480562,
      "loss": 5.0129,
      "mean_token_accuracy": 0.2150563642382622,
      "num_tokens": 18082979.0,
      "step": 7885
    },
    {
      "entropy": 5.306754159927368,
      "epoch": 0.7579250720461095,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004950598982599306,
      "loss": 5.0721,
      "mean_token_accuracy": 0.21355952471494674,
      "num_tokens": 18093889.0,
      "step": 7890
    },
    {
      "entropy": 5.303635692596435,
      "epoch": 0.7584053794428435,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004950527510626675,
      "loss": 5.1202,
      "mean_token_accuracy": 0.21013883501291275,
      "num_tokens": 18105559.0,
      "step": 7895
    },
    {
      "entropy": 5.3475329875946045,
      "epoch": 0.7588856868395774,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004950455987564329,
      "loss": 5.1876,
      "mean_token_accuracy": 0.20690033286809922,
      "num_tokens": 18116316.0,
      "step": 7900
    },
    {
      "entropy": 5.367655611038208,
      "epoch": 0.7593659942363112,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004950384413413931,
      "loss": 5.084,
      "mean_token_accuracy": 0.21405460834503173,
      "num_tokens": 18126851.0,
      "step": 7905
    },
    {
      "entropy": 5.409810495376587,
      "epoch": 0.7598463016330451,
      "grad_norm": 1.0,
      "learning_rate": 0.0004950312788177139,
      "loss": 5.1744,
      "mean_token_accuracy": 0.20579453110694884,
      "num_tokens": 18139571.0,
      "step": 7910
    },
    {
      "entropy": 5.401680421829224,
      "epoch": 0.7603266090297791,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004950241111855618,
      "loss": 5.1602,
      "mean_token_accuracy": 0.2073150560259819,
      "num_tokens": 18150680.0,
      "step": 7915
    },
    {
      "entropy": 5.349238586425781,
      "epoch": 0.760806916426513,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004950169384451031,
      "loss": 5.1037,
      "mean_token_accuracy": 0.21686834394931792,
      "num_tokens": 18161911.0,
      "step": 7920
    },
    {
      "entropy": 5.322481346130371,
      "epoch": 0.7612872238232469,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004950097605965045,
      "loss": 5.0981,
      "mean_token_accuracy": 0.21482323259115219,
      "num_tokens": 18172714.0,
      "step": 7925
    },
    {
      "entropy": 5.340502977371216,
      "epoch": 0.7617675312199808,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004950025776399326,
      "loss": 5.0822,
      "mean_token_accuracy": 0.21751550883054732,
      "num_tokens": 18184250.0,
      "step": 7930
    },
    {
      "entropy": 5.416499662399292,
      "epoch": 0.7622478386167147,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000494995389575554,
      "loss": 5.2061,
      "mean_token_accuracy": 0.20439111143350602,
      "num_tokens": 18195299.0,
      "step": 7935
    },
    {
      "entropy": 5.417877340316773,
      "epoch": 0.7627281460134486,
      "grad_norm": 1.0,
      "learning_rate": 0.0004949881964035357,
      "loss": 5.2643,
      "mean_token_accuracy": 0.20470178574323655,
      "num_tokens": 18206863.0,
      "step": 7940
    },
    {
      "entropy": 5.378323221206665,
      "epoch": 0.7632084534101825,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004949809981240448,
      "loss": 5.1785,
      "mean_token_accuracy": 0.2075530543923378,
      "num_tokens": 18219643.0,
      "step": 7945
    },
    {
      "entropy": 5.325662612915039,
      "epoch": 0.7636887608069164,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004949737947372483,
      "loss": 5.0208,
      "mean_token_accuracy": 0.21761458963155747,
      "num_tokens": 18230461.0,
      "step": 7950
    },
    {
      "entropy": 5.356977033615112,
      "epoch": 0.7641690682036504,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004949665862433134,
      "loss": 5.1652,
      "mean_token_accuracy": 0.2047246813774109,
      "num_tokens": 18243768.0,
      "step": 7955
    },
    {
      "entropy": 5.338019752502442,
      "epoch": 0.7646493756003843,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004949593726424077,
      "loss": 5.0716,
      "mean_token_accuracy": 0.21247661113739014,
      "num_tokens": 18255322.0,
      "step": 7960
    },
    {
      "entropy": 5.326325607299805,
      "epoch": 0.7651296829971181,
      "grad_norm": 1.0,
      "learning_rate": 0.0004949521539346985,
      "loss": 5.1115,
      "mean_token_accuracy": 0.21496090292930603,
      "num_tokens": 18267262.0,
      "step": 7965
    },
    {
      "entropy": 5.325637435913086,
      "epoch": 0.765609990393852,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004949449301203533,
      "loss": 5.02,
      "mean_token_accuracy": 0.2124694988131523,
      "num_tokens": 18277332.0,
      "step": 7970
    },
    {
      "entropy": 5.328628253936768,
      "epoch": 0.766090297790586,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004949377011995399,
      "loss": 5.1493,
      "mean_token_accuracy": 0.2142286017537117,
      "num_tokens": 18289547.0,
      "step": 7975
    },
    {
      "entropy": 5.334627962112426,
      "epoch": 0.7665706051873199,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004949304671724263,
      "loss": 5.0492,
      "mean_token_accuracy": 0.21425910890102387,
      "num_tokens": 18302097.0,
      "step": 7980
    },
    {
      "entropy": 5.358500671386719,
      "epoch": 0.7670509125840538,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004949232280391802,
      "loss": 5.1348,
      "mean_token_accuracy": 0.20893050283193587,
      "num_tokens": 18314151.0,
      "step": 7985
    },
    {
      "entropy": 5.300828313827514,
      "epoch": 0.7675312199807877,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004949159837999698,
      "loss": 5.0731,
      "mean_token_accuracy": 0.2079789936542511,
      "num_tokens": 18326085.0,
      "step": 7990
    },
    {
      "entropy": 5.332225370407104,
      "epoch": 0.7680115273775217,
      "grad_norm": 0.875,
      "learning_rate": 0.0004949087344549633,
      "loss": 5.0661,
      "mean_token_accuracy": 0.2095819741487503,
      "num_tokens": 18338158.0,
      "step": 7995
    },
    {
      "entropy": 5.340604019165039,
      "epoch": 0.7684918347742555,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000494901480004329,
      "loss": 5.072,
      "mean_token_accuracy": 0.21303831338882445,
      "num_tokens": 18349418.0,
      "step": 8000
    },
    {
      "entropy": 5.335345029830933,
      "epoch": 0.7689721421709894,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004948942204482351,
      "loss": 5.0913,
      "mean_token_accuracy": 0.21118417233228684,
      "num_tokens": 18361964.0,
      "step": 8005
    },
    {
      "entropy": 5.2611064434051515,
      "epoch": 0.7694524495677233,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004948869557868506,
      "loss": 4.9816,
      "mean_token_accuracy": 0.22380157858133315,
      "num_tokens": 18373108.0,
      "step": 8010
    },
    {
      "entropy": 5.20665922164917,
      "epoch": 0.7699327569644573,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004948796860203439,
      "loss": 5.0039,
      "mean_token_accuracy": 0.21662286221981047,
      "num_tokens": 18385310.0,
      "step": 8015
    },
    {
      "entropy": 5.35377984046936,
      "epoch": 0.7704130643611912,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004948724111488838,
      "loss": 5.1958,
      "mean_token_accuracy": 0.20525239855051042,
      "num_tokens": 18396132.0,
      "step": 8020
    },
    {
      "entropy": 5.476380968093872,
      "epoch": 0.770893371757925,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004948651311726391,
      "loss": 5.3076,
      "mean_token_accuracy": 0.2004573792219162,
      "num_tokens": 18409930.0,
      "step": 8025
    },
    {
      "entropy": 5.424106931686401,
      "epoch": 0.7713736791546589,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004948578460917789,
      "loss": 5.1777,
      "mean_token_accuracy": 0.20446532666683198,
      "num_tokens": 18421204.0,
      "step": 8030
    },
    {
      "entropy": 5.346558856964111,
      "epoch": 0.7718539865513929,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004948505559064725,
      "loss": 5.0687,
      "mean_token_accuracy": 0.2158183068037033,
      "num_tokens": 18433194.0,
      "step": 8035
    },
    {
      "entropy": 5.272424125671387,
      "epoch": 0.7723342939481268,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004948432606168889,
      "loss": 5.0364,
      "mean_token_accuracy": 0.21654528975486756,
      "num_tokens": 18445282.0,
      "step": 8040
    },
    {
      "entropy": 5.305673980712891,
      "epoch": 0.7728146013448607,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004948359602231976,
      "loss": 5.0673,
      "mean_token_accuracy": 0.21816591918468475,
      "num_tokens": 18456264.0,
      "step": 8045
    },
    {
      "entropy": 5.286828231811524,
      "epoch": 0.7732949087415946,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004948286547255681,
      "loss": 5.0004,
      "mean_token_accuracy": 0.21798124611377717,
      "num_tokens": 18467573.0,
      "step": 8050
    },
    {
      "entropy": 5.309575033187866,
      "epoch": 0.7737752161383286,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00049482134412417,
      "loss": 4.9823,
      "mean_token_accuracy": 0.2214351251721382,
      "num_tokens": 18478333.0,
      "step": 8055
    },
    {
      "entropy": 5.313509607315064,
      "epoch": 0.7742555235350624,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000494814028419173,
      "loss": 5.0132,
      "mean_token_accuracy": 0.21748400628566741,
      "num_tokens": 18489106.0,
      "step": 8060
    },
    {
      "entropy": 5.337710571289063,
      "epoch": 0.7747358309317963,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000494806707610747,
      "loss": 5.1422,
      "mean_token_accuracy": 0.20957866162061692,
      "num_tokens": 18500090.0,
      "step": 8065
    },
    {
      "entropy": 5.376635408401489,
      "epoch": 0.7752161383285303,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000494799381699062,
      "loss": 5.1037,
      "mean_token_accuracy": 0.21432682424783706,
      "num_tokens": 18510305.0,
      "step": 8070
    },
    {
      "entropy": 5.354851531982422,
      "epoch": 0.7756964457252642,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004947920506842879,
      "loss": 5.109,
      "mean_token_accuracy": 0.20744638592004777,
      "num_tokens": 18523048.0,
      "step": 8075
    },
    {
      "entropy": 5.252260065078735,
      "epoch": 0.7761767531219981,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004947847145665951,
      "loss": 5.0212,
      "mean_token_accuracy": 0.21565694659948348,
      "num_tokens": 18534145.0,
      "step": 8080
    },
    {
      "entropy": 5.320717811584473,
      "epoch": 0.776657060518732,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004947773733461539,
      "loss": 5.0631,
      "mean_token_accuracy": 0.21510692089796066,
      "num_tokens": 18545045.0,
      "step": 8085
    },
    {
      "entropy": 5.335205030441284,
      "epoch": 0.777137367915466,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004947700270231347,
      "loss": 5.0702,
      "mean_token_accuracy": 0.21586762219667435,
      "num_tokens": 18557531.0,
      "step": 8090
    },
    {
      "entropy": 5.325165128707885,
      "epoch": 0.7776176753121998,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004947626755977079,
      "loss": 5.099,
      "mean_token_accuracy": 0.21161842197179795,
      "num_tokens": 18569127.0,
      "step": 8095
    },
    {
      "entropy": 5.3843793869018555,
      "epoch": 0.7780979827089337,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004947553190700444,
      "loss": 5.1227,
      "mean_token_accuracy": 0.21503574401140213,
      "num_tokens": 18580606.0,
      "step": 8100
    },
    {
      "entropy": 5.438429355621338,
      "epoch": 0.7785782901056676,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000494747957440315,
      "loss": 5.2419,
      "mean_token_accuracy": 0.20055583566427232,
      "num_tokens": 18592330.0,
      "step": 8105
    },
    {
      "entropy": 5.394421195983886,
      "epoch": 0.7790585975024016,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004947405907086905,
      "loss": 5.207,
      "mean_token_accuracy": 0.20635333955287932,
      "num_tokens": 18604721.0,
      "step": 8110
    },
    {
      "entropy": 5.3272958278656,
      "epoch": 0.7795389048991355,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004947332188753419,
      "loss": 5.0818,
      "mean_token_accuracy": 0.2168866366147995,
      "num_tokens": 18616814.0,
      "step": 8115
    },
    {
      "entropy": 5.369074058532715,
      "epoch": 0.7800192122958693,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004947258419404405,
      "loss": 5.0782,
      "mean_token_accuracy": 0.21367238312959672,
      "num_tokens": 18628224.0,
      "step": 8120
    },
    {
      "entropy": 5.341666793823242,
      "epoch": 0.7804995196926032,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004947184599041576,
      "loss": 5.1201,
      "mean_token_accuracy": 0.2109402135014534,
      "num_tokens": 18639777.0,
      "step": 8125
    },
    {
      "entropy": 5.262123155593872,
      "epoch": 0.7809798270893372,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004947110727666644,
      "loss": 5.047,
      "mean_token_accuracy": 0.20775717198848725,
      "num_tokens": 18651044.0,
      "step": 8130
    },
    {
      "entropy": 5.378970146179199,
      "epoch": 0.7814601344860711,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004947036805281325,
      "loss": 5.1182,
      "mean_token_accuracy": 0.21273023933172225,
      "num_tokens": 18663142.0,
      "step": 8135
    },
    {
      "entropy": 5.46126880645752,
      "epoch": 0.781940441882805,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004946962831887336,
      "loss": 5.184,
      "mean_token_accuracy": 0.20995586663484572,
      "num_tokens": 18674079.0,
      "step": 8140
    },
    {
      "entropy": 5.357724142074585,
      "epoch": 0.7824207492795389,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004946888807486393,
      "loss": 5.1638,
      "mean_token_accuracy": 0.20491517782211305,
      "num_tokens": 18685744.0,
      "step": 8145
    },
    {
      "entropy": 5.25427393913269,
      "epoch": 0.7829010566762729,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004946814732080214,
      "loss": 5.0496,
      "mean_token_accuracy": 0.21469540596008302,
      "num_tokens": 18697049.0,
      "step": 8150
    },
    {
      "entropy": 5.287244415283203,
      "epoch": 0.7833813640730067,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004946740605670523,
      "loss": 5.0683,
      "mean_token_accuracy": 0.2153855100274086,
      "num_tokens": 18708765.0,
      "step": 8155
    },
    {
      "entropy": 5.338965272903442,
      "epoch": 0.7838616714697406,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004946666428259037,
      "loss": 5.054,
      "mean_token_accuracy": 0.21777086108922958,
      "num_tokens": 18719819.0,
      "step": 8160
    },
    {
      "entropy": 5.383521318435669,
      "epoch": 0.7843419788664745,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004946592199847478,
      "loss": 5.123,
      "mean_token_accuracy": 0.20879859924316407,
      "num_tokens": 18730668.0,
      "step": 8165
    },
    {
      "entropy": 5.223228931427002,
      "epoch": 0.7848222862632085,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004946517920437571,
      "loss": 4.9226,
      "mean_token_accuracy": 0.21504913419485092,
      "num_tokens": 18741991.0,
      "step": 8170
    },
    {
      "entropy": 5.29312801361084,
      "epoch": 0.7853025936599424,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004946443590031041,
      "loss": 5.1491,
      "mean_token_accuracy": 0.2096063584089279,
      "num_tokens": 18753919.0,
      "step": 8175
    },
    {
      "entropy": 5.399734401702881,
      "epoch": 0.7857829010566763,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004946369208629613,
      "loss": 5.0561,
      "mean_token_accuracy": 0.22159929424524308,
      "num_tokens": 18764342.0,
      "step": 8180
    },
    {
      "entropy": 5.25146107673645,
      "epoch": 0.7862632084534101,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004946294776235013,
      "loss": 5.0712,
      "mean_token_accuracy": 0.20978716462850572,
      "num_tokens": 18776044.0,
      "step": 8185
    },
    {
      "entropy": 5.380789184570313,
      "epoch": 0.7867435158501441,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004946220292848971,
      "loss": 5.1423,
      "mean_token_accuracy": 0.21196742206811905,
      "num_tokens": 18787354.0,
      "step": 8190
    },
    {
      "entropy": 5.347150278091431,
      "epoch": 0.787223823246878,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004946145758473214,
      "loss": 5.0312,
      "mean_token_accuracy": 0.20693592727184296,
      "num_tokens": 18797845.0,
      "step": 8195
    },
    {
      "entropy": 5.370972061157227,
      "epoch": 0.7877041306436119,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004946071173109475,
      "loss": 5.1206,
      "mean_token_accuracy": 0.20758142024278642,
      "num_tokens": 18809253.0,
      "step": 8200
    },
    {
      "entropy": 5.293522930145263,
      "epoch": 0.7881844380403458,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004945996536759484,
      "loss": 4.9741,
      "mean_token_accuracy": 0.22032079249620437,
      "num_tokens": 18819768.0,
      "step": 8205
    },
    {
      "entropy": 5.395445680618286,
      "epoch": 0.7886647454370798,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004945921849424974,
      "loss": 5.2296,
      "mean_token_accuracy": 0.20500316172838212,
      "num_tokens": 18831151.0,
      "step": 8210
    },
    {
      "entropy": 5.31566424369812,
      "epoch": 0.7891450528338136,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004945847111107679,
      "loss": 5.0087,
      "mean_token_accuracy": 0.21720090508460999,
      "num_tokens": 18842133.0,
      "step": 8215
    },
    {
      "entropy": 5.346229505538941,
      "epoch": 0.7896253602305475,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004945772321809334,
      "loss": 5.1648,
      "mean_token_accuracy": 0.2060374140739441,
      "num_tokens": 18853295.0,
      "step": 8220
    },
    {
      "entropy": 5.3284361362457275,
      "epoch": 0.7901056676272814,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004945697481531677,
      "loss": 5.1093,
      "mean_token_accuracy": 0.21173674166202544,
      "num_tokens": 18865802.0,
      "step": 8225
    },
    {
      "entropy": 5.311218690872193,
      "epoch": 0.7905859750240154,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004945622590276443,
      "loss": 5.032,
      "mean_token_accuracy": 0.21996215879917144,
      "num_tokens": 18877693.0,
      "step": 8230
    },
    {
      "entropy": 5.346083736419677,
      "epoch": 0.7910662824207493,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004945547648045373,
      "loss": 5.0943,
      "mean_token_accuracy": 0.2140740066766739,
      "num_tokens": 18888549.0,
      "step": 8235
    },
    {
      "entropy": 5.291837453842163,
      "epoch": 0.7915465898174832,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004945472654840206,
      "loss": 5.0245,
      "mean_token_accuracy": 0.2168283298611641,
      "num_tokens": 18899132.0,
      "step": 8240
    },
    {
      "entropy": 5.2041699409484865,
      "epoch": 0.7920268972142172,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004945397610662683,
      "loss": 5.0389,
      "mean_token_accuracy": 0.20819741487503052,
      "num_tokens": 18911774.0,
      "step": 8245
    },
    {
      "entropy": 5.378843021392822,
      "epoch": 0.792507204610951,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004945322515514547,
      "loss": 5.1127,
      "mean_token_accuracy": 0.21274399608373643,
      "num_tokens": 18922806.0,
      "step": 8250
    },
    {
      "entropy": 5.363301086425781,
      "epoch": 0.7929875120076849,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000494524736939754,
      "loss": 5.0864,
      "mean_token_accuracy": 0.22166195064783095,
      "num_tokens": 18934861.0,
      "step": 8255
    },
    {
      "entropy": 5.319985294342041,
      "epoch": 0.7934678194044188,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004945172172313408,
      "loss": 5.0606,
      "mean_token_accuracy": 0.21769434064626694,
      "num_tokens": 18944269.0,
      "step": 8260
    },
    {
      "entropy": 5.335051441192627,
      "epoch": 0.7939481268011528,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004945096924263896,
      "loss": 5.0943,
      "mean_token_accuracy": 0.21137671172618866,
      "num_tokens": 18957197.0,
      "step": 8265
    },
    {
      "entropy": 5.365131044387818,
      "epoch": 0.7944284341978867,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004945021625250753,
      "loss": 5.1208,
      "mean_token_accuracy": 0.2078988566994667,
      "num_tokens": 18968254.0,
      "step": 8270
    },
    {
      "entropy": 5.2728620052337645,
      "epoch": 0.7949087415946205,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004944946275275724,
      "loss": 5.0079,
      "mean_token_accuracy": 0.22354944944381713,
      "num_tokens": 18979372.0,
      "step": 8275
    },
    {
      "entropy": 5.305412864685058,
      "epoch": 0.7953890489913544,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004944870874340561,
      "loss": 5.0152,
      "mean_token_accuracy": 0.21727642118930818,
      "num_tokens": 18991075.0,
      "step": 8280
    },
    {
      "entropy": 5.239272832870483,
      "epoch": 0.7958693563880884,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004944795422447013,
      "loss": 4.9616,
      "mean_token_accuracy": 0.21712242513895036,
      "num_tokens": 19002324.0,
      "step": 8285
    },
    {
      "entropy": 5.36112904548645,
      "epoch": 0.7963496637848223,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004944719919596835,
      "loss": 5.1357,
      "mean_token_accuracy": 0.2119611993432045,
      "num_tokens": 19014406.0,
      "step": 8290
    },
    {
      "entropy": 5.266506242752075,
      "epoch": 0.7968299711815562,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004944644365791776,
      "loss": 5.0109,
      "mean_token_accuracy": 0.21952706724405288,
      "num_tokens": 19025984.0,
      "step": 8295
    },
    {
      "entropy": 5.356861162185669,
      "epoch": 0.7973102785782901,
      "grad_norm": 0.984375,
      "learning_rate": 0.000494456876103359,
      "loss": 5.1221,
      "mean_token_accuracy": 0.21321585327386855,
      "num_tokens": 19036189.0,
      "step": 8300
    },
    {
      "entropy": 5.349044847488403,
      "epoch": 0.7977905859750241,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004944493105324035,
      "loss": 5.1126,
      "mean_token_accuracy": 0.2092265710234642,
      "num_tokens": 19047587.0,
      "step": 8305
    },
    {
      "entropy": 5.368412446975708,
      "epoch": 0.7982708933717579,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004944417398664866,
      "loss": 5.0865,
      "mean_token_accuracy": 0.2098768413066864,
      "num_tokens": 19058467.0,
      "step": 8310
    },
    {
      "entropy": 5.319743537902832,
      "epoch": 0.7987512007684918,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004944341641057843,
      "loss": 5.1592,
      "mean_token_accuracy": 0.20655251890420914,
      "num_tokens": 19070235.0,
      "step": 8315
    },
    {
      "entropy": 5.373353815078735,
      "epoch": 0.7992315081652257,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004944265832504721,
      "loss": 5.1265,
      "mean_token_accuracy": 0.21016550213098525,
      "num_tokens": 19082005.0,
      "step": 8320
    },
    {
      "entropy": 5.381409645080566,
      "epoch": 0.7997118155619597,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004944189973007262,
      "loss": 5.1286,
      "mean_token_accuracy": 0.20851912647485732,
      "num_tokens": 19092922.0,
      "step": 8325
    },
    {
      "entropy": 5.331025552749634,
      "epoch": 0.8001921229586936,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004944114062567229,
      "loss": 5.1183,
      "mean_token_accuracy": 0.2098314180970192,
      "num_tokens": 19104832.0,
      "step": 8330
    },
    {
      "entropy": 5.350796508789062,
      "epoch": 0.8006724303554275,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004944038101186381,
      "loss": 5.1563,
      "mean_token_accuracy": 0.21115050017833709,
      "num_tokens": 19116261.0,
      "step": 8335
    },
    {
      "entropy": 5.384184217453003,
      "epoch": 0.8011527377521613,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004943962088866483,
      "loss": 5.1275,
      "mean_token_accuracy": 0.2100977197289467,
      "num_tokens": 19127195.0,
      "step": 8340
    },
    {
      "entropy": 5.3867803573608395,
      "epoch": 0.8016330451488953,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004943886025609301,
      "loss": 5.1825,
      "mean_token_accuracy": 0.20455964505672455,
      "num_tokens": 19138164.0,
      "step": 8345
    },
    {
      "entropy": 5.335765314102173,
      "epoch": 0.8021133525456292,
      "grad_norm": 0.96875,
      "learning_rate": 0.00049438099114166,
      "loss": 5.135,
      "mean_token_accuracy": 0.20542529970407486,
      "num_tokens": 19149153.0,
      "step": 8350
    },
    {
      "entropy": 5.36065936088562,
      "epoch": 0.8025936599423631,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004943733746290147,
      "loss": 5.1476,
      "mean_token_accuracy": 0.2054949954152107,
      "num_tokens": 19161023.0,
      "step": 8355
    },
    {
      "entropy": 5.297316360473633,
      "epoch": 0.803073967339097,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000494365753023171,
      "loss": 4.9966,
      "mean_token_accuracy": 0.22194321006536483,
      "num_tokens": 19172466.0,
      "step": 8360
    },
    {
      "entropy": 5.355871915817261,
      "epoch": 0.803554274735831,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004943581263243059,
      "loss": 5.0845,
      "mean_token_accuracy": 0.21780402660369874,
      "num_tokens": 19183684.0,
      "step": 8365
    },
    {
      "entropy": 5.277257871627808,
      "epoch": 0.8040345821325648,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004943504945325965,
      "loss": 4.9927,
      "mean_token_accuracy": 0.221344518661499,
      "num_tokens": 19194479.0,
      "step": 8370
    },
    {
      "entropy": 5.368801832199097,
      "epoch": 0.8045148895292987,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004943428576482198,
      "loss": 5.0784,
      "mean_token_accuracy": 0.20726759284734725,
      "num_tokens": 19206323.0,
      "step": 8375
    },
    {
      "entropy": 5.389929342269897,
      "epoch": 0.8049951969260326,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004943352156713535,
      "loss": 5.1593,
      "mean_token_accuracy": 0.20489852130413055,
      "num_tokens": 19218849.0,
      "step": 8380
    },
    {
      "entropy": 5.280133438110352,
      "epoch": 0.8054755043227666,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004943275686021747,
      "loss": 4.975,
      "mean_token_accuracy": 0.21554317623376845,
      "num_tokens": 19229603.0,
      "step": 8385
    },
    {
      "entropy": 5.354938077926636,
      "epoch": 0.8059558117195005,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000494319916440861,
      "loss": 5.1644,
      "mean_token_accuracy": 0.20746131092309952,
      "num_tokens": 19241318.0,
      "step": 8390
    },
    {
      "entropy": 5.339200258255005,
      "epoch": 0.8064361191162344,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004943122591875901,
      "loss": 5.0495,
      "mean_token_accuracy": 0.21261312961578369,
      "num_tokens": 19252640.0,
      "step": 8395
    },
    {
      "entropy": 5.28208646774292,
      "epoch": 0.8069164265129684,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004943045968425398,
      "loss": 5.0387,
      "mean_token_accuracy": 0.2161882311105728,
      "num_tokens": 19262971.0,
      "step": 8400
    },
    {
      "entropy": 5.335348129272461,
      "epoch": 0.8073967339097022,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004942969294058878,
      "loss": 5.1338,
      "mean_token_accuracy": 0.21078867763280867,
      "num_tokens": 19274426.0,
      "step": 8405
    },
    {
      "entropy": 5.329405164718628,
      "epoch": 0.8078770413064361,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004942892568778125,
      "loss": 5.0889,
      "mean_token_accuracy": 0.2150216817855835,
      "num_tokens": 19286806.0,
      "step": 8410
    },
    {
      "entropy": 5.294596529006958,
      "epoch": 0.80835734870317,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004942815792584917,
      "loss": 5.0652,
      "mean_token_accuracy": 0.21479784101247787,
      "num_tokens": 19297997.0,
      "step": 8415
    },
    {
      "entropy": 5.298877191543579,
      "epoch": 0.808837656099904,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004942738965481038,
      "loss": 5.0407,
      "mean_token_accuracy": 0.2120647981762886,
      "num_tokens": 19309789.0,
      "step": 8420
    },
    {
      "entropy": 5.310087585449219,
      "epoch": 0.8093179634966379,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004942662087468272,
      "loss": 5.0664,
      "mean_token_accuracy": 0.2190921738743782,
      "num_tokens": 19320688.0,
      "step": 8425
    },
    {
      "entropy": 5.300725936889648,
      "epoch": 0.8097982708933718,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004942585158548402,
      "loss": 5.096,
      "mean_token_accuracy": 0.21266603618860244,
      "num_tokens": 19331670.0,
      "step": 8430
    },
    {
      "entropy": 5.353588581085205,
      "epoch": 0.8102785782901056,
      "grad_norm": 1.0,
      "learning_rate": 0.0004942508178723214,
      "loss": 5.0435,
      "mean_token_accuracy": 0.2116592988371849,
      "num_tokens": 19343578.0,
      "step": 8435
    },
    {
      "entropy": 5.35286431312561,
      "epoch": 0.8107588856868396,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004942431147994499,
      "loss": 5.0988,
      "mean_token_accuracy": 0.21038884222507476,
      "num_tokens": 19354875.0,
      "step": 8440
    },
    {
      "entropy": 5.294489812850952,
      "epoch": 0.8112391930835735,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004942354066364042,
      "loss": 5.0434,
      "mean_token_accuracy": 0.21754383891820908,
      "num_tokens": 19366353.0,
      "step": 8445
    },
    {
      "entropy": 5.222540330886841,
      "epoch": 0.8117195004803074,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004942276933833634,
      "loss": 4.9713,
      "mean_token_accuracy": 0.22078879177570343,
      "num_tokens": 19377534.0,
      "step": 8450
    },
    {
      "entropy": 5.311097049713135,
      "epoch": 0.8121998078770413,
      "grad_norm": 1.0,
      "learning_rate": 0.0004942199750405064,
      "loss": 5.1121,
      "mean_token_accuracy": 0.20324969440698623,
      "num_tokens": 19388188.0,
      "step": 8455
    },
    {
      "entropy": 5.2128173828125,
      "epoch": 0.8126801152737753,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004942122516080127,
      "loss": 4.9474,
      "mean_token_accuracy": 0.22659370303153992,
      "num_tokens": 19399910.0,
      "step": 8460
    },
    {
      "entropy": 5.347034454345703,
      "epoch": 0.8131604226705091,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004942045230860614,
      "loss": 5.097,
      "mean_token_accuracy": 0.2169678568840027,
      "num_tokens": 19411715.0,
      "step": 8465
    },
    {
      "entropy": 5.397089576721191,
      "epoch": 0.813640730067243,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004941967894748319,
      "loss": 5.2014,
      "mean_token_accuracy": 0.20122881829738617,
      "num_tokens": 19423275.0,
      "step": 8470
    },
    {
      "entropy": 5.352301979064942,
      "epoch": 0.8141210374639769,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004941890507745039,
      "loss": 5.1198,
      "mean_token_accuracy": 0.2074746236205101,
      "num_tokens": 19436035.0,
      "step": 8475
    },
    {
      "entropy": 5.256072378158569,
      "epoch": 0.8146013448607109,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004941813069852569,
      "loss": 5.0778,
      "mean_token_accuracy": 0.208758844435215,
      "num_tokens": 19447755.0,
      "step": 8480
    },
    {
      "entropy": 5.34722113609314,
      "epoch": 0.8150816522574448,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004941735581072708,
      "loss": 5.0444,
      "mean_token_accuracy": 0.21877842992544175,
      "num_tokens": 19459044.0,
      "step": 8485
    },
    {
      "entropy": 5.319392919540405,
      "epoch": 0.8155619596541787,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004941658041407255,
      "loss": 5.0445,
      "mean_token_accuracy": 0.21560748517513276,
      "num_tokens": 19471486.0,
      "step": 8490
    },
    {
      "entropy": 5.23969407081604,
      "epoch": 0.8160422670509125,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000494158045085801,
      "loss": 4.9668,
      "mean_token_accuracy": 0.22304799407720566,
      "num_tokens": 19482428.0,
      "step": 8495
    },
    {
      "entropy": 5.298281669616699,
      "epoch": 0.8165225744476465,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004941502809426776,
      "loss": 5.1325,
      "mean_token_accuracy": 0.21189234107732774,
      "num_tokens": 19494616.0,
      "step": 8500
    },
    {
      "entropy": 5.322142028808594,
      "epoch": 0.8170028818443804,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004941425117115354,
      "loss": 5.0279,
      "mean_token_accuracy": 0.21516261249780655,
      "num_tokens": 19504953.0,
      "step": 8505
    },
    {
      "entropy": 5.344069623947144,
      "epoch": 0.8174831892411143,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004941347373925547,
      "loss": 5.1183,
      "mean_token_accuracy": 0.2153186947107315,
      "num_tokens": 19517408.0,
      "step": 8510
    },
    {
      "entropy": 5.260396909713745,
      "epoch": 0.8179634966378482,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004941269579859161,
      "loss": 5.0485,
      "mean_token_accuracy": 0.22386470288038254,
      "num_tokens": 19529190.0,
      "step": 8515
    },
    {
      "entropy": 5.295646095275879,
      "epoch": 0.8184438040345822,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004941191734918002,
      "loss": 5.0853,
      "mean_token_accuracy": 0.21131166815757751,
      "num_tokens": 19540278.0,
      "step": 8520
    },
    {
      "entropy": 5.2985352039337155,
      "epoch": 0.818924111431316,
      "grad_norm": 0.96875,
      "learning_rate": 0.000494111383910388,
      "loss": 5.0668,
      "mean_token_accuracy": 0.21477940827608108,
      "num_tokens": 19551224.0,
      "step": 8525
    },
    {
      "entropy": 5.413621044158935,
      "epoch": 0.8194044188280499,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004941035892418597,
      "loss": 5.0978,
      "mean_token_accuracy": 0.21000590771436692,
      "num_tokens": 19564386.0,
      "step": 8530
    },
    {
      "entropy": 5.35599045753479,
      "epoch": 0.8198847262247838,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004940957894863968,
      "loss": 5.0677,
      "mean_token_accuracy": 0.2179810106754303,
      "num_tokens": 19576170.0,
      "step": 8535
    },
    {
      "entropy": 5.303261089324951,
      "epoch": 0.8203650336215178,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004940879846441804,
      "loss": 5.0672,
      "mean_token_accuracy": 0.21418049484491347,
      "num_tokens": 19587220.0,
      "step": 8540
    },
    {
      "entropy": 5.311101293563842,
      "epoch": 0.8208453410182517,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004940801747153914,
      "loss": 5.0612,
      "mean_token_accuracy": 0.2150135889649391,
      "num_tokens": 19598649.0,
      "step": 8545
    },
    {
      "entropy": 5.326148223876953,
      "epoch": 0.8213256484149856,
      "grad_norm": 1.0,
      "learning_rate": 0.0004940723597002113,
      "loss": 5.1095,
      "mean_token_accuracy": 0.21055997759103776,
      "num_tokens": 19610243.0,
      "step": 8550
    },
    {
      "entropy": 5.339675760269165,
      "epoch": 0.8218059558117194,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004940645395988216,
      "loss": 5.1264,
      "mean_token_accuracy": 0.20540111064910888,
      "num_tokens": 19621467.0,
      "step": 8555
    },
    {
      "entropy": 5.43748722076416,
      "epoch": 0.8222862632084534,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004940567144114036,
      "loss": 5.1911,
      "mean_token_accuracy": 0.20521468967199324,
      "num_tokens": 19633367.0,
      "step": 8560
    },
    {
      "entropy": 5.280074691772461,
      "epoch": 0.8227665706051873,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004940488841381393,
      "loss": 5.05,
      "mean_token_accuracy": 0.21650393456220626,
      "num_tokens": 19643144.0,
      "step": 8565
    },
    {
      "entropy": 5.322038555145264,
      "epoch": 0.8232468780019212,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004940410487792103,
      "loss": 5.1168,
      "mean_token_accuracy": 0.21338038444519042,
      "num_tokens": 19654501.0,
      "step": 8570
    },
    {
      "entropy": 5.361709308624268,
      "epoch": 0.8237271853986552,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004940332083347986,
      "loss": 5.0909,
      "mean_token_accuracy": 0.20908380001783372,
      "num_tokens": 19665382.0,
      "step": 8575
    },
    {
      "entropy": 5.306746339797973,
      "epoch": 0.8242074927953891,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004940253628050861,
      "loss": 4.936,
      "mean_token_accuracy": 0.22473569065332413,
      "num_tokens": 19677222.0,
      "step": 8580
    },
    {
      "entropy": 5.251218366622925,
      "epoch": 0.824687800192123,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004940175121902552,
      "loss": 5.0229,
      "mean_token_accuracy": 0.21675127148628234,
      "num_tokens": 19688550.0,
      "step": 8585
    },
    {
      "entropy": 5.372262859344483,
      "epoch": 0.8251681075888568,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000494009656490488,
      "loss": 5.1515,
      "mean_token_accuracy": 0.2072849303483963,
      "num_tokens": 19700125.0,
      "step": 8590
    },
    {
      "entropy": 5.377041482925415,
      "epoch": 0.8256484149855908,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004940017957059668,
      "loss": 5.1397,
      "mean_token_accuracy": 0.20963060855865479,
      "num_tokens": 19711969.0,
      "step": 8595
    },
    {
      "entropy": 5.31815013885498,
      "epoch": 0.8261287223823247,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004939939298368742,
      "loss": 5.0993,
      "mean_token_accuracy": 0.21465936601161956,
      "num_tokens": 19723813.0,
      "step": 8600
    },
    {
      "entropy": 5.344383239746094,
      "epoch": 0.8266090297790586,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004939860588833929,
      "loss": 5.1196,
      "mean_token_accuracy": 0.20407067835330964,
      "num_tokens": 19735539.0,
      "step": 8605
    },
    {
      "entropy": 5.457013177871704,
      "epoch": 0.8270893371757925,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004939781828457055,
      "loss": 5.1854,
      "mean_token_accuracy": 0.207156802713871,
      "num_tokens": 19747106.0,
      "step": 8610
    },
    {
      "entropy": 5.327617025375366,
      "epoch": 0.8275696445725265,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000493970301723995,
      "loss": 5.0977,
      "mean_token_accuracy": 0.22132242172956468,
      "num_tokens": 19760156.0,
      "step": 8615
    },
    {
      "entropy": 5.2592857837677,
      "epoch": 0.8280499519692603,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004939624155184443,
      "loss": 4.9854,
      "mean_token_accuracy": 0.2137436553835869,
      "num_tokens": 19771256.0,
      "step": 8620
    },
    {
      "entropy": 5.317019081115722,
      "epoch": 0.8285302593659942,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004939545242292365,
      "loss": 5.0075,
      "mean_token_accuracy": 0.21734109967947007,
      "num_tokens": 19781332.0,
      "step": 8625
    },
    {
      "entropy": 5.366267538070678,
      "epoch": 0.8290105667627281,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004939466278565547,
      "loss": 5.1698,
      "mean_token_accuracy": 0.2046363517642021,
      "num_tokens": 19793573.0,
      "step": 8630
    },
    {
      "entropy": 5.300094032287598,
      "epoch": 0.8294908741594621,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004939387264005825,
      "loss": 5.0553,
      "mean_token_accuracy": 0.22160843908786773,
      "num_tokens": 19803421.0,
      "step": 8635
    },
    {
      "entropy": 5.274855470657348,
      "epoch": 0.829971181556196,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004939308198615031,
      "loss": 4.9411,
      "mean_token_accuracy": 0.22383784204721452,
      "num_tokens": 19814440.0,
      "step": 8640
    },
    {
      "entropy": 5.291015243530273,
      "epoch": 0.8304514889529299,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004939229082395001,
      "loss": 5.0772,
      "mean_token_accuracy": 0.21296679377555847,
      "num_tokens": 19825721.0,
      "step": 8645
    },
    {
      "entropy": 5.294602918624878,
      "epoch": 0.8309317963496637,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004939149915347573,
      "loss": 5.0091,
      "mean_token_accuracy": 0.22048348635435105,
      "num_tokens": 19837273.0,
      "step": 8650
    },
    {
      "entropy": 5.322047472000122,
      "epoch": 0.8314121037463977,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004939070697474585,
      "loss": 5.0627,
      "mean_token_accuracy": 0.2189880520105362,
      "num_tokens": 19849436.0,
      "step": 8655
    },
    {
      "entropy": 5.262040901184082,
      "epoch": 0.8318924111431316,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004938991428777875,
      "loss": 4.9283,
      "mean_token_accuracy": 0.2296365961432457,
      "num_tokens": 19860222.0,
      "step": 8660
    },
    {
      "entropy": 5.303294992446899,
      "epoch": 0.8323727185398655,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004938912109259284,
      "loss": 5.0134,
      "mean_token_accuracy": 0.22054938077926636,
      "num_tokens": 19870934.0,
      "step": 8665
    },
    {
      "entropy": 5.334577989578247,
      "epoch": 0.8328530259365994,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004938832738920654,
      "loss": 5.0891,
      "mean_token_accuracy": 0.20727307349443436,
      "num_tokens": 19882149.0,
      "step": 8670
    },
    {
      "entropy": 5.352105045318604,
      "epoch": 0.8333333333333334,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004938753317763826,
      "loss": 5.0107,
      "mean_token_accuracy": 0.2173917070031166,
      "num_tokens": 19893276.0,
      "step": 8675
    },
    {
      "entropy": 5.369452667236328,
      "epoch": 0.8338136407300673,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004938673845790646,
      "loss": 5.192,
      "mean_token_accuracy": 0.20366691648960114,
      "num_tokens": 19904723.0,
      "step": 8680
    },
    {
      "entropy": 5.336636304855347,
      "epoch": 0.8342939481268011,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004938594323002957,
      "loss": 5.0841,
      "mean_token_accuracy": 0.21069223880767823,
      "num_tokens": 19915226.0,
      "step": 8685
    },
    {
      "entropy": 5.302987289428711,
      "epoch": 0.834774255523535,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004938514749402609,
      "loss": 5.0932,
      "mean_token_accuracy": 0.2127903327345848,
      "num_tokens": 19927911.0,
      "step": 8690
    },
    {
      "entropy": 5.342841100692749,
      "epoch": 0.835254562920269,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004938435124991447,
      "loss": 5.1459,
      "mean_token_accuracy": 0.21555902510881425,
      "num_tokens": 19940356.0,
      "step": 8695
    },
    {
      "entropy": 5.369471168518066,
      "epoch": 0.8357348703170029,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004938355449771318,
      "loss": 5.1068,
      "mean_token_accuracy": 0.20579011142253875,
      "num_tokens": 19951108.0,
      "step": 8700
    },
    {
      "entropy": 5.26868748664856,
      "epoch": 0.8362151777137368,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004938275723744075,
      "loss": 5.044,
      "mean_token_accuracy": 0.21689395755529403,
      "num_tokens": 19962427.0,
      "step": 8705
    },
    {
      "entropy": 5.396554660797119,
      "epoch": 0.8366954851104706,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004938195946911567,
      "loss": 5.1491,
      "mean_token_accuracy": 0.20994247645139694,
      "num_tokens": 19973476.0,
      "step": 8710
    },
    {
      "entropy": 5.3372321128845215,
      "epoch": 0.8371757925072046,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004938116119275645,
      "loss": 5.0856,
      "mean_token_accuracy": 0.21437105536460876,
      "num_tokens": 19984034.0,
      "step": 8715
    },
    {
      "entropy": 5.349392986297607,
      "epoch": 0.8376560999039385,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004938036240838166,
      "loss": 5.1156,
      "mean_token_accuracy": 0.21108293831348418,
      "num_tokens": 19996035.0,
      "step": 8720
    },
    {
      "entropy": 5.351335144042968,
      "epoch": 0.8381364073006724,
      "grad_norm": 1.0,
      "learning_rate": 0.0004937956311600983,
      "loss": 5.0344,
      "mean_token_accuracy": 0.21916403919458388,
      "num_tokens": 20007259.0,
      "step": 8725
    },
    {
      "entropy": 5.320114088058472,
      "epoch": 0.8386167146974063,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004937876331565951,
      "loss": 5.0514,
      "mean_token_accuracy": 0.2123742774128914,
      "num_tokens": 20018673.0,
      "step": 8730
    },
    {
      "entropy": 5.293909978866577,
      "epoch": 0.8390970220941403,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004937796300734926,
      "loss": 5.0931,
      "mean_token_accuracy": 0.21447703391313552,
      "num_tokens": 20030767.0,
      "step": 8735
    },
    {
      "entropy": 5.374973201751709,
      "epoch": 0.8395773294908742,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004937716219109769,
      "loss": 5.1203,
      "mean_token_accuracy": 0.2089406043291092,
      "num_tokens": 20041268.0,
      "step": 8740
    },
    {
      "entropy": 5.292406320571899,
      "epoch": 0.840057636887608,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004937636086692336,
      "loss": 4.9928,
      "mean_token_accuracy": 0.2205505609512329,
      "num_tokens": 20052285.0,
      "step": 8745
    },
    {
      "entropy": 5.37066388130188,
      "epoch": 0.840537944284342,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000493755590348449,
      "loss": 5.1899,
      "mean_token_accuracy": 0.20929153263568878,
      "num_tokens": 20065113.0,
      "step": 8750
    },
    {
      "entropy": 5.455632162094116,
      "epoch": 0.8410182516810759,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004937475669488091,
      "loss": 5.1563,
      "mean_token_accuracy": 0.21504518389701843,
      "num_tokens": 20076151.0,
      "step": 8755
    },
    {
      "entropy": 5.304359483718872,
      "epoch": 0.8414985590778098,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004937395384705004,
      "loss": 5.0843,
      "mean_token_accuracy": 0.21089454740285873,
      "num_tokens": 20088195.0,
      "step": 8760
    },
    {
      "entropy": 5.246131372451782,
      "epoch": 0.8419788664745437,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004937315049137089,
      "loss": 4.9576,
      "mean_token_accuracy": 0.22410656064748763,
      "num_tokens": 20098576.0,
      "step": 8765
    },
    {
      "entropy": 5.392441177368164,
      "epoch": 0.8424591738712777,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004937234662786216,
      "loss": 5.1599,
      "mean_token_accuracy": 0.20310539305210112,
      "num_tokens": 20110176.0,
      "step": 8770
    },
    {
      "entropy": 5.37788577079773,
      "epoch": 0.8429394812680115,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004937154225654246,
      "loss": 5.0819,
      "mean_token_accuracy": 0.21563199907541275,
      "num_tokens": 20121713.0,
      "step": 8775
    },
    {
      "entropy": 5.409587812423706,
      "epoch": 0.8434197886647454,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004937073737743051,
      "loss": 5.1762,
      "mean_token_accuracy": 0.20503998696804046,
      "num_tokens": 20134126.0,
      "step": 8780
    },
    {
      "entropy": 5.296372509002685,
      "epoch": 0.8439000960614793,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004936993199054499,
      "loss": 5.0648,
      "mean_token_accuracy": 0.21902117878198624,
      "num_tokens": 20144791.0,
      "step": 8785
    },
    {
      "entropy": 5.379805946350098,
      "epoch": 0.8443804034582133,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004936912609590458,
      "loss": 5.2054,
      "mean_token_accuracy": 0.21019366681575774,
      "num_tokens": 20157214.0,
      "step": 8790
    },
    {
      "entropy": 5.450400161743164,
      "epoch": 0.8448607108549472,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00049368319693528,
      "loss": 5.2171,
      "mean_token_accuracy": 0.19938351958990097,
      "num_tokens": 20168952.0,
      "step": 8795
    },
    {
      "entropy": 5.350177001953125,
      "epoch": 0.8453410182516811,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004936751278343397,
      "loss": 5.0845,
      "mean_token_accuracy": 0.21037333756685256,
      "num_tokens": 20181829.0,
      "step": 8800
    },
    {
      "entropy": 5.387225246429443,
      "epoch": 0.845821325648415,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004936670536564123,
      "loss": 5.2402,
      "mean_token_accuracy": 0.19706704914569856,
      "num_tokens": 20193362.0,
      "step": 8805
    },
    {
      "entropy": 5.338515043258667,
      "epoch": 0.8463016330451489,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004936589744016853,
      "loss": 5.1415,
      "mean_token_accuracy": 0.2130685493350029,
      "num_tokens": 20205159.0,
      "step": 8810
    },
    {
      "entropy": 5.326151418685913,
      "epoch": 0.8467819404418828,
      "grad_norm": 0.96875,
      "learning_rate": 0.000493650890070346,
      "loss": 5.0295,
      "mean_token_accuracy": 0.22603174149990082,
      "num_tokens": 20217075.0,
      "step": 8815
    },
    {
      "entropy": 5.335449981689453,
      "epoch": 0.8472622478386167,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004936428006625824,
      "loss": 5.0445,
      "mean_token_accuracy": 0.2198606699705124,
      "num_tokens": 20227901.0,
      "step": 8820
    },
    {
      "entropy": 5.347858715057373,
      "epoch": 0.8477425552353506,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004936347061785823,
      "loss": 5.0476,
      "mean_token_accuracy": 0.21948010176420213,
      "num_tokens": 20239945.0,
      "step": 8825
    },
    {
      "entropy": 5.393839931488037,
      "epoch": 0.8482228626320846,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004936266066185334,
      "loss": 5.1434,
      "mean_token_accuracy": 0.20632718950510026,
      "num_tokens": 20251503.0,
      "step": 8830
    },
    {
      "entropy": 5.29513463973999,
      "epoch": 0.8487031700288185,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004936185019826239,
      "loss": 5.0587,
      "mean_token_accuracy": 0.2160756379365921,
      "num_tokens": 20263044.0,
      "step": 8835
    },
    {
      "entropy": 5.344191026687622,
      "epoch": 0.8491834774255523,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004936103922710419,
      "loss": 5.0376,
      "mean_token_accuracy": 0.2183619663119316,
      "num_tokens": 20273563.0,
      "step": 8840
    },
    {
      "entropy": 5.333452129364014,
      "epoch": 0.8496637848222862,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004936022774839759,
      "loss": 5.0882,
      "mean_token_accuracy": 0.21377312242984772,
      "num_tokens": 20285482.0,
      "step": 8845
    },
    {
      "entropy": 5.443333387374878,
      "epoch": 0.8501440922190202,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004935941576216141,
      "loss": 5.268,
      "mean_token_accuracy": 0.20341867357492446,
      "num_tokens": 20296666.0,
      "step": 8850
    },
    {
      "entropy": 5.22430272102356,
      "epoch": 0.8506243996157541,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000493586032684145,
      "loss": 4.9551,
      "mean_token_accuracy": 0.22342516332864762,
      "num_tokens": 20308978.0,
      "step": 8855
    },
    {
      "entropy": 5.261478471755981,
      "epoch": 0.851104707012488,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004935779026717573,
      "loss": 5.0398,
      "mean_token_accuracy": 0.21408196687698364,
      "num_tokens": 20321488.0,
      "step": 8860
    },
    {
      "entropy": 5.2735799789428714,
      "epoch": 0.8515850144092219,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004935697675846396,
      "loss": 4.8971,
      "mean_token_accuracy": 0.23052898645401002,
      "num_tokens": 20332938.0,
      "step": 8865
    },
    {
      "entropy": 5.272325849533081,
      "epoch": 0.8520653218059558,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004935616274229811,
      "loss": 4.9956,
      "mean_token_accuracy": 0.22203793078660966,
      "num_tokens": 20342986.0,
      "step": 8870
    },
    {
      "entropy": 5.300751733779907,
      "epoch": 0.8525456292026897,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004935534821869705,
      "loss": 5.0555,
      "mean_token_accuracy": 0.2123943343758583,
      "num_tokens": 20355791.0,
      "step": 8875
    },
    {
      "entropy": 5.233129358291626,
      "epoch": 0.8530259365994236,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004935453318767971,
      "loss": 4.9309,
      "mean_token_accuracy": 0.22178226709365845,
      "num_tokens": 20367080.0,
      "step": 8880
    },
    {
      "entropy": 5.330565071105957,
      "epoch": 0.8535062439961575,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00049353717649265,
      "loss": 5.0483,
      "mean_token_accuracy": 0.22484754920005798,
      "num_tokens": 20376697.0,
      "step": 8885
    },
    {
      "entropy": 5.3550821304321286,
      "epoch": 0.8539865513928915,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004935290160347185,
      "loss": 5.0587,
      "mean_token_accuracy": 0.21934617012739183,
      "num_tokens": 20387430.0,
      "step": 8890
    },
    {
      "entropy": 5.3011399269104,
      "epoch": 0.8544668587896254,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004935208505031922,
      "loss": 5.0336,
      "mean_token_accuracy": 0.21876584887504577,
      "num_tokens": 20398488.0,
      "step": 8895
    },
    {
      "entropy": 5.251365756988525,
      "epoch": 0.8549471661863592,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004935126798982606,
      "loss": 5.0694,
      "mean_token_accuracy": 0.21767811328172684,
      "num_tokens": 20410316.0,
      "step": 8900
    },
    {
      "entropy": 5.318483972549439,
      "epoch": 0.8554274735830932,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004935045042201135,
      "loss": 4.9889,
      "mean_token_accuracy": 0.2249806195497513,
      "num_tokens": 20421645.0,
      "step": 8905
    },
    {
      "entropy": 5.3637378215789795,
      "epoch": 0.8559077809798271,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004934963234689407,
      "loss": 5.1071,
      "mean_token_accuracy": 0.2078435719013214,
      "num_tokens": 20433397.0,
      "step": 8910
    },
    {
      "entropy": 5.272715377807617,
      "epoch": 0.856388088376561,
      "grad_norm": 0.96875,
      "learning_rate": 0.000493488137644932,
      "loss": 5.0126,
      "mean_token_accuracy": 0.22031760662794114,
      "num_tokens": 20446218.0,
      "step": 8915
    },
    {
      "entropy": 5.3629358291625975,
      "epoch": 0.8568683957732949,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004934799467482774,
      "loss": 5.1104,
      "mean_token_accuracy": 0.2142086386680603,
      "num_tokens": 20457265.0,
      "step": 8920
    },
    {
      "entropy": 5.325270271301269,
      "epoch": 0.8573487031700289,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004934717507791673,
      "loss": 5.1028,
      "mean_token_accuracy": 0.2102256551384926,
      "num_tokens": 20468748.0,
      "step": 8925
    },
    {
      "entropy": 5.350523042678833,
      "epoch": 0.8578290105667628,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004934635497377919,
      "loss": 5.1024,
      "mean_token_accuracy": 0.21354436427354812,
      "num_tokens": 20481416.0,
      "step": 8930
    },
    {
      "entropy": 5.272500276565552,
      "epoch": 0.8583093179634966,
      "grad_norm": 0.875,
      "learning_rate": 0.0004934553436243415,
      "loss": 5.0154,
      "mean_token_accuracy": 0.2177797794342041,
      "num_tokens": 20493063.0,
      "step": 8935
    },
    {
      "entropy": 5.249647235870361,
      "epoch": 0.8587896253602305,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004934471324390067,
      "loss": 5.0015,
      "mean_token_accuracy": 0.2156543180346489,
      "num_tokens": 20504000.0,
      "step": 8940
    },
    {
      "entropy": 5.33812108039856,
      "epoch": 0.8592699327569645,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004934389161819783,
      "loss": 5.1044,
      "mean_token_accuracy": 0.21617372930049897,
      "num_tokens": 20516419.0,
      "step": 8945
    },
    {
      "entropy": 5.331470489501953,
      "epoch": 0.8597502401536984,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004934306948534467,
      "loss": 5.0546,
      "mean_token_accuracy": 0.2139761984348297,
      "num_tokens": 20527385.0,
      "step": 8950
    },
    {
      "entropy": 5.33019003868103,
      "epoch": 0.8602305475504323,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004934224684536031,
      "loss": 5.0523,
      "mean_token_accuracy": 0.21539317667484284,
      "num_tokens": 20538051.0,
      "step": 8955
    },
    {
      "entropy": 5.3348424434661865,
      "epoch": 0.8607108549471661,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004934142369826382,
      "loss": 5.0694,
      "mean_token_accuracy": 0.2150691881775856,
      "num_tokens": 20550321.0,
      "step": 8960
    },
    {
      "entropy": 5.349039459228516,
      "epoch": 0.8611911623439001,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004934060004407434,
      "loss": 5.0583,
      "mean_token_accuracy": 0.2163606807589531,
      "num_tokens": 20561229.0,
      "step": 8965
    },
    {
      "entropy": 5.298765516281128,
      "epoch": 0.861671469740634,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004933977588281099,
      "loss": 5.0059,
      "mean_token_accuracy": 0.21803612411022186,
      "num_tokens": 20572040.0,
      "step": 8970
    },
    {
      "entropy": 5.263881349563599,
      "epoch": 0.8621517771373679,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004933895121449288,
      "loss": 5.0304,
      "mean_token_accuracy": 0.22013145685195923,
      "num_tokens": 20583110.0,
      "step": 8975
    },
    {
      "entropy": 5.310286951065064,
      "epoch": 0.8626320845341018,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004933812603913917,
      "loss": 5.0316,
      "mean_token_accuracy": 0.2132230743765831,
      "num_tokens": 20593647.0,
      "step": 8980
    },
    {
      "entropy": 5.244890308380127,
      "epoch": 0.8631123919308358,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004933730035676903,
      "loss": 4.9104,
      "mean_token_accuracy": 0.2285411074757576,
      "num_tokens": 20604428.0,
      "step": 8985
    },
    {
      "entropy": 5.208324146270752,
      "epoch": 0.8635926993275697,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004933647416740161,
      "loss": 4.9729,
      "mean_token_accuracy": 0.22819437831640244,
      "num_tokens": 20615811.0,
      "step": 8990
    },
    {
      "entropy": 5.338266658782959,
      "epoch": 0.8640730067243035,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000493356474710561,
      "loss": 5.0611,
      "mean_token_accuracy": 0.21221038401126863,
      "num_tokens": 20627679.0,
      "step": 8995
    },
    {
      "entropy": 5.377738285064697,
      "epoch": 0.8645533141210374,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000493348202677517,
      "loss": 5.0993,
      "mean_token_accuracy": 0.21800316274166107,
      "num_tokens": 20638879.0,
      "step": 9000
    },
    {
      "epoch": 0.8645533141210374,
      "eval_entropy": 5.128290109318213,
      "eval_loss": 5.07960844039917,
      "eval_mean_token_accuracy": 0.22225927816443272,
      "eval_num_tokens": 20638879.0,
      "eval_runtime": 20.6364,
      "eval_samples_per_second": 1590.153,
      "eval_steps_per_second": 198.775,
      "step": 9000
    },
    {
      "entropy": 5.304322862625122,
      "epoch": 0.8650336215177714,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004933399255750761,
      "loss": 5.0933,
      "mean_token_accuracy": 0.2189257949590683,
      "num_tokens": 20649729.0,
      "step": 9005
    },
    {
      "entropy": 5.367015981674195,
      "epoch": 0.8655139289145053,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004933316434034304,
      "loss": 5.1238,
      "mean_token_accuracy": 0.22117715328931808,
      "num_tokens": 20660473.0,
      "step": 9010
    },
    {
      "entropy": 5.373179006576538,
      "epoch": 0.8659942363112392,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004933233561627723,
      "loss": 5.0914,
      "mean_token_accuracy": 0.21476134955883025,
      "num_tokens": 20671776.0,
      "step": 9015
    },
    {
      "entropy": 5.324523973464966,
      "epoch": 0.866474543707973,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004933150638532942,
      "loss": 5.0805,
      "mean_token_accuracy": 0.2174440935254097,
      "num_tokens": 20684147.0,
      "step": 9020
    },
    {
      "entropy": 5.310344314575195,
      "epoch": 0.866954851104707,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004933067664751885,
      "loss": 5.0355,
      "mean_token_accuracy": 0.21691005527973176,
      "num_tokens": 20695248.0,
      "step": 9025
    },
    {
      "entropy": 5.311398458480835,
      "epoch": 0.8674351585014409,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000493298464028648,
      "loss": 5.0673,
      "mean_token_accuracy": 0.2179734766483307,
      "num_tokens": 20707772.0,
      "step": 9030
    },
    {
      "entropy": 5.314677667617798,
      "epoch": 0.8679154658981748,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004932901565138653,
      "loss": 5.0359,
      "mean_token_accuracy": 0.21230170279741287,
      "num_tokens": 20718813.0,
      "step": 9035
    },
    {
      "entropy": 5.255127382278443,
      "epoch": 0.8683957732949087,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004932818439310334,
      "loss": 5.0028,
      "mean_token_accuracy": 0.22453548312187194,
      "num_tokens": 20730939.0,
      "step": 9040
    },
    {
      "entropy": 5.365116882324219,
      "epoch": 0.8688760806916427,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004932735262803452,
      "loss": 5.1114,
      "mean_token_accuracy": 0.20979399234056473,
      "num_tokens": 20742990.0,
      "step": 9045
    },
    {
      "entropy": 5.282762479782105,
      "epoch": 0.8693563880883766,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004932652035619939,
      "loss": 4.9764,
      "mean_token_accuracy": 0.21699778139591216,
      "num_tokens": 20754076.0,
      "step": 9050
    },
    {
      "entropy": 5.3042949676513675,
      "epoch": 0.8698366954851104,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004932568757761727,
      "loss": 5.0184,
      "mean_token_accuracy": 0.2197915717959404,
      "num_tokens": 20765538.0,
      "step": 9055
    },
    {
      "entropy": 5.321621656417847,
      "epoch": 0.8703170028818443,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004932485429230748,
      "loss": 5.0941,
      "mean_token_accuracy": 0.209049491584301,
      "num_tokens": 20776359.0,
      "step": 9060
    },
    {
      "entropy": 5.314424800872803,
      "epoch": 0.8707973102785783,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000493240205002894,
      "loss": 5.0477,
      "mean_token_accuracy": 0.21981824338436126,
      "num_tokens": 20787581.0,
      "step": 9065
    },
    {
      "entropy": 5.36282000541687,
      "epoch": 0.8712776176753122,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004932318620158235,
      "loss": 5.1045,
      "mean_token_accuracy": 0.2082442060112953,
      "num_tokens": 20799904.0,
      "step": 9070
    },
    {
      "entropy": 5.345914220809936,
      "epoch": 0.8717579250720461,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004932235139620574,
      "loss": 5.0299,
      "mean_token_accuracy": 0.21915481686592103,
      "num_tokens": 20810238.0,
      "step": 9075
    },
    {
      "entropy": 5.2967604637146,
      "epoch": 0.8722382324687801,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004932151608417892,
      "loss": 4.9859,
      "mean_token_accuracy": 0.2188950315117836,
      "num_tokens": 20821349.0,
      "step": 9080
    },
    {
      "entropy": 5.240472078323364,
      "epoch": 0.872718539865514,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004932068026552127,
      "loss": 5.0265,
      "mean_token_accuracy": 0.2168602854013443,
      "num_tokens": 20834788.0,
      "step": 9085
    },
    {
      "entropy": 5.3542228698730465,
      "epoch": 0.8731988472622478,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004931984394025224,
      "loss": 5.1033,
      "mean_token_accuracy": 0.21255614310503007,
      "num_tokens": 20845571.0,
      "step": 9090
    },
    {
      "entropy": 5.375140523910522,
      "epoch": 0.8736791546589817,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004931900710839123,
      "loss": 5.0948,
      "mean_token_accuracy": 0.21259163916110993,
      "num_tokens": 20857209.0,
      "step": 9095
    },
    {
      "entropy": 5.360235691070557,
      "epoch": 0.8741594620557157,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004931816976995766,
      "loss": 5.1542,
      "mean_token_accuracy": 0.21156563013792037,
      "num_tokens": 20870624.0,
      "step": 9100
    },
    {
      "entropy": 5.292970657348633,
      "epoch": 0.8746397694524496,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004931733192497097,
      "loss": 5.0477,
      "mean_token_accuracy": 0.22114946991205214,
      "num_tokens": 20881769.0,
      "step": 9105
    },
    {
      "entropy": 5.272651815414429,
      "epoch": 0.8751200768491835,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004931649357345062,
      "loss": 5.0107,
      "mean_token_accuracy": 0.21319330334663392,
      "num_tokens": 20892817.0,
      "step": 9110
    },
    {
      "entropy": 5.320811128616333,
      "epoch": 0.8756003842459174,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004931565471541606,
      "loss": 4.9931,
      "mean_token_accuracy": 0.21794378459453584,
      "num_tokens": 20903042.0,
      "step": 9115
    },
    {
      "entropy": 5.222180557250977,
      "epoch": 0.8760806916426513,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004931481535088679,
      "loss": 4.9553,
      "mean_token_accuracy": 0.2280317261815071,
      "num_tokens": 20914684.0,
      "step": 9120
    },
    {
      "entropy": 5.237375926971436,
      "epoch": 0.8765609990393852,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004931397547988229,
      "loss": 5.0232,
      "mean_token_accuracy": 0.22466280162334443,
      "num_tokens": 20926585.0,
      "step": 9125
    },
    {
      "entropy": 5.372783613204956,
      "epoch": 0.8770413064361191,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004931313510242204,
      "loss": 5.0824,
      "mean_token_accuracy": 0.2110675424337387,
      "num_tokens": 20939729.0,
      "step": 9130
    },
    {
      "entropy": 5.289358282089234,
      "epoch": 0.877521613832853,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004931229421852557,
      "loss": 4.9955,
      "mean_token_accuracy": 0.21599567234516143,
      "num_tokens": 20951697.0,
      "step": 9135
    },
    {
      "entropy": 5.3184737205505375,
      "epoch": 0.878001921229587,
      "grad_norm": 1.0,
      "learning_rate": 0.000493114528282124,
      "loss": 5.1312,
      "mean_token_accuracy": 0.21179965734481812,
      "num_tokens": 20962729.0,
      "step": 9140
    },
    {
      "entropy": 5.26534628868103,
      "epoch": 0.8784822286263209,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004931061093150206,
      "loss": 4.9783,
      "mean_token_accuracy": 0.2254866987466812,
      "num_tokens": 20973331.0,
      "step": 9145
    },
    {
      "entropy": 5.259491205215454,
      "epoch": 0.8789625360230547,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004930976852841409,
      "loss": 5.0015,
      "mean_token_accuracy": 0.2175792545080185,
      "num_tokens": 20985609.0,
      "step": 9150
    },
    {
      "entropy": 5.286224222183227,
      "epoch": 0.8794428434197886,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004930892561896806,
      "loss": 5.0189,
      "mean_token_accuracy": 0.21264548152685164,
      "num_tokens": 20997231.0,
      "step": 9155
    },
    {
      "entropy": 5.3339275360107425,
      "epoch": 0.8799231508165226,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004930808220318354,
      "loss": 5.0288,
      "mean_token_accuracy": 0.22238517850637435,
      "num_tokens": 21008511.0,
      "step": 9160
    },
    {
      "entropy": 5.271183824539184,
      "epoch": 0.8804034582132565,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004930723828108012,
      "loss": 4.9773,
      "mean_token_accuracy": 0.21766621321439744,
      "num_tokens": 21019108.0,
      "step": 9165
    },
    {
      "entropy": 5.279416465759278,
      "epoch": 0.8808837656099904,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004930639385267736,
      "loss": 5.0311,
      "mean_token_accuracy": 0.2213197886943817,
      "num_tokens": 21030621.0,
      "step": 9170
    },
    {
      "entropy": 5.310615968704224,
      "epoch": 0.8813640730067243,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000493055489179949,
      "loss": 4.996,
      "mean_token_accuracy": 0.22043120115995407,
      "num_tokens": 21041778.0,
      "step": 9175
    },
    {
      "entropy": 5.296990442276001,
      "epoch": 0.8818443804034583,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004930470347705234,
      "loss": 5.0172,
      "mean_token_accuracy": 0.22189473062753678,
      "num_tokens": 21054257.0,
      "step": 9180
    },
    {
      "entropy": 5.336837434768677,
      "epoch": 0.8823246878001921,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000493038575298693,
      "loss": 5.0694,
      "mean_token_accuracy": 0.2096772536635399,
      "num_tokens": 21066378.0,
      "step": 9185
    },
    {
      "entropy": 5.232914733886719,
      "epoch": 0.882804995196926,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004930301107646545,
      "loss": 5.0089,
      "mean_token_accuracy": 0.21363568156957627,
      "num_tokens": 21078913.0,
      "step": 9190
    },
    {
      "entropy": 5.328195667266845,
      "epoch": 0.8832853025936599,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004930216411686042,
      "loss": 5.055,
      "mean_token_accuracy": 0.2074598103761673,
      "num_tokens": 21090500.0,
      "step": 9195
    },
    {
      "entropy": 5.322371149063111,
      "epoch": 0.8837656099903939,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004930131665107387,
      "loss": 5.0165,
      "mean_token_accuracy": 0.21313245594501495,
      "num_tokens": 21102793.0,
      "step": 9200
    },
    {
      "entropy": 5.395883417129516,
      "epoch": 0.8842459173871278,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000493004686791255,
      "loss": 5.1268,
      "mean_token_accuracy": 0.20592979043722154,
      "num_tokens": 21114504.0,
      "step": 9205
    },
    {
      "entropy": 5.281560707092285,
      "epoch": 0.8847262247838616,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004929962020103496,
      "loss": 4.9981,
      "mean_token_accuracy": 0.2196553975343704,
      "num_tokens": 21126733.0,
      "step": 9210
    },
    {
      "entropy": 5.245408201217652,
      "epoch": 0.8852065321805955,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004929877121682198,
      "loss": 5.0043,
      "mean_token_accuracy": 0.2142224684357643,
      "num_tokens": 21138045.0,
      "step": 9215
    },
    {
      "entropy": 5.264658212661743,
      "epoch": 0.8856868395773295,
      "grad_norm": 0.875,
      "learning_rate": 0.0004929792172650627,
      "loss": 5.0701,
      "mean_token_accuracy": 0.2094641298055649,
      "num_tokens": 21151562.0,
      "step": 9220
    },
    {
      "entropy": 5.312583160400391,
      "epoch": 0.8861671469740634,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004929707173010753,
      "loss": 5.0355,
      "mean_token_accuracy": 0.2133541002869606,
      "num_tokens": 21162943.0,
      "step": 9225
    },
    {
      "entropy": 5.284322690963745,
      "epoch": 0.8866474543707973,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004929622122764552,
      "loss": 5.067,
      "mean_token_accuracy": 0.2122477889060974,
      "num_tokens": 21174392.0,
      "step": 9230
    },
    {
      "entropy": 5.2570521354675295,
      "epoch": 0.8871277617675313,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004929537021913997,
      "loss": 4.9079,
      "mean_token_accuracy": 0.2274593859910965,
      "num_tokens": 21185372.0,
      "step": 9235
    },
    {
      "entropy": 5.310981607437133,
      "epoch": 0.8876080691642652,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004929451870461064,
      "loss": 5.0608,
      "mean_token_accuracy": 0.2103777438402176,
      "num_tokens": 21197044.0,
      "step": 9240
    },
    {
      "entropy": 5.256484508514404,
      "epoch": 0.888088376560999,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004929366668407731,
      "loss": 5.0014,
      "mean_token_accuracy": 0.22313978374004365,
      "num_tokens": 21207729.0,
      "step": 9245
    },
    {
      "entropy": 5.341896963119507,
      "epoch": 0.8885686839577329,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004929281415755974,
      "loss": 5.0381,
      "mean_token_accuracy": 0.21489929258823395,
      "num_tokens": 21218909.0,
      "step": 9250
    },
    {
      "entropy": 5.373255968093872,
      "epoch": 0.8890489913544669,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004929196112507775,
      "loss": 5.1288,
      "mean_token_accuracy": 0.2102525308728218,
      "num_tokens": 21230543.0,
      "step": 9255
    },
    {
      "entropy": 5.319366931915283,
      "epoch": 0.8895292987512008,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004929110758665112,
      "loss": 5.0718,
      "mean_token_accuracy": 0.21341250389814376,
      "num_tokens": 21242064.0,
      "step": 9260
    },
    {
      "entropy": 5.3429302215576175,
      "epoch": 0.8900096061479347,
      "grad_norm": 0.875,
      "learning_rate": 0.0004929025354229969,
      "loss": 5.0959,
      "mean_token_accuracy": 0.2134937137365341,
      "num_tokens": 21254321.0,
      "step": 9265
    },
    {
      "entropy": 5.200224161148071,
      "epoch": 0.8904899135446686,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004928939899204326,
      "loss": 4.9228,
      "mean_token_accuracy": 0.2240544319152832,
      "num_tokens": 21264741.0,
      "step": 9270
    },
    {
      "entropy": 5.305225276947022,
      "epoch": 0.8909702209414025,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000492885439359017,
      "loss": 5.0321,
      "mean_token_accuracy": 0.21736638247966766,
      "num_tokens": 21276834.0,
      "step": 9275
    },
    {
      "entropy": 5.265902757644653,
      "epoch": 0.8914505283381364,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004928768837389485,
      "loss": 5.0068,
      "mean_token_accuracy": 0.22131659984588622,
      "num_tokens": 21287108.0,
      "step": 9280
    },
    {
      "entropy": 5.261452007293701,
      "epoch": 0.8919308357348703,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004928683230604257,
      "loss": 5.0248,
      "mean_token_accuracy": 0.21500168442726136,
      "num_tokens": 21299942.0,
      "step": 9285
    },
    {
      "entropy": 5.415303564071655,
      "epoch": 0.8924111431316042,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004928597573236474,
      "loss": 5.1665,
      "mean_token_accuracy": 0.21139634251594544,
      "num_tokens": 21311243.0,
      "step": 9290
    },
    {
      "entropy": 5.321933794021606,
      "epoch": 0.8928914505283382,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004928511865288123,
      "loss": 4.9879,
      "mean_token_accuracy": 0.2260330691933632,
      "num_tokens": 21322291.0,
      "step": 9295
    },
    {
      "entropy": 5.294135284423828,
      "epoch": 0.8933717579250721,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004928426106761197,
      "loss": 5.1031,
      "mean_token_accuracy": 0.21394744217395784,
      "num_tokens": 21333257.0,
      "step": 9300
    },
    {
      "entropy": 5.2416738986969,
      "epoch": 0.8938520653218059,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004928340297657685,
      "loss": 5.0005,
      "mean_token_accuracy": 0.2148350179195404,
      "num_tokens": 21345848.0,
      "step": 9305
    },
    {
      "entropy": 5.326335811614991,
      "epoch": 0.8943323727185398,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004928254437979578,
      "loss": 5.031,
      "mean_token_accuracy": 0.21932677626609803,
      "num_tokens": 21357693.0,
      "step": 9310
    },
    {
      "entropy": 5.2927850723266605,
      "epoch": 0.8948126801152738,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004928168527728873,
      "loss": 5.1364,
      "mean_token_accuracy": 0.20861892253160477,
      "num_tokens": 21369653.0,
      "step": 9315
    },
    {
      "entropy": 5.388954401016235,
      "epoch": 0.8952929875120077,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004928082566907562,
      "loss": 5.0755,
      "mean_token_accuracy": 0.21415806710720062,
      "num_tokens": 21383924.0,
      "step": 9320
    },
    {
      "entropy": 5.34713978767395,
      "epoch": 0.8957732949087416,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004927996555517642,
      "loss": 5.0453,
      "mean_token_accuracy": 0.21313219368457795,
      "num_tokens": 21395963.0,
      "step": 9325
    },
    {
      "entropy": 5.303074550628662,
      "epoch": 0.8962536023054755,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004927910493561109,
      "loss": 5.0407,
      "mean_token_accuracy": 0.2162101998925209,
      "num_tokens": 21408200.0,
      "step": 9330
    },
    {
      "entropy": 5.376826810836792,
      "epoch": 0.8967339097022095,
      "grad_norm": 0.953125,
      "learning_rate": 0.000492782438103996,
      "loss": 5.1584,
      "mean_token_accuracy": 0.20834774225950242,
      "num_tokens": 21419963.0,
      "step": 9335
    },
    {
      "entropy": 5.336347961425782,
      "epoch": 0.8972142170989433,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004927738217956197,
      "loss": 5.1155,
      "mean_token_accuracy": 0.2120037257671356,
      "num_tokens": 21431824.0,
      "step": 9340
    },
    {
      "entropy": 5.2245738983154295,
      "epoch": 0.8976945244956772,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004927652004311819,
      "loss": 4.8709,
      "mean_token_accuracy": 0.2259211078286171,
      "num_tokens": 21442354.0,
      "step": 9345
    },
    {
      "entropy": 5.332903957366943,
      "epoch": 0.8981748318924111,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004927565740108828,
      "loss": 5.0398,
      "mean_token_accuracy": 0.21106244921684264,
      "num_tokens": 21453734.0,
      "step": 9350
    },
    {
      "entropy": 5.322707557678223,
      "epoch": 0.8986551392891451,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004927479425349226,
      "loss": 5.0593,
      "mean_token_accuracy": 0.21989485621452332,
      "num_tokens": 21465471.0,
      "step": 9355
    },
    {
      "entropy": 5.368767786026001,
      "epoch": 0.899135446685879,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004927393060035018,
      "loss": 5.1897,
      "mean_token_accuracy": 0.2053573101758957,
      "num_tokens": 21477775.0,
      "step": 9360
    },
    {
      "entropy": 5.260050106048584,
      "epoch": 0.8996157540826129,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004927306644168207,
      "loss": 4.9686,
      "mean_token_accuracy": 0.22822231501340867,
      "num_tokens": 21489319.0,
      "step": 9365
    },
    {
      "entropy": 5.422834205627441,
      "epoch": 0.9000960614793467,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004927220177750803,
      "loss": 5.1968,
      "mean_token_accuracy": 0.21390132904052733,
      "num_tokens": 21499742.0,
      "step": 9370
    },
    {
      "entropy": 5.297619342803955,
      "epoch": 0.9005763688760807,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004927133660784811,
      "loss": 4.9733,
      "mean_token_accuracy": 0.2223273202776909,
      "num_tokens": 21511063.0,
      "step": 9375
    },
    {
      "entropy": 5.292022466659546,
      "epoch": 0.9010566762728146,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004927047093272241,
      "loss": 4.9909,
      "mean_token_accuracy": 0.21932114511728287,
      "num_tokens": 21522500.0,
      "step": 9380
    },
    {
      "entropy": 5.360550308227539,
      "epoch": 0.9015369836695485,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00049269604752151,
      "loss": 5.0842,
      "mean_token_accuracy": 0.21180106997489928,
      "num_tokens": 21533578.0,
      "step": 9385
    },
    {
      "entropy": 5.300341463088989,
      "epoch": 0.9020172910662824,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004926873806615403,
      "loss": 5.068,
      "mean_token_accuracy": 0.21274211257696152,
      "num_tokens": 21544296.0,
      "step": 9390
    },
    {
      "entropy": 5.33724799156189,
      "epoch": 0.9024975984630164,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004926787087475158,
      "loss": 5.1581,
      "mean_token_accuracy": 0.20258792787790297,
      "num_tokens": 21555386.0,
      "step": 9395
    },
    {
      "entropy": 5.281759786605835,
      "epoch": 0.9029779058597502,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004926700317796382,
      "loss": 4.9137,
      "mean_token_accuracy": 0.22768477499485015,
      "num_tokens": 21566527.0,
      "step": 9400
    },
    {
      "entropy": 5.378341150283814,
      "epoch": 0.9034582132564841,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004926613497581088,
      "loss": 5.0617,
      "mean_token_accuracy": 0.21062430441379548,
      "num_tokens": 21576870.0,
      "step": 9405
    },
    {
      "entropy": 5.302308368682861,
      "epoch": 0.9039385206532181,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004926526626831292,
      "loss": 5.0274,
      "mean_token_accuracy": 0.21775817573070527,
      "num_tokens": 21588113.0,
      "step": 9410
    },
    {
      "entropy": 5.281714200973511,
      "epoch": 0.904418828049952,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004926439705549011,
      "loss": 5.0409,
      "mean_token_accuracy": 0.21568123996257782,
      "num_tokens": 21599307.0,
      "step": 9415
    },
    {
      "entropy": 5.285175752639771,
      "epoch": 0.9048991354466859,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004926352733736262,
      "loss": 4.9947,
      "mean_token_accuracy": 0.2229517564177513,
      "num_tokens": 21609961.0,
      "step": 9420
    },
    {
      "entropy": 5.262327337265015,
      "epoch": 0.9053794428434198,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004926265711395065,
      "loss": 5.0344,
      "mean_token_accuracy": 0.21594463288784027,
      "num_tokens": 21622222.0,
      "step": 9425
    },
    {
      "entropy": 5.295537900924683,
      "epoch": 0.9058597502401537,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000492617863852744,
      "loss": 5.0184,
      "mean_token_accuracy": 0.2234782561659813,
      "num_tokens": 21632843.0,
      "step": 9430
    },
    {
      "entropy": 5.3120135307312015,
      "epoch": 0.9063400576368876,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004926091515135409,
      "loss": 5.0552,
      "mean_token_accuracy": 0.21013385355472564,
      "num_tokens": 21645387.0,
      "step": 9435
    },
    {
      "entropy": 5.294017171859741,
      "epoch": 0.9068203650336215,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004926004341220995,
      "loss": 5.0126,
      "mean_token_accuracy": 0.2199779376387596,
      "num_tokens": 21656787.0,
      "step": 9440
    },
    {
      "entropy": 5.328347396850586,
      "epoch": 0.9073006724303554,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004925917116786222,
      "loss": 5.0574,
      "mean_token_accuracy": 0.214459191262722,
      "num_tokens": 21667800.0,
      "step": 9445
    },
    {
      "entropy": 5.335863542556763,
      "epoch": 0.9077809798270894,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004925829841833114,
      "loss": 5.0869,
      "mean_token_accuracy": 0.21195723712444306,
      "num_tokens": 21679297.0,
      "step": 9450
    },
    {
      "entropy": 5.32595043182373,
      "epoch": 0.9082612872238233,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004925742516363699,
      "loss": 5.1115,
      "mean_token_accuracy": 0.20532667338848115,
      "num_tokens": 21692956.0,
      "step": 9455
    },
    {
      "entropy": 5.277609825134277,
      "epoch": 0.9087415946205571,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004925655140380002,
      "loss": 4.8805,
      "mean_token_accuracy": 0.2253878116607666,
      "num_tokens": 21704852.0,
      "step": 9460
    },
    {
      "entropy": 5.29805965423584,
      "epoch": 0.909221902017291,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004925567713884054,
      "loss": 5.0349,
      "mean_token_accuracy": 0.21731006503105163,
      "num_tokens": 21715886.0,
      "step": 9465
    },
    {
      "entropy": 5.276641845703125,
      "epoch": 0.909702209414025,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004925480236877884,
      "loss": 5.003,
      "mean_token_accuracy": 0.22124453783035278,
      "num_tokens": 21727157.0,
      "step": 9470
    },
    {
      "entropy": 5.302370071411133,
      "epoch": 0.9101825168107589,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004925392709363522,
      "loss": 5.0364,
      "mean_token_accuracy": 0.2185185119509697,
      "num_tokens": 21738232.0,
      "step": 9475
    },
    {
      "entropy": 5.287598466873169,
      "epoch": 0.9106628242074928,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004925305131343001,
      "loss": 5.0504,
      "mean_token_accuracy": 0.21443340480327605,
      "num_tokens": 21749029.0,
      "step": 9480
    },
    {
      "entropy": 5.353913640975952,
      "epoch": 0.9111431316042267,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004925217502818355,
      "loss": 5.0668,
      "mean_token_accuracy": 0.2134115591645241,
      "num_tokens": 21761243.0,
      "step": 9485
    },
    {
      "entropy": 5.300025892257691,
      "epoch": 0.9116234390009607,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004925129823791616,
      "loss": 5.0326,
      "mean_token_accuracy": 0.21262899935245513,
      "num_tokens": 21772623.0,
      "step": 9490
    },
    {
      "entropy": 5.369426488876343,
      "epoch": 0.9121037463976945,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004925042094264822,
      "loss": 5.0629,
      "mean_token_accuracy": 0.21896116584539413,
      "num_tokens": 21782611.0,
      "step": 9495
    },
    {
      "entropy": 5.22397027015686,
      "epoch": 0.9125840537944284,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000492495431424001,
      "loss": 4.9353,
      "mean_token_accuracy": 0.22842116206884383,
      "num_tokens": 21793946.0,
      "step": 9500
    },
    {
      "entropy": 5.387156534194946,
      "epoch": 0.9130643611911623,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004924866483719216,
      "loss": 5.1061,
      "mean_token_accuracy": 0.20594156682491302,
      "num_tokens": 21803878.0,
      "step": 9505
    },
    {
      "entropy": 5.296429920196533,
      "epoch": 0.9135446685878963,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004924778602704481,
      "loss": 5.0479,
      "mean_token_accuracy": 0.21579761654138566,
      "num_tokens": 21815187.0,
      "step": 9510
    },
    {
      "entropy": 5.309639263153076,
      "epoch": 0.9140249759846302,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004924690671197845,
      "loss": 5.0315,
      "mean_token_accuracy": 0.2213830441236496,
      "num_tokens": 21825597.0,
      "step": 9515
    },
    {
      "entropy": 5.2714544296264645,
      "epoch": 0.914505283381364,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004924602689201348,
      "loss": 4.9977,
      "mean_token_accuracy": 0.2193887785077095,
      "num_tokens": 21837110.0,
      "step": 9520
    },
    {
      "entropy": 5.477096748352051,
      "epoch": 0.9149855907780979,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004924514656717034,
      "loss": 5.2517,
      "mean_token_accuracy": 0.20803963840007783,
      "num_tokens": 21847754.0,
      "step": 9525
    },
    {
      "entropy": 5.287465810775757,
      "epoch": 0.9154658981748319,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004924426573746948,
      "loss": 5.0353,
      "mean_token_accuracy": 0.21161038279533387,
      "num_tokens": 21859162.0,
      "step": 9530
    },
    {
      "entropy": 5.327201271057129,
      "epoch": 0.9159462055715658,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004924338440293131,
      "loss": 5.0151,
      "mean_token_accuracy": 0.22045329958200455,
      "num_tokens": 21870826.0,
      "step": 9535
    },
    {
      "entropy": 5.291822719573974,
      "epoch": 0.9164265129682997,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004924250256357635,
      "loss": 5.0609,
      "mean_token_accuracy": 0.2096717819571495,
      "num_tokens": 21882003.0,
      "step": 9540
    },
    {
      "entropy": 5.29669508934021,
      "epoch": 0.9169068203650336,
      "grad_norm": 0.875,
      "learning_rate": 0.0004924162021942502,
      "loss": 5.0111,
      "mean_token_accuracy": 0.21527315080165862,
      "num_tokens": 21894132.0,
      "step": 9545
    },
    {
      "entropy": 5.2839117527008055,
      "epoch": 0.9173871277617676,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004924073737049784,
      "loss": 4.985,
      "mean_token_accuracy": 0.22348141223192214,
      "num_tokens": 21904951.0,
      "step": 9550
    },
    {
      "entropy": 5.250293111801147,
      "epoch": 0.9178674351585014,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004923985401681528,
      "loss": 5.0285,
      "mean_token_accuracy": 0.21841436624526978,
      "num_tokens": 21917100.0,
      "step": 9555
    },
    {
      "entropy": 5.346006631851196,
      "epoch": 0.9183477425552353,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004923897015839788,
      "loss": 5.0562,
      "mean_token_accuracy": 0.21544896215200424,
      "num_tokens": 21927541.0,
      "step": 9560
    },
    {
      "entropy": 5.316199064254761,
      "epoch": 0.9188280499519692,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004923808579526613,
      "loss": 5.0016,
      "mean_token_accuracy": 0.2158419594168663,
      "num_tokens": 21938723.0,
      "step": 9565
    },
    {
      "entropy": 5.252933311462402,
      "epoch": 0.9193083573487032,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004923720092744059,
      "loss": 4.9157,
      "mean_token_accuracy": 0.22935039401054383,
      "num_tokens": 21950424.0,
      "step": 9570
    },
    {
      "entropy": 5.329428434371948,
      "epoch": 0.9197886647454371,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004923631555494179,
      "loss": 5.0918,
      "mean_token_accuracy": 0.2169760212302208,
      "num_tokens": 21961030.0,
      "step": 9575
    },
    {
      "entropy": 5.321308851242065,
      "epoch": 0.920268972142171,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004923542967779028,
      "loss": 4.99,
      "mean_token_accuracy": 0.22132398039102555,
      "num_tokens": 21971625.0,
      "step": 9580
    },
    {
      "entropy": 5.3300539493560795,
      "epoch": 0.920749279538905,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004923454329600664,
      "loss": 5.0168,
      "mean_token_accuracy": 0.22263507694005966,
      "num_tokens": 21983733.0,
      "step": 9585
    },
    {
      "entropy": 5.245372009277344,
      "epoch": 0.9212295869356388,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004923365640961143,
      "loss": 4.9647,
      "mean_token_accuracy": 0.2244121789932251,
      "num_tokens": 21995621.0,
      "step": 9590
    },
    {
      "entropy": 5.338005924224854,
      "epoch": 0.9217098943323727,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004923276901862526,
      "loss": 5.0418,
      "mean_token_accuracy": 0.21422896832227706,
      "num_tokens": 22007325.0,
      "step": 9595
    },
    {
      "entropy": 5.312576150894165,
      "epoch": 0.9221902017291066,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004923188112306874,
      "loss": 5.0005,
      "mean_token_accuracy": 0.21406269818544388,
      "num_tokens": 22017733.0,
      "step": 9600
    },
    {
      "entropy": 5.291594314575195,
      "epoch": 0.9226705091258406,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004923099272296246,
      "loss": 5.077,
      "mean_token_accuracy": 0.21990929991006852,
      "num_tokens": 22030451.0,
      "step": 9605
    },
    {
      "entropy": 5.288282108306885,
      "epoch": 0.9231508165225745,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004923010381832706,
      "loss": 5.05,
      "mean_token_accuracy": 0.21142314970493317,
      "num_tokens": 22042626.0,
      "step": 9610
    },
    {
      "entropy": 5.274527645111084,
      "epoch": 0.9236311239193083,
      "grad_norm": 1.0,
      "learning_rate": 0.0004922921440918318,
      "loss": 5.0295,
      "mean_token_accuracy": 0.2114730805158615,
      "num_tokens": 22053314.0,
      "step": 9615
    },
    {
      "entropy": 5.374013614654541,
      "epoch": 0.9241114313160422,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004922832449555144,
      "loss": 5.1209,
      "mean_token_accuracy": 0.20706668198108674,
      "num_tokens": 22064395.0,
      "step": 9620
    },
    {
      "entropy": 5.321868133544922,
      "epoch": 0.9245917387127762,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004922743407745255,
      "loss": 4.9881,
      "mean_token_accuracy": 0.21041023433208467,
      "num_tokens": 22075960.0,
      "step": 9625
    },
    {
      "entropy": 5.335540628433227,
      "epoch": 0.9250720461095101,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004922654315490714,
      "loss": 5.0836,
      "mean_token_accuracy": 0.2134032741189003,
      "num_tokens": 22086034.0,
      "step": 9630
    },
    {
      "entropy": 5.3098639965057375,
      "epoch": 0.925552353506244,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004922565172793593,
      "loss": 5.0966,
      "mean_token_accuracy": 0.21248842775821686,
      "num_tokens": 22096184.0,
      "step": 9635
    },
    {
      "entropy": 5.341927194595337,
      "epoch": 0.9260326609029779,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004922475979655958,
      "loss": 5.0605,
      "mean_token_accuracy": 0.21449708789587021,
      "num_tokens": 22108795.0,
      "step": 9640
    },
    {
      "entropy": 5.27025260925293,
      "epoch": 0.9265129682997119,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004922386736079883,
      "loss": 5.0224,
      "mean_token_accuracy": 0.2129576712846756,
      "num_tokens": 22119608.0,
      "step": 9645
    },
    {
      "entropy": 5.285499334335327,
      "epoch": 0.9269932756964457,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004922297442067438,
      "loss": 5.0037,
      "mean_token_accuracy": 0.2271778926253319,
      "num_tokens": 22131621.0,
      "step": 9650
    },
    {
      "entropy": 5.3321291446685795,
      "epoch": 0.9274735830931796,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004922208097620697,
      "loss": 4.9729,
      "mean_token_accuracy": 0.21801707297563552,
      "num_tokens": 22142745.0,
      "step": 9655
    },
    {
      "entropy": 5.2967113018035885,
      "epoch": 0.9279538904899135,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004922118702741735,
      "loss": 5.1588,
      "mean_token_accuracy": 0.21353457272052764,
      "num_tokens": 22155457.0,
      "step": 9660
    },
    {
      "entropy": 5.3383396625518795,
      "epoch": 0.9284341978866475,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004922029257432625,
      "loss": 5.028,
      "mean_token_accuracy": 0.22553674578666688,
      "num_tokens": 22165955.0,
      "step": 9665
    },
    {
      "entropy": 5.301980018615723,
      "epoch": 0.9289145052833814,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004921939761695446,
      "loss": 4.9701,
      "mean_token_accuracy": 0.2177879497408867,
      "num_tokens": 22178142.0,
      "step": 9670
    },
    {
      "entropy": 5.191862678527832,
      "epoch": 0.9293948126801153,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004921850215532275,
      "loss": 4.951,
      "mean_token_accuracy": 0.21580154448747635,
      "num_tokens": 22190315.0,
      "step": 9675
    },
    {
      "entropy": 5.2612996101379395,
      "epoch": 0.9298751200768491,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004921760618945192,
      "loss": 5.012,
      "mean_token_accuracy": 0.22119226455688476,
      "num_tokens": 22201785.0,
      "step": 9680
    },
    {
      "entropy": 5.336707925796508,
      "epoch": 0.9303554274735831,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004921670971936276,
      "loss": 4.9725,
      "mean_token_accuracy": 0.22037847191095353,
      "num_tokens": 22212471.0,
      "step": 9685
    },
    {
      "entropy": 5.318843841552734,
      "epoch": 0.930835734870317,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004921581274507607,
      "loss": 5.0549,
      "mean_token_accuracy": 0.22015346735715866,
      "num_tokens": 22223188.0,
      "step": 9690
    },
    {
      "entropy": 5.240539455413819,
      "epoch": 0.9313160422670509,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000492149152666127,
      "loss": 5.0535,
      "mean_token_accuracy": 0.2153894707560539,
      "num_tokens": 22233978.0,
      "step": 9695
    },
    {
      "entropy": 5.2962329387664795,
      "epoch": 0.9317963496637848,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004921401728399348,
      "loss": 5.0052,
      "mean_token_accuracy": 0.22538379430770875,
      "num_tokens": 22244713.0,
      "step": 9700
    },
    {
      "entropy": 5.313981342315674,
      "epoch": 0.9322766570605188,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004921311879723926,
      "loss": 5.0612,
      "mean_token_accuracy": 0.21742804497480392,
      "num_tokens": 22256192.0,
      "step": 9705
    },
    {
      "entropy": 5.320638275146484,
      "epoch": 0.9327569644572526,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004921221980637088,
      "loss": 5.0253,
      "mean_token_accuracy": 0.21348996013402938,
      "num_tokens": 22268294.0,
      "step": 9710
    },
    {
      "entropy": 5.345013046264649,
      "epoch": 0.9332372718539865,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004921132031140925,
      "loss": 5.0619,
      "mean_token_accuracy": 0.22159128934144973,
      "num_tokens": 22278952.0,
      "step": 9715
    },
    {
      "entropy": 5.298283100128174,
      "epoch": 0.9337175792507204,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004921042031237521,
      "loss": 5.0077,
      "mean_token_accuracy": 0.214227357506752,
      "num_tokens": 22291057.0,
      "step": 9720
    },
    {
      "entropy": 5.362701845169068,
      "epoch": 0.9341978866474544,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004920951980928969,
      "loss": 5.1001,
      "mean_token_accuracy": 0.21576779037714006,
      "num_tokens": 22302479.0,
      "step": 9725
    },
    {
      "entropy": 5.40646915435791,
      "epoch": 0.9346781940441883,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004920861880217359,
      "loss": 5.1498,
      "mean_token_accuracy": 0.21108359694480897,
      "num_tokens": 22315116.0,
      "step": 9730
    },
    {
      "entropy": 5.362285614013672,
      "epoch": 0.9351585014409222,
      "grad_norm": 0.875,
      "learning_rate": 0.0004920771729104781,
      "loss": 5.0711,
      "mean_token_accuracy": 0.21583196371793748,
      "num_tokens": 22327548.0,
      "step": 9735
    },
    {
      "entropy": 5.310701847076416,
      "epoch": 0.9356388088376562,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004920681527593329,
      "loss": 4.9447,
      "mean_token_accuracy": 0.22339875102043152,
      "num_tokens": 22339154.0,
      "step": 9740
    },
    {
      "entropy": 5.272175693511963,
      "epoch": 0.93611911623439,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004920591275685098,
      "loss": 5.0547,
      "mean_token_accuracy": 0.21745239645242692,
      "num_tokens": 22350781.0,
      "step": 9745
    },
    {
      "entropy": 5.330118989944458,
      "epoch": 0.9365994236311239,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004920500973382184,
      "loss": 5.059,
      "mean_token_accuracy": 0.21584096401929856,
      "num_tokens": 22361990.0,
      "step": 9750
    },
    {
      "entropy": 5.311127805709839,
      "epoch": 0.9370797310278578,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004920410620686682,
      "loss": 4.9467,
      "mean_token_accuracy": 0.22903422266244888,
      "num_tokens": 22372973.0,
      "step": 9755
    },
    {
      "entropy": 5.224628543853759,
      "epoch": 0.9375600384245918,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004920320217600689,
      "loss": 4.9639,
      "mean_token_accuracy": 0.2249605804681778,
      "num_tokens": 22384369.0,
      "step": 9760
    },
    {
      "entropy": 5.29384708404541,
      "epoch": 0.9380403458213257,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004920229764126306,
      "loss": 5.0578,
      "mean_token_accuracy": 0.22132847607135772,
      "num_tokens": 22395792.0,
      "step": 9765
    },
    {
      "entropy": 5.314130878448486,
      "epoch": 0.9385206532180596,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004920139260265632,
      "loss": 5.0452,
      "mean_token_accuracy": 0.21164032071828842,
      "num_tokens": 22408182.0,
      "step": 9770
    },
    {
      "entropy": 5.407832193374634,
      "epoch": 0.9390009606147934,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004920048706020769,
      "loss": 5.1843,
      "mean_token_accuracy": 0.2085844174027443,
      "num_tokens": 22419774.0,
      "step": 9775
    },
    {
      "entropy": 5.2361040115356445,
      "epoch": 0.9394812680115274,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004919958101393817,
      "loss": 4.8937,
      "mean_token_accuracy": 0.22411441206932067,
      "num_tokens": 22430210.0,
      "step": 9780
    },
    {
      "entropy": 5.3042370796203615,
      "epoch": 0.9399615754082613,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004919867446386883,
      "loss": 5.0594,
      "mean_token_accuracy": 0.21930066645145416,
      "num_tokens": 22442444.0,
      "step": 9785
    },
    {
      "entropy": 5.253676652908325,
      "epoch": 0.9404418828049952,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000491977674100207,
      "loss": 5.0051,
      "mean_token_accuracy": 0.2222920313477516,
      "num_tokens": 22455521.0,
      "step": 9790
    },
    {
      "entropy": 5.351623296737671,
      "epoch": 0.9409221902017291,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004919685985241483,
      "loss": 5.0587,
      "mean_token_accuracy": 0.2203366920351982,
      "num_tokens": 22466997.0,
      "step": 9795
    },
    {
      "entropy": 5.2852089405059814,
      "epoch": 0.9414024975984631,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000491959517910723,
      "loss": 4.9935,
      "mean_token_accuracy": 0.22110982835292817,
      "num_tokens": 22477851.0,
      "step": 9800
    },
    {
      "entropy": 5.231943035125733,
      "epoch": 0.9418828049951969,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004919504322601421,
      "loss": 5.0692,
      "mean_token_accuracy": 0.2192753002047539,
      "num_tokens": 22489319.0,
      "step": 9805
    },
    {
      "entropy": 5.226250076293946,
      "epoch": 0.9423631123919308,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004919413415726162,
      "loss": 4.9336,
      "mean_token_accuracy": 0.22262371182441712,
      "num_tokens": 22500847.0,
      "step": 9810
    },
    {
      "entropy": 5.337717771530151,
      "epoch": 0.9428434197886647,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004919322458483566,
      "loss": 5.0197,
      "mean_token_accuracy": 0.21989092826843262,
      "num_tokens": 22512719.0,
      "step": 9815
    },
    {
      "entropy": 5.229172563552856,
      "epoch": 0.9433237271853987,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004919231450875745,
      "loss": 4.955,
      "mean_token_accuracy": 0.22381351739168168,
      "num_tokens": 22522984.0,
      "step": 9820
    },
    {
      "entropy": 5.271284198760986,
      "epoch": 0.9438040345821326,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004919140392904809,
      "loss": 5.0181,
      "mean_token_accuracy": 0.22189277559518814,
      "num_tokens": 22534816.0,
      "step": 9825
    },
    {
      "entropy": 5.3265495777130125,
      "epoch": 0.9442843419788665,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004919049284572875,
      "loss": 5.0014,
      "mean_token_accuracy": 0.2158041477203369,
      "num_tokens": 22545753.0,
      "step": 9830
    },
    {
      "entropy": 5.355762577056884,
      "epoch": 0.9447646493756003,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004918958125882058,
      "loss": 5.0852,
      "mean_token_accuracy": 0.2120089739561081,
      "num_tokens": 22557237.0,
      "step": 9835
    },
    {
      "entropy": 5.320783710479736,
      "epoch": 0.9452449567723343,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004918866916834474,
      "loss": 5.0002,
      "mean_token_accuracy": 0.21973955184221267,
      "num_tokens": 22568909.0,
      "step": 9840
    },
    {
      "entropy": 5.40159239768982,
      "epoch": 0.9457252641690682,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004918775657432239,
      "loss": 5.1431,
      "mean_token_accuracy": 0.21086762696504593,
      "num_tokens": 22582162.0,
      "step": 9845
    },
    {
      "entropy": 5.275126314163208,
      "epoch": 0.9462055715658021,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004918684347677474,
      "loss": 5.0001,
      "mean_token_accuracy": 0.2167303204536438,
      "num_tokens": 22592405.0,
      "step": 9850
    },
    {
      "entropy": 5.306456708908081,
      "epoch": 0.946685878962536,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004918592987572298,
      "loss": 5.0259,
      "mean_token_accuracy": 0.2159822553396225,
      "num_tokens": 22603588.0,
      "step": 9855
    },
    {
      "entropy": 5.424379158020019,
      "epoch": 0.94716618635927,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004918501577118832,
      "loss": 5.0999,
      "mean_token_accuracy": 0.2184132531285286,
      "num_tokens": 22614995.0,
      "step": 9860
    },
    {
      "entropy": 5.301549100875855,
      "epoch": 0.9476464937560038,
      "grad_norm": 0.9375,
      "learning_rate": 0.00049184101163192,
      "loss": 5.0532,
      "mean_token_accuracy": 0.2093043655157089,
      "num_tokens": 22627556.0,
      "step": 9865
    },
    {
      "entropy": 5.32760272026062,
      "epoch": 0.9481268011527377,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004918318605175522,
      "loss": 5.0392,
      "mean_token_accuracy": 0.22332019060850145,
      "num_tokens": 22638339.0,
      "step": 9870
    },
    {
      "entropy": 5.356701183319092,
      "epoch": 0.9486071085494716,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004918227043689924,
      "loss": 4.9975,
      "mean_token_accuracy": 0.22180181443691255,
      "num_tokens": 22648922.0,
      "step": 9875
    },
    {
      "entropy": 5.272241592407227,
      "epoch": 0.9490874159462056,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004918135431864534,
      "loss": 5.0233,
      "mean_token_accuracy": 0.22174060791730882,
      "num_tokens": 22662006.0,
      "step": 9880
    },
    {
      "entropy": 5.2540536403656,
      "epoch": 0.9495677233429395,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004918043769701478,
      "loss": 4.9785,
      "mean_token_accuracy": 0.22149344980716706,
      "num_tokens": 22674649.0,
      "step": 9885
    },
    {
      "entropy": 5.365615797042847,
      "epoch": 0.9500480307396734,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004917952057202882,
      "loss": 5.0827,
      "mean_token_accuracy": 0.20941140800714492,
      "num_tokens": 22685971.0,
      "step": 9890
    },
    {
      "entropy": 5.295986700057983,
      "epoch": 0.9505283381364072,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004917860294370877,
      "loss": 4.9563,
      "mean_token_accuracy": 0.2219228759407997,
      "num_tokens": 22696174.0,
      "step": 9895
    },
    {
      "entropy": 5.343828630447388,
      "epoch": 0.9510086455331412,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004917768481207593,
      "loss": 5.0676,
      "mean_token_accuracy": 0.21522822678089143,
      "num_tokens": 22706983.0,
      "step": 9900
    },
    {
      "entropy": 5.323217821121216,
      "epoch": 0.9514889529298751,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004917676617715162,
      "loss": 4.9996,
      "mean_token_accuracy": 0.21866356134414672,
      "num_tokens": 22718251.0,
      "step": 9905
    },
    {
      "entropy": 5.244712257385254,
      "epoch": 0.951969260326609,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004917584703895717,
      "loss": 4.9765,
      "mean_token_accuracy": 0.22329760491847991,
      "num_tokens": 22730071.0,
      "step": 9910
    },
    {
      "entropy": 5.273686122894287,
      "epoch": 0.952449567723343,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004917492739751391,
      "loss": 5.0273,
      "mean_token_accuracy": 0.21593200266361237,
      "num_tokens": 22742027.0,
      "step": 9915
    },
    {
      "entropy": 5.231245136260986,
      "epoch": 0.9529298751200769,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000491740072528432,
      "loss": 4.9182,
      "mean_token_accuracy": 0.2306719332933426,
      "num_tokens": 22752946.0,
      "step": 9920
    },
    {
      "entropy": 5.329925537109375,
      "epoch": 0.9534101825168108,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000491730866049664,
      "loss": 4.9973,
      "mean_token_accuracy": 0.2207105800509453,
      "num_tokens": 22763944.0,
      "step": 9925
    },
    {
      "entropy": 5.239139032363892,
      "epoch": 0.9538904899135446,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004917216545390489,
      "loss": 4.9723,
      "mean_token_accuracy": 0.22325474321842192,
      "num_tokens": 22774414.0,
      "step": 9930
    },
    {
      "entropy": 5.212150573730469,
      "epoch": 0.9543707973102786,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004917124379968004,
      "loss": 4.9188,
      "mean_token_accuracy": 0.23533675372600554,
      "num_tokens": 22785533.0,
      "step": 9935
    },
    {
      "entropy": 5.2332391262054445,
      "epoch": 0.9548511047070125,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004917032164231327,
      "loss": 4.8796,
      "mean_token_accuracy": 0.22827497869729996,
      "num_tokens": 22795809.0,
      "step": 9940
    },
    {
      "entropy": 5.300201463699341,
      "epoch": 0.9553314121037464,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004916939898182598,
      "loss": 5.1133,
      "mean_token_accuracy": 0.2133575603365898,
      "num_tokens": 22807705.0,
      "step": 9945
    },
    {
      "entropy": 5.310309267044067,
      "epoch": 0.9558117195004803,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004916847581823958,
      "loss": 5.0087,
      "mean_token_accuracy": 0.22103045880794525,
      "num_tokens": 22818852.0,
      "step": 9950
    },
    {
      "entropy": 5.249653244018555,
      "epoch": 0.9562920268972143,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004916755215157552,
      "loss": 4.9293,
      "mean_token_accuracy": 0.22121626734733582,
      "num_tokens": 22829146.0,
      "step": 9955
    },
    {
      "entropy": 5.290464591979981,
      "epoch": 0.9567723342939481,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004916662798185524,
      "loss": 5.0095,
      "mean_token_accuracy": 0.22800589352846146,
      "num_tokens": 22840088.0,
      "step": 9960
    },
    {
      "entropy": 5.284116554260254,
      "epoch": 0.957252641690682,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004916570330910019,
      "loss": 5.034,
      "mean_token_accuracy": 0.22295115888118744,
      "num_tokens": 22852470.0,
      "step": 9965
    },
    {
      "entropy": 5.332837724685669,
      "epoch": 0.9577329490874159,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004916477813333185,
      "loss": 5.0598,
      "mean_token_accuracy": 0.21266402453184127,
      "num_tokens": 22863673.0,
      "step": 9970
    },
    {
      "entropy": 5.287660074234009,
      "epoch": 0.9582132564841499,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004916385245457168,
      "loss": 5.0561,
      "mean_token_accuracy": 0.2156519278883934,
      "num_tokens": 22874888.0,
      "step": 9975
    },
    {
      "entropy": 5.321747875213623,
      "epoch": 0.9586935638808838,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000491629262728412,
      "loss": 5.0567,
      "mean_token_accuracy": 0.2118343621492386,
      "num_tokens": 22886811.0,
      "step": 9980
    },
    {
      "entropy": 5.342364549636841,
      "epoch": 0.9591738712776177,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004916199958816188,
      "loss": 5.0106,
      "mean_token_accuracy": 0.21754788160324096,
      "num_tokens": 22898777.0,
      "step": 9985
    },
    {
      "entropy": 5.311126327514648,
      "epoch": 0.9596541786743515,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004916107240055527,
      "loss": 4.9666,
      "mean_token_accuracy": 0.22761310338974,
      "num_tokens": 22910804.0,
      "step": 9990
    },
    {
      "entropy": 5.354531717300415,
      "epoch": 0.9601344860710855,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004916014471004287,
      "loss": 5.1254,
      "mean_token_accuracy": 0.220575250685215,
      "num_tokens": 22922002.0,
      "step": 9995
    },
    {
      "entropy": 5.325109243392944,
      "epoch": 0.9606147934678194,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004915921651664622,
      "loss": 5.0089,
      "mean_token_accuracy": 0.21408329159021378,
      "num_tokens": 22933471.0,
      "step": 10000
    },
    {
      "entropy": 5.224257469177246,
      "epoch": 0.9610951008645533,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000491582878203869,
      "loss": 4.9326,
      "mean_token_accuracy": 0.22505528032779692,
      "num_tokens": 22945303.0,
      "step": 10005
    },
    {
      "entropy": 5.292037534713745,
      "epoch": 0.9615754082612872,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004915735862128643,
      "loss": 5.0657,
      "mean_token_accuracy": 0.20910924673080444,
      "num_tokens": 22956620.0,
      "step": 10010
    },
    {
      "entropy": 5.314778804779053,
      "epoch": 0.9620557156580212,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004915642891936641,
      "loss": 5.0578,
      "mean_token_accuracy": 0.2219095468521118,
      "num_tokens": 22968941.0,
      "step": 10015
    },
    {
      "entropy": 5.33321442604065,
      "epoch": 0.962536023054755,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004915549871464841,
      "loss": 5.0361,
      "mean_token_accuracy": 0.21429357081651687,
      "num_tokens": 22980222.0,
      "step": 10020
    },
    {
      "entropy": 5.377556276321411,
      "epoch": 0.9630163304514889,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004915456800715403,
      "loss": 5.0425,
      "mean_token_accuracy": 0.21123910546302796,
      "num_tokens": 22991156.0,
      "step": 10025
    },
    {
      "entropy": 5.336756849288941,
      "epoch": 0.9634966378482228,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000491536367969049,
      "loss": 5.1813,
      "mean_token_accuracy": 0.21061207503080367,
      "num_tokens": 23002939.0,
      "step": 10030
    },
    {
      "entropy": 5.394953393936158,
      "epoch": 0.9639769452449568,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004915270508392261,
      "loss": 5.0385,
      "mean_token_accuracy": 0.21745241582393646,
      "num_tokens": 23015590.0,
      "step": 10035
    },
    {
      "entropy": 5.252877998352051,
      "epoch": 0.9644572526416907,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000491517728682288,
      "loss": 4.9266,
      "mean_token_accuracy": 0.2228121280670166,
      "num_tokens": 23026387.0,
      "step": 10040
    },
    {
      "entropy": 5.219428205490113,
      "epoch": 0.9649375600384246,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004915084014984512,
      "loss": 4.9925,
      "mean_token_accuracy": 0.22173692882061005,
      "num_tokens": 23037475.0,
      "step": 10045
    },
    {
      "entropy": 5.2481975078582765,
      "epoch": 0.9654178674351584,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004914990692879322,
      "loss": 4.9651,
      "mean_token_accuracy": 0.2259371742606163,
      "num_tokens": 23049305.0,
      "step": 10050
    },
    {
      "entropy": 5.300817632675171,
      "epoch": 0.9658981748318924,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004914897320509478,
      "loss": 4.9908,
      "mean_token_accuracy": 0.22172350883483888,
      "num_tokens": 23061765.0,
      "step": 10055
    },
    {
      "entropy": 5.257356834411621,
      "epoch": 0.9663784822286263,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004914803897877146,
      "loss": 4.9985,
      "mean_token_accuracy": 0.21440304666757584,
      "num_tokens": 23072355.0,
      "step": 10060
    },
    {
      "entropy": 5.32854437828064,
      "epoch": 0.9668587896253602,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004914710424984495,
      "loss": 5.0043,
      "mean_token_accuracy": 0.2121011108160019,
      "num_tokens": 23085583.0,
      "step": 10065
    },
    {
      "entropy": 5.378414011001587,
      "epoch": 0.9673390970220941,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004914616901833696,
      "loss": 5.0624,
      "mean_token_accuracy": 0.2151917889714241,
      "num_tokens": 23095942.0,
      "step": 10070
    },
    {
      "entropy": 5.302213287353515,
      "epoch": 0.9678194044188281,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000491452332842692,
      "loss": 4.9753,
      "mean_token_accuracy": 0.22557428926229478,
      "num_tokens": 23106540.0,
      "step": 10075
    },
    {
      "entropy": 5.283980798721314,
      "epoch": 0.968299711815562,
      "grad_norm": 0.953125,
      "learning_rate": 0.000491442970476634,
      "loss": 5.0369,
      "mean_token_accuracy": 0.22101497948169707,
      "num_tokens": 23118006.0,
      "step": 10080
    },
    {
      "entropy": 5.269079351425171,
      "epoch": 0.9687800192122958,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004914336030854129,
      "loss": 4.9404,
      "mean_token_accuracy": 0.22801474779844283,
      "num_tokens": 23129103.0,
      "step": 10085
    },
    {
      "entropy": 5.251669120788574,
      "epoch": 0.9692603266090298,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004914242306692461,
      "loss": 4.9605,
      "mean_token_accuracy": 0.22190286964178085,
      "num_tokens": 23140009.0,
      "step": 10090
    },
    {
      "entropy": 5.317932891845703,
      "epoch": 0.9697406340057637,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004914148532283516,
      "loss": 5.0005,
      "mean_token_accuracy": 0.22181001603603362,
      "num_tokens": 23150982.0,
      "step": 10095
    },
    {
      "entropy": 5.299000930786133,
      "epoch": 0.9702209414024976,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004914054707629466,
      "loss": 5.0184,
      "mean_token_accuracy": 0.21632662266492844,
      "num_tokens": 23161834.0,
      "step": 10100
    },
    {
      "entropy": 5.334769248962402,
      "epoch": 0.9707012487992315,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004913960832732493,
      "loss": 5.1415,
      "mean_token_accuracy": 0.21129993945360184,
      "num_tokens": 23173355.0,
      "step": 10105
    },
    {
      "entropy": 5.347547817230224,
      "epoch": 0.9711815561959655,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004913866907594774,
      "loss": 5.0729,
      "mean_token_accuracy": 0.21839730739593505,
      "num_tokens": 23185075.0,
      "step": 10110
    },
    {
      "entropy": 5.298357248306274,
      "epoch": 0.9716618635926993,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004913772932218491,
      "loss": 5.0306,
      "mean_token_accuracy": 0.22525582313537598,
      "num_tokens": 23195590.0,
      "step": 10115
    },
    {
      "entropy": 5.363512706756592,
      "epoch": 0.9721421709894332,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004913678906605825,
      "loss": 5.0112,
      "mean_token_accuracy": 0.21946504414081575,
      "num_tokens": 23207668.0,
      "step": 10120
    },
    {
      "entropy": 5.2490180969238285,
      "epoch": 0.9726224783861671,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004913584830758961,
      "loss": 4.9907,
      "mean_token_accuracy": 0.22626605480909348,
      "num_tokens": 23218497.0,
      "step": 10125
    },
    {
      "entropy": 5.360607147216797,
      "epoch": 0.9731027857829011,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004913490704680081,
      "loss": 5.0589,
      "mean_token_accuracy": 0.21183543801307678,
      "num_tokens": 23230575.0,
      "step": 10130
    },
    {
      "entropy": 5.322029542922974,
      "epoch": 0.973583093179635,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004913396528371371,
      "loss": 5.0718,
      "mean_token_accuracy": 0.21645451486110687,
      "num_tokens": 23242348.0,
      "step": 10135
    },
    {
      "entropy": 5.349451732635498,
      "epoch": 0.9740634005763689,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004913302301835018,
      "loss": 5.0349,
      "mean_token_accuracy": 0.22246104925870896,
      "num_tokens": 23253297.0,
      "step": 10140
    },
    {
      "entropy": 5.266198921203613,
      "epoch": 0.9745437079731027,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000491320802507321,
      "loss": 4.9783,
      "mean_token_accuracy": 0.22720368206501007,
      "num_tokens": 23265830.0,
      "step": 10145
    },
    {
      "entropy": 5.308753538131714,
      "epoch": 0.9750240153698367,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004913113698088133,
      "loss": 5.0473,
      "mean_token_accuracy": 0.21321402490139008,
      "num_tokens": 23275591.0,
      "step": 10150
    },
    {
      "entropy": 5.287076950073242,
      "epoch": 0.9755043227665706,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000491301932088198,
      "loss": 4.9457,
      "mean_token_accuracy": 0.223526331782341,
      "num_tokens": 23286685.0,
      "step": 10155
    },
    {
      "entropy": 5.229479455947876,
      "epoch": 0.9759846301633045,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004912924893456942,
      "loss": 4.9596,
      "mean_token_accuracy": 0.22354679554700851,
      "num_tokens": 23298776.0,
      "step": 10160
    },
    {
      "entropy": 5.281533288955688,
      "epoch": 0.9764649375600384,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000491283041581521,
      "loss": 4.9605,
      "mean_token_accuracy": 0.22061575800180436,
      "num_tokens": 23310498.0,
      "step": 10165
    },
    {
      "entropy": 5.298981666564941,
      "epoch": 0.9769452449567724,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004912735887958978,
      "loss": 5.0132,
      "mean_token_accuracy": 0.2203960821032524,
      "num_tokens": 23321089.0,
      "step": 10170
    },
    {
      "entropy": 5.2984107494354244,
      "epoch": 0.9774255523535063,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004912641309890441,
      "loss": 4.9974,
      "mean_token_accuracy": 0.217452335357666,
      "num_tokens": 23332142.0,
      "step": 10175
    },
    {
      "entropy": 5.231074714660645,
      "epoch": 0.9779058597502401,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004912546681611794,
      "loss": 4.962,
      "mean_token_accuracy": 0.22657897919416428,
      "num_tokens": 23343014.0,
      "step": 10180
    },
    {
      "entropy": 5.295417165756225,
      "epoch": 0.978386167146974,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004912452003125234,
      "loss": 4.9422,
      "mean_token_accuracy": 0.2236812576651573,
      "num_tokens": 23354611.0,
      "step": 10185
    },
    {
      "entropy": 5.2809806823730465,
      "epoch": 0.978866474543708,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000491235727443296,
      "loss": 4.9964,
      "mean_token_accuracy": 0.21665359735488893,
      "num_tokens": 23365608.0,
      "step": 10190
    },
    {
      "entropy": 5.321546363830566,
      "epoch": 0.9793467819404419,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004912262495537171,
      "loss": 5.0302,
      "mean_token_accuracy": 0.2251007616519928,
      "num_tokens": 23377884.0,
      "step": 10195
    },
    {
      "entropy": 5.2129639148712155,
      "epoch": 0.9798270893371758,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004912167666440068,
      "loss": 4.9361,
      "mean_token_accuracy": 0.22044955641031266,
      "num_tokens": 23389553.0,
      "step": 10200
    },
    {
      "entropy": 5.258600664138794,
      "epoch": 0.9803073967339097,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004912072787143852,
      "loss": 4.9195,
      "mean_token_accuracy": 0.22222509384155273,
      "num_tokens": 23401079.0,
      "step": 10205
    },
    {
      "entropy": 5.245012474060059,
      "epoch": 0.9807877041306436,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004911977857650725,
      "loss": 4.9865,
      "mean_token_accuracy": 0.21984685361385345,
      "num_tokens": 23412886.0,
      "step": 10210
    },
    {
      "entropy": 5.319645833969116,
      "epoch": 0.9812680115273775,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004911882877962893,
      "loss": 5.0518,
      "mean_token_accuracy": 0.21402455121278763,
      "num_tokens": 23424758.0,
      "step": 10215
    },
    {
      "entropy": 5.27735447883606,
      "epoch": 0.9817483189241114,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004911787848082559,
      "loss": 4.9176,
      "mean_token_accuracy": 0.23004138171672822,
      "num_tokens": 23435552.0,
      "step": 10220
    },
    {
      "entropy": 5.216567468643189,
      "epoch": 0.9822286263208453,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004911692768011931,
      "loss": 4.9271,
      "mean_token_accuracy": 0.22384416908025742,
      "num_tokens": 23446584.0,
      "step": 10225
    },
    {
      "entropy": 5.354017972946167,
      "epoch": 0.9827089337175793,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004911597637753217,
      "loss": 5.1484,
      "mean_token_accuracy": 0.20666585266590118,
      "num_tokens": 23458452.0,
      "step": 10230
    },
    {
      "entropy": 5.327365684509277,
      "epoch": 0.9831892411143132,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004911502457308623,
      "loss": 5.012,
      "mean_token_accuracy": 0.21689023822546005,
      "num_tokens": 23470310.0,
      "step": 10235
    },
    {
      "entropy": 5.359573411941528,
      "epoch": 0.983669548511047,
      "grad_norm": 1.015625,
      "learning_rate": 0.000491140722668036,
      "loss": 5.0612,
      "mean_token_accuracy": 0.21777289360761642,
      "num_tokens": 23481166.0,
      "step": 10240
    },
    {
      "entropy": 5.289134979248047,
      "epoch": 0.984149855907781,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000491131194587064,
      "loss": 5.0402,
      "mean_token_accuracy": 0.21946453899145127,
      "num_tokens": 23493134.0,
      "step": 10245
    },
    {
      "entropy": 5.227468347549438,
      "epoch": 0.9846301633045149,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004911216614881675,
      "loss": 4.909,
      "mean_token_accuracy": 0.2278345912694931,
      "num_tokens": 23504983.0,
      "step": 10250
    },
    {
      "entropy": 5.361302852630615,
      "epoch": 0.9851104707012488,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004911121233715677,
      "loss": 5.002,
      "mean_token_accuracy": 0.2184489369392395,
      "num_tokens": 23516119.0,
      "step": 10255
    },
    {
      "entropy": 5.37627682685852,
      "epoch": 0.9855907780979827,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004911025802374861,
      "loss": 5.0589,
      "mean_token_accuracy": 0.2115888014435768,
      "num_tokens": 23528242.0,
      "step": 10260
    },
    {
      "entropy": 5.233339643478393,
      "epoch": 0.9860710854947167,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004910930320861442,
      "loss": 4.9268,
      "mean_token_accuracy": 0.22927811294794082,
      "num_tokens": 23539738.0,
      "step": 10265
    },
    {
      "entropy": 5.133146905899048,
      "epoch": 0.9865513928914506,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004910834789177639,
      "loss": 4.953,
      "mean_token_accuracy": 0.2210186243057251,
      "num_tokens": 23551228.0,
      "step": 10270
    },
    {
      "entropy": 5.326025152206421,
      "epoch": 0.9870317002881844,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004910739207325668,
      "loss": 5.0016,
      "mean_token_accuracy": 0.2216675505042076,
      "num_tokens": 23563084.0,
      "step": 10275
    },
    {
      "entropy": 5.279853010177613,
      "epoch": 0.9875120076849183,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004910643575307749,
      "loss": 4.9741,
      "mean_token_accuracy": 0.22098210602998733,
      "num_tokens": 23574328.0,
      "step": 10280
    },
    {
      "entropy": 5.2325310707092285,
      "epoch": 0.9879923150816523,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004910547893126102,
      "loss": 4.9681,
      "mean_token_accuracy": 0.22121353298425675,
      "num_tokens": 23585230.0,
      "step": 10285
    },
    {
      "entropy": 5.300854921340942,
      "epoch": 0.9884726224783862,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004910452160782948,
      "loss": 4.9891,
      "mean_token_accuracy": 0.21553015559911728,
      "num_tokens": 23596951.0,
      "step": 10290
    },
    {
      "entropy": 5.352536344528199,
      "epoch": 0.9889529298751201,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000491035637828051,
      "loss": 5.0182,
      "mean_token_accuracy": 0.21842366605997085,
      "num_tokens": 23607759.0,
      "step": 10295
    },
    {
      "entropy": 5.296850347518921,
      "epoch": 0.989433237271854,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004910260545621012,
      "loss": 5.033,
      "mean_token_accuracy": 0.21345939189195634,
      "num_tokens": 23619631.0,
      "step": 10300
    },
    {
      "entropy": 5.2701770782470705,
      "epoch": 0.9899135446685879,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004910164662806679,
      "loss": 5.035,
      "mean_token_accuracy": 0.2240593209862709,
      "num_tokens": 23630601.0,
      "step": 10305
    },
    {
      "entropy": 5.320000696182251,
      "epoch": 0.9903938520653218,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004910068729839736,
      "loss": 4.9518,
      "mean_token_accuracy": 0.22623386085033417,
      "num_tokens": 23641330.0,
      "step": 10310
    },
    {
      "entropy": 5.3434319496154785,
      "epoch": 0.9908741594620557,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004909972746722413,
      "loss": 5.0332,
      "mean_token_accuracy": 0.2199174076318741,
      "num_tokens": 23651492.0,
      "step": 10315
    },
    {
      "entropy": 5.2613872528076175,
      "epoch": 0.9913544668587896,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004909876713456935,
      "loss": 4.9264,
      "mean_token_accuracy": 0.22344893962144852,
      "num_tokens": 23661773.0,
      "step": 10320
    },
    {
      "entropy": 5.268007755279541,
      "epoch": 0.9918347742555236,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004909780630045534,
      "loss": 4.9735,
      "mean_token_accuracy": 0.22084975391626357,
      "num_tokens": 23673534.0,
      "step": 10325
    },
    {
      "entropy": 5.32368655204773,
      "epoch": 0.9923150816522575,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000490968449649044,
      "loss": 5.0769,
      "mean_token_accuracy": 0.21700199991464614,
      "num_tokens": 23684892.0,
      "step": 10330
    },
    {
      "entropy": 5.367602586746216,
      "epoch": 0.9927953890489913,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004909588312793884,
      "loss": 5.0538,
      "mean_token_accuracy": 0.2168782666325569,
      "num_tokens": 23696076.0,
      "step": 10335
    },
    {
      "entropy": 5.330792045593261,
      "epoch": 0.9932756964457252,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004909492078958101,
      "loss": 5.0358,
      "mean_token_accuracy": 0.21044176518917085,
      "num_tokens": 23707795.0,
      "step": 10340
    },
    {
      "entropy": 5.263197565078736,
      "epoch": 0.9937560038424592,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004909395794985324,
      "loss": 4.9559,
      "mean_token_accuracy": 0.22215351909399034,
      "num_tokens": 23720802.0,
      "step": 10345
    },
    {
      "entropy": 5.279692363739014,
      "epoch": 0.9942363112391931,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004909299460877788,
      "loss": 4.9792,
      "mean_token_accuracy": 0.21437558382749558,
      "num_tokens": 23732854.0,
      "step": 10350
    },
    {
      "entropy": 5.348536968231201,
      "epoch": 0.994716618635927,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004909203076637732,
      "loss": 5.0792,
      "mean_token_accuracy": 0.21412832736968995,
      "num_tokens": 23743593.0,
      "step": 10355
    },
    {
      "entropy": 5.320701646804809,
      "epoch": 0.9951969260326609,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004909106642267392,
      "loss": 5.05,
      "mean_token_accuracy": 0.21275997012853623,
      "num_tokens": 23755447.0,
      "step": 10360
    },
    {
      "entropy": 5.282635164260864,
      "epoch": 0.9956772334293948,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004909010157769006,
      "loss": 5.0341,
      "mean_token_accuracy": 0.2189592108130455,
      "num_tokens": 23767181.0,
      "step": 10365
    },
    {
      "entropy": 5.410617780685425,
      "epoch": 0.9961575408261287,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004908913623144814,
      "loss": 5.1141,
      "mean_token_accuracy": 0.20751374810934067,
      "num_tokens": 23776356.0,
      "step": 10370
    },
    {
      "entropy": 5.290426349639892,
      "epoch": 0.9966378482228626,
      "grad_norm": 1.0,
      "learning_rate": 0.000490881703839706,
      "loss": 5.059,
      "mean_token_accuracy": 0.2208680361509323,
      "num_tokens": 23787965.0,
      "step": 10375
    },
    {
      "entropy": 5.245562410354614,
      "epoch": 0.9971181556195965,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004908720403527984,
      "loss": 4.9236,
      "mean_token_accuracy": 0.2275167480111122,
      "num_tokens": 23800327.0,
      "step": 10380
    },
    {
      "entropy": 5.283764028549195,
      "epoch": 0.9975984630163305,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000490862371853983,
      "loss": 4.9346,
      "mean_token_accuracy": 0.222221177816391,
      "num_tokens": 23812845.0,
      "step": 10385
    },
    {
      "entropy": 5.3816490650177,
      "epoch": 0.9980787704130644,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004908526983434844,
      "loss": 5.0894,
      "mean_token_accuracy": 0.21485405117273332,
      "num_tokens": 23824831.0,
      "step": 10390
    },
    {
      "entropy": 5.292706298828125,
      "epoch": 0.9985590778097982,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000490843019821527,
      "loss": 4.9345,
      "mean_token_accuracy": 0.22274068295955657,
      "num_tokens": 23836697.0,
      "step": 10395
    },
    {
      "entropy": 5.2798878192901615,
      "epoch": 0.9990393852065321,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004908333362883358,
      "loss": 4.9837,
      "mean_token_accuracy": 0.22151907086372374,
      "num_tokens": 23847112.0,
      "step": 10400
    },
    {
      "entropy": 5.361454105377197,
      "epoch": 0.9995196926032661,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004908236477441353,
      "loss": 5.0687,
      "mean_token_accuracy": 0.2196926862001419,
      "num_tokens": 23858185.0,
      "step": 10405
    },
    {
      "entropy": 5.221156263351441,
      "epoch": 1.0,
      "grad_norm": 1.125,
      "learning_rate": 0.0004908139541891505,
      "loss": 4.8482,
      "mean_token_accuracy": 0.2361444428563118,
      "num_tokens": 23868536.0,
      "step": 10410
    },
    {
      "entropy": 5.3278686046600345,
      "epoch": 1.0004803073967339,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004908042556236066,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2296917513012886,
      "num_tokens": 23880283.0,
      "step": 10415
    },
    {
      "entropy": 5.3632895946502686,
      "epoch": 1.0009606147934678,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004907945520477286,
      "loss": 4.978,
      "mean_token_accuracy": 0.22007014751434326,
      "num_tokens": 23892413.0,
      "step": 10420
    },
    {
      "entropy": 5.396367597579956,
      "epoch": 1.0014409221902016,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004907848434617419,
      "loss": 5.0655,
      "mean_token_accuracy": 0.2130151569843292,
      "num_tokens": 23903977.0,
      "step": 10425
    },
    {
      "entropy": 5.35911979675293,
      "epoch": 1.0019212295869357,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000490775129865872,
      "loss": 5.0215,
      "mean_token_accuracy": 0.21769970357418061,
      "num_tokens": 23915153.0,
      "step": 10430
    },
    {
      "entropy": 5.3030928611755375,
      "epoch": 1.0024015369836696,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004907654112603442,
      "loss": 4.9131,
      "mean_token_accuracy": 0.22039141207933427,
      "num_tokens": 23926043.0,
      "step": 10435
    },
    {
      "entropy": 5.2299731254577635,
      "epoch": 1.0028818443804035,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004907556876453843,
      "loss": 4.8102,
      "mean_token_accuracy": 0.2317795604467392,
      "num_tokens": 23936658.0,
      "step": 10440
    },
    {
      "entropy": 5.266319370269775,
      "epoch": 1.0033621517771374,
      "grad_norm": 1.0,
      "learning_rate": 0.000490745959021218,
      "loss": 4.9298,
      "mean_token_accuracy": 0.21730142533779145,
      "num_tokens": 23947676.0,
      "step": 10445
    },
    {
      "entropy": 5.311759614944458,
      "epoch": 1.0038424591738713,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004907362253880711,
      "loss": 4.9118,
      "mean_token_accuracy": 0.2232137978076935,
      "num_tokens": 23959130.0,
      "step": 10450
    },
    {
      "entropy": 5.390711832046509,
      "epoch": 1.0043227665706052,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004907264867461697,
      "loss": 5.0287,
      "mean_token_accuracy": 0.213955120742321,
      "num_tokens": 23969905.0,
      "step": 10455
    },
    {
      "entropy": 5.271139764785767,
      "epoch": 1.004803073967339,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004907167430957399,
      "loss": 4.8921,
      "mean_token_accuracy": 0.22693864554166793,
      "num_tokens": 23982016.0,
      "step": 10460
    },
    {
      "entropy": 5.315487670898437,
      "epoch": 1.005283381364073,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004907069944370077,
      "loss": 5.0371,
      "mean_token_accuracy": 0.21980830430984497,
      "num_tokens": 23994200.0,
      "step": 10465
    },
    {
      "entropy": 5.387299728393555,
      "epoch": 1.005763688760807,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004906972407701998,
      "loss": 4.9349,
      "mean_token_accuracy": 0.23154017627239226,
      "num_tokens": 24004695.0,
      "step": 10470
    },
    {
      "entropy": 5.2524463653564455,
      "epoch": 1.006243996157541,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004906874820955423,
      "loss": 4.8884,
      "mean_token_accuracy": 0.22934781163930892,
      "num_tokens": 24015922.0,
      "step": 10475
    },
    {
      "entropy": 5.238456726074219,
      "epoch": 1.0067243035542748,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004906777184132621,
      "loss": 4.8841,
      "mean_token_accuracy": 0.22528507709503173,
      "num_tokens": 24026759.0,
      "step": 10480
    },
    {
      "entropy": 5.33406662940979,
      "epoch": 1.0072046109510087,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004906679497235856,
      "loss": 4.9721,
      "mean_token_accuracy": 0.21279727071523666,
      "num_tokens": 24037988.0,
      "step": 10485
    },
    {
      "entropy": 5.313938236236572,
      "epoch": 1.0076849183477425,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004906581760267397,
      "loss": 4.9959,
      "mean_token_accuracy": 0.22539696842432022,
      "num_tokens": 24050837.0,
      "step": 10490
    },
    {
      "entropy": 5.2948326587677,
      "epoch": 1.0081652257444764,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004906483973229513,
      "loss": 4.8822,
      "mean_token_accuracy": 0.22920510470867156,
      "num_tokens": 24063085.0,
      "step": 10495
    },
    {
      "entropy": 5.2527330875396725,
      "epoch": 1.0086455331412103,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004906386136124476,
      "loss": 4.9176,
      "mean_token_accuracy": 0.2195865496993065,
      "num_tokens": 24074082.0,
      "step": 10500
    },
    {
      "entropy": 5.337040948867798,
      "epoch": 1.0091258405379442,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004906288248954554,
      "loss": 4.9793,
      "mean_token_accuracy": 0.2203507363796234,
      "num_tokens": 24085050.0,
      "step": 10505
    },
    {
      "entropy": 5.305181550979614,
      "epoch": 1.0096061479346783,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004906190311722023,
      "loss": 4.9664,
      "mean_token_accuracy": 0.2254604294896126,
      "num_tokens": 24095523.0,
      "step": 10510
    },
    {
      "entropy": 5.2492616176605225,
      "epoch": 1.0100864553314122,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004906092324429155,
      "loss": 4.911,
      "mean_token_accuracy": 0.2163439065217972,
      "num_tokens": 24107157.0,
      "step": 10515
    },
    {
      "entropy": 5.332018613815308,
      "epoch": 1.010566762728146,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004905994287078227,
      "loss": 4.9192,
      "mean_token_accuracy": 0.2261065348982811,
      "num_tokens": 24118668.0,
      "step": 10520
    },
    {
      "entropy": 5.291033363342285,
      "epoch": 1.01104707012488,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004905896199671512,
      "loss": 4.9798,
      "mean_token_accuracy": 0.22180853933095931,
      "num_tokens": 24129563.0,
      "step": 10525
    },
    {
      "entropy": 5.392156076431275,
      "epoch": 1.0115273775216138,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000490579806221129,
      "loss": 4.9913,
      "mean_token_accuracy": 0.21278861165046692,
      "num_tokens": 24139965.0,
      "step": 10530
    },
    {
      "entropy": 5.362678909301758,
      "epoch": 1.0120076849183477,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004905699874699838,
      "loss": 4.9683,
      "mean_token_accuracy": 0.21862077564001084,
      "num_tokens": 24152109.0,
      "step": 10535
    },
    {
      "entropy": 5.299379587173462,
      "epoch": 1.0124879923150816,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004905601637139436,
      "loss": 4.9534,
      "mean_token_accuracy": 0.2159527361392975,
      "num_tokens": 24162558.0,
      "step": 10540
    },
    {
      "entropy": 5.208910703659058,
      "epoch": 1.0129682997118155,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004905503349532365,
      "loss": 4.8411,
      "mean_token_accuracy": 0.22960131466388703,
      "num_tokens": 24173963.0,
      "step": 10545
    },
    {
      "entropy": 5.262447166442871,
      "epoch": 1.0134486071085496,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004905405011880906,
      "loss": 4.8929,
      "mean_token_accuracy": 0.2306346595287323,
      "num_tokens": 24184369.0,
      "step": 10550
    },
    {
      "entropy": 5.368165636062622,
      "epoch": 1.0139289145052834,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004905306624187343,
      "loss": 5.039,
      "mean_token_accuracy": 0.21562075167894362,
      "num_tokens": 24194692.0,
      "step": 10555
    },
    {
      "entropy": 5.329478931427002,
      "epoch": 1.0144092219020173,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004905208186453961,
      "loss": 4.9601,
      "mean_token_accuracy": 0.22259898632764816,
      "num_tokens": 24205363.0,
      "step": 10560
    },
    {
      "entropy": 5.347483205795288,
      "epoch": 1.0148895292987512,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004905109698683044,
      "loss": 4.9754,
      "mean_token_accuracy": 0.2167939454317093,
      "num_tokens": 24216678.0,
      "step": 10565
    },
    {
      "entropy": 5.245152950286865,
      "epoch": 1.015369836695485,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004905011160876878,
      "loss": 4.8742,
      "mean_token_accuracy": 0.22317345887422563,
      "num_tokens": 24227541.0,
      "step": 10570
    },
    {
      "entropy": 5.304044437408447,
      "epoch": 1.015850144092219,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004904912573037753,
      "loss": 4.8951,
      "mean_token_accuracy": 0.22681118547916412,
      "num_tokens": 24238118.0,
      "step": 10575
    },
    {
      "entropy": 5.346661806106567,
      "epoch": 1.0163304514889528,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004904813935167957,
      "loss": 4.999,
      "mean_token_accuracy": 0.21668680161237716,
      "num_tokens": 24250044.0,
      "step": 10580
    },
    {
      "entropy": 5.3276934146881105,
      "epoch": 1.0168107588856867,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004904715247269779,
      "loss": 5.0155,
      "mean_token_accuracy": 0.20891176015138627,
      "num_tokens": 24262805.0,
      "step": 10585
    },
    {
      "entropy": 5.262403964996338,
      "epoch": 1.0172910662824208,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004904616509345514,
      "loss": 4.9503,
      "mean_token_accuracy": 0.21291984766721725,
      "num_tokens": 24274287.0,
      "step": 10590
    },
    {
      "entropy": 5.1879736423492435,
      "epoch": 1.0177713736791547,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004904517721397449,
      "loss": 4.8193,
      "mean_token_accuracy": 0.22884566485881805,
      "num_tokens": 24284839.0,
      "step": 10595
    },
    {
      "entropy": 5.269995594024659,
      "epoch": 1.0182516810758886,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004904418883427881,
      "loss": 4.939,
      "mean_token_accuracy": 0.22381459772586823,
      "num_tokens": 24295812.0,
      "step": 10600
    },
    {
      "entropy": 5.321586179733276,
      "epoch": 1.0187319884726225,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004904319995439104,
      "loss": 4.9144,
      "mean_token_accuracy": 0.22940877079963684,
      "num_tokens": 24306365.0,
      "step": 10605
    },
    {
      "entropy": 5.292356204986572,
      "epoch": 1.0192122958693564,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004904221057433412,
      "loss": 4.9131,
      "mean_token_accuracy": 0.2165081039071083,
      "num_tokens": 24317557.0,
      "step": 10610
    },
    {
      "entropy": 5.259749460220337,
      "epoch": 1.0196926032660902,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004904122069413105,
      "loss": 4.8533,
      "mean_token_accuracy": 0.23203216195106507,
      "num_tokens": 24328874.0,
      "step": 10615
    },
    {
      "entropy": 5.3840187072753904,
      "epoch": 1.0201729106628241,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000490402303138048,
      "loss": 5.0655,
      "mean_token_accuracy": 0.21525616496801375,
      "num_tokens": 24340918.0,
      "step": 10620
    },
    {
      "entropy": 5.327084875106811,
      "epoch": 1.0206532180595582,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004903923943337836,
      "loss": 4.9725,
      "mean_token_accuracy": 0.2169102594256401,
      "num_tokens": 24352642.0,
      "step": 10625
    },
    {
      "entropy": 5.259642648696899,
      "epoch": 1.021133525456292,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004903824805287475,
      "loss": 4.8681,
      "mean_token_accuracy": 0.2313620164990425,
      "num_tokens": 24364874.0,
      "step": 10630
    },
    {
      "entropy": 5.25417685508728,
      "epoch": 1.021613832853026,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004903725617231696,
      "loss": 4.9809,
      "mean_token_accuracy": 0.220237572491169,
      "num_tokens": 24376566.0,
      "step": 10635
    },
    {
      "entropy": 5.289783382415772,
      "epoch": 1.0220941402497599,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004903626379172805,
      "loss": 4.8124,
      "mean_token_accuracy": 0.23486493080854415,
      "num_tokens": 24387283.0,
      "step": 10640
    },
    {
      "entropy": 5.338705682754517,
      "epoch": 1.0225744476464937,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004903527091113102,
      "loss": 5.0114,
      "mean_token_accuracy": 0.21624980717897416,
      "num_tokens": 24397970.0,
      "step": 10645
    },
    {
      "entropy": 5.268766736984253,
      "epoch": 1.0230547550432276,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004903427753054897,
      "loss": 4.9473,
      "mean_token_accuracy": 0.22086566388607026,
      "num_tokens": 24411223.0,
      "step": 10650
    },
    {
      "entropy": 5.327949380874633,
      "epoch": 1.0235350624399615,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004903328365000492,
      "loss": 4.9692,
      "mean_token_accuracy": 0.2191864550113678,
      "num_tokens": 24422117.0,
      "step": 10655
    },
    {
      "entropy": 5.318518495559692,
      "epoch": 1.0240153698366954,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004903228926952199,
      "loss": 4.9391,
      "mean_token_accuracy": 0.22112109363079072,
      "num_tokens": 24434142.0,
      "step": 10660
    },
    {
      "entropy": 5.278940296173095,
      "epoch": 1.0244956772334295,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004903129438912322,
      "loss": 4.8303,
      "mean_token_accuracy": 0.2299876794219017,
      "num_tokens": 24445295.0,
      "step": 10665
    },
    {
      "entropy": 5.241746234893799,
      "epoch": 1.0249759846301634,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004903029900883174,
      "loss": 4.9163,
      "mean_token_accuracy": 0.22946201115846634,
      "num_tokens": 24456092.0,
      "step": 10670
    },
    {
      "entropy": 5.2859790325164795,
      "epoch": 1.0254562920268973,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004902930312867063,
      "loss": 5.0143,
      "mean_token_accuracy": 0.21171343624591826,
      "num_tokens": 24467653.0,
      "step": 10675
    },
    {
      "entropy": 5.349200105667114,
      "epoch": 1.0259365994236311,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004902830674866306,
      "loss": 4.9573,
      "mean_token_accuracy": 0.22305092364549636,
      "num_tokens": 24479164.0,
      "step": 10680
    },
    {
      "entropy": 5.268773603439331,
      "epoch": 1.026416906820365,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004902730986883211,
      "loss": 4.8374,
      "mean_token_accuracy": 0.23050859570503235,
      "num_tokens": 24489785.0,
      "step": 10685
    },
    {
      "entropy": 5.2912109851837155,
      "epoch": 1.026897214217099,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004902631248920096,
      "loss": 4.9194,
      "mean_token_accuracy": 0.23076211661100388,
      "num_tokens": 24500158.0,
      "step": 10690
    },
    {
      "entropy": 5.346887731552124,
      "epoch": 1.0273775216138328,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004902531460979274,
      "loss": 5.031,
      "mean_token_accuracy": 0.21305167078971862,
      "num_tokens": 24512851.0,
      "step": 10695
    },
    {
      "entropy": 5.332522630691528,
      "epoch": 1.0278578290105667,
      "grad_norm": 1.0,
      "learning_rate": 0.0004902431623063065,
      "loss": 4.974,
      "mean_token_accuracy": 0.22368449866771697,
      "num_tokens": 24524016.0,
      "step": 10700
    },
    {
      "entropy": 5.220294666290283,
      "epoch": 1.0283381364073008,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004902331735173785,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22112229019403457,
      "num_tokens": 24536348.0,
      "step": 10705
    },
    {
      "entropy": 5.223771381378174,
      "epoch": 1.0288184438040346,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004902231797313752,
      "loss": 4.9284,
      "mean_token_accuracy": 0.22263671904802323,
      "num_tokens": 24548718.0,
      "step": 10710
    },
    {
      "entropy": 5.239733505249023,
      "epoch": 1.0292987512007685,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004902131809485288,
      "loss": 4.8367,
      "mean_token_accuracy": 0.23678794503211975,
      "num_tokens": 24560567.0,
      "step": 10715
    },
    {
      "entropy": 5.271087789535523,
      "epoch": 1.0297790585975024,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004902031771690713,
      "loss": 4.8692,
      "mean_token_accuracy": 0.22583483755588532,
      "num_tokens": 24572610.0,
      "step": 10720
    },
    {
      "entropy": 5.278970193862915,
      "epoch": 1.0302593659942363,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004901931683932352,
      "loss": 4.9166,
      "mean_token_accuracy": 0.22547948509454727,
      "num_tokens": 24584738.0,
      "step": 10725
    },
    {
      "entropy": 5.347322845458985,
      "epoch": 1.0307396733909702,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004901831546212526,
      "loss": 4.9705,
      "mean_token_accuracy": 0.2246183454990387,
      "num_tokens": 24596603.0,
      "step": 10730
    },
    {
      "entropy": 5.251077604293823,
      "epoch": 1.031219980787704,
      "grad_norm": 1.0,
      "learning_rate": 0.0004901731358533562,
      "loss": 4.8521,
      "mean_token_accuracy": 0.22551156729459762,
      "num_tokens": 24607061.0,
      "step": 10735
    },
    {
      "entropy": 5.220140171051026,
      "epoch": 1.031700288184438,
      "grad_norm": 0.875,
      "learning_rate": 0.0004901631120897785,
      "loss": 4.9258,
      "mean_token_accuracy": 0.22389309108257294,
      "num_tokens": 24619177.0,
      "step": 10740
    },
    {
      "entropy": 5.217163038253784,
      "epoch": 1.032180595581172,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004901530833307522,
      "loss": 4.8405,
      "mean_token_accuracy": 0.2244423046708107,
      "num_tokens": 24631336.0,
      "step": 10745
    },
    {
      "entropy": 5.268964052200317,
      "epoch": 1.032660902977906,
      "grad_norm": 1.0,
      "learning_rate": 0.0004901430495765103,
      "loss": 4.8229,
      "mean_token_accuracy": 0.23508888781070708,
      "num_tokens": 24641743.0,
      "step": 10750
    },
    {
      "entropy": 5.2789037227630615,
      "epoch": 1.0331412103746398,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004901330108272855,
      "loss": 4.8706,
      "mean_token_accuracy": 0.2270166367292404,
      "num_tokens": 24652318.0,
      "step": 10755
    },
    {
      "entropy": 5.240129184722901,
      "epoch": 1.0336215177713737,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004901229670833111,
      "loss": 4.8744,
      "mean_token_accuracy": 0.22638727128505706,
      "num_tokens": 24664129.0,
      "step": 10760
    },
    {
      "entropy": 5.324405145645142,
      "epoch": 1.0341018251681076,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004901129183448201,
      "loss": 4.9571,
      "mean_token_accuracy": 0.21583176106214524,
      "num_tokens": 24674921.0,
      "step": 10765
    },
    {
      "entropy": 5.251188087463379,
      "epoch": 1.0345821325648414,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004901028646120459,
      "loss": 4.9109,
      "mean_token_accuracy": 0.22119202166795732,
      "num_tokens": 24686052.0,
      "step": 10770
    },
    {
      "entropy": 5.24521164894104,
      "epoch": 1.0350624399615753,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000490092805885222,
      "loss": 4.8965,
      "mean_token_accuracy": 0.2218001216650009,
      "num_tokens": 24698625.0,
      "step": 10775
    },
    {
      "entropy": 5.260120153427124,
      "epoch": 1.0355427473583094,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004900827421645816,
      "loss": 4.8671,
      "mean_token_accuracy": 0.23538533449172974,
      "num_tokens": 24709322.0,
      "step": 10780
    },
    {
      "entropy": 5.217743396759033,
      "epoch": 1.0360230547550433,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004900726734503589,
      "loss": 4.8464,
      "mean_token_accuracy": 0.22876954972743987,
      "num_tokens": 24719512.0,
      "step": 10785
    },
    {
      "entropy": 5.238106822967529,
      "epoch": 1.0365033621517772,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004900625997427872,
      "loss": 4.8734,
      "mean_token_accuracy": 0.22652555257081985,
      "num_tokens": 24729947.0,
      "step": 10790
    },
    {
      "entropy": 5.199485349655151,
      "epoch": 1.036983669548511,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004900525210421006,
      "loss": 4.8768,
      "mean_token_accuracy": 0.23508709222078322,
      "num_tokens": 24741423.0,
      "step": 10795
    },
    {
      "entropy": 5.28390588760376,
      "epoch": 1.037463976945245,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004900424373485329,
      "loss": 4.9054,
      "mean_token_accuracy": 0.2212287589907646,
      "num_tokens": 24753403.0,
      "step": 10800
    },
    {
      "entropy": 5.351353311538697,
      "epoch": 1.0379442843419788,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004900323486623185,
      "loss": 5.0179,
      "mean_token_accuracy": 0.21868869364261628,
      "num_tokens": 24763660.0,
      "step": 10805
    },
    {
      "entropy": 5.29876184463501,
      "epoch": 1.0384245917387127,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004900222549836914,
      "loss": 4.919,
      "mean_token_accuracy": 0.22647754549980165,
      "num_tokens": 24775061.0,
      "step": 10810
    },
    {
      "entropy": 5.283719730377197,
      "epoch": 1.0389048991354466,
      "grad_norm": 1.0,
      "learning_rate": 0.000490012156312886,
      "loss": 4.9185,
      "mean_token_accuracy": 0.21527404338121414,
      "num_tokens": 24785248.0,
      "step": 10815
    },
    {
      "entropy": 5.225417757034302,
      "epoch": 1.0393852065321807,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004900020526501369,
      "loss": 4.8292,
      "mean_token_accuracy": 0.23071327805519104,
      "num_tokens": 24797024.0,
      "step": 10820
    },
    {
      "entropy": 5.379131650924682,
      "epoch": 1.0398655139289146,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004899919439956785,
      "loss": 5.0744,
      "mean_token_accuracy": 0.21304948329925538,
      "num_tokens": 24808500.0,
      "step": 10825
    },
    {
      "entropy": 5.373821496963501,
      "epoch": 1.0403458213256485,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004899818303497455,
      "loss": 5.0216,
      "mean_token_accuracy": 0.21425938457250596,
      "num_tokens": 24818805.0,
      "step": 10830
    },
    {
      "entropy": 5.332788419723511,
      "epoch": 1.0408261287223823,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004899717117125728,
      "loss": 4.9366,
      "mean_token_accuracy": 0.2206803485751152,
      "num_tokens": 24829247.0,
      "step": 10835
    },
    {
      "entropy": 5.200869131088257,
      "epoch": 1.0413064361191162,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004899615880843953,
      "loss": 4.8863,
      "mean_token_accuracy": 0.23112350404262544,
      "num_tokens": 24840139.0,
      "step": 10840
    },
    {
      "entropy": 5.300251054763794,
      "epoch": 1.04178674351585,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004899514594654481,
      "loss": 4.9826,
      "mean_token_accuracy": 0.2162712574005127,
      "num_tokens": 24851734.0,
      "step": 10845
    },
    {
      "entropy": 5.338700151443481,
      "epoch": 1.042267050912584,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004899413258559662,
      "loss": 4.939,
      "mean_token_accuracy": 0.2226698502898216,
      "num_tokens": 24863424.0,
      "step": 10850
    },
    {
      "entropy": 5.2999755382537845,
      "epoch": 1.0427473583093179,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004899311872561849,
      "loss": 4.964,
      "mean_token_accuracy": 0.2218521788716316,
      "num_tokens": 24875086.0,
      "step": 10855
    },
    {
      "entropy": 5.244442510604858,
      "epoch": 1.043227665706052,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004899210436663398,
      "loss": 4.8982,
      "mean_token_accuracy": 0.22415481507778168,
      "num_tokens": 24888408.0,
      "step": 10860
    },
    {
      "entropy": 5.21670184135437,
      "epoch": 1.0437079731027858,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004899108950866661,
      "loss": 4.8334,
      "mean_token_accuracy": 0.2331397995352745,
      "num_tokens": 24900357.0,
      "step": 10865
    },
    {
      "entropy": 5.273863363265991,
      "epoch": 1.0441882804995197,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004899007415173997,
      "loss": 4.9327,
      "mean_token_accuracy": 0.2236193984746933,
      "num_tokens": 24910790.0,
      "step": 10870
    },
    {
      "entropy": 5.28215446472168,
      "epoch": 1.0446685878962536,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004898905829587762,
      "loss": 4.9141,
      "mean_token_accuracy": 0.22417525798082352,
      "num_tokens": 24922124.0,
      "step": 10875
    },
    {
      "entropy": 5.204834413528443,
      "epoch": 1.0451488952929875,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004898804194110313,
      "loss": 4.8204,
      "mean_token_accuracy": 0.23332640677690505,
      "num_tokens": 24933591.0,
      "step": 10880
    },
    {
      "entropy": 5.265139770507813,
      "epoch": 1.0456292026897214,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004898702508744012,
      "loss": 4.8424,
      "mean_token_accuracy": 0.22899567186832429,
      "num_tokens": 24944708.0,
      "step": 10885
    },
    {
      "entropy": 5.240614509582519,
      "epoch": 1.0461095100864553,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004898600773491221,
      "loss": 4.8188,
      "mean_token_accuracy": 0.23244248926639557,
      "num_tokens": 24955966.0,
      "step": 10890
    },
    {
      "entropy": 5.3070580005645756,
      "epoch": 1.0465898174831891,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004898498988354297,
      "loss": 4.9266,
      "mean_token_accuracy": 0.22492922693490983,
      "num_tokens": 24967292.0,
      "step": 10895
    },
    {
      "entropy": 5.296743440628052,
      "epoch": 1.0470701248799232,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004898397153335608,
      "loss": 4.9808,
      "mean_token_accuracy": 0.22152854800224303,
      "num_tokens": 24977407.0,
      "step": 10900
    },
    {
      "entropy": 5.3793113231658936,
      "epoch": 1.0475504322766571,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004898295268437517,
      "loss": 5.0345,
      "mean_token_accuracy": 0.22074057757854462,
      "num_tokens": 24988804.0,
      "step": 10905
    },
    {
      "entropy": 5.308307313919068,
      "epoch": 1.048030739673391,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004898193333662388,
      "loss": 4.9787,
      "mean_token_accuracy": 0.2194735199213028,
      "num_tokens": 25000297.0,
      "step": 10910
    },
    {
      "entropy": 5.2945903778076175,
      "epoch": 1.0485110470701249,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004898091349012588,
      "loss": 4.908,
      "mean_token_accuracy": 0.2232706993818283,
      "num_tokens": 25012135.0,
      "step": 10915
    },
    {
      "entropy": 5.225506830215454,
      "epoch": 1.0489913544668588,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004897989314490486,
      "loss": 4.8842,
      "mean_token_accuracy": 0.23087289780378342,
      "num_tokens": 25023572.0,
      "step": 10920
    },
    {
      "entropy": 5.283700752258301,
      "epoch": 1.0494716618635926,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004897887230098451,
      "loss": 4.9383,
      "mean_token_accuracy": 0.22301181703805922,
      "num_tokens": 25035015.0,
      "step": 10925
    },
    {
      "entropy": 5.284937524795533,
      "epoch": 1.0499519692603265,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004897785095838852,
      "loss": 4.8943,
      "mean_token_accuracy": 0.2228178307414055,
      "num_tokens": 25045931.0,
      "step": 10930
    },
    {
      "entropy": 5.332826709747314,
      "epoch": 1.0504322766570606,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004897682911714061,
      "loss": 4.9287,
      "mean_token_accuracy": 0.2277660548686981,
      "num_tokens": 25056767.0,
      "step": 10935
    },
    {
      "entropy": 5.207847452163696,
      "epoch": 1.0509125840537945,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000489758067772645,
      "loss": 4.8822,
      "mean_token_accuracy": 0.23213577568531035,
      "num_tokens": 25068731.0,
      "step": 10940
    },
    {
      "entropy": 5.293945407867431,
      "epoch": 1.0513928914505284,
      "grad_norm": 0.875,
      "learning_rate": 0.0004897478393878392,
      "loss": 4.9579,
      "mean_token_accuracy": 0.21727697402238846,
      "num_tokens": 25081268.0,
      "step": 10945
    },
    {
      "entropy": 5.20956711769104,
      "epoch": 1.0518731988472623,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004897376060172264,
      "loss": 4.8674,
      "mean_token_accuracy": 0.2320805683732033,
      "num_tokens": 25093105.0,
      "step": 10950
    },
    {
      "entropy": 5.236131095886231,
      "epoch": 1.0523535062439962,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004897273676610438,
      "loss": 4.8731,
      "mean_token_accuracy": 0.23257286846637726,
      "num_tokens": 25103766.0,
      "step": 10955
    },
    {
      "entropy": 5.319604778289795,
      "epoch": 1.05283381364073,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004897171243195295,
      "loss": 5.0151,
      "mean_token_accuracy": 0.22169419080018998,
      "num_tokens": 25115675.0,
      "step": 10960
    },
    {
      "entropy": 5.293093729019165,
      "epoch": 1.053314121037464,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000489706875992921,
      "loss": 4.8743,
      "mean_token_accuracy": 0.2256014570593834,
      "num_tokens": 25127907.0,
      "step": 10965
    },
    {
      "entropy": 5.287135791778565,
      "epoch": 1.0537944284341978,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004896966226814565,
      "loss": 4.9054,
      "mean_token_accuracy": 0.2241995304822922,
      "num_tokens": 25139675.0,
      "step": 10970
    },
    {
      "entropy": 5.358683776855469,
      "epoch": 1.054274735830932,
      "grad_norm": 1.0,
      "learning_rate": 0.0004896863643853739,
      "loss": 5.0216,
      "mean_token_accuracy": 0.22017746269702912,
      "num_tokens": 25150960.0,
      "step": 10975
    },
    {
      "entropy": 5.343587827682495,
      "epoch": 1.0547550432276658,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004896761011049114,
      "loss": 4.9979,
      "mean_token_accuracy": 0.2200980231165886,
      "num_tokens": 25163676.0,
      "step": 10980
    },
    {
      "entropy": 5.233455419540405,
      "epoch": 1.0552353506243997,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004896658328403074,
      "loss": 4.9248,
      "mean_token_accuracy": 0.22182402163743972,
      "num_tokens": 25174317.0,
      "step": 10985
    },
    {
      "entropy": 5.242417049407959,
      "epoch": 1.0557156580211335,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004896555595918001,
      "loss": 4.9349,
      "mean_token_accuracy": 0.22319017648696898,
      "num_tokens": 25186585.0,
      "step": 10990
    },
    {
      "entropy": 5.311861848831176,
      "epoch": 1.0561959654178674,
      "grad_norm": 1.125,
      "learning_rate": 0.0004896452813596281,
      "loss": 4.9506,
      "mean_token_accuracy": 0.22000137716531754,
      "num_tokens": 25199014.0,
      "step": 10995
    },
    {
      "entropy": 5.345607757568359,
      "epoch": 1.0566762728146013,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004896349981440301,
      "loss": 5.0329,
      "mean_token_accuracy": 0.21464401334524155,
      "num_tokens": 25210544.0,
      "step": 11000
    },
    {
      "entropy": 5.299417638778687,
      "epoch": 1.0571565802113352,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004896247099452447,
      "loss": 4.9293,
      "mean_token_accuracy": 0.2301488548517227,
      "num_tokens": 25221583.0,
      "step": 11005
    },
    {
      "entropy": 5.326053762435913,
      "epoch": 1.057636887608069,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004896144167635108,
      "loss": 4.9051,
      "mean_token_accuracy": 0.22906794399023056,
      "num_tokens": 25231724.0,
      "step": 11010
    },
    {
      "entropy": 5.2919032096862795,
      "epoch": 1.0581171950048032,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004896041185990675,
      "loss": 5.0082,
      "mean_token_accuracy": 0.21718186438083648,
      "num_tokens": 25243021.0,
      "step": 11015
    },
    {
      "entropy": 5.325321531295776,
      "epoch": 1.058597502401537,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004895938154521538,
      "loss": 4.986,
      "mean_token_accuracy": 0.22029063552618028,
      "num_tokens": 25254189.0,
      "step": 11020
    },
    {
      "entropy": 5.360890436172485,
      "epoch": 1.059077809798271,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004895835073230089,
      "loss": 5.0106,
      "mean_token_accuracy": 0.2243978813290596,
      "num_tokens": 25265556.0,
      "step": 11025
    },
    {
      "entropy": 5.211797285079956,
      "epoch": 1.0595581171950048,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004895731942118722,
      "loss": 4.8478,
      "mean_token_accuracy": 0.2227287083864212,
      "num_tokens": 25276789.0,
      "step": 11030
    },
    {
      "entropy": 5.289017152786255,
      "epoch": 1.0600384245917387,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004895628761189829,
      "loss": 5.0134,
      "mean_token_accuracy": 0.21681419312953948,
      "num_tokens": 25288505.0,
      "step": 11035
    },
    {
      "entropy": 5.251034164428711,
      "epoch": 1.0605187319884726,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004895525530445809,
      "loss": 4.8577,
      "mean_token_accuracy": 0.2378873959183693,
      "num_tokens": 25301490.0,
      "step": 11040
    },
    {
      "entropy": 5.277717924118042,
      "epoch": 1.0609990393852065,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004895422249889057,
      "loss": 4.9531,
      "mean_token_accuracy": 0.22043091505765916,
      "num_tokens": 25313303.0,
      "step": 11045
    },
    {
      "entropy": 5.35553879737854,
      "epoch": 1.0614793467819403,
      "grad_norm": 1.0,
      "learning_rate": 0.0004895318919521971,
      "loss": 5.0131,
      "mean_token_accuracy": 0.21087596118450164,
      "num_tokens": 25324379.0,
      "step": 11050
    },
    {
      "entropy": 5.227749395370483,
      "epoch": 1.0619596541786744,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004895215539346949,
      "loss": 4.854,
      "mean_token_accuracy": 0.22884634584188462,
      "num_tokens": 25335834.0,
      "step": 11055
    },
    {
      "entropy": 5.256495094299316,
      "epoch": 1.0624399615754083,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004895112109366393,
      "loss": 4.9143,
      "mean_token_accuracy": 0.2239677742123604,
      "num_tokens": 25347591.0,
      "step": 11060
    },
    {
      "entropy": 5.286858415603637,
      "epoch": 1.0629202689721422,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004895008629582703,
      "loss": 4.901,
      "mean_token_accuracy": 0.22555171549320222,
      "num_tokens": 25358483.0,
      "step": 11065
    },
    {
      "entropy": 5.255856227874756,
      "epoch": 1.063400576368876,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004894905099998283,
      "loss": 4.9136,
      "mean_token_accuracy": 0.22227107584476472,
      "num_tokens": 25369434.0,
      "step": 11070
    },
    {
      "entropy": 5.286383056640625,
      "epoch": 1.06388088376561,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004894801520615535,
      "loss": 4.9164,
      "mean_token_accuracy": 0.22479838728904725,
      "num_tokens": 25381007.0,
      "step": 11075
    },
    {
      "entropy": 5.275960683822632,
      "epoch": 1.0643611911623438,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004894697891436863,
      "loss": 4.9175,
      "mean_token_accuracy": 0.22316204756498337,
      "num_tokens": 25393809.0,
      "step": 11080
    },
    {
      "entropy": 5.322711420059204,
      "epoch": 1.0648414985590777,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004894594212464676,
      "loss": 4.9383,
      "mean_token_accuracy": 0.2315184861421585,
      "num_tokens": 25404967.0,
      "step": 11085
    },
    {
      "entropy": 5.289861345291138,
      "epoch": 1.0653218059558118,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004894490483701381,
      "loss": 4.89,
      "mean_token_accuracy": 0.2353300780057907,
      "num_tokens": 25417092.0,
      "step": 11090
    },
    {
      "entropy": 5.3253110408782955,
      "epoch": 1.0658021133525457,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004894386705149382,
      "loss": 4.9852,
      "mean_token_accuracy": 0.21821215748786926,
      "num_tokens": 25428425.0,
      "step": 11095
    },
    {
      "entropy": 5.260989809036255,
      "epoch": 1.0662824207492796,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004894282876811093,
      "loss": 4.8228,
      "mean_token_accuracy": 0.23943477869033813,
      "num_tokens": 25440134.0,
      "step": 11100
    },
    {
      "entropy": 5.2422929286956785,
      "epoch": 1.0667627281460135,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004894178998688921,
      "loss": 4.9046,
      "mean_token_accuracy": 0.22531140595674515,
      "num_tokens": 25452222.0,
      "step": 11105
    },
    {
      "entropy": 5.238423538208008,
      "epoch": 1.0672430355427474,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004894075070785281,
      "loss": 4.9543,
      "mean_token_accuracy": 0.22202043384313583,
      "num_tokens": 25464541.0,
      "step": 11110
    },
    {
      "entropy": 5.3626092910766605,
      "epoch": 1.0677233429394812,
      "grad_norm": 1.0,
      "learning_rate": 0.0004893971093102585,
      "loss": 4.9415,
      "mean_token_accuracy": 0.21499006003141402,
      "num_tokens": 25476537.0,
      "step": 11115
    },
    {
      "entropy": 5.377402496337891,
      "epoch": 1.0682036503362151,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004893867065643245,
      "loss": 5.0029,
      "mean_token_accuracy": 0.2135973557829857,
      "num_tokens": 25486737.0,
      "step": 11120
    },
    {
      "entropy": 5.2622960090637205,
      "epoch": 1.068683957732949,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004893762988409678,
      "loss": 4.9332,
      "mean_token_accuracy": 0.22087916880846023,
      "num_tokens": 25497278.0,
      "step": 11125
    },
    {
      "entropy": 5.275923252105713,
      "epoch": 1.069164265129683,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004893658861404301,
      "loss": 4.9136,
      "mean_token_accuracy": 0.21977146416902543,
      "num_tokens": 25508716.0,
      "step": 11130
    },
    {
      "entropy": 5.369044399261474,
      "epoch": 1.069644572526417,
      "grad_norm": 1.0,
      "learning_rate": 0.0004893554684629529,
      "loss": 4.997,
      "mean_token_accuracy": 0.21598267257213594,
      "num_tokens": 25519439.0,
      "step": 11135
    },
    {
      "entropy": 5.245421981811523,
      "epoch": 1.0701248799231509,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004893450458087784,
      "loss": 4.8045,
      "mean_token_accuracy": 0.2322712168097496,
      "num_tokens": 25530911.0,
      "step": 11140
    },
    {
      "entropy": 5.307769632339477,
      "epoch": 1.0706051873198847,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004893346181781483,
      "loss": 5.0352,
      "mean_token_accuracy": 0.21192309260368347,
      "num_tokens": 25542452.0,
      "step": 11145
    },
    {
      "entropy": 5.302496719360351,
      "epoch": 1.0710854947166186,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004893241855713048,
      "loss": 4.9465,
      "mean_token_accuracy": 0.21189783960580827,
      "num_tokens": 25554105.0,
      "step": 11150
    },
    {
      "entropy": 5.424971342086792,
      "epoch": 1.0715658021133525,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004893137479884903,
      "loss": 5.0056,
      "mean_token_accuracy": 0.22016999274492263,
      "num_tokens": 25564806.0,
      "step": 11155
    },
    {
      "entropy": 5.2566863059997555,
      "epoch": 1.0720461095100864,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004893033054299468,
      "loss": 4.8947,
      "mean_token_accuracy": 0.22889358699321746,
      "num_tokens": 25575664.0,
      "step": 11160
    },
    {
      "entropy": 5.229598426818848,
      "epoch": 1.0725264169068203,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000489292857895917,
      "loss": 4.8275,
      "mean_token_accuracy": 0.23152268975973128,
      "num_tokens": 25586227.0,
      "step": 11165
    },
    {
      "entropy": 5.32275424003601,
      "epoch": 1.0730067243035544,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004892824053866432,
      "loss": 5.0001,
      "mean_token_accuracy": 0.21985238492488862,
      "num_tokens": 25597475.0,
      "step": 11170
    },
    {
      "entropy": 5.24053373336792,
      "epoch": 1.0734870317002883,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004892719479023683,
      "loss": 4.8813,
      "mean_token_accuracy": 0.2295086681842804,
      "num_tokens": 25608098.0,
      "step": 11175
    },
    {
      "entropy": 5.334474992752075,
      "epoch": 1.0739673390970221,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000489261485443335,
      "loss": 4.9879,
      "mean_token_accuracy": 0.21846836656332017,
      "num_tokens": 25620053.0,
      "step": 11180
    },
    {
      "entropy": 5.326720237731934,
      "epoch": 1.074447646493756,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004892510180097863,
      "loss": 4.9078,
      "mean_token_accuracy": 0.21928017139434813,
      "num_tokens": 25630534.0,
      "step": 11185
    },
    {
      "entropy": 5.331039953231811,
      "epoch": 1.07492795389049,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004892405456019651,
      "loss": 4.9562,
      "mean_token_accuracy": 0.2267116293311119,
      "num_tokens": 25641413.0,
      "step": 11190
    },
    {
      "entropy": 5.2167181968688965,
      "epoch": 1.0754082612872238,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004892300682201147,
      "loss": 4.8733,
      "mean_token_accuracy": 0.23322367072105407,
      "num_tokens": 25652081.0,
      "step": 11195
    },
    {
      "entropy": 5.3137329578399655,
      "epoch": 1.0758885686839577,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004892195858644782,
      "loss": 4.9692,
      "mean_token_accuracy": 0.22998296469449997,
      "num_tokens": 25664282.0,
      "step": 11200
    },
    {
      "entropy": 5.345920991897583,
      "epoch": 1.0763688760806915,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000489209098535299,
      "loss": 5.0395,
      "mean_token_accuracy": 0.22130452394485473,
      "num_tokens": 25675310.0,
      "step": 11205
    },
    {
      "entropy": 5.385965394973755,
      "epoch": 1.0768491834774256,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004891986062328205,
      "loss": 5.0212,
      "mean_token_accuracy": 0.22202289551496507,
      "num_tokens": 25686895.0,
      "step": 11210
    },
    {
      "entropy": 5.277755212783814,
      "epoch": 1.0773294908741595,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004891881089572865,
      "loss": 4.8269,
      "mean_token_accuracy": 0.22484762370586395,
      "num_tokens": 25697778.0,
      "step": 11215
    },
    {
      "entropy": 5.270056295394897,
      "epoch": 1.0778097982708934,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004891776067089406,
      "loss": 4.9765,
      "mean_token_accuracy": 0.21236666887998581,
      "num_tokens": 25708602.0,
      "step": 11220
    },
    {
      "entropy": 5.260102272033691,
      "epoch": 1.0782901056676273,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004891670994880266,
      "loss": 4.8626,
      "mean_token_accuracy": 0.2270296648144722,
      "num_tokens": 25719671.0,
      "step": 11225
    },
    {
      "entropy": 5.339909791946411,
      "epoch": 1.0787704130643612,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004891565872947888,
      "loss": 5.0351,
      "mean_token_accuracy": 0.21836673021316527,
      "num_tokens": 25731797.0,
      "step": 11230
    },
    {
      "entropy": 5.25717077255249,
      "epoch": 1.079250720461095,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004891460701294706,
      "loss": 4.894,
      "mean_token_accuracy": 0.22847594171762467,
      "num_tokens": 25743984.0,
      "step": 11235
    },
    {
      "entropy": 5.240834045410156,
      "epoch": 1.079731027857829,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004891355479923167,
      "loss": 4.9069,
      "mean_token_accuracy": 0.22459755688905716,
      "num_tokens": 25755252.0,
      "step": 11240
    },
    {
      "entropy": 5.22453818321228,
      "epoch": 1.080211335254563,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004891250208835712,
      "loss": 4.8201,
      "mean_token_accuracy": 0.2254481941461563,
      "num_tokens": 25765715.0,
      "step": 11245
    },
    {
      "entropy": 5.34910740852356,
      "epoch": 1.080691642651297,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004891144888034784,
      "loss": 4.9508,
      "mean_token_accuracy": 0.22036246657371522,
      "num_tokens": 25777866.0,
      "step": 11250
    },
    {
      "entropy": 5.287067890167236,
      "epoch": 1.0811719500480308,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004891039517522832,
      "loss": 4.8614,
      "mean_token_accuracy": 0.2264195367693901,
      "num_tokens": 25791199.0,
      "step": 11255
    },
    {
      "entropy": 5.190082263946533,
      "epoch": 1.0816522574447647,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004890934097302299,
      "loss": 4.8217,
      "mean_token_accuracy": 0.23926398009061814,
      "num_tokens": 25802979.0,
      "step": 11260
    },
    {
      "entropy": 5.152024602890014,
      "epoch": 1.0821325648414986,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004890828627375632,
      "loss": 4.8549,
      "mean_token_accuracy": 0.22393021434545518,
      "num_tokens": 25814696.0,
      "step": 11265
    },
    {
      "entropy": 5.3096537590026855,
      "epoch": 1.0826128722382324,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004890723107745283,
      "loss": 4.959,
      "mean_token_accuracy": 0.21532270908355713,
      "num_tokens": 25825376.0,
      "step": 11270
    },
    {
      "entropy": 5.332683372497558,
      "epoch": 1.0830931796349663,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004890617538413699,
      "loss": 4.9061,
      "mean_token_accuracy": 0.22685956209897995,
      "num_tokens": 25835491.0,
      "step": 11275
    },
    {
      "entropy": 5.293481636047363,
      "epoch": 1.0835734870317002,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004890511919383333,
      "loss": 4.9689,
      "mean_token_accuracy": 0.21982008814811707,
      "num_tokens": 25848154.0,
      "step": 11280
    },
    {
      "entropy": 5.280985593795776,
      "epoch": 1.084053794428434,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004890406250656636,
      "loss": 4.9333,
      "mean_token_accuracy": 0.22627804130315782,
      "num_tokens": 25859471.0,
      "step": 11285
    },
    {
      "entropy": 5.252989053726196,
      "epoch": 1.0845341018251682,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004890300532236062,
      "loss": 4.8706,
      "mean_token_accuracy": 0.22963903695344925,
      "num_tokens": 25869460.0,
      "step": 11290
    },
    {
      "entropy": 5.293918228149414,
      "epoch": 1.085014409221902,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004890194764124064,
      "loss": 4.8859,
      "mean_token_accuracy": 0.22546039372682572,
      "num_tokens": 25881892.0,
      "step": 11295
    },
    {
      "entropy": 5.314129877090454,
      "epoch": 1.085494716618636,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004890088946323099,
      "loss": 5.0051,
      "mean_token_accuracy": 0.21368849277496338,
      "num_tokens": 25893774.0,
      "step": 11300
    },
    {
      "entropy": 5.213082981109619,
      "epoch": 1.0859750240153698,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004889983078835623,
      "loss": 4.8255,
      "mean_token_accuracy": 0.22466489374637605,
      "num_tokens": 25904758.0,
      "step": 11305
    },
    {
      "entropy": 5.224869012832642,
      "epoch": 1.0864553314121037,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004889877161664096,
      "loss": 4.9348,
      "mean_token_accuracy": 0.21893048286437988,
      "num_tokens": 25917013.0,
      "step": 11310
    },
    {
      "entropy": 5.389647436141968,
      "epoch": 1.0869356388088376,
      "grad_norm": 1.0,
      "learning_rate": 0.0004889771194810974,
      "loss": 4.9859,
      "mean_token_accuracy": 0.21903414875268937,
      "num_tokens": 25927780.0,
      "step": 11315
    },
    {
      "entropy": 5.227621746063233,
      "epoch": 1.0874159462055715,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004889665178278719,
      "loss": 4.8885,
      "mean_token_accuracy": 0.22977528870105743,
      "num_tokens": 25939339.0,
      "step": 11320
    },
    {
      "entropy": 5.244958591461182,
      "epoch": 1.0878962536023056,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004889559112069792,
      "loss": 4.8731,
      "mean_token_accuracy": 0.227946874499321,
      "num_tokens": 25950440.0,
      "step": 11325
    },
    {
      "entropy": 5.282286834716797,
      "epoch": 1.0883765609990395,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004889452996186657,
      "loss": 4.9159,
      "mean_token_accuracy": 0.21897949278354645,
      "num_tokens": 25962849.0,
      "step": 11330
    },
    {
      "entropy": 5.278962326049805,
      "epoch": 1.0888568683957733,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004889346830631774,
      "loss": 4.9532,
      "mean_token_accuracy": 0.22255521267652512,
      "num_tokens": 25973616.0,
      "step": 11335
    },
    {
      "entropy": 5.252470016479492,
      "epoch": 1.0893371757925072,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000488924061540761,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2288068115711212,
      "num_tokens": 25984727.0,
      "step": 11340
    },
    {
      "entropy": 5.3238630294799805,
      "epoch": 1.089817483189241,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004889134350516633,
      "loss": 4.9593,
      "mean_token_accuracy": 0.2259926065802574,
      "num_tokens": 25996431.0,
      "step": 11345
    },
    {
      "entropy": 5.293342542648316,
      "epoch": 1.090297790585975,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004889028035961308,
      "loss": 4.9168,
      "mean_token_accuracy": 0.2246268644928932,
      "num_tokens": 26008936.0,
      "step": 11350
    },
    {
      "entropy": 5.336949682235717,
      "epoch": 1.0907780979827089,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004888921671744103,
      "loss": 4.9744,
      "mean_token_accuracy": 0.21497151404619216,
      "num_tokens": 26019308.0,
      "step": 11355
    },
    {
      "entropy": 5.306262826919555,
      "epoch": 1.0912584053794427,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004888815257867488,
      "loss": 4.8761,
      "mean_token_accuracy": 0.23164073675870894,
      "num_tokens": 26030705.0,
      "step": 11360
    },
    {
      "entropy": 5.2742046356201175,
      "epoch": 1.0917387127761768,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004888708794333934,
      "loss": 4.8991,
      "mean_token_accuracy": 0.22692825198173522,
      "num_tokens": 26042759.0,
      "step": 11365
    },
    {
      "entropy": 5.274525928497314,
      "epoch": 1.0922190201729107,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004888602281145913,
      "loss": 4.9453,
      "mean_token_accuracy": 0.2244431659579277,
      "num_tokens": 26054719.0,
      "step": 11370
    },
    {
      "entropy": 5.26015796661377,
      "epoch": 1.0926993275696446,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004888495718305897,
      "loss": 4.9509,
      "mean_token_accuracy": 0.21566283702850342,
      "num_tokens": 26065765.0,
      "step": 11375
    },
    {
      "entropy": 5.16221513748169,
      "epoch": 1.0931796349663785,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000488838910581636,
      "loss": 4.8471,
      "mean_token_accuracy": 0.22685923129320146,
      "num_tokens": 26077719.0,
      "step": 11380
    },
    {
      "entropy": 5.335711812973022,
      "epoch": 1.0936599423631124,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004888282443679777,
      "loss": 4.9868,
      "mean_token_accuracy": 0.2173221752047539,
      "num_tokens": 26089924.0,
      "step": 11385
    },
    {
      "entropy": 5.4440351009368895,
      "epoch": 1.0941402497598463,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004888175731898627,
      "loss": 5.054,
      "mean_token_accuracy": 0.21265479922294617,
      "num_tokens": 26100312.0,
      "step": 11390
    },
    {
      "entropy": 5.2555005073547365,
      "epoch": 1.0946205571565801,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004888068970475384,
      "loss": 4.9497,
      "mean_token_accuracy": 0.22392383813858033,
      "num_tokens": 26111932.0,
      "step": 11395
    },
    {
      "entropy": 5.267872285842896,
      "epoch": 1.0951008645533142,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004887962159412529,
      "loss": 4.8638,
      "mean_token_accuracy": 0.22638123333454133,
      "num_tokens": 26123989.0,
      "step": 11400
    },
    {
      "entropy": 5.353457689285278,
      "epoch": 1.0955811719500481,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004887855298712541,
      "loss": 4.9658,
      "mean_token_accuracy": 0.22385124564170839,
      "num_tokens": 26135589.0,
      "step": 11405
    },
    {
      "entropy": 5.180663681030273,
      "epoch": 1.096061479346782,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048877483883779,
      "loss": 4.8837,
      "mean_token_accuracy": 0.2282148852944374,
      "num_tokens": 26147069.0,
      "step": 11410
    },
    {
      "entropy": 5.233529758453369,
      "epoch": 1.0965417867435159,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004887641428411091,
      "loss": 4.9292,
      "mean_token_accuracy": 0.2226127043366432,
      "num_tokens": 26159331.0,
      "step": 11415
    },
    {
      "entropy": 5.303081274032593,
      "epoch": 1.0970220941402498,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004887534418814595,
      "loss": 4.9337,
      "mean_token_accuracy": 0.2245740696787834,
      "num_tokens": 26169863.0,
      "step": 11420
    },
    {
      "entropy": 5.286851596832276,
      "epoch": 1.0975024015369836,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004887427359590897,
      "loss": 4.9072,
      "mean_token_accuracy": 0.22017393559217452,
      "num_tokens": 26182888.0,
      "step": 11425
    },
    {
      "entropy": 5.267059326171875,
      "epoch": 1.0979827089337175,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004887320250742482,
      "loss": 4.9101,
      "mean_token_accuracy": 0.2294606864452362,
      "num_tokens": 26194979.0,
      "step": 11430
    },
    {
      "entropy": 5.225768804550171,
      "epoch": 1.0984630163304514,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004887213092271838,
      "loss": 4.9153,
      "mean_token_accuracy": 0.23525336235761643,
      "num_tokens": 26207309.0,
      "step": 11435
    },
    {
      "entropy": 5.255485010147095,
      "epoch": 1.0989433237271853,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004887105884181451,
      "loss": 4.938,
      "mean_token_accuracy": 0.2270049422979355,
      "num_tokens": 26219231.0,
      "step": 11440
    },
    {
      "entropy": 5.349270343780518,
      "epoch": 1.0994236311239194,
      "grad_norm": 1.0,
      "learning_rate": 0.0004886998626473813,
      "loss": 4.9697,
      "mean_token_accuracy": 0.22255009710788726,
      "num_tokens": 26229355.0,
      "step": 11445
    },
    {
      "entropy": 5.315190267562866,
      "epoch": 1.0999039385206533,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004886891319151411,
      "loss": 4.8609,
      "mean_token_accuracy": 0.22966767996549606,
      "num_tokens": 26239069.0,
      "step": 11450
    },
    {
      "entropy": 5.30435881614685,
      "epoch": 1.1003842459173871,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004886783962216738,
      "loss": 5.0077,
      "mean_token_accuracy": 0.21714320927858352,
      "num_tokens": 26250403.0,
      "step": 11455
    },
    {
      "entropy": 5.322808933258057,
      "epoch": 1.100864553314121,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004886676555672287,
      "loss": 4.893,
      "mean_token_accuracy": 0.2236496776342392,
      "num_tokens": 26262926.0,
      "step": 11460
    },
    {
      "entropy": 5.38580732345581,
      "epoch": 1.101344860710855,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004886569099520551,
      "loss": 5.0414,
      "mean_token_accuracy": 0.21133878380060195,
      "num_tokens": 26274030.0,
      "step": 11465
    },
    {
      "entropy": 5.227356958389282,
      "epoch": 1.1018251681075888,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004886461593764024,
      "loss": 4.9185,
      "mean_token_accuracy": 0.2280497133731842,
      "num_tokens": 26284799.0,
      "step": 11470
    },
    {
      "entropy": 5.152918529510498,
      "epoch": 1.1023054755043227,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004886354038405204,
      "loss": 4.8033,
      "mean_token_accuracy": 0.23828006088733672,
      "num_tokens": 26295154.0,
      "step": 11475
    },
    {
      "entropy": 5.330612087249756,
      "epoch": 1.1027857829010568,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004886246433446586,
      "loss": 4.9505,
      "mean_token_accuracy": 0.21783934384584427,
      "num_tokens": 26306181.0,
      "step": 11480
    },
    {
      "entropy": 5.401623392105103,
      "epoch": 1.1032660902977907,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004886138778890669,
      "loss": 5.0358,
      "mean_token_accuracy": 0.21328163594007493,
      "num_tokens": 26318674.0,
      "step": 11485
    },
    {
      "entropy": 5.346351861953735,
      "epoch": 1.1037463976945245,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004886031074739953,
      "loss": 4.9594,
      "mean_token_accuracy": 0.21103494614362717,
      "num_tokens": 26330257.0,
      "step": 11490
    },
    {
      "entropy": 5.294247055053711,
      "epoch": 1.1042267050912584,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004885923320996938,
      "loss": 4.8954,
      "mean_token_accuracy": 0.2256893113255501,
      "num_tokens": 26342035.0,
      "step": 11495
    },
    {
      "entropy": 5.306338310241699,
      "epoch": 1.1047070124879923,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004885815517664127,
      "loss": 4.9356,
      "mean_token_accuracy": 0.22292064130306244,
      "num_tokens": 26352703.0,
      "step": 11500
    },
    {
      "entropy": 5.344753646850586,
      "epoch": 1.1051873198847262,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000488570766474402,
      "loss": 4.9464,
      "mean_token_accuracy": 0.22787559032440186,
      "num_tokens": 26362324.0,
      "step": 11505
    },
    {
      "entropy": 5.314879417419434,
      "epoch": 1.10566762728146,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004885599762239124,
      "loss": 4.9675,
      "mean_token_accuracy": 0.21332177966833116,
      "num_tokens": 26373540.0,
      "step": 11510
    },
    {
      "entropy": 5.261144733428955,
      "epoch": 1.106147934678194,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004885491810151943,
      "loss": 4.9468,
      "mean_token_accuracy": 0.22136803716421127,
      "num_tokens": 26384259.0,
      "step": 11515
    },
    {
      "entropy": 5.229436159133911,
      "epoch": 1.106628242074928,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004885383808484982,
      "loss": 4.9121,
      "mean_token_accuracy": 0.22861432731151582,
      "num_tokens": 26395193.0,
      "step": 11520
    },
    {
      "entropy": 5.24059042930603,
      "epoch": 1.107108549471662,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004885275757240751,
      "loss": 4.861,
      "mean_token_accuracy": 0.22366121113300325,
      "num_tokens": 26408556.0,
      "step": 11525
    },
    {
      "entropy": 5.261339473724365,
      "epoch": 1.1075888568683958,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004885167656421757,
      "loss": 4.939,
      "mean_token_accuracy": 0.22771972864866258,
      "num_tokens": 26420066.0,
      "step": 11530
    },
    {
      "entropy": 5.254054212570191,
      "epoch": 1.1080691642651297,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000488505950603051,
      "loss": 4.9025,
      "mean_token_accuracy": 0.22542301714420318,
      "num_tokens": 26432533.0,
      "step": 11535
    },
    {
      "entropy": 5.243907499313354,
      "epoch": 1.1085494716618636,
      "grad_norm": 1.078125,
      "learning_rate": 0.000488495130606952,
      "loss": 4.8964,
      "mean_token_accuracy": 0.22167578041553498,
      "num_tokens": 26443878.0,
      "step": 11540
    },
    {
      "entropy": 5.291236448287964,
      "epoch": 1.1090297790585975,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004884843056541302,
      "loss": 4.9075,
      "mean_token_accuracy": 0.22179460972547532,
      "num_tokens": 26455111.0,
      "step": 11545
    },
    {
      "entropy": 5.25293550491333,
      "epoch": 1.1095100864553313,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004884734757448367,
      "loss": 4.9347,
      "mean_token_accuracy": 0.22688528001308442,
      "num_tokens": 26466577.0,
      "step": 11550
    },
    {
      "entropy": 5.25384521484375,
      "epoch": 1.1099903938520654,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000488462640879323,
      "loss": 4.855,
      "mean_token_accuracy": 0.22806688398122787,
      "num_tokens": 26479637.0,
      "step": 11555
    },
    {
      "entropy": 5.247193241119385,
      "epoch": 1.1104707012487993,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004884518010578405,
      "loss": 4.8874,
      "mean_token_accuracy": 0.22609785795211793,
      "num_tokens": 26489923.0,
      "step": 11560
    },
    {
      "entropy": 5.331763458251953,
      "epoch": 1.1109510086455332,
      "grad_norm": 1.0,
      "learning_rate": 0.0004884409562806411,
      "loss": 4.9949,
      "mean_token_accuracy": 0.2185052290558815,
      "num_tokens": 26501236.0,
      "step": 11565
    },
    {
      "entropy": 5.284239339828491,
      "epoch": 1.111431316042267,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004884301065479765,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22361597120761872,
      "num_tokens": 26511509.0,
      "step": 11570
    },
    {
      "entropy": 5.2041707038879395,
      "epoch": 1.111911623439001,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004884192518600986,
      "loss": 4.8344,
      "mean_token_accuracy": 0.23340863436460496,
      "num_tokens": 26520931.0,
      "step": 11575
    },
    {
      "entropy": 5.20103178024292,
      "epoch": 1.1123919308357348,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004884083922172593,
      "loss": 4.8436,
      "mean_token_accuracy": 0.23373724818229674,
      "num_tokens": 26531333.0,
      "step": 11580
    },
    {
      "entropy": 5.205853462219238,
      "epoch": 1.1128722382324687,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004883975276197108,
      "loss": 4.9108,
      "mean_token_accuracy": 0.22486654669046402,
      "num_tokens": 26543696.0,
      "step": 11585
    },
    {
      "entropy": 5.277018785476685,
      "epoch": 1.1133525456292026,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004883866580677055,
      "loss": 4.8691,
      "mean_token_accuracy": 0.23109534680843352,
      "num_tokens": 26556292.0,
      "step": 11590
    },
    {
      "entropy": 5.3012731075286865,
      "epoch": 1.1138328530259365,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004883757835614956,
      "loss": 4.8974,
      "mean_token_accuracy": 0.21795395016670227,
      "num_tokens": 26568023.0,
      "step": 11595
    },
    {
      "entropy": 5.219055986404419,
      "epoch": 1.1143131604226706,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004883649041013335,
      "loss": 4.8853,
      "mean_token_accuracy": 0.23042893558740615,
      "num_tokens": 26579469.0,
      "step": 11600
    },
    {
      "entropy": 5.338874673843383,
      "epoch": 1.1147934678194045,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000488354019687472,
      "loss": 4.9401,
      "mean_token_accuracy": 0.21997726410627366,
      "num_tokens": 26591592.0,
      "step": 11605
    },
    {
      "entropy": 5.305851316452026,
      "epoch": 1.1152737752161384,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004883431303201636,
      "loss": 4.9032,
      "mean_token_accuracy": 0.22865972369909288,
      "num_tokens": 26604075.0,
      "step": 11610
    },
    {
      "entropy": 5.176166248321533,
      "epoch": 1.1157540826128722,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004883322359996613,
      "loss": 4.821,
      "mean_token_accuracy": 0.2349545180797577,
      "num_tokens": 26616448.0,
      "step": 11615
    },
    {
      "entropy": 5.268229198455811,
      "epoch": 1.1162343900096061,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004883213367262179,
      "loss": 4.9146,
      "mean_token_accuracy": 0.23280704468488694,
      "num_tokens": 26628749.0,
      "step": 11620
    },
    {
      "entropy": 5.268984746932984,
      "epoch": 1.11671469740634,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004883104325000866,
      "loss": 4.863,
      "mean_token_accuracy": 0.232693350315094,
      "num_tokens": 26640163.0,
      "step": 11625
    },
    {
      "entropy": 5.294044399261475,
      "epoch": 1.1171950048030739,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004882995233215203,
      "loss": 4.9195,
      "mean_token_accuracy": 0.2248235136270523,
      "num_tokens": 26650729.0,
      "step": 11630
    },
    {
      "entropy": 5.245526266098023,
      "epoch": 1.117675312199808,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004882886091907726,
      "loss": 4.9424,
      "mean_token_accuracy": 0.22403739392757416,
      "num_tokens": 26661995.0,
      "step": 11635
    },
    {
      "entropy": 5.233623123168945,
      "epoch": 1.1181556195965419,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048827769010809666,
      "loss": 4.8642,
      "mean_token_accuracy": 0.22833341658115386,
      "num_tokens": 26673224.0,
      "step": 11640
    },
    {
      "entropy": 5.298730659484863,
      "epoch": 1.1186359269932757,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048826676607374606,
      "loss": 4.8939,
      "mean_token_accuracy": 0.22179376482963561,
      "num_tokens": 26686331.0,
      "step": 11645
    },
    {
      "entropy": 5.35620379447937,
      "epoch": 1.1191162343900096,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048825583708797434,
      "loss": 4.9512,
      "mean_token_accuracy": 0.22299579828977584,
      "num_tokens": 26696816.0,
      "step": 11650
    },
    {
      "entropy": 5.245871686935425,
      "epoch": 1.1195965417867435,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004882449031510354,
      "loss": 4.8982,
      "mean_token_accuracy": 0.2305406704545021,
      "num_tokens": 26708126.0,
      "step": 11655
    },
    {
      "entropy": 5.337231731414795,
      "epoch": 1.1200768491834774,
      "grad_norm": 0.953125,
      "learning_rate": 0.000488233964263183,
      "loss": 5.0133,
      "mean_token_accuracy": 0.22279638648033143,
      "num_tokens": 26718951.0,
      "step": 11660
    },
    {
      "entropy": 5.344915866851807,
      "epoch": 1.1205571565802113,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048822302042467115,
      "loss": 4.9736,
      "mean_token_accuracy": 0.2190208688378334,
      "num_tokens": 26730550.0,
      "step": 11665
    },
    {
      "entropy": 5.210195398330688,
      "epoch": 1.1210374639769451,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004882120716357539,
      "loss": 4.7998,
      "mean_token_accuracy": 0.23389629125595093,
      "num_tokens": 26741485.0,
      "step": 11670
    },
    {
      "entropy": 5.181673336029053,
      "epoch": 1.1215177713736793,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004882011178966854,
      "loss": 4.8052,
      "mean_token_accuracy": 0.2290264055132866,
      "num_tokens": 26753947.0,
      "step": 11675
    },
    {
      "entropy": 5.250951957702637,
      "epoch": 1.1219980787704131,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004881901592077201,
      "loss": 4.8626,
      "mean_token_accuracy": 0.23917226046323775,
      "num_tokens": 26764921.0,
      "step": 11680
    },
    {
      "entropy": 5.263632965087891,
      "epoch": 1.122478386167147,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004881791955691123,
      "loss": 4.8448,
      "mean_token_accuracy": 0.22826866805553436,
      "num_tokens": 26777384.0,
      "step": 11685
    },
    {
      "entropy": 5.300962448120117,
      "epoch": 1.122958693563881,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048816822698111655,
      "loss": 4.9626,
      "mean_token_accuracy": 0.22107819765806197,
      "num_tokens": 26789916.0,
      "step": 11690
    },
    {
      "entropy": 5.262270832061768,
      "epoch": 1.1234390009606148,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048815725344398766,
      "loss": 4.8585,
      "mean_token_accuracy": 0.23172865957021713,
      "num_tokens": 26801167.0,
      "step": 11695
    },
    {
      "entropy": 5.344625997543335,
      "epoch": 1.1239193083573487,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048814627495798017,
      "loss": 4.965,
      "mean_token_accuracy": 0.22059017717838286,
      "num_tokens": 26813235.0,
      "step": 11700
    },
    {
      "entropy": 5.328128719329834,
      "epoch": 1.1243996157540825,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004881352915233492,
      "loss": 4.9574,
      "mean_token_accuracy": 0.22624506950378417,
      "num_tokens": 26824758.0,
      "step": 11705
    },
    {
      "entropy": 5.271256923675537,
      "epoch": 1.1248799231508164,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048812430314034956,
      "loss": 4.9398,
      "mean_token_accuracy": 0.2202802777290344,
      "num_tokens": 26836877.0,
      "step": 11710
    },
    {
      "entropy": 5.191570472717285,
      "epoch": 1.1253602305475505,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004881133098092365,
      "loss": 4.7814,
      "mean_token_accuracy": 0.2398412346839905,
      "num_tokens": 26848394.0,
      "step": 11715
    },
    {
      "entropy": 5.236995029449463,
      "epoch": 1.1258405379442844,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004881023115302652,
      "loss": 4.9128,
      "mean_token_accuracy": 0.22459905296564103,
      "num_tokens": 26859064.0,
      "step": 11720
    },
    {
      "entropy": 5.346769046783447,
      "epoch": 1.1263208453410183,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000488091308303691,
      "loss": 5.0282,
      "mean_token_accuracy": 0.21090710610151292,
      "num_tokens": 26870401.0,
      "step": 11725
    },
    {
      "entropy": 5.308807516098023,
      "epoch": 1.1268011527377522,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004880803001297694,
      "loss": 4.9202,
      "mean_token_accuracy": 0.2240648478269577,
      "num_tokens": 26881767.0,
      "step": 11730
    },
    {
      "entropy": 5.235432052612305,
      "epoch": 1.127281460134486,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004880692870087559,
      "loss": 4.835,
      "mean_token_accuracy": 0.2254103273153305,
      "num_tokens": 26892854.0,
      "step": 11735
    },
    {
      "entropy": 5.367609310150146,
      "epoch": 1.12776176753122,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048805826894090626,
      "loss": 4.9824,
      "mean_token_accuracy": 0.2195629969239235,
      "num_tokens": 26905079.0,
      "step": 11740
    },
    {
      "entropy": 5.363622236251831,
      "epoch": 1.1282420749279538,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048804724592647626,
      "loss": 5.0124,
      "mean_token_accuracy": 0.21763491481542588,
      "num_tokens": 26915640.0,
      "step": 11745
    },
    {
      "entropy": 5.292817115783691,
      "epoch": 1.1287223823246877,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004880362179657218,
      "loss": 4.8925,
      "mean_token_accuracy": 0.22654586136341096,
      "num_tokens": 26927039.0,
      "step": 11750
    },
    {
      "entropy": 5.247404479980469,
      "epoch": 1.1292026897214218,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048802518505889904,
      "loss": 4.8727,
      "mean_token_accuracy": 0.22370742708444596,
      "num_tokens": 26939368.0,
      "step": 11755
    },
    {
      "entropy": 5.24517126083374,
      "epoch": 1.1296829971181557,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048801414720626404,
      "loss": 5.0515,
      "mean_token_accuracy": 0.20830000191926956,
      "num_tokens": 26952051.0,
      "step": 11760
    },
    {
      "entropy": 5.258783912658691,
      "epoch": 1.1301633045148896,
      "grad_norm": 1.0,
      "learning_rate": 0.00048800310440807294,
      "loss": 4.8731,
      "mean_token_accuracy": 0.22336975634098052,
      "num_tokens": 26963680.0,
      "step": 11765
    },
    {
      "entropy": 5.210563278198242,
      "epoch": 1.1306436119116234,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004879920566645823,
      "loss": 4.7843,
      "mean_token_accuracy": 0.2323192059993744,
      "num_tokens": 26973899.0,
      "step": 11770
    },
    {
      "entropy": 5.252214574813843,
      "epoch": 1.1311239193083573,
      "grad_norm": 1.0,
      "learning_rate": 0.0004879810039760486,
      "loss": 4.9232,
      "mean_token_accuracy": 0.22012716829776763,
      "num_tokens": 26985692.0,
      "step": 11775
    },
    {
      "entropy": 5.292316484451294,
      "epoch": 1.1316042267050912,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004879699463427284,
      "loss": 4.8588,
      "mean_token_accuracy": 0.22603437006473542,
      "num_tokens": 26996354.0,
      "step": 11780
    },
    {
      "entropy": 5.310973119735718,
      "epoch": 1.132084534101825,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004879588837648785,
      "loss": 4.8996,
      "mean_token_accuracy": 0.22527918219566345,
      "num_tokens": 27007925.0,
      "step": 11785
    },
    {
      "entropy": 5.228297185897827,
      "epoch": 1.1325648414985592,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048794781624275554,
      "loss": 4.8679,
      "mean_token_accuracy": 0.23088593035936356,
      "num_tokens": 27019222.0,
      "step": 11790
    },
    {
      "entropy": 5.183725261688233,
      "epoch": 1.133045148895293,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048793674377661664,
      "loss": 4.8551,
      "mean_token_accuracy": 0.23071279227733613,
      "num_tokens": 27029973.0,
      "step": 11795
    },
    {
      "entropy": 5.331638145446777,
      "epoch": 1.133525456292027,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048792566636671886,
      "loss": 5.0155,
      "mean_token_accuracy": 0.22305222153663634,
      "num_tokens": 27042002.0,
      "step": 11800
    },
    {
      "entropy": 5.299916791915893,
      "epoch": 1.1340057636887608,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004879145840133194,
      "loss": 4.9147,
      "mean_token_accuracy": 0.23431730419397354,
      "num_tokens": 27052770.0,
      "step": 11805
    },
    {
      "entropy": 5.265444087982178,
      "epoch": 1.1344860710854947,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004879034967166755,
      "loss": 4.9487,
      "mean_token_accuracy": 0.2243955463171005,
      "num_tokens": 27063375.0,
      "step": 11810
    },
    {
      "entropy": 5.204526472091675,
      "epoch": 1.1349663784822286,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004878924044770446,
      "loss": 4.8885,
      "mean_token_accuracy": 0.22880009561777115,
      "num_tokens": 27075908.0,
      "step": 11815
    },
    {
      "entropy": 5.270445871353149,
      "epoch": 1.1354466858789625,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004878813072946843,
      "loss": 4.9309,
      "mean_token_accuracy": 0.2226850301027298,
      "num_tokens": 27087590.0,
      "step": 11820
    },
    {
      "entropy": 5.289422607421875,
      "epoch": 1.1359269932756964,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048787020516985203,
      "loss": 4.8552,
      "mean_token_accuracy": 0.23421876281499862,
      "num_tokens": 27098076.0,
      "step": 11825
    },
    {
      "entropy": 5.226479387283325,
      "epoch": 1.1364073006724302,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048785909810280576,
      "loss": 4.8515,
      "mean_token_accuracy": 0.2359784185886383,
      "num_tokens": 27109851.0,
      "step": 11830
    },
    {
      "entropy": 5.260421323776245,
      "epoch": 1.1368876080691643,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004878479860938033,
      "loss": 4.8215,
      "mean_token_accuracy": 0.23131102770566941,
      "num_tokens": 27121288.0,
      "step": 11835
    },
    {
      "entropy": 5.26934585571289,
      "epoch": 1.1373679154658982,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048783686914310266,
      "loss": 4.8999,
      "mean_token_accuracy": 0.2264281153678894,
      "num_tokens": 27133877.0,
      "step": 11840
    },
    {
      "entropy": 5.114246082305908,
      "epoch": 1.137848222862632,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004878257472509619,
      "loss": 4.7265,
      "mean_token_accuracy": 0.23701368421316146,
      "num_tokens": 27145616.0,
      "step": 11845
    },
    {
      "entropy": 5.1711156368255615,
      "epoch": 1.138328530259366,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004878146204176392,
      "loss": 4.8665,
      "mean_token_accuracy": 0.2223424568772316,
      "num_tokens": 27157182.0,
      "step": 11850
    },
    {
      "entropy": 5.287313032150268,
      "epoch": 1.1388088376560999,
      "grad_norm": 1.015625,
      "learning_rate": 0.000487803488643393,
      "loss": 4.8234,
      "mean_token_accuracy": 0.2310245379805565,
      "num_tokens": 27167609.0,
      "step": 11855
    },
    {
      "entropy": 5.350826120376587,
      "epoch": 1.1392891450528337,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048779235192848166,
      "loss": 4.9776,
      "mean_token_accuracy": 0.21891916841268538,
      "num_tokens": 27180043.0,
      "step": 11860
    },
    {
      "entropy": 5.204944849014282,
      "epoch": 1.1397694524495678,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004877812102731638,
      "loss": 4.7965,
      "mean_token_accuracy": 0.23278993368148804,
      "num_tokens": 27191870.0,
      "step": 11865
    },
    {
      "entropy": 5.251076745986938,
      "epoch": 1.1402497598463017,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048777006367769804,
      "loss": 4.9228,
      "mean_token_accuracy": 0.22674772292375564,
      "num_tokens": 27203291.0,
      "step": 11870
    },
    {
      "entropy": 5.262463045120239,
      "epoch": 1.1407300672430356,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004877589121423432,
      "loss": 4.9318,
      "mean_token_accuracy": 0.22514677196741104,
      "num_tokens": 27214607.0,
      "step": 11875
    },
    {
      "entropy": 5.237838506698608,
      "epoch": 1.1412103746397695,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004877477556673582,
      "loss": 4.8094,
      "mean_token_accuracy": 0.23124697506427766,
      "num_tokens": 27224658.0,
      "step": 11880
    },
    {
      "entropy": 5.188950490951538,
      "epoch": 1.1416906820365034,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000487736594253002,
      "loss": 4.9069,
      "mean_token_accuracy": 0.2195824682712555,
      "num_tokens": 27235306.0,
      "step": 11885
    },
    {
      "entropy": 5.291343355178833,
      "epoch": 1.1421709894332372,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048772542789953384,
      "loss": 4.8972,
      "mean_token_accuracy": 0.22390957325696945,
      "num_tokens": 27246679.0,
      "step": 11890
    },
    {
      "entropy": 5.367614650726319,
      "epoch": 1.1426512968299711,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048771425660721284,
      "loss": 4.9541,
      "mean_token_accuracy": 0.21486895084381102,
      "num_tokens": 27257799.0,
      "step": 11895
    },
    {
      "entropy": 5.227843332290649,
      "epoch": 1.143131604226705,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048770308037629853,
      "loss": 4.9326,
      "mean_token_accuracy": 0.2198262929916382,
      "num_tokens": 27268651.0,
      "step": 11900
    },
    {
      "entropy": 5.209802865982056,
      "epoch": 1.143611911623439,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004876918992070502,
      "loss": 4.893,
      "mean_token_accuracy": 0.22881580591201783,
      "num_tokens": 27280655.0,
      "step": 11905
    },
    {
      "entropy": 5.337397050857544,
      "epoch": 1.144092219020173,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004876807130997276,
      "loss": 4.9716,
      "mean_token_accuracy": 0.21730809658765793,
      "num_tokens": 27292664.0,
      "step": 11910
    },
    {
      "entropy": 5.287272357940674,
      "epoch": 1.1445725264169069,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004876695220545903,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2241139143705368,
      "num_tokens": 27306143.0,
      "step": 11915
    },
    {
      "entropy": 5.239326333999633,
      "epoch": 1.1450528338136408,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048765832607189824,
      "loss": 4.8437,
      "mean_token_accuracy": 0.22911227941513063,
      "num_tokens": 27316926.0,
      "step": 11920
    },
    {
      "entropy": 5.350647401809693,
      "epoch": 1.1455331412103746,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048764712515191136,
      "loss": 4.9638,
      "mean_token_accuracy": 0.22819378972053528,
      "num_tokens": 27327472.0,
      "step": 11925
    },
    {
      "entropy": 5.32572226524353,
      "epoch": 1.1460134486071085,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048763591929488966,
      "loss": 4.9308,
      "mean_token_accuracy": 0.22831723988056182,
      "num_tokens": 27338388.0,
      "step": 11930
    },
    {
      "entropy": 5.245071649551392,
      "epoch": 1.1464937560038424,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004876247085010933,
      "loss": 4.8443,
      "mean_token_accuracy": 0.22653891146183014,
      "num_tokens": 27349879.0,
      "step": 11935
    },
    {
      "entropy": 5.226050901412964,
      "epoch": 1.1469740634005763,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048761349277078253,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2275835782289505,
      "num_tokens": 27361226.0,
      "step": 11940
    },
    {
      "entropy": 5.251747131347656,
      "epoch": 1.1474543707973104,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048760227210421775,
      "loss": 4.8482,
      "mean_token_accuracy": 0.2297546699643135,
      "num_tokens": 27373512.0,
      "step": 11945
    },
    {
      "entropy": 5.223568773269653,
      "epoch": 1.1479346781940443,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004875910465016596,
      "loss": 4.8845,
      "mean_token_accuracy": 0.2332419291138649,
      "num_tokens": 27384950.0,
      "step": 11950
    },
    {
      "entropy": 5.376255941390991,
      "epoch": 1.1484149855907781,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004875798159633686,
      "loss": 4.9514,
      "mean_token_accuracy": 0.22477176338434218,
      "num_tokens": 27395877.0,
      "step": 11955
    },
    {
      "entropy": 5.3011030673980715,
      "epoch": 1.148895292987512,
      "grad_norm": 1.0,
      "learning_rate": 0.0004875685804896055,
      "loss": 4.9845,
      "mean_token_accuracy": 0.2234122261404991,
      "num_tokens": 27407783.0,
      "step": 11960
    },
    {
      "entropy": 5.228872632980346,
      "epoch": 1.149375600384246,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004875573400806312,
      "loss": 4.867,
      "mean_token_accuracy": 0.2225528836250305,
      "num_tokens": 27420598.0,
      "step": 11965
    },
    {
      "entropy": 5.276258087158203,
      "epoch": 1.1498559077809798,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048754609473670654,
      "loss": 4.9545,
      "mean_token_accuracy": 0.22420434951782225,
      "num_tokens": 27431961.0,
      "step": 11970
    },
    {
      "entropy": 5.2426292419433596,
      "epoch": 1.1503362151777137,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004875348444580927,
      "loss": 4.8196,
      "mean_token_accuracy": 0.230824775993824,
      "num_tokens": 27444010.0,
      "step": 11975
    },
    {
      "entropy": 5.315143585205078,
      "epoch": 1.1508165225744476,
      "grad_norm": 0.953125,
      "learning_rate": 0.000487523589245051,
      "loss": 4.8887,
      "mean_token_accuracy": 0.225459223985672,
      "num_tokens": 27455576.0,
      "step": 11980
    },
    {
      "entropy": 5.24446816444397,
      "epoch": 1.1512968299711814,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004875123290978425,
      "loss": 4.9127,
      "mean_token_accuracy": 0.22304505556821824,
      "num_tokens": 27468957.0,
      "step": 11985
    },
    {
      "entropy": 5.247353363037109,
      "epoch": 1.1517771373679155,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048750106401672876,
      "loss": 4.9046,
      "mean_token_accuracy": 0.23000882416963578,
      "num_tokens": 27479992.0,
      "step": 11990
    },
    {
      "entropy": 5.317772912979126,
      "epoch": 1.1522574447646494,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048748979400197134,
      "loss": 4.9728,
      "mean_token_accuracy": 0.21948516964912415,
      "num_tokens": 27490878.0,
      "step": 11995
    },
    {
      "entropy": 5.2117226123809814,
      "epoch": 1.1527377521613833,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048747851905383183,
      "loss": 4.8075,
      "mean_token_accuracy": 0.23611709475517273,
      "num_tokens": 27502009.0,
      "step": 12000
    },
    {
      "epoch": 1.1527377521613833,
      "eval_entropy": 5.037428198762779,
      "eval_loss": 4.992321968078613,
      "eval_mean_token_accuracy": 0.22976471865227285,
      "eval_num_tokens": 27502009.0,
      "eval_runtime": 20.4844,
      "eval_samples_per_second": 1601.952,
      "eval_steps_per_second": 200.25,
      "step": 12000
    },
    {
      "entropy": 5.257339668273926,
      "epoch": 1.1532180595581172,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004874672391725721,
      "loss": 4.9963,
      "mean_token_accuracy": 0.2174884110689163,
      "num_tokens": 27513376.0,
      "step": 12005
    },
    {
      "entropy": 5.2297076225280765,
      "epoch": 1.153698366954851,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004874559543584539,
      "loss": 4.8542,
      "mean_token_accuracy": 0.2263937324285507,
      "num_tokens": 27525166.0,
      "step": 12010
    },
    {
      "entropy": 5.213230037689209,
      "epoch": 1.154178674351585,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004874446646117394,
      "loss": 4.7773,
      "mean_token_accuracy": 0.23873990029096603,
      "num_tokens": 27535994.0,
      "step": 12015
    },
    {
      "entropy": 5.234176921844482,
      "epoch": 1.154658981748319,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004874333699326906,
      "loss": 4.8686,
      "mean_token_accuracy": 0.22671267986297608,
      "num_tokens": 27546883.0,
      "step": 12020
    },
    {
      "entropy": 5.325826025009155,
      "epoch": 1.155139289145053,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004874220703215697,
      "loss": 4.9838,
      "mean_token_accuracy": 0.21815072149038314,
      "num_tokens": 27558840.0,
      "step": 12025
    },
    {
      "entropy": 5.250137615203857,
      "epoch": 1.1556195965417868,
      "grad_norm": 1.0,
      "learning_rate": 0.0004874107657786391,
      "loss": 4.9038,
      "mean_token_accuracy": 0.22472755014896392,
      "num_tokens": 27569776.0,
      "step": 12030
    },
    {
      "entropy": 5.294756460189819,
      "epoch": 1.1560999039385207,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048739945630416124,
      "loss": 4.856,
      "mean_token_accuracy": 0.22741033285856246,
      "num_tokens": 27580097.0,
      "step": 12035
    },
    {
      "entropy": 5.343919181823731,
      "epoch": 1.1565802113352546,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004873881418983987,
      "loss": 4.9876,
      "mean_token_accuracy": 0.22281960099935533,
      "num_tokens": 27592035.0,
      "step": 12040
    },
    {
      "entropy": 5.302003669738769,
      "epoch": 1.1570605187319885,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004873768225616141,
      "loss": 4.9305,
      "mean_token_accuracy": 0.22886396944522858,
      "num_tokens": 27604409.0,
      "step": 12045
    },
    {
      "entropy": 5.280681037902832,
      "epoch": 1.1575408261287223,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048736549829407047,
      "loss": 4.8603,
      "mean_token_accuracy": 0.23030827641487123,
      "num_tokens": 27615019.0,
      "step": 12050
    },
    {
      "entropy": 5.353492975234985,
      "epoch": 1.1580211335254562,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004873541690960305,
      "loss": 5.0051,
      "mean_token_accuracy": 0.2199113830924034,
      "num_tokens": 27627924.0,
      "step": 12055
    },
    {
      "entropy": 5.259052324295044,
      "epoch": 1.15850144092219,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004873428349677573,
      "loss": 4.8163,
      "mean_token_accuracy": 0.22819008380174638,
      "num_tokens": 27639188.0,
      "step": 12060
    },
    {
      "entropy": 5.352970218658447,
      "epoch": 1.1589817483189242,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000487331495909514,
      "loss": 5.0281,
      "mean_token_accuracy": 0.21241912245750427,
      "num_tokens": 27652621.0,
      "step": 12065
    },
    {
      "entropy": 5.30935320854187,
      "epoch": 1.159462055715658,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048732015192156383,
      "loss": 4.9593,
      "mean_token_accuracy": 0.21591803133487703,
      "num_tokens": 27665726.0,
      "step": 12070
    },
    {
      "entropy": 5.271768426895141,
      "epoch": 1.159942363112392,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048730880300417015,
      "loss": 4.92,
      "mean_token_accuracy": 0.21938228905200957,
      "num_tokens": 27676984.0,
      "step": 12075
    },
    {
      "entropy": 5.283555412292481,
      "epoch": 1.1604226705091258,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048729744915759657,
      "loss": 4.9635,
      "mean_token_accuracy": 0.22618515491485597,
      "num_tokens": 27688812.0,
      "step": 12080
    },
    {
      "entropy": 5.195539236068726,
      "epoch": 1.1609029779058597,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048728609038210655,
      "loss": 4.7888,
      "mean_token_accuracy": 0.23063700497150422,
      "num_tokens": 27699152.0,
      "step": 12085
    },
    {
      "entropy": 5.266979026794433,
      "epoch": 1.1613832853025936,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048727472667796395,
      "loss": 4.9478,
      "mean_token_accuracy": 0.22452907115221024,
      "num_tokens": 27710347.0,
      "step": 12090
    },
    {
      "entropy": 5.210976457595825,
      "epoch": 1.1618635926993275,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004872633580454325,
      "loss": 4.8626,
      "mean_token_accuracy": 0.22892893552780152,
      "num_tokens": 27721112.0,
      "step": 12095
    },
    {
      "entropy": 5.302061223983765,
      "epoch": 1.1623439000960616,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048725198448477616,
      "loss": 4.9401,
      "mean_token_accuracy": 0.21984083354473113,
      "num_tokens": 27731766.0,
      "step": 12100
    },
    {
      "entropy": 5.2563975811004635,
      "epoch": 1.1628242074927955,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048724060599625893,
      "loss": 4.8858,
      "mean_token_accuracy": 0.22780081927776336,
      "num_tokens": 27743718.0,
      "step": 12105
    },
    {
      "entropy": 5.293583869934082,
      "epoch": 1.1633045148895294,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048722922258014506,
      "loss": 4.9402,
      "mean_token_accuracy": 0.2237210363149643,
      "num_tokens": 27754999.0,
      "step": 12110
    },
    {
      "entropy": 5.3243388652801515,
      "epoch": 1.1637848222862632,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004872178342366989,
      "loss": 5.0286,
      "mean_token_accuracy": 0.2146253451704979,
      "num_tokens": 27767684.0,
      "step": 12115
    },
    {
      "entropy": 5.267794275283814,
      "epoch": 1.1642651296829971,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048720644096618475,
      "loss": 4.9077,
      "mean_token_accuracy": 0.22877496480941772,
      "num_tokens": 27779103.0,
      "step": 12120
    },
    {
      "entropy": 5.25468807220459,
      "epoch": 1.164745437079731,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004871950427688672,
      "loss": 4.9307,
      "mean_token_accuracy": 0.22676109224557878,
      "num_tokens": 27791223.0,
      "step": 12125
    },
    {
      "entropy": 5.262859630584717,
      "epoch": 1.1652257444764649,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048718363964501087,
      "loss": 4.8648,
      "mean_token_accuracy": 0.22908487319946289,
      "num_tokens": 27802660.0,
      "step": 12130
    },
    {
      "entropy": 5.326535320281982,
      "epoch": 1.1657060518731988,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004871722315948805,
      "loss": 4.9367,
      "mean_token_accuracy": 0.21966375261545182,
      "num_tokens": 27814370.0,
      "step": 12135
    },
    {
      "entropy": 5.271794986724854,
      "epoch": 1.1661863592699326,
      "grad_norm": 1.0,
      "learning_rate": 0.0004871608186187408,
      "loss": 4.8545,
      "mean_token_accuracy": 0.23157301545143127,
      "num_tokens": 27825035.0,
      "step": 12140
    },
    {
      "entropy": 5.181306028366089,
      "epoch": 1.1666666666666667,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048714940071685703,
      "loss": 4.7855,
      "mean_token_accuracy": 0.22974889874458312,
      "num_tokens": 27837276.0,
      "step": 12145
    },
    {
      "entropy": 5.351789951324463,
      "epoch": 1.1671469740634006,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048713797788949405,
      "loss": 4.9811,
      "mean_token_accuracy": 0.2161165863275528,
      "num_tokens": 27848618.0,
      "step": 12150
    },
    {
      "entropy": 5.233269309997558,
      "epoch": 1.1676272814601345,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048712655013691714,
      "loss": 4.9157,
      "mean_token_accuracy": 0.22098602205514908,
      "num_tokens": 27861556.0,
      "step": 12155
    },
    {
      "entropy": 5.312176322937011,
      "epoch": 1.1681075888568684,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048711511745939165,
      "loss": 4.918,
      "mean_token_accuracy": 0.2209663763642311,
      "num_tokens": 27873211.0,
      "step": 12160
    },
    {
      "entropy": 5.30955867767334,
      "epoch": 1.1685878962536023,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000487103679857183,
      "loss": 4.9094,
      "mean_token_accuracy": 0.22887077182531357,
      "num_tokens": 27883365.0,
      "step": 12165
    },
    {
      "entropy": 5.177827215194702,
      "epoch": 1.1690682036503361,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004870922373305567,
      "loss": 4.7983,
      "mean_token_accuracy": 0.23087912052869797,
      "num_tokens": 27894669.0,
      "step": 12170
    },
    {
      "entropy": 5.24456467628479,
      "epoch": 1.1695485110470702,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00048708078987977837,
      "loss": 4.9052,
      "mean_token_accuracy": 0.22964977324008942,
      "num_tokens": 27906480.0,
      "step": 12175
    },
    {
      "entropy": 5.248481559753418,
      "epoch": 1.1700288184438041,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048706933750511394,
      "loss": 4.8679,
      "mean_token_accuracy": 0.23251143246889114,
      "num_tokens": 27918383.0,
      "step": 12180
    },
    {
      "entropy": 5.314880895614624,
      "epoch": 1.170509125840538,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004870578802068292,
      "loss": 4.9089,
      "mean_token_accuracy": 0.22570340931415558,
      "num_tokens": 27928944.0,
      "step": 12185
    },
    {
      "entropy": 5.203420639038086,
      "epoch": 1.170989433237272,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048704641798519006,
      "loss": 4.8338,
      "mean_token_accuracy": 0.22706536203622818,
      "num_tokens": 27941105.0,
      "step": 12190
    },
    {
      "entropy": 5.24574089050293,
      "epoch": 1.1714697406340058,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048703495084046286,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2327367126941681,
      "num_tokens": 27952925.0,
      "step": 12195
    },
    {
      "entropy": 5.216184902191162,
      "epoch": 1.1719500480307397,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004870234787729137,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2336528331041336,
      "num_tokens": 27965504.0,
      "step": 12200
    },
    {
      "entropy": 5.232403326034546,
      "epoch": 1.1724303554274735,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004870120017828089,
      "loss": 4.8555,
      "mean_token_accuracy": 0.23489184975624083,
      "num_tokens": 27975985.0,
      "step": 12205
    },
    {
      "entropy": 5.383049964904785,
      "epoch": 1.1729106628242074,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000487000519870415,
      "loss": 4.9603,
      "mean_token_accuracy": 0.21262255162000657,
      "num_tokens": 27987784.0,
      "step": 12210
    },
    {
      "entropy": 5.2907867431640625,
      "epoch": 1.1733909702209413,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004869890330359986,
      "loss": 4.8786,
      "mean_token_accuracy": 0.22434595376253127,
      "num_tokens": 27999623.0,
      "step": 12215
    },
    {
      "entropy": 5.260912752151489,
      "epoch": 1.1738712776176754,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004869775412798262,
      "loss": 4.8809,
      "mean_token_accuracy": 0.22460998892784118,
      "num_tokens": 28011561.0,
      "step": 12220
    },
    {
      "entropy": 5.275447225570678,
      "epoch": 1.1743515850144093,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048696604460216476,
      "loss": 4.9302,
      "mean_token_accuracy": 0.22801712900400162,
      "num_tokens": 28022108.0,
      "step": 12225
    },
    {
      "entropy": 5.291474866867065,
      "epoch": 1.1748318924111432,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048695454300328123,
      "loss": 4.9096,
      "mean_token_accuracy": 0.22193693816661836,
      "num_tokens": 28033497.0,
      "step": 12230
    },
    {
      "entropy": 5.340562152862549,
      "epoch": 1.175312199807877,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048694303648344256,
      "loss": 4.9292,
      "mean_token_accuracy": 0.23044474571943283,
      "num_tokens": 28044790.0,
      "step": 12235
    },
    {
      "entropy": 5.217104959487915,
      "epoch": 1.175792507204611,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048693152504291595,
      "loss": 4.8709,
      "mean_token_accuracy": 0.22987905591726304,
      "num_tokens": 28056000.0,
      "step": 12240
    },
    {
      "entropy": 5.236352205276489,
      "epoch": 1.1762728146013448,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004869200086819686,
      "loss": 4.9015,
      "mean_token_accuracy": 0.22358250468969346,
      "num_tokens": 28068264.0,
      "step": 12245
    },
    {
      "entropy": 5.319037914276123,
      "epoch": 1.1767531219980787,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048690848740086796,
      "loss": 4.8214,
      "mean_token_accuracy": 0.22945654392242432,
      "num_tokens": 28080220.0,
      "step": 12250
    },
    {
      "entropy": 5.30375599861145,
      "epoch": 1.1772334293948128,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004868969611998814,
      "loss": 4.8585,
      "mean_token_accuracy": 0.2302807867527008,
      "num_tokens": 28091531.0,
      "step": 12255
    },
    {
      "entropy": 5.22531795501709,
      "epoch": 1.1777137367915467,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004868854300792767,
      "loss": 4.8631,
      "mean_token_accuracy": 0.22490509152412413,
      "num_tokens": 28102982.0,
      "step": 12260
    },
    {
      "entropy": 5.249143791198731,
      "epoch": 1.1781940441882806,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048687389403932144,
      "loss": 4.9219,
      "mean_token_accuracy": 0.2242107704281807,
      "num_tokens": 28114131.0,
      "step": 12265
    },
    {
      "entropy": 5.268837928771973,
      "epoch": 1.1786743515850144,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004868623530802835,
      "loss": 4.8441,
      "mean_token_accuracy": 0.226272052526474,
      "num_tokens": 28125637.0,
      "step": 12270
    },
    {
      "entropy": 5.35436782836914,
      "epoch": 1.1791546589817483,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048685080720243086,
      "loss": 4.9091,
      "mean_token_accuracy": 0.2267795830965042,
      "num_tokens": 28137284.0,
      "step": 12275
    },
    {
      "entropy": 5.291742420196533,
      "epoch": 1.1796349663784822,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004868392564060315,
      "loss": 4.9195,
      "mean_token_accuracy": 0.23213468343019486,
      "num_tokens": 28148909.0,
      "step": 12280
    },
    {
      "entropy": 5.297422075271607,
      "epoch": 1.180115273775216,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004868277006913537,
      "loss": 4.9155,
      "mean_token_accuracy": 0.22935260683298112,
      "num_tokens": 28160392.0,
      "step": 12285
    },
    {
      "entropy": 5.323032569885254,
      "epoch": 1.18059558117195,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004868161400586656,
      "loss": 4.9775,
      "mean_token_accuracy": 0.21787267029285431,
      "num_tokens": 28172957.0,
      "step": 12290
    },
    {
      "entropy": 5.389630603790283,
      "epoch": 1.1810758885686838,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004868045745082357,
      "loss": 4.9655,
      "mean_token_accuracy": 0.21511545926332473,
      "num_tokens": 28183239.0,
      "step": 12295
    },
    {
      "entropy": 5.241728019714356,
      "epoch": 1.181556195965418,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004867930040403326,
      "loss": 4.8807,
      "mean_token_accuracy": 0.22663789391517639,
      "num_tokens": 28195382.0,
      "step": 12300
    },
    {
      "entropy": 5.252456951141357,
      "epoch": 1.1820365033621518,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048678142865522475,
      "loss": 4.9416,
      "mean_token_accuracy": 0.2186951220035553,
      "num_tokens": 28206645.0,
      "step": 12305
    },
    {
      "entropy": 5.329221963882446,
      "epoch": 1.1825168107588857,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000486769848353181,
      "loss": 4.9269,
      "mean_token_accuracy": 0.22751059681177138,
      "num_tokens": 28218420.0,
      "step": 12310
    },
    {
      "entropy": 5.317015504837036,
      "epoch": 1.1829971181556196,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048675826313447027,
      "loss": 4.9435,
      "mean_token_accuracy": 0.22257424592971803,
      "num_tokens": 28229458.0,
      "step": 12315
    },
    {
      "entropy": 5.296759128570557,
      "epoch": 1.1834774255523535,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048674667299936135,
      "loss": 4.9288,
      "mean_token_accuracy": 0.22483680695295333,
      "num_tokens": 28242240.0,
      "step": 12320
    },
    {
      "entropy": 5.415534830093383,
      "epoch": 1.1839577329490873,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048673507794812356,
      "loss": 5.026,
      "mean_token_accuracy": 0.21760628819465638,
      "num_tokens": 28254597.0,
      "step": 12325
    },
    {
      "entropy": 5.288152170181275,
      "epoch": 1.1844380403458212,
      "grad_norm": 0.875,
      "learning_rate": 0.0004867234779810259,
      "loss": 4.955,
      "mean_token_accuracy": 0.2214645639061928,
      "num_tokens": 28266674.0,
      "step": 12330
    },
    {
      "entropy": 5.2395241260528564,
      "epoch": 1.1849183477425553,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004867118730983378,
      "loss": 4.9064,
      "mean_token_accuracy": 0.22530634105205535,
      "num_tokens": 28278671.0,
      "step": 12335
    },
    {
      "entropy": 5.297441387176514,
      "epoch": 1.1853986551392892,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004867002633003286,
      "loss": 4.8374,
      "mean_token_accuracy": 0.23156213909387588,
      "num_tokens": 28291085.0,
      "step": 12340
    },
    {
      "entropy": 5.240581846237182,
      "epoch": 1.185878962536023,
      "grad_norm": 1.125,
      "learning_rate": 0.000486688648587268,
      "loss": 4.9087,
      "mean_token_accuracy": 0.22988221645355225,
      "num_tokens": 28303457.0,
      "step": 12345
    },
    {
      "entropy": 5.286825466156006,
      "epoch": 1.186359269932757,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004866770289594256,
      "loss": 4.9397,
      "mean_token_accuracy": 0.22337566465139388,
      "num_tokens": 28314599.0,
      "step": 12350
    },
    {
      "entropy": 5.410071659088135,
      "epoch": 1.1868395773294909,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048666540441707107,
      "loss": 5.0059,
      "mean_token_accuracy": 0.2223011761903763,
      "num_tokens": 28326266.0,
      "step": 12355
    },
    {
      "entropy": 5.285570907592773,
      "epoch": 1.1873198847262247,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004866537749604744,
      "loss": 4.875,
      "mean_token_accuracy": 0.22331466972827912,
      "num_tokens": 28337238.0,
      "step": 12360
    },
    {
      "entropy": 5.201148653030396,
      "epoch": 1.1878001921229586,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048664214058990546,
      "loss": 4.8653,
      "mean_token_accuracy": 0.2285682052373886,
      "num_tokens": 28348060.0,
      "step": 12365
    },
    {
      "entropy": 5.272304058074951,
      "epoch": 1.1882804995196925,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004866305013056346,
      "loss": 4.9209,
      "mean_token_accuracy": 0.22376569956541062,
      "num_tokens": 28359442.0,
      "step": 12370
    },
    {
      "entropy": 5.318316173553467,
      "epoch": 1.1887608069164266,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004866188571079318,
      "loss": 4.9485,
      "mean_token_accuracy": 0.22598656117916108,
      "num_tokens": 28370827.0,
      "step": 12375
    },
    {
      "entropy": 5.3819609642028805,
      "epoch": 1.1892411143131605,
      "grad_norm": 1.0,
      "learning_rate": 0.0004866072079970676,
      "loss": 4.9855,
      "mean_token_accuracy": 0.21506985127925873,
      "num_tokens": 28383115.0,
      "step": 12380
    },
    {
      "entropy": 5.285931634902954,
      "epoch": 1.1897214217098944,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048659555397331236,
      "loss": 4.8766,
      "mean_token_accuracy": 0.2303728625178337,
      "num_tokens": 28394904.0,
      "step": 12385
    },
    {
      "entropy": 5.256549739837647,
      "epoch": 1.1902017291066282,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004865838950369366,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22239051163196563,
      "num_tokens": 28407357.0,
      "step": 12390
    },
    {
      "entropy": 5.259527397155762,
      "epoch": 1.1906820365033621,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048657223118821116,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22825814932584762,
      "num_tokens": 28418088.0,
      "step": 12395
    },
    {
      "entropy": 5.14347734451294,
      "epoch": 1.191162343900096,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048656056242740665,
      "loss": 4.7781,
      "mean_token_accuracy": 0.23603992611169816,
      "num_tokens": 28430022.0,
      "step": 12400
    },
    {
      "entropy": 5.273916387557984,
      "epoch": 1.19164265129683,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004865488887547942,
      "loss": 4.8889,
      "mean_token_accuracy": 0.22520145624876023,
      "num_tokens": 28440530.0,
      "step": 12405
    },
    {
      "entropy": 5.333625555038452,
      "epoch": 1.192122958693564,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004865372101706446,
      "loss": 4.9761,
      "mean_token_accuracy": 0.22075784355401992,
      "num_tokens": 28452707.0,
      "step": 12410
    },
    {
      "entropy": 5.180582523345947,
      "epoch": 1.1926032660902979,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004865255266752292,
      "loss": 4.7821,
      "mean_token_accuracy": 0.2379097431898117,
      "num_tokens": 28465131.0,
      "step": 12415
    },
    {
      "entropy": 5.174480056762695,
      "epoch": 1.1930835734870318,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004865138382688191,
      "loss": 4.912,
      "mean_token_accuracy": 0.22855011075735093,
      "num_tokens": 28476136.0,
      "step": 12420
    },
    {
      "entropy": 5.279729270935059,
      "epoch": 1.1935638808837656,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004865021449516859,
      "loss": 4.8298,
      "mean_token_accuracy": 0.2275603711605072,
      "num_tokens": 28488374.0,
      "step": 12425
    },
    {
      "entropy": 5.2462316989898685,
      "epoch": 1.1940441882804995,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004864904467241008,
      "loss": 4.8746,
      "mean_token_accuracy": 0.22224860042333602,
      "num_tokens": 28499585.0,
      "step": 12430
    },
    {
      "entropy": 5.175922060012818,
      "epoch": 1.1945244956772334,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048647874358633556,
      "loss": 4.8253,
      "mean_token_accuracy": 0.23099935352802276,
      "num_tokens": 28510707.0,
      "step": 12435
    },
    {
      "entropy": 5.333666706085205,
      "epoch": 1.1950048030739673,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048646703553866183,
      "loss": 4.9687,
      "mean_token_accuracy": 0.21973738223314285,
      "num_tokens": 28522398.0,
      "step": 12440
    },
    {
      "entropy": 5.244618034362793,
      "epoch": 1.1954851104707012,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004864553225813515,
      "loss": 4.8546,
      "mean_token_accuracy": 0.22881031334400176,
      "num_tokens": 28532949.0,
      "step": 12445
    },
    {
      "entropy": 5.176388597488403,
      "epoch": 1.195965417867435,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004864436047146765,
      "loss": 4.8082,
      "mean_token_accuracy": 0.23415758162736894,
      "num_tokens": 28544292.0,
      "step": 12450
    },
    {
      "entropy": 5.299740314483643,
      "epoch": 1.1964457252641691,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048643188193890874,
      "loss": 4.9672,
      "mean_token_accuracy": 0.22111070156097412,
      "num_tokens": 28556430.0,
      "step": 12455
    },
    {
      "entropy": 5.325401973724365,
      "epoch": 1.196926032660903,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004864201542543206,
      "loss": 4.9412,
      "mean_token_accuracy": 0.2247442424297333,
      "num_tokens": 28568475.0,
      "step": 12460
    },
    {
      "entropy": 5.307481479644776,
      "epoch": 1.197406340057637,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004864084216611843,
      "loss": 4.9011,
      "mean_token_accuracy": 0.22981150448322296,
      "num_tokens": 28579653.0,
      "step": 12465
    },
    {
      "entropy": 5.283983898162842,
      "epoch": 1.1978866474543708,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048639668415977207,
      "loss": 4.8434,
      "mean_token_accuracy": 0.22291150093078613,
      "num_tokens": 28590108.0,
      "step": 12470
    },
    {
      "entropy": 5.255195903778076,
      "epoch": 1.1983669548511047,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048638494175035665,
      "loss": 4.855,
      "mean_token_accuracy": 0.23315632343292236,
      "num_tokens": 28602008.0,
      "step": 12475
    },
    {
      "entropy": 5.2289731979370115,
      "epoch": 1.1988472622478386,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004863731944332105,
      "loss": 4.7756,
      "mean_token_accuracy": 0.23319611549377442,
      "num_tokens": 28613286.0,
      "step": 12480
    },
    {
      "entropy": 5.228581476211548,
      "epoch": 1.1993275696445724,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004863614422086065,
      "loss": 4.7987,
      "mean_token_accuracy": 0.23187640011310579,
      "num_tokens": 28625497.0,
      "step": 12485
    },
    {
      "entropy": 5.254920721054077,
      "epoch": 1.1998078770413065,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004863496850768174,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22485039830207826,
      "num_tokens": 28638046.0,
      "step": 12490
    },
    {
      "entropy": 5.19256820678711,
      "epoch": 1.2002881844380404,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004863379230381162,
      "loss": 4.7925,
      "mean_token_accuracy": 0.2258091762661934,
      "num_tokens": 28647923.0,
      "step": 12495
    },
    {
      "entropy": 5.237299156188965,
      "epoch": 1.2007684918347743,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000486326156092776,
      "loss": 4.8511,
      "mean_token_accuracy": 0.23233419209718703,
      "num_tokens": 28660023.0,
      "step": 12500
    },
    {
      "entropy": 5.233293056488037,
      "epoch": 1.2012487992315082,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048631438424106985,
      "loss": 4.8681,
      "mean_token_accuracy": 0.22824403792619705,
      "num_tokens": 28671568.0,
      "step": 12505
    },
    {
      "entropy": 5.205038118362427,
      "epoch": 1.201729106628242,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048630260748327124,
      "loss": 4.8406,
      "mean_token_accuracy": 0.2351243793964386,
      "num_tokens": 28682897.0,
      "step": 12510
    },
    {
      "entropy": 5.285437965393067,
      "epoch": 1.202209414024976,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048629082581965355,
      "loss": 4.9107,
      "mean_token_accuracy": 0.22458701729774475,
      "num_tokens": 28694067.0,
      "step": 12515
    },
    {
      "entropy": 5.235145187377929,
      "epoch": 1.2026897214217098,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048627903925049033,
      "loss": 4.8589,
      "mean_token_accuracy": 0.22398596853017808,
      "num_tokens": 28705738.0,
      "step": 12520
    },
    {
      "entropy": 5.182246208190918,
      "epoch": 1.2031700288184437,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048626724777605507,
      "loss": 4.797,
      "mean_token_accuracy": 0.2313511461019516,
      "num_tokens": 28717419.0,
      "step": 12525
    },
    {
      "entropy": 5.3119792461395265,
      "epoch": 1.2036503362151778,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004862554513966217,
      "loss": 4.9907,
      "mean_token_accuracy": 0.21465240567922592,
      "num_tokens": 28728587.0,
      "step": 12530
    },
    {
      "entropy": 5.356195783615112,
      "epoch": 1.2041306436119117,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048624365011246405,
      "loss": 4.9795,
      "mean_token_accuracy": 0.21918516606092453,
      "num_tokens": 28740818.0,
      "step": 12535
    },
    {
      "entropy": 5.236780738830566,
      "epoch": 1.2046109510086456,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004862318439238561,
      "loss": 4.8181,
      "mean_token_accuracy": 0.238843110203743,
      "num_tokens": 28751936.0,
      "step": 12540
    },
    {
      "entropy": 5.2428741455078125,
      "epoch": 1.2050912584053795,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000486220032831072,
      "loss": 4.8793,
      "mean_token_accuracy": 0.22560443580150605,
      "num_tokens": 28763271.0,
      "step": 12545
    },
    {
      "entropy": 5.314275026321411,
      "epoch": 1.2055715658021133,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004862082168343859,
      "loss": 4.9154,
      "mean_token_accuracy": 0.22781168669462204,
      "num_tokens": 28774282.0,
      "step": 12550
    },
    {
      "entropy": 5.3074757099151615,
      "epoch": 1.2060518731988472,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004861963959340722,
      "loss": 4.9828,
      "mean_token_accuracy": 0.2203218534588814,
      "num_tokens": 28785826.0,
      "step": 12555
    },
    {
      "entropy": 5.198665904998779,
      "epoch": 1.206532180595581,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004861845701304053,
      "loss": 4.792,
      "mean_token_accuracy": 0.23408025354146958,
      "num_tokens": 28797669.0,
      "step": 12560
    },
    {
      "entropy": 5.294032049179077,
      "epoch": 1.2070124879923152,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048617273942365977,
      "loss": 4.9646,
      "mean_token_accuracy": 0.21765471249818802,
      "num_tokens": 28808438.0,
      "step": 12565
    },
    {
      "entropy": 5.223355150222778,
      "epoch": 1.207492795389049,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004861609038141103,
      "loss": 4.8001,
      "mean_token_accuracy": 0.2369787111878395,
      "num_tokens": 28819707.0,
      "step": 12570
    },
    {
      "entropy": 5.241160202026367,
      "epoch": 1.207973102785783,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048614906330203165,
      "loss": 4.9217,
      "mean_token_accuracy": 0.22261269390583038,
      "num_tokens": 28831829.0,
      "step": 12575
    },
    {
      "entropy": 5.23832139968872,
      "epoch": 1.2084534101825168,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004861372178876987,
      "loss": 4.8762,
      "mean_token_accuracy": 0.22937853783369064,
      "num_tokens": 28843429.0,
      "step": 12580
    },
    {
      "entropy": 5.218038177490234,
      "epoch": 1.2089337175792507,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048612536757138653,
      "loss": 4.8244,
      "mean_token_accuracy": 0.2332033097743988,
      "num_tokens": 28856239.0,
      "step": 12585
    },
    {
      "entropy": 5.2844311714172365,
      "epoch": 1.2094140249759846,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004861135123533702,
      "loss": 4.8864,
      "mean_token_accuracy": 0.23117202371358872,
      "num_tokens": 28868599.0,
      "step": 12590
    },
    {
      "entropy": 5.233556461334229,
      "epoch": 1.2098943323727185,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048610165223392503,
      "loss": 4.9077,
      "mean_token_accuracy": 0.2252286419272423,
      "num_tokens": 28880234.0,
      "step": 12595
    },
    {
      "entropy": 5.322301816940308,
      "epoch": 1.2103746397694524,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004860897872133263,
      "loss": 4.9368,
      "mean_token_accuracy": 0.22366511672735215,
      "num_tokens": 28892198.0,
      "step": 12600
    },
    {
      "entropy": 5.292739486694336,
      "epoch": 1.2108549471661862,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004860779172918496,
      "loss": 4.9713,
      "mean_token_accuracy": 0.21923016756772995,
      "num_tokens": 28904153.0,
      "step": 12605
    },
    {
      "entropy": 5.298227167129516,
      "epoch": 1.2113352545629203,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004860660424697704,
      "loss": 4.9393,
      "mean_token_accuracy": 0.22530335634946824,
      "num_tokens": 28915284.0,
      "step": 12610
    },
    {
      "entropy": 5.275113201141357,
      "epoch": 1.2118155619596542,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048605416274736434,
      "loss": 4.9044,
      "mean_token_accuracy": 0.22288039922714234,
      "num_tokens": 28928168.0,
      "step": 12615
    },
    {
      "entropy": 5.233430433273315,
      "epoch": 1.2122958693563881,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048604227812490744,
      "loss": 4.7924,
      "mean_token_accuracy": 0.2359822452068329,
      "num_tokens": 28938548.0,
      "step": 12620
    },
    {
      "entropy": 5.262873554229737,
      "epoch": 1.212776176753122,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048603038860267546,
      "loss": 4.8925,
      "mean_token_accuracy": 0.22611672878265382,
      "num_tokens": 28949252.0,
      "step": 12625
    },
    {
      "entropy": 5.296838092803955,
      "epoch": 1.2132564841498559,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004860184941809445,
      "loss": 4.9095,
      "mean_token_accuracy": 0.22626104652881623,
      "num_tokens": 28960758.0,
      "step": 12630
    },
    {
      "entropy": 5.283544969558716,
      "epoch": 1.2137367915465898,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048600659485999073,
      "loss": 4.8766,
      "mean_token_accuracy": 0.22739771902561187,
      "num_tokens": 28972604.0,
      "step": 12635
    },
    {
      "entropy": 5.2510387897491455,
      "epoch": 1.2142170989433236,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048599469064009027,
      "loss": 4.8801,
      "mean_token_accuracy": 0.22893189638853073,
      "num_tokens": 28983617.0,
      "step": 12640
    },
    {
      "entropy": 5.24014105796814,
      "epoch": 1.2146974063400577,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048598278152151974,
      "loss": 4.8971,
      "mean_token_accuracy": 0.22809119522571564,
      "num_tokens": 28994277.0,
      "step": 12645
    },
    {
      "entropy": 5.326847887039184,
      "epoch": 1.2151777137367916,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004859708675045555,
      "loss": 4.9412,
      "mean_token_accuracy": 0.22625575810670853,
      "num_tokens": 29004844.0,
      "step": 12650
    },
    {
      "entropy": 5.180233764648437,
      "epoch": 1.2156580211335255,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004859589485894741,
      "loss": 4.802,
      "mean_token_accuracy": 0.23248205929994584,
      "num_tokens": 29015978.0,
      "step": 12655
    },
    {
      "entropy": 5.324327993392944,
      "epoch": 1.2161383285302594,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004859470247765524,
      "loss": 4.9296,
      "mean_token_accuracy": 0.22249450385570527,
      "num_tokens": 29026966.0,
      "step": 12660
    },
    {
      "entropy": 5.299567794799804,
      "epoch": 1.2166186359269933,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004859350960660671,
      "loss": 4.8412,
      "mean_token_accuracy": 0.23241887986660004,
      "num_tokens": 29037943.0,
      "step": 12665
    },
    {
      "entropy": 5.30787878036499,
      "epoch": 1.2170989433237271,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004859231624582953,
      "loss": 4.914,
      "mean_token_accuracy": 0.21967006623744964,
      "num_tokens": 29049939.0,
      "step": 12670
    },
    {
      "entropy": 5.2285425662994385,
      "epoch": 1.217579250720461,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048591122395351394,
      "loss": 4.914,
      "mean_token_accuracy": 0.22539721727371215,
      "num_tokens": 29062442.0,
      "step": 12675
    },
    {
      "entropy": 5.195257902145386,
      "epoch": 1.218059558117195,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004858992805520003,
      "loss": 4.8864,
      "mean_token_accuracy": 0.2229933500289917,
      "num_tokens": 29074240.0,
      "step": 12680
    },
    {
      "entropy": 5.338029956817627,
      "epoch": 1.218539865513929,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048588733225403153,
      "loss": 4.9231,
      "mean_token_accuracy": 0.22241293042898178,
      "num_tokens": 29085550.0,
      "step": 12685
    },
    {
      "entropy": 5.256054782867432,
      "epoch": 1.219020172910663,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004858753790598851,
      "loss": 4.8565,
      "mean_token_accuracy": 0.23010126650333404,
      "num_tokens": 29097092.0,
      "step": 12690
    },
    {
      "entropy": 5.2294456481933596,
      "epoch": 1.2195004803073968,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004858634209698386,
      "loss": 4.8923,
      "mean_token_accuracy": 0.23201584666967393,
      "num_tokens": 29108119.0,
      "step": 12695
    },
    {
      "entropy": 5.25562801361084,
      "epoch": 1.2199807877041307,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048585145798416956,
      "loss": 4.8658,
      "mean_token_accuracy": 0.2246290534734726,
      "num_tokens": 29120605.0,
      "step": 12700
    },
    {
      "entropy": 5.303462409973145,
      "epoch": 1.2204610951008645,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004858394901031558,
      "loss": 4.8737,
      "mean_token_accuracy": 0.2242195725440979,
      "num_tokens": 29131582.0,
      "step": 12705
    },
    {
      "entropy": 5.326866245269775,
      "epoch": 1.2209414024975984,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004858275173270751,
      "loss": 4.9028,
      "mean_token_accuracy": 0.21921043545007707,
      "num_tokens": 29143436.0,
      "step": 12710
    },
    {
      "entropy": 5.2524255275726315,
      "epoch": 1.2214217098943323,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048581553965620553,
      "loss": 4.8186,
      "mean_token_accuracy": 0.2284674435853958,
      "num_tokens": 29154445.0,
      "step": 12715
    },
    {
      "entropy": 5.298490524291992,
      "epoch": 1.2219020172910664,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048580355709082506,
      "loss": 4.9411,
      "mean_token_accuracy": 0.22572432905435563,
      "num_tokens": 29164599.0,
      "step": 12720
    },
    {
      "entropy": 5.241838836669922,
      "epoch": 1.2223823246878003,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000485791569631212,
      "loss": 4.8901,
      "mean_token_accuracy": 0.22316294461488723,
      "num_tokens": 29176275.0,
      "step": 12725
    },
    {
      "entropy": 5.3148106098175045,
      "epoch": 1.2228626320845342,
      "grad_norm": 0.875,
      "learning_rate": 0.0004857795772776446,
      "loss": 4.883,
      "mean_token_accuracy": 0.22649589776992798,
      "num_tokens": 29189102.0,
      "step": 12730
    },
    {
      "entropy": 5.298193645477295,
      "epoch": 1.223342939481268,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048576758003040127,
      "loss": 4.9291,
      "mean_token_accuracy": 0.22039455324411392,
      "num_tokens": 29200953.0,
      "step": 12735
    },
    {
      "entropy": 5.2784913063049315,
      "epoch": 1.223823246878002,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048575557788976066,
      "loss": 4.9398,
      "mean_token_accuracy": 0.2164348617196083,
      "num_tokens": 29212942.0,
      "step": 12740
    },
    {
      "entropy": 5.197840929031372,
      "epoch": 1.2243035542747358,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004857435708560013,
      "loss": 4.819,
      "mean_token_accuracy": 0.23091818541288375,
      "num_tokens": 29224949.0,
      "step": 12745
    },
    {
      "entropy": 5.3517271995544435,
      "epoch": 1.2247838616714697,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048573155892940204,
      "loss": 4.9825,
      "mean_token_accuracy": 0.2224832609295845,
      "num_tokens": 29236044.0,
      "step": 12750
    },
    {
      "entropy": 5.275476026535034,
      "epoch": 1.2252641690682036,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00048571954211024164,
      "loss": 4.8727,
      "mean_token_accuracy": 0.2272889330983162,
      "num_tokens": 29248084.0,
      "step": 12755
    },
    {
      "entropy": 5.26051435470581,
      "epoch": 1.2257444764649374,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048570752039879924,
      "loss": 4.7803,
      "mean_token_accuracy": 0.23204691410064698,
      "num_tokens": 29258710.0,
      "step": 12760
    },
    {
      "entropy": 5.330449724197388,
      "epoch": 1.2262247838616716,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004856954937953539,
      "loss": 4.9707,
      "mean_token_accuracy": 0.21625088453292846,
      "num_tokens": 29270173.0,
      "step": 12765
    },
    {
      "entropy": 5.336314630508423,
      "epoch": 1.2267050912584054,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004856834623001848,
      "loss": 4.9539,
      "mean_token_accuracy": 0.22372423261404037,
      "num_tokens": 29280407.0,
      "step": 12770
    },
    {
      "entropy": 5.229439783096313,
      "epoch": 1.2271853986551393,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004856714259135713,
      "loss": 4.8303,
      "mean_token_accuracy": 0.23523038774728774,
      "num_tokens": 29292287.0,
      "step": 12775
    },
    {
      "entropy": 5.30071530342102,
      "epoch": 1.2276657060518732,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004856593846357929,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21936024576425553,
      "num_tokens": 29303099.0,
      "step": 12780
    },
    {
      "entropy": 5.370990467071533,
      "epoch": 1.228146013448607,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004856473384671291,
      "loss": 4.9892,
      "mean_token_accuracy": 0.22630964815616608,
      "num_tokens": 29314445.0,
      "step": 12785
    },
    {
      "entropy": 5.240963125228882,
      "epoch": 1.228626320845341,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048563528740785955,
      "loss": 4.867,
      "mean_token_accuracy": 0.22688735127449036,
      "num_tokens": 29325309.0,
      "step": 12790
    },
    {
      "entropy": 5.2361877918243405,
      "epoch": 1.2291066282420748,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048562323145826414,
      "loss": 4.8126,
      "mean_token_accuracy": 0.23488469272851945,
      "num_tokens": 29338582.0,
      "step": 12795
    },
    {
      "entropy": 5.2982172012329105,
      "epoch": 1.229586935638809,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004856111706186227,
      "loss": 4.8866,
      "mean_token_accuracy": 0.22783713638782502,
      "num_tokens": 29349875.0,
      "step": 12800
    },
    {
      "entropy": 5.363876581192017,
      "epoch": 1.2300672430355428,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048559910488921534,
      "loss": 4.9951,
      "mean_token_accuracy": 0.21884061098098756,
      "num_tokens": 29361800.0,
      "step": 12805
    },
    {
      "entropy": 5.256297492980957,
      "epoch": 1.2305475504322767,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000485587034270322,
      "loss": 4.8616,
      "mean_token_accuracy": 0.22844843715429305,
      "num_tokens": 29372795.0,
      "step": 12810
    },
    {
      "entropy": 5.290785455703736,
      "epoch": 1.2310278578290106,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000485574958762223,
      "loss": 4.9512,
      "mean_token_accuracy": 0.22574531584978103,
      "num_tokens": 29385391.0,
      "step": 12815
    },
    {
      "entropy": 5.204898357391357,
      "epoch": 1.2315081652257445,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048556287836519886,
      "loss": 4.8019,
      "mean_token_accuracy": 0.2362285614013672,
      "num_tokens": 29397113.0,
      "step": 12820
    },
    {
      "entropy": 5.380714654922485,
      "epoch": 1.2319884726224783,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004855507930795299,
      "loss": 5.0066,
      "mean_token_accuracy": 0.22712673097848893,
      "num_tokens": 29407552.0,
      "step": 12825
    },
    {
      "entropy": 5.296839618682862,
      "epoch": 1.2324687800192122,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048553870290549665,
      "loss": 4.9065,
      "mean_token_accuracy": 0.22649119794368744,
      "num_tokens": 29418500.0,
      "step": 12830
    },
    {
      "entropy": 5.3432575225830075,
      "epoch": 1.232949087415946,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048552660784338,
      "loss": 4.9153,
      "mean_token_accuracy": 0.22437922060489654,
      "num_tokens": 29430335.0,
      "step": 12835
    },
    {
      "entropy": 5.267501878738403,
      "epoch": 1.23342939481268,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004855145078934606,
      "loss": 4.8954,
      "mean_token_accuracy": 0.21701440960168839,
      "num_tokens": 29441435.0,
      "step": 12840
    },
    {
      "entropy": 5.271999025344849,
      "epoch": 1.233909702209414,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004855024030560195,
      "loss": 4.8956,
      "mean_token_accuracy": 0.22557250708341597,
      "num_tokens": 29453140.0,
      "step": 12845
    },
    {
      "entropy": 5.225580072402954,
      "epoch": 1.234390009606148,
      "grad_norm": 1.0,
      "learning_rate": 0.0004854902933313376,
      "loss": 4.8248,
      "mean_token_accuracy": 0.2356981471180916,
      "num_tokens": 29464572.0,
      "step": 12850
    },
    {
      "entropy": 5.358086585998535,
      "epoch": 1.2348703170028819,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048547817871969607,
      "loss": 5.0086,
      "mean_token_accuracy": 0.21887802630662917,
      "num_tokens": 29477069.0,
      "step": 12855
    },
    {
      "entropy": 5.37074613571167,
      "epoch": 1.2353506243996157,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048546605922137633,
      "loss": 4.9196,
      "mean_token_accuracy": 0.22431710064411164,
      "num_tokens": 29486860.0,
      "step": 12860
    },
    {
      "entropy": 5.396909189224243,
      "epoch": 1.2358309317963496,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004854539348366596,
      "loss": 4.9622,
      "mean_token_accuracy": 0.22197997570037842,
      "num_tokens": 29499129.0,
      "step": 12865
    },
    {
      "entropy": 5.2085799217224125,
      "epoch": 1.2363112391930835,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004854418055658274,
      "loss": 4.8399,
      "mean_token_accuracy": 0.23631833791732787,
      "num_tokens": 29510764.0,
      "step": 12870
    },
    {
      "entropy": 5.239119052886963,
      "epoch": 1.2367915465898176,
      "grad_norm": 1.0,
      "learning_rate": 0.00048542967140916134,
      "loss": 4.9265,
      "mean_token_accuracy": 0.2252615809440613,
      "num_tokens": 29522882.0,
      "step": 12875
    },
    {
      "entropy": 5.314130973815918,
      "epoch": 1.2372718539865515,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004854175323669432,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2247759684920311,
      "num_tokens": 29533348.0,
      "step": 12880
    },
    {
      "entropy": 5.2806788921356205,
      "epoch": 1.2377521613832854,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004854053884394547,
      "loss": 4.8618,
      "mean_token_accuracy": 0.22817662805318834,
      "num_tokens": 29545649.0,
      "step": 12885
    },
    {
      "entropy": 5.310440492630005,
      "epoch": 1.2382324687800192,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048539323962697796,
      "loss": 4.8944,
      "mean_token_accuracy": 0.2233073055744171,
      "num_tokens": 29558252.0,
      "step": 12890
    },
    {
      "entropy": 5.363599824905395,
      "epoch": 1.2387127761767531,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004853810859297949,
      "loss": 5.0038,
      "mean_token_accuracy": 0.22439903020858765,
      "num_tokens": 29569495.0,
      "step": 12895
    },
    {
      "entropy": 5.268115663528443,
      "epoch": 1.239193083573487,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048536892734818773,
      "loss": 4.8917,
      "mean_token_accuracy": 0.22725153267383574,
      "num_tokens": 29582167.0,
      "step": 12900
    },
    {
      "entropy": 5.311830520629883,
      "epoch": 1.239673390970221,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004853567638824387,
      "loss": 4.9167,
      "mean_token_accuracy": 0.22679216265678406,
      "num_tokens": 29593964.0,
      "step": 12905
    },
    {
      "entropy": 5.283583641052246,
      "epoch": 1.2401536983669548,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048534459553283026,
      "loss": 4.8691,
      "mean_token_accuracy": 0.2311812534928322,
      "num_tokens": 29604844.0,
      "step": 12910
    },
    {
      "entropy": 5.312759828567505,
      "epoch": 1.2406340057636887,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004853324222996449,
      "loss": 4.9144,
      "mean_token_accuracy": 0.22637501060962678,
      "num_tokens": 29615219.0,
      "step": 12915
    },
    {
      "entropy": 5.392581987380981,
      "epoch": 1.2411143131604228,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048532024418316525,
      "loss": 4.9917,
      "mean_token_accuracy": 0.21924567371606826,
      "num_tokens": 29626472.0,
      "step": 12920
    },
    {
      "entropy": 5.264899873733521,
      "epoch": 1.2415946205571566,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004853080611836741,
      "loss": 4.8673,
      "mean_token_accuracy": 0.2329067125916481,
      "num_tokens": 29637966.0,
      "step": 12925
    },
    {
      "entropy": 5.212028551101684,
      "epoch": 1.2420749279538905,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048529587330145427,
      "loss": 4.8158,
      "mean_token_accuracy": 0.22972645312547685,
      "num_tokens": 29648730.0,
      "step": 12930
    },
    {
      "entropy": 5.220188093185425,
      "epoch": 1.2425552353506244,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048528368053678863,
      "loss": 4.8334,
      "mean_token_accuracy": 0.2270681917667389,
      "num_tokens": 29660576.0,
      "step": 12935
    },
    {
      "entropy": 5.233583688735962,
      "epoch": 1.2430355427473583,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004852714828899604,
      "loss": 4.8495,
      "mean_token_accuracy": 0.23300411254167558,
      "num_tokens": 29672906.0,
      "step": 12940
    },
    {
      "entropy": 5.251200342178345,
      "epoch": 1.2435158501440922,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048525928036125264,
      "loss": 4.8408,
      "mean_token_accuracy": 0.23513444364070893,
      "num_tokens": 29685360.0,
      "step": 12945
    },
    {
      "entropy": 5.28692946434021,
      "epoch": 1.243996157540826,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048524707295094884,
      "loss": 4.8589,
      "mean_token_accuracy": 0.219436913728714,
      "num_tokens": 29697257.0,
      "step": 12950
    },
    {
      "entropy": 5.225015258789062,
      "epoch": 1.2444764649375601,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004852348606593322,
      "loss": 4.8222,
      "mean_token_accuracy": 0.22740707397460938,
      "num_tokens": 29707877.0,
      "step": 12955
    },
    {
      "entropy": 5.281049680709839,
      "epoch": 1.244956772334294,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048522264348668646,
      "loss": 4.8808,
      "mean_token_accuracy": 0.22606362998485566,
      "num_tokens": 29719358.0,
      "step": 12960
    },
    {
      "entropy": 5.221434831619263,
      "epoch": 1.245437079731028,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004852104214332951,
      "loss": 4.7549,
      "mean_token_accuracy": 0.2435452163219452,
      "num_tokens": 29730383.0,
      "step": 12965
    },
    {
      "entropy": 5.302107429504394,
      "epoch": 1.2459173871277618,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048519819449944205,
      "loss": 4.897,
      "mean_token_accuracy": 0.23018603026866913,
      "num_tokens": 29741142.0,
      "step": 12970
    },
    {
      "entropy": 5.204427480697632,
      "epoch": 1.2463976945244957,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000485185962685411,
      "loss": 4.8562,
      "mean_token_accuracy": 0.23069577515125275,
      "num_tokens": 29754618.0,
      "step": 12975
    },
    {
      "entropy": 5.28589277267456,
      "epoch": 1.2468780019212296,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000485173725991486,
      "loss": 4.8616,
      "mean_token_accuracy": 0.23017289042472838,
      "num_tokens": 29767115.0,
      "step": 12980
    },
    {
      "entropy": 5.320121908187867,
      "epoch": 1.2473583093179634,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048516148441795124,
      "loss": 4.9419,
      "mean_token_accuracy": 0.22126028835773467,
      "num_tokens": 29778165.0,
      "step": 12985
    },
    {
      "entropy": 5.31697449684143,
      "epoch": 1.2478386167146973,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004851492379650908,
      "loss": 5.0105,
      "mean_token_accuracy": 0.21841544508934022,
      "num_tokens": 29790528.0,
      "step": 12990
    },
    {
      "entropy": 5.260398483276367,
      "epoch": 1.2483189241114312,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004851369866331891,
      "loss": 4.8146,
      "mean_token_accuracy": 0.22849899530410767,
      "num_tokens": 29801709.0,
      "step": 12995
    },
    {
      "entropy": 5.305760478973388,
      "epoch": 1.2487992315081653,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004851247304225306,
      "loss": 4.8989,
      "mean_token_accuracy": 0.22332677841186524,
      "num_tokens": 29812963.0,
      "step": 13000
    },
    {
      "entropy": 5.334535932540893,
      "epoch": 1.2492795389048992,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004851124693333997,
      "loss": 4.8725,
      "mean_token_accuracy": 0.22705477327108384,
      "num_tokens": 29823711.0,
      "step": 13005
    },
    {
      "entropy": 5.259051513671875,
      "epoch": 1.249759846301633,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004851002033660812,
      "loss": 4.8451,
      "mean_token_accuracy": 0.23193236142396928,
      "num_tokens": 29834038.0,
      "step": 13010
    },
    {
      "entropy": 5.2705339908599855,
      "epoch": 1.250240153698367,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048508793252085994,
      "loss": 4.8683,
      "mean_token_accuracy": 0.2312237486243248,
      "num_tokens": 29844759.0,
      "step": 13015
    },
    {
      "entropy": 5.166585683822632,
      "epoch": 1.2507204610951008,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004850756567980206,
      "loss": 4.7388,
      "mean_token_accuracy": 0.23773652613162993,
      "num_tokens": 29855643.0,
      "step": 13020
    },
    {
      "entropy": 5.3025603771209715,
      "epoch": 1.2512007684918347,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048506337619784836,
      "loss": 4.9453,
      "mean_token_accuracy": 0.22435982674360275,
      "num_tokens": 29866917.0,
      "step": 13025
    },
    {
      "entropy": 5.34097695350647,
      "epoch": 1.2516810758885688,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004850510907206283,
      "loss": 5.0271,
      "mean_token_accuracy": 0.2181246742606163,
      "num_tokens": 29878937.0,
      "step": 13030
    },
    {
      "entropy": 5.217430591583252,
      "epoch": 1.2521613832853027,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048503880036664555,
      "loss": 4.8311,
      "mean_token_accuracy": 0.23065926730632783,
      "num_tokens": 29889544.0,
      "step": 13035
    },
    {
      "entropy": 5.228662919998169,
      "epoch": 1.2526416906820366,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004850265051361857,
      "loss": 4.8496,
      "mean_token_accuracy": 0.22884223610162735,
      "num_tokens": 29901919.0,
      "step": 13040
    },
    {
      "entropy": 5.2675981521606445,
      "epoch": 1.2531219980787704,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004850142050295339,
      "loss": 4.8975,
      "mean_token_accuracy": 0.22509329617023469,
      "num_tokens": 29913870.0,
      "step": 13045
    },
    {
      "entropy": 5.2230833053588865,
      "epoch": 1.2536023054755043,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048500190004697595,
      "loss": 4.8399,
      "mean_token_accuracy": 0.23583188205957412,
      "num_tokens": 29925180.0,
      "step": 13050
    },
    {
      "entropy": 5.16978759765625,
      "epoch": 1.2540826128722382,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004849895901887974,
      "loss": 4.7873,
      "mean_token_accuracy": 0.23792642951011658,
      "num_tokens": 29936433.0,
      "step": 13055
    },
    {
      "entropy": 5.327881813049316,
      "epoch": 1.254562920268972,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004849772754552842,
      "loss": 4.9437,
      "mean_token_accuracy": 0.22192156463861465,
      "num_tokens": 29948891.0,
      "step": 13060
    },
    {
      "entropy": 5.240267705917359,
      "epoch": 1.255043227665706,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048496495584672214,
      "loss": 4.7633,
      "mean_token_accuracy": 0.23425445854663848,
      "num_tokens": 29960113.0,
      "step": 13065
    },
    {
      "entropy": 5.294047451019287,
      "epoch": 1.2555235350624399,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048495263136339725,
      "loss": 4.9089,
      "mean_token_accuracy": 0.2252771481871605,
      "num_tokens": 29972168.0,
      "step": 13070
    },
    {
      "entropy": 5.2826474666595455,
      "epoch": 1.2560038424591737,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004849403020055956,
      "loss": 4.9193,
      "mean_token_accuracy": 0.2262105628848076,
      "num_tokens": 29982996.0,
      "step": 13075
    },
    {
      "entropy": 5.2494645595550535,
      "epoch": 1.2564841498559078,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048492796777360373,
      "loss": 4.9021,
      "mean_token_accuracy": 0.22427692413330078,
      "num_tokens": 29994088.0,
      "step": 13080
    },
    {
      "entropy": 5.362490224838257,
      "epoch": 1.2569644572526417,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048491562866770767,
      "loss": 4.9645,
      "mean_token_accuracy": 0.22577129751443864,
      "num_tokens": 30005403.0,
      "step": 13085
    },
    {
      "entropy": 5.286747694015503,
      "epoch": 1.2574447646493756,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048490328468819404,
      "loss": 4.8853,
      "mean_token_accuracy": 0.23747577220201493,
      "num_tokens": 30015961.0,
      "step": 13090
    },
    {
      "entropy": 5.221665239334106,
      "epoch": 1.2579250720461095,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048489093583534945,
      "loss": 4.8435,
      "mean_token_accuracy": 0.22955506443977355,
      "num_tokens": 30026670.0,
      "step": 13095
    },
    {
      "entropy": 5.233548974990844,
      "epoch": 1.2584053794428434,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004848785821094606,
      "loss": 4.866,
      "mean_token_accuracy": 0.22804583758115768,
      "num_tokens": 30036711.0,
      "step": 13100
    },
    {
      "entropy": 5.271080827713012,
      "epoch": 1.2588856868395775,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004848662235108142,
      "loss": 4.9387,
      "mean_token_accuracy": 0.2245746746659279,
      "num_tokens": 30047587.0,
      "step": 13105
    },
    {
      "entropy": 5.251093101501465,
      "epoch": 1.2593659942363113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004848538600396973,
      "loss": 4.8853,
      "mean_token_accuracy": 0.22791888564825058,
      "num_tokens": 30059348.0,
      "step": 13110
    },
    {
      "entropy": 5.31338415145874,
      "epoch": 1.2598463016330452,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048484149169639694,
      "loss": 4.8975,
      "mean_token_accuracy": 0.22811620980501174,
      "num_tokens": 30070485.0,
      "step": 13115
    },
    {
      "entropy": 5.2312568664550785,
      "epoch": 1.260326609029779,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004848291184812003,
      "loss": 4.8123,
      "mean_token_accuracy": 0.23672863841056824,
      "num_tokens": 30081114.0,
      "step": 13120
    },
    {
      "entropy": 5.259006023406982,
      "epoch": 1.260806916426513,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004848167403943945,
      "loss": 4.9411,
      "mean_token_accuracy": 0.22369351387023925,
      "num_tokens": 30092634.0,
      "step": 13125
    },
    {
      "entropy": 5.258079099655151,
      "epoch": 1.2612872238232469,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048480435743626703,
      "loss": 4.8793,
      "mean_token_accuracy": 0.23210976719856263,
      "num_tokens": 30104205.0,
      "step": 13130
    },
    {
      "entropy": 5.290403699874878,
      "epoch": 1.2617675312199808,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004847919696071054,
      "loss": 4.8977,
      "mean_token_accuracy": 0.22457866817712785,
      "num_tokens": 30116978.0,
      "step": 13135
    },
    {
      "entropy": 5.235513734817505,
      "epoch": 1.2622478386167146,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048477957690719716,
      "loss": 4.792,
      "mean_token_accuracy": 0.23143470883369446,
      "num_tokens": 30128549.0,
      "step": 13140
    },
    {
      "entropy": 5.2985937118530275,
      "epoch": 1.2627281460134485,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004847671793368301,
      "loss": 4.9281,
      "mean_token_accuracy": 0.22890064120292664,
      "num_tokens": 30139492.0,
      "step": 13145
    },
    {
      "entropy": 5.2714262962341305,
      "epoch": 1.2632084534101824,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000484754776896292,
      "loss": 4.8651,
      "mean_token_accuracy": 0.22884595841169358,
      "num_tokens": 30150450.0,
      "step": 13150
    },
    {
      "entropy": 5.2889257907867435,
      "epoch": 1.2636887608069165,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004847423695858708,
      "loss": 4.9202,
      "mean_token_accuracy": 0.2286774769425392,
      "num_tokens": 30162204.0,
      "step": 13155
    },
    {
      "entropy": 5.3097892761230465,
      "epoch": 1.2641690682036504,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048472995740585456,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22797740995883942,
      "num_tokens": 30172574.0,
      "step": 13160
    },
    {
      "entropy": 5.184846878051758,
      "epoch": 1.2646493756003843,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004847175403565316,
      "loss": 4.7842,
      "mean_token_accuracy": 0.2362135887145996,
      "num_tokens": 30183957.0,
      "step": 13165
    },
    {
      "entropy": 5.2937885284423825,
      "epoch": 1.2651296829971181,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048470511843818996,
      "loss": 4.8621,
      "mean_token_accuracy": 0.23150158524513245,
      "num_tokens": 30194207.0,
      "step": 13170
    },
    {
      "entropy": 5.207911777496338,
      "epoch": 1.265609990393852,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004846926916511182,
      "loss": 4.8231,
      "mean_token_accuracy": 0.24067401736974717,
      "num_tokens": 30205180.0,
      "step": 13175
    },
    {
      "entropy": 5.244674587249756,
      "epoch": 1.266090297790586,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004846802599956048,
      "loss": 4.8759,
      "mean_token_accuracy": 0.2271193727850914,
      "num_tokens": 30217734.0,
      "step": 13180
    },
    {
      "entropy": 5.352945280075073,
      "epoch": 1.26657060518732,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048466782347193847,
      "loss": 5.0931,
      "mean_token_accuracy": 0.2140210285782814,
      "num_tokens": 30229835.0,
      "step": 13185
    },
    {
      "entropy": 5.313801002502442,
      "epoch": 1.267050912584054,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048465538208040775,
      "loss": 4.9198,
      "mean_token_accuracy": 0.22783443182706833,
      "num_tokens": 30241932.0,
      "step": 13190
    },
    {
      "entropy": 5.312128162384033,
      "epoch": 1.2675312199807878,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048464293582130166,
      "loss": 4.9354,
      "mean_token_accuracy": 0.22157906889915466,
      "num_tokens": 30253149.0,
      "step": 13195
    },
    {
      "entropy": 5.308742618560791,
      "epoch": 1.2680115273775217,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004846304846949091,
      "loss": 4.8381,
      "mean_token_accuracy": 0.22973738610744476,
      "num_tokens": 30264083.0,
      "step": 13200
    },
    {
      "entropy": 5.2179765701293945,
      "epoch": 1.2684918347742555,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048461802870151916,
      "loss": 4.821,
      "mean_token_accuracy": 0.2336911678314209,
      "num_tokens": 30274832.0,
      "step": 13205
    },
    {
      "entropy": 5.207124900817871,
      "epoch": 1.2689721421709894,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048460556784142106,
      "loss": 4.8464,
      "mean_token_accuracy": 0.22988076359033585,
      "num_tokens": 30284945.0,
      "step": 13210
    },
    {
      "entropy": 5.325016641616822,
      "epoch": 1.2694524495677233,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048459310211490406,
      "loss": 4.8828,
      "mean_token_accuracy": 0.2339743807911873,
      "num_tokens": 30295133.0,
      "step": 13215
    },
    {
      "entropy": 5.2552345275878904,
      "epoch": 1.2699327569644572,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004845806315222576,
      "loss": 4.9334,
      "mean_token_accuracy": 0.22475293427705764,
      "num_tokens": 30305268.0,
      "step": 13220
    },
    {
      "entropy": 5.196200275421143,
      "epoch": 1.270413064361191,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004845681560637711,
      "loss": 4.8166,
      "mean_token_accuracy": 0.23148071467876435,
      "num_tokens": 30317118.0,
      "step": 13225
    },
    {
      "entropy": 5.306098890304566,
      "epoch": 1.270893371757925,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004845556757397344,
      "loss": 4.9647,
      "mean_token_accuracy": 0.2200284257531166,
      "num_tokens": 30328684.0,
      "step": 13230
    },
    {
      "entropy": 5.339798450469971,
      "epoch": 1.271373679154659,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004845431905504372,
      "loss": 4.9724,
      "mean_token_accuracy": 0.22104789018630983,
      "num_tokens": 30341433.0,
      "step": 13235
    },
    {
      "entropy": 5.273892116546631,
      "epoch": 1.271853986551393,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048453070049616926,
      "loss": 4.8474,
      "mean_token_accuracy": 0.2382711336016655,
      "num_tokens": 30353159.0,
      "step": 13240
    },
    {
      "entropy": 5.245979070663452,
      "epoch": 1.2723342939481268,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048451820557722064,
      "loss": 4.9277,
      "mean_token_accuracy": 0.2237758755683899,
      "num_tokens": 30363251.0,
      "step": 13245
    },
    {
      "entropy": 5.373185729980468,
      "epoch": 1.2728146013448607,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004845057057938815,
      "loss": 4.9683,
      "mean_token_accuracy": 0.22629047185182571,
      "num_tokens": 30375850.0,
      "step": 13250
    },
    {
      "entropy": 5.338417482376099,
      "epoch": 1.2732949087415946,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048449320114644185,
      "loss": 4.9568,
      "mean_token_accuracy": 0.21990599483251572,
      "num_tokens": 30386839.0,
      "step": 13255
    },
    {
      "entropy": 5.267644882202148,
      "epoch": 1.2737752161383284,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004844806916351922,
      "loss": 4.9378,
      "mean_token_accuracy": 0.21919597089290618,
      "num_tokens": 30398872.0,
      "step": 13260
    },
    {
      "entropy": 5.28463397026062,
      "epoch": 1.2742555235350626,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004844681772604229,
      "loss": 4.8689,
      "mean_token_accuracy": 0.2361966997385025,
      "num_tokens": 30409581.0,
      "step": 13265
    },
    {
      "entropy": 5.264424657821655,
      "epoch": 1.2747358309317964,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048445565802242454,
      "loss": 4.8788,
      "mean_token_accuracy": 0.2281175285577774,
      "num_tokens": 30420209.0,
      "step": 13270
    },
    {
      "entropy": 5.311786031723022,
      "epoch": 1.2752161383285303,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004844431339214878,
      "loss": 4.9297,
      "mean_token_accuracy": 0.23335610777139665,
      "num_tokens": 30432093.0,
      "step": 13275
    },
    {
      "entropy": 5.3059694290161135,
      "epoch": 1.2756964457252642,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004844306049579034,
      "loss": 4.8436,
      "mean_token_accuracy": 0.22806389182806014,
      "num_tokens": 30442796.0,
      "step": 13280
    },
    {
      "entropy": 5.264642667770386,
      "epoch": 1.276176753121998,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048441807113196216,
      "loss": 4.8689,
      "mean_token_accuracy": 0.22561000734567643,
      "num_tokens": 30455226.0,
      "step": 13285
    },
    {
      "entropy": 5.286166858673096,
      "epoch": 1.276657060518732,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048440553244395517,
      "loss": 4.9788,
      "mean_token_accuracy": 0.2189204588532448,
      "num_tokens": 30467082.0,
      "step": 13290
    },
    {
      "entropy": 5.356497383117675,
      "epoch": 1.2771373679154658,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048439298889417357,
      "loss": 4.8874,
      "mean_token_accuracy": 0.22182206213474273,
      "num_tokens": 30479051.0,
      "step": 13295
    },
    {
      "entropy": 5.332766962051392,
      "epoch": 1.2776176753121997,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048438044048290847,
      "loss": 4.9207,
      "mean_token_accuracy": 0.22814682871103287,
      "num_tokens": 30489989.0,
      "step": 13300
    },
    {
      "entropy": 5.266085290908814,
      "epoch": 1.2780979827089336,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048436788721045135,
      "loss": 4.8231,
      "mean_token_accuracy": 0.23183311223983766,
      "num_tokens": 30501533.0,
      "step": 13305
    },
    {
      "entropy": 5.276353454589843,
      "epoch": 1.2785782901056677,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004843553290770935,
      "loss": 4.9194,
      "mean_token_accuracy": 0.2249412104487419,
      "num_tokens": 30512231.0,
      "step": 13310
    },
    {
      "entropy": 5.260562372207642,
      "epoch": 1.2790585975024016,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004843427660831266,
      "loss": 4.8919,
      "mean_token_accuracy": 0.22729096412658692,
      "num_tokens": 30523204.0,
      "step": 13315
    },
    {
      "entropy": 5.368788385391236,
      "epoch": 1.2795389048991355,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048433019822884235,
      "loss": 5.0104,
      "mean_token_accuracy": 0.21593842804431915,
      "num_tokens": 30534956.0,
      "step": 13320
    },
    {
      "entropy": 5.24132137298584,
      "epoch": 1.2800192122958693,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004843176255145325,
      "loss": 4.8603,
      "mean_token_accuracy": 0.23657788038253785,
      "num_tokens": 30545938.0,
      "step": 13325
    },
    {
      "entropy": 5.266680431365967,
      "epoch": 1.2804995196926032,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004843050479404888,
      "loss": 4.7993,
      "mean_token_accuracy": 0.23469994217157364,
      "num_tokens": 30557323.0,
      "step": 13330
    },
    {
      "entropy": 5.293450736999512,
      "epoch": 1.280979827089337,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048429246550700343,
      "loss": 4.8683,
      "mean_token_accuracy": 0.2319569021463394,
      "num_tokens": 30569960.0,
      "step": 13335
    },
    {
      "entropy": 5.240808439254761,
      "epoch": 1.2814601344860712,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004842798782143686,
      "loss": 4.9061,
      "mean_token_accuracy": 0.2297001376748085,
      "num_tokens": 30581904.0,
      "step": 13340
    },
    {
      "entropy": 5.275762748718262,
      "epoch": 1.281940441882805,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048426728606287627,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2276452124118805,
      "num_tokens": 30592955.0,
      "step": 13345
    },
    {
      "entropy": 5.315696430206299,
      "epoch": 1.282420749279539,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004842546890528191,
      "loss": 4.9004,
      "mean_token_accuracy": 0.22609656900167466,
      "num_tokens": 30604020.0,
      "step": 13350
    },
    {
      "entropy": 5.260344505310059,
      "epoch": 1.2829010566762729,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004842420871844893,
      "loss": 4.8598,
      "mean_token_accuracy": 0.22626498490571975,
      "num_tokens": 30615623.0,
      "step": 13355
    },
    {
      "entropy": 5.3137672424316404,
      "epoch": 1.2833813640730067,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004842294804581796,
      "loss": 4.9115,
      "mean_token_accuracy": 0.22739534825086594,
      "num_tokens": 30626258.0,
      "step": 13360
    },
    {
      "entropy": 5.300073719024658,
      "epoch": 1.2838616714697406,
      "grad_norm": 0.875,
      "learning_rate": 0.00048421686887418266,
      "loss": 4.9022,
      "mean_token_accuracy": 0.22163822501897812,
      "num_tokens": 30637861.0,
      "step": 13365
    },
    {
      "entropy": 5.327418184280395,
      "epoch": 1.2843419788664745,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004842042524327912,
      "loss": 4.9076,
      "mean_token_accuracy": 0.22947875112295152,
      "num_tokens": 30648835.0,
      "step": 13370
    },
    {
      "entropy": 5.365424108505249,
      "epoch": 1.2848222862632084,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004841916311342983,
      "loss": 4.9729,
      "mean_token_accuracy": 0.21744562685489655,
      "num_tokens": 30659117.0,
      "step": 13375
    },
    {
      "entropy": 5.2717735290527346,
      "epoch": 1.2853025936599423,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004841790049789969,
      "loss": 4.8861,
      "mean_token_accuracy": 0.22931284308433533,
      "num_tokens": 30670282.0,
      "step": 13380
    },
    {
      "entropy": 5.266284322738647,
      "epoch": 1.2857829010566761,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048416637396718004,
      "loss": 4.9123,
      "mean_token_accuracy": 0.21836647689342498,
      "num_tokens": 30681967.0,
      "step": 13385
    },
    {
      "entropy": 5.355836868286133,
      "epoch": 1.2862632084534102,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004841537380991411,
      "loss": 4.8901,
      "mean_token_accuracy": 0.22283714413642883,
      "num_tokens": 30692803.0,
      "step": 13390
    },
    {
      "entropy": 5.26470799446106,
      "epoch": 1.2867435158501441,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048414109737517346,
      "loss": 4.7916,
      "mean_token_accuracy": 0.23633354902267456,
      "num_tokens": 30704097.0,
      "step": 13395
    },
    {
      "entropy": 5.296727705001831,
      "epoch": 1.287223823246878,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004841284517955706,
      "loss": 4.9521,
      "mean_token_accuracy": 0.21994349956512452,
      "num_tokens": 30717983.0,
      "step": 13400
    },
    {
      "entropy": 5.176114130020141,
      "epoch": 1.2877041306436119,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000484115801360626,
      "loss": 4.7525,
      "mean_token_accuracy": 0.2392497777938843,
      "num_tokens": 30728446.0,
      "step": 13405
    },
    {
      "entropy": 5.196917009353638,
      "epoch": 1.2881844380403458,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004841031460706335,
      "loss": 4.7191,
      "mean_token_accuracy": 0.2364543244242668,
      "num_tokens": 30739587.0,
      "step": 13410
    },
    {
      "entropy": 5.242478942871093,
      "epoch": 1.2886647454370797,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048409048592588683,
      "loss": 4.8536,
      "mean_token_accuracy": 0.2295059084892273,
      "num_tokens": 30750093.0,
      "step": 13415
    },
    {
      "entropy": 5.266240835189819,
      "epoch": 1.2891450528338138,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004840778209266799,
      "loss": 4.8833,
      "mean_token_accuracy": 0.23077643066644668,
      "num_tokens": 30761692.0,
      "step": 13420
    },
    {
      "entropy": 5.23282823562622,
      "epoch": 1.2896253602305476,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048406515107330685,
      "loss": 4.8243,
      "mean_token_accuracy": 0.233749595284462,
      "num_tokens": 30773474.0,
      "step": 13425
    },
    {
      "entropy": 5.37356595993042,
      "epoch": 1.2901056676272815,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048405247636606173,
      "loss": 4.9843,
      "mean_token_accuracy": 0.21992174088954924,
      "num_tokens": 30785464.0,
      "step": 13430
    },
    {
      "entropy": 5.277270650863647,
      "epoch": 1.2905859750240154,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048403979680523894,
      "loss": 4.8963,
      "mean_token_accuracy": 0.2226998522877693,
      "num_tokens": 30796343.0,
      "step": 13435
    },
    {
      "entropy": 5.311119937896729,
      "epoch": 1.2910662824207493,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004840271123911328,
      "loss": 4.8672,
      "mean_token_accuracy": 0.23047288954257966,
      "num_tokens": 30807795.0,
      "step": 13440
    },
    {
      "entropy": 5.287996387481689,
      "epoch": 1.2915465898174832,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004840144231240377,
      "loss": 4.8793,
      "mean_token_accuracy": 0.226242034137249,
      "num_tokens": 30819629.0,
      "step": 13445
    },
    {
      "entropy": 5.222072267532349,
      "epoch": 1.292026897214217,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004840017290042484,
      "loss": 4.8243,
      "mean_token_accuracy": 0.23122052401304244,
      "num_tokens": 30831545.0,
      "step": 13450
    },
    {
      "entropy": 5.264734697341919,
      "epoch": 1.292507204610951,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048398903003205957,
      "loss": 4.8829,
      "mean_token_accuracy": 0.22700888365507127,
      "num_tokens": 30843614.0,
      "step": 13455
    },
    {
      "entropy": 5.253847408294678,
      "epoch": 1.2929875120076848,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048397632620776604,
      "loss": 4.7963,
      "mean_token_accuracy": 0.23053706288337708,
      "num_tokens": 30853749.0,
      "step": 13460
    },
    {
      "entropy": 5.2437526226043705,
      "epoch": 1.293467819404419,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048396361753166276,
      "loss": 4.8182,
      "mean_token_accuracy": 0.23231911808252334,
      "num_tokens": 30865065.0,
      "step": 13465
    },
    {
      "entropy": 5.276921653747559,
      "epoch": 1.2939481268011528,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048395090400404466,
      "loss": 4.8953,
      "mean_token_accuracy": 0.2245018735527992,
      "num_tokens": 30876746.0,
      "step": 13470
    },
    {
      "entropy": 5.332549524307251,
      "epoch": 1.2944284341978867,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048393818562520715,
      "loss": 4.9432,
      "mean_token_accuracy": 0.22525248527526856,
      "num_tokens": 30889543.0,
      "step": 13475
    },
    {
      "entropy": 5.286140966415405,
      "epoch": 1.2949087415946205,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048392546239544535,
      "loss": 4.9233,
      "mean_token_accuracy": 0.22417937070131302,
      "num_tokens": 30901949.0,
      "step": 13480
    },
    {
      "entropy": 5.291250467300415,
      "epoch": 1.2953890489913544,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004839127343150547,
      "loss": 4.9122,
      "mean_token_accuracy": 0.22542797178030013,
      "num_tokens": 30912660.0,
      "step": 13485
    },
    {
      "entropy": 5.27012529373169,
      "epoch": 1.2958693563880883,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004839000013843307,
      "loss": 4.8852,
      "mean_token_accuracy": 0.2297861710190773,
      "num_tokens": 30924276.0,
      "step": 13490
    },
    {
      "entropy": 5.280837106704712,
      "epoch": 1.2963496637848224,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048388726360356894,
      "loss": 4.8155,
      "mean_token_accuracy": 0.23490381687879563,
      "num_tokens": 30935169.0,
      "step": 13495
    },
    {
      "entropy": 5.375510931015015,
      "epoch": 1.2968299711815563,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004838745209730653,
      "loss": 4.9881,
      "mean_token_accuracy": 0.21939805299043655,
      "num_tokens": 30946116.0,
      "step": 13500
    },
    {
      "entropy": 5.309035873413086,
      "epoch": 1.2973102785782902,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048386177349311535,
      "loss": 4.9138,
      "mean_token_accuracy": 0.2286698967218399,
      "num_tokens": 30958509.0,
      "step": 13505
    },
    {
      "entropy": 5.320603656768799,
      "epoch": 1.297790585975024,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004838490211640153,
      "loss": 4.9178,
      "mean_token_accuracy": 0.23016474545001983,
      "num_tokens": 30969501.0,
      "step": 13510
    },
    {
      "entropy": 5.266346454620361,
      "epoch": 1.298270893371758,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004838362639860611,
      "loss": 4.8306,
      "mean_token_accuracy": 0.2308288484811783,
      "num_tokens": 30979768.0,
      "step": 13515
    },
    {
      "entropy": 5.207309818267822,
      "epoch": 1.2987512007684918,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000483823501959549,
      "loss": 4.8281,
      "mean_token_accuracy": 0.23510258942842482,
      "num_tokens": 30992020.0,
      "step": 13520
    },
    {
      "entropy": 5.337653350830078,
      "epoch": 1.2992315081652257,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048381073508477527,
      "loss": 4.957,
      "mean_token_accuracy": 0.22191729694604873,
      "num_tokens": 31004081.0,
      "step": 13525
    },
    {
      "entropy": 5.405849599838257,
      "epoch": 1.2997118155619596,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048379796336203625,
      "loss": 4.9953,
      "mean_token_accuracy": 0.21514866948127748,
      "num_tokens": 31015776.0,
      "step": 13530
    },
    {
      "entropy": 5.282814168930054,
      "epoch": 1.3001921229586935,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004837851867916286,
      "loss": 4.814,
      "mean_token_accuracy": 0.22947711497545242,
      "num_tokens": 31028066.0,
      "step": 13535
    },
    {
      "entropy": 5.216132593154907,
      "epoch": 1.3006724303554273,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004837724053738488,
      "loss": 4.862,
      "mean_token_accuracy": 0.2343861311674118,
      "num_tokens": 31040234.0,
      "step": 13540
    },
    {
      "entropy": 5.301957893371582,
      "epoch": 1.3011527377521614,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048375961910899373,
      "loss": 4.9069,
      "mean_token_accuracy": 0.22507015168666838,
      "num_tokens": 31051158.0,
      "step": 13545
    },
    {
      "entropy": 5.363882780075073,
      "epoch": 1.3016330451488953,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004837468279973602,
      "loss": 4.9098,
      "mean_token_accuracy": 0.23003066927194596,
      "num_tokens": 31063404.0,
      "step": 13550
    },
    {
      "entropy": 5.28774790763855,
      "epoch": 1.3021133525456292,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004837340320392451,
      "loss": 4.8736,
      "mean_token_accuracy": 0.23003602474927903,
      "num_tokens": 31073845.0,
      "step": 13555
    },
    {
      "entropy": 5.238974905014038,
      "epoch": 1.302593659942363,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048372123123494563,
      "loss": 4.8953,
      "mean_token_accuracy": 0.22424570918083192,
      "num_tokens": 31086914.0,
      "step": 13560
    },
    {
      "entropy": 5.292280530929565,
      "epoch": 1.303073967339097,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000483708425584759,
      "loss": 4.8312,
      "mean_token_accuracy": 0.23561133742332457,
      "num_tokens": 31098627.0,
      "step": 13565
    },
    {
      "entropy": 5.383846569061279,
      "epoch": 1.3035542747358309,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004836956150889825,
      "loss": 5.084,
      "mean_token_accuracy": 0.22011065781116484,
      "num_tokens": 31110064.0,
      "step": 13570
    },
    {
      "entropy": 5.2750279903411865,
      "epoch": 1.304034582132565,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004836827997479134,
      "loss": 4.8379,
      "mean_token_accuracy": 0.23587071746587754,
      "num_tokens": 31121572.0,
      "step": 13575
    },
    {
      "entropy": 5.366293716430664,
      "epoch": 1.3045148895292988,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004836699795618494,
      "loss": 4.9631,
      "mean_token_accuracy": 0.22241220474243165,
      "num_tokens": 31133276.0,
      "step": 13580
    },
    {
      "entropy": 5.325075149536133,
      "epoch": 1.3049951969260327,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004836571545310881,
      "loss": 4.9384,
      "mean_token_accuracy": 0.22693736106157303,
      "num_tokens": 31144524.0,
      "step": 13585
    },
    {
      "entropy": 5.258250188827515,
      "epoch": 1.3054755043227666,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048364432465592723,
      "loss": 4.8147,
      "mean_token_accuracy": 0.2377645492553711,
      "num_tokens": 31155775.0,
      "step": 13590
    },
    {
      "entropy": 5.265634775161743,
      "epoch": 1.3059558117195005,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004836314899366647,
      "loss": 4.809,
      "mean_token_accuracy": 0.2340649649500847,
      "num_tokens": 31166176.0,
      "step": 13595
    },
    {
      "entropy": 5.30874285697937,
      "epoch": 1.3064361191162344,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048361865037359846,
      "loss": 4.857,
      "mean_token_accuracy": 0.2338636502623558,
      "num_tokens": 31178568.0,
      "step": 13600
    },
    {
      "entropy": 5.334344387054443,
      "epoch": 1.3069164265129682,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048360580596702664,
      "loss": 4.9321,
      "mean_token_accuracy": 0.2239208310842514,
      "num_tokens": 31188897.0,
      "step": 13605
    },
    {
      "entropy": 5.290353298187256,
      "epoch": 1.3073967339097021,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048359295671724744,
      "loss": 4.8936,
      "mean_token_accuracy": 0.22775744199752807,
      "num_tokens": 31201370.0,
      "step": 13610
    },
    {
      "entropy": 5.263905429840088,
      "epoch": 1.307877041306436,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004835801026245592,
      "loss": 4.9568,
      "mean_token_accuracy": 0.22285640984773636,
      "num_tokens": 31212704.0,
      "step": 13615
    },
    {
      "entropy": 5.293863010406494,
      "epoch": 1.30835734870317,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048356724368926035,
      "loss": 4.9166,
      "mean_token_accuracy": 0.2302747517824173,
      "num_tokens": 31224447.0,
      "step": 13620
    },
    {
      "entropy": 5.3539848804473875,
      "epoch": 1.308837656099904,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048355437991164937,
      "loss": 4.9408,
      "mean_token_accuracy": 0.22849018424749373,
      "num_tokens": 31235648.0,
      "step": 13625
    },
    {
      "entropy": 5.267015409469605,
      "epoch": 1.3093179634966379,
      "grad_norm": 1.015625,
      "learning_rate": 0.000483541511292025,
      "loss": 4.9381,
      "mean_token_accuracy": 0.22321582436561585,
      "num_tokens": 31248330.0,
      "step": 13630
    },
    {
      "entropy": 5.320993328094483,
      "epoch": 1.3097982708933718,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048352863783068594,
      "loss": 4.971,
      "mean_token_accuracy": 0.22274261713027954,
      "num_tokens": 31260036.0,
      "step": 13635
    },
    {
      "entropy": 5.227537012100219,
      "epoch": 1.3102785782901056,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048351575952793117,
      "loss": 4.7396,
      "mean_token_accuracy": 0.23415720015764235,
      "num_tokens": 31271473.0,
      "step": 13640
    },
    {
      "entropy": 5.205772399902344,
      "epoch": 1.3107588856868395,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004835028763840595,
      "loss": 4.7409,
      "mean_token_accuracy": 0.23972004950046538,
      "num_tokens": 31282045.0,
      "step": 13645
    },
    {
      "entropy": 5.317965650558472,
      "epoch": 1.3112391930835736,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004834899883993703,
      "loss": 5.0056,
      "mean_token_accuracy": 0.2181426152586937,
      "num_tokens": 31292644.0,
      "step": 13650
    },
    {
      "entropy": 5.335716772079468,
      "epoch": 1.3117195004803075,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048347709557416263,
      "loss": 4.9132,
      "mean_token_accuracy": 0.23307441174983978,
      "num_tokens": 31304422.0,
      "step": 13655
    },
    {
      "entropy": 5.259875154495239,
      "epoch": 1.3121998078770414,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004834641979087359,
      "loss": 4.8325,
      "mean_token_accuracy": 0.2267339825630188,
      "num_tokens": 31314845.0,
      "step": 13660
    },
    {
      "entropy": 5.349634408950806,
      "epoch": 1.3126801152737753,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004834512954033894,
      "loss": 4.9741,
      "mean_token_accuracy": 0.22460563629865646,
      "num_tokens": 31325669.0,
      "step": 13665
    },
    {
      "entropy": 5.305160140991211,
      "epoch": 1.3131604226705091,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048343838805842284,
      "loss": 4.8604,
      "mean_token_accuracy": 0.23805071413516998,
      "num_tokens": 31336023.0,
      "step": 13670
    },
    {
      "entropy": 5.326093435287476,
      "epoch": 1.313640730067243,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048342547587413583,
      "loss": 4.9462,
      "mean_token_accuracy": 0.22223210632801055,
      "num_tokens": 31347146.0,
      "step": 13675
    },
    {
      "entropy": 5.302976131439209,
      "epoch": 1.314121037463977,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004834125588508282,
      "loss": 4.9395,
      "mean_token_accuracy": 0.22895164042711258,
      "num_tokens": 31357616.0,
      "step": 13680
    },
    {
      "entropy": 5.347755193710327,
      "epoch": 1.3146013448607108,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004833996369887998,
      "loss": 4.9516,
      "mean_token_accuracy": 0.22688707262277602,
      "num_tokens": 31369189.0,
      "step": 13685
    },
    {
      "entropy": 5.295722341537475,
      "epoch": 1.3150816522574447,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048338671028835063,
      "loss": 4.8215,
      "mean_token_accuracy": 0.23288591653108598,
      "num_tokens": 31380841.0,
      "step": 13690
    },
    {
      "entropy": 5.2210533142089846,
      "epoch": 1.3155619596541785,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048337377874978086,
      "loss": 4.827,
      "mean_token_accuracy": 0.2313823476433754,
      "num_tokens": 31393092.0,
      "step": 13695
    },
    {
      "entropy": 5.268663787841797,
      "epoch": 1.3160422670509127,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048336084237339067,
      "loss": 4.8847,
      "mean_token_accuracy": 0.23121354579925538,
      "num_tokens": 31404228.0,
      "step": 13700
    },
    {
      "entropy": 5.358303451538086,
      "epoch": 1.3165225744476465,
      "grad_norm": 1.0,
      "learning_rate": 0.0004833479011594804,
      "loss": 4.9847,
      "mean_token_accuracy": 0.22208735346794128,
      "num_tokens": 31416233.0,
      "step": 13705
    },
    {
      "entropy": 5.368725872039795,
      "epoch": 1.3170028818443804,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048333495510835057,
      "loss": 4.9238,
      "mean_token_accuracy": 0.22617635279893875,
      "num_tokens": 31427787.0,
      "step": 13710
    },
    {
      "entropy": 5.150783586502075,
      "epoch": 1.3174831892411143,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048332200422030163,
      "loss": 4.7517,
      "mean_token_accuracy": 0.23224983662366866,
      "num_tokens": 31438736.0,
      "step": 13715
    },
    {
      "entropy": 5.170078897476197,
      "epoch": 1.3179634966378482,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004833090484956345,
      "loss": 4.7609,
      "mean_token_accuracy": 0.23098889142274856,
      "num_tokens": 31449225.0,
      "step": 13720
    },
    {
      "entropy": 5.342968225479126,
      "epoch": 1.318443804034582,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048329608793464966,
      "loss": 4.9721,
      "mean_token_accuracy": 0.22556157857179643,
      "num_tokens": 31461223.0,
      "step": 13725
    },
    {
      "entropy": 5.402387380599976,
      "epoch": 1.3189241114313162,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004832831225376482,
      "loss": 4.9975,
      "mean_token_accuracy": 0.22288862317800523,
      "num_tokens": 31472397.0,
      "step": 13730
    },
    {
      "entropy": 5.221269702911377,
      "epoch": 1.31940441882805,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004832701523049312,
      "loss": 4.8175,
      "mean_token_accuracy": 0.23640452623367308,
      "num_tokens": 31484079.0,
      "step": 13735
    },
    {
      "entropy": 5.237677812576294,
      "epoch": 1.319884726224784,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004832571772367997,
      "loss": 4.8359,
      "mean_token_accuracy": 0.2255745232105255,
      "num_tokens": 31496732.0,
      "step": 13740
    },
    {
      "entropy": 5.3093877792358395,
      "epoch": 1.3203650336215178,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048324419733355485,
      "loss": 4.8058,
      "mean_token_accuracy": 0.2365250915288925,
      "num_tokens": 31507601.0,
      "step": 13745
    },
    {
      "entropy": 5.251601648330689,
      "epoch": 1.3208453410182517,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048323121259549805,
      "loss": 4.8535,
      "mean_token_accuracy": 0.22918368130922318,
      "num_tokens": 31519551.0,
      "step": 13750
    },
    {
      "entropy": 5.2553027153015135,
      "epoch": 1.3213256484149856,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048321822302293095,
      "loss": 4.9155,
      "mean_token_accuracy": 0.23115395605564118,
      "num_tokens": 31531531.0,
      "step": 13755
    },
    {
      "entropy": 5.274306344985962,
      "epoch": 1.3218059558117194,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004832052286161549,
      "loss": 4.8405,
      "mean_token_accuracy": 0.2278796136379242,
      "num_tokens": 31542386.0,
      "step": 13760
    },
    {
      "entropy": 5.305922269821167,
      "epoch": 1.3222862632084533,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048319222937547176,
      "loss": 4.8634,
      "mean_token_accuracy": 0.23021839410066605,
      "num_tokens": 31554142.0,
      "step": 13765
    },
    {
      "entropy": 5.291234302520752,
      "epoch": 1.3227665706051872,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048317922530118323,
      "loss": 4.8955,
      "mean_token_accuracy": 0.2234797179698944,
      "num_tokens": 31566687.0,
      "step": 13770
    },
    {
      "entropy": 5.300694417953491,
      "epoch": 1.3232468780019213,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004831662163935912,
      "loss": 4.9643,
      "mean_token_accuracy": 0.22140786647796631,
      "num_tokens": 31577893.0,
      "step": 13775
    },
    {
      "entropy": 5.392210912704468,
      "epoch": 1.3237271853986552,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048315320265299784,
      "loss": 5.0274,
      "mean_token_accuracy": 0.21902917176485062,
      "num_tokens": 31590483.0,
      "step": 13780
    },
    {
      "entropy": 5.332138204574585,
      "epoch": 1.324207492795389,
      "grad_norm": 1.0,
      "learning_rate": 0.00048314018407970516,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2252175837755203,
      "num_tokens": 31601373.0,
      "step": 13785
    },
    {
      "entropy": 5.2918332576751705,
      "epoch": 1.324687800192123,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048312716067401535,
      "loss": 4.926,
      "mean_token_accuracy": 0.22004558593034745,
      "num_tokens": 31613981.0,
      "step": 13790
    },
    {
      "entropy": 5.335304641723633,
      "epoch": 1.3251681075888568,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000483114132436231,
      "loss": 4.9259,
      "mean_token_accuracy": 0.22104025781154632,
      "num_tokens": 31626727.0,
      "step": 13795
    },
    {
      "entropy": 5.2277909278869625,
      "epoch": 1.3256484149855907,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004831010993666545,
      "loss": 4.7926,
      "mean_token_accuracy": 0.2344595268368721,
      "num_tokens": 31638519.0,
      "step": 13800
    },
    {
      "entropy": 5.201468658447266,
      "epoch": 1.3261287223823248,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004830880614655884,
      "loss": 4.819,
      "mean_token_accuracy": 0.23099566847085953,
      "num_tokens": 31649184.0,
      "step": 13805
    },
    {
      "entropy": 5.31439528465271,
      "epoch": 1.3266090297790587,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048307501873333527,
      "loss": 4.913,
      "mean_token_accuracy": 0.2248839646577835,
      "num_tokens": 31660266.0,
      "step": 13810
    },
    {
      "entropy": 5.269160985946655,
      "epoch": 1.3270893371757926,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004830619711701982,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2287002608180046,
      "num_tokens": 31670841.0,
      "step": 13815
    },
    {
      "entropy": 5.2656652450561525,
      "epoch": 1.3275696445725265,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004830489187764799,
      "loss": 4.8539,
      "mean_token_accuracy": 0.22434525191783905,
      "num_tokens": 31680786.0,
      "step": 13820
    },
    {
      "entropy": 5.197905445098877,
      "epoch": 1.3280499519692603,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004830358615524835,
      "loss": 4.8482,
      "mean_token_accuracy": 0.22958334386348725,
      "num_tokens": 31692639.0,
      "step": 13825
    },
    {
      "entropy": 5.39703335762024,
      "epoch": 1.3285302593659942,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048302279949851215,
      "loss": 4.9557,
      "mean_token_accuracy": 0.22690303027629852,
      "num_tokens": 31703520.0,
      "step": 13830
    },
    {
      "entropy": 5.2889323234558105,
      "epoch": 1.329010566762728,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048300973261486906,
      "loss": 4.8191,
      "mean_token_accuracy": 0.24196283370256425,
      "num_tokens": 31715659.0,
      "step": 13835
    },
    {
      "entropy": 5.283125734329223,
      "epoch": 1.329490874159462,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004829966609018577,
      "loss": 4.8869,
      "mean_token_accuracy": 0.22775750011205673,
      "num_tokens": 31727896.0,
      "step": 13840
    },
    {
      "entropy": 5.268806457519531,
      "epoch": 1.3299711815561959,
      "grad_norm": 1.0,
      "learning_rate": 0.00048298358435978146,
      "loss": 4.8551,
      "mean_token_accuracy": 0.2319924294948578,
      "num_tokens": 31738618.0,
      "step": 13845
    },
    {
      "entropy": 5.342721748352051,
      "epoch": 1.3304514889529298,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048297050298894394,
      "loss": 4.9328,
      "mean_token_accuracy": 0.22385592460632325,
      "num_tokens": 31750306.0,
      "step": 13850
    },
    {
      "entropy": 5.395067834854126,
      "epoch": 1.3309317963496639,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004829574167896489,
      "loss": 4.9759,
      "mean_token_accuracy": 0.21937171816825868,
      "num_tokens": 31762015.0,
      "step": 13855
    },
    {
      "entropy": 5.230828475952149,
      "epoch": 1.3314121037463977,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048294432576220027,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22865526229143143,
      "num_tokens": 31774120.0,
      "step": 13860
    },
    {
      "entropy": 5.235819387435913,
      "epoch": 1.3318924111431316,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004829312299069017,
      "loss": 4.8236,
      "mean_token_accuracy": 0.2374514862895012,
      "num_tokens": 31785485.0,
      "step": 13865
    },
    {
      "entropy": 5.330244779586792,
      "epoch": 1.3323727185398655,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048291812922405755,
      "loss": 4.9189,
      "mean_token_accuracy": 0.22538455575704575,
      "num_tokens": 31795621.0,
      "step": 13870
    },
    {
      "entropy": 5.199333000183105,
      "epoch": 1.3328530259365994,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004829050237139717,
      "loss": 4.8021,
      "mean_token_accuracy": 0.2365028977394104,
      "num_tokens": 31806761.0,
      "step": 13875
    },
    {
      "entropy": 5.317231893539429,
      "epoch": 1.3333333333333333,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004828919133769486,
      "loss": 4.8899,
      "mean_token_accuracy": 0.22403468042612076,
      "num_tokens": 31817996.0,
      "step": 13880
    },
    {
      "entropy": 5.352569484710694,
      "epoch": 1.3338136407300674,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004828787982132926,
      "loss": 4.958,
      "mean_token_accuracy": 0.22435901314020157,
      "num_tokens": 31829888.0,
      "step": 13885
    },
    {
      "entropy": 5.337896013259888,
      "epoch": 1.3342939481268012,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048286567822330815,
      "loss": 4.9084,
      "mean_token_accuracy": 0.22616616934537886,
      "num_tokens": 31842582.0,
      "step": 13890
    },
    {
      "entropy": 5.202333641052246,
      "epoch": 1.3347742555235351,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004828525534072999,
      "loss": 4.7778,
      "mean_token_accuracy": 0.22988406121730803,
      "num_tokens": 31854765.0,
      "step": 13895
    },
    {
      "entropy": 5.298439884185791,
      "epoch": 1.335254562920269,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048283942376557254,
      "loss": 4.8459,
      "mean_token_accuracy": 0.2252200201153755,
      "num_tokens": 31865505.0,
      "step": 13900
    },
    {
      "entropy": 5.230467033386231,
      "epoch": 1.3357348703170029,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048282628929843097,
      "loss": 4.8652,
      "mean_token_accuracy": 0.23085143715143203,
      "num_tokens": 31876599.0,
      "step": 13905
    },
    {
      "entropy": 5.308727693557739,
      "epoch": 1.3362151777137368,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048281315000617996,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22780879735946655,
      "num_tokens": 31887292.0,
      "step": 13910
    },
    {
      "entropy": 5.31342306137085,
      "epoch": 1.3366954851104706,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004828000058891248,
      "loss": 4.9066,
      "mean_token_accuracy": 0.22340749055147172,
      "num_tokens": 31899018.0,
      "step": 13915
    },
    {
      "entropy": 5.254479169845581,
      "epoch": 1.3371757925072045,
      "grad_norm": 1.0,
      "learning_rate": 0.0004827868569475706,
      "loss": 4.8236,
      "mean_token_accuracy": 0.23178063631057738,
      "num_tokens": 31910402.0,
      "step": 13920
    },
    {
      "entropy": 5.339969158172607,
      "epoch": 1.3376560999039384,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048277370318182243,
      "loss": 4.9644,
      "mean_token_accuracy": 0.21924670934677123,
      "num_tokens": 31921792.0,
      "step": 13925
    },
    {
      "entropy": 5.353150367736816,
      "epoch": 1.3381364073006723,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048276054459218596,
      "loss": 4.9573,
      "mean_token_accuracy": 0.22011711746454238,
      "num_tokens": 31932786.0,
      "step": 13930
    },
    {
      "entropy": 5.332780265808106,
      "epoch": 1.3386167146974064,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048274738117896643,
      "loss": 4.9047,
      "mean_token_accuracy": 0.2252891719341278,
      "num_tokens": 31944286.0,
      "step": 13935
    },
    {
      "entropy": 5.2982954502105715,
      "epoch": 1.3390970220941403,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048273421294246966,
      "loss": 4.9816,
      "mean_token_accuracy": 0.21585046350955964,
      "num_tokens": 31956304.0,
      "step": 13940
    },
    {
      "entropy": 5.2724692821502686,
      "epoch": 1.3395773294908742,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00048272103988300134,
      "loss": 4.8846,
      "mean_token_accuracy": 0.23008790761232376,
      "num_tokens": 31968978.0,
      "step": 13945
    },
    {
      "entropy": 5.33787956237793,
      "epoch": 1.340057636887608,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004827078620008672,
      "loss": 4.8836,
      "mean_token_accuracy": 0.22187694013118744,
      "num_tokens": 31980133.0,
      "step": 13950
    },
    {
      "entropy": 5.2687465190887455,
      "epoch": 1.340537944284342,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048269467929637337,
      "loss": 4.8408,
      "mean_token_accuracy": 0.23840171545743943,
      "num_tokens": 31990331.0,
      "step": 13955
    },
    {
      "entropy": 5.305340814590454,
      "epoch": 1.341018251681076,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048268149176982576,
      "loss": 4.8917,
      "mean_token_accuracy": 0.22483019977808,
      "num_tokens": 32001956.0,
      "step": 13960
    },
    {
      "entropy": 5.3418107509613035,
      "epoch": 1.34149855907781,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048266829942153055,
      "loss": 4.9381,
      "mean_token_accuracy": 0.22709191888570784,
      "num_tokens": 32013577.0,
      "step": 13965
    },
    {
      "entropy": 5.404013967514038,
      "epoch": 1.3419788664745438,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048265510225179413,
      "loss": 4.9913,
      "mean_token_accuracy": 0.22226700335741043,
      "num_tokens": 32025751.0,
      "step": 13970
    },
    {
      "entropy": 5.304413557052612,
      "epoch": 1.3424591738712777,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004826419002609229,
      "loss": 4.952,
      "mean_token_accuracy": 0.22278677970170974,
      "num_tokens": 32036592.0,
      "step": 13975
    },
    {
      "entropy": 5.400818109512329,
      "epoch": 1.3429394812680115,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048262869344922326,
      "loss": 4.9162,
      "mean_token_accuracy": 0.22806359380483626,
      "num_tokens": 32048122.0,
      "step": 13980
    },
    {
      "entropy": 5.298269367218017,
      "epoch": 1.3434197886647454,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048261548181700186,
      "loss": 4.8805,
      "mean_token_accuracy": 0.22628283351659775,
      "num_tokens": 32058673.0,
      "step": 13985
    },
    {
      "entropy": 5.17995138168335,
      "epoch": 1.3439000960614793,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004826022653645655,
      "loss": 4.8171,
      "mean_token_accuracy": 0.23226406425237656,
      "num_tokens": 32070219.0,
      "step": 13990
    },
    {
      "entropy": 5.324084711074829,
      "epoch": 1.3443804034582132,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000482589044092221,
      "loss": 4.9384,
      "mean_token_accuracy": 0.21984881162643433,
      "num_tokens": 32081883.0,
      "step": 13995
    },
    {
      "entropy": 5.397384786605835,
      "epoch": 1.344860710854947,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048257581800027527,
      "loss": 4.964,
      "mean_token_accuracy": 0.22106391340494155,
      "num_tokens": 32094107.0,
      "step": 14000
    },
    {
      "entropy": 5.296050357818603,
      "epoch": 1.345341018251681,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004825625870890354,
      "loss": 4.842,
      "mean_token_accuracy": 0.2294775903224945,
      "num_tokens": 32105242.0,
      "step": 14005
    },
    {
      "entropy": 5.30046763420105,
      "epoch": 1.345821325648415,
      "grad_norm": 1.0,
      "learning_rate": 0.0004825493513588086,
      "loss": 4.8476,
      "mean_token_accuracy": 0.23135566115379333,
      "num_tokens": 32117326.0,
      "step": 14010
    },
    {
      "entropy": 5.301973819732666,
      "epoch": 1.346301633045149,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048253611080990226,
      "loss": 4.9257,
      "mean_token_accuracy": 0.22462474554777145,
      "num_tokens": 32129002.0,
      "step": 14015
    },
    {
      "entropy": 5.299451303482056,
      "epoch": 1.3467819404418828,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004825228654426236,
      "loss": 4.9373,
      "mean_token_accuracy": 0.2273844838142395,
      "num_tokens": 32140380.0,
      "step": 14020
    },
    {
      "entropy": 5.264945936203003,
      "epoch": 1.3472622478386167,
      "grad_norm": 1.0,
      "learning_rate": 0.0004825096152572803,
      "loss": 4.8078,
      "mean_token_accuracy": 0.23440905958414077,
      "num_tokens": 32151806.0,
      "step": 14025
    },
    {
      "entropy": 5.311148166656494,
      "epoch": 1.3477425552353506,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048249636025417974,
      "loss": 5.0088,
      "mean_token_accuracy": 0.22276712507009505,
      "num_tokens": 32163426.0,
      "step": 14030
    },
    {
      "entropy": 5.306394863128662,
      "epoch": 1.3482228626320845,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048248310043362997,
      "loss": 4.8873,
      "mean_token_accuracy": 0.2344762086868286,
      "num_tokens": 32174349.0,
      "step": 14035
    },
    {
      "entropy": 5.368053245544433,
      "epoch": 1.3487031700288186,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004824698357959386,
      "loss": 4.9701,
      "mean_token_accuracy": 0.22140346318483353,
      "num_tokens": 32185382.0,
      "step": 14040
    },
    {
      "entropy": 5.298127746582031,
      "epoch": 1.3491834774255524,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048245656634141385,
      "loss": 4.8549,
      "mean_token_accuracy": 0.22934520989656448,
      "num_tokens": 32195687.0,
      "step": 14045
    },
    {
      "entropy": 5.278028249740601,
      "epoch": 1.3496637848222863,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048244329207036354,
      "loss": 4.8106,
      "mean_token_accuracy": 0.23638486862182617,
      "num_tokens": 32205973.0,
      "step": 14050
    },
    {
      "entropy": 5.274392032623291,
      "epoch": 1.3501440922190202,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048243001298309604,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22828484028577806,
      "num_tokens": 32217469.0,
      "step": 14055
    },
    {
      "entropy": 5.271539640426636,
      "epoch": 1.350624399615754,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048241672907991954,
      "loss": 4.923,
      "mean_token_accuracy": 0.22847344726324081,
      "num_tokens": 32228257.0,
      "step": 14060
    },
    {
      "entropy": 5.302990102767945,
      "epoch": 1.351104707012488,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004824034403611424,
      "loss": 4.8908,
      "mean_token_accuracy": 0.22954811304807662,
      "num_tokens": 32239420.0,
      "step": 14065
    },
    {
      "entropy": 5.345232629776001,
      "epoch": 1.3515850144092219,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004823901468270733,
      "loss": 4.9102,
      "mean_token_accuracy": 0.23065542429685593,
      "num_tokens": 32250962.0,
      "step": 14070
    },
    {
      "entropy": 5.273244714736938,
      "epoch": 1.3520653218059557,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004823768484780209,
      "loss": 4.8996,
      "mean_token_accuracy": 0.23094652593135834,
      "num_tokens": 32262310.0,
      "step": 14075
    },
    {
      "entropy": 5.227424240112304,
      "epoch": 1.3525456292026896,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048236354531429375,
      "loss": 4.7922,
      "mean_token_accuracy": 0.23424913585186005,
      "num_tokens": 32273373.0,
      "step": 14080
    },
    {
      "entropy": 5.360067987442017,
      "epoch": 1.3530259365994235,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004823502373362009,
      "loss": 4.9292,
      "mean_token_accuracy": 0.22526446729898453,
      "num_tokens": 32285020.0,
      "step": 14085
    },
    {
      "entropy": 5.305191278457642,
      "epoch": 1.3535062439961576,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004823369245440512,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22486730366945268,
      "num_tokens": 32296224.0,
      "step": 14090
    },
    {
      "entropy": 5.272822761535645,
      "epoch": 1.3539865513928915,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048232360693815387,
      "loss": 4.8943,
      "mean_token_accuracy": 0.23144407868385314,
      "num_tokens": 32306913.0,
      "step": 14095
    },
    {
      "entropy": 5.297202682495117,
      "epoch": 1.3544668587896254,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048231028451881786,
      "loss": 4.8421,
      "mean_token_accuracy": 0.23436393439769745,
      "num_tokens": 32317981.0,
      "step": 14100
    },
    {
      "entropy": 5.195507478713989,
      "epoch": 1.3549471661863592,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004822969572863527,
      "loss": 4.7814,
      "mean_token_accuracy": 0.23572082817554474,
      "num_tokens": 32329656.0,
      "step": 14105
    },
    {
      "entropy": 5.338408994674682,
      "epoch": 1.3554274735830931,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048228362524106776,
      "loss": 4.9515,
      "mean_token_accuracy": 0.2277660772204399,
      "num_tokens": 32339460.0,
      "step": 14110
    },
    {
      "entropy": 5.2624366760253904,
      "epoch": 1.3559077809798272,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048227028838327253,
      "loss": 4.8778,
      "mean_token_accuracy": 0.22880464643239976,
      "num_tokens": 32351237.0,
      "step": 14115
    },
    {
      "entropy": 5.302978134155273,
      "epoch": 1.356388088376561,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048225694671327665,
      "loss": 4.818,
      "mean_token_accuracy": 0.22864389717578887,
      "num_tokens": 32362368.0,
      "step": 14120
    },
    {
      "entropy": 5.1929412364959715,
      "epoch": 1.356868395773295,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004822436002313899,
      "loss": 4.8464,
      "mean_token_accuracy": 0.23378677070140838,
      "num_tokens": 32373738.0,
      "step": 14125
    },
    {
      "entropy": 5.282374286651612,
      "epoch": 1.3573487031700289,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004822302489379222,
      "loss": 4.9027,
      "mean_token_accuracy": 0.22398252338171004,
      "num_tokens": 32384274.0,
      "step": 14130
    },
    {
      "entropy": 5.257030010223389,
      "epoch": 1.3578290105667628,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048221689283318335,
      "loss": 4.8057,
      "mean_token_accuracy": 0.23504820913076402,
      "num_tokens": 32395692.0,
      "step": 14135
    },
    {
      "entropy": 5.241180181503296,
      "epoch": 1.3583093179634966,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004822035319174837,
      "loss": 4.8781,
      "mean_token_accuracy": 0.2330724075436592,
      "num_tokens": 32407367.0,
      "step": 14140
    },
    {
      "entropy": 5.297163486480713,
      "epoch": 1.3587896253602305,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004821901661911332,
      "loss": 4.8945,
      "mean_token_accuracy": 0.2306876763701439,
      "num_tokens": 32418753.0,
      "step": 14145
    },
    {
      "entropy": 5.330021762847901,
      "epoch": 1.3592699327569644,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004821767956544423,
      "loss": 4.9061,
      "mean_token_accuracy": 0.22500046342611313,
      "num_tokens": 32429499.0,
      "step": 14150
    },
    {
      "entropy": 5.240992212295533,
      "epoch": 1.3597502401536983,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004821634203077214,
      "loss": 4.854,
      "mean_token_accuracy": 0.22698822021484374,
      "num_tokens": 32441246.0,
      "step": 14155
    },
    {
      "entropy": 5.322040176391601,
      "epoch": 1.3602305475504322,
      "grad_norm": 1.0625,
      "learning_rate": 0.000482150040151281,
      "loss": 4.9222,
      "mean_token_accuracy": 0.2234741196036339,
      "num_tokens": 32452953.0,
      "step": 14160
    },
    {
      "entropy": 5.335481309890747,
      "epoch": 1.3607108549471663,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004821366551854318,
      "loss": 4.8994,
      "mean_token_accuracy": 0.22852758765220643,
      "num_tokens": 32464030.0,
      "step": 14165
    },
    {
      "entropy": 5.2378545761108395,
      "epoch": 1.3611911623439001,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004821232654104845,
      "loss": 4.7119,
      "mean_token_accuracy": 0.23504256904125215,
      "num_tokens": 32475091.0,
      "step": 14170
    },
    {
      "entropy": 5.253159666061402,
      "epoch": 1.361671469740634,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048210987082675005,
      "loss": 4.9498,
      "mean_token_accuracy": 0.22817568331956864,
      "num_tokens": 32486047.0,
      "step": 14175
    },
    {
      "entropy": 5.253031635284424,
      "epoch": 1.362151777137368,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048209647143453946,
      "loss": 4.823,
      "mean_token_accuracy": 0.23249447494745254,
      "num_tokens": 32497141.0,
      "step": 14180
    },
    {
      "entropy": 5.307148265838623,
      "epoch": 1.3626320845341018,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048208306723416356,
      "loss": 4.8681,
      "mean_token_accuracy": 0.22868854254484178,
      "num_tokens": 32509282.0,
      "step": 14185
    },
    {
      "entropy": 5.350787019729614,
      "epoch": 1.3631123919308357,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004820696582259339,
      "loss": 4.9446,
      "mean_token_accuracy": 0.22276135981082917,
      "num_tokens": 32521383.0,
      "step": 14190
    },
    {
      "entropy": 5.258606433868408,
      "epoch": 1.3635926993275698,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004820562444101616,
      "loss": 4.8322,
      "mean_token_accuracy": 0.22747501581907273,
      "num_tokens": 32533168.0,
      "step": 14195
    },
    {
      "entropy": 5.239515256881714,
      "epoch": 1.3640730067243036,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004820428257871581,
      "loss": 4.8791,
      "mean_token_accuracy": 0.2272891879081726,
      "num_tokens": 32545333.0,
      "step": 14200
    },
    {
      "entropy": 5.235069704055786,
      "epoch": 1.3645533141210375,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004820294023572351,
      "loss": 4.841,
      "mean_token_accuracy": 0.23414137065410615,
      "num_tokens": 32556665.0,
      "step": 14205
    },
    {
      "entropy": 5.326712894439697,
      "epoch": 1.3650336215177714,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000482015974120704,
      "loss": 5.0079,
      "mean_token_accuracy": 0.21975035071372986,
      "num_tokens": 32568318.0,
      "step": 14210
    },
    {
      "entropy": 5.319150924682617,
      "epoch": 1.3655139289145053,
      "grad_norm": 1.0,
      "learning_rate": 0.00048200254107787677,
      "loss": 4.8278,
      "mean_token_accuracy": 0.2272895410656929,
      "num_tokens": 32580010.0,
      "step": 14215
    },
    {
      "entropy": 5.342264842987061,
      "epoch": 1.3659942363112392,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048198910322906516,
      "loss": 4.9743,
      "mean_token_accuracy": 0.216232268512249,
      "num_tokens": 32592075.0,
      "step": 14220
    },
    {
      "entropy": 5.394401359558105,
      "epoch": 1.366474543707973,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048197566057458125,
      "loss": 5.0063,
      "mean_token_accuracy": 0.22330859452486038,
      "num_tokens": 32604548.0,
      "step": 14225
    },
    {
      "entropy": 5.33350133895874,
      "epoch": 1.366954851104707,
      "grad_norm": 0.953125,
      "learning_rate": 0.000481962213114737,
      "loss": 4.904,
      "mean_token_accuracy": 0.22722495198249817,
      "num_tokens": 32616036.0,
      "step": 14230
    },
    {
      "entropy": 5.239726543426514,
      "epoch": 1.3674351585014408,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004819487608498448,
      "loss": 4.8642,
      "mean_token_accuracy": 0.22242840081453324,
      "num_tokens": 32628157.0,
      "step": 14235
    },
    {
      "entropy": 5.354171848297119,
      "epoch": 1.3679154658981747,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048193530378021687,
      "loss": 4.9518,
      "mean_token_accuracy": 0.22765274047851564,
      "num_tokens": 32640571.0,
      "step": 14240
    },
    {
      "entropy": 5.3244517803192135,
      "epoch": 1.3683957732949088,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048192184190616567,
      "loss": 4.9136,
      "mean_token_accuracy": 0.22483347058296205,
      "num_tokens": 32652005.0,
      "step": 14245
    },
    {
      "entropy": 5.284851312637329,
      "epoch": 1.3688760806916427,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004819083752280037,
      "loss": 4.848,
      "mean_token_accuracy": 0.22708419859409332,
      "num_tokens": 32663655.0,
      "step": 14250
    },
    {
      "entropy": 5.2339483261108395,
      "epoch": 1.3693563880883766,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048189490374604373,
      "loss": 4.9241,
      "mean_token_accuracy": 0.22166304290294647,
      "num_tokens": 32675419.0,
      "step": 14255
    },
    {
      "entropy": 5.273736476898193,
      "epoch": 1.3698366954851104,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004818814274605983,
      "loss": 4.84,
      "mean_token_accuracy": 0.23122029155492782,
      "num_tokens": 32686680.0,
      "step": 14260
    },
    {
      "entropy": 5.2717455387115475,
      "epoch": 1.3703170028818443,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004818679463719805,
      "loss": 4.7685,
      "mean_token_accuracy": 0.2367350533604622,
      "num_tokens": 32697321.0,
      "step": 14265
    },
    {
      "entropy": 5.251150178909302,
      "epoch": 1.3707973102785782,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004818544604805033,
      "loss": 4.8869,
      "mean_token_accuracy": 0.22150780707597734,
      "num_tokens": 32708885.0,
      "step": 14270
    },
    {
      "entropy": 5.317039585113525,
      "epoch": 1.3712776176753123,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004818409697864798,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2314225137233734,
      "num_tokens": 32720517.0,
      "step": 14275
    },
    {
      "entropy": 5.281079864501953,
      "epoch": 1.3717579250720462,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048182747429022303,
      "loss": 4.8615,
      "mean_token_accuracy": 0.22813022285699844,
      "num_tokens": 32731072.0,
      "step": 14280
    },
    {
      "entropy": 5.379647541046142,
      "epoch": 1.37223823246878,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004818139739920465,
      "loss": 4.9507,
      "mean_token_accuracy": 0.22585344314575195,
      "num_tokens": 32742117.0,
      "step": 14285
    },
    {
      "entropy": 5.280153274536133,
      "epoch": 1.372718539865514,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004818004688922637,
      "loss": 4.9218,
      "mean_token_accuracy": 0.22037615925073623,
      "num_tokens": 32753754.0,
      "step": 14290
    },
    {
      "entropy": 5.255582809448242,
      "epoch": 1.3731988472622478,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000481786958991188,
      "loss": 4.8557,
      "mean_token_accuracy": 0.23131667226552963,
      "num_tokens": 32766511.0,
      "step": 14295
    },
    {
      "entropy": 5.423879957199096,
      "epoch": 1.3736791546589817,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048177344428913316,
      "loss": 4.9347,
      "mean_token_accuracy": 0.22357359230518342,
      "num_tokens": 32777715.0,
      "step": 14300
    },
    {
      "entropy": 5.302725601196289,
      "epoch": 1.3741594620557156,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048175992478641293,
      "loss": 4.8752,
      "mean_token_accuracy": 0.2304987534880638,
      "num_tokens": 32789132.0,
      "step": 14305
    },
    {
      "entropy": 5.369827318191528,
      "epoch": 1.3746397694524495,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004817464004833412,
      "loss": 4.9445,
      "mean_token_accuracy": 0.2308490738272667,
      "num_tokens": 32800439.0,
      "step": 14310
    },
    {
      "entropy": 5.308284091949463,
      "epoch": 1.3751200768491834,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048173287138023204,
      "loss": 4.8511,
      "mean_token_accuracy": 0.22442493736743926,
      "num_tokens": 32813605.0,
      "step": 14315
    },
    {
      "entropy": 5.389495325088501,
      "epoch": 1.3756003842459175,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004817193374773993,
      "loss": 4.94,
      "mean_token_accuracy": 0.2217870682477951,
      "num_tokens": 32824225.0,
      "step": 14320
    },
    {
      "entropy": 5.364447402954101,
      "epoch": 1.3760806916426513,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048170579877515753,
      "loss": 5.0024,
      "mean_token_accuracy": 0.21266071498394012,
      "num_tokens": 32836917.0,
      "step": 14325
    },
    {
      "entropy": 5.280249547958374,
      "epoch": 1.3765609990393852,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004816922552738209,
      "loss": 4.8444,
      "mean_token_accuracy": 0.2293942853808403,
      "num_tokens": 32847972.0,
      "step": 14330
    },
    {
      "entropy": 5.20649471282959,
      "epoch": 1.377041306436119,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048167870697370373,
      "loss": 4.8109,
      "mean_token_accuracy": 0.23405249416828156,
      "num_tokens": 32858922.0,
      "step": 14335
    },
    {
      "entropy": 5.261417722702026,
      "epoch": 1.377521613832853,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004816651538751207,
      "loss": 4.7829,
      "mean_token_accuracy": 0.23046064376831055,
      "num_tokens": 32869788.0,
      "step": 14340
    },
    {
      "entropy": 5.462363910675049,
      "epoch": 1.3780019212295869,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048165159597838664,
      "loss": 4.997,
      "mean_token_accuracy": 0.21188186407089232,
      "num_tokens": 32881678.0,
      "step": 14345
    },
    {
      "entropy": 5.309618949890137,
      "epoch": 1.378482228626321,
      "grad_norm": 0.953125,
      "learning_rate": 0.000481638033283816,
      "loss": 4.8981,
      "mean_token_accuracy": 0.22378532439470292,
      "num_tokens": 32894183.0,
      "step": 14350
    },
    {
      "entropy": 5.262823390960693,
      "epoch": 1.3789625360230549,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048162446579172387,
      "loss": 4.8372,
      "mean_token_accuracy": 0.22888760268688202,
      "num_tokens": 32906001.0,
      "step": 14355
    },
    {
      "entropy": 5.304659128189087,
      "epoch": 1.3794428434197887,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004816108935024252,
      "loss": 4.8819,
      "mean_token_accuracy": 0.2275289922952652,
      "num_tokens": 32917236.0,
      "step": 14360
    },
    {
      "entropy": 5.313546419143677,
      "epoch": 1.3799231508165226,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048159731641623507,
      "loss": 4.9161,
      "mean_token_accuracy": 0.22366948574781417,
      "num_tokens": 32929710.0,
      "step": 14365
    },
    {
      "entropy": 5.274649524688721,
      "epoch": 1.3804034582132565,
      "grad_norm": 1.125,
      "learning_rate": 0.0004815837345334687,
      "loss": 4.9446,
      "mean_token_accuracy": 0.22530733346939086,
      "num_tokens": 32941565.0,
      "step": 14370
    },
    {
      "entropy": 5.374508380889893,
      "epoch": 1.3808837656099904,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004815701478544415,
      "loss": 4.9846,
      "mean_token_accuracy": 0.22207273095846175,
      "num_tokens": 32952730.0,
      "step": 14375
    },
    {
      "entropy": 5.276336908340454,
      "epoch": 1.3813640730067243,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048155655637946876,
      "loss": 4.833,
      "mean_token_accuracy": 0.22827137857675553,
      "num_tokens": 32963589.0,
      "step": 14380
    },
    {
      "entropy": 5.2532624244689945,
      "epoch": 1.3818443804034581,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004815429601088662,
      "loss": 4.7988,
      "mean_token_accuracy": 0.23190557062625886,
      "num_tokens": 32975008.0,
      "step": 14385
    },
    {
      "entropy": 5.303989744186401,
      "epoch": 1.382324687800192,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004815293590429494,
      "loss": 4.9936,
      "mean_token_accuracy": 0.2149255469441414,
      "num_tokens": 32986497.0,
      "step": 14390
    },
    {
      "entropy": 5.252300453186035,
      "epoch": 1.382804995196926,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048151575318203417,
      "loss": 4.7232,
      "mean_token_accuracy": 0.2425052911043167,
      "num_tokens": 32998298.0,
      "step": 14395
    },
    {
      "entropy": 5.394147348403931,
      "epoch": 1.38328530259366,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048150214252643637,
      "loss": 4.9974,
      "mean_token_accuracy": 0.2214560329914093,
      "num_tokens": 33010943.0,
      "step": 14400
    },
    {
      "entropy": 5.296246719360352,
      "epoch": 1.3837656099903939,
      "grad_norm": 0.984375,
      "learning_rate": 0.000481488527076472,
      "loss": 4.8899,
      "mean_token_accuracy": 0.2263984888792038,
      "num_tokens": 33022139.0,
      "step": 14405
    },
    {
      "entropy": 5.35267481803894,
      "epoch": 1.3842459173871278,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004814749068324571,
      "loss": 4.9714,
      "mean_token_accuracy": 0.22736300081014632,
      "num_tokens": 33034653.0,
      "step": 14410
    },
    {
      "entropy": 5.216076946258545,
      "epoch": 1.3847262247838616,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048146128179470804,
      "loss": 4.6715,
      "mean_token_accuracy": 0.24685922265052795,
      "num_tokens": 33044339.0,
      "step": 14415
    },
    {
      "entropy": 5.268087673187256,
      "epoch": 1.3852065321805955,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004814476519635411,
      "loss": 4.887,
      "mean_token_accuracy": 0.23384263962507248,
      "num_tokens": 33055418.0,
      "step": 14420
    },
    {
      "entropy": 5.298843288421631,
      "epoch": 1.3856868395773294,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048143401733927274,
      "loss": 4.9625,
      "mean_token_accuracy": 0.2245676949620247,
      "num_tokens": 33067239.0,
      "step": 14425
    },
    {
      "entropy": 5.2854948997497555,
      "epoch": 1.3861671469740635,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048142037792221943,
      "loss": 4.7664,
      "mean_token_accuracy": 0.23389922827482224,
      "num_tokens": 33079101.0,
      "step": 14430
    },
    {
      "entropy": 5.3123688220977785,
      "epoch": 1.3866474543707974,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004814067337126978,
      "loss": 4.8661,
      "mean_token_accuracy": 0.23377363979816437,
      "num_tokens": 33090265.0,
      "step": 14435
    },
    {
      "entropy": 5.269676780700683,
      "epoch": 1.3871277617675313,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004813930847110248,
      "loss": 4.8129,
      "mean_token_accuracy": 0.23596663028001785,
      "num_tokens": 33101724.0,
      "step": 14440
    },
    {
      "entropy": 5.234242630004883,
      "epoch": 1.3876080691642652,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004813794309175173,
      "loss": 4.8085,
      "mean_token_accuracy": 0.23423583805561066,
      "num_tokens": 33113111.0,
      "step": 14445
    },
    {
      "entropy": 5.376258420944214,
      "epoch": 1.388088376560999,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048136577233249205,
      "loss": 4.9806,
      "mean_token_accuracy": 0.21598962545394898,
      "num_tokens": 33123925.0,
      "step": 14450
    },
    {
      "entropy": 5.322758388519287,
      "epoch": 1.388568683957733,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004813521089562666,
      "loss": 4.9185,
      "mean_token_accuracy": 0.2252228707075119,
      "num_tokens": 33137400.0,
      "step": 14455
    },
    {
      "entropy": 5.178587293624878,
      "epoch": 1.3890489913544668,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004813384407891577,
      "loss": 4.7612,
      "mean_token_accuracy": 0.24200093299150466,
      "num_tokens": 33149326.0,
      "step": 14460
    },
    {
      "entropy": 5.368178462982177,
      "epoch": 1.3895292987512007,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000481324767831483,
      "loss": 4.9556,
      "mean_token_accuracy": 0.21476072520017625,
      "num_tokens": 33159904.0,
      "step": 14465
    },
    {
      "entropy": 5.265328598022461,
      "epoch": 1.3900096061479346,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004813110900835598,
      "loss": 4.8441,
      "mean_token_accuracy": 0.23006946891546248,
      "num_tokens": 33171701.0,
      "step": 14470
    },
    {
      "entropy": 5.282948780059814,
      "epoch": 1.3904899135446687,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004812974075457058,
      "loss": 4.7615,
      "mean_token_accuracy": 0.23180748969316484,
      "num_tokens": 33181598.0,
      "step": 14475
    },
    {
      "entropy": 5.262138795852661,
      "epoch": 1.3909702209414025,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048128372021823845,
      "loss": 4.7939,
      "mean_token_accuracy": 0.23494381457567215,
      "num_tokens": 33192674.0,
      "step": 14480
    },
    {
      "entropy": 5.255914545059204,
      "epoch": 1.3914505283381364,
      "grad_norm": 1.0,
      "learning_rate": 0.00048127002810147574,
      "loss": 4.8494,
      "mean_token_accuracy": 0.23133856058120728,
      "num_tokens": 33203356.0,
      "step": 14485
    },
    {
      "entropy": 5.296570491790772,
      "epoch": 1.3919308357348703,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004812563311957355,
      "loss": 4.8774,
      "mean_token_accuracy": 0.22688755244016648,
      "num_tokens": 33215471.0,
      "step": 14490
    },
    {
      "entropy": 5.293786430358887,
      "epoch": 1.3924111431316042,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004812426295013356,
      "loss": 4.8629,
      "mean_token_accuracy": 0.22339128702878952,
      "num_tokens": 33227656.0,
      "step": 14495
    },
    {
      "entropy": 5.355456686019897,
      "epoch": 1.392891450528338,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048122892301859433,
      "loss": 4.9814,
      "mean_token_accuracy": 0.2230197459459305,
      "num_tokens": 33239752.0,
      "step": 14500
    },
    {
      "entropy": 5.292311573028565,
      "epoch": 1.3933717579250722,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048121521174782983,
      "loss": 4.9031,
      "mean_token_accuracy": 0.22118544280529023,
      "num_tokens": 33251352.0,
      "step": 14505
    },
    {
      "entropy": 5.316899251937866,
      "epoch": 1.393852065321806,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048120149568936044,
      "loss": 4.9089,
      "mean_token_accuracy": 0.22172236442565918,
      "num_tokens": 33262276.0,
      "step": 14510
    },
    {
      "entropy": 5.367921257019043,
      "epoch": 1.39433237271854,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004811877748435046,
      "loss": 4.9467,
      "mean_token_accuracy": 0.22397078424692154,
      "num_tokens": 33273615.0,
      "step": 14515
    },
    {
      "entropy": 5.32791485786438,
      "epoch": 1.3948126801152738,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004811740492105809,
      "loss": 4.8695,
      "mean_token_accuracy": 0.23079843968153,
      "num_tokens": 33283990.0,
      "step": 14520
    },
    {
      "entropy": 5.24510612487793,
      "epoch": 1.3952929875120077,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000481160318790908,
      "loss": 4.867,
      "mean_token_accuracy": 0.23262408822774888,
      "num_tokens": 33295472.0,
      "step": 14525
    },
    {
      "entropy": 5.310835695266723,
      "epoch": 1.3957732949087416,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048114658358480467,
      "loss": 4.9191,
      "mean_token_accuracy": 0.22403932064771653,
      "num_tokens": 33306742.0,
      "step": 14530
    },
    {
      "entropy": 5.352105331420899,
      "epoch": 1.3962536023054755,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048113284359258977,
      "loss": 4.9043,
      "mean_token_accuracy": 0.22386656999588012,
      "num_tokens": 33317737.0,
      "step": 14535
    },
    {
      "entropy": 5.314499282836914,
      "epoch": 1.3967339097022093,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048111909881458234,
      "loss": 4.9116,
      "mean_token_accuracy": 0.22386102080345155,
      "num_tokens": 33329673.0,
      "step": 14540
    },
    {
      "entropy": 5.189743423461914,
      "epoch": 1.3972142170989432,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048110534925110146,
      "loss": 4.7398,
      "mean_token_accuracy": 0.24067392498254775,
      "num_tokens": 33342047.0,
      "step": 14545
    },
    {
      "entropy": 5.255878591537476,
      "epoch": 1.397694524495677,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004810915949024664,
      "loss": 4.8408,
      "mean_token_accuracy": 0.2376130625605583,
      "num_tokens": 33353287.0,
      "step": 14550
    },
    {
      "entropy": 5.325775861740112,
      "epoch": 1.3981748318924112,
      "grad_norm": 0.875,
      "learning_rate": 0.0004810778357689965,
      "loss": 4.9184,
      "mean_token_accuracy": 0.2302737906575203,
      "num_tokens": 33365465.0,
      "step": 14555
    },
    {
      "entropy": 5.265552377700805,
      "epoch": 1.398655139289145,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048106407185101116,
      "loss": 4.8532,
      "mean_token_accuracy": 0.22942604422569274,
      "num_tokens": 33376680.0,
      "step": 14560
    },
    {
      "entropy": 5.248045444488525,
      "epoch": 1.399135446685879,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048105030314883,
      "loss": 4.8737,
      "mean_token_accuracy": 0.2303225129842758,
      "num_tokens": 33387995.0,
      "step": 14565
    },
    {
      "entropy": 5.27147331237793,
      "epoch": 1.3996157540826129,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004810365296627725,
      "loss": 4.8295,
      "mean_token_accuracy": 0.22726923376321792,
      "num_tokens": 33400455.0,
      "step": 14570
    },
    {
      "entropy": 5.297858047485351,
      "epoch": 1.4000960614793467,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004810227513931587,
      "loss": 4.888,
      "mean_token_accuracy": 0.2206079751253128,
      "num_tokens": 33413264.0,
      "step": 14575
    },
    {
      "entropy": 5.267934894561767,
      "epoch": 1.4005763688760806,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004810089683403084,
      "loss": 4.769,
      "mean_token_accuracy": 0.24050407707691193,
      "num_tokens": 33423516.0,
      "step": 14580
    },
    {
      "entropy": 5.3227698802948,
      "epoch": 1.4010566762728147,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004809951805045415,
      "loss": 4.9253,
      "mean_token_accuracy": 0.22132017314434052,
      "num_tokens": 33434952.0,
      "step": 14585
    },
    {
      "entropy": 5.272095489501953,
      "epoch": 1.4015369836695486,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048098138788617815,
      "loss": 4.8605,
      "mean_token_accuracy": 0.22687692642211915,
      "num_tokens": 33447025.0,
      "step": 14590
    },
    {
      "entropy": 5.3293595790863035,
      "epoch": 1.4020172910662825,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004809675904855387,
      "loss": 4.8543,
      "mean_token_accuracy": 0.22839802056550979,
      "num_tokens": 33459990.0,
      "step": 14595
    },
    {
      "entropy": 5.286959838867188,
      "epoch": 1.4024975984630164,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048095378830294343,
      "loss": 4.8011,
      "mean_token_accuracy": 0.2337146133184433,
      "num_tokens": 33471334.0,
      "step": 14600
    },
    {
      "entropy": 5.3244202613830565,
      "epoch": 1.4029779058597502,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048093998133871276,
      "loss": 4.9189,
      "mean_token_accuracy": 0.23044095933437347,
      "num_tokens": 33483744.0,
      "step": 14605
    },
    {
      "entropy": 5.30485200881958,
      "epoch": 1.4034582132564841,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004809261695931671,
      "loss": 4.9418,
      "mean_token_accuracy": 0.2281404986977577,
      "num_tokens": 33496532.0,
      "step": 14610
    },
    {
      "entropy": 5.236634874343872,
      "epoch": 1.403938520653218,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004809123530666273,
      "loss": 4.8776,
      "mean_token_accuracy": 0.23158711642026902,
      "num_tokens": 33507545.0,
      "step": 14615
    },
    {
      "entropy": 5.326156997680664,
      "epoch": 1.4044188280499519,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004808985317594142,
      "loss": 4.8688,
      "mean_token_accuracy": 0.2337332397699356,
      "num_tokens": 33519232.0,
      "step": 14620
    },
    {
      "entropy": 5.283540916442871,
      "epoch": 1.4048991354466858,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048088470567184854,
      "loss": 4.8431,
      "mean_token_accuracy": 0.22654474824666976,
      "num_tokens": 33531088.0,
      "step": 14625
    },
    {
      "entropy": 5.316868829727173,
      "epoch": 1.4053794428434199,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048087087480425133,
      "loss": 4.9138,
      "mean_token_accuracy": 0.22268874049186707,
      "num_tokens": 33543991.0,
      "step": 14630
    },
    {
      "entropy": 5.370850419998169,
      "epoch": 1.4058597502401537,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004808570391569437,
      "loss": 4.9658,
      "mean_token_accuracy": 0.21773944348096846,
      "num_tokens": 33555303.0,
      "step": 14635
    },
    {
      "entropy": 5.348700618743896,
      "epoch": 1.4063400576368876,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048084319873024694,
      "loss": 4.9347,
      "mean_token_accuracy": 0.22135068476200104,
      "num_tokens": 33565587.0,
      "step": 14640
    },
    {
      "entropy": 5.268626594543457,
      "epoch": 1.4068203650336215,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004808293535244823,
      "loss": 4.796,
      "mean_token_accuracy": 0.23036053776741028,
      "num_tokens": 33576874.0,
      "step": 14645
    },
    {
      "entropy": 5.240434741973877,
      "epoch": 1.4073006724303554,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004808155035399712,
      "loss": 4.812,
      "mean_token_accuracy": 0.23767502158880233,
      "num_tokens": 33587703.0,
      "step": 14650
    },
    {
      "entropy": 5.242403650283814,
      "epoch": 1.4077809798270893,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004808016487770354,
      "loss": 4.7868,
      "mean_token_accuracy": 0.23517387807369233,
      "num_tokens": 33598487.0,
      "step": 14655
    },
    {
      "entropy": 5.290165615081787,
      "epoch": 1.4082612872238234,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048078778923599637,
      "loss": 4.9204,
      "mean_token_accuracy": 0.22201876789331437,
      "num_tokens": 33610838.0,
      "step": 14660
    },
    {
      "entropy": 5.306839895248413,
      "epoch": 1.4087415946205573,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048077392491717593,
      "loss": 4.8746,
      "mean_token_accuracy": 0.2279173970222473,
      "num_tokens": 33622726.0,
      "step": 14665
    },
    {
      "entropy": 5.247614049911499,
      "epoch": 1.4092219020172911,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048076005582089597,
      "loss": 4.8509,
      "mean_token_accuracy": 0.22881805002689362,
      "num_tokens": 33635922.0,
      "step": 14670
    },
    {
      "entropy": 5.401433753967285,
      "epoch": 1.409702209414025,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048074618194747845,
      "loss": 4.9841,
      "mean_token_accuracy": 0.2192772373557091,
      "num_tokens": 33648486.0,
      "step": 14675
    },
    {
      "entropy": 5.343288898468018,
      "epoch": 1.410182516810759,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004807323032972456,
      "loss": 4.8853,
      "mean_token_accuracy": 0.2222984865307808,
      "num_tokens": 33659812.0,
      "step": 14680
    },
    {
      "entropy": 5.309471273422242,
      "epoch": 1.4106628242074928,
      "grad_norm": 1.0,
      "learning_rate": 0.0004807184198705196,
      "loss": 4.8588,
      "mean_token_accuracy": 0.2305012747645378,
      "num_tokens": 33671878.0,
      "step": 14685
    },
    {
      "entropy": 5.225200510025024,
      "epoch": 1.4111431316042267,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004807045316676226,
      "loss": 4.7643,
      "mean_token_accuracy": 0.23440636396408082,
      "num_tokens": 33683759.0,
      "step": 14690
    },
    {
      "entropy": 5.370955562591552,
      "epoch": 1.4116234390009605,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004806906386888773,
      "loss": 4.9704,
      "mean_token_accuracy": 0.22036536037921906,
      "num_tokens": 33694085.0,
      "step": 14695
    },
    {
      "entropy": 5.309545660018921,
      "epoch": 1.4121037463976944,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048067674093460607,
      "loss": 4.821,
      "mean_token_accuracy": 0.22646492272615432,
      "num_tokens": 33705810.0,
      "step": 14700
    },
    {
      "entropy": 5.336349773406982,
      "epoch": 1.4125840537944283,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048066283840513175,
      "loss": 4.9066,
      "mean_token_accuracy": 0.2304420456290245,
      "num_tokens": 33716798.0,
      "step": 14705
    },
    {
      "entropy": 5.248374986648559,
      "epoch": 1.4130643611911624,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004806489311007769,
      "loss": 4.8632,
      "mean_token_accuracy": 0.23660283982753755,
      "num_tokens": 33728515.0,
      "step": 14710
    },
    {
      "entropy": 5.2993871688842775,
      "epoch": 1.4135446685878963,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00048063501902186463,
      "loss": 4.8758,
      "mean_token_accuracy": 0.22869690358638764,
      "num_tokens": 33740684.0,
      "step": 14715
    },
    {
      "entropy": 5.296594572067261,
      "epoch": 1.4140249759846302,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048062110216871775,
      "loss": 4.8586,
      "mean_token_accuracy": 0.2369449257850647,
      "num_tokens": 33753017.0,
      "step": 14720
    },
    {
      "entropy": 5.293589067459107,
      "epoch": 1.414505283381364,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048060718054165945,
      "loss": 4.8854,
      "mean_token_accuracy": 0.22540443688631057,
      "num_tokens": 33764393.0,
      "step": 14725
    },
    {
      "entropy": 5.344416570663452,
      "epoch": 1.414985590778098,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000480593254141013,
      "loss": 4.9897,
      "mean_token_accuracy": 0.21315707564353942,
      "num_tokens": 33774941.0,
      "step": 14730
    },
    {
      "entropy": 5.2065235614776615,
      "epoch": 1.4154658981748318,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048057932296710165,
      "loss": 4.7389,
      "mean_token_accuracy": 0.24507845044136048,
      "num_tokens": 33786534.0,
      "step": 14735
    },
    {
      "entropy": 5.382257032394409,
      "epoch": 1.415946205571566,
      "grad_norm": 1.0,
      "learning_rate": 0.0004805653870202489,
      "loss": 5.0014,
      "mean_token_accuracy": 0.2190493419766426,
      "num_tokens": 33798339.0,
      "step": 14740
    },
    {
      "entropy": 5.4149984359741214,
      "epoch": 1.4164265129682998,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048055144630077825,
      "loss": 5.0161,
      "mean_token_accuracy": 0.21654149889945984,
      "num_tokens": 33810368.0,
      "step": 14745
    },
    {
      "entropy": 5.343071508407593,
      "epoch": 1.4169068203650337,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048053750080901336,
      "loss": 4.8583,
      "mean_token_accuracy": 0.22541316002607345,
      "num_tokens": 33821111.0,
      "step": 14750
    },
    {
      "entropy": 5.249497413635254,
      "epoch": 1.4173871277617676,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048052355054527794,
      "loss": 4.852,
      "mean_token_accuracy": 0.2298365592956543,
      "num_tokens": 33833629.0,
      "step": 14755
    },
    {
      "entropy": 5.326733541488648,
      "epoch": 1.4178674351585014,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048050959550989606,
      "loss": 4.9268,
      "mean_token_accuracy": 0.22754903733730317,
      "num_tokens": 33846531.0,
      "step": 14760
    },
    {
      "entropy": 5.242965412139893,
      "epoch": 1.4183477425552353,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004804956357031916,
      "loss": 4.8461,
      "mean_token_accuracy": 0.22400427907705306,
      "num_tokens": 33857251.0,
      "step": 14765
    },
    {
      "entropy": 5.2794678688049315,
      "epoch": 1.4188280499519692,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048048167112548873,
      "loss": 4.8481,
      "mean_token_accuracy": 0.231631276011467,
      "num_tokens": 33869314.0,
      "step": 14770
    },
    {
      "entropy": 5.371252584457397,
      "epoch": 1.419308357348703,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048046770177711157,
      "loss": 4.9889,
      "mean_token_accuracy": 0.21860482543706894,
      "num_tokens": 33880203.0,
      "step": 14775
    },
    {
      "entropy": 5.288868951797485,
      "epoch": 1.419788664745437,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004804537276583844,
      "loss": 4.7902,
      "mean_token_accuracy": 0.2370796397328377,
      "num_tokens": 33891379.0,
      "step": 14780
    },
    {
      "entropy": 5.264895582199097,
      "epoch": 1.420268972142171,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004804397487696319,
      "loss": 4.8347,
      "mean_token_accuracy": 0.23265964686870574,
      "num_tokens": 33902788.0,
      "step": 14785
    },
    {
      "entropy": 5.2729803085327145,
      "epoch": 1.420749279538905,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004804257651111783,
      "loss": 4.8668,
      "mean_token_accuracy": 0.23387159854173661,
      "num_tokens": 33913609.0,
      "step": 14790
    },
    {
      "entropy": 5.290232181549072,
      "epoch": 1.4212295869356388,
      "grad_norm": 1.0,
      "learning_rate": 0.00048041177668334853,
      "loss": 4.7793,
      "mean_token_accuracy": 0.2309631049633026,
      "num_tokens": 33924379.0,
      "step": 14795
    },
    {
      "entropy": 5.290174198150635,
      "epoch": 1.4217098943323727,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004803977834864672,
      "loss": 4.8945,
      "mean_token_accuracy": 0.22657026201486588,
      "num_tokens": 33936209.0,
      "step": 14800
    },
    {
      "entropy": 5.337809753417969,
      "epoch": 1.4221902017291066,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048038378552085927,
      "loss": 4.8588,
      "mean_token_accuracy": 0.22344097793102263,
      "num_tokens": 33947679.0,
      "step": 14805
    },
    {
      "entropy": 5.298593950271607,
      "epoch": 1.4226705091258405,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048036978278684974,
      "loss": 4.868,
      "mean_token_accuracy": 0.22663102746009828,
      "num_tokens": 33959474.0,
      "step": 14810
    },
    {
      "entropy": 5.403438234329224,
      "epoch": 1.4231508165225746,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004803557752847636,
      "loss": 5.0039,
      "mean_token_accuracy": 0.2193470925092697,
      "num_tokens": 33970831.0,
      "step": 14815
    },
    {
      "entropy": 5.398932886123657,
      "epoch": 1.4236311239193085,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048034176301492616,
      "loss": 4.9537,
      "mean_token_accuracy": 0.2201936274766922,
      "num_tokens": 33981765.0,
      "step": 14820
    },
    {
      "entropy": 5.308623361587524,
      "epoch": 1.4241114313160423,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004803277459776628,
      "loss": 4.8384,
      "mean_token_accuracy": 0.22936079651117325,
      "num_tokens": 33992435.0,
      "step": 14825
    },
    {
      "entropy": 5.209544658660889,
      "epoch": 1.4245917387127762,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048031372417329875,
      "loss": 4.8211,
      "mean_token_accuracy": 0.23474671095609664,
      "num_tokens": 34004570.0,
      "step": 14830
    },
    {
      "entropy": 5.302250528335572,
      "epoch": 1.42507204610951,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004802996976021598,
      "loss": 4.8042,
      "mean_token_accuracy": 0.23451599180698396,
      "num_tokens": 34015494.0,
      "step": 14835
    },
    {
      "entropy": 5.3437244415283205,
      "epoch": 1.425552353506244,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048028566626457145,
      "loss": 4.8779,
      "mean_token_accuracy": 0.229156494140625,
      "num_tokens": 34026684.0,
      "step": 14840
    },
    {
      "entropy": 5.279656267166137,
      "epoch": 1.4260326609029779,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048027163016085947,
      "loss": 4.8689,
      "mean_token_accuracy": 0.22592695951461791,
      "num_tokens": 34038948.0,
      "step": 14845
    },
    {
      "entropy": 5.355035877227783,
      "epoch": 1.4265129682997117,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048025758929134976,
      "loss": 5.0223,
      "mean_token_accuracy": 0.21761560440063477,
      "num_tokens": 34052216.0,
      "step": 14850
    },
    {
      "entropy": 5.363251543045044,
      "epoch": 1.4269932756964456,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004802435436563684,
      "loss": 4.8696,
      "mean_token_accuracy": 0.23204109817743301,
      "num_tokens": 34062602.0,
      "step": 14855
    },
    {
      "entropy": 5.337154388427734,
      "epoch": 1.4274735830931795,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048022949325624134,
      "loss": 4.8529,
      "mean_token_accuracy": 0.22917574346065522,
      "num_tokens": 34075049.0,
      "step": 14860
    },
    {
      "entropy": 5.314659166336059,
      "epoch": 1.4279538904899136,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048021543809129483,
      "loss": 4.9395,
      "mean_token_accuracy": 0.22196658104658126,
      "num_tokens": 34086719.0,
      "step": 14865
    },
    {
      "entropy": 5.310436344146728,
      "epoch": 1.4284341978866475,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004802013781618552,
      "loss": 4.8524,
      "mean_token_accuracy": 0.228761288523674,
      "num_tokens": 34098439.0,
      "step": 14870
    },
    {
      "entropy": 5.33615083694458,
      "epoch": 1.4289145052833814,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048018731346824895,
      "loss": 4.8628,
      "mean_token_accuracy": 0.23306185156106948,
      "num_tokens": 34110711.0,
      "step": 14875
    },
    {
      "entropy": 5.236464023590088,
      "epoch": 1.4293948126801153,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004801732440108026,
      "loss": 4.8267,
      "mean_token_accuracy": 0.22710946053266526,
      "num_tokens": 34122191.0,
      "step": 14880
    },
    {
      "entropy": 5.187019348144531,
      "epoch": 1.4298751200768491,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004801591697898427,
      "loss": 4.7777,
      "mean_token_accuracy": 0.23513223230838776,
      "num_tokens": 34132838.0,
      "step": 14885
    },
    {
      "entropy": 5.309075212478637,
      "epoch": 1.430355427473583,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004801450908056961,
      "loss": 4.8934,
      "mean_token_accuracy": 0.22493908256292344,
      "num_tokens": 34143394.0,
      "step": 14890
    },
    {
      "entropy": 5.440160703659058,
      "epoch": 1.4308357348703171,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004801310070586896,
      "loss": 4.957,
      "mean_token_accuracy": 0.22244300842285156,
      "num_tokens": 34155934.0,
      "step": 14895
    },
    {
      "entropy": 5.191530466079712,
      "epoch": 1.431316042267051,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004801169185491503,
      "loss": 4.7519,
      "mean_token_accuracy": 0.23938858956098558,
      "num_tokens": 34167949.0,
      "step": 14900
    },
    {
      "entropy": 5.347885274887085,
      "epoch": 1.4317963496637849,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048010282527740516,
      "loss": 5.0267,
      "mean_token_accuracy": 0.21898865550756455,
      "num_tokens": 34179733.0,
      "step": 14905
    },
    {
      "entropy": 5.3085263729095455,
      "epoch": 1.4322766570605188,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048008872724378146,
      "loss": 4.7996,
      "mean_token_accuracy": 0.2345488965511322,
      "num_tokens": 34190513.0,
      "step": 14910
    },
    {
      "entropy": 5.266893482208252,
      "epoch": 1.4327569644572526,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004800746244486065,
      "loss": 4.8255,
      "mean_token_accuracy": 0.2369401603937149,
      "num_tokens": 34201562.0,
      "step": 14915
    },
    {
      "entropy": 5.20132303237915,
      "epoch": 1.4332372718539865,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004800605168922077,
      "loss": 4.8098,
      "mean_token_accuracy": 0.2317703530192375,
      "num_tokens": 34212637.0,
      "step": 14920
    },
    {
      "entropy": 5.273764562606812,
      "epoch": 1.4337175792507204,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048004640457491267,
      "loss": 4.8592,
      "mean_token_accuracy": 0.23041097372770308,
      "num_tokens": 34225394.0,
      "step": 14925
    },
    {
      "entropy": 5.34603533744812,
      "epoch": 1.4341978866474543,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000480032287497049,
      "loss": 4.8411,
      "mean_token_accuracy": 0.2289062738418579,
      "num_tokens": 34236977.0,
      "step": 14930
    },
    {
      "entropy": 5.3204319953918455,
      "epoch": 1.4346781940441882,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048001816565894427,
      "loss": 4.9313,
      "mean_token_accuracy": 0.22781244218349456,
      "num_tokens": 34247486.0,
      "step": 14935
    },
    {
      "entropy": 5.270957612991333,
      "epoch": 1.435158501440922,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004800040390609267,
      "loss": 4.8228,
      "mean_token_accuracy": 0.237338151037693,
      "num_tokens": 34259404.0,
      "step": 14940
    },
    {
      "entropy": 5.284013080596924,
      "epoch": 1.4356388088376562,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047998990770332396,
      "loss": 4.8243,
      "mean_token_accuracy": 0.23784033507108687,
      "num_tokens": 34270388.0,
      "step": 14945
    },
    {
      "entropy": 5.347184085845948,
      "epoch": 1.43611911623439,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004799757715864643,
      "loss": 4.9033,
      "mean_token_accuracy": 0.22315224260091782,
      "num_tokens": 34281321.0,
      "step": 14950
    },
    {
      "entropy": 5.373961591720581,
      "epoch": 1.436599423631124,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004799616307106759,
      "loss": 4.9443,
      "mean_token_accuracy": 0.2180386945605278,
      "num_tokens": 34293177.0,
      "step": 14955
    },
    {
      "entropy": 5.370635986328125,
      "epoch": 1.4370797310278578,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000479947485076287,
      "loss": 4.9312,
      "mean_token_accuracy": 0.22635638564825059,
      "num_tokens": 34305175.0,
      "step": 14960
    },
    {
      "entropy": 5.312832880020141,
      "epoch": 1.4375600384245917,
      "grad_norm": 1.0,
      "learning_rate": 0.00047993333468362607,
      "loss": 4.9248,
      "mean_token_accuracy": 0.2191855490207672,
      "num_tokens": 34317969.0,
      "step": 14965
    },
    {
      "entropy": 5.389577436447143,
      "epoch": 1.4380403458213258,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047991917953302173,
      "loss": 4.9422,
      "mean_token_accuracy": 0.22101296335458756,
      "num_tokens": 34329913.0,
      "step": 14970
    },
    {
      "entropy": 5.36714539527893,
      "epoch": 1.4385206532180597,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047990501962480236,
      "loss": 4.8768,
      "mean_token_accuracy": 0.23410227447748183,
      "num_tokens": 34341656.0,
      "step": 14975
    },
    {
      "entropy": 5.285263919830323,
      "epoch": 1.4390009606147935,
      "grad_norm": 1.109375,
      "learning_rate": 0.000479890854959297,
      "loss": 4.8539,
      "mean_token_accuracy": 0.22962389886379242,
      "num_tokens": 34351767.0,
      "step": 14980
    },
    {
      "entropy": 5.275706243515015,
      "epoch": 1.4394812680115274,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004798766855368344,
      "loss": 4.8506,
      "mean_token_accuracy": 0.22923009544610978,
      "num_tokens": 34363437.0,
      "step": 14985
    },
    {
      "entropy": 5.395393657684326,
      "epoch": 1.4399615754082613,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047986251135774343,
      "loss": 4.9358,
      "mean_token_accuracy": 0.2311631441116333,
      "num_tokens": 34374991.0,
      "step": 14990
    },
    {
      "entropy": 5.281896543502808,
      "epoch": 1.4404418828049952,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004798483324223533,
      "loss": 4.8627,
      "mean_token_accuracy": 0.23248424828052522,
      "num_tokens": 34386218.0,
      "step": 14995
    },
    {
      "entropy": 5.3347405910491945,
      "epoch": 1.440922190201729,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004798341487309932,
      "loss": 4.8768,
      "mean_token_accuracy": 0.22712049782276153,
      "num_tokens": 34396287.0,
      "step": 15000
    },
    {
      "epoch": 1.440922190201729,
      "eval_entropy": 5.056596125003363,
      "eval_loss": 4.932506084442139,
      "eval_mean_token_accuracy": 0.23529131853383584,
      "eval_num_tokens": 34396287.0,
      "eval_runtime": 20.6517,
      "eval_samples_per_second": 1588.975,
      "eval_steps_per_second": 198.628,
      "step": 15000
    },
    {
      "entropy": 5.322837495803833,
      "epoch": 1.441402497598463,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047981996028399233,
      "loss": 4.8939,
      "mean_token_accuracy": 0.2239050894975662,
      "num_tokens": 34407251.0,
      "step": 15005
    },
    {
      "entropy": 5.4101053237915036,
      "epoch": 1.4418828049951968,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004798057670816802,
      "loss": 5.0177,
      "mean_token_accuracy": 0.2196619912981987,
      "num_tokens": 34419185.0,
      "step": 15010
    },
    {
      "entropy": 5.313039970397949,
      "epoch": 1.4423631123919307,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004797915691243863,
      "loss": 4.8656,
      "mean_token_accuracy": 0.22862207591533662,
      "num_tokens": 34431159.0,
      "step": 15015
    },
    {
      "entropy": 5.2857646465301515,
      "epoch": 1.4428434197886648,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004797773664124403,
      "loss": 4.8434,
      "mean_token_accuracy": 0.23019345998764038,
      "num_tokens": 34442614.0,
      "step": 15020
    },
    {
      "entropy": 5.248080205917359,
      "epoch": 1.4433237271853987,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047976315894617195,
      "loss": 4.8217,
      "mean_token_accuracy": 0.23618856817483902,
      "num_tokens": 34453984.0,
      "step": 15025
    },
    {
      "entropy": 5.257301902770996,
      "epoch": 1.4438040345821326,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000479748946725911,
      "loss": 4.8306,
      "mean_token_accuracy": 0.23076163232326508,
      "num_tokens": 34466296.0,
      "step": 15030
    },
    {
      "entropy": 5.407183122634888,
      "epoch": 1.4442843419788665,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004797347297519875,
      "loss": 5.0356,
      "mean_token_accuracy": 0.22263554632663726,
      "num_tokens": 34478088.0,
      "step": 15035
    },
    {
      "entropy": 5.325309896469117,
      "epoch": 1.4447646493756003,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047972050802473154,
      "loss": 4.9296,
      "mean_token_accuracy": 0.21698679327964782,
      "num_tokens": 34491664.0,
      "step": 15040
    },
    {
      "entropy": 5.245135831832886,
      "epoch": 1.4452449567723342,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004797062815444733,
      "loss": 4.8254,
      "mean_token_accuracy": 0.23286427259445192,
      "num_tokens": 34502977.0,
      "step": 15045
    },
    {
      "entropy": 5.3038969993591305,
      "epoch": 1.4457252641690683,
      "grad_norm": 1.03125,
      "learning_rate": 0.000479692050311543,
      "loss": 4.8001,
      "mean_token_accuracy": 0.23515771478414535,
      "num_tokens": 34515067.0,
      "step": 15050
    },
    {
      "entropy": 5.347664403915405,
      "epoch": 1.4462055715658022,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004796778143262711,
      "loss": 4.9461,
      "mean_token_accuracy": 0.22438474446535112,
      "num_tokens": 34525012.0,
      "step": 15055
    },
    {
      "entropy": 5.206511497497559,
      "epoch": 1.446685878962536,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004796635735889882,
      "loss": 4.763,
      "mean_token_accuracy": 0.24257690757513045,
      "num_tokens": 34535789.0,
      "step": 15060
    },
    {
      "entropy": 5.334837818145752,
      "epoch": 1.44716618635927,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047964932810002476,
      "loss": 4.8831,
      "mean_token_accuracy": 0.23461017906665801,
      "num_tokens": 34546276.0,
      "step": 15065
    },
    {
      "entropy": 5.30002908706665,
      "epoch": 1.4476464937560038,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004796350778597117,
      "loss": 4.8664,
      "mean_token_accuracy": 0.22426819652318955,
      "num_tokens": 34558361.0,
      "step": 15070
    },
    {
      "entropy": 5.234620857238769,
      "epoch": 1.4481268011527377,
      "grad_norm": 1.0,
      "learning_rate": 0.0004796208228683796,
      "loss": 4.8076,
      "mean_token_accuracy": 0.23360914438962938,
      "num_tokens": 34569482.0,
      "step": 15075
    },
    {
      "entropy": 5.362079191207886,
      "epoch": 1.4486071085494716,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047960656312635977,
      "loss": 4.9004,
      "mean_token_accuracy": 0.2324800208210945,
      "num_tokens": 34580128.0,
      "step": 15080
    },
    {
      "entropy": 5.2715034008026125,
      "epoch": 1.4490874159462055,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004795922986339831,
      "loss": 4.8449,
      "mean_token_accuracy": 0.2252741649746895,
      "num_tokens": 34591105.0,
      "step": 15085
    },
    {
      "entropy": 5.323545694351196,
      "epoch": 1.4495677233429394,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047957802939158057,
      "loss": 4.8843,
      "mean_token_accuracy": 0.22690114229917527,
      "num_tokens": 34602618.0,
      "step": 15090
    },
    {
      "entropy": 5.331810140609742,
      "epoch": 1.4500480307396733,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004795637553994838,
      "loss": 4.9364,
      "mean_token_accuracy": 0.22098395228385925,
      "num_tokens": 34614179.0,
      "step": 15095
    },
    {
      "entropy": 5.395627212524414,
      "epoch": 1.4505283381364074,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047954947665802404,
      "loss": 4.9688,
      "mean_token_accuracy": 0.22174616158008575,
      "num_tokens": 34625456.0,
      "step": 15100
    },
    {
      "entropy": 5.308123397827148,
      "epoch": 1.4510086455331412,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004795351931675329,
      "loss": 4.8326,
      "mean_token_accuracy": 0.23669068217277528,
      "num_tokens": 34636268.0,
      "step": 15105
    },
    {
      "entropy": 5.258298587799072,
      "epoch": 1.4514889529298751,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004795209049283419,
      "loss": 4.8059,
      "mean_token_accuracy": 0.23799711167812349,
      "num_tokens": 34647665.0,
      "step": 15110
    },
    {
      "entropy": 5.299715423583985,
      "epoch": 1.451969260326609,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004795066119407827,
      "loss": 4.9052,
      "mean_token_accuracy": 0.22525591105222703,
      "num_tokens": 34659965.0,
      "step": 15115
    },
    {
      "entropy": 5.337926721572876,
      "epoch": 1.4524495677233429,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004794923142051873,
      "loss": 4.9386,
      "mean_token_accuracy": 0.21996494084596635,
      "num_tokens": 34671724.0,
      "step": 15120
    },
    {
      "entropy": 5.3093891620635985,
      "epoch": 1.452929875120077,
      "grad_norm": 1.0,
      "learning_rate": 0.00047947801172188755,
      "loss": 4.8145,
      "mean_token_accuracy": 0.23379482179880143,
      "num_tokens": 34682445.0,
      "step": 15125
    },
    {
      "entropy": 5.210580635070801,
      "epoch": 1.4534101825168109,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004794637044912155,
      "loss": 4.7479,
      "mean_token_accuracy": 0.23569637387990952,
      "num_tokens": 34693621.0,
      "step": 15130
    },
    {
      "entropy": 5.232554244995117,
      "epoch": 1.4538904899135447,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004794493925135034,
      "loss": 4.7471,
      "mean_token_accuracy": 0.2431368872523308,
      "num_tokens": 34703978.0,
      "step": 15135
    },
    {
      "entropy": 5.310867404937744,
      "epoch": 1.4543707973102786,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047943507578908357,
      "loss": 4.9263,
      "mean_token_accuracy": 0.23298752009868623,
      "num_tokens": 34715468.0,
      "step": 15140
    },
    {
      "entropy": 5.349620008468628,
      "epoch": 1.4548511047070125,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004794207543182883,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22547394931316375,
      "num_tokens": 34726383.0,
      "step": 15145
    },
    {
      "entropy": 5.349632024765015,
      "epoch": 1.4553314121037464,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047940642810145005,
      "loss": 4.9115,
      "mean_token_accuracy": 0.22655945420265197,
      "num_tokens": 34737123.0,
      "step": 15150
    },
    {
      "entropy": 5.232617282867432,
      "epoch": 1.4558117195004803,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047939209713890156,
      "loss": 4.8169,
      "mean_token_accuracy": 0.23605723083019256,
      "num_tokens": 34749197.0,
      "step": 15155
    },
    {
      "entropy": 5.301115417480469,
      "epoch": 1.4562920268972142,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047937776143097547,
      "loss": 4.8904,
      "mean_token_accuracy": 0.2282451570034027,
      "num_tokens": 34759785.0,
      "step": 15160
    },
    {
      "entropy": 5.209020757675171,
      "epoch": 1.456772334293948,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004793634209780047,
      "loss": 4.7466,
      "mean_token_accuracy": 0.23839640766382217,
      "num_tokens": 34770938.0,
      "step": 15165
    },
    {
      "entropy": 5.291518688201904,
      "epoch": 1.457252641690682,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004793490757803221,
      "loss": 4.7814,
      "mean_token_accuracy": 0.2331702455878258,
      "num_tokens": 34782126.0,
      "step": 15170
    },
    {
      "entropy": 5.240386629104615,
      "epoch": 1.457732949087416,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047933472583826063,
      "loss": 4.8235,
      "mean_token_accuracy": 0.2282555028796196,
      "num_tokens": 34793802.0,
      "step": 15175
    },
    {
      "entropy": 5.250314235687256,
      "epoch": 1.45821325648415,
      "grad_norm": 1.0,
      "learning_rate": 0.0004793203711521537,
      "loss": 4.8769,
      "mean_token_accuracy": 0.23565060645341873,
      "num_tokens": 34805696.0,
      "step": 15180
    },
    {
      "entropy": 5.385017061233521,
      "epoch": 1.4586935638808838,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047930601172233446,
      "loss": 4.9906,
      "mean_token_accuracy": 0.2219552740454674,
      "num_tokens": 34818679.0,
      "step": 15185
    },
    {
      "entropy": 5.394923353195191,
      "epoch": 1.4591738712776177,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047929164754913624,
      "loss": 4.8925,
      "mean_token_accuracy": 0.229435732960701,
      "num_tokens": 34830528.0,
      "step": 15190
    },
    {
      "entropy": 5.318212461471558,
      "epoch": 1.4596541786743515,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004792772786328926,
      "loss": 4.7664,
      "mean_token_accuracy": 0.2387414202094078,
      "num_tokens": 34841621.0,
      "step": 15195
    },
    {
      "entropy": 5.211922454833984,
      "epoch": 1.4601344860710854,
      "grad_norm": 0.890625,
      "learning_rate": 0.00047926290497393714,
      "loss": 4.7739,
      "mean_token_accuracy": 0.23202792406082154,
      "num_tokens": 34854448.0,
      "step": 15200
    },
    {
      "entropy": 5.304074048995972,
      "epoch": 1.4606147934678195,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004792485265726036,
      "loss": 4.8763,
      "mean_token_accuracy": 0.2296251803636551,
      "num_tokens": 34866492.0,
      "step": 15205
    },
    {
      "entropy": 5.279980039596557,
      "epoch": 1.4610951008645534,
      "grad_norm": 1.0,
      "learning_rate": 0.0004792341434292257,
      "loss": 4.8691,
      "mean_token_accuracy": 0.22673982232809067,
      "num_tokens": 34876869.0,
      "step": 15210
    },
    {
      "entropy": 5.227858304977417,
      "epoch": 1.4615754082612873,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004792197555441374,
      "loss": 4.7718,
      "mean_token_accuracy": 0.24674406498670579,
      "num_tokens": 34888449.0,
      "step": 15215
    },
    {
      "entropy": 5.232436895370483,
      "epoch": 1.4620557156580212,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004792053629176729,
      "loss": 4.8163,
      "mean_token_accuracy": 0.2390949994325638,
      "num_tokens": 34898124.0,
      "step": 15220
    },
    {
      "entropy": 5.286084461212158,
      "epoch": 1.462536023054755,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004791909655501662,
      "loss": 4.7877,
      "mean_token_accuracy": 0.24076079577207565,
      "num_tokens": 34909128.0,
      "step": 15225
    },
    {
      "entropy": 5.236679935455323,
      "epoch": 1.463016330451489,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004791765634419516,
      "loss": 4.8333,
      "mean_token_accuracy": 0.23409875333309174,
      "num_tokens": 34920541.0,
      "step": 15230
    },
    {
      "entropy": 5.219994306564331,
      "epoch": 1.4634966378482228,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047916215659336343,
      "loss": 4.7952,
      "mean_token_accuracy": 0.23523704558610917,
      "num_tokens": 34931605.0,
      "step": 15235
    },
    {
      "entropy": 5.273922157287598,
      "epoch": 1.4639769452449567,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004791477450047363,
      "loss": 4.8206,
      "mean_token_accuracy": 0.23168710768222808,
      "num_tokens": 34943057.0,
      "step": 15240
    },
    {
      "entropy": 5.329599857330322,
      "epoch": 1.4644572526416906,
      "grad_norm": 1.0,
      "learning_rate": 0.00047913332867640464,
      "loss": 4.9803,
      "mean_token_accuracy": 0.2242401957511902,
      "num_tokens": 34954386.0,
      "step": 15245
    },
    {
      "entropy": 5.253720760345459,
      "epoch": 1.4649375600384245,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004791189076087033,
      "loss": 4.8046,
      "mean_token_accuracy": 0.2318597435951233,
      "num_tokens": 34965874.0,
      "step": 15250
    },
    {
      "entropy": 5.293386173248291,
      "epoch": 1.4654178674351586,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047910448180196703,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2267417848110199,
      "num_tokens": 34977408.0,
      "step": 15255
    },
    {
      "entropy": 5.269009113311768,
      "epoch": 1.4658981748318924,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004790900512565307,
      "loss": 4.7545,
      "mean_token_accuracy": 0.23654140383005143,
      "num_tokens": 34987788.0,
      "step": 15260
    },
    {
      "entropy": 5.306836032867432,
      "epoch": 1.4663784822286263,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004790756159727294,
      "loss": 4.9206,
      "mean_token_accuracy": 0.22820985913276673,
      "num_tokens": 35001051.0,
      "step": 15265
    },
    {
      "entropy": 5.315717172622681,
      "epoch": 1.4668587896253602,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047906117595089835,
      "loss": 4.862,
      "mean_token_accuracy": 0.2356676608324051,
      "num_tokens": 35012621.0,
      "step": 15270
    },
    {
      "entropy": 5.2959192276000975,
      "epoch": 1.467339097022094,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004790467311913727,
      "loss": 4.8614,
      "mean_token_accuracy": 0.2269326478242874,
      "num_tokens": 35023789.0,
      "step": 15275
    },
    {
      "entropy": 5.236650514602661,
      "epoch": 1.4678194044188282,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004790322816944879,
      "loss": 4.8497,
      "mean_token_accuracy": 0.23749999701976776,
      "num_tokens": 35035839.0,
      "step": 15280
    },
    {
      "entropy": 5.336534404754639,
      "epoch": 1.468299711815562,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004790178274605793,
      "loss": 4.9335,
      "mean_token_accuracy": 0.22153427749872207,
      "num_tokens": 35047823.0,
      "step": 15285
    },
    {
      "entropy": 5.315057706832886,
      "epoch": 1.468780019212296,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047900336848998254,
      "loss": 4.9576,
      "mean_token_accuracy": 0.21916618794202805,
      "num_tokens": 35058597.0,
      "step": 15290
    },
    {
      "entropy": 5.237282848358154,
      "epoch": 1.4692603266090298,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004789889047830334,
      "loss": 4.8297,
      "mean_token_accuracy": 0.2327761247754097,
      "num_tokens": 35069822.0,
      "step": 15295
    },
    {
      "entropy": 5.342449378967285,
      "epoch": 1.4697406340057637,
      "grad_norm": 0.875,
      "learning_rate": 0.00047897443634006766,
      "loss": 4.8676,
      "mean_token_accuracy": 0.23215243220329285,
      "num_tokens": 35081423.0,
      "step": 15300
    },
    {
      "entropy": 5.364402532577515,
      "epoch": 1.4702209414024976,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004789599631614211,
      "loss": 4.9085,
      "mean_token_accuracy": 0.2241699516773224,
      "num_tokens": 35092565.0,
      "step": 15305
    },
    {
      "entropy": 5.260176420211792,
      "epoch": 1.4707012487992315,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004789454852474298,
      "loss": 4.8492,
      "mean_token_accuracy": 0.23020370304584503,
      "num_tokens": 35103811.0,
      "step": 15310
    },
    {
      "entropy": 5.356721544265747,
      "epoch": 1.4711815561959654,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004789310025984299,
      "loss": 4.9728,
      "mean_token_accuracy": 0.22102905362844466,
      "num_tokens": 35115516.0,
      "step": 15315
    },
    {
      "entropy": 5.3009645462036135,
      "epoch": 1.4716618635926992,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047891651521475776,
      "loss": 4.8982,
      "mean_token_accuracy": 0.22946638017892837,
      "num_tokens": 35127285.0,
      "step": 15320
    },
    {
      "entropy": 5.2919567108154295,
      "epoch": 1.4721421709894331,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047890202309674963,
      "loss": 4.8293,
      "mean_token_accuracy": 0.23283894658088683,
      "num_tokens": 35137884.0,
      "step": 15325
    },
    {
      "entropy": 5.372288370132447,
      "epoch": 1.4726224783861672,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047888752624474195,
      "loss": 4.9851,
      "mean_token_accuracy": 0.21955863982439042,
      "num_tokens": 35149935.0,
      "step": 15330
    },
    {
      "entropy": 5.37389349937439,
      "epoch": 1.473102785782901,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004788730246590714,
      "loss": 4.9384,
      "mean_token_accuracy": 0.22437623739242554,
      "num_tokens": 35162610.0,
      "step": 15335
    },
    {
      "entropy": 5.304444694519043,
      "epoch": 1.473583093179635,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047885851834007456,
      "loss": 4.8023,
      "mean_token_accuracy": 0.23396918773651124,
      "num_tokens": 35174799.0,
      "step": 15340
    },
    {
      "entropy": 5.271753978729248,
      "epoch": 1.4740634005763689,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047884400728808824,
      "loss": 4.8021,
      "mean_token_accuracy": 0.2358495771884918,
      "num_tokens": 35186004.0,
      "step": 15345
    },
    {
      "entropy": 5.300626993179321,
      "epoch": 1.4745437079731027,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004788294915034494,
      "loss": 4.8453,
      "mean_token_accuracy": 0.23323777616024016,
      "num_tokens": 35197310.0,
      "step": 15350
    },
    {
      "entropy": 5.2111467838287355,
      "epoch": 1.4750240153698366,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000478814970986495,
      "loss": 4.8053,
      "mean_token_accuracy": 0.23739207834005355,
      "num_tokens": 35208703.0,
      "step": 15355
    },
    {
      "entropy": 5.262756967544556,
      "epoch": 1.4755043227665707,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047880044573756213,
      "loss": 4.8194,
      "mean_token_accuracy": 0.2368703991174698,
      "num_tokens": 35219927.0,
      "step": 15360
    },
    {
      "entropy": 5.267511177062988,
      "epoch": 1.4759846301633046,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047878591575698816,
      "loss": 4.809,
      "mean_token_accuracy": 0.23296742588281633,
      "num_tokens": 35231077.0,
      "step": 15365
    },
    {
      "entropy": 5.334640455245972,
      "epoch": 1.4764649375600385,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004787713810451103,
      "loss": 4.9101,
      "mean_token_accuracy": 0.2241204112768173,
      "num_tokens": 35241984.0,
      "step": 15370
    },
    {
      "entropy": 5.254921960830688,
      "epoch": 1.4769452449567724,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047875684160226606,
      "loss": 4.8332,
      "mean_token_accuracy": 0.2390206679701805,
      "num_tokens": 35252717.0,
      "step": 15375
    },
    {
      "entropy": 5.3078563690185545,
      "epoch": 1.4774255523535063,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000478742297428793,
      "loss": 4.8983,
      "mean_token_accuracy": 0.22824815660715103,
      "num_tokens": 35263916.0,
      "step": 15380
    },
    {
      "entropy": 5.292064476013183,
      "epoch": 1.4779058597502401,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047872774852502877,
      "loss": 4.8291,
      "mean_token_accuracy": 0.23399220108985902,
      "num_tokens": 35274772.0,
      "step": 15385
    },
    {
      "entropy": 5.203951072692871,
      "epoch": 1.478386167146974,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004787131948913112,
      "loss": 4.7679,
      "mean_token_accuracy": 0.24168988913297654,
      "num_tokens": 35287150.0,
      "step": 15390
    },
    {
      "entropy": 5.251729345321655,
      "epoch": 1.478866474543708,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047869863652797806,
      "loss": 4.7897,
      "mean_token_accuracy": 0.23650373369455338,
      "num_tokens": 35298538.0,
      "step": 15395
    },
    {
      "entropy": 5.232923936843872,
      "epoch": 1.4793467819404418,
      "grad_norm": 1.0,
      "learning_rate": 0.0004786840734353675,
      "loss": 4.7912,
      "mean_token_accuracy": 0.23600231409072875,
      "num_tokens": 35309395.0,
      "step": 15400
    },
    {
      "entropy": 5.277243757247925,
      "epoch": 1.4798270893371757,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047866950561381756,
      "loss": 4.8201,
      "mean_token_accuracy": 0.23137278854846954,
      "num_tokens": 35320741.0,
      "step": 15405
    },
    {
      "entropy": 5.405085182189941,
      "epoch": 1.4803073967339098,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004786549330636665,
      "loss": 4.9455,
      "mean_token_accuracy": 0.22242363393306733,
      "num_tokens": 35331895.0,
      "step": 15410
    },
    {
      "entropy": 5.21675329208374,
      "epoch": 1.4807877041306436,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047864035578525256,
      "loss": 4.7326,
      "mean_token_accuracy": 0.24284151047468186,
      "num_tokens": 35343775.0,
      "step": 15415
    },
    {
      "entropy": 5.220193862915039,
      "epoch": 1.4812680115273775,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004786257737789143,
      "loss": 4.9093,
      "mean_token_accuracy": 0.22198485136032103,
      "num_tokens": 35355043.0,
      "step": 15420
    },
    {
      "entropy": 5.386041498184204,
      "epoch": 1.4817483189241114,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004786111870449902,
      "loss": 4.9032,
      "mean_token_accuracy": 0.22457415908575057,
      "num_tokens": 35365387.0,
      "step": 15425
    },
    {
      "entropy": 5.335860252380371,
      "epoch": 1.4822286263208453,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047859659558381894,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23600376695394515,
      "num_tokens": 35376400.0,
      "step": 15430
    },
    {
      "entropy": 5.2802282810211185,
      "epoch": 1.4827089337175792,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047858199939573935,
      "loss": 4.8908,
      "mean_token_accuracy": 0.22946794480085372,
      "num_tokens": 35387315.0,
      "step": 15435
    },
    {
      "entropy": 5.319474220275879,
      "epoch": 1.4831892411143133,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047856739848109014,
      "loss": 4.8823,
      "mean_token_accuracy": 0.23261700570583344,
      "num_tokens": 35398666.0,
      "step": 15440
    },
    {
      "entropy": 5.310687494277954,
      "epoch": 1.4836695485110472,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047855279284021046,
      "loss": 4.8509,
      "mean_token_accuracy": 0.23163353353738786,
      "num_tokens": 35409192.0,
      "step": 15445
    },
    {
      "entropy": 5.320069074630737,
      "epoch": 1.484149855907781,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047853818247343933,
      "loss": 4.9146,
      "mean_token_accuracy": 0.220777527987957,
      "num_tokens": 35419812.0,
      "step": 15450
    },
    {
      "entropy": 5.343916702270508,
      "epoch": 1.484630163304515,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047852356738111606,
      "loss": 4.8329,
      "mean_token_accuracy": 0.24104802161455155,
      "num_tokens": 35430875.0,
      "step": 15455
    },
    {
      "entropy": 5.2792346477508545,
      "epoch": 1.4851104707012488,
      "grad_norm": 1.125,
      "learning_rate": 0.0004785089475635799,
      "loss": 4.8121,
      "mean_token_accuracy": 0.2362471416592598,
      "num_tokens": 35441065.0,
      "step": 15460
    },
    {
      "entropy": 5.268859577178955,
      "epoch": 1.4855907780979827,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047849432302117024,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22677670568227767,
      "num_tokens": 35452164.0,
      "step": 15465
    },
    {
      "entropy": 5.33691143989563,
      "epoch": 1.4860710854947166,
      "grad_norm": 1.125,
      "learning_rate": 0.00047847969375422656,
      "loss": 4.9598,
      "mean_token_accuracy": 0.21776237934827805,
      "num_tokens": 35463158.0,
      "step": 15470
    },
    {
      "entropy": 5.228866195678711,
      "epoch": 1.4865513928914504,
      "grad_norm": 1.0,
      "learning_rate": 0.0004784650597630887,
      "loss": 4.7679,
      "mean_token_accuracy": 0.2441105827689171,
      "num_tokens": 35474153.0,
      "step": 15475
    },
    {
      "entropy": 5.319555759429932,
      "epoch": 1.4870317002881843,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047845042104809635,
      "loss": 4.8524,
      "mean_token_accuracy": 0.22858181446790696,
      "num_tokens": 35485680.0,
      "step": 15480
    },
    {
      "entropy": 5.348137760162354,
      "epoch": 1.4875120076849184,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004784357776095892,
      "loss": 4.9046,
      "mean_token_accuracy": 0.23375765830278397,
      "num_tokens": 35497271.0,
      "step": 15485
    },
    {
      "entropy": 5.266527414321899,
      "epoch": 1.4879923150816523,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004784211294479075,
      "loss": 4.8249,
      "mean_token_accuracy": 0.23728204965591432,
      "num_tokens": 35509166.0,
      "step": 15490
    },
    {
      "entropy": 5.247035551071167,
      "epoch": 1.4884726224783862,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004784064765633912,
      "loss": 4.9197,
      "mean_token_accuracy": 0.23268962353467942,
      "num_tokens": 35521289.0,
      "step": 15495
    },
    {
      "entropy": 5.330571222305298,
      "epoch": 1.48895292987512,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047839181895638057,
      "loss": 4.8431,
      "mean_token_accuracy": 0.23650385141372682,
      "num_tokens": 35532179.0,
      "step": 15500
    },
    {
      "entropy": 5.39588303565979,
      "epoch": 1.489433237271854,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047837715662721575,
      "loss": 5.015,
      "mean_token_accuracy": 0.22179284691810608,
      "num_tokens": 35544703.0,
      "step": 15505
    },
    {
      "entropy": 5.351811742782592,
      "epoch": 1.4899135446685878,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004783624895762372,
      "loss": 5.0063,
      "mean_token_accuracy": 0.21626497358083724,
      "num_tokens": 35557853.0,
      "step": 15510
    },
    {
      "entropy": 5.265878438949585,
      "epoch": 1.490393852065322,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047834781780378563,
      "loss": 4.7339,
      "mean_token_accuracy": 0.23818091601133345,
      "num_tokens": 35570340.0,
      "step": 15515
    },
    {
      "entropy": 5.347053194046021,
      "epoch": 1.4908741594620558,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004783331413102015,
      "loss": 4.9276,
      "mean_token_accuracy": 0.22817194014787673,
      "num_tokens": 35582387.0,
      "step": 15520
    },
    {
      "entropy": 5.382496786117554,
      "epoch": 1.4913544668587897,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047831846009582557,
      "loss": 4.97,
      "mean_token_accuracy": 0.21992394030094148,
      "num_tokens": 35595105.0,
      "step": 15525
    },
    {
      "entropy": 5.321152687072754,
      "epoch": 1.4918347742555236,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004783037741609988,
      "loss": 4.9413,
      "mean_token_accuracy": 0.22543869018554688,
      "num_tokens": 35607160.0,
      "step": 15530
    },
    {
      "entropy": 5.284582853317261,
      "epoch": 1.4923150816522575,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004782890835060621,
      "loss": 4.8137,
      "mean_token_accuracy": 0.23793631345033645,
      "num_tokens": 35619097.0,
      "step": 15535
    },
    {
      "entropy": 5.198253011703491,
      "epoch": 1.4927953890489913,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004782743881313564,
      "loss": 4.72,
      "mean_token_accuracy": 0.2396555334329605,
      "num_tokens": 35629868.0,
      "step": 15540
    },
    {
      "entropy": 5.261683940887451,
      "epoch": 1.4932756964457252,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047825968803722315,
      "loss": 4.88,
      "mean_token_accuracy": 0.23469069451093674,
      "num_tokens": 35640622.0,
      "step": 15545
    },
    {
      "entropy": 5.311580562591553,
      "epoch": 1.493756003842459,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004782449832240035,
      "loss": 4.8612,
      "mean_token_accuracy": 0.2268459230661392,
      "num_tokens": 35652383.0,
      "step": 15550
    },
    {
      "entropy": 5.2440534114837645,
      "epoch": 1.494236311239193,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004782302736920387,
      "loss": 4.8013,
      "mean_token_accuracy": 0.23204883188009262,
      "num_tokens": 35663838.0,
      "step": 15555
    },
    {
      "entropy": 5.306398677825928,
      "epoch": 1.4947166186359269,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004782155594416705,
      "loss": 4.8501,
      "mean_token_accuracy": 0.23239926248788834,
      "num_tokens": 35674564.0,
      "step": 15560
    },
    {
      "entropy": 5.366567945480346,
      "epoch": 1.495196926032661,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047820084047324045,
      "loss": 4.8582,
      "mean_token_accuracy": 0.2314916342496872,
      "num_tokens": 35685518.0,
      "step": 15565
    },
    {
      "entropy": 5.3587563037872314,
      "epoch": 1.4956772334293948,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047818611678709027,
      "loss": 4.8821,
      "mean_token_accuracy": 0.23761725574731826,
      "num_tokens": 35696597.0,
      "step": 15570
    },
    {
      "entropy": 5.23171443939209,
      "epoch": 1.4961575408261287,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004781713883835618,
      "loss": 4.7958,
      "mean_token_accuracy": 0.24107930064201355,
      "num_tokens": 35707229.0,
      "step": 15575
    },
    {
      "entropy": 5.261811923980713,
      "epoch": 1.4966378482228626,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047815665526299695,
      "loss": 4.8686,
      "mean_token_accuracy": 0.22915585041046144,
      "num_tokens": 35719440.0,
      "step": 15580
    },
    {
      "entropy": 5.335789918899536,
      "epoch": 1.4971181556195965,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004781419174257378,
      "loss": 4.8827,
      "mean_token_accuracy": 0.22214479446411134,
      "num_tokens": 35731611.0,
      "step": 15585
    },
    {
      "entropy": 5.266561031341553,
      "epoch": 1.4975984630163304,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004781271748721266,
      "loss": 4.8669,
      "mean_token_accuracy": 0.2292900875210762,
      "num_tokens": 35743047.0,
      "step": 15590
    },
    {
      "entropy": 5.325171995162964,
      "epoch": 1.4980787704130645,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004781124276025055,
      "loss": 4.7978,
      "mean_token_accuracy": 0.2474538967013359,
      "num_tokens": 35753499.0,
      "step": 15595
    },
    {
      "entropy": 5.321884393692017,
      "epoch": 1.4985590778097984,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000478097675617217,
      "loss": 5.0284,
      "mean_token_accuracy": 0.22085419595241546,
      "num_tokens": 35764682.0,
      "step": 15600
    },
    {
      "entropy": 5.33892240524292,
      "epoch": 1.4990393852065322,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047808291891660357,
      "loss": 4.8216,
      "mean_token_accuracy": 0.2385969877243042,
      "num_tokens": 35775160.0,
      "step": 15605
    },
    {
      "entropy": 5.340634489059449,
      "epoch": 1.4995196926032661,
      "grad_norm": 1.0,
      "learning_rate": 0.00047806815750100774,
      "loss": 4.8751,
      "mean_token_accuracy": 0.2270866870880127,
      "num_tokens": 35786089.0,
      "step": 15610
    },
    {
      "entropy": 5.308981657028198,
      "epoch": 1.5,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004780533913707723,
      "loss": 4.8822,
      "mean_token_accuracy": 0.22809898406267165,
      "num_tokens": 35796851.0,
      "step": 15615
    },
    {
      "entropy": 5.326308298110962,
      "epoch": 1.5004803073967339,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047803862052624006,
      "loss": 4.9013,
      "mean_token_accuracy": 0.22453001588582994,
      "num_tokens": 35808553.0,
      "step": 15620
    },
    {
      "entropy": 5.389265346527099,
      "epoch": 1.5009606147934678,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047802384496775397,
      "loss": 5.0467,
      "mean_token_accuracy": 0.22353046089410783,
      "num_tokens": 35820108.0,
      "step": 15625
    },
    {
      "entropy": 5.269623041152954,
      "epoch": 1.5014409221902016,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004780090646956571,
      "loss": 4.7801,
      "mean_token_accuracy": 0.23806541860103608,
      "num_tokens": 35831672.0,
      "step": 15630
    },
    {
      "entropy": 5.3786945819854735,
      "epoch": 1.5019212295869355,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047799427971029245,
      "loss": 4.9817,
      "mean_token_accuracy": 0.22599853575229645,
      "num_tokens": 35843164.0,
      "step": 15635
    },
    {
      "entropy": 5.345219516754151,
      "epoch": 1.5024015369836694,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004779794900120034,
      "loss": 4.9628,
      "mean_token_accuracy": 0.22293103039264678,
      "num_tokens": 35854677.0,
      "step": 15640
    },
    {
      "entropy": 5.317985534667969,
      "epoch": 1.5028818443804035,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004779646956011334,
      "loss": 4.864,
      "mean_token_accuracy": 0.2339224174618721,
      "num_tokens": 35865956.0,
      "step": 15645
    },
    {
      "entropy": 5.296347713470459,
      "epoch": 1.5033621517771374,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047794989647802574,
      "loss": 4.8596,
      "mean_token_accuracy": 0.23177069127559663,
      "num_tokens": 35877451.0,
      "step": 15650
    },
    {
      "entropy": 5.287661123275757,
      "epoch": 1.5038424591738713,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047793509264302424,
      "loss": 4.8524,
      "mean_token_accuracy": 0.23361632227897644,
      "num_tokens": 35888436.0,
      "step": 15655
    },
    {
      "entropy": 5.293042707443237,
      "epoch": 1.5043227665706052,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047792028409647237,
      "loss": 4.8871,
      "mean_token_accuracy": 0.22954732179641724,
      "num_tokens": 35901010.0,
      "step": 15660
    },
    {
      "entropy": 5.335793113708496,
      "epoch": 1.5048030739673393,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047790547083871414,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22609178274869918,
      "num_tokens": 35912697.0,
      "step": 15665
    },
    {
      "entropy": 5.2677881717681885,
      "epoch": 1.5052833813640731,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047789065287009335,
      "loss": 4.7897,
      "mean_token_accuracy": 0.23934204280376434,
      "num_tokens": 35924614.0,
      "step": 15670
    },
    {
      "entropy": 5.330773639678955,
      "epoch": 1.505763688760807,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004778758301909542,
      "loss": 4.8835,
      "mean_token_accuracy": 0.23043995946645737,
      "num_tokens": 35935492.0,
      "step": 15675
    },
    {
      "entropy": 5.222614479064942,
      "epoch": 1.506243996157541,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004778610028016405,
      "loss": 4.887,
      "mean_token_accuracy": 0.22976403832435607,
      "num_tokens": 35947901.0,
      "step": 15680
    },
    {
      "entropy": 5.214390468597412,
      "epoch": 1.5067243035542748,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004778461707024967,
      "loss": 4.718,
      "mean_token_accuracy": 0.24358439594507217,
      "num_tokens": 35959690.0,
      "step": 15685
    },
    {
      "entropy": 5.301652574539185,
      "epoch": 1.5072046109510087,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004778313338938672,
      "loss": 4.8064,
      "mean_token_accuracy": 0.24143880009651184,
      "num_tokens": 35971209.0,
      "step": 15690
    },
    {
      "entropy": 5.258506631851196,
      "epoch": 1.5076849183477425,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047781649237609643,
      "loss": 4.7232,
      "mean_token_accuracy": 0.23969053030014037,
      "num_tokens": 35981795.0,
      "step": 15695
    },
    {
      "entropy": 5.258210515975952,
      "epoch": 1.5081652257444764,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004778016461495289,
      "loss": 4.8611,
      "mean_token_accuracy": 0.22966285645961762,
      "num_tokens": 35993358.0,
      "step": 15700
    },
    {
      "entropy": 5.276451492309571,
      "epoch": 1.5086455331412103,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004777867952145094,
      "loss": 4.8908,
      "mean_token_accuracy": 0.23202537447214128,
      "num_tokens": 36005353.0,
      "step": 15705
    },
    {
      "entropy": 5.300562858581543,
      "epoch": 1.5091258405379442,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004777719395713826,
      "loss": 4.8492,
      "mean_token_accuracy": 0.22776930779218674,
      "num_tokens": 36017510.0,
      "step": 15710
    },
    {
      "entropy": 5.289818048477173,
      "epoch": 1.509606147934678,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047775707922049354,
      "loss": 4.8365,
      "mean_token_accuracy": 0.2318302646279335,
      "num_tokens": 36028687.0,
      "step": 15715
    },
    {
      "entropy": 5.2685675621032715,
      "epoch": 1.510086455331412,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004777422141621871,
      "loss": 4.7386,
      "mean_token_accuracy": 0.23630355447530746,
      "num_tokens": 36039851.0,
      "step": 15720
    },
    {
      "entropy": 5.314839124679565,
      "epoch": 1.510566762728146,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004777273443968085,
      "loss": 4.9362,
      "mean_token_accuracy": 0.22915293276309967,
      "num_tokens": 36050776.0,
      "step": 15725
    },
    {
      "entropy": 5.223689699172974,
      "epoch": 1.51104707012488,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004777124699247029,
      "loss": 4.7584,
      "mean_token_accuracy": 0.24307429641485215,
      "num_tokens": 36062151.0,
      "step": 15730
    },
    {
      "entropy": 5.316832733154297,
      "epoch": 1.5115273775216138,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004776975907462157,
      "loss": 4.8875,
      "mean_token_accuracy": 0.23343409597873688,
      "num_tokens": 36074601.0,
      "step": 15735
    },
    {
      "entropy": 5.279083585739135,
      "epoch": 1.5120076849183477,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004776827068616924,
      "loss": 4.7781,
      "mean_token_accuracy": 0.2385477364063263,
      "num_tokens": 36085436.0,
      "step": 15740
    },
    {
      "entropy": 5.352189731597901,
      "epoch": 1.5124879923150818,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004776678182714785,
      "loss": 4.9391,
      "mean_token_accuracy": 0.22232407182455063,
      "num_tokens": 36097589.0,
      "step": 15745
    },
    {
      "entropy": 5.2991596221923825,
      "epoch": 1.5129682997118157,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047765292497591955,
      "loss": 4.866,
      "mean_token_accuracy": 0.23065272271633147,
      "num_tokens": 36109993.0,
      "step": 15750
    },
    {
      "entropy": 5.285720348358154,
      "epoch": 1.5134486071085496,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047763802697536146,
      "loss": 4.8275,
      "mean_token_accuracy": 0.2289637506008148,
      "num_tokens": 36122439.0,
      "step": 15755
    },
    {
      "entropy": 5.292698717117309,
      "epoch": 1.5139289145052834,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047762312427015015,
      "loss": 4.8469,
      "mean_token_accuracy": 0.22954070270061494,
      "num_tokens": 36133867.0,
      "step": 15760
    },
    {
      "entropy": 5.367386436462402,
      "epoch": 1.5144092219020173,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047760821686063153,
      "loss": 4.9477,
      "mean_token_accuracy": 0.2252997025847435,
      "num_tokens": 36144202.0,
      "step": 15765
    },
    {
      "entropy": 5.309964942932129,
      "epoch": 1.5148895292987512,
      "grad_norm": 1.0,
      "learning_rate": 0.00047759330474715173,
      "loss": 4.8435,
      "mean_token_accuracy": 0.23548685759305954,
      "num_tokens": 36154863.0,
      "step": 15770
    },
    {
      "entropy": 5.337285470962525,
      "epoch": 1.515369836695485,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047757838793005704,
      "loss": 4.8975,
      "mean_token_accuracy": 0.22361056953668595,
      "num_tokens": 36166360.0,
      "step": 15775
    },
    {
      "entropy": 5.219887733459473,
      "epoch": 1.515850144092219,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047756346640969366,
      "loss": 4.7713,
      "mean_token_accuracy": 0.2394397646188736,
      "num_tokens": 36177477.0,
      "step": 15780
    },
    {
      "entropy": 5.339203691482544,
      "epoch": 1.5163304514889528,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047754854018640803,
      "loss": 4.892,
      "mean_token_accuracy": 0.22861256897449495,
      "num_tokens": 36188510.0,
      "step": 15785
    },
    {
      "entropy": 5.295318651199341,
      "epoch": 1.5168107588856867,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047753360926054684,
      "loss": 4.8168,
      "mean_token_accuracy": 0.23766406625509262,
      "num_tokens": 36199084.0,
      "step": 15790
    },
    {
      "entropy": 5.243965816497803,
      "epoch": 1.5172910662824206,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047751867363245653,
      "loss": 4.8166,
      "mean_token_accuracy": 0.2291777491569519,
      "num_tokens": 36211265.0,
      "step": 15795
    },
    {
      "entropy": 5.415978050231933,
      "epoch": 1.5177713736791547,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004775037333024841,
      "loss": 4.9716,
      "mean_token_accuracy": 0.217196124792099,
      "num_tokens": 36223070.0,
      "step": 15800
    },
    {
      "entropy": 5.318770170211792,
      "epoch": 1.5182516810758886,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004774887882709762,
      "loss": 4.8975,
      "mean_token_accuracy": 0.23457422405481337,
      "num_tokens": 36235021.0,
      "step": 15805
    },
    {
      "entropy": 5.3214949607849125,
      "epoch": 1.5187319884726225,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047747383853827995,
      "loss": 4.8664,
      "mean_token_accuracy": 0.22893532812595369,
      "num_tokens": 36245647.0,
      "step": 15810
    },
    {
      "entropy": 5.339529514312744,
      "epoch": 1.5192122958693564,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004774588841047424,
      "loss": 4.9311,
      "mean_token_accuracy": 0.22789707630872727,
      "num_tokens": 36257470.0,
      "step": 15815
    },
    {
      "entropy": 5.291301155090332,
      "epoch": 1.5196926032660905,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004774439249707108,
      "loss": 4.804,
      "mean_token_accuracy": 0.2306607887148857,
      "num_tokens": 36268839.0,
      "step": 15820
    },
    {
      "entropy": 5.366343593597412,
      "epoch": 1.5201729106628243,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004774289611365323,
      "loss": 4.9319,
      "mean_token_accuracy": 0.22647317349910737,
      "num_tokens": 36280624.0,
      "step": 15825
    },
    {
      "entropy": 5.307565498352051,
      "epoch": 1.5206532180595582,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047741399260255434,
      "loss": 4.8427,
      "mean_token_accuracy": 0.2340143084526062,
      "num_tokens": 36292696.0,
      "step": 15830
    },
    {
      "entropy": 5.317500829696655,
      "epoch": 1.521133525456292,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047739901936912467,
      "loss": 4.8527,
      "mean_token_accuracy": 0.23304020166397094,
      "num_tokens": 36303612.0,
      "step": 15835
    },
    {
      "entropy": 5.24335241317749,
      "epoch": 1.521613832853026,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004773840414365907,
      "loss": 4.8121,
      "mean_token_accuracy": 0.23160793036222457,
      "num_tokens": 36314511.0,
      "step": 15840
    },
    {
      "entropy": 5.362295055389405,
      "epoch": 1.5220941402497599,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047736905880530026,
      "loss": 4.9499,
      "mean_token_accuracy": 0.22204620391130447,
      "num_tokens": 36327276.0,
      "step": 15845
    },
    {
      "entropy": 5.342337942123413,
      "epoch": 1.5225744476464937,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004773540714756012,
      "loss": 4.8165,
      "mean_token_accuracy": 0.23153068572282792,
      "num_tokens": 36339373.0,
      "step": 15850
    },
    {
      "entropy": 5.359478092193603,
      "epoch": 1.5230547550432276,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047733907944784144,
      "loss": 4.9381,
      "mean_token_accuracy": 0.22310269623994827,
      "num_tokens": 36351863.0,
      "step": 15855
    },
    {
      "entropy": 5.286734533309937,
      "epoch": 1.5235350624399615,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004773240827223691,
      "loss": 4.8437,
      "mean_token_accuracy": 0.23715512752532958,
      "num_tokens": 36363961.0,
      "step": 15860
    },
    {
      "entropy": 5.299909448623657,
      "epoch": 1.5240153698366954,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004773090812995323,
      "loss": 4.8432,
      "mean_token_accuracy": 0.23781957924365998,
      "num_tokens": 36374738.0,
      "step": 15865
    },
    {
      "entropy": 5.283557558059693,
      "epoch": 1.5244956772334293,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047729407517967945,
      "loss": 4.7596,
      "mean_token_accuracy": 0.2381643608212471,
      "num_tokens": 36386472.0,
      "step": 15870
    },
    {
      "entropy": 5.270206451416016,
      "epoch": 1.5249759846301632,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047727906436315884,
      "loss": 4.8684,
      "mean_token_accuracy": 0.2405390202999115,
      "num_tokens": 36397042.0,
      "step": 15875
    },
    {
      "entropy": 5.2528444766998295,
      "epoch": 1.5254562920268973,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047726404885031895,
      "loss": 4.8211,
      "mean_token_accuracy": 0.23520954698324203,
      "num_tokens": 36408720.0,
      "step": 15880
    },
    {
      "entropy": 5.327558851242065,
      "epoch": 1.5259365994236311,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047724902864150845,
      "loss": 4.899,
      "mean_token_accuracy": 0.227407369017601,
      "num_tokens": 36420885.0,
      "step": 15885
    },
    {
      "entropy": 5.3699760913848875,
      "epoch": 1.526416906820365,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00047723400373707607,
      "loss": 4.9056,
      "mean_token_accuracy": 0.22553061693906784,
      "num_tokens": 36433678.0,
      "step": 15890
    },
    {
      "entropy": 5.354255342483521,
      "epoch": 1.526897214217099,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004772189741373707,
      "loss": 4.9168,
      "mean_token_accuracy": 0.23086422979831694,
      "num_tokens": 36445143.0,
      "step": 15895
    },
    {
      "entropy": 5.2422120571136475,
      "epoch": 1.527377521613833,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047720393984274117,
      "loss": 4.8316,
      "mean_token_accuracy": 0.22710686326026916,
      "num_tokens": 36456214.0,
      "step": 15900
    },
    {
      "entropy": 5.311274862289428,
      "epoch": 1.5278578290105669,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047718890085353654,
      "loss": 4.8661,
      "mean_token_accuracy": 0.23183921873569488,
      "num_tokens": 36466767.0,
      "step": 15905
    },
    {
      "entropy": 5.327906465530395,
      "epoch": 1.5283381364073008,
      "grad_norm": 1.109375,
      "learning_rate": 0.000477173857170106,
      "loss": 4.8172,
      "mean_token_accuracy": 0.23434820622205735,
      "num_tokens": 36478150.0,
      "step": 15910
    },
    {
      "entropy": 5.307441091537475,
      "epoch": 1.5288184438040346,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047715880879279894,
      "loss": 4.8609,
      "mean_token_accuracy": 0.22793294489383698,
      "num_tokens": 36488280.0,
      "step": 15915
    },
    {
      "entropy": 5.371834564208984,
      "epoch": 1.5292987512007685,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004771437557219646,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22581709027290345,
      "num_tokens": 36498505.0,
      "step": 15920
    },
    {
      "entropy": 5.285241556167603,
      "epoch": 1.5297790585975024,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004771286979579524,
      "loss": 4.8599,
      "mean_token_accuracy": 0.2285178780555725,
      "num_tokens": 36509271.0,
      "step": 15925
    },
    {
      "entropy": 5.333609294891358,
      "epoch": 1.5302593659942363,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004771136355011121,
      "loss": 4.8991,
      "mean_token_accuracy": 0.22658884078264235,
      "num_tokens": 36520022.0,
      "step": 15930
    },
    {
      "entropy": 5.31448712348938,
      "epoch": 1.5307396733909702,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00047709856835179333,
      "loss": 4.833,
      "mean_token_accuracy": 0.2357179790735245,
      "num_tokens": 36532769.0,
      "step": 15935
    },
    {
      "entropy": 5.351951265335083,
      "epoch": 1.531219980787704,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047708349651034586,
      "loss": 4.884,
      "mean_token_accuracy": 0.23616425096988677,
      "num_tokens": 36544454.0,
      "step": 15940
    },
    {
      "entropy": 5.273876571655274,
      "epoch": 1.531700288184438,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047706841997711974,
      "loss": 4.8505,
      "mean_token_accuracy": 0.23204512596130372,
      "num_tokens": 36555916.0,
      "step": 15945
    },
    {
      "entropy": 5.265675640106201,
      "epoch": 1.5321805955811718,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047705333875246495,
      "loss": 4.8442,
      "mean_token_accuracy": 0.23218553215265275,
      "num_tokens": 36567829.0,
      "step": 15950
    },
    {
      "entropy": 5.351208877563477,
      "epoch": 1.532660902977906,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047703825283673153,
      "loss": 4.9415,
      "mean_token_accuracy": 0.2301754742860794,
      "num_tokens": 36578216.0,
      "step": 15955
    },
    {
      "entropy": 5.34692759513855,
      "epoch": 1.5331412103746398,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004770231622302699,
      "loss": 4.8585,
      "mean_token_accuracy": 0.23205514252185822,
      "num_tokens": 36589945.0,
      "step": 15960
    },
    {
      "entropy": 5.384651517868042,
      "epoch": 1.5336215177713737,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047700806693343016,
      "loss": 4.8795,
      "mean_token_accuracy": 0.228605979681015,
      "num_tokens": 36600724.0,
      "step": 15965
    },
    {
      "entropy": 5.208251810073852,
      "epoch": 1.5341018251681076,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047699296694656316,
      "loss": 4.6993,
      "mean_token_accuracy": 0.24553751349449157,
      "num_tokens": 36611154.0,
      "step": 15970
    },
    {
      "entropy": 5.285090970993042,
      "epoch": 1.5345821325648417,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004769778622700192,
      "loss": 4.8096,
      "mean_token_accuracy": 0.22953053712844848,
      "num_tokens": 36621750.0,
      "step": 15975
    },
    {
      "entropy": 5.326995992660523,
      "epoch": 1.5350624399615755,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047696275290414885,
      "loss": 4.8243,
      "mean_token_accuracy": 0.23363364338874817,
      "num_tokens": 36633294.0,
      "step": 15980
    },
    {
      "entropy": 5.24855751991272,
      "epoch": 1.5355427473583094,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047694763884930324,
      "loss": 4.7912,
      "mean_token_accuracy": 0.23363662362098694,
      "num_tokens": 36646377.0,
      "step": 15985
    },
    {
      "entropy": 5.244277572631836,
      "epoch": 1.5360230547550433,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047693252010583314,
      "loss": 4.7515,
      "mean_token_accuracy": 0.2385295510292053,
      "num_tokens": 36656159.0,
      "step": 15990
    },
    {
      "entropy": 5.25462589263916,
      "epoch": 1.5365033621517772,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004769173966740895,
      "loss": 4.8946,
      "mean_token_accuracy": 0.22452451139688492,
      "num_tokens": 36667522.0,
      "step": 15995
    },
    {
      "entropy": 5.38155083656311,
      "epoch": 1.536983669548511,
      "grad_norm": 1.0,
      "learning_rate": 0.00047690226855442346,
      "loss": 4.8986,
      "mean_token_accuracy": 0.2294987827539444,
      "num_tokens": 36678662.0,
      "step": 16000
    },
    {
      "entropy": 5.315290546417236,
      "epoch": 1.537463976945245,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004768871357471863,
      "loss": 4.8697,
      "mean_token_accuracy": 0.23045988082885743,
      "num_tokens": 36689876.0,
      "step": 16005
    },
    {
      "entropy": 5.286601209640503,
      "epoch": 1.5379442843419788,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047687199825272936,
      "loss": 4.8151,
      "mean_token_accuracy": 0.23384485244750977,
      "num_tokens": 36701140.0,
      "step": 16010
    },
    {
      "entropy": 5.347996377944947,
      "epoch": 1.5384245917387127,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047685685607140403,
      "loss": 4.9575,
      "mean_token_accuracy": 0.22559704631567,
      "num_tokens": 36711077.0,
      "step": 16015
    },
    {
      "entropy": 5.336470222473144,
      "epoch": 1.5389048991354466,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047684170920356185,
      "loss": 4.8364,
      "mean_token_accuracy": 0.23338758945465088,
      "num_tokens": 36722705.0,
      "step": 16020
    },
    {
      "entropy": 5.266508007049561,
      "epoch": 1.5393852065321805,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004768265576495546,
      "loss": 4.7887,
      "mean_token_accuracy": 0.23464114665985109,
      "num_tokens": 36733446.0,
      "step": 16025
    },
    {
      "entropy": 5.236494827270508,
      "epoch": 1.5398655139289144,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047681140140973396,
      "loss": 4.8282,
      "mean_token_accuracy": 0.2391822576522827,
      "num_tokens": 36744529.0,
      "step": 16030
    },
    {
      "entropy": 5.203745794296265,
      "epoch": 1.5403458213256485,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004767962404844517,
      "loss": 4.8283,
      "mean_token_accuracy": 0.23280325829982756,
      "num_tokens": 36756248.0,
      "step": 16035
    },
    {
      "entropy": 5.37889142036438,
      "epoch": 1.5408261287223823,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047678107487406015,
      "loss": 4.9203,
      "mean_token_accuracy": 0.22914234846830367,
      "num_tokens": 36768097.0,
      "step": 16040
    },
    {
      "entropy": 5.3123187065124515,
      "epoch": 1.5413064361191162,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047676590457891116,
      "loss": 4.845,
      "mean_token_accuracy": 0.23210765570402145,
      "num_tokens": 36780395.0,
      "step": 16045
    },
    {
      "entropy": 5.2832153797149655,
      "epoch": 1.54178674351585,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004767507295993569,
      "loss": 4.8235,
      "mean_token_accuracy": 0.22845427095890045,
      "num_tokens": 36791897.0,
      "step": 16050
    },
    {
      "entropy": 5.321641635894776,
      "epoch": 1.5422670509125842,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004767355499357498,
      "loss": 4.8175,
      "mean_token_accuracy": 0.24046906381845473,
      "num_tokens": 36802716.0,
      "step": 16055
    },
    {
      "entropy": 5.286835718154907,
      "epoch": 1.542747358309318,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004767203655884423,
      "loss": 4.8754,
      "mean_token_accuracy": 0.22742675095796586,
      "num_tokens": 36814471.0,
      "step": 16060
    },
    {
      "entropy": 5.2745355606079105,
      "epoch": 1.543227665706052,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004767051765577869,
      "loss": 4.801,
      "mean_token_accuracy": 0.2397352248430252,
      "num_tokens": 36825975.0,
      "step": 16065
    },
    {
      "entropy": 5.352064847946167,
      "epoch": 1.5437079731027858,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047668998284413624,
      "loss": 4.9313,
      "mean_token_accuracy": 0.22164277583360673,
      "num_tokens": 36837990.0,
      "step": 16070
    },
    {
      "entropy": 5.32035756111145,
      "epoch": 1.5441882804995197,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047667478444784306,
      "loss": 4.9231,
      "mean_token_accuracy": 0.22442851662635804,
      "num_tokens": 36849115.0,
      "step": 16075
    },
    {
      "entropy": 5.311487102508545,
      "epoch": 1.5446685878962536,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004766595813692602,
      "loss": 4.7617,
      "mean_token_accuracy": 0.23538754731416703,
      "num_tokens": 36860626.0,
      "step": 16080
    },
    {
      "entropy": 5.3318342685699465,
      "epoch": 1.5451488952929875,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047664437360874076,
      "loss": 4.9379,
      "mean_token_accuracy": 0.22516281753778458,
      "num_tokens": 36871926.0,
      "step": 16085
    },
    {
      "entropy": 5.298501920700073,
      "epoch": 1.5456292026897214,
      "grad_norm": 1.0,
      "learning_rate": 0.00047662916116663766,
      "loss": 4.8375,
      "mean_token_accuracy": 0.23412572592496872,
      "num_tokens": 36883511.0,
      "step": 16090
    },
    {
      "entropy": 5.3957396984100345,
      "epoch": 1.5461095100864553,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047661394404330417,
      "loss": 4.9061,
      "mean_token_accuracy": 0.23284603506326676,
      "num_tokens": 36895468.0,
      "step": 16095
    },
    {
      "entropy": 5.297923421859741,
      "epoch": 1.5465898174831891,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047659872223909357,
      "loss": 4.8113,
      "mean_token_accuracy": 0.23344242870807647,
      "num_tokens": 36906957.0,
      "step": 16100
    },
    {
      "entropy": 5.302119779586792,
      "epoch": 1.547070124879923,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004765834957543592,
      "loss": 4.8348,
      "mean_token_accuracy": 0.23169482499361038,
      "num_tokens": 36916495.0,
      "step": 16105
    },
    {
      "entropy": 5.263612508773804,
      "epoch": 1.547550432276657,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047656826458945475,
      "loss": 4.8338,
      "mean_token_accuracy": 0.22775180488824845,
      "num_tokens": 36927301.0,
      "step": 16110
    },
    {
      "entropy": 5.281066656112671,
      "epoch": 1.548030739673391,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047655302874473365,
      "loss": 4.8058,
      "mean_token_accuracy": 0.22971298843622207,
      "num_tokens": 36938116.0,
      "step": 16115
    },
    {
      "entropy": 5.353935813903808,
      "epoch": 1.5485110470701249,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004765377882205498,
      "loss": 4.8552,
      "mean_token_accuracy": 0.23418129086494446,
      "num_tokens": 36950292.0,
      "step": 16120
    },
    {
      "entropy": 5.25476770401001,
      "epoch": 1.5489913544668588,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004765225430172568,
      "loss": 4.8751,
      "mean_token_accuracy": 0.23433665484189986,
      "num_tokens": 36962017.0,
      "step": 16125
    },
    {
      "entropy": 5.3294517517089846,
      "epoch": 1.5494716618635929,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004765072931352089,
      "loss": 4.8864,
      "mean_token_accuracy": 0.22833587378263473,
      "num_tokens": 36972831.0,
      "step": 16130
    },
    {
      "entropy": 5.309158039093018,
      "epoch": 1.5499519692603267,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047649203857476,
      "loss": 4.7749,
      "mean_token_accuracy": 0.2415161684155464,
      "num_tokens": 36983324.0,
      "step": 16135
    },
    {
      "entropy": 5.342039155960083,
      "epoch": 1.5504322766570606,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047647677933626423,
      "loss": 5.0235,
      "mean_token_accuracy": 0.2159330815076828,
      "num_tokens": 36996093.0,
      "step": 16140
    },
    {
      "entropy": 5.296381187438965,
      "epoch": 1.5509125840537945,
      "grad_norm": 1.0,
      "learning_rate": 0.00047646151542007583,
      "loss": 4.84,
      "mean_token_accuracy": 0.23520279079675674,
      "num_tokens": 37006548.0,
      "step": 16145
    },
    {
      "entropy": 5.299122619628906,
      "epoch": 1.5513928914505284,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004764462468265494,
      "loss": 4.847,
      "mean_token_accuracy": 0.2285194218158722,
      "num_tokens": 37019181.0,
      "step": 16150
    },
    {
      "entropy": 5.354714107513428,
      "epoch": 1.5518731988472623,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047643097355603913,
      "loss": 4.8779,
      "mean_token_accuracy": 0.23139499574899675,
      "num_tokens": 37029808.0,
      "step": 16155
    },
    {
      "entropy": 5.296012353897095,
      "epoch": 1.5523535062439962,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004764156956088999,
      "loss": 4.8492,
      "mean_token_accuracy": 0.2287864089012146,
      "num_tokens": 37040352.0,
      "step": 16160
    },
    {
      "entropy": 5.26588454246521,
      "epoch": 1.55283381364073,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004764004129854863,
      "loss": 4.8847,
      "mean_token_accuracy": 0.2347981035709381,
      "num_tokens": 37050780.0,
      "step": 16165
    },
    {
      "entropy": 5.299176454544067,
      "epoch": 1.553314121037464,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047638512568615307,
      "loss": 4.8185,
      "mean_token_accuracy": 0.23486877232789993,
      "num_tokens": 37062608.0,
      "step": 16170
    },
    {
      "entropy": 5.342418241500854,
      "epoch": 1.5537944284341978,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004763698337112553,
      "loss": 4.9244,
      "mean_token_accuracy": 0.22450561970472335,
      "num_tokens": 37074578.0,
      "step": 16175
    },
    {
      "entropy": 5.3515702247619625,
      "epoch": 1.5542747358309317,
      "grad_norm": 1.0,
      "learning_rate": 0.0004763545370611479,
      "loss": 4.9555,
      "mean_token_accuracy": 0.22807884216308594,
      "num_tokens": 37086487.0,
      "step": 16180
    },
    {
      "entropy": 5.306707048416138,
      "epoch": 1.5547550432276656,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047633923573618605,
      "loss": 4.8202,
      "mean_token_accuracy": 0.23750915825366975,
      "num_tokens": 37097249.0,
      "step": 16185
    },
    {
      "entropy": 5.41294002532959,
      "epoch": 1.5552353506243997,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000476323929736725,
      "loss": 5.0215,
      "mean_token_accuracy": 0.22022175639867783,
      "num_tokens": 37109274.0,
      "step": 16190
    },
    {
      "entropy": 5.3014778137207035,
      "epoch": 1.5557156580211335,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047630861906312004,
      "loss": 4.8278,
      "mean_token_accuracy": 0.23897305876016617,
      "num_tokens": 37120100.0,
      "step": 16195
    },
    {
      "entropy": 5.2865681648254395,
      "epoch": 1.5561959654178674,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004762933037157268,
      "loss": 4.8327,
      "mean_token_accuracy": 0.22644709646701813,
      "num_tokens": 37131889.0,
      "step": 16200
    },
    {
      "entropy": 5.242715835571289,
      "epoch": 1.5566762728146013,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047627798369490076,
      "loss": 4.7965,
      "mean_token_accuracy": 0.23715779036283494,
      "num_tokens": 37142538.0,
      "step": 16205
    },
    {
      "entropy": 5.386628580093384,
      "epoch": 1.5571565802113354,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047626265900099757,
      "loss": 4.9129,
      "mean_token_accuracy": 0.23105997443199158,
      "num_tokens": 37153639.0,
      "step": 16210
    },
    {
      "entropy": 5.333990001678467,
      "epoch": 1.5576368876080693,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00047624732963437314,
      "loss": 4.8545,
      "mean_token_accuracy": 0.23212917894124985,
      "num_tokens": 37164029.0,
      "step": 16215
    },
    {
      "entropy": 5.321325206756592,
      "epoch": 1.5581171950048032,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047623199559538324,
      "loss": 4.9352,
      "mean_token_accuracy": 0.2245692327618599,
      "num_tokens": 37175880.0,
      "step": 16220
    },
    {
      "entropy": 5.341187953948975,
      "epoch": 1.558597502401537,
      "grad_norm": 1.015625,
      "learning_rate": 0.000476216656884384,
      "loss": 4.8642,
      "mean_token_accuracy": 0.22681307792663574,
      "num_tokens": 37187483.0,
      "step": 16225
    },
    {
      "entropy": 5.312691926956177,
      "epoch": 1.559077809798271,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047620131350173135,
      "loss": 4.8809,
      "mean_token_accuracy": 0.23059872090816497,
      "num_tokens": 37198978.0,
      "step": 16230
    },
    {
      "entropy": 5.3702775001525875,
      "epoch": 1.5595581171950048,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004761859654477817,
      "loss": 4.9024,
      "mean_token_accuracy": 0.21752983331680298,
      "num_tokens": 37210054.0,
      "step": 16235
    },
    {
      "entropy": 5.336034488677979,
      "epoch": 1.5600384245917387,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004761706127228914,
      "loss": 4.8615,
      "mean_token_accuracy": 0.23263344466686248,
      "num_tokens": 37221744.0,
      "step": 16240
    },
    {
      "entropy": 5.281007862091064,
      "epoch": 1.5605187319884726,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004761552553274168,
      "loss": 4.8858,
      "mean_token_accuracy": 0.2331802323460579,
      "num_tokens": 37234629.0,
      "step": 16245
    },
    {
      "entropy": 5.368319845199585,
      "epoch": 1.5609990393852065,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004761398932617144,
      "loss": 4.9603,
      "mean_token_accuracy": 0.22229765653610228,
      "num_tokens": 37246960.0,
      "step": 16250
    },
    {
      "entropy": 5.277141761779785,
      "epoch": 1.5614793467819403,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000476124526526141,
      "loss": 4.8214,
      "mean_token_accuracy": 0.23109882175922394,
      "num_tokens": 37259326.0,
      "step": 16255
    },
    {
      "entropy": 5.282608556747436,
      "epoch": 1.5619596541786742,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047610915512105327,
      "loss": 4.8596,
      "mean_token_accuracy": 0.22780956774950029,
      "num_tokens": 37271436.0,
      "step": 16260
    },
    {
      "entropy": 5.336248064041138,
      "epoch": 1.562439961575408,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004760937790468082,
      "loss": 4.9345,
      "mean_token_accuracy": 0.22453770488500596,
      "num_tokens": 37283230.0,
      "step": 16265
    },
    {
      "entropy": 5.32380862236023,
      "epoch": 1.5629202689721422,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004760783983037627,
      "loss": 4.8333,
      "mean_token_accuracy": 0.23718595653772354,
      "num_tokens": 37295185.0,
      "step": 16270
    },
    {
      "entropy": 5.346075057983398,
      "epoch": 1.563400576368876,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004760630128922738,
      "loss": 4.8901,
      "mean_token_accuracy": 0.2300552397966385,
      "num_tokens": 37307573.0,
      "step": 16275
    },
    {
      "entropy": 5.390898036956787,
      "epoch": 1.56388088376561,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004760476228126989,
      "loss": 4.926,
      "mean_token_accuracy": 0.227378211915493,
      "num_tokens": 37319084.0,
      "step": 16280
    },
    {
      "entropy": 5.227478790283203,
      "epoch": 1.5643611911623438,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004760322280653951,
      "loss": 4.8144,
      "mean_token_accuracy": 0.23728471249341965,
      "num_tokens": 37330447.0,
      "step": 16285
    },
    {
      "entropy": 5.260581636428833,
      "epoch": 1.564841498559078,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004760168286507199,
      "loss": 4.7553,
      "mean_token_accuracy": 0.2303868427872658,
      "num_tokens": 37341184.0,
      "step": 16290
    },
    {
      "entropy": 5.331058835983276,
      "epoch": 1.5653218059558118,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047600142456903085,
      "loss": 4.8784,
      "mean_token_accuracy": 0.22778766006231307,
      "num_tokens": 37352712.0,
      "step": 16295
    },
    {
      "entropy": 5.368314933776856,
      "epoch": 1.5658021133525457,
      "grad_norm": 1.0,
      "learning_rate": 0.00047598601582068555,
      "loss": 4.9538,
      "mean_token_accuracy": 0.22532380521297454,
      "num_tokens": 37365001.0,
      "step": 16300
    },
    {
      "entropy": 5.388139009475708,
      "epoch": 1.5662824207492796,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004759706024060418,
      "loss": 4.9896,
      "mean_token_accuracy": 0.22012481093406677,
      "num_tokens": 37377005.0,
      "step": 16305
    },
    {
      "entropy": 5.32170672416687,
      "epoch": 1.5667627281460135,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004759551843254575,
      "loss": 4.8367,
      "mean_token_accuracy": 0.23388337790966035,
      "num_tokens": 37387329.0,
      "step": 16310
    },
    {
      "entropy": 5.415711498260498,
      "epoch": 1.5672430355427474,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047593976157929034,
      "loss": 4.9102,
      "mean_token_accuracy": 0.2313454568386078,
      "num_tokens": 37398894.0,
      "step": 16315
    },
    {
      "entropy": 5.3518674850463865,
      "epoch": 1.5677233429394812,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004759243341678987,
      "loss": 4.8369,
      "mean_token_accuracy": 0.23481279760599136,
      "num_tokens": 37409913.0,
      "step": 16320
    },
    {
      "entropy": 5.3368377685546875,
      "epoch": 1.5682036503362151,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004759089020916407,
      "loss": 4.8088,
      "mean_token_accuracy": 0.2272825673222542,
      "num_tokens": 37421859.0,
      "step": 16325
    },
    {
      "entropy": 5.325921106338501,
      "epoch": 1.568683957732949,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047589346535087444,
      "loss": 4.8651,
      "mean_token_accuracy": 0.22855456322431564,
      "num_tokens": 37432827.0,
      "step": 16330
    },
    {
      "entropy": 5.302629995346069,
      "epoch": 1.5691642651296829,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004758780239459586,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2396998643875122,
      "num_tokens": 37444171.0,
      "step": 16335
    },
    {
      "entropy": 5.2967935562133786,
      "epoch": 1.5696445725264168,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004758625778772514,
      "loss": 4.836,
      "mean_token_accuracy": 0.22940060794353484,
      "num_tokens": 37455194.0,
      "step": 16340
    },
    {
      "entropy": 5.214334630966187,
      "epoch": 1.5701248799231509,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047584712714511166,
      "loss": 4.7727,
      "mean_token_accuracy": 0.23967199474573136,
      "num_tokens": 37465839.0,
      "step": 16345
    },
    {
      "entropy": 5.325697231292724,
      "epoch": 1.5706051873198847,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047583167174989797,
      "loss": 4.8427,
      "mean_token_accuracy": 0.23536853939294816,
      "num_tokens": 37476948.0,
      "step": 16350
    },
    {
      "entropy": 5.340693616867066,
      "epoch": 1.5710854947166186,
      "grad_norm": 1.0,
      "learning_rate": 0.0004758162116919692,
      "loss": 4.85,
      "mean_token_accuracy": 0.22908419817686082,
      "num_tokens": 37487382.0,
      "step": 16355
    },
    {
      "entropy": 5.263298034667969,
      "epoch": 1.5715658021133525,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047580074697168434,
      "loss": 4.8173,
      "mean_token_accuracy": 0.23902253061532974,
      "num_tokens": 37498347.0,
      "step": 16360
    },
    {
      "entropy": 5.311135196685791,
      "epoch": 1.5720461095100866,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047578527758940236,
      "loss": 4.9131,
      "mean_token_accuracy": 0.2248442590236664,
      "num_tokens": 37509620.0,
      "step": 16365
    },
    {
      "entropy": 5.321051406860351,
      "epoch": 1.5725264169068205,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004757698035454825,
      "loss": 4.8065,
      "mean_token_accuracy": 0.23575814813375473,
      "num_tokens": 37519768.0,
      "step": 16370
    },
    {
      "entropy": 5.284927034378052,
      "epoch": 1.5730067243035544,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004757543248402839,
      "loss": 4.7748,
      "mean_token_accuracy": 0.2311959981918335,
      "num_tokens": 37530476.0,
      "step": 16375
    },
    {
      "entropy": 5.304613018035889,
      "epoch": 1.5734870317002883,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047573884147416597,
      "loss": 4.8908,
      "mean_token_accuracy": 0.23279555439949035,
      "num_tokens": 37541702.0,
      "step": 16380
    },
    {
      "entropy": 5.354133749008179,
      "epoch": 1.5739673390970221,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004757233534474883,
      "loss": 4.8699,
      "mean_token_accuracy": 0.22893128097057341,
      "num_tokens": 37553323.0,
      "step": 16385
    },
    {
      "entropy": 5.353621482849121,
      "epoch": 1.574447646493756,
      "grad_norm": 1.125,
      "learning_rate": 0.0004757078607606103,
      "loss": 4.9341,
      "mean_token_accuracy": 0.21915322095155715,
      "num_tokens": 37564882.0,
      "step": 16390
    },
    {
      "entropy": 5.290194845199585,
      "epoch": 1.57492795389049,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004756923634138918,
      "loss": 4.8605,
      "mean_token_accuracy": 0.229867622256279,
      "num_tokens": 37575612.0,
      "step": 16395
    },
    {
      "entropy": 5.2875903129577635,
      "epoch": 1.5754082612872238,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047567686140769264,
      "loss": 4.8708,
      "mean_token_accuracy": 0.2304797053337097,
      "num_tokens": 37587089.0,
      "step": 16400
    },
    {
      "entropy": 5.344682025909424,
      "epoch": 1.5758885686839577,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047566135474237247,
      "loss": 4.8242,
      "mean_token_accuracy": 0.230683134496212,
      "num_tokens": 37598429.0,
      "step": 16405
    },
    {
      "entropy": 5.376549291610718,
      "epoch": 1.5763688760806915,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047564584341829166,
      "loss": 4.9342,
      "mean_token_accuracy": 0.22744107693433763,
      "num_tokens": 37609769.0,
      "step": 16410
    },
    {
      "entropy": 5.333934688568116,
      "epoch": 1.5768491834774254,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047563032743581,
      "loss": 4.971,
      "mean_token_accuracy": 0.2224176123738289,
      "num_tokens": 37622425.0,
      "step": 16415
    },
    {
      "entropy": 5.322865533828735,
      "epoch": 1.5773294908741593,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00047561480679528804,
      "loss": 4.8871,
      "mean_token_accuracy": 0.23216375857591628,
      "num_tokens": 37634819.0,
      "step": 16420
    },
    {
      "entropy": 5.31697998046875,
      "epoch": 1.5778097982708934,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004755992814970859,
      "loss": 4.7615,
      "mean_token_accuracy": 0.23547240644693374,
      "num_tokens": 37645531.0,
      "step": 16425
    },
    {
      "entropy": 5.267870044708252,
      "epoch": 1.5782901056676273,
      "grad_norm": 1.0,
      "learning_rate": 0.000475583751541564,
      "loss": 4.8218,
      "mean_token_accuracy": 0.2314648449420929,
      "num_tokens": 37657243.0,
      "step": 16430
    },
    {
      "entropy": 5.36224365234375,
      "epoch": 1.5787704130643612,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047556821692908315,
      "loss": 4.8921,
      "mean_token_accuracy": 0.2308179259300232,
      "num_tokens": 37668218.0,
      "step": 16435
    },
    {
      "entropy": 5.301560735702514,
      "epoch": 1.579250720461095,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004755526776600038,
      "loss": 4.8181,
      "mean_token_accuracy": 0.23066598325967788,
      "num_tokens": 37680173.0,
      "step": 16440
    },
    {
      "entropy": 5.296558094024658,
      "epoch": 1.5797310278578292,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047553713373468684,
      "loss": 4.8528,
      "mean_token_accuracy": 0.2263149306178093,
      "num_tokens": 37691281.0,
      "step": 16445
    },
    {
      "entropy": 5.358935165405273,
      "epoch": 1.580211335254563,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047552158515349306,
      "loss": 4.9292,
      "mean_token_accuracy": 0.22107756435871123,
      "num_tokens": 37702320.0,
      "step": 16450
    },
    {
      "entropy": 5.3044578552246096,
      "epoch": 1.580691642651297,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047550603191678356,
      "loss": 4.8419,
      "mean_token_accuracy": 0.2271473526954651,
      "num_tokens": 37713724.0,
      "step": 16455
    },
    {
      "entropy": 5.3035468578338625,
      "epoch": 1.5811719500480308,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004754904740249194,
      "loss": 4.8347,
      "mean_token_accuracy": 0.23166981041431428,
      "num_tokens": 37724532.0,
      "step": 16460
    },
    {
      "entropy": 5.3303100109100345,
      "epoch": 1.5816522574447647,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047547491147826156,
      "loss": 4.7967,
      "mean_token_accuracy": 0.24148535430431367,
      "num_tokens": 37735174.0,
      "step": 16465
    },
    {
      "entropy": 5.307122707366943,
      "epoch": 1.5821325648414986,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004754593442771718,
      "loss": 4.942,
      "mean_token_accuracy": 0.22628413289785385,
      "num_tokens": 37746981.0,
      "step": 16470
    },
    {
      "entropy": 5.338410997390747,
      "epoch": 1.5826128722382324,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047544377242201115,
      "loss": 4.8756,
      "mean_token_accuracy": 0.23064044415950774,
      "num_tokens": 37757576.0,
      "step": 16475
    },
    {
      "entropy": 5.361675167083741,
      "epoch": 1.5830931796349663,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047542819591314136,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22386698424816132,
      "num_tokens": 37769053.0,
      "step": 16480
    },
    {
      "entropy": 5.3916079044342045,
      "epoch": 1.5835734870317002,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004754126147509241,
      "loss": 4.8657,
      "mean_token_accuracy": 0.2248765856027603,
      "num_tokens": 37780517.0,
      "step": 16485
    },
    {
      "entropy": 5.424533176422119,
      "epoch": 1.584053794428434,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047539702893572086,
      "loss": 4.9455,
      "mean_token_accuracy": 0.22334675192832948,
      "num_tokens": 37791333.0,
      "step": 16490
    },
    {
      "entropy": 5.298500871658325,
      "epoch": 1.584534101825168,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047538143846789376,
      "loss": 4.8064,
      "mean_token_accuracy": 0.23906953483819962,
      "num_tokens": 37802635.0,
      "step": 16495
    },
    {
      "entropy": 5.241256856918335,
      "epoch": 1.585014409221902,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004753658433478047,
      "loss": 4.7196,
      "mean_token_accuracy": 0.24023620635271073,
      "num_tokens": 37814788.0,
      "step": 16500
    },
    {
      "entropy": 5.308159399032593,
      "epoch": 1.585494716618636,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047535024357581564,
      "loss": 4.8245,
      "mean_token_accuracy": 0.2347756326198578,
      "num_tokens": 37826650.0,
      "step": 16505
    },
    {
      "entropy": 5.3455774784088135,
      "epoch": 1.5859750240153698,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004753346391522889,
      "loss": 4.9135,
      "mean_token_accuracy": 0.2256648525595665,
      "num_tokens": 37839362.0,
      "step": 16510
    },
    {
      "entropy": 5.346570682525635,
      "epoch": 1.5864553314121037,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047531903007758667,
      "loss": 4.8755,
      "mean_token_accuracy": 0.2271828681230545,
      "num_tokens": 37851057.0,
      "step": 16515
    },
    {
      "entropy": 5.286417722702026,
      "epoch": 1.5869356388088378,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004753034163520714,
      "loss": 4.7907,
      "mean_token_accuracy": 0.24084875136613845,
      "num_tokens": 37863507.0,
      "step": 16520
    },
    {
      "entropy": 5.333589220046997,
      "epoch": 1.5874159462055717,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047528779797610557,
      "loss": 4.8683,
      "mean_token_accuracy": 0.2257616102695465,
      "num_tokens": 37874439.0,
      "step": 16525
    },
    {
      "entropy": 5.3353879928588865,
      "epoch": 1.5878962536023056,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047527217495005184,
      "loss": 4.8308,
      "mean_token_accuracy": 0.23060812652111054,
      "num_tokens": 37886720.0,
      "step": 16530
    },
    {
      "entropy": 5.350425338745117,
      "epoch": 1.5883765609990395,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047525654727427285,
      "loss": 4.8922,
      "mean_token_accuracy": 0.22747205942869186,
      "num_tokens": 37897919.0,
      "step": 16535
    },
    {
      "entropy": 5.346773147583008,
      "epoch": 1.5888568683957733,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004752409149491315,
      "loss": 4.9522,
      "mean_token_accuracy": 0.2296921730041504,
      "num_tokens": 37911061.0,
      "step": 16540
    },
    {
      "entropy": 5.3656598091125485,
      "epoch": 1.5893371757925072,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047522527797499075,
      "loss": 4.9094,
      "mean_token_accuracy": 0.22531614303588868,
      "num_tokens": 37922677.0,
      "step": 16545
    },
    {
      "entropy": 5.213374614715576,
      "epoch": 1.589817483189241,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004752096363522135,
      "loss": 4.72,
      "mean_token_accuracy": 0.23779584765434264,
      "num_tokens": 37932802.0,
      "step": 16550
    },
    {
      "entropy": 5.2971234798431395,
      "epoch": 1.590297790585975,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047519399008116305,
      "loss": 4.8419,
      "mean_token_accuracy": 0.23157459497451782,
      "num_tokens": 37944782.0,
      "step": 16555
    },
    {
      "entropy": 5.315040397644043,
      "epoch": 1.5907780979827089,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004751783391622026,
      "loss": 4.8494,
      "mean_token_accuracy": 0.2319590762257576,
      "num_tokens": 37956577.0,
      "step": 16560
    },
    {
      "entropy": 5.37283992767334,
      "epoch": 1.5912584053794427,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004751626835956955,
      "loss": 4.9094,
      "mean_token_accuracy": 0.2282887727022171,
      "num_tokens": 37967937.0,
      "step": 16565
    },
    {
      "entropy": 5.282363510131836,
      "epoch": 1.5917387127761766,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004751470233820053,
      "loss": 4.7972,
      "mean_token_accuracy": 0.24421661645174025,
      "num_tokens": 37978113.0,
      "step": 16570
    },
    {
      "entropy": 5.235817813873291,
      "epoch": 1.5922190201729105,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004751313585214955,
      "loss": 4.7773,
      "mean_token_accuracy": 0.2378433659672737,
      "num_tokens": 37988852.0,
      "step": 16575
    },
    {
      "entropy": 5.314593315124512,
      "epoch": 1.5926993275696446,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004751156890145298,
      "loss": 4.8645,
      "mean_token_accuracy": 0.22487269937992097,
      "num_tokens": 38000531.0,
      "step": 16580
    },
    {
      "entropy": 5.270912742614746,
      "epoch": 1.5931796349663785,
      "grad_norm": 1.125,
      "learning_rate": 0.000475100014861472,
      "loss": 4.8474,
      "mean_token_accuracy": 0.2319420948624611,
      "num_tokens": 38011818.0,
      "step": 16585
    },
    {
      "entropy": 5.336111783981323,
      "epoch": 1.5936599423631124,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004750843360626861,
      "loss": 4.8678,
      "mean_token_accuracy": 0.23008830696344376,
      "num_tokens": 38022455.0,
      "step": 16590
    },
    {
      "entropy": 5.282876491546631,
      "epoch": 1.5941402497598463,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004750686526185359,
      "loss": 4.8362,
      "mean_token_accuracy": 0.24025724530220033,
      "num_tokens": 38033126.0,
      "step": 16595
    },
    {
      "entropy": 5.159281301498413,
      "epoch": 1.5946205571565804,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047505296452938584,
      "loss": 4.7732,
      "mean_token_accuracy": 0.23416064083576202,
      "num_tokens": 38044935.0,
      "step": 16600
    },
    {
      "entropy": 5.328322410583496,
      "epoch": 1.5951008645533142,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047503727179559995,
      "loss": 4.8803,
      "mean_token_accuracy": 0.23503536581993104,
      "num_tokens": 38056080.0,
      "step": 16605
    },
    {
      "entropy": 5.349979400634766,
      "epoch": 1.5955811719500481,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047502157441754256,
      "loss": 4.8914,
      "mean_token_accuracy": 0.2258293956518173,
      "num_tokens": 38066788.0,
      "step": 16610
    },
    {
      "entropy": 5.351882839202881,
      "epoch": 1.596061479346782,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004750058723955781,
      "loss": 4.9091,
      "mean_token_accuracy": 0.23327139019966125,
      "num_tokens": 38079529.0,
      "step": 16615
    },
    {
      "entropy": 5.313291358947754,
      "epoch": 1.5965417867435159,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004749901657300713,
      "loss": 4.8654,
      "mean_token_accuracy": 0.22945188730955124,
      "num_tokens": 38091484.0,
      "step": 16620
    },
    {
      "entropy": 5.361520433425904,
      "epoch": 1.5970220941402498,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047497445442138667,
      "loss": 4.8081,
      "mean_token_accuracy": 0.23376710563898087,
      "num_tokens": 38103102.0,
      "step": 16625
    },
    {
      "entropy": 5.302425861358643,
      "epoch": 1.5975024015369836,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047495873846988896,
      "loss": 4.9273,
      "mean_token_accuracy": 0.22163014709949494,
      "num_tokens": 38115565.0,
      "step": 16630
    },
    {
      "entropy": 5.318923425674439,
      "epoch": 1.5979827089337175,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004749430178759431,
      "loss": 4.851,
      "mean_token_accuracy": 0.23588893413543702,
      "num_tokens": 38126741.0,
      "step": 16635
    },
    {
      "entropy": 5.300588130950928,
      "epoch": 1.5984630163304514,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047492729263991413,
      "loss": 4.8341,
      "mean_token_accuracy": 0.23471156358718873,
      "num_tokens": 38137149.0,
      "step": 16640
    },
    {
      "entropy": 5.265857887268067,
      "epoch": 1.5989433237271853,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047491156276216695,
      "loss": 4.7847,
      "mean_token_accuracy": 0.2450176551938057,
      "num_tokens": 38148537.0,
      "step": 16645
    },
    {
      "entropy": 5.28712100982666,
      "epoch": 1.5994236311239192,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047489582824306704,
      "loss": 4.8231,
      "mean_token_accuracy": 0.23426364958286286,
      "num_tokens": 38159097.0,
      "step": 16650
    },
    {
      "entropy": 5.276128196716309,
      "epoch": 1.5999039385206533,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047488008908297955,
      "loss": 4.796,
      "mean_token_accuracy": 0.2390404999256134,
      "num_tokens": 38171400.0,
      "step": 16655
    },
    {
      "entropy": 5.369984102249146,
      "epoch": 1.6003842459173871,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004748643452822699,
      "loss": 4.8497,
      "mean_token_accuracy": 0.23602927923202516,
      "num_tokens": 38181876.0,
      "step": 16660
    },
    {
      "entropy": 5.3271177291870115,
      "epoch": 1.600864553314121,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004748485968413036,
      "loss": 4.846,
      "mean_token_accuracy": 0.2314800962805748,
      "num_tokens": 38193680.0,
      "step": 16665
    },
    {
      "entropy": 5.248681354522705,
      "epoch": 1.601344860710855,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047483284376044634,
      "loss": 4.8145,
      "mean_token_accuracy": 0.23897212594747544,
      "num_tokens": 38204944.0,
      "step": 16670
    },
    {
      "entropy": 5.260244607925415,
      "epoch": 1.601825168107589,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004748170860400638,
      "loss": 4.7632,
      "mean_token_accuracy": 0.24003359228372573,
      "num_tokens": 38216381.0,
      "step": 16675
    },
    {
      "entropy": 5.314260721206665,
      "epoch": 1.602305475504323,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047480132368052185,
      "loss": 4.8063,
      "mean_token_accuracy": 0.23932712525129318,
      "num_tokens": 38227420.0,
      "step": 16680
    },
    {
      "entropy": 5.319372749328613,
      "epoch": 1.6027857829010568,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047478555668218643,
      "loss": 4.9028,
      "mean_token_accuracy": 0.23502586632966996,
      "num_tokens": 38237869.0,
      "step": 16685
    },
    {
      "entropy": 5.29118275642395,
      "epoch": 1.6032660902977907,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004747697850454237,
      "loss": 4.8715,
      "mean_token_accuracy": 0.229049950838089,
      "num_tokens": 38250362.0,
      "step": 16690
    },
    {
      "entropy": 5.418159294128418,
      "epoch": 1.6037463976945245,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004747540087705997,
      "loss": 4.9605,
      "mean_token_accuracy": 0.22362844944000243,
      "num_tokens": 38262887.0,
      "step": 16695
    },
    {
      "entropy": 5.339011240005493,
      "epoch": 1.6042267050912584,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004747382278580808,
      "loss": 4.812,
      "mean_token_accuracy": 0.23529215902090073,
      "num_tokens": 38273206.0,
      "step": 16700
    },
    {
      "entropy": 5.312827253341675,
      "epoch": 1.6047070124879923,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004747224423082333,
      "loss": 4.8215,
      "mean_token_accuracy": 0.23166944533586503,
      "num_tokens": 38283307.0,
      "step": 16705
    },
    {
      "entropy": 5.338718843460083,
      "epoch": 1.6051873198847262,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047470665212142384,
      "loss": 4.8971,
      "mean_token_accuracy": 0.22853039503097533,
      "num_tokens": 38293830.0,
      "step": 16710
    },
    {
      "entropy": 5.331242322921753,
      "epoch": 1.60566762728146,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047469085729801887,
      "loss": 4.8838,
      "mean_token_accuracy": 0.23334264904260635,
      "num_tokens": 38306523.0,
      "step": 16715
    },
    {
      "entropy": 5.311114978790283,
      "epoch": 1.606147934678194,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047467505783838515,
      "loss": 4.8534,
      "mean_token_accuracy": 0.2376910075545311,
      "num_tokens": 38318468.0,
      "step": 16720
    },
    {
      "entropy": 5.344900369644165,
      "epoch": 1.6066282420749278,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004746592537428895,
      "loss": 4.795,
      "mean_token_accuracy": 0.24126742482185365,
      "num_tokens": 38329852.0,
      "step": 16725
    },
    {
      "entropy": 5.218759155273437,
      "epoch": 1.6071085494716617,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047464344501189877,
      "loss": 4.752,
      "mean_token_accuracy": 0.24235767424106597,
      "num_tokens": 38340951.0,
      "step": 16730
    },
    {
      "entropy": 5.354107999801636,
      "epoch": 1.6075888568683958,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047462763164578015,
      "loss": 4.9398,
      "mean_token_accuracy": 0.2226387158036232,
      "num_tokens": 38351490.0,
      "step": 16735
    },
    {
      "entropy": 5.345208024978637,
      "epoch": 1.6080691642651297,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004746118136449007,
      "loss": 4.8151,
      "mean_token_accuracy": 0.23409664928913115,
      "num_tokens": 38362364.0,
      "step": 16740
    },
    {
      "entropy": 5.317148637771607,
      "epoch": 1.6085494716618636,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004745959910096276,
      "loss": 4.9135,
      "mean_token_accuracy": 0.23519372045993805,
      "num_tokens": 38374132.0,
      "step": 16745
    },
    {
      "entropy": 5.310082244873047,
      "epoch": 1.6090297790585975,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047458016374032837,
      "loss": 4.8703,
      "mean_token_accuracy": 0.23073885440826417,
      "num_tokens": 38384608.0,
      "step": 16750
    },
    {
      "entropy": 5.299181747436523,
      "epoch": 1.6095100864553316,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004745643318373703,
      "loss": 4.8878,
      "mean_token_accuracy": 0.23246051520109176,
      "num_tokens": 38396639.0,
      "step": 16755
    },
    {
      "entropy": 5.326464509963989,
      "epoch": 1.6099903938520654,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047454849530112106,
      "loss": 4.8844,
      "mean_token_accuracy": 0.22019701898097993,
      "num_tokens": 38407954.0,
      "step": 16760
    },
    {
      "entropy": 5.371581602096557,
      "epoch": 1.6104707012487993,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047453265413194826,
      "loss": 4.8529,
      "mean_token_accuracy": 0.2304906129837036,
      "num_tokens": 38418939.0,
      "step": 16765
    },
    {
      "entropy": 5.3418220520019535,
      "epoch": 1.6109510086455332,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047451680833021973,
      "loss": 4.8103,
      "mean_token_accuracy": 0.2345270186662674,
      "num_tokens": 38429717.0,
      "step": 16770
    },
    {
      "entropy": 5.273714637756347,
      "epoch": 1.611431316042267,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004745009578963034,
      "loss": 4.7943,
      "mean_token_accuracy": 0.24046590626239778,
      "num_tokens": 38441054.0,
      "step": 16775
    },
    {
      "entropy": 5.2340507984161375,
      "epoch": 1.611911623439001,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047448510283056716,
      "loss": 4.8146,
      "mean_token_accuracy": 0.23847273886203765,
      "num_tokens": 38451329.0,
      "step": 16780
    },
    {
      "entropy": 5.304728221893311,
      "epoch": 1.6123919308357348,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047446924313337925,
      "loss": 4.885,
      "mean_token_accuracy": 0.23136307448148727,
      "num_tokens": 38462118.0,
      "step": 16785
    },
    {
      "entropy": 5.3354720115661625,
      "epoch": 1.6128722382324687,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047445337880510773,
      "loss": 4.8432,
      "mean_token_accuracy": 0.23683312237262727,
      "num_tokens": 38472642.0,
      "step": 16790
    },
    {
      "entropy": 5.235729646682739,
      "epoch": 1.6133525456292026,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004744375098461211,
      "loss": 4.783,
      "mean_token_accuracy": 0.23568097203969957,
      "num_tokens": 38483866.0,
      "step": 16795
    },
    {
      "entropy": 5.253613233566284,
      "epoch": 1.6138328530259365,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004744216362567876,
      "loss": 4.754,
      "mean_token_accuracy": 0.24031510353088378,
      "num_tokens": 38494164.0,
      "step": 16800
    },
    {
      "entropy": 5.369624376296997,
      "epoch": 1.6143131604226704,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047440575803747595,
      "loss": 4.9678,
      "mean_token_accuracy": 0.2244415208697319,
      "num_tokens": 38505340.0,
      "step": 16805
    },
    {
      "entropy": 5.454342985153199,
      "epoch": 1.6147934678194045,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047438987518855463,
      "loss": 4.9775,
      "mean_token_accuracy": 0.2219114273786545,
      "num_tokens": 38516164.0,
      "step": 16810
    },
    {
      "entropy": 5.4084104061126705,
      "epoch": 1.6152737752161384,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004743739877103926,
      "loss": 4.8448,
      "mean_token_accuracy": 0.2289230152964592,
      "num_tokens": 38526545.0,
      "step": 16815
    },
    {
      "entropy": 5.279346132278443,
      "epoch": 1.6157540826128722,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004743580956033585,
      "loss": 4.8105,
      "mean_token_accuracy": 0.23393609672784804,
      "num_tokens": 38538102.0,
      "step": 16820
    },
    {
      "entropy": 5.3187871932983395,
      "epoch": 1.6162343900096061,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047434219886782135,
      "loss": 4.9193,
      "mean_token_accuracy": 0.22329402714967728,
      "num_tokens": 38550594.0,
      "step": 16825
    },
    {
      "entropy": 5.318387413024903,
      "epoch": 1.6167146974063402,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004743262975041504,
      "loss": 4.8021,
      "mean_token_accuracy": 0.23455502986907958,
      "num_tokens": 38561666.0,
      "step": 16830
    },
    {
      "entropy": 5.30687460899353,
      "epoch": 1.617195004803074,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004743103915127146,
      "loss": 4.7757,
      "mean_token_accuracy": 0.23611160069704057,
      "num_tokens": 38572923.0,
      "step": 16835
    },
    {
      "entropy": 5.318736219406128,
      "epoch": 1.617675312199808,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047429448089388336,
      "loss": 4.8196,
      "mean_token_accuracy": 0.23803658932447433,
      "num_tokens": 38584108.0,
      "step": 16840
    },
    {
      "entropy": 5.420387506484985,
      "epoch": 1.6181556195965419,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047427856564802605,
      "loss": 4.9472,
      "mean_token_accuracy": 0.22166336327791214,
      "num_tokens": 38594974.0,
      "step": 16845
    },
    {
      "entropy": 5.370044708251953,
      "epoch": 1.6186359269932757,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004742626457755122,
      "loss": 4.8935,
      "mean_token_accuracy": 0.23090327233076097,
      "num_tokens": 38606436.0,
      "step": 16850
    },
    {
      "entropy": 5.312457799911499,
      "epoch": 1.6191162343900096,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004742467212767114,
      "loss": 4.7959,
      "mean_token_accuracy": 0.23226692974567414,
      "num_tokens": 38618760.0,
      "step": 16855
    },
    {
      "entropy": 5.267887973785401,
      "epoch": 1.6195965417867435,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004742307921519933,
      "loss": 4.7595,
      "mean_token_accuracy": 0.24203593134880066,
      "num_tokens": 38629901.0,
      "step": 16860
    },
    {
      "entropy": 5.375607061386108,
      "epoch": 1.6200768491834774,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047421485840172794,
      "loss": 4.954,
      "mean_token_accuracy": 0.21832100450992584,
      "num_tokens": 38640798.0,
      "step": 16865
    },
    {
      "entropy": 5.377183485031128,
      "epoch": 1.6205571565802113,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004741989200262851,
      "loss": 4.8785,
      "mean_token_accuracy": 0.2285748153924942,
      "num_tokens": 38653048.0,
      "step": 16870
    },
    {
      "entropy": 5.373303318023682,
      "epoch": 1.6210374639769451,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004741829770260347,
      "loss": 4.9253,
      "mean_token_accuracy": 0.2282385289669037,
      "num_tokens": 38666219.0,
      "step": 16875
    },
    {
      "entropy": 5.310598468780517,
      "epoch": 1.621517771373679,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047416702940134714,
      "loss": 4.8035,
      "mean_token_accuracy": 0.24008783400058747,
      "num_tokens": 38678584.0,
      "step": 16880
    },
    {
      "entropy": 5.320067691802978,
      "epoch": 1.621998078770413,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047415107715259255,
      "loss": 4.8316,
      "mean_token_accuracy": 0.23007622212171555,
      "num_tokens": 38690164.0,
      "step": 16885
    },
    {
      "entropy": 5.341851806640625,
      "epoch": 1.622478386167147,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047413512028014125,
      "loss": 4.9128,
      "mean_token_accuracy": 0.23446644097566605,
      "num_tokens": 38702035.0,
      "step": 16890
    },
    {
      "entropy": 5.362625312805176,
      "epoch": 1.622958693563881,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004741191587843638,
      "loss": 4.9291,
      "mean_token_accuracy": 0.22950348258018494,
      "num_tokens": 38713802.0,
      "step": 16895
    },
    {
      "entropy": 5.314287233352661,
      "epoch": 1.6234390009606148,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004741031926656308,
      "loss": 4.8096,
      "mean_token_accuracy": 0.24604994356632232,
      "num_tokens": 38725682.0,
      "step": 16900
    },
    {
      "entropy": 5.296239995956421,
      "epoch": 1.6239193083573487,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004740872219243128,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23781765699386598,
      "num_tokens": 38737897.0,
      "step": 16905
    },
    {
      "entropy": 5.3836602687835695,
      "epoch": 1.6243996157540828,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004740712465607807,
      "loss": 4.9143,
      "mean_token_accuracy": 0.22907067835330963,
      "num_tokens": 38749446.0,
      "step": 16910
    },
    {
      "entropy": 5.351065635681152,
      "epoch": 1.6248799231508166,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004740552665754054,
      "loss": 4.8746,
      "mean_token_accuracy": 0.22515296190977097,
      "num_tokens": 38761406.0,
      "step": 16915
    },
    {
      "entropy": 5.366115427017212,
      "epoch": 1.6253602305475505,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047403928196855776,
      "loss": 4.8651,
      "mean_token_accuracy": 0.23380886167287826,
      "num_tokens": 38772574.0,
      "step": 16920
    },
    {
      "entropy": 5.311499404907226,
      "epoch": 1.6258405379442844,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047402329274060916,
      "loss": 4.8367,
      "mean_token_accuracy": 0.24524975568056107,
      "num_tokens": 38782932.0,
      "step": 16925
    },
    {
      "entropy": 5.337351369857788,
      "epoch": 1.6263208453410183,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004740072988919306,
      "loss": 4.9054,
      "mean_token_accuracy": 0.22993927597999572,
      "num_tokens": 38793799.0,
      "step": 16930
    },
    {
      "entropy": 5.257449579238892,
      "epoch": 1.6268011527377522,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004739913004228936,
      "loss": 4.733,
      "mean_token_accuracy": 0.23873628824949264,
      "num_tokens": 38804899.0,
      "step": 16935
    },
    {
      "entropy": 5.337667846679688,
      "epoch": 1.627281460134486,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004739752973338694,
      "loss": 4.9396,
      "mean_token_accuracy": 0.218874229490757,
      "num_tokens": 38816455.0,
      "step": 16940
    },
    {
      "entropy": 5.349135208129883,
      "epoch": 1.62776176753122,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047395928962522965,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2329275369644165,
      "num_tokens": 38827735.0,
      "step": 16945
    },
    {
      "entropy": 5.35344820022583,
      "epoch": 1.6282420749279538,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047394327729734595,
      "loss": 4.8912,
      "mean_token_accuracy": 0.22794464081525803,
      "num_tokens": 38838536.0,
      "step": 16950
    },
    {
      "entropy": 5.4143304347991945,
      "epoch": 1.6287223823246877,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004739272603505901,
      "loss": 4.9203,
      "mean_token_accuracy": 0.22855342477560042,
      "num_tokens": 38849577.0,
      "step": 16955
    },
    {
      "entropy": 5.3888733863830565,
      "epoch": 1.6292026897214216,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000473911238785334,
      "loss": 4.8788,
      "mean_token_accuracy": 0.22385758608579637,
      "num_tokens": 38861370.0,
      "step": 16960
    },
    {
      "entropy": 5.304583024978638,
      "epoch": 1.6296829971181557,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004738952126019496,
      "loss": 4.8319,
      "mean_token_accuracy": 0.22880436927080156,
      "num_tokens": 38872885.0,
      "step": 16965
    },
    {
      "entropy": 5.314086532592773,
      "epoch": 1.6301633045148896,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004738791818008089,
      "loss": 4.8884,
      "mean_token_accuracy": 0.2355169877409935,
      "num_tokens": 38885440.0,
      "step": 16970
    },
    {
      "entropy": 5.322174596786499,
      "epoch": 1.6306436119116234,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004738631463822841,
      "loss": 4.7923,
      "mean_token_accuracy": 0.23654236197471618,
      "num_tokens": 38897108.0,
      "step": 16975
    },
    {
      "entropy": 5.3021636486053465,
      "epoch": 1.6311239193083573,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047384710634674766,
      "loss": 4.8646,
      "mean_token_accuracy": 0.23019543886184693,
      "num_tokens": 38908020.0,
      "step": 16980
    },
    {
      "entropy": 5.296778917312622,
      "epoch": 1.6316042267050914,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047383106169457184,
      "loss": 4.8817,
      "mean_token_accuracy": 0.23354432433843614,
      "num_tokens": 38919558.0,
      "step": 16985
    },
    {
      "entropy": 5.300839567184449,
      "epoch": 1.6320845341018253,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004738150124261292,
      "loss": 4.8651,
      "mean_token_accuracy": 0.23240365087985992,
      "num_tokens": 38931008.0,
      "step": 16990
    },
    {
      "entropy": 5.247205924987793,
      "epoch": 1.6325648414985592,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047379895854179226,
      "loss": 4.7747,
      "mean_token_accuracy": 0.23725308775901793,
      "num_tokens": 38943474.0,
      "step": 16995
    },
    {
      "entropy": 5.357652091979981,
      "epoch": 1.633045148895293,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004737829000419338,
      "loss": 4.9239,
      "mean_token_accuracy": 0.22566596269607545,
      "num_tokens": 38954592.0,
      "step": 17000
    },
    {
      "entropy": 5.327875137329102,
      "epoch": 1.633525456292027,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047376683692692666,
      "loss": 4.8224,
      "mean_token_accuracy": 0.2367486521601677,
      "num_tokens": 38965627.0,
      "step": 17005
    },
    {
      "entropy": 5.233381032943726,
      "epoch": 1.6340057636887608,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004737507691971439,
      "loss": 4.7988,
      "mean_token_accuracy": 0.2399650976061821,
      "num_tokens": 38975731.0,
      "step": 17010
    },
    {
      "entropy": 5.312558698654175,
      "epoch": 1.6344860710854947,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047373469685295833,
      "loss": 4.8052,
      "mean_token_accuracy": 0.23812069594860077,
      "num_tokens": 38987101.0,
      "step": 17015
    },
    {
      "entropy": 5.34988169670105,
      "epoch": 1.6349663784822286,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047371861989474326,
      "loss": 4.8522,
      "mean_token_accuracy": 0.23584005683660508,
      "num_tokens": 38999584.0,
      "step": 17020
    },
    {
      "entropy": 5.268465614318847,
      "epoch": 1.6354466858789625,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004737025383228719,
      "loss": 4.8409,
      "mean_token_accuracy": 0.23688526898622514,
      "num_tokens": 39012025.0,
      "step": 17025
    },
    {
      "entropy": 5.216385555267334,
      "epoch": 1.6359269932756964,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047368645213771764,
      "loss": 4.7196,
      "mean_token_accuracy": 0.24150403290987016,
      "num_tokens": 39024102.0,
      "step": 17030
    },
    {
      "entropy": 5.335909080505371,
      "epoch": 1.6364073006724302,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000473670361339654,
      "loss": 4.8634,
      "mean_token_accuracy": 0.22700446993112564,
      "num_tokens": 39035295.0,
      "step": 17035
    },
    {
      "entropy": 5.292139101028442,
      "epoch": 1.6368876080691641,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004736542659290544,
      "loss": 4.7669,
      "mean_token_accuracy": 0.23769527673721313,
      "num_tokens": 39046493.0,
      "step": 17040
    },
    {
      "entropy": 5.310463953018188,
      "epoch": 1.6373679154658982,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004736381659062927,
      "loss": 4.8076,
      "mean_token_accuracy": 0.23666122257709504,
      "num_tokens": 39056746.0,
      "step": 17045
    },
    {
      "entropy": 5.328650236129761,
      "epoch": 1.637848222862632,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047362206127174255,
      "loss": 4.8589,
      "mean_token_accuracy": 0.2272334098815918,
      "num_tokens": 39068263.0,
      "step": 17050
    },
    {
      "entropy": 5.346367025375367,
      "epoch": 1.638328530259366,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047360595202577786,
      "loss": 4.8483,
      "mean_token_accuracy": 0.23044548332691192,
      "num_tokens": 39080666.0,
      "step": 17055
    },
    {
      "entropy": 5.2979566097259525,
      "epoch": 1.6388088376560999,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047358983816877284,
      "loss": 4.849,
      "mean_token_accuracy": 0.22956279814243316,
      "num_tokens": 39091851.0,
      "step": 17060
    },
    {
      "entropy": 5.349362230300903,
      "epoch": 1.639289145052834,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004735737197011015,
      "loss": 4.8466,
      "mean_token_accuracy": 0.23494686633348466,
      "num_tokens": 39102893.0,
      "step": 17065
    },
    {
      "entropy": 5.2646032810211185,
      "epoch": 1.6397694524495678,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047355759662313793,
      "loss": 4.7679,
      "mean_token_accuracy": 0.23781567364931105,
      "num_tokens": 39113582.0,
      "step": 17070
    },
    {
      "entropy": 5.385623025894165,
      "epoch": 1.6402497598463017,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004735414689352566,
      "loss": 4.9726,
      "mean_token_accuracy": 0.22290729433298112,
      "num_tokens": 39124616.0,
      "step": 17075
    },
    {
      "entropy": 5.253909397125244,
      "epoch": 1.6407300672430356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004735253366378318,
      "loss": 4.7646,
      "mean_token_accuracy": 0.2430978611111641,
      "num_tokens": 39135537.0,
      "step": 17080
    },
    {
      "entropy": 5.304638910293579,
      "epoch": 1.6412103746397695,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004735091997312383,
      "loss": 4.8418,
      "mean_token_accuracy": 0.23836640864610673,
      "num_tokens": 39146363.0,
      "step": 17085
    },
    {
      "entropy": 5.3078656673431395,
      "epoch": 1.6416906820365034,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047349305821585067,
      "loss": 4.8217,
      "mean_token_accuracy": 0.23962389081716537,
      "num_tokens": 39157319.0,
      "step": 17090
    },
    {
      "entropy": 5.3185149192810055,
      "epoch": 1.6421709894332372,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004734769120920435,
      "loss": 4.8273,
      "mean_token_accuracy": 0.23501784205436707,
      "num_tokens": 39169352.0,
      "step": 17095
    },
    {
      "entropy": 5.349051189422608,
      "epoch": 1.6426512968299711,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004734607613601919,
      "loss": 4.8132,
      "mean_token_accuracy": 0.23369207233190536,
      "num_tokens": 39180989.0,
      "step": 17100
    },
    {
      "entropy": 5.338135147094727,
      "epoch": 1.643131604226705,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047344460602067077,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2323719248175621,
      "num_tokens": 39192309.0,
      "step": 17105
    },
    {
      "entropy": 5.261227035522461,
      "epoch": 1.643611911623439,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000473428446073855,
      "loss": 4.8195,
      "mean_token_accuracy": 0.22920311838388444,
      "num_tokens": 39203586.0,
      "step": 17110
    },
    {
      "entropy": 5.296448516845703,
      "epoch": 1.6440922190201728,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047341228152012003,
      "loss": 4.8385,
      "mean_token_accuracy": 0.22713108956813813,
      "num_tokens": 39216248.0,
      "step": 17115
    },
    {
      "entropy": 5.355902147293091,
      "epoch": 1.6445725264169067,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000473396112359841,
      "loss": 4.8139,
      "mean_token_accuracy": 0.2326844647526741,
      "num_tokens": 39227244.0,
      "step": 17120
    },
    {
      "entropy": 5.280712509155274,
      "epoch": 1.6450528338136408,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047337993859339334,
      "loss": 4.7594,
      "mean_token_accuracy": 0.24350704699754716,
      "num_tokens": 39238769.0,
      "step": 17125
    },
    {
      "entropy": 5.296138286590576,
      "epoch": 1.6455331412103746,
      "grad_norm": 1.0,
      "learning_rate": 0.00047336376022115255,
      "loss": 4.8536,
      "mean_token_accuracy": 0.23425736874341965,
      "num_tokens": 39251424.0,
      "step": 17130
    },
    {
      "entropy": 5.271192979812622,
      "epoch": 1.6460134486071085,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047334757724349437,
      "loss": 4.8148,
      "mean_token_accuracy": 0.23912425488233566,
      "num_tokens": 39262451.0,
      "step": 17135
    },
    {
      "entropy": 5.294335508346558,
      "epoch": 1.6464937560038426,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004733313896607943,
      "loss": 4.8191,
      "mean_token_accuracy": 0.23237797021865844,
      "num_tokens": 39275129.0,
      "step": 17140
    },
    {
      "entropy": 5.4056634426116945,
      "epoch": 1.6469740634005765,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004733151974734284,
      "loss": 4.9424,
      "mean_token_accuracy": 0.22798584252595902,
      "num_tokens": 39287399.0,
      "step": 17145
    },
    {
      "entropy": 5.327386903762817,
      "epoch": 1.6474543707973104,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047329900068177245,
      "loss": 4.794,
      "mean_token_accuracy": 0.2394887313246727,
      "num_tokens": 39297755.0,
      "step": 17150
    },
    {
      "entropy": 5.35604681968689,
      "epoch": 1.6479346781940443,
      "grad_norm": 1.0,
      "learning_rate": 0.00047328279928620244,
      "loss": 4.9155,
      "mean_token_accuracy": 0.22415244281291963,
      "num_tokens": 39311012.0,
      "step": 17155
    },
    {
      "entropy": 5.362229919433593,
      "epoch": 1.6484149855907781,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004732665932870947,
      "loss": 4.8463,
      "mean_token_accuracy": 0.23218199163675307,
      "num_tokens": 39321797.0,
      "step": 17160
    },
    {
      "entropy": 5.3225359439849855,
      "epoch": 1.648895292987512,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047325038268482544,
      "loss": 4.793,
      "mean_token_accuracy": 0.24264750480651856,
      "num_tokens": 39333598.0,
      "step": 17165
    },
    {
      "entropy": 5.364239978790283,
      "epoch": 1.649375600384246,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004732341674797709,
      "loss": 4.879,
      "mean_token_accuracy": 0.23115923553705214,
      "num_tokens": 39345578.0,
      "step": 17170
    },
    {
      "entropy": 5.244977617263794,
      "epoch": 1.6498559077809798,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047321794767230766,
      "loss": 4.748,
      "mean_token_accuracy": 0.23771919161081315,
      "num_tokens": 39356744.0,
      "step": 17175
    },
    {
      "entropy": 5.323004531860351,
      "epoch": 1.6503362151777137,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047320172326281224,
      "loss": 4.8781,
      "mean_token_accuracy": 0.23042381554841995,
      "num_tokens": 39368138.0,
      "step": 17180
    },
    {
      "entropy": 5.263118648529053,
      "epoch": 1.6508165225744476,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047318549425166134,
      "loss": 4.7306,
      "mean_token_accuracy": 0.23570469319820403,
      "num_tokens": 39380117.0,
      "step": 17185
    },
    {
      "entropy": 5.2993724822998045,
      "epoch": 1.6512968299711814,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004731692606392318,
      "loss": 4.8044,
      "mean_token_accuracy": 0.2330299973487854,
      "num_tokens": 39390913.0,
      "step": 17190
    },
    {
      "entropy": 5.3888942241668705,
      "epoch": 1.6517771373679153,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004731530224259004,
      "loss": 4.939,
      "mean_token_accuracy": 0.22200429290533066,
      "num_tokens": 39401001.0,
      "step": 17195
    },
    {
      "entropy": 5.320006656646728,
      "epoch": 1.6522574447646494,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004731367796120442,
      "loss": 4.8197,
      "mean_token_accuracy": 0.23030412942171097,
      "num_tokens": 39411982.0,
      "step": 17200
    },
    {
      "entropy": 5.372227478027344,
      "epoch": 1.6527377521613833,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004731205321980404,
      "loss": 4.8566,
      "mean_token_accuracy": 0.23348181694746017,
      "num_tokens": 39423363.0,
      "step": 17205
    },
    {
      "entropy": 5.329045200347901,
      "epoch": 1.6532180595581172,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047310428018426616,
      "loss": 4.901,
      "mean_token_accuracy": 0.22835494577884674,
      "num_tokens": 39434002.0,
      "step": 17210
    },
    {
      "entropy": 5.296520233154297,
      "epoch": 1.653698366954851,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004730880235710987,
      "loss": 4.7851,
      "mean_token_accuracy": 0.2382122188806534,
      "num_tokens": 39444629.0,
      "step": 17215
    },
    {
      "entropy": 5.3008300304412845,
      "epoch": 1.6541786743515852,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004730717623589155,
      "loss": 4.7554,
      "mean_token_accuracy": 0.24046491980552673,
      "num_tokens": 39454803.0,
      "step": 17220
    },
    {
      "entropy": 5.304701519012451,
      "epoch": 1.654658981748319,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004730554965480942,
      "loss": 4.9166,
      "mean_token_accuracy": 0.22917634397745132,
      "num_tokens": 39466931.0,
      "step": 17225
    },
    {
      "entropy": 5.329890298843384,
      "epoch": 1.655139289145053,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004730392261390124,
      "loss": 4.8507,
      "mean_token_accuracy": 0.2320278599858284,
      "num_tokens": 39478127.0,
      "step": 17230
    },
    {
      "entropy": 5.346811962127686,
      "epoch": 1.6556195965417868,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004730229511320478,
      "loss": 4.8786,
      "mean_token_accuracy": 0.22637103348970414,
      "num_tokens": 39488188.0,
      "step": 17235
    },
    {
      "entropy": 5.337345886230469,
      "epoch": 1.6560999039385207,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047300667152757827,
      "loss": 4.8211,
      "mean_token_accuracy": 0.23502825796604157,
      "num_tokens": 39500035.0,
      "step": 17240
    },
    {
      "entropy": 5.360410213470459,
      "epoch": 1.6565802113352546,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047299038732598184,
      "loss": 4.901,
      "mean_token_accuracy": 0.22539904415607454,
      "num_tokens": 39510922.0,
      "step": 17245
    },
    {
      "entropy": 5.322126007080078,
      "epoch": 1.6570605187319885,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047297409852763644,
      "loss": 4.8932,
      "mean_token_accuracy": 0.23233979046344758,
      "num_tokens": 39521319.0,
      "step": 17250
    },
    {
      "entropy": 5.3503893375396725,
      "epoch": 1.6575408261287223,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004729578051329204,
      "loss": 4.8664,
      "mean_token_accuracy": 0.23262367993593216,
      "num_tokens": 39532408.0,
      "step": 17255
    },
    {
      "entropy": 5.354845666885376,
      "epoch": 1.6580211335254562,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047294150714221185,
      "loss": 4.9029,
      "mean_token_accuracy": 0.2282740890979767,
      "num_tokens": 39543323.0,
      "step": 17260
    },
    {
      "entropy": 5.359320878982544,
      "epoch": 1.65850144092219,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004729252045558894,
      "loss": 4.8628,
      "mean_token_accuracy": 0.23087600469589234,
      "num_tokens": 39554133.0,
      "step": 17265
    },
    {
      "entropy": 5.260821104049683,
      "epoch": 1.658981748318924,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047290889737433133,
      "loss": 4.7449,
      "mean_token_accuracy": 0.23878293335437775,
      "num_tokens": 39565990.0,
      "step": 17270
    },
    {
      "entropy": 5.336102485656738,
      "epoch": 1.6594620557156579,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047289258559791633,
      "loss": 4.8478,
      "mean_token_accuracy": 0.2368147000670433,
      "num_tokens": 39577591.0,
      "step": 17275
    },
    {
      "entropy": 5.31400146484375,
      "epoch": 1.659942363112392,
      "grad_norm": 1.0,
      "learning_rate": 0.00047287626922702317,
      "loss": 4.8139,
      "mean_token_accuracy": 0.23309870064258575,
      "num_tokens": 39589922.0,
      "step": 17280
    },
    {
      "entropy": 5.297982501983642,
      "epoch": 1.6604226705091258,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047285994826203054,
      "loss": 4.8479,
      "mean_token_accuracy": 0.24006879031658174,
      "num_tokens": 39602183.0,
      "step": 17285
    },
    {
      "entropy": 5.339755535125732,
      "epoch": 1.6609029779058597,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004728436227033175,
      "loss": 4.8449,
      "mean_token_accuracy": 0.235623861849308,
      "num_tokens": 39612922.0,
      "step": 17290
    },
    {
      "entropy": 5.3567280769348145,
      "epoch": 1.6613832853025938,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047282729255126294,
      "loss": 4.8501,
      "mean_token_accuracy": 0.23482128977775574,
      "num_tokens": 39624475.0,
      "step": 17295
    },
    {
      "entropy": 5.228869915008545,
      "epoch": 1.6618635926993277,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004728109578062461,
      "loss": 4.6999,
      "mean_token_accuracy": 0.24482650458812713,
      "num_tokens": 39635230.0,
      "step": 17300
    },
    {
      "entropy": 5.251270961761475,
      "epoch": 1.6623439000960616,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047279461846864626,
      "loss": 4.8279,
      "mean_token_accuracy": 0.23365410715341567,
      "num_tokens": 39646788.0,
      "step": 17305
    },
    {
      "entropy": 5.303299474716186,
      "epoch": 1.6628242074927955,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047277827453884265,
      "loss": 4.7973,
      "mean_token_accuracy": 0.23407886177301407,
      "num_tokens": 39657531.0,
      "step": 17310
    },
    {
      "entropy": 5.416685247421265,
      "epoch": 1.6633045148895294,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047276192601721477,
      "loss": 4.9691,
      "mean_token_accuracy": 0.22036943584680557,
      "num_tokens": 39669008.0,
      "step": 17315
    },
    {
      "entropy": 5.3273687839508055,
      "epoch": 1.6637848222862632,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004727455729041422,
      "loss": 4.7977,
      "mean_token_accuracy": 0.2378709211945534,
      "num_tokens": 39680772.0,
      "step": 17320
    },
    {
      "entropy": 5.381401634216308,
      "epoch": 1.6642651296829971,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047272921520000465,
      "loss": 4.9644,
      "mean_token_accuracy": 0.22087036371231078,
      "num_tokens": 39694107.0,
      "step": 17325
    },
    {
      "entropy": 5.327172756195068,
      "epoch": 1.664745437079731,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004727128529051819,
      "loss": 4.8135,
      "mean_token_accuracy": 0.22996855527162552,
      "num_tokens": 39705440.0,
      "step": 17330
    },
    {
      "entropy": 5.3319000720977785,
      "epoch": 1.6652257444764649,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004726964860200537,
      "loss": 4.8266,
      "mean_token_accuracy": 0.2367278128862381,
      "num_tokens": 39716246.0,
      "step": 17335
    },
    {
      "entropy": 5.316368341445923,
      "epoch": 1.6657060518731988,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004726801145450002,
      "loss": 4.8394,
      "mean_token_accuracy": 0.229177725315094,
      "num_tokens": 39727233.0,
      "step": 17340
    },
    {
      "entropy": 5.302699613571167,
      "epoch": 1.6661863592699326,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004726637384804014,
      "loss": 4.792,
      "mean_token_accuracy": 0.232194122672081,
      "num_tokens": 39738929.0,
      "step": 17345
    },
    {
      "entropy": 5.294046688079834,
      "epoch": 1.6666666666666665,
      "grad_norm": 1.0,
      "learning_rate": 0.0004726473578266375,
      "loss": 4.839,
      "mean_token_accuracy": 0.23290661424398423,
      "num_tokens": 39750714.0,
      "step": 17350
    },
    {
      "entropy": 5.307856464385987,
      "epoch": 1.6671469740634006,
      "grad_norm": 1.125,
      "learning_rate": 0.00047263097258408893,
      "loss": 4.8748,
      "mean_token_accuracy": 0.23050604313611983,
      "num_tokens": 39762810.0,
      "step": 17355
    },
    {
      "entropy": 5.404884386062622,
      "epoch": 1.6676272814601345,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004726145827531359,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22739558964967727,
      "num_tokens": 39774611.0,
      "step": 17360
    },
    {
      "entropy": 5.28033938407898,
      "epoch": 1.6681075888568684,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047259818833415916,
      "loss": 4.7488,
      "mean_token_accuracy": 0.24294953793287277,
      "num_tokens": 39784900.0,
      "step": 17365
    },
    {
      "entropy": 5.362227487564087,
      "epoch": 1.6685878962536023,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047258178932753917,
      "loss": 4.8696,
      "mean_token_accuracy": 0.23685087114572526,
      "num_tokens": 39794805.0,
      "step": 17370
    },
    {
      "entropy": 5.359183597564697,
      "epoch": 1.6690682036503364,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047256538573365675,
      "loss": 4.8628,
      "mean_token_accuracy": 0.22694920152425765,
      "num_tokens": 39806444.0,
      "step": 17375
    },
    {
      "entropy": 5.480212688446045,
      "epoch": 1.6695485110470702,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004725489775528928,
      "loss": 5.0167,
      "mean_token_accuracy": 0.21847590804100037,
      "num_tokens": 39817724.0,
      "step": 17380
    },
    {
      "entropy": 5.317599487304688,
      "epoch": 1.6700288184438041,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047253256478562805,
      "loss": 4.8135,
      "mean_token_accuracy": 0.23028451055288315,
      "num_tokens": 39828944.0,
      "step": 17385
    },
    {
      "entropy": 5.271615314483642,
      "epoch": 1.670509125840538,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047251614743224374,
      "loss": 4.832,
      "mean_token_accuracy": 0.23533270061016082,
      "num_tokens": 39839978.0,
      "step": 17390
    },
    {
      "entropy": 5.293868207931519,
      "epoch": 1.670989433237272,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047249972549312107,
      "loss": 4.8337,
      "mean_token_accuracy": 0.23712214529514314,
      "num_tokens": 39850787.0,
      "step": 17395
    },
    {
      "entropy": 5.381713819503784,
      "epoch": 1.6714697406340058,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004724832989686411,
      "loss": 4.8972,
      "mean_token_accuracy": 0.23155915588140488,
      "num_tokens": 39862245.0,
      "step": 17400
    },
    {
      "entropy": 5.30219669342041,
      "epoch": 1.6719500480307397,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047246686785918545,
      "loss": 4.8206,
      "mean_token_accuracy": 0.2375132530927658,
      "num_tokens": 39872295.0,
      "step": 17405
    },
    {
      "entropy": 5.284198808670044,
      "epoch": 1.6724303554274735,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047245043216513546,
      "loss": 4.7502,
      "mean_token_accuracy": 0.23475297838449477,
      "num_tokens": 39882839.0,
      "step": 17410
    },
    {
      "entropy": 5.234178686141968,
      "epoch": 1.6729106628242074,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004724339918868727,
      "loss": 4.7601,
      "mean_token_accuracy": 0.2366567462682724,
      "num_tokens": 39894767.0,
      "step": 17415
    },
    {
      "entropy": 5.332140159606934,
      "epoch": 1.6733909702209413,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000472417547024779,
      "loss": 4.9134,
      "mean_token_accuracy": 0.22562889158725738,
      "num_tokens": 39907492.0,
      "step": 17420
    },
    {
      "entropy": 5.238862037658691,
      "epoch": 1.6738712776176752,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047240109757923593,
      "loss": 4.6921,
      "mean_token_accuracy": 0.24220912158489227,
      "num_tokens": 39919005.0,
      "step": 17425
    },
    {
      "entropy": 5.271146965026856,
      "epoch": 1.674351585014409,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004723846435506256,
      "loss": 4.8148,
      "mean_token_accuracy": 0.23619162142276764,
      "num_tokens": 39930417.0,
      "step": 17430
    },
    {
      "entropy": 5.268793439865112,
      "epoch": 1.6748318924111432,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047236818493932994,
      "loss": 4.8468,
      "mean_token_accuracy": 0.23346664160490035,
      "num_tokens": 39941873.0,
      "step": 17435
    },
    {
      "entropy": 5.34087986946106,
      "epoch": 1.675312199807877,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004723517217457311,
      "loss": 4.8623,
      "mean_token_accuracy": 0.2266439214348793,
      "num_tokens": 39953280.0,
      "step": 17440
    },
    {
      "entropy": 5.399102449417114,
      "epoch": 1.675792507204611,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004723352539702113,
      "loss": 4.9244,
      "mean_token_accuracy": 0.22922951132059097,
      "num_tokens": 39964898.0,
      "step": 17445
    },
    {
      "entropy": 5.314057159423828,
      "epoch": 1.6762728146013448,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004723187816131529,
      "loss": 4.7841,
      "mean_token_accuracy": 0.24029537439346313,
      "num_tokens": 39976945.0,
      "step": 17450
    },
    {
      "entropy": 5.347735595703125,
      "epoch": 1.676753121998079,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004723023046749383,
      "loss": 4.8985,
      "mean_token_accuracy": 0.23211033046245574,
      "num_tokens": 39987760.0,
      "step": 17455
    },
    {
      "entropy": 5.261370372772217,
      "epoch": 1.6772334293948128,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047228582315595,
      "loss": 4.7043,
      "mean_token_accuracy": 0.24647301584482192,
      "num_tokens": 39998171.0,
      "step": 17460
    },
    {
      "entropy": 5.302586078643799,
      "epoch": 1.6777137367915467,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004722693370565708,
      "loss": 4.8203,
      "mean_token_accuracy": 0.2327909380197525,
      "num_tokens": 40010124.0,
      "step": 17465
    },
    {
      "entropy": 5.266671419143677,
      "epoch": 1.6781940441882806,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047225284637718323,
      "loss": 4.8407,
      "mean_token_accuracy": 0.23009913563728332,
      "num_tokens": 40022599.0,
      "step": 17470
    },
    {
      "entropy": 5.315566492080689,
      "epoch": 1.6786743515850144,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004722363511181703,
      "loss": 4.7901,
      "mean_token_accuracy": 0.23841935098171235,
      "num_tokens": 40033738.0,
      "step": 17475
    },
    {
      "entropy": 5.343968296051026,
      "epoch": 1.6791546589817483,
      "grad_norm": 1.078125,
      "learning_rate": 0.000472219851279915,
      "loss": 4.84,
      "mean_token_accuracy": 0.2377188980579376,
      "num_tokens": 40044148.0,
      "step": 17480
    },
    {
      "entropy": 5.384026384353637,
      "epoch": 1.6796349663784822,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004722033468628004,
      "loss": 4.8386,
      "mean_token_accuracy": 0.23580623120069505,
      "num_tokens": 40055430.0,
      "step": 17485
    },
    {
      "entropy": 5.295413303375244,
      "epoch": 1.680115273775216,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004721868378672098,
      "loss": 4.841,
      "mean_token_accuracy": 0.23453572243452073,
      "num_tokens": 40066147.0,
      "step": 17490
    },
    {
      "entropy": 5.307233667373657,
      "epoch": 1.68059558117195,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004721703242935261,
      "loss": 4.8076,
      "mean_token_accuracy": 0.2318134590983391,
      "num_tokens": 40077567.0,
      "step": 17495
    },
    {
      "entropy": 5.381526327133178,
      "epoch": 1.6810758885686838,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004721538061421331,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2322109267115593,
      "num_tokens": 40089544.0,
      "step": 17500
    },
    {
      "entropy": 5.255374765396118,
      "epoch": 1.6815561959654177,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047213728341341407,
      "loss": 4.7812,
      "mean_token_accuracy": 0.23968904316425324,
      "num_tokens": 40100557.0,
      "step": 17505
    },
    {
      "entropy": 5.299778270721435,
      "epoch": 1.6820365033621518,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004721207561077527,
      "loss": 4.8369,
      "mean_token_accuracy": 0.23468495011329651,
      "num_tokens": 40112052.0,
      "step": 17510
    },
    {
      "entropy": 5.507605648040771,
      "epoch": 1.6825168107588857,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004721042242255327,
      "loss": 5.0174,
      "mean_token_accuracy": 0.2219519644975662,
      "num_tokens": 40123830.0,
      "step": 17515
    },
    {
      "entropy": 5.366515684127807,
      "epoch": 1.6829971181556196,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047208768776713805,
      "loss": 4.901,
      "mean_token_accuracy": 0.23318385630846022,
      "num_tokens": 40134222.0,
      "step": 17520
    },
    {
      "entropy": 5.290441846847534,
      "epoch": 1.6834774255523535,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004720711467329523,
      "loss": 4.8713,
      "mean_token_accuracy": 0.23073187619447708,
      "num_tokens": 40145552.0,
      "step": 17525
    },
    {
      "entropy": 5.288965368270874,
      "epoch": 1.6839577329490876,
      "grad_norm": 1.0,
      "learning_rate": 0.0004720546011233599,
      "loss": 4.7278,
      "mean_token_accuracy": 0.23642816096544267,
      "num_tokens": 40156615.0,
      "step": 17530
    },
    {
      "entropy": 5.323382043838501,
      "epoch": 1.6844380403458215,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004720380509387446,
      "loss": 4.8437,
      "mean_token_accuracy": 0.23848306089639665,
      "num_tokens": 40168825.0,
      "step": 17535
    },
    {
      "entropy": 5.341952896118164,
      "epoch": 1.6849183477425553,
      "grad_norm": 1.015625,
      "learning_rate": 0.000472021496179491,
      "loss": 4.8323,
      "mean_token_accuracy": 0.23419066816568374,
      "num_tokens": 40179523.0,
      "step": 17540
    },
    {
      "entropy": 5.2746118068695065,
      "epoch": 1.6853986551392892,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047200493684598316,
      "loss": 4.7869,
      "mean_token_accuracy": 0.23777954429388046,
      "num_tokens": 40191362.0,
      "step": 17545
    },
    {
      "entropy": 5.305330371856689,
      "epoch": 1.685878962536023,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047198837293860573,
      "loss": 4.8816,
      "mean_token_accuracy": 0.23072271198034286,
      "num_tokens": 40202274.0,
      "step": 17550
    },
    {
      "entropy": 5.358941221237183,
      "epoch": 1.686359269932757,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004719718044577432,
      "loss": 4.8917,
      "mean_token_accuracy": 0.2273821637034416,
      "num_tokens": 40213907.0,
      "step": 17555
    },
    {
      "entropy": 5.291792869567871,
      "epoch": 1.6868395773294909,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047195523140378034,
      "loss": 4.8476,
      "mean_token_accuracy": 0.2324736773967743,
      "num_tokens": 40225300.0,
      "step": 17560
    },
    {
      "entropy": 5.238188886642456,
      "epoch": 1.6873198847262247,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047193865377710177,
      "loss": 4.7429,
      "mean_token_accuracy": 0.24181754142045975,
      "num_tokens": 40236197.0,
      "step": 17565
    },
    {
      "entropy": 5.406441354751587,
      "epoch": 1.6878001921229586,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047192207157809246,
      "loss": 4.9477,
      "mean_token_accuracy": 0.22480100393295288,
      "num_tokens": 40247887.0,
      "step": 17570
    },
    {
      "entropy": 5.400984811782837,
      "epoch": 1.6882804995196925,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047190548480713736,
      "loss": 4.7988,
      "mean_token_accuracy": 0.23655859827995301,
      "num_tokens": 40258262.0,
      "step": 17575
    },
    {
      "entropy": 5.3512553691864015,
      "epoch": 1.6887608069164264,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047188889346462163,
      "loss": 4.8603,
      "mean_token_accuracy": 0.23276210576295853,
      "num_tokens": 40268735.0,
      "step": 17580
    },
    {
      "entropy": 5.315930652618408,
      "epoch": 1.6892411143131603,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047187229755093037,
      "loss": 4.9384,
      "mean_token_accuracy": 0.23035197407007219,
      "num_tokens": 40279905.0,
      "step": 17585
    },
    {
      "entropy": 5.3425459384918215,
      "epoch": 1.6897214217098944,
      "grad_norm": 1.03125,
      "learning_rate": 0.000471855697066449,
      "loss": 4.8416,
      "mean_token_accuracy": 0.23243938982486725,
      "num_tokens": 40290580.0,
      "step": 17590
    },
    {
      "entropy": 5.318258476257324,
      "epoch": 1.6902017291066282,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047183909201156297,
      "loss": 4.8818,
      "mean_token_accuracy": 0.22819837927818298,
      "num_tokens": 40302472.0,
      "step": 17595
    },
    {
      "entropy": 5.398356151580811,
      "epoch": 1.6906820365033621,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004718224823866576,
      "loss": 4.8924,
      "mean_token_accuracy": 0.22608058005571366,
      "num_tokens": 40314238.0,
      "step": 17600
    },
    {
      "entropy": 5.329414749145508,
      "epoch": 1.691162343900096,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004718058681921186,
      "loss": 4.828,
      "mean_token_accuracy": 0.23109890520572662,
      "num_tokens": 40326147.0,
      "step": 17605
    },
    {
      "entropy": 5.306858968734741,
      "epoch": 1.6916426512968301,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047178924942833185,
      "loss": 4.8128,
      "mean_token_accuracy": 0.23422625362873079,
      "num_tokens": 40338210.0,
      "step": 17610
    },
    {
      "entropy": 5.289654302597046,
      "epoch": 1.692122958693564,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004717726260956831,
      "loss": 4.772,
      "mean_token_accuracy": 0.2395482122898102,
      "num_tokens": 40349293.0,
      "step": 17615
    },
    {
      "entropy": 5.286178064346314,
      "epoch": 1.6926032660902979,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004717559981945581,
      "loss": 4.804,
      "mean_token_accuracy": 0.23183785229921341,
      "num_tokens": 40360916.0,
      "step": 17620
    },
    {
      "entropy": 5.288455295562744,
      "epoch": 1.6930835734870318,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004717393657253432,
      "loss": 4.8762,
      "mean_token_accuracy": 0.22808484882116317,
      "num_tokens": 40373525.0,
      "step": 17625
    },
    {
      "entropy": 5.362302684783936,
      "epoch": 1.6935638808837656,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004717227286884243,
      "loss": 4.8746,
      "mean_token_accuracy": 0.23195552676916123,
      "num_tokens": 40385286.0,
      "step": 17630
    },
    {
      "entropy": 5.313257360458374,
      "epoch": 1.6940441882804995,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004717060870841879,
      "loss": 4.7639,
      "mean_token_accuracy": 0.23926417231559755,
      "num_tokens": 40396371.0,
      "step": 17635
    },
    {
      "entropy": 5.294438171386719,
      "epoch": 1.6945244956772334,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004716894409130202,
      "loss": 4.8274,
      "mean_token_accuracy": 0.23649212419986726,
      "num_tokens": 40407304.0,
      "step": 17640
    },
    {
      "entropy": 5.338647890090942,
      "epoch": 1.6950048030739673,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004716727901753078,
      "loss": 4.7911,
      "mean_token_accuracy": 0.23941706717014313,
      "num_tokens": 40418384.0,
      "step": 17645
    },
    {
      "entropy": 5.364147329330445,
      "epoch": 1.6954851104707012,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004716561348714371,
      "loss": 4.8969,
      "mean_token_accuracy": 0.2316227525472641,
      "num_tokens": 40430603.0,
      "step": 17650
    },
    {
      "entropy": 5.288480949401856,
      "epoch": 1.695965417867435,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047163947500179494,
      "loss": 4.7888,
      "mean_token_accuracy": 0.23532818704843522,
      "num_tokens": 40442597.0,
      "step": 17655
    },
    {
      "entropy": 5.232913398742676,
      "epoch": 1.696445725264169,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004716228105667681,
      "loss": 4.7224,
      "mean_token_accuracy": 0.24629324972629546,
      "num_tokens": 40454078.0,
      "step": 17660
    },
    {
      "entropy": 5.276931095123291,
      "epoch": 1.696926032660903,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004716061415667435,
      "loss": 4.7848,
      "mean_token_accuracy": 0.24215866178274154,
      "num_tokens": 40465561.0,
      "step": 17665
    },
    {
      "entropy": 5.340037202835083,
      "epoch": 1.697406340057637,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000471589468002108,
      "loss": 4.8895,
      "mean_token_accuracy": 0.2328166753053665,
      "num_tokens": 40476883.0,
      "step": 17670
    },
    {
      "entropy": 5.348902750015259,
      "epoch": 1.6978866474543708,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004715727898732488,
      "loss": 4.8399,
      "mean_token_accuracy": 0.23621284514665603,
      "num_tokens": 40488139.0,
      "step": 17675
    },
    {
      "entropy": 5.344909238815307,
      "epoch": 1.6983669548511047,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047155610718055315,
      "loss": 4.8346,
      "mean_token_accuracy": 0.23081368952989578,
      "num_tokens": 40499367.0,
      "step": 17680
    },
    {
      "entropy": 5.254923534393311,
      "epoch": 1.6988472622478388,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047153941992440833,
      "loss": 4.792,
      "mean_token_accuracy": 0.22740429043769836,
      "num_tokens": 40510628.0,
      "step": 17685
    },
    {
      "entropy": 5.261801528930664,
      "epoch": 1.6993275696445727,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004715227281052018,
      "loss": 4.7927,
      "mean_token_accuracy": 0.23776306509971618,
      "num_tokens": 40522680.0,
      "step": 17690
    },
    {
      "entropy": 5.28656792640686,
      "epoch": 1.6998078770413065,
      "grad_norm": 1.0,
      "learning_rate": 0.0004715060317233211,
      "loss": 4.735,
      "mean_token_accuracy": 0.24563343822956085,
      "num_tokens": 40533139.0,
      "step": 17695
    },
    {
      "entropy": 5.343411445617676,
      "epoch": 1.7002881844380404,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004714893307791538,
      "loss": 4.8854,
      "mean_token_accuracy": 0.23631310164928437,
      "num_tokens": 40544578.0,
      "step": 17700
    },
    {
      "entropy": 5.3925618648529055,
      "epoch": 1.7007684918347743,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047147262527308766,
      "loss": 4.8246,
      "mean_token_accuracy": 0.23148826509714127,
      "num_tokens": 40555667.0,
      "step": 17705
    },
    {
      "entropy": 5.2753969669342045,
      "epoch": 1.7012487992315082,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004714559152055106,
      "loss": 4.8635,
      "mean_token_accuracy": 0.2393338680267334,
      "num_tokens": 40567123.0,
      "step": 17710
    },
    {
      "entropy": 5.28863525390625,
      "epoch": 1.701729106628242,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004714392005768106,
      "loss": 4.8402,
      "mean_token_accuracy": 0.22861011177301407,
      "num_tokens": 40579692.0,
      "step": 17715
    },
    {
      "entropy": 5.306768321990967,
      "epoch": 1.702209414024976,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004714224813873756,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23919534981250762,
      "num_tokens": 40590989.0,
      "step": 17720
    },
    {
      "entropy": 5.301641511917114,
      "epoch": 1.7026897214217098,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047140575763759393,
      "loss": 4.8268,
      "mean_token_accuracy": 0.23671967536211014,
      "num_tokens": 40602050.0,
      "step": 17725
    },
    {
      "entropy": 5.281772613525391,
      "epoch": 1.7031700288184437,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047138902932785363,
      "loss": 4.8052,
      "mean_token_accuracy": 0.238541279733181,
      "num_tokens": 40614552.0,
      "step": 17730
    },
    {
      "entropy": 5.328154373168945,
      "epoch": 1.7036503362151776,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047137229645854333,
      "loss": 4.791,
      "mean_token_accuracy": 0.2408497840166092,
      "num_tokens": 40625903.0,
      "step": 17735
    },
    {
      "entropy": 5.3102137565612795,
      "epoch": 1.7041306436119115,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004713555590300513,
      "loss": 4.8653,
      "mean_token_accuracy": 0.23436218202114106,
      "num_tokens": 40638634.0,
      "step": 17740
    },
    {
      "entropy": 5.353314924240112,
      "epoch": 1.7046109510086456,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004713388170427664,
      "loss": 4.8463,
      "mean_token_accuracy": 0.2332760736346245,
      "num_tokens": 40651279.0,
      "step": 17745
    },
    {
      "entropy": 5.348843812942505,
      "epoch": 1.7050912584053795,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004713220704970771,
      "loss": 4.836,
      "mean_token_accuracy": 0.23317035138607026,
      "num_tokens": 40662306.0,
      "step": 17750
    },
    {
      "entropy": 5.211499834060669,
      "epoch": 1.7055715658021133,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047130531939337236,
      "loss": 4.6836,
      "mean_token_accuracy": 0.2462380960583687,
      "num_tokens": 40672290.0,
      "step": 17755
    },
    {
      "entropy": 5.314338493347168,
      "epoch": 1.7060518731988472,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047128856373204086,
      "loss": 4.8307,
      "mean_token_accuracy": 0.23511174768209459,
      "num_tokens": 40683447.0,
      "step": 17760
    },
    {
      "entropy": 5.3971656322479244,
      "epoch": 1.7065321805955813,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047127180351347184,
      "loss": 4.9511,
      "mean_token_accuracy": 0.2256905272603035,
      "num_tokens": 40695230.0,
      "step": 17765
    },
    {
      "entropy": 5.322312879562378,
      "epoch": 1.7070124879923152,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004712550387380544,
      "loss": 4.8032,
      "mean_token_accuracy": 0.2367397964000702,
      "num_tokens": 40707311.0,
      "step": 17770
    },
    {
      "entropy": 5.3410632610321045,
      "epoch": 1.707492795389049,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004712382694061776,
      "loss": 4.8196,
      "mean_token_accuracy": 0.2334221363067627,
      "num_tokens": 40717928.0,
      "step": 17775
    },
    {
      "entropy": 5.303009080886841,
      "epoch": 1.707973102785783,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047122149551823096,
      "loss": 4.8493,
      "mean_token_accuracy": 0.2291560634970665,
      "num_tokens": 40730355.0,
      "step": 17780
    },
    {
      "entropy": 5.351337146759033,
      "epoch": 1.7084534101825168,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004712047170746039,
      "loss": 4.9181,
      "mean_token_accuracy": 0.234482604265213,
      "num_tokens": 40741412.0,
      "step": 17785
    },
    {
      "entropy": 5.318236923217773,
      "epoch": 1.7089337175792507,
      "grad_norm": 1.0,
      "learning_rate": 0.00047118793407568586,
      "loss": 4.809,
      "mean_token_accuracy": 0.2423871174454689,
      "num_tokens": 40753491.0,
      "step": 17790
    },
    {
      "entropy": 5.288894844055176,
      "epoch": 1.7094140249759846,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047117114652186657,
      "loss": 4.7611,
      "mean_token_accuracy": 0.23849527090787886,
      "num_tokens": 40765209.0,
      "step": 17795
    },
    {
      "entropy": 5.270639371871948,
      "epoch": 1.7098943323727185,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047115435441353573,
      "loss": 4.836,
      "mean_token_accuracy": 0.23104438185691833,
      "num_tokens": 40778065.0,
      "step": 17800
    },
    {
      "entropy": 5.293634700775146,
      "epoch": 1.7103746397694524,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047113755775108333,
      "loss": 4.831,
      "mean_token_accuracy": 0.23301342278718948,
      "num_tokens": 40789149.0,
      "step": 17805
    },
    {
      "entropy": 5.291439771652222,
      "epoch": 1.7108549471661862,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047112075653489913,
      "loss": 4.815,
      "mean_token_accuracy": 0.2405300572514534,
      "num_tokens": 40800340.0,
      "step": 17810
    },
    {
      "entropy": 5.326860237121582,
      "epoch": 1.7113352545629201,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004711039507653734,
      "loss": 4.8605,
      "mean_token_accuracy": 0.23034507185220718,
      "num_tokens": 40811866.0,
      "step": 17815
    },
    {
      "entropy": 5.2893288135528564,
      "epoch": 1.7118155619596542,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004710871404428961,
      "loss": 4.8645,
      "mean_token_accuracy": 0.2254443198442459,
      "num_tokens": 40825850.0,
      "step": 17820
    },
    {
      "entropy": 5.3773431301116945,
      "epoch": 1.7122958693563881,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047107032556785786,
      "loss": 4.8173,
      "mean_token_accuracy": 0.23356731981039047,
      "num_tokens": 40836688.0,
      "step": 17825
    },
    {
      "entropy": 5.385855865478516,
      "epoch": 1.712776176753122,
      "grad_norm": 1.0,
      "learning_rate": 0.00047105350614064874,
      "loss": 4.8237,
      "mean_token_accuracy": 0.2342352107167244,
      "num_tokens": 40847803.0,
      "step": 17830
    },
    {
      "entropy": 5.357816600799561,
      "epoch": 1.7132564841498559,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047103668216165944,
      "loss": 4.8807,
      "mean_token_accuracy": 0.23057694137096404,
      "num_tokens": 40859099.0,
      "step": 17835
    },
    {
      "entropy": 5.274760961532593,
      "epoch": 1.71373679154659,
      "grad_norm": 1.0,
      "learning_rate": 0.00047101985363128045,
      "loss": 4.747,
      "mean_token_accuracy": 0.24494433999061585,
      "num_tokens": 40870440.0,
      "step": 17840
    },
    {
      "entropy": 5.418163919448853,
      "epoch": 1.7142170989433239,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047100302054990255,
      "loss": 4.8959,
      "mean_token_accuracy": 0.22737378180027007,
      "num_tokens": 40882329.0,
      "step": 17845
    },
    {
      "entropy": 5.340195894241333,
      "epoch": 1.7146974063400577,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004709861829179165,
      "loss": 4.9144,
      "mean_token_accuracy": 0.2322133421897888,
      "num_tokens": 40893458.0,
      "step": 17850
    },
    {
      "entropy": 5.239327049255371,
      "epoch": 1.7151777137367916,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047096934073571325,
      "loss": 4.7429,
      "mean_token_accuracy": 0.24615864008665084,
      "num_tokens": 40904626.0,
      "step": 17855
    },
    {
      "entropy": 5.256077814102173,
      "epoch": 1.7156580211335255,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047095249400368384,
      "loss": 4.7613,
      "mean_token_accuracy": 0.24246371388435364,
      "num_tokens": 40916005.0,
      "step": 17860
    },
    {
      "entropy": 5.271145582199097,
      "epoch": 1.7161383285302594,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047093564272221927,
      "loss": 4.7434,
      "mean_token_accuracy": 0.23880276530981065,
      "num_tokens": 40927470.0,
      "step": 17865
    },
    {
      "entropy": 5.361915874481201,
      "epoch": 1.7166186359269933,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047091878689171105,
      "loss": 4.8748,
      "mean_token_accuracy": 0.23655554801225662,
      "num_tokens": 40938968.0,
      "step": 17870
    },
    {
      "entropy": 5.216669321060181,
      "epoch": 1.7170989433237271,
      "grad_norm": 0.875,
      "learning_rate": 0.0004709019265125502,
      "loss": 4.7611,
      "mean_token_accuracy": 0.24814742803573608,
      "num_tokens": 40952636.0,
      "step": 17875
    },
    {
      "entropy": 5.3332010269165036,
      "epoch": 1.717579250720461,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047088506158512837,
      "loss": 4.8777,
      "mean_token_accuracy": 0.23712229281663894,
      "num_tokens": 40964816.0,
      "step": 17880
    },
    {
      "entropy": 5.383740425109863,
      "epoch": 1.718059558117195,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047086819210983714,
      "loss": 4.8994,
      "mean_token_accuracy": 0.22948394417762757,
      "num_tokens": 40977152.0,
      "step": 17885
    },
    {
      "entropy": 5.303059530258179,
      "epoch": 1.7185398655139288,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047085131808706813,
      "loss": 4.8231,
      "mean_token_accuracy": 0.23625145107507706,
      "num_tokens": 40987506.0,
      "step": 17890
    },
    {
      "entropy": 5.343436670303345,
      "epoch": 1.7190201729106627,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004708344395172129,
      "loss": 4.8707,
      "mean_token_accuracy": 0.23138151466846466,
      "num_tokens": 40998598.0,
      "step": 17895
    },
    {
      "entropy": 5.319497871398926,
      "epoch": 1.7195004803073968,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004708175564006636,
      "loss": 4.8625,
      "mean_token_accuracy": 0.23078042417764663,
      "num_tokens": 41009644.0,
      "step": 17900
    },
    {
      "entropy": 5.29230694770813,
      "epoch": 1.7199807877041307,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004708006687378121,
      "loss": 4.7531,
      "mean_token_accuracy": 0.23724479377269744,
      "num_tokens": 41021816.0,
      "step": 17905
    },
    {
      "entropy": 5.245496559143066,
      "epoch": 1.7204610951008645,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004707837765290505,
      "loss": 4.7805,
      "mean_token_accuracy": 0.23711299300193786,
      "num_tokens": 41033482.0,
      "step": 17910
    },
    {
      "entropy": 5.316038942337036,
      "epoch": 1.7209414024975984,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004707668797747709,
      "loss": 4.7625,
      "mean_token_accuracy": 0.23921154737472533,
      "num_tokens": 41043854.0,
      "step": 17915
    },
    {
      "entropy": 5.308507680892944,
      "epoch": 1.7214217098943325,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004707499784753657,
      "loss": 4.8976,
      "mean_token_accuracy": 0.22625094503164292,
      "num_tokens": 41056956.0,
      "step": 17920
    },
    {
      "entropy": 5.387033271789551,
      "epoch": 1.7219020172910664,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004707330726312273,
      "loss": 4.9179,
      "mean_token_accuracy": 0.22587489187717438,
      "num_tokens": 41068170.0,
      "step": 17925
    },
    {
      "entropy": 5.32667989730835,
      "epoch": 1.7223823246878003,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047071616224274803,
      "loss": 4.849,
      "mean_token_accuracy": 0.23338990956544875,
      "num_tokens": 41079149.0,
      "step": 17930
    },
    {
      "entropy": 5.361198949813843,
      "epoch": 1.7228626320845342,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004706992473103207,
      "loss": 4.8773,
      "mean_token_accuracy": 0.22841697335243225,
      "num_tokens": 41091039.0,
      "step": 17935
    },
    {
      "entropy": 5.298228311538696,
      "epoch": 1.723342939481268,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047068232783433806,
      "loss": 4.8641,
      "mean_token_accuracy": 0.2364045038819313,
      "num_tokens": 41103318.0,
      "step": 17940
    },
    {
      "entropy": 5.267046594619751,
      "epoch": 1.723823246878002,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004706654038151927,
      "loss": 4.7534,
      "mean_token_accuracy": 0.24060569554567338,
      "num_tokens": 41114235.0,
      "step": 17945
    },
    {
      "entropy": 5.275362205505371,
      "epoch": 1.7243035542747358,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004706484752532777,
      "loss": 4.8219,
      "mean_token_accuracy": 0.23771378695964812,
      "num_tokens": 41126008.0,
      "step": 17950
    },
    {
      "entropy": 5.274692010879517,
      "epoch": 1.7247838616714697,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004706315421489861,
      "loss": 4.8222,
      "mean_token_accuracy": 0.2335173860192299,
      "num_tokens": 41138819.0,
      "step": 17955
    },
    {
      "entropy": 5.321470880508423,
      "epoch": 1.7252641690682036,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004706146045027109,
      "loss": 4.7936,
      "mean_token_accuracy": 0.23493523001670838,
      "num_tokens": 41149389.0,
      "step": 17960
    },
    {
      "entropy": 5.312166595458985,
      "epoch": 1.7257444764649374,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004705976623148455,
      "loss": 4.79,
      "mean_token_accuracy": 0.22997956573963166,
      "num_tokens": 41161810.0,
      "step": 17965
    },
    {
      "entropy": 5.334911346435547,
      "epoch": 1.7262247838616713,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047058071558578324,
      "loss": 4.8006,
      "mean_token_accuracy": 0.2390309303998947,
      "num_tokens": 41172903.0,
      "step": 17970
    },
    {
      "entropy": 5.341552114486694,
      "epoch": 1.7267050912584054,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004705637643159175,
      "loss": 4.8789,
      "mean_token_accuracy": 0.23691878318786622,
      "num_tokens": 41183905.0,
      "step": 17975
    },
    {
      "entropy": 5.324447774887085,
      "epoch": 1.7271853986551393,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047054680850564185,
      "loss": 4.7972,
      "mean_token_accuracy": 0.23483333587646485,
      "num_tokens": 41195921.0,
      "step": 17980
    },
    {
      "entropy": 5.237176179885864,
      "epoch": 1.7276657060518732,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004705298481553499,
      "loss": 4.8383,
      "mean_token_accuracy": 0.23883955627679826,
      "num_tokens": 41208287.0,
      "step": 17985
    },
    {
      "entropy": 5.3274205207824705,
      "epoch": 1.728146013448607,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047051288326543553,
      "loss": 4.8505,
      "mean_token_accuracy": 0.22868515700101852,
      "num_tokens": 41219864.0,
      "step": 17990
    },
    {
      "entropy": 5.326673412322998,
      "epoch": 1.7286263208453412,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047049591383629247,
      "loss": 4.7887,
      "mean_token_accuracy": 0.23629366159439086,
      "num_tokens": 41230640.0,
      "step": 17995
    },
    {
      "entropy": 5.332422828674316,
      "epoch": 1.729106628242075,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047047893986831493,
      "loss": 4.8245,
      "mean_token_accuracy": 0.23889733105897903,
      "num_tokens": 41242413.0,
      "step": 18000
    },
    {
      "epoch": 1.729106628242075,
      "eval_entropy": 5.094374594718639,
      "eval_loss": 4.888225555419922,
      "eval_mean_token_accuracy": 0.2402204139269834,
      "eval_num_tokens": 41242413.0,
      "eval_runtime": 20.6691,
      "eval_samples_per_second": 1587.637,
      "eval_steps_per_second": 198.461,
      "step": 18000
    },
    {
      "entropy": 5.334175300598145,
      "epoch": 1.729586935638809,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047046196136189686,
      "loss": 4.8516,
      "mean_token_accuracy": 0.22889852672815322,
      "num_tokens": 41254400.0,
      "step": 18005
    },
    {
      "entropy": 5.312136507034301,
      "epoch": 1.7300672430355428,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004704449783174323,
      "loss": 4.8116,
      "mean_token_accuracy": 0.22989710569381713,
      "num_tokens": 41266849.0,
      "step": 18010
    },
    {
      "entropy": 5.337336492538452,
      "epoch": 1.7305475504322767,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004704279907353158,
      "loss": 4.8905,
      "mean_token_accuracy": 0.23011037707328796,
      "num_tokens": 41278179.0,
      "step": 18015
    },
    {
      "entropy": 5.317425680160523,
      "epoch": 1.7310278578290106,
      "grad_norm": 1.0,
      "learning_rate": 0.00047041099861594167,
      "loss": 4.826,
      "mean_token_accuracy": 0.23897365480661392,
      "num_tokens": 41289904.0,
      "step": 18020
    },
    {
      "entropy": 5.418792390823365,
      "epoch": 1.7315081652257445,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004703940019597044,
      "loss": 4.9647,
      "mean_token_accuracy": 0.22421171069145202,
      "num_tokens": 41301861.0,
      "step": 18025
    },
    {
      "entropy": 5.309838438034058,
      "epoch": 1.7319884726224783,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047037700076699857,
      "loss": 4.8088,
      "mean_token_accuracy": 0.2330406576395035,
      "num_tokens": 41312420.0,
      "step": 18030
    },
    {
      "entropy": 5.279628562927246,
      "epoch": 1.7324687800192122,
      "grad_norm": 1.109375,
      "learning_rate": 0.000470359995038219,
      "loss": 4.807,
      "mean_token_accuracy": 0.23825510889291762,
      "num_tokens": 41323264.0,
      "step": 18035
    },
    {
      "entropy": 5.359568738937378,
      "epoch": 1.732949087415946,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004703429847737604,
      "loss": 4.8808,
      "mean_token_accuracy": 0.23212892413139344,
      "num_tokens": 41335438.0,
      "step": 18040
    },
    {
      "entropy": 5.35673189163208,
      "epoch": 1.73342939481268,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004703259699740177,
      "loss": 4.8877,
      "mean_token_accuracy": 0.22486549615859985,
      "num_tokens": 41347295.0,
      "step": 18045
    },
    {
      "entropy": 5.32749752998352,
      "epoch": 1.7339097022094139,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047030895063938607,
      "loss": 4.8848,
      "mean_token_accuracy": 0.2280786246061325,
      "num_tokens": 41358886.0,
      "step": 18050
    },
    {
      "entropy": 5.271694374084473,
      "epoch": 1.734390009606148,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047029192677026043,
      "loss": 4.6956,
      "mean_token_accuracy": 0.2455468326807022,
      "num_tokens": 41369402.0,
      "step": 18055
    },
    {
      "entropy": 5.427206516265869,
      "epoch": 1.7348703170028819,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004702748983670363,
      "loss": 4.922,
      "mean_token_accuracy": 0.22963193953037261,
      "num_tokens": 41379772.0,
      "step": 18060
    },
    {
      "entropy": 5.3801178455352785,
      "epoch": 1.7353506243996157,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004702578654301088,
      "loss": 4.8571,
      "mean_token_accuracy": 0.2245788961648941,
      "num_tokens": 41389776.0,
      "step": 18065
    },
    {
      "entropy": 5.2236815929412845,
      "epoch": 1.7358309317963496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004702408279598734,
      "loss": 4.7552,
      "mean_token_accuracy": 0.23249728828668595,
      "num_tokens": 41400640.0,
      "step": 18070
    },
    {
      "entropy": 5.22925386428833,
      "epoch": 1.7363112391930837,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004702237859567258,
      "loss": 4.7788,
      "mean_token_accuracy": 0.2315487265586853,
      "num_tokens": 41412653.0,
      "step": 18075
    },
    {
      "entropy": 5.421981000900269,
      "epoch": 1.7367915465898176,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004702067394210616,
      "loss": 4.9211,
      "mean_token_accuracy": 0.22489297837018968,
      "num_tokens": 41423815.0,
      "step": 18080
    },
    {
      "entropy": 5.352878665924072,
      "epoch": 1.7372718539865515,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047018968835327643,
      "loss": 4.8351,
      "mean_token_accuracy": 0.2285082072019577,
      "num_tokens": 41436016.0,
      "step": 18085
    },
    {
      "entropy": 5.249156761169433,
      "epoch": 1.7377521613832854,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004701726327537664,
      "loss": 4.7944,
      "mean_token_accuracy": 0.23746320903301238,
      "num_tokens": 41447837.0,
      "step": 18090
    },
    {
      "entropy": 5.328136157989502,
      "epoch": 1.7382324687800192,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004701555726229274,
      "loss": 4.8591,
      "mean_token_accuracy": 0.23421545326709747,
      "num_tokens": 41459816.0,
      "step": 18095
    },
    {
      "entropy": 5.345780849456787,
      "epoch": 1.7387127761767531,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004701385079611555,
      "loss": 4.7965,
      "mean_token_accuracy": 0.23594623953104019,
      "num_tokens": 41471461.0,
      "step": 18100
    },
    {
      "entropy": 5.278888845443726,
      "epoch": 1.739193083573487,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047012143876884677,
      "loss": 4.7872,
      "mean_token_accuracy": 0.24245323687791825,
      "num_tokens": 41481886.0,
      "step": 18105
    },
    {
      "entropy": 5.272528028488159,
      "epoch": 1.739673390970221,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004701043650463977,
      "loss": 4.8262,
      "mean_token_accuracy": 0.23916192650794982,
      "num_tokens": 41493698.0,
      "step": 18110
    },
    {
      "entropy": 5.288167667388916,
      "epoch": 1.7401536983669548,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004700872867942046,
      "loss": 4.7738,
      "mean_token_accuracy": 0.24749969094991683,
      "num_tokens": 41504744.0,
      "step": 18115
    },
    {
      "entropy": 5.321786975860595,
      "epoch": 1.7406340057636887,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000470070204012664,
      "loss": 4.8488,
      "mean_token_accuracy": 0.23671911656856537,
      "num_tokens": 41517129.0,
      "step": 18120
    },
    {
      "entropy": 5.380322742462158,
      "epoch": 1.7411143131604225,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047005311670217256,
      "loss": 4.8485,
      "mean_token_accuracy": 0.2241552472114563,
      "num_tokens": 41529969.0,
      "step": 18125
    },
    {
      "entropy": 5.320414781570435,
      "epoch": 1.7415946205571564,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047003602486312687,
      "loss": 4.7921,
      "mean_token_accuracy": 0.2373647004365921,
      "num_tokens": 41541515.0,
      "step": 18130
    },
    {
      "entropy": 5.279444837570191,
      "epoch": 1.7420749279538905,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004700189284959238,
      "loss": 4.8586,
      "mean_token_accuracy": 0.23054289817810059,
      "num_tokens": 41553045.0,
      "step": 18135
    },
    {
      "entropy": 5.318865394592285,
      "epoch": 1.7425552353506244,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047000182760096037,
      "loss": 4.878,
      "mean_token_accuracy": 0.22913806289434432,
      "num_tokens": 41565376.0,
      "step": 18140
    },
    {
      "entropy": 5.321056222915649,
      "epoch": 1.7430355427473583,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004699847221786335,
      "loss": 4.8518,
      "mean_token_accuracy": 0.22313832491636276,
      "num_tokens": 41577515.0,
      "step": 18145
    },
    {
      "entropy": 5.278144836425781,
      "epoch": 1.7435158501440924,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004699676122293403,
      "loss": 4.7445,
      "mean_token_accuracy": 0.2371098056435585,
      "num_tokens": 41588589.0,
      "step": 18150
    },
    {
      "entropy": 5.361829471588135,
      "epoch": 1.7439961575408263,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004699504977534782,
      "loss": 4.902,
      "mean_token_accuracy": 0.22619068175554274,
      "num_tokens": 41600727.0,
      "step": 18155
    },
    {
      "entropy": 5.363212966918946,
      "epoch": 1.7444764649375601,
      "grad_norm": 1.0,
      "learning_rate": 0.0004699333787514444,
      "loss": 4.8187,
      "mean_token_accuracy": 0.2400011122226715,
      "num_tokens": 41612177.0,
      "step": 18160
    },
    {
      "entropy": 5.342322063446045,
      "epoch": 1.744956772334294,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004699162552236363,
      "loss": 4.8436,
      "mean_token_accuracy": 0.23816806375980376,
      "num_tokens": 41623749.0,
      "step": 18165
    },
    {
      "entropy": 5.349839735031128,
      "epoch": 1.745437079731028,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046989912717045165,
      "loss": 4.8569,
      "mean_token_accuracy": 0.23741406500339507,
      "num_tokens": 41635549.0,
      "step": 18170
    },
    {
      "entropy": 5.382081651687622,
      "epoch": 1.7459173871277618,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046988199459228793,
      "loss": 4.9226,
      "mean_token_accuracy": 0.228255432844162,
      "num_tokens": 41647219.0,
      "step": 18175
    },
    {
      "entropy": 5.416785097122192,
      "epoch": 1.7463976945244957,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004698648574895429,
      "loss": 4.8538,
      "mean_token_accuracy": 0.23395963460206987,
      "num_tokens": 41658850.0,
      "step": 18180
    },
    {
      "entropy": 5.407287740707398,
      "epoch": 1.7468780019212296,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046984771586261465,
      "loss": 4.873,
      "mean_token_accuracy": 0.2280627280473709,
      "num_tokens": 41669304.0,
      "step": 18185
    },
    {
      "entropy": 5.387171649932862,
      "epoch": 1.7473583093179634,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000469830569711901,
      "loss": 4.9075,
      "mean_token_accuracy": 0.23798957467079163,
      "num_tokens": 41681652.0,
      "step": 18190
    },
    {
      "entropy": 5.358082246780396,
      "epoch": 1.7478386167146973,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004698134190377999,
      "loss": 4.8483,
      "mean_token_accuracy": 0.2314437747001648,
      "num_tokens": 41692548.0,
      "step": 18195
    },
    {
      "entropy": 5.352532815933228,
      "epoch": 1.7483189241114312,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046979626384070983,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23135943114757537,
      "num_tokens": 41704528.0,
      "step": 18200
    },
    {
      "entropy": 5.303145790100098,
      "epoch": 1.748799231508165,
      "grad_norm": 0.9375,
      "learning_rate": 0.000469779104121029,
      "loss": 4.812,
      "mean_token_accuracy": 0.23737801611423492,
      "num_tokens": 41717314.0,
      "step": 18205
    },
    {
      "entropy": 5.424673366546631,
      "epoch": 1.7492795389048992,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046976193987915553,
      "loss": 4.908,
      "mean_token_accuracy": 0.2301422983407974,
      "num_tokens": 41729193.0,
      "step": 18210
    },
    {
      "entropy": 5.345047473907471,
      "epoch": 1.749759846301633,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004697447711154883,
      "loss": 4.7846,
      "mean_token_accuracy": 0.23849915713071823,
      "num_tokens": 41741208.0,
      "step": 18215
    },
    {
      "entropy": 5.316779088973999,
      "epoch": 1.750240153698367,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046972759783042576,
      "loss": 4.7922,
      "mean_token_accuracy": 0.238743594288826,
      "num_tokens": 41752792.0,
      "step": 18220
    },
    {
      "entropy": 5.345543622970581,
      "epoch": 1.7507204610951008,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004697104200243666,
      "loss": 4.8077,
      "mean_token_accuracy": 0.24057514369487762,
      "num_tokens": 41763629.0,
      "step": 18225
    },
    {
      "entropy": 5.2756541728973385,
      "epoch": 1.751200768491835,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004696932376977096,
      "loss": 4.7838,
      "mean_token_accuracy": 0.24175293296575545,
      "num_tokens": 41774502.0,
      "step": 18230
    },
    {
      "entropy": 5.32549090385437,
      "epoch": 1.7516810758885688,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004696760508508538,
      "loss": 4.8417,
      "mean_token_accuracy": 0.2297625094652176,
      "num_tokens": 41785700.0,
      "step": 18235
    },
    {
      "entropy": 5.24150915145874,
      "epoch": 1.7521613832853027,
      "grad_norm": 1.0,
      "learning_rate": 0.00046965885948419814,
      "loss": 4.8085,
      "mean_token_accuracy": 0.2284177765250206,
      "num_tokens": 41797347.0,
      "step": 18240
    },
    {
      "entropy": 5.293800973892212,
      "epoch": 1.7526416906820366,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004696416635981418,
      "loss": 4.8533,
      "mean_token_accuracy": 0.2355642169713974,
      "num_tokens": 41810866.0,
      "step": 18245
    },
    {
      "entropy": 5.370873260498047,
      "epoch": 1.7531219980787704,
      "grad_norm": 1.0625,
      "learning_rate": 0.000469624463193084,
      "loss": 4.8607,
      "mean_token_accuracy": 0.2358009099960327,
      "num_tokens": 41821786.0,
      "step": 18250
    },
    {
      "entropy": 5.365944910049438,
      "epoch": 1.7536023054755043,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000469607258269424,
      "loss": 4.8493,
      "mean_token_accuracy": 0.2310372069478035,
      "num_tokens": 41833591.0,
      "step": 18255
    },
    {
      "entropy": 5.285492706298828,
      "epoch": 1.7540826128722382,
      "grad_norm": 1.0,
      "learning_rate": 0.0004695900488275614,
      "loss": 4.7579,
      "mean_token_accuracy": 0.2433210387825966,
      "num_tokens": 41844659.0,
      "step": 18260
    },
    {
      "entropy": 5.288796424865723,
      "epoch": 1.754562920268972,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004695728348678957,
      "loss": 4.8066,
      "mean_token_accuracy": 0.23284512907266616,
      "num_tokens": 41856378.0,
      "step": 18265
    },
    {
      "entropy": 5.305641174316406,
      "epoch": 1.755043227665706,
      "grad_norm": 1.0,
      "learning_rate": 0.0004695556163908265,
      "loss": 4.7967,
      "mean_token_accuracy": 0.23265099674463272,
      "num_tokens": 41867624.0,
      "step": 18270
    },
    {
      "entropy": 5.325054931640625,
      "epoch": 1.7555235350624399,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004695383933967536,
      "loss": 4.8324,
      "mean_token_accuracy": 0.24321859180927277,
      "num_tokens": 41879250.0,
      "step": 18275
    },
    {
      "entropy": 5.322017812728882,
      "epoch": 1.7560038424591737,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046952116588607694,
      "loss": 4.8982,
      "mean_token_accuracy": 0.2350863352417946,
      "num_tokens": 41890713.0,
      "step": 18280
    },
    {
      "entropy": 5.3584897994995115,
      "epoch": 1.7564841498559076,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004695039338591963,
      "loss": 4.8777,
      "mean_token_accuracy": 0.23481959849596024,
      "num_tokens": 41902924.0,
      "step": 18285
    },
    {
      "entropy": 5.3814935207366945,
      "epoch": 1.7569644572526417,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000469486697316512,
      "loss": 4.8829,
      "mean_token_accuracy": 0.23306052535772323,
      "num_tokens": 41914498.0,
      "step": 18290
    },
    {
      "entropy": 5.309173107147217,
      "epoch": 1.7574447646493756,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000469469456258424,
      "loss": 4.8129,
      "mean_token_accuracy": 0.23048830181360244,
      "num_tokens": 41925258.0,
      "step": 18295
    },
    {
      "entropy": 5.311397743225098,
      "epoch": 1.7579250720461095,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004694522106853327,
      "loss": 4.8436,
      "mean_token_accuracy": 0.22938225269317628,
      "num_tokens": 41938176.0,
      "step": 18300
    },
    {
      "entropy": 5.287655973434449,
      "epoch": 1.7584053794428436,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046943496059763845,
      "loss": 4.6951,
      "mean_token_accuracy": 0.24524835199117662,
      "num_tokens": 41949160.0,
      "step": 18305
    },
    {
      "entropy": 5.3232372283935545,
      "epoch": 1.7588856868395775,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046941770599574176,
      "loss": 4.8771,
      "mean_token_accuracy": 0.23405975699424744,
      "num_tokens": 41960981.0,
      "step": 18310
    },
    {
      "entropy": 5.2687348365783695,
      "epoch": 1.7593659942363113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004694004468800433,
      "loss": 4.8437,
      "mean_token_accuracy": 0.23284367918968202,
      "num_tokens": 41971755.0,
      "step": 18315
    },
    {
      "entropy": 5.30984845161438,
      "epoch": 1.7598463016330452,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004693831832509437,
      "loss": 4.8113,
      "mean_token_accuracy": 0.24259986430406572,
      "num_tokens": 41984337.0,
      "step": 18320
    },
    {
      "entropy": 5.288689470291137,
      "epoch": 1.760326609029779,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046936591510884375,
      "loss": 4.8176,
      "mean_token_accuracy": 0.23725712448358535,
      "num_tokens": 41995648.0,
      "step": 18325
    },
    {
      "entropy": 5.2736903667449955,
      "epoch": 1.760806916426513,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046934864245414443,
      "loss": 4.7802,
      "mean_token_accuracy": 0.24014419913291932,
      "num_tokens": 42006995.0,
      "step": 18330
    },
    {
      "entropy": 5.387620067596435,
      "epoch": 1.7612872238232469,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046933136528724676,
      "loss": 4.919,
      "mean_token_accuracy": 0.22679372876882553,
      "num_tokens": 42020074.0,
      "step": 18335
    },
    {
      "entropy": 5.291437578201294,
      "epoch": 1.7617675312199808,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004693140836085518,
      "loss": 4.7348,
      "mean_token_accuracy": 0.24398934245109558,
      "num_tokens": 42032053.0,
      "step": 18340
    },
    {
      "entropy": 5.279104471206665,
      "epoch": 1.7622478386167146,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046929679741846076,
      "loss": 4.773,
      "mean_token_accuracy": 0.24521418809890747,
      "num_tokens": 42042223.0,
      "step": 18345
    },
    {
      "entropy": 5.295570755004883,
      "epoch": 1.7627281460134485,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046927950671737505,
      "loss": 4.8329,
      "mean_token_accuracy": 0.2436547890305519,
      "num_tokens": 42053988.0,
      "step": 18350
    },
    {
      "entropy": 5.229391622543335,
      "epoch": 1.7632084534101824,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046926221150569617,
      "loss": 4.7457,
      "mean_token_accuracy": 0.24266680479049682,
      "num_tokens": 42064451.0,
      "step": 18355
    },
    {
      "entropy": 5.313588285446167,
      "epoch": 1.7636887608069163,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004692449117838255,
      "loss": 4.9195,
      "mean_token_accuracy": 0.22560426443815232,
      "num_tokens": 42076503.0,
      "step": 18360
    },
    {
      "entropy": 5.373799896240234,
      "epoch": 1.7641690682036504,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004692276075521648,
      "loss": 4.8435,
      "mean_token_accuracy": 0.22682348191738128,
      "num_tokens": 42088341.0,
      "step": 18365
    },
    {
      "entropy": 5.3279211044311525,
      "epoch": 1.7646493756003843,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004692102988111158,
      "loss": 4.8,
      "mean_token_accuracy": 0.23738011568784714,
      "num_tokens": 42099862.0,
      "step": 18370
    },
    {
      "entropy": 5.358805751800537,
      "epoch": 1.7651296829971181,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046919298556108023,
      "loss": 4.9132,
      "mean_token_accuracy": 0.2350209340453148,
      "num_tokens": 42112117.0,
      "step": 18375
    },
    {
      "entropy": 5.375319957733154,
      "epoch": 1.765609990393852,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046917566780246036,
      "loss": 4.836,
      "mean_token_accuracy": 0.2308885559439659,
      "num_tokens": 42123093.0,
      "step": 18380
    },
    {
      "entropy": 5.3673210620880125,
      "epoch": 1.7660902977905861,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046915834553565793,
      "loss": 4.9187,
      "mean_token_accuracy": 0.22658791095018388,
      "num_tokens": 42135266.0,
      "step": 18385
    },
    {
      "entropy": 5.442306327819824,
      "epoch": 1.76657060518732,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004691410187610753,
      "loss": 4.9594,
      "mean_token_accuracy": 0.22727010250091553,
      "num_tokens": 42145743.0,
      "step": 18390
    },
    {
      "entropy": 5.268835973739624,
      "epoch": 1.767050912584054,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046912368747911465,
      "loss": 4.7555,
      "mean_token_accuracy": 0.24157587587833404,
      "num_tokens": 42157604.0,
      "step": 18395
    },
    {
      "entropy": 5.309444999694824,
      "epoch": 1.7675312199807878,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046910635169017845,
      "loss": 4.7649,
      "mean_token_accuracy": 0.2404394879937172,
      "num_tokens": 42168738.0,
      "step": 18400
    },
    {
      "entropy": 5.366307306289673,
      "epoch": 1.7680115273775217,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004690890113946691,
      "loss": 4.8748,
      "mean_token_accuracy": 0.2256273850798607,
      "num_tokens": 42179699.0,
      "step": 18405
    },
    {
      "entropy": 5.312521982192993,
      "epoch": 1.7684918347742555,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004690716665929893,
      "loss": 4.8105,
      "mean_token_accuracy": 0.2302168533205986,
      "num_tokens": 42190512.0,
      "step": 18410
    },
    {
      "entropy": 5.368676567077637,
      "epoch": 1.7689721421709894,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046905431728554164,
      "loss": 4.863,
      "mean_token_accuracy": 0.22884422689676284,
      "num_tokens": 42200996.0,
      "step": 18415
    },
    {
      "entropy": 5.2759167671203615,
      "epoch": 1.7694524495677233,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046903696347272894,
      "loss": 4.7853,
      "mean_token_accuracy": 0.24196382462978364,
      "num_tokens": 42213032.0,
      "step": 18420
    },
    {
      "entropy": 5.305151271820068,
      "epoch": 1.7699327569644572,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046901960515495413,
      "loss": 4.8794,
      "mean_token_accuracy": 0.22996368110179902,
      "num_tokens": 42224366.0,
      "step": 18425
    },
    {
      "entropy": 5.256261396408081,
      "epoch": 1.770413064361191,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004690022423326202,
      "loss": 4.7283,
      "mean_token_accuracy": 0.24411833435297012,
      "num_tokens": 42235540.0,
      "step": 18430
    },
    {
      "entropy": 5.28740873336792,
      "epoch": 1.770893371757925,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004689848750061303,
      "loss": 4.7901,
      "mean_token_accuracy": 0.23745082318782806,
      "num_tokens": 42246755.0,
      "step": 18435
    },
    {
      "entropy": 5.287445449829102,
      "epoch": 1.7713736791546588,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004689675031758876,
      "loss": 4.7867,
      "mean_token_accuracy": 0.23765382915735245,
      "num_tokens": 42259536.0,
      "step": 18440
    },
    {
      "entropy": 5.29618730545044,
      "epoch": 1.771853986551393,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004689501268422954,
      "loss": 4.7899,
      "mean_token_accuracy": 0.23535023033618926,
      "num_tokens": 42270813.0,
      "step": 18445
    },
    {
      "entropy": 5.449173545837402,
      "epoch": 1.7723342939481268,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046893274600575725,
      "loss": 4.9535,
      "mean_token_accuracy": 0.22201204895973206,
      "num_tokens": 42281953.0,
      "step": 18450
    },
    {
      "entropy": 5.389143180847168,
      "epoch": 1.7728146013448607,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004689153606666765,
      "loss": 4.9451,
      "mean_token_accuracy": 0.22534324675798417,
      "num_tokens": 42294041.0,
      "step": 18455
    },
    {
      "entropy": 5.321201372146606,
      "epoch": 1.7732949087415946,
      "grad_norm": 1.0,
      "learning_rate": 0.000468897970825457,
      "loss": 4.7735,
      "mean_token_accuracy": 0.23683712333440782,
      "num_tokens": 42305812.0,
      "step": 18460
    },
    {
      "entropy": 5.303789854049683,
      "epoch": 1.7737752161383287,
      "grad_norm": 1.0,
      "learning_rate": 0.00046888057648250233,
      "loss": 4.7476,
      "mean_token_accuracy": 0.23850935846567153,
      "num_tokens": 42317585.0,
      "step": 18465
    },
    {
      "entropy": 5.32672963142395,
      "epoch": 1.7742555235350626,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004688631776382164,
      "loss": 4.8078,
      "mean_token_accuracy": 0.2321149528026581,
      "num_tokens": 42328723.0,
      "step": 18470
    },
    {
      "entropy": 5.2855912208557125,
      "epoch": 1.7747358309317964,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046884577429300305,
      "loss": 4.821,
      "mean_token_accuracy": 0.24797983467578888,
      "num_tokens": 42340741.0,
      "step": 18475
    },
    {
      "entropy": 5.31296215057373,
      "epoch": 1.7752161383285303,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004688283664472665,
      "loss": 4.8616,
      "mean_token_accuracy": 0.230816313624382,
      "num_tokens": 42351504.0,
      "step": 18480
    },
    {
      "entropy": 5.338756704330445,
      "epoch": 1.7756964457252642,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046881095410141084,
      "loss": 4.8166,
      "mean_token_accuracy": 0.2317630559206009,
      "num_tokens": 42362920.0,
      "step": 18485
    },
    {
      "entropy": 5.289158296585083,
      "epoch": 1.776176753121998,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046879353725584036,
      "loss": 4.764,
      "mean_token_accuracy": 0.23623269498348237,
      "num_tokens": 42374523.0,
      "step": 18490
    },
    {
      "entropy": 5.363108921051025,
      "epoch": 1.776657060518732,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046877611591095923,
      "loss": 4.8578,
      "mean_token_accuracy": 0.23155148327350616,
      "num_tokens": 42384988.0,
      "step": 18495
    },
    {
      "entropy": 5.236738824844361,
      "epoch": 1.7771373679154658,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046875869006717224,
      "loss": 4.753,
      "mean_token_accuracy": 0.24231024980545043,
      "num_tokens": 42396503.0,
      "step": 18500
    },
    {
      "entropy": 5.3931622982025145,
      "epoch": 1.7776176753121997,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046874125972488375,
      "loss": 4.8622,
      "mean_token_accuracy": 0.2318602591753006,
      "num_tokens": 42406877.0,
      "step": 18505
    },
    {
      "entropy": 5.322722053527832,
      "epoch": 1.7780979827089336,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046872382488449853,
      "loss": 4.8054,
      "mean_token_accuracy": 0.23636471182107927,
      "num_tokens": 42418010.0,
      "step": 18510
    },
    {
      "entropy": 5.3079310894012455,
      "epoch": 1.7785782901056675,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046870638554642133,
      "loss": 4.7733,
      "mean_token_accuracy": 0.23646242320537567,
      "num_tokens": 42429056.0,
      "step": 18515
    },
    {
      "entropy": 5.240431022644043,
      "epoch": 1.7790585975024016,
      "grad_norm": 1.0,
      "learning_rate": 0.00046868894171105704,
      "loss": 4.7696,
      "mean_token_accuracy": 0.23970983028411866,
      "num_tokens": 42442294.0,
      "step": 18520
    },
    {
      "entropy": 5.335350561141968,
      "epoch": 1.7795389048991355,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004686714933788107,
      "loss": 4.8598,
      "mean_token_accuracy": 0.22822721302509308,
      "num_tokens": 42454064.0,
      "step": 18525
    },
    {
      "entropy": 5.326143598556518,
      "epoch": 1.7800192122958693,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004686540405500873,
      "loss": 4.8143,
      "mean_token_accuracy": 0.2421652138233185,
      "num_tokens": 42465289.0,
      "step": 18530
    },
    {
      "entropy": 5.262371015548706,
      "epoch": 1.7804995196926032,
      "grad_norm": 1.0,
      "learning_rate": 0.0004686365832252922,
      "loss": 4.7172,
      "mean_token_accuracy": 0.24332476705312728,
      "num_tokens": 42476662.0,
      "step": 18535
    },
    {
      "entropy": 5.294908857345581,
      "epoch": 1.7809798270893373,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046861912140483056,
      "loss": 4.7979,
      "mean_token_accuracy": 0.24014018774032592,
      "num_tokens": 42486893.0,
      "step": 18540
    },
    {
      "entropy": 5.3018640041351315,
      "epoch": 1.7814601344860712,
      "grad_norm": 1.0,
      "learning_rate": 0.00046860165508910787,
      "loss": 4.799,
      "mean_token_accuracy": 0.23657313287258147,
      "num_tokens": 42498202.0,
      "step": 18545
    },
    {
      "entropy": 5.270104885101318,
      "epoch": 1.781940441882805,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004685841842785296,
      "loss": 4.6859,
      "mean_token_accuracy": 0.2501818150281906,
      "num_tokens": 42509378.0,
      "step": 18550
    },
    {
      "entropy": 5.335957622528076,
      "epoch": 1.782420749279539,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004685667089735014,
      "loss": 4.8601,
      "mean_token_accuracy": 0.23777783960103988,
      "num_tokens": 42519955.0,
      "step": 18555
    },
    {
      "entropy": 5.3702586650848385,
      "epoch": 1.7829010566762729,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046854922917442907,
      "loss": 4.9175,
      "mean_token_accuracy": 0.2292545959353447,
      "num_tokens": 42531896.0,
      "step": 18560
    },
    {
      "entropy": 5.347605800628662,
      "epoch": 1.7833813640730067,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004685317448817182,
      "loss": 4.8199,
      "mean_token_accuracy": 0.23771380484104157,
      "num_tokens": 42542123.0,
      "step": 18565
    },
    {
      "entropy": 5.341956615447998,
      "epoch": 1.7838616714697406,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004685142560957751,
      "loss": 4.8151,
      "mean_token_accuracy": 0.2317924052476883,
      "num_tokens": 42554034.0,
      "step": 18570
    },
    {
      "entropy": 5.290957593917847,
      "epoch": 1.7843419788664745,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004684967628170054,
      "loss": 4.7691,
      "mean_token_accuracy": 0.24121385663747788,
      "num_tokens": 42564532.0,
      "step": 18575
    },
    {
      "entropy": 5.304830884933471,
      "epoch": 1.7848222862632084,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00046847926504581553,
      "loss": 4.7701,
      "mean_token_accuracy": 0.2322750687599182,
      "num_tokens": 42576864.0,
      "step": 18580
    },
    {
      "entropy": 5.250645160675049,
      "epoch": 1.7853025936599423,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004684617627826116,
      "loss": 4.7555,
      "mean_token_accuracy": 0.24496842473745345,
      "num_tokens": 42588571.0,
      "step": 18585
    },
    {
      "entropy": 5.360170173645019,
      "epoch": 1.7857829010566761,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004684442560278001,
      "loss": 4.8885,
      "mean_token_accuracy": 0.22591403126716614,
      "num_tokens": 42600912.0,
      "step": 18590
    },
    {
      "entropy": 5.355820369720459,
      "epoch": 1.78626320845341,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046842674478178727,
      "loss": 4.7603,
      "mean_token_accuracy": 0.23570075929164885,
      "num_tokens": 42612929.0,
      "step": 18595
    },
    {
      "entropy": 5.316252279281616,
      "epoch": 1.7867435158501441,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004684092290449798,
      "loss": 4.8378,
      "mean_token_accuracy": 0.24060069173574447,
      "num_tokens": 42623688.0,
      "step": 18600
    },
    {
      "entropy": 5.418286561965942,
      "epoch": 1.787223823246878,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004683917088177844,
      "loss": 4.9683,
      "mean_token_accuracy": 0.2215547889471054,
      "num_tokens": 42635773.0,
      "step": 18605
    },
    {
      "entropy": 5.358272647857666,
      "epoch": 1.7877041306436119,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004683741841006077,
      "loss": 4.8025,
      "mean_token_accuracy": 0.23297586143016816,
      "num_tokens": 42647210.0,
      "step": 18610
    },
    {
      "entropy": 5.337363576889038,
      "epoch": 1.7881844380403458,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004683566548938567,
      "loss": 4.8457,
      "mean_token_accuracy": 0.23124426752328872,
      "num_tokens": 42658029.0,
      "step": 18615
    },
    {
      "entropy": 5.277897500991822,
      "epoch": 1.7886647454370799,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004683391211979383,
      "loss": 4.8565,
      "mean_token_accuracy": 0.23412678092718125,
      "num_tokens": 42669327.0,
      "step": 18620
    },
    {
      "entropy": 5.342040967941284,
      "epoch": 1.7891450528338138,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004683215830132597,
      "loss": 4.7657,
      "mean_token_accuracy": 0.2439696729183197,
      "num_tokens": 42679794.0,
      "step": 18625
    },
    {
      "entropy": 5.345557641983032,
      "epoch": 1.7896253602305476,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046830404034022786,
      "loss": 4.7909,
      "mean_token_accuracy": 0.24250876903533936,
      "num_tokens": 42689838.0,
      "step": 18630
    },
    {
      "entropy": 5.3300461769104,
      "epoch": 1.7901056676272815,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004682864931792502,
      "loss": 4.8592,
      "mean_token_accuracy": 0.22953216284513472,
      "num_tokens": 42700935.0,
      "step": 18635
    },
    {
      "entropy": 5.418390417098999,
      "epoch": 1.7905859750240154,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004682689415307342,
      "loss": 4.9285,
      "mean_token_accuracy": 0.22354381382465363,
      "num_tokens": 42712533.0,
      "step": 18640
    },
    {
      "entropy": 5.35843300819397,
      "epoch": 1.7910662824207493,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004682513853950872,
      "loss": 4.8268,
      "mean_token_accuracy": 0.23694084584712982,
      "num_tokens": 42723604.0,
      "step": 18645
    },
    {
      "entropy": 5.3644201278686525,
      "epoch": 1.7915465898174832,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000468233824772717,
      "loss": 4.8299,
      "mean_token_accuracy": 0.23742058426141738,
      "num_tokens": 42735957.0,
      "step": 18650
    },
    {
      "entropy": 5.374178504943847,
      "epoch": 1.792026897214217,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000468216259664031,
      "loss": 4.9461,
      "mean_token_accuracy": 0.22973705381155013,
      "num_tokens": 42747287.0,
      "step": 18655
    },
    {
      "entropy": 5.312151861190796,
      "epoch": 1.792507204610951,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046819869006943727,
      "loss": 4.8015,
      "mean_token_accuracy": 0.2389428585767746,
      "num_tokens": 42759270.0,
      "step": 18660
    },
    {
      "entropy": 5.251245737075806,
      "epoch": 1.7929875120076848,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004681811159893436,
      "loss": 4.7584,
      "mean_token_accuracy": 0.23988056778907776,
      "num_tokens": 42771955.0,
      "step": 18665
    },
    {
      "entropy": 5.433082485198975,
      "epoch": 1.7934678194044187,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046816353742415814,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2298687919974327,
      "num_tokens": 42784610.0,
      "step": 18670
    },
    {
      "entropy": 5.389463996887207,
      "epoch": 1.7939481268011528,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046814595437428885,
      "loss": 4.8221,
      "mean_token_accuracy": 0.23702902793884278,
      "num_tokens": 42795221.0,
      "step": 18675
    },
    {
      "entropy": 5.242968654632568,
      "epoch": 1.7944284341978867,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000468128366840144,
      "loss": 4.797,
      "mean_token_accuracy": 0.23541533946990967,
      "num_tokens": 42807475.0,
      "step": 18680
    },
    {
      "entropy": 5.355447149276733,
      "epoch": 1.7949087415946205,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000468110774822132,
      "loss": 4.7955,
      "mean_token_accuracy": 0.2357812449336052,
      "num_tokens": 42819193.0,
      "step": 18685
    },
    {
      "entropy": 5.375124931335449,
      "epoch": 1.7953890489913544,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004680931783206612,
      "loss": 4.8704,
      "mean_token_accuracy": 0.2359850063920021,
      "num_tokens": 42831267.0,
      "step": 18690
    },
    {
      "entropy": 5.353081464767456,
      "epoch": 1.7958693563880885,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046807557733614014,
      "loss": 4.9109,
      "mean_token_accuracy": 0.22933330684900283,
      "num_tokens": 42843066.0,
      "step": 18695
    },
    {
      "entropy": 5.307684087753296,
      "epoch": 1.7963496637848224,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046805797186897757,
      "loss": 4.7742,
      "mean_token_accuracy": 0.2406655639410019,
      "num_tokens": 42854630.0,
      "step": 18700
    },
    {
      "entropy": 5.420975685119629,
      "epoch": 1.7968299711815563,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046804036191958206,
      "loss": 4.8657,
      "mean_token_accuracy": 0.22765026092529297,
      "num_tokens": 42865986.0,
      "step": 18705
    },
    {
      "entropy": 5.406425142288208,
      "epoch": 1.7973102785782902,
      "grad_norm": 1.0,
      "learning_rate": 0.00046802274748836267,
      "loss": 4.888,
      "mean_token_accuracy": 0.2329917773604393,
      "num_tokens": 42877533.0,
      "step": 18710
    },
    {
      "entropy": 5.301871013641358,
      "epoch": 1.797790585975024,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004680051285757281,
      "loss": 4.7596,
      "mean_token_accuracy": 0.24334810376167298,
      "num_tokens": 42889114.0,
      "step": 18715
    },
    {
      "entropy": 5.393255424499512,
      "epoch": 1.798270893371758,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004679875051820877,
      "loss": 4.9446,
      "mean_token_accuracy": 0.22610963582992555,
      "num_tokens": 42899483.0,
      "step": 18720
    },
    {
      "entropy": 5.329036903381348,
      "epoch": 1.7987512007684918,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004679698773078503,
      "loss": 4.803,
      "mean_token_accuracy": 0.24018712490797042,
      "num_tokens": 42910267.0,
      "step": 18725
    },
    {
      "entropy": 5.303273487091064,
      "epoch": 1.7992315081652257,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00046795224495342554,
      "loss": 4.8117,
      "mean_token_accuracy": 0.2356491506099701,
      "num_tokens": 42922440.0,
      "step": 18730
    },
    {
      "entropy": 5.30610842704773,
      "epoch": 1.7997118155619596,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046793460811922255,
      "loss": 4.7801,
      "mean_token_accuracy": 0.24123239368200303,
      "num_tokens": 42933967.0,
      "step": 18735
    },
    {
      "entropy": 5.359523344039917,
      "epoch": 1.8001921229586935,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046791696680565075,
      "loss": 4.8046,
      "mean_token_accuracy": 0.23364455848932267,
      "num_tokens": 42945049.0,
      "step": 18740
    },
    {
      "entropy": 5.383444356918335,
      "epoch": 1.8006724303554273,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046789932101312003,
      "loss": 4.9201,
      "mean_token_accuracy": 0.22682029902935028,
      "num_tokens": 42956062.0,
      "step": 18745
    },
    {
      "entropy": 5.366603708267212,
      "epoch": 1.8011527377521612,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004678816707420397,
      "loss": 4.8911,
      "mean_token_accuracy": 0.2340461477637291,
      "num_tokens": 42967687.0,
      "step": 18750
    },
    {
      "entropy": 5.391021299362182,
      "epoch": 1.8016330451488953,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004678640159928198,
      "loss": 4.8718,
      "mean_token_accuracy": 0.22202206254005433,
      "num_tokens": 42979916.0,
      "step": 18755
    },
    {
      "entropy": 5.389227104187012,
      "epoch": 1.8021133525456292,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004678463567658701,
      "loss": 4.8583,
      "mean_token_accuracy": 0.23270570188760759,
      "num_tokens": 42991639.0,
      "step": 18760
    },
    {
      "entropy": 5.3586420059204105,
      "epoch": 1.802593659942363,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004678286930616006,
      "loss": 4.8353,
      "mean_token_accuracy": 0.2409009486436844,
      "num_tokens": 43001843.0,
      "step": 18765
    },
    {
      "entropy": 5.254851818084717,
      "epoch": 1.803073967339097,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004678110248804215,
      "loss": 4.7816,
      "mean_token_accuracy": 0.23013591319322585,
      "num_tokens": 43013249.0,
      "step": 18770
    },
    {
      "entropy": 5.304228115081787,
      "epoch": 1.803554274735831,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046779335222274293,
      "loss": 4.8277,
      "mean_token_accuracy": 0.23691774010658265,
      "num_tokens": 43025867.0,
      "step": 18775
    },
    {
      "entropy": 5.3752845287322994,
      "epoch": 1.804034582132565,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046777567508897515,
      "loss": 4.8936,
      "mean_token_accuracy": 0.22844956070184708,
      "num_tokens": 43037530.0,
      "step": 18780
    },
    {
      "entropy": 5.347773170471191,
      "epoch": 1.8045148895292988,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046775799347952864,
      "loss": 4.8634,
      "mean_token_accuracy": 0.23357735127210616,
      "num_tokens": 43048675.0,
      "step": 18785
    },
    {
      "entropy": 5.405790996551514,
      "epoch": 1.8049951969260327,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004677403073948139,
      "loss": 4.8884,
      "mean_token_accuracy": 0.2322069838643074,
      "num_tokens": 43059592.0,
      "step": 18790
    },
    {
      "entropy": 5.262194347381592,
      "epoch": 1.8054755043227666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004677226168352416,
      "loss": 4.6948,
      "mean_token_accuracy": 0.23749196529388428,
      "num_tokens": 43071755.0,
      "step": 18795
    },
    {
      "entropy": 5.367287063598633,
      "epoch": 1.8059558117195005,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004677049218012223,
      "loss": 4.8663,
      "mean_token_accuracy": 0.2293684095144272,
      "num_tokens": 43081801.0,
      "step": 18800
    },
    {
      "entropy": 5.299804162979126,
      "epoch": 1.8064361191162344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000467687222293167,
      "loss": 4.7684,
      "mean_token_accuracy": 0.24213323891162872,
      "num_tokens": 43094160.0,
      "step": 18805
    },
    {
      "entropy": 5.320735168457031,
      "epoch": 1.8069164265129682,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004676695183114866,
      "loss": 4.8234,
      "mean_token_accuracy": 0.23752613812685014,
      "num_tokens": 43106082.0,
      "step": 18810
    },
    {
      "entropy": 5.351252174377441,
      "epoch": 1.8073967339097021,
      "grad_norm": 1.0625,
      "learning_rate": 0.000467651809856592,
      "loss": 4.823,
      "mean_token_accuracy": 0.23210612535476685,
      "num_tokens": 43117215.0,
      "step": 18815
    },
    {
      "entropy": 5.424026155471802,
      "epoch": 1.807877041306436,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046763409692889446,
      "loss": 5.0004,
      "mean_token_accuracy": 0.2224865138530731,
      "num_tokens": 43127858.0,
      "step": 18820
    },
    {
      "entropy": 5.357612180709839,
      "epoch": 1.8083573487031699,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046761637952880516,
      "loss": 4.8097,
      "mean_token_accuracy": 0.23704468458890915,
      "num_tokens": 43139952.0,
      "step": 18825
    },
    {
      "entropy": 5.294938802719116,
      "epoch": 1.808837656099904,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046759865765673555,
      "loss": 4.7305,
      "mean_token_accuracy": 0.24506887942552566,
      "num_tokens": 43150900.0,
      "step": 18830
    },
    {
      "entropy": 5.272110939025879,
      "epoch": 1.8093179634966379,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000467580931313097,
      "loss": 4.7895,
      "mean_token_accuracy": 0.23697066456079482,
      "num_tokens": 43163199.0,
      "step": 18835
    },
    {
      "entropy": 5.323736572265625,
      "epoch": 1.8097982708933718,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046756320049830106,
      "loss": 4.7733,
      "mean_token_accuracy": 0.23210617005825043,
      "num_tokens": 43174147.0,
      "step": 18840
    },
    {
      "entropy": 5.27902626991272,
      "epoch": 1.8102785782901056,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004675454652127594,
      "loss": 4.7582,
      "mean_token_accuracy": 0.23874192982912062,
      "num_tokens": 43186191.0,
      "step": 18845
    },
    {
      "entropy": 5.373548221588135,
      "epoch": 1.8107588856868397,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046752772545688377,
      "loss": 4.9356,
      "mean_token_accuracy": 0.22663432210683823,
      "num_tokens": 43197541.0,
      "step": 18850
    },
    {
      "entropy": 5.411742258071899,
      "epoch": 1.8112391930835736,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004675099812310861,
      "loss": 4.8405,
      "mean_token_accuracy": 0.22984518408775328,
      "num_tokens": 43208792.0,
      "step": 18855
    },
    {
      "entropy": 5.360199022293091,
      "epoch": 1.8117195004803075,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004674922325357782,
      "loss": 4.9034,
      "mean_token_accuracy": 0.22692606896162032,
      "num_tokens": 43219914.0,
      "step": 18860
    },
    {
      "entropy": 5.231616830825805,
      "epoch": 1.8121998078770414,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046747447937137235,
      "loss": 4.7557,
      "mean_token_accuracy": 0.24353164285421372,
      "num_tokens": 43232179.0,
      "step": 18865
    },
    {
      "entropy": 5.315025758743286,
      "epoch": 1.8126801152737753,
      "grad_norm": 1.0,
      "learning_rate": 0.00046745672173828057,
      "loss": 4.7852,
      "mean_token_accuracy": 0.23169963657855988,
      "num_tokens": 43243196.0,
      "step": 18870
    },
    {
      "entropy": 5.425253963470459,
      "epoch": 1.8131604226705091,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004674389596369151,
      "loss": 4.8596,
      "mean_token_accuracy": 0.23150249272584916,
      "num_tokens": 43255573.0,
      "step": 18875
    },
    {
      "entropy": 5.32325930595398,
      "epoch": 1.813640730067243,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046742119306768855,
      "loss": 4.7998,
      "mean_token_accuracy": 0.23495101779699326,
      "num_tokens": 43267679.0,
      "step": 18880
    },
    {
      "entropy": 5.416063976287842,
      "epoch": 1.814121037463977,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004674034220310132,
      "loss": 4.9215,
      "mean_token_accuracy": 0.22412729114294053,
      "num_tokens": 43278857.0,
      "step": 18885
    },
    {
      "entropy": 5.369078159332275,
      "epoch": 1.8146013448607108,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046738564652730176,
      "loss": 4.8261,
      "mean_token_accuracy": 0.2393363445997238,
      "num_tokens": 43291099.0,
      "step": 18890
    },
    {
      "entropy": 5.355206394195557,
      "epoch": 1.8150816522574447,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004673678665569669,
      "loss": 4.8397,
      "mean_token_accuracy": 0.23117140382528306,
      "num_tokens": 43302863.0,
      "step": 18895
    },
    {
      "entropy": 5.316800546646118,
      "epoch": 1.8155619596541785,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004673500821204213,
      "loss": 4.7758,
      "mean_token_accuracy": 0.24110138416290283,
      "num_tokens": 43314673.0,
      "step": 18900
    },
    {
      "entropy": 5.315600204467773,
      "epoch": 1.8160422670509124,
      "grad_norm": 1.015625,
      "learning_rate": 0.000467332293218078,
      "loss": 4.774,
      "mean_token_accuracy": 0.2435930073261261,
      "num_tokens": 43325927.0,
      "step": 18905
    },
    {
      "entropy": 5.305320978164673,
      "epoch": 1.8165225744476465,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046731449985035,
      "loss": 4.8346,
      "mean_token_accuracy": 0.2365132138133049,
      "num_tokens": 43337615.0,
      "step": 18910
    },
    {
      "entropy": 5.386141729354859,
      "epoch": 1.8170028818443804,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046729670201765036,
      "loss": 4.8981,
      "mean_token_accuracy": 0.23362567722797395,
      "num_tokens": 43349195.0,
      "step": 18915
    },
    {
      "entropy": 5.413286590576172,
      "epoch": 1.8174831892411143,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046727889972039227,
      "loss": 4.8699,
      "mean_token_accuracy": 0.2347696140408516,
      "num_tokens": 43361274.0,
      "step": 18920
    },
    {
      "entropy": 5.306696796417237,
      "epoch": 1.8179634966378482,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046726109295898904,
      "loss": 4.8193,
      "mean_token_accuracy": 0.24112956523895263,
      "num_tokens": 43372843.0,
      "step": 18925
    },
    {
      "entropy": 5.337078619003296,
      "epoch": 1.8184438040345823,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004672432817338542,
      "loss": 4.7988,
      "mean_token_accuracy": 0.23738375753164292,
      "num_tokens": 43383492.0,
      "step": 18930
    },
    {
      "entropy": 5.340841913223267,
      "epoch": 1.8189241114313162,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046722546604540115,
      "loss": 4.8064,
      "mean_token_accuracy": 0.22923208475112916,
      "num_tokens": 43395669.0,
      "step": 18935
    },
    {
      "entropy": 5.264487266540527,
      "epoch": 1.81940441882805,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004672076458940436,
      "loss": 4.7132,
      "mean_token_accuracy": 0.23562943786382676,
      "num_tokens": 43407340.0,
      "step": 18940
    },
    {
      "entropy": 5.294338417053223,
      "epoch": 1.819884726224784,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046718982128019534,
      "loss": 4.8456,
      "mean_token_accuracy": 0.23393580615520476,
      "num_tokens": 43418607.0,
      "step": 18945
    },
    {
      "entropy": 5.4234520435333256,
      "epoch": 1.8203650336215178,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046717199220427003,
      "loss": 4.9496,
      "mean_token_accuracy": 0.23031118214130403,
      "num_tokens": 43429286.0,
      "step": 18950
    },
    {
      "entropy": 5.296292161941528,
      "epoch": 1.8208453410182517,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046715415866668163,
      "loss": 4.7935,
      "mean_token_accuracy": 0.23442893773317336,
      "num_tokens": 43440794.0,
      "step": 18955
    },
    {
      "entropy": 5.315075445175171,
      "epoch": 1.8213256484149856,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004671363206678443,
      "loss": 4.8486,
      "mean_token_accuracy": 0.23503599017858506,
      "num_tokens": 43452184.0,
      "step": 18960
    },
    {
      "entropy": 5.289494848251342,
      "epoch": 1.8218059558117194,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046711847820817215,
      "loss": 4.8004,
      "mean_token_accuracy": 0.24075008779764176,
      "num_tokens": 43463361.0,
      "step": 18965
    },
    {
      "entropy": 5.380526542663574,
      "epoch": 1.8222862632084533,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004671006312880794,
      "loss": 4.8358,
      "mean_token_accuracy": 0.23367957174777984,
      "num_tokens": 43474802.0,
      "step": 18970
    },
    {
      "entropy": 5.35435905456543,
      "epoch": 1.8227665706051872,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004670827799079805,
      "loss": 4.7938,
      "mean_token_accuracy": 0.23141216188669206,
      "num_tokens": 43486962.0,
      "step": 18975
    },
    {
      "entropy": 5.3372049808502195,
      "epoch": 1.823246878001921,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046706492406828966,
      "loss": 4.8164,
      "mean_token_accuracy": 0.23355931043624878,
      "num_tokens": 43498761.0,
      "step": 18980
    },
    {
      "entropy": 5.354840278625488,
      "epoch": 1.8237271853986552,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004670470637694217,
      "loss": 4.8818,
      "mean_token_accuracy": 0.23630741834640503,
      "num_tokens": 43511294.0,
      "step": 18985
    },
    {
      "entropy": 5.345191431045532,
      "epoch": 1.824207492795389,
      "grad_norm": 1.0,
      "learning_rate": 0.0004670291990117912,
      "loss": 4.7971,
      "mean_token_accuracy": 0.24137774258852004,
      "num_tokens": 43522858.0,
      "step": 18990
    },
    {
      "entropy": 5.354160261154175,
      "epoch": 1.824687800192123,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004670113297958128,
      "loss": 4.807,
      "mean_token_accuracy": 0.23499049693346025,
      "num_tokens": 43533453.0,
      "step": 18995
    },
    {
      "entropy": 5.356452417373657,
      "epoch": 1.8251681075888568,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046699345612190155,
      "loss": 4.7738,
      "mean_token_accuracy": 0.24402957856655122,
      "num_tokens": 43543889.0,
      "step": 19000
    },
    {
      "entropy": 5.283263301849365,
      "epoch": 1.825648414985591,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046697557799047233,
      "loss": 4.7735,
      "mean_token_accuracy": 0.24124203473329545,
      "num_tokens": 43554815.0,
      "step": 19005
    },
    {
      "entropy": 5.257811260223389,
      "epoch": 1.8261287223823248,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004669576954019403,
      "loss": 4.6974,
      "mean_token_accuracy": 0.2465030536055565,
      "num_tokens": 43565258.0,
      "step": 19010
    },
    {
      "entropy": 5.349430418014526,
      "epoch": 1.8266090297790587,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004669398083567205,
      "loss": 4.84,
      "mean_token_accuracy": 0.23269233703613282,
      "num_tokens": 43576710.0,
      "step": 19015
    },
    {
      "entropy": 5.3422716617584225,
      "epoch": 1.8270893371757926,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004669219168552284,
      "loss": 4.8151,
      "mean_token_accuracy": 0.23747183233499528,
      "num_tokens": 43587590.0,
      "step": 19020
    },
    {
      "entropy": 5.335068607330323,
      "epoch": 1.8275696445725265,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046690402089787916,
      "loss": 4.8292,
      "mean_token_accuracy": 0.23599403500556945,
      "num_tokens": 43599675.0,
      "step": 19025
    },
    {
      "entropy": 5.279744672775268,
      "epoch": 1.8280499519692603,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004668861204850884,
      "loss": 4.697,
      "mean_token_accuracy": 0.24680829495191575,
      "num_tokens": 43612369.0,
      "step": 19030
    },
    {
      "entropy": 5.298574304580688,
      "epoch": 1.8285302593659942,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046686821561727176,
      "loss": 4.7624,
      "mean_token_accuracy": 0.24757715463638305,
      "num_tokens": 43624807.0,
      "step": 19035
    },
    {
      "entropy": 5.280199575424194,
      "epoch": 1.829010566762728,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004668503062948449,
      "loss": 4.8309,
      "mean_token_accuracy": 0.2334493324160576,
      "num_tokens": 43635389.0,
      "step": 19040
    },
    {
      "entropy": 5.313683748245239,
      "epoch": 1.829490874159462,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004668323925182236,
      "loss": 4.8007,
      "mean_token_accuracy": 0.23520201593637466,
      "num_tokens": 43646807.0,
      "step": 19045
    },
    {
      "entropy": 5.371701383590699,
      "epoch": 1.8299711815561959,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046681447428782377,
      "loss": 4.8528,
      "mean_token_accuracy": 0.23070276379585267,
      "num_tokens": 43657910.0,
      "step": 19050
    },
    {
      "entropy": 5.318098497390747,
      "epoch": 1.8304514889529298,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004667965516040613,
      "loss": 4.8022,
      "mean_token_accuracy": 0.23870700895786284,
      "num_tokens": 43669376.0,
      "step": 19055
    },
    {
      "entropy": 5.296724605560303,
      "epoch": 1.8309317963496636,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004667786244673526,
      "loss": 4.7821,
      "mean_token_accuracy": 0.23811755627393721,
      "num_tokens": 43681702.0,
      "step": 19060
    },
    {
      "entropy": 5.381584644317627,
      "epoch": 1.8314121037463977,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046676069287811365,
      "loss": 4.8817,
      "mean_token_accuracy": 0.2378578394651413,
      "num_tokens": 43694795.0,
      "step": 19065
    },
    {
      "entropy": 5.4090433597564695,
      "epoch": 1.8318924111431316,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004667427568367607,
      "loss": 4.9009,
      "mean_token_accuracy": 0.23528937697410585,
      "num_tokens": 43707197.0,
      "step": 19070
    },
    {
      "entropy": 5.372711801528931,
      "epoch": 1.8323727185398655,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046672481634371047,
      "loss": 4.8903,
      "mean_token_accuracy": 0.22935521006584167,
      "num_tokens": 43719066.0,
      "step": 19075
    },
    {
      "entropy": 5.335252523422241,
      "epoch": 1.8328530259365994,
      "grad_norm": 1.0,
      "learning_rate": 0.00046670687139937925,
      "loss": 4.8024,
      "mean_token_accuracy": 0.23671803027391433,
      "num_tokens": 43731080.0,
      "step": 19080
    },
    {
      "entropy": 5.37404842376709,
      "epoch": 1.8333333333333335,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004666889220041837,
      "loss": 4.849,
      "mean_token_accuracy": 0.23210448920726776,
      "num_tokens": 43743048.0,
      "step": 19085
    },
    {
      "entropy": 5.313898086547852,
      "epoch": 1.8338136407300674,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046667096815854056,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23117799013853074,
      "num_tokens": 43754701.0,
      "step": 19090
    },
    {
      "entropy": 5.360203218460083,
      "epoch": 1.8342939481268012,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004666530098628667,
      "loss": 4.8365,
      "mean_token_accuracy": 0.23412107080221176,
      "num_tokens": 43765696.0,
      "step": 19095
    },
    {
      "entropy": 5.3765143871307375,
      "epoch": 1.8347742555235351,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004666350471175791,
      "loss": 4.9151,
      "mean_token_accuracy": 0.22697183936834336,
      "num_tokens": 43777162.0,
      "step": 19100
    },
    {
      "entropy": 5.365682935714721,
      "epoch": 1.835254562920269,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004666170799230947,
      "loss": 4.8302,
      "mean_token_accuracy": 0.23285940587520598,
      "num_tokens": 43789340.0,
      "step": 19105
    },
    {
      "entropy": 5.287034702301026,
      "epoch": 1.8357348703170029,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004665991082798307,
      "loss": 4.7285,
      "mean_token_accuracy": 0.24818031191825868,
      "num_tokens": 43800098.0,
      "step": 19110
    },
    {
      "entropy": 5.330290222167969,
      "epoch": 1.8362151777137368,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004665811321882043,
      "loss": 4.8022,
      "mean_token_accuracy": 0.23856159001588823,
      "num_tokens": 43811403.0,
      "step": 19115
    },
    {
      "entropy": 5.3786664485931395,
      "epoch": 1.8366954851104706,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046656315164863297,
      "loss": 4.8418,
      "mean_token_accuracy": 0.23901668190956116,
      "num_tokens": 43821946.0,
      "step": 19120
    },
    {
      "entropy": 5.26647629737854,
      "epoch": 1.8371757925072045,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046654516666153403,
      "loss": 4.7586,
      "mean_token_accuracy": 0.23568009883165358,
      "num_tokens": 43832638.0,
      "step": 19125
    },
    {
      "entropy": 5.295343112945557,
      "epoch": 1.8376560999039384,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004665271772273251,
      "loss": 4.7579,
      "mean_token_accuracy": 0.2399964988231659,
      "num_tokens": 43843616.0,
      "step": 19130
    },
    {
      "entropy": 5.371764755249023,
      "epoch": 1.8381364073006723,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004665091833464239,
      "loss": 4.7749,
      "mean_token_accuracy": 0.23898979276418686,
      "num_tokens": 43854680.0,
      "step": 19135
    },
    {
      "entropy": 5.288388013839722,
      "epoch": 1.8386167146974062,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046649118501924805,
      "loss": 4.7979,
      "mean_token_accuracy": 0.2386866569519043,
      "num_tokens": 43866683.0,
      "step": 19140
    },
    {
      "entropy": 5.364487409591675,
      "epoch": 1.8390970220941403,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004664731822462154,
      "loss": 4.8656,
      "mean_token_accuracy": 0.2338874876499176,
      "num_tokens": 43878080.0,
      "step": 19145
    },
    {
      "entropy": 5.368899822235107,
      "epoch": 1.8395773294908742,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046645517502774415,
      "loss": 4.823,
      "mean_token_accuracy": 0.2370073527097702,
      "num_tokens": 43888633.0,
      "step": 19150
    },
    {
      "entropy": 5.427195596694946,
      "epoch": 1.840057636887608,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046643716336425224,
      "loss": 4.8394,
      "mean_token_accuracy": 0.23397401571273804,
      "num_tokens": 43899678.0,
      "step": 19155
    },
    {
      "entropy": 5.316941642761231,
      "epoch": 1.8405379442843421,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004664191472561578,
      "loss": 4.8093,
      "mean_token_accuracy": 0.23949493318796158,
      "num_tokens": 43910744.0,
      "step": 19160
    },
    {
      "entropy": 5.339258909225464,
      "epoch": 1.841018251681076,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004664011267038792,
      "loss": 4.9255,
      "mean_token_accuracy": 0.2319337770342827,
      "num_tokens": 43922467.0,
      "step": 19165
    },
    {
      "entropy": 5.314611959457397,
      "epoch": 1.84149855907781,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046638310170783476,
      "loss": 4.7716,
      "mean_token_accuracy": 0.2406027615070343,
      "num_tokens": 43934919.0,
      "step": 19170
    },
    {
      "entropy": 5.349668359756469,
      "epoch": 1.8419788664745438,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000466365072268443,
      "loss": 4.8301,
      "mean_token_accuracy": 0.236470727622509,
      "num_tokens": 43947737.0,
      "step": 19175
    },
    {
      "entropy": 5.263504505157471,
      "epoch": 1.8424591738712777,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004663470383861225,
      "loss": 4.7632,
      "mean_token_accuracy": 0.23781696408987046,
      "num_tokens": 43959656.0,
      "step": 19180
    },
    {
      "entropy": 5.321942377090454,
      "epoch": 1.8429394812680115,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000466329000061292,
      "loss": 4.824,
      "mean_token_accuracy": 0.23539715260267258,
      "num_tokens": 43970732.0,
      "step": 19185
    },
    {
      "entropy": 5.292244005203247,
      "epoch": 1.8434197886647454,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004663109572943702,
      "loss": 4.7931,
      "mean_token_accuracy": 0.23052601516246796,
      "num_tokens": 43982369.0,
      "step": 19190
    },
    {
      "entropy": 5.397276210784912,
      "epoch": 1.8439000960614793,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000466292910085776,
      "loss": 4.8422,
      "mean_token_accuracy": 0.2315011739730835,
      "num_tokens": 43993751.0,
      "step": 19195
    },
    {
      "entropy": 5.290119886398315,
      "epoch": 1.8443804034582132,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046627485843592854,
      "loss": 4.7928,
      "mean_token_accuracy": 0.23402820974588395,
      "num_tokens": 44005021.0,
      "step": 19200
    },
    {
      "entropy": 5.258894443511963,
      "epoch": 1.844860710854947,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046625680234524674,
      "loss": 4.7801,
      "mean_token_accuracy": 0.23940657526254655,
      "num_tokens": 44015571.0,
      "step": 19205
    },
    {
      "entropy": 5.350990724563599,
      "epoch": 1.845341018251681,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046623874181414993,
      "loss": 4.8501,
      "mean_token_accuracy": 0.23670962899923326,
      "num_tokens": 44026363.0,
      "step": 19210
    },
    {
      "entropy": 5.286859846115112,
      "epoch": 1.8458213256484148,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004662206768430574,
      "loss": 4.7435,
      "mean_token_accuracy": 0.24639168679714202,
      "num_tokens": 44038042.0,
      "step": 19215
    },
    {
      "entropy": 5.373714780807495,
      "epoch": 1.846301633045149,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004662026074323886,
      "loss": 4.8318,
      "mean_token_accuracy": 0.2261994630098343,
      "num_tokens": 44049086.0,
      "step": 19220
    },
    {
      "entropy": 5.326414012908936,
      "epoch": 1.8467819404418828,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046618453358256303,
      "loss": 4.835,
      "mean_token_accuracy": 0.23724903911352158,
      "num_tokens": 44060816.0,
      "step": 19225
    },
    {
      "entropy": 5.263651180267334,
      "epoch": 1.8472622478386167,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046616645529400026,
      "loss": 4.7094,
      "mean_token_accuracy": 0.23911830186843872,
      "num_tokens": 44071877.0,
      "step": 19230
    },
    {
      "entropy": 5.369073152542114,
      "epoch": 1.8477425552353506,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004661483725671201,
      "loss": 4.8173,
      "mean_token_accuracy": 0.23581244200468063,
      "num_tokens": 44082618.0,
      "step": 19235
    },
    {
      "entropy": 5.302669858932495,
      "epoch": 1.8482228626320847,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046613028540234226,
      "loss": 4.769,
      "mean_token_accuracy": 0.2411525934934616,
      "num_tokens": 44093411.0,
      "step": 19240
    },
    {
      "entropy": 5.249357128143311,
      "epoch": 1.8487031700288186,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004661121938000867,
      "loss": 4.7008,
      "mean_token_accuracy": 0.24516435712575912,
      "num_tokens": 44103595.0,
      "step": 19245
    },
    {
      "entropy": 5.275901508331299,
      "epoch": 1.8491834774255524,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004660940977607736,
      "loss": 4.7892,
      "mean_token_accuracy": 0.24561985731124877,
      "num_tokens": 44114740.0,
      "step": 19250
    },
    {
      "entropy": 5.281003093719482,
      "epoch": 1.8496637848222863,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046607599728482285,
      "loss": 4.7466,
      "mean_token_accuracy": 0.2441767856478691,
      "num_tokens": 44125415.0,
      "step": 19255
    },
    {
      "entropy": 5.276584959030151,
      "epoch": 1.8501440922190202,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046605789237265496,
      "loss": 4.7867,
      "mean_token_accuracy": 0.23682679682970048,
      "num_tokens": 44137399.0,
      "step": 19260
    },
    {
      "entropy": 5.2770264625549315,
      "epoch": 1.850624399615754,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046603978302469,
      "loss": 4.7358,
      "mean_token_accuracy": 0.23698250502347945,
      "num_tokens": 44148354.0,
      "step": 19265
    },
    {
      "entropy": 5.268051719665527,
      "epoch": 1.851104707012488,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004660216692413486,
      "loss": 4.7635,
      "mean_token_accuracy": 0.24118833988904953,
      "num_tokens": 44160496.0,
      "step": 19270
    },
    {
      "entropy": 5.301895236968994,
      "epoch": 1.8515850144092219,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004660035510230513,
      "loss": 4.8487,
      "mean_token_accuracy": 0.23391092866659163,
      "num_tokens": 44171832.0,
      "step": 19275
    },
    {
      "entropy": 5.378113794326782,
      "epoch": 1.8520653218059557,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004659854283702186,
      "loss": 4.805,
      "mean_token_accuracy": 0.24087000489234925,
      "num_tokens": 44182850.0,
      "step": 19280
    },
    {
      "entropy": 5.339982891082764,
      "epoch": 1.8525456292026896,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004659673012832715,
      "loss": 4.8339,
      "mean_token_accuracy": 0.23404190838336944,
      "num_tokens": 44194343.0,
      "step": 19285
    },
    {
      "entropy": 5.3670577049255375,
      "epoch": 1.8530259365994235,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004659491697626306,
      "loss": 4.9124,
      "mean_token_accuracy": 0.22554775476455688,
      "num_tokens": 44206294.0,
      "step": 19290
    },
    {
      "entropy": 5.324203968048096,
      "epoch": 1.8535062439961574,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00046593103380871705,
      "loss": 4.7966,
      "mean_token_accuracy": 0.23547422289848327,
      "num_tokens": 44218576.0,
      "step": 19295
    },
    {
      "entropy": 5.325460481643677,
      "epoch": 1.8539865513928915,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046591289342195184,
      "loss": 4.8177,
      "mean_token_accuracy": 0.23303375542163848,
      "num_tokens": 44230480.0,
      "step": 19300
    },
    {
      "entropy": 5.2933878898620605,
      "epoch": 1.8544668587896254,
      "grad_norm": 1.0,
      "learning_rate": 0.0004658947486027562,
      "loss": 4.7132,
      "mean_token_accuracy": 0.2437806636095047,
      "num_tokens": 44240907.0,
      "step": 19305
    },
    {
      "entropy": 5.355522155761719,
      "epoch": 1.8549471661863592,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046587659935155124,
      "loss": 4.9118,
      "mean_token_accuracy": 0.2282836005091667,
      "num_tokens": 44251469.0,
      "step": 19310
    },
    {
      "entropy": 5.399255418777466,
      "epoch": 1.8554274735830933,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046585844566875845,
      "loss": 4.8989,
      "mean_token_accuracy": 0.2344025894999504,
      "num_tokens": 44263875.0,
      "step": 19315
    },
    {
      "entropy": 5.385026693344116,
      "epoch": 1.8559077809798272,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004658402875547993,
      "loss": 4.8896,
      "mean_token_accuracy": 0.23391070514917373,
      "num_tokens": 44276163.0,
      "step": 19320
    },
    {
      "entropy": 5.308988618850708,
      "epoch": 1.856388088376561,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046582212501009533,
      "loss": 4.8217,
      "mean_token_accuracy": 0.2344052568078041,
      "num_tokens": 44287985.0,
      "step": 19325
    },
    {
      "entropy": 5.357990026473999,
      "epoch": 1.856868395773295,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046580395803506825,
      "loss": 4.9079,
      "mean_token_accuracy": 0.23311863988637924,
      "num_tokens": 44299360.0,
      "step": 19330
    },
    {
      "entropy": 5.359539890289307,
      "epoch": 1.8573487031700289,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004657857866301399,
      "loss": 4.7925,
      "mean_token_accuracy": 0.24603033512830735,
      "num_tokens": 44310459.0,
      "step": 19335
    },
    {
      "entropy": 5.427111721038818,
      "epoch": 1.8578290105667628,
      "grad_norm": 2.625,
      "learning_rate": 0.00046576761079573204,
      "loss": 4.923,
      "mean_token_accuracy": 0.22831496000289916,
      "num_tokens": 44321832.0,
      "step": 19340
    },
    {
      "entropy": 5.350720834732056,
      "epoch": 1.8583093179634966,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004657494305322667,
      "loss": 4.897,
      "mean_token_accuracy": 0.22792275995016098,
      "num_tokens": 44332829.0,
      "step": 19345
    },
    {
      "entropy": 5.335819673538208,
      "epoch": 1.8587896253602305,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004657312458401661,
      "loss": 4.7853,
      "mean_token_accuracy": 0.23219239413738252,
      "num_tokens": 44344321.0,
      "step": 19350
    },
    {
      "entropy": 5.452392530441284,
      "epoch": 1.8592699327569644,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004657130567198522,
      "loss": 4.8764,
      "mean_token_accuracy": 0.22953421026468276,
      "num_tokens": 44357183.0,
      "step": 19355
    },
    {
      "entropy": 5.428354167938233,
      "epoch": 1.8597502401536983,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046569486317174746,
      "loss": 4.8992,
      "mean_token_accuracy": 0.23140558153390883,
      "num_tokens": 44367118.0,
      "step": 19360
    },
    {
      "entropy": 5.344645023345947,
      "epoch": 1.8602305475504322,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004656766651962742,
      "loss": 4.885,
      "mean_token_accuracy": 0.23166532218456268,
      "num_tokens": 44377699.0,
      "step": 19365
    },
    {
      "entropy": 5.330774259567261,
      "epoch": 1.860710854947166,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004656584627938551,
      "loss": 4.8285,
      "mean_token_accuracy": 0.23279052674770356,
      "num_tokens": 44389457.0,
      "step": 19370
    },
    {
      "entropy": 5.3003928661346436,
      "epoch": 1.8611911623439001,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046564025596491254,
      "loss": 4.7605,
      "mean_token_accuracy": 0.24009636491537095,
      "num_tokens": 44400140.0,
      "step": 19375
    },
    {
      "entropy": 5.280788707733154,
      "epoch": 1.861671469740634,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004656220447098693,
      "loss": 4.7431,
      "mean_token_accuracy": 0.24891536980867385,
      "num_tokens": 44411088.0,
      "step": 19380
    },
    {
      "entropy": 5.326315927505493,
      "epoch": 1.862151777137368,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004656038290291483,
      "loss": 4.7578,
      "mean_token_accuracy": 0.23753364980220795,
      "num_tokens": 44422515.0,
      "step": 19385
    },
    {
      "entropy": 5.289603042602539,
      "epoch": 1.8626320845341018,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004655856089231723,
      "loss": 4.8118,
      "mean_token_accuracy": 0.24016451537609101,
      "num_tokens": 44433711.0,
      "step": 19390
    },
    {
      "entropy": 5.338762998580933,
      "epoch": 1.8631123919308359,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004655673843923644,
      "loss": 4.855,
      "mean_token_accuracy": 0.23614740669727324,
      "num_tokens": 44446157.0,
      "step": 19395
    },
    {
      "entropy": 5.448622798919677,
      "epoch": 1.8635926993275698,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004655491554371477,
      "loss": 4.9414,
      "mean_token_accuracy": 0.22583320289850234,
      "num_tokens": 44457506.0,
      "step": 19400
    },
    {
      "entropy": 5.356912755966187,
      "epoch": 1.8640730067243036,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046553092205794543,
      "loss": 4.9401,
      "mean_token_accuracy": 0.23019564002752305,
      "num_tokens": 44469572.0,
      "step": 19405
    },
    {
      "entropy": 5.356427764892578,
      "epoch": 1.8645533141210375,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046551268425518096,
      "loss": 4.8577,
      "mean_token_accuracy": 0.23300497978925705,
      "num_tokens": 44481932.0,
      "step": 19410
    },
    {
      "entropy": 5.312144088745117,
      "epoch": 1.8650336215177714,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004654944420292776,
      "loss": 4.7698,
      "mean_token_accuracy": 0.2370797023177147,
      "num_tokens": 44493081.0,
      "step": 19415
    },
    {
      "entropy": 5.3543023586273195,
      "epoch": 1.8655139289145053,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000465476195380659,
      "loss": 4.817,
      "mean_token_accuracy": 0.2373598948121071,
      "num_tokens": 44503919.0,
      "step": 19420
    },
    {
      "entropy": 5.337809324264526,
      "epoch": 1.8659942363112392,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004654579443097487,
      "loss": 4.7962,
      "mean_token_accuracy": 0.24034407138824462,
      "num_tokens": 44515493.0,
      "step": 19425
    },
    {
      "entropy": 5.389287900924683,
      "epoch": 1.866474543707973,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004654396888169705,
      "loss": 4.8517,
      "mean_token_accuracy": 0.22846542447805404,
      "num_tokens": 44527432.0,
      "step": 19430
    },
    {
      "entropy": 5.3036237239837645,
      "epoch": 1.866954851104707,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046542142890274816,
      "loss": 4.8082,
      "mean_token_accuracy": 0.23790030777454377,
      "num_tokens": 44537634.0,
      "step": 19435
    },
    {
      "entropy": 5.265326404571534,
      "epoch": 1.8674351585014408,
      "grad_norm": 1.0,
      "learning_rate": 0.0004654031645675057,
      "loss": 4.863,
      "mean_token_accuracy": 0.2319808691740036,
      "num_tokens": 44549856.0,
      "step": 19440
    },
    {
      "entropy": 5.269454050064087,
      "epoch": 1.8679154658981747,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004653848958116672,
      "loss": 4.7637,
      "mean_token_accuracy": 0.23406877666711806,
      "num_tokens": 44563214.0,
      "step": 19445
    },
    {
      "entropy": 5.3021434307098385,
      "epoch": 1.8683957732949086,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046536662263565667,
      "loss": 4.7916,
      "mean_token_accuracy": 0.22978847622871398,
      "num_tokens": 44572982.0,
      "step": 19450
    },
    {
      "entropy": 5.308111715316772,
      "epoch": 1.8688760806916427,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004653483450398985,
      "loss": 4.7566,
      "mean_token_accuracy": 0.23778516203165054,
      "num_tokens": 44583664.0,
      "step": 19455
    },
    {
      "entropy": 5.263463401794434,
      "epoch": 1.8693563880883766,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046533006302481694,
      "loss": 4.6628,
      "mean_token_accuracy": 0.2494708850979805,
      "num_tokens": 44594987.0,
      "step": 19460
    },
    {
      "entropy": 5.303329801559448,
      "epoch": 1.8698366954851104,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004653117765908365,
      "loss": 4.848,
      "mean_token_accuracy": 0.23907629400491714,
      "num_tokens": 44607123.0,
      "step": 19465
    },
    {
      "entropy": 5.310296630859375,
      "epoch": 1.8703170028818443,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004652934857383816,
      "loss": 4.753,
      "mean_token_accuracy": 0.23591097444295883,
      "num_tokens": 44618736.0,
      "step": 19470
    },
    {
      "entropy": 5.31084361076355,
      "epoch": 1.8707973102785784,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004652751904678772,
      "loss": 4.7445,
      "mean_token_accuracy": 0.24765047281980515,
      "num_tokens": 44630681.0,
      "step": 19475
    },
    {
      "entropy": 5.309070348739624,
      "epoch": 1.8712776176753123,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046525689077974775,
      "loss": 4.7707,
      "mean_token_accuracy": 0.2449563130736351,
      "num_tokens": 44641641.0,
      "step": 19480
    },
    {
      "entropy": 5.2390765190124515,
      "epoch": 1.8717579250720462,
      "grad_norm": 1.0,
      "learning_rate": 0.00046523858667441834,
      "loss": 4.7756,
      "mean_token_accuracy": 0.23614326417446135,
      "num_tokens": 44652489.0,
      "step": 19485
    },
    {
      "entropy": 5.369136571884155,
      "epoch": 1.87223823246878,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004652202781523138,
      "loss": 4.841,
      "mean_token_accuracy": 0.23318687379360198,
      "num_tokens": 44665659.0,
      "step": 19490
    },
    {
      "entropy": 5.335555982589722,
      "epoch": 1.872718539865514,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004652019652138592,
      "loss": 4.7563,
      "mean_token_accuracy": 0.24647001475095748,
      "num_tokens": 44677300.0,
      "step": 19495
    },
    {
      "entropy": 5.307165956497192,
      "epoch": 1.8731988472622478,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004651836478594798,
      "loss": 4.8236,
      "mean_token_accuracy": 0.2332539439201355,
      "num_tokens": 44690084.0,
      "step": 19500
    },
    {
      "entropy": 5.322757863998413,
      "epoch": 1.8736791546589817,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004651653260896008,
      "loss": 4.8246,
      "mean_token_accuracy": 0.24637349098920822,
      "num_tokens": 44701208.0,
      "step": 19505
    },
    {
      "entropy": 5.4297788619995115,
      "epoch": 1.8741594620557156,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00046514699990464763,
      "loss": 4.9322,
      "mean_token_accuracy": 0.21855489015579224,
      "num_tokens": 44713546.0,
      "step": 19510
    },
    {
      "entropy": 5.538158273696899,
      "epoch": 1.8746397694524495,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004651286693050458,
      "loss": 5.0286,
      "mean_token_accuracy": 0.21638088822364807,
      "num_tokens": 44724802.0,
      "step": 19515
    },
    {
      "entropy": 5.341595983505249,
      "epoch": 1.8751200768491834,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004651103342912207,
      "loss": 4.8136,
      "mean_token_accuracy": 0.23857346624135972,
      "num_tokens": 44736625.0,
      "step": 19520
    },
    {
      "entropy": 5.284909963607788,
      "epoch": 1.8756003842459172,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046509199486359824,
      "loss": 4.757,
      "mean_token_accuracy": 0.22877490371465684,
      "num_tokens": 44750451.0,
      "step": 19525
    },
    {
      "entropy": 5.276770210266113,
      "epoch": 1.8760806916426513,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046507365102260403,
      "loss": 4.8051,
      "mean_token_accuracy": 0.23690056502819062,
      "num_tokens": 44762043.0,
      "step": 19530
    },
    {
      "entropy": 5.366952180862427,
      "epoch": 1.8765609990393852,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046505530276866417,
      "loss": 4.8701,
      "mean_token_accuracy": 0.23183054625988006,
      "num_tokens": 44774329.0,
      "step": 19535
    },
    {
      "entropy": 5.409959602355957,
      "epoch": 1.877041306436119,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046503695010220443,
      "loss": 4.8468,
      "mean_token_accuracy": 0.23679649084806442,
      "num_tokens": 44786352.0,
      "step": 19540
    },
    {
      "entropy": 5.2541663646698,
      "epoch": 1.877521613832853,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000465018593023651,
      "loss": 4.6771,
      "mean_token_accuracy": 0.2470657393336296,
      "num_tokens": 44796829.0,
      "step": 19545
    },
    {
      "entropy": 5.361844825744629,
      "epoch": 1.878001921229587,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004650002315334302,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22766576409339906,
      "num_tokens": 44808900.0,
      "step": 19550
    },
    {
      "entropy": 5.292666101455689,
      "epoch": 1.878482228626321,
      "grad_norm": 1.125,
      "learning_rate": 0.000464981865631968,
      "loss": 4.8063,
      "mean_token_accuracy": 0.24270494282245636,
      "num_tokens": 44820638.0,
      "step": 19555
    },
    {
      "entropy": 5.288909912109375,
      "epoch": 1.8789625360230549,
      "grad_norm": 1.0,
      "learning_rate": 0.0004649634953196912,
      "loss": 4.7779,
      "mean_token_accuracy": 0.2384965419769287,
      "num_tokens": 44833370.0,
      "step": 19560
    },
    {
      "entropy": 5.310856723785401,
      "epoch": 1.8794428434197887,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046494512059702605,
      "loss": 4.832,
      "mean_token_accuracy": 0.23839040100574493,
      "num_tokens": 44844291.0,
      "step": 19565
    },
    {
      "entropy": 5.299627161026001,
      "epoch": 1.8799231508165226,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004649267414643992,
      "loss": 4.7679,
      "mean_token_accuracy": 0.23895736634731293,
      "num_tokens": 44855303.0,
      "step": 19570
    },
    {
      "entropy": 5.2506927967071535,
      "epoch": 1.8804034582132565,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004649083579222374,
      "loss": 4.73,
      "mean_token_accuracy": 0.23802496641874313,
      "num_tokens": 44866593.0,
      "step": 19575
    },
    {
      "entropy": 5.2946474075317385,
      "epoch": 1.8808837656099904,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046488996997096744,
      "loss": 4.8133,
      "mean_token_accuracy": 0.23758843839168547,
      "num_tokens": 44879124.0,
      "step": 19580
    },
    {
      "entropy": 5.296995973587036,
      "epoch": 1.8813640730067243,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004648715776110162,
      "loss": 4.7744,
      "mean_token_accuracy": 0.25116845965385437,
      "num_tokens": 44889942.0,
      "step": 19585
    },
    {
      "entropy": 5.27787594795227,
      "epoch": 1.8818443804034581,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004648531808428108,
      "loss": 4.7412,
      "mean_token_accuracy": 0.24343717098236084,
      "num_tokens": 44901002.0,
      "step": 19590
    },
    {
      "entropy": 5.314434623718261,
      "epoch": 1.882324687800192,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004648347796667782,
      "loss": 4.8792,
      "mean_token_accuracy": 0.2295316979289055,
      "num_tokens": 44913495.0,
      "step": 19595
    },
    {
      "entropy": 5.332977151870727,
      "epoch": 1.882804995196926,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004648163740833458,
      "loss": 4.7802,
      "mean_token_accuracy": 0.2315352201461792,
      "num_tokens": 44924624.0,
      "step": 19600
    },
    {
      "entropy": 5.349265289306641,
      "epoch": 1.8832853025936598,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046479796409294076,
      "loss": 4.8512,
      "mean_token_accuracy": 0.23954532593488692,
      "num_tokens": 44935539.0,
      "step": 19605
    },
    {
      "entropy": 5.313002252578736,
      "epoch": 1.8837656099903939,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004647795496959907,
      "loss": 4.7753,
      "mean_token_accuracy": 0.24047466069459916,
      "num_tokens": 44945979.0,
      "step": 19610
    },
    {
      "entropy": 5.334032249450684,
      "epoch": 1.8842459173871278,
      "grad_norm": 0.9375,
      "learning_rate": 0.00046476113089292286,
      "loss": 4.8457,
      "mean_token_accuracy": 0.23045680522918702,
      "num_tokens": 44958294.0,
      "step": 19615
    },
    {
      "entropy": 5.400491428375244,
      "epoch": 1.8847262247838616,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004647427076841651,
      "loss": 4.9162,
      "mean_token_accuracy": 0.23163335025310516,
      "num_tokens": 44970261.0,
      "step": 19620
    },
    {
      "entropy": 5.338990783691406,
      "epoch": 1.8852065321805955,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046472428007014515,
      "loss": 4.8294,
      "mean_token_accuracy": 0.22968354523181916,
      "num_tokens": 44981515.0,
      "step": 19625
    },
    {
      "entropy": 5.259257745742798,
      "epoch": 1.8856868395773296,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004647058480512907,
      "loss": 4.7271,
      "mean_token_accuracy": 0.24266520738601685,
      "num_tokens": 44992473.0,
      "step": 19630
    },
    {
      "entropy": 5.380437421798706,
      "epoch": 1.8861671469740635,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046468741162802987,
      "loss": 4.8568,
      "mean_token_accuracy": 0.23369975835084916,
      "num_tokens": 45004264.0,
      "step": 19635
    },
    {
      "entropy": 5.3055156707763675,
      "epoch": 1.8866474543707974,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004646689708007905,
      "loss": 4.7395,
      "mean_token_accuracy": 0.24595486372709274,
      "num_tokens": 45015601.0,
      "step": 19640
    },
    {
      "entropy": 5.36806354522705,
      "epoch": 1.8871277617675313,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046465052557000087,
      "loss": 4.9054,
      "mean_token_accuracy": 0.2364653930068016,
      "num_tokens": 45028044.0,
      "step": 19645
    },
    {
      "entropy": 5.39833083152771,
      "epoch": 1.8876080691642652,
      "grad_norm": 1.78125,
      "learning_rate": 0.00046463207593608916,
      "loss": 4.8932,
      "mean_token_accuracy": 0.2369382455945015,
      "num_tokens": 45040144.0,
      "step": 19650
    },
    {
      "entropy": 5.461945295333862,
      "epoch": 1.888088376560999,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004646136218994837,
      "loss": 4.9222,
      "mean_token_accuracy": 0.22617594599723817,
      "num_tokens": 45051844.0,
      "step": 19655
    },
    {
      "entropy": 5.3864953994750975,
      "epoch": 1.888568683957733,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046459516346061304,
      "loss": 4.8863,
      "mean_token_accuracy": 0.23207226693630217,
      "num_tokens": 45062684.0,
      "step": 19660
    },
    {
      "entropy": 5.321763181686402,
      "epoch": 1.8890489913544668,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046457670061990564,
      "loss": 4.8144,
      "mean_token_accuracy": 0.23836594671010972,
      "num_tokens": 45073327.0,
      "step": 19665
    },
    {
      "entropy": 5.341830253601074,
      "epoch": 1.8895292987512007,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046455823337779024,
      "loss": 4.8258,
      "mean_token_accuracy": 0.23639376908540727,
      "num_tokens": 45085309.0,
      "step": 19670
    },
    {
      "entropy": 5.399540185928345,
      "epoch": 1.8900096061479346,
      "grad_norm": 1.0,
      "learning_rate": 0.0004645397617346954,
      "loss": 4.8859,
      "mean_token_accuracy": 0.22966315150260924,
      "num_tokens": 45095934.0,
      "step": 19675
    },
    {
      "entropy": 5.3351686000823975,
      "epoch": 1.8904899135446684,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004645212856910502,
      "loss": 4.8083,
      "mean_token_accuracy": 0.2418196514248848,
      "num_tokens": 45106220.0,
      "step": 19680
    },
    {
      "entropy": 5.329040479660034,
      "epoch": 1.8909702209414025,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004645028052472835,
      "loss": 4.7678,
      "mean_token_accuracy": 0.2370263695716858,
      "num_tokens": 45117861.0,
      "step": 19685
    },
    {
      "entropy": 5.2990975856781,
      "epoch": 1.8914505283381364,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046448432040382444,
      "loss": 4.7694,
      "mean_token_accuracy": 0.24930208921432495,
      "num_tokens": 45129248.0,
      "step": 19690
    },
    {
      "entropy": 5.387988805770874,
      "epoch": 1.8919308357348703,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000464465831161102,
      "loss": 4.8171,
      "mean_token_accuracy": 0.23184303492307662,
      "num_tokens": 45140610.0,
      "step": 19695
    },
    {
      "entropy": 5.315452957153321,
      "epoch": 1.8924111431316042,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004644473375195456,
      "loss": 4.7576,
      "mean_token_accuracy": 0.24340281784534454,
      "num_tokens": 45150697.0,
      "step": 19700
    },
    {
      "entropy": 5.379933786392212,
      "epoch": 1.8928914505283383,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046442883947958466,
      "loss": 4.8239,
      "mean_token_accuracy": 0.23173193037509918,
      "num_tokens": 45162627.0,
      "step": 19705
    },
    {
      "entropy": 5.252563190460205,
      "epoch": 1.8933717579250722,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046441033704164845,
      "loss": 4.6498,
      "mean_token_accuracy": 0.2532480522990227,
      "num_tokens": 45173030.0,
      "step": 19710
    },
    {
      "entropy": 5.351161861419678,
      "epoch": 1.893852065321806,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004643918302061666,
      "loss": 4.8309,
      "mean_token_accuracy": 0.2346762239933014,
      "num_tokens": 45183588.0,
      "step": 19715
    },
    {
      "entropy": 5.3788610935211185,
      "epoch": 1.89433237271854,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004643733189735689,
      "loss": 4.8344,
      "mean_token_accuracy": 0.2349509745836258,
      "num_tokens": 45195289.0,
      "step": 19720
    },
    {
      "entropy": 5.3375091552734375,
      "epoch": 1.8948126801152738,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000464354803344285,
      "loss": 4.8008,
      "mean_token_accuracy": 0.23909063786268234,
      "num_tokens": 45207001.0,
      "step": 19725
    },
    {
      "entropy": 5.306853866577148,
      "epoch": 1.8952929875120077,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046433628331874496,
      "loss": 4.8248,
      "mean_token_accuracy": 0.23974661529064178,
      "num_tokens": 45217447.0,
      "step": 19730
    },
    {
      "entropy": 5.363127136230469,
      "epoch": 1.8957732949087416,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004643177588973785,
      "loss": 4.79,
      "mean_token_accuracy": 0.2330551639199257,
      "num_tokens": 45227588.0,
      "step": 19735
    },
    {
      "entropy": 5.336324548721313,
      "epoch": 1.8962536023054755,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004642992300806159,
      "loss": 4.7994,
      "mean_token_accuracy": 0.24114888310432434,
      "num_tokens": 45239969.0,
      "step": 19740
    },
    {
      "entropy": 5.348459005355835,
      "epoch": 1.8967339097022093,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004642806968688873,
      "loss": 4.7889,
      "mean_token_accuracy": 0.23573863208293916,
      "num_tokens": 45250877.0,
      "step": 19745
    },
    {
      "entropy": 5.397932052612305,
      "epoch": 1.8972142170989432,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046426215926262295,
      "loss": 4.8505,
      "mean_token_accuracy": 0.2393438547849655,
      "num_tokens": 45262846.0,
      "step": 19750
    },
    {
      "entropy": 5.282821321487427,
      "epoch": 1.897694524495677,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004642436172622532,
      "loss": 4.8009,
      "mean_token_accuracy": 0.23859634697437287,
      "num_tokens": 45274198.0,
      "step": 19755
    },
    {
      "entropy": 5.262491035461426,
      "epoch": 1.898174831892411,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004642250708682086,
      "loss": 4.8042,
      "mean_token_accuracy": 0.2422293245792389,
      "num_tokens": 45285803.0,
      "step": 19760
    },
    {
      "entropy": 5.370756006240844,
      "epoch": 1.898655139289145,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046420652008091984,
      "loss": 4.8245,
      "mean_token_accuracy": 0.2318570613861084,
      "num_tokens": 45298242.0,
      "step": 19765
    },
    {
      "entropy": 5.3204974174499515,
      "epoch": 1.899135446685879,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004641879649008174,
      "loss": 4.7881,
      "mean_token_accuracy": 0.23800161778926848,
      "num_tokens": 45309567.0,
      "step": 19770
    },
    {
      "entropy": 5.374288892745971,
      "epoch": 1.8996157540826129,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004641694053283323,
      "loss": 4.8992,
      "mean_token_accuracy": 0.23522718697786332,
      "num_tokens": 45321693.0,
      "step": 19775
    },
    {
      "entropy": 5.4371476650238035,
      "epoch": 1.9000960614793467,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046415084136389525,
      "loss": 4.855,
      "mean_token_accuracy": 0.23592697978019714,
      "num_tokens": 45333094.0,
      "step": 19780
    },
    {
      "entropy": 5.329966640472412,
      "epoch": 1.9005763688760808,
      "grad_norm": 1.0,
      "learning_rate": 0.0004641322730079374,
      "loss": 4.7912,
      "mean_token_accuracy": 0.242096446454525,
      "num_tokens": 45344549.0,
      "step": 19785
    },
    {
      "entropy": 5.363551807403565,
      "epoch": 1.9010566762728147,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004641137002608897,
      "loss": 4.8627,
      "mean_token_accuracy": 0.22879418432712556,
      "num_tokens": 45356433.0,
      "step": 19790
    },
    {
      "entropy": 5.344631719589233,
      "epoch": 1.9015369836695486,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046409512312318345,
      "loss": 4.8006,
      "mean_token_accuracy": 0.2370127335190773,
      "num_tokens": 45368224.0,
      "step": 19795
    },
    {
      "entropy": 5.233707332611084,
      "epoch": 1.9020172910662825,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046407654159524994,
      "loss": 4.7288,
      "mean_token_accuracy": 0.24250718504190444,
      "num_tokens": 45379673.0,
      "step": 19800
    },
    {
      "entropy": 5.394478797912598,
      "epoch": 1.9024975984630164,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046405795567752055,
      "loss": 4.8603,
      "mean_token_accuracy": 0.2359165906906128,
      "num_tokens": 45391613.0,
      "step": 19805
    },
    {
      "entropy": 5.378146028518676,
      "epoch": 1.9029779058597502,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046403936537042686,
      "loss": 4.8159,
      "mean_token_accuracy": 0.23334169387817383,
      "num_tokens": 45403693.0,
      "step": 19810
    },
    {
      "entropy": 5.337565088272095,
      "epoch": 1.9034582132564841,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046402077067440043,
      "loss": 4.8482,
      "mean_token_accuracy": 0.23233731240034103,
      "num_tokens": 45416151.0,
      "step": 19815
    },
    {
      "entropy": 5.340458679199219,
      "epoch": 1.903938520653218,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046400217158987293,
      "loss": 4.8693,
      "mean_token_accuracy": 0.22975389659404755,
      "num_tokens": 45428276.0,
      "step": 19820
    },
    {
      "entropy": 5.284849405288696,
      "epoch": 1.9044188280499519,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046398356811727626,
      "loss": 4.7423,
      "mean_token_accuracy": 0.24498820900917054,
      "num_tokens": 45439877.0,
      "step": 19825
    },
    {
      "entropy": 5.335544204711914,
      "epoch": 1.9048991354466858,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004639649602570423,
      "loss": 4.8608,
      "mean_token_accuracy": 0.24109353870153427,
      "num_tokens": 45451459.0,
      "step": 19830
    },
    {
      "entropy": 5.26503324508667,
      "epoch": 1.9053794428434196,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046394634800960314,
      "loss": 4.7025,
      "mean_token_accuracy": 0.23932120352983474,
      "num_tokens": 45463341.0,
      "step": 19835
    },
    {
      "entropy": 5.391911315917969,
      "epoch": 1.9058597502401537,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046392773137539074,
      "loss": 4.8112,
      "mean_token_accuracy": 0.23890299797058107,
      "num_tokens": 45475585.0,
      "step": 19840
    },
    {
      "entropy": 5.365194511413574,
      "epoch": 1.9063400576368876,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046390911035483744,
      "loss": 4.7988,
      "mean_token_accuracy": 0.2359558254480362,
      "num_tokens": 45487178.0,
      "step": 19845
    },
    {
      "entropy": 5.3801226139068605,
      "epoch": 1.9068203650336215,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004638904849483756,
      "loss": 4.8795,
      "mean_token_accuracy": 0.23579144924879075,
      "num_tokens": 45498832.0,
      "step": 19850
    },
    {
      "entropy": 5.341107988357544,
      "epoch": 1.9073006724303554,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046387185515643756,
      "loss": 4.8578,
      "mean_token_accuracy": 0.23463883697986604,
      "num_tokens": 45511549.0,
      "step": 19855
    },
    {
      "entropy": 5.364961767196656,
      "epoch": 1.9077809798270895,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004638532209794559,
      "loss": 4.849,
      "mean_token_accuracy": 0.2324466437101364,
      "num_tokens": 45523421.0,
      "step": 19860
    },
    {
      "entropy": 5.386903572082519,
      "epoch": 1.9082612872238234,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004638345824178631,
      "loss": 4.8462,
      "mean_token_accuracy": 0.23422902822494507,
      "num_tokens": 45535247.0,
      "step": 19865
    },
    {
      "entropy": 5.293197202682495,
      "epoch": 1.9087415946205573,
      "grad_norm": 1.125,
      "learning_rate": 0.00046381593947209215,
      "loss": 4.7063,
      "mean_token_accuracy": 0.24662705659866332,
      "num_tokens": 45546296.0,
      "step": 19870
    },
    {
      "entropy": 5.341130352020263,
      "epoch": 1.9092219020172911,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004637972921425757,
      "loss": 4.873,
      "mean_token_accuracy": 0.2299395829439163,
      "num_tokens": 45557360.0,
      "step": 19875
    },
    {
      "entropy": 5.340062475204467,
      "epoch": 1.909702209414025,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046377864042974675,
      "loss": 4.8077,
      "mean_token_accuracy": 0.24167460054159165,
      "num_tokens": 45568500.0,
      "step": 19880
    },
    {
      "entropy": 5.393922472000122,
      "epoch": 1.910182516810759,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004637599843340384,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22844699174165725,
      "num_tokens": 45579513.0,
      "step": 19885
    },
    {
      "entropy": 5.345752859115601,
      "epoch": 1.9106628242074928,
      "grad_norm": 1.0,
      "learning_rate": 0.00046374132385588356,
      "loss": 4.7793,
      "mean_token_accuracy": 0.23754032850265502,
      "num_tokens": 45591144.0,
      "step": 19890
    },
    {
      "entropy": 5.366381883621216,
      "epoch": 1.9111431316042267,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046372265899571576,
      "loss": 4.8447,
      "mean_token_accuracy": 0.23262392729520798,
      "num_tokens": 45602487.0,
      "step": 19895
    },
    {
      "entropy": 5.442900562286377,
      "epoch": 1.9116234390009605,
      "grad_norm": 1.0,
      "learning_rate": 0.00046370398975396817,
      "loss": 4.8736,
      "mean_token_accuracy": 0.2273546874523163,
      "num_tokens": 45614448.0,
      "step": 19900
    },
    {
      "entropy": 5.369155740737915,
      "epoch": 1.9121037463976944,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004636853161310743,
      "loss": 4.8239,
      "mean_token_accuracy": 0.23091223537921907,
      "num_tokens": 45625660.0,
      "step": 19905
    },
    {
      "entropy": 5.25907711982727,
      "epoch": 1.9125840537944283,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046366663812746764,
      "loss": 4.7249,
      "mean_token_accuracy": 0.24713721126317978,
      "num_tokens": 45638602.0,
      "step": 19910
    },
    {
      "entropy": 5.3136687755584715,
      "epoch": 1.9130643611911622,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004636479557435818,
      "loss": 4.7214,
      "mean_token_accuracy": 0.24980825632810594,
      "num_tokens": 45649460.0,
      "step": 19915
    },
    {
      "entropy": 5.2597757339477536,
      "epoch": 1.9135446685878963,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046362926897985067,
      "loss": 4.6978,
      "mean_token_accuracy": 0.24582015424966813,
      "num_tokens": 45659995.0,
      "step": 19920
    },
    {
      "entropy": 5.281824159622192,
      "epoch": 1.9140249759846302,
      "grad_norm": 1.046875,
      "learning_rate": 0.000463610577836708,
      "loss": 4.7807,
      "mean_token_accuracy": 0.23559925258159636,
      "num_tokens": 45672262.0,
      "step": 19925
    },
    {
      "entropy": 5.358142852783203,
      "epoch": 1.914505283381364,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046359188231458783,
      "loss": 4.8374,
      "mean_token_accuracy": 0.23558975160121917,
      "num_tokens": 45685210.0,
      "step": 19930
    },
    {
      "entropy": 5.3540058612823485,
      "epoch": 1.914985590778098,
      "grad_norm": 1.0,
      "learning_rate": 0.00046357318241392414,
      "loss": 4.8675,
      "mean_token_accuracy": 0.2335409089922905,
      "num_tokens": 45696418.0,
      "step": 19935
    },
    {
      "entropy": 5.3596251010894775,
      "epoch": 1.915465898174832,
      "grad_norm": 1.046875,
      "learning_rate": 0.000463554478135151,
      "loss": 4.8721,
      "mean_token_accuracy": 0.232419216632843,
      "num_tokens": 45708393.0,
      "step": 19940
    },
    {
      "entropy": 5.392849349975586,
      "epoch": 1.915946205571566,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004635357694787029,
      "loss": 4.8532,
      "mean_token_accuracy": 0.22789739221334457,
      "num_tokens": 45721287.0,
      "step": 19945
    },
    {
      "entropy": 5.31380443572998,
      "epoch": 1.9164265129682998,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000463517056445014,
      "loss": 4.7057,
      "mean_token_accuracy": 0.24682828038930893,
      "num_tokens": 45732846.0,
      "step": 19950
    },
    {
      "entropy": 5.367904281616211,
      "epoch": 1.9169068203650337,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046349833903451884,
      "loss": 4.8784,
      "mean_token_accuracy": 0.22242726534605026,
      "num_tokens": 45743931.0,
      "step": 19955
    },
    {
      "entropy": 5.356407880783081,
      "epoch": 1.9173871277617676,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046347961724765196,
      "loss": 4.7908,
      "mean_token_accuracy": 0.23706873804330825,
      "num_tokens": 45755512.0,
      "step": 19960
    },
    {
      "entropy": 5.360158061981201,
      "epoch": 1.9178674351585014,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046346089108484806,
      "loss": 4.7796,
      "mean_token_accuracy": 0.23853152096271515,
      "num_tokens": 45766793.0,
      "step": 19965
    },
    {
      "entropy": 5.371113443374634,
      "epoch": 1.9183477425552353,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046344216054654193,
      "loss": 4.85,
      "mean_token_accuracy": 0.2355587363243103,
      "num_tokens": 45778214.0,
      "step": 19970
    },
    {
      "entropy": 5.472319412231445,
      "epoch": 1.9188280499519692,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046342342563316833,
      "loss": 4.999,
      "mean_token_accuracy": 0.2201638400554657,
      "num_tokens": 45791255.0,
      "step": 19975
    },
    {
      "entropy": 5.415157890319824,
      "epoch": 1.919308357348703,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046340468634516223,
      "loss": 4.8484,
      "mean_token_accuracy": 0.22834677696228028,
      "num_tokens": 45803351.0,
      "step": 19980
    },
    {
      "entropy": 5.303845643997192,
      "epoch": 1.919788664745437,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046338594268295884,
      "loss": 4.7976,
      "mean_token_accuracy": 0.24340311586856841,
      "num_tokens": 45814986.0,
      "step": 19985
    },
    {
      "entropy": 5.269008684158325,
      "epoch": 1.9202689721421708,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004633671946469933,
      "loss": 4.7849,
      "mean_token_accuracy": 0.23593875169754028,
      "num_tokens": 45827073.0,
      "step": 19990
    },
    {
      "entropy": 5.349352264404297,
      "epoch": 1.920749279538905,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046334844223770076,
      "loss": 4.8162,
      "mean_token_accuracy": 0.2462590217590332,
      "num_tokens": 45837808.0,
      "step": 19995
    },
    {
      "entropy": 5.328284025192261,
      "epoch": 1.9212295869356388,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046332968545551674,
      "loss": 4.8583,
      "mean_token_accuracy": 0.23802825361490249,
      "num_tokens": 45849932.0,
      "step": 20000
    },
    {
      "entropy": 5.336858701705933,
      "epoch": 1.9217098943323727,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004633109243008765,
      "loss": 4.809,
      "mean_token_accuracy": 0.23452580571174622,
      "num_tokens": 45860961.0,
      "step": 20005
    },
    {
      "entropy": 5.300229883193969,
      "epoch": 1.9221902017291066,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004632921587742159,
      "loss": 4.7546,
      "mean_token_accuracy": 0.243807153403759,
      "num_tokens": 45872251.0,
      "step": 20010
    },
    {
      "entropy": 5.443514728546143,
      "epoch": 1.9226705091258407,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046327338887597043,
      "loss": 4.8722,
      "mean_token_accuracy": 0.23636229038238527,
      "num_tokens": 45883234.0,
      "step": 20015
    },
    {
      "entropy": 5.349977302551269,
      "epoch": 1.9231508165225746,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000463254614606576,
      "loss": 4.7891,
      "mean_token_accuracy": 0.2392379656434059,
      "num_tokens": 45894003.0,
      "step": 20020
    },
    {
      "entropy": 5.256306600570679,
      "epoch": 1.9236311239193085,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004632358359664683,
      "loss": 4.7423,
      "mean_token_accuracy": 0.24469906240701675,
      "num_tokens": 45906109.0,
      "step": 20025
    },
    {
      "entropy": 5.321726512908936,
      "epoch": 1.9241114313160423,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046321705295608356,
      "loss": 4.8604,
      "mean_token_accuracy": 0.23638131618499755,
      "num_tokens": 45918372.0,
      "step": 20030
    },
    {
      "entropy": 5.399338054656982,
      "epoch": 1.9245917387127762,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046319826557585764,
      "loss": 4.9026,
      "mean_token_accuracy": 0.2265574023127556,
      "num_tokens": 45928750.0,
      "step": 20035
    },
    {
      "entropy": 5.4325597286224365,
      "epoch": 1.92507204610951,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004631794738262269,
      "loss": 4.8746,
      "mean_token_accuracy": 0.24359854757785798,
      "num_tokens": 45939297.0,
      "step": 20040
    },
    {
      "entropy": 5.361817646026611,
      "epoch": 1.925552353506244,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004631606777076275,
      "loss": 4.8113,
      "mean_token_accuracy": 0.23584813326597215,
      "num_tokens": 45949986.0,
      "step": 20045
    },
    {
      "entropy": 5.344836711883545,
      "epoch": 1.9260326609029779,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046314187722049587,
      "loss": 4.8285,
      "mean_token_accuracy": 0.23275830298662187,
      "num_tokens": 45960747.0,
      "step": 20050
    },
    {
      "entropy": 5.364686155319214,
      "epoch": 1.9265129682997117,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046312307236526863,
      "loss": 4.8286,
      "mean_token_accuracy": 0.23959849774837494,
      "num_tokens": 45972260.0,
      "step": 20055
    },
    {
      "entropy": 5.344644975662232,
      "epoch": 1.9269932756964456,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046310426314238217,
      "loss": 4.7592,
      "mean_token_accuracy": 0.24234011322259902,
      "num_tokens": 45983589.0,
      "step": 20060
    },
    {
      "entropy": 5.274921941757202,
      "epoch": 1.9274735830931795,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004630854495522733,
      "loss": 4.7974,
      "mean_token_accuracy": 0.2399603620171547,
      "num_tokens": 45994776.0,
      "step": 20065
    },
    {
      "entropy": 5.38647313117981,
      "epoch": 1.9279538904899134,
      "grad_norm": 1.0,
      "learning_rate": 0.00046306663159537874,
      "loss": 4.8747,
      "mean_token_accuracy": 0.22968562096357345,
      "num_tokens": 46005530.0,
      "step": 20070
    },
    {
      "entropy": 5.3619239807128904,
      "epoch": 1.9284341978866475,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046304780927213554,
      "loss": 4.7706,
      "mean_token_accuracy": 0.24525422751903533,
      "num_tokens": 46017538.0,
      "step": 20075
    },
    {
      "entropy": 5.3402856349945065,
      "epoch": 1.9289145052833814,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046302898258298046,
      "loss": 4.756,
      "mean_token_accuracy": 0.2356465756893158,
      "num_tokens": 46029011.0,
      "step": 20080
    },
    {
      "entropy": 5.260745334625244,
      "epoch": 1.9293948126801153,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004630101515283509,
      "loss": 4.7906,
      "mean_token_accuracy": 0.2404362291097641,
      "num_tokens": 46040354.0,
      "step": 20085
    },
    {
      "entropy": 5.325657892227173,
      "epoch": 1.9298751200768491,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046299131610868377,
      "loss": 4.7929,
      "mean_token_accuracy": 0.24076392203569413,
      "num_tokens": 46051127.0,
      "step": 20090
    },
    {
      "entropy": 5.34856915473938,
      "epoch": 1.9303554274735832,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004629724763244165,
      "loss": 4.8175,
      "mean_token_accuracy": 0.2397377982735634,
      "num_tokens": 46062343.0,
      "step": 20095
    },
    {
      "entropy": 5.259103441238404,
      "epoch": 1.9308357348703171,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004629536321759866,
      "loss": 4.7614,
      "mean_token_accuracy": 0.24164892584085465,
      "num_tokens": 46074193.0,
      "step": 20100
    },
    {
      "entropy": 5.363073444366455,
      "epoch": 1.931316042267051,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046293478366383133,
      "loss": 4.7973,
      "mean_token_accuracy": 0.24354219436645508,
      "num_tokens": 46085930.0,
      "step": 20105
    },
    {
      "entropy": 5.420891189575196,
      "epoch": 1.9317963496637849,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004629159307883885,
      "loss": 4.8917,
      "mean_token_accuracy": 0.2286633789539337,
      "num_tokens": 46099759.0,
      "step": 20110
    },
    {
      "entropy": 5.467717981338501,
      "epoch": 1.9322766570605188,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004628970735500958,
      "loss": 4.9363,
      "mean_token_accuracy": 0.22301208227872849,
      "num_tokens": 46110907.0,
      "step": 20115
    },
    {
      "entropy": 5.374463748931885,
      "epoch": 1.9327569644572526,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046287821194939094,
      "loss": 4.8379,
      "mean_token_accuracy": 0.23561217337846757,
      "num_tokens": 46122167.0,
      "step": 20120
    },
    {
      "entropy": 5.382249927520752,
      "epoch": 1.9332372718539865,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046285934598671186,
      "loss": 4.9105,
      "mean_token_accuracy": 0.233339524269104,
      "num_tokens": 46132948.0,
      "step": 20125
    },
    {
      "entropy": 5.383736228942871,
      "epoch": 1.9337175792507204,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00046284047566249665,
      "loss": 4.8852,
      "mean_token_accuracy": 0.2355683520436287,
      "num_tokens": 46144275.0,
      "step": 20130
    },
    {
      "entropy": 5.332909870147705,
      "epoch": 1.9341978866474543,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046282160097718336,
      "loss": 4.8441,
      "mean_token_accuracy": 0.23685766607522965,
      "num_tokens": 46156138.0,
      "step": 20135
    },
    {
      "entropy": 5.363868618011475,
      "epoch": 1.9346781940441882,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004628027219312102,
      "loss": 4.8604,
      "mean_token_accuracy": 0.23534936159849168,
      "num_tokens": 46169437.0,
      "step": 20140
    },
    {
      "entropy": 5.339114952087402,
      "epoch": 1.935158501440922,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004627838385250155,
      "loss": 4.7753,
      "mean_token_accuracy": 0.24139665365219115,
      "num_tokens": 46182150.0,
      "step": 20145
    },
    {
      "entropy": 5.3675790309906,
      "epoch": 1.9356388088376562,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046276495075903764,
      "loss": 4.8149,
      "mean_token_accuracy": 0.2452509492635727,
      "num_tokens": 46193945.0,
      "step": 20150
    },
    {
      "entropy": 5.379150772094727,
      "epoch": 1.93611911623439,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046274605863371517,
      "loss": 4.9201,
      "mean_token_accuracy": 0.2328202784061432,
      "num_tokens": 46206130.0,
      "step": 20155
    },
    {
      "entropy": 5.28761510848999,
      "epoch": 1.936599423631124,
      "grad_norm": 1.0,
      "learning_rate": 0.0004627271621494868,
      "loss": 4.7598,
      "mean_token_accuracy": 0.24176323562860488,
      "num_tokens": 46217284.0,
      "step": 20160
    },
    {
      "entropy": 5.318743944168091,
      "epoch": 1.9370797310278578,
      "grad_norm": 1.015625,
      "learning_rate": 0.000462708261306791,
      "loss": 4.8007,
      "mean_token_accuracy": 0.23690288364887238,
      "num_tokens": 46227949.0,
      "step": 20165
    },
    {
      "entropy": 5.346001529693604,
      "epoch": 1.937560038424592,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004626893561060669,
      "loss": 4.7567,
      "mean_token_accuracy": 0.23342611491680146,
      "num_tokens": 46239101.0,
      "step": 20170
    },
    {
      "entropy": 5.392166423797607,
      "epoch": 1.9380403458213258,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046267044654775324,
      "loss": 4.9783,
      "mean_token_accuracy": 0.23080155849456788,
      "num_tokens": 46251272.0,
      "step": 20175
    },
    {
      "entropy": 5.360622215270996,
      "epoch": 1.9385206532180597,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000462651532632289,
      "loss": 4.8984,
      "mean_token_accuracy": 0.23023003488779067,
      "num_tokens": 46262603.0,
      "step": 20180
    },
    {
      "entropy": 5.347624015808106,
      "epoch": 1.9390009606147935,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046263261436011344,
      "loss": 4.7991,
      "mean_token_accuracy": 0.2355230674147606,
      "num_tokens": 46275142.0,
      "step": 20185
    },
    {
      "entropy": 5.331681442260742,
      "epoch": 1.9394812680115274,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004626136917316657,
      "loss": 4.8323,
      "mean_token_accuracy": 0.23914351910352707,
      "num_tokens": 46286339.0,
      "step": 20190
    },
    {
      "entropy": 5.329445123672485,
      "epoch": 1.9399615754082613,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046259476474738514,
      "loss": 4.783,
      "mean_token_accuracy": 0.23875517547130584,
      "num_tokens": 46297514.0,
      "step": 20195
    },
    {
      "entropy": 5.290621519088745,
      "epoch": 1.9404418828049952,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046257583340771123,
      "loss": 4.7013,
      "mean_token_accuracy": 0.24338473081588746,
      "num_tokens": 46309249.0,
      "step": 20200
    },
    {
      "entropy": 5.38100037574768,
      "epoch": 1.940922190201729,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004625568977130835,
      "loss": 4.8686,
      "mean_token_accuracy": 0.23066089749336244,
      "num_tokens": 46320759.0,
      "step": 20205
    },
    {
      "entropy": 5.434899616241455,
      "epoch": 1.941402497598463,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004625379576639414,
      "loss": 4.873,
      "mean_token_accuracy": 0.2320779785513878,
      "num_tokens": 46332113.0,
      "step": 20210
    },
    {
      "entropy": 5.351282119750977,
      "epoch": 1.9418828049951968,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046251901326072487,
      "loss": 4.8367,
      "mean_token_accuracy": 0.24284330755472183,
      "num_tokens": 46342648.0,
      "step": 20215
    },
    {
      "entropy": 5.289817905426025,
      "epoch": 1.9423631123919307,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046250006450387367,
      "loss": 4.7603,
      "mean_token_accuracy": 0.2387276977300644,
      "num_tokens": 46354066.0,
      "step": 20220
    },
    {
      "entropy": 5.399824905395508,
      "epoch": 1.9428434197886646,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004624811113938277,
      "loss": 4.8689,
      "mean_token_accuracy": 0.23045812994241716,
      "num_tokens": 46366586.0,
      "step": 20225
    },
    {
      "entropy": 5.369870615005493,
      "epoch": 1.9433237271853987,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004624621539310271,
      "loss": 4.7966,
      "mean_token_accuracy": 0.2432965785264969,
      "num_tokens": 46378155.0,
      "step": 20230
    },
    {
      "entropy": 5.368222713470459,
      "epoch": 1.9438040345821326,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046244319211591193,
      "loss": 4.8556,
      "mean_token_accuracy": 0.23281151801347733,
      "num_tokens": 46389631.0,
      "step": 20235
    },
    {
      "entropy": 5.4045090675354,
      "epoch": 1.9442843419788665,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004624242259489223,
      "loss": 4.8837,
      "mean_token_accuracy": 0.22764550745487214,
      "num_tokens": 46401915.0,
      "step": 20240
    },
    {
      "entropy": 5.425702905654907,
      "epoch": 1.9447646493756003,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046240525543049884,
      "loss": 4.8326,
      "mean_token_accuracy": 0.2353322833776474,
      "num_tokens": 46412396.0,
      "step": 20245
    },
    {
      "entropy": 5.272420024871826,
      "epoch": 1.9452449567723344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046238628056108176,
      "loss": 4.7484,
      "mean_token_accuracy": 0.24524901807308197,
      "num_tokens": 46423089.0,
      "step": 20250
    },
    {
      "entropy": 5.386928939819336,
      "epoch": 1.9457252641690683,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046236730134111166,
      "loss": 4.854,
      "mean_token_accuracy": 0.22576792687177658,
      "num_tokens": 46434858.0,
      "step": 20255
    },
    {
      "entropy": 5.314341926574707,
      "epoch": 1.9462055715658022,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004623483177710291,
      "loss": 4.7866,
      "mean_token_accuracy": 0.2478043705224991,
      "num_tokens": 46445628.0,
      "step": 20260
    },
    {
      "entropy": 5.3217638492584225,
      "epoch": 1.946685878962536,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004623293298512751,
      "loss": 4.7935,
      "mean_token_accuracy": 0.2385061964392662,
      "num_tokens": 46457382.0,
      "step": 20265
    },
    {
      "entropy": 5.3332127094268795,
      "epoch": 1.94716618635927,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046231033758229026,
      "loss": 4.8328,
      "mean_token_accuracy": 0.2301131308078766,
      "num_tokens": 46469765.0,
      "step": 20270
    },
    {
      "entropy": 5.475966835021973,
      "epoch": 1.9476464937560038,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004622913409645154,
      "loss": 4.9361,
      "mean_token_accuracy": 0.2241200476884842,
      "num_tokens": 46481451.0,
      "step": 20275
    },
    {
      "entropy": 5.360985136032104,
      "epoch": 1.9481268011527377,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004622723399983919,
      "loss": 4.72,
      "mean_token_accuracy": 0.2424934983253479,
      "num_tokens": 46491585.0,
      "step": 20280
    },
    {
      "entropy": 5.353017997741699,
      "epoch": 1.9486071085494716,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046225333468436077,
      "loss": 4.93,
      "mean_token_accuracy": 0.22804316133260727,
      "num_tokens": 46503789.0,
      "step": 20285
    },
    {
      "entropy": 5.433307266235351,
      "epoch": 1.9490874159462055,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046223432502286323,
      "loss": 4.8912,
      "mean_token_accuracy": 0.23260654211044313,
      "num_tokens": 46516647.0,
      "step": 20290
    },
    {
      "entropy": 5.407173442840576,
      "epoch": 1.9495677233429394,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046221531101434056,
      "loss": 4.8435,
      "mean_token_accuracy": 0.22540843337774277,
      "num_tokens": 46527432.0,
      "step": 20295
    },
    {
      "entropy": 5.38223237991333,
      "epoch": 1.9500480307396733,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004621962926592343,
      "loss": 4.8659,
      "mean_token_accuracy": 0.23220161348581314,
      "num_tokens": 46538990.0,
      "step": 20300
    },
    {
      "entropy": 5.36404390335083,
      "epoch": 1.9505283381364071,
      "grad_norm": 1.09375,
      "learning_rate": 0.000462177269957986,
      "loss": 4.8393,
      "mean_token_accuracy": 0.2339954659342766,
      "num_tokens": 46551291.0,
      "step": 20305
    },
    {
      "entropy": 5.3489098072052,
      "epoch": 1.9510086455331412,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004621582429110373,
      "loss": 4.7589,
      "mean_token_accuracy": 0.24161833226680757,
      "num_tokens": 46562415.0,
      "step": 20310
    },
    {
      "entropy": 5.406794261932373,
      "epoch": 1.9514889529298751,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046213921151883,
      "loss": 4.8914,
      "mean_token_accuracy": 0.24213654398918152,
      "num_tokens": 46572544.0,
      "step": 20315
    },
    {
      "entropy": 5.324662494659424,
      "epoch": 1.951969260326609,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004621201757818059,
      "loss": 4.7903,
      "mean_token_accuracy": 0.23972784876823425,
      "num_tokens": 46584694.0,
      "step": 20320
    },
    {
      "entropy": 5.391635465621948,
      "epoch": 1.952449567723343,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046210113570040683,
      "loss": 4.822,
      "mean_token_accuracy": 0.2402570888400078,
      "num_tokens": 46595086.0,
      "step": 20325
    },
    {
      "entropy": 5.4260388851165775,
      "epoch": 1.952929875120077,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000462082091275075,
      "loss": 4.8476,
      "mean_token_accuracy": 0.2334176331758499,
      "num_tokens": 46606627.0,
      "step": 20330
    },
    {
      "entropy": 5.397216320037842,
      "epoch": 1.9534101825168109,
      "grad_norm": 1.25,
      "learning_rate": 0.0004620630425062526,
      "loss": 4.8589,
      "mean_token_accuracy": 0.22950285375118257,
      "num_tokens": 46618560.0,
      "step": 20335
    },
    {
      "entropy": 5.3590456485748295,
      "epoch": 1.9538904899135447,
      "grad_norm": 1.0,
      "learning_rate": 0.0004620439893943817,
      "loss": 4.8163,
      "mean_token_accuracy": 0.22947212010622026,
      "num_tokens": 46630109.0,
      "step": 20340
    },
    {
      "entropy": 5.315780544281006,
      "epoch": 1.9543707973102786,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004620249319399049,
      "loss": 4.8061,
      "mean_token_accuracy": 0.2439057022333145,
      "num_tokens": 46642247.0,
      "step": 20345
    },
    {
      "entropy": 5.410763883590699,
      "epoch": 1.9548511047070125,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046200587014326455,
      "loss": 4.8682,
      "mean_token_accuracy": 0.23265372663736344,
      "num_tokens": 46653119.0,
      "step": 20350
    },
    {
      "entropy": 5.314918184280396,
      "epoch": 1.9553314121037464,
      "grad_norm": 1.125,
      "learning_rate": 0.0004619868040049031,
      "loss": 4.7803,
      "mean_token_accuracy": 0.2401042804121971,
      "num_tokens": 46664392.0,
      "step": 20355
    },
    {
      "entropy": 5.323352241516114,
      "epoch": 1.9558117195004803,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004619677335252633,
      "loss": 4.789,
      "mean_token_accuracy": 0.24389179795980453,
      "num_tokens": 46674981.0,
      "step": 20360
    },
    {
      "entropy": 5.366019582748413,
      "epoch": 1.9562920268972142,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00046194865870478793,
      "loss": 4.8782,
      "mean_token_accuracy": 0.23237283676862716,
      "num_tokens": 46688692.0,
      "step": 20365
    },
    {
      "entropy": 5.2496335983276365,
      "epoch": 1.956772334293948,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046192957954391983,
      "loss": 4.7086,
      "mean_token_accuracy": 0.24947577714920044,
      "num_tokens": 46699799.0,
      "step": 20370
    },
    {
      "entropy": 5.299591112136841,
      "epoch": 1.957252641690682,
      "grad_norm": 1.046875,
      "learning_rate": 0.000461910496043102,
      "loss": 4.8037,
      "mean_token_accuracy": 0.23547885566949844,
      "num_tokens": 46711994.0,
      "step": 20375
    },
    {
      "entropy": 5.35178804397583,
      "epoch": 1.9577329490874158,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004618914082027773,
      "loss": 4.8558,
      "mean_token_accuracy": 0.23428200632333757,
      "num_tokens": 46723247.0,
      "step": 20380
    },
    {
      "entropy": 5.453936052322388,
      "epoch": 1.95821325648415,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046187231602338926,
      "loss": 4.8876,
      "mean_token_accuracy": 0.22858227491378785,
      "num_tokens": 46732824.0,
      "step": 20385
    },
    {
      "entropy": 5.32496600151062,
      "epoch": 1.9586935638808838,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046185321950538086,
      "loss": 4.8058,
      "mean_token_accuracy": 0.23688641786575318,
      "num_tokens": 46744640.0,
      "step": 20390
    },
    {
      "entropy": 5.317246913909912,
      "epoch": 1.9591738712776177,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004618341186491955,
      "loss": 4.8224,
      "mean_token_accuracy": 0.24122364073991776,
      "num_tokens": 46757513.0,
      "step": 20395
    },
    {
      "entropy": 5.315438508987427,
      "epoch": 1.9596541786743515,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004618150134552768,
      "loss": 4.7603,
      "mean_token_accuracy": 0.24414074867963792,
      "num_tokens": 46769021.0,
      "step": 20400
    },
    {
      "entropy": 5.310075759887695,
      "epoch": 1.9601344860710856,
      "grad_norm": 0.984375,
      "learning_rate": 0.000461795903924068,
      "loss": 4.725,
      "mean_token_accuracy": 0.23971999138593675,
      "num_tokens": 46781175.0,
      "step": 20405
    },
    {
      "entropy": 5.333147859573364,
      "epoch": 1.9606147934678195,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046177679005601313,
      "loss": 4.7366,
      "mean_token_accuracy": 0.2390782728791237,
      "num_tokens": 46792256.0,
      "step": 20410
    },
    {
      "entropy": 5.287149429321289,
      "epoch": 1.9610951008645534,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004617576718515558,
      "loss": 4.7372,
      "mean_token_accuracy": 0.24306785464286804,
      "num_tokens": 46803215.0,
      "step": 20415
    },
    {
      "entropy": 5.386393404006958,
      "epoch": 1.9615754082612873,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004617385493111399,
      "loss": 4.8642,
      "mean_token_accuracy": 0.2420525297522545,
      "num_tokens": 46813770.0,
      "step": 20420
    },
    {
      "entropy": 5.332234048843384,
      "epoch": 1.9620557156580212,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004617194224352093,
      "loss": 4.8176,
      "mean_token_accuracy": 0.23516337871551513,
      "num_tokens": 46826923.0,
      "step": 20425
    },
    {
      "entropy": 5.288569736480713,
      "epoch": 1.962536023054755,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004617002912242083,
      "loss": 4.7529,
      "mean_token_accuracy": 0.24247635304927825,
      "num_tokens": 46837761.0,
      "step": 20430
    },
    {
      "entropy": 5.345312643051147,
      "epoch": 1.963016330451489,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046168115567858084,
      "loss": 4.8214,
      "mean_token_accuracy": 0.2344038724899292,
      "num_tokens": 46848190.0,
      "step": 20435
    },
    {
      "entropy": 5.279523706436157,
      "epoch": 1.9634966378482228,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046166201579877125,
      "loss": 4.7241,
      "mean_token_accuracy": 0.24518334716558457,
      "num_tokens": 46858643.0,
      "step": 20440
    },
    {
      "entropy": 5.343136644363403,
      "epoch": 1.9639769452449567,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004616428715852239,
      "loss": 4.8134,
      "mean_token_accuracy": 0.24062685966491698,
      "num_tokens": 46868711.0,
      "step": 20445
    },
    {
      "entropy": 5.370600175857544,
      "epoch": 1.9644572526416906,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004616237230383833,
      "loss": 4.8909,
      "mean_token_accuracy": 0.23237680941820144,
      "num_tokens": 46880045.0,
      "step": 20450
    },
    {
      "entropy": 5.336586761474609,
      "epoch": 1.9649375600384245,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046160457015869414,
      "loss": 4.8548,
      "mean_token_accuracy": 0.23151590675115585,
      "num_tokens": 46890680.0,
      "step": 20455
    },
    {
      "entropy": 5.324366331100464,
      "epoch": 1.9654178674351583,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046158541294660083,
      "loss": 4.7794,
      "mean_token_accuracy": 0.23879679441452026,
      "num_tokens": 46904807.0,
      "step": 20460
    },
    {
      "entropy": 5.394479131698608,
      "epoch": 1.9658981748318924,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004615662514025483,
      "loss": 4.7807,
      "mean_token_accuracy": 0.24052320122718812,
      "num_tokens": 46916039.0,
      "step": 20465
    },
    {
      "entropy": 5.372701215744018,
      "epoch": 1.9663784822286263,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046154708552698147,
      "loss": 4.7988,
      "mean_token_accuracy": 0.23545903265476226,
      "num_tokens": 46927175.0,
      "step": 20470
    },
    {
      "entropy": 5.277928400039673,
      "epoch": 1.9668587896253602,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046152791532034517,
      "loss": 4.6978,
      "mean_token_accuracy": 0.24975655525922774,
      "num_tokens": 46938711.0,
      "step": 20475
    },
    {
      "entropy": 5.336776542663574,
      "epoch": 1.967339097022094,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00046150874078308463,
      "loss": 4.8092,
      "mean_token_accuracy": 0.23568939715623854,
      "num_tokens": 46951002.0,
      "step": 20480
    },
    {
      "entropy": 5.408925485610962,
      "epoch": 1.9678194044188282,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004614895619156449,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2365707501769066,
      "num_tokens": 46962673.0,
      "step": 20485
    },
    {
      "entropy": 5.356033039093018,
      "epoch": 1.968299711815562,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046147037871847125,
      "loss": 4.7928,
      "mean_token_accuracy": 0.234052412211895,
      "num_tokens": 46973955.0,
      "step": 20490
    },
    {
      "entropy": 5.374083089828491,
      "epoch": 1.968780019212296,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004614511911920092,
      "loss": 4.9166,
      "mean_token_accuracy": 0.23103232234716414,
      "num_tokens": 46985791.0,
      "step": 20495
    },
    {
      "entropy": 5.478353357315063,
      "epoch": 1.9692603266090298,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046143199933670404,
      "loss": 4.9685,
      "mean_token_accuracy": 0.22179683893918992,
      "num_tokens": 46998437.0,
      "step": 20500
    },
    {
      "entropy": 5.3344331741333,
      "epoch": 1.9697406340057637,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004614128031530015,
      "loss": 4.7712,
      "mean_token_accuracy": 0.23573401272296907,
      "num_tokens": 47009666.0,
      "step": 20505
    },
    {
      "entropy": 5.389057540893555,
      "epoch": 1.9702209414024976,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046139360264134724,
      "loss": 4.8591,
      "mean_token_accuracy": 0.2322443827986717,
      "num_tokens": 47021170.0,
      "step": 20510
    },
    {
      "entropy": 5.3606383323669435,
      "epoch": 1.9707012487992315,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004613743978021869,
      "loss": 4.7956,
      "mean_token_accuracy": 0.24760405272245406,
      "num_tokens": 47033077.0,
      "step": 20515
    },
    {
      "entropy": 5.365155363082886,
      "epoch": 1.9711815561959654,
      "grad_norm": 1.0,
      "learning_rate": 0.00046135518863596654,
      "loss": 4.8524,
      "mean_token_accuracy": 0.23260460644960404,
      "num_tokens": 47044858.0,
      "step": 20520
    },
    {
      "entropy": 5.42169303894043,
      "epoch": 1.9716618635926992,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046133597514313204,
      "loss": 4.9354,
      "mean_token_accuracy": 0.22503034621477128,
      "num_tokens": 47057587.0,
      "step": 20525
    },
    {
      "entropy": 5.431663703918457,
      "epoch": 1.9721421709894331,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004613167573241295,
      "loss": 4.8633,
      "mean_token_accuracy": 0.22571469843387604,
      "num_tokens": 47068690.0,
      "step": 20530
    },
    {
      "entropy": 5.344943714141846,
      "epoch": 1.972622478386167,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004612975351794051,
      "loss": 4.815,
      "mean_token_accuracy": 0.23921523839235306,
      "num_tokens": 47081279.0,
      "step": 20535
    },
    {
      "entropy": 5.368409919738769,
      "epoch": 1.973102785782901,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004612783087094051,
      "loss": 4.8761,
      "mean_token_accuracy": 0.23760145604610444,
      "num_tokens": 47093372.0,
      "step": 20540
    },
    {
      "entropy": 5.450712728500366,
      "epoch": 1.973583093179635,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046125907791457594,
      "loss": 4.9563,
      "mean_token_accuracy": 0.2253487750887871,
      "num_tokens": 47104332.0,
      "step": 20545
    },
    {
      "entropy": 5.372053575515747,
      "epoch": 1.9740634005763689,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046123984279536405,
      "loss": 4.8203,
      "mean_token_accuracy": 0.22933131009340285,
      "num_tokens": 47116941.0,
      "step": 20550
    },
    {
      "entropy": 5.406744289398193,
      "epoch": 1.9745437079731027,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046122060335221604,
      "loss": 4.8703,
      "mean_token_accuracy": 0.23210972547531128,
      "num_tokens": 47128473.0,
      "step": 20555
    },
    {
      "entropy": 5.3354418754577635,
      "epoch": 1.9750240153698368,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046120135958557855,
      "loss": 4.826,
      "mean_token_accuracy": 0.22706248462200165,
      "num_tokens": 47140586.0,
      "step": 20560
    },
    {
      "entropy": 5.325597047805786,
      "epoch": 1.9755043227665707,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046118211149589843,
      "loss": 4.7732,
      "mean_token_accuracy": 0.2444738507270813,
      "num_tokens": 47150875.0,
      "step": 20565
    },
    {
      "entropy": 5.322201251983643,
      "epoch": 1.9759846301633046,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004611628590836225,
      "loss": 4.7291,
      "mean_token_accuracy": 0.2462625563144684,
      "num_tokens": 47162074.0,
      "step": 20570
    },
    {
      "entropy": 5.299340629577637,
      "epoch": 1.9764649375600385,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004611436023491979,
      "loss": 4.7638,
      "mean_token_accuracy": 0.24526274502277373,
      "num_tokens": 47174544.0,
      "step": 20575
    },
    {
      "entropy": 5.392938327789307,
      "epoch": 1.9769452449567724,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004611243412930714,
      "loss": 4.8578,
      "mean_token_accuracy": 0.23189646899700164,
      "num_tokens": 47186260.0,
      "step": 20580
    },
    {
      "entropy": 5.379665184020996,
      "epoch": 1.9774255523535063,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046110507591569047,
      "loss": 4.7953,
      "mean_token_accuracy": 0.2351905345916748,
      "num_tokens": 47198059.0,
      "step": 20585
    },
    {
      "entropy": 5.338194417953491,
      "epoch": 1.9779058597502401,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004610858062175023,
      "loss": 4.7412,
      "mean_token_accuracy": 0.2419790357351303,
      "num_tokens": 47208300.0,
      "step": 20590
    },
    {
      "entropy": 5.333975219726563,
      "epoch": 1.978386167146974,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046106653219895417,
      "loss": 4.7967,
      "mean_token_accuracy": 0.2414494961500168,
      "num_tokens": 47220135.0,
      "step": 20595
    },
    {
      "entropy": 5.447124719619751,
      "epoch": 1.978866474543708,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004610472538604938,
      "loss": 4.8814,
      "mean_token_accuracy": 0.22667128443717957,
      "num_tokens": 47230385.0,
      "step": 20600
    },
    {
      "entropy": 5.262587833404541,
      "epoch": 1.9793467819404418,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046102797120256854,
      "loss": 4.7178,
      "mean_token_accuracy": 0.2454138681292534,
      "num_tokens": 47242593.0,
      "step": 20605
    },
    {
      "entropy": 5.346311759948731,
      "epoch": 1.9798270893371757,
      "grad_norm": 1.125,
      "learning_rate": 0.0004610086842256262,
      "loss": 4.8621,
      "mean_token_accuracy": 0.23621759712696075,
      "num_tokens": 47254223.0,
      "step": 20610
    },
    {
      "entropy": 5.363051033020019,
      "epoch": 1.9803073967339095,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004609893929301146,
      "loss": 4.8706,
      "mean_token_accuracy": 0.2401515170931816,
      "num_tokens": 47266559.0,
      "step": 20615
    },
    {
      "entropy": 5.432278299331665,
      "epoch": 1.9807877041306436,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004609700973164816,
      "loss": 4.8483,
      "mean_token_accuracy": 0.23838646560907364,
      "num_tokens": 47276438.0,
      "step": 20620
    },
    {
      "entropy": 5.338067150115966,
      "epoch": 1.9812680115273775,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004609507973851751,
      "loss": 4.8562,
      "mean_token_accuracy": 0.23061287850141526,
      "num_tokens": 47287791.0,
      "step": 20625
    },
    {
      "entropy": 5.281011247634888,
      "epoch": 1.9817483189241114,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046093149313664316,
      "loss": 4.811,
      "mean_token_accuracy": 0.23800179362297058,
      "num_tokens": 47300282.0,
      "step": 20630
    },
    {
      "entropy": 5.4391289234161375,
      "epoch": 1.9822286263208453,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004609121845713342,
      "loss": 4.8717,
      "mean_token_accuracy": 0.24332231879234315,
      "num_tokens": 47310473.0,
      "step": 20635
    },
    {
      "entropy": 5.3338133811950685,
      "epoch": 1.9827089337175794,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004608928716896963,
      "loss": 4.78,
      "mean_token_accuracy": 0.2434871032834053,
      "num_tokens": 47322245.0,
      "step": 20640
    },
    {
      "entropy": 5.396986579895019,
      "epoch": 1.9831892411143133,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004608735544921778,
      "loss": 4.823,
      "mean_token_accuracy": 0.2421170637011528,
      "num_tokens": 47334316.0,
      "step": 20645
    },
    {
      "entropy": 5.423580646514893,
      "epoch": 1.9836695485110472,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046085423297922745,
      "loss": 4.8595,
      "mean_token_accuracy": 0.23548156768083572,
      "num_tokens": 47344909.0,
      "step": 20650
    },
    {
      "entropy": 5.2697875022888185,
      "epoch": 1.984149855907781,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046083490715129367,
      "loss": 4.738,
      "mean_token_accuracy": 0.24617200791835786,
      "num_tokens": 47356682.0,
      "step": 20655
    },
    {
      "entropy": 5.311660814285278,
      "epoch": 1.984630163304515,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004608155770088251,
      "loss": 4.7172,
      "mean_token_accuracy": 0.24628447443246843,
      "num_tokens": 47368362.0,
      "step": 20660
    },
    {
      "entropy": 5.4142707824707035,
      "epoch": 1.9851104707012488,
      "grad_norm": 1.125,
      "learning_rate": 0.00046079624255227066,
      "loss": 4.8172,
      "mean_token_accuracy": 0.23526561856269837,
      "num_tokens": 47379932.0,
      "step": 20665
    },
    {
      "entropy": 5.40438985824585,
      "epoch": 1.9855907780979827,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004607769037820791,
      "loss": 4.9365,
      "mean_token_accuracy": 0.2272479146718979,
      "num_tokens": 47391244.0,
      "step": 20670
    },
    {
      "entropy": 5.357568645477295,
      "epoch": 1.9860710854947166,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004607575606986995,
      "loss": 4.8431,
      "mean_token_accuracy": 0.2362509161233902,
      "num_tokens": 47404005.0,
      "step": 20675
    },
    {
      "entropy": 5.282725238800049,
      "epoch": 1.9865513928914504,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004607382133025809,
      "loss": 4.735,
      "mean_token_accuracy": 0.2423497185111046,
      "num_tokens": 47414555.0,
      "step": 20680
    },
    {
      "entropy": 5.348944807052613,
      "epoch": 1.9870317002881843,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046071886159417257,
      "loss": 4.8761,
      "mean_token_accuracy": 0.23038425594568251,
      "num_tokens": 47425902.0,
      "step": 20685
    },
    {
      "entropy": 5.2835489273071286,
      "epoch": 1.9875120076849182,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004606995055739238,
      "loss": 4.7438,
      "mean_token_accuracy": 0.24230887740850449,
      "num_tokens": 47438068.0,
      "step": 20690
    },
    {
      "entropy": 5.433887147903443,
      "epoch": 1.9879923150816523,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046068014524228374,
      "loss": 4.8846,
      "mean_token_accuracy": 0.23057420700788497,
      "num_tokens": 47450322.0,
      "step": 20695
    },
    {
      "entropy": 5.307609033584595,
      "epoch": 1.9884726224783862,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046066078059970217,
      "loss": 4.8025,
      "mean_token_accuracy": 0.2398951083421707,
      "num_tokens": 47462615.0,
      "step": 20700
    },
    {
      "entropy": 5.352494573593139,
      "epoch": 1.98895292987512,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004606414116466286,
      "loss": 4.7789,
      "mean_token_accuracy": 0.23670558631420135,
      "num_tokens": 47472530.0,
      "step": 20705
    },
    {
      "entropy": 5.2940788745880125,
      "epoch": 1.989433237271854,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046062203838351267,
      "loss": 4.7548,
      "mean_token_accuracy": 0.2420820876955986,
      "num_tokens": 47484217.0,
      "step": 20710
    },
    {
      "entropy": 5.401149892807007,
      "epoch": 1.989913544668588,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046060266081080414,
      "loss": 4.8216,
      "mean_token_accuracy": 0.2340631052851677,
      "num_tokens": 47496365.0,
      "step": 20715
    },
    {
      "entropy": 5.354928684234619,
      "epoch": 1.990393852065322,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004605832789289531,
      "loss": 4.8195,
      "mean_token_accuracy": 0.2480039194226265,
      "num_tokens": 47509437.0,
      "step": 20720
    },
    {
      "entropy": 5.3869099617004395,
      "epoch": 1.9908741594620558,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004605638927384093,
      "loss": 4.866,
      "mean_token_accuracy": 0.23057839125394822,
      "num_tokens": 47520913.0,
      "step": 20725
    },
    {
      "entropy": 5.423606634140015,
      "epoch": 1.9913544668587897,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046054450223962284,
      "loss": 4.8547,
      "mean_token_accuracy": 0.2293355494737625,
      "num_tokens": 47532015.0,
      "step": 20730
    },
    {
      "entropy": 5.450711393356324,
      "epoch": 1.9918347742555236,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046052510743304405,
      "loss": 4.9485,
      "mean_token_accuracy": 0.2284650668501854,
      "num_tokens": 47543675.0,
      "step": 20735
    },
    {
      "entropy": 5.356999826431275,
      "epoch": 1.9923150816522575,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004605057083191232,
      "loss": 4.8042,
      "mean_token_accuracy": 0.23203132301568985,
      "num_tokens": 47555757.0,
      "step": 20740
    },
    {
      "entropy": 5.401649904251099,
      "epoch": 1.9927953890489913,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004604863048983106,
      "loss": 4.9398,
      "mean_token_accuracy": 0.23058196604251863,
      "num_tokens": 47568043.0,
      "step": 20745
    },
    {
      "entropy": 5.39755072593689,
      "epoch": 1.9932756964457252,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004604668971710568,
      "loss": 4.7996,
      "mean_token_accuracy": 0.24258084893226622,
      "num_tokens": 47577811.0,
      "step": 20750
    },
    {
      "entropy": 5.463502693176269,
      "epoch": 1.993756003842459,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004604474851378124,
      "loss": 4.906,
      "mean_token_accuracy": 0.22441911697387695,
      "num_tokens": 47588309.0,
      "step": 20755
    },
    {
      "entropy": 5.476667737960815,
      "epoch": 1.994236311239193,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046042806879902803,
      "loss": 4.9802,
      "mean_token_accuracy": 0.2218422472476959,
      "num_tokens": 47599814.0,
      "step": 20760
    },
    {
      "entropy": 5.4068738460540775,
      "epoch": 1.9947166186359269,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004604086481551546,
      "loss": 4.8762,
      "mean_token_accuracy": 0.23246559351682664,
      "num_tokens": 47611851.0,
      "step": 20765
    },
    {
      "entropy": 5.3370061874389645,
      "epoch": 1.9951969260326607,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004603892232066428,
      "loss": 4.7271,
      "mean_token_accuracy": 0.24287843704223633,
      "num_tokens": 47624056.0,
      "step": 20770
    },
    {
      "entropy": 5.357308769226075,
      "epoch": 1.9956772334293948,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046036979395394374,
      "loss": 4.8421,
      "mean_token_accuracy": 0.2403175100684166,
      "num_tokens": 47637214.0,
      "step": 20775
    },
    {
      "entropy": 5.379515933990478,
      "epoch": 1.9961575408261287,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004603503603975085,
      "loss": 4.8039,
      "mean_token_accuracy": 0.23682049363851548,
      "num_tokens": 47647498.0,
      "step": 20780
    },
    {
      "entropy": 5.351403665542603,
      "epoch": 1.9966378482228626,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004603309225377883,
      "loss": 4.8008,
      "mean_token_accuracy": 0.2361682891845703,
      "num_tokens": 47659576.0,
      "step": 20785
    },
    {
      "entropy": 5.3923849105834964,
      "epoch": 1.9971181556195965,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004603114803752344,
      "loss": 4.8067,
      "mean_token_accuracy": 0.23475031405687333,
      "num_tokens": 47670631.0,
      "step": 20790
    },
    {
      "entropy": 5.336964893341064,
      "epoch": 1.9975984630163306,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046029203391029813,
      "loss": 4.7334,
      "mean_token_accuracy": 0.23478687107563018,
      "num_tokens": 47681647.0,
      "step": 20795
    },
    {
      "entropy": 5.3020288944244385,
      "epoch": 1.9980787704130645,
      "grad_norm": 1.0,
      "learning_rate": 0.00046027258314343107,
      "loss": 4.7786,
      "mean_token_accuracy": 0.241782608628273,
      "num_tokens": 47694386.0,
      "step": 20800
    },
    {
      "entropy": 5.390986347198487,
      "epoch": 1.9985590778097984,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046025312807508487,
      "loss": 4.8195,
      "mean_token_accuracy": 0.23522856682538987,
      "num_tokens": 47704897.0,
      "step": 20805
    },
    {
      "entropy": 5.38504114151001,
      "epoch": 1.9990393852065322,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046023366870571097,
      "loss": 4.869,
      "mean_token_accuracy": 0.23877871185541152,
      "num_tokens": 47715771.0,
      "step": 20810
    },
    {
      "entropy": 5.386336708068848,
      "epoch": 1.9995196926032661,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046021420503576145,
      "loss": 4.7533,
      "mean_token_accuracy": 0.2434265583753586,
      "num_tokens": 47726295.0,
      "step": 20815
    },
    {
      "entropy": 5.298949956893921,
      "epoch": 2.0,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004601947370656879,
      "loss": 4.8486,
      "mean_token_accuracy": 0.23349688053131104,
      "num_tokens": 47737072.0,
      "step": 20820
    },
    {
      "entropy": 5.383020305633545,
      "epoch": 2.000480307396734,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004601752647959426,
      "loss": 4.8023,
      "mean_token_accuracy": 0.2329091712832451,
      "num_tokens": 47750379.0,
      "step": 20825
    },
    {
      "entropy": 5.453515243530274,
      "epoch": 2.0009606147934678,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004601557882269775,
      "loss": 4.8344,
      "mean_token_accuracy": 0.23468188643455506,
      "num_tokens": 47763059.0,
      "step": 20830
    },
    {
      "entropy": 5.3504842758178714,
      "epoch": 2.0014409221902016,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004601363073592447,
      "loss": 4.731,
      "mean_token_accuracy": 0.2441660925745964,
      "num_tokens": 47775513.0,
      "step": 20835
    },
    {
      "entropy": 5.269275188446045,
      "epoch": 2.0019212295869355,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004601168221931967,
      "loss": 4.6787,
      "mean_token_accuracy": 0.24839745014905928,
      "num_tokens": 47786215.0,
      "step": 20840
    },
    {
      "entropy": 5.334464740753174,
      "epoch": 2.0024015369836694,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004600973327292857,
      "loss": 4.7592,
      "mean_token_accuracy": 0.24104916602373122,
      "num_tokens": 47796894.0,
      "step": 20845
    },
    {
      "entropy": 5.332740020751953,
      "epoch": 2.0028818443804033,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046007783896796436,
      "loss": 4.7824,
      "mean_token_accuracy": 0.23955798298120498,
      "num_tokens": 47808505.0,
      "step": 20850
    },
    {
      "entropy": 5.341226291656494,
      "epoch": 2.003362151777137,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004600583409096851,
      "loss": 4.6721,
      "mean_token_accuracy": 0.24632181525230407,
      "num_tokens": 47819617.0,
      "step": 20855
    },
    {
      "entropy": 5.307547950744629,
      "epoch": 2.0038424591738715,
      "grad_norm": 1.0,
      "learning_rate": 0.00046003883855490066,
      "loss": 4.7102,
      "mean_token_accuracy": 0.24946338981389998,
      "num_tokens": 47830655.0,
      "step": 20860
    },
    {
      "entropy": 5.287535762786865,
      "epoch": 2.0043227665706054,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004600193319040638,
      "loss": 4.6497,
      "mean_token_accuracy": 0.24971108883619308,
      "num_tokens": 47842439.0,
      "step": 20865
    },
    {
      "entropy": 5.23764328956604,
      "epoch": 2.0048030739673393,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045999982095762756,
      "loss": 4.6277,
      "mean_token_accuracy": 0.25381664633750917,
      "num_tokens": 47853327.0,
      "step": 20870
    },
    {
      "entropy": 5.370103454589843,
      "epoch": 2.005283381364073,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00045998030571604473,
      "loss": 4.7965,
      "mean_token_accuracy": 0.24174266010522844,
      "num_tokens": 47867155.0,
      "step": 20875
    },
    {
      "entropy": 5.353168535232544,
      "epoch": 2.005763688760807,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004599607861797685,
      "loss": 4.7481,
      "mean_token_accuracy": 0.24198179841041564,
      "num_tokens": 47879641.0,
      "step": 20880
    },
    {
      "entropy": 5.378943252563476,
      "epoch": 2.006243996157541,
      "grad_norm": 1.0,
      "learning_rate": 0.00045994126234925203,
      "loss": 4.7666,
      "mean_token_accuracy": 0.2377865269780159,
      "num_tokens": 47890512.0,
      "step": 20885
    },
    {
      "entropy": 5.399324035644531,
      "epoch": 2.006724303554275,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045992173422494865,
      "loss": 4.8124,
      "mean_token_accuracy": 0.23698397129774093,
      "num_tokens": 47902575.0,
      "step": 20890
    },
    {
      "entropy": 5.3433715343475345,
      "epoch": 2.0072046109510087,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004599022018073117,
      "loss": 4.6941,
      "mean_token_accuracy": 0.2514435604214668,
      "num_tokens": 47915447.0,
      "step": 20895
    },
    {
      "entropy": 5.350863218307495,
      "epoch": 2.0076849183477425,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004598826650967946,
      "loss": 4.7477,
      "mean_token_accuracy": 0.2387835279107094,
      "num_tokens": 47925702.0,
      "step": 20900
    },
    {
      "entropy": 5.353581333160401,
      "epoch": 2.0081652257444764,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045986312409385105,
      "loss": 4.7464,
      "mean_token_accuracy": 0.24300419986248017,
      "num_tokens": 47936621.0,
      "step": 20905
    },
    {
      "entropy": 5.297792673110962,
      "epoch": 2.0086455331412103,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004598435787989347,
      "loss": 4.6807,
      "mean_token_accuracy": 0.251143741607666,
      "num_tokens": 47948026.0,
      "step": 20910
    },
    {
      "entropy": 5.418527746200562,
      "epoch": 2.009125840537944,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045982402921249934,
      "loss": 4.7931,
      "mean_token_accuracy": 0.24130697697401046,
      "num_tokens": 47959976.0,
      "step": 20915
    },
    {
      "entropy": 5.401682949066162,
      "epoch": 2.009606147934678,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004598044753349988,
      "loss": 4.7944,
      "mean_token_accuracy": 0.23238254338502884,
      "num_tokens": 47972071.0,
      "step": 20920
    },
    {
      "entropy": 5.378173160552978,
      "epoch": 2.010086455331412,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00045978491716688706,
      "loss": 4.7079,
      "mean_token_accuracy": 0.24170660823583603,
      "num_tokens": 47984789.0,
      "step": 20925
    },
    {
      "entropy": 5.33974027633667,
      "epoch": 2.010566762728146,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004597653547086184,
      "loss": 4.7291,
      "mean_token_accuracy": 0.24780184030532837,
      "num_tokens": 47995661.0,
      "step": 20930
    },
    {
      "entropy": 5.395277261734009,
      "epoch": 2.0110470701248797,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004597457879606467,
      "loss": 4.8078,
      "mean_token_accuracy": 0.2367477908730507,
      "num_tokens": 48006043.0,
      "step": 20935
    },
    {
      "entropy": 5.386736726760864,
      "epoch": 2.011527377521614,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045972621692342636,
      "loss": 4.8354,
      "mean_token_accuracy": 0.24112777262926102,
      "num_tokens": 48016904.0,
      "step": 20940
    },
    {
      "entropy": 5.374846887588501,
      "epoch": 2.012007684918348,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045970664159741186,
      "loss": 4.76,
      "mean_token_accuracy": 0.23858283311128617,
      "num_tokens": 48026533.0,
      "step": 20945
    },
    {
      "entropy": 5.319024658203125,
      "epoch": 2.012487992315082,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045968706198305765,
      "loss": 4.7085,
      "mean_token_accuracy": 0.24728355407714844,
      "num_tokens": 48038280.0,
      "step": 20950
    },
    {
      "entropy": 5.259371042251587,
      "epoch": 2.0129682997118157,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045966747808081824,
      "loss": 4.6876,
      "mean_token_accuracy": 0.2481056272983551,
      "num_tokens": 48049663.0,
      "step": 20955
    },
    {
      "entropy": 5.35407190322876,
      "epoch": 2.0134486071085496,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004596478898911483,
      "loss": 4.7575,
      "mean_token_accuracy": 0.24126793295145035,
      "num_tokens": 48059826.0,
      "step": 20960
    },
    {
      "entropy": 5.365375804901123,
      "epoch": 2.0139289145052834,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045962829741450265,
      "loss": 4.6623,
      "mean_token_accuracy": 0.25181821286678313,
      "num_tokens": 48071097.0,
      "step": 20965
    },
    {
      "entropy": 5.358314371109008,
      "epoch": 2.0144092219020173,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004596087006513361,
      "loss": 4.7088,
      "mean_token_accuracy": 0.24086989015340804,
      "num_tokens": 48082297.0,
      "step": 20970
    },
    {
      "entropy": 5.292452001571656,
      "epoch": 2.014889529298751,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045958909960210385,
      "loss": 4.6752,
      "mean_token_accuracy": 0.243099245429039,
      "num_tokens": 48092946.0,
      "step": 20975
    },
    {
      "entropy": 5.275531911849976,
      "epoch": 2.015369836695485,
      "grad_norm": 0.921875,
      "learning_rate": 0.00045956949426726075,
      "loss": 4.7373,
      "mean_token_accuracy": 0.24587941765785218,
      "num_tokens": 48105000.0,
      "step": 20980
    },
    {
      "entropy": 5.307825183868408,
      "epoch": 2.015850144092219,
      "grad_norm": 1.0,
      "learning_rate": 0.00045954988464726203,
      "loss": 4.6917,
      "mean_token_accuracy": 0.24250535815954208,
      "num_tokens": 48117513.0,
      "step": 20985
    },
    {
      "entropy": 5.434242153167725,
      "epoch": 2.016330451488953,
      "grad_norm": 1.046875,
      "learning_rate": 0.000459530270742563,
      "loss": 4.9093,
      "mean_token_accuracy": 0.2257519245147705,
      "num_tokens": 48130895.0,
      "step": 20990
    },
    {
      "entropy": 5.338216352462768,
      "epoch": 2.0168107588856867,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045951065255361905,
      "loss": 4.6884,
      "mean_token_accuracy": 0.2425951763987541,
      "num_tokens": 48143348.0,
      "step": 20995
    },
    {
      "entropy": 5.354166793823242,
      "epoch": 2.0172910662824206,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004594910300808856,
      "loss": 4.7277,
      "mean_token_accuracy": 0.24661253094673158,
      "num_tokens": 48154959.0,
      "step": 21000
    },
    {
      "epoch": 2.0172910662824206,
      "eval_entropy": 5.109812836016985,
      "eval_loss": 4.855212211608887,
      "eval_mean_token_accuracy": 0.24462091985695802,
      "eval_num_tokens": 48154959.0,
      "eval_runtime": 20.6022,
      "eval_samples_per_second": 1592.791,
      "eval_steps_per_second": 199.105,
      "step": 21000
    },
    {
      "entropy": 5.290034484863281,
      "epoch": 2.0177713736791545,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004594714033248183,
      "loss": 4.6019,
      "mean_token_accuracy": 0.25204839557409286,
      "num_tokens": 48165227.0,
      "step": 21005
    },
    {
      "entropy": 5.397037363052368,
      "epoch": 2.0182516810758884,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004594517722858728,
      "loss": 4.8253,
      "mean_token_accuracy": 0.23692800253629684,
      "num_tokens": 48175133.0,
      "step": 21010
    },
    {
      "entropy": 5.396946954727173,
      "epoch": 2.0187319884726227,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045943213696450475,
      "loss": 4.7415,
      "mean_token_accuracy": 0.24556162506341933,
      "num_tokens": 48187554.0,
      "step": 21015
    },
    {
      "entropy": 5.386456441879273,
      "epoch": 2.0192122958693566,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045941249736117023,
      "loss": 4.7725,
      "mean_token_accuracy": 0.2372765824198723,
      "num_tokens": 48199410.0,
      "step": 21020
    },
    {
      "entropy": 5.28778657913208,
      "epoch": 2.0196926032660905,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004593928534763251,
      "loss": 4.7113,
      "mean_token_accuracy": 0.24394305795431137,
      "num_tokens": 48211124.0,
      "step": 21025
    },
    {
      "entropy": 5.308722496032715,
      "epoch": 2.0201729106628243,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004593732053104254,
      "loss": 4.6705,
      "mean_token_accuracy": 0.24703284502029418,
      "num_tokens": 48222690.0,
      "step": 21030
    },
    {
      "entropy": 5.330176258087159,
      "epoch": 2.020653218059558,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045935355286392735,
      "loss": 4.7081,
      "mean_token_accuracy": 0.24366649091243744,
      "num_tokens": 48235054.0,
      "step": 21035
    },
    {
      "entropy": 5.2703159809112545,
      "epoch": 2.021133525456292,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004593338961372873,
      "loss": 4.6548,
      "mean_token_accuracy": 0.25128018856048584,
      "num_tokens": 48246645.0,
      "step": 21040
    },
    {
      "entropy": 5.274085330963135,
      "epoch": 2.021613832853026,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004593142351309614,
      "loss": 4.6818,
      "mean_token_accuracy": 0.2496226519346237,
      "num_tokens": 48258149.0,
      "step": 21045
    },
    {
      "entropy": 5.359761905670166,
      "epoch": 2.02209414024976,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004592945698454064,
      "loss": 4.83,
      "mean_token_accuracy": 0.2371765524148941,
      "num_tokens": 48269892.0,
      "step": 21050
    },
    {
      "entropy": 5.414676475524902,
      "epoch": 2.0225744476464937,
      "grad_norm": 1.0,
      "learning_rate": 0.00045927490028107866,
      "loss": 4.8067,
      "mean_token_accuracy": 0.23603057265281677,
      "num_tokens": 48281087.0,
      "step": 21055
    },
    {
      "entropy": 5.302771329879761,
      "epoch": 2.0230547550432276,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000459255226438435,
      "loss": 4.6641,
      "mean_token_accuracy": 0.24952913373708724,
      "num_tokens": 48292177.0,
      "step": 21060
    },
    {
      "entropy": 5.293143653869629,
      "epoch": 2.0235350624399615,
      "grad_norm": 1.015625,
      "learning_rate": 0.000459235548317932,
      "loss": 4.7074,
      "mean_token_accuracy": 0.24041447341442107,
      "num_tokens": 48304281.0,
      "step": 21065
    },
    {
      "entropy": 5.305316638946533,
      "epoch": 2.0240153698366954,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045921586592002667,
      "loss": 4.6901,
      "mean_token_accuracy": 0.24445006400346755,
      "num_tokens": 48316758.0,
      "step": 21070
    },
    {
      "entropy": 5.427260017395019,
      "epoch": 2.0244956772334293,
      "grad_norm": 1.203125,
      "learning_rate": 0.000459196179245176,
      "loss": 4.829,
      "mean_token_accuracy": 0.23701524138450622,
      "num_tokens": 48327455.0,
      "step": 21075
    },
    {
      "entropy": 5.343733453750611,
      "epoch": 2.024975984630163,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004591764882938369,
      "loss": 4.7519,
      "mean_token_accuracy": 0.23701979368925094,
      "num_tokens": 48338105.0,
      "step": 21080
    },
    {
      "entropy": 5.372756433486939,
      "epoch": 2.025456292026897,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004591567930664667,
      "loss": 4.7958,
      "mean_token_accuracy": 0.23564729541540147,
      "num_tokens": 48349599.0,
      "step": 21085
    },
    {
      "entropy": 5.36819863319397,
      "epoch": 2.025936599423631,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004591370935635226,
      "loss": 4.8382,
      "mean_token_accuracy": 0.23259318470954896,
      "num_tokens": 48361488.0,
      "step": 21090
    },
    {
      "entropy": 5.400428581237793,
      "epoch": 2.0264169068203652,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004591173897854619,
      "loss": 4.7612,
      "mean_token_accuracy": 0.2373006075620651,
      "num_tokens": 48372049.0,
      "step": 21095
    },
    {
      "entropy": 5.321510314941406,
      "epoch": 2.026897214217099,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004590976817327422,
      "loss": 4.6604,
      "mean_token_accuracy": 0.2508836522698402,
      "num_tokens": 48385599.0,
      "step": 21100
    },
    {
      "entropy": 5.376580715179443,
      "epoch": 2.027377521613833,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004590779694058209,
      "loss": 4.7593,
      "mean_token_accuracy": 0.24639811664819716,
      "num_tokens": 48397390.0,
      "step": 21105
    },
    {
      "entropy": 5.344245958328247,
      "epoch": 2.027857829010567,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045905825280515586,
      "loss": 4.7895,
      "mean_token_accuracy": 0.23948260694742202,
      "num_tokens": 48409837.0,
      "step": 21110
    },
    {
      "entropy": 5.301026916503906,
      "epoch": 2.0283381364073008,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045903853193120464,
      "loss": 4.6317,
      "mean_token_accuracy": 0.25185379236936567,
      "num_tokens": 48420765.0,
      "step": 21115
    },
    {
      "entropy": 5.358480596542359,
      "epoch": 2.0288184438040346,
      "grad_norm": 1.0,
      "learning_rate": 0.00045901880678442524,
      "loss": 4.774,
      "mean_token_accuracy": 0.2450431615114212,
      "num_tokens": 48431939.0,
      "step": 21120
    },
    {
      "entropy": 5.411695957183838,
      "epoch": 2.0292987512007685,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045899907736527556,
      "loss": 4.8237,
      "mean_token_accuracy": 0.2332577645778656,
      "num_tokens": 48444820.0,
      "step": 21125
    },
    {
      "entropy": 5.3155049800872805,
      "epoch": 2.0297790585975024,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045897934367421364,
      "loss": 4.7582,
      "mean_token_accuracy": 0.24120995849370958,
      "num_tokens": 48457315.0,
      "step": 21130
    },
    {
      "entropy": 5.449026346206665,
      "epoch": 2.0302593659942363,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004589596057116977,
      "loss": 4.83,
      "mean_token_accuracy": 0.23113423734903335,
      "num_tokens": 48468612.0,
      "step": 21135
    },
    {
      "entropy": 5.319980239868164,
      "epoch": 2.03073967339097,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045893986347818593,
      "loss": 4.6781,
      "mean_token_accuracy": 0.24789415299892426,
      "num_tokens": 48479453.0,
      "step": 21140
    },
    {
      "entropy": 5.30315580368042,
      "epoch": 2.031219980787704,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004589201169741368,
      "loss": 4.7129,
      "mean_token_accuracy": 0.2522516489028931,
      "num_tokens": 48491656.0,
      "step": 21145
    },
    {
      "entropy": 5.2849365234375,
      "epoch": 2.031700288184438,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045890036620000856,
      "loss": 4.6742,
      "mean_token_accuracy": 0.23446068912744522,
      "num_tokens": 48503431.0,
      "step": 21150
    },
    {
      "entropy": 5.309907388687134,
      "epoch": 2.032180595581172,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045888061115626,
      "loss": 4.6932,
      "mean_token_accuracy": 0.24660854190587997,
      "num_tokens": 48513922.0,
      "step": 21155
    },
    {
      "entropy": 5.25793423652649,
      "epoch": 2.0326609029779057,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004588608518433496,
      "loss": 4.6774,
      "mean_token_accuracy": 0.24650042206048967,
      "num_tokens": 48524715.0,
      "step": 21160
    },
    {
      "entropy": 5.36615743637085,
      "epoch": 2.0331412103746396,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004588410882617362,
      "loss": 4.8103,
      "mean_token_accuracy": 0.24000100791454315,
      "num_tokens": 48536696.0,
      "step": 21165
    },
    {
      "entropy": 5.4124753952026365,
      "epoch": 2.0336215177713735,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004588213204118786,
      "loss": 4.8395,
      "mean_token_accuracy": 0.23888108879327774,
      "num_tokens": 48550290.0,
      "step": 21170
    },
    {
      "entropy": 5.338668537139893,
      "epoch": 2.034101825168108,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045880154829423586,
      "loss": 4.7944,
      "mean_token_accuracy": 0.24348049610853195,
      "num_tokens": 48561047.0,
      "step": 21175
    },
    {
      "entropy": 5.220439004898071,
      "epoch": 2.0345821325648417,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004587817719092668,
      "loss": 4.5697,
      "mean_token_accuracy": 0.25948929488658906,
      "num_tokens": 48571268.0,
      "step": 21180
    },
    {
      "entropy": 5.300819683074951,
      "epoch": 2.0350624399615755,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045876199125743087,
      "loss": 4.7071,
      "mean_token_accuracy": 0.2422064945101738,
      "num_tokens": 48583158.0,
      "step": 21185
    },
    {
      "entropy": 5.364986801147461,
      "epoch": 2.0355427473583094,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004587422063391871,
      "loss": 4.7752,
      "mean_token_accuracy": 0.24221312552690505,
      "num_tokens": 48593419.0,
      "step": 21190
    },
    {
      "entropy": 5.36146240234375,
      "epoch": 2.0360230547550433,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004587224171549949,
      "loss": 4.7536,
      "mean_token_accuracy": 0.2443439021706581,
      "num_tokens": 48605845.0,
      "step": 21195
    },
    {
      "entropy": 5.352344131469726,
      "epoch": 2.036503362151777,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045870262370531376,
      "loss": 4.7033,
      "mean_token_accuracy": 0.24835161119699478,
      "num_tokens": 48616503.0,
      "step": 21200
    },
    {
      "entropy": 5.389885091781617,
      "epoch": 2.036983669548511,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045868282599060314,
      "loss": 4.8023,
      "mean_token_accuracy": 0.24115507006645204,
      "num_tokens": 48628410.0,
      "step": 21205
    },
    {
      "entropy": 5.288307428359985,
      "epoch": 2.037463976945245,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004586630240113227,
      "loss": 4.6957,
      "mean_token_accuracy": 0.24389824271202087,
      "num_tokens": 48639457.0,
      "step": 21210
    },
    {
      "entropy": 5.29772253036499,
      "epoch": 2.037944284341979,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004586432177679322,
      "loss": 4.7499,
      "mean_token_accuracy": 0.23929579108953475,
      "num_tokens": 48652250.0,
      "step": 21215
    },
    {
      "entropy": 5.335483551025391,
      "epoch": 2.0384245917387127,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045862340726089153,
      "loss": 4.8062,
      "mean_token_accuracy": 0.2381945386528969,
      "num_tokens": 48663386.0,
      "step": 21220
    },
    {
      "entropy": 5.387906551361084,
      "epoch": 2.0389048991354466,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004586035924906606,
      "loss": 4.7903,
      "mean_token_accuracy": 0.23735613077878953,
      "num_tokens": 48675111.0,
      "step": 21225
    },
    {
      "entropy": 5.294706583023071,
      "epoch": 2.0393852065321805,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045858377345769946,
      "loss": 4.693,
      "mean_token_accuracy": 0.2424813374876976,
      "num_tokens": 48686264.0,
      "step": 21230
    },
    {
      "entropy": 5.346692323684692,
      "epoch": 2.0398655139289144,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004585639501624682,
      "loss": 4.7385,
      "mean_token_accuracy": 0.23811812102794647,
      "num_tokens": 48697249.0,
      "step": 21235
    },
    {
      "entropy": 5.2865574836730955,
      "epoch": 2.0403458213256482,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000458544122605427,
      "loss": 4.6946,
      "mean_token_accuracy": 0.24472663104534148,
      "num_tokens": 48708808.0,
      "step": 21240
    },
    {
      "entropy": 5.3886514663696286,
      "epoch": 2.040826128722382,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045852429078703646,
      "loss": 4.815,
      "mean_token_accuracy": 0.23943982571363448,
      "num_tokens": 48719879.0,
      "step": 21245
    },
    {
      "entropy": 5.366911697387695,
      "epoch": 2.0413064361191164,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045850445470775673,
      "loss": 4.7839,
      "mean_token_accuracy": 0.23906283527612687,
      "num_tokens": 48731110.0,
      "step": 21250
    },
    {
      "entropy": 5.319228219985962,
      "epoch": 2.0417867435158503,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004584846143680485,
      "loss": 4.7337,
      "mean_token_accuracy": 0.24477429538965226,
      "num_tokens": 48742799.0,
      "step": 21255
    },
    {
      "entropy": 5.457876873016358,
      "epoch": 2.042267050912584,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004584647697683723,
      "loss": 4.8734,
      "mean_token_accuracy": 0.23283659666776657,
      "num_tokens": 48753773.0,
      "step": 21260
    },
    {
      "entropy": 5.362517070770264,
      "epoch": 2.042747358309318,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045844492090918904,
      "loss": 4.7399,
      "mean_token_accuracy": 0.24527440071105958,
      "num_tokens": 48764561.0,
      "step": 21265
    },
    {
      "entropy": 5.256858825683594,
      "epoch": 2.043227665706052,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045842506779095936,
      "loss": 4.6303,
      "mean_token_accuracy": 0.24584768265485762,
      "num_tokens": 48775319.0,
      "step": 21270
    },
    {
      "entropy": 5.429414987564087,
      "epoch": 2.043707973102786,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004584052104141443,
      "loss": 4.8721,
      "mean_token_accuracy": 0.22454794049263,
      "num_tokens": 48787595.0,
      "step": 21275
    },
    {
      "entropy": 5.427281618118286,
      "epoch": 2.0441882804995197,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004583853487792048,
      "loss": 4.7333,
      "mean_token_accuracy": 0.24210079461336137,
      "num_tokens": 48800015.0,
      "step": 21280
    },
    {
      "entropy": 5.382694244384766,
      "epoch": 2.0446685878962536,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004583654828866021,
      "loss": 4.7765,
      "mean_token_accuracy": 0.2355405107140541,
      "num_tokens": 48811651.0,
      "step": 21285
    },
    {
      "entropy": 5.3641656875610355,
      "epoch": 2.0451488952929875,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004583456127367973,
      "loss": 4.7698,
      "mean_token_accuracy": 0.24159234017133713,
      "num_tokens": 48823561.0,
      "step": 21290
    },
    {
      "entropy": 5.352596473693848,
      "epoch": 2.0456292026897214,
      "grad_norm": 1.125,
      "learning_rate": 0.0004583257383302519,
      "loss": 4.8158,
      "mean_token_accuracy": 0.23239808082580565,
      "num_tokens": 48835385.0,
      "step": 21295
    },
    {
      "entropy": 5.398534774780273,
      "epoch": 2.0461095100864553,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004583058596674271,
      "loss": 4.8089,
      "mean_token_accuracy": 0.23776627480983734,
      "num_tokens": 48846899.0,
      "step": 21300
    },
    {
      "entropy": 5.378354549407959,
      "epoch": 2.046589817483189,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004582859767487846,
      "loss": 4.7432,
      "mean_token_accuracy": 0.24431652426719666,
      "num_tokens": 48857850.0,
      "step": 21305
    },
    {
      "entropy": 5.33493824005127,
      "epoch": 2.047070124879923,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045826608957478604,
      "loss": 4.7394,
      "mean_token_accuracy": 0.24140822887420654,
      "num_tokens": 48869192.0,
      "step": 21310
    },
    {
      "entropy": 5.312749290466309,
      "epoch": 2.047550432276657,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045824619814589297,
      "loss": 4.7093,
      "mean_token_accuracy": 0.2477246940135956,
      "num_tokens": 48881314.0,
      "step": 21315
    },
    {
      "entropy": 5.305272245407105,
      "epoch": 2.0480307396733908,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004582263024625674,
      "loss": 4.7801,
      "mean_token_accuracy": 0.23990575820207596,
      "num_tokens": 48893292.0,
      "step": 21320
    },
    {
      "entropy": 5.319245433807373,
      "epoch": 2.048511047070125,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004582064025252711,
      "loss": 4.8383,
      "mean_token_accuracy": 0.23401868492364883,
      "num_tokens": 48905607.0,
      "step": 21325
    },
    {
      "entropy": 5.389092445373535,
      "epoch": 2.048991354466859,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004581864983344661,
      "loss": 4.7209,
      "mean_token_accuracy": 0.24498866200447084,
      "num_tokens": 48916194.0,
      "step": 21330
    },
    {
      "entropy": 5.329754972457886,
      "epoch": 2.049471661863593,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004581665898906147,
      "loss": 4.7268,
      "mean_token_accuracy": 0.2440531447529793,
      "num_tokens": 48927560.0,
      "step": 21335
    },
    {
      "entropy": 5.289850521087646,
      "epoch": 2.0499519692603267,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045814667719417887,
      "loss": 4.7296,
      "mean_token_accuracy": 0.24423727095127107,
      "num_tokens": 48938284.0,
      "step": 21340
    },
    {
      "entropy": 5.354289817810058,
      "epoch": 2.0504322766570606,
      "grad_norm": 1.078125,
      "learning_rate": 0.000458126760245621,
      "loss": 4.7317,
      "mean_token_accuracy": 0.2472059205174446,
      "num_tokens": 48948325.0,
      "step": 21345
    },
    {
      "entropy": 5.333071565628051,
      "epoch": 2.0509125840537945,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004581068390454036,
      "loss": 4.729,
      "mean_token_accuracy": 0.2400486320257187,
      "num_tokens": 48960020.0,
      "step": 21350
    },
    {
      "entropy": 5.314674425125122,
      "epoch": 2.0513928914505284,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045808691359398905,
      "loss": 4.6974,
      "mean_token_accuracy": 0.2422620713710785,
      "num_tokens": 48970833.0,
      "step": 21355
    },
    {
      "entropy": 5.337756395339966,
      "epoch": 2.0518731988472623,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004580669838918401,
      "loss": 4.7558,
      "mean_token_accuracy": 0.24695815593004228,
      "num_tokens": 48981821.0,
      "step": 21360
    },
    {
      "entropy": 5.39783034324646,
      "epoch": 2.052353506243996,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045804704993941935,
      "loss": 4.7473,
      "mean_token_accuracy": 0.23907109647989272,
      "num_tokens": 48992341.0,
      "step": 21365
    },
    {
      "entropy": 5.37365231513977,
      "epoch": 2.05283381364073,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045802711173718966,
      "loss": 4.8349,
      "mean_token_accuracy": 0.2349153757095337,
      "num_tokens": 49003063.0,
      "step": 21370
    },
    {
      "entropy": 5.331434822082519,
      "epoch": 2.053314121037464,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004580071692856138,
      "loss": 4.8181,
      "mean_token_accuracy": 0.2420615017414093,
      "num_tokens": 49015353.0,
      "step": 21375
    },
    {
      "entropy": 5.295809316635132,
      "epoch": 2.053794428434198,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045798722258515504,
      "loss": 4.6989,
      "mean_token_accuracy": 0.24534988552331924,
      "num_tokens": 49026405.0,
      "step": 21380
    },
    {
      "entropy": 5.412870264053344,
      "epoch": 2.0542747358309317,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045796727163627623,
      "loss": 4.7921,
      "mean_token_accuracy": 0.2366405889391899,
      "num_tokens": 49038433.0,
      "step": 21385
    },
    {
      "entropy": 5.298480892181397,
      "epoch": 2.0547550432276656,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004579473164394408,
      "loss": 4.6389,
      "mean_token_accuracy": 0.25175449401140215,
      "num_tokens": 49049463.0,
      "step": 21390
    },
    {
      "entropy": 5.184693050384522,
      "epoch": 2.0552353506243994,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045792735699511176,
      "loss": 4.6211,
      "mean_token_accuracy": 0.24900815188884734,
      "num_tokens": 49061607.0,
      "step": 21395
    },
    {
      "entropy": 5.37518801689148,
      "epoch": 2.0557156580211333,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045790739330375276,
      "loss": 4.7872,
      "mean_token_accuracy": 0.23516640812158585,
      "num_tokens": 49074687.0,
      "step": 21400
    },
    {
      "entropy": 5.334755325317383,
      "epoch": 2.0561959654178676,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045788742536582717,
      "loss": 4.6644,
      "mean_token_accuracy": 0.248280930519104,
      "num_tokens": 49086151.0,
      "step": 21405
    },
    {
      "entropy": 5.301209354400635,
      "epoch": 2.0566762728146015,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045786745318179866,
      "loss": 4.7785,
      "mean_token_accuracy": 0.2352563664317131,
      "num_tokens": 49097418.0,
      "step": 21410
    },
    {
      "entropy": 5.268042469024659,
      "epoch": 2.0571565802113354,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004578474767521309,
      "loss": 4.703,
      "mean_token_accuracy": 0.2509147360920906,
      "num_tokens": 49108570.0,
      "step": 21415
    },
    {
      "entropy": 5.275310707092285,
      "epoch": 2.0576368876080693,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045782749607728765,
      "loss": 4.76,
      "mean_token_accuracy": 0.23310210704803466,
      "num_tokens": 49120906.0,
      "step": 21420
    },
    {
      "entropy": 5.3213889598846436,
      "epoch": 2.058117195004803,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045780751115773286,
      "loss": 4.6831,
      "mean_token_accuracy": 0.24659384340047835,
      "num_tokens": 49132040.0,
      "step": 21425
    },
    {
      "entropy": 5.353364753723144,
      "epoch": 2.058597502401537,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004577875219939304,
      "loss": 4.8016,
      "mean_token_accuracy": 0.24267542660236358,
      "num_tokens": 49143427.0,
      "step": 21430
    },
    {
      "entropy": 5.368893098831177,
      "epoch": 2.059077809798271,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004577675285863446,
      "loss": 4.7529,
      "mean_token_accuracy": 0.23651560246944428,
      "num_tokens": 49155351.0,
      "step": 21435
    },
    {
      "entropy": 5.3643182754516605,
      "epoch": 2.059558117195005,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045774753093543943,
      "loss": 4.7693,
      "mean_token_accuracy": 0.24268012046813964,
      "num_tokens": 49167898.0,
      "step": 21440
    },
    {
      "entropy": 5.316374158859253,
      "epoch": 2.0600384245917387,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004577275290416791,
      "loss": 4.7636,
      "mean_token_accuracy": 0.23579960465431213,
      "num_tokens": 49179209.0,
      "step": 21445
    },
    {
      "entropy": 5.315546941757202,
      "epoch": 2.0605187319884726,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004577075229055283,
      "loss": 4.6895,
      "mean_token_accuracy": 0.24915789663791657,
      "num_tokens": 49190382.0,
      "step": 21450
    },
    {
      "entropy": 5.309725570678711,
      "epoch": 2.0609990393852065,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045768751252745133,
      "loss": 4.7354,
      "mean_token_accuracy": 0.24884034395217897,
      "num_tokens": 49203511.0,
      "step": 21455
    },
    {
      "entropy": 5.414013385772705,
      "epoch": 2.0614793467819403,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045766749790791274,
      "loss": 4.7616,
      "mean_token_accuracy": 0.24029064029455185,
      "num_tokens": 49214276.0,
      "step": 21460
    },
    {
      "entropy": 5.2196526527404785,
      "epoch": 2.061959654178674,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004576474790473773,
      "loss": 4.6351,
      "mean_token_accuracy": 0.25165160894393923,
      "num_tokens": 49225803.0,
      "step": 21465
    },
    {
      "entropy": 5.313275671005249,
      "epoch": 2.062439961575408,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045762745594630973,
      "loss": 4.769,
      "mean_token_accuracy": 0.23772428929805756,
      "num_tokens": 49237187.0,
      "step": 21470
    },
    {
      "entropy": 5.435977458953857,
      "epoch": 2.062920268972142,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004576074286051749,
      "loss": 4.8083,
      "mean_token_accuracy": 0.23641644418239594,
      "num_tokens": 49250151.0,
      "step": 21475
    },
    {
      "entropy": 5.296788167953491,
      "epoch": 2.063400576368876,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045758739702443787,
      "loss": 4.6545,
      "mean_token_accuracy": 0.2556405529379845,
      "num_tokens": 49261970.0,
      "step": 21480
    },
    {
      "entropy": 5.323768138885498,
      "epoch": 2.06388088376561,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004575673612045636,
      "loss": 4.7147,
      "mean_token_accuracy": 0.2497069552540779,
      "num_tokens": 49271427.0,
      "step": 21485
    },
    {
      "entropy": 5.335181808471679,
      "epoch": 2.064361191162344,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004575473211460173,
      "loss": 4.7518,
      "mean_token_accuracy": 0.24204548448324203,
      "num_tokens": 49285102.0,
      "step": 21490
    },
    {
      "entropy": 5.362382650375366,
      "epoch": 2.064841498559078,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004575272768492644,
      "loss": 4.7466,
      "mean_token_accuracy": 0.24451406300067902,
      "num_tokens": 49297569.0,
      "step": 21495
    },
    {
      "entropy": 5.336537504196167,
      "epoch": 2.065321805955812,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045750722831476993,
      "loss": 4.6817,
      "mean_token_accuracy": 0.24896909445524215,
      "num_tokens": 49308795.0,
      "step": 21500
    },
    {
      "entropy": 5.375980520248413,
      "epoch": 2.0658021133525457,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045748717554299964,
      "loss": 4.7898,
      "mean_token_accuracy": 0.24192394614219664,
      "num_tokens": 49320971.0,
      "step": 21505
    },
    {
      "entropy": 5.422058153152466,
      "epoch": 2.0662824207492796,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000457467118534419,
      "loss": 4.8571,
      "mean_token_accuracy": 0.23217440247535706,
      "num_tokens": 49333624.0,
      "step": 21510
    },
    {
      "entropy": 5.461378145217895,
      "epoch": 2.0667627281460135,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004574470572894938,
      "loss": 4.8583,
      "mean_token_accuracy": 0.23416798561811447,
      "num_tokens": 49344863.0,
      "step": 21515
    },
    {
      "entropy": 5.348563146591187,
      "epoch": 2.0672430355427474,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004574269918086895,
      "loss": 4.7283,
      "mean_token_accuracy": 0.24484729766845703,
      "num_tokens": 49355501.0,
      "step": 21520
    },
    {
      "entropy": 5.339051914215088,
      "epoch": 2.0677233429394812,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004574069220924722,
      "loss": 4.7855,
      "mean_token_accuracy": 0.24667538702487946,
      "num_tokens": 49366384.0,
      "step": 21525
    },
    {
      "entropy": 5.414738941192627,
      "epoch": 2.068203650336215,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004573868481413079,
      "loss": 4.887,
      "mean_token_accuracy": 0.2265598878264427,
      "num_tokens": 49378670.0,
      "step": 21530
    },
    {
      "entropy": 5.352966022491455,
      "epoch": 2.068683957732949,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045736676995566244,
      "loss": 4.7179,
      "mean_token_accuracy": 0.24022359102964402,
      "num_tokens": 49390273.0,
      "step": 21535
    },
    {
      "entropy": 5.387016773223877,
      "epoch": 2.069164265129683,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045734668753600217,
      "loss": 4.7874,
      "mean_token_accuracy": 0.2407538115978241,
      "num_tokens": 49402503.0,
      "step": 21540
    },
    {
      "entropy": 5.361164331436157,
      "epoch": 2.0696445725264168,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045732660088279326,
      "loss": 4.7396,
      "mean_token_accuracy": 0.2388540983200073,
      "num_tokens": 49413948.0,
      "step": 21545
    },
    {
      "entropy": 5.318202066421509,
      "epoch": 2.0701248799231506,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045730650999650216,
      "loss": 4.6997,
      "mean_token_accuracy": 0.2502593740820885,
      "num_tokens": 49424164.0,
      "step": 21550
    },
    {
      "entropy": 5.4151145935058596,
      "epoch": 2.0706051873198845,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045728641487759506,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22442456036806108,
      "num_tokens": 49435102.0,
      "step": 21555
    },
    {
      "entropy": 5.298818492889405,
      "epoch": 2.071085494716619,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004572663155265388,
      "loss": 4.7265,
      "mean_token_accuracy": 0.24677687287330627,
      "num_tokens": 49445445.0,
      "step": 21560
    },
    {
      "entropy": 5.308383178710938,
      "epoch": 2.0715658021133527,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004572462119437999,
      "loss": 4.7173,
      "mean_token_accuracy": 0.2426982581615448,
      "num_tokens": 49457578.0,
      "step": 21565
    },
    {
      "entropy": 5.26834282875061,
      "epoch": 2.0720461095100866,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045722610412984513,
      "loss": 4.7089,
      "mean_token_accuracy": 0.25143365412950514,
      "num_tokens": 49470253.0,
      "step": 21570
    },
    {
      "entropy": 5.418248558044434,
      "epoch": 2.0725264169068205,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004572059920851412,
      "loss": 4.7915,
      "mean_token_accuracy": 0.2350108504295349,
      "num_tokens": 49481854.0,
      "step": 21575
    },
    {
      "entropy": 5.440323734283448,
      "epoch": 2.0730067243035544,
      "grad_norm": 0.921875,
      "learning_rate": 0.00045718587581015534,
      "loss": 4.8113,
      "mean_token_accuracy": 0.23503387868404388,
      "num_tokens": 49494311.0,
      "step": 21580
    },
    {
      "entropy": 5.355147123336792,
      "epoch": 2.0734870317002883,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004571657553053544,
      "loss": 4.7466,
      "mean_token_accuracy": 0.24655626267194747,
      "num_tokens": 49506186.0,
      "step": 21585
    },
    {
      "entropy": 5.3490739345550535,
      "epoch": 2.073967339097022,
      "grad_norm": 1.0,
      "learning_rate": 0.0004571456305712055,
      "loss": 4.7552,
      "mean_token_accuracy": 0.23621722161769867,
      "num_tokens": 49517841.0,
      "step": 21590
    },
    {
      "entropy": 5.351836442947388,
      "epoch": 2.074447646493756,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004571255016081759,
      "loss": 4.7192,
      "mean_token_accuracy": 0.24269696325063705,
      "num_tokens": 49529260.0,
      "step": 21595
    },
    {
      "entropy": 5.32754693031311,
      "epoch": 2.07492795389049,
      "grad_norm": 1.046875,
      "learning_rate": 0.000457105368416733,
      "loss": 4.7134,
      "mean_token_accuracy": 0.24901265501976014,
      "num_tokens": 49540892.0,
      "step": 21600
    },
    {
      "entropy": 5.426615238189697,
      "epoch": 2.0754082612872238,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045708523099734417,
      "loss": 4.7699,
      "mean_token_accuracy": 0.24187403172254562,
      "num_tokens": 49550924.0,
      "step": 21605
    },
    {
      "entropy": 5.341100215911865,
      "epoch": 2.0758885686839577,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045706508935047693,
      "loss": 4.7762,
      "mean_token_accuracy": 0.24325562119483948,
      "num_tokens": 49562424.0,
      "step": 21610
    },
    {
      "entropy": 5.278396034240723,
      "epoch": 2.0763688760806915,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000457044943476599,
      "loss": 4.7024,
      "mean_token_accuracy": 0.24477345049381255,
      "num_tokens": 49573341.0,
      "step": 21615
    },
    {
      "entropy": 5.390311574935913,
      "epoch": 2.0768491834774254,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045702479337617795,
      "loss": 4.6856,
      "mean_token_accuracy": 0.24397279918193818,
      "num_tokens": 49584047.0,
      "step": 21620
    },
    {
      "entropy": 5.292179441452026,
      "epoch": 2.0773294908741593,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004570046390496818,
      "loss": 4.716,
      "mean_token_accuracy": 0.24419501721858977,
      "num_tokens": 49595898.0,
      "step": 21625
    },
    {
      "entropy": 5.31299319267273,
      "epoch": 2.077809798270893,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004569844804975783,
      "loss": 4.7591,
      "mean_token_accuracy": 0.2418929934501648,
      "num_tokens": 49607865.0,
      "step": 21630
    },
    {
      "entropy": 5.424623203277588,
      "epoch": 2.0782901056676275,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004569643177203356,
      "loss": 4.8202,
      "mean_token_accuracy": 0.23764758259058,
      "num_tokens": 49619513.0,
      "step": 21635
    },
    {
      "entropy": 5.357638597488403,
      "epoch": 2.0787704130643614,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004569441507184218,
      "loss": 4.7099,
      "mean_token_accuracy": 0.2506066381931305,
      "num_tokens": 49631232.0,
      "step": 21640
    },
    {
      "entropy": 5.342900276184082,
      "epoch": 2.0792507204610953,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045692397949230495,
      "loss": 4.7208,
      "mean_token_accuracy": 0.24328146427869796,
      "num_tokens": 49642653.0,
      "step": 21645
    },
    {
      "entropy": 5.309677410125732,
      "epoch": 2.079731027857829,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045690380404245364,
      "loss": 4.6896,
      "mean_token_accuracy": 0.24417025595903397,
      "num_tokens": 49654479.0,
      "step": 21650
    },
    {
      "entropy": 5.334396314620972,
      "epoch": 2.080211335254563,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00045688362436933607,
      "loss": 4.7577,
      "mean_token_accuracy": 0.23897801339626312,
      "num_tokens": 49666461.0,
      "step": 21655
    },
    {
      "entropy": 5.337480211257935,
      "epoch": 2.080691642651297,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004568634404734208,
      "loss": 4.797,
      "mean_token_accuracy": 0.24040066599845886,
      "num_tokens": 49676602.0,
      "step": 21660
    },
    {
      "entropy": 5.298612022399903,
      "epoch": 2.081171950048031,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004568432523551765,
      "loss": 4.7242,
      "mean_token_accuracy": 0.2443593665957451,
      "num_tokens": 49689225.0,
      "step": 21665
    },
    {
      "entropy": 5.333560276031494,
      "epoch": 2.0816522574447647,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004568230600150718,
      "loss": 4.7562,
      "mean_token_accuracy": 0.24433075040578842,
      "num_tokens": 49700827.0,
      "step": 21670
    },
    {
      "entropy": 5.35960545539856,
      "epoch": 2.0821325648414986,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004568028634535757,
      "loss": 4.7812,
      "mean_token_accuracy": 0.2398593693971634,
      "num_tokens": 49712777.0,
      "step": 21675
    },
    {
      "entropy": 5.325171566009521,
      "epoch": 2.0826128722382324,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004567826626711568,
      "loss": 4.7433,
      "mean_token_accuracy": 0.24779264479875565,
      "num_tokens": 49725234.0,
      "step": 21680
    },
    {
      "entropy": 5.322592163085938,
      "epoch": 2.0830931796349663,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004567624576682843,
      "loss": 4.7057,
      "mean_token_accuracy": 0.2460240200161934,
      "num_tokens": 49736933.0,
      "step": 21685
    },
    {
      "entropy": 5.437100839614868,
      "epoch": 2.0835734870317,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004567422484454273,
      "loss": 4.7889,
      "mean_token_accuracy": 0.23864771872758866,
      "num_tokens": 49748279.0,
      "step": 21690
    },
    {
      "entropy": 5.282914686203003,
      "epoch": 2.084053794428434,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045672203500305493,
      "loss": 4.7059,
      "mean_token_accuracy": 0.2536537691950798,
      "num_tokens": 49760874.0,
      "step": 21695
    },
    {
      "entropy": 5.278042221069336,
      "epoch": 2.084534101825168,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045670181734163654,
      "loss": 4.6358,
      "mean_token_accuracy": 0.24929590672254562,
      "num_tokens": 49772520.0,
      "step": 21700
    },
    {
      "entropy": 5.301714134216309,
      "epoch": 2.085014409221902,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004566815954616414,
      "loss": 4.7174,
      "mean_token_accuracy": 0.24473112672567368,
      "num_tokens": 49783019.0,
      "step": 21705
    },
    {
      "entropy": 5.2461323738098145,
      "epoch": 2.0854947166186357,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045666136936353913,
      "loss": 4.7115,
      "mean_token_accuracy": 0.2468599408864975,
      "num_tokens": 49795111.0,
      "step": 21710
    },
    {
      "entropy": 5.276230573654175,
      "epoch": 2.08597502401537,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004566411390477993,
      "loss": 4.6442,
      "mean_token_accuracy": 0.2562761977314949,
      "num_tokens": 49807022.0,
      "step": 21715
    },
    {
      "entropy": 5.3702209949493405,
      "epoch": 2.086455331412104,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045662090451489156,
      "loss": 4.7794,
      "mean_token_accuracy": 0.23602993190288543,
      "num_tokens": 49819225.0,
      "step": 21720
    },
    {
      "entropy": 5.360667181015015,
      "epoch": 2.086935638808838,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045660066576528577,
      "loss": 4.7346,
      "mean_token_accuracy": 0.2475792035460472,
      "num_tokens": 49830989.0,
      "step": 21725
    },
    {
      "entropy": 5.348718452453613,
      "epoch": 2.0874159462055717,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004565804227994518,
      "loss": 4.7182,
      "mean_token_accuracy": 0.24425677061080933,
      "num_tokens": 49841618.0,
      "step": 21730
    },
    {
      "entropy": 5.280185985565185,
      "epoch": 2.0878962536023056,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004565601756178595,
      "loss": 4.5908,
      "mean_token_accuracy": 0.2579324170947075,
      "num_tokens": 49851985.0,
      "step": 21735
    },
    {
      "entropy": 5.331367826461792,
      "epoch": 2.0883765609990395,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004565399242209791,
      "loss": 4.8064,
      "mean_token_accuracy": 0.23768426477909088,
      "num_tokens": 49863383.0,
      "step": 21740
    },
    {
      "entropy": 5.326331901550293,
      "epoch": 2.0888568683957733,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004565196686092807,
      "loss": 4.7229,
      "mean_token_accuracy": 0.24766570329666138,
      "num_tokens": 49875757.0,
      "step": 21745
    },
    {
      "entropy": 5.3431929588317875,
      "epoch": 2.089337175792507,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004564994087832346,
      "loss": 4.7157,
      "mean_token_accuracy": 0.24904207289218902,
      "num_tokens": 49888419.0,
      "step": 21750
    },
    {
      "entropy": 5.353397035598755,
      "epoch": 2.089817483189241,
      "grad_norm": 1.125,
      "learning_rate": 0.00045647914474331123,
      "loss": 4.7783,
      "mean_token_accuracy": 0.23832125812768937,
      "num_tokens": 49899629.0,
      "step": 21755
    },
    {
      "entropy": 5.4490087032318115,
      "epoch": 2.090297790585975,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045645887648998094,
      "loss": 4.7838,
      "mean_token_accuracy": 0.24203052520751953,
      "num_tokens": 49909858.0,
      "step": 21760
    },
    {
      "entropy": 5.324327039718628,
      "epoch": 2.090778097982709,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045643860402371433,
      "loss": 4.7573,
      "mean_token_accuracy": 0.24026572853326797,
      "num_tokens": 49921305.0,
      "step": 21765
    },
    {
      "entropy": 5.341726064682007,
      "epoch": 2.0912584053794427,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004564183273449821,
      "loss": 4.7644,
      "mean_token_accuracy": 0.24171933084726333,
      "num_tokens": 49931966.0,
      "step": 21770
    },
    {
      "entropy": 5.34996542930603,
      "epoch": 2.0917387127761766,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004563980464542551,
      "loss": 4.6861,
      "mean_token_accuracy": 0.23853109925985336,
      "num_tokens": 49943781.0,
      "step": 21775
    },
    {
      "entropy": 5.373552227020264,
      "epoch": 2.0922190201729105,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045637776135200406,
      "loss": 4.7348,
      "mean_token_accuracy": 0.24122607558965684,
      "num_tokens": 49954149.0,
      "step": 21780
    },
    {
      "entropy": 5.412948322296143,
      "epoch": 2.0926993275696444,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004563574720386999,
      "loss": 4.7847,
      "mean_token_accuracy": 0.24245233684778214,
      "num_tokens": 49965542.0,
      "step": 21785
    },
    {
      "entropy": 5.349530267715454,
      "epoch": 2.0931796349663783,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004563371785148139,
      "loss": 4.7294,
      "mean_token_accuracy": 0.2406831279397011,
      "num_tokens": 49977444.0,
      "step": 21790
    },
    {
      "entropy": 5.307690286636353,
      "epoch": 2.0936599423631126,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045631688078081695,
      "loss": 4.7072,
      "mean_token_accuracy": 0.24434132874011993,
      "num_tokens": 49988409.0,
      "step": 21795
    },
    {
      "entropy": 5.350734758377075,
      "epoch": 2.0941402497598465,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004562965788371805,
      "loss": 4.7689,
      "mean_token_accuracy": 0.2389935925602913,
      "num_tokens": 49999639.0,
      "step": 21800
    },
    {
      "entropy": 5.359672546386719,
      "epoch": 2.0946205571565804,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004562762726843758,
      "loss": 4.7132,
      "mean_token_accuracy": 0.24335898011922835,
      "num_tokens": 50010964.0,
      "step": 21805
    },
    {
      "entropy": 5.433216142654419,
      "epoch": 2.0951008645533142,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045625596232287436,
      "loss": 4.8481,
      "mean_token_accuracy": 0.2299287810921669,
      "num_tokens": 50024487.0,
      "step": 21810
    },
    {
      "entropy": 5.416201066970825,
      "epoch": 2.095581171950048,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004562356477531477,
      "loss": 4.8904,
      "mean_token_accuracy": 0.23180729150772095,
      "num_tokens": 50037202.0,
      "step": 21815
    },
    {
      "entropy": 5.340386056900025,
      "epoch": 2.096061479346782,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004562153289756674,
      "loss": 4.6626,
      "mean_token_accuracy": 0.2501459583640099,
      "num_tokens": 50048520.0,
      "step": 21820
    },
    {
      "entropy": 5.409684944152832,
      "epoch": 2.096541786743516,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004561950059909053,
      "loss": 4.7916,
      "mean_token_accuracy": 0.2371048077940941,
      "num_tokens": 50058523.0,
      "step": 21825
    },
    {
      "entropy": 5.378475141525269,
      "epoch": 2.0970220941402498,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004561746787993332,
      "loss": 4.7637,
      "mean_token_accuracy": 0.2397240623831749,
      "num_tokens": 50070241.0,
      "step": 21830
    },
    {
      "entropy": 5.365769624710083,
      "epoch": 2.0975024015369836,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045615434740142307,
      "loss": 4.7831,
      "mean_token_accuracy": 0.24456010907888412,
      "num_tokens": 50081326.0,
      "step": 21835
    },
    {
      "entropy": 5.28947377204895,
      "epoch": 2.0979827089337175,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045613401179764686,
      "loss": 4.6413,
      "mean_token_accuracy": 0.25025185495615004,
      "num_tokens": 50091767.0,
      "step": 21840
    },
    {
      "entropy": 5.345547246932983,
      "epoch": 2.0984630163304514,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045611367198847676,
      "loss": 4.7886,
      "mean_token_accuracy": 0.2369891256093979,
      "num_tokens": 50103435.0,
      "step": 21845
    },
    {
      "entropy": 5.393610429763794,
      "epoch": 2.0989433237271853,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000456093327974385,
      "loss": 4.7446,
      "mean_token_accuracy": 0.24280955046415328,
      "num_tokens": 50114127.0,
      "step": 21850
    },
    {
      "entropy": 5.359191465377807,
      "epoch": 2.099423631123919,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004560729797558438,
      "loss": 4.7621,
      "mean_token_accuracy": 0.24238914400339126,
      "num_tokens": 50126328.0,
      "step": 21855
    },
    {
      "entropy": 5.376317501068115,
      "epoch": 2.099903938520653,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004560526273333259,
      "loss": 4.7236,
      "mean_token_accuracy": 0.24391075521707534,
      "num_tokens": 50137600.0,
      "step": 21860
    },
    {
      "entropy": 5.390663385391235,
      "epoch": 2.100384245917387,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045603227070730346,
      "loss": 4.8034,
      "mean_token_accuracy": 0.23654671013355255,
      "num_tokens": 50148711.0,
      "step": 21865
    },
    {
      "entropy": 5.341365003585816,
      "epoch": 2.1008645533141213,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00045601190987824933,
      "loss": 4.7686,
      "mean_token_accuracy": 0.23718400597572326,
      "num_tokens": 50161141.0,
      "step": 21870
    },
    {
      "entropy": 5.360894775390625,
      "epoch": 2.101344860710855,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045599154484663606,
      "loss": 4.7705,
      "mean_token_accuracy": 0.24140380471944808,
      "num_tokens": 50173145.0,
      "step": 21875
    },
    {
      "entropy": 5.351045703887939,
      "epoch": 2.101825168107589,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045597117561293663,
      "loss": 4.7441,
      "mean_token_accuracy": 0.2420676350593567,
      "num_tokens": 50184074.0,
      "step": 21880
    },
    {
      "entropy": 5.509739875793457,
      "epoch": 2.102305475504323,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004559508021776238,
      "loss": 4.9832,
      "mean_token_accuracy": 0.22515525817871093,
      "num_tokens": 50195004.0,
      "step": 21885
    },
    {
      "entropy": 5.343635177612304,
      "epoch": 2.1027857829010568,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004559304245411707,
      "loss": 4.71,
      "mean_token_accuracy": 0.2452000305056572,
      "num_tokens": 50205329.0,
      "step": 21890
    },
    {
      "entropy": 5.366356372833252,
      "epoch": 2.1032660902977907,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045591004270405044,
      "loss": 4.7628,
      "mean_token_accuracy": 0.2399482935667038,
      "num_tokens": 50218346.0,
      "step": 21895
    },
    {
      "entropy": 5.33423490524292,
      "epoch": 2.1037463976945245,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004558896566667361,
      "loss": 4.7292,
      "mean_token_accuracy": 0.24340257346630095,
      "num_tokens": 50229790.0,
      "step": 21900
    },
    {
      "entropy": 5.455516958236695,
      "epoch": 2.1042267050912584,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045586926642970113,
      "loss": 4.8984,
      "mean_token_accuracy": 0.239687243103981,
      "num_tokens": 50240547.0,
      "step": 21905
    },
    {
      "entropy": 5.368298053741455,
      "epoch": 2.1047070124879923,
      "grad_norm": 1.125,
      "learning_rate": 0.0004558488719934188,
      "loss": 4.7139,
      "mean_token_accuracy": 0.2503134712576866,
      "num_tokens": 50251889.0,
      "step": 21910
    },
    {
      "entropy": 5.308863687515259,
      "epoch": 2.105187319884726,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004558284733583627,
      "loss": 4.7842,
      "mean_token_accuracy": 0.2366962268948555,
      "num_tokens": 50263599.0,
      "step": 21915
    },
    {
      "entropy": 5.359787273406982,
      "epoch": 2.10566762728146,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045580807052500645,
      "loss": 4.8216,
      "mean_token_accuracy": 0.23999680429697037,
      "num_tokens": 50275645.0,
      "step": 21920
    },
    {
      "entropy": 5.331399631500244,
      "epoch": 2.106147934678194,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004557876634938236,
      "loss": 4.693,
      "mean_token_accuracy": 0.244863560795784,
      "num_tokens": 50287640.0,
      "step": 21925
    },
    {
      "entropy": 5.36338472366333,
      "epoch": 2.106628242074928,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004557672522652881,
      "loss": 4.8066,
      "mean_token_accuracy": 0.23576122373342515,
      "num_tokens": 50299119.0,
      "step": 21930
    },
    {
      "entropy": 5.3552783012390135,
      "epoch": 2.1071085494716617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004557468368398738,
      "loss": 4.7352,
      "mean_token_accuracy": 0.24158240407705306,
      "num_tokens": 50311377.0,
      "step": 21935
    },
    {
      "entropy": 5.33909502029419,
      "epoch": 2.1075888568683956,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004557264172180546,
      "loss": 4.7804,
      "mean_token_accuracy": 0.24107198864221574,
      "num_tokens": 50323185.0,
      "step": 21940
    },
    {
      "entropy": 5.347428798675537,
      "epoch": 2.1080691642651295,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004557059934003046,
      "loss": 4.8379,
      "mean_token_accuracy": 0.24203363060951233,
      "num_tokens": 50333992.0,
      "step": 21945
    },
    {
      "entropy": 5.416202163696289,
      "epoch": 2.108549471661864,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004556855653870981,
      "loss": 4.8013,
      "mean_token_accuracy": 0.23655421584844588,
      "num_tokens": 50346087.0,
      "step": 21950
    },
    {
      "entropy": 5.428376293182373,
      "epoch": 2.1090297790585977,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004556651331789092,
      "loss": 4.8182,
      "mean_token_accuracy": 0.23533014059066773,
      "num_tokens": 50358274.0,
      "step": 21955
    },
    {
      "entropy": 5.364601278305054,
      "epoch": 2.1095100864553316,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004556446967762125,
      "loss": 4.7454,
      "mean_token_accuracy": 0.24022976458072662,
      "num_tokens": 50367784.0,
      "step": 21960
    },
    {
      "entropy": 5.363576602935791,
      "epoch": 2.1099903938520654,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045562425617948226,
      "loss": 4.8259,
      "mean_token_accuracy": 0.23598181456327438,
      "num_tokens": 50377877.0,
      "step": 21965
    },
    {
      "entropy": 5.3720948696136475,
      "epoch": 2.1104707012487993,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045560381138919315,
      "loss": 4.8138,
      "mean_token_accuracy": 0.23854826688766478,
      "num_tokens": 50391023.0,
      "step": 21970
    },
    {
      "entropy": 5.384488725662232,
      "epoch": 2.110951008645533,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045558336240581984,
      "loss": 4.7658,
      "mean_token_accuracy": 0.2368793323636055,
      "num_tokens": 50404417.0,
      "step": 21975
    },
    {
      "entropy": 5.363023281097412,
      "epoch": 2.111431316042267,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045556290922983705,
      "loss": 4.7602,
      "mean_token_accuracy": 0.24107472002506256,
      "num_tokens": 50416153.0,
      "step": 21980
    },
    {
      "entropy": 5.329302072525024,
      "epoch": 2.111911623439001,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004555424518617197,
      "loss": 4.7322,
      "mean_token_accuracy": 0.24618351310491562,
      "num_tokens": 50426860.0,
      "step": 21985
    },
    {
      "entropy": 5.335509252548218,
      "epoch": 2.112391930835735,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045552199030194274,
      "loss": 4.6864,
      "mean_token_accuracy": 0.2497244119644165,
      "num_tokens": 50437262.0,
      "step": 21990
    },
    {
      "entropy": 5.300484132766724,
      "epoch": 2.1128722382324687,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045550152455098113,
      "loss": 4.6933,
      "mean_token_accuracy": 0.2464381605386734,
      "num_tokens": 50448564.0,
      "step": 21995
    },
    {
      "entropy": 5.307597398757935,
      "epoch": 2.1133525456292026,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004554810546093102,
      "loss": 4.6721,
      "mean_token_accuracy": 0.2573861375451088,
      "num_tokens": 50459416.0,
      "step": 22000
    },
    {
      "entropy": 5.376848030090332,
      "epoch": 2.1138328530259365,
      "grad_norm": 1.125,
      "learning_rate": 0.000455460580477405,
      "loss": 4.758,
      "mean_token_accuracy": 0.2479401260614395,
      "num_tokens": 50470934.0,
      "step": 22005
    },
    {
      "entropy": 5.428783273696899,
      "epoch": 2.1143131604226704,
      "grad_norm": 1.03125,
      "learning_rate": 0.000455440102155741,
      "loss": 4.7882,
      "mean_token_accuracy": 0.23781522661447524,
      "num_tokens": 50483050.0,
      "step": 22010
    },
    {
      "entropy": 5.364634466171265,
      "epoch": 2.1147934678194042,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004554196196447937,
      "loss": 4.7872,
      "mean_token_accuracy": 0.23326063752174378,
      "num_tokens": 50494655.0,
      "step": 22015
    },
    {
      "entropy": 5.391149616241455,
      "epoch": 2.115273775216138,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004553991329450385,
      "loss": 4.7225,
      "mean_token_accuracy": 0.24041213989257812,
      "num_tokens": 50506302.0,
      "step": 22020
    },
    {
      "entropy": 5.408211660385132,
      "epoch": 2.115754082612872,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045537864205695116,
      "loss": 4.7656,
      "mean_token_accuracy": 0.2441441982984543,
      "num_tokens": 50517979.0,
      "step": 22025
    },
    {
      "entropy": 5.383596181869507,
      "epoch": 2.1162343900096063,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004553581469810073,
      "loss": 4.8394,
      "mean_token_accuracy": 0.23017778545618056,
      "num_tokens": 50530054.0,
      "step": 22030
    },
    {
      "entropy": 5.348245143890381,
      "epoch": 2.11671469740634,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045533764771768287,
      "loss": 4.741,
      "mean_token_accuracy": 0.23665117025375365,
      "num_tokens": 50540664.0,
      "step": 22035
    },
    {
      "entropy": 5.4022626876831055,
      "epoch": 2.117195004803074,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045531714426745373,
      "loss": 4.8047,
      "mean_token_accuracy": 0.2346808448433876,
      "num_tokens": 50551444.0,
      "step": 22040
    },
    {
      "entropy": 5.312077522277832,
      "epoch": 2.117675312199808,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004552966366307959,
      "loss": 4.7352,
      "mean_token_accuracy": 0.2416164830327034,
      "num_tokens": 50562371.0,
      "step": 22045
    },
    {
      "entropy": 5.296939277648926,
      "epoch": 2.118155619596542,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004552761248081856,
      "loss": 4.7083,
      "mean_token_accuracy": 0.24459124803543092,
      "num_tokens": 50573545.0,
      "step": 22050
    },
    {
      "entropy": 5.413045215606689,
      "epoch": 2.1186359269932757,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000455255608800099,
      "loss": 4.8404,
      "mean_token_accuracy": 0.2340160295367241,
      "num_tokens": 50584799.0,
      "step": 22055
    },
    {
      "entropy": 5.393403244018555,
      "epoch": 2.1191162343900096,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045523508860701237,
      "loss": 4.7729,
      "mean_token_accuracy": 0.2372686058282852,
      "num_tokens": 50596435.0,
      "step": 22060
    },
    {
      "entropy": 5.359182643890381,
      "epoch": 2.1195965417867435,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004552145642294021,
      "loss": 4.6845,
      "mean_token_accuracy": 0.25223897248506544,
      "num_tokens": 50606921.0,
      "step": 22065
    },
    {
      "entropy": 5.365457820892334,
      "epoch": 2.1200768491834774,
      "grad_norm": 1.125,
      "learning_rate": 0.00045519403566774493,
      "loss": 4.8775,
      "mean_token_accuracy": 0.22896389216184615,
      "num_tokens": 50618107.0,
      "step": 22070
    },
    {
      "entropy": 5.367590141296387,
      "epoch": 2.1205571565802113,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004551735029225172,
      "loss": 4.7765,
      "mean_token_accuracy": 0.23838126212358474,
      "num_tokens": 50628958.0,
      "step": 22075
    },
    {
      "entropy": 5.385892581939697,
      "epoch": 2.121037463976945,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045515296599419583,
      "loss": 4.7408,
      "mean_token_accuracy": 0.24018959254026412,
      "num_tokens": 50640557.0,
      "step": 22080
    },
    {
      "entropy": 5.386507415771485,
      "epoch": 2.121517771373679,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004551324248832574,
      "loss": 4.773,
      "mean_token_accuracy": 0.2409815952181816,
      "num_tokens": 50652699.0,
      "step": 22085
    },
    {
      "entropy": 5.318156957626343,
      "epoch": 2.121998078770413,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004551118795901791,
      "loss": 4.6826,
      "mean_token_accuracy": 0.23484003245830537,
      "num_tokens": 50663735.0,
      "step": 22090
    },
    {
      "entropy": 5.317165374755859,
      "epoch": 2.122478386167147,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004550913301154376,
      "loss": 4.7448,
      "mean_token_accuracy": 0.24152814596891403,
      "num_tokens": 50676384.0,
      "step": 22095
    },
    {
      "entropy": 5.3907591819763185,
      "epoch": 2.1229586935638807,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004550707764595103,
      "loss": 4.7784,
      "mean_token_accuracy": 0.23375765085220337,
      "num_tokens": 50687775.0,
      "step": 22100
    },
    {
      "entropy": 5.408683729171753,
      "epoch": 2.123439000960615,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045505021862287434,
      "loss": 4.8303,
      "mean_token_accuracy": 0.23292600065469743,
      "num_tokens": 50698522.0,
      "step": 22105
    },
    {
      "entropy": 5.298542118072509,
      "epoch": 2.123919308357349,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045502965660600684,
      "loss": 4.7285,
      "mean_token_accuracy": 0.2545346483588219,
      "num_tokens": 50709994.0,
      "step": 22110
    },
    {
      "entropy": 5.3288689136505125,
      "epoch": 2.1243996157540828,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004550090904093853,
      "loss": 4.7132,
      "mean_token_accuracy": 0.24237690269947051,
      "num_tokens": 50720777.0,
      "step": 22115
    },
    {
      "entropy": 5.336025238037109,
      "epoch": 2.1248799231508166,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004549885200334872,
      "loss": 4.704,
      "mean_token_accuracy": 0.24558757096529008,
      "num_tokens": 50732836.0,
      "step": 22120
    },
    {
      "entropy": 5.358902168273926,
      "epoch": 2.1253602305475505,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004549679454787901,
      "loss": 4.7495,
      "mean_token_accuracy": 0.24522778242826462,
      "num_tokens": 50743854.0,
      "step": 22125
    },
    {
      "entropy": 5.272014570236206,
      "epoch": 2.1258405379442844,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045494736674577175,
      "loss": 4.6743,
      "mean_token_accuracy": 0.24982055872678757,
      "num_tokens": 50755524.0,
      "step": 22130
    },
    {
      "entropy": 5.341372871398926,
      "epoch": 2.1263208453410183,
      "grad_norm": 1.125,
      "learning_rate": 0.0004549267838349099,
      "loss": 4.7039,
      "mean_token_accuracy": 0.24524527937173843,
      "num_tokens": 50766909.0,
      "step": 22135
    },
    {
      "entropy": 5.2916099548339846,
      "epoch": 2.126801152737752,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004549061967466823,
      "loss": 4.6848,
      "mean_token_accuracy": 0.24639111310243605,
      "num_tokens": 50778065.0,
      "step": 22140
    },
    {
      "entropy": 5.359102058410644,
      "epoch": 2.127281460134486,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004548856054815671,
      "loss": 4.7762,
      "mean_token_accuracy": 0.24102241098880767,
      "num_tokens": 50790122.0,
      "step": 22145
    },
    {
      "entropy": 5.34037504196167,
      "epoch": 2.12776176753122,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045486501004004225,
      "loss": 4.7787,
      "mean_token_accuracy": 0.2322711393237114,
      "num_tokens": 50802661.0,
      "step": 22150
    },
    {
      "entropy": 5.373083257675171,
      "epoch": 2.128242074927954,
      "grad_norm": 1.03125,
      "learning_rate": 0.000454844410422586,
      "loss": 4.8027,
      "mean_token_accuracy": 0.24615223556756974,
      "num_tokens": 50814279.0,
      "step": 22155
    },
    {
      "entropy": 5.288534164428711,
      "epoch": 2.1287223823246877,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045482380662967655,
      "loss": 4.6464,
      "mean_token_accuracy": 0.2508205622434616,
      "num_tokens": 50826186.0,
      "step": 22160
    },
    {
      "entropy": 5.386482667922974,
      "epoch": 2.1292026897214216,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004548031986617923,
      "loss": 4.7829,
      "mean_token_accuracy": 0.23078881949186325,
      "num_tokens": 50838185.0,
      "step": 22165
    },
    {
      "entropy": 5.327685260772705,
      "epoch": 2.1296829971181555,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004547825865194117,
      "loss": 4.731,
      "mean_token_accuracy": 0.24447422623634338,
      "num_tokens": 50850461.0,
      "step": 22170
    },
    {
      "entropy": 5.383919906616211,
      "epoch": 2.1301633045148893,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045476197020301323,
      "loss": 4.8074,
      "mean_token_accuracy": 0.24052779078483583,
      "num_tokens": 50861859.0,
      "step": 22175
    },
    {
      "entropy": 5.361906909942627,
      "epoch": 2.1306436119116237,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045474134971307554,
      "loss": 4.7123,
      "mean_token_accuracy": 0.24450083523988725,
      "num_tokens": 50872776.0,
      "step": 22180
    },
    {
      "entropy": 5.361825466156006,
      "epoch": 2.1311239193083575,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004547207250500775,
      "loss": 4.7036,
      "mean_token_accuracy": 0.2426493212580681,
      "num_tokens": 50885333.0,
      "step": 22185
    },
    {
      "entropy": 5.35405387878418,
      "epoch": 2.1316042267050914,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000454700096214498,
      "loss": 4.7262,
      "mean_token_accuracy": 0.23289585262537002,
      "num_tokens": 50897256.0,
      "step": 22190
    },
    {
      "entropy": 5.289123678207398,
      "epoch": 2.1320845341018253,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045467946320681567,
      "loss": 4.715,
      "mean_token_accuracy": 0.2408551275730133,
      "num_tokens": 50909558.0,
      "step": 22195
    },
    {
      "entropy": 5.340747404098511,
      "epoch": 2.132564841498559,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004546588260275098,
      "loss": 4.7615,
      "mean_token_accuracy": 0.2445739194750786,
      "num_tokens": 50920617.0,
      "step": 22200
    },
    {
      "entropy": 5.327092599868775,
      "epoch": 2.133045148895293,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045463818467705955,
      "loss": 4.7244,
      "mean_token_accuracy": 0.24114726781845092,
      "num_tokens": 50931323.0,
      "step": 22205
    },
    {
      "entropy": 5.299500751495361,
      "epoch": 2.133525456292027,
      "grad_norm": 1.078125,
      "learning_rate": 0.000454617539155944,
      "loss": 4.6355,
      "mean_token_accuracy": 0.2533891752362251,
      "num_tokens": 50942798.0,
      "step": 22210
    },
    {
      "entropy": 5.278481817245483,
      "epoch": 2.134005763688761,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045459688946464255,
      "loss": 4.6257,
      "mean_token_accuracy": 0.24638168066740035,
      "num_tokens": 50954877.0,
      "step": 22215
    },
    {
      "entropy": 5.3988118171691895,
      "epoch": 2.1344860710854947,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004545762356036346,
      "loss": 4.6996,
      "mean_token_accuracy": 0.24510339200496672,
      "num_tokens": 50966133.0,
      "step": 22220
    },
    {
      "entropy": 5.349482297897339,
      "epoch": 2.1349663784822286,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004545555775733998,
      "loss": 4.7022,
      "mean_token_accuracy": 0.25301606953144073,
      "num_tokens": 50977334.0,
      "step": 22225
    },
    {
      "entropy": 5.35176272392273,
      "epoch": 2.1354466858789625,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045453491537441747,
      "loss": 4.7805,
      "mean_token_accuracy": 0.23289942741394043,
      "num_tokens": 50988743.0,
      "step": 22230
    },
    {
      "entropy": 5.367421293258667,
      "epoch": 2.1359269932756964,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045451424900716763,
      "loss": 4.7165,
      "mean_token_accuracy": 0.2434825822710991,
      "num_tokens": 51000591.0,
      "step": 22235
    },
    {
      "entropy": 5.460317945480346,
      "epoch": 2.1364073006724302,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045449357847212994,
      "loss": 4.9218,
      "mean_token_accuracy": 0.23702692985534668,
      "num_tokens": 51011565.0,
      "step": 22240
    },
    {
      "entropy": 5.391738653182983,
      "epoch": 2.136887608069164,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004544729037697844,
      "loss": 4.7866,
      "mean_token_accuracy": 0.23577540069818498,
      "num_tokens": 51022755.0,
      "step": 22245
    },
    {
      "entropy": 5.392456674575806,
      "epoch": 2.137367915465898,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045445222490061093,
      "loss": 4.8225,
      "mean_token_accuracy": 0.236035318672657,
      "num_tokens": 51032839.0,
      "step": 22250
    },
    {
      "entropy": 5.3465581893920895,
      "epoch": 2.1378482228626323,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004544315418650897,
      "loss": 4.7659,
      "mean_token_accuracy": 0.24447442591190338,
      "num_tokens": 51045020.0,
      "step": 22255
    },
    {
      "entropy": 5.334083223342896,
      "epoch": 2.138328530259366,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004544108546637008,
      "loss": 4.6759,
      "mean_token_accuracy": 0.24963700473308564,
      "num_tokens": 51056765.0,
      "step": 22260
    },
    {
      "entropy": 5.309499788284302,
      "epoch": 2.1388088376561,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004543901632969247,
      "loss": 4.7039,
      "mean_token_accuracy": 0.24235988557338714,
      "num_tokens": 51069559.0,
      "step": 22265
    },
    {
      "entropy": 5.33808217048645,
      "epoch": 2.139289145052834,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045436946776524157,
      "loss": 4.7331,
      "mean_token_accuracy": 0.24002631902694702,
      "num_tokens": 51080972.0,
      "step": 22270
    },
    {
      "entropy": 5.294958448410034,
      "epoch": 2.139769452449568,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045434876806913204,
      "loss": 4.681,
      "mean_token_accuracy": 0.24800195395946503,
      "num_tokens": 51092101.0,
      "step": 22275
    },
    {
      "entropy": 5.417356157302857,
      "epoch": 2.1402497598463017,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004543280642090767,
      "loss": 4.886,
      "mean_token_accuracy": 0.23672794550657272,
      "num_tokens": 51103372.0,
      "step": 22280
    },
    {
      "entropy": 5.394207763671875,
      "epoch": 2.1407300672430356,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004543073561855562,
      "loss": 4.7817,
      "mean_token_accuracy": 0.23206587731838227,
      "num_tokens": 51114856.0,
      "step": 22285
    },
    {
      "entropy": 5.446911716461182,
      "epoch": 2.1412103746397695,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004542866439990513,
      "loss": 4.833,
      "mean_token_accuracy": 0.2395749494433403,
      "num_tokens": 51126354.0,
      "step": 22290
    },
    {
      "entropy": 5.425191354751587,
      "epoch": 2.1416906820365034,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004542659276500429,
      "loss": 4.8085,
      "mean_token_accuracy": 0.23967182487249375,
      "num_tokens": 51138341.0,
      "step": 22295
    },
    {
      "entropy": 5.304317283630371,
      "epoch": 2.1421709894332372,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045424520713901204,
      "loss": 4.6726,
      "mean_token_accuracy": 0.2510075345635414,
      "num_tokens": 51150969.0,
      "step": 22300
    },
    {
      "entropy": 5.2972650051116945,
      "epoch": 2.142651296829971,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004542244824664396,
      "loss": 4.7157,
      "mean_token_accuracy": 0.24297581911087035,
      "num_tokens": 51162199.0,
      "step": 22305
    },
    {
      "entropy": 5.353101396560669,
      "epoch": 2.143131604226705,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045420375363280696,
      "loss": 4.7952,
      "mean_token_accuracy": 0.23679332435131073,
      "num_tokens": 51173129.0,
      "step": 22310
    },
    {
      "entropy": 5.302580833435059,
      "epoch": 2.143611911623439,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045418302063859526,
      "loss": 4.6963,
      "mean_token_accuracy": 0.24769868403673173,
      "num_tokens": 51184593.0,
      "step": 22315
    },
    {
      "entropy": 5.346362161636352,
      "epoch": 2.1440922190201728,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045416228348428583,
      "loss": 4.7373,
      "mean_token_accuracy": 0.24488938152790068,
      "num_tokens": 51196524.0,
      "step": 22320
    },
    {
      "entropy": 5.3225513935089115,
      "epoch": 2.1445725264169067,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045414154217036023,
      "loss": 4.8024,
      "mean_token_accuracy": 0.2400071695446968,
      "num_tokens": 51209307.0,
      "step": 22325
    },
    {
      "entropy": 5.288712406158448,
      "epoch": 2.1450528338136405,
      "grad_norm": 1.125,
      "learning_rate": 0.00045412079669730006,
      "loss": 4.7131,
      "mean_token_accuracy": 0.25196952521800997,
      "num_tokens": 51220919.0,
      "step": 22330
    },
    {
      "entropy": 5.340694284439087,
      "epoch": 2.1455331412103744,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004541000470655867,
      "loss": 4.6639,
      "mean_token_accuracy": 0.24791524857282637,
      "num_tokens": 51232930.0,
      "step": 22335
    },
    {
      "entropy": 5.318307733535766,
      "epoch": 2.1460134486071087,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045407929327570215,
      "loss": 4.7681,
      "mean_token_accuracy": 0.24373000413179396,
      "num_tokens": 51244697.0,
      "step": 22340
    },
    {
      "entropy": 5.336845731735229,
      "epoch": 2.1464937560038426,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004540585353281282,
      "loss": 4.7699,
      "mean_token_accuracy": 0.24696495532989501,
      "num_tokens": 51255333.0,
      "step": 22345
    },
    {
      "entropy": 5.402605772018433,
      "epoch": 2.1469740634005765,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004540377732233467,
      "loss": 4.7625,
      "mean_token_accuracy": 0.2408861204981804,
      "num_tokens": 51267649.0,
      "step": 22350
    },
    {
      "entropy": 5.360998296737671,
      "epoch": 2.1474543707973104,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004540170069618397,
      "loss": 4.6852,
      "mean_token_accuracy": 0.24558133035898208,
      "num_tokens": 51279576.0,
      "step": 22355
    },
    {
      "entropy": 5.378098011016846,
      "epoch": 2.1479346781940443,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045399623654408946,
      "loss": 4.7526,
      "mean_token_accuracy": 0.23477070778608322,
      "num_tokens": 51290974.0,
      "step": 22360
    },
    {
      "entropy": 5.47357177734375,
      "epoch": 2.148414985590778,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000453975461970578,
      "loss": 4.8129,
      "mean_token_accuracy": 0.2378174766898155,
      "num_tokens": 51301405.0,
      "step": 22365
    },
    {
      "entropy": 5.454221773147583,
      "epoch": 2.148895292987512,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004539546832417879,
      "loss": 4.8603,
      "mean_token_accuracy": 0.23202855437994002,
      "num_tokens": 51313158.0,
      "step": 22370
    },
    {
      "entropy": 5.367579412460327,
      "epoch": 2.149375600384246,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045393390035820136,
      "loss": 4.7485,
      "mean_token_accuracy": 0.2508128896355629,
      "num_tokens": 51323789.0,
      "step": 22375
    },
    {
      "entropy": 5.293633270263672,
      "epoch": 2.14985590778098,
      "grad_norm": 1.046875,
      "learning_rate": 0.000453913113320301,
      "loss": 4.7726,
      "mean_token_accuracy": 0.2400583416223526,
      "num_tokens": 51334864.0,
      "step": 22380
    },
    {
      "entropy": 5.41011381149292,
      "epoch": 2.1503362151777137,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004538923221285694,
      "loss": 4.6813,
      "mean_token_accuracy": 0.23894115090370177,
      "num_tokens": 51345473.0,
      "step": 22385
    },
    {
      "entropy": 5.467349720001221,
      "epoch": 2.1508165225744476,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004538715267834893,
      "loss": 4.8302,
      "mean_token_accuracy": 0.23579642474651336,
      "num_tokens": 51356755.0,
      "step": 22390
    },
    {
      "entropy": 5.233108377456665,
      "epoch": 2.1512968299711814,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004538507272855434,
      "loss": 4.6365,
      "mean_token_accuracy": 0.2477122128009796,
      "num_tokens": 51368351.0,
      "step": 22395
    },
    {
      "entropy": 5.416212892532348,
      "epoch": 2.1517771373679153,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045382992363521486,
      "loss": 4.891,
      "mean_token_accuracy": 0.23613059818744658,
      "num_tokens": 51379144.0,
      "step": 22400
    },
    {
      "entropy": 5.337558174133301,
      "epoch": 2.152257444764649,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045380911583298633,
      "loss": 4.6924,
      "mean_token_accuracy": 0.24536923915147782,
      "num_tokens": 51390732.0,
      "step": 22405
    },
    {
      "entropy": 5.330533838272094,
      "epoch": 2.152737752161383,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045378830387934123,
      "loss": 4.6735,
      "mean_token_accuracy": 0.2479643329977989,
      "num_tokens": 51401185.0,
      "step": 22410
    },
    {
      "entropy": 5.396887922286988,
      "epoch": 2.1532180595581174,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004537674877747626,
      "loss": 4.7743,
      "mean_token_accuracy": 0.2382538616657257,
      "num_tokens": 51412598.0,
      "step": 22415
    },
    {
      "entropy": 5.343439817428589,
      "epoch": 2.1536983669548513,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045374666751973365,
      "loss": 4.6747,
      "mean_token_accuracy": 0.24617379158735275,
      "num_tokens": 51423248.0,
      "step": 22420
    },
    {
      "entropy": 5.264202070236206,
      "epoch": 2.154178674351585,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045372584311473784,
      "loss": 4.6433,
      "mean_token_accuracy": 0.25315298438072203,
      "num_tokens": 51433862.0,
      "step": 22425
    },
    {
      "entropy": 5.282385015487671,
      "epoch": 2.154658981748319,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004537050145602587,
      "loss": 4.7371,
      "mean_token_accuracy": 0.2444470703601837,
      "num_tokens": 51445311.0,
      "step": 22430
    },
    {
      "entropy": 5.269491815567017,
      "epoch": 2.155139289145053,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004536841818567798,
      "loss": 4.6544,
      "mean_token_accuracy": 0.2510728150606155,
      "num_tokens": 51456550.0,
      "step": 22435
    },
    {
      "entropy": 5.345057678222656,
      "epoch": 2.155619596541787,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004536633450047847,
      "loss": 4.7355,
      "mean_token_accuracy": 0.24746509790420532,
      "num_tokens": 51467645.0,
      "step": 22440
    },
    {
      "entropy": 5.335013151168823,
      "epoch": 2.1560999039385207,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045364250400475734,
      "loss": 4.7322,
      "mean_token_accuracy": 0.24695896208286286,
      "num_tokens": 51478275.0,
      "step": 22445
    },
    {
      "entropy": 5.308736896514892,
      "epoch": 2.1565802113352546,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004536216588571814,
      "loss": 4.7051,
      "mean_token_accuracy": 0.2487216606736183,
      "num_tokens": 51489634.0,
      "step": 22450
    },
    {
      "entropy": 5.30159854888916,
      "epoch": 2.1570605187319885,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000453600809562541,
      "loss": 4.7264,
      "mean_token_accuracy": 0.24681274741888046,
      "num_tokens": 51500245.0,
      "step": 22455
    },
    {
      "entropy": 5.310759353637695,
      "epoch": 2.1575408261287223,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004535799561213202,
      "loss": 4.7238,
      "mean_token_accuracy": 0.2447451874613762,
      "num_tokens": 51511996.0,
      "step": 22460
    },
    {
      "entropy": 5.327337741851807,
      "epoch": 2.158021133525456,
      "grad_norm": 0.9375,
      "learning_rate": 0.000453559098534003,
      "loss": 4.6977,
      "mean_token_accuracy": 0.24798343926668168,
      "num_tokens": 51523522.0,
      "step": 22465
    },
    {
      "entropy": 5.366113615036011,
      "epoch": 2.15850144092219,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004535382368010738,
      "loss": 4.7233,
      "mean_token_accuracy": 0.24666790366172792,
      "num_tokens": 51535261.0,
      "step": 22470
    },
    {
      "entropy": 5.318097877502441,
      "epoch": 2.158981748318924,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045351737092301676,
      "loss": 4.6656,
      "mean_token_accuracy": 0.2442825347185135,
      "num_tokens": 51546965.0,
      "step": 22475
    },
    {
      "entropy": 5.335655403137207,
      "epoch": 2.159462055715658,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004534965009003165,
      "loss": 4.701,
      "mean_token_accuracy": 0.2482134759426117,
      "num_tokens": 51558976.0,
      "step": 22480
    },
    {
      "entropy": 5.435312938690186,
      "epoch": 2.1599423631123917,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004534756267334576,
      "loss": 4.8211,
      "mean_token_accuracy": 0.23551032990217208,
      "num_tokens": 51571625.0,
      "step": 22485
    },
    {
      "entropy": 5.443209171295166,
      "epoch": 2.160422670509126,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045345474842292455,
      "loss": 4.8239,
      "mean_token_accuracy": 0.23554776161909102,
      "num_tokens": 51582732.0,
      "step": 22490
    },
    {
      "entropy": 5.334451150894165,
      "epoch": 2.16090297790586,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004534338659692022,
      "loss": 4.743,
      "mean_token_accuracy": 0.24029989093542098,
      "num_tokens": 51594346.0,
      "step": 22495
    },
    {
      "entropy": 5.349223947525024,
      "epoch": 2.161383285302594,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004534129793727753,
      "loss": 4.7394,
      "mean_token_accuracy": 0.2408968210220337,
      "num_tokens": 51606909.0,
      "step": 22500
    },
    {
      "entropy": 5.335518836975098,
      "epoch": 2.1618635926993277,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004533920886341288,
      "loss": 4.7265,
      "mean_token_accuracy": 0.24453021585941315,
      "num_tokens": 51617596.0,
      "step": 22505
    },
    {
      "entropy": 5.30107569694519,
      "epoch": 2.1623439000960616,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004533711937537477,
      "loss": 4.6231,
      "mean_token_accuracy": 0.2519013717770576,
      "num_tokens": 51627573.0,
      "step": 22510
    },
    {
      "entropy": 5.2957581043243405,
      "epoch": 2.1628242074927955,
      "grad_norm": 1.125,
      "learning_rate": 0.0004533502947321171,
      "loss": 4.6941,
      "mean_token_accuracy": 0.2363706722855568,
      "num_tokens": 51640070.0,
      "step": 22515
    },
    {
      "entropy": 5.351886892318726,
      "epoch": 2.1633045148895294,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004533293915697223,
      "loss": 4.6532,
      "mean_token_accuracy": 0.24689853489398955,
      "num_tokens": 51651139.0,
      "step": 22520
    },
    {
      "entropy": 5.424572372436524,
      "epoch": 2.1637848222862632,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045330848426704853,
      "loss": 4.808,
      "mean_token_accuracy": 0.24071638882160187,
      "num_tokens": 51661893.0,
      "step": 22525
    },
    {
      "entropy": 5.335722970962524,
      "epoch": 2.164265129682997,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004532875728245813,
      "loss": 4.7174,
      "mean_token_accuracy": 0.24333761483430863,
      "num_tokens": 51674266.0,
      "step": 22530
    },
    {
      "entropy": 5.348578453063965,
      "epoch": 2.164745437079731,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045326665724280594,
      "loss": 4.7523,
      "mean_token_accuracy": 0.2464014247059822,
      "num_tokens": 51685450.0,
      "step": 22535
    },
    {
      "entropy": 5.400900793075562,
      "epoch": 2.165225744476465,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045324573752220814,
      "loss": 4.7557,
      "mean_token_accuracy": 0.24519071131944656,
      "num_tokens": 51696779.0,
      "step": 22540
    },
    {
      "entropy": 5.402159023284912,
      "epoch": 2.1657060518731988,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045322481366327365,
      "loss": 4.7695,
      "mean_token_accuracy": 0.2406044125556946,
      "num_tokens": 51709387.0,
      "step": 22545
    },
    {
      "entropy": 5.332025575637817,
      "epoch": 2.1661863592699326,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004532038856664882,
      "loss": 4.7451,
      "mean_token_accuracy": 0.244800141453743,
      "num_tokens": 51720978.0,
      "step": 22550
    },
    {
      "entropy": 5.463725280761719,
      "epoch": 2.1666666666666665,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004531829535323376,
      "loss": 4.8551,
      "mean_token_accuracy": 0.23565064668655394,
      "num_tokens": 51733510.0,
      "step": 22555
    },
    {
      "entropy": 5.289333868026733,
      "epoch": 2.1671469740634004,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000453162017261308,
      "loss": 4.6595,
      "mean_token_accuracy": 0.25486769527196884,
      "num_tokens": 51743660.0,
      "step": 22560
    },
    {
      "entropy": 5.3499373435974125,
      "epoch": 2.1676272814601343,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004531410768538854,
      "loss": 4.6826,
      "mean_token_accuracy": 0.2468963548541069,
      "num_tokens": 51754592.0,
      "step": 22565
    },
    {
      "entropy": 5.313680648803711,
      "epoch": 2.168107588856868,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045312013231055596,
      "loss": 4.6872,
      "mean_token_accuracy": 0.25352584272623063,
      "num_tokens": 51766241.0,
      "step": 22570
    },
    {
      "entropy": 5.400108718872071,
      "epoch": 2.1685878962536025,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045309918363180593,
      "loss": 4.721,
      "mean_token_accuracy": 0.240858493745327,
      "num_tokens": 51777718.0,
      "step": 22575
    },
    {
      "entropy": 5.248170185089111,
      "epoch": 2.1690682036503364,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045307823081812166,
      "loss": 4.6506,
      "mean_token_accuracy": 0.25142836272716523,
      "num_tokens": 51788290.0,
      "step": 22580
    },
    {
      "entropy": 5.310922479629516,
      "epoch": 2.1695485110470702,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045305727386998977,
      "loss": 4.6795,
      "mean_token_accuracy": 0.24602738320827483,
      "num_tokens": 51800080.0,
      "step": 22585
    },
    {
      "entropy": 5.394365835189819,
      "epoch": 2.170028818443804,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004530363127878966,
      "loss": 4.7463,
      "mean_token_accuracy": 0.24155033826828004,
      "num_tokens": 51810155.0,
      "step": 22590
    },
    {
      "entropy": 5.356741046905517,
      "epoch": 2.170509125840538,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045301534757232885,
      "loss": 4.7651,
      "mean_token_accuracy": 0.2461069941520691,
      "num_tokens": 51823082.0,
      "step": 22595
    },
    {
      "entropy": 5.371994400024414,
      "epoch": 2.170989433237272,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004529943782237735,
      "loss": 4.8177,
      "mean_token_accuracy": 0.23945444226264953,
      "num_tokens": 51833280.0,
      "step": 22600
    },
    {
      "entropy": 5.347340488433838,
      "epoch": 2.1714697406340058,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045297340474271717,
      "loss": 4.7289,
      "mean_token_accuracy": 0.24307978600263597,
      "num_tokens": 51845105.0,
      "step": 22605
    },
    {
      "entropy": 5.419521045684815,
      "epoch": 2.1719500480307397,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004529524271296468,
      "loss": 4.7222,
      "mean_token_accuracy": 0.24897104054689406,
      "num_tokens": 51856232.0,
      "step": 22610
    },
    {
      "entropy": 5.365228319168091,
      "epoch": 2.1724303554274735,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045293144538504943,
      "loss": 4.7367,
      "mean_token_accuracy": 0.24987443387508393,
      "num_tokens": 51866218.0,
      "step": 22615
    },
    {
      "entropy": 5.319495391845703,
      "epoch": 2.1729106628242074,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004529104595094124,
      "loss": 4.7518,
      "mean_token_accuracy": 0.24271588176488876,
      "num_tokens": 51877425.0,
      "step": 22620
    },
    {
      "entropy": 5.368311786651612,
      "epoch": 2.1733909702209413,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045288946950322264,
      "loss": 4.7592,
      "mean_token_accuracy": 0.2420559510588646,
      "num_tokens": 51888818.0,
      "step": 22625
    },
    {
      "entropy": 5.359784746170044,
      "epoch": 2.173871277617675,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004528684753669677,
      "loss": 4.7118,
      "mean_token_accuracy": 0.24027734994888306,
      "num_tokens": 51899875.0,
      "step": 22630
    },
    {
      "entropy": 5.372973775863647,
      "epoch": 2.174351585014409,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004528474771011349,
      "loss": 4.7266,
      "mean_token_accuracy": 0.24124199151992798,
      "num_tokens": 51911084.0,
      "step": 22635
    },
    {
      "entropy": 5.310886812210083,
      "epoch": 2.174831892411143,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045282647470621176,
      "loss": 4.7417,
      "mean_token_accuracy": 0.24308115094900132,
      "num_tokens": 51922165.0,
      "step": 22640
    },
    {
      "entropy": 5.352889633178711,
      "epoch": 2.175312199807877,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045280546818268595,
      "loss": 4.7544,
      "mean_token_accuracy": 0.24395983219146727,
      "num_tokens": 51934827.0,
      "step": 22645
    },
    {
      "entropy": 5.3935784816741945,
      "epoch": 2.175792507204611,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004527844575310452,
      "loss": 4.7774,
      "mean_token_accuracy": 0.23333912938833237,
      "num_tokens": 51947997.0,
      "step": 22650
    },
    {
      "entropy": 5.389936876296997,
      "epoch": 2.176272814601345,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045276344275177715,
      "loss": 4.7881,
      "mean_token_accuracy": 0.23995428532361984,
      "num_tokens": 51959639.0,
      "step": 22655
    },
    {
      "entropy": 5.403819227218628,
      "epoch": 2.176753121998079,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045274242384536984,
      "loss": 4.8237,
      "mean_token_accuracy": 0.23671771734952926,
      "num_tokens": 51970716.0,
      "step": 22660
    },
    {
      "entropy": 5.35613055229187,
      "epoch": 2.177233429394813,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004527214008123113,
      "loss": 4.7349,
      "mean_token_accuracy": 0.246605084836483,
      "num_tokens": 51982556.0,
      "step": 22665
    },
    {
      "entropy": 5.3798277378082275,
      "epoch": 2.1777137367915467,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004527003736530895,
      "loss": 4.7186,
      "mean_token_accuracy": 0.24709362536668777,
      "num_tokens": 51994894.0,
      "step": 22670
    },
    {
      "entropy": 5.358087778091431,
      "epoch": 2.1781940441882806,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045267934236819265,
      "loss": 4.7929,
      "mean_token_accuracy": 0.23217078894376755,
      "num_tokens": 52007128.0,
      "step": 22675
    },
    {
      "entropy": 5.365379905700683,
      "epoch": 2.1786743515850144,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004526583069581091,
      "loss": 4.8856,
      "mean_token_accuracy": 0.23357589542865753,
      "num_tokens": 52019598.0,
      "step": 22680
    },
    {
      "entropy": 5.415447330474853,
      "epoch": 2.1791546589817483,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004526372674233272,
      "loss": 4.7576,
      "mean_token_accuracy": 0.2449882820248604,
      "num_tokens": 52031445.0,
      "step": 22685
    },
    {
      "entropy": 5.444033241271972,
      "epoch": 2.179634966378482,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045261622376433543,
      "loss": 4.7946,
      "mean_token_accuracy": 0.24207804799079896,
      "num_tokens": 52042548.0,
      "step": 22690
    },
    {
      "entropy": 5.317544794082641,
      "epoch": 2.180115273775216,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045259517598162237,
      "loss": 4.7158,
      "mean_token_accuracy": 0.24300388544797896,
      "num_tokens": 52054147.0,
      "step": 22695
    },
    {
      "entropy": 5.320989418029785,
      "epoch": 2.18059558117195,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004525741240756766,
      "loss": 4.7087,
      "mean_token_accuracy": 0.2524859309196472,
      "num_tokens": 52065556.0,
      "step": 22700
    },
    {
      "entropy": 5.41361141204834,
      "epoch": 2.181075888568684,
      "grad_norm": 1.125,
      "learning_rate": 0.0004525530680469871,
      "loss": 4.7807,
      "mean_token_accuracy": 0.24154318124055862,
      "num_tokens": 52077543.0,
      "step": 22705
    },
    {
      "entropy": 5.362244129180908,
      "epoch": 2.1815561959654177,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045253200789604245,
      "loss": 4.7368,
      "mean_token_accuracy": 0.24387475997209548,
      "num_tokens": 52088376.0,
      "step": 22710
    },
    {
      "entropy": 5.344343137741089,
      "epoch": 2.1820365033621516,
      "grad_norm": 1.0,
      "learning_rate": 0.00045251094362333186,
      "loss": 4.7554,
      "mean_token_accuracy": 0.2392376258969307,
      "num_tokens": 52100019.0,
      "step": 22715
    },
    {
      "entropy": 5.317023849487304,
      "epoch": 2.1825168107588855,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004524898752293441,
      "loss": 4.7296,
      "mean_token_accuracy": 0.2423124611377716,
      "num_tokens": 52110643.0,
      "step": 22720
    },
    {
      "entropy": 5.280483865737915,
      "epoch": 2.18299711815562,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045246880271456857,
      "loss": 4.6498,
      "mean_token_accuracy": 0.24819518774747848,
      "num_tokens": 52123092.0,
      "step": 22725
    },
    {
      "entropy": 5.449012613296508,
      "epoch": 2.1834774255523537,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004524477260794944,
      "loss": 4.8297,
      "mean_token_accuracy": 0.23696319311857222,
      "num_tokens": 52135371.0,
      "step": 22730
    },
    {
      "entropy": 5.371112632751465,
      "epoch": 2.1839577329490876,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045242664532461094,
      "loss": 4.7006,
      "mean_token_accuracy": 0.251247750222683,
      "num_tokens": 52145877.0,
      "step": 22735
    },
    {
      "entropy": 5.309965515136719,
      "epoch": 2.1844380403458215,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045240556045040767,
      "loss": 4.7351,
      "mean_token_accuracy": 0.2434508442878723,
      "num_tokens": 52158228.0,
      "step": 22740
    },
    {
      "entropy": 5.467838287353516,
      "epoch": 2.1849183477425553,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045238447145737397,
      "loss": 4.8445,
      "mean_token_accuracy": 0.22892505377531053,
      "num_tokens": 52169515.0,
      "step": 22745
    },
    {
      "entropy": 5.412427616119385,
      "epoch": 2.185398655139289,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045236337834599966,
      "loss": 4.7923,
      "mean_token_accuracy": 0.23637601286172866,
      "num_tokens": 52179846.0,
      "step": 22750
    },
    {
      "entropy": 5.22548999786377,
      "epoch": 2.185878962536023,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045234228111677434,
      "loss": 4.6366,
      "mean_token_accuracy": 0.2481299787759781,
      "num_tokens": 52191095.0,
      "step": 22755
    },
    {
      "entropy": 5.384168577194214,
      "epoch": 2.186359269932757,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004523211797701878,
      "loss": 4.7986,
      "mean_token_accuracy": 0.23907318860292434,
      "num_tokens": 52202289.0,
      "step": 22760
    },
    {
      "entropy": 5.326616668701172,
      "epoch": 2.186839577329491,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045230007430673014,
      "loss": 4.7268,
      "mean_token_accuracy": 0.25085781663656237,
      "num_tokens": 52214424.0,
      "step": 22765
    },
    {
      "entropy": 5.441716718673706,
      "epoch": 2.1873198847262247,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004522789647268911,
      "loss": 4.7802,
      "mean_token_accuracy": 0.2377971038222313,
      "num_tokens": 52225446.0,
      "step": 22770
    },
    {
      "entropy": 5.362802124023437,
      "epoch": 2.1878001921229586,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004522578510311609,
      "loss": 4.6978,
      "mean_token_accuracy": 0.2405153274536133,
      "num_tokens": 52236063.0,
      "step": 22775
    },
    {
      "entropy": 5.303202962875366,
      "epoch": 2.1882804995196925,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045223673322002984,
      "loss": 4.6969,
      "mean_token_accuracy": 0.2475898504257202,
      "num_tokens": 52248618.0,
      "step": 22780
    },
    {
      "entropy": 5.3797464847564695,
      "epoch": 2.1887608069164264,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045221561129398804,
      "loss": 4.8398,
      "mean_token_accuracy": 0.23435880690813066,
      "num_tokens": 52261913.0,
      "step": 22785
    },
    {
      "entropy": 5.417964744567871,
      "epoch": 2.1892411143131603,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000452194485253526,
      "loss": 4.7213,
      "mean_token_accuracy": 0.24504962414503098,
      "num_tokens": 52273838.0,
      "step": 22790
    },
    {
      "entropy": 5.325524234771729,
      "epoch": 2.189721421709894,
      "grad_norm": 1.125,
      "learning_rate": 0.0004521733550991342,
      "loss": 4.7677,
      "mean_token_accuracy": 0.24584855884313583,
      "num_tokens": 52285464.0,
      "step": 22795
    },
    {
      "entropy": 5.309654569625854,
      "epoch": 2.1902017291066285,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045215222083130316,
      "loss": 4.7027,
      "mean_token_accuracy": 0.24640196114778518,
      "num_tokens": 52295800.0,
      "step": 22800
    },
    {
      "entropy": 5.387327575683594,
      "epoch": 2.1906820365033624,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004521310824505236,
      "loss": 4.8113,
      "mean_token_accuracy": 0.24023323655128478,
      "num_tokens": 52307243.0,
      "step": 22805
    },
    {
      "entropy": 5.3488575458526615,
      "epoch": 2.1911623439000962,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045210993995728623,
      "loss": 4.7249,
      "mean_token_accuracy": 0.245066037774086,
      "num_tokens": 52317065.0,
      "step": 22810
    },
    {
      "entropy": 5.354053115844726,
      "epoch": 2.19164265129683,
      "grad_norm": 1.109375,
      "learning_rate": 0.000452088793352082,
      "loss": 4.7399,
      "mean_token_accuracy": 0.23981525003910065,
      "num_tokens": 52328385.0,
      "step": 22815
    },
    {
      "entropy": 5.276601791381836,
      "epoch": 2.192122958693564,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004520676426354018,
      "loss": 4.6806,
      "mean_token_accuracy": 0.24795373231172563,
      "num_tokens": 52340046.0,
      "step": 22820
    },
    {
      "entropy": 5.283549737930298,
      "epoch": 2.192603266090298,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004520464878077368,
      "loss": 4.6951,
      "mean_token_accuracy": 0.24655756652355193,
      "num_tokens": 52352071.0,
      "step": 22825
    },
    {
      "entropy": 5.458646535873413,
      "epoch": 2.1930835734870318,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045202532886957805,
      "loss": 4.8136,
      "mean_token_accuracy": 0.24131188541650772,
      "num_tokens": 52364173.0,
      "step": 22830
    },
    {
      "entropy": 5.339965724945069,
      "epoch": 2.1935638808837656,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045200416582141676,
      "loss": 4.6554,
      "mean_token_accuracy": 0.24562277048826217,
      "num_tokens": 52374690.0,
      "step": 22835
    },
    {
      "entropy": 5.409131193161011,
      "epoch": 2.1940441882804995,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004519829986637444,
      "loss": 4.9023,
      "mean_token_accuracy": 0.23564806282520295,
      "num_tokens": 52385661.0,
      "step": 22840
    },
    {
      "entropy": 5.37903938293457,
      "epoch": 2.1945244956772334,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004519618273970523,
      "loss": 4.8105,
      "mean_token_accuracy": 0.23963706344366073,
      "num_tokens": 52396759.0,
      "step": 22845
    },
    {
      "entropy": 5.475943613052368,
      "epoch": 2.1950048030739673,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045194065202183205,
      "loss": 4.8648,
      "mean_token_accuracy": 0.238334122300148,
      "num_tokens": 52409122.0,
      "step": 22850
    },
    {
      "entropy": 5.505629348754883,
      "epoch": 2.195485110470701,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004519194725385752,
      "loss": 4.838,
      "mean_token_accuracy": 0.2373344272375107,
      "num_tokens": 52421318.0,
      "step": 22855
    },
    {
      "entropy": 5.265140724182129,
      "epoch": 2.195965417867435,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045189828894777364,
      "loss": 4.707,
      "mean_token_accuracy": 0.24899120777845382,
      "num_tokens": 52432595.0,
      "step": 22860
    },
    {
      "entropy": 5.310952472686767,
      "epoch": 2.196445725264169,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045187710124991904,
      "loss": 4.715,
      "mean_token_accuracy": 0.24419403523206712,
      "num_tokens": 52444423.0,
      "step": 22865
    },
    {
      "entropy": 5.425644016265869,
      "epoch": 2.196926032660903,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004518559094455034,
      "loss": 4.8323,
      "mean_token_accuracy": 0.2365770161151886,
      "num_tokens": 52457183.0,
      "step": 22870
    },
    {
      "entropy": 5.3973705768585205,
      "epoch": 2.1974063400576367,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004518347135350187,
      "loss": 4.7569,
      "mean_token_accuracy": 0.237899848818779,
      "num_tokens": 52469624.0,
      "step": 22875
    },
    {
      "entropy": 5.438340187072754,
      "epoch": 2.1978866474543706,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045181351351895703,
      "loss": 4.8653,
      "mean_token_accuracy": 0.23445328027009965,
      "num_tokens": 52481369.0,
      "step": 22880
    },
    {
      "entropy": 5.356675338745117,
      "epoch": 2.198366954851105,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004517923093978106,
      "loss": 4.6695,
      "mean_token_accuracy": 0.24875199645757676,
      "num_tokens": 52493315.0,
      "step": 22885
    },
    {
      "entropy": 5.355348062515259,
      "epoch": 2.1988472622478388,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004517711011720717,
      "loss": 4.7738,
      "mean_token_accuracy": 0.24247583150863647,
      "num_tokens": 52504929.0,
      "step": 22890
    },
    {
      "entropy": 5.3211784839630125,
      "epoch": 2.1993275696445727,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004517498888422327,
      "loss": 4.6691,
      "mean_token_accuracy": 0.24199117720127106,
      "num_tokens": 52518364.0,
      "step": 22895
    },
    {
      "entropy": 5.346278238296509,
      "epoch": 2.1998078770413065,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004517286724087862,
      "loss": 4.7214,
      "mean_token_accuracy": 0.24185704290866852,
      "num_tokens": 52530359.0,
      "step": 22900
    },
    {
      "entropy": 5.337815046310425,
      "epoch": 2.2002881844380404,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004517074518722247,
      "loss": 4.7555,
      "mean_token_accuracy": 0.24640928953886032,
      "num_tokens": 52542242.0,
      "step": 22905
    },
    {
      "entropy": 5.415160512924194,
      "epoch": 2.2007684918347743,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045168622723304084,
      "loss": 4.7942,
      "mean_token_accuracy": 0.24159536510705948,
      "num_tokens": 52552838.0,
      "step": 22910
    },
    {
      "entropy": 5.366970205307007,
      "epoch": 2.201248799231508,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004516649984917274,
      "loss": 4.7171,
      "mean_token_accuracy": 0.24212263971567155,
      "num_tokens": 52564568.0,
      "step": 22915
    },
    {
      "entropy": 5.394775104522705,
      "epoch": 2.201729106628242,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045164376564877734,
      "loss": 4.7958,
      "mean_token_accuracy": 0.2362925186753273,
      "num_tokens": 52577563.0,
      "step": 22920
    },
    {
      "entropy": 5.390401077270508,
      "epoch": 2.202209414024976,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045162252870468354,
      "loss": 4.8002,
      "mean_token_accuracy": 0.2367830127477646,
      "num_tokens": 52590295.0,
      "step": 22925
    },
    {
      "entropy": 5.367273807525635,
      "epoch": 2.20268972142171,
      "grad_norm": 1.0,
      "learning_rate": 0.0004516012876599391,
      "loss": 4.7283,
      "mean_token_accuracy": 0.24799194931983948,
      "num_tokens": 52602727.0,
      "step": 22930
    },
    {
      "entropy": 5.33521032333374,
      "epoch": 2.2031700288184437,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045158004251503715,
      "loss": 4.6494,
      "mean_token_accuracy": 0.25390335023403166,
      "num_tokens": 52613422.0,
      "step": 22935
    },
    {
      "entropy": 5.284107685089111,
      "epoch": 2.2036503362151776,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045155879327047087,
      "loss": 4.6756,
      "mean_token_accuracy": 0.25340172052383425,
      "num_tokens": 52624907.0,
      "step": 22940
    },
    {
      "entropy": 5.402729749679565,
      "epoch": 2.2041306436119115,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004515375399267338,
      "loss": 4.8002,
      "mean_token_accuracy": 0.2358044669032097,
      "num_tokens": 52635864.0,
      "step": 22945
    },
    {
      "entropy": 5.410293245315552,
      "epoch": 2.2046109510086453,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045151628248431925,
      "loss": 4.8362,
      "mean_token_accuracy": 0.2375938504934311,
      "num_tokens": 52647023.0,
      "step": 22950
    },
    {
      "entropy": 5.400002574920654,
      "epoch": 2.2050912584053792,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045149502094372077,
      "loss": 4.7985,
      "mean_token_accuracy": 0.23991550654172897,
      "num_tokens": 52658846.0,
      "step": 22955
    },
    {
      "entropy": 5.300663757324219,
      "epoch": 2.2055715658021136,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045147375530543195,
      "loss": 4.5863,
      "mean_token_accuracy": 0.25751482099294665,
      "num_tokens": 52671363.0,
      "step": 22960
    },
    {
      "entropy": 5.411427688598633,
      "epoch": 2.2060518731988474,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045145248556994653,
      "loss": 4.8222,
      "mean_token_accuracy": 0.23961656242609025,
      "num_tokens": 52682734.0,
      "step": 22965
    },
    {
      "entropy": 5.344496488571167,
      "epoch": 2.2065321805955813,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004514312117377584,
      "loss": 4.6781,
      "mean_token_accuracy": 0.25134668350219724,
      "num_tokens": 52693272.0,
      "step": 22970
    },
    {
      "entropy": 5.336364698410034,
      "epoch": 2.207012487992315,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004514099338093613,
      "loss": 4.7669,
      "mean_token_accuracy": 0.23930230587720872,
      "num_tokens": 52703395.0,
      "step": 22975
    },
    {
      "entropy": 5.316155099868775,
      "epoch": 2.207492795389049,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004513886517852496,
      "loss": 4.6602,
      "mean_token_accuracy": 0.2469403252005577,
      "num_tokens": 52715870.0,
      "step": 22980
    },
    {
      "entropy": 5.369192600250244,
      "epoch": 2.207973102785783,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000451367365665917,
      "loss": 4.7575,
      "mean_token_accuracy": 0.24813996255397797,
      "num_tokens": 52726922.0,
      "step": 22985
    },
    {
      "entropy": 5.366548776626587,
      "epoch": 2.208453410182517,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045134607545185785,
      "loss": 4.7546,
      "mean_token_accuracy": 0.24247103780508042,
      "num_tokens": 52738211.0,
      "step": 22990
    },
    {
      "entropy": 5.391791486740113,
      "epoch": 2.2089337175792507,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004513247811435666,
      "loss": 4.7484,
      "mean_token_accuracy": 0.24238137900829315,
      "num_tokens": 52749267.0,
      "step": 22995
    },
    {
      "entropy": 5.272260904312134,
      "epoch": 2.2094140249759846,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045130348274153735,
      "loss": 4.7133,
      "mean_token_accuracy": 0.24371498823165894,
      "num_tokens": 52760520.0,
      "step": 23000
    },
    {
      "entropy": 5.298842811584473,
      "epoch": 2.2098943323727185,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045128218024626486,
      "loss": 4.686,
      "mean_token_accuracy": 0.2566222339868546,
      "num_tokens": 52772865.0,
      "step": 23005
    },
    {
      "entropy": 5.3887992858886715,
      "epoch": 2.2103746397694524,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004512608736582436,
      "loss": 4.7242,
      "mean_token_accuracy": 0.24449869841337205,
      "num_tokens": 52783935.0,
      "step": 23010
    },
    {
      "entropy": 5.2871020317077635,
      "epoch": 2.2108549471661862,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004512395629779682,
      "loss": 4.6948,
      "mean_token_accuracy": 0.248355533182621,
      "num_tokens": 52794980.0,
      "step": 23015
    },
    {
      "entropy": 5.429094409942627,
      "epoch": 2.21133525456292,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004512182482059335,
      "loss": 4.8417,
      "mean_token_accuracy": 0.23077597022056578,
      "num_tokens": 52806426.0,
      "step": 23020
    },
    {
      "entropy": 5.267021942138672,
      "epoch": 2.211815561959654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004511969293426343,
      "loss": 4.6813,
      "mean_token_accuracy": 0.2502855613827705,
      "num_tokens": 52816832.0,
      "step": 23025
    },
    {
      "entropy": 5.353589725494385,
      "epoch": 2.212295869356388,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045117560638856567,
      "loss": 4.7213,
      "mean_token_accuracy": 0.24701728522777558,
      "num_tokens": 52829913.0,
      "step": 23030
    },
    {
      "entropy": 5.390307426452637,
      "epoch": 2.212776176753122,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004511542793442225,
      "loss": 4.7676,
      "mean_token_accuracy": 0.2443980947136879,
      "num_tokens": 52841300.0,
      "step": 23035
    },
    {
      "entropy": 5.3457804203033445,
      "epoch": 2.213256484149856,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004511329482101001,
      "loss": 4.6975,
      "mean_token_accuracy": 0.24186426103115083,
      "num_tokens": 52852862.0,
      "step": 23040
    },
    {
      "entropy": 5.37824592590332,
      "epoch": 2.21373679154659,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004511116129866936,
      "loss": 4.7669,
      "mean_token_accuracy": 0.23803505301475525,
      "num_tokens": 52863499.0,
      "step": 23045
    },
    {
      "entropy": 5.434776544570923,
      "epoch": 2.214217098943324,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045109027367449845,
      "loss": 4.8045,
      "mean_token_accuracy": 0.24557927548885344,
      "num_tokens": 52875459.0,
      "step": 23050
    },
    {
      "entropy": 5.344277715682983,
      "epoch": 2.2146974063400577,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045106893027400995,
      "loss": 4.6927,
      "mean_token_accuracy": 0.2430109202861786,
      "num_tokens": 52886758.0,
      "step": 23055
    },
    {
      "entropy": 5.362405681610108,
      "epoch": 2.2151777137367916,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045104758278572375,
      "loss": 4.7048,
      "mean_token_accuracy": 0.24113674610853195,
      "num_tokens": 52898703.0,
      "step": 23060
    },
    {
      "entropy": 5.396484613418579,
      "epoch": 2.2156580211335255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004510262312101355,
      "loss": 4.749,
      "mean_token_accuracy": 0.23642770946025848,
      "num_tokens": 52910455.0,
      "step": 23065
    },
    {
      "entropy": 5.273646688461303,
      "epoch": 2.2161383285302594,
      "grad_norm": 1.0,
      "learning_rate": 0.0004510048755477407,
      "loss": 4.6541,
      "mean_token_accuracy": 0.24454946517944337,
      "num_tokens": 52921992.0,
      "step": 23070
    },
    {
      "entropy": 5.349497604370117,
      "epoch": 2.2166186359269933,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004509835157990354,
      "loss": 4.7877,
      "mean_token_accuracy": 0.240719336271286,
      "num_tokens": 52934994.0,
      "step": 23075
    },
    {
      "entropy": 5.422077655792236,
      "epoch": 2.217098943323727,
      "grad_norm": 1.125,
      "learning_rate": 0.00045096215196451547,
      "loss": 4.858,
      "mean_token_accuracy": 0.2349883183836937,
      "num_tokens": 52945745.0,
      "step": 23080
    },
    {
      "entropy": 5.40758261680603,
      "epoch": 2.217579250720461,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045094078404467683,
      "loss": 4.8401,
      "mean_token_accuracy": 0.2356007143855095,
      "num_tokens": 52957470.0,
      "step": 23085
    },
    {
      "entropy": 5.422402572631836,
      "epoch": 2.218059558117195,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045091941204001564,
      "loss": 4.7651,
      "mean_token_accuracy": 0.24167854636907576,
      "num_tokens": 52968584.0,
      "step": 23090
    },
    {
      "entropy": 5.35484733581543,
      "epoch": 2.218539865513929,
      "grad_norm": 1.03125,
      "learning_rate": 0.000450898035951028,
      "loss": 4.7484,
      "mean_token_accuracy": 0.24415334016084672,
      "num_tokens": 52980187.0,
      "step": 23095
    },
    {
      "entropy": 5.351089429855347,
      "epoch": 2.2190201729106627,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045087665577821034,
      "loss": 4.697,
      "mean_token_accuracy": 0.24790607541799545,
      "num_tokens": 52990478.0,
      "step": 23100
    },
    {
      "entropy": 5.3712762832641605,
      "epoch": 2.2195004803073966,
      "grad_norm": 1.203125,
      "learning_rate": 0.000450855271522059,
      "loss": 4.8376,
      "mean_token_accuracy": 0.23397158086299896,
      "num_tokens": 53001130.0,
      "step": 23105
    },
    {
      "entropy": 5.424617099761963,
      "epoch": 2.219980787704131,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045083388318307044,
      "loss": 4.8085,
      "mean_token_accuracy": 0.23893664926290512,
      "num_tokens": 53013057.0,
      "step": 23110
    },
    {
      "entropy": 5.454006767272949,
      "epoch": 2.2204610951008648,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004508124907617411,
      "loss": 4.8224,
      "mean_token_accuracy": 0.240672504901886,
      "num_tokens": 53024333.0,
      "step": 23115
    },
    {
      "entropy": 5.336420726776123,
      "epoch": 2.2209414024975986,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004507910942585679,
      "loss": 4.7513,
      "mean_token_accuracy": 0.2397530809044838,
      "num_tokens": 53036723.0,
      "step": 23120
    },
    {
      "entropy": 5.392857599258423,
      "epoch": 2.2214217098943325,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004507696936740475,
      "loss": 4.8025,
      "mean_token_accuracy": 0.24131497740745544,
      "num_tokens": 53048183.0,
      "step": 23125
    },
    {
      "entropy": 5.354008722305298,
      "epoch": 2.2219020172910664,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004507482890086767,
      "loss": 4.7593,
      "mean_token_accuracy": 0.2451452359557152,
      "num_tokens": 53061467.0,
      "step": 23130
    },
    {
      "entropy": 5.3471739292144775,
      "epoch": 2.2223823246878003,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004507268802629525,
      "loss": 4.7596,
      "mean_token_accuracy": 0.24305087924003602,
      "num_tokens": 53072833.0,
      "step": 23135
    },
    {
      "entropy": 5.3908086776733395,
      "epoch": 2.222862632084534,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004507054674373719,
      "loss": 4.7074,
      "mean_token_accuracy": 0.24224669188261033,
      "num_tokens": 53084109.0,
      "step": 23140
    },
    {
      "entropy": 5.390872526168823,
      "epoch": 2.223342939481268,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045068405053243216,
      "loss": 4.7567,
      "mean_token_accuracy": 0.24402753710746766,
      "num_tokens": 53095997.0,
      "step": 23145
    },
    {
      "entropy": 5.326051330566406,
      "epoch": 2.223823246878002,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004506626295486304,
      "loss": 4.6698,
      "mean_token_accuracy": 0.2524714425206184,
      "num_tokens": 53107518.0,
      "step": 23150
    },
    {
      "entropy": 5.340564823150634,
      "epoch": 2.224303554274736,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045064120448646405,
      "loss": 4.7407,
      "mean_token_accuracy": 0.2447209596633911,
      "num_tokens": 53118908.0,
      "step": 23155
    },
    {
      "entropy": 5.3743256568908695,
      "epoch": 2.2247838616714697,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004506197753464304,
      "loss": 4.7303,
      "mean_token_accuracy": 0.24782205373048782,
      "num_tokens": 53129404.0,
      "step": 23160
    },
    {
      "entropy": 5.415022993087769,
      "epoch": 2.2252641690682036,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045059834212902707,
      "loss": 4.8231,
      "mean_token_accuracy": 0.23813145160675048,
      "num_tokens": 53141364.0,
      "step": 23165
    },
    {
      "entropy": 5.434082794189453,
      "epoch": 2.2257444764649374,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045057690483475167,
      "loss": 4.7212,
      "mean_token_accuracy": 0.24736321717500687,
      "num_tokens": 53153617.0,
      "step": 23170
    },
    {
      "entropy": 5.388888931274414,
      "epoch": 2.2262247838616713,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004505554634641019,
      "loss": 4.8146,
      "mean_token_accuracy": 0.2386634424328804,
      "num_tokens": 53166537.0,
      "step": 23175
    },
    {
      "entropy": 5.357851934432984,
      "epoch": 2.226705091258405,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045053401801757554,
      "loss": 4.708,
      "mean_token_accuracy": 0.24122390449047088,
      "num_tokens": 53178979.0,
      "step": 23180
    },
    {
      "entropy": 5.374495267868042,
      "epoch": 2.227185398655139,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045051256849567054,
      "loss": 4.7798,
      "mean_token_accuracy": 0.24311446398496628,
      "num_tokens": 53190177.0,
      "step": 23185
    },
    {
      "entropy": 5.365670394897461,
      "epoch": 2.227665706051873,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045049111489888486,
      "loss": 4.7468,
      "mean_token_accuracy": 0.24519308656454086,
      "num_tokens": 53201398.0,
      "step": 23190
    },
    {
      "entropy": 5.427424907684326,
      "epoch": 2.2281460134486073,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004504696572277165,
      "loss": 4.7775,
      "mean_token_accuracy": 0.23783231377601624,
      "num_tokens": 53213795.0,
      "step": 23195
    },
    {
      "entropy": 5.379670333862305,
      "epoch": 2.228626320845341,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045044819548266385,
      "loss": 4.8142,
      "mean_token_accuracy": 0.2334773987531662,
      "num_tokens": 53226831.0,
      "step": 23200
    },
    {
      "entropy": 5.432452630996704,
      "epoch": 2.229106628242075,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045042672966422506,
      "loss": 4.8296,
      "mean_token_accuracy": 0.24416715353727342,
      "num_tokens": 53238558.0,
      "step": 23205
    },
    {
      "entropy": 5.394069719314575,
      "epoch": 2.229586935638809,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045040525977289847,
      "loss": 4.7338,
      "mean_token_accuracy": 0.23982976377010345,
      "num_tokens": 53250376.0,
      "step": 23210
    },
    {
      "entropy": 5.3844420433044435,
      "epoch": 2.230067243035543,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004503837858091826,
      "loss": 4.7358,
      "mean_token_accuracy": 0.2401161789894104,
      "num_tokens": 53260872.0,
      "step": 23215
    },
    {
      "entropy": 5.351884984970093,
      "epoch": 2.2305475504322767,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045036230777357604,
      "loss": 4.7339,
      "mean_token_accuracy": 0.2454468473792076,
      "num_tokens": 53271363.0,
      "step": 23220
    },
    {
      "entropy": 5.423594284057617,
      "epoch": 2.2310278578290106,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004503408256665774,
      "loss": 4.857,
      "mean_token_accuracy": 0.2338402658700943,
      "num_tokens": 53282356.0,
      "step": 23225
    },
    {
      "entropy": 5.385366058349609,
      "epoch": 2.2315081652257445,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045031933948868545,
      "loss": 4.7706,
      "mean_token_accuracy": 0.23995175659656526,
      "num_tokens": 53293675.0,
      "step": 23230
    },
    {
      "entropy": 5.321435165405274,
      "epoch": 2.2319884726224783,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045029784924039903,
      "loss": 4.6685,
      "mean_token_accuracy": 0.251647575199604,
      "num_tokens": 53305193.0,
      "step": 23235
    },
    {
      "entropy": 5.382462406158448,
      "epoch": 2.2324687800192122,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045027635492221716,
      "loss": 4.6707,
      "mean_token_accuracy": 0.253790020942688,
      "num_tokens": 53315729.0,
      "step": 23240
    },
    {
      "entropy": 5.339874124526977,
      "epoch": 2.232949087415946,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045025485653463866,
      "loss": 4.7408,
      "mean_token_accuracy": 0.24487038999795913,
      "num_tokens": 53327220.0,
      "step": 23245
    },
    {
      "entropy": 5.358764791488648,
      "epoch": 2.23342939481268,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004502333540781628,
      "loss": 4.7783,
      "mean_token_accuracy": 0.2372737020254135,
      "num_tokens": 53337565.0,
      "step": 23250
    },
    {
      "entropy": 5.409684705734253,
      "epoch": 2.233909702209414,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004502118475532888,
      "loss": 4.7533,
      "mean_token_accuracy": 0.2465962365269661,
      "num_tokens": 53348791.0,
      "step": 23255
    },
    {
      "entropy": 5.331168079376221,
      "epoch": 2.2343900096061478,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000450190336960516,
      "loss": 4.7068,
      "mean_token_accuracy": 0.24385990351438522,
      "num_tokens": 53361341.0,
      "step": 23260
    },
    {
      "entropy": 5.3605423927307125,
      "epoch": 2.2348703170028816,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004501688223003438,
      "loss": 4.7809,
      "mean_token_accuracy": 0.23969177007675171,
      "num_tokens": 53372085.0,
      "step": 23265
    },
    {
      "entropy": 5.373791742324829,
      "epoch": 2.235350624399616,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004501473035732717,
      "loss": 4.7322,
      "mean_token_accuracy": 0.24274193793535231,
      "num_tokens": 53383665.0,
      "step": 23270
    },
    {
      "entropy": 5.3592729568481445,
      "epoch": 2.23583093179635,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004501257807797993,
      "loss": 4.73,
      "mean_token_accuracy": 0.23759220391511918,
      "num_tokens": 53394697.0,
      "step": 23275
    },
    {
      "entropy": 5.3164391040802,
      "epoch": 2.2363112391930837,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045010425392042624,
      "loss": 4.7557,
      "mean_token_accuracy": 0.24005646854639054,
      "num_tokens": 53405538.0,
      "step": 23280
    },
    {
      "entropy": 5.460251665115356,
      "epoch": 2.2367915465898176,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004500827229956524,
      "loss": 4.8582,
      "mean_token_accuracy": 0.2309081584215164,
      "num_tokens": 53416515.0,
      "step": 23285
    },
    {
      "entropy": 5.385433292388916,
      "epoch": 2.2372718539865515,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045006118800597757,
      "loss": 4.7117,
      "mean_token_accuracy": 0.23580761551856994,
      "num_tokens": 53426530.0,
      "step": 23290
    },
    {
      "entropy": 5.35495080947876,
      "epoch": 2.2377521613832854,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045003964895190177,
      "loss": 4.8184,
      "mean_token_accuracy": 0.239141084253788,
      "num_tokens": 53438514.0,
      "step": 23295
    },
    {
      "entropy": 5.318082189559936,
      "epoch": 2.2382324687800192,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004500181058339251,
      "loss": 4.6713,
      "mean_token_accuracy": 0.2513399675488472,
      "num_tokens": 53448845.0,
      "step": 23300
    },
    {
      "entropy": 5.393799877166748,
      "epoch": 2.238712776176753,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004499965586525478,
      "loss": 4.787,
      "mean_token_accuracy": 0.2417538657784462,
      "num_tokens": 53460526.0,
      "step": 23305
    },
    {
      "entropy": 5.496137285232544,
      "epoch": 2.239193083573487,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044997500740826993,
      "loss": 4.8259,
      "mean_token_accuracy": 0.24251023232936858,
      "num_tokens": 53471997.0,
      "step": 23310
    },
    {
      "entropy": 5.366572475433349,
      "epoch": 2.239673390970221,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000449953452101592,
      "loss": 4.7913,
      "mean_token_accuracy": 0.2391184613108635,
      "num_tokens": 53483097.0,
      "step": 23315
    },
    {
      "entropy": 5.32829008102417,
      "epoch": 2.2401536983669548,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044993189273301445,
      "loss": 4.706,
      "mean_token_accuracy": 0.2559645101428032,
      "num_tokens": 53494877.0,
      "step": 23320
    },
    {
      "entropy": 5.319917678833008,
      "epoch": 2.2406340057636887,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004499103293030377,
      "loss": 4.7164,
      "mean_token_accuracy": 0.24611409306526183,
      "num_tokens": 53507065.0,
      "step": 23325
    },
    {
      "entropy": 5.405312585830688,
      "epoch": 2.2411143131604225,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004498887618121625,
      "loss": 4.7929,
      "mean_token_accuracy": 0.24261318892240524,
      "num_tokens": 53518188.0,
      "step": 23330
    },
    {
      "entropy": 5.308767938613892,
      "epoch": 2.2415946205571564,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004498671902608897,
      "loss": 4.6537,
      "mean_token_accuracy": 0.24955554753541948,
      "num_tokens": 53529872.0,
      "step": 23335
    },
    {
      "entropy": 5.377794504165649,
      "epoch": 2.2420749279538903,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004498456146497198,
      "loss": 4.7344,
      "mean_token_accuracy": 0.23558282107114792,
      "num_tokens": 53540175.0,
      "step": 23340
    },
    {
      "entropy": 5.30223708152771,
      "epoch": 2.2425552353506246,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044982403497915405,
      "loss": 4.6643,
      "mean_token_accuracy": 0.24464835077524186,
      "num_tokens": 53552520.0,
      "step": 23345
    },
    {
      "entropy": 5.342087364196777,
      "epoch": 2.2430355427473585,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044980245124969333,
      "loss": 4.7421,
      "mean_token_accuracy": 0.24208449870347976,
      "num_tokens": 53563937.0,
      "step": 23350
    },
    {
      "entropy": 5.344053268432617,
      "epoch": 2.2435158501440924,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004497808634618386,
      "loss": 4.7232,
      "mean_token_accuracy": 0.2384442836046219,
      "num_tokens": 53574654.0,
      "step": 23355
    },
    {
      "entropy": 5.371211814880371,
      "epoch": 2.2439961575408263,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004497592716160913,
      "loss": 4.7642,
      "mean_token_accuracy": 0.2418166369199753,
      "num_tokens": 53586848.0,
      "step": 23360
    },
    {
      "entropy": 5.3486950397491455,
      "epoch": 2.24447646493756,
      "grad_norm": 1.125,
      "learning_rate": 0.00044973767571295273,
      "loss": 4.7446,
      "mean_token_accuracy": 0.24233797192573547,
      "num_tokens": 53598200.0,
      "step": 23365
    },
    {
      "entropy": 5.347919082641601,
      "epoch": 2.244956772334294,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004497160757529241,
      "loss": 4.6974,
      "mean_token_accuracy": 0.2465250849723816,
      "num_tokens": 53609957.0,
      "step": 23370
    },
    {
      "entropy": 5.3781211376190186,
      "epoch": 2.245437079731028,
      "grad_norm": 1.125,
      "learning_rate": 0.00044969447173650695,
      "loss": 4.7663,
      "mean_token_accuracy": 0.2489526614546776,
      "num_tokens": 53621029.0,
      "step": 23375
    },
    {
      "entropy": 5.347478246688842,
      "epoch": 2.245917387127762,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000449672863664203,
      "loss": 4.7445,
      "mean_token_accuracy": 0.24210285693407058,
      "num_tokens": 53633279.0,
      "step": 23380
    },
    {
      "entropy": 5.353676223754883,
      "epoch": 2.2463976945244957,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044965125153651375,
      "loss": 4.724,
      "mean_token_accuracy": 0.24776716530323029,
      "num_tokens": 53643959.0,
      "step": 23385
    },
    {
      "entropy": 5.346708345413208,
      "epoch": 2.2468780019212296,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004496296353539411,
      "loss": 4.7265,
      "mean_token_accuracy": 0.24483227580785752,
      "num_tokens": 53655215.0,
      "step": 23390
    },
    {
      "entropy": 5.402083826065064,
      "epoch": 2.2473583093179634,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004496080151169869,
      "loss": 4.7446,
      "mean_token_accuracy": 0.24383836686611177,
      "num_tokens": 53666557.0,
      "step": 23395
    },
    {
      "entropy": 5.39082350730896,
      "epoch": 2.2478386167146973,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044958639082615294,
      "loss": 4.8023,
      "mean_token_accuracy": 0.24004835784435272,
      "num_tokens": 53679626.0,
      "step": 23400
    },
    {
      "entropy": 5.349415588378906,
      "epoch": 2.248318924111431,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004495647624819415,
      "loss": 4.7388,
      "mean_token_accuracy": 0.2475324422121048,
      "num_tokens": 53691681.0,
      "step": 23405
    },
    {
      "entropy": 5.353669357299805,
      "epoch": 2.248799231508165,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044954313008485457,
      "loss": 4.6992,
      "mean_token_accuracy": 0.2472623586654663,
      "num_tokens": 53703320.0,
      "step": 23410
    },
    {
      "entropy": 5.325284004211426,
      "epoch": 2.249279538904899,
      "grad_norm": 1.0,
      "learning_rate": 0.00044952149363539453,
      "loss": 4.6807,
      "mean_token_accuracy": 0.2459624856710434,
      "num_tokens": 53714867.0,
      "step": 23415
    },
    {
      "entropy": 5.362803745269775,
      "epoch": 2.249759846301633,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004494998531340635,
      "loss": 4.7992,
      "mean_token_accuracy": 0.24262806177139282,
      "num_tokens": 53726121.0,
      "step": 23420
    },
    {
      "entropy": 5.418250560760498,
      "epoch": 2.2502401536983667,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004494782085813641,
      "loss": 4.7876,
      "mean_token_accuracy": 0.23684019446372986,
      "num_tokens": 53736966.0,
      "step": 23425
    },
    {
      "entropy": 5.335140657424927,
      "epoch": 2.250720461095101,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004494565599777988,
      "loss": 4.6808,
      "mean_token_accuracy": 0.25315539389848707,
      "num_tokens": 53747446.0,
      "step": 23430
    },
    {
      "entropy": 5.341621541976929,
      "epoch": 2.251200768491835,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044943490732387025,
      "loss": 4.7989,
      "mean_token_accuracy": 0.23252898454666138,
      "num_tokens": 53759588.0,
      "step": 23435
    },
    {
      "entropy": 5.402499866485596,
      "epoch": 2.251681075888569,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004494132506200811,
      "loss": 4.7889,
      "mean_token_accuracy": 0.24071105271577836,
      "num_tokens": 53770314.0,
      "step": 23440
    },
    {
      "entropy": 5.35519232749939,
      "epoch": 2.2521613832853027,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004493915898669341,
      "loss": 4.7153,
      "mean_token_accuracy": 0.24488306194543838,
      "num_tokens": 53781251.0,
      "step": 23445
    },
    {
      "entropy": 5.425385046005249,
      "epoch": 2.2526416906820366,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004493699250649323,
      "loss": 4.8261,
      "mean_token_accuracy": 0.2376004710793495,
      "num_tokens": 53792676.0,
      "step": 23450
    },
    {
      "entropy": 5.432702493667603,
      "epoch": 2.2531219980787704,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004493482562145786,
      "loss": 4.8189,
      "mean_token_accuracy": 0.23628537207841874,
      "num_tokens": 53802714.0,
      "step": 23455
    },
    {
      "entropy": 5.333294534683228,
      "epoch": 2.2536023054755043,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044932658331637605,
      "loss": 4.7805,
      "mean_token_accuracy": 0.23623937368392944,
      "num_tokens": 53815043.0,
      "step": 23460
    },
    {
      "entropy": 5.381735992431641,
      "epoch": 2.254082612872238,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004493049063708279,
      "loss": 4.7579,
      "mean_token_accuracy": 0.2326949343085289,
      "num_tokens": 53827903.0,
      "step": 23465
    },
    {
      "entropy": 5.448827314376831,
      "epoch": 2.254562920268972,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044928322537843746,
      "loss": 4.8416,
      "mean_token_accuracy": 0.2404489442706108,
      "num_tokens": 53839574.0,
      "step": 23470
    },
    {
      "entropy": 5.36792688369751,
      "epoch": 2.255043227665706,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044926154033970793,
      "loss": 4.6914,
      "mean_token_accuracy": 0.25175562500953674,
      "num_tokens": 53850488.0,
      "step": 23475
    },
    {
      "entropy": 5.383304405212402,
      "epoch": 2.25552353506244,
      "grad_norm": 1.09375,
      "learning_rate": 0.000449239851255143,
      "loss": 4.7153,
      "mean_token_accuracy": 0.2464928597211838,
      "num_tokens": 53860766.0,
      "step": 23480
    },
    {
      "entropy": 5.323852252960205,
      "epoch": 2.2560038424591737,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044921815812524606,
      "loss": 4.7279,
      "mean_token_accuracy": 0.24258071184158325,
      "num_tokens": 53872088.0,
      "step": 23485
    },
    {
      "entropy": 5.422620820999145,
      "epoch": 2.2564841498559076,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044919646095052077,
      "loss": 4.7569,
      "mean_token_accuracy": 0.24211328774690627,
      "num_tokens": 53883302.0,
      "step": 23490
    },
    {
      "entropy": 5.2782533168792725,
      "epoch": 2.2569644572526415,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000449174759731471,
      "loss": 4.6517,
      "mean_token_accuracy": 0.2557285025715828,
      "num_tokens": 53894310.0,
      "step": 23495
    },
    {
      "entropy": 5.378604364395142,
      "epoch": 2.2574447646493754,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044915305446860046,
      "loss": 4.7851,
      "mean_token_accuracy": 0.23200318813323975,
      "num_tokens": 53906760.0,
      "step": 23500
    },
    {
      "entropy": 5.373687219619751,
      "epoch": 2.2579250720461097,
      "grad_norm": 1.125,
      "learning_rate": 0.00044913134516241305,
      "loss": 4.7754,
      "mean_token_accuracy": 0.2394435316324234,
      "num_tokens": 53918758.0,
      "step": 23505
    },
    {
      "entropy": 5.306469583511353,
      "epoch": 2.2584053794428436,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004491096318134129,
      "loss": 4.6987,
      "mean_token_accuracy": 0.24450825601816178,
      "num_tokens": 53932574.0,
      "step": 23510
    },
    {
      "entropy": 5.417883014678955,
      "epoch": 2.2588856868395775,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004490879144221042,
      "loss": 4.7999,
      "mean_token_accuracy": 0.24212246984243393,
      "num_tokens": 53943801.0,
      "step": 23515
    },
    {
      "entropy": 5.368294095993042,
      "epoch": 2.2593659942363113,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044906619298899097,
      "loss": 4.7336,
      "mean_token_accuracy": 0.2522010147571564,
      "num_tokens": 53955458.0,
      "step": 23520
    },
    {
      "entropy": 5.361940479278564,
      "epoch": 2.2598463016330452,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004490444675145777,
      "loss": 4.769,
      "mean_token_accuracy": 0.23688667118549347,
      "num_tokens": 53967242.0,
      "step": 23525
    },
    {
      "entropy": 5.377488517761231,
      "epoch": 2.260326609029779,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004490227379993686,
      "loss": 4.79,
      "mean_token_accuracy": 0.24150825738906861,
      "num_tokens": 53978953.0,
      "step": 23530
    },
    {
      "entropy": 5.441363286972046,
      "epoch": 2.260806916426513,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004490010044438683,
      "loss": 4.777,
      "mean_token_accuracy": 0.23673059344291686,
      "num_tokens": 53990553.0,
      "step": 23535
    },
    {
      "entropy": 5.308207082748413,
      "epoch": 2.261287223823247,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044897926684858133,
      "loss": 4.7816,
      "mean_token_accuracy": 0.24735689014196396,
      "num_tokens": 54002187.0,
      "step": 23540
    },
    {
      "entropy": 5.3834826946258545,
      "epoch": 2.2617675312199808,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044895752521401246,
      "loss": 4.754,
      "mean_token_accuracy": 0.2371496468782425,
      "num_tokens": 54014202.0,
      "step": 23545
    },
    {
      "entropy": 5.357447767257691,
      "epoch": 2.2622478386167146,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004489357795406663,
      "loss": 4.7518,
      "mean_token_accuracy": 0.24634327590465546,
      "num_tokens": 54026286.0,
      "step": 23550
    },
    {
      "entropy": 5.349040699005127,
      "epoch": 2.2627281460134485,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004489140298290479,
      "loss": 4.756,
      "mean_token_accuracy": 0.24483545124530792,
      "num_tokens": 54038845.0,
      "step": 23555
    },
    {
      "entropy": 5.45167293548584,
      "epoch": 2.2632084534101824,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044889227607966217,
      "loss": 4.8245,
      "mean_token_accuracy": 0.2407233566045761,
      "num_tokens": 54049991.0,
      "step": 23560
    },
    {
      "entropy": 5.363229036331177,
      "epoch": 2.2636887608069163,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044887051829301406,
      "loss": 4.7242,
      "mean_token_accuracy": 0.24055353105068206,
      "num_tokens": 54061408.0,
      "step": 23565
    },
    {
      "entropy": 5.372832775115967,
      "epoch": 2.26416906820365,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044884875646960886,
      "loss": 4.7223,
      "mean_token_accuracy": 0.24083716720342635,
      "num_tokens": 54072670.0,
      "step": 23570
    },
    {
      "entropy": 5.414704322814941,
      "epoch": 2.264649375600384,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044882699060995175,
      "loss": 4.7498,
      "mean_token_accuracy": 0.23819730281829835,
      "num_tokens": 54084555.0,
      "step": 23575
    },
    {
      "entropy": 5.326103258132934,
      "epoch": 2.2651296829971184,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004488052207145481,
      "loss": 4.6931,
      "mean_token_accuracy": 0.2510451957583427,
      "num_tokens": 54096764.0,
      "step": 23580
    },
    {
      "entropy": 5.410356235504151,
      "epoch": 2.2656099903938522,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044878344678390324,
      "loss": 4.7762,
      "mean_token_accuracy": 0.24725412279367448,
      "num_tokens": 54109101.0,
      "step": 23585
    },
    {
      "entropy": 5.396112585067749,
      "epoch": 2.266090297790586,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044876166881852286,
      "loss": 4.7852,
      "mean_token_accuracy": 0.23819769024848939,
      "num_tokens": 54120217.0,
      "step": 23590
    },
    {
      "entropy": 5.4100696563720705,
      "epoch": 2.26657060518732,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004487398868189125,
      "loss": 4.7729,
      "mean_token_accuracy": 0.24125591218471526,
      "num_tokens": 54132059.0,
      "step": 23595
    },
    {
      "entropy": 5.45318717956543,
      "epoch": 2.267050912584054,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044871810078557777,
      "loss": 4.8231,
      "mean_token_accuracy": 0.23826612085103988,
      "num_tokens": 54141928.0,
      "step": 23600
    },
    {
      "entropy": 5.379159641265869,
      "epoch": 2.2675312199807878,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004486963107190247,
      "loss": 4.8471,
      "mean_token_accuracy": 0.23512840420007705,
      "num_tokens": 54153052.0,
      "step": 23605
    },
    {
      "entropy": 5.458614349365234,
      "epoch": 2.2680115273775217,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044867451661975894,
      "loss": 4.7862,
      "mean_token_accuracy": 0.2377358168363571,
      "num_tokens": 54163539.0,
      "step": 23610
    },
    {
      "entropy": 5.315417671203614,
      "epoch": 2.2684918347742555,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044865271848828673,
      "loss": 4.7328,
      "mean_token_accuracy": 0.24632785022258757,
      "num_tokens": 54176313.0,
      "step": 23615
    },
    {
      "entropy": 5.4021241664886475,
      "epoch": 2.2689721421709894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000448630916325114,
      "loss": 4.8183,
      "mean_token_accuracy": 0.23479231148958207,
      "num_tokens": 54187080.0,
      "step": 23620
    },
    {
      "entropy": 5.38437671661377,
      "epoch": 2.2694524495677233,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000448609110130747,
      "loss": 4.7866,
      "mean_token_accuracy": 0.23882126808166504,
      "num_tokens": 54197856.0,
      "step": 23625
    },
    {
      "entropy": 5.419962692260742,
      "epoch": 2.269932756964457,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044858729990569193,
      "loss": 4.7905,
      "mean_token_accuracy": 0.23693884760141373,
      "num_tokens": 54210865.0,
      "step": 23630
    },
    {
      "entropy": 5.302602672576905,
      "epoch": 2.270413064361191,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044856548565045523,
      "loss": 4.6837,
      "mean_token_accuracy": 0.2407823011279106,
      "num_tokens": 54222719.0,
      "step": 23635
    },
    {
      "entropy": 5.357883167266846,
      "epoch": 2.270893371757925,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044854366736554323,
      "loss": 4.7306,
      "mean_token_accuracy": 0.24650098085403443,
      "num_tokens": 54233874.0,
      "step": 23640
    },
    {
      "entropy": 5.297213792800903,
      "epoch": 2.271373679154659,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044852184505146274,
      "loss": 4.6488,
      "mean_token_accuracy": 0.2427629292011261,
      "num_tokens": 54245966.0,
      "step": 23645
    },
    {
      "entropy": 5.353053617477417,
      "epoch": 2.2718539865513927,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004485000187087202,
      "loss": 4.7264,
      "mean_token_accuracy": 0.2504178449511528,
      "num_tokens": 54256468.0,
      "step": 23650
    },
    {
      "entropy": 5.3082081317901615,
      "epoch": 2.272334293948127,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004484781883378224,
      "loss": 4.6688,
      "mean_token_accuracy": 0.23639036864042282,
      "num_tokens": 54266362.0,
      "step": 23655
    },
    {
      "entropy": 5.380661869049073,
      "epoch": 2.2728146013448605,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044845635393927623,
      "loss": 4.8552,
      "mean_token_accuracy": 0.23412602245807648,
      "num_tokens": 54278365.0,
      "step": 23660
    },
    {
      "entropy": 5.405053615570068,
      "epoch": 2.273294908741595,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004484345155135886,
      "loss": 4.8412,
      "mean_token_accuracy": 0.24001600593328476,
      "num_tokens": 54289310.0,
      "step": 23665
    },
    {
      "entropy": 5.344498348236084,
      "epoch": 2.2737752161383287,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004484126730612664,
      "loss": 4.6892,
      "mean_token_accuracy": 0.24424464106559754,
      "num_tokens": 54300421.0,
      "step": 23670
    },
    {
      "entropy": 5.336839056015014,
      "epoch": 2.2742555235350626,
      "grad_norm": 1.078125,
      "learning_rate": 0.000448390826582817,
      "loss": 4.7484,
      "mean_token_accuracy": 0.24691020399332048,
      "num_tokens": 54312803.0,
      "step": 23675
    },
    {
      "entropy": 5.412374258041382,
      "epoch": 2.2747358309317964,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044836897607874744,
      "loss": 4.7931,
      "mean_token_accuracy": 0.24158488661050798,
      "num_tokens": 54324651.0,
      "step": 23680
    },
    {
      "entropy": 5.332076597213745,
      "epoch": 2.2752161383285303,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000448347121549565,
      "loss": 4.6945,
      "mean_token_accuracy": 0.2533195361495018,
      "num_tokens": 54334841.0,
      "step": 23685
    },
    {
      "entropy": 5.381388568878174,
      "epoch": 2.275696445725264,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004483252629957771,
      "loss": 4.74,
      "mean_token_accuracy": 0.2433346003293991,
      "num_tokens": 54346685.0,
      "step": 23690
    },
    {
      "entropy": 5.331369209289551,
      "epoch": 2.276176753121998,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044830340041789133,
      "loss": 4.7924,
      "mean_token_accuracy": 0.23552144914865494,
      "num_tokens": 54358710.0,
      "step": 23695
    },
    {
      "entropy": 5.377756071090698,
      "epoch": 2.276657060518732,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004482815338164152,
      "loss": 4.7969,
      "mean_token_accuracy": 0.24372245967388154,
      "num_tokens": 54369695.0,
      "step": 23700
    },
    {
      "entropy": 5.308825731277466,
      "epoch": 2.277137367915466,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004482596631918564,
      "loss": 4.7179,
      "mean_token_accuracy": 0.242182257771492,
      "num_tokens": 54381605.0,
      "step": 23705
    },
    {
      "entropy": 5.351789331436157,
      "epoch": 2.2776176753121997,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044823778854472267,
      "loss": 4.738,
      "mean_token_accuracy": 0.2453562095761299,
      "num_tokens": 54392309.0,
      "step": 23710
    },
    {
      "entropy": 5.3613427639007565,
      "epoch": 2.2780979827089336,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004482159098755219,
      "loss": 4.7741,
      "mean_token_accuracy": 0.24419075548648833,
      "num_tokens": 54403195.0,
      "step": 23715
    },
    {
      "entropy": 5.294277286529541,
      "epoch": 2.2785782901056675,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004481940271847621,
      "loss": 4.711,
      "mean_token_accuracy": 0.25550367385149003,
      "num_tokens": 54415105.0,
      "step": 23720
    },
    {
      "entropy": 5.346685981750488,
      "epoch": 2.2790585975024014,
      "grad_norm": 1.125,
      "learning_rate": 0.0004481721404729513,
      "loss": 4.7345,
      "mean_token_accuracy": 0.2450408712029457,
      "num_tokens": 54425798.0,
      "step": 23725
    },
    {
      "entropy": 5.4033355712890625,
      "epoch": 2.2795389048991357,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004481502497405975,
      "loss": 4.7873,
      "mean_token_accuracy": 0.238650906085968,
      "num_tokens": 54437254.0,
      "step": 23730
    },
    {
      "entropy": 5.383609676361084,
      "epoch": 2.280019212295869,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004481283549882091,
      "loss": 4.7533,
      "mean_token_accuracy": 0.2338685631752014,
      "num_tokens": 54448152.0,
      "step": 23735
    },
    {
      "entropy": 5.379763269424439,
      "epoch": 2.2804995196926034,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044810645621629443,
      "loss": 4.7889,
      "mean_token_accuracy": 0.23679124414920807,
      "num_tokens": 54460066.0,
      "step": 23740
    },
    {
      "entropy": 5.417216968536377,
      "epoch": 2.2809798270893373,
      "grad_norm": 1.0,
      "learning_rate": 0.00044808455342536176,
      "loss": 4.8116,
      "mean_token_accuracy": 0.2412234902381897,
      "num_tokens": 54471769.0,
      "step": 23745
    },
    {
      "entropy": 5.320485258102417,
      "epoch": 2.281460134486071,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044806264661591976,
      "loss": 4.6506,
      "mean_token_accuracy": 0.25938552916049956,
      "num_tokens": 54483630.0,
      "step": 23750
    },
    {
      "entropy": 5.40175838470459,
      "epoch": 2.281940441882805,
      "grad_norm": 1.0,
      "learning_rate": 0.0004480407357884771,
      "loss": 4.755,
      "mean_token_accuracy": 0.24135053902864456,
      "num_tokens": 54495696.0,
      "step": 23755
    },
    {
      "entropy": 5.33389983177185,
      "epoch": 2.282420749279539,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044801882094354226,
      "loss": 4.7727,
      "mean_token_accuracy": 0.23880326747894287,
      "num_tokens": 54507153.0,
      "step": 23760
    },
    {
      "entropy": 5.402502536773682,
      "epoch": 2.282901056676273,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004479969020816242,
      "loss": 4.8088,
      "mean_token_accuracy": 0.23793846219778061,
      "num_tokens": 54519922.0,
      "step": 23765
    },
    {
      "entropy": 5.374177360534668,
      "epoch": 2.2833813640730067,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044797497920323175,
      "loss": 4.7242,
      "mean_token_accuracy": 0.24292647391557692,
      "num_tokens": 54531857.0,
      "step": 23770
    },
    {
      "entropy": 5.362651443481445,
      "epoch": 2.2838616714697406,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004479530523088739,
      "loss": 4.7347,
      "mean_token_accuracy": 0.2488892987370491,
      "num_tokens": 54542079.0,
      "step": 23775
    },
    {
      "entropy": 5.2990048885345455,
      "epoch": 2.2843419788664745,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004479311213990598,
      "loss": 4.7102,
      "mean_token_accuracy": 0.24719039350748062,
      "num_tokens": 54552722.0,
      "step": 23780
    },
    {
      "entropy": 5.302193880081177,
      "epoch": 2.2848222862632084,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00044790918647429854,
      "loss": 4.7077,
      "mean_token_accuracy": 0.24708641171455384,
      "num_tokens": 54564224.0,
      "step": 23785
    },
    {
      "entropy": 5.377292156219482,
      "epoch": 2.2853025936599423,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044788724753509935,
      "loss": 4.6954,
      "mean_token_accuracy": 0.2455615445971489,
      "num_tokens": 54575553.0,
      "step": 23790
    },
    {
      "entropy": 5.404565095901489,
      "epoch": 2.285782901056676,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004478653045819717,
      "loss": 4.854,
      "mean_token_accuracy": 0.23626711070537568,
      "num_tokens": 54587315.0,
      "step": 23795
    },
    {
      "entropy": 5.42647795677185,
      "epoch": 2.28626320845341,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004478433576154249,
      "loss": 4.7651,
      "mean_token_accuracy": 0.24680692106485366,
      "num_tokens": 54598776.0,
      "step": 23800
    },
    {
      "entropy": 5.447894620895386,
      "epoch": 2.286743515850144,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004478214066359687,
      "loss": 4.8477,
      "mean_token_accuracy": 0.22989751249551774,
      "num_tokens": 54610272.0,
      "step": 23805
    },
    {
      "entropy": 5.390677213668823,
      "epoch": 2.287223823246878,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044779945164411254,
      "loss": 4.7974,
      "mean_token_accuracy": 0.2439188316464424,
      "num_tokens": 54622105.0,
      "step": 23810
    },
    {
      "entropy": 5.4279053688049315,
      "epoch": 2.287704130643612,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004477774926403662,
      "loss": 4.7574,
      "mean_token_accuracy": 0.24718030095100402,
      "num_tokens": 54631771.0,
      "step": 23815
    },
    {
      "entropy": 5.34711446762085,
      "epoch": 2.288184438040346,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004477555296252396,
      "loss": 4.7004,
      "mean_token_accuracy": 0.24802278578281403,
      "num_tokens": 54642724.0,
      "step": 23820
    },
    {
      "entropy": 5.295652961730957,
      "epoch": 2.28866474543708,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044773356259924255,
      "loss": 4.7133,
      "mean_token_accuracy": 0.24654761254787444,
      "num_tokens": 54655126.0,
      "step": 23825
    },
    {
      "entropy": 5.303250122070312,
      "epoch": 2.2891450528338138,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044771159156288505,
      "loss": 4.6775,
      "mean_token_accuracy": 0.250502310693264,
      "num_tokens": 54665334.0,
      "step": 23830
    },
    {
      "entropy": 5.3510094165802,
      "epoch": 2.2896253602305476,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004476896165166772,
      "loss": 4.6803,
      "mean_token_accuracy": 0.2460148498415947,
      "num_tokens": 54676205.0,
      "step": 23835
    },
    {
      "entropy": 5.419126653671265,
      "epoch": 2.2901056676272815,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044766763746112936,
      "loss": 4.7729,
      "mean_token_accuracy": 0.24228078424930571,
      "num_tokens": 54686586.0,
      "step": 23840
    },
    {
      "entropy": 5.383961391448975,
      "epoch": 2.2905859750240154,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004476456543967517,
      "loss": 4.7489,
      "mean_token_accuracy": 0.24086440056562425,
      "num_tokens": 54698879.0,
      "step": 23845
    },
    {
      "entropy": 5.345009994506836,
      "epoch": 2.2910662824207493,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044762366732405454,
      "loss": 4.7231,
      "mean_token_accuracy": 0.24357127845287324,
      "num_tokens": 54710140.0,
      "step": 23850
    },
    {
      "entropy": 5.313773059844971,
      "epoch": 2.291546589817483,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004476016762435484,
      "loss": 4.6441,
      "mean_token_accuracy": 0.25309446454048157,
      "num_tokens": 54721317.0,
      "step": 23855
    },
    {
      "entropy": 5.450733661651611,
      "epoch": 2.292026897214217,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004475796811557439,
      "loss": 4.906,
      "mean_token_accuracy": 0.22944701164960862,
      "num_tokens": 54733329.0,
      "step": 23860
    },
    {
      "entropy": 5.462297344207764,
      "epoch": 2.292507204610951,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044755768206115155,
      "loss": 4.8002,
      "mean_token_accuracy": 0.23181714117527008,
      "num_tokens": 54745230.0,
      "step": 23865
    },
    {
      "entropy": 5.4061564922332765,
      "epoch": 2.292987512007685,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004475356789602823,
      "loss": 4.7398,
      "mean_token_accuracy": 0.2424248367547989,
      "num_tokens": 54757428.0,
      "step": 23870
    },
    {
      "entropy": 5.314087438583374,
      "epoch": 2.2934678194044187,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044751367185364696,
      "loss": 4.6651,
      "mean_token_accuracy": 0.24992069453001023,
      "num_tokens": 54768886.0,
      "step": 23875
    },
    {
      "entropy": 5.4232447147369385,
      "epoch": 2.2939481268011526,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044749166074175634,
      "loss": 4.87,
      "mean_token_accuracy": 0.2338674858212471,
      "num_tokens": 54780243.0,
      "step": 23880
    },
    {
      "entropy": 5.3358580589294435,
      "epoch": 2.2944284341978864,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044746964562512154,
      "loss": 4.6592,
      "mean_token_accuracy": 0.24369479417800904,
      "num_tokens": 54792462.0,
      "step": 23885
    },
    {
      "entropy": 5.412147140502929,
      "epoch": 2.2949087415946208,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044744762650425376,
      "loss": 4.7787,
      "mean_token_accuracy": 0.23887651711702346,
      "num_tokens": 54804412.0,
      "step": 23890
    },
    {
      "entropy": 5.402648305892944,
      "epoch": 2.2953890489913547,
      "grad_norm": 1.125,
      "learning_rate": 0.00044742560337966415,
      "loss": 4.7412,
      "mean_token_accuracy": 0.25097568333148956,
      "num_tokens": 54814910.0,
      "step": 23895
    },
    {
      "entropy": 5.340834760665894,
      "epoch": 2.2958693563880885,
      "grad_norm": 1.015625,
      "learning_rate": 0.000447403576251864,
      "loss": 4.7258,
      "mean_token_accuracy": 0.24566648602485658,
      "num_tokens": 54826500.0,
      "step": 23900
    },
    {
      "entropy": 5.364181709289551,
      "epoch": 2.2963496637848224,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004473815451213648,
      "loss": 4.7543,
      "mean_token_accuracy": 0.23718868792057038,
      "num_tokens": 54837682.0,
      "step": 23905
    },
    {
      "entropy": 5.341012954711914,
      "epoch": 2.2968299711815563,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004473595099886779,
      "loss": 4.6929,
      "mean_token_accuracy": 0.24641891568899155,
      "num_tokens": 54848937.0,
      "step": 23910
    },
    {
      "entropy": 5.406399869918824,
      "epoch": 2.29731027857829,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004473374708543151,
      "loss": 4.6633,
      "mean_token_accuracy": 0.2553804650902748,
      "num_tokens": 54859384.0,
      "step": 23915
    },
    {
      "entropy": 5.347112607955933,
      "epoch": 2.297790585975024,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004473154277187878,
      "loss": 4.6849,
      "mean_token_accuracy": 0.2510125756263733,
      "num_tokens": 54871280.0,
      "step": 23920
    },
    {
      "entropy": 5.278886032104492,
      "epoch": 2.298270893371758,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044729338058260805,
      "loss": 4.6743,
      "mean_token_accuracy": 0.24774739742279053,
      "num_tokens": 54882699.0,
      "step": 23925
    },
    {
      "entropy": 5.40600004196167,
      "epoch": 2.298751200768492,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004472713294462876,
      "loss": 4.7996,
      "mean_token_accuracy": 0.24133175164461135,
      "num_tokens": 54894041.0,
      "step": 23930
    },
    {
      "entropy": 5.44746732711792,
      "epoch": 2.2992315081652257,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044724927431033843,
      "loss": 4.7846,
      "mean_token_accuracy": 0.23625168800354004,
      "num_tokens": 54905007.0,
      "step": 23935
    },
    {
      "entropy": 5.43376612663269,
      "epoch": 2.2997118155619596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004472272151752725,
      "loss": 4.7874,
      "mean_token_accuracy": 0.24219250679016113,
      "num_tokens": 54915735.0,
      "step": 23940
    },
    {
      "entropy": 5.288920497894287,
      "epoch": 2.3001921229586935,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004472051520416022,
      "loss": 4.6237,
      "mean_token_accuracy": 0.24623878449201583,
      "num_tokens": 54926350.0,
      "step": 23945
    },
    {
      "entropy": 5.366663408279419,
      "epoch": 2.3006724303554273,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004471830849098395,
      "loss": 4.7839,
      "mean_token_accuracy": 0.23704888969659804,
      "num_tokens": 54938354.0,
      "step": 23950
    },
    {
      "entropy": 5.38896164894104,
      "epoch": 2.3011527377521612,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044716101378049683,
      "loss": 4.7685,
      "mean_token_accuracy": 0.23570197075605392,
      "num_tokens": 54949440.0,
      "step": 23955
    },
    {
      "entropy": 5.372972297668457,
      "epoch": 2.301633045148895,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044713893865408667,
      "loss": 4.7137,
      "mean_token_accuracy": 0.24763110876083375,
      "num_tokens": 54961751.0,
      "step": 23960
    },
    {
      "entropy": 5.430323219299316,
      "epoch": 2.3021133525456294,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004471168595311215,
      "loss": 4.8977,
      "mean_token_accuracy": 0.22562142759561538,
      "num_tokens": 54975415.0,
      "step": 23965
    },
    {
      "entropy": 5.420666933059692,
      "epoch": 2.302593659942363,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044709477641211395,
      "loss": 4.723,
      "mean_token_accuracy": 0.24551412910223008,
      "num_tokens": 54986599.0,
      "step": 23970
    },
    {
      "entropy": 5.371295070648193,
      "epoch": 2.303073967339097,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004470726892975767,
      "loss": 4.7186,
      "mean_token_accuracy": 0.2457256332039833,
      "num_tokens": 54997428.0,
      "step": 23975
    },
    {
      "entropy": 5.374204111099243,
      "epoch": 2.303554274735831,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044705059818802255,
      "loss": 4.8126,
      "mean_token_accuracy": 0.2337118938565254,
      "num_tokens": 55008434.0,
      "step": 23980
    },
    {
      "entropy": 5.311865949630738,
      "epoch": 2.304034582132565,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004470285030839644,
      "loss": 4.7281,
      "mean_token_accuracy": 0.24439775496721267,
      "num_tokens": 55019834.0,
      "step": 23985
    },
    {
      "entropy": 5.366216993331909,
      "epoch": 2.304514889529299,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044700640398591526,
      "loss": 4.7188,
      "mean_token_accuracy": 0.24998879134655,
      "num_tokens": 55030432.0,
      "step": 23990
    },
    {
      "entropy": 5.470173740386963,
      "epoch": 2.3049951969260327,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004469843008943881,
      "loss": 4.8107,
      "mean_token_accuracy": 0.23871708810329437,
      "num_tokens": 55041482.0,
      "step": 23995
    },
    {
      "entropy": 5.378698205947876,
      "epoch": 2.3054755043227666,
      "grad_norm": 1.0,
      "learning_rate": 0.0004469621938098962,
      "loss": 4.664,
      "mean_token_accuracy": 0.2528966799378395,
      "num_tokens": 55051894.0,
      "step": 24000
    },
    {
      "epoch": 2.3054755043227666,
      "eval_entropy": 5.142419308816323,
      "eval_loss": 4.8432159423828125,
      "eval_mean_token_accuracy": 0.24652676420083777,
      "eval_num_tokens": 55051894.0,
      "eval_runtime": 20.6089,
      "eval_samples_per_second": 1592.276,
      "eval_steps_per_second": 199.041,
      "step": 24000
    },
    {
      "entropy": 5.405510759353637,
      "epoch": 2.3059558117195005,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004469400827329528,
      "loss": 4.7659,
      "mean_token_accuracy": 0.24693724513053894,
      "num_tokens": 55063907.0,
      "step": 24005
    },
    {
      "entropy": 5.365446329116821,
      "epoch": 2.3064361191162344,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004469179676640712,
      "loss": 4.7455,
      "mean_token_accuracy": 0.24159053415060044,
      "num_tokens": 55076004.0,
      "step": 24010
    },
    {
      "entropy": 5.315970134735108,
      "epoch": 2.3069164265129682,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004468958486037649,
      "loss": 4.654,
      "mean_token_accuracy": 0.2559589847922325,
      "num_tokens": 55087506.0,
      "step": 24015
    },
    {
      "entropy": 5.357095384597779,
      "epoch": 2.307396733909702,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004468737255525474,
      "loss": 4.7466,
      "mean_token_accuracy": 0.23781119883060456,
      "num_tokens": 55099406.0,
      "step": 24020
    },
    {
      "entropy": 5.488614273071289,
      "epoch": 2.307877041306436,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004468515985109324,
      "loss": 4.8299,
      "mean_token_accuracy": 0.22946524173021315,
      "num_tokens": 55111733.0,
      "step": 24025
    },
    {
      "entropy": 5.401976776123047,
      "epoch": 2.30835734870317,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004468294674794335,
      "loss": 4.7339,
      "mean_token_accuracy": 0.24159752279520036,
      "num_tokens": 55123455.0,
      "step": 24030
    },
    {
      "entropy": 5.408126640319824,
      "epoch": 2.3088376560999038,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004468073324585647,
      "loss": 4.7698,
      "mean_token_accuracy": 0.2403123989701271,
      "num_tokens": 55134284.0,
      "step": 24035
    },
    {
      "entropy": 5.375697803497315,
      "epoch": 2.309317963496638,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004467851934488398,
      "loss": 4.7375,
      "mean_token_accuracy": 0.2478972926735878,
      "num_tokens": 55147224.0,
      "step": 24040
    },
    {
      "entropy": 5.3420837879180905,
      "epoch": 2.3097982708933715,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004467630504507727,
      "loss": 4.6995,
      "mean_token_accuracy": 0.24112358540296555,
      "num_tokens": 55158144.0,
      "step": 24045
    },
    {
      "entropy": 5.356317472457886,
      "epoch": 2.310278578290106,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004467409034648776,
      "loss": 4.8162,
      "mean_token_accuracy": 0.24342479556798935,
      "num_tokens": 55171522.0,
      "step": 24050
    },
    {
      "entropy": 5.386338520050049,
      "epoch": 2.3107588856868397,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004467187524916688,
      "loss": 4.7264,
      "mean_token_accuracy": 0.24797234833240508,
      "num_tokens": 55182097.0,
      "step": 24055
    },
    {
      "entropy": 5.4180131435394285,
      "epoch": 2.3112391930835736,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004466965975316604,
      "loss": 4.7883,
      "mean_token_accuracy": 0.2370381996035576,
      "num_tokens": 55192287.0,
      "step": 24060
    },
    {
      "entropy": 5.384656381607056,
      "epoch": 2.3117195004803075,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044667443858536685,
      "loss": 4.793,
      "mean_token_accuracy": 0.24547805190086364,
      "num_tokens": 55204049.0,
      "step": 24065
    },
    {
      "entropy": 5.300537633895874,
      "epoch": 2.3121998078770414,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004466522756533026,
      "loss": 4.6203,
      "mean_token_accuracy": 0.2517087772488594,
      "num_tokens": 55214969.0,
      "step": 24070
    },
    {
      "entropy": 5.436120223999024,
      "epoch": 2.3126801152737753,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004466301087359822,
      "loss": 4.8196,
      "mean_token_accuracy": 0.24630804657936095,
      "num_tokens": 55226362.0,
      "step": 24075
    },
    {
      "entropy": 5.432807588577271,
      "epoch": 2.313160422670509,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044660793783392035,
      "loss": 4.8074,
      "mean_token_accuracy": 0.24228545278310776,
      "num_tokens": 55237345.0,
      "step": 24080
    },
    {
      "entropy": 5.421674585342407,
      "epoch": 2.313640730067243,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004465857629476317,
      "loss": 4.8446,
      "mean_token_accuracy": 0.24017298370599746,
      "num_tokens": 55248426.0,
      "step": 24085
    },
    {
      "entropy": 5.35838623046875,
      "epoch": 2.314121037463977,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004465635840776312,
      "loss": 4.7899,
      "mean_token_accuracy": 0.2415303111076355,
      "num_tokens": 55259017.0,
      "step": 24090
    },
    {
      "entropy": 5.433893728256225,
      "epoch": 2.314601344860711,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044654140122443373,
      "loss": 4.8094,
      "mean_token_accuracy": 0.2426323264837265,
      "num_tokens": 55270074.0,
      "step": 24095
    },
    {
      "entropy": 5.360273694992065,
      "epoch": 2.3150816522574447,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004465192143885542,
      "loss": 4.7393,
      "mean_token_accuracy": 0.24461957663297654,
      "num_tokens": 55282291.0,
      "step": 24100
    },
    {
      "entropy": 5.441861248016357,
      "epoch": 2.3155619596541785,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044649702357050787,
      "loss": 4.8771,
      "mean_token_accuracy": 0.2427952527999878,
      "num_tokens": 55293113.0,
      "step": 24105
    },
    {
      "entropy": 5.409810304641724,
      "epoch": 2.3160422670509124,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004464748287708099,
      "loss": 4.7791,
      "mean_token_accuracy": 0.24135114401578903,
      "num_tokens": 55304182.0,
      "step": 24110
    },
    {
      "entropy": 5.38310489654541,
      "epoch": 2.3165225744476463,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044645262998997557,
      "loss": 4.7158,
      "mean_token_accuracy": 0.24660283625125884,
      "num_tokens": 55315292.0,
      "step": 24115
    },
    {
      "entropy": 5.348039150238037,
      "epoch": 2.31700288184438,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044643042722852024,
      "loss": 4.7186,
      "mean_token_accuracy": 0.24397215694189073,
      "num_tokens": 55327000.0,
      "step": 24120
    },
    {
      "entropy": 5.33282790184021,
      "epoch": 2.3174831892411145,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004464082204869595,
      "loss": 4.6875,
      "mean_token_accuracy": 0.24936299324035643,
      "num_tokens": 55337506.0,
      "step": 24125
    },
    {
      "entropy": 5.31218991279602,
      "epoch": 2.3179634966378484,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004463860097658088,
      "loss": 4.641,
      "mean_token_accuracy": 0.25338819473981855,
      "num_tokens": 55348538.0,
      "step": 24130
    },
    {
      "entropy": 5.43988265991211,
      "epoch": 2.3184438040345823,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004463637950655839,
      "loss": 4.793,
      "mean_token_accuracy": 0.23874622583389282,
      "num_tokens": 55360501.0,
      "step": 24135
    },
    {
      "entropy": 5.368517160415649,
      "epoch": 2.318924111431316,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044634157638680054,
      "loss": 4.7446,
      "mean_token_accuracy": 0.24226535409688948,
      "num_tokens": 55371616.0,
      "step": 24140
    },
    {
      "entropy": 5.391592073440552,
      "epoch": 2.31940441882805,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044631935372997455,
      "loss": 4.8266,
      "mean_token_accuracy": 0.2380247965455055,
      "num_tokens": 55382702.0,
      "step": 24145
    },
    {
      "entropy": 5.384558439254761,
      "epoch": 2.319884726224784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004462971270956219,
      "loss": 4.7203,
      "mean_token_accuracy": 0.24133380204439164,
      "num_tokens": 55394826.0,
      "step": 24150
    },
    {
      "entropy": 5.341002607345581,
      "epoch": 2.320365033621518,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004462748964842586,
      "loss": 4.6755,
      "mean_token_accuracy": 0.2551581904292107,
      "num_tokens": 55406584.0,
      "step": 24155
    },
    {
      "entropy": 5.3673559665679935,
      "epoch": 2.3208453410182517,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004462526618964008,
      "loss": 4.7934,
      "mean_token_accuracy": 0.23824091404676437,
      "num_tokens": 55419640.0,
      "step": 24160
    },
    {
      "entropy": 5.337840032577515,
      "epoch": 2.3213256484149856,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004462304233325647,
      "loss": 4.7409,
      "mean_token_accuracy": 0.2470174953341484,
      "num_tokens": 55430197.0,
      "step": 24165
    },
    {
      "entropy": 5.382766056060791,
      "epoch": 2.3218059558117194,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004462081807932666,
      "loss": 4.7761,
      "mean_token_accuracy": 0.2415039986371994,
      "num_tokens": 55441832.0,
      "step": 24170
    },
    {
      "entropy": 5.339666938781738,
      "epoch": 2.3222862632084533,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004461859342790229,
      "loss": 4.6729,
      "mean_token_accuracy": 0.25101673007011416,
      "num_tokens": 55452656.0,
      "step": 24175
    },
    {
      "entropy": 5.274402523040772,
      "epoch": 2.322766570605187,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004461636837903501,
      "loss": 4.6137,
      "mean_token_accuracy": 0.2533323511481285,
      "num_tokens": 55463936.0,
      "step": 24180
    },
    {
      "entropy": 5.415198802947998,
      "epoch": 2.323246878001921,
      "grad_norm": 1.125,
      "learning_rate": 0.0004461414293277649,
      "loss": 4.8072,
      "mean_token_accuracy": 0.2453521877527237,
      "num_tokens": 55475885.0,
      "step": 24185
    },
    {
      "entropy": 5.360448837280273,
      "epoch": 2.323727185398655,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004461191708917839,
      "loss": 4.7139,
      "mean_token_accuracy": 0.23911530822515487,
      "num_tokens": 55489022.0,
      "step": 24190
    },
    {
      "entropy": 5.3602581977844235,
      "epoch": 2.324207492795389,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044609690848292376,
      "loss": 4.715,
      "mean_token_accuracy": 0.24276641458272935,
      "num_tokens": 55500008.0,
      "step": 24195
    },
    {
      "entropy": 5.378081464767456,
      "epoch": 2.324687800192123,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004460746421017016,
      "loss": 4.8607,
      "mean_token_accuracy": 0.22956019043922424,
      "num_tokens": 55511024.0,
      "step": 24200
    },
    {
      "entropy": 5.42891674041748,
      "epoch": 2.325168107588857,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044605237174863405,
      "loss": 4.7689,
      "mean_token_accuracy": 0.2468690648674965,
      "num_tokens": 55522199.0,
      "step": 24205
    },
    {
      "entropy": 5.374897909164429,
      "epoch": 2.325648414985591,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004460300974242386,
      "loss": 4.7051,
      "mean_token_accuracy": 0.24730193018913268,
      "num_tokens": 55533197.0,
      "step": 24210
    },
    {
      "entropy": 5.344883584976197,
      "epoch": 2.326128722382325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004460078191290319,
      "loss": 4.7509,
      "mean_token_accuracy": 0.2441083997488022,
      "num_tokens": 55544317.0,
      "step": 24215
    },
    {
      "entropy": 5.380463171005249,
      "epoch": 2.3266090297790587,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044598553686353153,
      "loss": 4.785,
      "mean_token_accuracy": 0.2415432870388031,
      "num_tokens": 55555388.0,
      "step": 24220
    },
    {
      "entropy": 5.396922636032104,
      "epoch": 2.3270893371757926,
      "grad_norm": 1.125,
      "learning_rate": 0.00044596325062825476,
      "loss": 4.7291,
      "mean_token_accuracy": 0.24088791757822037,
      "num_tokens": 55566189.0,
      "step": 24225
    },
    {
      "entropy": 5.3566121578216555,
      "epoch": 2.3275696445725265,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004459409604237189,
      "loss": 4.7071,
      "mean_token_accuracy": 0.2531862512230873,
      "num_tokens": 55578235.0,
      "step": 24230
    },
    {
      "entropy": 5.262808179855346,
      "epoch": 2.3280499519692603,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044591866625044154,
      "loss": 4.6873,
      "mean_token_accuracy": 0.24664416462182998,
      "num_tokens": 55590300.0,
      "step": 24235
    },
    {
      "entropy": 5.419283819198609,
      "epoch": 2.3285302593659942,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004458963681089403,
      "loss": 4.7646,
      "mean_token_accuracy": 0.24560531228780746,
      "num_tokens": 55601017.0,
      "step": 24240
    },
    {
      "entropy": 5.449910449981689,
      "epoch": 2.329010566762728,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004458740659997328,
      "loss": 4.7662,
      "mean_token_accuracy": 0.24293055683374404,
      "num_tokens": 55612411.0,
      "step": 24245
    },
    {
      "entropy": 5.41005220413208,
      "epoch": 2.329490874159462,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004458517599233369,
      "loss": 4.8121,
      "mean_token_accuracy": 0.2436547189950943,
      "num_tokens": 55623872.0,
      "step": 24250
    },
    {
      "entropy": 5.460035085678101,
      "epoch": 2.329971181556196,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004458294498802706,
      "loss": 4.8104,
      "mean_token_accuracy": 0.23224240094423293,
      "num_tokens": 55634887.0,
      "step": 24255
    },
    {
      "entropy": 5.377669811248779,
      "epoch": 2.3304514889529298,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004458071358710516,
      "loss": 4.7553,
      "mean_token_accuracy": 0.2428038537502289,
      "num_tokens": 55646286.0,
      "step": 24260
    },
    {
      "entropy": 5.358595705032348,
      "epoch": 2.3309317963496636,
      "grad_norm": 1.0,
      "learning_rate": 0.0004457848178961981,
      "loss": 4.7655,
      "mean_token_accuracy": 0.24074459969997405,
      "num_tokens": 55657647.0,
      "step": 24265
    },
    {
      "entropy": 5.374101972579956,
      "epoch": 2.3314121037463975,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044576249595622833,
      "loss": 4.745,
      "mean_token_accuracy": 0.24135944247245789,
      "num_tokens": 55669000.0,
      "step": 24270
    },
    {
      "entropy": 5.35329647064209,
      "epoch": 2.331892411143132,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004457401700516603,
      "loss": 4.7413,
      "mean_token_accuracy": 0.23942579627037047,
      "num_tokens": 55680678.0,
      "step": 24275
    },
    {
      "entropy": 5.399015045166015,
      "epoch": 2.3323727185398653,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044571784018301267,
      "loss": 4.7447,
      "mean_token_accuracy": 0.2424183025956154,
      "num_tokens": 55692429.0,
      "step": 24280
    },
    {
      "entropy": 5.409781408309937,
      "epoch": 2.3328530259365996,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044569550635080365,
      "loss": 4.8255,
      "mean_token_accuracy": 0.23680232763290404,
      "num_tokens": 55704659.0,
      "step": 24285
    },
    {
      "entropy": 5.322394514083863,
      "epoch": 2.3333333333333335,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044567316855555184,
      "loss": 4.6371,
      "mean_token_accuracy": 0.25267155915498735,
      "num_tokens": 55715584.0,
      "step": 24290
    },
    {
      "entropy": 5.457093238830566,
      "epoch": 2.3338136407300674,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004456508267977759,
      "loss": 4.8432,
      "mean_token_accuracy": 0.241513592004776,
      "num_tokens": 55726715.0,
      "step": 24295
    },
    {
      "entropy": 5.335708951950073,
      "epoch": 2.3342939481268012,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044562848107799444,
      "loss": 4.7099,
      "mean_token_accuracy": 0.24930669665336608,
      "num_tokens": 55737069.0,
      "step": 24300
    },
    {
      "entropy": 5.346783638000488,
      "epoch": 2.334774255523535,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044560613139672627,
      "loss": 4.6963,
      "mean_token_accuracy": 0.25421649515628814,
      "num_tokens": 55747774.0,
      "step": 24305
    },
    {
      "entropy": 5.395544958114624,
      "epoch": 2.335254562920269,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044558377775449036,
      "loss": 4.7449,
      "mean_token_accuracy": 0.24664495140314102,
      "num_tokens": 55758209.0,
      "step": 24310
    },
    {
      "entropy": 5.46163420677185,
      "epoch": 2.335734870317003,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044556142015180573,
      "loss": 4.7736,
      "mean_token_accuracy": 0.23464470952749253,
      "num_tokens": 55768668.0,
      "step": 24315
    },
    {
      "entropy": 5.53148889541626,
      "epoch": 2.3362151777137368,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044553905858919134,
      "loss": 4.8685,
      "mean_token_accuracy": 0.23433106541633605,
      "num_tokens": 55780254.0,
      "step": 24320
    },
    {
      "entropy": 5.41887149810791,
      "epoch": 2.3366954851104706,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004455166930671664,
      "loss": 4.7747,
      "mean_token_accuracy": 0.24235983937978745,
      "num_tokens": 55790241.0,
      "step": 24325
    },
    {
      "entropy": 5.452752923965454,
      "epoch": 2.3371757925072045,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044549432358625014,
      "loss": 4.8501,
      "mean_token_accuracy": 0.23614850342273713,
      "num_tokens": 55802040.0,
      "step": 24330
    },
    {
      "entropy": 5.385617303848266,
      "epoch": 2.3376560999039384,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000445471950146962,
      "loss": 4.7472,
      "mean_token_accuracy": 0.2435455933213234,
      "num_tokens": 55813926.0,
      "step": 24335
    },
    {
      "entropy": 5.365787696838379,
      "epoch": 2.3381364073006723,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004454495727498214,
      "loss": 4.7442,
      "mean_token_accuracy": 0.2459506869316101,
      "num_tokens": 55825281.0,
      "step": 24340
    },
    {
      "entropy": 5.360678482055664,
      "epoch": 2.338616714697406,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004454271913953478,
      "loss": 4.7136,
      "mean_token_accuracy": 0.24871269166469573,
      "num_tokens": 55836295.0,
      "step": 24345
    },
    {
      "entropy": 5.326148271560669,
      "epoch": 2.3390970220941405,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044540480608406093,
      "loss": 4.7218,
      "mean_token_accuracy": 0.25169241428375244,
      "num_tokens": 55847150.0,
      "step": 24350
    },
    {
      "entropy": 5.3880736351013185,
      "epoch": 2.339577329490874,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004453824168164804,
      "loss": 4.814,
      "mean_token_accuracy": 0.2398724615573883,
      "num_tokens": 55858135.0,
      "step": 24355
    },
    {
      "entropy": 5.476385641098022,
      "epoch": 2.3400576368876083,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004453600235931261,
      "loss": 4.8389,
      "mean_token_accuracy": 0.23863284140825272,
      "num_tokens": 55869319.0,
      "step": 24360
    },
    {
      "entropy": 5.412109375,
      "epoch": 2.340537944284342,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000445337626414518,
      "loss": 4.8129,
      "mean_token_accuracy": 0.24201759099960327,
      "num_tokens": 55881383.0,
      "step": 24365
    },
    {
      "entropy": 5.332005834579467,
      "epoch": 2.341018251681076,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044531522528117593,
      "loss": 4.7303,
      "mean_token_accuracy": 0.2432515174150467,
      "num_tokens": 55893726.0,
      "step": 24370
    },
    {
      "entropy": 5.385066366195678,
      "epoch": 2.34149855907781,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044529282019362007,
      "loss": 4.7794,
      "mean_token_accuracy": 0.24230920225381852,
      "num_tokens": 55905260.0,
      "step": 24375
    },
    {
      "entropy": 5.433302021026611,
      "epoch": 2.341978866474544,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044527041115237056,
      "loss": 4.8037,
      "mean_token_accuracy": 0.24223033338785172,
      "num_tokens": 55915404.0,
      "step": 24380
    },
    {
      "entropy": 5.3678652286529545,
      "epoch": 2.3424591738712777,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004452479981579477,
      "loss": 4.7083,
      "mean_token_accuracy": 0.24183041751384735,
      "num_tokens": 55925972.0,
      "step": 24385
    },
    {
      "entropy": 5.365244770050049,
      "epoch": 2.3429394812680115,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004452255812108719,
      "loss": 4.7295,
      "mean_token_accuracy": 0.23950830101966858,
      "num_tokens": 55937545.0,
      "step": 24390
    },
    {
      "entropy": 5.327963352203369,
      "epoch": 2.3434197886647454,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004452031603116636,
      "loss": 4.6816,
      "mean_token_accuracy": 0.254170224070549,
      "num_tokens": 55949934.0,
      "step": 24395
    },
    {
      "entropy": 5.380492305755615,
      "epoch": 2.3439000960614793,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044518073546084326,
      "loss": 4.7905,
      "mean_token_accuracy": 0.2501857474446297,
      "num_tokens": 55960414.0,
      "step": 24400
    },
    {
      "entropy": 5.338408041000366,
      "epoch": 2.344380403458213,
      "grad_norm": 1.125,
      "learning_rate": 0.0004451583066589316,
      "loss": 4.7219,
      "mean_token_accuracy": 0.24372323900461196,
      "num_tokens": 55970999.0,
      "step": 24405
    },
    {
      "entropy": 5.390763950347901,
      "epoch": 2.344860710854947,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044513587390644925,
      "loss": 4.7404,
      "mean_token_accuracy": 0.245574688911438,
      "num_tokens": 55982801.0,
      "step": 24410
    },
    {
      "entropy": 5.441425800323486,
      "epoch": 2.345341018251681,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044511343720391724,
      "loss": 4.7815,
      "mean_token_accuracy": 0.24260494261980056,
      "num_tokens": 55995143.0,
      "step": 24415
    },
    {
      "entropy": 5.348770189285278,
      "epoch": 2.345821325648415,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004450909965518563,
      "loss": 4.7801,
      "mean_token_accuracy": 0.2403816670179367,
      "num_tokens": 56006368.0,
      "step": 24420
    },
    {
      "entropy": 5.415327644348144,
      "epoch": 2.3463016330451487,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044506855195078755,
      "loss": 4.8306,
      "mean_token_accuracy": 0.24338632076978683,
      "num_tokens": 56019245.0,
      "step": 24425
    },
    {
      "entropy": 5.43805341720581,
      "epoch": 2.3467819404418826,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044504610340123185,
      "loss": 4.8203,
      "mean_token_accuracy": 0.23409497141838073,
      "num_tokens": 56030982.0,
      "step": 24430
    },
    {
      "entropy": 5.377782773971558,
      "epoch": 2.347262247838617,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044502365090371066,
      "loss": 4.7801,
      "mean_token_accuracy": 0.24158954471349717,
      "num_tokens": 56042871.0,
      "step": 24435
    },
    {
      "entropy": 5.446085739135742,
      "epoch": 2.347742555235351,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004450011944587452,
      "loss": 4.8039,
      "mean_token_accuracy": 0.23252429515123368,
      "num_tokens": 56054099.0,
      "step": 24440
    },
    {
      "entropy": 5.3344316482543945,
      "epoch": 2.3482228626320847,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044497873406685673,
      "loss": 4.726,
      "mean_token_accuracy": 0.2443087413907051,
      "num_tokens": 56064315.0,
      "step": 24445
    },
    {
      "entropy": 5.311272048950196,
      "epoch": 2.3487031700288186,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004449562697285669,
      "loss": 4.7281,
      "mean_token_accuracy": 0.24245982319116594,
      "num_tokens": 56076446.0,
      "step": 24450
    },
    {
      "entropy": 5.473170328140259,
      "epoch": 2.3491834774255524,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044493380144439707,
      "loss": 4.8266,
      "mean_token_accuracy": 0.23882719725370408,
      "num_tokens": 56089356.0,
      "step": 24455
    },
    {
      "entropy": 5.40914363861084,
      "epoch": 2.3496637848222863,
      "grad_norm": 1.140625,
      "learning_rate": 0.000444911329214869,
      "loss": 4.8354,
      "mean_token_accuracy": 0.23410024791955947,
      "num_tokens": 56101245.0,
      "step": 24460
    },
    {
      "entropy": 5.265419960021973,
      "epoch": 2.35014409221902,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044488885304050434,
      "loss": 4.6456,
      "mean_token_accuracy": 0.25217376053333285,
      "num_tokens": 56112756.0,
      "step": 24465
    },
    {
      "entropy": 5.401223373413086,
      "epoch": 2.350624399615754,
      "grad_norm": 1.046875,
      "learning_rate": 0.000444866372921825,
      "loss": 4.7822,
      "mean_token_accuracy": 0.2440278246998787,
      "num_tokens": 56124316.0,
      "step": 24470
    },
    {
      "entropy": 5.326864862442017,
      "epoch": 2.351104707012488,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044484388885935287,
      "loss": 4.7021,
      "mean_token_accuracy": 0.2539144232869148,
      "num_tokens": 56135972.0,
      "step": 24475
    },
    {
      "entropy": 5.410159635543823,
      "epoch": 2.351585014409222,
      "grad_norm": 1.125,
      "learning_rate": 0.00044482140085361005,
      "loss": 4.8001,
      "mean_token_accuracy": 0.23848803341388702,
      "num_tokens": 56147335.0,
      "step": 24480
    },
    {
      "entropy": 5.371070814132691,
      "epoch": 2.3520653218059557,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044479890890511853,
      "loss": 4.6773,
      "mean_token_accuracy": 0.25424722880125045,
      "num_tokens": 56158294.0,
      "step": 24485
    },
    {
      "entropy": 5.338474702835083,
      "epoch": 2.3525456292026896,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044477641301440054,
      "loss": 4.6211,
      "mean_token_accuracy": 0.25962951481342317,
      "num_tokens": 56169468.0,
      "step": 24490
    },
    {
      "entropy": 5.382090473175049,
      "epoch": 2.3530259365994235,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004447539131819784,
      "loss": 4.7018,
      "mean_token_accuracy": 0.24220379292964936,
      "num_tokens": 56180112.0,
      "step": 24495
    },
    {
      "entropy": 5.423308038711548,
      "epoch": 2.3535062439961574,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044473140940837436,
      "loss": 4.7378,
      "mean_token_accuracy": 0.2505744680762291,
      "num_tokens": 56191290.0,
      "step": 24500
    },
    {
      "entropy": 5.421861696243286,
      "epoch": 2.3539865513928913,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044470890169411107,
      "loss": 4.8367,
      "mean_token_accuracy": 0.23475950360298156,
      "num_tokens": 56202447.0,
      "step": 24505
    },
    {
      "entropy": 5.43946099281311,
      "epoch": 2.3544668587896256,
      "grad_norm": 1.109375,
      "learning_rate": 0.000444686390039711,
      "loss": 4.8097,
      "mean_token_accuracy": 0.2330123171210289,
      "num_tokens": 56213802.0,
      "step": 24510
    },
    {
      "entropy": 5.362165355682373,
      "epoch": 2.3549471661863595,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004446638744456968,
      "loss": 4.7436,
      "mean_token_accuracy": 0.23978333920240402,
      "num_tokens": 56225243.0,
      "step": 24515
    },
    {
      "entropy": 5.418326807022095,
      "epoch": 2.3554274735830933,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044464135491259135,
      "loss": 4.6883,
      "mean_token_accuracy": 0.24532321989536285,
      "num_tokens": 56235116.0,
      "step": 24520
    },
    {
      "entropy": 5.35143461227417,
      "epoch": 2.3559077809798272,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004446188314409172,
      "loss": 4.7613,
      "mean_token_accuracy": 0.23596743047237395,
      "num_tokens": 56246654.0,
      "step": 24525
    },
    {
      "entropy": 5.403826332092285,
      "epoch": 2.356388088376561,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004445963040311975,
      "loss": 4.753,
      "mean_token_accuracy": 0.24527794867753983,
      "num_tokens": 56257787.0,
      "step": 24530
    },
    {
      "entropy": 5.291752862930298,
      "epoch": 2.356868395773295,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044457377268395526,
      "loss": 4.6555,
      "mean_token_accuracy": 0.2462114468216896,
      "num_tokens": 56268881.0,
      "step": 24535
    },
    {
      "entropy": 5.261334323883057,
      "epoch": 2.357348703170029,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044455123739971355,
      "loss": 4.6646,
      "mean_token_accuracy": 0.24906189888715743,
      "num_tokens": 56281515.0,
      "step": 24540
    },
    {
      "entropy": 5.297220039367676,
      "epoch": 2.3578290105667628,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044452869817899554,
      "loss": 4.6515,
      "mean_token_accuracy": 0.25063349306583405,
      "num_tokens": 56293226.0,
      "step": 24545
    },
    {
      "entropy": 5.384673118591309,
      "epoch": 2.3583093179634966,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004445061550223246,
      "loss": 4.7519,
      "mean_token_accuracy": 0.2440799966454506,
      "num_tokens": 56303408.0,
      "step": 24550
    },
    {
      "entropy": 5.381060791015625,
      "epoch": 2.3587896253602305,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044448360793022403,
      "loss": 4.742,
      "mean_token_accuracy": 0.23596848100423812,
      "num_tokens": 56315379.0,
      "step": 24555
    },
    {
      "entropy": 5.396325397491455,
      "epoch": 2.3592699327569644,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004444610569032174,
      "loss": 4.7408,
      "mean_token_accuracy": 0.24325896352529525,
      "num_tokens": 56326496.0,
      "step": 24560
    },
    {
      "entropy": 5.298276090621949,
      "epoch": 2.3597502401536983,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004444385019418281,
      "loss": 4.7011,
      "mean_token_accuracy": 0.2442005157470703,
      "num_tokens": 56339416.0,
      "step": 24565
    },
    {
      "entropy": 5.39544448852539,
      "epoch": 2.360230547550432,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044441594304658004,
      "loss": 4.771,
      "mean_token_accuracy": 0.2399225115776062,
      "num_tokens": 56352305.0,
      "step": 24570
    },
    {
      "entropy": 5.372354173660279,
      "epoch": 2.360710854947166,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004443933802179968,
      "loss": 4.6844,
      "mean_token_accuracy": 0.24682267755270004,
      "num_tokens": 56364598.0,
      "step": 24575
    },
    {
      "entropy": 5.446901035308838,
      "epoch": 2.3611911623439,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044437081345660224,
      "loss": 4.8408,
      "mean_token_accuracy": 0.229458686709404,
      "num_tokens": 56376651.0,
      "step": 24580
    },
    {
      "entropy": 5.311614990234375,
      "epoch": 2.3616714697406342,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004443482427629204,
      "loss": 4.714,
      "mean_token_accuracy": 0.24857954531908036,
      "num_tokens": 56388038.0,
      "step": 24585
    },
    {
      "entropy": 5.375942373275757,
      "epoch": 2.3621517771373677,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004443256681374751,
      "loss": 4.7325,
      "mean_token_accuracy": 0.24614741355180741,
      "num_tokens": 56398889.0,
      "step": 24590
    },
    {
      "entropy": 5.352107000350952,
      "epoch": 2.362632084534102,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004443030895807907,
      "loss": 4.7793,
      "mean_token_accuracy": 0.24085396975278855,
      "num_tokens": 56411548.0,
      "step": 24595
    },
    {
      "entropy": 5.389865684509277,
      "epoch": 2.363112391930836,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044428050709339117,
      "loss": 4.6573,
      "mean_token_accuracy": 0.24860192239284515,
      "num_tokens": 56423292.0,
      "step": 24600
    },
    {
      "entropy": 5.341977739334107,
      "epoch": 2.3635926993275698,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000444257920675801,
      "loss": 4.6899,
      "mean_token_accuracy": 0.2508048310875893,
      "num_tokens": 56434834.0,
      "step": 24605
    },
    {
      "entropy": 5.328068685531616,
      "epoch": 2.3640730067243036,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044423533032854454,
      "loss": 4.7837,
      "mean_token_accuracy": 0.23819314539432526,
      "num_tokens": 56445554.0,
      "step": 24610
    },
    {
      "entropy": 5.345740509033203,
      "epoch": 2.3645533141210375,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004442127360521462,
      "loss": 4.7316,
      "mean_token_accuracy": 0.24981726109981536,
      "num_tokens": 56457748.0,
      "step": 24615
    },
    {
      "entropy": 5.319956588745117,
      "epoch": 2.3650336215177714,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004441901378471306,
      "loss": 4.6609,
      "mean_token_accuracy": 0.24875420778989793,
      "num_tokens": 56469650.0,
      "step": 24620
    },
    {
      "entropy": 5.337354850769043,
      "epoch": 2.3655139289145053,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044416753571402233,
      "loss": 4.6996,
      "mean_token_accuracy": 0.24815977215766907,
      "num_tokens": 56482061.0,
      "step": 24625
    },
    {
      "entropy": 5.395077800750732,
      "epoch": 2.365994236311239,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004441449296533462,
      "loss": 4.7994,
      "mean_token_accuracy": 0.23985088467597962,
      "num_tokens": 56493599.0,
      "step": 24630
    },
    {
      "entropy": 5.414036655426026,
      "epoch": 2.366474543707973,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00044412231966562717,
      "loss": 4.8297,
      "mean_token_accuracy": 0.23988510966300963,
      "num_tokens": 56505842.0,
      "step": 24635
    },
    {
      "entropy": 5.346997451782227,
      "epoch": 2.366954851104707,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044409970575139,
      "loss": 4.6978,
      "mean_token_accuracy": 0.2505576014518738,
      "num_tokens": 56517688.0,
      "step": 24640
    },
    {
      "entropy": 5.403812408447266,
      "epoch": 2.367435158501441,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004440770879111598,
      "loss": 4.7513,
      "mean_token_accuracy": 0.2473706066608429,
      "num_tokens": 56528665.0,
      "step": 24645
    },
    {
      "entropy": 5.252115964889526,
      "epoch": 2.3679154658981747,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044405446614546163,
      "loss": 4.6502,
      "mean_token_accuracy": 0.24611236602067948,
      "num_tokens": 56539097.0,
      "step": 24650
    },
    {
      "entropy": 5.417644739151001,
      "epoch": 2.3683957732949086,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004440318404548208,
      "loss": 4.8232,
      "mean_token_accuracy": 0.24274299740791322,
      "num_tokens": 56550373.0,
      "step": 24655
    },
    {
      "entropy": 5.376272058486938,
      "epoch": 2.3688760806916425,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044400921083976246,
      "loss": 4.7916,
      "mean_token_accuracy": 0.2359602451324463,
      "num_tokens": 56561289.0,
      "step": 24660
    },
    {
      "entropy": 5.3942617893219,
      "epoch": 2.3693563880883763,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004439865773008122,
      "loss": 4.7919,
      "mean_token_accuracy": 0.24262021481990814,
      "num_tokens": 56572331.0,
      "step": 24665
    },
    {
      "entropy": 5.3463828563690186,
      "epoch": 2.3698366954851107,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004439639398384953,
      "loss": 4.6879,
      "mean_token_accuracy": 0.2508407562971115,
      "num_tokens": 56584375.0,
      "step": 24670
    },
    {
      "entropy": 5.308524322509766,
      "epoch": 2.3703170028818445,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044394129845333756,
      "loss": 4.7088,
      "mean_token_accuracy": 0.2437274932861328,
      "num_tokens": 56595825.0,
      "step": 24675
    },
    {
      "entropy": 5.329288387298584,
      "epoch": 2.3707973102785784,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004439186531458645,
      "loss": 4.7774,
      "mean_token_accuracy": 0.2479529321193695,
      "num_tokens": 56607842.0,
      "step": 24680
    },
    {
      "entropy": 5.429071712493896,
      "epoch": 2.3712776176753123,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044389600391660185,
      "loss": 4.8031,
      "mean_token_accuracy": 0.24310345202684402,
      "num_tokens": 56618509.0,
      "step": 24685
    },
    {
      "entropy": 5.337828540802002,
      "epoch": 2.371757925072046,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044387335076607554,
      "loss": 4.7026,
      "mean_token_accuracy": 0.24605659395456314,
      "num_tokens": 56630293.0,
      "step": 24690
    },
    {
      "entropy": 5.442462205886841,
      "epoch": 2.37223823246878,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004438506936948115,
      "loss": 4.8567,
      "mean_token_accuracy": 0.23513938784599303,
      "num_tokens": 56641852.0,
      "step": 24695
    },
    {
      "entropy": 5.498385381698609,
      "epoch": 2.372718539865514,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044382803270333565,
      "loss": 4.7855,
      "mean_token_accuracy": 0.24585455358028413,
      "num_tokens": 56654409.0,
      "step": 24700
    },
    {
      "entropy": 5.406163358688355,
      "epoch": 2.373198847262248,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004438053677921743,
      "loss": 4.7457,
      "mean_token_accuracy": 0.24594021886587142,
      "num_tokens": 56665227.0,
      "step": 24705
    },
    {
      "entropy": 5.389076900482178,
      "epoch": 2.3736791546589817,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044378269896185344,
      "loss": 4.7812,
      "mean_token_accuracy": 0.24169406443834304,
      "num_tokens": 56677924.0,
      "step": 24710
    },
    {
      "entropy": 5.376308345794678,
      "epoch": 2.3741594620557156,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004437600262128996,
      "loss": 4.6871,
      "mean_token_accuracy": 0.2468394234776497,
      "num_tokens": 56689789.0,
      "step": 24715
    },
    {
      "entropy": 5.43754014968872,
      "epoch": 2.3746397694524495,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000443737349545839,
      "loss": 4.8359,
      "mean_token_accuracy": 0.23991654217243194,
      "num_tokens": 56700595.0,
      "step": 24720
    },
    {
      "entropy": 5.437364292144776,
      "epoch": 2.3751200768491834,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044371466896119823,
      "loss": 4.8497,
      "mean_token_accuracy": 0.2384314626455307,
      "num_tokens": 56712554.0,
      "step": 24725
    },
    {
      "entropy": 5.401896238327026,
      "epoch": 2.3756003842459172,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044369198445950384,
      "loss": 4.7396,
      "mean_token_accuracy": 0.24134009778499604,
      "num_tokens": 56723698.0,
      "step": 24730
    },
    {
      "entropy": 5.359768676757812,
      "epoch": 2.376080691642651,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004436692960412824,
      "loss": 4.7424,
      "mean_token_accuracy": 0.25375120490789416,
      "num_tokens": 56734660.0,
      "step": 24735
    },
    {
      "entropy": 5.357243299484253,
      "epoch": 2.376560999039385,
      "grad_norm": 1.125,
      "learning_rate": 0.0004436466037070608,
      "loss": 4.7665,
      "mean_token_accuracy": 0.2420510768890381,
      "num_tokens": 56746150.0,
      "step": 24740
    },
    {
      "entropy": 5.470491790771485,
      "epoch": 2.3770413064361193,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044362390745736585,
      "loss": 4.8671,
      "mean_token_accuracy": 0.2311130866408348,
      "num_tokens": 56758018.0,
      "step": 24745
    },
    {
      "entropy": 5.3474507331848145,
      "epoch": 2.377521613832853,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004436012072927245,
      "loss": 4.6847,
      "mean_token_accuracy": 0.2527194768190384,
      "num_tokens": 56769129.0,
      "step": 24750
    },
    {
      "entropy": 5.37448353767395,
      "epoch": 2.378001921229587,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044357850321366375,
      "loss": 4.758,
      "mean_token_accuracy": 0.24415764808654786,
      "num_tokens": 56780445.0,
      "step": 24755
    },
    {
      "entropy": 5.34340271949768,
      "epoch": 2.378482228626321,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004435557952207107,
      "loss": 4.7464,
      "mean_token_accuracy": 0.24567800760269165,
      "num_tokens": 56792338.0,
      "step": 24760
    },
    {
      "entropy": 5.392950344085693,
      "epoch": 2.378962536023055,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044353308331439257,
      "loss": 4.7727,
      "mean_token_accuracy": 0.2448149263858795,
      "num_tokens": 56802345.0,
      "step": 24765
    },
    {
      "entropy": 5.444286727905274,
      "epoch": 2.3794428434197887,
      "grad_norm": 1.25,
      "learning_rate": 0.0004435103674952367,
      "loss": 4.787,
      "mean_token_accuracy": 0.24694550633430482,
      "num_tokens": 56813865.0,
      "step": 24770
    },
    {
      "entropy": 5.343532991409302,
      "epoch": 2.3799231508165226,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044348764776377047,
      "loss": 4.6919,
      "mean_token_accuracy": 0.24489250779151917,
      "num_tokens": 56824135.0,
      "step": 24775
    },
    {
      "entropy": 5.342560577392578,
      "epoch": 2.3804034582132565,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004434649241205214,
      "loss": 4.7346,
      "mean_token_accuracy": 0.24225799292325972,
      "num_tokens": 56836104.0,
      "step": 24780
    },
    {
      "entropy": 5.368499040603638,
      "epoch": 2.3808837656099904,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044344219656601704,
      "loss": 4.7078,
      "mean_token_accuracy": 0.2502123937010765,
      "num_tokens": 56848308.0,
      "step": 24785
    },
    {
      "entropy": 5.361868524551392,
      "epoch": 2.3813640730067243,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000443419465100785,
      "loss": 4.7649,
      "mean_token_accuracy": 0.242922405898571,
      "num_tokens": 56859351.0,
      "step": 24790
    },
    {
      "entropy": 5.4045041561126705,
      "epoch": 2.381844380403458,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004433967297253531,
      "loss": 4.7864,
      "mean_token_accuracy": 0.24021644443273543,
      "num_tokens": 56870645.0,
      "step": 24795
    },
    {
      "entropy": 5.386020946502685,
      "epoch": 2.382324687800192,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044337399044024924,
      "loss": 4.7545,
      "mean_token_accuracy": 0.23957474231719972,
      "num_tokens": 56881246.0,
      "step": 24800
    },
    {
      "entropy": 5.344377136230468,
      "epoch": 2.382804995196926,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004433512472460012,
      "loss": 4.7125,
      "mean_token_accuracy": 0.2453825980424881,
      "num_tokens": 56893477.0,
      "step": 24805
    },
    {
      "entropy": 5.3220109939575195,
      "epoch": 2.38328530259366,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044332850014313713,
      "loss": 4.7491,
      "mean_token_accuracy": 0.24275881350040435,
      "num_tokens": 56904813.0,
      "step": 24810
    },
    {
      "entropy": 5.343864297866821,
      "epoch": 2.3837656099903937,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004433057491321851,
      "loss": 4.7013,
      "mean_token_accuracy": 0.24222712367773055,
      "num_tokens": 56916160.0,
      "step": 24815
    },
    {
      "entropy": 5.334198045730591,
      "epoch": 2.384245917387128,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044328299421367333,
      "loss": 4.7032,
      "mean_token_accuracy": 0.24905670285224915,
      "num_tokens": 56927791.0,
      "step": 24820
    },
    {
      "entropy": 5.390904188156128,
      "epoch": 2.3847262247838614,
      "grad_norm": 1.0,
      "learning_rate": 0.0004432602353881302,
      "loss": 4.7854,
      "mean_token_accuracy": 0.24086306095123292,
      "num_tokens": 56938848.0,
      "step": 24825
    },
    {
      "entropy": 5.378338098526001,
      "epoch": 2.3852065321805958,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044323747265608395,
      "loss": 4.7672,
      "mean_token_accuracy": 0.24482038021087646,
      "num_tokens": 56950116.0,
      "step": 24830
    },
    {
      "entropy": 5.369629716873169,
      "epoch": 2.3856868395773296,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004432147060180632,
      "loss": 4.7243,
      "mean_token_accuracy": 0.24448232501745223,
      "num_tokens": 56961911.0,
      "step": 24835
    },
    {
      "entropy": 5.421646356582642,
      "epoch": 2.3861671469740635,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044319193547459645,
      "loss": 4.7759,
      "mean_token_accuracy": 0.230791337788105,
      "num_tokens": 56974105.0,
      "step": 24840
    },
    {
      "entropy": 5.354376125335693,
      "epoch": 2.3866474543707974,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004431691610262124,
      "loss": 4.7245,
      "mean_token_accuracy": 0.2449526771903038,
      "num_tokens": 56985564.0,
      "step": 24845
    },
    {
      "entropy": 5.341426181793213,
      "epoch": 2.3871277617675313,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044314638267343976,
      "loss": 4.677,
      "mean_token_accuracy": 0.25597117394208907,
      "num_tokens": 56995835.0,
      "step": 24850
    },
    {
      "entropy": 5.262914085388184,
      "epoch": 2.387608069164265,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004431236004168075,
      "loss": 4.6333,
      "mean_token_accuracy": 0.24734197854995726,
      "num_tokens": 57008114.0,
      "step": 24855
    },
    {
      "entropy": 5.323145914077759,
      "epoch": 2.388088376560999,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004431008142568444,
      "loss": 4.7017,
      "mean_token_accuracy": 0.24549409002065659,
      "num_tokens": 57018968.0,
      "step": 24860
    },
    {
      "entropy": 5.365915060043335,
      "epoch": 2.388568683957733,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044307802419407954,
      "loss": 4.7524,
      "mean_token_accuracy": 0.2434718757867813,
      "num_tokens": 57031297.0,
      "step": 24865
    },
    {
      "entropy": 5.407230424880981,
      "epoch": 2.389048991354467,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004430552302290421,
      "loss": 4.7609,
      "mean_token_accuracy": 0.2429988607764244,
      "num_tokens": 57043457.0,
      "step": 24870
    },
    {
      "entropy": 5.321143722534179,
      "epoch": 2.3895292987512007,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004430324323622611,
      "loss": 4.6853,
      "mean_token_accuracy": 0.24727607220411302,
      "num_tokens": 57054474.0,
      "step": 24875
    },
    {
      "entropy": 5.353217601776123,
      "epoch": 2.3900096061479346,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044300963059426605,
      "loss": 4.7351,
      "mean_token_accuracy": 0.24782684594392776,
      "num_tokens": 57065120.0,
      "step": 24880
    },
    {
      "entropy": 5.3495735168457035,
      "epoch": 2.3904899135446684,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044298682492558637,
      "loss": 4.7297,
      "mean_token_accuracy": 0.2464601770043373,
      "num_tokens": 57076712.0,
      "step": 24885
    },
    {
      "entropy": 5.474457693099976,
      "epoch": 2.3909702209414023,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044296401535675136,
      "loss": 4.8416,
      "mean_token_accuracy": 0.23395977467298507,
      "num_tokens": 57087894.0,
      "step": 24890
    },
    {
      "entropy": 5.461782646179199,
      "epoch": 2.3914505283381366,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044294120188829056,
      "loss": 4.9106,
      "mean_token_accuracy": 0.22417801916599273,
      "num_tokens": 57101029.0,
      "step": 24895
    },
    {
      "entropy": 5.451648902893067,
      "epoch": 2.39193083573487,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004429183845207339,
      "loss": 4.7594,
      "mean_token_accuracy": 0.2416039764881134,
      "num_tokens": 57112893.0,
      "step": 24900
    },
    {
      "entropy": 5.38688473701477,
      "epoch": 2.3924111431316044,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004428955632546108,
      "loss": 4.7943,
      "mean_token_accuracy": 0.24550238251686096,
      "num_tokens": 57123899.0,
      "step": 24905
    },
    {
      "entropy": 5.357917642593383,
      "epoch": 2.3928914505283383,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004428727380904514,
      "loss": 4.7919,
      "mean_token_accuracy": 0.24541237950325012,
      "num_tokens": 57134896.0,
      "step": 24910
    },
    {
      "entropy": 5.263505554199218,
      "epoch": 2.393371757925072,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044284990902878545,
      "loss": 4.6028,
      "mean_token_accuracy": 0.24901290535926818,
      "num_tokens": 57146278.0,
      "step": 24915
    },
    {
      "entropy": 5.400255489349365,
      "epoch": 2.393852065321806,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044282707607014304,
      "loss": 4.733,
      "mean_token_accuracy": 0.24695148468017578,
      "num_tokens": 57157663.0,
      "step": 24920
    },
    {
      "entropy": 5.400369787216187,
      "epoch": 2.39433237271854,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044280423921505427,
      "loss": 4.6964,
      "mean_token_accuracy": 0.25263994485139846,
      "num_tokens": 57169131.0,
      "step": 24925
    },
    {
      "entropy": 5.364775800704956,
      "epoch": 2.394812680115274,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004427813984640493,
      "loss": 4.7377,
      "mean_token_accuracy": 0.23674312233924866,
      "num_tokens": 57181653.0,
      "step": 24930
    },
    {
      "entropy": 5.315454912185669,
      "epoch": 2.3952929875120077,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004427585538176585,
      "loss": 4.7284,
      "mean_token_accuracy": 0.24706507921218873,
      "num_tokens": 57192817.0,
      "step": 24935
    },
    {
      "entropy": 5.426525640487671,
      "epoch": 2.3957732949087416,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044273570527641223,
      "loss": 4.7942,
      "mean_token_accuracy": 0.24260084182024003,
      "num_tokens": 57203570.0,
      "step": 24940
    },
    {
      "entropy": 5.364018869400025,
      "epoch": 2.3962536023054755,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044271285284084097,
      "loss": 4.6806,
      "mean_token_accuracy": 0.24823171496391297,
      "num_tokens": 57214388.0,
      "step": 24945
    },
    {
      "entropy": 5.393468189239502,
      "epoch": 2.3967339097022093,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004426899965114752,
      "loss": 4.7681,
      "mean_token_accuracy": 0.2396840900182724,
      "num_tokens": 57225004.0,
      "step": 24950
    },
    {
      "entropy": 5.309605312347412,
      "epoch": 2.3972142170989432,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044266713628884566,
      "loss": 4.6996,
      "mean_token_accuracy": 0.2489694818854332,
      "num_tokens": 57236643.0,
      "step": 24955
    },
    {
      "entropy": 5.328159236907959,
      "epoch": 2.397694524495677,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044264427217348315,
      "loss": 4.7128,
      "mean_token_accuracy": 0.2452095165848732,
      "num_tokens": 57247502.0,
      "step": 24960
    },
    {
      "entropy": 5.449368190765381,
      "epoch": 2.398174831892411,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004426214041659184,
      "loss": 4.8665,
      "mean_token_accuracy": 0.23875218480825425,
      "num_tokens": 57259539.0,
      "step": 24965
    },
    {
      "entropy": 5.516855144500733,
      "epoch": 2.398655139289145,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004425985322666824,
      "loss": 4.8388,
      "mean_token_accuracy": 0.2363383024930954,
      "num_tokens": 57270946.0,
      "step": 24970
    },
    {
      "entropy": 5.367375278472901,
      "epoch": 2.3991354466858787,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004425756564763061,
      "loss": 4.7769,
      "mean_token_accuracy": 0.24190166741609573,
      "num_tokens": 57282855.0,
      "step": 24975
    },
    {
      "entropy": 5.421717691421509,
      "epoch": 2.399615754082613,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044255277679532075,
      "loss": 4.804,
      "mean_token_accuracy": 0.23556292653083802,
      "num_tokens": 57292638.0,
      "step": 24980
    },
    {
      "entropy": 5.415770101547241,
      "epoch": 2.400096061479347,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044252989322425735,
      "loss": 4.7601,
      "mean_token_accuracy": 0.24347348809242247,
      "num_tokens": 57304088.0,
      "step": 24985
    },
    {
      "entropy": 5.4424052238464355,
      "epoch": 2.400576368876081,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044250700576364734,
      "loss": 4.7371,
      "mean_token_accuracy": 0.24567839950323106,
      "num_tokens": 57314534.0,
      "step": 24990
    },
    {
      "entropy": 5.389139127731323,
      "epoch": 2.4010566762728147,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000442484114414022,
      "loss": 4.8144,
      "mean_token_accuracy": 0.2386138767004013,
      "num_tokens": 57326787.0,
      "step": 24995
    },
    {
      "entropy": 5.332731246948242,
      "epoch": 2.4015369836695486,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004424612191759129,
      "loss": 4.7259,
      "mean_token_accuracy": 0.24312745928764343,
      "num_tokens": 57338935.0,
      "step": 25000
    },
    {
      "entropy": 5.449759531021118,
      "epoch": 2.4020172910662825,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004424383200498515,
      "loss": 4.7679,
      "mean_token_accuracy": 0.24413973838090897,
      "num_tokens": 57349963.0,
      "step": 25005
    },
    {
      "entropy": 5.394966506958008,
      "epoch": 2.4024975984630164,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004424154170363696,
      "loss": 4.7635,
      "mean_token_accuracy": 0.24339204728603364,
      "num_tokens": 57361305.0,
      "step": 25010
    },
    {
      "entropy": 5.3406812191009525,
      "epoch": 2.4029779058597502,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004423925101359987,
      "loss": 4.7255,
      "mean_token_accuracy": 0.23943083584308625,
      "num_tokens": 57371582.0,
      "step": 25015
    },
    {
      "entropy": 5.383817672729492,
      "epoch": 2.403458213256484,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004423695993492709,
      "loss": 4.7322,
      "mean_token_accuracy": 0.23945662677288054,
      "num_tokens": 57382222.0,
      "step": 25020
    },
    {
      "entropy": 5.344041442871093,
      "epoch": 2.403938520653218,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004423466846767179,
      "loss": 4.7241,
      "mean_token_accuracy": 0.24130001217126845,
      "num_tokens": 57393939.0,
      "step": 25025
    },
    {
      "entropy": 5.398882818222046,
      "epoch": 2.404418828049952,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044232376611887185,
      "loss": 4.7303,
      "mean_token_accuracy": 0.24250920712947846,
      "num_tokens": 57404889.0,
      "step": 25030
    },
    {
      "entropy": 5.3964825630187985,
      "epoch": 2.4048991354466858,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044230084367626477,
      "loss": 4.8095,
      "mean_token_accuracy": 0.23977378606796265,
      "num_tokens": 57416296.0,
      "step": 25035
    },
    {
      "entropy": 5.3234028816223145,
      "epoch": 2.4053794428434196,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004422779173494288,
      "loss": 4.7725,
      "mean_token_accuracy": 0.24205521941185,
      "num_tokens": 57427926.0,
      "step": 25040
    },
    {
      "entropy": 5.4025167465209964,
      "epoch": 2.4058597502401535,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004422549871388965,
      "loss": 4.7382,
      "mean_token_accuracy": 0.2456643059849739,
      "num_tokens": 57439397.0,
      "step": 25045
    },
    {
      "entropy": 5.424097919464112,
      "epoch": 2.4063400576368874,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044223205304519994,
      "loss": 4.802,
      "mean_token_accuracy": 0.24320348352193832,
      "num_tokens": 57449976.0,
      "step": 25050
    },
    {
      "entropy": 5.430766963958741,
      "epoch": 2.4068203650336217,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004422091150688717,
      "loss": 4.8284,
      "mean_token_accuracy": 0.23413482457399368,
      "num_tokens": 57461578.0,
      "step": 25055
    },
    {
      "entropy": 5.3661009788513185,
      "epoch": 2.4073006724303556,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004421861732104443,
      "loss": 4.7768,
      "mean_token_accuracy": 0.2405850514769554,
      "num_tokens": 57471832.0,
      "step": 25060
    },
    {
      "entropy": 5.447331809997559,
      "epoch": 2.4077809798270895,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004421632274704504,
      "loss": 4.7936,
      "mean_token_accuracy": 0.2422275260090828,
      "num_tokens": 57482854.0,
      "step": 25065
    },
    {
      "entropy": 5.4147998809814455,
      "epoch": 2.4082612872238234,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004421402778494227,
      "loss": 4.8311,
      "mean_token_accuracy": 0.2378895878791809,
      "num_tokens": 57494243.0,
      "step": 25070
    },
    {
      "entropy": 5.413335704803467,
      "epoch": 2.4087415946205573,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004421173243478941,
      "loss": 4.7053,
      "mean_token_accuracy": 0.24278289526700975,
      "num_tokens": 57506185.0,
      "step": 25075
    },
    {
      "entropy": 5.404696416854859,
      "epoch": 2.409221902017291,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044209436696639745,
      "loss": 4.7418,
      "mean_token_accuracy": 0.24618547707796096,
      "num_tokens": 57517980.0,
      "step": 25080
    },
    {
      "entropy": 5.2989014148712155,
      "epoch": 2.409702209414025,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044207140570546574,
      "loss": 4.6285,
      "mean_token_accuracy": 0.253798845410347,
      "num_tokens": 57529331.0,
      "step": 25085
    },
    {
      "entropy": 5.372100067138672,
      "epoch": 2.410182516810759,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044204844056563216,
      "loss": 4.7154,
      "mean_token_accuracy": 0.245811827480793,
      "num_tokens": 57540463.0,
      "step": 25090
    },
    {
      "entropy": 5.430115652084351,
      "epoch": 2.410662824207493,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004420254715474297,
      "loss": 4.8643,
      "mean_token_accuracy": 0.23662404865026473,
      "num_tokens": 57550896.0,
      "step": 25095
    },
    {
      "entropy": 5.39035930633545,
      "epoch": 2.4111431316042267,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044200249865139187,
      "loss": 4.7256,
      "mean_token_accuracy": 0.251804019510746,
      "num_tokens": 57562041.0,
      "step": 25100
    },
    {
      "entropy": 5.490136671066284,
      "epoch": 2.4116234390009605,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044197952187805185,
      "loss": 4.8748,
      "mean_token_accuracy": 0.2374124139547348,
      "num_tokens": 57573771.0,
      "step": 25105
    },
    {
      "entropy": 5.373491811752319,
      "epoch": 2.4121037463976944,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00044195654122794324,
      "loss": 4.7139,
      "mean_token_accuracy": 0.24144657850265502,
      "num_tokens": 57585651.0,
      "step": 25110
    },
    {
      "entropy": 5.347451114654541,
      "epoch": 2.4125840537944283,
      "grad_norm": 1.125,
      "learning_rate": 0.0004419335567015994,
      "loss": 4.7335,
      "mean_token_accuracy": 0.24801772385835646,
      "num_tokens": 57596328.0,
      "step": 25115
    },
    {
      "entropy": 5.3098366260528564,
      "epoch": 2.413064361191162,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004419105682995542,
      "loss": 4.6734,
      "mean_token_accuracy": 0.24192522913217546,
      "num_tokens": 57608012.0,
      "step": 25120
    },
    {
      "entropy": 5.4525322914123535,
      "epoch": 2.413544668587896,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004418875760223411,
      "loss": 4.7517,
      "mean_token_accuracy": 0.23932038396596908,
      "num_tokens": 57619711.0,
      "step": 25125
    },
    {
      "entropy": 5.405850887298584,
      "epoch": 2.4140249759846304,
      "grad_norm": 1.125,
      "learning_rate": 0.00044186457987049405,
      "loss": 4.731,
      "mean_token_accuracy": 0.2425232082605362,
      "num_tokens": 57630263.0,
      "step": 25130
    },
    {
      "entropy": 5.350929880142212,
      "epoch": 2.414505283381364,
      "grad_norm": 1.15625,
      "learning_rate": 0.000441841579844547,
      "loss": 4.7492,
      "mean_token_accuracy": 0.23773858100175857,
      "num_tokens": 57640973.0,
      "step": 25135
    },
    {
      "entropy": 5.366160249710083,
      "epoch": 2.414985590778098,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004418185759450338,
      "loss": 4.6877,
      "mean_token_accuracy": 0.2452749103307724,
      "num_tokens": 57651698.0,
      "step": 25140
    },
    {
      "entropy": 5.363884162902832,
      "epoch": 2.415465898174832,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004417955681724887,
      "loss": 4.7374,
      "mean_token_accuracy": 0.24178969115018845,
      "num_tokens": 57662547.0,
      "step": 25145
    },
    {
      "entropy": 5.429005813598633,
      "epoch": 2.415946205571566,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044177255652744576,
      "loss": 4.8366,
      "mean_token_accuracy": 0.23525419533252717,
      "num_tokens": 57675665.0,
      "step": 25150
    },
    {
      "entropy": 5.35930027961731,
      "epoch": 2.4164265129683,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044174954101043926,
      "loss": 4.6942,
      "mean_token_accuracy": 0.2501408115029335,
      "num_tokens": 57687765.0,
      "step": 25155
    },
    {
      "entropy": 5.414928293228149,
      "epoch": 2.4169068203650337,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044172652162200354,
      "loss": 4.7778,
      "mean_token_accuracy": 0.23721953630447387,
      "num_tokens": 57699575.0,
      "step": 25160
    },
    {
      "entropy": 5.457152462005615,
      "epoch": 2.4173871277617676,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004417034983626731,
      "loss": 4.7968,
      "mean_token_accuracy": 0.24390049725770951,
      "num_tokens": 57710014.0,
      "step": 25165
    },
    {
      "entropy": 5.3275071144104,
      "epoch": 2.4178674351585014,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004416804712329825,
      "loss": 4.6527,
      "mean_token_accuracy": 0.25452283024787903,
      "num_tokens": 57720763.0,
      "step": 25170
    },
    {
      "entropy": 5.3909680366516115,
      "epoch": 2.4183477425552353,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044165744023346614,
      "loss": 4.8028,
      "mean_token_accuracy": 0.2379068687558174,
      "num_tokens": 57731581.0,
      "step": 25175
    },
    {
      "entropy": 5.435700368881226,
      "epoch": 2.418828049951969,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044163440536465904,
      "loss": 4.8349,
      "mean_token_accuracy": 0.23993818163871766,
      "num_tokens": 57742473.0,
      "step": 25180
    },
    {
      "entropy": 5.442564249038696,
      "epoch": 2.419308357348703,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044161136662709577,
      "loss": 4.8448,
      "mean_token_accuracy": 0.23644158840179444,
      "num_tokens": 57754317.0,
      "step": 25185
    },
    {
      "entropy": 5.354081487655639,
      "epoch": 2.419788664745437,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044158832402131133,
      "loss": 4.7498,
      "mean_token_accuracy": 0.24501430690288545,
      "num_tokens": 57765882.0,
      "step": 25190
    },
    {
      "entropy": 5.386537504196167,
      "epoch": 2.420268972142171,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044156527754784066,
      "loss": 4.7145,
      "mean_token_accuracy": 0.24881352633237838,
      "num_tokens": 57776977.0,
      "step": 25195
    },
    {
      "entropy": 5.417631435394287,
      "epoch": 2.4207492795389047,
      "grad_norm": 1.0,
      "learning_rate": 0.00044154222720721887,
      "loss": 4.7889,
      "mean_token_accuracy": 0.2485474705696106,
      "num_tokens": 57788652.0,
      "step": 25200
    },
    {
      "entropy": 5.464466762542725,
      "epoch": 2.421229586935639,
      "grad_norm": 1.125,
      "learning_rate": 0.000441519172999981,
      "loss": 4.7938,
      "mean_token_accuracy": 0.235762557387352,
      "num_tokens": 57799660.0,
      "step": 25205
    },
    {
      "entropy": 5.418867444992065,
      "epoch": 2.4217098943323725,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004414961149266625,
      "loss": 4.8526,
      "mean_token_accuracy": 0.2281285047531128,
      "num_tokens": 57811754.0,
      "step": 25210
    },
    {
      "entropy": 5.427315664291382,
      "epoch": 2.422190201729107,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044147305298779856,
      "loss": 4.7607,
      "mean_token_accuracy": 0.24710265249013902,
      "num_tokens": 57823470.0,
      "step": 25215
    },
    {
      "entropy": 5.484676456451416,
      "epoch": 2.4226705091258407,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004414499871839247,
      "loss": 4.8642,
      "mean_token_accuracy": 0.2375214084982872,
      "num_tokens": 57834917.0,
      "step": 25220
    },
    {
      "entropy": 5.401324510574341,
      "epoch": 2.4231508165225746,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004414269175155763,
      "loss": 4.7675,
      "mean_token_accuracy": 0.23911667764186859,
      "num_tokens": 57844655.0,
      "step": 25225
    },
    {
      "entropy": 5.432307767868042,
      "epoch": 2.4236311239193085,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004414038439832891,
      "loss": 4.799,
      "mean_token_accuracy": 0.24375952482223512,
      "num_tokens": 57856019.0,
      "step": 25230
    },
    {
      "entropy": 5.392108535766601,
      "epoch": 2.4241114313160423,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004413807665875988,
      "loss": 4.7544,
      "mean_token_accuracy": 0.24336008429527284,
      "num_tokens": 57867661.0,
      "step": 25235
    },
    {
      "entropy": 5.3954095363616945,
      "epoch": 2.4245917387127762,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044135768532904104,
      "loss": 4.7602,
      "mean_token_accuracy": 0.2406143844127655,
      "num_tokens": 57879899.0,
      "step": 25240
    },
    {
      "entropy": 5.469573450088501,
      "epoch": 2.42507204610951,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000441334600208152,
      "loss": 4.8251,
      "mean_token_accuracy": 0.23193451166152954,
      "num_tokens": 57891168.0,
      "step": 25245
    },
    {
      "entropy": 5.38267240524292,
      "epoch": 2.425552353506244,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044131151122546724,
      "loss": 4.706,
      "mean_token_accuracy": 0.2416976884007454,
      "num_tokens": 57901997.0,
      "step": 25250
    },
    {
      "entropy": 5.395679092407226,
      "epoch": 2.426032660902978,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044128841838152313,
      "loss": 4.7654,
      "mean_token_accuracy": 0.24526615887880326,
      "num_tokens": 57913380.0,
      "step": 25255
    },
    {
      "entropy": 5.345044136047363,
      "epoch": 2.4265129682997117,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004412653216768558,
      "loss": 4.6691,
      "mean_token_accuracy": 0.2484130784869194,
      "num_tokens": 57923675.0,
      "step": 25260
    },
    {
      "entropy": 5.4317787170410154,
      "epoch": 2.4269932756964456,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004412422211120013,
      "loss": 4.7798,
      "mean_token_accuracy": 0.24325423836708068,
      "num_tokens": 57935885.0,
      "step": 25265
    },
    {
      "entropy": 5.378539562225342,
      "epoch": 2.4274735830931795,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004412191166874961,
      "loss": 4.7446,
      "mean_token_accuracy": 0.24204291999340058,
      "num_tokens": 57946551.0,
      "step": 25270
    },
    {
      "entropy": 5.34529128074646,
      "epoch": 2.4279538904899134,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004411960084038766,
      "loss": 4.6611,
      "mean_token_accuracy": 0.2494047164916992,
      "num_tokens": 57957778.0,
      "step": 25275
    },
    {
      "entropy": 5.381160831451416,
      "epoch": 2.4284341978866473,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044117289626167917,
      "loss": 4.7809,
      "mean_token_accuracy": 0.2446133404970169,
      "num_tokens": 57969256.0,
      "step": 25280
    },
    {
      "entropy": 5.393218040466309,
      "epoch": 2.428914505283381,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004411497802614406,
      "loss": 4.714,
      "mean_token_accuracy": 0.2489876016974449,
      "num_tokens": 57981368.0,
      "step": 25285
    },
    {
      "entropy": 5.37966480255127,
      "epoch": 2.4293948126801155,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004411266604036975,
      "loss": 4.8247,
      "mean_token_accuracy": 0.23801179528236388,
      "num_tokens": 57992862.0,
      "step": 25290
    },
    {
      "entropy": 5.324821138381958,
      "epoch": 2.4298751200768494,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044110353668898674,
      "loss": 4.6493,
      "mean_token_accuracy": 0.2516516610980034,
      "num_tokens": 58003492.0,
      "step": 25295
    },
    {
      "entropy": 5.343811655044556,
      "epoch": 2.4303554274735832,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004410804091178449,
      "loss": 4.663,
      "mean_token_accuracy": 0.2490598663687706,
      "num_tokens": 58014623.0,
      "step": 25300
    },
    {
      "entropy": 5.458928918838501,
      "epoch": 2.430835734870317,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004410572776908092,
      "loss": 4.8101,
      "mean_token_accuracy": 0.2447643220424652,
      "num_tokens": 58024886.0,
      "step": 25305
    },
    {
      "entropy": 5.417258262634277,
      "epoch": 2.431316042267051,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044103414240841664,
      "loss": 4.8051,
      "mean_token_accuracy": 0.24110288619995118,
      "num_tokens": 58035998.0,
      "step": 25310
    },
    {
      "entropy": 5.429548168182373,
      "epoch": 2.431796349663785,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004410110032712043,
      "loss": 4.7586,
      "mean_token_accuracy": 0.250640332698822,
      "num_tokens": 58046577.0,
      "step": 25315
    },
    {
      "entropy": 5.335135793685913,
      "epoch": 2.4322766570605188,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004409878602797094,
      "loss": 4.721,
      "mean_token_accuracy": 0.24538574665784835,
      "num_tokens": 58057711.0,
      "step": 25320
    },
    {
      "entropy": 5.3735435009002686,
      "epoch": 2.4327569644572526,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044096471343446923,
      "loss": 4.692,
      "mean_token_accuracy": 0.24724264591932296,
      "num_tokens": 58069657.0,
      "step": 25325
    },
    {
      "entropy": 5.406388616561889,
      "epoch": 2.4332372718539865,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004409415627360213,
      "loss": 4.7659,
      "mean_token_accuracy": 0.24297385960817336,
      "num_tokens": 58081081.0,
      "step": 25330
    },
    {
      "entropy": 5.44472975730896,
      "epoch": 2.4337175792507204,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044091840818490303,
      "loss": 4.7106,
      "mean_token_accuracy": 0.23477858752012254,
      "num_tokens": 58091561.0,
      "step": 25335
    },
    {
      "entropy": 5.346974611282349,
      "epoch": 2.4341978866474543,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044089524978165197,
      "loss": 4.7304,
      "mean_token_accuracy": 0.2459243655204773,
      "num_tokens": 58102640.0,
      "step": 25340
    },
    {
      "entropy": 5.378867864608765,
      "epoch": 2.434678194044188,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044087208752680577,
      "loss": 4.8432,
      "mean_token_accuracy": 0.23132905066013337,
      "num_tokens": 58113356.0,
      "step": 25345
    },
    {
      "entropy": 5.450962781906128,
      "epoch": 2.435158501440922,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004408489214209023,
      "loss": 4.8569,
      "mean_token_accuracy": 0.23311981409788132,
      "num_tokens": 58125437.0,
      "step": 25350
    },
    {
      "entropy": 5.499756622314453,
      "epoch": 2.435638808837656,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004408257514644793,
      "loss": 4.8224,
      "mean_token_accuracy": 0.23946654349565505,
      "num_tokens": 58138491.0,
      "step": 25355
    },
    {
      "entropy": 5.37842583656311,
      "epoch": 2.43611911623439,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044080257765807476,
      "loss": 4.7598,
      "mean_token_accuracy": 0.23774637877941132,
      "num_tokens": 58150200.0,
      "step": 25360
    },
    {
      "entropy": 5.389662265777588,
      "epoch": 2.436599423631124,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004407794000022267,
      "loss": 4.8648,
      "mean_token_accuracy": 0.2430683434009552,
      "num_tokens": 58162467.0,
      "step": 25365
    },
    {
      "entropy": 5.503590154647827,
      "epoch": 2.437079731027858,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004407562184974732,
      "loss": 4.8687,
      "mean_token_accuracy": 0.2286500945687294,
      "num_tokens": 58175052.0,
      "step": 25370
    },
    {
      "entropy": 5.436009550094605,
      "epoch": 2.437560038424592,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004407330331443526,
      "loss": 4.7688,
      "mean_token_accuracy": 0.2451614946126938,
      "num_tokens": 58186999.0,
      "step": 25375
    },
    {
      "entropy": 5.360422611236572,
      "epoch": 2.438040345821326,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000440709843943403,
      "loss": 4.672,
      "mean_token_accuracy": 0.24780011177062988,
      "num_tokens": 58197811.0,
      "step": 25380
    },
    {
      "entropy": 5.470729684829712,
      "epoch": 2.4385206532180597,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000440686650895163,
      "loss": 4.878,
      "mean_token_accuracy": 0.2380610227584839,
      "num_tokens": 58209545.0,
      "step": 25385
    },
    {
      "entropy": 5.353319692611694,
      "epoch": 2.4390009606147935,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044066345400017084,
      "loss": 4.6715,
      "mean_token_accuracy": 0.2495146229863167,
      "num_tokens": 58220994.0,
      "step": 25390
    },
    {
      "entropy": 5.3160679817199705,
      "epoch": 2.4394812680115274,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044064025325896524,
      "loss": 4.6277,
      "mean_token_accuracy": 0.25380575060844424,
      "num_tokens": 58232778.0,
      "step": 25395
    },
    {
      "entropy": 5.429329347610474,
      "epoch": 2.4399615754082613,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044061704867208484,
      "loss": 4.7912,
      "mean_token_accuracy": 0.24779620468616487,
      "num_tokens": 58243889.0,
      "step": 25400
    },
    {
      "entropy": 5.332193803787232,
      "epoch": 2.440441882804995,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044059384024006825,
      "loss": 4.7042,
      "mean_token_accuracy": 0.24877294301986694,
      "num_tokens": 58256144.0,
      "step": 25405
    },
    {
      "entropy": 5.396083736419678,
      "epoch": 2.440922190201729,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004405706279634545,
      "loss": 4.7357,
      "mean_token_accuracy": 0.2432622343301773,
      "num_tokens": 58267509.0,
      "step": 25410
    },
    {
      "entropy": 5.341105937957764,
      "epoch": 2.441402497598463,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044054741184278243,
      "loss": 4.6801,
      "mean_token_accuracy": 0.25425427258014677,
      "num_tokens": 58278880.0,
      "step": 25415
    },
    {
      "entropy": 5.428200387954712,
      "epoch": 2.441882804995197,
      "grad_norm": 1.125,
      "learning_rate": 0.00044052419187859095,
      "loss": 4.7228,
      "mean_token_accuracy": 0.2408788815140724,
      "num_tokens": 58290046.0,
      "step": 25420
    },
    {
      "entropy": 5.3101543426513675,
      "epoch": 2.4423631123919307,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004405009680714193,
      "loss": 4.6049,
      "mean_token_accuracy": 0.2592446908354759,
      "num_tokens": 58301406.0,
      "step": 25425
    },
    {
      "entropy": 5.353295707702637,
      "epoch": 2.4428434197886646,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004404777404218065,
      "loss": 4.6763,
      "mean_token_accuracy": 0.2531026616692543,
      "num_tokens": 58313064.0,
      "step": 25430
    },
    {
      "entropy": 5.375765943527222,
      "epoch": 2.4433237271853985,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000440454508930292,
      "loss": 4.8211,
      "mean_token_accuracy": 0.23146286755800247,
      "num_tokens": 58324779.0,
      "step": 25435
    },
    {
      "entropy": 5.4561323642730715,
      "epoch": 2.443804034582133,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004404312735974152,
      "loss": 4.8095,
      "mean_token_accuracy": 0.24069816321134568,
      "num_tokens": 58336064.0,
      "step": 25440
    },
    {
      "entropy": 5.365040588378906,
      "epoch": 2.4442843419788662,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044040803442371533,
      "loss": 4.6708,
      "mean_token_accuracy": 0.24879835397005082,
      "num_tokens": 58347497.0,
      "step": 25445
    },
    {
      "entropy": 5.29216775894165,
      "epoch": 2.4447646493756006,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004403847914097321,
      "loss": 4.6631,
      "mean_token_accuracy": 0.25356304496526716,
      "num_tokens": 58357716.0,
      "step": 25450
    },
    {
      "entropy": 5.390027236938477,
      "epoch": 2.4452449567723344,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044036154455600517,
      "loss": 4.781,
      "mean_token_accuracy": 0.24298169016838073,
      "num_tokens": 58368451.0,
      "step": 25455
    },
    {
      "entropy": 5.3990724086761475,
      "epoch": 2.4457252641690683,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004403382938630741,
      "loss": 4.7601,
      "mean_token_accuracy": 0.24161727726459503,
      "num_tokens": 58378915.0,
      "step": 25460
    },
    {
      "entropy": 5.390514278411866,
      "epoch": 2.446205571565802,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044031503933147887,
      "loss": 4.6656,
      "mean_token_accuracy": 0.2610498383641243,
      "num_tokens": 58390251.0,
      "step": 25465
    },
    {
      "entropy": 5.350936460494995,
      "epoch": 2.446685878962536,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044029178096175934,
      "loss": 4.6944,
      "mean_token_accuracy": 0.24399091750383378,
      "num_tokens": 58401390.0,
      "step": 25470
    },
    {
      "entropy": 5.333190107345581,
      "epoch": 2.44716618635927,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004402685187544554,
      "loss": 4.7304,
      "mean_token_accuracy": 0.2539989486336708,
      "num_tokens": 58412530.0,
      "step": 25475
    },
    {
      "entropy": 5.430140638351441,
      "epoch": 2.447646493756004,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004402452527101072,
      "loss": 4.7939,
      "mean_token_accuracy": 0.24286381006240845,
      "num_tokens": 58424437.0,
      "step": 25480
    },
    {
      "entropy": 5.45406289100647,
      "epoch": 2.4481268011527377,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004402219828292549,
      "loss": 4.7828,
      "mean_token_accuracy": 0.2382509380578995,
      "num_tokens": 58435661.0,
      "step": 25485
    },
    {
      "entropy": 5.457407093048095,
      "epoch": 2.4486071085494716,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004401987091124388,
      "loss": 4.8676,
      "mean_token_accuracy": 0.2303012266755104,
      "num_tokens": 58445840.0,
      "step": 25490
    },
    {
      "entropy": 5.403382730484009,
      "epoch": 2.4490874159462055,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004401754315601992,
      "loss": 4.819,
      "mean_token_accuracy": 0.23290482461452483,
      "num_tokens": 58460160.0,
      "step": 25495
    },
    {
      "entropy": 5.351606941223144,
      "epoch": 2.4495677233429394,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004401521501730765,
      "loss": 4.6469,
      "mean_token_accuracy": 0.2521843835711479,
      "num_tokens": 58471257.0,
      "step": 25500
    },
    {
      "entropy": 5.298886251449585,
      "epoch": 2.4500480307396733,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044012886495161144,
      "loss": 4.6948,
      "mean_token_accuracy": 0.24499523490667344,
      "num_tokens": 58482830.0,
      "step": 25505
    },
    {
      "entropy": 5.378184747695923,
      "epoch": 2.450528338136407,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004401055758963443,
      "loss": 4.7165,
      "mean_token_accuracy": 0.2532515212893486,
      "num_tokens": 58494278.0,
      "step": 25510
    },
    {
      "entropy": 5.368350982666016,
      "epoch": 2.451008645533141,
      "grad_norm": 1.125,
      "learning_rate": 0.000440082283007816,
      "loss": 4.7208,
      "mean_token_accuracy": 0.23849379718303682,
      "num_tokens": 58505653.0,
      "step": 25515
    },
    {
      "entropy": 5.419531202316284,
      "epoch": 2.451488952929875,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044005898628656734,
      "loss": 4.7227,
      "mean_token_accuracy": 0.24240888357162477,
      "num_tokens": 58516906.0,
      "step": 25520
    },
    {
      "entropy": 5.411347723007202,
      "epoch": 2.4519692603266092,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000440035685733139,
      "loss": 4.761,
      "mean_token_accuracy": 0.24563549160957338,
      "num_tokens": 58527326.0,
      "step": 25525
    },
    {
      "entropy": 5.367625665664673,
      "epoch": 2.452449567723343,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004400123813480722,
      "loss": 4.6904,
      "mean_token_accuracy": 0.25090546905994415,
      "num_tokens": 58537749.0,
      "step": 25530
    },
    {
      "entropy": 5.370323848724365,
      "epoch": 2.452929875120077,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043998907313190787,
      "loss": 4.7468,
      "mean_token_accuracy": 0.23745930790901185,
      "num_tokens": 58549479.0,
      "step": 25535
    },
    {
      "entropy": 5.3792094707489015,
      "epoch": 2.453410182516811,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004399657610851873,
      "loss": 4.7311,
      "mean_token_accuracy": 0.24654944241046906,
      "num_tokens": 58561051.0,
      "step": 25540
    },
    {
      "entropy": 5.417632102966309,
      "epoch": 2.4538904899135447,
      "grad_norm": 1.125,
      "learning_rate": 0.00043994244520845146,
      "loss": 4.8423,
      "mean_token_accuracy": 0.23531692028045653,
      "num_tokens": 58571482.0,
      "step": 25545
    },
    {
      "entropy": 5.396629571914673,
      "epoch": 2.4543707973102786,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004399191255022418,
      "loss": 4.7148,
      "mean_token_accuracy": 0.24531390964984895,
      "num_tokens": 58584369.0,
      "step": 25550
    },
    {
      "entropy": 5.401398706436157,
      "epoch": 2.4548511047070125,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004398958019670998,
      "loss": 4.7109,
      "mean_token_accuracy": 0.24777349829673767,
      "num_tokens": 58595786.0,
      "step": 25555
    },
    {
      "entropy": 5.324465417861939,
      "epoch": 2.4553314121037464,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043987247460356696,
      "loss": 4.7085,
      "mean_token_accuracy": 0.24785555452108382,
      "num_tokens": 58608536.0,
      "step": 25560
    },
    {
      "entropy": 5.288117456436157,
      "epoch": 2.4558117195004803,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004398491434121847,
      "loss": 4.662,
      "mean_token_accuracy": 0.24949141591787338,
      "num_tokens": 58620981.0,
      "step": 25565
    },
    {
      "entropy": 5.439028549194336,
      "epoch": 2.456292026897214,
      "grad_norm": 1.203125,
      "learning_rate": 0.000439825808393495,
      "loss": 4.7418,
      "mean_token_accuracy": 0.24214016646146774,
      "num_tokens": 58632802.0,
      "step": 25570
    },
    {
      "entropy": 5.421792697906494,
      "epoch": 2.456772334293948,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004398024695480394,
      "loss": 4.7985,
      "mean_token_accuracy": 0.24561038613319397,
      "num_tokens": 58644577.0,
      "step": 25575
    },
    {
      "entropy": 5.41075553894043,
      "epoch": 2.457252641690682,
      "grad_norm": 1.125,
      "learning_rate": 0.0004397791268763598,
      "loss": 4.7333,
      "mean_token_accuracy": 0.24557553231716156,
      "num_tokens": 58654970.0,
      "step": 25580
    },
    {
      "entropy": 5.390444040298462,
      "epoch": 2.457732949087416,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043975578037899814,
      "loss": 4.8245,
      "mean_token_accuracy": 0.23927068561315537,
      "num_tokens": 58666534.0,
      "step": 25585
    },
    {
      "entropy": 5.371790075302124,
      "epoch": 2.4582132564841497,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004397324300564966,
      "loss": 4.7173,
      "mean_token_accuracy": 0.24546075314283372,
      "num_tokens": 58677451.0,
      "step": 25590
    },
    {
      "entropy": 5.400810432434082,
      "epoch": 2.4586935638808836,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004397090759093971,
      "loss": 4.7268,
      "mean_token_accuracy": 0.2488635376095772,
      "num_tokens": 58689179.0,
      "step": 25595
    },
    {
      "entropy": 5.389435720443726,
      "epoch": 2.459173871277618,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043968571793824194,
      "loss": 4.8144,
      "mean_token_accuracy": 0.24551637321710587,
      "num_tokens": 58701939.0,
      "step": 25600
    },
    {
      "entropy": 5.307760858535767,
      "epoch": 2.4596541786743518,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004396623561435734,
      "loss": 4.5959,
      "mean_token_accuracy": 0.25961717367172243,
      "num_tokens": 58714302.0,
      "step": 25605
    },
    {
      "entropy": 5.376425886154175,
      "epoch": 2.4601344860710856,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004396389905259339,
      "loss": 4.7457,
      "mean_token_accuracy": 0.2428629592061043,
      "num_tokens": 58727619.0,
      "step": 25610
    },
    {
      "entropy": 5.427079057693481,
      "epoch": 2.4606147934678195,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043961562108586603,
      "loss": 4.7855,
      "mean_token_accuracy": 0.23633432239294053,
      "num_tokens": 58738929.0,
      "step": 25615
    },
    {
      "entropy": 5.4626524448394775,
      "epoch": 2.4610951008645534,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043959224782391215,
      "loss": 4.8251,
      "mean_token_accuracy": 0.24330597817897798,
      "num_tokens": 58751771.0,
      "step": 25620
    },
    {
      "entropy": 5.315464496612549,
      "epoch": 2.4615754082612873,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000439568870740615,
      "loss": 4.662,
      "mean_token_accuracy": 0.2574248418211937,
      "num_tokens": 58762304.0,
      "step": 25625
    },
    {
      "entropy": 5.335001277923584,
      "epoch": 2.462055715658021,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004395454898365174,
      "loss": 4.72,
      "mean_token_accuracy": 0.23894039988517762,
      "num_tokens": 58775222.0,
      "step": 25630
    },
    {
      "entropy": 5.345364665985107,
      "epoch": 2.462536023054755,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043952210511216205,
      "loss": 4.7059,
      "mean_token_accuracy": 0.2470898449420929,
      "num_tokens": 58785904.0,
      "step": 25635
    },
    {
      "entropy": 5.311428642272949,
      "epoch": 2.463016330451489,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043949871656809205,
      "loss": 4.7196,
      "mean_token_accuracy": 0.2505138576030731,
      "num_tokens": 58796186.0,
      "step": 25640
    },
    {
      "entropy": 5.403340911865234,
      "epoch": 2.463496637848223,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043947532420485024,
      "loss": 4.7728,
      "mean_token_accuracy": 0.24650858491659164,
      "num_tokens": 58807317.0,
      "step": 25645
    },
    {
      "entropy": 5.454269790649414,
      "epoch": 2.4639769452449567,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004394519280229798,
      "loss": 4.8433,
      "mean_token_accuracy": 0.23962317854166032,
      "num_tokens": 58819108.0,
      "step": 25650
    },
    {
      "entropy": 5.361330461502075,
      "epoch": 2.4644572526416906,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043942852802302397,
      "loss": 4.7612,
      "mean_token_accuracy": 0.24043281227350236,
      "num_tokens": 58830962.0,
      "step": 25655
    },
    {
      "entropy": 5.441388654708862,
      "epoch": 2.4649375600384245,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004394051242055259,
      "loss": 4.8664,
      "mean_token_accuracy": 0.23230114132165908,
      "num_tokens": 58843236.0,
      "step": 25660
    },
    {
      "entropy": 5.361920928955078,
      "epoch": 2.4654178674351583,
      "grad_norm": 1.125,
      "learning_rate": 0.0004393817165710291,
      "loss": 4.6984,
      "mean_token_accuracy": 0.24878550469875335,
      "num_tokens": 58853995.0,
      "step": 25665
    },
    {
      "entropy": 5.35396409034729,
      "epoch": 2.465898174831892,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043935830512007687,
      "loss": 4.693,
      "mean_token_accuracy": 0.25704448819160464,
      "num_tokens": 58866563.0,
      "step": 25670
    },
    {
      "entropy": 5.447143936157227,
      "epoch": 2.4663784822286265,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00043933488985321286,
      "loss": 4.8401,
      "mean_token_accuracy": 0.23659503757953643,
      "num_tokens": 58878660.0,
      "step": 25675
    },
    {
      "entropy": 5.3441990375518795,
      "epoch": 2.46685878962536,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004393114707709807,
      "loss": 4.6939,
      "mean_token_accuracy": 0.24830996245145798,
      "num_tokens": 58890332.0,
      "step": 25680
    },
    {
      "entropy": 5.4060649394989015,
      "epoch": 2.4673390970220943,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004392880478739241,
      "loss": 4.79,
      "mean_token_accuracy": 0.23591913282871246,
      "num_tokens": 58902228.0,
      "step": 25685
    },
    {
      "entropy": 5.427274179458618,
      "epoch": 2.467819404418828,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004392646211625869,
      "loss": 4.7647,
      "mean_token_accuracy": 0.24449009895324708,
      "num_tokens": 58914452.0,
      "step": 25690
    },
    {
      "entropy": 5.386216020584106,
      "epoch": 2.468299711815562,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004392411906375129,
      "loss": 4.738,
      "mean_token_accuracy": 0.2456927329301834,
      "num_tokens": 58926079.0,
      "step": 25695
    },
    {
      "entropy": 5.357523345947266,
      "epoch": 2.468780019212296,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043921775629924615,
      "loss": 4.7186,
      "mean_token_accuracy": 0.2530702963471413,
      "num_tokens": 58937292.0,
      "step": 25700
    },
    {
      "entropy": 5.374798774719238,
      "epoch": 2.46926032660903,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043919431814833077,
      "loss": 4.7766,
      "mean_token_accuracy": 0.241981141269207,
      "num_tokens": 58949425.0,
      "step": 25705
    },
    {
      "entropy": 5.433634662628174,
      "epoch": 2.4697406340057637,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043917087618531084,
      "loss": 4.7744,
      "mean_token_accuracy": 0.24196288734674454,
      "num_tokens": 58961620.0,
      "step": 25710
    },
    {
      "entropy": 5.423659801483154,
      "epoch": 2.4702209414024976,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004391474304107307,
      "loss": 4.6956,
      "mean_token_accuracy": 0.24756777733564378,
      "num_tokens": 58973043.0,
      "step": 25715
    },
    {
      "entropy": 5.404456758499146,
      "epoch": 2.4707012487992315,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043912398082513463,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22713396400213243,
      "num_tokens": 58984516.0,
      "step": 25720
    },
    {
      "entropy": 5.355918502807617,
      "epoch": 2.4711815561959654,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000439100527429067,
      "loss": 4.6881,
      "mean_token_accuracy": 0.2490299105644226,
      "num_tokens": 58996473.0,
      "step": 25725
    },
    {
      "entropy": 5.388628673553467,
      "epoch": 2.4716618635926992,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043907707022307243,
      "loss": 4.7,
      "mean_token_accuracy": 0.24315328449010848,
      "num_tokens": 59009001.0,
      "step": 25730
    },
    {
      "entropy": 5.454033708572387,
      "epoch": 2.472142170989433,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043905360920769553,
      "loss": 4.8068,
      "mean_token_accuracy": 0.24469457268714906,
      "num_tokens": 59021238.0,
      "step": 25735
    },
    {
      "entropy": 5.443097829818726,
      "epoch": 2.472622478386167,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000439030144383481,
      "loss": 4.7989,
      "mean_token_accuracy": 0.24826587438583375,
      "num_tokens": 59031311.0,
      "step": 25740
    },
    {
      "entropy": 5.37871994972229,
      "epoch": 2.473102785782901,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043900667575097355,
      "loss": 4.7034,
      "mean_token_accuracy": 0.2477331981062889,
      "num_tokens": 59043151.0,
      "step": 25745
    },
    {
      "entropy": 5.451038789749146,
      "epoch": 2.473583093179635,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004389832033107181,
      "loss": 4.8172,
      "mean_token_accuracy": 0.2431465744972229,
      "num_tokens": 59054513.0,
      "step": 25750
    },
    {
      "entropy": 5.40445876121521,
      "epoch": 2.4740634005763686,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043895972706325953,
      "loss": 4.7111,
      "mean_token_accuracy": 0.24367939829826354,
      "num_tokens": 59064909.0,
      "step": 25755
    },
    {
      "entropy": 5.353833484649658,
      "epoch": 2.474543707973103,
      "grad_norm": 1.078125,
      "learning_rate": 0.000438936247009143,
      "loss": 4.7695,
      "mean_token_accuracy": 0.23789274990558623,
      "num_tokens": 59077089.0,
      "step": 25760
    },
    {
      "entropy": 5.383675575256348,
      "epoch": 2.475024015369837,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043891276314891365,
      "loss": 4.7166,
      "mean_token_accuracy": 0.24366047531366347,
      "num_tokens": 59089188.0,
      "step": 25765
    },
    {
      "entropy": 5.466149616241455,
      "epoch": 2.4755043227665707,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004388892754831166,
      "loss": 4.7767,
      "mean_token_accuracy": 0.23934666365385054,
      "num_tokens": 59101384.0,
      "step": 25770
    },
    {
      "entropy": 5.421213579177857,
      "epoch": 2.4759846301633046,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004388657840122973,
      "loss": 4.7467,
      "mean_token_accuracy": 0.24034998714923858,
      "num_tokens": 59112327.0,
      "step": 25775
    },
    {
      "entropy": 5.409308815002442,
      "epoch": 2.4764649375600385,
      "grad_norm": 1.203125,
      "learning_rate": 0.000438842288737001,
      "loss": 4.7793,
      "mean_token_accuracy": 0.24119100123643875,
      "num_tokens": 59123577.0,
      "step": 25780
    },
    {
      "entropy": 5.345993518829346,
      "epoch": 2.4769452449567724,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043881878965777325,
      "loss": 4.6903,
      "mean_token_accuracy": 0.25301083624362947,
      "num_tokens": 59135359.0,
      "step": 25785
    },
    {
      "entropy": 5.355593872070313,
      "epoch": 2.4774255523535063,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043879528677515973,
      "loss": 4.6197,
      "mean_token_accuracy": 0.2531038269400597,
      "num_tokens": 59146679.0,
      "step": 25790
    },
    {
      "entropy": 5.373365688323974,
      "epoch": 2.47790585975024,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043877178008970596,
      "loss": 4.7713,
      "mean_token_accuracy": 0.24171495884656907,
      "num_tokens": 59159893.0,
      "step": 25795
    },
    {
      "entropy": 5.313538980484009,
      "epoch": 2.478386167146974,
      "grad_norm": 1.125,
      "learning_rate": 0.0004387482696019578,
      "loss": 4.6985,
      "mean_token_accuracy": 0.25267437398433684,
      "num_tokens": 59172592.0,
      "step": 25800
    },
    {
      "entropy": 5.3608804702758786,
      "epoch": 2.478866474543708,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043872475531246105,
      "loss": 4.7342,
      "mean_token_accuracy": 0.24810693562030792,
      "num_tokens": 59183650.0,
      "step": 25805
    },
    {
      "entropy": 5.418859529495239,
      "epoch": 2.479346781940442,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043870123722176166,
      "loss": 4.784,
      "mean_token_accuracy": 0.24691410660743712,
      "num_tokens": 59195423.0,
      "step": 25810
    },
    {
      "entropy": 5.383276319503784,
      "epoch": 2.4798270893371757,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004386777153304056,
      "loss": 4.7779,
      "mean_token_accuracy": 0.24038495272397994,
      "num_tokens": 59208486.0,
      "step": 25815
    },
    {
      "entropy": 5.344905757904053,
      "epoch": 2.4803073967339095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043865418963893896,
      "loss": 4.7321,
      "mean_token_accuracy": 0.24595056772232055,
      "num_tokens": 59220230.0,
      "step": 25820
    },
    {
      "entropy": 5.365811109542847,
      "epoch": 2.4807877041306434,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004386306601479081,
      "loss": 4.7685,
      "mean_token_accuracy": 0.24147001653909683,
      "num_tokens": 59231524.0,
      "step": 25825
    },
    {
      "entropy": 5.3919964790344235,
      "epoch": 2.4812680115273773,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004386071268578591,
      "loss": 4.8374,
      "mean_token_accuracy": 0.2363841250538826,
      "num_tokens": 59243525.0,
      "step": 25830
    },
    {
      "entropy": 5.400684261322022,
      "epoch": 2.4817483189241116,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043858358976933844,
      "loss": 4.6888,
      "mean_token_accuracy": 0.24802614003419876,
      "num_tokens": 59253774.0,
      "step": 25835
    },
    {
      "entropy": 5.355800628662109,
      "epoch": 2.4822286263208455,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043856004888289264,
      "loss": 4.7392,
      "mean_token_accuracy": 0.24552376121282576,
      "num_tokens": 59265640.0,
      "step": 25840
    },
    {
      "entropy": 5.3478264808654785,
      "epoch": 2.4827089337175794,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004385365041990681,
      "loss": 4.7943,
      "mean_token_accuracy": 0.24037551134824753,
      "num_tokens": 59277513.0,
      "step": 25845
    },
    {
      "entropy": 5.46808934211731,
      "epoch": 2.4831892411143133,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004385129557184116,
      "loss": 4.7672,
      "mean_token_accuracy": 0.2448406472802162,
      "num_tokens": 59289446.0,
      "step": 25850
    },
    {
      "entropy": 5.460447120666504,
      "epoch": 2.483669548511047,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043848940344146976,
      "loss": 4.7902,
      "mean_token_accuracy": 0.24794330596923828,
      "num_tokens": 59300156.0,
      "step": 25855
    },
    {
      "entropy": 5.343347120285034,
      "epoch": 2.484149855907781,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004384658473687894,
      "loss": 4.7339,
      "mean_token_accuracy": 0.24673579186201094,
      "num_tokens": 59311500.0,
      "step": 25860
    },
    {
      "entropy": 5.4473051071167,
      "epoch": 2.484630163304515,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004384422875009176,
      "loss": 4.829,
      "mean_token_accuracy": 0.2453504055738449,
      "num_tokens": 59323299.0,
      "step": 25865
    },
    {
      "entropy": 5.442589855194091,
      "epoch": 2.485110470701249,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004384187238384011,
      "loss": 4.798,
      "mean_token_accuracy": 0.24006806164979935,
      "num_tokens": 59335955.0,
      "step": 25870
    },
    {
      "entropy": 5.334350156784057,
      "epoch": 2.4855907780979827,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004383951563817871,
      "loss": 4.6534,
      "mean_token_accuracy": 0.24766553789377213,
      "num_tokens": 59347571.0,
      "step": 25875
    },
    {
      "entropy": 5.384089469909668,
      "epoch": 2.4860710854947166,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004383715851316227,
      "loss": 4.7337,
      "mean_token_accuracy": 0.24527962058782576,
      "num_tokens": 59358932.0,
      "step": 25880
    },
    {
      "entropy": 5.28055329322815,
      "epoch": 2.4865513928914504,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043834801008845527,
      "loss": 4.6586,
      "mean_token_accuracy": 0.25466011464595795,
      "num_tokens": 59371425.0,
      "step": 25885
    },
    {
      "entropy": 5.327711057662964,
      "epoch": 2.4870317002881843,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004383244312528321,
      "loss": 4.6912,
      "mean_token_accuracy": 0.24259742796421052,
      "num_tokens": 59382917.0,
      "step": 25890
    },
    {
      "entropy": 5.44249119758606,
      "epoch": 2.487512007684918,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004383008486253006,
      "loss": 4.8204,
      "mean_token_accuracy": 0.23373837769031525,
      "num_tokens": 59395014.0,
      "step": 25895
    },
    {
      "entropy": 5.40719952583313,
      "epoch": 2.487992315081652,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043827726220640827,
      "loss": 4.7902,
      "mean_token_accuracy": 0.24128009378910065,
      "num_tokens": 59407115.0,
      "step": 25900
    },
    {
      "entropy": 5.485486698150635,
      "epoch": 2.488472622478386,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043825367199670274,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2380875289440155,
      "num_tokens": 59418744.0,
      "step": 25905
    },
    {
      "entropy": 5.353435373306274,
      "epoch": 2.4889529298751203,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004382300779967318,
      "loss": 4.678,
      "mean_token_accuracy": 0.25460988134145734,
      "num_tokens": 59429950.0,
      "step": 25910
    },
    {
      "entropy": 5.424906158447266,
      "epoch": 2.489433237271854,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043820648020704303,
      "loss": 4.8637,
      "mean_token_accuracy": 0.2404331997036934,
      "num_tokens": 59441335.0,
      "step": 25915
    },
    {
      "entropy": 5.460147333145142,
      "epoch": 2.489913544668588,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043818287862818444,
      "loss": 4.7637,
      "mean_token_accuracy": 0.24254172146320344,
      "num_tokens": 59452325.0,
      "step": 25920
    },
    {
      "entropy": 5.468817853927613,
      "epoch": 2.490393852065322,
      "grad_norm": 0.953125,
      "learning_rate": 0.000438159273260704,
      "loss": 4.8488,
      "mean_token_accuracy": 0.24479147046804428,
      "num_tokens": 59464352.0,
      "step": 25925
    },
    {
      "entropy": 5.382438468933105,
      "epoch": 2.490874159462056,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004381356641051497,
      "loss": 4.7406,
      "mean_token_accuracy": 0.2466382712125778,
      "num_tokens": 59476048.0,
      "step": 25930
    },
    {
      "entropy": 5.4692668437957765,
      "epoch": 2.4913544668587897,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004381120511620697,
      "loss": 4.7975,
      "mean_token_accuracy": 0.23528916239738465,
      "num_tokens": 59488139.0,
      "step": 25935
    },
    {
      "entropy": 5.40494737625122,
      "epoch": 2.4918347742555236,
      "grad_norm": 1.125,
      "learning_rate": 0.00043808843443201217,
      "loss": 4.834,
      "mean_token_accuracy": 0.23749298304319383,
      "num_tokens": 59498279.0,
      "step": 25940
    },
    {
      "entropy": 5.403119802474976,
      "epoch": 2.4923150816522575,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004380648139155255,
      "loss": 4.75,
      "mean_token_accuracy": 0.24258187264204026,
      "num_tokens": 59509914.0,
      "step": 25945
    },
    {
      "entropy": 5.523049068450928,
      "epoch": 2.4927953890489913,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004380411896131581,
      "loss": 4.9119,
      "mean_token_accuracy": 0.2295467585325241,
      "num_tokens": 59521441.0,
      "step": 25950
    },
    {
      "entropy": 5.442595672607422,
      "epoch": 2.493275696445725,
      "grad_norm": 1.125,
      "learning_rate": 0.00043801756152545836,
      "loss": 4.8054,
      "mean_token_accuracy": 0.24178927689790725,
      "num_tokens": 59533086.0,
      "step": 25955
    },
    {
      "entropy": 5.410498476028442,
      "epoch": 2.493756003842459,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043799392965297496,
      "loss": 4.7929,
      "mean_token_accuracy": 0.23916533589363098,
      "num_tokens": 59545165.0,
      "step": 25960
    },
    {
      "entropy": 5.3977570056915285,
      "epoch": 2.494236311239193,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004379702939962564,
      "loss": 4.7834,
      "mean_token_accuracy": 0.2368497684597969,
      "num_tokens": 59556016.0,
      "step": 25965
    },
    {
      "entropy": 5.42678599357605,
      "epoch": 2.494716618635927,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004379466545558516,
      "loss": 4.7717,
      "mean_token_accuracy": 0.24144032895565032,
      "num_tokens": 59566431.0,
      "step": 25970
    },
    {
      "entropy": 5.4518684387207035,
      "epoch": 2.4951969260326607,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043792301133230933,
      "loss": 4.8149,
      "mean_token_accuracy": 0.23704971075057985,
      "num_tokens": 59579371.0,
      "step": 25975
    },
    {
      "entropy": 5.391521406173706,
      "epoch": 2.4956772334293946,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004378993643261785,
      "loss": 4.7398,
      "mean_token_accuracy": 0.24513784795999527,
      "num_tokens": 59590088.0,
      "step": 25980
    },
    {
      "entropy": 5.347687816619873,
      "epoch": 2.496157540826129,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043787571353800814,
      "loss": 4.686,
      "mean_token_accuracy": 0.2568423166871071,
      "num_tokens": 59600973.0,
      "step": 25985
    },
    {
      "entropy": 5.428435516357422,
      "epoch": 2.4966378482228624,
      "grad_norm": 1.0,
      "learning_rate": 0.0004378520589683475,
      "loss": 4.7508,
      "mean_token_accuracy": 0.23972638249397277,
      "num_tokens": 59613194.0,
      "step": 25990
    },
    {
      "entropy": 5.428511047363282,
      "epoch": 2.4971181556195967,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043782840061774544,
      "loss": 4.7469,
      "mean_token_accuracy": 0.23990225046873093,
      "num_tokens": 59623932.0,
      "step": 25995
    },
    {
      "entropy": 5.406115579605102,
      "epoch": 2.4975984630163306,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043780473848675143,
      "loss": 4.8181,
      "mean_token_accuracy": 0.23697038143873214,
      "num_tokens": 59635095.0,
      "step": 26000
    },
    {
      "entropy": 5.418171548843384,
      "epoch": 2.4980787704130645,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004377810725759149,
      "loss": 4.7546,
      "mean_token_accuracy": 0.24139282256364822,
      "num_tokens": 59646930.0,
      "step": 26005
    },
    {
      "entropy": 5.404428386688233,
      "epoch": 2.4985590778097984,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00043775740288578516,
      "loss": 4.7143,
      "mean_token_accuracy": 0.24402257800102234,
      "num_tokens": 59658729.0,
      "step": 26010
    },
    {
      "entropy": 5.3774933338165285,
      "epoch": 2.4990393852065322,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004377337294169118,
      "loss": 4.7922,
      "mean_token_accuracy": 0.23786480128765106,
      "num_tokens": 59670619.0,
      "step": 26015
    },
    {
      "entropy": 5.453701686859131,
      "epoch": 2.499519692603266,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043771005216984457,
      "loss": 4.8709,
      "mean_token_accuracy": 0.2352215752005577,
      "num_tokens": 59682307.0,
      "step": 26020
    },
    {
      "entropy": 5.406220531463623,
      "epoch": 2.5,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000437686371145133,
      "loss": 4.7367,
      "mean_token_accuracy": 0.241387240588665,
      "num_tokens": 59693515.0,
      "step": 26025
    },
    {
      "entropy": 5.4120158672332765,
      "epoch": 2.500480307396734,
      "grad_norm": 1.046875,
      "learning_rate": 0.000437662686343327,
      "loss": 4.7925,
      "mean_token_accuracy": 0.23513654619455338,
      "num_tokens": 59705012.0,
      "step": 26030
    },
    {
      "entropy": 5.483615350723267,
      "epoch": 2.5009606147934678,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004376389977649764,
      "loss": 4.8617,
      "mean_token_accuracy": 0.239364455640316,
      "num_tokens": 59716776.0,
      "step": 26035
    },
    {
      "entropy": 5.382342100143433,
      "epoch": 2.5014409221902016,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004376153054106313,
      "loss": 4.7552,
      "mean_token_accuracy": 0.24385649263858794,
      "num_tokens": 59727526.0,
      "step": 26040
    },
    {
      "entropy": 5.384013366699219,
      "epoch": 2.5019212295869355,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004375916092808416,
      "loss": 4.7343,
      "mean_token_accuracy": 0.24766200929880142,
      "num_tokens": 59738861.0,
      "step": 26045
    },
    {
      "entropy": 5.528242588043213,
      "epoch": 2.5024015369836694,
      "grad_norm": 1.125,
      "learning_rate": 0.0004375679093761575,
      "loss": 4.9438,
      "mean_token_accuracy": 0.22891802191734315,
      "num_tokens": 59751019.0,
      "step": 26050
    },
    {
      "entropy": 5.426020336151123,
      "epoch": 2.5028818443804033,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043754420569712925,
      "loss": 4.7631,
      "mean_token_accuracy": 0.2513150379061699,
      "num_tokens": 59762936.0,
      "step": 26055
    },
    {
      "entropy": 5.413517904281616,
      "epoch": 2.5033621517771376,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043752049824430736,
      "loss": 4.7205,
      "mean_token_accuracy": 0.2434921994805336,
      "num_tokens": 59772909.0,
      "step": 26060
    },
    {
      "entropy": 5.290168619155883,
      "epoch": 2.503842459173871,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00043749678701824197,
      "loss": 4.6959,
      "mean_token_accuracy": 0.24723115861415862,
      "num_tokens": 59784003.0,
      "step": 26065
    },
    {
      "entropy": 5.3953193664550785,
      "epoch": 2.5043227665706054,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004374730720194837,
      "loss": 4.8033,
      "mean_token_accuracy": 0.2419385090470314,
      "num_tokens": 59796200.0,
      "step": 26070
    },
    {
      "entropy": 5.431125211715698,
      "epoch": 2.5048030739673393,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004374493532485832,
      "loss": 4.743,
      "mean_token_accuracy": 0.24489697962999343,
      "num_tokens": 59807559.0,
      "step": 26075
    },
    {
      "entropy": 5.40878210067749,
      "epoch": 2.505283381364073,
      "grad_norm": 1.109375,
      "learning_rate": 0.000437425630706091,
      "loss": 4.7944,
      "mean_token_accuracy": 0.23980665504932402,
      "num_tokens": 59818213.0,
      "step": 26080
    },
    {
      "entropy": 5.38883695602417,
      "epoch": 2.505763688760807,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000437401904392558,
      "loss": 4.7737,
      "mean_token_accuracy": 0.24356097728013992,
      "num_tokens": 59829485.0,
      "step": 26085
    },
    {
      "entropy": 5.3506917476654055,
      "epoch": 2.506243996157541,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043737817430853504,
      "loss": 4.8278,
      "mean_token_accuracy": 0.24365534037351608,
      "num_tokens": 59842187.0,
      "step": 26090
    },
    {
      "entropy": 5.481291437149048,
      "epoch": 2.506724303554275,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043735444045457303,
      "loss": 4.8936,
      "mean_token_accuracy": 0.23358812779188157,
      "num_tokens": 59852783.0,
      "step": 26095
    },
    {
      "entropy": 5.462952089309693,
      "epoch": 2.5072046109510087,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043733070283122306,
      "loss": 4.7292,
      "mean_token_accuracy": 0.24072492569684983,
      "num_tokens": 59863661.0,
      "step": 26100
    },
    {
      "entropy": 5.352629661560059,
      "epoch": 2.5076849183477425,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043730696143903607,
      "loss": 4.7172,
      "mean_token_accuracy": 0.2486035332083702,
      "num_tokens": 59876109.0,
      "step": 26105
    },
    {
      "entropy": 5.302836608886719,
      "epoch": 2.5081652257444764,
      "grad_norm": 1.0,
      "learning_rate": 0.0004372832162785635,
      "loss": 4.5947,
      "mean_token_accuracy": 0.25356592833995817,
      "num_tokens": 59886322.0,
      "step": 26110
    },
    {
      "entropy": 5.330037641525268,
      "epoch": 2.5086455331412103,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004372594673503565,
      "loss": 4.6962,
      "mean_token_accuracy": 0.24216514825820923,
      "num_tokens": 59897307.0,
      "step": 26115
    },
    {
      "entropy": 5.345793151855469,
      "epoch": 2.509125840537944,
      "grad_norm": 1.125,
      "learning_rate": 0.0004372357146549664,
      "loss": 4.6557,
      "mean_token_accuracy": 0.24758805483579635,
      "num_tokens": 59908709.0,
      "step": 26120
    },
    {
      "entropy": 5.363976049423218,
      "epoch": 2.509606147934678,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043721195819294487,
      "loss": 4.6992,
      "mean_token_accuracy": 0.2530563846230507,
      "num_tokens": 59920082.0,
      "step": 26125
    },
    {
      "entropy": 5.3595599174499515,
      "epoch": 2.510086455331412,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004371881979648433,
      "loss": 4.7306,
      "mean_token_accuracy": 0.2431192234158516,
      "num_tokens": 59931632.0,
      "step": 26130
    },
    {
      "entropy": 5.311651611328125,
      "epoch": 2.5105667627281463,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004371644339712133,
      "loss": 4.6959,
      "mean_token_accuracy": 0.24440656006336212,
      "num_tokens": 59944093.0,
      "step": 26135
    },
    {
      "entropy": 5.386670541763306,
      "epoch": 2.5110470701248797,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004371406662126067,
      "loss": 4.7476,
      "mean_token_accuracy": 0.23917070478200914,
      "num_tokens": 59954879.0,
      "step": 26140
    },
    {
      "entropy": 5.374618053436279,
      "epoch": 2.511527377521614,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043711689468957534,
      "loss": 4.745,
      "mean_token_accuracy": 0.24223673343658447,
      "num_tokens": 59967074.0,
      "step": 26145
    },
    {
      "entropy": 5.46340651512146,
      "epoch": 2.5120076849183475,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043709311940267107,
      "loss": 4.7825,
      "mean_token_accuracy": 0.24325994700193404,
      "num_tokens": 59978939.0,
      "step": 26150
    },
    {
      "entropy": 5.455024003982544,
      "epoch": 2.512487992315082,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004370693403524458,
      "loss": 4.7609,
      "mean_token_accuracy": 0.24458985030651093,
      "num_tokens": 59989251.0,
      "step": 26155
    },
    {
      "entropy": 5.326098442077637,
      "epoch": 2.5129682997118157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004370455575394518,
      "loss": 4.7319,
      "mean_token_accuracy": 0.24381151795387268,
      "num_tokens": 60000267.0,
      "step": 26160
    },
    {
      "entropy": 5.445640420913696,
      "epoch": 2.5134486071085496,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004370217709642411,
      "loss": 4.7714,
      "mean_token_accuracy": 0.23925601541996003,
      "num_tokens": 60011190.0,
      "step": 26165
    },
    {
      "entropy": 5.411395454406739,
      "epoch": 2.5139289145052834,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000436997980627366,
      "loss": 4.8399,
      "mean_token_accuracy": 0.23552492558956145,
      "num_tokens": 60023496.0,
      "step": 26170
    },
    {
      "entropy": 5.4673620700836185,
      "epoch": 2.5144092219020173,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043697418652937877,
      "loss": 4.8346,
      "mean_token_accuracy": 0.23520189821720122,
      "num_tokens": 60034585.0,
      "step": 26175
    },
    {
      "entropy": 5.473264455795288,
      "epoch": 2.514889529298751,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004369503886708319,
      "loss": 4.8162,
      "mean_token_accuracy": 0.23601806908845901,
      "num_tokens": 60046333.0,
      "step": 26180
    },
    {
      "entropy": 5.3879374980926515,
      "epoch": 2.515369836695485,
      "grad_norm": 1.125,
      "learning_rate": 0.00043692658705227796,
      "loss": 4.8089,
      "mean_token_accuracy": 0.23890177756547928,
      "num_tokens": 60058744.0,
      "step": 26185
    },
    {
      "entropy": 5.50883846282959,
      "epoch": 2.515850144092219,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043690278167426945,
      "loss": 4.8643,
      "mean_token_accuracy": 0.23429105132818223,
      "num_tokens": 60069303.0,
      "step": 26190
    },
    {
      "entropy": 5.363683414459229,
      "epoch": 2.516330451488953,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004368789725373591,
      "loss": 4.7083,
      "mean_token_accuracy": 0.24661850184202194,
      "num_tokens": 60080604.0,
      "step": 26195
    },
    {
      "entropy": 5.390801572799683,
      "epoch": 2.5168107588856867,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043685515964209977,
      "loss": 4.7574,
      "mean_token_accuracy": 0.24561378061771394,
      "num_tokens": 60090151.0,
      "step": 26200
    },
    {
      "entropy": 5.436556148529053,
      "epoch": 2.5172910662824206,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004368313429890441,
      "loss": 4.8463,
      "mean_token_accuracy": 0.23860891163349152,
      "num_tokens": 60102528.0,
      "step": 26205
    },
    {
      "entropy": 5.352163600921631,
      "epoch": 2.517771373679155,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004368075225787453,
      "loss": 4.696,
      "mean_token_accuracy": 0.25002682358026507,
      "num_tokens": 60113078.0,
      "step": 26210
    },
    {
      "entropy": 5.308748817443847,
      "epoch": 2.5182516810758884,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004367836984117562,
      "loss": 4.6948,
      "mean_token_accuracy": 0.24448444247245787,
      "num_tokens": 60124728.0,
      "step": 26215
    },
    {
      "entropy": 5.286151361465454,
      "epoch": 2.5187319884726227,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043675987048863,
      "loss": 4.6746,
      "mean_token_accuracy": 0.24954676181077956,
      "num_tokens": 60136224.0,
      "step": 26220
    },
    {
      "entropy": 5.432214403152466,
      "epoch": 2.519212295869356,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004367360388099201,
      "loss": 4.7893,
      "mean_token_accuracy": 0.24063422977924348,
      "num_tokens": 60148654.0,
      "step": 26225
    },
    {
      "entropy": 5.476292896270752,
      "epoch": 2.5196926032660905,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004367122033761796,
      "loss": 4.795,
      "mean_token_accuracy": 0.2362612470984459,
      "num_tokens": 60159740.0,
      "step": 26230
    },
    {
      "entropy": 5.397809886932373,
      "epoch": 2.5201729106628243,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004366883641879618,
      "loss": 4.7952,
      "mean_token_accuracy": 0.23468934148550033,
      "num_tokens": 60171387.0,
      "step": 26235
    },
    {
      "entropy": 5.345549249649048,
      "epoch": 2.520653218059558,
      "grad_norm": 1.125,
      "learning_rate": 0.00043666452124582034,
      "loss": 4.7138,
      "mean_token_accuracy": 0.25029412657022476,
      "num_tokens": 60181562.0,
      "step": 26240
    },
    {
      "entropy": 5.307652187347412,
      "epoch": 2.521133525456292,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004366406745503088,
      "loss": 4.666,
      "mean_token_accuracy": 0.24811489284038543,
      "num_tokens": 60191187.0,
      "step": 26245
    },
    {
      "entropy": 5.27403998374939,
      "epoch": 2.521613832853026,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004366168241019807,
      "loss": 4.6969,
      "mean_token_accuracy": 0.24928056001663207,
      "num_tokens": 60203792.0,
      "step": 26250
    },
    {
      "entropy": 5.476580429077148,
      "epoch": 2.52209414024976,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004365929699013899,
      "loss": 4.764,
      "mean_token_accuracy": 0.24602048397064208,
      "num_tokens": 60215523.0,
      "step": 26255
    },
    {
      "entropy": 5.451271867752075,
      "epoch": 2.5225744476464937,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004365691119490902,
      "loss": 4.8065,
      "mean_token_accuracy": 0.24529763162136078,
      "num_tokens": 60226113.0,
      "step": 26260
    },
    {
      "entropy": 5.477841997146607,
      "epoch": 2.5230547550432276,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004365452502456354,
      "loss": 4.8748,
      "mean_token_accuracy": 0.23135137557983398,
      "num_tokens": 60238521.0,
      "step": 26265
    },
    {
      "entropy": 5.310695695877075,
      "epoch": 2.5235350624399615,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004365213847915796,
      "loss": 4.6512,
      "mean_token_accuracy": 0.25126627534627916,
      "num_tokens": 60249717.0,
      "step": 26270
    },
    {
      "entropy": 5.428727531433106,
      "epoch": 2.5240153698366954,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043649751558747695,
      "loss": 4.8557,
      "mean_token_accuracy": 0.23051680624485016,
      "num_tokens": 60261749.0,
      "step": 26275
    },
    {
      "entropy": 5.45498480796814,
      "epoch": 2.5244956772334293,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043647364263388143,
      "loss": 4.8317,
      "mean_token_accuracy": 0.23963929563760758,
      "num_tokens": 60272925.0,
      "step": 26280
    },
    {
      "entropy": 5.459753465652466,
      "epoch": 2.524975984630163,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004364497659313475,
      "loss": 4.7823,
      "mean_token_accuracy": 0.24077234268188477,
      "num_tokens": 60285081.0,
      "step": 26285
    },
    {
      "entropy": 5.399181747436524,
      "epoch": 2.525456292026897,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004364258854804294,
      "loss": 4.7745,
      "mean_token_accuracy": 0.24604720622301102,
      "num_tokens": 60296796.0,
      "step": 26290
    },
    {
      "entropy": 5.4235923290252686,
      "epoch": 2.5259365994236314,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004364020012816815,
      "loss": 4.8569,
      "mean_token_accuracy": 0.23703459203243255,
      "num_tokens": 60308600.0,
      "step": 26295
    },
    {
      "entropy": 5.405559110641479,
      "epoch": 2.526416906820365,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004363781133356584,
      "loss": 4.7136,
      "mean_token_accuracy": 0.2434051975607872,
      "num_tokens": 60319188.0,
      "step": 26300
    },
    {
      "entropy": 5.448766374588013,
      "epoch": 2.526897214217099,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004363542216429147,
      "loss": 4.815,
      "mean_token_accuracy": 0.2368224412202835,
      "num_tokens": 60329912.0,
      "step": 26305
    },
    {
      "entropy": 5.427968883514405,
      "epoch": 2.527377521613833,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004363303262040051,
      "loss": 4.7848,
      "mean_token_accuracy": 0.24137367457151412,
      "num_tokens": 60339760.0,
      "step": 26310
    },
    {
      "entropy": 5.293443202972412,
      "epoch": 2.527857829010567,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043630642701948446,
      "loss": 4.6374,
      "mean_token_accuracy": 0.25378872603178027,
      "num_tokens": 60349947.0,
      "step": 26315
    },
    {
      "entropy": 5.393341588973999,
      "epoch": 2.5283381364073008,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043628252408990756,
      "loss": 4.8311,
      "mean_token_accuracy": 0.23416674733161927,
      "num_tokens": 60362620.0,
      "step": 26320
    },
    {
      "entropy": 5.472264623641967,
      "epoch": 2.5288184438040346,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043625861741582926,
      "loss": 4.8057,
      "mean_token_accuracy": 0.24401302635669708,
      "num_tokens": 60373729.0,
      "step": 26325
    },
    {
      "entropy": 5.449048042297363,
      "epoch": 2.5292987512007685,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043623470699780483,
      "loss": 4.7454,
      "mean_token_accuracy": 0.24043226838111878,
      "num_tokens": 60386423.0,
      "step": 26330
    },
    {
      "entropy": 5.404461193084717,
      "epoch": 2.5297790585975024,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004362107928363892,
      "loss": 4.7818,
      "mean_token_accuracy": 0.24169419705867767,
      "num_tokens": 60398364.0,
      "step": 26335
    },
    {
      "entropy": 5.405173397064209,
      "epoch": 2.5302593659942363,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004361868749321377,
      "loss": 4.8064,
      "mean_token_accuracy": 0.2403477042913437,
      "num_tokens": 60408423.0,
      "step": 26340
    },
    {
      "entropy": 5.447754716873169,
      "epoch": 2.53073967339097,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004361629532856055,
      "loss": 4.8307,
      "mean_token_accuracy": 0.23803750872612,
      "num_tokens": 60420229.0,
      "step": 26345
    },
    {
      "entropy": 5.510746860504151,
      "epoch": 2.531219980787704,
      "grad_norm": 1.125,
      "learning_rate": 0.00043613902789734816,
      "loss": 4.852,
      "mean_token_accuracy": 0.24019477069377898,
      "num_tokens": 60430711.0,
      "step": 26350
    },
    {
      "entropy": 5.466444301605224,
      "epoch": 2.531700288184438,
      "grad_norm": 1.09375,
      "learning_rate": 0.000436115098767921,
      "loss": 4.8151,
      "mean_token_accuracy": 0.23854459673166276,
      "num_tokens": 60442293.0,
      "step": 26355
    },
    {
      "entropy": 5.356484842300415,
      "epoch": 2.532180595581172,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043609116589787974,
      "loss": 4.7391,
      "mean_token_accuracy": 0.2467636615037918,
      "num_tokens": 60452616.0,
      "step": 26360
    },
    {
      "entropy": 5.404138851165771,
      "epoch": 2.5326609029779057,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004360672292877799,
      "loss": 4.7309,
      "mean_token_accuracy": 0.24109349995851517,
      "num_tokens": 60463254.0,
      "step": 26365
    },
    {
      "entropy": 5.430644559860229,
      "epoch": 2.53314121037464,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043604328893817726,
      "loss": 4.7598,
      "mean_token_accuracy": 0.24327847957611085,
      "num_tokens": 60473725.0,
      "step": 26370
    },
    {
      "entropy": 5.41340274810791,
      "epoch": 2.5336215177713735,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043601934484962775,
      "loss": 4.7789,
      "mean_token_accuracy": 0.24553025811910628,
      "num_tokens": 60484502.0,
      "step": 26375
    },
    {
      "entropy": 5.319082975387573,
      "epoch": 2.534101825168108,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004359953970226871,
      "loss": 4.7285,
      "mean_token_accuracy": 0.2484838545322418,
      "num_tokens": 60496134.0,
      "step": 26380
    },
    {
      "entropy": 5.436810159683228,
      "epoch": 2.5345821325648417,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043597144545791134,
      "loss": 4.7973,
      "mean_token_accuracy": 0.2448647364974022,
      "num_tokens": 60507236.0,
      "step": 26385
    },
    {
      "entropy": 5.373466682434082,
      "epoch": 2.5350624399615755,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004359474901558567,
      "loss": 4.7443,
      "mean_token_accuracy": 0.24590823501348497,
      "num_tokens": 60517499.0,
      "step": 26390
    },
    {
      "entropy": 5.403765344619751,
      "epoch": 2.5355427473583094,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004359235311170792,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23556326776742936,
      "num_tokens": 60529434.0,
      "step": 26395
    },
    {
      "entropy": 5.427843332290649,
      "epoch": 2.5360230547550433,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004358995683421352,
      "loss": 4.8068,
      "mean_token_accuracy": 0.24145500361919403,
      "num_tokens": 60542571.0,
      "step": 26400
    },
    {
      "entropy": 5.438608932495117,
      "epoch": 2.536503362151777,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043587560183158095,
      "loss": 4.7307,
      "mean_token_accuracy": 0.23853669911623002,
      "num_tokens": 60553826.0,
      "step": 26405
    },
    {
      "entropy": 5.360276222229004,
      "epoch": 2.536983669548511,
      "grad_norm": 1.09375,
      "learning_rate": 0.000435851631585973,
      "loss": 4.7483,
      "mean_token_accuracy": 0.23980994373559952,
      "num_tokens": 60565173.0,
      "step": 26410
    },
    {
      "entropy": 5.348510026931763,
      "epoch": 2.537463976945245,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004358276576058677,
      "loss": 4.7289,
      "mean_token_accuracy": 0.2515919581055641,
      "num_tokens": 60577267.0,
      "step": 26415
    },
    {
      "entropy": 5.396236419677734,
      "epoch": 2.537944284341979,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004358036798918218,
      "loss": 4.7798,
      "mean_token_accuracy": 0.23711578398942948,
      "num_tokens": 60587441.0,
      "step": 26420
    },
    {
      "entropy": 5.490083789825439,
      "epoch": 2.5384245917387127,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004357796984443919,
      "loss": 4.8166,
      "mean_token_accuracy": 0.24640152603387833,
      "num_tokens": 60597173.0,
      "step": 26425
    },
    {
      "entropy": 5.349520444869995,
      "epoch": 2.5389048991354466,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043575571326413484,
      "loss": 4.6979,
      "mean_token_accuracy": 0.2411357954144478,
      "num_tokens": 60608299.0,
      "step": 26430
    },
    {
      "entropy": 5.482545661926269,
      "epoch": 2.5393852065321805,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004357317243516075,
      "loss": 4.8762,
      "mean_token_accuracy": 0.23880002945661544,
      "num_tokens": 60619106.0,
      "step": 26435
    },
    {
      "entropy": 5.383048725128174,
      "epoch": 2.5398655139289144,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043570773170736676,
      "loss": 4.6959,
      "mean_token_accuracy": 0.24706820845603944,
      "num_tokens": 60629339.0,
      "step": 26440
    },
    {
      "entropy": 5.4033629417419435,
      "epoch": 2.5403458213256487,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043568373533196976,
      "loss": 4.7123,
      "mean_token_accuracy": 0.2532023414969444,
      "num_tokens": 60640162.0,
      "step": 26445
    },
    {
      "entropy": 5.34549765586853,
      "epoch": 2.540826128722382,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043565973522597344,
      "loss": 4.7267,
      "mean_token_accuracy": 0.24083796441555022,
      "num_tokens": 60652529.0,
      "step": 26450
    },
    {
      "entropy": 5.3181699275970455,
      "epoch": 2.5413064361191164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043563573138993524,
      "loss": 4.6896,
      "mean_token_accuracy": 0.24639348834753036,
      "num_tokens": 60663035.0,
      "step": 26455
    },
    {
      "entropy": 5.364248371124267,
      "epoch": 2.54178674351585,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004356117238244123,
      "loss": 4.6969,
      "mean_token_accuracy": 0.2492280811071396,
      "num_tokens": 60674430.0,
      "step": 26460
    },
    {
      "entropy": 5.352776336669922,
      "epoch": 2.542267050912584,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043558771252996204,
      "loss": 4.7013,
      "mean_token_accuracy": 0.249556565284729,
      "num_tokens": 60685923.0,
      "step": 26465
    },
    {
      "entropy": 5.354985904693604,
      "epoch": 2.542747358309318,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000435563697507142,
      "loss": 4.7606,
      "mean_token_accuracy": 0.24579499959945678,
      "num_tokens": 60696344.0,
      "step": 26470
    },
    {
      "entropy": 5.390136289596557,
      "epoch": 2.543227665706052,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004355396787565096,
      "loss": 4.7772,
      "mean_token_accuracy": 0.24470880776643752,
      "num_tokens": 60709941.0,
      "step": 26475
    },
    {
      "entropy": 5.421936655044556,
      "epoch": 2.543707973102786,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043551565627862257,
      "loss": 4.7424,
      "mean_token_accuracy": 0.24229725301265717,
      "num_tokens": 60721245.0,
      "step": 26480
    },
    {
      "entropy": 5.403618621826172,
      "epoch": 2.5441882804995197,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004354916300740387,
      "loss": 4.7312,
      "mean_token_accuracy": 0.24174535125494004,
      "num_tokens": 60732763.0,
      "step": 26485
    },
    {
      "entropy": 5.3577147960662845,
      "epoch": 2.5446685878962536,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004354676001433157,
      "loss": 4.7198,
      "mean_token_accuracy": 0.24710991978645325,
      "num_tokens": 60743192.0,
      "step": 26490
    },
    {
      "entropy": 5.298436069488526,
      "epoch": 2.5451488952929875,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004354435664870116,
      "loss": 4.6459,
      "mean_token_accuracy": 0.24920875281095506,
      "num_tokens": 60754115.0,
      "step": 26495
    },
    {
      "entropy": 5.475417613983154,
      "epoch": 2.5456292026897214,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043541952910568417,
      "loss": 4.7812,
      "mean_token_accuracy": 0.24674731343984604,
      "num_tokens": 60765516.0,
      "step": 26500
    },
    {
      "entropy": 5.441589593887329,
      "epoch": 2.5461095100864553,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004353954879998916,
      "loss": 4.7595,
      "mean_token_accuracy": 0.24230417013168334,
      "num_tokens": 60776206.0,
      "step": 26505
    },
    {
      "entropy": 5.359499216079712,
      "epoch": 2.546589817483189,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004353714431701922,
      "loss": 4.6969,
      "mean_token_accuracy": 0.2443567007780075,
      "num_tokens": 60786992.0,
      "step": 26510
    },
    {
      "entropy": 5.362868356704712,
      "epoch": 2.547070124879923,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004353473946171441,
      "loss": 4.7732,
      "mean_token_accuracy": 0.23602970838546752,
      "num_tokens": 60797765.0,
      "step": 26515
    },
    {
      "entropy": 5.424197006225586,
      "epoch": 2.547550432276657,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043532334234130547,
      "loss": 4.799,
      "mean_token_accuracy": 0.23326411098241806,
      "num_tokens": 60808873.0,
      "step": 26520
    },
    {
      "entropy": 5.4764244556427,
      "epoch": 2.5480307396733908,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043529928634323503,
      "loss": 4.8104,
      "mean_token_accuracy": 0.24119106978178023,
      "num_tokens": 60820245.0,
      "step": 26525
    },
    {
      "entropy": 5.378897047042846,
      "epoch": 2.548511047070125,
      "grad_norm": 1.125,
      "learning_rate": 0.00043527522662349113,
      "loss": 4.7782,
      "mean_token_accuracy": 0.24554303288459778,
      "num_tokens": 60832159.0,
      "step": 26530
    },
    {
      "entropy": 5.367926597595215,
      "epoch": 2.5489913544668585,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004352511631826324,
      "loss": 4.7359,
      "mean_token_accuracy": 0.24373805075883864,
      "num_tokens": 60842957.0,
      "step": 26535
    },
    {
      "entropy": 5.366412687301636,
      "epoch": 2.549471661863593,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004352270960212175,
      "loss": 4.7374,
      "mean_token_accuracy": 0.24511200487613677,
      "num_tokens": 60852537.0,
      "step": 26540
    },
    {
      "entropy": 5.477486896514892,
      "epoch": 2.5499519692603267,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004352030251398052,
      "loss": 4.8761,
      "mean_token_accuracy": 0.23133329898118973,
      "num_tokens": 60865761.0,
      "step": 26545
    },
    {
      "entropy": 5.435649967193603,
      "epoch": 2.5504322766570606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043517895053895434,
      "loss": 4.7468,
      "mean_token_accuracy": 0.2497814938426018,
      "num_tokens": 60877805.0,
      "step": 26550
    },
    {
      "entropy": 5.4090588092803955,
      "epoch": 2.5509125840537945,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004351548722192239,
      "loss": 4.7779,
      "mean_token_accuracy": 0.24240620732307433,
      "num_tokens": 60889868.0,
      "step": 26555
    },
    {
      "entropy": 5.372077178955078,
      "epoch": 2.5513928914505284,
      "grad_norm": 1.125,
      "learning_rate": 0.0004351307901811729,
      "loss": 4.7504,
      "mean_token_accuracy": 0.2507741376757622,
      "num_tokens": 60902046.0,
      "step": 26560
    },
    {
      "entropy": 5.397452545166016,
      "epoch": 2.5518731988472623,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004351067044253605,
      "loss": 4.7672,
      "mean_token_accuracy": 0.23708975613117217,
      "num_tokens": 60913447.0,
      "step": 26565
    },
    {
      "entropy": 5.408935594558716,
      "epoch": 2.552353506243996,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043508261495234577,
      "loss": 4.7552,
      "mean_token_accuracy": 0.25014561116695405,
      "num_tokens": 60924920.0,
      "step": 26570
    },
    {
      "entropy": 5.339941120147705,
      "epoch": 2.55283381364073,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000435058521762688,
      "loss": 4.6898,
      "mean_token_accuracy": 0.2526871860027313,
      "num_tokens": 60936937.0,
      "step": 26575
    },
    {
      "entropy": 5.331101799011231,
      "epoch": 2.553314121037464,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004350344248569467,
      "loss": 4.7083,
      "mean_token_accuracy": 0.2554743319749832,
      "num_tokens": 60947836.0,
      "step": 26580
    },
    {
      "entropy": 5.35800518989563,
      "epoch": 2.553794428434198,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004350103242356813,
      "loss": 4.8074,
      "mean_token_accuracy": 0.24301904290914536,
      "num_tokens": 60959380.0,
      "step": 26585
    },
    {
      "entropy": 5.455740356445313,
      "epoch": 2.5542747358309317,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004349862198994513,
      "loss": 4.7919,
      "mean_token_accuracy": 0.24020466655492784,
      "num_tokens": 60970148.0,
      "step": 26590
    },
    {
      "entropy": 5.415919065475464,
      "epoch": 2.5547550432276656,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004349621118488163,
      "loss": 4.7207,
      "mean_token_accuracy": 0.25130564868450167,
      "num_tokens": 60981139.0,
      "step": 26595
    },
    {
      "entropy": 5.338467979431153,
      "epoch": 2.5552353506243994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004349380000843361,
      "loss": 4.7192,
      "mean_token_accuracy": 0.2441763922572136,
      "num_tokens": 60992204.0,
      "step": 26600
    },
    {
      "entropy": 5.340367412567138,
      "epoch": 2.5557156580211338,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004349138846065704,
      "loss": 4.7294,
      "mean_token_accuracy": 0.2442852884531021,
      "num_tokens": 61002198.0,
      "step": 26605
    },
    {
      "entropy": 5.461579084396362,
      "epoch": 2.556195965417867,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004348897654160791,
      "loss": 4.8322,
      "mean_token_accuracy": 0.23854032307863235,
      "num_tokens": 61012987.0,
      "step": 26610
    },
    {
      "entropy": 5.376810693740845,
      "epoch": 2.5566762728146015,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004348656425134223,
      "loss": 4.6202,
      "mean_token_accuracy": 0.25457109659910204,
      "num_tokens": 61023668.0,
      "step": 26615
    },
    {
      "entropy": 5.438167476654053,
      "epoch": 2.5571565802113354,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043484151589916,
      "loss": 4.6665,
      "mean_token_accuracy": 0.2539594069123268,
      "num_tokens": 61034375.0,
      "step": 26620
    },
    {
      "entropy": 5.394752168655396,
      "epoch": 2.5576368876080693,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004348173855738523,
      "loss": 4.7659,
      "mean_token_accuracy": 0.23951872289180756,
      "num_tokens": 61045625.0,
      "step": 26625
    },
    {
      "entropy": 5.3566123962402346,
      "epoch": 2.558117195004803,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004347932515380595,
      "loss": 4.7952,
      "mean_token_accuracy": 0.24655058830976487,
      "num_tokens": 61056359.0,
      "step": 26630
    },
    {
      "entropy": 5.425376749038696,
      "epoch": 2.558597502401537,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004347691137923418,
      "loss": 4.7278,
      "mean_token_accuracy": 0.2470552310347557,
      "num_tokens": 61067194.0,
      "step": 26635
    },
    {
      "entropy": 5.4770286083221436,
      "epoch": 2.559077809798271,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004347449723372598,
      "loss": 4.8243,
      "mean_token_accuracy": 0.23305363357067108,
      "num_tokens": 61079341.0,
      "step": 26640
    },
    {
      "entropy": 5.425379419326783,
      "epoch": 2.559558117195005,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004347208271733738,
      "loss": 4.8085,
      "mean_token_accuracy": 0.2430064395070076,
      "num_tokens": 61090843.0,
      "step": 26645
    },
    {
      "entropy": 5.368398714065552,
      "epoch": 2.5600384245917387,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004346966783012445,
      "loss": 4.7249,
      "mean_token_accuracy": 0.245524038374424,
      "num_tokens": 61101780.0,
      "step": 26650
    },
    {
      "entropy": 5.404880046844482,
      "epoch": 2.5605187319884726,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043467252572143247,
      "loss": 4.7941,
      "mean_token_accuracy": 0.2466665968298912,
      "num_tokens": 61113026.0,
      "step": 26655
    },
    {
      "entropy": 5.440839338302612,
      "epoch": 2.5609990393852065,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00043464836943449866,
      "loss": 4.7664,
      "mean_token_accuracy": 0.24620601087808608,
      "num_tokens": 61124772.0,
      "step": 26660
    },
    {
      "entropy": 5.4018789768219,
      "epoch": 2.5614793467819403,
      "grad_norm": 1.125,
      "learning_rate": 0.0004346242094410036,
      "loss": 4.7725,
      "mean_token_accuracy": 0.2440594419836998,
      "num_tokens": 61136683.0,
      "step": 26665
    },
    {
      "entropy": 5.359942483901977,
      "epoch": 2.561959654178674,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004346000457415085,
      "loss": 4.7185,
      "mean_token_accuracy": 0.248667673766613,
      "num_tokens": 61148047.0,
      "step": 26670
    },
    {
      "entropy": 5.364991998672485,
      "epoch": 2.562439961575408,
      "grad_norm": 0.953125,
      "learning_rate": 0.00043457587833657424,
      "loss": 4.6789,
      "mean_token_accuracy": 0.24786294549703597,
      "num_tokens": 61160680.0,
      "step": 26675
    },
    {
      "entropy": 5.4134870052337645,
      "epoch": 2.5629202689721424,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043455170722676194,
      "loss": 4.7774,
      "mean_token_accuracy": 0.23341585248708724,
      "num_tokens": 61172624.0,
      "step": 26680
    },
    {
      "entropy": 5.284338474273682,
      "epoch": 2.563400576368876,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004345275324126328,
      "loss": 4.6757,
      "mean_token_accuracy": 0.24423374533653258,
      "num_tokens": 61183971.0,
      "step": 26685
    },
    {
      "entropy": 5.381948804855346,
      "epoch": 2.56388088376561,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043450335389474796,
      "loss": 4.6666,
      "mean_token_accuracy": 0.25926610082387924,
      "num_tokens": 61194511.0,
      "step": 26690
    },
    {
      "entropy": 5.376442289352417,
      "epoch": 2.5643611911623436,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004344791716736689,
      "loss": 4.6935,
      "mean_token_accuracy": 0.24549823999404907,
      "num_tokens": 61205235.0,
      "step": 26695
    },
    {
      "entropy": 5.294037771224976,
      "epoch": 2.564841498559078,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043445498574995705,
      "loss": 4.7196,
      "mean_token_accuracy": 0.24851959496736525,
      "num_tokens": 61217460.0,
      "step": 26700
    },
    {
      "entropy": 5.38119592666626,
      "epoch": 2.565321805955812,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043443079612417394,
      "loss": 4.7061,
      "mean_token_accuracy": 0.24079620242118835,
      "num_tokens": 61229917.0,
      "step": 26705
    },
    {
      "entropy": 5.462753009796143,
      "epoch": 2.5658021133525457,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004344066027968812,
      "loss": 4.8562,
      "mean_token_accuracy": 0.23331095427274703,
      "num_tokens": 61241826.0,
      "step": 26710
    },
    {
      "entropy": 5.502343463897705,
      "epoch": 2.5662824207492796,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043438240576864034,
      "loss": 4.8485,
      "mean_token_accuracy": 0.24407794326543808,
      "num_tokens": 61254095.0,
      "step": 26715
    },
    {
      "entropy": 5.425324392318726,
      "epoch": 2.5667627281460135,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004343582050400134,
      "loss": 4.7995,
      "mean_token_accuracy": 0.2457335487008095,
      "num_tokens": 61266114.0,
      "step": 26720
    },
    {
      "entropy": 5.365182304382325,
      "epoch": 2.5672430355427474,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004343340006115621,
      "loss": 4.7361,
      "mean_token_accuracy": 0.24902541935443878,
      "num_tokens": 61277289.0,
      "step": 26725
    },
    {
      "entropy": 5.517322826385498,
      "epoch": 2.5677233429394812,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004343097924838483,
      "loss": 4.8432,
      "mean_token_accuracy": 0.23656417280435563,
      "num_tokens": 61289424.0,
      "step": 26730
    },
    {
      "entropy": 5.457433795928955,
      "epoch": 2.568203650336215,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004342855806574344,
      "loss": 4.7634,
      "mean_token_accuracy": 0.2436715617775917,
      "num_tokens": 61300821.0,
      "step": 26735
    },
    {
      "entropy": 5.378152465820312,
      "epoch": 2.568683957732949,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004342613651328822,
      "loss": 4.7254,
      "mean_token_accuracy": 0.2481303483247757,
      "num_tokens": 61312097.0,
      "step": 26740
    },
    {
      "entropy": 5.325612020492554,
      "epoch": 2.569164265129683,
      "grad_norm": 1.0625,
      "learning_rate": 0.000434237145910754,
      "loss": 4.7355,
      "mean_token_accuracy": 0.249875009059906,
      "num_tokens": 61324144.0,
      "step": 26745
    },
    {
      "entropy": 5.380805015563965,
      "epoch": 2.5696445725264168,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043421292299161213,
      "loss": 4.7103,
      "mean_token_accuracy": 0.24853265285491943,
      "num_tokens": 61335809.0,
      "step": 26750
    },
    {
      "entropy": 5.388551092147827,
      "epoch": 2.570124879923151,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043418869637601887,
      "loss": 4.7218,
      "mean_token_accuracy": 0.24585588574409484,
      "num_tokens": 61346980.0,
      "step": 26755
    },
    {
      "entropy": 5.427833175659179,
      "epoch": 2.5706051873198845,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043416446606453686,
      "loss": 4.8529,
      "mean_token_accuracy": 0.24366180896759032,
      "num_tokens": 61358401.0,
      "step": 26760
    },
    {
      "entropy": 5.354063653945923,
      "epoch": 2.571085494716619,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004341402320577285,
      "loss": 4.7168,
      "mean_token_accuracy": 0.24620661437511443,
      "num_tokens": 61369383.0,
      "step": 26765
    },
    {
      "entropy": 5.429807233810425,
      "epoch": 2.5715658021133523,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004341159943561566,
      "loss": 4.7594,
      "mean_token_accuracy": 0.2377810999751091,
      "num_tokens": 61380439.0,
      "step": 26770
    },
    {
      "entropy": 5.410635757446289,
      "epoch": 2.5720461095100866,
      "grad_norm": 1.125,
      "learning_rate": 0.0004340917529603837,
      "loss": 4.7063,
      "mean_token_accuracy": 0.2507499396800995,
      "num_tokens": 61391218.0,
      "step": 26775
    },
    {
      "entropy": 5.3791828632354735,
      "epoch": 2.5725264169068205,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004340675078709728,
      "loss": 4.693,
      "mean_token_accuracy": 0.24795616269111634,
      "num_tokens": 61402715.0,
      "step": 26780
    },
    {
      "entropy": 5.436404609680176,
      "epoch": 2.5730067243035544,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004340432590884866,
      "loss": 4.7785,
      "mean_token_accuracy": 0.23952916413545608,
      "num_tokens": 61414361.0,
      "step": 26785
    },
    {
      "entropy": 5.387647247314453,
      "epoch": 2.5734870317002883,
      "grad_norm": 1.0,
      "learning_rate": 0.00043401900661348825,
      "loss": 4.6968,
      "mean_token_accuracy": 0.25190409570932387,
      "num_tokens": 61426817.0,
      "step": 26790
    },
    {
      "entropy": 5.353986644744873,
      "epoch": 2.573967339097022,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043399475044654073,
      "loss": 4.7078,
      "mean_token_accuracy": 0.24744145572185516,
      "num_tokens": 61438142.0,
      "step": 26795
    },
    {
      "entropy": 5.422077083587647,
      "epoch": 2.574447646493756,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043397049058820726,
      "loss": 4.7761,
      "mean_token_accuracy": 0.24240230321884154,
      "num_tokens": 61450708.0,
      "step": 26800
    },
    {
      "entropy": 5.420721387863159,
      "epoch": 2.57492795389049,
      "grad_norm": 1.03125,
      "learning_rate": 0.000433946227039051,
      "loss": 4.7606,
      "mean_token_accuracy": 0.24458420723676683,
      "num_tokens": 61462266.0,
      "step": 26805
    },
    {
      "entropy": 5.3840230941772464,
      "epoch": 2.5754082612872238,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004339219597996353,
      "loss": 4.7584,
      "mean_token_accuracy": 0.24211145788431168,
      "num_tokens": 61473690.0,
      "step": 26810
    },
    {
      "entropy": 5.28098783493042,
      "epoch": 2.5758885686839577,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004338976888705236,
      "loss": 4.6421,
      "mean_token_accuracy": 0.25010411590337756,
      "num_tokens": 61484997.0,
      "step": 26815
    },
    {
      "entropy": 5.336399793624878,
      "epoch": 2.5763688760806915,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043387341425227944,
      "loss": 4.7857,
      "mean_token_accuracy": 0.24360421895980836,
      "num_tokens": 61497792.0,
      "step": 26820
    },
    {
      "entropy": 5.398309421539307,
      "epoch": 2.5768491834774254,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004338491359454664,
      "loss": 4.6525,
      "mean_token_accuracy": 0.2510547533631325,
      "num_tokens": 61509060.0,
      "step": 26825
    },
    {
      "entropy": 5.4404332637786865,
      "epoch": 2.5773294908741593,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043382485395064796,
      "loss": 4.7662,
      "mean_token_accuracy": 0.2506387040019035,
      "num_tokens": 61519577.0,
      "step": 26830
    },
    {
      "entropy": 5.353124618530273,
      "epoch": 2.577809798270893,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004338005682683881,
      "loss": 4.7123,
      "mean_token_accuracy": 0.24984124153852463,
      "num_tokens": 61531956.0,
      "step": 26835
    },
    {
      "entropy": 5.411020183563233,
      "epoch": 2.5782901056676275,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043377627889925057,
      "loss": 4.8007,
      "mean_token_accuracy": 0.2450212359428406,
      "num_tokens": 61543801.0,
      "step": 26840
    },
    {
      "entropy": 5.385021162033081,
      "epoch": 2.578770413064361,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004337519858437994,
      "loss": 4.8285,
      "mean_token_accuracy": 0.2368527188897133,
      "num_tokens": 61555899.0,
      "step": 26845
    },
    {
      "entropy": 5.317285537719727,
      "epoch": 2.5792507204610953,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004337276891025984,
      "loss": 4.6472,
      "mean_token_accuracy": 0.2569920912384987,
      "num_tokens": 61567280.0,
      "step": 26850
    },
    {
      "entropy": 5.348791599273682,
      "epoch": 2.579731027857829,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043370338867621184,
      "loss": 4.7706,
      "mean_token_accuracy": 0.243095101416111,
      "num_tokens": 61578026.0,
      "step": 26855
    },
    {
      "entropy": 5.436596918106079,
      "epoch": 2.580211335254563,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043367908456520387,
      "loss": 4.7517,
      "mean_token_accuracy": 0.24303337633609773,
      "num_tokens": 61589399.0,
      "step": 26860
    },
    {
      "entropy": 5.45786075592041,
      "epoch": 2.580691642651297,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004336547767701387,
      "loss": 4.7925,
      "mean_token_accuracy": 0.2391694739460945,
      "num_tokens": 61601025.0,
      "step": 26865
    },
    {
      "entropy": 5.491956615447998,
      "epoch": 2.581171950048031,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043363046529158077,
      "loss": 4.8686,
      "mean_token_accuracy": 0.23651119321584702,
      "num_tokens": 61612529.0,
      "step": 26870
    },
    {
      "entropy": 5.402074146270752,
      "epoch": 2.5816522574447647,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004336061501300944,
      "loss": 4.7948,
      "mean_token_accuracy": 0.2399053692817688,
      "num_tokens": 61624423.0,
      "step": 26875
    },
    {
      "entropy": 5.435030317306518,
      "epoch": 2.5821325648414986,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004335818312862442,
      "loss": 4.8337,
      "mean_token_accuracy": 0.23949942290782927,
      "num_tokens": 61635002.0,
      "step": 26880
    },
    {
      "entropy": 5.321755886077881,
      "epoch": 2.5826128722382324,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043355750876059485,
      "loss": 4.7035,
      "mean_token_accuracy": 0.25082610100507735,
      "num_tokens": 61645911.0,
      "step": 26885
    },
    {
      "entropy": 5.379904890060425,
      "epoch": 2.5830931796349663,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043353318255371086,
      "loss": 4.7385,
      "mean_token_accuracy": 0.24205119758844376,
      "num_tokens": 61657545.0,
      "step": 26890
    },
    {
      "entropy": 5.468476009368897,
      "epoch": 2.5835734870317,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004335088526661571,
      "loss": 4.8118,
      "mean_token_accuracy": 0.23892753571271896,
      "num_tokens": 61669567.0,
      "step": 26895
    },
    {
      "entropy": 5.489011001586914,
      "epoch": 2.584053794428434,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043348451909849855,
      "loss": 4.7816,
      "mean_token_accuracy": 0.2450801834464073,
      "num_tokens": 61681725.0,
      "step": 26900
    },
    {
      "entropy": 5.4333521842956545,
      "epoch": 2.584534101825168,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043346018185130006,
      "loss": 4.8117,
      "mean_token_accuracy": 0.24731603860855103,
      "num_tokens": 61692503.0,
      "step": 26905
    },
    {
      "entropy": 5.338430118560791,
      "epoch": 2.585014409221902,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043343584092512665,
      "loss": 4.6821,
      "mean_token_accuracy": 0.25278599858283995,
      "num_tokens": 61704567.0,
      "step": 26910
    },
    {
      "entropy": 5.379977512359619,
      "epoch": 2.585494716618636,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004334114963205435,
      "loss": 4.7136,
      "mean_token_accuracy": 0.24970741271972657,
      "num_tokens": 61716155.0,
      "step": 26915
    },
    {
      "entropy": 5.360914087295532,
      "epoch": 2.5859750240153696,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004333871480381158,
      "loss": 4.8381,
      "mean_token_accuracy": 0.24026299715042115,
      "num_tokens": 61727489.0,
      "step": 26920
    },
    {
      "entropy": 5.387317752838134,
      "epoch": 2.586455331412104,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004333627960784088,
      "loss": 4.7579,
      "mean_token_accuracy": 0.24654409289360046,
      "num_tokens": 61739792.0,
      "step": 26925
    },
    {
      "entropy": 5.428094387054443,
      "epoch": 2.586935638808838,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004333384404419879,
      "loss": 4.7818,
      "mean_token_accuracy": 0.24289190769195557,
      "num_tokens": 61750891.0,
      "step": 26930
    },
    {
      "entropy": 5.354785490036011,
      "epoch": 2.5874159462055717,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004333140811294186,
      "loss": 4.7156,
      "mean_token_accuracy": 0.25256241858005524,
      "num_tokens": 61761691.0,
      "step": 26935
    },
    {
      "entropy": 5.370416259765625,
      "epoch": 2.5878962536023056,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004332897181412664,
      "loss": 4.7412,
      "mean_token_accuracy": 0.24235396981239318,
      "num_tokens": 61772371.0,
      "step": 26940
    },
    {
      "entropy": 5.369528818130493,
      "epoch": 2.5883765609990395,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043326535147809696,
      "loss": 4.7214,
      "mean_token_accuracy": 0.24525033831596374,
      "num_tokens": 61784908.0,
      "step": 26945
    },
    {
      "entropy": 5.417806339263916,
      "epoch": 2.5888568683957733,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043324098114047604,
      "loss": 4.806,
      "mean_token_accuracy": 0.23758330047130585,
      "num_tokens": 61795948.0,
      "step": 26950
    },
    {
      "entropy": 5.441598796844483,
      "epoch": 2.589337175792507,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004332166071289695,
      "loss": 4.7519,
      "mean_token_accuracy": 0.24092639237642288,
      "num_tokens": 61807528.0,
      "step": 26955
    },
    {
      "entropy": 5.385370922088623,
      "epoch": 2.589817483189241,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043319222944414304,
      "loss": 4.7075,
      "mean_token_accuracy": 0.2527859404683113,
      "num_tokens": 61819971.0,
      "step": 26960
    },
    {
      "entropy": 5.317042446136474,
      "epoch": 2.590297790585975,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043316784808656276,
      "loss": 4.6986,
      "mean_token_accuracy": 0.25076976269483564,
      "num_tokens": 61832284.0,
      "step": 26965
    },
    {
      "entropy": 5.390976476669311,
      "epoch": 2.590778097982709,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043314346305679477,
      "loss": 4.7972,
      "mean_token_accuracy": 0.23816056400537491,
      "num_tokens": 61844350.0,
      "step": 26970
    },
    {
      "entropy": 5.429300594329834,
      "epoch": 2.5912584053794427,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043311907435540517,
      "loss": 4.7304,
      "mean_token_accuracy": 0.25098297744989395,
      "num_tokens": 61855180.0,
      "step": 26975
    },
    {
      "entropy": 5.408500051498413,
      "epoch": 2.5917387127761766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004330946819829601,
      "loss": 4.7902,
      "mean_token_accuracy": 0.2379859447479248,
      "num_tokens": 61866963.0,
      "step": 26980
    },
    {
      "entropy": 5.44175090789795,
      "epoch": 2.5922190201729105,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043307028594002597,
      "loss": 4.8154,
      "mean_token_accuracy": 0.24287242889404298,
      "num_tokens": 61877503.0,
      "step": 26985
    },
    {
      "entropy": 5.357765102386475,
      "epoch": 2.592699327569645,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043304588622716924,
      "loss": 4.7149,
      "mean_token_accuracy": 0.25312431901693344,
      "num_tokens": 61888426.0,
      "step": 26990
    },
    {
      "entropy": 5.396273899078369,
      "epoch": 2.5931796349663783,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004330214828449563,
      "loss": 4.713,
      "mean_token_accuracy": 0.23927657306194305,
      "num_tokens": 61899952.0,
      "step": 26995
    },
    {
      "entropy": 5.323684597015381,
      "epoch": 2.5936599423631126,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043299707579395365,
      "loss": 4.721,
      "mean_token_accuracy": 0.25136542320251465,
      "num_tokens": 61911956.0,
      "step": 27000
    },
    {
      "epoch": 2.5936599423631126,
      "eval_entropy": 5.114557192303505,
      "eval_loss": 4.827047824859619,
      "eval_mean_token_accuracy": 0.24942060406326608,
      "eval_num_tokens": 61911956.0,
      "eval_runtime": 20.6736,
      "eval_samples_per_second": 1587.29,
      "eval_steps_per_second": 198.417,
      "step": 27000
    },
    {
      "entropy": 5.365941286087036,
      "epoch": 2.594140249759846,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004329726650747282,
      "loss": 4.7411,
      "mean_token_accuracy": 0.2444704607129097,
      "num_tokens": 61922916.0,
      "step": 27005
    },
    {
      "entropy": 5.442340135574341,
      "epoch": 2.5946205571565804,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004329482506878465,
      "loss": 4.7881,
      "mean_token_accuracy": 0.24097278714179993,
      "num_tokens": 61933683.0,
      "step": 27010
    },
    {
      "entropy": 5.371219682693481,
      "epoch": 2.5951008645533142,
      "grad_norm": 1.125,
      "learning_rate": 0.00043292383263387536,
      "loss": 4.8252,
      "mean_token_accuracy": 0.24567444920539855,
      "num_tokens": 61945063.0,
      "step": 27015
    },
    {
      "entropy": 5.377324056625366,
      "epoch": 2.595581171950048,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043289941091338187,
      "loss": 4.778,
      "mean_token_accuracy": 0.2369828298687935,
      "num_tokens": 61955919.0,
      "step": 27020
    },
    {
      "entropy": 5.561729001998901,
      "epoch": 2.596061479346782,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004328749855269329,
      "loss": 4.9135,
      "mean_token_accuracy": 0.23482038229703903,
      "num_tokens": 61967083.0,
      "step": 27025
    },
    {
      "entropy": 5.378096437454223,
      "epoch": 2.596541786743516,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004328505564750955,
      "loss": 4.6877,
      "mean_token_accuracy": 0.2520021513104439,
      "num_tokens": 61978659.0,
      "step": 27030
    },
    {
      "entropy": 5.366757535934449,
      "epoch": 2.5970220941402498,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000432826123758437,
      "loss": 4.6867,
      "mean_token_accuracy": 0.2520362287759781,
      "num_tokens": 61990966.0,
      "step": 27035
    },
    {
      "entropy": 5.34021406173706,
      "epoch": 2.5975024015369836,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004328016873775244,
      "loss": 4.7154,
      "mean_token_accuracy": 0.24646880626678466,
      "num_tokens": 62002417.0,
      "step": 27040
    },
    {
      "entropy": 5.459745025634765,
      "epoch": 2.5979827089337175,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043277724733292527,
      "loss": 4.7851,
      "mean_token_accuracy": 0.24165849834680558,
      "num_tokens": 62014698.0,
      "step": 27045
    },
    {
      "entropy": 5.332755517959595,
      "epoch": 2.5984630163304514,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004327528036252071,
      "loss": 4.6812,
      "mean_token_accuracy": 0.2507682338356972,
      "num_tokens": 62025298.0,
      "step": 27050
    },
    {
      "entropy": 5.344302988052368,
      "epoch": 2.5989433237271853,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004327283562549371,
      "loss": 4.6851,
      "mean_token_accuracy": 0.25215565264225004,
      "num_tokens": 62036419.0,
      "step": 27055
    },
    {
      "entropy": 5.447202396392822,
      "epoch": 2.599423631123919,
      "grad_norm": 1.171875,
      "learning_rate": 0.000432703905222683,
      "loss": 4.809,
      "mean_token_accuracy": 0.24314012080430986,
      "num_tokens": 62046800.0,
      "step": 27060
    },
    {
      "entropy": 5.372847557067871,
      "epoch": 2.5999039385206535,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043267945052901264,
      "loss": 4.7523,
      "mean_token_accuracy": 0.24354317486286164,
      "num_tokens": 62058519.0,
      "step": 27065
    },
    {
      "entropy": 5.380311298370361,
      "epoch": 2.600384245917387,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004326549921744935,
      "loss": 4.7525,
      "mean_token_accuracy": 0.244274365901947,
      "num_tokens": 62070445.0,
      "step": 27070
    },
    {
      "entropy": 5.423142433166504,
      "epoch": 2.6008645533141213,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004326305301596937,
      "loss": 4.7861,
      "mean_token_accuracy": 0.23655037134885787,
      "num_tokens": 62083084.0,
      "step": 27075
    },
    {
      "entropy": 5.424830198287964,
      "epoch": 2.6013448607108547,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043260606448518096,
      "loss": 4.7459,
      "mean_token_accuracy": 0.23972322344779967,
      "num_tokens": 62094334.0,
      "step": 27080
    },
    {
      "entropy": 5.37874207496643,
      "epoch": 2.601825168107589,
      "grad_norm": 1.125,
      "learning_rate": 0.00043258159515152347,
      "loss": 4.7612,
      "mean_token_accuracy": 0.24274174124002457,
      "num_tokens": 62105065.0,
      "step": 27085
    },
    {
      "entropy": 5.432070302963257,
      "epoch": 2.602305475504323,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004325571221592892,
      "loss": 4.8315,
      "mean_token_accuracy": 0.2402794197201729,
      "num_tokens": 62116990.0,
      "step": 27090
    },
    {
      "entropy": 5.385702276229859,
      "epoch": 2.6027857829010568,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043253264550904646,
      "loss": 4.6874,
      "mean_token_accuracy": 0.24798618704080583,
      "num_tokens": 62128096.0,
      "step": 27095
    },
    {
      "entropy": 5.503657388687134,
      "epoch": 2.6032660902977907,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004325081652013635,
      "loss": 4.8517,
      "mean_token_accuracy": 0.23686621636152266,
      "num_tokens": 62141139.0,
      "step": 27100
    },
    {
      "entropy": 5.321858263015747,
      "epoch": 2.6037463976945245,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043248368123680855,
      "loss": 4.6124,
      "mean_token_accuracy": 0.25344028472900393,
      "num_tokens": 62152522.0,
      "step": 27105
    },
    {
      "entropy": 5.34559588432312,
      "epoch": 2.6042267050912584,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043245919361595026,
      "loss": 4.6901,
      "mean_token_accuracy": 0.2520501405000687,
      "num_tokens": 62162919.0,
      "step": 27110
    },
    {
      "entropy": 5.395872783660889,
      "epoch": 2.6047070124879923,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043243470233935696,
      "loss": 4.7276,
      "mean_token_accuracy": 0.24843762069940567,
      "num_tokens": 62174936.0,
      "step": 27115
    },
    {
      "entropy": 5.435598134994507,
      "epoch": 2.605187319884726,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004324102074075973,
      "loss": 4.7909,
      "mean_token_accuracy": 0.24091539680957794,
      "num_tokens": 62186260.0,
      "step": 27120
    },
    {
      "entropy": 5.389654731750488,
      "epoch": 2.60566762728146,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004323857088212402,
      "loss": 4.6809,
      "mean_token_accuracy": 0.25344342589378355,
      "num_tokens": 62198616.0,
      "step": 27125
    },
    {
      "entropy": 5.3632011890411375,
      "epoch": 2.606147934678194,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004323612065808541,
      "loss": 4.6931,
      "mean_token_accuracy": 0.24599785655736922,
      "num_tokens": 62209318.0,
      "step": 27130
    },
    {
      "entropy": 5.415614461898803,
      "epoch": 2.606628242074928,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043233670068700827,
      "loss": 4.8313,
      "mean_token_accuracy": 0.23850579857826232,
      "num_tokens": 62220510.0,
      "step": 27135
    },
    {
      "entropy": 5.440539360046387,
      "epoch": 2.6071085494716617,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004323121911402713,
      "loss": 4.8455,
      "mean_token_accuracy": 0.23486332297325135,
      "num_tokens": 62233676.0,
      "step": 27140
    },
    {
      "entropy": 5.45298490524292,
      "epoch": 2.6075888568683956,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043228767794121245,
      "loss": 4.7657,
      "mean_token_accuracy": 0.23865221738815307,
      "num_tokens": 62244604.0,
      "step": 27145
    },
    {
      "entropy": 5.393918657302857,
      "epoch": 2.60806916426513,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004322631610904006,
      "loss": 4.7829,
      "mean_token_accuracy": 0.2407765120267868,
      "num_tokens": 62255431.0,
      "step": 27150
    },
    {
      "entropy": 5.488351535797119,
      "epoch": 2.6085494716618634,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043223864058840534,
      "loss": 4.9135,
      "mean_token_accuracy": 0.2326049402356148,
      "num_tokens": 62267050.0,
      "step": 27155
    },
    {
      "entropy": 5.4513195037841795,
      "epoch": 2.6090297790585977,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043221411643579557,
      "loss": 4.8073,
      "mean_token_accuracy": 0.24196969866752624,
      "num_tokens": 62278939.0,
      "step": 27160
    },
    {
      "entropy": 5.3630777359008786,
      "epoch": 2.6095100864553316,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043218958863314096,
      "loss": 4.7365,
      "mean_token_accuracy": 0.2450277626514435,
      "num_tokens": 62291275.0,
      "step": 27165
    },
    {
      "entropy": 5.3461730003356935,
      "epoch": 2.6099903938520654,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004321650571810109,
      "loss": 4.6804,
      "mean_token_accuracy": 0.24995003938674926,
      "num_tokens": 62302191.0,
      "step": 27170
    },
    {
      "entropy": 5.422148418426514,
      "epoch": 2.6104707012487993,
      "grad_norm": 1.125,
      "learning_rate": 0.0004321405220799747,
      "loss": 4.6814,
      "mean_token_accuracy": 0.25394339859485626,
      "num_tokens": 62313678.0,
      "step": 27175
    },
    {
      "entropy": 5.364298391342163,
      "epoch": 2.610951008645533,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004321159833306024,
      "loss": 4.7152,
      "mean_token_accuracy": 0.2458559826016426,
      "num_tokens": 62324324.0,
      "step": 27180
    },
    {
      "entropy": 5.467964172363281,
      "epoch": 2.611431316042267,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004320914409334634,
      "loss": 4.8411,
      "mean_token_accuracy": 0.2400656074285507,
      "num_tokens": 62335385.0,
      "step": 27185
    },
    {
      "entropy": 5.447521734237671,
      "epoch": 2.611911623439001,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004320668948891276,
      "loss": 4.8084,
      "mean_token_accuracy": 0.23488404601812363,
      "num_tokens": 62346733.0,
      "step": 27190
    },
    {
      "entropy": 5.426527452468872,
      "epoch": 2.612391930835735,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043204234519816486,
      "loss": 4.7686,
      "mean_token_accuracy": 0.24817940592765808,
      "num_tokens": 62358674.0,
      "step": 27195
    },
    {
      "entropy": 5.419166183471679,
      "epoch": 2.6128722382324687,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004320177918611453,
      "loss": 4.7375,
      "mean_token_accuracy": 0.24740142822265626,
      "num_tokens": 62370431.0,
      "step": 27200
    },
    {
      "entropy": 5.355650758743286,
      "epoch": 2.6133525456292026,
      "grad_norm": 1.125,
      "learning_rate": 0.00043199323487863876,
      "loss": 4.6886,
      "mean_token_accuracy": 0.24634982645511627,
      "num_tokens": 62381098.0,
      "step": 27205
    },
    {
      "entropy": 5.367639780044556,
      "epoch": 2.6138328530259365,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043196867425121554,
      "loss": 4.7089,
      "mean_token_accuracy": 0.24472171515226365,
      "num_tokens": 62393446.0,
      "step": 27210
    },
    {
      "entropy": 5.4458471775054935,
      "epoch": 2.6143131604226704,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00043194410997944577,
      "loss": 4.7769,
      "mean_token_accuracy": 0.24370328485965728,
      "num_tokens": 62406581.0,
      "step": 27215
    },
    {
      "entropy": 5.4719915866851805,
      "epoch": 2.6147934678194042,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043191954206389985,
      "loss": 4.8012,
      "mean_token_accuracy": 0.24267559498548508,
      "num_tokens": 62419516.0,
      "step": 27220
    },
    {
      "entropy": 5.390590667724609,
      "epoch": 2.6152737752161386,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004318949705051481,
      "loss": 4.7261,
      "mean_token_accuracy": 0.24639428853988649,
      "num_tokens": 62431577.0,
      "step": 27225
    },
    {
      "entropy": 5.332197475433349,
      "epoch": 2.615754082612872,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004318703953037609,
      "loss": 4.7369,
      "mean_token_accuracy": 0.24618161618709564,
      "num_tokens": 62444507.0,
      "step": 27230
    },
    {
      "entropy": 5.343217515945435,
      "epoch": 2.6162343900096063,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004318458164603091,
      "loss": 4.7073,
      "mean_token_accuracy": 0.23854710459709166,
      "num_tokens": 62455692.0,
      "step": 27235
    },
    {
      "entropy": 5.479034376144409,
      "epoch": 2.61671469740634,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004318212339753631,
      "loss": 4.812,
      "mean_token_accuracy": 0.237315633893013,
      "num_tokens": 62467215.0,
      "step": 27240
    },
    {
      "entropy": 5.404189825057983,
      "epoch": 2.617195004803074,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043179664784949375,
      "loss": 4.8351,
      "mean_token_accuracy": 0.23694666028022765,
      "num_tokens": 62480016.0,
      "step": 27245
    },
    {
      "entropy": 5.4408926486969,
      "epoch": 2.617675312199808,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004317720580832717,
      "loss": 4.82,
      "mean_token_accuracy": 0.24277731031179428,
      "num_tokens": 62491088.0,
      "step": 27250
    },
    {
      "entropy": 5.467700004577637,
      "epoch": 2.618155619596542,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004317474646772681,
      "loss": 4.769,
      "mean_token_accuracy": 0.24393269866704942,
      "num_tokens": 62502405.0,
      "step": 27255
    },
    {
      "entropy": 5.43391227722168,
      "epoch": 2.6186359269932757,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004317228676320539,
      "loss": 4.7581,
      "mean_token_accuracy": 0.24345581531524657,
      "num_tokens": 62512968.0,
      "step": 27260
    },
    {
      "entropy": 5.359237861633301,
      "epoch": 2.6191162343900096,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043169826694819987,
      "loss": 4.6481,
      "mean_token_accuracy": 0.25387070924043653,
      "num_tokens": 62524242.0,
      "step": 27265
    },
    {
      "entropy": 5.4246735095977785,
      "epoch": 2.6195965417867435,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004316736626262775,
      "loss": 4.7955,
      "mean_token_accuracy": 0.24734617322683333,
      "num_tokens": 62536556.0,
      "step": 27270
    },
    {
      "entropy": 5.435672187805176,
      "epoch": 2.6200768491834774,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004316490546668579,
      "loss": 4.7908,
      "mean_token_accuracy": 0.24606167078018187,
      "num_tokens": 62547897.0,
      "step": 27275
    },
    {
      "entropy": 5.411454582214356,
      "epoch": 2.6205571565802113,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004316244430705124,
      "loss": 4.7868,
      "mean_token_accuracy": 0.2390949860215187,
      "num_tokens": 62560190.0,
      "step": 27280
    },
    {
      "entropy": 5.443369197845459,
      "epoch": 2.621037463976945,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004315998278378123,
      "loss": 4.8511,
      "mean_token_accuracy": 0.23609217554330825,
      "num_tokens": 62571885.0,
      "step": 27285
    },
    {
      "entropy": 5.399615621566772,
      "epoch": 2.621517771373679,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043157520896932943,
      "loss": 4.7784,
      "mean_token_accuracy": 0.2398118108510971,
      "num_tokens": 62583698.0,
      "step": 27290
    },
    {
      "entropy": 5.495315074920654,
      "epoch": 2.621998078770413,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000431550586465635,
      "loss": 4.8086,
      "mean_token_accuracy": 0.23951950669288635,
      "num_tokens": 62594813.0,
      "step": 27295
    },
    {
      "entropy": 5.343179559707641,
      "epoch": 2.6224783861671472,
      "grad_norm": 1.125,
      "learning_rate": 0.00043152596032730085,
      "loss": 4.6092,
      "mean_token_accuracy": 0.2526554957032204,
      "num_tokens": 62604868.0,
      "step": 27300
    },
    {
      "entropy": 5.306625461578369,
      "epoch": 2.6229586935638807,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043150133055489865,
      "loss": 4.6914,
      "mean_token_accuracy": 0.24760037958621978,
      "num_tokens": 62616553.0,
      "step": 27305
    },
    {
      "entropy": 5.4035509586334225,
      "epoch": 2.623439000960615,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004314766971490003,
      "loss": 4.6688,
      "mean_token_accuracy": 0.2479391574859619,
      "num_tokens": 62627468.0,
      "step": 27310
    },
    {
      "entropy": 5.436639881134033,
      "epoch": 2.6239193083573484,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004314520601101776,
      "loss": 4.77,
      "mean_token_accuracy": 0.24703161269426346,
      "num_tokens": 62639255.0,
      "step": 27315
    },
    {
      "entropy": 5.383675670623779,
      "epoch": 2.6243996157540828,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043142741943900275,
      "loss": 4.7344,
      "mean_token_accuracy": 0.24184447079896926,
      "num_tokens": 62650553.0,
      "step": 27320
    },
    {
      "entropy": 5.447794103622437,
      "epoch": 2.6248799231508166,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043140277513604763,
      "loss": 4.8492,
      "mean_token_accuracy": 0.24215661883354186,
      "num_tokens": 62662653.0,
      "step": 27325
    },
    {
      "entropy": 5.402947807312012,
      "epoch": 2.6253602305475505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004313781272018845,
      "loss": 4.7175,
      "mean_token_accuracy": 0.24550085961818696,
      "num_tokens": 62673618.0,
      "step": 27330
    },
    {
      "entropy": 5.42490816116333,
      "epoch": 2.6258405379442844,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004313534756370857,
      "loss": 4.699,
      "mean_token_accuracy": 0.248326900601387,
      "num_tokens": 62685221.0,
      "step": 27335
    },
    {
      "entropy": 5.371766853332519,
      "epoch": 2.6263208453410183,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043132882044222336,
      "loss": 4.606,
      "mean_token_accuracy": 0.2561478689312935,
      "num_tokens": 62695544.0,
      "step": 27340
    },
    {
      "entropy": 5.257101917266846,
      "epoch": 2.626801152737752,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043130416161787005,
      "loss": 4.5663,
      "mean_token_accuracy": 0.2601322948932648,
      "num_tokens": 62706945.0,
      "step": 27345
    },
    {
      "entropy": 5.3827964782714846,
      "epoch": 2.627281460134486,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043127949916459823,
      "loss": 4.7281,
      "mean_token_accuracy": 0.24910924583673477,
      "num_tokens": 62717821.0,
      "step": 27350
    },
    {
      "entropy": 5.351065444946289,
      "epoch": 2.62776176753122,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043125483308298053,
      "loss": 4.6469,
      "mean_token_accuracy": 0.25450560450553894,
      "num_tokens": 62730477.0,
      "step": 27355
    },
    {
      "entropy": 5.394227695465088,
      "epoch": 2.628242074927954,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004312301633735895,
      "loss": 4.7255,
      "mean_token_accuracy": 0.249906724691391,
      "num_tokens": 62741725.0,
      "step": 27360
    },
    {
      "entropy": 5.389869976043701,
      "epoch": 2.6287223823246877,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004312054900369981,
      "loss": 4.823,
      "mean_token_accuracy": 0.2375815376639366,
      "num_tokens": 62753600.0,
      "step": 27365
    },
    {
      "entropy": 5.387724351882935,
      "epoch": 2.6292026897214216,
      "grad_norm": 1.0625,
      "learning_rate": 0.000431180813073779,
      "loss": 4.841,
      "mean_token_accuracy": 0.24022828489542009,
      "num_tokens": 62766218.0,
      "step": 27370
    },
    {
      "entropy": 5.389410591125488,
      "epoch": 2.629682997118156,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004311561324845051,
      "loss": 4.7194,
      "mean_token_accuracy": 0.24608094990253448,
      "num_tokens": 62777723.0,
      "step": 27375
    },
    {
      "entropy": 5.495397520065308,
      "epoch": 2.6301633045148893,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004311314482697496,
      "loss": 4.8745,
      "mean_token_accuracy": 0.23949940204620362,
      "num_tokens": 62789711.0,
      "step": 27380
    },
    {
      "entropy": 5.357630586624145,
      "epoch": 2.6306436119116237,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004311067604300855,
      "loss": 4.7029,
      "mean_token_accuracy": 0.2495878666639328,
      "num_tokens": 62800600.0,
      "step": 27385
    },
    {
      "entropy": 5.411387586593628,
      "epoch": 2.631123919308357,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004310820689660859,
      "loss": 4.7744,
      "mean_token_accuracy": 0.24528637677431106,
      "num_tokens": 62811872.0,
      "step": 27390
    },
    {
      "entropy": 5.402470493316651,
      "epoch": 2.6316042267050914,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004310573738783242,
      "loss": 4.7542,
      "mean_token_accuracy": 0.25010149329900744,
      "num_tokens": 62822886.0,
      "step": 27395
    },
    {
      "entropy": 5.3554459571838375,
      "epoch": 2.6320845341018253,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004310326751673736,
      "loss": 4.6931,
      "mean_token_accuracy": 0.24558376520872116,
      "num_tokens": 62833911.0,
      "step": 27400
    },
    {
      "entropy": 5.366436910629273,
      "epoch": 2.632564841498559,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043100797283380756,
      "loss": 4.7291,
      "mean_token_accuracy": 0.24541582614183427,
      "num_tokens": 62844756.0,
      "step": 27405
    },
    {
      "entropy": 5.398586416244507,
      "epoch": 2.633045148895293,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043098326687819973,
      "loss": 4.7347,
      "mean_token_accuracy": 0.24726848304271698,
      "num_tokens": 62856072.0,
      "step": 27410
    },
    {
      "entropy": 5.477467679977417,
      "epoch": 2.633525456292027,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004309585573011236,
      "loss": 4.8121,
      "mean_token_accuracy": 0.2364842787384987,
      "num_tokens": 62868334.0,
      "step": 27415
    },
    {
      "entropy": 5.368943548202514,
      "epoch": 2.634005763688761,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004309338441031528,
      "loss": 4.6383,
      "mean_token_accuracy": 0.24979076087474822,
      "num_tokens": 62879459.0,
      "step": 27420
    },
    {
      "entropy": 5.300130558013916,
      "epoch": 2.6344860710854947,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043090912728486135,
      "loss": 4.635,
      "mean_token_accuracy": 0.25776236057281493,
      "num_tokens": 62890137.0,
      "step": 27425
    },
    {
      "entropy": 5.376676034927368,
      "epoch": 2.6349663784822286,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004308844068468228,
      "loss": 4.7795,
      "mean_token_accuracy": 0.23624580949544907,
      "num_tokens": 62902688.0,
      "step": 27430
    },
    {
      "entropy": 5.351876640319825,
      "epoch": 2.6354466858789625,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043085968278961116,
      "loss": 4.7524,
      "mean_token_accuracy": 0.24708005636930466,
      "num_tokens": 62913008.0,
      "step": 27435
    },
    {
      "entropy": 5.433205318450928,
      "epoch": 2.6359269932756964,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043083495511380055,
      "loss": 4.7835,
      "mean_token_accuracy": 0.23897168338298796,
      "num_tokens": 62924894.0,
      "step": 27440
    },
    {
      "entropy": 5.328454780578613,
      "epoch": 2.6364073006724302,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043081022381996506,
      "loss": 4.6091,
      "mean_token_accuracy": 0.2629702866077423,
      "num_tokens": 62935709.0,
      "step": 27445
    },
    {
      "entropy": 5.447645759582519,
      "epoch": 2.636887608069164,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004307854889086787,
      "loss": 4.7936,
      "mean_token_accuracy": 0.24263509958982468,
      "num_tokens": 62947122.0,
      "step": 27450
    },
    {
      "entropy": 5.372140407562256,
      "epoch": 2.637367915465898,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043076075038051605,
      "loss": 4.7029,
      "mean_token_accuracy": 0.24746685326099396,
      "num_tokens": 62957993.0,
      "step": 27455
    },
    {
      "entropy": 5.470898294448853,
      "epoch": 2.6378482228626323,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004307360082360512,
      "loss": 4.8091,
      "mean_token_accuracy": 0.23998838067054748,
      "num_tokens": 62969645.0,
      "step": 27460
    },
    {
      "entropy": 5.416840028762818,
      "epoch": 2.6383285302593658,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043071126247585866,
      "loss": 4.7304,
      "mean_token_accuracy": 0.245679834485054,
      "num_tokens": 62980770.0,
      "step": 27465
    },
    {
      "entropy": 5.348116254806518,
      "epoch": 2.6388088376561,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000430686513100513,
      "loss": 4.6151,
      "mean_token_accuracy": 0.26015287935733794,
      "num_tokens": 62991418.0,
      "step": 27470
    },
    {
      "entropy": 5.34922194480896,
      "epoch": 2.639289145052834,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043066176011058877,
      "loss": 4.6812,
      "mean_token_accuracy": 0.24959785789251326,
      "num_tokens": 63002228.0,
      "step": 27475
    },
    {
      "entropy": 5.3801661968231205,
      "epoch": 2.639769452449568,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043063700350666066,
      "loss": 4.7385,
      "mean_token_accuracy": 0.23810098618268966,
      "num_tokens": 63013502.0,
      "step": 27480
    },
    {
      "entropy": 5.433741617202759,
      "epoch": 2.6402497598463017,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004306122432893036,
      "loss": 4.8374,
      "mean_token_accuracy": 0.24044195115566253,
      "num_tokens": 63024530.0,
      "step": 27485
    },
    {
      "entropy": 5.369090414047241,
      "epoch": 2.6407300672430356,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043058747945909224,
      "loss": 4.7187,
      "mean_token_accuracy": 0.25310765355825426,
      "num_tokens": 63035668.0,
      "step": 27490
    },
    {
      "entropy": 5.374751234054566,
      "epoch": 2.6412103746397695,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043056271201660166,
      "loss": 4.7063,
      "mean_token_accuracy": 0.24774042069911956,
      "num_tokens": 63048020.0,
      "step": 27495
    },
    {
      "entropy": 5.40585036277771,
      "epoch": 2.6416906820365034,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004305379409624068,
      "loss": 4.7186,
      "mean_token_accuracy": 0.24656472206115723,
      "num_tokens": 63060019.0,
      "step": 27500
    },
    {
      "entropy": 5.347212076187134,
      "epoch": 2.6421709894332372,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004305131662970827,
      "loss": 4.703,
      "mean_token_accuracy": 0.24932971596717834,
      "num_tokens": 63070151.0,
      "step": 27505
    },
    {
      "entropy": 5.338239860534668,
      "epoch": 2.642651296829971,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004304883880212048,
      "loss": 4.7098,
      "mean_token_accuracy": 0.24854695349931716,
      "num_tokens": 63082174.0,
      "step": 27510
    },
    {
      "entropy": 5.386590957641602,
      "epoch": 2.643131604226705,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004304636061353482,
      "loss": 4.734,
      "mean_token_accuracy": 0.24927013367414474,
      "num_tokens": 63093478.0,
      "step": 27515
    },
    {
      "entropy": 5.300673389434815,
      "epoch": 2.643611911623439,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004304388206400883,
      "loss": 4.6488,
      "mean_token_accuracy": 0.2504815489053726,
      "num_tokens": 63104153.0,
      "step": 27520
    },
    {
      "entropy": 5.315580749511719,
      "epoch": 2.6440922190201728,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004304140315360005,
      "loss": 4.6384,
      "mean_token_accuracy": 0.25897945910692216,
      "num_tokens": 63115469.0,
      "step": 27525
    },
    {
      "entropy": 5.344649839401245,
      "epoch": 2.6445725264169067,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004303892388236604,
      "loss": 4.7907,
      "mean_token_accuracy": 0.24048371165990828,
      "num_tokens": 63126917.0,
      "step": 27530
    },
    {
      "entropy": 5.374996614456177,
      "epoch": 2.645052833813641,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004303644425036436,
      "loss": 4.7538,
      "mean_token_accuracy": 0.24419547468423844,
      "num_tokens": 63138865.0,
      "step": 27535
    },
    {
      "entropy": 5.313395309448242,
      "epoch": 2.6455331412103744,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043033964257652575,
      "loss": 4.5799,
      "mean_token_accuracy": 0.25590559244155886,
      "num_tokens": 63148937.0,
      "step": 27540
    },
    {
      "entropy": 5.340604543685913,
      "epoch": 2.6460134486071087,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004303148390428827,
      "loss": 4.6707,
      "mean_token_accuracy": 0.2502760201692581,
      "num_tokens": 63160850.0,
      "step": 27545
    },
    {
      "entropy": 5.440329694747925,
      "epoch": 2.6464937560038426,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043029003190329023,
      "loss": 4.7799,
      "mean_token_accuracy": 0.24458541423082353,
      "num_tokens": 63171660.0,
      "step": 27550
    },
    {
      "entropy": 5.4093639850616455,
      "epoch": 2.6469740634005765,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004302652211583244,
      "loss": 4.758,
      "mean_token_accuracy": 0.2418833389878273,
      "num_tokens": 63182291.0,
      "step": 27555
    },
    {
      "entropy": 5.364452266693116,
      "epoch": 2.6474543707973104,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004302404068085612,
      "loss": 4.6504,
      "mean_token_accuracy": 0.24869536757469177,
      "num_tokens": 63193646.0,
      "step": 27560
    },
    {
      "entropy": 5.364268064498901,
      "epoch": 2.6479346781940443,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004302155888545766,
      "loss": 4.8107,
      "mean_token_accuracy": 0.23560923635959624,
      "num_tokens": 63205021.0,
      "step": 27565
    },
    {
      "entropy": 5.421589994430542,
      "epoch": 2.648414985590778,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000430190767296947,
      "loss": 4.7693,
      "mean_token_accuracy": 0.2419068083167076,
      "num_tokens": 63217315.0,
      "step": 27570
    },
    {
      "entropy": 5.409494018554687,
      "epoch": 2.648895292987512,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004301659421362486,
      "loss": 4.7311,
      "mean_token_accuracy": 0.2459936887025833,
      "num_tokens": 63229082.0,
      "step": 27575
    },
    {
      "entropy": 5.505275583267212,
      "epoch": 2.649375600384246,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004301411133730578,
      "loss": 4.8512,
      "mean_token_accuracy": 0.24338161200284958,
      "num_tokens": 63240498.0,
      "step": 27580
    },
    {
      "entropy": 5.438808298110962,
      "epoch": 2.64985590778098,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043011628100795093,
      "loss": 4.8109,
      "mean_token_accuracy": 0.23912405222654343,
      "num_tokens": 63251149.0,
      "step": 27585
    },
    {
      "entropy": 5.477717161178589,
      "epoch": 2.6503362151777137,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004300914450415047,
      "loss": 4.7959,
      "mean_token_accuracy": 0.24633644372224808,
      "num_tokens": 63261903.0,
      "step": 27590
    },
    {
      "entropy": 5.336872673034668,
      "epoch": 2.6508165225744476,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043006660547429565,
      "loss": 4.7219,
      "mean_token_accuracy": 0.24443302899599076,
      "num_tokens": 63274281.0,
      "step": 27595
    },
    {
      "entropy": 5.319478988647461,
      "epoch": 2.6512968299711814,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004300417623069005,
      "loss": 4.6252,
      "mean_token_accuracy": 0.25826826244592666,
      "num_tokens": 63285421.0,
      "step": 27600
    },
    {
      "entropy": 5.430746126174927,
      "epoch": 2.6517771373679153,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004300169155398959,
      "loss": 4.7423,
      "mean_token_accuracy": 0.24500790536403655,
      "num_tokens": 63296828.0,
      "step": 27605
    },
    {
      "entropy": 5.469449996948242,
      "epoch": 2.6522574447646496,
      "grad_norm": 1.125,
      "learning_rate": 0.00042999206517385885,
      "loss": 4.7839,
      "mean_token_accuracy": 0.24201326668262482,
      "num_tokens": 63307132.0,
      "step": 27610
    },
    {
      "entropy": 5.430924749374389,
      "epoch": 2.652737752161383,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004299672112093663,
      "loss": 4.8183,
      "mean_token_accuracy": 0.24299897998571396,
      "num_tokens": 63318911.0,
      "step": 27615
    },
    {
      "entropy": 5.481614589691162,
      "epoch": 2.6532180595581174,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042994235364699526,
      "loss": 4.8254,
      "mean_token_accuracy": 0.2391984209418297,
      "num_tokens": 63329973.0,
      "step": 27620
    },
    {
      "entropy": 5.41963381767273,
      "epoch": 2.653698366954851,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004299174924873229,
      "loss": 4.7672,
      "mean_token_accuracy": 0.23557716310024263,
      "num_tokens": 63341576.0,
      "step": 27625
    },
    {
      "entropy": 5.36252064704895,
      "epoch": 2.654178674351585,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004298926277309263,
      "loss": 4.7615,
      "mean_token_accuracy": 0.24077270179986954,
      "num_tokens": 63353988.0,
      "step": 27630
    },
    {
      "entropy": 5.445815324783325,
      "epoch": 2.654658981748319,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042986775937838283,
      "loss": 4.7616,
      "mean_token_accuracy": 0.2503779575228691,
      "num_tokens": 63364746.0,
      "step": 27635
    },
    {
      "entropy": 5.493074131011963,
      "epoch": 2.655139289145053,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004298428874302699,
      "loss": 4.7729,
      "mean_token_accuracy": 0.2363354295492172,
      "num_tokens": 63375608.0,
      "step": 27640
    },
    {
      "entropy": 5.356559467315674,
      "epoch": 2.655619596541787,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004298180118871649,
      "loss": 4.6848,
      "mean_token_accuracy": 0.2468680188059807,
      "num_tokens": 63386581.0,
      "step": 27645
    },
    {
      "entropy": 5.390760660171509,
      "epoch": 2.6560999039385207,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042979313274964535,
      "loss": 4.7792,
      "mean_token_accuracy": 0.24268116503953935,
      "num_tokens": 63398844.0,
      "step": 27650
    },
    {
      "entropy": 5.432166624069214,
      "epoch": 2.6565802113352546,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042976825001828897,
      "loss": 4.7611,
      "mean_token_accuracy": 0.24832434207201004,
      "num_tokens": 63409972.0,
      "step": 27655
    },
    {
      "entropy": 5.403826522827148,
      "epoch": 2.6570605187319885,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042974336369367333,
      "loss": 4.6949,
      "mean_token_accuracy": 0.24698113650083542,
      "num_tokens": 63420598.0,
      "step": 27660
    },
    {
      "entropy": 5.460781764984131,
      "epoch": 2.6575408261287223,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004297184737763763,
      "loss": 4.855,
      "mean_token_accuracy": 0.23927136659622192,
      "num_tokens": 63431943.0,
      "step": 27665
    },
    {
      "entropy": 5.377767419815063,
      "epoch": 2.658021133525456,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042969358026697567,
      "loss": 4.6587,
      "mean_token_accuracy": 0.25359688997268676,
      "num_tokens": 63442539.0,
      "step": 27670
    },
    {
      "entropy": 5.396117305755615,
      "epoch": 2.65850144092219,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004296686831660495,
      "loss": 4.6932,
      "mean_token_accuracy": 0.24513079077005387,
      "num_tokens": 63453100.0,
      "step": 27675
    },
    {
      "entropy": 5.382018089294434,
      "epoch": 2.658981748318924,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004296437824741758,
      "loss": 4.7719,
      "mean_token_accuracy": 0.24763492196798326,
      "num_tokens": 63464226.0,
      "step": 27680
    },
    {
      "entropy": 5.409978246688842,
      "epoch": 2.659462055715658,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042961887819193263,
      "loss": 4.6865,
      "mean_token_accuracy": 0.2519227162003517,
      "num_tokens": 63475159.0,
      "step": 27685
    },
    {
      "entropy": 5.456892251968384,
      "epoch": 2.6599423631123917,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004295939703198983,
      "loss": 4.8886,
      "mean_token_accuracy": 0.23541590124368666,
      "num_tokens": 63488015.0,
      "step": 27690
    },
    {
      "entropy": 5.3947278499603275,
      "epoch": 2.660422670509126,
      "grad_norm": 1.109375,
      "learning_rate": 0.000429569058858651,
      "loss": 4.7298,
      "mean_token_accuracy": 0.24965973943471909,
      "num_tokens": 63501216.0,
      "step": 27695
    },
    {
      "entropy": 5.475032377243042,
      "epoch": 2.6609029779058595,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042954414380876906,
      "loss": 4.8033,
      "mean_token_accuracy": 0.24117963016033173,
      "num_tokens": 63511560.0,
      "step": 27700
    },
    {
      "entropy": 5.491408824920654,
      "epoch": 2.661383285302594,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042951922517083104,
      "loss": 4.8671,
      "mean_token_accuracy": 0.24012646228075027,
      "num_tokens": 63522762.0,
      "step": 27705
    },
    {
      "entropy": 5.425256252288818,
      "epoch": 2.6618635926993277,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004294943029454155,
      "loss": 4.8134,
      "mean_token_accuracy": 0.2458910048007965,
      "num_tokens": 63535095.0,
      "step": 27710
    },
    {
      "entropy": 5.374281930923462,
      "epoch": 2.6623439000960616,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042946937713310093,
      "loss": 4.6535,
      "mean_token_accuracy": 0.25604122281074526,
      "num_tokens": 63545256.0,
      "step": 27715
    },
    {
      "entropy": 5.414144515991211,
      "epoch": 2.6628242074927955,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004294444477344661,
      "loss": 4.7531,
      "mean_token_accuracy": 0.24247779995203017,
      "num_tokens": 63557269.0,
      "step": 27720
    },
    {
      "entropy": 5.344027280807495,
      "epoch": 2.6633045148895294,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004294195147500898,
      "loss": 4.6466,
      "mean_token_accuracy": 0.25068314373493195,
      "num_tokens": 63568452.0,
      "step": 27725
    },
    {
      "entropy": 5.43699312210083,
      "epoch": 2.6637848222862632,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042939457818055095,
      "loss": 4.8105,
      "mean_token_accuracy": 0.23725928068161012,
      "num_tokens": 63581653.0,
      "step": 27730
    },
    {
      "entropy": 5.375607347488403,
      "epoch": 2.664265129682997,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042936963802642843,
      "loss": 4.6486,
      "mean_token_accuracy": 0.24675327092409133,
      "num_tokens": 63592862.0,
      "step": 27735
    },
    {
      "entropy": 5.462741851806641,
      "epoch": 2.664745437079731,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004293446942883013,
      "loss": 4.8029,
      "mean_token_accuracy": 0.240984009206295,
      "num_tokens": 63603306.0,
      "step": 27740
    },
    {
      "entropy": 5.486707735061645,
      "epoch": 2.665225744476465,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042931974696674866,
      "loss": 4.8064,
      "mean_token_accuracy": 0.2391876682639122,
      "num_tokens": 63615031.0,
      "step": 27745
    },
    {
      "entropy": 5.467939329147339,
      "epoch": 2.6657060518731988,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004292947960623497,
      "loss": 4.7987,
      "mean_token_accuracy": 0.2454229176044464,
      "num_tokens": 63627004.0,
      "step": 27750
    },
    {
      "entropy": 5.447974681854248,
      "epoch": 2.6661863592699326,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042926984157568384,
      "loss": 4.7766,
      "mean_token_accuracy": 0.24386601001024247,
      "num_tokens": 63637712.0,
      "step": 27755
    },
    {
      "entropy": 5.396505546569824,
      "epoch": 2.6666666666666665,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042924488350733024,
      "loss": 4.7804,
      "mean_token_accuracy": 0.24363623410463334,
      "num_tokens": 63649631.0,
      "step": 27760
    },
    {
      "entropy": 5.456443977355957,
      "epoch": 2.6671469740634004,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042921992185786847,
      "loss": 4.7244,
      "mean_token_accuracy": 0.2459566041827202,
      "num_tokens": 63661929.0,
      "step": 27765
    },
    {
      "entropy": 5.463167715072632,
      "epoch": 2.6676272814601347,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042919495662787813,
      "loss": 4.8198,
      "mean_token_accuracy": 0.2383720561861992,
      "num_tokens": 63674411.0,
      "step": 27770
    },
    {
      "entropy": 5.378346729278564,
      "epoch": 2.668107588856868,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004291699878179387,
      "loss": 4.688,
      "mean_token_accuracy": 0.24489396661520005,
      "num_tokens": 63685182.0,
      "step": 27775
    },
    {
      "entropy": 5.41718020439148,
      "epoch": 2.6685878962536025,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004291450154286299,
      "loss": 4.7822,
      "mean_token_accuracy": 0.24077709317207335,
      "num_tokens": 63696374.0,
      "step": 27780
    },
    {
      "entropy": 5.400509262084961,
      "epoch": 2.6690682036503364,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004291200394605317,
      "loss": 4.8011,
      "mean_token_accuracy": 0.2416760578751564,
      "num_tokens": 63708092.0,
      "step": 27785
    },
    {
      "entropy": 5.348237133026123,
      "epoch": 2.6695485110470702,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004290950599142237,
      "loss": 4.6859,
      "mean_token_accuracy": 0.24980639815330505,
      "num_tokens": 63721143.0,
      "step": 27790
    },
    {
      "entropy": 5.444928407669067,
      "epoch": 2.670028818443804,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000429070076790286,
      "loss": 4.7862,
      "mean_token_accuracy": 0.24139244258403778,
      "num_tokens": 63732346.0,
      "step": 27795
    },
    {
      "entropy": 5.373377275466919,
      "epoch": 2.670509125840538,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042904509008929873,
      "loss": 4.6913,
      "mean_token_accuracy": 0.2578283488750458,
      "num_tokens": 63744008.0,
      "step": 27800
    },
    {
      "entropy": 5.424270296096802,
      "epoch": 2.670989433237272,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004290200998118417,
      "loss": 4.7307,
      "mean_token_accuracy": 0.2510578155517578,
      "num_tokens": 63754792.0,
      "step": 27805
    },
    {
      "entropy": 5.384313726425171,
      "epoch": 2.6714697406340058,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042899510595849544,
      "loss": 4.6859,
      "mean_token_accuracy": 0.2462619960308075,
      "num_tokens": 63767194.0,
      "step": 27810
    },
    {
      "entropy": 5.394554138183594,
      "epoch": 2.6719500480307397,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042897010852984004,
      "loss": 4.7205,
      "mean_token_accuracy": 0.2480889931321144,
      "num_tokens": 63778641.0,
      "step": 27815
    },
    {
      "entropy": 5.422869062423706,
      "epoch": 2.6724303554274735,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042894510752645586,
      "loss": 4.7422,
      "mean_token_accuracy": 0.2402460053563118,
      "num_tokens": 63791447.0,
      "step": 27820
    },
    {
      "entropy": 5.465722227096558,
      "epoch": 2.6729106628242074,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004289201029489235,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23711801767349244,
      "num_tokens": 63802162.0,
      "step": 27825
    },
    {
      "entropy": 5.426036548614502,
      "epoch": 2.6733909702209413,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004288950947978234,
      "loss": 4.7132,
      "mean_token_accuracy": 0.24456358104944229,
      "num_tokens": 63812689.0,
      "step": 27830
    },
    {
      "entropy": 5.462690639495849,
      "epoch": 2.673871277617675,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004288700830737361,
      "loss": 4.7917,
      "mean_token_accuracy": 0.24169176518917085,
      "num_tokens": 63822915.0,
      "step": 27835
    },
    {
      "entropy": 5.3502744197845455,
      "epoch": 2.674351585014409,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042884506777724244,
      "loss": 4.7549,
      "mean_token_accuracy": 0.24702333956956862,
      "num_tokens": 63834350.0,
      "step": 27840
    },
    {
      "entropy": 5.319619083404541,
      "epoch": 2.6748318924111434,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004288200489089231,
      "loss": 4.6847,
      "mean_token_accuracy": 0.24957484155893325,
      "num_tokens": 63846150.0,
      "step": 27845
    },
    {
      "entropy": 5.459547901153565,
      "epoch": 2.675312199807877,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000428795026469359,
      "loss": 4.7179,
      "mean_token_accuracy": 0.24470028132200242,
      "num_tokens": 63857675.0,
      "step": 27850
    },
    {
      "entropy": 5.4961363792419435,
      "epoch": 2.675792507204611,
      "grad_norm": 1.1875,
      "learning_rate": 0.000428770000459131,
      "loss": 4.8524,
      "mean_token_accuracy": 0.23379655927419662,
      "num_tokens": 63868073.0,
      "step": 27855
    },
    {
      "entropy": 5.434722280502319,
      "epoch": 2.6762728146013446,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004287449708788202,
      "loss": 4.7501,
      "mean_token_accuracy": 0.25394485890865326,
      "num_tokens": 63879085.0,
      "step": 27860
    },
    {
      "entropy": 5.34103684425354,
      "epoch": 2.676753121998079,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004287199377290077,
      "loss": 4.6746,
      "mean_token_accuracy": 0.24774724692106248,
      "num_tokens": 63890885.0,
      "step": 27865
    },
    {
      "entropy": 5.413989162445068,
      "epoch": 2.677233429394813,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004286949010102748,
      "loss": 4.7311,
      "mean_token_accuracy": 0.24412297159433366,
      "num_tokens": 63902421.0,
      "step": 27870
    },
    {
      "entropy": 5.429949808120727,
      "epoch": 2.6777137367915467,
      "grad_norm": 1.125,
      "learning_rate": 0.0004286698607232026,
      "loss": 4.7501,
      "mean_token_accuracy": 0.25011225044727325,
      "num_tokens": 63913845.0,
      "step": 27875
    },
    {
      "entropy": 5.307811069488525,
      "epoch": 2.6781940441882806,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042864481686837253,
      "loss": 4.6368,
      "mean_token_accuracy": 0.24601849019527436,
      "num_tokens": 63925429.0,
      "step": 27880
    },
    {
      "entropy": 5.346835279464722,
      "epoch": 2.6786743515850144,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042861976944636604,
      "loss": 4.7363,
      "mean_token_accuracy": 0.2463357701897621,
      "num_tokens": 63939451.0,
      "step": 27885
    },
    {
      "entropy": 5.537296485900879,
      "epoch": 2.6791546589817483,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004285947184577647,
      "loss": 4.8403,
      "mean_token_accuracy": 0.23259590417146683,
      "num_tokens": 63949077.0,
      "step": 27890
    },
    {
      "entropy": 5.51677451133728,
      "epoch": 2.679634966378482,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042856966390315013,
      "loss": 4.7763,
      "mean_token_accuracy": 0.24243636876344682,
      "num_tokens": 63960553.0,
      "step": 27895
    },
    {
      "entropy": 5.463659429550171,
      "epoch": 2.680115273775216,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004285446057831039,
      "loss": 4.8009,
      "mean_token_accuracy": 0.2354699119925499,
      "num_tokens": 63973086.0,
      "step": 27900
    },
    {
      "entropy": 5.3934777736663815,
      "epoch": 2.68059558117195,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004285195440982078,
      "loss": 4.7728,
      "mean_token_accuracy": 0.23404837101697923,
      "num_tokens": 63983933.0,
      "step": 27905
    },
    {
      "entropy": 5.3292436599731445,
      "epoch": 2.681075888568684,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004284944788490439,
      "loss": 4.6452,
      "mean_token_accuracy": 0.25834558457136153,
      "num_tokens": 63994557.0,
      "step": 27910
    },
    {
      "entropy": 5.479689455032348,
      "epoch": 2.6815561959654177,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004284694100361938,
      "loss": 4.8548,
      "mean_token_accuracy": 0.23803412020206452,
      "num_tokens": 64006368.0,
      "step": 27915
    },
    {
      "entropy": 5.421801519393921,
      "epoch": 2.682036503362152,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004284443376602398,
      "loss": 4.7023,
      "mean_token_accuracy": 0.24684924632310867,
      "num_tokens": 64018336.0,
      "step": 27920
    },
    {
      "entropy": 5.431493711471558,
      "epoch": 2.6825168107588855,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004284192617217639,
      "loss": 4.7978,
      "mean_token_accuracy": 0.24570002555847167,
      "num_tokens": 64030329.0,
      "step": 27925
    },
    {
      "entropy": 5.417697811126709,
      "epoch": 2.68299711815562,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004283941822213484,
      "loss": 4.7539,
      "mean_token_accuracy": 0.2446535274386406,
      "num_tokens": 64042502.0,
      "step": 27930
    },
    {
      "entropy": 5.410272693634033,
      "epoch": 2.6834774255523532,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004283690991595754,
      "loss": 4.7581,
      "mean_token_accuracy": 0.24353795349597931,
      "num_tokens": 64054217.0,
      "step": 27935
    },
    {
      "entropy": 5.532155084609985,
      "epoch": 2.6839577329490876,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042834401253702734,
      "loss": 4.8023,
      "mean_token_accuracy": 0.24236811697483063,
      "num_tokens": 64065019.0,
      "step": 27940
    },
    {
      "entropy": 5.3871711730957035,
      "epoch": 2.6844380403458215,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004283189223542866,
      "loss": 4.6854,
      "mean_token_accuracy": 0.2469455733895302,
      "num_tokens": 64076251.0,
      "step": 27945
    },
    {
      "entropy": 5.3920269966125485,
      "epoch": 2.6849183477425553,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042829382861193585,
      "loss": 4.7422,
      "mean_token_accuracy": 0.24888666570186616,
      "num_tokens": 64087257.0,
      "step": 27950
    },
    {
      "entropy": 5.420381927490235,
      "epoch": 2.685398655139289,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004282687313105575,
      "loss": 4.7834,
      "mean_token_accuracy": 0.23902465403079987,
      "num_tokens": 64098304.0,
      "step": 27955
    },
    {
      "entropy": 5.401509141921997,
      "epoch": 2.685878962536023,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042824363045073434,
      "loss": 4.7219,
      "mean_token_accuracy": 0.25229850262403486,
      "num_tokens": 64108839.0,
      "step": 27960
    },
    {
      "entropy": 5.458192014694214,
      "epoch": 2.686359269932757,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004282185260330491,
      "loss": 4.8431,
      "mean_token_accuracy": 0.23673263639211656,
      "num_tokens": 64121579.0,
      "step": 27965
    },
    {
      "entropy": 5.332287549972534,
      "epoch": 2.686839577329491,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042819341805808473,
      "loss": 4.6686,
      "mean_token_accuracy": 0.25608350336551666,
      "num_tokens": 64132220.0,
      "step": 27970
    },
    {
      "entropy": 5.364251661300659,
      "epoch": 2.6873198847262247,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042816830652642396,
      "loss": 4.7407,
      "mean_token_accuracy": 0.2448322668671608,
      "num_tokens": 64143580.0,
      "step": 27975
    },
    {
      "entropy": 5.422649002075195,
      "epoch": 2.6878001921229586,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004281431914386501,
      "loss": 4.7388,
      "mean_token_accuracy": 0.23970117419958115,
      "num_tokens": 64155042.0,
      "step": 27980
    },
    {
      "entropy": 5.535814189910889,
      "epoch": 2.6882804995196925,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000428118072795346,
      "loss": 4.8376,
      "mean_token_accuracy": 0.2417844444513321,
      "num_tokens": 64167755.0,
      "step": 27985
    },
    {
      "entropy": 5.440021419525147,
      "epoch": 2.6887608069164264,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042809295059709483,
      "loss": 4.7923,
      "mean_token_accuracy": 0.2473200798034668,
      "num_tokens": 64180930.0,
      "step": 27990
    },
    {
      "entropy": 5.399223184585571,
      "epoch": 2.6892411143131603,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042806782484448,
      "loss": 4.769,
      "mean_token_accuracy": 0.24244466572999954,
      "num_tokens": 64192994.0,
      "step": 27995
    },
    {
      "entropy": 5.455280303955078,
      "epoch": 2.689721421709894,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004280426955380848,
      "loss": 4.7877,
      "mean_token_accuracy": 0.24309344589710236,
      "num_tokens": 64204572.0,
      "step": 28000
    },
    {
      "entropy": 5.392292594909668,
      "epoch": 2.6902017291066285,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042801756267849266,
      "loss": 4.7603,
      "mean_token_accuracy": 0.2442528322339058,
      "num_tokens": 64216361.0,
      "step": 28005
    },
    {
      "entropy": 5.395213842391968,
      "epoch": 2.690682036503362,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004279924262662871,
      "loss": 4.6879,
      "mean_token_accuracy": 0.2510628789663315,
      "num_tokens": 64227074.0,
      "step": 28010
    },
    {
      "entropy": 5.347905063629151,
      "epoch": 2.6911623439000962,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004279672863020517,
      "loss": 4.7126,
      "mean_token_accuracy": 0.24995876997709274,
      "num_tokens": 64239507.0,
      "step": 28015
    },
    {
      "entropy": 5.436480903625489,
      "epoch": 2.69164265129683,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042794214278637013,
      "loss": 4.8152,
      "mean_token_accuracy": 0.23526376634836196,
      "num_tokens": 64251307.0,
      "step": 28020
    },
    {
      "entropy": 5.376623201370239,
      "epoch": 2.692122958693564,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042791699571982606,
      "loss": 4.7866,
      "mean_token_accuracy": 0.24253678023815156,
      "num_tokens": 64263988.0,
      "step": 28025
    },
    {
      "entropy": 5.439928245544434,
      "epoch": 2.692603266090298,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004278918451030035,
      "loss": 4.7782,
      "mean_token_accuracy": 0.2378925859928131,
      "num_tokens": 64275311.0,
      "step": 28030
    },
    {
      "entropy": 5.493709945678711,
      "epoch": 2.6930835734870318,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004278666909364863,
      "loss": 4.7945,
      "mean_token_accuracy": 0.24652593433856965,
      "num_tokens": 64286055.0,
      "step": 28035
    },
    {
      "entropy": 5.414661026000976,
      "epoch": 2.6935638808837656,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004278415332208584,
      "loss": 4.708,
      "mean_token_accuracy": 0.24607889205217362,
      "num_tokens": 64297286.0,
      "step": 28040
    },
    {
      "entropy": 5.4170403480529785,
      "epoch": 2.6940441882804995,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042781637195670396,
      "loss": 4.7835,
      "mean_token_accuracy": 0.24288019239902497,
      "num_tokens": 64308277.0,
      "step": 28045
    },
    {
      "entropy": 5.5572264194488525,
      "epoch": 2.6945244956772334,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004277912071446072,
      "loss": 4.9332,
      "mean_token_accuracy": 0.22636699974536895,
      "num_tokens": 64319260.0,
      "step": 28050
    },
    {
      "entropy": 5.532470083236694,
      "epoch": 2.6950048030739673,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004277660387851522,
      "loss": 4.8404,
      "mean_token_accuracy": 0.24142902046442033,
      "num_tokens": 64329840.0,
      "step": 28055
    },
    {
      "entropy": 5.513100242614746,
      "epoch": 2.695485110470701,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004277408668789235,
      "loss": 4.9027,
      "mean_token_accuracy": 0.2347881928086281,
      "num_tokens": 64342375.0,
      "step": 28060
    },
    {
      "entropy": 5.323594999313355,
      "epoch": 2.695965417867435,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004277156914265054,
      "loss": 4.6539,
      "mean_token_accuracy": 0.2560312822461128,
      "num_tokens": 64353396.0,
      "step": 28065
    },
    {
      "entropy": 5.478079319000244,
      "epoch": 2.696445725264169,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004276905124284824,
      "loss": 4.7944,
      "mean_token_accuracy": 0.24180107116699218,
      "num_tokens": 64364802.0,
      "step": 28070
    },
    {
      "entropy": 5.486475801467895,
      "epoch": 2.696926032660903,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004276653298854391,
      "loss": 4.765,
      "mean_token_accuracy": 0.23637621700763703,
      "num_tokens": 64376379.0,
      "step": 28075
    },
    {
      "entropy": 5.4181602001190186,
      "epoch": 2.697406340057637,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004276401437979601,
      "loss": 4.7939,
      "mean_token_accuracy": 0.24150921106338502,
      "num_tokens": 64388039.0,
      "step": 28080
    },
    {
      "entropy": 5.428839635848999,
      "epoch": 2.6978866474543706,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004276149541666303,
      "loss": 4.7426,
      "mean_token_accuracy": 0.2429282233119011,
      "num_tokens": 64401288.0,
      "step": 28085
    },
    {
      "entropy": 5.441208553314209,
      "epoch": 2.698366954851105,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042758976099203444,
      "loss": 4.7229,
      "mean_token_accuracy": 0.24525045454502106,
      "num_tokens": 64411392.0,
      "step": 28090
    },
    {
      "entropy": 5.430397081375122,
      "epoch": 2.6988472622478388,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042756456427475736,
      "loss": 4.821,
      "mean_token_accuracy": 0.24045581370592117,
      "num_tokens": 64421955.0,
      "step": 28095
    },
    {
      "entropy": 5.367854118347168,
      "epoch": 2.6993275696445727,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004275393640153842,
      "loss": 4.7288,
      "mean_token_accuracy": 0.24875780791044236,
      "num_tokens": 64434365.0,
      "step": 28100
    },
    {
      "entropy": 5.378961849212646,
      "epoch": 2.6998078770413065,
      "grad_norm": 1.25,
      "learning_rate": 0.00042751416021449986,
      "loss": 4.6993,
      "mean_token_accuracy": 0.2490269958972931,
      "num_tokens": 64444846.0,
      "step": 28105
    },
    {
      "entropy": 5.407178544998169,
      "epoch": 2.7002881844380404,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004274889528726896,
      "loss": 4.6835,
      "mean_token_accuracy": 0.25027810633182523,
      "num_tokens": 64455215.0,
      "step": 28110
    },
    {
      "entropy": 5.42086877822876,
      "epoch": 2.7007684918347743,
      "grad_norm": 1.125,
      "learning_rate": 0.0004274637419905388,
      "loss": 4.7309,
      "mean_token_accuracy": 0.24757531583309172,
      "num_tokens": 64465810.0,
      "step": 28115
    },
    {
      "entropy": 5.36236457824707,
      "epoch": 2.701248799231508,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042743852756863253,
      "loss": 4.7085,
      "mean_token_accuracy": 0.24996331036090852,
      "num_tokens": 64476615.0,
      "step": 28120
    },
    {
      "entropy": 5.408087539672851,
      "epoch": 2.701729106628242,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004274133096075563,
      "loss": 4.7498,
      "mean_token_accuracy": 0.24691009372472764,
      "num_tokens": 64490137.0,
      "step": 28125
    },
    {
      "entropy": 5.407566118240356,
      "epoch": 2.702209414024976,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004273880881078956,
      "loss": 4.7539,
      "mean_token_accuracy": 0.24741583168506623,
      "num_tokens": 64501357.0,
      "step": 28130
    },
    {
      "entropy": 5.288990449905396,
      "epoch": 2.70268972142171,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004273628630702359,
      "loss": 4.5898,
      "mean_token_accuracy": 0.25804840326309203,
      "num_tokens": 64512784.0,
      "step": 28135
    },
    {
      "entropy": 5.399630165100097,
      "epoch": 2.7031700288184437,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042733763449516313,
      "loss": 4.7383,
      "mean_token_accuracy": 0.2513707116246223,
      "num_tokens": 64523800.0,
      "step": 28140
    },
    {
      "entropy": 5.394446086883545,
      "epoch": 2.7036503362151776,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042731240238326273,
      "loss": 4.7293,
      "mean_token_accuracy": 0.24637897312641144,
      "num_tokens": 64535551.0,
      "step": 28145
    },
    {
      "entropy": 5.374680900573731,
      "epoch": 2.7041306436119115,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00042728716673512065,
      "loss": 4.737,
      "mean_token_accuracy": 0.25050972402095795,
      "num_tokens": 64547892.0,
      "step": 28150
    },
    {
      "entropy": 5.291445732116699,
      "epoch": 2.704610951008646,
      "grad_norm": 1.0,
      "learning_rate": 0.00042726192755132276,
      "loss": 4.6709,
      "mean_token_accuracy": 0.2510423496365547,
      "num_tokens": 64559687.0,
      "step": 28155
    },
    {
      "entropy": 5.379894256591797,
      "epoch": 2.7050912584053792,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042723668483245496,
      "loss": 4.6547,
      "mean_token_accuracy": 0.25457797646522523,
      "num_tokens": 64571691.0,
      "step": 28160
    },
    {
      "entropy": 5.477201366424561,
      "epoch": 2.7055715658021136,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004272114385791035,
      "loss": 4.8069,
      "mean_token_accuracy": 0.2409012109041214,
      "num_tokens": 64582570.0,
      "step": 28165
    },
    {
      "entropy": 5.46386685371399,
      "epoch": 2.706051873198847,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042718618879185435,
      "loss": 4.74,
      "mean_token_accuracy": 0.24740103483200074,
      "num_tokens": 64593610.0,
      "step": 28170
    },
    {
      "entropy": 5.397436952590942,
      "epoch": 2.7065321805955813,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004271609354712938,
      "loss": 4.712,
      "mean_token_accuracy": 0.25090051591396334,
      "num_tokens": 64604802.0,
      "step": 28175
    },
    {
      "entropy": 5.4009569644927975,
      "epoch": 2.707012487992315,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004271356786180082,
      "loss": 4.7258,
      "mean_token_accuracy": 0.24021450877189637,
      "num_tokens": 64615816.0,
      "step": 28180
    },
    {
      "entropy": 5.3387322425842285,
      "epoch": 2.707492795389049,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004271104182325838,
      "loss": 4.6759,
      "mean_token_accuracy": 0.24658720940351486,
      "num_tokens": 64628890.0,
      "step": 28185
    },
    {
      "entropy": 5.383990907669068,
      "epoch": 2.707973102785783,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042708515431560723,
      "loss": 4.6987,
      "mean_token_accuracy": 0.2510429695248604,
      "num_tokens": 64639494.0,
      "step": 28190
    },
    {
      "entropy": 5.48203935623169,
      "epoch": 2.708453410182517,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000427059886867665,
      "loss": 4.9104,
      "mean_token_accuracy": 0.23250313550233842,
      "num_tokens": 64651235.0,
      "step": 28195
    },
    {
      "entropy": 5.424621868133545,
      "epoch": 2.7089337175792507,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004270346158893436,
      "loss": 4.7722,
      "mean_token_accuracy": 0.2443312868475914,
      "num_tokens": 64662517.0,
      "step": 28200
    },
    {
      "entropy": 5.433714246749878,
      "epoch": 2.7094140249759846,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042700934138123004,
      "loss": 4.8184,
      "mean_token_accuracy": 0.2287725478410721,
      "num_tokens": 64674521.0,
      "step": 28205
    },
    {
      "entropy": 5.470170164108277,
      "epoch": 2.7098943323727185,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042698406334391084,
      "loss": 4.9083,
      "mean_token_accuracy": 0.22971467822790145,
      "num_tokens": 64686894.0,
      "step": 28210
    },
    {
      "entropy": 5.5204150676727295,
      "epoch": 2.7103746397694524,
      "grad_norm": 1.046875,
      "learning_rate": 0.000426958781777973,
      "loss": 4.8367,
      "mean_token_accuracy": 0.2435731753706932,
      "num_tokens": 64697948.0,
      "step": 28215
    },
    {
      "entropy": 5.428372144699097,
      "epoch": 2.7108549471661862,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004269334966840035,
      "loss": 4.7018,
      "mean_token_accuracy": 0.24844202548265457,
      "num_tokens": 64709429.0,
      "step": 28220
    },
    {
      "entropy": 5.41628131866455,
      "epoch": 2.71133525456292,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042690820806258933,
      "loss": 4.6984,
      "mean_token_accuracy": 0.2502831846475601,
      "num_tokens": 64720529.0,
      "step": 28225
    },
    {
      "entropy": 5.351749897003174,
      "epoch": 2.7118155619596545,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004268829159143176,
      "loss": 4.6842,
      "mean_token_accuracy": 0.25175545513629916,
      "num_tokens": 64731602.0,
      "step": 28230
    },
    {
      "entropy": 5.474798536300659,
      "epoch": 2.712295869356388,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004268576202397757,
      "loss": 4.6825,
      "mean_token_accuracy": 0.24939606189727784,
      "num_tokens": 64742746.0,
      "step": 28235
    },
    {
      "entropy": 5.410719537734986,
      "epoch": 2.712776176753122,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004268323210395506,
      "loss": 4.7469,
      "mean_token_accuracy": 0.24969069063663482,
      "num_tokens": 64754643.0,
      "step": 28240
    },
    {
      "entropy": 5.5170186996459964,
      "epoch": 2.7132564841498557,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042680701831423004,
      "loss": 4.7947,
      "mean_token_accuracy": 0.2432980790734291,
      "num_tokens": 64766349.0,
      "step": 28245
    },
    {
      "entropy": 5.375593280792236,
      "epoch": 2.71373679154659,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004267817120644012,
      "loss": 4.7455,
      "mean_token_accuracy": 0.24474356025457383,
      "num_tokens": 64777645.0,
      "step": 28250
    },
    {
      "entropy": 5.39573040008545,
      "epoch": 2.714217098943324,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042675640229065167,
      "loss": 4.7752,
      "mean_token_accuracy": 0.24167141020298005,
      "num_tokens": 64788065.0,
      "step": 28255
    },
    {
      "entropy": 5.454219818115234,
      "epoch": 2.7146974063400577,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042673108899356915,
      "loss": 4.7197,
      "mean_token_accuracy": 0.25385880619287493,
      "num_tokens": 64798938.0,
      "step": 28260
    },
    {
      "entropy": 5.406842851638794,
      "epoch": 2.7151777137367916,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004267057721737413,
      "loss": 4.7354,
      "mean_token_accuracy": 0.25110981464385984,
      "num_tokens": 64810572.0,
      "step": 28265
    },
    {
      "entropy": 5.388420104980469,
      "epoch": 2.7156580211335255,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004266804518317559,
      "loss": 4.7753,
      "mean_token_accuracy": 0.2510840862989426,
      "num_tokens": 64821877.0,
      "step": 28270
    },
    {
      "entropy": 5.4074255466461185,
      "epoch": 2.7161383285302594,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004266551279682008,
      "loss": 4.6901,
      "mean_token_accuracy": 0.25527661889791486,
      "num_tokens": 64832737.0,
      "step": 28275
    },
    {
      "entropy": 5.424128437042237,
      "epoch": 2.7166186359269933,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004266298005836639,
      "loss": 4.7562,
      "mean_token_accuracy": 0.2440018519759178,
      "num_tokens": 64844713.0,
      "step": 28280
    },
    {
      "entropy": 5.416880369186401,
      "epoch": 2.717098943323727,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042660446967873327,
      "loss": 4.7553,
      "mean_token_accuracy": 0.24758695513010026,
      "num_tokens": 64855896.0,
      "step": 28285
    },
    {
      "entropy": 5.445542860031128,
      "epoch": 2.717579250720461,
      "grad_norm": 1.25,
      "learning_rate": 0.00042657913525399703,
      "loss": 4.8339,
      "mean_token_accuracy": 0.2384026676416397,
      "num_tokens": 64866529.0,
      "step": 28290
    },
    {
      "entropy": 5.418431901931763,
      "epoch": 2.718059558117195,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004265537973100435,
      "loss": 4.7518,
      "mean_token_accuracy": 0.2391757145524025,
      "num_tokens": 64876691.0,
      "step": 28295
    },
    {
      "entropy": 5.38100962638855,
      "epoch": 2.718539865513929,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004265284558474607,
      "loss": 4.7198,
      "mean_token_accuracy": 0.2450137257575989,
      "num_tokens": 64887271.0,
      "step": 28300
    },
    {
      "entropy": 5.440728712081909,
      "epoch": 2.7190201729106627,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00042650311086683715,
      "loss": 4.8092,
      "mean_token_accuracy": 0.2339702621102333,
      "num_tokens": 64900698.0,
      "step": 28305
    },
    {
      "entropy": 5.463139343261719,
      "epoch": 2.7195004803073966,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004264777623687612,
      "loss": 4.7545,
      "mean_token_accuracy": 0.24520614594221116,
      "num_tokens": 64911349.0,
      "step": 28310
    },
    {
      "entropy": 5.423392391204834,
      "epoch": 2.719980787704131,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004264524103538214,
      "loss": 4.7411,
      "mean_token_accuracy": 0.24840528070926665,
      "num_tokens": 64922522.0,
      "step": 28315
    },
    {
      "entropy": 5.434117412567138,
      "epoch": 2.7204610951008643,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004264270548226064,
      "loss": 4.7774,
      "mean_token_accuracy": 0.2431645303964615,
      "num_tokens": 64934796.0,
      "step": 28320
    },
    {
      "entropy": 5.4388164520263675,
      "epoch": 2.7209414024975986,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004264016957757048,
      "loss": 4.7631,
      "mean_token_accuracy": 0.24915701746940613,
      "num_tokens": 64946717.0,
      "step": 28325
    },
    {
      "entropy": 5.450743818283081,
      "epoch": 2.7214217098943325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042637633321370545,
      "loss": 4.8288,
      "mean_token_accuracy": 0.23467555940151213,
      "num_tokens": 64958818.0,
      "step": 28330
    },
    {
      "entropy": 5.449873399734497,
      "epoch": 2.7219020172910664,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004263509671371971,
      "loss": 4.715,
      "mean_token_accuracy": 0.23738178312778474,
      "num_tokens": 64968974.0,
      "step": 28335
    },
    {
      "entropy": 5.451811599731445,
      "epoch": 2.7223823246878003,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042632559754676865,
      "loss": 4.7695,
      "mean_token_accuracy": 0.24037060290575027,
      "num_tokens": 64980347.0,
      "step": 28340
    },
    {
      "entropy": 5.470523357391357,
      "epoch": 2.722862632084534,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004263002244430092,
      "loss": 4.7419,
      "mean_token_accuracy": 0.24410362243652345,
      "num_tokens": 64992916.0,
      "step": 28345
    },
    {
      "entropy": 5.43455548286438,
      "epoch": 2.723342939481268,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004262748478265078,
      "loss": 4.8241,
      "mean_token_accuracy": 0.24208626449108123,
      "num_tokens": 65003382.0,
      "step": 28350
    },
    {
      "entropy": 5.3437333583831785,
      "epoch": 2.723823246878002,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004262494676978537,
      "loss": 4.6047,
      "mean_token_accuracy": 0.25605958998203276,
      "num_tokens": 65014064.0,
      "step": 28355
    },
    {
      "entropy": 5.452087306976319,
      "epoch": 2.724303554274736,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042622408405763607,
      "loss": 4.8103,
      "mean_token_accuracy": 0.23948317021131516,
      "num_tokens": 65025701.0,
      "step": 28360
    },
    {
      "entropy": 5.470452880859375,
      "epoch": 2.7247838616714697,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004261986969064442,
      "loss": 4.7484,
      "mean_token_accuracy": 0.24871542006731034,
      "num_tokens": 65038270.0,
      "step": 28365
    },
    {
      "entropy": 5.4126063823699955,
      "epoch": 2.7252641690682036,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042617330624486753,
      "loss": 4.7173,
      "mean_token_accuracy": 0.2516777545213699,
      "num_tokens": 65049453.0,
      "step": 28370
    },
    {
      "entropy": 5.2899651527404785,
      "epoch": 2.7257444764649374,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004261479120734956,
      "loss": 4.5786,
      "mean_token_accuracy": 0.26093064844608305,
      "num_tokens": 65061211.0,
      "step": 28375
    },
    {
      "entropy": 5.361912202835083,
      "epoch": 2.7262247838616713,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000426122514392918,
      "loss": 4.6493,
      "mean_token_accuracy": 0.25357838422060014,
      "num_tokens": 65073073.0,
      "step": 28380
    },
    {
      "entropy": 5.400931024551392,
      "epoch": 2.726705091258405,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042609711320372435,
      "loss": 4.7173,
      "mean_token_accuracy": 0.2570881173014641,
      "num_tokens": 65084647.0,
      "step": 28385
    },
    {
      "entropy": 5.425582647323608,
      "epoch": 2.7271853986551395,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004260717085065045,
      "loss": 4.7131,
      "mean_token_accuracy": 0.2563871756196022,
      "num_tokens": 65095405.0,
      "step": 28390
    },
    {
      "entropy": 5.463880586624145,
      "epoch": 2.727665706051873,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042604630030184797,
      "loss": 4.7653,
      "mean_token_accuracy": 0.24285699874162675,
      "num_tokens": 65106981.0,
      "step": 28395
    },
    {
      "entropy": 5.507924318313599,
      "epoch": 2.7281460134486073,
      "grad_norm": 1.0625,
      "learning_rate": 0.000426020888590345,
      "loss": 4.8513,
      "mean_token_accuracy": 0.23626574575901033,
      "num_tokens": 65118957.0,
      "step": 28400
    },
    {
      "entropy": 5.4085915088653564,
      "epoch": 2.728626320845341,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00042599547337258536,
      "loss": 4.6864,
      "mean_token_accuracy": 0.25474683195352554,
      "num_tokens": 65130897.0,
      "step": 28405
    },
    {
      "entropy": 5.314285230636597,
      "epoch": 2.729106628242075,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042597005464915924,
      "loss": 4.5606,
      "mean_token_accuracy": 0.2593643024563789,
      "num_tokens": 65142828.0,
      "step": 28410
    },
    {
      "entropy": 5.351073551177978,
      "epoch": 2.729586935638809,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042594463242065674,
      "loss": 4.6933,
      "mean_token_accuracy": 0.24999530762434005,
      "num_tokens": 65154787.0,
      "step": 28415
    },
    {
      "entropy": 5.402349662780762,
      "epoch": 2.730067243035543,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004259192066876681,
      "loss": 4.7049,
      "mean_token_accuracy": 0.24523156583309175,
      "num_tokens": 65165530.0,
      "step": 28420
    },
    {
      "entropy": 5.464373445510864,
      "epoch": 2.7305475504322767,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042589377745078354,
      "loss": 4.829,
      "mean_token_accuracy": 0.2409464344382286,
      "num_tokens": 65176809.0,
      "step": 28425
    },
    {
      "entropy": 5.500715351104736,
      "epoch": 2.7310278578290106,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042586834471059366,
      "loss": 4.8517,
      "mean_token_accuracy": 0.23619853109121322,
      "num_tokens": 65189251.0,
      "step": 28430
    },
    {
      "entropy": 5.339837265014649,
      "epoch": 2.7315081652257445,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042584290846768867,
      "loss": 4.643,
      "mean_token_accuracy": 0.2540797472000122,
      "num_tokens": 65200918.0,
      "step": 28435
    },
    {
      "entropy": 5.436923456192017,
      "epoch": 2.7319884726224783,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004258174687226593,
      "loss": 4.7232,
      "mean_token_accuracy": 0.24621603786945342,
      "num_tokens": 65213283.0,
      "step": 28440
    },
    {
      "entropy": 5.446203947067261,
      "epoch": 2.7324687800192122,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004257920254760962,
      "loss": 4.7677,
      "mean_token_accuracy": 0.25008384585380555,
      "num_tokens": 65223680.0,
      "step": 28445
    },
    {
      "entropy": 5.459822750091552,
      "epoch": 2.732949087415946,
      "grad_norm": 1.125,
      "learning_rate": 0.0004257665787285899,
      "loss": 4.7418,
      "mean_token_accuracy": 0.2489818051457405,
      "num_tokens": 65234570.0,
      "step": 28450
    },
    {
      "entropy": 5.422098064422608,
      "epoch": 2.73342939481268,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042574112848073147,
      "loss": 4.7116,
      "mean_token_accuracy": 0.2457915335893631,
      "num_tokens": 65246616.0,
      "step": 28455
    },
    {
      "entropy": 5.365636253356934,
      "epoch": 2.733909702209414,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042571567473311157,
      "loss": 4.732,
      "mean_token_accuracy": 0.24309368878602983,
      "num_tokens": 65257853.0,
      "step": 28460
    },
    {
      "entropy": 5.422026491165161,
      "epoch": 2.734390009606148,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004256902174863213,
      "loss": 4.7192,
      "mean_token_accuracy": 0.24572359323501586,
      "num_tokens": 65268843.0,
      "step": 28465
    },
    {
      "entropy": 5.522764492034912,
      "epoch": 2.7348703170028816,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042566475674095155,
      "loss": 4.8797,
      "mean_token_accuracy": 0.2280285432934761,
      "num_tokens": 65279824.0,
      "step": 28470
    },
    {
      "entropy": 5.388922119140625,
      "epoch": 2.735350624399616,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004256392924975936,
      "loss": 4.79,
      "mean_token_accuracy": 0.24348122924566268,
      "num_tokens": 65291318.0,
      "step": 28475
    },
    {
      "entropy": 5.309589004516601,
      "epoch": 2.7358309317963494,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042561382475683854,
      "loss": 4.7181,
      "mean_token_accuracy": 0.24986278116703034,
      "num_tokens": 65302997.0,
      "step": 28480
    },
    {
      "entropy": 5.343360424041748,
      "epoch": 2.7363112391930837,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004255883535192777,
      "loss": 4.7086,
      "mean_token_accuracy": 0.2551179096102715,
      "num_tokens": 65314336.0,
      "step": 28485
    },
    {
      "entropy": 5.472641658782959,
      "epoch": 2.7367915465898176,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004255628787855025,
      "loss": 4.7684,
      "mean_token_accuracy": 0.24537987411022186,
      "num_tokens": 65325138.0,
      "step": 28490
    },
    {
      "entropy": 5.481514310836792,
      "epoch": 2.7372718539865515,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004255374005561043,
      "loss": 4.7306,
      "mean_token_accuracy": 0.24480615109205245,
      "num_tokens": 65337067.0,
      "step": 28495
    },
    {
      "entropy": 5.433556318283081,
      "epoch": 2.7377521613832854,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042551191883167464,
      "loss": 4.7579,
      "mean_token_accuracy": 0.2411099448800087,
      "num_tokens": 65349598.0,
      "step": 28500
    },
    {
      "entropy": 5.491288900375366,
      "epoch": 2.7382324687800192,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004254864336128052,
      "loss": 4.8314,
      "mean_token_accuracy": 0.23637872338294982,
      "num_tokens": 65361052.0,
      "step": 28505
    },
    {
      "entropy": 5.465780782699585,
      "epoch": 2.738712776176753,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042546094490008765,
      "loss": 4.8257,
      "mean_token_accuracy": 0.2428350880742073,
      "num_tokens": 65373246.0,
      "step": 28510
    },
    {
      "entropy": 5.452096843719483,
      "epoch": 2.739193083573487,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004254354526941136,
      "loss": 4.7,
      "mean_token_accuracy": 0.2462410345673561,
      "num_tokens": 65384349.0,
      "step": 28515
    },
    {
      "entropy": 5.332217454910278,
      "epoch": 2.739673390970221,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004254099569954751,
      "loss": 4.7236,
      "mean_token_accuracy": 0.24366410076618195,
      "num_tokens": 65396241.0,
      "step": 28520
    },
    {
      "entropy": 5.360094594955444,
      "epoch": 2.7401536983669548,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004253844578047641,
      "loss": 4.7737,
      "mean_token_accuracy": 0.2388931319117546,
      "num_tokens": 65408583.0,
      "step": 28525
    },
    {
      "entropy": 5.427208948135376,
      "epoch": 2.7406340057636887,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004253589551225725,
      "loss": 4.694,
      "mean_token_accuracy": 0.2494916573166847,
      "num_tokens": 65419845.0,
      "step": 28530
    },
    {
      "entropy": 5.463818550109863,
      "epoch": 2.7411143131604225,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042533344894949245,
      "loss": 4.7507,
      "mean_token_accuracy": 0.2352234125137329,
      "num_tokens": 65433790.0,
      "step": 28535
    },
    {
      "entropy": 5.44460654258728,
      "epoch": 2.7415946205571564,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042530793928611605,
      "loss": 4.7896,
      "mean_token_accuracy": 0.24361498951911925,
      "num_tokens": 65445958.0,
      "step": 28540
    },
    {
      "entropy": 5.409924650192261,
      "epoch": 2.7420749279538903,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004252824261330357,
      "loss": 4.761,
      "mean_token_accuracy": 0.24712521433830262,
      "num_tokens": 65457973.0,
      "step": 28545
    },
    {
      "entropy": 5.40349326133728,
      "epoch": 2.7425552353506246,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042525690949084364,
      "loss": 4.7574,
      "mean_token_accuracy": 0.24913830310106277,
      "num_tokens": 65470245.0,
      "step": 28550
    },
    {
      "entropy": 5.3300261974334715,
      "epoch": 2.743035542747358,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042523138936013233,
      "loss": 4.7366,
      "mean_token_accuracy": 0.25012079924345015,
      "num_tokens": 65482541.0,
      "step": 28555
    },
    {
      "entropy": 5.499741315841675,
      "epoch": 2.7435158501440924,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042520586574149423,
      "loss": 4.8323,
      "mean_token_accuracy": 0.23532106429338456,
      "num_tokens": 65494499.0,
      "step": 28560
    },
    {
      "entropy": 5.555266809463501,
      "epoch": 2.7439961575408263,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042518033863552185,
      "loss": 4.866,
      "mean_token_accuracy": 0.23328568041324615,
      "num_tokens": 65506377.0,
      "step": 28565
    },
    {
      "entropy": 5.4571826457977295,
      "epoch": 2.74447646493756,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004251548080428081,
      "loss": 4.8222,
      "mean_token_accuracy": 0.23736741840839387,
      "num_tokens": 65517476.0,
      "step": 28570
    },
    {
      "entropy": 5.430370903015136,
      "epoch": 2.744956772334294,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004251292739639455,
      "loss": 4.7382,
      "mean_token_accuracy": 0.24124278426170348,
      "num_tokens": 65529829.0,
      "step": 28575
    },
    {
      "entropy": 5.316912078857422,
      "epoch": 2.745437079731028,
      "grad_norm": 1.125,
      "learning_rate": 0.00042510373639952694,
      "loss": 4.638,
      "mean_token_accuracy": 0.25318251848220824,
      "num_tokens": 65541030.0,
      "step": 28580
    },
    {
      "entropy": 5.4001201629638675,
      "epoch": 2.745917387127762,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042507819535014547,
      "loss": 4.7411,
      "mean_token_accuracy": 0.2467976599931717,
      "num_tokens": 65551727.0,
      "step": 28585
    },
    {
      "entropy": 5.362022018432617,
      "epoch": 2.7463976945244957,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042505265081639376,
      "loss": 4.7245,
      "mean_token_accuracy": 0.25286721885204316,
      "num_tokens": 65563085.0,
      "step": 28590
    },
    {
      "entropy": 5.391195392608642,
      "epoch": 2.7468780019212296,
      "grad_norm": 1.125,
      "learning_rate": 0.0004250271027988652,
      "loss": 4.7245,
      "mean_token_accuracy": 0.24974887669086457,
      "num_tokens": 65574909.0,
      "step": 28595
    },
    {
      "entropy": 5.38725790977478,
      "epoch": 2.7473583093179634,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042500155129815274,
      "loss": 4.7091,
      "mean_token_accuracy": 0.2546834394335747,
      "num_tokens": 65585515.0,
      "step": 28600
    },
    {
      "entropy": 5.3609028339385985,
      "epoch": 2.7478386167146973,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042497599631484965,
      "loss": 4.713,
      "mean_token_accuracy": 0.24348908811807632,
      "num_tokens": 65598160.0,
      "step": 28605
    },
    {
      "entropy": 5.428983211517334,
      "epoch": 2.748318924111431,
      "grad_norm": 1.125,
      "learning_rate": 0.00042495043784954926,
      "loss": 4.769,
      "mean_token_accuracy": 0.24007164537906647,
      "num_tokens": 65609886.0,
      "step": 28610
    },
    {
      "entropy": 5.41954460144043,
      "epoch": 2.748799231508165,
      "grad_norm": 1.15625,
      "learning_rate": 0.000424924875902845,
      "loss": 4.7336,
      "mean_token_accuracy": 0.24491978883743287,
      "num_tokens": 65620256.0,
      "step": 28615
    },
    {
      "entropy": 5.4435615062713625,
      "epoch": 2.749279538904899,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004248993104753303,
      "loss": 4.7665,
      "mean_token_accuracy": 0.24866771101951599,
      "num_tokens": 65631108.0,
      "step": 28620
    },
    {
      "entropy": 5.429860973358155,
      "epoch": 2.7497598463016333,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004248737415675987,
      "loss": 4.7565,
      "mean_token_accuracy": 0.2473422884941101,
      "num_tokens": 65642423.0,
      "step": 28625
    },
    {
      "entropy": 5.372056770324707,
      "epoch": 2.7502401536983667,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004248481691802439,
      "loss": 4.7469,
      "mean_token_accuracy": 0.24539847522974015,
      "num_tokens": 65654105.0,
      "step": 28630
    },
    {
      "entropy": 5.477809000015259,
      "epoch": 2.750720461095101,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004248225933138595,
      "loss": 4.7997,
      "mean_token_accuracy": 0.24738354682922364,
      "num_tokens": 65664738.0,
      "step": 28635
    },
    {
      "entropy": 5.327704811096192,
      "epoch": 2.751200768491835,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042479701396903945,
      "loss": 4.6737,
      "mean_token_accuracy": 0.2477193668484688,
      "num_tokens": 65676815.0,
      "step": 28640
    },
    {
      "entropy": 5.454016876220703,
      "epoch": 2.751681075888569,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004247714311463775,
      "loss": 4.7876,
      "mean_token_accuracy": 0.24802256375551224,
      "num_tokens": 65687234.0,
      "step": 28645
    },
    {
      "entropy": 5.480003118515015,
      "epoch": 2.7521613832853027,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042474584484646766,
      "loss": 4.8588,
      "mean_token_accuracy": 0.23649338632822037,
      "num_tokens": 65700356.0,
      "step": 28650
    },
    {
      "entropy": 5.326114463806152,
      "epoch": 2.7526416906820366,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004247202550699039,
      "loss": 4.5907,
      "mean_token_accuracy": 0.2592442288994789,
      "num_tokens": 65711320.0,
      "step": 28655
    },
    {
      "entropy": 5.390331792831421,
      "epoch": 2.7531219980787704,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004246946618172805,
      "loss": 4.7763,
      "mean_token_accuracy": 0.24326636493206025,
      "num_tokens": 65722461.0,
      "step": 28660
    },
    {
      "entropy": 5.351155090332031,
      "epoch": 2.7536023054755043,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004246690650891915,
      "loss": 4.6424,
      "mean_token_accuracy": 0.2501160204410553,
      "num_tokens": 65732311.0,
      "step": 28665
    },
    {
      "entropy": 5.433306121826172,
      "epoch": 2.754082612872238,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004246434648862312,
      "loss": 4.7682,
      "mean_token_accuracy": 0.24781240075826644,
      "num_tokens": 65743830.0,
      "step": 28670
    },
    {
      "entropy": 5.44104495048523,
      "epoch": 2.754562920268972,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004246178612089941,
      "loss": 4.816,
      "mean_token_accuracy": 0.23382602632045746,
      "num_tokens": 65754851.0,
      "step": 28675
    },
    {
      "entropy": 5.482121133804322,
      "epoch": 2.755043227665706,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004245922540580744,
      "loss": 4.818,
      "mean_token_accuracy": 0.24116883873939515,
      "num_tokens": 65765228.0,
      "step": 28680
    },
    {
      "entropy": 5.478459644317627,
      "epoch": 2.75552353506244,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004245666434340668,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2316098541021347,
      "num_tokens": 65775334.0,
      "step": 28685
    },
    {
      "entropy": 5.3326962947845455,
      "epoch": 2.7560038424591737,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004245410293375659,
      "loss": 4.6664,
      "mean_token_accuracy": 0.26340180188417434,
      "num_tokens": 65786256.0,
      "step": 28690
    },
    {
      "entropy": 5.412775754928589,
      "epoch": 2.7564841498559076,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004245154117691664,
      "loss": 4.7244,
      "mean_token_accuracy": 0.2506538763642311,
      "num_tokens": 65798831.0,
      "step": 28695
    },
    {
      "entropy": 5.403091669082642,
      "epoch": 2.756964457252642,
      "grad_norm": 1.125,
      "learning_rate": 0.0004244897907294628,
      "loss": 4.7676,
      "mean_token_accuracy": 0.24207725673913955,
      "num_tokens": 65810256.0,
      "step": 28700
    },
    {
      "entropy": 5.565428829193115,
      "epoch": 2.7574447646493754,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004244641662190504,
      "loss": 4.9103,
      "mean_token_accuracy": 0.2355375900864601,
      "num_tokens": 65821752.0,
      "step": 28705
    },
    {
      "entropy": 5.499311923980713,
      "epoch": 2.7579250720461097,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042443853823852376,
      "loss": 4.8054,
      "mean_token_accuracy": 0.23901659548282622,
      "num_tokens": 65833056.0,
      "step": 28710
    },
    {
      "entropy": 5.433590841293335,
      "epoch": 2.7584053794428436,
      "grad_norm": 1.140625,
      "learning_rate": 0.000424412906788478,
      "loss": 4.7077,
      "mean_token_accuracy": 0.24740938246250152,
      "num_tokens": 65845081.0,
      "step": 28715
    },
    {
      "entropy": 5.427774810791016,
      "epoch": 2.7588856868395775,
      "grad_norm": 1.25,
      "learning_rate": 0.0004243872718695082,
      "loss": 4.752,
      "mean_token_accuracy": 0.2413619726896286,
      "num_tokens": 65856024.0,
      "step": 28720
    },
    {
      "entropy": 5.375940275192261,
      "epoch": 2.7593659942363113,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042436163348220956,
      "loss": 4.6474,
      "mean_token_accuracy": 0.258437180519104,
      "num_tokens": 65868282.0,
      "step": 28725
    },
    {
      "entropy": 5.337905406951904,
      "epoch": 2.7598463016330452,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004243359916271773,
      "loss": 4.717,
      "mean_token_accuracy": 0.24722606986761092,
      "num_tokens": 65879835.0,
      "step": 28730
    },
    {
      "entropy": 5.3398253440856935,
      "epoch": 2.760326609029779,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004243103463050067,
      "loss": 4.6873,
      "mean_token_accuracy": 0.24480577409267426,
      "num_tokens": 65890916.0,
      "step": 28735
    },
    {
      "entropy": 5.444711494445801,
      "epoch": 2.760806916426513,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004242846975162933,
      "loss": 4.7775,
      "mean_token_accuracy": 0.2406430274248123,
      "num_tokens": 65901716.0,
      "step": 28740
    },
    {
      "entropy": 5.456648826599121,
      "epoch": 2.761287223823247,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042425904526163246,
      "loss": 4.8048,
      "mean_token_accuracy": 0.2417793557047844,
      "num_tokens": 65914366.0,
      "step": 28745
    },
    {
      "entropy": 5.527831649780273,
      "epoch": 2.7617675312199808,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004242333895416198,
      "loss": 4.9162,
      "mean_token_accuracy": 0.2347020298242569,
      "num_tokens": 65926396.0,
      "step": 28750
    },
    {
      "entropy": 5.413573980331421,
      "epoch": 2.7622478386167146,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000424207730356851,
      "loss": 4.7334,
      "mean_token_accuracy": 0.2480783686041832,
      "num_tokens": 65939200.0,
      "step": 28755
    },
    {
      "entropy": 5.4263811111450195,
      "epoch": 2.7627281460134485,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004241820677079218,
      "loss": 4.7553,
      "mean_token_accuracy": 0.2441684350371361,
      "num_tokens": 65948978.0,
      "step": 28760
    },
    {
      "entropy": 5.417643976211548,
      "epoch": 2.7632084534101824,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042415640159542783,
      "loss": 4.6949,
      "mean_token_accuracy": 0.2520849481225014,
      "num_tokens": 65959661.0,
      "step": 28765
    },
    {
      "entropy": 5.342396211624146,
      "epoch": 2.7636887608069163,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004241307320199652,
      "loss": 4.6145,
      "mean_token_accuracy": 0.26022802740335466,
      "num_tokens": 65970208.0,
      "step": 28770
    },
    {
      "entropy": 5.404763269424438,
      "epoch": 2.7641690682036506,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004241050589821298,
      "loss": 4.7744,
      "mean_token_accuracy": 0.24365601539611817,
      "num_tokens": 65982606.0,
      "step": 28775
    },
    {
      "entropy": 5.3715637683868405,
      "epoch": 2.764649375600384,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004240793824825177,
      "loss": 4.6565,
      "mean_token_accuracy": 0.2501645922660828,
      "num_tokens": 65993984.0,
      "step": 28780
    },
    {
      "entropy": 5.471664333343506,
      "epoch": 2.7651296829971184,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042405370252172496,
      "loss": 4.7378,
      "mean_token_accuracy": 0.24825451970100404,
      "num_tokens": 66004740.0,
      "step": 28785
    },
    {
      "entropy": 5.427179622650146,
      "epoch": 2.765609990393852,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004240280191003479,
      "loss": 4.8082,
      "mean_token_accuracy": 0.24536763727664948,
      "num_tokens": 66019265.0,
      "step": 28790
    },
    {
      "entropy": 5.416748523712158,
      "epoch": 2.766090297790586,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004240023322189828,
      "loss": 4.7143,
      "mean_token_accuracy": 0.25034407526254654,
      "num_tokens": 66030434.0,
      "step": 28795
    },
    {
      "entropy": 5.453929328918457,
      "epoch": 2.76657060518732,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004239766418782258,
      "loss": 4.7326,
      "mean_token_accuracy": 0.23904861360788346,
      "num_tokens": 66041100.0,
      "step": 28800
    },
    {
      "entropy": 5.380207633972168,
      "epoch": 2.767050912584054,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004239509480786737,
      "loss": 4.6996,
      "mean_token_accuracy": 0.24596095830202103,
      "num_tokens": 66053046.0,
      "step": 28805
    },
    {
      "entropy": 5.479794979095459,
      "epoch": 2.7675312199807878,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042392525082092286,
      "loss": 4.8548,
      "mean_token_accuracy": 0.23605933487415315,
      "num_tokens": 66063508.0,
      "step": 28810
    },
    {
      "entropy": 5.493244934082031,
      "epoch": 2.7680115273775217,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004238995501055699,
      "loss": 4.8455,
      "mean_token_accuracy": 0.23872555941343307,
      "num_tokens": 66076135.0,
      "step": 28815
    },
    {
      "entropy": 5.435422420501709,
      "epoch": 2.7684918347742555,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004238738459332115,
      "loss": 4.801,
      "mean_token_accuracy": 0.23495250344276428,
      "num_tokens": 66087565.0,
      "step": 28820
    },
    {
      "entropy": 5.422157430648804,
      "epoch": 2.7689721421709894,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004238481383044445,
      "loss": 4.7981,
      "mean_token_accuracy": 0.24723170101642608,
      "num_tokens": 66099430.0,
      "step": 28825
    },
    {
      "entropy": 5.407952547073364,
      "epoch": 2.7694524495677233,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042382242721986573,
      "loss": 4.7058,
      "mean_token_accuracy": 0.24221314638853073,
      "num_tokens": 66111060.0,
      "step": 28830
    },
    {
      "entropy": 5.288201093673706,
      "epoch": 2.769932756964457,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042379671268007207,
      "loss": 4.6857,
      "mean_token_accuracy": 0.25341973453760147,
      "num_tokens": 66123367.0,
      "step": 28835
    },
    {
      "entropy": 5.41419358253479,
      "epoch": 2.770413064361191,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004237709946856607,
      "loss": 4.6961,
      "mean_token_accuracy": 0.2573114261031151,
      "num_tokens": 66134198.0,
      "step": 28840
    },
    {
      "entropy": 5.446699333190918,
      "epoch": 2.770893371757925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042374527323722836,
      "loss": 4.7312,
      "mean_token_accuracy": 0.24385600239038469,
      "num_tokens": 66145991.0,
      "step": 28845
    },
    {
      "entropy": 5.481243085861206,
      "epoch": 2.771373679154659,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042371954833537263,
      "loss": 4.7715,
      "mean_token_accuracy": 0.24091633558273315,
      "num_tokens": 66157447.0,
      "step": 28850
    },
    {
      "entropy": 5.365038251876831,
      "epoch": 2.7718539865513927,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042369381998069055,
      "loss": 4.785,
      "mean_token_accuracy": 0.24732874929904938,
      "num_tokens": 66169259.0,
      "step": 28855
    },
    {
      "entropy": 5.398494005203247,
      "epoch": 2.772334293948127,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004236680881737795,
      "loss": 4.7359,
      "mean_token_accuracy": 0.24609700888395308,
      "num_tokens": 66181221.0,
      "step": 28860
    },
    {
      "entropy": 5.437147092819214,
      "epoch": 2.7728146013448605,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004236423529152369,
      "loss": 4.7773,
      "mean_token_accuracy": 0.24274158328771592,
      "num_tokens": 66193513.0,
      "step": 28865
    },
    {
      "entropy": 5.385577011108398,
      "epoch": 2.773294908741595,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004236166142056602,
      "loss": 4.6516,
      "mean_token_accuracy": 0.25319864153862,
      "num_tokens": 66204951.0,
      "step": 28870
    },
    {
      "entropy": 5.349120712280273,
      "epoch": 2.7737752161383287,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004235908720456471,
      "loss": 4.7332,
      "mean_token_accuracy": 0.25508856773376465,
      "num_tokens": 66215845.0,
      "step": 28875
    },
    {
      "entropy": 5.453661680221558,
      "epoch": 2.7742555235350626,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004235651264357951,
      "loss": 4.8625,
      "mean_token_accuracy": 0.2371811032295227,
      "num_tokens": 66226510.0,
      "step": 28880
    },
    {
      "entropy": 5.448530673980713,
      "epoch": 2.7747358309317964,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042353937737670206,
      "loss": 4.7575,
      "mean_token_accuracy": 0.243757925927639,
      "num_tokens": 66238134.0,
      "step": 28885
    },
    {
      "entropy": 5.453515148162841,
      "epoch": 2.7752161383285303,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004235136248689658,
      "loss": 4.7599,
      "mean_token_accuracy": 0.2401899665594101,
      "num_tokens": 66250526.0,
      "step": 28890
    },
    {
      "entropy": 5.500604581832886,
      "epoch": 2.775696445725264,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004234878689131841,
      "loss": 4.8121,
      "mean_token_accuracy": 0.24055665582418442,
      "num_tokens": 66262144.0,
      "step": 28895
    },
    {
      "entropy": 5.495979499816895,
      "epoch": 2.776176753121998,
      "grad_norm": 1.1875,
      "learning_rate": 0.000423462109509955,
      "loss": 4.8001,
      "mean_token_accuracy": 0.239180126786232,
      "num_tokens": 66273262.0,
      "step": 28900
    },
    {
      "entropy": 5.367092895507812,
      "epoch": 2.776657060518732,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004234363466598765,
      "loss": 4.7205,
      "mean_token_accuracy": 0.25420398712158204,
      "num_tokens": 66284127.0,
      "step": 28905
    },
    {
      "entropy": 5.383769607543945,
      "epoch": 2.777137367915466,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042341058036354687,
      "loss": 4.7372,
      "mean_token_accuracy": 0.24137601852416993,
      "num_tokens": 66295092.0,
      "step": 28910
    },
    {
      "entropy": 5.318414163589478,
      "epoch": 2.7776176753121997,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042338481062156424,
      "loss": 4.6029,
      "mean_token_accuracy": 0.25552314817905425,
      "num_tokens": 66306050.0,
      "step": 28915
    },
    {
      "entropy": 5.356041336059571,
      "epoch": 2.7780979827089336,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042335903743452694,
      "loss": 4.6917,
      "mean_token_accuracy": 0.24492626637220383,
      "num_tokens": 66317186.0,
      "step": 28920
    },
    {
      "entropy": 5.420587587356567,
      "epoch": 2.7785782901056675,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004233332608030333,
      "loss": 4.7717,
      "mean_token_accuracy": 0.24382143914699556,
      "num_tokens": 66328486.0,
      "step": 28925
    },
    {
      "entropy": 5.380883979797363,
      "epoch": 2.7790585975024014,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042330748072768183,
      "loss": 4.6715,
      "mean_token_accuracy": 0.2545857459306717,
      "num_tokens": 66340649.0,
      "step": 28930
    },
    {
      "entropy": 5.46318826675415,
      "epoch": 2.7795389048991357,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000423281697209071,
      "loss": 4.7951,
      "mean_token_accuracy": 0.23817380517721176,
      "num_tokens": 66353555.0,
      "step": 28935
    },
    {
      "entropy": 5.496905279159546,
      "epoch": 2.780019212295869,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004232559102477995,
      "loss": 4.8508,
      "mean_token_accuracy": 0.23567385077476502,
      "num_tokens": 66365445.0,
      "step": 28940
    },
    {
      "entropy": 5.390019702911377,
      "epoch": 2.7804995196926034,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000423230119844466,
      "loss": 4.6782,
      "mean_token_accuracy": 0.2517347872257233,
      "num_tokens": 66377203.0,
      "step": 28945
    },
    {
      "entropy": 5.395397090911866,
      "epoch": 2.7809798270893373,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004232043259996692,
      "loss": 4.7954,
      "mean_token_accuracy": 0.2419469103217125,
      "num_tokens": 66388637.0,
      "step": 28950
    },
    {
      "entropy": 5.410062503814697,
      "epoch": 2.781460134486071,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004231785287140081,
      "loss": 4.7852,
      "mean_token_accuracy": 0.24235988408327103,
      "num_tokens": 66400470.0,
      "step": 28955
    },
    {
      "entropy": 5.5018415451049805,
      "epoch": 2.781940441882805,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004231527279880816,
      "loss": 4.8069,
      "mean_token_accuracy": 0.2501769706606865,
      "num_tokens": 66411079.0,
      "step": 28960
    },
    {
      "entropy": 5.41577262878418,
      "epoch": 2.782420749279539,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004231269238224885,
      "loss": 4.7613,
      "mean_token_accuracy": 0.2417920634150505,
      "num_tokens": 66423706.0,
      "step": 28965
    },
    {
      "entropy": 5.336153507232666,
      "epoch": 2.782901056676273,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004231011162178282,
      "loss": 4.7164,
      "mean_token_accuracy": 0.24876337200403215,
      "num_tokens": 66434994.0,
      "step": 28970
    },
    {
      "entropy": 5.361925411224365,
      "epoch": 2.7833813640730067,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004230753051746998,
      "loss": 4.6343,
      "mean_token_accuracy": 0.25579093396663666,
      "num_tokens": 66446474.0,
      "step": 28975
    },
    {
      "entropy": 5.420636940002441,
      "epoch": 2.7838616714697406,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042304949069370246,
      "loss": 4.7154,
      "mean_token_accuracy": 0.24502642899751664,
      "num_tokens": 66457633.0,
      "step": 28980
    },
    {
      "entropy": 5.40522575378418,
      "epoch": 2.7843419788664745,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042302367277543553,
      "loss": 4.735,
      "mean_token_accuracy": 0.24755038022994996,
      "num_tokens": 66468160.0,
      "step": 28985
    },
    {
      "entropy": 5.397738981246948,
      "epoch": 2.7848222862632084,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042299785142049855,
      "loss": 4.7385,
      "mean_token_accuracy": 0.24192909449338912,
      "num_tokens": 66478609.0,
      "step": 28990
    },
    {
      "entropy": 5.425030374526978,
      "epoch": 2.7853025936599423,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004229720266294908,
      "loss": 4.7921,
      "mean_token_accuracy": 0.24385026544332505,
      "num_tokens": 66489992.0,
      "step": 28995
    },
    {
      "entropy": 5.475899934768677,
      "epoch": 2.785782901056676,
      "grad_norm": 1.15625,
      "learning_rate": 0.000422946198403012,
      "loss": 4.8228,
      "mean_token_accuracy": 0.2419847369194031,
      "num_tokens": 66501331.0,
      "step": 29000
    },
    {
      "entropy": 5.527988529205322,
      "epoch": 2.78626320845341,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004229203667416619,
      "loss": 4.8129,
      "mean_token_accuracy": 0.23928232342004777,
      "num_tokens": 66513336.0,
      "step": 29005
    },
    {
      "entropy": 5.38966736793518,
      "epoch": 2.7867435158501443,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042289453164604,
      "loss": 4.6608,
      "mean_token_accuracy": 0.2544807270169258,
      "num_tokens": 66523577.0,
      "step": 29010
    },
    {
      "entropy": 5.343083763122559,
      "epoch": 2.787223823246878,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004228686931167463,
      "loss": 4.7773,
      "mean_token_accuracy": 0.23747317045927047,
      "num_tokens": 66536181.0,
      "step": 29015
    },
    {
      "entropy": 5.387795114517212,
      "epoch": 2.787704130643612,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004228428511543806,
      "loss": 4.7258,
      "mean_token_accuracy": 0.24689608216285705,
      "num_tokens": 66546881.0,
      "step": 29020
    },
    {
      "entropy": 5.492980909347534,
      "epoch": 2.7881844380403455,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042281700575954283,
      "loss": 4.8548,
      "mean_token_accuracy": 0.24495659470558168,
      "num_tokens": 66559091.0,
      "step": 29025
    },
    {
      "entropy": 5.461927223205566,
      "epoch": 2.78866474543708,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004227911569328332,
      "loss": 4.8337,
      "mean_token_accuracy": 0.23905434161424638,
      "num_tokens": 66570584.0,
      "step": 29030
    },
    {
      "entropy": 5.4512686252594,
      "epoch": 2.7891450528338138,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004227653046748517,
      "loss": 4.7351,
      "mean_token_accuracy": 0.24474625140428544,
      "num_tokens": 66581921.0,
      "step": 29035
    },
    {
      "entropy": 5.441592502593994,
      "epoch": 2.7896253602305476,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042273944898619864,
      "loss": 4.8437,
      "mean_token_accuracy": 0.2405219316482544,
      "num_tokens": 66595132.0,
      "step": 29040
    },
    {
      "entropy": 5.4853252410888675,
      "epoch": 2.7901056676272815,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042271358986747427,
      "loss": 4.7955,
      "mean_token_accuracy": 0.24156619161367415,
      "num_tokens": 66607212.0,
      "step": 29045
    },
    {
      "entropy": 5.420843935012817,
      "epoch": 2.7905859750240154,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042268772731927895,
      "loss": 4.7547,
      "mean_token_accuracy": 0.246026349067688,
      "num_tokens": 66619502.0,
      "step": 29050
    },
    {
      "entropy": 5.427176427841187,
      "epoch": 2.7910662824207493,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042266186134221317,
      "loss": 4.7996,
      "mean_token_accuracy": 0.2348610818386078,
      "num_tokens": 66631813.0,
      "step": 29055
    },
    {
      "entropy": 5.4740111351013185,
      "epoch": 2.791546589817483,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004226359919368774,
      "loss": 4.8489,
      "mean_token_accuracy": 0.2355217680335045,
      "num_tokens": 66643879.0,
      "step": 29060
    },
    {
      "entropy": 5.516980409622192,
      "epoch": 2.792026897214217,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042261011910387224,
      "loss": 4.8339,
      "mean_token_accuracy": 0.2409004345536232,
      "num_tokens": 66656819.0,
      "step": 29065
    },
    {
      "entropy": 5.447045230865479,
      "epoch": 2.792507204610951,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004225842428437985,
      "loss": 4.6958,
      "mean_token_accuracy": 0.2489951729774475,
      "num_tokens": 66668745.0,
      "step": 29070
    },
    {
      "entropy": 5.391724729537964,
      "epoch": 2.792987512007685,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042255836315725694,
      "loss": 4.734,
      "mean_token_accuracy": 0.24833835363388063,
      "num_tokens": 66681620.0,
      "step": 29075
    },
    {
      "entropy": 5.403437948226928,
      "epoch": 2.7934678194044187,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004225324800448483,
      "loss": 4.6904,
      "mean_token_accuracy": 0.24848663061857224,
      "num_tokens": 66692232.0,
      "step": 29080
    },
    {
      "entropy": 5.379767942428589,
      "epoch": 2.793948126801153,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042250659350717343,
      "loss": 4.6974,
      "mean_token_accuracy": 0.24950201362371444,
      "num_tokens": 66703401.0,
      "step": 29085
    },
    {
      "entropy": 5.340684127807617,
      "epoch": 2.7944284341978864,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042248070354483354,
      "loss": 4.6457,
      "mean_token_accuracy": 0.25329732000827787,
      "num_tokens": 66714066.0,
      "step": 29090
    },
    {
      "entropy": 5.303261661529541,
      "epoch": 2.7949087415946208,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004224548101584297,
      "loss": 4.5995,
      "mean_token_accuracy": 0.2547901630401611,
      "num_tokens": 66724504.0,
      "step": 29095
    },
    {
      "entropy": 5.392637538909912,
      "epoch": 2.795389048991354,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000422428913348563,
      "loss": 4.7228,
      "mean_token_accuracy": 0.2395992875099182,
      "num_tokens": 66735772.0,
      "step": 29100
    },
    {
      "entropy": 5.439428043365479,
      "epoch": 2.7958693563880885,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004224030131158346,
      "loss": 4.7366,
      "mean_token_accuracy": 0.24921173006296157,
      "num_tokens": 66747956.0,
      "step": 29105
    },
    {
      "entropy": 5.426474857330322,
      "epoch": 2.7963496637848224,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004223771094608461,
      "loss": 4.7906,
      "mean_token_accuracy": 0.2460456073284149,
      "num_tokens": 66759112.0,
      "step": 29110
    },
    {
      "entropy": 5.429778051376343,
      "epoch": 2.7968299711815563,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004223512023841986,
      "loss": 4.7204,
      "mean_token_accuracy": 0.24944740235805513,
      "num_tokens": 66770927.0,
      "step": 29115
    },
    {
      "entropy": 5.455029678344727,
      "epoch": 2.79731027857829,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042232529188649374,
      "loss": 4.8269,
      "mean_token_accuracy": 0.24201827496290207,
      "num_tokens": 66781809.0,
      "step": 29120
    },
    {
      "entropy": 5.363722419738769,
      "epoch": 2.797790585975024,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004222993779683331,
      "loss": 4.7201,
      "mean_token_accuracy": 0.24183619171380996,
      "num_tokens": 66793495.0,
      "step": 29125
    },
    {
      "entropy": 5.400586938858032,
      "epoch": 2.798270893371758,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042227346063031837,
      "loss": 4.7462,
      "mean_token_accuracy": 0.2440822958946228,
      "num_tokens": 66804568.0,
      "step": 29130
    },
    {
      "entropy": 5.398395586013794,
      "epoch": 2.798751200768492,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004222475398730511,
      "loss": 4.686,
      "mean_token_accuracy": 0.2488371893763542,
      "num_tokens": 66816698.0,
      "step": 29135
    },
    {
      "entropy": 5.4423322677612305,
      "epoch": 2.7992315081652257,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004222216156971332,
      "loss": 4.7967,
      "mean_token_accuracy": 0.23946560621261598,
      "num_tokens": 66829299.0,
      "step": 29140
    },
    {
      "entropy": 5.330298995971679,
      "epoch": 2.7997118155619596,
      "grad_norm": 1.125,
      "learning_rate": 0.00042219568810316656,
      "loss": 4.602,
      "mean_token_accuracy": 0.25220405161380766,
      "num_tokens": 66839871.0,
      "step": 29145
    },
    {
      "entropy": 5.52785005569458,
      "epoch": 2.8001921229586935,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004221697570917531,
      "loss": 4.8082,
      "mean_token_accuracy": 0.2420360803604126,
      "num_tokens": 66850368.0,
      "step": 29150
    },
    {
      "entropy": 5.477698278427124,
      "epoch": 2.8006724303554273,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000422143822663495,
      "loss": 4.7891,
      "mean_token_accuracy": 0.24027273058891296,
      "num_tokens": 66861067.0,
      "step": 29155
    },
    {
      "entropy": 5.422284650802612,
      "epoch": 2.8011527377521612,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004221178848189941,
      "loss": 4.8389,
      "mean_token_accuracy": 0.23698890805244446,
      "num_tokens": 66873841.0,
      "step": 29160
    },
    {
      "entropy": 5.4304474830627445,
      "epoch": 2.801633045148895,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042209194355885283,
      "loss": 4.7864,
      "mean_token_accuracy": 0.24147802740335464,
      "num_tokens": 66885042.0,
      "step": 29165
    },
    {
      "entropy": 5.437592935562134,
      "epoch": 2.8021133525456294,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004220659988836734,
      "loss": 4.7694,
      "mean_token_accuracy": 0.24245473891496658,
      "num_tokens": 66897076.0,
      "step": 29170
    },
    {
      "entropy": 5.39712347984314,
      "epoch": 2.802593659942363,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004220400507940582,
      "loss": 4.6786,
      "mean_token_accuracy": 0.24912838339805604,
      "num_tokens": 66907879.0,
      "step": 29175
    },
    {
      "entropy": 5.4613447189331055,
      "epoch": 2.803073967339097,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042201409929060955,
      "loss": 4.824,
      "mean_token_accuracy": 0.2400746762752533,
      "num_tokens": 66920801.0,
      "step": 29180
    },
    {
      "entropy": 5.429881048202515,
      "epoch": 2.803554274735831,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004219881443739301,
      "loss": 4.6832,
      "mean_token_accuracy": 0.24397042393684387,
      "num_tokens": 66931411.0,
      "step": 29185
    },
    {
      "entropy": 5.426124095916748,
      "epoch": 2.804034582132565,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004219621860446225,
      "loss": 4.7615,
      "mean_token_accuracy": 0.24425334930419923,
      "num_tokens": 66942122.0,
      "step": 29190
    },
    {
      "entropy": 5.422949743270874,
      "epoch": 2.804514889529299,
      "grad_norm": 1.125,
      "learning_rate": 0.0004219362243032892,
      "loss": 4.7915,
      "mean_token_accuracy": 0.24012758284807206,
      "num_tokens": 66954096.0,
      "step": 29195
    },
    {
      "entropy": 5.488507175445557,
      "epoch": 2.8049951969260327,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042191025915053323,
      "loss": 4.824,
      "mean_token_accuracy": 0.23366222381591797,
      "num_tokens": 66965729.0,
      "step": 29200
    },
    {
      "entropy": 5.485562944412232,
      "epoch": 2.8054755043227666,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042188429058695714,
      "loss": 4.7701,
      "mean_token_accuracy": 0.236957086622715,
      "num_tokens": 66975819.0,
      "step": 29205
    },
    {
      "entropy": 5.36932864189148,
      "epoch": 2.8059558117195005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042185831861316406,
      "loss": 4.7099,
      "mean_token_accuracy": 0.2562666967511177,
      "num_tokens": 66987150.0,
      "step": 29210
    },
    {
      "entropy": 5.385476112365723,
      "epoch": 2.8064361191162344,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004218323432297568,
      "loss": 4.7164,
      "mean_token_accuracy": 0.2380199044942856,
      "num_tokens": 66999449.0,
      "step": 29215
    },
    {
      "entropy": 5.37556300163269,
      "epoch": 2.8069164265129682,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042180636443733864,
      "loss": 4.6929,
      "mean_token_accuracy": 0.2526826798915863,
      "num_tokens": 67011256.0,
      "step": 29220
    },
    {
      "entropy": 5.266582870483399,
      "epoch": 2.807396733909702,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042178038223651253,
      "loss": 4.588,
      "mean_token_accuracy": 0.2542863473296165,
      "num_tokens": 67021878.0,
      "step": 29225
    },
    {
      "entropy": 5.422073602676392,
      "epoch": 2.807877041306436,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042175439662788195,
      "loss": 4.8649,
      "mean_token_accuracy": 0.232040473818779,
      "num_tokens": 67032544.0,
      "step": 29230
    },
    {
      "entropy": 5.415950727462769,
      "epoch": 2.80835734870317,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042172840761204986,
      "loss": 4.749,
      "mean_token_accuracy": 0.2454808086156845,
      "num_tokens": 67042119.0,
      "step": 29235
    },
    {
      "entropy": 5.467488479614258,
      "epoch": 2.8088376560999038,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004217024151896199,
      "loss": 4.7909,
      "mean_token_accuracy": 0.25075800716876984,
      "num_tokens": 67053126.0,
      "step": 29240
    },
    {
      "entropy": 5.489912462234497,
      "epoch": 2.809317963496638,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042167641936119557,
      "loss": 4.846,
      "mean_token_accuracy": 0.23864096999168397,
      "num_tokens": 67063818.0,
      "step": 29245
    },
    {
      "entropy": 5.3805726051330565,
      "epoch": 2.8097982708933715,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004216504201273802,
      "loss": 4.7553,
      "mean_token_accuracy": 0.24466656744480134,
      "num_tokens": 67074477.0,
      "step": 29250
    },
    {
      "entropy": 5.447819375991822,
      "epoch": 2.810278578290106,
      "grad_norm": 1.125,
      "learning_rate": 0.0004216244174887776,
      "loss": 4.7943,
      "mean_token_accuracy": 0.2410410612821579,
      "num_tokens": 67085920.0,
      "step": 29255
    },
    {
      "entropy": 5.432564306259155,
      "epoch": 2.8107588856868397,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042159841144599145,
      "loss": 4.8184,
      "mean_token_accuracy": 0.24728379100561143,
      "num_tokens": 67097542.0,
      "step": 29260
    },
    {
      "entropy": 5.452751922607422,
      "epoch": 2.8112391930835736,
      "grad_norm": 1.25,
      "learning_rate": 0.00042157240199962537,
      "loss": 4.7975,
      "mean_token_accuracy": 0.24653913229703903,
      "num_tokens": 67108514.0,
      "step": 29265
    },
    {
      "entropy": 5.4052787780761715,
      "epoch": 2.8117195004803075,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004215463891502834,
      "loss": 4.647,
      "mean_token_accuracy": 0.250359995663166,
      "num_tokens": 67120263.0,
      "step": 29270
    },
    {
      "entropy": 5.4615801811218265,
      "epoch": 2.8121998078770414,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042152037289856954,
      "loss": 4.7429,
      "mean_token_accuracy": 0.24356600046157836,
      "num_tokens": 67133607.0,
      "step": 29275
    },
    {
      "entropy": 5.418210124969482,
      "epoch": 2.8126801152737753,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042149435324508755,
      "loss": 4.7196,
      "mean_token_accuracy": 0.25232277661561964,
      "num_tokens": 67145452.0,
      "step": 29280
    },
    {
      "entropy": 5.367259550094604,
      "epoch": 2.813160422670509,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004214683301904417,
      "loss": 4.743,
      "mean_token_accuracy": 0.24590889513492584,
      "num_tokens": 67159178.0,
      "step": 29285
    },
    {
      "entropy": 5.5308629989624025,
      "epoch": 2.813640730067243,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042144230373523624,
      "loss": 4.8482,
      "mean_token_accuracy": 0.23237972408533097,
      "num_tokens": 67171652.0,
      "step": 29290
    },
    {
      "entropy": 5.443948125839233,
      "epoch": 2.814121037463977,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004214162738800753,
      "loss": 4.7627,
      "mean_token_accuracy": 0.24827366024255754,
      "num_tokens": 67181844.0,
      "step": 29295
    },
    {
      "entropy": 5.408216953277588,
      "epoch": 2.814601344860711,
      "grad_norm": 1.125,
      "learning_rate": 0.0004213902406255632,
      "loss": 4.8054,
      "mean_token_accuracy": 0.2505427226424217,
      "num_tokens": 67193156.0,
      "step": 29300
    },
    {
      "entropy": 5.395747470855713,
      "epoch": 2.8150816522574447,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004213642039723044,
      "loss": 4.7487,
      "mean_token_accuracy": 0.24489939212799072,
      "num_tokens": 67205447.0,
      "step": 29305
    },
    {
      "entropy": 5.5281157970428465,
      "epoch": 2.8155619596541785,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042133816392090343,
      "loss": 4.828,
      "mean_token_accuracy": 0.2396499052643776,
      "num_tokens": 67216101.0,
      "step": 29310
    },
    {
      "entropy": 5.34780445098877,
      "epoch": 2.8160422670509124,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042131212047196484,
      "loss": 4.5766,
      "mean_token_accuracy": 0.25987788289785385,
      "num_tokens": 67226972.0,
      "step": 29315
    },
    {
      "entropy": 5.362637710571289,
      "epoch": 2.8165225744476468,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042128607362609317,
      "loss": 4.7348,
      "mean_token_accuracy": 0.2497376948595047,
      "num_tokens": 67237968.0,
      "step": 29320
    },
    {
      "entropy": 5.427807140350342,
      "epoch": 2.81700288184438,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042126002338389336,
      "loss": 4.8243,
      "mean_token_accuracy": 0.2441770017147064,
      "num_tokens": 67250817.0,
      "step": 29325
    },
    {
      "entropy": 5.49323878288269,
      "epoch": 2.8174831892411145,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042123396974597007,
      "loss": 4.7273,
      "mean_token_accuracy": 0.24442493617534639,
      "num_tokens": 67261491.0,
      "step": 29330
    },
    {
      "entropy": 5.399051094055176,
      "epoch": 2.817963496637848,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042120791271292823,
      "loss": 4.7596,
      "mean_token_accuracy": 0.2479822129011154,
      "num_tokens": 67272684.0,
      "step": 29335
    },
    {
      "entropy": 5.438144683837891,
      "epoch": 2.8184438040345823,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042118185228537283,
      "loss": 4.7681,
      "mean_token_accuracy": 0.2413681700825691,
      "num_tokens": 67284070.0,
      "step": 29340
    },
    {
      "entropy": 5.462632513046264,
      "epoch": 2.818924111431316,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042115578846390884,
      "loss": 4.8318,
      "mean_token_accuracy": 0.23746369928121566,
      "num_tokens": 67296302.0,
      "step": 29345
    },
    {
      "entropy": 5.458406925201416,
      "epoch": 2.81940441882805,
      "grad_norm": 1.125,
      "learning_rate": 0.0004211297212491414,
      "loss": 4.7807,
      "mean_token_accuracy": 0.24666290134191513,
      "num_tokens": 67307123.0,
      "step": 29350
    },
    {
      "entropy": 5.46578803062439,
      "epoch": 2.819884726224784,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004211036506416759,
      "loss": 4.7569,
      "mean_token_accuracy": 0.2459060326218605,
      "num_tokens": 67318832.0,
      "step": 29355
    },
    {
      "entropy": 5.434204959869385,
      "epoch": 2.820365033621518,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004210775766421173,
      "loss": 4.8414,
      "mean_token_accuracy": 0.2388100042939186,
      "num_tokens": 67329821.0,
      "step": 29360
    },
    {
      "entropy": 5.38563346862793,
      "epoch": 2.8208453410182517,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004210514992510713,
      "loss": 4.7316,
      "mean_token_accuracy": 0.24560409039258957,
      "num_tokens": 67341569.0,
      "step": 29365
    },
    {
      "entropy": 5.510251760482788,
      "epoch": 2.8213256484149856,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004210254184691431,
      "loss": 4.8316,
      "mean_token_accuracy": 0.24096498936414718,
      "num_tokens": 67352499.0,
      "step": 29370
    },
    {
      "entropy": 5.409310245513916,
      "epoch": 2.8218059558117194,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042099933429693814,
      "loss": 4.7154,
      "mean_token_accuracy": 0.24735905230045319,
      "num_tokens": 67363810.0,
      "step": 29375
    },
    {
      "entropy": 5.46429181098938,
      "epoch": 2.8222862632084533,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004209732467350624,
      "loss": 4.8318,
      "mean_token_accuracy": 0.24332145750522613,
      "num_tokens": 67374690.0,
      "step": 29380
    },
    {
      "entropy": 5.432605266571045,
      "epoch": 2.822766570605187,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004209471557841212,
      "loss": 4.7548,
      "mean_token_accuracy": 0.2427100121974945,
      "num_tokens": 67386297.0,
      "step": 29385
    },
    {
      "entropy": 5.396324968338012,
      "epoch": 2.823246878001921,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004209210614447204,
      "loss": 4.7041,
      "mean_token_accuracy": 0.24563615769147873,
      "num_tokens": 67397544.0,
      "step": 29390
    },
    {
      "entropy": 5.446240854263306,
      "epoch": 2.8237271853986554,
      "grad_norm": 1.03125,
      "learning_rate": 0.000420894963717466,
      "loss": 4.7642,
      "mean_token_accuracy": 0.2444387272000313,
      "num_tokens": 67408609.0,
      "step": 29395
    },
    {
      "entropy": 5.413319396972656,
      "epoch": 2.824207492795389,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004208688626029636,
      "loss": 4.7488,
      "mean_token_accuracy": 0.2409954771399498,
      "num_tokens": 67421091.0,
      "step": 29400
    },
    {
      "entropy": 5.432141923904419,
      "epoch": 2.824687800192123,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004208427581018194,
      "loss": 4.7568,
      "mean_token_accuracy": 0.24445250630378723,
      "num_tokens": 67433141.0,
      "step": 29405
    },
    {
      "entropy": 5.463276052474976,
      "epoch": 2.8251681075888566,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004208166502146394,
      "loss": 4.8197,
      "mean_token_accuracy": 0.2371688187122345,
      "num_tokens": 67443646.0,
      "step": 29410
    },
    {
      "entropy": 5.444390201568604,
      "epoch": 2.825648414985591,
      "grad_norm": 1.125,
      "learning_rate": 0.00042079053894202977,
      "loss": 4.7683,
      "mean_token_accuracy": 0.24503573179244995,
      "num_tokens": 67455783.0,
      "step": 29415
    },
    {
      "entropy": 5.4559227466583256,
      "epoch": 2.826128722382325,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004207644242845968,
      "loss": 4.732,
      "mean_token_accuracy": 0.24811291843652725,
      "num_tokens": 67467185.0,
      "step": 29420
    },
    {
      "entropy": 5.43085412979126,
      "epoch": 2.8266090297790587,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004207383062429467,
      "loss": 4.7398,
      "mean_token_accuracy": 0.24474017173051835,
      "num_tokens": 67478740.0,
      "step": 29425
    },
    {
      "entropy": 5.3660296440124515,
      "epoch": 2.8270893371757926,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004207121848176858,
      "loss": 4.7044,
      "mean_token_accuracy": 0.2538249909877777,
      "num_tokens": 67490492.0,
      "step": 29430
    },
    {
      "entropy": 5.361283874511718,
      "epoch": 2.8275696445725265,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042068606000942075,
      "loss": 4.6854,
      "mean_token_accuracy": 0.24867427945137024,
      "num_tokens": 67500861.0,
      "step": 29435
    },
    {
      "entropy": 5.424371290206909,
      "epoch": 2.8280499519692603,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042065993181875794,
      "loss": 4.7298,
      "mean_token_accuracy": 0.2438322901725769,
      "num_tokens": 67513577.0,
      "step": 29440
    },
    {
      "entropy": 5.417195606231689,
      "epoch": 2.8285302593659942,
      "grad_norm": 1.078125,
      "learning_rate": 0.000420633800246304,
      "loss": 4.7614,
      "mean_token_accuracy": 0.2491031050682068,
      "num_tokens": 67525147.0,
      "step": 29445
    },
    {
      "entropy": 5.410862922668457,
      "epoch": 2.829010566762728,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042060766529266577,
      "loss": 4.7537,
      "mean_token_accuracy": 0.24213724434375763,
      "num_tokens": 67536183.0,
      "step": 29450
    },
    {
      "entropy": 5.475987339019776,
      "epoch": 2.829490874159462,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042058152695844986,
      "loss": 4.758,
      "mean_token_accuracy": 0.2445371478796005,
      "num_tokens": 67546868.0,
      "step": 29455
    },
    {
      "entropy": 5.378584146499634,
      "epoch": 2.829971181556196,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042055538524426317,
      "loss": 4.6963,
      "mean_token_accuracy": 0.24906175434589387,
      "num_tokens": 67557243.0,
      "step": 29460
    },
    {
      "entropy": 5.375512552261353,
      "epoch": 2.8304514889529298,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004205292401507127,
      "loss": 4.749,
      "mean_token_accuracy": 0.24055611491203308,
      "num_tokens": 67568977.0,
      "step": 29465
    },
    {
      "entropy": 5.418300580978394,
      "epoch": 2.8309317963496636,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004205030916784053,
      "loss": 4.6971,
      "mean_token_accuracy": 0.24951165169477463,
      "num_tokens": 67581318.0,
      "step": 29470
    },
    {
      "entropy": 5.395383739471436,
      "epoch": 2.8314121037463975,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042047693982794824,
      "loss": 4.6706,
      "mean_token_accuracy": 0.2534430056810379,
      "num_tokens": 67592707.0,
      "step": 29475
    },
    {
      "entropy": 5.439543437957764,
      "epoch": 2.831892411143132,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042045078459994854,
      "loss": 4.755,
      "mean_token_accuracy": 0.24567740261554719,
      "num_tokens": 67602929.0,
      "step": 29480
    },
    {
      "entropy": 5.449347448348999,
      "epoch": 2.8323727185398653,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004204246259950136,
      "loss": 4.7507,
      "mean_token_accuracy": 0.24754078686237335,
      "num_tokens": 67613049.0,
      "step": 29485
    },
    {
      "entropy": 5.532525444030762,
      "epoch": 2.8328530259365996,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042039846401375065,
      "loss": 4.8429,
      "mean_token_accuracy": 0.23725205659866333,
      "num_tokens": 67623355.0,
      "step": 29490
    },
    {
      "entropy": 5.48448543548584,
      "epoch": 2.8333333333333335,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042037229865676714,
      "loss": 4.8315,
      "mean_token_accuracy": 0.24353027045726777,
      "num_tokens": 67634607.0,
      "step": 29495
    },
    {
      "entropy": 5.426722717285156,
      "epoch": 2.8338136407300674,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042034612992467046,
      "loss": 4.7281,
      "mean_token_accuracy": 0.2489796385169029,
      "num_tokens": 67646549.0,
      "step": 29500
    },
    {
      "entropy": 5.3744007587432865,
      "epoch": 2.8342939481268012,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004203199578180683,
      "loss": 4.6368,
      "mean_token_accuracy": 0.2562882736325264,
      "num_tokens": 67657785.0,
      "step": 29505
    },
    {
      "entropy": 5.432777214050293,
      "epoch": 2.834774255523535,
      "grad_norm": 1.125,
      "learning_rate": 0.0004202937823375682,
      "loss": 4.735,
      "mean_token_accuracy": 0.24833301603794097,
      "num_tokens": 67670415.0,
      "step": 29510
    },
    {
      "entropy": 5.4596200466156,
      "epoch": 2.835254562920269,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004202676034837779,
      "loss": 4.7496,
      "mean_token_accuracy": 0.2446795806288719,
      "num_tokens": 67682815.0,
      "step": 29515
    },
    {
      "entropy": 5.4494414806365965,
      "epoch": 2.835734870317003,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004202414212573052,
      "loss": 4.7463,
      "mean_token_accuracy": 0.24467545002698898,
      "num_tokens": 67693820.0,
      "step": 29520
    },
    {
      "entropy": 5.395065259933472,
      "epoch": 2.8362151777137368,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042021523565875796,
      "loss": 4.7707,
      "mean_token_accuracy": 0.23746475279331208,
      "num_tokens": 67705375.0,
      "step": 29525
    },
    {
      "entropy": 5.4599040985107425,
      "epoch": 2.8366954851104706,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004201890466887442,
      "loss": 4.7928,
      "mean_token_accuracy": 0.2412624552845955,
      "num_tokens": 67717549.0,
      "step": 29530
    },
    {
      "entropy": 5.430511808395385,
      "epoch": 2.8371757925072045,
      "grad_norm": 1.25,
      "learning_rate": 0.0004201628543478718,
      "loss": 4.7497,
      "mean_token_accuracy": 0.24554066956043244,
      "num_tokens": 67727462.0,
      "step": 29535
    },
    {
      "entropy": 5.374365043640137,
      "epoch": 2.8376560999039384,
      "grad_norm": 1.078125,
      "learning_rate": 0.000420136658636749,
      "loss": 4.7088,
      "mean_token_accuracy": 0.24623846113681794,
      "num_tokens": 67740059.0,
      "step": 29540
    },
    {
      "entropy": 5.427841520309448,
      "epoch": 2.8381364073006723,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004201104595559841,
      "loss": 4.7747,
      "mean_token_accuracy": 0.24435027241706847,
      "num_tokens": 67751010.0,
      "step": 29545
    },
    {
      "entropy": 5.418402910232544,
      "epoch": 2.838616714697406,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042008425710618507,
      "loss": 4.7612,
      "mean_token_accuracy": 0.24297545701265336,
      "num_tokens": 67762131.0,
      "step": 29550
    },
    {
      "entropy": 5.423785209655762,
      "epoch": 2.8390970220941405,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042005805128796043,
      "loss": 4.7871,
      "mean_token_accuracy": 0.2514567658305168,
      "num_tokens": 67773759.0,
      "step": 29555
    },
    {
      "entropy": 5.373593187332153,
      "epoch": 2.839577329490874,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004200318421019186,
      "loss": 4.6607,
      "mean_token_accuracy": 0.2508962169289589,
      "num_tokens": 67785009.0,
      "step": 29560
    },
    {
      "entropy": 5.38088026046753,
      "epoch": 2.8400576368876083,
      "grad_norm": 1.109375,
      "learning_rate": 0.000420005629548668,
      "loss": 4.7584,
      "mean_token_accuracy": 0.24059962183237077,
      "num_tokens": 67796588.0,
      "step": 29565
    },
    {
      "entropy": 5.424646854400635,
      "epoch": 2.840537944284342,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041997941362881735,
      "loss": 4.711,
      "mean_token_accuracy": 0.2474783629179001,
      "num_tokens": 67808777.0,
      "step": 29570
    },
    {
      "entropy": 5.477227020263672,
      "epoch": 2.841018251681076,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004199531943429752,
      "loss": 4.7727,
      "mean_token_accuracy": 0.24559730887413025,
      "num_tokens": 67818562.0,
      "step": 29575
    },
    {
      "entropy": 5.373538970947266,
      "epoch": 2.84149855907781,
      "grad_norm": 1.125,
      "learning_rate": 0.0004199269716917502,
      "loss": 4.7049,
      "mean_token_accuracy": 0.2561283096671104,
      "num_tokens": 67829938.0,
      "step": 29580
    },
    {
      "entropy": 5.433379030227661,
      "epoch": 2.841978866474544,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004199007456757513,
      "loss": 4.7306,
      "mean_token_accuracy": 0.24385513365268707,
      "num_tokens": 67841735.0,
      "step": 29585
    },
    {
      "entropy": 5.362236356735229,
      "epoch": 2.8424591738712777,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041987451629558743,
      "loss": 4.693,
      "mean_token_accuracy": 0.24472609162330627,
      "num_tokens": 67853545.0,
      "step": 29590
    },
    {
      "entropy": 5.4084230899810795,
      "epoch": 2.8429394812680115,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004198482835518674,
      "loss": 4.6592,
      "mean_token_accuracy": 0.2502295419573784,
      "num_tokens": 67864763.0,
      "step": 29595
    },
    {
      "entropy": 5.464794778823853,
      "epoch": 2.8434197886647454,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004198220474452004,
      "loss": 4.8042,
      "mean_token_accuracy": 0.243648524582386,
      "num_tokens": 67875837.0,
      "step": 29600
    },
    {
      "entropy": 5.450630807876587,
      "epoch": 2.8439000960614793,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004197958079761954,
      "loss": 4.7997,
      "mean_token_accuracy": 0.24125282913446428,
      "num_tokens": 67887623.0,
      "step": 29605
    },
    {
      "entropy": 5.412113761901855,
      "epoch": 2.844380403458213,
      "grad_norm": 1.125,
      "learning_rate": 0.00041976956514546185,
      "loss": 4.7388,
      "mean_token_accuracy": 0.24959108382463455,
      "num_tokens": 67898883.0,
      "step": 29610
    },
    {
      "entropy": 5.451898241043091,
      "epoch": 2.844860710854947,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041974331895360873,
      "loss": 4.7286,
      "mean_token_accuracy": 0.25425905734300613,
      "num_tokens": 67908871.0,
      "step": 29615
    },
    {
      "entropy": 5.400686025619507,
      "epoch": 2.845341018251681,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004197170694012456,
      "loss": 4.7235,
      "mean_token_accuracy": 0.24521445482969284,
      "num_tokens": 67920812.0,
      "step": 29620
    },
    {
      "entropy": 5.363453149795532,
      "epoch": 2.845821325648415,
      "grad_norm": 1.25,
      "learning_rate": 0.0004196908164889818,
      "loss": 4.7224,
      "mean_token_accuracy": 0.24754569083452224,
      "num_tokens": 67931703.0,
      "step": 29625
    },
    {
      "entropy": 5.468993139266968,
      "epoch": 2.846301633045149,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004196645602174269,
      "loss": 4.7895,
      "mean_token_accuracy": 0.24322633743286132,
      "num_tokens": 67943423.0,
      "step": 29630
    },
    {
      "entropy": 5.480159711837769,
      "epoch": 2.8467819404418826,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041963830058719046,
      "loss": 4.7985,
      "mean_token_accuracy": 0.2401154011487961,
      "num_tokens": 67954047.0,
      "step": 29635
    },
    {
      "entropy": 5.431506729125976,
      "epoch": 2.847262247838617,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004196120375988822,
      "loss": 4.7337,
      "mean_token_accuracy": 0.24923163950443267,
      "num_tokens": 67964833.0,
      "step": 29640
    },
    {
      "entropy": 5.414096212387085,
      "epoch": 2.8477425552353504,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004195857712531119,
      "loss": 4.8155,
      "mean_token_accuracy": 0.24353648275136947,
      "num_tokens": 67977370.0,
      "step": 29645
    },
    {
      "entropy": 5.441700267791748,
      "epoch": 2.8482228626320847,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004195595015504892,
      "loss": 4.7238,
      "mean_token_accuracy": 0.2467266783118248,
      "num_tokens": 67988680.0,
      "step": 29650
    },
    {
      "entropy": 5.43659200668335,
      "epoch": 2.8487031700288186,
      "grad_norm": 1.25,
      "learning_rate": 0.00041953322849162415,
      "loss": 4.8052,
      "mean_token_accuracy": 0.24505287259817124,
      "num_tokens": 67999927.0,
      "step": 29655
    },
    {
      "entropy": 5.426180219650268,
      "epoch": 2.8491834774255524,
      "grad_norm": 1.125,
      "learning_rate": 0.0004195069520771268,
      "loss": 4.837,
      "mean_token_accuracy": 0.23717934340238572,
      "num_tokens": 68012176.0,
      "step": 29660
    },
    {
      "entropy": 5.4870744228363035,
      "epoch": 2.8496637848222863,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041948067230760706,
      "loss": 4.7381,
      "mean_token_accuracy": 0.24375637769699096,
      "num_tokens": 68023561.0,
      "step": 29665
    },
    {
      "entropy": 5.4885783195495605,
      "epoch": 2.85014409221902,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041945438918367513,
      "loss": 4.8416,
      "mean_token_accuracy": 0.23815546780824662,
      "num_tokens": 68035768.0,
      "step": 29670
    },
    {
      "entropy": 5.521748828887939,
      "epoch": 2.850624399615754,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041942810270594115,
      "loss": 4.8333,
      "mean_token_accuracy": 0.23708852380514145,
      "num_tokens": 68049414.0,
      "step": 29675
    },
    {
      "entropy": 5.405167484283448,
      "epoch": 2.851104707012488,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004194018128750157,
      "loss": 4.6593,
      "mean_token_accuracy": 0.2544316783547401,
      "num_tokens": 68060743.0,
      "step": 29680
    },
    {
      "entropy": 5.426904630661011,
      "epoch": 2.851585014409222,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041937551969150873,
      "loss": 4.7003,
      "mean_token_accuracy": 0.2503191947937012,
      "num_tokens": 68070525.0,
      "step": 29685
    },
    {
      "entropy": 5.461566209793091,
      "epoch": 2.8520653218059557,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000419349223156031,
      "loss": 4.7832,
      "mean_token_accuracy": 0.244477479159832,
      "num_tokens": 68082160.0,
      "step": 29690
    },
    {
      "entropy": 5.374125146865845,
      "epoch": 2.8525456292026896,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004193229232691929,
      "loss": 4.7004,
      "mean_token_accuracy": 0.24467044919729233,
      "num_tokens": 68092473.0,
      "step": 29695
    },
    {
      "entropy": 5.277084016799927,
      "epoch": 2.8530259365994235,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041929662003160504,
      "loss": 4.6963,
      "mean_token_accuracy": 0.2558184489607811,
      "num_tokens": 68103820.0,
      "step": 29700
    },
    {
      "entropy": 5.332514381408691,
      "epoch": 2.8535062439961574,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041927031344387824,
      "loss": 4.6275,
      "mean_token_accuracy": 0.261365482211113,
      "num_tokens": 68116364.0,
      "step": 29705
    },
    {
      "entropy": 5.299054384231567,
      "epoch": 2.8539865513928913,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041924400350662304,
      "loss": 4.5776,
      "mean_token_accuracy": 0.25915678292512895,
      "num_tokens": 68127970.0,
      "step": 29710
    },
    {
      "entropy": 5.449429893493653,
      "epoch": 2.8544668587896256,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041921769022045045,
      "loss": 4.8044,
      "mean_token_accuracy": 0.23869973719120025,
      "num_tokens": 68139433.0,
      "step": 29715
    },
    {
      "entropy": 5.39155330657959,
      "epoch": 2.854947166186359,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00041919137358597137,
      "loss": 4.7158,
      "mean_token_accuracy": 0.24109620749950408,
      "num_tokens": 68153169.0,
      "step": 29720
    },
    {
      "entropy": 5.453806304931641,
      "epoch": 2.8554274735830933,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004191650536037967,
      "loss": 4.8065,
      "mean_token_accuracy": 0.23663766086101531,
      "num_tokens": 68163899.0,
      "step": 29725
    },
    {
      "entropy": 5.381336641311646,
      "epoch": 2.8559077809798272,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041913873027453756,
      "loss": 4.6552,
      "mean_token_accuracy": 0.252064049243927,
      "num_tokens": 68175342.0,
      "step": 29730
    },
    {
      "entropy": 5.410021591186523,
      "epoch": 2.856388088376561,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041911240359880517,
      "loss": 4.8082,
      "mean_token_accuracy": 0.24117662012577057,
      "num_tokens": 68186646.0,
      "step": 29735
    },
    {
      "entropy": 5.352937841415406,
      "epoch": 2.856868395773295,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041908607357721067,
      "loss": 4.6698,
      "mean_token_accuracy": 0.25147605538368223,
      "num_tokens": 68197464.0,
      "step": 29740
    },
    {
      "entropy": 5.289333248138428,
      "epoch": 2.857348703170029,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041905974021036533,
      "loss": 4.6333,
      "mean_token_accuracy": 0.256196328997612,
      "num_tokens": 68208846.0,
      "step": 29745
    },
    {
      "entropy": 5.491367435455322,
      "epoch": 2.8578290105667628,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041903340349888065,
      "loss": 4.8242,
      "mean_token_accuracy": 0.2464098036289215,
      "num_tokens": 68220273.0,
      "step": 29750
    },
    {
      "entropy": 5.53825798034668,
      "epoch": 2.8583093179634966,
      "grad_norm": 1.09375,
      "learning_rate": 0.000419007063443368,
      "loss": 4.9039,
      "mean_token_accuracy": 0.23608268648386002,
      "num_tokens": 68233871.0,
      "step": 29755
    },
    {
      "entropy": 5.424506998062133,
      "epoch": 2.8587896253602305,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041898072004443906,
      "loss": 4.7489,
      "mean_token_accuracy": 0.24402574598789215,
      "num_tokens": 68246155.0,
      "step": 29760
    },
    {
      "entropy": 5.419758033752442,
      "epoch": 2.8592699327569644,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004189543733027052,
      "loss": 4.7049,
      "mean_token_accuracy": 0.25870115011930467,
      "num_tokens": 68258201.0,
      "step": 29765
    },
    {
      "entropy": 5.377692413330078,
      "epoch": 2.8597502401536983,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004189280232187783,
      "loss": 4.6764,
      "mean_token_accuracy": 0.252983433008194,
      "num_tokens": 68270106.0,
      "step": 29770
    },
    {
      "entropy": 5.397055768966675,
      "epoch": 2.860230547550432,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004189016697932701,
      "loss": 4.7071,
      "mean_token_accuracy": 0.25147671699523927,
      "num_tokens": 68282549.0,
      "step": 29775
    },
    {
      "entropy": 5.421273422241211,
      "epoch": 2.860710854947166,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004188753130267924,
      "loss": 4.7603,
      "mean_token_accuracy": 0.24655990302562714,
      "num_tokens": 68294141.0,
      "step": 29780
    },
    {
      "entropy": 5.469683980941772,
      "epoch": 2.8611911623439,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004188489529199572,
      "loss": 4.7844,
      "mean_token_accuracy": 0.24677526205778122,
      "num_tokens": 68305520.0,
      "step": 29785
    },
    {
      "entropy": 5.377097034454346,
      "epoch": 2.8616714697406342,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041882258947337637,
      "loss": 4.7471,
      "mean_token_accuracy": 0.24819574803113936,
      "num_tokens": 68317906.0,
      "step": 29790
    },
    {
      "entropy": 5.4489552021026615,
      "epoch": 2.8621517771373677,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041879622268766207,
      "loss": 4.7768,
      "mean_token_accuracy": 0.24333668649196624,
      "num_tokens": 68328998.0,
      "step": 29795
    },
    {
      "entropy": 5.3898285865783695,
      "epoch": 2.862632084534102,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004187698525634266,
      "loss": 4.7576,
      "mean_token_accuracy": 0.2446745216846466,
      "num_tokens": 68341329.0,
      "step": 29800
    },
    {
      "entropy": 5.448543643951416,
      "epoch": 2.863112391930836,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041874347910128193,
      "loss": 4.7971,
      "mean_token_accuracy": 0.2376049906015396,
      "num_tokens": 68353791.0,
      "step": 29805
    },
    {
      "entropy": 5.422285413742065,
      "epoch": 2.8635926993275698,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004187171023018406,
      "loss": 4.7393,
      "mean_token_accuracy": 0.2493459552526474,
      "num_tokens": 68365493.0,
      "step": 29810
    },
    {
      "entropy": 5.36162633895874,
      "epoch": 2.8640730067243036,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041869072216571486,
      "loss": 4.6653,
      "mean_token_accuracy": 0.2532081097364426,
      "num_tokens": 68377658.0,
      "step": 29815
    },
    {
      "entropy": 5.423202800750732,
      "epoch": 2.8645533141210375,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041866433869351715,
      "loss": 4.6899,
      "mean_token_accuracy": 0.2523887425661087,
      "num_tokens": 68387710.0,
      "step": 29820
    },
    {
      "entropy": 5.492438220977784,
      "epoch": 2.8650336215177714,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004186379518858602,
      "loss": 4.8327,
      "mean_token_accuracy": 0.24215269833803177,
      "num_tokens": 68400726.0,
      "step": 29825
    },
    {
      "entropy": 5.539453744888306,
      "epoch": 2.8655139289145053,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004186115617433565,
      "loss": 4.891,
      "mean_token_accuracy": 0.23457389026880265,
      "num_tokens": 68413414.0,
      "step": 29830
    },
    {
      "entropy": 5.3896770000457765,
      "epoch": 2.865994236311239,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041858516826661876,
      "loss": 4.6749,
      "mean_token_accuracy": 0.25051148384809496,
      "num_tokens": 68424834.0,
      "step": 29835
    },
    {
      "entropy": 5.337043190002442,
      "epoch": 2.866474543707973,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041855877145625974,
      "loss": 4.6323,
      "mean_token_accuracy": 0.2545389398932457,
      "num_tokens": 68437220.0,
      "step": 29840
    },
    {
      "entropy": 5.404065275192261,
      "epoch": 2.866954851104707,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004185323713128924,
      "loss": 4.8296,
      "mean_token_accuracy": 0.23973434120416642,
      "num_tokens": 68449571.0,
      "step": 29845
    },
    {
      "entropy": 5.464690542221069,
      "epoch": 2.867435158501441,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041850596783712956,
      "loss": 4.7052,
      "mean_token_accuracy": 0.2444482907652855,
      "num_tokens": 68460592.0,
      "step": 29850
    },
    {
      "entropy": 5.432982397079468,
      "epoch": 2.8679154658981747,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004184795610295843,
      "loss": 4.7175,
      "mean_token_accuracy": 0.2512087881565094,
      "num_tokens": 68472268.0,
      "step": 29855
    },
    {
      "entropy": 5.383265113830566,
      "epoch": 2.8683957732949086,
      "grad_norm": 1.125,
      "learning_rate": 0.0004184531508908697,
      "loss": 4.7293,
      "mean_token_accuracy": 0.24083735495805741,
      "num_tokens": 68483382.0,
      "step": 29860
    },
    {
      "entropy": 5.4212806701660154,
      "epoch": 2.868876080691643,
      "grad_norm": 1.125,
      "learning_rate": 0.0004184267374215989,
      "loss": 4.7631,
      "mean_token_accuracy": 0.24688544273376464,
      "num_tokens": 68494421.0,
      "step": 29865
    },
    {
      "entropy": 5.464725685119629,
      "epoch": 2.8693563880883763,
      "grad_norm": 1.09375,
      "learning_rate": 0.000418400320622385,
      "loss": 4.7872,
      "mean_token_accuracy": 0.2413398653268814,
      "num_tokens": 68505810.0,
      "step": 29870
    },
    {
      "entropy": 5.348456811904907,
      "epoch": 2.8698366954851107,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004183739004938416,
      "loss": 4.5934,
      "mean_token_accuracy": 0.26091232150793076,
      "num_tokens": 68515632.0,
      "step": 29875
    },
    {
      "entropy": 5.408326816558838,
      "epoch": 2.870317002881844,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000418347477036582,
      "loss": 4.8238,
      "mean_token_accuracy": 0.24088083058595658,
      "num_tokens": 68526930.0,
      "step": 29880
    },
    {
      "entropy": 5.381545448303223,
      "epoch": 2.8707973102785784,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041832105025121956,
      "loss": 4.6854,
      "mean_token_accuracy": 0.2487562909722328,
      "num_tokens": 68537381.0,
      "step": 29885
    },
    {
      "entropy": 5.332199382781982,
      "epoch": 2.8712776176753123,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004182946201383679,
      "loss": 4.6175,
      "mean_token_accuracy": 0.2520546421408653,
      "num_tokens": 68548982.0,
      "step": 29890
    },
    {
      "entropy": 5.366804790496826,
      "epoch": 2.871757925072046,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004182681866986407,
      "loss": 4.6914,
      "mean_token_accuracy": 0.24447359889745712,
      "num_tokens": 68560572.0,
      "step": 29895
    },
    {
      "entropy": 5.433857536315918,
      "epoch": 2.87223823246878,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041824174993265165,
      "loss": 4.7497,
      "mean_token_accuracy": 0.24484139829874038,
      "num_tokens": 68572080.0,
      "step": 29900
    },
    {
      "entropy": 5.387609434127808,
      "epoch": 2.872718539865514,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041821530984101444,
      "loss": 4.711,
      "mean_token_accuracy": 0.24759194999933243,
      "num_tokens": 68584481.0,
      "step": 29905
    },
    {
      "entropy": 5.3896496295928955,
      "epoch": 2.873198847262248,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000418188866424343,
      "loss": 4.7061,
      "mean_token_accuracy": 0.2581990256905556,
      "num_tokens": 68595673.0,
      "step": 29910
    },
    {
      "entropy": 5.468475294113159,
      "epoch": 2.8736791546589817,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004181624196832513,
      "loss": 4.797,
      "mean_token_accuracy": 0.250538270175457,
      "num_tokens": 68607310.0,
      "step": 29915
    },
    {
      "entropy": 5.3739830493927006,
      "epoch": 2.8741594620557156,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041813596961835336,
      "loss": 4.7469,
      "mean_token_accuracy": 0.24565428644418716,
      "num_tokens": 68618579.0,
      "step": 29920
    },
    {
      "entropy": 5.324251508712768,
      "epoch": 2.8746397694524495,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041810951623026313,
      "loss": 4.607,
      "mean_token_accuracy": 0.25831740349531174,
      "num_tokens": 68629977.0,
      "step": 29925
    },
    {
      "entropy": 5.426117420196533,
      "epoch": 2.8751200768491834,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041808305951959496,
      "loss": 4.813,
      "mean_token_accuracy": 0.23945355117321016,
      "num_tokens": 68641740.0,
      "step": 29930
    },
    {
      "entropy": 5.410346126556396,
      "epoch": 2.8756003842459172,
      "grad_norm": 1.265625,
      "learning_rate": 0.000418056599486963,
      "loss": 4.7773,
      "mean_token_accuracy": 0.243172986805439,
      "num_tokens": 68653171.0,
      "step": 29935
    },
    {
      "entropy": 5.4464842796325685,
      "epoch": 2.8760806916426516,
      "grad_norm": 1.125,
      "learning_rate": 0.0004180301361329816,
      "loss": 4.7789,
      "mean_token_accuracy": 0.24666168689727783,
      "num_tokens": 68664658.0,
      "step": 29940
    },
    {
      "entropy": 5.413475131988525,
      "epoch": 2.876560999039385,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004180036694582651,
      "loss": 4.7131,
      "mean_token_accuracy": 0.2558938667178154,
      "num_tokens": 68675912.0,
      "step": 29945
    },
    {
      "entropy": 5.4056652069091795,
      "epoch": 2.8770413064361193,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041797719946342813,
      "loss": 4.7155,
      "mean_token_accuracy": 0.25067547857761385,
      "num_tokens": 68686586.0,
      "step": 29950
    },
    {
      "entropy": 5.4253888607025145,
      "epoch": 2.8775216138328528,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041795072614908503,
      "loss": 4.7662,
      "mean_token_accuracy": 0.23997806161642074,
      "num_tokens": 68698280.0,
      "step": 29955
    },
    {
      "entropy": 5.4065611362457275,
      "epoch": 2.878001921229587,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041792424951585055,
      "loss": 4.7227,
      "mean_token_accuracy": 0.2562417447566986,
      "num_tokens": 68709424.0,
      "step": 29960
    },
    {
      "entropy": 5.433400392532349,
      "epoch": 2.878482228626321,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041789776956433947,
      "loss": 4.7752,
      "mean_token_accuracy": 0.24879423528909683,
      "num_tokens": 68720785.0,
      "step": 29965
    },
    {
      "entropy": 5.3845641136169435,
      "epoch": 2.878962536023055,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041787128629516645,
      "loss": 4.7186,
      "mean_token_accuracy": 0.24488111436367035,
      "num_tokens": 68732157.0,
      "step": 29970
    },
    {
      "entropy": 5.371813535690308,
      "epoch": 2.8794428434197887,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004178447997089464,
      "loss": 4.7328,
      "mean_token_accuracy": 0.24753082245588304,
      "num_tokens": 68743604.0,
      "step": 29975
    },
    {
      "entropy": 5.4154760360717775,
      "epoch": 2.8799231508165226,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004178183098062943,
      "loss": 4.7651,
      "mean_token_accuracy": 0.241828154027462,
      "num_tokens": 68755056.0,
      "step": 29980
    },
    {
      "entropy": 5.431168127059936,
      "epoch": 2.8804034582132565,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004177918165878251,
      "loss": 4.7717,
      "mean_token_accuracy": 0.2428354188799858,
      "num_tokens": 68766390.0,
      "step": 29985
    },
    {
      "entropy": 5.468379783630371,
      "epoch": 2.8808837656099904,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004177653200541539,
      "loss": 4.7686,
      "mean_token_accuracy": 0.24843352288007736,
      "num_tokens": 68778780.0,
      "step": 29990
    },
    {
      "entropy": 5.398436784744263,
      "epoch": 2.8813640730067243,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000417738820205896,
      "loss": 4.6619,
      "mean_token_accuracy": 0.24657977521419525,
      "num_tokens": 68789545.0,
      "step": 29995
    },
    {
      "entropy": 5.391963958740234,
      "epoch": 2.881844380403458,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004177123170436665,
      "loss": 4.6911,
      "mean_token_accuracy": 0.24974467605352402,
      "num_tokens": 68802231.0,
      "step": 30000
    },
    {
      "epoch": 2.881844380403458,
      "eval_entropy": 5.1520640982586485,
      "eval_loss": 4.813636302947998,
      "eval_mean_token_accuracy": 0.25195895420838743,
      "eval_num_tokens": 68802231.0,
      "eval_runtime": 20.6317,
      "eval_samples_per_second": 1590.51,
      "eval_steps_per_second": 198.82,
      "step": 30000
    },
    {
      "entropy": 5.514973735809326,
      "epoch": 2.882324687800192,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004176858105680807,
      "loss": 4.8302,
      "mean_token_accuracy": 0.24749020487070084,
      "num_tokens": 68812424.0,
      "step": 30005
    },
    {
      "entropy": 5.476418209075928,
      "epoch": 2.882804995196926,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041765930077975415,
      "loss": 4.7159,
      "mean_token_accuracy": 0.24599323123693467,
      "num_tokens": 68823770.0,
      "step": 30010
    },
    {
      "entropy": 5.3775125503540036,
      "epoch": 2.88328530259366,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041763278767930213,
      "loss": 4.7013,
      "mean_token_accuracy": 0.24772650748491287,
      "num_tokens": 68835007.0,
      "step": 30015
    },
    {
      "entropy": 5.343743562698364,
      "epoch": 2.8837656099903937,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004176062712673404,
      "loss": 4.6363,
      "mean_token_accuracy": 0.24789452850818633,
      "num_tokens": 68847070.0,
      "step": 30020
    },
    {
      "entropy": 5.405806016921997,
      "epoch": 2.884245917387128,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004175797515444845,
      "loss": 4.7187,
      "mean_token_accuracy": 0.24122971296310425,
      "num_tokens": 68857197.0,
      "step": 30025
    },
    {
      "entropy": 5.441985416412353,
      "epoch": 2.8847262247838614,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004175532285113501,
      "loss": 4.7498,
      "mean_token_accuracy": 0.24473426342010499,
      "num_tokens": 68869719.0,
      "step": 30030
    },
    {
      "entropy": 5.390156984329224,
      "epoch": 2.8852065321805958,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004175267021685531,
      "loss": 4.7388,
      "mean_token_accuracy": 0.24955329447984695,
      "num_tokens": 68880300.0,
      "step": 30035
    },
    {
      "entropy": 5.385481357574463,
      "epoch": 2.8856868395773296,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041750017251670926,
      "loss": 4.701,
      "mean_token_accuracy": 0.24855161607265472,
      "num_tokens": 68892616.0,
      "step": 30040
    },
    {
      "entropy": 5.516348886489868,
      "epoch": 2.8861671469740635,
      "grad_norm": 1.125,
      "learning_rate": 0.0004174736395564345,
      "loss": 4.8929,
      "mean_token_accuracy": 0.2365196630358696,
      "num_tokens": 68904595.0,
      "step": 30045
    },
    {
      "entropy": 5.442195749282837,
      "epoch": 2.8866474543707974,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041744710328834493,
      "loss": 4.7147,
      "mean_token_accuracy": 0.2529665887355804,
      "num_tokens": 68917036.0,
      "step": 30050
    },
    {
      "entropy": 5.431843042373657,
      "epoch": 2.8871277617675313,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041742056371305665,
      "loss": 4.7282,
      "mean_token_accuracy": 0.24536866396665574,
      "num_tokens": 68928216.0,
      "step": 30055
    },
    {
      "entropy": 5.390303230285644,
      "epoch": 2.887608069164265,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041739402083118576,
      "loss": 4.748,
      "mean_token_accuracy": 0.2452323853969574,
      "num_tokens": 68939806.0,
      "step": 30060
    },
    {
      "entropy": 5.400676155090332,
      "epoch": 2.888088376560999,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004173674746433485,
      "loss": 4.6825,
      "mean_token_accuracy": 0.2523895099759102,
      "num_tokens": 68950402.0,
      "step": 30065
    },
    {
      "entropy": 5.437553358078003,
      "epoch": 2.888568683957733,
      "grad_norm": 1.125,
      "learning_rate": 0.00041734092515016127,
      "loss": 4.7647,
      "mean_token_accuracy": 0.24629093706607819,
      "num_tokens": 68961472.0,
      "step": 30070
    },
    {
      "entropy": 5.4502684593200685,
      "epoch": 2.889048991354467,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041731437235224036,
      "loss": 4.8446,
      "mean_token_accuracy": 0.24003095775842667,
      "num_tokens": 68972844.0,
      "step": 30075
    },
    {
      "entropy": 5.449607849121094,
      "epoch": 2.8895292987512007,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004172878162502023,
      "loss": 4.7952,
      "mean_token_accuracy": 0.24318084716796876,
      "num_tokens": 68985219.0,
      "step": 30080
    },
    {
      "entropy": 5.460611581802368,
      "epoch": 2.8900096061479346,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041726125684466374,
      "loss": 4.8001,
      "mean_token_accuracy": 0.2430158481001854,
      "num_tokens": 68997243.0,
      "step": 30085
    },
    {
      "entropy": 5.513708686828613,
      "epoch": 2.8904899135446684,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004172346941362412,
      "loss": 4.8422,
      "mean_token_accuracy": 0.23593346923589706,
      "num_tokens": 69008325.0,
      "step": 30090
    },
    {
      "entropy": 5.540022563934326,
      "epoch": 2.8909702209414023,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041720812812555137,
      "loss": 4.8723,
      "mean_token_accuracy": 0.23139505684375763,
      "num_tokens": 69020203.0,
      "step": 30095
    },
    {
      "entropy": 5.402646160125732,
      "epoch": 2.8914505283381366,
      "grad_norm": 1.0,
      "learning_rate": 0.0004171815588132111,
      "loss": 4.7228,
      "mean_token_accuracy": 0.24998842924833298,
      "num_tokens": 69032538.0,
      "step": 30100
    },
    {
      "entropy": 5.375315856933594,
      "epoch": 2.89193083573487,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004171549861998372,
      "loss": 4.7411,
      "mean_token_accuracy": 0.2489866718649864,
      "num_tokens": 69045106.0,
      "step": 30105
    },
    {
      "entropy": 5.413784027099609,
      "epoch": 2.8924111431316044,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004171284102860467,
      "loss": 4.7423,
      "mean_token_accuracy": 0.24649261832237243,
      "num_tokens": 69056526.0,
      "step": 30110
    },
    {
      "entropy": 5.429532909393311,
      "epoch": 2.8928914505283383,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004171018310724565,
      "loss": 4.7311,
      "mean_token_accuracy": 0.24577226042747496,
      "num_tokens": 69067497.0,
      "step": 30115
    },
    {
      "entropy": 5.483668661117553,
      "epoch": 2.893371757925072,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004170752485596838,
      "loss": 4.7105,
      "mean_token_accuracy": 0.24604740142822265,
      "num_tokens": 69078114.0,
      "step": 30120
    },
    {
      "entropy": 5.389729833602905,
      "epoch": 2.893852065321806,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041704866274834557,
      "loss": 4.7,
      "mean_token_accuracy": 0.24967650920152665,
      "num_tokens": 69089329.0,
      "step": 30125
    },
    {
      "entropy": 5.486542558670044,
      "epoch": 2.89433237271854,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041702207363905933,
      "loss": 4.8231,
      "mean_token_accuracy": 0.2308259502053261,
      "num_tokens": 69101487.0,
      "step": 30130
    },
    {
      "entropy": 5.4422448635101315,
      "epoch": 2.894812680115274,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041699548123244216,
      "loss": 4.6967,
      "mean_token_accuracy": 0.24779887348413468,
      "num_tokens": 69112297.0,
      "step": 30135
    },
    {
      "entropy": 5.483095169067383,
      "epoch": 2.8952929875120077,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004169688855291116,
      "loss": 4.8172,
      "mean_token_accuracy": 0.24148597866296767,
      "num_tokens": 69123628.0,
      "step": 30140
    },
    {
      "entropy": 5.339601612091064,
      "epoch": 2.8957732949087416,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004169422865296851,
      "loss": 4.6204,
      "mean_token_accuracy": 0.24666374325752258,
      "num_tokens": 69135040.0,
      "step": 30145
    },
    {
      "entropy": 5.5082015037536625,
      "epoch": 2.8962536023054755,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004169156842347802,
      "loss": 4.8474,
      "mean_token_accuracy": 0.23879437744617463,
      "num_tokens": 69146251.0,
      "step": 30150
    },
    {
      "entropy": 5.491968154907227,
      "epoch": 2.8967339097022093,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004168890786450144,
      "loss": 4.8575,
      "mean_token_accuracy": 0.234991554915905,
      "num_tokens": 69157040.0,
      "step": 30155
    },
    {
      "entropy": 5.487174034118652,
      "epoch": 2.8972142170989432,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004168624697610056,
      "loss": 4.7321,
      "mean_token_accuracy": 0.2500568926334381,
      "num_tokens": 69167330.0,
      "step": 30160
    },
    {
      "entropy": 5.3733296394348145,
      "epoch": 2.897694524495677,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041683585758337156,
      "loss": 4.6696,
      "mean_token_accuracy": 0.2557193458080292,
      "num_tokens": 69177903.0,
      "step": 30165
    },
    {
      "entropy": 5.347247838973999,
      "epoch": 2.898174831892411,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004168092421127299,
      "loss": 4.6917,
      "mean_token_accuracy": 0.24884424805641175,
      "num_tokens": 69189946.0,
      "step": 30170
    },
    {
      "entropy": 5.328969955444336,
      "epoch": 2.8986551392891453,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004167826233496989,
      "loss": 4.6416,
      "mean_token_accuracy": 0.2610753834247589,
      "num_tokens": 69201393.0,
      "step": 30175
    },
    {
      "entropy": 5.436554479598999,
      "epoch": 2.8991354466858787,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004167560012948963,
      "loss": 4.7996,
      "mean_token_accuracy": 0.23789606988430023,
      "num_tokens": 69213894.0,
      "step": 30180
    },
    {
      "entropy": 5.4939323425292965,
      "epoch": 2.899615754082613,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041672937594894034,
      "loss": 4.772,
      "mean_token_accuracy": 0.24254297018051146,
      "num_tokens": 69224798.0,
      "step": 30185
    },
    {
      "entropy": 5.450915050506592,
      "epoch": 2.9000960614793465,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041670274731244903,
      "loss": 4.7518,
      "mean_token_accuracy": 0.24557366222143173,
      "num_tokens": 69235893.0,
      "step": 30190
    },
    {
      "entropy": 5.358152770996094,
      "epoch": 2.900576368876081,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004166761153860408,
      "loss": 4.6708,
      "mean_token_accuracy": 0.25425933599472045,
      "num_tokens": 69246554.0,
      "step": 30195
    },
    {
      "entropy": 5.371938753128052,
      "epoch": 2.9010566762728147,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041664948017033383,
      "loss": 4.7057,
      "mean_token_accuracy": 0.25192324817180634,
      "num_tokens": 69259232.0,
      "step": 30200
    },
    {
      "entropy": 5.3735052108764645,
      "epoch": 2.9015369836695486,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004166228416659465,
      "loss": 4.6246,
      "mean_token_accuracy": 0.25761575996875763,
      "num_tokens": 69270143.0,
      "step": 30205
    },
    {
      "entropy": 5.418178129196167,
      "epoch": 2.9020172910662825,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041659619987349734,
      "loss": 4.7492,
      "mean_token_accuracy": 0.24759024977684022,
      "num_tokens": 69280702.0,
      "step": 30210
    },
    {
      "entropy": 5.4352504253387455,
      "epoch": 2.9024975984630164,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041656955479360487,
      "loss": 4.71,
      "mean_token_accuracy": 0.24845624715089798,
      "num_tokens": 69292605.0,
      "step": 30215
    },
    {
      "entropy": 5.414093017578125,
      "epoch": 2.9029779058597502,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004165429064268877,
      "loss": 4.6875,
      "mean_token_accuracy": 0.25057590305805205,
      "num_tokens": 69303243.0,
      "step": 30220
    },
    {
      "entropy": 5.418157291412354,
      "epoch": 2.903458213256484,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004165162547739646,
      "loss": 4.7819,
      "mean_token_accuracy": 0.2414577156305313,
      "num_tokens": 69315059.0,
      "step": 30225
    },
    {
      "entropy": 5.4563154697418215,
      "epoch": 2.903938520653218,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004164895998354542,
      "loss": 4.8123,
      "mean_token_accuracy": 0.2354363039135933,
      "num_tokens": 69327338.0,
      "step": 30230
    },
    {
      "entropy": 5.427417516708374,
      "epoch": 2.904418828049952,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004164629416119755,
      "loss": 4.7212,
      "mean_token_accuracy": 0.244558684527874,
      "num_tokens": 69339955.0,
      "step": 30235
    },
    {
      "entropy": 5.430848789215088,
      "epoch": 2.9048991354466858,
      "grad_norm": 1.25,
      "learning_rate": 0.00041643628010414735,
      "loss": 4.7304,
      "mean_token_accuracy": 0.24757471531629563,
      "num_tokens": 69350847.0,
      "step": 30240
    },
    {
      "entropy": 5.417345905303955,
      "epoch": 2.9053794428434196,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041640961531258877,
      "loss": 4.7305,
      "mean_token_accuracy": 0.24630888551473618,
      "num_tokens": 69362066.0,
      "step": 30245
    },
    {
      "entropy": 5.369820070266724,
      "epoch": 2.905859750240154,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004163829472379187,
      "loss": 4.7421,
      "mean_token_accuracy": 0.24125708192586898,
      "num_tokens": 69373239.0,
      "step": 30250
    },
    {
      "entropy": 5.410001754760742,
      "epoch": 2.9063400576368874,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041635627588075655,
      "loss": 4.7277,
      "mean_token_accuracy": 0.24294416904449462,
      "num_tokens": 69384094.0,
      "step": 30255
    },
    {
      "entropy": 5.375732946395874,
      "epoch": 2.9068203650336217,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004163296012417213,
      "loss": 4.7011,
      "mean_token_accuracy": 0.2515264838933945,
      "num_tokens": 69394667.0,
      "step": 30260
    },
    {
      "entropy": 5.423192834854126,
      "epoch": 2.907300672430355,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041630292332143245,
      "loss": 4.7718,
      "mean_token_accuracy": 0.24744143337011337,
      "num_tokens": 69405146.0,
      "step": 30265
    },
    {
      "entropy": 5.4813817024230955,
      "epoch": 2.9077809798270895,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004162762421205093,
      "loss": 4.7919,
      "mean_token_accuracy": 0.24890727549791336,
      "num_tokens": 69416147.0,
      "step": 30270
    },
    {
      "entropy": 5.448138618469239,
      "epoch": 2.9082612872238234,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041624955763957134,
      "loss": 4.734,
      "mean_token_accuracy": 0.2509457141160965,
      "num_tokens": 69426065.0,
      "step": 30275
    },
    {
      "entropy": 5.391021108627319,
      "epoch": 2.9087415946205573,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004162228698792381,
      "loss": 4.7171,
      "mean_token_accuracy": 0.25232072323560717,
      "num_tokens": 69437266.0,
      "step": 30280
    },
    {
      "entropy": 5.410219430923462,
      "epoch": 2.909221902017291,
      "grad_norm": 1.125,
      "learning_rate": 0.00041619617884012904,
      "loss": 4.7708,
      "mean_token_accuracy": 0.2525736287236214,
      "num_tokens": 69449539.0,
      "step": 30285
    },
    {
      "entropy": 5.468535852432251,
      "epoch": 2.909702209414025,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004161694845228641,
      "loss": 4.8132,
      "mean_token_accuracy": 0.2367036372423172,
      "num_tokens": 69460515.0,
      "step": 30290
    },
    {
      "entropy": 5.484281587600708,
      "epoch": 2.910182516810759,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004161427869280628,
      "loss": 4.8415,
      "mean_token_accuracy": 0.24684972018003465,
      "num_tokens": 69472997.0,
      "step": 30295
    },
    {
      "entropy": 5.459004306793213,
      "epoch": 2.910662824207493,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041611608605634517,
      "loss": 4.8169,
      "mean_token_accuracy": 0.238900026679039,
      "num_tokens": 69485118.0,
      "step": 30300
    },
    {
      "entropy": 5.420716571807861,
      "epoch": 2.9111431316042267,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000416089381908331,
      "loss": 4.7497,
      "mean_token_accuracy": 0.24910160303115844,
      "num_tokens": 69495696.0,
      "step": 30305
    },
    {
      "entropy": 5.358734035491944,
      "epoch": 2.9116234390009605,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004160626744846404,
      "loss": 4.6035,
      "mean_token_accuracy": 0.26398776918649675,
      "num_tokens": 69507832.0,
      "step": 30310
    },
    {
      "entropy": 5.424478006362915,
      "epoch": 2.9121037463976944,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004160359637858933,
      "loss": 4.8246,
      "mean_token_accuracy": 0.23932187259197235,
      "num_tokens": 69520180.0,
      "step": 30315
    },
    {
      "entropy": 5.392144966125488,
      "epoch": 2.9125840537944283,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041600924981270997,
      "loss": 4.7059,
      "mean_token_accuracy": 0.25186933726072314,
      "num_tokens": 69531201.0,
      "step": 30320
    },
    {
      "entropy": 5.470813989639282,
      "epoch": 2.913064361191162,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041598253256571057,
      "loss": 4.7809,
      "mean_token_accuracy": 0.24131005257368088,
      "num_tokens": 69542552.0,
      "step": 30325
    },
    {
      "entropy": 5.425634717941284,
      "epoch": 2.913544668587896,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004159558120455154,
      "loss": 4.7045,
      "mean_token_accuracy": 0.24742803275585173,
      "num_tokens": 69554424.0,
      "step": 30330
    },
    {
      "entropy": 5.310094404220581,
      "epoch": 2.9140249759846304,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004159290882527448,
      "loss": 4.5984,
      "mean_token_accuracy": 0.2595137134194374,
      "num_tokens": 69565043.0,
      "step": 30335
    },
    {
      "entropy": 5.398024368286133,
      "epoch": 2.914505283381364,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004159023611880192,
      "loss": 4.7278,
      "mean_token_accuracy": 0.24297191798686982,
      "num_tokens": 69576734.0,
      "step": 30340
    },
    {
      "entropy": 5.435925531387329,
      "epoch": 2.914985590778098,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004158756308519592,
      "loss": 4.7532,
      "mean_token_accuracy": 0.24881927520036698,
      "num_tokens": 69588565.0,
      "step": 30345
    },
    {
      "entropy": 5.4459997653961185,
      "epoch": 2.915465898174832,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041584889724518545,
      "loss": 4.7504,
      "mean_token_accuracy": 0.2476239338517189,
      "num_tokens": 69600595.0,
      "step": 30350
    },
    {
      "entropy": 5.353517150878906,
      "epoch": 2.915946205571566,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041582216036831844,
      "loss": 4.6611,
      "mean_token_accuracy": 0.2582151487469673,
      "num_tokens": 69611091.0,
      "step": 30355
    },
    {
      "entropy": 5.4893965244293215,
      "epoch": 2.9164265129683,
      "grad_norm": 1.109375,
      "learning_rate": 0.000415795420221979,
      "loss": 4.8017,
      "mean_token_accuracy": 0.23985295593738556,
      "num_tokens": 69621882.0,
      "step": 30360
    },
    {
      "entropy": 5.452169418334961,
      "epoch": 2.9169068203650337,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041576867680678803,
      "loss": 4.7346,
      "mean_token_accuracy": 0.24419746845960616,
      "num_tokens": 69634673.0,
      "step": 30365
    },
    {
      "entropy": 5.42421588897705,
      "epoch": 2.9173871277617676,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004157419301233664,
      "loss": 4.7712,
      "mean_token_accuracy": 0.24426280111074447,
      "num_tokens": 69644950.0,
      "step": 30370
    },
    {
      "entropy": 5.429913806915283,
      "epoch": 2.9178674351585014,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041571518017233505,
      "loss": 4.7641,
      "mean_token_accuracy": 0.25136164128780364,
      "num_tokens": 69656750.0,
      "step": 30375
    },
    {
      "entropy": 5.45133867263794,
      "epoch": 2.9183477425552353,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004156884269543151,
      "loss": 4.6984,
      "mean_token_accuracy": 0.2533477023243904,
      "num_tokens": 69667945.0,
      "step": 30380
    },
    {
      "entropy": 5.453699779510498,
      "epoch": 2.918828049951969,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004156616704699275,
      "loss": 4.7662,
      "mean_token_accuracy": 0.24665976762771608,
      "num_tokens": 69679995.0,
      "step": 30385
    },
    {
      "entropy": 5.393405628204346,
      "epoch": 2.919308357348703,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041563491071979375,
      "loss": 4.7185,
      "mean_token_accuracy": 0.24574214816093445,
      "num_tokens": 69691029.0,
      "step": 30390
    },
    {
      "entropy": 5.412744522094727,
      "epoch": 2.919788664745437,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041560814770453495,
      "loss": 4.6764,
      "mean_token_accuracy": 0.2520484894514084,
      "num_tokens": 69702497.0,
      "step": 30395
    },
    {
      "entropy": 5.444868326187134,
      "epoch": 2.920268972142171,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041558138142477235,
      "loss": 4.7612,
      "mean_token_accuracy": 0.24373957216739656,
      "num_tokens": 69713127.0,
      "step": 30400
    },
    {
      "entropy": 5.412368679046631,
      "epoch": 2.9207492795389047,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041555461188112763,
      "loss": 4.7227,
      "mean_token_accuracy": 0.2506107002496719,
      "num_tokens": 69725314.0,
      "step": 30405
    },
    {
      "entropy": 5.31435546875,
      "epoch": 2.921229586935639,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041552783907422217,
      "loss": 4.6703,
      "mean_token_accuracy": 0.2517583668231964,
      "num_tokens": 69736696.0,
      "step": 30410
    },
    {
      "entropy": 5.35591983795166,
      "epoch": 2.9217098943323725,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004155010630046775,
      "loss": 4.6967,
      "mean_token_accuracy": 0.2516411796212196,
      "num_tokens": 69748639.0,
      "step": 30415
    },
    {
      "entropy": 5.44840726852417,
      "epoch": 2.922190201729107,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004154742836731155,
      "loss": 4.8176,
      "mean_token_accuracy": 0.24347722232341767,
      "num_tokens": 69760124.0,
      "step": 30420
    },
    {
      "entropy": 5.402956247329712,
      "epoch": 2.9226705091258407,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004154475010801576,
      "loss": 4.759,
      "mean_token_accuracy": 0.2504714146256447,
      "num_tokens": 69772280.0,
      "step": 30425
    },
    {
      "entropy": 5.385808420181275,
      "epoch": 2.9231508165225746,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041542071522642583,
      "loss": 4.7111,
      "mean_token_accuracy": 0.2500303342938423,
      "num_tokens": 69784308.0,
      "step": 30430
    },
    {
      "entropy": 5.431158590316772,
      "epoch": 2.9236311239193085,
      "grad_norm": 1.125,
      "learning_rate": 0.0004153939261125421,
      "loss": 4.7769,
      "mean_token_accuracy": 0.24424721747636796,
      "num_tokens": 69795761.0,
      "step": 30435
    },
    {
      "entropy": 5.390278387069702,
      "epoch": 2.9241114313160423,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004153671337391281,
      "loss": 4.6377,
      "mean_token_accuracy": 0.26064167320728304,
      "num_tokens": 69805581.0,
      "step": 30440
    },
    {
      "entropy": 5.390604400634766,
      "epoch": 2.9245917387127762,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004153403381068062,
      "loss": 4.7413,
      "mean_token_accuracy": 0.24600663632154465,
      "num_tokens": 69816593.0,
      "step": 30445
    },
    {
      "entropy": 5.3855781078338625,
      "epoch": 2.92507204610951,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041531353921619833,
      "loss": 4.7223,
      "mean_token_accuracy": 0.2501136511564255,
      "num_tokens": 69828424.0,
      "step": 30450
    },
    {
      "entropy": 5.450203990936279,
      "epoch": 2.925552353506244,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004152867370679267,
      "loss": 4.7322,
      "mean_token_accuracy": 0.2444000229239464,
      "num_tokens": 69839250.0,
      "step": 30455
    },
    {
      "entropy": 5.371061229705811,
      "epoch": 2.926032660902978,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041525993166261366,
      "loss": 4.6649,
      "mean_token_accuracy": 0.25219952315092087,
      "num_tokens": 69850910.0,
      "step": 30460
    },
    {
      "entropy": 5.410703659057617,
      "epoch": 2.9265129682997117,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004152331230008814,
      "loss": 4.7377,
      "mean_token_accuracy": 0.2431185618042946,
      "num_tokens": 69863048.0,
      "step": 30465
    },
    {
      "entropy": 5.4433393478393555,
      "epoch": 2.9269932756964456,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041520631108335254,
      "loss": 4.785,
      "mean_token_accuracy": 0.23952400535345078,
      "num_tokens": 69873512.0,
      "step": 30470
    },
    {
      "entropy": 5.4971456050872805,
      "epoch": 2.9274735830931795,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004151794959106494,
      "loss": 4.8708,
      "mean_token_accuracy": 0.23444258123636247,
      "num_tokens": 69884813.0,
      "step": 30475
    },
    {
      "entropy": 5.424994945526123,
      "epoch": 2.9279538904899134,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004151526774833947,
      "loss": 4.7538,
      "mean_token_accuracy": 0.24824733138084412,
      "num_tokens": 69896832.0,
      "step": 30480
    },
    {
      "entropy": 5.392922973632812,
      "epoch": 2.9284341978866477,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041512585580221086,
      "loss": 4.7234,
      "mean_token_accuracy": 0.25170409232378005,
      "num_tokens": 69908181.0,
      "step": 30485
    },
    {
      "entropy": 5.477452802658081,
      "epoch": 2.928914505283381,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004150990308677208,
      "loss": 4.776,
      "mean_token_accuracy": 0.23943832367658616,
      "num_tokens": 69918839.0,
      "step": 30490
    },
    {
      "entropy": 5.5110362529754635,
      "epoch": 2.9293948126801155,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041507220268054737,
      "loss": 4.8289,
      "mean_token_accuracy": 0.23796805143356323,
      "num_tokens": 69929637.0,
      "step": 30495
    },
    {
      "entropy": 5.4602303981781,
      "epoch": 2.929875120076849,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004150453712413131,
      "loss": 4.7658,
      "mean_token_accuracy": 0.24476047605276108,
      "num_tokens": 69940202.0,
      "step": 30500
    },
    {
      "entropy": 5.433092737197876,
      "epoch": 2.9303554274735832,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041501853655064134,
      "loss": 4.7234,
      "mean_token_accuracy": 0.24303341805934905,
      "num_tokens": 69951049.0,
      "step": 30505
    },
    {
      "entropy": 5.393680238723755,
      "epoch": 2.930835734870317,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004149916986091548,
      "loss": 4.7587,
      "mean_token_accuracy": 0.24686580747365952,
      "num_tokens": 69962950.0,
      "step": 30510
    },
    {
      "entropy": 5.4453733444213865,
      "epoch": 2.931316042267051,
      "grad_norm": 1.125,
      "learning_rate": 0.0004149648574174768,
      "loss": 4.8053,
      "mean_token_accuracy": 0.2425194576382637,
      "num_tokens": 69974961.0,
      "step": 30515
    },
    {
      "entropy": 5.500875854492188,
      "epoch": 2.931796349663785,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004149380129762304,
      "loss": 4.7385,
      "mean_token_accuracy": 0.25076675564050677,
      "num_tokens": 69985184.0,
      "step": 30520
    },
    {
      "entropy": 5.3248227596282955,
      "epoch": 2.9322766570605188,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004149111652860389,
      "loss": 4.6251,
      "mean_token_accuracy": 0.25515489280223846,
      "num_tokens": 69996773.0,
      "step": 30525
    },
    {
      "entropy": 5.473872709274292,
      "epoch": 2.9327569644572526,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004148843143475255,
      "loss": 4.7495,
      "mean_token_accuracy": 0.24799790233373642,
      "num_tokens": 70008834.0,
      "step": 30530
    },
    {
      "entropy": 5.394053220748901,
      "epoch": 2.9332372718539865,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004148574601613137,
      "loss": 4.6988,
      "mean_token_accuracy": 0.2491510197520256,
      "num_tokens": 70019708.0,
      "step": 30535
    },
    {
      "entropy": 5.540844535827636,
      "epoch": 2.9337175792507204,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004148306027280271,
      "loss": 4.926,
      "mean_token_accuracy": 0.23159885108470918,
      "num_tokens": 70029680.0,
      "step": 30540
    },
    {
      "entropy": 5.372378492355347,
      "epoch": 2.9341978866474543,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041480374204828896,
      "loss": 4.6718,
      "mean_token_accuracy": 0.25628808736801145,
      "num_tokens": 70041473.0,
      "step": 30545
    },
    {
      "entropy": 5.458177804946899,
      "epoch": 2.934678194044188,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041477687812272314,
      "loss": 4.7675,
      "mean_token_accuracy": 0.24512702971696854,
      "num_tokens": 70051646.0,
      "step": 30550
    },
    {
      "entropy": 5.447763872146607,
      "epoch": 2.935158501440922,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041475001095195324,
      "loss": 4.7852,
      "mean_token_accuracy": 0.24060483872890473,
      "num_tokens": 70063795.0,
      "step": 30555
    },
    {
      "entropy": 5.480457353591919,
      "epoch": 2.9356388088376564,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004147231405366031,
      "loss": 4.7059,
      "mean_token_accuracy": 0.25443692654371264,
      "num_tokens": 70074481.0,
      "step": 30560
    },
    {
      "entropy": 5.356074666976928,
      "epoch": 2.93611911623439,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004146962668772965,
      "loss": 4.7265,
      "mean_token_accuracy": 0.24825941920280456,
      "num_tokens": 70085939.0,
      "step": 30565
    },
    {
      "entropy": 5.399203681945801,
      "epoch": 2.936599423631124,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041466938997465744,
      "loss": 4.7701,
      "mean_token_accuracy": 0.2399018496274948,
      "num_tokens": 70097466.0,
      "step": 30570
    },
    {
      "entropy": 5.400314998626709,
      "epoch": 2.9370797310278576,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041464250982930974,
      "loss": 4.6733,
      "mean_token_accuracy": 0.24862549602985382,
      "num_tokens": 70109720.0,
      "step": 30575
    },
    {
      "entropy": 5.512042284011841,
      "epoch": 2.937560038424592,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041461562644187777,
      "loss": 4.8727,
      "mean_token_accuracy": 0.24435559362173082,
      "num_tokens": 70121101.0,
      "step": 30580
    },
    {
      "entropy": 5.477317905426025,
      "epoch": 2.938040345821326,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041458873981298547,
      "loss": 4.7816,
      "mean_token_accuracy": 0.2502028927206993,
      "num_tokens": 70131512.0,
      "step": 30585
    },
    {
      "entropy": 5.469296216964722,
      "epoch": 2.9385206532180597,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041456184994325714,
      "loss": 4.8325,
      "mean_token_accuracy": 0.23967812359333038,
      "num_tokens": 70142873.0,
      "step": 30590
    },
    {
      "entropy": 5.41158127784729,
      "epoch": 2.9390009606147935,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041453495683331694,
      "loss": 4.7904,
      "mean_token_accuracy": 0.23982602655887603,
      "num_tokens": 70154476.0,
      "step": 30595
    },
    {
      "entropy": 5.480956554412842,
      "epoch": 2.9394812680115274,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041450806048378954,
      "loss": 4.7922,
      "mean_token_accuracy": 0.24508068561553956,
      "num_tokens": 70165439.0,
      "step": 30600
    },
    {
      "entropy": 5.447005367279052,
      "epoch": 2.9399615754082613,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004144811608952992,
      "loss": 4.7337,
      "mean_token_accuracy": 0.2463529020547867,
      "num_tokens": 70176515.0,
      "step": 30605
    },
    {
      "entropy": 5.393143129348755,
      "epoch": 2.940441882804995,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041445425806847043,
      "loss": 4.7444,
      "mean_token_accuracy": 0.24794113785028457,
      "num_tokens": 70189064.0,
      "step": 30610
    },
    {
      "entropy": 5.3850609302520756,
      "epoch": 2.940922190201729,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041442735200392783,
      "loss": 4.736,
      "mean_token_accuracy": 0.24263548254966735,
      "num_tokens": 70200984.0,
      "step": 30615
    },
    {
      "entropy": 5.332991695404052,
      "epoch": 2.941402497598463,
      "grad_norm": 1.125,
      "learning_rate": 0.0004144004427022962,
      "loss": 4.7202,
      "mean_token_accuracy": 0.24138085544109344,
      "num_tokens": 70212071.0,
      "step": 30620
    },
    {
      "entropy": 5.4781359195709225,
      "epoch": 2.941882804995197,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041437353016420025,
      "loss": 4.8268,
      "mean_token_accuracy": 0.24210086166858674,
      "num_tokens": 70223395.0,
      "step": 30625
    },
    {
      "entropy": 5.406235599517823,
      "epoch": 2.9423631123919307,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004143466143902648,
      "loss": 4.6646,
      "mean_token_accuracy": 0.2497050344944,
      "num_tokens": 70233130.0,
      "step": 30630
    },
    {
      "entropy": 5.4559413433074955,
      "epoch": 2.9428434197886646,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041431969538111463,
      "loss": 4.797,
      "mean_token_accuracy": 0.24286259561777115,
      "num_tokens": 70245227.0,
      "step": 30635
    },
    {
      "entropy": 5.4242527961730955,
      "epoch": 2.9433237271853985,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004142927731373749,
      "loss": 4.702,
      "mean_token_accuracy": 0.24425833970308303,
      "num_tokens": 70256483.0,
      "step": 30640
    },
    {
      "entropy": 5.415040159225464,
      "epoch": 2.943804034582133,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004142658476596706,
      "loss": 4.7451,
      "mean_token_accuracy": 0.24522114545106888,
      "num_tokens": 70267484.0,
      "step": 30645
    },
    {
      "entropy": 5.5432604312896725,
      "epoch": 2.9442843419788662,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041423891894862687,
      "loss": 4.8721,
      "mean_token_accuracy": 0.23590061515569688,
      "num_tokens": 70278137.0,
      "step": 30650
    },
    {
      "entropy": 5.4455554485321045,
      "epoch": 2.9447646493756006,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004142119870048689,
      "loss": 4.751,
      "mean_token_accuracy": 0.24108364433050156,
      "num_tokens": 70289255.0,
      "step": 30655
    },
    {
      "entropy": 5.440363502502441,
      "epoch": 2.9452449567723344,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004141850518290219,
      "loss": 4.7952,
      "mean_token_accuracy": 0.24837689250707626,
      "num_tokens": 70300672.0,
      "step": 30660
    },
    {
      "entropy": 5.416042232513428,
      "epoch": 2.9457252641690683,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041415811342171134,
      "loss": 4.7427,
      "mean_token_accuracy": 0.25017111748456955,
      "num_tokens": 70311964.0,
      "step": 30665
    },
    {
      "entropy": 5.382091379165649,
      "epoch": 2.946205571565802,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004141311717835625,
      "loss": 4.7007,
      "mean_token_accuracy": 0.2460867792367935,
      "num_tokens": 70323354.0,
      "step": 30670
    },
    {
      "entropy": 5.4736621379852295,
      "epoch": 2.946685878962536,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041410422691520114,
      "loss": 4.8294,
      "mean_token_accuracy": 0.2310547411441803,
      "num_tokens": 70335432.0,
      "step": 30675
    },
    {
      "entropy": 5.470238065719604,
      "epoch": 2.94716618635927,
      "grad_norm": 1.125,
      "learning_rate": 0.00041407727881725265,
      "loss": 4.7893,
      "mean_token_accuracy": 0.24098130762577058,
      "num_tokens": 70347012.0,
      "step": 30680
    },
    {
      "entropy": 5.446294832229614,
      "epoch": 2.947646493756004,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004140503274903426,
      "loss": 4.7344,
      "mean_token_accuracy": 0.2503788724541664,
      "num_tokens": 70358680.0,
      "step": 30685
    },
    {
      "entropy": 5.399000406265259,
      "epoch": 2.9481268011527377,
      "grad_norm": 1.203125,
      "learning_rate": 0.000414023372935097,
      "loss": 4.619,
      "mean_token_accuracy": 0.25993958711624143,
      "num_tokens": 70369081.0,
      "step": 30690
    },
    {
      "entropy": 5.437582159042359,
      "epoch": 2.9486071085494716,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041399641515214137,
      "loss": 4.7952,
      "mean_token_accuracy": 0.24031443446874617,
      "num_tokens": 70380831.0,
      "step": 30695
    },
    {
      "entropy": 5.406020212173462,
      "epoch": 2.9490874159462055,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004139694541421018,
      "loss": 4.7271,
      "mean_token_accuracy": 0.25234911143779754,
      "num_tokens": 70392403.0,
      "step": 30700
    },
    {
      "entropy": 5.434829187393189,
      "epoch": 2.9495677233429394,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004139424899056042,
      "loss": 4.8027,
      "mean_token_accuracy": 0.24474362432956695,
      "num_tokens": 70404332.0,
      "step": 30705
    },
    {
      "entropy": 5.391707181930542,
      "epoch": 2.9500480307396733,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041391552244327446,
      "loss": 4.7195,
      "mean_token_accuracy": 0.24536866694688797,
      "num_tokens": 70416119.0,
      "step": 30710
    },
    {
      "entropy": 5.517967605590821,
      "epoch": 2.950528338136407,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004138885517557387,
      "loss": 4.79,
      "mean_token_accuracy": 0.24000992625951767,
      "num_tokens": 70425874.0,
      "step": 30715
    },
    {
      "entropy": 5.462554407119751,
      "epoch": 2.9510086455331415,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004138615778436234,
      "loss": 4.8124,
      "mean_token_accuracy": 0.24143174290657043,
      "num_tokens": 70435755.0,
      "step": 30720
    },
    {
      "entropy": 5.323880529403686,
      "epoch": 2.951488952929875,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041383460070755447,
      "loss": 4.6216,
      "mean_token_accuracy": 0.25338154286146164,
      "num_tokens": 70446764.0,
      "step": 30725
    },
    {
      "entropy": 5.4062248229980465,
      "epoch": 2.9519692603266092,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041380762034815834,
      "loss": 4.6632,
      "mean_token_accuracy": 0.2583295270800591,
      "num_tokens": 70457635.0,
      "step": 30730
    },
    {
      "entropy": 5.511537742614746,
      "epoch": 2.952449567723343,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041378063676606147,
      "loss": 4.8383,
      "mean_token_accuracy": 0.2393524467945099,
      "num_tokens": 70469082.0,
      "step": 30735
    },
    {
      "entropy": 5.415397119522095,
      "epoch": 2.952929875120077,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041375364996189035,
      "loss": 4.7369,
      "mean_token_accuracy": 0.2491852179169655,
      "num_tokens": 70481634.0,
      "step": 30740
    },
    {
      "entropy": 5.442045831680298,
      "epoch": 2.953410182516811,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041372665993627143,
      "loss": 4.8073,
      "mean_token_accuracy": 0.24377822130918503,
      "num_tokens": 70493320.0,
      "step": 30745
    },
    {
      "entropy": 5.4400434494018555,
      "epoch": 2.9538904899135447,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041369966668983144,
      "loss": 4.7251,
      "mean_token_accuracy": 0.2449456349015236,
      "num_tokens": 70504998.0,
      "step": 30750
    },
    {
      "entropy": 5.441867828369141,
      "epoch": 2.9543707973102786,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041367267022319706,
      "loss": 4.7368,
      "mean_token_accuracy": 0.2487803503870964,
      "num_tokens": 70516096.0,
      "step": 30755
    },
    {
      "entropy": 5.459111118316651,
      "epoch": 2.9548511047070125,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000413645670536995,
      "loss": 4.8083,
      "mean_token_accuracy": 0.24525668174028398,
      "num_tokens": 70527193.0,
      "step": 30760
    },
    {
      "entropy": 5.53522047996521,
      "epoch": 2.9553314121037464,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004136186676318522,
      "loss": 4.8532,
      "mean_token_accuracy": 0.23459171652793884,
      "num_tokens": 70538733.0,
      "step": 30765
    },
    {
      "entropy": 5.440821981430053,
      "epoch": 2.9558117195004803,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004135916615083956,
      "loss": 4.69,
      "mean_token_accuracy": 0.2437302753329277,
      "num_tokens": 70549381.0,
      "step": 30770
    },
    {
      "entropy": 5.394108724594116,
      "epoch": 2.956292026897214,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041356465216725195,
      "loss": 4.7622,
      "mean_token_accuracy": 0.24223081916570663,
      "num_tokens": 70560576.0,
      "step": 30775
    },
    {
      "entropy": 5.431447505950928,
      "epoch": 2.956772334293948,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041353763960904873,
      "loss": 4.7452,
      "mean_token_accuracy": 0.2424158573150635,
      "num_tokens": 70572093.0,
      "step": 30780
    },
    {
      "entropy": 5.471800041198731,
      "epoch": 2.957252641690682,
      "grad_norm": 1.25,
      "learning_rate": 0.00041351062383441286,
      "loss": 4.7239,
      "mean_token_accuracy": 0.24854634404182435,
      "num_tokens": 70581567.0,
      "step": 30785
    },
    {
      "entropy": 5.391196298599243,
      "epoch": 2.957732949087416,
      "grad_norm": 1.25,
      "learning_rate": 0.0004134836048439716,
      "loss": 4.7472,
      "mean_token_accuracy": 0.2502668142318726,
      "num_tokens": 70593729.0,
      "step": 30790
    },
    {
      "entropy": 5.3966224670410154,
      "epoch": 2.95821325648415,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041345658263835215,
      "loss": 4.75,
      "mean_token_accuracy": 0.24660700857639312,
      "num_tokens": 70605816.0,
      "step": 30795
    },
    {
      "entropy": 5.435596609115601,
      "epoch": 2.9586935638808836,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041342955721818207,
      "loss": 4.7449,
      "mean_token_accuracy": 0.2458964094519615,
      "num_tokens": 70618310.0,
      "step": 30800
    },
    {
      "entropy": 5.458064460754395,
      "epoch": 2.959173871277618,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041340252858408866,
      "loss": 4.7675,
      "mean_token_accuracy": 0.24310930371284484,
      "num_tokens": 70629842.0,
      "step": 30805
    },
    {
      "entropy": 5.436635589599609,
      "epoch": 2.9596541786743513,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041337549673669963,
      "loss": 4.8483,
      "mean_token_accuracy": 0.23876065760850906,
      "num_tokens": 70642564.0,
      "step": 30810
    },
    {
      "entropy": 5.4864894390106205,
      "epoch": 2.9601344860710856,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004133484616766423,
      "loss": 4.8107,
      "mean_token_accuracy": 0.24540838599205017,
      "num_tokens": 70653225.0,
      "step": 30815
    },
    {
      "entropy": 5.512187814712524,
      "epoch": 2.9606147934678195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041332142340454463,
      "loss": 4.7846,
      "mean_token_accuracy": 0.24153124392032624,
      "num_tokens": 70665428.0,
      "step": 30820
    },
    {
      "entropy": 5.455508470535278,
      "epoch": 2.9610951008645534,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004132943819210342,
      "loss": 4.8413,
      "mean_token_accuracy": 0.24151257872581483,
      "num_tokens": 70676441.0,
      "step": 30825
    },
    {
      "entropy": 5.342586088180542,
      "epoch": 2.9615754082612873,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041326733722673876,
      "loss": 4.6221,
      "mean_token_accuracy": 0.257737335562706,
      "num_tokens": 70687561.0,
      "step": 30830
    },
    {
      "entropy": 5.445098447799682,
      "epoch": 2.962055715658021,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041324028932228645,
      "loss": 4.7725,
      "mean_token_accuracy": 0.25210694819688795,
      "num_tokens": 70699949.0,
      "step": 30835
    },
    {
      "entropy": 5.385701751708984,
      "epoch": 2.962536023054755,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004132132382083051,
      "loss": 4.7024,
      "mean_token_accuracy": 0.25421572774648665,
      "num_tokens": 70711176.0,
      "step": 30840
    },
    {
      "entropy": 5.514336967468262,
      "epoch": 2.963016330451489,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041318618388542274,
      "loss": 4.7634,
      "mean_token_accuracy": 0.24097488224506378,
      "num_tokens": 70722974.0,
      "step": 30845
    },
    {
      "entropy": 5.349890041351318,
      "epoch": 2.963496637848223,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004131591263542675,
      "loss": 4.6124,
      "mean_token_accuracy": 0.25449867099523543,
      "num_tokens": 70735158.0,
      "step": 30850
    },
    {
      "entropy": 5.403360843658447,
      "epoch": 2.9639769452449567,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004131320656154676,
      "loss": 4.7441,
      "mean_token_accuracy": 0.24556258767843248,
      "num_tokens": 70746444.0,
      "step": 30855
    },
    {
      "entropy": 5.489116287231445,
      "epoch": 2.9644572526416906,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004131050016696514,
      "loss": 4.7347,
      "mean_token_accuracy": 0.24677797853946687,
      "num_tokens": 70757985.0,
      "step": 30860
    },
    {
      "entropy": 5.47293529510498,
      "epoch": 2.9649375600384245,
      "grad_norm": 1.109375,
      "learning_rate": 0.000413077934517447,
      "loss": 4.7701,
      "mean_token_accuracy": 0.2424607440829277,
      "num_tokens": 70769867.0,
      "step": 30865
    },
    {
      "entropy": 5.441319751739502,
      "epoch": 2.9654178674351583,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004130508641594831,
      "loss": 4.8106,
      "mean_token_accuracy": 0.24050588756799698,
      "num_tokens": 70781610.0,
      "step": 30870
    },
    {
      "entropy": 5.4482197761535645,
      "epoch": 2.965898174831892,
      "grad_norm": 1.125,
      "learning_rate": 0.00041302379059638794,
      "loss": 4.7322,
      "mean_token_accuracy": 0.24705844074487687,
      "num_tokens": 70793336.0,
      "step": 30875
    },
    {
      "entropy": 5.4060248851776125,
      "epoch": 2.9663784822286265,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041299671382879024,
      "loss": 4.7126,
      "mean_token_accuracy": 0.24994179010391235,
      "num_tokens": 70802714.0,
      "step": 30880
    },
    {
      "entropy": 5.503444194793701,
      "epoch": 2.96685878962536,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004129696338573187,
      "loss": 4.8615,
      "mean_token_accuracy": 0.23677269518375396,
      "num_tokens": 70813300.0,
      "step": 30885
    },
    {
      "entropy": 5.514606618881226,
      "epoch": 2.9673390970220943,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004129425506826019,
      "loss": 4.8746,
      "mean_token_accuracy": 0.23198026567697524,
      "num_tokens": 70824828.0,
      "step": 30890
    },
    {
      "entropy": 5.452273178100586,
      "epoch": 2.967819404418828,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041291546430526863,
      "loss": 4.7937,
      "mean_token_accuracy": 0.2489680156111717,
      "num_tokens": 70836788.0,
      "step": 30895
    },
    {
      "entropy": 5.4527263164520265,
      "epoch": 2.968299711815562,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004128883747259478,
      "loss": 4.7314,
      "mean_token_accuracy": 0.25041339099407195,
      "num_tokens": 70849308.0,
      "step": 30900
    },
    {
      "entropy": 5.564817142486572,
      "epoch": 2.968780019212296,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004128612819452684,
      "loss": 4.8952,
      "mean_token_accuracy": 0.23172715157270432,
      "num_tokens": 70859987.0,
      "step": 30905
    },
    {
      "entropy": 5.510544872283935,
      "epoch": 2.96926032660903,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041283418596385944,
      "loss": 4.7909,
      "mean_token_accuracy": 0.24994247853755952,
      "num_tokens": 70872096.0,
      "step": 30910
    },
    {
      "entropy": 5.40047516822815,
      "epoch": 2.9697406340057637,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004128070867823499,
      "loss": 4.7044,
      "mean_token_accuracy": 0.2490798532962799,
      "num_tokens": 70884336.0,
      "step": 30915
    },
    {
      "entropy": 5.38635663986206,
      "epoch": 2.9702209414024976,
      "grad_norm": 1.34375,
      "learning_rate": 0.000412779984401369,
      "loss": 4.7001,
      "mean_token_accuracy": 0.25363219529390335,
      "num_tokens": 70894707.0,
      "step": 30920
    },
    {
      "entropy": 5.476804494857788,
      "epoch": 2.9707012487992315,
      "grad_norm": 1.234375,
      "learning_rate": 0.000412752878821546,
      "loss": 4.7739,
      "mean_token_accuracy": 0.24442863315343857,
      "num_tokens": 70906461.0,
      "step": 30925
    },
    {
      "entropy": 5.402764844894409,
      "epoch": 2.9711815561959654,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041272577004351026,
      "loss": 4.6662,
      "mean_token_accuracy": 0.25148017704486847,
      "num_tokens": 70917282.0,
      "step": 30930
    },
    {
      "entropy": 5.400554895401001,
      "epoch": 2.9716618635926992,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041269865806789095,
      "loss": 4.6896,
      "mean_token_accuracy": 0.2525585725903511,
      "num_tokens": 70928253.0,
      "step": 30935
    },
    {
      "entropy": 5.421302080154419,
      "epoch": 2.972142170989433,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004126715428953178,
      "loss": 4.7401,
      "mean_token_accuracy": 0.23740800470113754,
      "num_tokens": 70939089.0,
      "step": 30940
    },
    {
      "entropy": 5.436180782318115,
      "epoch": 2.972622478386167,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004126444245264202,
      "loss": 4.6843,
      "mean_token_accuracy": 0.25023125410079955,
      "num_tokens": 70949533.0,
      "step": 30945
    },
    {
      "entropy": 5.425344562530517,
      "epoch": 2.973102785782901,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004126173029618278,
      "loss": 4.6917,
      "mean_token_accuracy": 0.2452595204114914,
      "num_tokens": 70961449.0,
      "step": 30950
    },
    {
      "entropy": 5.396829462051391,
      "epoch": 2.973583093179635,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004125901782021702,
      "loss": 4.7155,
      "mean_token_accuracy": 0.24633262008428575,
      "num_tokens": 70973008.0,
      "step": 30955
    },
    {
      "entropy": 5.404406213760376,
      "epoch": 2.9740634005763686,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004125630502480773,
      "loss": 4.7138,
      "mean_token_accuracy": 0.25075856149196624,
      "num_tokens": 70984170.0,
      "step": 30960
    },
    {
      "entropy": 5.455488109588623,
      "epoch": 2.974543707973103,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004125359191001788,
      "loss": 4.7781,
      "mean_token_accuracy": 0.24196144789457322,
      "num_tokens": 70994900.0,
      "step": 30965
    },
    {
      "entropy": 5.431443166732788,
      "epoch": 2.975024015369837,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004125087847591047,
      "loss": 4.7134,
      "mean_token_accuracy": 0.2528527855873108,
      "num_tokens": 71007430.0,
      "step": 30970
    },
    {
      "entropy": 5.4473213195800785,
      "epoch": 2.9755043227665707,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041248164722548493,
      "loss": 4.8125,
      "mean_token_accuracy": 0.24202187061309816,
      "num_tokens": 71020306.0,
      "step": 30975
    },
    {
      "entropy": 5.396373319625854,
      "epoch": 2.9759846301633046,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004124545064999495,
      "loss": 4.6606,
      "mean_token_accuracy": 0.2554009616374969,
      "num_tokens": 71031114.0,
      "step": 30980
    },
    {
      "entropy": 5.332390546798706,
      "epoch": 2.9764649375600385,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041242736258312866,
      "loss": 4.6821,
      "mean_token_accuracy": 0.2539800703525543,
      "num_tokens": 71043289.0,
      "step": 30985
    },
    {
      "entropy": 5.366827964782715,
      "epoch": 2.9769452449567724,
      "grad_norm": 1.125,
      "learning_rate": 0.0004124002154756525,
      "loss": 4.7063,
      "mean_token_accuracy": 0.2509092390537262,
      "num_tokens": 71055163.0,
      "step": 30990
    },
    {
      "entropy": 5.402624225616455,
      "epoch": 2.9774255523535063,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041237306517815124,
      "loss": 4.6795,
      "mean_token_accuracy": 0.24970801323652267,
      "num_tokens": 71067830.0,
      "step": 30995
    },
    {
      "entropy": 5.404392433166504,
      "epoch": 2.97790585975024,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004123459116912554,
      "loss": 4.674,
      "mean_token_accuracy": 0.2534033939242363,
      "num_tokens": 71079242.0,
      "step": 31000
    },
    {
      "entropy": 5.447406339645386,
      "epoch": 2.978386167146974,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041231875501559535,
      "loss": 4.7741,
      "mean_token_accuracy": 0.24447084367275237,
      "num_tokens": 71091076.0,
      "step": 31005
    },
    {
      "entropy": 5.487995052337647,
      "epoch": 2.978866474543708,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041229159515180155,
      "loss": 4.6822,
      "mean_token_accuracy": 0.24779572784900666,
      "num_tokens": 71101226.0,
      "step": 31010
    },
    {
      "entropy": 5.430442523956299,
      "epoch": 2.979346781940442,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004122644321005046,
      "loss": 4.652,
      "mean_token_accuracy": 0.2501703456044197,
      "num_tokens": 71112689.0,
      "step": 31015
    },
    {
      "entropy": 5.401924896240234,
      "epoch": 2.9798270893371757,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041223726586233505,
      "loss": 4.7618,
      "mean_token_accuracy": 0.2372576802968979,
      "num_tokens": 71124256.0,
      "step": 31020
    },
    {
      "entropy": 5.436739206314087,
      "epoch": 2.9803073967339095,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041221009643792377,
      "loss": 4.7634,
      "mean_token_accuracy": 0.2500620663166046,
      "num_tokens": 71134636.0,
      "step": 31025
    },
    {
      "entropy": 5.4077401638031,
      "epoch": 2.980787704130644,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004121829238279014,
      "loss": 4.6553,
      "mean_token_accuracy": 0.25030392557382586,
      "num_tokens": 71145969.0,
      "step": 31030
    },
    {
      "entropy": 5.3913304805755615,
      "epoch": 2.9812680115273773,
      "grad_norm": 1.125,
      "learning_rate": 0.00041215574803289896,
      "loss": 4.7602,
      "mean_token_accuracy": 0.2438112258911133,
      "num_tokens": 71157491.0,
      "step": 31035
    },
    {
      "entropy": 5.400703430175781,
      "epoch": 2.9817483189241116,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004121285690535473,
      "loss": 4.7951,
      "mean_token_accuracy": 0.2435734450817108,
      "num_tokens": 71169325.0,
      "step": 31040
    },
    {
      "entropy": 5.407777214050293,
      "epoch": 2.982228626320845,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041210138689047745,
      "loss": 4.7586,
      "mean_token_accuracy": 0.24633133858442308,
      "num_tokens": 71181179.0,
      "step": 31045
    },
    {
      "entropy": 5.417395496368409,
      "epoch": 2.9827089337175794,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004120742015443206,
      "loss": 4.7418,
      "mean_token_accuracy": 0.2508948281407356,
      "num_tokens": 71193250.0,
      "step": 31050
    },
    {
      "entropy": 5.365188360214233,
      "epoch": 2.9831892411143133,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004120470130157077,
      "loss": 4.652,
      "mean_token_accuracy": 0.2552367463707924,
      "num_tokens": 71207544.0,
      "step": 31055
    },
    {
      "entropy": 5.456047868728637,
      "epoch": 2.983669548511047,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041201982130527006,
      "loss": 4.7692,
      "mean_token_accuracy": 0.24849779456853865,
      "num_tokens": 71219989.0,
      "step": 31060
    },
    {
      "entropy": 5.418975496292115,
      "epoch": 2.984149855907781,
      "grad_norm": 1.125,
      "learning_rate": 0.00041199262641363914,
      "loss": 4.7542,
      "mean_token_accuracy": 0.24078051149845123,
      "num_tokens": 71232455.0,
      "step": 31065
    },
    {
      "entropy": 5.440390634536743,
      "epoch": 2.984630163304515,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041196542834144617,
      "loss": 4.6683,
      "mean_token_accuracy": 0.2555010810494423,
      "num_tokens": 71244179.0,
      "step": 31070
    },
    {
      "entropy": 5.377125263214111,
      "epoch": 2.985110470701249,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041193822708932265,
      "loss": 4.7272,
      "mean_token_accuracy": 0.2512971624732018,
      "num_tokens": 71255468.0,
      "step": 31075
    },
    {
      "entropy": 5.475847959518433,
      "epoch": 2.9855907780979827,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004119110226579002,
      "loss": 4.7769,
      "mean_token_accuracy": 0.24636635035276414,
      "num_tokens": 71266192.0,
      "step": 31080
    },
    {
      "entropy": 5.380905246734619,
      "epoch": 2.9860710854947166,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041188381504781026,
      "loss": 4.681,
      "mean_token_accuracy": 0.24854741394519805,
      "num_tokens": 71277166.0,
      "step": 31085
    },
    {
      "entropy": 5.38548526763916,
      "epoch": 2.9865513928914504,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004118566042596846,
      "loss": 4.692,
      "mean_token_accuracy": 0.24676567763090135,
      "num_tokens": 71288077.0,
      "step": 31090
    },
    {
      "entropy": 5.341213941574097,
      "epoch": 2.9870317002881843,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000411829390294155,
      "loss": 4.6756,
      "mean_token_accuracy": 0.25623353719711306,
      "num_tokens": 71300423.0,
      "step": 31095
    },
    {
      "entropy": 5.4563820362091064,
      "epoch": 2.987512007684918,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041180217315185333,
      "loss": 4.6999,
      "mean_token_accuracy": 0.24675726592540742,
      "num_tokens": 71312265.0,
      "step": 31100
    },
    {
      "entropy": 5.422246646881104,
      "epoch": 2.9879923150816525,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041177495283341124,
      "loss": 4.6639,
      "mean_token_accuracy": 0.25029499679803846,
      "num_tokens": 71323369.0,
      "step": 31105
    },
    {
      "entropy": 5.443338537216187,
      "epoch": 2.988472622478386,
      "grad_norm": 1.203125,
      "learning_rate": 0.000411747729339461,
      "loss": 4.7169,
      "mean_token_accuracy": 0.2557911142706871,
      "num_tokens": 71333977.0,
      "step": 31110
    },
    {
      "entropy": 5.333019256591797,
      "epoch": 2.9889529298751203,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004117205026706345,
      "loss": 4.7213,
      "mean_token_accuracy": 0.25177475810050964,
      "num_tokens": 71345176.0,
      "step": 31115
    },
    {
      "entropy": 5.428236627578736,
      "epoch": 2.9894332372718537,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041169327282756396,
      "loss": 4.8113,
      "mean_token_accuracy": 0.2398540571331978,
      "num_tokens": 71357310.0,
      "step": 31120
    },
    {
      "entropy": 5.53222165107727,
      "epoch": 2.989913544668588,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004116660398108815,
      "loss": 4.8715,
      "mean_token_accuracy": 0.23938356190919877,
      "num_tokens": 71369610.0,
      "step": 31125
    },
    {
      "entropy": 5.502241468429565,
      "epoch": 2.990393852065322,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004116388036212194,
      "loss": 4.856,
      "mean_token_accuracy": 0.23990450352430343,
      "num_tokens": 71381973.0,
      "step": 31130
    },
    {
      "entropy": 5.364748430252075,
      "epoch": 2.990874159462056,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041161156425921004,
      "loss": 4.7317,
      "mean_token_accuracy": 0.24429362267255783,
      "num_tokens": 71393739.0,
      "step": 31135
    },
    {
      "entropy": 5.473902416229248,
      "epoch": 2.9913544668587897,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041158432172548577,
      "loss": 4.8088,
      "mean_token_accuracy": 0.23955023288726807,
      "num_tokens": 71405205.0,
      "step": 31140
    },
    {
      "entropy": 5.4109046936035154,
      "epoch": 2.9918347742555236,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041155707602067923,
      "loss": 4.6862,
      "mean_token_accuracy": 0.24627838283777237,
      "num_tokens": 71416645.0,
      "step": 31145
    },
    {
      "entropy": 5.455294513702393,
      "epoch": 2.9923150816522575,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004115298271454227,
      "loss": 4.7769,
      "mean_token_accuracy": 0.23832102864980698,
      "num_tokens": 71428854.0,
      "step": 31150
    },
    {
      "entropy": 5.431166887283325,
      "epoch": 2.9927953890489913,
      "grad_norm": 1.25,
      "learning_rate": 0.0004115025751003491,
      "loss": 4.7239,
      "mean_token_accuracy": 0.24914879500865936,
      "num_tokens": 71439950.0,
      "step": 31155
    },
    {
      "entropy": 5.406118583679199,
      "epoch": 2.993275696445725,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000411475319886091,
      "loss": 4.7274,
      "mean_token_accuracy": 0.24593147337436677,
      "num_tokens": 71451901.0,
      "step": 31160
    },
    {
      "entropy": 5.512956666946411,
      "epoch": 2.993756003842459,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041144806150328117,
      "loss": 4.8645,
      "mean_token_accuracy": 0.24414031505584716,
      "num_tokens": 71463688.0,
      "step": 31165
    },
    {
      "entropy": 5.5157266616821286,
      "epoch": 2.994236311239193,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004114207999525527,
      "loss": 4.821,
      "mean_token_accuracy": 0.23856604993343353,
      "num_tokens": 71475392.0,
      "step": 31170
    },
    {
      "entropy": 5.444282579421997,
      "epoch": 2.994716618635927,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041139353523453814,
      "loss": 4.7706,
      "mean_token_accuracy": 0.24700376093387605,
      "num_tokens": 71486004.0,
      "step": 31175
    },
    {
      "entropy": 5.4562098503112795,
      "epoch": 2.9951969260326607,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004113662673498708,
      "loss": 4.8446,
      "mean_token_accuracy": 0.24112093299627305,
      "num_tokens": 71497850.0,
      "step": 31180
    },
    {
      "entropy": 5.441487741470337,
      "epoch": 2.9956772334293946,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041133899629918364,
      "loss": 4.7224,
      "mean_token_accuracy": 0.25248983353376386,
      "num_tokens": 71507778.0,
      "step": 31185
    },
    {
      "entropy": 5.450409650802612,
      "epoch": 2.996157540826129,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041131172208310986,
      "loss": 4.773,
      "mean_token_accuracy": 0.24468955099582673,
      "num_tokens": 71519506.0,
      "step": 31190
    },
    {
      "entropy": 5.379032468795776,
      "epoch": 2.9966378482228624,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041128444470228253,
      "loss": 4.6975,
      "mean_token_accuracy": 0.2522758975625038,
      "num_tokens": 71530523.0,
      "step": 31195
    },
    {
      "entropy": 5.472422361373901,
      "epoch": 2.9971181556195967,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041125716415733524,
      "loss": 4.7973,
      "mean_token_accuracy": 0.23988354504108428,
      "num_tokens": 71541473.0,
      "step": 31200
    },
    {
      "entropy": 5.418289041519165,
      "epoch": 2.9975984630163306,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004112298804489011,
      "loss": 4.7385,
      "mean_token_accuracy": 0.25063555389642717,
      "num_tokens": 71552012.0,
      "step": 31205
    },
    {
      "entropy": 5.437396574020386,
      "epoch": 2.9980787704130645,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004112025935776137,
      "loss": 4.7567,
      "mean_token_accuracy": 0.2454693540930748,
      "num_tokens": 71562239.0,
      "step": 31210
    },
    {
      "entropy": 5.388843393325805,
      "epoch": 2.9985590778097984,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041117530354410647,
      "loss": 4.6864,
      "mean_token_accuracy": 0.24955980181694032,
      "num_tokens": 71573296.0,
      "step": 31215
    },
    {
      "entropy": 5.433870697021485,
      "epoch": 2.9990393852065322,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004111480103490131,
      "loss": 4.8142,
      "mean_token_accuracy": 0.23995545655488967,
      "num_tokens": 71584659.0,
      "step": 31220
    },
    {
      "entropy": 5.396944284439087,
      "epoch": 2.999519692603266,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041112071399296724,
      "loss": 4.7142,
      "mean_token_accuracy": 0.24749199897050858,
      "num_tokens": 71595672.0,
      "step": 31225
    },
    {
      "entropy": 5.406460618972778,
      "epoch": 3.0,
      "grad_norm": 1.25,
      "learning_rate": 0.0004110934144766025,
      "loss": 4.6589,
      "mean_token_accuracy": 0.25702732503414155,
      "num_tokens": 71605608.0,
      "step": 31230
    },
    {
      "entropy": 5.43630084991455,
      "epoch": 3.000480307396734,
      "grad_norm": 1.25,
      "learning_rate": 0.00041106611180055284,
      "loss": 4.7515,
      "mean_token_accuracy": 0.24345263540744783,
      "num_tokens": 71617542.0,
      "step": 31235
    },
    {
      "entropy": 5.371315145492554,
      "epoch": 3.0009606147934678,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041103880596545206,
      "loss": 4.6173,
      "mean_token_accuracy": 0.24850389510393142,
      "num_tokens": 71629087.0,
      "step": 31240
    },
    {
      "entropy": 5.452855396270752,
      "epoch": 3.0014409221902016,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004110114969719342,
      "loss": 4.711,
      "mean_token_accuracy": 0.25310745239257815,
      "num_tokens": 71640929.0,
      "step": 31245
    },
    {
      "entropy": 5.433626270294189,
      "epoch": 3.0019212295869355,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004109841848206333,
      "loss": 4.6958,
      "mean_token_accuracy": 0.2533930137753487,
      "num_tokens": 71652641.0,
      "step": 31250
    },
    {
      "entropy": 5.389298915863037,
      "epoch": 3.0024015369836694,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004109568695121833,
      "loss": 4.6305,
      "mean_token_accuracy": 0.25690782964229586,
      "num_tokens": 71664482.0,
      "step": 31255
    },
    {
      "entropy": 5.334581804275513,
      "epoch": 3.0028818443804033,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004109295510472186,
      "loss": 4.5986,
      "mean_token_accuracy": 0.25831114649772646,
      "num_tokens": 71675568.0,
      "step": 31260
    },
    {
      "entropy": 5.487595415115356,
      "epoch": 3.003362151777137,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041090222942637323,
      "loss": 4.7529,
      "mean_token_accuracy": 0.2454451784491539,
      "num_tokens": 71685731.0,
      "step": 31265
    },
    {
      "entropy": 5.486229801177979,
      "epoch": 3.0038424591738715,
      "grad_norm": 1.25,
      "learning_rate": 0.00041087490465028175,
      "loss": 4.8367,
      "mean_token_accuracy": 0.23815142065286637,
      "num_tokens": 71697212.0,
      "step": 31270
    },
    {
      "entropy": 5.373194789886474,
      "epoch": 3.0043227665706054,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041084757671957844,
      "loss": 4.6445,
      "mean_token_accuracy": 0.2507548496127129,
      "num_tokens": 71708444.0,
      "step": 31275
    },
    {
      "entropy": 5.442957830429077,
      "epoch": 3.0048030739673393,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041082024563489773,
      "loss": 4.7006,
      "mean_token_accuracy": 0.2480404794216156,
      "num_tokens": 71718645.0,
      "step": 31280
    },
    {
      "entropy": 5.42732801437378,
      "epoch": 3.005283381364073,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004107929113968743,
      "loss": 4.6467,
      "mean_token_accuracy": 0.25192674100399015,
      "num_tokens": 71728798.0,
      "step": 31285
    },
    {
      "entropy": 5.4356742858886715,
      "epoch": 3.005763688760807,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004107655740061427,
      "loss": 4.7803,
      "mean_token_accuracy": 0.23909824788570405,
      "num_tokens": 71741237.0,
      "step": 31290
    },
    {
      "entropy": 5.433341836929321,
      "epoch": 3.006243996157541,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004107382334633375,
      "loss": 4.7421,
      "mean_token_accuracy": 0.25329892337322235,
      "num_tokens": 71753755.0,
      "step": 31295
    },
    {
      "entropy": 5.505087852478027,
      "epoch": 3.006724303554275,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004107108897690936,
      "loss": 4.7341,
      "mean_token_accuracy": 0.24800524413585662,
      "num_tokens": 71765447.0,
      "step": 31300
    },
    {
      "entropy": 5.461794376373291,
      "epoch": 3.0072046109510087,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004106835429240458,
      "loss": 4.7352,
      "mean_token_accuracy": 0.24467562586069108,
      "num_tokens": 71777106.0,
      "step": 31305
    },
    {
      "entropy": 5.377772045135498,
      "epoch": 3.0076849183477425,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004106561929288292,
      "loss": 4.6498,
      "mean_token_accuracy": 0.25095162987709047,
      "num_tokens": 71789226.0,
      "step": 31310
    },
    {
      "entropy": 5.451354694366455,
      "epoch": 3.0081652257444764,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041062883978407844,
      "loss": 4.7372,
      "mean_token_accuracy": 0.2477993845939636,
      "num_tokens": 71800754.0,
      "step": 31315
    },
    {
      "entropy": 5.4422777652740475,
      "epoch": 3.0086455331412103,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041060148349042876,
      "loss": 4.7895,
      "mean_token_accuracy": 0.24823420196771623,
      "num_tokens": 71812107.0,
      "step": 31320
    },
    {
      "entropy": 5.447896671295166,
      "epoch": 3.009125840537944,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041057412404851536,
      "loss": 4.6825,
      "mean_token_accuracy": 0.24816567897796632,
      "num_tokens": 71822031.0,
      "step": 31325
    },
    {
      "entropy": 5.448412322998047,
      "epoch": 3.009606147934678,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004105467614589734,
      "loss": 4.6948,
      "mean_token_accuracy": 0.24769560843706132,
      "num_tokens": 71833744.0,
      "step": 31330
    },
    {
      "entropy": 5.451032018661499,
      "epoch": 3.010086455331412,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004105193957224381,
      "loss": 4.6648,
      "mean_token_accuracy": 0.2575537532567978,
      "num_tokens": 71845385.0,
      "step": 31335
    },
    {
      "entropy": 5.382905340194702,
      "epoch": 3.010566762728146,
      "grad_norm": 2.0,
      "learning_rate": 0.00041049202683954473,
      "loss": 4.5901,
      "mean_token_accuracy": 0.26583255231380465,
      "num_tokens": 71858065.0,
      "step": 31340
    },
    {
      "entropy": 5.372502326965332,
      "epoch": 3.0110470701248797,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041046465481092893,
      "loss": 4.6501,
      "mean_token_accuracy": 0.25050911754369737,
      "num_tokens": 71869553.0,
      "step": 31345
    },
    {
      "entropy": 5.3847941875457765,
      "epoch": 3.011527377521614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041043727963722607,
      "loss": 4.6886,
      "mean_token_accuracy": 0.24886180013418197,
      "num_tokens": 71881196.0,
      "step": 31350
    },
    {
      "entropy": 5.422232818603516,
      "epoch": 3.012007684918348,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004104099013190718,
      "loss": 4.7193,
      "mean_token_accuracy": 0.24333220422267915,
      "num_tokens": 71893323.0,
      "step": 31355
    },
    {
      "entropy": 5.4189855575561525,
      "epoch": 3.012487992315082,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041038251985710164,
      "loss": 4.6646,
      "mean_token_accuracy": 0.2528666228055954,
      "num_tokens": 71904679.0,
      "step": 31360
    },
    {
      "entropy": 5.394376564025879,
      "epoch": 3.0129682997118157,
      "grad_norm": 1.125,
      "learning_rate": 0.0004103551352519514,
      "loss": 4.7316,
      "mean_token_accuracy": 0.2554255336523056,
      "num_tokens": 71916478.0,
      "step": 31365
    },
    {
      "entropy": 5.449830961227417,
      "epoch": 3.0134486071085496,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041032774750425683,
      "loss": 4.7389,
      "mean_token_accuracy": 0.24960014522075652,
      "num_tokens": 71928203.0,
      "step": 31370
    },
    {
      "entropy": 5.4411736011505125,
      "epoch": 3.0139289145052834,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004103003566146538,
      "loss": 4.6456,
      "mean_token_accuracy": 0.25319170504808425,
      "num_tokens": 71939118.0,
      "step": 31375
    },
    {
      "entropy": 5.473449802398681,
      "epoch": 3.0144092219020173,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004102729625837782,
      "loss": 4.7656,
      "mean_token_accuracy": 0.2473948910832405,
      "num_tokens": 71950964.0,
      "step": 31380
    },
    {
      "entropy": 5.339174461364746,
      "epoch": 3.014889529298751,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004102455654122662,
      "loss": 4.6228,
      "mean_token_accuracy": 0.2579213738441467,
      "num_tokens": 71960587.0,
      "step": 31385
    },
    {
      "entropy": 5.417657947540283,
      "epoch": 3.015369836695485,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041021816510075366,
      "loss": 4.6528,
      "mean_token_accuracy": 0.25429451912641526,
      "num_tokens": 71972014.0,
      "step": 31390
    },
    {
      "entropy": 5.487690782546997,
      "epoch": 3.015850144092219,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041019076164987696,
      "loss": 4.7085,
      "mean_token_accuracy": 0.2522218063473701,
      "num_tokens": 71982842.0,
      "step": 31395
    },
    {
      "entropy": 5.441136407852173,
      "epoch": 3.016330451488953,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004101633550602721,
      "loss": 4.6602,
      "mean_token_accuracy": 0.24545825272798538,
      "num_tokens": 71994450.0,
      "step": 31400
    },
    {
      "entropy": 5.3491809368133545,
      "epoch": 3.0168107588856867,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004101359453325755,
      "loss": 4.6017,
      "mean_token_accuracy": 0.25238874554634094,
      "num_tokens": 72005490.0,
      "step": 31405
    },
    {
      "entropy": 5.420967531204224,
      "epoch": 3.0172910662824206,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041010853246742357,
      "loss": 4.7078,
      "mean_token_accuracy": 0.2482004463672638,
      "num_tokens": 72016723.0,
      "step": 31410
    },
    {
      "entropy": 5.37503662109375,
      "epoch": 3.0177713736791545,
      "grad_norm": 1.125,
      "learning_rate": 0.0004100811164654527,
      "loss": 4.6266,
      "mean_token_accuracy": 0.25839925557374954,
      "num_tokens": 72027189.0,
      "step": 31415
    },
    {
      "entropy": 5.478568601608276,
      "epoch": 3.0182516810758884,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004100536973272994,
      "loss": 4.7633,
      "mean_token_accuracy": 0.24480478912591935,
      "num_tokens": 72039662.0,
      "step": 31420
    },
    {
      "entropy": 5.418169784545898,
      "epoch": 3.0187319884726227,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004100262750536003,
      "loss": 4.6819,
      "mean_token_accuracy": 0.248098886013031,
      "num_tokens": 72050769.0,
      "step": 31425
    },
    {
      "entropy": 5.401988887786866,
      "epoch": 3.0192122958693566,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040999884964499196,
      "loss": 4.7135,
      "mean_token_accuracy": 0.24972352087497712,
      "num_tokens": 72062657.0,
      "step": 31430
    },
    {
      "entropy": 5.362568330764771,
      "epoch": 3.0196926032660905,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040997142110211127,
      "loss": 4.6728,
      "mean_token_accuracy": 0.25458624958992004,
      "num_tokens": 72074805.0,
      "step": 31435
    },
    {
      "entropy": 5.395764112472534,
      "epoch": 3.0201729106628243,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040994398942559496,
      "loss": 4.6698,
      "mean_token_accuracy": 0.2578009977936745,
      "num_tokens": 72085842.0,
      "step": 31440
    },
    {
      "entropy": 5.49265422821045,
      "epoch": 3.020653218059558,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004099165546160799,
      "loss": 4.7529,
      "mean_token_accuracy": 0.24584387093782425,
      "num_tokens": 72096750.0,
      "step": 31445
    },
    {
      "entropy": 5.395013761520386,
      "epoch": 3.021133525456292,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040988911667420305,
      "loss": 4.6852,
      "mean_token_accuracy": 0.24220336973667145,
      "num_tokens": 72108842.0,
      "step": 31450
    },
    {
      "entropy": 5.452181386947632,
      "epoch": 3.021613832853026,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004098616756006015,
      "loss": 4.7485,
      "mean_token_accuracy": 0.24758126884698867,
      "num_tokens": 72119582.0,
      "step": 31455
    },
    {
      "entropy": 5.475160455703735,
      "epoch": 3.02209414024976,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004098342313959122,
      "loss": 4.7817,
      "mean_token_accuracy": 0.24983332306146622,
      "num_tokens": 72129984.0,
      "step": 31460
    },
    {
      "entropy": 5.442875099182129,
      "epoch": 3.0225744476464937,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004098067840607725,
      "loss": 4.7286,
      "mean_token_accuracy": 0.24402923434972762,
      "num_tokens": 72142160.0,
      "step": 31465
    },
    {
      "entropy": 5.509760761260987,
      "epoch": 3.0230547550432276,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004097793335958195,
      "loss": 4.7949,
      "mean_token_accuracy": 0.23985414803028107,
      "num_tokens": 72153691.0,
      "step": 31470
    },
    {
      "entropy": 5.442243385314941,
      "epoch": 3.0235350624399615,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040975188000169074,
      "loss": 4.706,
      "mean_token_accuracy": 0.25244276970624924,
      "num_tokens": 72165123.0,
      "step": 31475
    },
    {
      "entropy": 5.397859191894531,
      "epoch": 3.0240153698366954,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040972442327902325,
      "loss": 4.6925,
      "mean_token_accuracy": 0.24930056780576706,
      "num_tokens": 72177006.0,
      "step": 31480
    },
    {
      "entropy": 5.463469171524048,
      "epoch": 3.0244956772334293,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004096969634284549,
      "loss": 4.7724,
      "mean_token_accuracy": 0.24546630084514617,
      "num_tokens": 72189185.0,
      "step": 31485
    },
    {
      "entropy": 5.4196553230285645,
      "epoch": 3.024975984630163,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004096695004506229,
      "loss": 4.7265,
      "mean_token_accuracy": 0.24816420376300813,
      "num_tokens": 72201530.0,
      "step": 31490
    },
    {
      "entropy": 5.408088636398316,
      "epoch": 3.025456292026897,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040964203434616496,
      "loss": 4.6746,
      "mean_token_accuracy": 0.2599488914012909,
      "num_tokens": 72213514.0,
      "step": 31495
    },
    {
      "entropy": 5.392025184631348,
      "epoch": 3.025936599423631,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004096145651157189,
      "loss": 4.6696,
      "mean_token_accuracy": 0.25082807838916776,
      "num_tokens": 72224811.0,
      "step": 31500
    },
    {
      "entropy": 5.477365493774414,
      "epoch": 3.0264169068203652,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004095870927599223,
      "loss": 4.8065,
      "mean_token_accuracy": 0.24190468192100525,
      "num_tokens": 72236201.0,
      "step": 31505
    },
    {
      "entropy": 5.462470865249633,
      "epoch": 3.026897214217099,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040955961727941306,
      "loss": 4.7417,
      "mean_token_accuracy": 0.24195805788040162,
      "num_tokens": 72248963.0,
      "step": 31510
    },
    {
      "entropy": 5.475694465637207,
      "epoch": 3.027377521613833,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004095321386748291,
      "loss": 4.7485,
      "mean_token_accuracy": 0.25027197450399397,
      "num_tokens": 72259489.0,
      "step": 31515
    },
    {
      "entropy": 5.3750862121582035,
      "epoch": 3.027857829010567,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040950465694680825,
      "loss": 4.7167,
      "mean_token_accuracy": 0.24904489368200303,
      "num_tokens": 72271124.0,
      "step": 31520
    },
    {
      "entropy": 5.471638965606689,
      "epoch": 3.0283381364073008,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040947717209598877,
      "loss": 4.7464,
      "mean_token_accuracy": 0.24667309820652009,
      "num_tokens": 72281144.0,
      "step": 31525
    },
    {
      "entropy": 5.41636815071106,
      "epoch": 3.0288184438040346,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040944968412300867,
      "loss": 4.6903,
      "mean_token_accuracy": 0.25304004102945327,
      "num_tokens": 72293422.0,
      "step": 31530
    },
    {
      "entropy": 5.410150241851807,
      "epoch": 3.0292987512007685,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040942219302850605,
      "loss": 4.722,
      "mean_token_accuracy": 0.2530945032835007,
      "num_tokens": 72304450.0,
      "step": 31535
    },
    {
      "entropy": 5.427745246887207,
      "epoch": 3.0297790585975024,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004093946988131193,
      "loss": 4.6718,
      "mean_token_accuracy": 0.258553808927536,
      "num_tokens": 72315000.0,
      "step": 31540
    },
    {
      "entropy": 5.459435987472534,
      "epoch": 3.0302593659942363,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004093672014774868,
      "loss": 4.7343,
      "mean_token_accuracy": 0.24740231186151504,
      "num_tokens": 72327268.0,
      "step": 31545
    },
    {
      "entropy": 5.3902990341186525,
      "epoch": 3.03073967339097,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040933970102224675,
      "loss": 4.6467,
      "mean_token_accuracy": 0.25879779905080796,
      "num_tokens": 72338371.0,
      "step": 31550
    },
    {
      "entropy": 5.4619041919708256,
      "epoch": 3.031219980787704,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040931219744803774,
      "loss": 4.6951,
      "mean_token_accuracy": 0.2499908283352852,
      "num_tokens": 72349211.0,
      "step": 31555
    },
    {
      "entropy": 5.480604410171509,
      "epoch": 3.031700288184438,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004092846907554985,
      "loss": 4.7599,
      "mean_token_accuracy": 0.2451999694108963,
      "num_tokens": 72361634.0,
      "step": 31560
    },
    {
      "entropy": 5.403137350082398,
      "epoch": 3.032180595581172,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040925718094526724,
      "loss": 4.716,
      "mean_token_accuracy": 0.2498915895819664,
      "num_tokens": 72374071.0,
      "step": 31565
    },
    {
      "entropy": 5.50739541053772,
      "epoch": 3.0326609029779057,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040922966801798305,
      "loss": 4.8194,
      "mean_token_accuracy": 0.24797720909118653,
      "num_tokens": 72386272.0,
      "step": 31570
    },
    {
      "entropy": 5.469808292388916,
      "epoch": 3.0331412103746396,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040920215197428456,
      "loss": 4.705,
      "mean_token_accuracy": 0.25158767551183703,
      "num_tokens": 72398832.0,
      "step": 31575
    },
    {
      "entropy": 5.529854679107666,
      "epoch": 3.0336215177713735,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040917463281481053,
      "loss": 4.7938,
      "mean_token_accuracy": 0.2387840762734413,
      "num_tokens": 72409970.0,
      "step": 31580
    },
    {
      "entropy": 5.434739351272583,
      "epoch": 3.034101825168108,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004091471105402,
      "loss": 4.7015,
      "mean_token_accuracy": 0.2549155533313751,
      "num_tokens": 72420640.0,
      "step": 31585
    },
    {
      "entropy": 5.455982637405396,
      "epoch": 3.0345821325648417,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004091195851510918,
      "loss": 4.757,
      "mean_token_accuracy": 0.24885769188404083,
      "num_tokens": 72432111.0,
      "step": 31590
    },
    {
      "entropy": 5.3829183101654055,
      "epoch": 3.0350624399615755,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004090920566481252,
      "loss": 4.6605,
      "mean_token_accuracy": 0.25713742077350615,
      "num_tokens": 72444427.0,
      "step": 31595
    },
    {
      "entropy": 5.487593412399292,
      "epoch": 3.0355427473583094,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004090645250319392,
      "loss": 4.7855,
      "mean_token_accuracy": 0.24672870635986327,
      "num_tokens": 72454955.0,
      "step": 31600
    },
    {
      "entropy": 5.393697881698609,
      "epoch": 3.0360230547550433,
      "grad_norm": 1.171875,
      "learning_rate": 0.000409036990303173,
      "loss": 4.6128,
      "mean_token_accuracy": 0.259036922454834,
      "num_tokens": 72464914.0,
      "step": 31605
    },
    {
      "entropy": 5.405201816558838,
      "epoch": 3.036503362151777,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004090094524624658,
      "loss": 4.7457,
      "mean_token_accuracy": 0.2552265331149101,
      "num_tokens": 72476379.0,
      "step": 31610
    },
    {
      "entropy": 5.482214879989624,
      "epoch": 3.036983669548511,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040898191151045717,
      "loss": 4.7279,
      "mean_token_accuracy": 0.24659708738327027,
      "num_tokens": 72488249.0,
      "step": 31615
    },
    {
      "entropy": 5.439580678939819,
      "epoch": 3.037463976945245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004089543674477864,
      "loss": 4.6864,
      "mean_token_accuracy": 0.24815485924482344,
      "num_tokens": 72499365.0,
      "step": 31620
    },
    {
      "entropy": 5.3949812889099125,
      "epoch": 3.037944284341979,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004089268202750929,
      "loss": 4.6826,
      "mean_token_accuracy": 0.25092879831790926,
      "num_tokens": 72511231.0,
      "step": 31625
    },
    {
      "entropy": 5.403468608856201,
      "epoch": 3.0384245917387127,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040889926999301634,
      "loss": 4.6784,
      "mean_token_accuracy": 0.25134842842817307,
      "num_tokens": 72524357.0,
      "step": 31630
    },
    {
      "entropy": 5.447069835662842,
      "epoch": 3.0389048991354466,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004088717166021964,
      "loss": 4.7148,
      "mean_token_accuracy": 0.2477050691843033,
      "num_tokens": 72537170.0,
      "step": 31635
    },
    {
      "entropy": 5.406735467910766,
      "epoch": 3.0393852065321805,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004088441601032727,
      "loss": 4.6439,
      "mean_token_accuracy": 0.25877716541290285,
      "num_tokens": 72548631.0,
      "step": 31640
    },
    {
      "entropy": 5.38604736328125,
      "epoch": 3.0398655139289144,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004088166004968849,
      "loss": 4.6534,
      "mean_token_accuracy": 0.24561428874731064,
      "num_tokens": 72561787.0,
      "step": 31645
    },
    {
      "entropy": 5.392182779312134,
      "epoch": 3.0403458213256482,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040878903778367317,
      "loss": 4.7029,
      "mean_token_accuracy": 0.25232123136520385,
      "num_tokens": 72572480.0,
      "step": 31650
    },
    {
      "entropy": 5.422531366348267,
      "epoch": 3.040826128722382,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004087614719642772,
      "loss": 4.7308,
      "mean_token_accuracy": 0.24184340834617615,
      "num_tokens": 72584636.0,
      "step": 31655
    },
    {
      "entropy": 5.49031629562378,
      "epoch": 3.0413064361191164,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040873390303933693,
      "loss": 4.7527,
      "mean_token_accuracy": 0.24294841885566712,
      "num_tokens": 72595425.0,
      "step": 31660
    },
    {
      "entropy": 5.511763429641723,
      "epoch": 3.0417867435158503,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040870633100949266,
      "loss": 4.7629,
      "mean_token_accuracy": 0.24646903127431868,
      "num_tokens": 72607344.0,
      "step": 31665
    },
    {
      "entropy": 5.404758262634277,
      "epoch": 3.042267050912584,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040867875587538436,
      "loss": 4.7642,
      "mean_token_accuracy": 0.25239389538764956,
      "num_tokens": 72619704.0,
      "step": 31670
    },
    {
      "entropy": 5.401260089874268,
      "epoch": 3.042747358309318,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004086511776376523,
      "loss": 4.6737,
      "mean_token_accuracy": 0.25864939838647844,
      "num_tokens": 72632143.0,
      "step": 31675
    },
    {
      "entropy": 5.405474138259888,
      "epoch": 3.043227665706052,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040862359629693684,
      "loss": 4.699,
      "mean_token_accuracy": 0.24872822165489197,
      "num_tokens": 72644581.0,
      "step": 31680
    },
    {
      "entropy": 5.470728492736816,
      "epoch": 3.043707973102786,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004085960118538781,
      "loss": 4.7724,
      "mean_token_accuracy": 0.24473333358764648,
      "num_tokens": 72656309.0,
      "step": 31685
    },
    {
      "entropy": 5.473985862731934,
      "epoch": 3.0441882804995197,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004085684243091168,
      "loss": 4.7221,
      "mean_token_accuracy": 0.25143611878156663,
      "num_tokens": 72668347.0,
      "step": 31690
    },
    {
      "entropy": 5.371982049942017,
      "epoch": 3.0446685878962536,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004085408336632933,
      "loss": 4.6597,
      "mean_token_accuracy": 0.2534921243786812,
      "num_tokens": 72679375.0,
      "step": 31695
    },
    {
      "entropy": 5.411968755722046,
      "epoch": 3.0451488952929875,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040851323991704803,
      "loss": 4.7111,
      "mean_token_accuracy": 0.24489734023809434,
      "num_tokens": 72691020.0,
      "step": 31700
    },
    {
      "entropy": 5.405348873138427,
      "epoch": 3.0456292026897214,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004084856430710219,
      "loss": 4.6694,
      "mean_token_accuracy": 0.25402964651584625,
      "num_tokens": 72703215.0,
      "step": 31705
    },
    {
      "entropy": 5.452715682983398,
      "epoch": 3.0461095100864553,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004084580431258555,
      "loss": 4.6877,
      "mean_token_accuracy": 0.24958547949790955,
      "num_tokens": 72715107.0,
      "step": 31710
    },
    {
      "entropy": 5.43934440612793,
      "epoch": 3.046589817483189,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004084304400821896,
      "loss": 4.7689,
      "mean_token_accuracy": 0.24831286519765855,
      "num_tokens": 72726301.0,
      "step": 31715
    },
    {
      "entropy": 5.405560684204102,
      "epoch": 3.047070124879923,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004084028339406651,
      "loss": 4.7083,
      "mean_token_accuracy": 0.25415685921907427,
      "num_tokens": 72738102.0,
      "step": 31720
    },
    {
      "entropy": 5.439062261581421,
      "epoch": 3.047550432276657,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040837522470192297,
      "loss": 4.7502,
      "mean_token_accuracy": 0.25139717757701874,
      "num_tokens": 72748840.0,
      "step": 31725
    },
    {
      "entropy": 5.338397216796875,
      "epoch": 3.0480307396733908,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004083476123666041,
      "loss": 4.5963,
      "mean_token_accuracy": 0.2661815360188484,
      "num_tokens": 72760480.0,
      "step": 31730
    },
    {
      "entropy": 5.419046545028687,
      "epoch": 3.048511047070125,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004083199969353496,
      "loss": 4.6519,
      "mean_token_accuracy": 0.25800025165081025,
      "num_tokens": 72770446.0,
      "step": 31735
    },
    {
      "entropy": 5.39647159576416,
      "epoch": 3.048991354466859,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040829237840880075,
      "loss": 4.6916,
      "mean_token_accuracy": 0.24629432260990142,
      "num_tokens": 72782262.0,
      "step": 31740
    },
    {
      "entropy": 5.441455364227295,
      "epoch": 3.049471661863593,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040826475678759855,
      "loss": 4.6889,
      "mean_token_accuracy": 0.255404444038868,
      "num_tokens": 72793857.0,
      "step": 31745
    },
    {
      "entropy": 5.413947105407715,
      "epoch": 3.0499519692603267,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004082371320723845,
      "loss": 4.7395,
      "mean_token_accuracy": 0.2454444020986557,
      "num_tokens": 72805396.0,
      "step": 31750
    },
    {
      "entropy": 5.4392200946807865,
      "epoch": 3.0504322766570606,
      "grad_norm": 1.125,
      "learning_rate": 0.00040820950426379986,
      "loss": 4.7418,
      "mean_token_accuracy": 0.24927812218666076,
      "num_tokens": 72816946.0,
      "step": 31755
    },
    {
      "entropy": 5.421645641326904,
      "epoch": 3.0509125840537945,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004081818733624861,
      "loss": 4.6349,
      "mean_token_accuracy": 0.2552956581115723,
      "num_tokens": 72827589.0,
      "step": 31760
    },
    {
      "entropy": 5.385325622558594,
      "epoch": 3.0513928914505284,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004081542393690847,
      "loss": 4.6892,
      "mean_token_accuracy": 0.2490872859954834,
      "num_tokens": 72838333.0,
      "step": 31765
    },
    {
      "entropy": 5.392614030838013,
      "epoch": 3.0518731988472623,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004081266022842372,
      "loss": 4.6777,
      "mean_token_accuracy": 0.24939001053571702,
      "num_tokens": 72850212.0,
      "step": 31770
    },
    {
      "entropy": 5.4534789562225345,
      "epoch": 3.052353506243996,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040809896210858537,
      "loss": 4.6406,
      "mean_token_accuracy": 0.25279958099126815,
      "num_tokens": 72861406.0,
      "step": 31775
    },
    {
      "entropy": 5.421908760070801,
      "epoch": 3.05283381364073,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040807131884277085,
      "loss": 4.7069,
      "mean_token_accuracy": 0.2563543230295181,
      "num_tokens": 72872563.0,
      "step": 31780
    },
    {
      "entropy": 5.445491695404053,
      "epoch": 3.053314121037464,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004080436724874354,
      "loss": 4.7532,
      "mean_token_accuracy": 0.24389543384313583,
      "num_tokens": 72883718.0,
      "step": 31785
    },
    {
      "entropy": 5.494697427749633,
      "epoch": 3.053794428434198,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040801602304322095,
      "loss": 4.7733,
      "mean_token_accuracy": 0.24995149224996566,
      "num_tokens": 72896423.0,
      "step": 31790
    },
    {
      "entropy": 5.5271608352661135,
      "epoch": 3.0542747358309317,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040798837051076944,
      "loss": 4.8277,
      "mean_token_accuracy": 0.23709163665771485,
      "num_tokens": 72907681.0,
      "step": 31795
    },
    {
      "entropy": 5.3535003662109375,
      "epoch": 3.0547550432276656,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040796071489072286,
      "loss": 4.5914,
      "mean_token_accuracy": 0.2591127544641495,
      "num_tokens": 72919782.0,
      "step": 31800
    },
    {
      "entropy": 5.4183971881866455,
      "epoch": 3.0552353506243994,
      "grad_norm": 1.125,
      "learning_rate": 0.0004079330561837233,
      "loss": 4.6611,
      "mean_token_accuracy": 0.25479639917612074,
      "num_tokens": 72931458.0,
      "step": 31805
    },
    {
      "entropy": 5.4595225811004635,
      "epoch": 3.0557156580211333,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040790539439041287,
      "loss": 4.7578,
      "mean_token_accuracy": 0.24716752916574478,
      "num_tokens": 72942415.0,
      "step": 31810
    },
    {
      "entropy": 5.367463445663452,
      "epoch": 3.0561959654178676,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040787772951143386,
      "loss": 4.5278,
      "mean_token_accuracy": 0.26195010393857954,
      "num_tokens": 72953134.0,
      "step": 31815
    },
    {
      "entropy": 5.358659124374389,
      "epoch": 3.0566762728146015,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004078500615474285,
      "loss": 4.6509,
      "mean_token_accuracy": 0.25069466084241865,
      "num_tokens": 72964390.0,
      "step": 31820
    },
    {
      "entropy": 5.376478147506714,
      "epoch": 3.0571565802113354,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040782239049903926,
      "loss": 4.6923,
      "mean_token_accuracy": 0.2502450466156006,
      "num_tokens": 72976899.0,
      "step": 31825
    },
    {
      "entropy": 5.454514694213867,
      "epoch": 3.0576368876080693,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040779471636690845,
      "loss": 4.7579,
      "mean_token_accuracy": 0.25231370627880095,
      "num_tokens": 72989074.0,
      "step": 31830
    },
    {
      "entropy": 5.451037883758545,
      "epoch": 3.058117195004803,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040776703915167866,
      "loss": 4.6965,
      "mean_token_accuracy": 0.24929951131343842,
      "num_tokens": 73000434.0,
      "step": 31835
    },
    {
      "entropy": 5.400342512130737,
      "epoch": 3.058597502401537,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040773935885399254,
      "loss": 4.6696,
      "mean_token_accuracy": 0.2528528407216072,
      "num_tokens": 73011902.0,
      "step": 31840
    },
    {
      "entropy": 5.476351547241211,
      "epoch": 3.059077809798271,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004077116754744926,
      "loss": 4.7851,
      "mean_token_accuracy": 0.24447929710149766,
      "num_tokens": 73022349.0,
      "step": 31845
    },
    {
      "entropy": 5.389696741104126,
      "epoch": 3.059558117195005,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040768398901382157,
      "loss": 4.6408,
      "mean_token_accuracy": 0.25907291322946546,
      "num_tokens": 73033958.0,
      "step": 31850
    },
    {
      "entropy": 5.424142551422119,
      "epoch": 3.0600384245917387,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004076562994726223,
      "loss": 4.6458,
      "mean_token_accuracy": 0.2594060435891151,
      "num_tokens": 73045807.0,
      "step": 31855
    },
    {
      "entropy": 5.410061550140381,
      "epoch": 3.0605187319884726,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004076286068515378,
      "loss": 4.6454,
      "mean_token_accuracy": 0.25745212882757185,
      "num_tokens": 73056883.0,
      "step": 31860
    },
    {
      "entropy": 5.439346742630005,
      "epoch": 3.0609990393852065,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004076009111512108,
      "loss": 4.6678,
      "mean_token_accuracy": 0.25388701260089874,
      "num_tokens": 73067924.0,
      "step": 31865
    },
    {
      "entropy": 5.431147956848145,
      "epoch": 3.0614793467819403,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004075732123722844,
      "loss": 4.714,
      "mean_token_accuracy": 0.24784445762634277,
      "num_tokens": 73079965.0,
      "step": 31870
    },
    {
      "entropy": 5.410074043273926,
      "epoch": 3.061959654178674,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004075455105154016,
      "loss": 4.727,
      "mean_token_accuracy": 0.24908010214567183,
      "num_tokens": 73091986.0,
      "step": 31875
    },
    {
      "entropy": 5.470567274093628,
      "epoch": 3.062439961575408,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040751780558120573,
      "loss": 4.7346,
      "mean_token_accuracy": 0.24198924303054808,
      "num_tokens": 73102445.0,
      "step": 31880
    },
    {
      "entropy": 5.44150013923645,
      "epoch": 3.062920268972142,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004074900975703398,
      "loss": 4.6739,
      "mean_token_accuracy": 0.25238869041204454,
      "num_tokens": 73112921.0,
      "step": 31885
    },
    {
      "entropy": 5.4417212963104244,
      "epoch": 3.063400576368876,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004074623864834473,
      "loss": 4.6865,
      "mean_token_accuracy": 0.24847902953624726,
      "num_tokens": 73123653.0,
      "step": 31890
    },
    {
      "entropy": 5.443913173675537,
      "epoch": 3.06388088376561,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004074346723211715,
      "loss": 4.6893,
      "mean_token_accuracy": 0.25654398947954177,
      "num_tokens": 73135568.0,
      "step": 31895
    },
    {
      "entropy": 5.400175380706787,
      "epoch": 3.064361191162344,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040740695508415583,
      "loss": 4.6597,
      "mean_token_accuracy": 0.25320352911949157,
      "num_tokens": 73145964.0,
      "step": 31900
    },
    {
      "entropy": 5.338721227645874,
      "epoch": 3.064841498559078,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040737923477304386,
      "loss": 4.58,
      "mean_token_accuracy": 0.26180414110422134,
      "num_tokens": 73156941.0,
      "step": 31905
    },
    {
      "entropy": 5.382403945922851,
      "epoch": 3.065321805955812,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040735151138847917,
      "loss": 4.6228,
      "mean_token_accuracy": 0.25594266802072524,
      "num_tokens": 73167928.0,
      "step": 31910
    },
    {
      "entropy": 5.3222486019134525,
      "epoch": 3.0658021133525457,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004073237849311053,
      "loss": 4.6179,
      "mean_token_accuracy": 0.2580364987254143,
      "num_tokens": 73179861.0,
      "step": 31915
    },
    {
      "entropy": 5.404750013351441,
      "epoch": 3.0662824207492796,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004072960554015661,
      "loss": 4.7335,
      "mean_token_accuracy": 0.24946896433830262,
      "num_tokens": 73191152.0,
      "step": 31920
    },
    {
      "entropy": 5.40966763496399,
      "epoch": 3.0667627281460135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004072683228005055,
      "loss": 4.684,
      "mean_token_accuracy": 0.25537008345127105,
      "num_tokens": 73202952.0,
      "step": 31925
    },
    {
      "entropy": 5.343868827819824,
      "epoch": 3.0672430355427474,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040724058712856697,
      "loss": 4.5928,
      "mean_token_accuracy": 0.25902306437492373,
      "num_tokens": 73214899.0,
      "step": 31930
    },
    {
      "entropy": 5.4017585754394535,
      "epoch": 3.0677233429394812,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004072128483863948,
      "loss": 4.7377,
      "mean_token_accuracy": 0.25197691917419435,
      "num_tokens": 73226599.0,
      "step": 31935
    },
    {
      "entropy": 5.366008710861206,
      "epoch": 3.068203650336215,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004071851065746328,
      "loss": 4.6092,
      "mean_token_accuracy": 0.2522598564624786,
      "num_tokens": 73238588.0,
      "step": 31940
    },
    {
      "entropy": 5.479011297225952,
      "epoch": 3.068683957732949,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004071573616939252,
      "loss": 4.7201,
      "mean_token_accuracy": 0.2491963431239128,
      "num_tokens": 73250727.0,
      "step": 31945
    },
    {
      "entropy": 5.396910953521728,
      "epoch": 3.069164265129683,
      "grad_norm": 1.125,
      "learning_rate": 0.0004071296137449161,
      "loss": 4.6483,
      "mean_token_accuracy": 0.26274738311767576,
      "num_tokens": 73261937.0,
      "step": 31950
    },
    {
      "entropy": 5.347153854370117,
      "epoch": 3.0696445725264168,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040710186272824967,
      "loss": 4.6514,
      "mean_token_accuracy": 0.2518037989735603,
      "num_tokens": 73274853.0,
      "step": 31955
    },
    {
      "entropy": 5.463469076156616,
      "epoch": 3.0701248799231506,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004070741086445703,
      "loss": 4.7387,
      "mean_token_accuracy": 0.24824900627136232,
      "num_tokens": 73284901.0,
      "step": 31960
    },
    {
      "entropy": 5.388898611068726,
      "epoch": 3.0706051873198845,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040704635149452223,
      "loss": 4.6991,
      "mean_token_accuracy": 0.24568756222724913,
      "num_tokens": 73295989.0,
      "step": 31965
    },
    {
      "entropy": 5.512873125076294,
      "epoch": 3.071085494716619,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040701859127875,
      "loss": 4.7844,
      "mean_token_accuracy": 0.24035003185272216,
      "num_tokens": 73306661.0,
      "step": 31970
    },
    {
      "entropy": 5.468945837020874,
      "epoch": 3.0715658021133527,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040699082799789814,
      "loss": 4.7709,
      "mean_token_accuracy": 0.24048351645469665,
      "num_tokens": 73318456.0,
      "step": 31975
    },
    {
      "entropy": 5.462014245986938,
      "epoch": 3.0720461095100866,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040696306165261117,
      "loss": 4.7797,
      "mean_token_accuracy": 0.24688344299793244,
      "num_tokens": 73330577.0,
      "step": 31980
    },
    {
      "entropy": 5.476660203933716,
      "epoch": 3.0725264169068205,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004069352922435337,
      "loss": 4.7267,
      "mean_token_accuracy": 0.25395661890506743,
      "num_tokens": 73341682.0,
      "step": 31985
    },
    {
      "entropy": 5.409141969680786,
      "epoch": 3.0730067243035544,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004069075197713106,
      "loss": 4.6999,
      "mean_token_accuracy": 0.2579110190272331,
      "num_tokens": 73352924.0,
      "step": 31990
    },
    {
      "entropy": 5.41199426651001,
      "epoch": 3.0734870317002883,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040687974423658655,
      "loss": 4.7083,
      "mean_token_accuracy": 0.25305051505565646,
      "num_tokens": 73364904.0,
      "step": 31995
    },
    {
      "entropy": 5.467185592651367,
      "epoch": 3.073967339097022,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040685196564000644,
      "loss": 4.7437,
      "mean_token_accuracy": 0.25133312940597535,
      "num_tokens": 73376680.0,
      "step": 32000
    },
    {
      "entropy": 5.4096832275390625,
      "epoch": 3.074447646493756,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040682418398221517,
      "loss": 4.6828,
      "mean_token_accuracy": 0.25394040048122407,
      "num_tokens": 73388631.0,
      "step": 32005
    },
    {
      "entropy": 5.394646406173706,
      "epoch": 3.07492795389049,
      "grad_norm": 1.125,
      "learning_rate": 0.00040679639926385783,
      "loss": 4.6556,
      "mean_token_accuracy": 0.2569452330470085,
      "num_tokens": 73400953.0,
      "step": 32010
    },
    {
      "entropy": 5.398634624481201,
      "epoch": 3.0754082612872238,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004067686114855794,
      "loss": 4.6041,
      "mean_token_accuracy": 0.2564701303839684,
      "num_tokens": 73411167.0,
      "step": 32015
    },
    {
      "entropy": 5.308679008483887,
      "epoch": 3.0758885686839577,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040674082064802507,
      "loss": 4.6254,
      "mean_token_accuracy": 0.25669390261173247,
      "num_tokens": 73422434.0,
      "step": 32020
    },
    {
      "entropy": 5.473992919921875,
      "epoch": 3.0763688760806915,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004067130267518401,
      "loss": 4.7266,
      "mean_token_accuracy": 0.24629221260547637,
      "num_tokens": 73433543.0,
      "step": 32025
    },
    {
      "entropy": 5.5504766464233395,
      "epoch": 3.0768491834774254,
      "grad_norm": 1.25,
      "learning_rate": 0.0004066852297976698,
      "loss": 4.8945,
      "mean_token_accuracy": 0.2339830905199051,
      "num_tokens": 73444624.0,
      "step": 32030
    },
    {
      "entropy": 5.421989631652832,
      "epoch": 3.0773294908741593,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004066574297861595,
      "loss": 4.6851,
      "mean_token_accuracy": 0.2580222517251968,
      "num_tokens": 73454446.0,
      "step": 32035
    },
    {
      "entropy": 5.402618455886841,
      "epoch": 3.077809798270893,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040662962671795454,
      "loss": 4.6853,
      "mean_token_accuracy": 0.24495853185653688,
      "num_tokens": 73465557.0,
      "step": 32040
    },
    {
      "entropy": 5.463791608810425,
      "epoch": 3.0782901056676275,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004066018205937006,
      "loss": 4.6853,
      "mean_token_accuracy": 0.24628733843564987,
      "num_tokens": 73477939.0,
      "step": 32045
    },
    {
      "entropy": 5.40993914604187,
      "epoch": 3.0787704130643614,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004065740114140431,
      "loss": 4.7429,
      "mean_token_accuracy": 0.2548402935266495,
      "num_tokens": 73490715.0,
      "step": 32050
    },
    {
      "entropy": 5.395765495300293,
      "epoch": 3.0792507204610953,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040654619917962774,
      "loss": 4.6436,
      "mean_token_accuracy": 0.25523015260696413,
      "num_tokens": 73503147.0,
      "step": 32055
    },
    {
      "entropy": 5.476238822937011,
      "epoch": 3.079731027857829,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004065183838911003,
      "loss": 4.7447,
      "mean_token_accuracy": 0.24487211108207702,
      "num_tokens": 73514750.0,
      "step": 32060
    },
    {
      "entropy": 5.41054859161377,
      "epoch": 3.080211335254563,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004064905655491065,
      "loss": 4.6948,
      "mean_token_accuracy": 0.2520566269755363,
      "num_tokens": 73525845.0,
      "step": 32065
    },
    {
      "entropy": 5.47278356552124,
      "epoch": 3.080691642651297,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040646274415429224,
      "loss": 4.7527,
      "mean_token_accuracy": 0.2443629801273346,
      "num_tokens": 73537086.0,
      "step": 32070
    },
    {
      "entropy": 5.380621433258057,
      "epoch": 3.081171950048031,
      "grad_norm": 1.125,
      "learning_rate": 0.0004064349197073033,
      "loss": 4.6512,
      "mean_token_accuracy": 0.25645876228809356,
      "num_tokens": 73548708.0,
      "step": 32075
    },
    {
      "entropy": 5.476252794265747,
      "epoch": 3.0816522574447647,
      "grad_norm": 1.125,
      "learning_rate": 0.0004064070922087859,
      "loss": 4.7459,
      "mean_token_accuracy": 0.24463370442390442,
      "num_tokens": 73559703.0,
      "step": 32080
    },
    {
      "entropy": 5.373406314849854,
      "epoch": 3.0821325648414986,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040637926165938606,
      "loss": 4.6297,
      "mean_token_accuracy": 0.25520001649856566,
      "num_tokens": 73570858.0,
      "step": 32085
    },
    {
      "entropy": 5.4235001564025875,
      "epoch": 3.0826128722382324,
      "grad_norm": 1.125,
      "learning_rate": 0.00040635142805974986,
      "loss": 4.7147,
      "mean_token_accuracy": 0.2496036022901535,
      "num_tokens": 73583442.0,
      "step": 32090
    },
    {
      "entropy": 5.446864080429077,
      "epoch": 3.0830931796349663,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004063235914105235,
      "loss": 4.7245,
      "mean_token_accuracy": 0.24680494517087936,
      "num_tokens": 73595336.0,
      "step": 32095
    },
    {
      "entropy": 5.377295351028442,
      "epoch": 3.0835734870317,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040629575171235327,
      "loss": 4.7108,
      "mean_token_accuracy": 0.24217517524957657,
      "num_tokens": 73606582.0,
      "step": 32100
    },
    {
      "entropy": 5.42228331565857,
      "epoch": 3.084053794428434,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004062679089658856,
      "loss": 4.7198,
      "mean_token_accuracy": 0.252319772541523,
      "num_tokens": 73618432.0,
      "step": 32105
    },
    {
      "entropy": 5.408816146850586,
      "epoch": 3.084534101825168,
      "grad_norm": 1.125,
      "learning_rate": 0.00040624006317176685,
      "loss": 4.6016,
      "mean_token_accuracy": 0.25887449979782107,
      "num_tokens": 73628880.0,
      "step": 32110
    },
    {
      "entropy": 5.412018585205078,
      "epoch": 3.085014409221902,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040621221433064354,
      "loss": 4.7032,
      "mean_token_accuracy": 0.2445244774222374,
      "num_tokens": 73639813.0,
      "step": 32115
    },
    {
      "entropy": 5.561542129516601,
      "epoch": 3.0854947166186357,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004061843624431623,
      "loss": 4.8721,
      "mean_token_accuracy": 0.23925027549266814,
      "num_tokens": 73651276.0,
      "step": 32120
    },
    {
      "entropy": 5.3947512149810795,
      "epoch": 3.08597502401537,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040615650750996956,
      "loss": 4.6413,
      "mean_token_accuracy": 0.25298803448677065,
      "num_tokens": 73663869.0,
      "step": 32125
    },
    {
      "entropy": 5.405317974090576,
      "epoch": 3.086455331412104,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040612864953171223,
      "loss": 4.7232,
      "mean_token_accuracy": 0.24248720109462737,
      "num_tokens": 73676089.0,
      "step": 32130
    },
    {
      "entropy": 5.392791271209717,
      "epoch": 3.086935638808838,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040610078850903715,
      "loss": 4.6935,
      "mean_token_accuracy": 0.25081963539123536,
      "num_tokens": 73688718.0,
      "step": 32135
    },
    {
      "entropy": 5.387832832336426,
      "epoch": 3.0874159462055717,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040607292444259094,
      "loss": 4.6498,
      "mean_token_accuracy": 0.25693250447511673,
      "num_tokens": 73700363.0,
      "step": 32140
    },
    {
      "entropy": 5.302104473114014,
      "epoch": 3.0878962536023056,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004060450573330206,
      "loss": 4.5606,
      "mean_token_accuracy": 0.25906648933887483,
      "num_tokens": 73711873.0,
      "step": 32145
    },
    {
      "entropy": 5.389756393432617,
      "epoch": 3.0883765609990395,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040601718718097325,
      "loss": 4.6155,
      "mean_token_accuracy": 0.2518511891365051,
      "num_tokens": 73722466.0,
      "step": 32150
    },
    {
      "entropy": 5.424039649963379,
      "epoch": 3.0888568683957733,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040598931398709576,
      "loss": 4.6907,
      "mean_token_accuracy": 0.25375115275383,
      "num_tokens": 73734712.0,
      "step": 32155
    },
    {
      "entropy": 5.5696734428405765,
      "epoch": 3.089337175792507,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040596143775203534,
      "loss": 4.8613,
      "mean_token_accuracy": 0.2359880045056343,
      "num_tokens": 73746367.0,
      "step": 32160
    },
    {
      "entropy": 5.399368619918823,
      "epoch": 3.089817483189241,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040593355847643933,
      "loss": 4.6292,
      "mean_token_accuracy": 0.2647946417331696,
      "num_tokens": 73758223.0,
      "step": 32165
    },
    {
      "entropy": 5.415778350830078,
      "epoch": 3.090297790585975,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004059056761609548,
      "loss": 4.7209,
      "mean_token_accuracy": 0.24606593400239946,
      "num_tokens": 73769278.0,
      "step": 32170
    },
    {
      "entropy": 5.417026662826538,
      "epoch": 3.090778097982709,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004058777908062292,
      "loss": 4.684,
      "mean_token_accuracy": 0.2490689292550087,
      "num_tokens": 73778999.0,
      "step": 32175
    },
    {
      "entropy": 5.409389209747315,
      "epoch": 3.0912584053794427,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004058499024129099,
      "loss": 4.6853,
      "mean_token_accuracy": 0.2494792729616165,
      "num_tokens": 73789556.0,
      "step": 32180
    },
    {
      "entropy": 5.430225658416748,
      "epoch": 3.0917387127761766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040582201098164443,
      "loss": 4.692,
      "mean_token_accuracy": 0.25642050355672835,
      "num_tokens": 73800691.0,
      "step": 32185
    },
    {
      "entropy": 5.3930583000183105,
      "epoch": 3.0922190201729105,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040579411651308034,
      "loss": 4.764,
      "mean_token_accuracy": 0.24685028493404387,
      "num_tokens": 73811353.0,
      "step": 32190
    },
    {
      "entropy": 5.421144723892212,
      "epoch": 3.0926993275696444,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040576621900786523,
      "loss": 4.6891,
      "mean_token_accuracy": 0.25326237380504607,
      "num_tokens": 73822531.0,
      "step": 32195
    },
    {
      "entropy": 5.432612705230713,
      "epoch": 3.0931796349663783,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004057383184666468,
      "loss": 4.7417,
      "mean_token_accuracy": 0.24348176568746566,
      "num_tokens": 73833857.0,
      "step": 32200
    },
    {
      "entropy": 5.456097841262817,
      "epoch": 3.0936599423631126,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040571041489007286,
      "loss": 4.7209,
      "mean_token_accuracy": 0.24818669706583024,
      "num_tokens": 73845664.0,
      "step": 32205
    },
    {
      "entropy": 5.4222992897033695,
      "epoch": 3.0941402497598465,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040568250827879127,
      "loss": 4.5823,
      "mean_token_accuracy": 0.2606431424617767,
      "num_tokens": 73856860.0,
      "step": 32210
    },
    {
      "entropy": 5.399226713180542,
      "epoch": 3.0946205571565804,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004056545986334497,
      "loss": 4.7079,
      "mean_token_accuracy": 0.25444191247224807,
      "num_tokens": 73867532.0,
      "step": 32215
    },
    {
      "entropy": 5.404533052444458,
      "epoch": 3.0951008645533142,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004056266859546965,
      "loss": 4.7439,
      "mean_token_accuracy": 0.24687238335609435,
      "num_tokens": 73880458.0,
      "step": 32220
    },
    {
      "entropy": 5.482771158218384,
      "epoch": 3.095581171950048,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004055987702431795,
      "loss": 4.7613,
      "mean_token_accuracy": 0.24875346273183824,
      "num_tokens": 73892842.0,
      "step": 32225
    },
    {
      "entropy": 5.456459712982178,
      "epoch": 3.096061479346782,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040557085149954677,
      "loss": 4.7348,
      "mean_token_accuracy": 0.24850642681121826,
      "num_tokens": 73903491.0,
      "step": 32230
    },
    {
      "entropy": 5.388722467422485,
      "epoch": 3.096541786743516,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040554292972444663,
      "loss": 4.6818,
      "mean_token_accuracy": 0.24719207286834716,
      "num_tokens": 73915255.0,
      "step": 32235
    },
    {
      "entropy": 5.373990726470947,
      "epoch": 3.0970220941402498,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040551500491852735,
      "loss": 4.5856,
      "mean_token_accuracy": 0.2642291486263275,
      "num_tokens": 73925571.0,
      "step": 32240
    },
    {
      "entropy": 5.49576153755188,
      "epoch": 3.0975024015369836,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004054870770824371,
      "loss": 4.7687,
      "mean_token_accuracy": 0.24462023824453355,
      "num_tokens": 73936957.0,
      "step": 32245
    },
    {
      "entropy": 5.4745849609375,
      "epoch": 3.0979827089337175,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040545914621682445,
      "loss": 4.63,
      "mean_token_accuracy": 0.2604548841714859,
      "num_tokens": 73948018.0,
      "step": 32250
    },
    {
      "entropy": 5.4173437595367435,
      "epoch": 3.0984630163304514,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004054312123223378,
      "loss": 4.6983,
      "mean_token_accuracy": 0.24737781882286072,
      "num_tokens": 73958713.0,
      "step": 32255
    },
    {
      "entropy": 5.427838897705078,
      "epoch": 3.0989433237271853,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040540327539962567,
      "loss": 4.762,
      "mean_token_accuracy": 0.2467810958623886,
      "num_tokens": 73969299.0,
      "step": 32260
    },
    {
      "entropy": 5.535404634475708,
      "epoch": 3.099423631123919,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040537533544933674,
      "loss": 4.8271,
      "mean_token_accuracy": 0.23553924709558488,
      "num_tokens": 73980149.0,
      "step": 32265
    },
    {
      "entropy": 5.441357660293579,
      "epoch": 3.099903938520653,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004053473924721197,
      "loss": 4.652,
      "mean_token_accuracy": 0.25293754786252975,
      "num_tokens": 73991462.0,
      "step": 32270
    },
    {
      "entropy": 5.47279543876648,
      "epoch": 3.100384245917387,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004053194464686232,
      "loss": 4.7439,
      "mean_token_accuracy": 0.24772390276193618,
      "num_tokens": 74003442.0,
      "step": 32275
    },
    {
      "entropy": 5.312168788909912,
      "epoch": 3.1008645533141213,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004052914974394961,
      "loss": 4.5663,
      "mean_token_accuracy": 0.253504590690136,
      "num_tokens": 74014394.0,
      "step": 32280
    },
    {
      "entropy": 5.487759590148926,
      "epoch": 3.101344860710855,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040526354538538735,
      "loss": 4.8105,
      "mean_token_accuracy": 0.2447645992040634,
      "num_tokens": 74025377.0,
      "step": 32285
    },
    {
      "entropy": 5.433785724639892,
      "epoch": 3.101825168107589,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004052355903069459,
      "loss": 4.6525,
      "mean_token_accuracy": 0.25777705311775206,
      "num_tokens": 74037094.0,
      "step": 32290
    },
    {
      "entropy": 5.339253091812134,
      "epoch": 3.102305475504323,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004052076322048207,
      "loss": 4.5601,
      "mean_token_accuracy": 0.26288551837205887,
      "num_tokens": 74048043.0,
      "step": 32295
    },
    {
      "entropy": 5.388340282440185,
      "epoch": 3.1027857829010568,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040517967107966095,
      "loss": 4.7383,
      "mean_token_accuracy": 0.24906097501516342,
      "num_tokens": 74061026.0,
      "step": 32300
    },
    {
      "entropy": 5.489827394485474,
      "epoch": 3.1032660902977907,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040515170693211584,
      "loss": 4.7797,
      "mean_token_accuracy": 0.24222379177808762,
      "num_tokens": 74072930.0,
      "step": 32305
    },
    {
      "entropy": 5.460755157470703,
      "epoch": 3.1037463976945245,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004051237397628345,
      "loss": 4.7261,
      "mean_token_accuracy": 0.2472880557179451,
      "num_tokens": 74083847.0,
      "step": 32310
    },
    {
      "entropy": 5.445835971832276,
      "epoch": 3.1042267050912584,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004050957695724663,
      "loss": 4.6865,
      "mean_token_accuracy": 0.24893891513347627,
      "num_tokens": 74096187.0,
      "step": 32315
    },
    {
      "entropy": 5.4027598857879635,
      "epoch": 3.1047070124879923,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004050677963616607,
      "loss": 4.6672,
      "mean_token_accuracy": 0.25313600897789,
      "num_tokens": 74108894.0,
      "step": 32320
    },
    {
      "entropy": 5.408213758468628,
      "epoch": 3.105187319884726,
      "grad_norm": 1.125,
      "learning_rate": 0.00040503982013106706,
      "loss": 4.7575,
      "mean_token_accuracy": 0.24820259511470794,
      "num_tokens": 74119941.0,
      "step": 32325
    },
    {
      "entropy": 5.423283624649048,
      "epoch": 3.10566762728146,
      "grad_norm": 1.125,
      "learning_rate": 0.000405011840881335,
      "loss": 4.6637,
      "mean_token_accuracy": 0.25507993400096896,
      "num_tokens": 74131550.0,
      "step": 32330
    },
    {
      "entropy": 5.393262910842895,
      "epoch": 3.106147934678194,
      "grad_norm": 1.125,
      "learning_rate": 0.0004049838586131139,
      "loss": 4.6139,
      "mean_token_accuracy": 0.25770925283432006,
      "num_tokens": 74142530.0,
      "step": 32335
    },
    {
      "entropy": 5.400250625610352,
      "epoch": 3.106628242074928,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004049558733270537,
      "loss": 4.6852,
      "mean_token_accuracy": 0.2505195066332817,
      "num_tokens": 74155117.0,
      "step": 32340
    },
    {
      "entropy": 5.49473614692688,
      "epoch": 3.1071085494716617,
      "grad_norm": 1.046875,
      "learning_rate": 0.000404927885023804,
      "loss": 4.7971,
      "mean_token_accuracy": 0.24157607257366182,
      "num_tokens": 74165880.0,
      "step": 32345
    },
    {
      "entropy": 5.291862630844117,
      "epoch": 3.1075888568683956,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040489989370401456,
      "loss": 4.5006,
      "mean_token_accuracy": 0.2701272860169411,
      "num_tokens": 74175777.0,
      "step": 32350
    },
    {
      "entropy": 5.381004381179809,
      "epoch": 3.1080691642651295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004048718993683353,
      "loss": 4.6747,
      "mean_token_accuracy": 0.2544348746538162,
      "num_tokens": 74187745.0,
      "step": 32355
    },
    {
      "entropy": 5.380965375900269,
      "epoch": 3.108549471661864,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040484390201741627,
      "loss": 4.679,
      "mean_token_accuracy": 0.24462247043848037,
      "num_tokens": 74198606.0,
      "step": 32360
    },
    {
      "entropy": 5.543749141693115,
      "epoch": 3.1090297790585977,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004048159016519073,
      "loss": 4.7999,
      "mean_token_accuracy": 0.23559768199920655,
      "num_tokens": 74210530.0,
      "step": 32365
    },
    {
      "entropy": 5.420763778686523,
      "epoch": 3.1095100864553316,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004047878982724586,
      "loss": 4.6561,
      "mean_token_accuracy": 0.25371604710817336,
      "num_tokens": 74222155.0,
      "step": 32370
    },
    {
      "entropy": 5.556166505813598,
      "epoch": 3.1099903938520654,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040475989187972034,
      "loss": 4.8194,
      "mean_token_accuracy": 0.24210400730371476,
      "num_tokens": 74233003.0,
      "step": 32375
    },
    {
      "entropy": 5.415853071212768,
      "epoch": 3.1104707012487993,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040473188247434265,
      "loss": 4.6244,
      "mean_token_accuracy": 0.2580511674284935,
      "num_tokens": 74244013.0,
      "step": 32380
    },
    {
      "entropy": 5.425964736938477,
      "epoch": 3.110951008645533,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040470387005697587,
      "loss": 4.7017,
      "mean_token_accuracy": 0.2497740939259529,
      "num_tokens": 74256010.0,
      "step": 32385
    },
    {
      "entropy": 5.532007026672363,
      "epoch": 3.111431316042267,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004046758546282704,
      "loss": 4.7516,
      "mean_token_accuracy": 0.24304653853178024,
      "num_tokens": 74268680.0,
      "step": 32390
    },
    {
      "entropy": 5.523661470413208,
      "epoch": 3.111911623439001,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004046478361888766,
      "loss": 4.8127,
      "mean_token_accuracy": 0.24179320335388182,
      "num_tokens": 74280807.0,
      "step": 32395
    },
    {
      "entropy": 5.404564094543457,
      "epoch": 3.112391930835735,
      "grad_norm": 1.15625,
      "learning_rate": 0.000404619814739445,
      "loss": 4.7394,
      "mean_token_accuracy": 0.24654528498649597,
      "num_tokens": 74292076.0,
      "step": 32400
    },
    {
      "entropy": 5.452355241775512,
      "epoch": 3.1128722382324687,
      "grad_norm": 1.125,
      "learning_rate": 0.0004045917902806263,
      "loss": 4.73,
      "mean_token_accuracy": 0.2507376194000244,
      "num_tokens": 74303965.0,
      "step": 32405
    },
    {
      "entropy": 5.358909988403321,
      "epoch": 3.1133525456292026,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000404563762813071,
      "loss": 4.6054,
      "mean_token_accuracy": 0.259626230597496,
      "num_tokens": 74315774.0,
      "step": 32410
    },
    {
      "entropy": 5.370004463195801,
      "epoch": 3.1138328530259365,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004045357323374298,
      "loss": 4.6654,
      "mean_token_accuracy": 0.2550672650337219,
      "num_tokens": 74326624.0,
      "step": 32415
    },
    {
      "entropy": 5.408546209335327,
      "epoch": 3.1143131604226704,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040450769885435364,
      "loss": 4.6978,
      "mean_token_accuracy": 0.2477891758084297,
      "num_tokens": 74337595.0,
      "step": 32420
    },
    {
      "entropy": 5.457337951660156,
      "epoch": 3.1147934678194042,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040447966236449313,
      "loss": 4.7,
      "mean_token_accuracy": 0.2532836630940437,
      "num_tokens": 74347695.0,
      "step": 32425
    },
    {
      "entropy": 5.365638780593872,
      "epoch": 3.115273775216138,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040445162286849935,
      "loss": 4.6028,
      "mean_token_accuracy": 0.2596740037202835,
      "num_tokens": 74358491.0,
      "step": 32430
    },
    {
      "entropy": 5.463245153427124,
      "epoch": 3.115754082612872,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040442358036702343,
      "loss": 4.766,
      "mean_token_accuracy": 0.24098825454711914,
      "num_tokens": 74371038.0,
      "step": 32435
    },
    {
      "entropy": 5.443546104431152,
      "epoch": 3.1162343900096063,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004043955348607161,
      "loss": 4.7234,
      "mean_token_accuracy": 0.24897630661725997,
      "num_tokens": 74381449.0,
      "step": 32440
    },
    {
      "entropy": 5.420851516723633,
      "epoch": 3.11671469740634,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004043674863502288,
      "loss": 4.6709,
      "mean_token_accuracy": 0.25514269173145293,
      "num_tokens": 74392970.0,
      "step": 32445
    },
    {
      "entropy": 5.365319490432739,
      "epoch": 3.117195004803074,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040433943483621253,
      "loss": 4.6031,
      "mean_token_accuracy": 0.25846096575260163,
      "num_tokens": 74404243.0,
      "step": 32450
    },
    {
      "entropy": 5.483207893371582,
      "epoch": 3.117675312199808,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004043113803193187,
      "loss": 4.7889,
      "mean_token_accuracy": 0.24763226509094238,
      "num_tokens": 74415546.0,
      "step": 32455
    },
    {
      "entropy": 5.3800084590911865,
      "epoch": 3.118155619596542,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040428332280019864,
      "loss": 4.6884,
      "mean_token_accuracy": 0.2540818199515343,
      "num_tokens": 74427138.0,
      "step": 32460
    },
    {
      "entropy": 5.377131557464599,
      "epoch": 3.1186359269932757,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004042552622795036,
      "loss": 4.6122,
      "mean_token_accuracy": 0.2500847652554512,
      "num_tokens": 74438088.0,
      "step": 32465
    },
    {
      "entropy": 5.428716659545898,
      "epoch": 3.1191162343900096,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004042271987578852,
      "loss": 4.747,
      "mean_token_accuracy": 0.24965788573026657,
      "num_tokens": 74450310.0,
      "step": 32470
    },
    {
      "entropy": 5.436430406570435,
      "epoch": 3.1195965417867435,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040419913223599505,
      "loss": 4.6125,
      "mean_token_accuracy": 0.25162733644247054,
      "num_tokens": 74461128.0,
      "step": 32475
    },
    {
      "entropy": 5.4371236801147464,
      "epoch": 3.1200768491834774,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040417106271448464,
      "loss": 4.7234,
      "mean_token_accuracy": 0.24975120574235915,
      "num_tokens": 74472459.0,
      "step": 32480
    },
    {
      "entropy": 5.340144205093384,
      "epoch": 3.1205571565802113,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004041429901940057,
      "loss": 4.6343,
      "mean_token_accuracy": 0.2554701313376427,
      "num_tokens": 74485257.0,
      "step": 32485
    },
    {
      "entropy": 5.3753440380096436,
      "epoch": 3.121037463976945,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040411491467521,
      "loss": 4.6337,
      "mean_token_accuracy": 0.25329829156398775,
      "num_tokens": 74496277.0,
      "step": 32490
    },
    {
      "entropy": 5.444793748855591,
      "epoch": 3.121517771373679,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004040868361587494,
      "loss": 4.7206,
      "mean_token_accuracy": 0.24398766905069352,
      "num_tokens": 74508083.0,
      "step": 32495
    },
    {
      "entropy": 5.474699592590332,
      "epoch": 3.121998078770413,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004040587546452758,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2341634064912796,
      "num_tokens": 74520403.0,
      "step": 32500
    },
    {
      "entropy": 5.433988904953003,
      "epoch": 3.122478386167147,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040403067013544116,
      "loss": 4.7069,
      "mean_token_accuracy": 0.24594936072826384,
      "num_tokens": 74531657.0,
      "step": 32505
    },
    {
      "entropy": 5.472259616851806,
      "epoch": 3.1229586935638807,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040400258262989744,
      "loss": 4.6868,
      "mean_token_accuracy": 0.25376867353916166,
      "num_tokens": 74541841.0,
      "step": 32510
    },
    {
      "entropy": 5.491916275024414,
      "epoch": 3.123439000960615,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040397449212929676,
      "loss": 4.7879,
      "mean_token_accuracy": 0.24109724462032317,
      "num_tokens": 74553312.0,
      "step": 32515
    },
    {
      "entropy": 5.4748311042785645,
      "epoch": 3.123919308357349,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004039463986342914,
      "loss": 4.732,
      "mean_token_accuracy": 0.2503294199705124,
      "num_tokens": 74564470.0,
      "step": 32520
    },
    {
      "entropy": 5.4497767925262455,
      "epoch": 3.1243996157540828,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040391830214553365,
      "loss": 4.7498,
      "mean_token_accuracy": 0.2464900016784668,
      "num_tokens": 74575548.0,
      "step": 32525
    },
    {
      "entropy": 5.414874601364136,
      "epoch": 3.1248799231508166,
      "grad_norm": 1.125,
      "learning_rate": 0.0004038902026636756,
      "loss": 4.7179,
      "mean_token_accuracy": 0.25215829312801363,
      "num_tokens": 74587470.0,
      "step": 32530
    },
    {
      "entropy": 5.4513537883758545,
      "epoch": 3.1253602305475505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004038621001893698,
      "loss": 4.6529,
      "mean_token_accuracy": 0.2581662848591805,
      "num_tokens": 74598743.0,
      "step": 32535
    },
    {
      "entropy": 5.44598388671875,
      "epoch": 3.1258405379442844,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040383399472326874,
      "loss": 4.6885,
      "mean_token_accuracy": 0.2487691283226013,
      "num_tokens": 74609057.0,
      "step": 32540
    },
    {
      "entropy": 5.505279111862182,
      "epoch": 3.1263208453410183,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040380588626602484,
      "loss": 4.7775,
      "mean_token_accuracy": 0.24384137541055678,
      "num_tokens": 74620938.0,
      "step": 32545
    },
    {
      "entropy": 5.4435227394104,
      "epoch": 3.126801152737752,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004037777748182907,
      "loss": 4.6584,
      "mean_token_accuracy": 0.25599451661109923,
      "num_tokens": 74631877.0,
      "step": 32550
    },
    {
      "entropy": 5.437601804733276,
      "epoch": 3.127281460134486,
      "grad_norm": 1.125,
      "learning_rate": 0.0004037496603807191,
      "loss": 4.7436,
      "mean_token_accuracy": 0.24635605961084367,
      "num_tokens": 74641921.0,
      "step": 32555
    },
    {
      "entropy": 5.38299298286438,
      "epoch": 3.12776176753122,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004037215429539626,
      "loss": 4.6779,
      "mean_token_accuracy": 0.252654929459095,
      "num_tokens": 74652950.0,
      "step": 32560
    },
    {
      "entropy": 5.4296728610992435,
      "epoch": 3.128242074927954,
      "grad_norm": 1.125,
      "learning_rate": 0.00040369342253867413,
      "loss": 4.6846,
      "mean_token_accuracy": 0.2565008491277695,
      "num_tokens": 74665512.0,
      "step": 32565
    },
    {
      "entropy": 5.475854539871216,
      "epoch": 3.1287223823246877,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004036652991355066,
      "loss": 4.7092,
      "mean_token_accuracy": 0.25639738142490387,
      "num_tokens": 74675833.0,
      "step": 32570
    },
    {
      "entropy": 5.299305009841919,
      "epoch": 3.1292026897214216,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004036371727451128,
      "loss": 4.6095,
      "mean_token_accuracy": 0.2562756299972534,
      "num_tokens": 74688356.0,
      "step": 32575
    },
    {
      "entropy": 5.3467484474182125,
      "epoch": 3.1296829971181555,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040360904336814586,
      "loss": 4.663,
      "mean_token_accuracy": 0.25575137734413145,
      "num_tokens": 74699617.0,
      "step": 32580
    },
    {
      "entropy": 5.336010885238648,
      "epoch": 3.1301633045148893,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004035809110052588,
      "loss": 4.6076,
      "mean_token_accuracy": 0.2560220927000046,
      "num_tokens": 74711319.0,
      "step": 32585
    },
    {
      "entropy": 5.347618150711059,
      "epoch": 3.1306436119116237,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004035527756571048,
      "loss": 4.6522,
      "mean_token_accuracy": 0.25357957035303114,
      "num_tokens": 74723378.0,
      "step": 32590
    },
    {
      "entropy": 5.4204216480255125,
      "epoch": 3.1311239193083575,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040352463732433707,
      "loss": 4.7086,
      "mean_token_accuracy": 0.2479709044098854,
      "num_tokens": 74734011.0,
      "step": 32595
    },
    {
      "entropy": 5.393408346176147,
      "epoch": 3.1316042267050914,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040349649600760894,
      "loss": 4.6781,
      "mean_token_accuracy": 0.2584321931004524,
      "num_tokens": 74745534.0,
      "step": 32600
    },
    {
      "entropy": 5.440715169906616,
      "epoch": 3.1320845341018253,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004034683517075737,
      "loss": 4.7133,
      "mean_token_accuracy": 0.2534827947616577,
      "num_tokens": 74757288.0,
      "step": 32605
    },
    {
      "entropy": 5.391999578475952,
      "epoch": 3.132564841498559,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040344020442488476,
      "loss": 4.6905,
      "mean_token_accuracy": 0.248069129884243,
      "num_tokens": 74770520.0,
      "step": 32610
    },
    {
      "entropy": 5.34910249710083,
      "epoch": 3.133045148895293,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040341205416019577,
      "loss": 4.6641,
      "mean_token_accuracy": 0.2557758703827858,
      "num_tokens": 74781967.0,
      "step": 32615
    },
    {
      "entropy": 5.422588777542114,
      "epoch": 3.133525456292027,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004033839009141601,
      "loss": 4.6981,
      "mean_token_accuracy": 0.2464147463440895,
      "num_tokens": 74793022.0,
      "step": 32620
    },
    {
      "entropy": 5.449282550811768,
      "epoch": 3.134005763688761,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040335574468743145,
      "loss": 4.6779,
      "mean_token_accuracy": 0.2538814529776573,
      "num_tokens": 74805165.0,
      "step": 32625
    },
    {
      "entropy": 5.397122383117676,
      "epoch": 3.1344860710854947,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004033275854806636,
      "loss": 4.7368,
      "mean_token_accuracy": 0.25267460495233535,
      "num_tokens": 74816607.0,
      "step": 32630
    },
    {
      "entropy": 5.398761367797851,
      "epoch": 3.1349663784822286,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004032994232945103,
      "loss": 4.7247,
      "mean_token_accuracy": 0.2474468782544136,
      "num_tokens": 74827606.0,
      "step": 32635
    },
    {
      "entropy": 5.419905281066894,
      "epoch": 3.1354466858789625,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004032712581296253,
      "loss": 4.648,
      "mean_token_accuracy": 0.2528204545378685,
      "num_tokens": 74837507.0,
      "step": 32640
    },
    {
      "entropy": 5.423628854751587,
      "epoch": 3.1359269932756964,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040324308998666267,
      "loss": 4.7383,
      "mean_token_accuracy": 0.24390976130962372,
      "num_tokens": 74848164.0,
      "step": 32645
    },
    {
      "entropy": 5.414722871780396,
      "epoch": 3.1364073006724302,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040321491886627614,
      "loss": 4.6879,
      "mean_token_accuracy": 0.2474554196000099,
      "num_tokens": 74858010.0,
      "step": 32650
    },
    {
      "entropy": 5.414615440368652,
      "epoch": 3.136887608069164,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040318674476912006,
      "loss": 4.6417,
      "mean_token_accuracy": 0.25117197185754775,
      "num_tokens": 74868654.0,
      "step": 32655
    },
    {
      "entropy": 5.475795030593872,
      "epoch": 3.137367915465898,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004031585676958483,
      "loss": 4.7918,
      "mean_token_accuracy": 0.24407100677490234,
      "num_tokens": 74879751.0,
      "step": 32660
    },
    {
      "entropy": 5.465546321868897,
      "epoch": 3.1378482228626323,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040313038764711517,
      "loss": 4.7051,
      "mean_token_accuracy": 0.2486040085554123,
      "num_tokens": 74890569.0,
      "step": 32665
    },
    {
      "entropy": 5.526501417160034,
      "epoch": 3.138328530259366,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040310220462357494,
      "loss": 4.7484,
      "mean_token_accuracy": 0.23897335678339005,
      "num_tokens": 74902078.0,
      "step": 32670
    },
    {
      "entropy": 5.4246588230133055,
      "epoch": 3.1388088376561,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004030740186258819,
      "loss": 4.6923,
      "mean_token_accuracy": 0.2509327709674835,
      "num_tokens": 74913228.0,
      "step": 32675
    },
    {
      "entropy": 5.384682607650757,
      "epoch": 3.139289145052834,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004030458296546905,
      "loss": 4.7005,
      "mean_token_accuracy": 0.2564899265766144,
      "num_tokens": 74925419.0,
      "step": 32680
    },
    {
      "entropy": 5.451738166809082,
      "epoch": 3.139769452449568,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040301763771065504,
      "loss": 4.7125,
      "mean_token_accuracy": 0.2460775688290596,
      "num_tokens": 74936219.0,
      "step": 32685
    },
    {
      "entropy": 5.410635042190552,
      "epoch": 3.1402497598463017,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004029894427944302,
      "loss": 4.6978,
      "mean_token_accuracy": 0.25235638171434405,
      "num_tokens": 74948015.0,
      "step": 32690
    },
    {
      "entropy": 5.403562736511231,
      "epoch": 3.1407300672430356,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040296124490667065,
      "loss": 4.6822,
      "mean_token_accuracy": 0.2498397335410118,
      "num_tokens": 74959753.0,
      "step": 32695
    },
    {
      "entropy": 5.389874267578125,
      "epoch": 3.1412103746397695,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004029330440480308,
      "loss": 4.6357,
      "mean_token_accuracy": 0.2565514504909515,
      "num_tokens": 74970696.0,
      "step": 32700
    },
    {
      "entropy": 5.473667621612549,
      "epoch": 3.1416906820365034,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004029048402191656,
      "loss": 4.7431,
      "mean_token_accuracy": 0.24811418503522872,
      "num_tokens": 74982070.0,
      "step": 32705
    },
    {
      "entropy": 5.4255297660827635,
      "epoch": 3.1421709894332372,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004028766334207299,
      "loss": 4.6979,
      "mean_token_accuracy": 0.2573621287941933,
      "num_tokens": 74992823.0,
      "step": 32710
    },
    {
      "entropy": 5.295782518386841,
      "epoch": 3.142651296829971,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004028484236533784,
      "loss": 4.588,
      "mean_token_accuracy": 0.25656102448701856,
      "num_tokens": 75004456.0,
      "step": 32715
    },
    {
      "entropy": 5.4826456069946286,
      "epoch": 3.143131604226705,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040282021091776624,
      "loss": 4.8063,
      "mean_token_accuracy": 0.23789096772670745,
      "num_tokens": 75017039.0,
      "step": 32720
    },
    {
      "entropy": 5.450796508789063,
      "epoch": 3.143611911623439,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004027919952145482,
      "loss": 4.6668,
      "mean_token_accuracy": 0.2532722741365433,
      "num_tokens": 75028155.0,
      "step": 32725
    },
    {
      "entropy": 5.418787145614624,
      "epoch": 3.1440922190201728,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004027637765443795,
      "loss": 4.6931,
      "mean_token_accuracy": 0.2501083791255951,
      "num_tokens": 75039529.0,
      "step": 32730
    },
    {
      "entropy": 5.458703422546387,
      "epoch": 3.1445725264169067,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040273555490791534,
      "loss": 4.7195,
      "mean_token_accuracy": 0.24809072762727738,
      "num_tokens": 75051205.0,
      "step": 32735
    },
    {
      "entropy": 5.4633745670318605,
      "epoch": 3.1450528338136405,
      "grad_norm": 1.125,
      "learning_rate": 0.0004027073303058109,
      "loss": 4.7279,
      "mean_token_accuracy": 0.2539056032896042,
      "num_tokens": 75063379.0,
      "step": 32740
    },
    {
      "entropy": 5.404110288619995,
      "epoch": 3.1455331412103744,
      "grad_norm": 1.125,
      "learning_rate": 0.0004026791027387214,
      "loss": 4.7083,
      "mean_token_accuracy": 0.24633872807025908,
      "num_tokens": 75074034.0,
      "step": 32745
    },
    {
      "entropy": 5.433588981628418,
      "epoch": 3.1460134486071087,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004026508722073024,
      "loss": 4.721,
      "mean_token_accuracy": 0.2505216091871262,
      "num_tokens": 75087004.0,
      "step": 32750
    },
    {
      "entropy": 5.479429149627686,
      "epoch": 3.1464937560038426,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040262263871220904,
      "loss": 4.7533,
      "mean_token_accuracy": 0.24750628620386123,
      "num_tokens": 75100270.0,
      "step": 32755
    },
    {
      "entropy": 5.414999055862427,
      "epoch": 3.1469740634005765,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004025944022540971,
      "loss": 4.6822,
      "mean_token_accuracy": 0.2548674926161766,
      "num_tokens": 75113116.0,
      "step": 32760
    },
    {
      "entropy": 5.4617725849151615,
      "epoch": 3.1474543707973104,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040256616283362195,
      "loss": 4.7761,
      "mean_token_accuracy": 0.2421977326273918,
      "num_tokens": 75124225.0,
      "step": 32765
    },
    {
      "entropy": 5.393592119216919,
      "epoch": 3.1479346781940443,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040253792045143926,
      "loss": 4.6275,
      "mean_token_accuracy": 0.2613397866487503,
      "num_tokens": 75135859.0,
      "step": 32770
    },
    {
      "entropy": 5.542902708053589,
      "epoch": 3.148414985590778,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004025096751082048,
      "loss": 4.8256,
      "mean_token_accuracy": 0.2458980530500412,
      "num_tokens": 75148626.0,
      "step": 32775
    },
    {
      "entropy": 5.454569482803345,
      "epoch": 3.148895292987512,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004024814268045743,
      "loss": 4.7167,
      "mean_token_accuracy": 0.24564780443906784,
      "num_tokens": 75159539.0,
      "step": 32780
    },
    {
      "entropy": 5.390147495269775,
      "epoch": 3.149375600384246,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040245317554120363,
      "loss": 4.7115,
      "mean_token_accuracy": 0.2507906839251518,
      "num_tokens": 75172393.0,
      "step": 32785
    },
    {
      "entropy": 5.465893793106079,
      "epoch": 3.14985590778098,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004024249213187487,
      "loss": 4.7388,
      "mean_token_accuracy": 0.24862713664770125,
      "num_tokens": 75183777.0,
      "step": 32790
    },
    {
      "entropy": 5.387915229797363,
      "epoch": 3.1503362151777137,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004023966641378655,
      "loss": 4.6357,
      "mean_token_accuracy": 0.2561225160956383,
      "num_tokens": 75194976.0,
      "step": 32795
    },
    {
      "entropy": 5.372679615020752,
      "epoch": 3.1508165225744476,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040236840399920996,
      "loss": 4.6595,
      "mean_token_accuracy": 0.24864391684532167,
      "num_tokens": 75207709.0,
      "step": 32800
    },
    {
      "entropy": 5.352499055862427,
      "epoch": 3.1512968299711814,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040234014090343833,
      "loss": 4.6155,
      "mean_token_accuracy": 0.26191352903842924,
      "num_tokens": 75217848.0,
      "step": 32805
    },
    {
      "entropy": 5.465042448043823,
      "epoch": 3.1517771373679153,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004023118748512068,
      "loss": 4.7478,
      "mean_token_accuracy": 0.25220310389995576,
      "num_tokens": 75229670.0,
      "step": 32810
    },
    {
      "entropy": 5.363207054138184,
      "epoch": 3.152257444764649,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004022836058431715,
      "loss": 4.6464,
      "mean_token_accuracy": 0.25446137338876723,
      "num_tokens": 75241752.0,
      "step": 32815
    },
    {
      "entropy": 5.40836353302002,
      "epoch": 3.152737752161383,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00040225533387998883,
      "loss": 4.6488,
      "mean_token_accuracy": 0.25337919443845747,
      "num_tokens": 75252818.0,
      "step": 32820
    },
    {
      "entropy": 5.438999795913697,
      "epoch": 3.1532180595581174,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004022270589623152,
      "loss": 4.6799,
      "mean_token_accuracy": 0.24939750432968139,
      "num_tokens": 75264076.0,
      "step": 32825
    },
    {
      "entropy": 5.445992660522461,
      "epoch": 3.1536983669548513,
      "grad_norm": 1.171875,
      "learning_rate": 0.000402198781090807,
      "loss": 4.7241,
      "mean_token_accuracy": 0.2463959902524948,
      "num_tokens": 75275281.0,
      "step": 32830
    },
    {
      "entropy": 5.376329708099365,
      "epoch": 3.154178674351585,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004021705002661208,
      "loss": 4.6842,
      "mean_token_accuracy": 0.24443005621433259,
      "num_tokens": 75287443.0,
      "step": 32835
    },
    {
      "entropy": 5.402212429046631,
      "epoch": 3.154658981748319,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004021422164889133,
      "loss": 4.6931,
      "mean_token_accuracy": 0.2449648454785347,
      "num_tokens": 75299256.0,
      "step": 32840
    },
    {
      "entropy": 5.451338624954223,
      "epoch": 3.155139289145053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000402113929759841,
      "loss": 4.7304,
      "mean_token_accuracy": 0.24946227371692659,
      "num_tokens": 75310485.0,
      "step": 32845
    },
    {
      "entropy": 5.490271663665771,
      "epoch": 3.155619596541787,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040208564007956075,
      "loss": 4.7046,
      "mean_token_accuracy": 0.2513514205813408,
      "num_tokens": 75322826.0,
      "step": 32850
    },
    {
      "entropy": 5.429955768585205,
      "epoch": 3.1560999039385207,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004020573474487293,
      "loss": 4.6721,
      "mean_token_accuracy": 0.2548229739069939,
      "num_tokens": 75334826.0,
      "step": 32855
    },
    {
      "entropy": 5.333129596710205,
      "epoch": 3.1565802113352546,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040202905186800347,
      "loss": 4.6452,
      "mean_token_accuracy": 0.25517977923154833,
      "num_tokens": 75345997.0,
      "step": 32860
    },
    {
      "entropy": 5.476942920684815,
      "epoch": 3.1570605187319885,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004020007533380403,
      "loss": 4.7203,
      "mean_token_accuracy": 0.239910326898098,
      "num_tokens": 75357018.0,
      "step": 32865
    },
    {
      "entropy": 5.420318365097046,
      "epoch": 3.1575408261287223,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004019724518594967,
      "loss": 4.6865,
      "mean_token_accuracy": 0.2481824904680252,
      "num_tokens": 75368190.0,
      "step": 32870
    },
    {
      "entropy": 5.429897212982178,
      "epoch": 3.158021133525456,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004019441474330298,
      "loss": 4.6439,
      "mean_token_accuracy": 0.25252472013235094,
      "num_tokens": 75380284.0,
      "step": 32875
    },
    {
      "entropy": 5.373954916000367,
      "epoch": 3.15850144092219,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040191584005929684,
      "loss": 4.6657,
      "mean_token_accuracy": 0.25456233471632006,
      "num_tokens": 75392543.0,
      "step": 32880
    },
    {
      "entropy": 5.395052099227906,
      "epoch": 3.158981748318924,
      "grad_norm": 1.125,
      "learning_rate": 0.0004018875297389549,
      "loss": 4.6693,
      "mean_token_accuracy": 0.2491029605269432,
      "num_tokens": 75404883.0,
      "step": 32885
    },
    {
      "entropy": 5.429851961135864,
      "epoch": 3.159462055715658,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040185921647266126,
      "loss": 4.7066,
      "mean_token_accuracy": 0.25163571536540985,
      "num_tokens": 75416717.0,
      "step": 32890
    },
    {
      "entropy": 5.439846038818359,
      "epoch": 3.1599423631123917,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040183090026107326,
      "loss": 4.6558,
      "mean_token_accuracy": 0.25849063992500304,
      "num_tokens": 75428103.0,
      "step": 32895
    },
    {
      "entropy": 5.404587364196777,
      "epoch": 3.160422670509126,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040180258110484847,
      "loss": 4.6275,
      "mean_token_accuracy": 0.25965452939271927,
      "num_tokens": 75438333.0,
      "step": 32900
    },
    {
      "entropy": 5.441953325271607,
      "epoch": 3.16090297790586,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004017742590046442,
      "loss": 4.7081,
      "mean_token_accuracy": 0.2537092223763466,
      "num_tokens": 75450023.0,
      "step": 32905
    },
    {
      "entropy": 5.41506781578064,
      "epoch": 3.161383285302594,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040174593396111814,
      "loss": 4.6919,
      "mean_token_accuracy": 0.2511137962341309,
      "num_tokens": 75460724.0,
      "step": 32910
    },
    {
      "entropy": 5.440560436248779,
      "epoch": 3.1618635926993277,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040171760597492785,
      "loss": 4.7072,
      "mean_token_accuracy": 0.24921204149723053,
      "num_tokens": 75471752.0,
      "step": 32915
    },
    {
      "entropy": 5.369786739349365,
      "epoch": 3.1623439000960616,
      "grad_norm": 1.125,
      "learning_rate": 0.00040168927504673094,
      "loss": 4.6315,
      "mean_token_accuracy": 0.25100278556346894,
      "num_tokens": 75483748.0,
      "step": 32920
    },
    {
      "entropy": 5.470439767837524,
      "epoch": 3.1628242074927955,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004016609411771853,
      "loss": 4.7848,
      "mean_token_accuracy": 0.24488767236471176,
      "num_tokens": 75495497.0,
      "step": 32925
    },
    {
      "entropy": 5.46571364402771,
      "epoch": 3.1633045148895294,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040163260436694876,
      "loss": 4.7429,
      "mean_token_accuracy": 0.24293323010206222,
      "num_tokens": 75506852.0,
      "step": 32930
    },
    {
      "entropy": 5.440067720413208,
      "epoch": 3.1637848222862632,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004016042646166791,
      "loss": 4.6948,
      "mean_token_accuracy": 0.2569209381937981,
      "num_tokens": 75518633.0,
      "step": 32935
    },
    {
      "entropy": 5.388991403579712,
      "epoch": 3.164265129682997,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004015759219270344,
      "loss": 4.665,
      "mean_token_accuracy": 0.24998752623796464,
      "num_tokens": 75529668.0,
      "step": 32940
    },
    {
      "entropy": 5.422969818115234,
      "epoch": 3.164745437079731,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004015475762986726,
      "loss": 4.6879,
      "mean_token_accuracy": 0.2593930095434189,
      "num_tokens": 75541603.0,
      "step": 32945
    },
    {
      "entropy": 5.450471353530884,
      "epoch": 3.165225744476465,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040151922773225187,
      "loss": 4.7372,
      "mean_token_accuracy": 0.2515485122799873,
      "num_tokens": 75553489.0,
      "step": 32950
    },
    {
      "entropy": 5.385764837265015,
      "epoch": 3.1657060518731988,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004014908762284303,
      "loss": 4.6851,
      "mean_token_accuracy": 0.25144273191690447,
      "num_tokens": 75564809.0,
      "step": 32955
    },
    {
      "entropy": 5.454453945159912,
      "epoch": 3.1661863592699326,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040146252178786633,
      "loss": 4.7113,
      "mean_token_accuracy": 0.25179846584796906,
      "num_tokens": 75575886.0,
      "step": 32960
    },
    {
      "entropy": 5.429741430282593,
      "epoch": 3.1666666666666665,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000401434164411218,
      "loss": 4.7111,
      "mean_token_accuracy": 0.24498990774154664,
      "num_tokens": 75586518.0,
      "step": 32965
    },
    {
      "entropy": 5.336591386795044,
      "epoch": 3.1671469740634004,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040140580409914385,
      "loss": 4.5601,
      "mean_token_accuracy": 0.260009703040123,
      "num_tokens": 75598819.0,
      "step": 32970
    },
    {
      "entropy": 5.413063812255859,
      "epoch": 3.1676272814601343,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040137744085230227,
      "loss": 4.7398,
      "mean_token_accuracy": 0.2553927004337311,
      "num_tokens": 75610319.0,
      "step": 32975
    },
    {
      "entropy": 5.577210092544556,
      "epoch": 3.168107588856868,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004013490746713518,
      "loss": 4.7919,
      "mean_token_accuracy": 0.24699200689792633,
      "num_tokens": 75622479.0,
      "step": 32980
    },
    {
      "entropy": 5.427000904083252,
      "epoch": 3.1685878962536025,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040132070555695096,
      "loss": 4.6973,
      "mean_token_accuracy": 0.24908494502305983,
      "num_tokens": 75634114.0,
      "step": 32985
    },
    {
      "entropy": 5.427808952331543,
      "epoch": 3.1690682036503364,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040129233350975847,
      "loss": 4.8037,
      "mean_token_accuracy": 0.24554310292005538,
      "num_tokens": 75645828.0,
      "step": 32990
    },
    {
      "entropy": 5.428324604034424,
      "epoch": 3.1695485110470702,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040126395853043293,
      "loss": 4.7197,
      "mean_token_accuracy": 0.2504788234829903,
      "num_tokens": 75657335.0,
      "step": 32995
    },
    {
      "entropy": 5.44268569946289,
      "epoch": 3.170028818443804,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004012355806196332,
      "loss": 4.6583,
      "mean_token_accuracy": 0.24811361730098724,
      "num_tokens": 75668886.0,
      "step": 33000
    },
    {
      "epoch": 3.170028818443804,
      "eval_entropy": 5.181510149822301,
      "eval_loss": 4.8020219802856445,
      "eval_mean_token_accuracy": 0.2542905479093795,
      "eval_num_tokens": 75668886.0,
      "eval_runtime": 20.4573,
      "eval_samples_per_second": 1604.072,
      "eval_steps_per_second": 200.515,
      "step": 33000
    },
    {
      "entropy": 5.508826017379761,
      "epoch": 3.170509125840538,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040120719977801823,
      "loss": 4.774,
      "mean_token_accuracy": 0.2481457233428955,
      "num_tokens": 75679392.0,
      "step": 33005
    },
    {
      "entropy": 5.521950149536133,
      "epoch": 3.170989433237272,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040117881600624676,
      "loss": 4.7829,
      "mean_token_accuracy": 0.2465755209326744,
      "num_tokens": 75692028.0,
      "step": 33010
    },
    {
      "entropy": 5.377668714523315,
      "epoch": 3.1714697406340058,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040115042930497787,
      "loss": 4.6958,
      "mean_token_accuracy": 0.24940951019525529,
      "num_tokens": 75702879.0,
      "step": 33015
    },
    {
      "entropy": 5.420516872406006,
      "epoch": 3.1719500480307397,
      "grad_norm": 1.25,
      "learning_rate": 0.00040112203967487066,
      "loss": 4.674,
      "mean_token_accuracy": 0.24765269905328752,
      "num_tokens": 75713564.0,
      "step": 33020
    },
    {
      "entropy": 5.375752544403076,
      "epoch": 3.1724303554274735,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040109364711658416,
      "loss": 4.6508,
      "mean_token_accuracy": 0.2527853071689606,
      "num_tokens": 75724577.0,
      "step": 33025
    },
    {
      "entropy": 5.463782835006714,
      "epoch": 3.1729106628242074,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040106525163077756,
      "loss": 4.7101,
      "mean_token_accuracy": 0.2557879686355591,
      "num_tokens": 75736349.0,
      "step": 33030
    },
    {
      "entropy": 5.402255582809448,
      "epoch": 3.1733909702209413,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004010368532181102,
      "loss": 4.7241,
      "mean_token_accuracy": 0.25012126863002776,
      "num_tokens": 75747465.0,
      "step": 33035
    },
    {
      "entropy": 5.533922863006592,
      "epoch": 3.173871277617675,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004010084518792413,
      "loss": 4.8254,
      "mean_token_accuracy": 0.24469562768936157,
      "num_tokens": 75759014.0,
      "step": 33040
    },
    {
      "entropy": 5.413418865203857,
      "epoch": 3.174351585014409,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040098004761483037,
      "loss": 4.6588,
      "mean_token_accuracy": 0.24878330230712892,
      "num_tokens": 75770354.0,
      "step": 33045
    },
    {
      "entropy": 5.494133186340332,
      "epoch": 3.174831892411143,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004009516404255368,
      "loss": 4.7841,
      "mean_token_accuracy": 0.24581747949123384,
      "num_tokens": 75781530.0,
      "step": 33050
    },
    {
      "entropy": 5.412806797027588,
      "epoch": 3.175312199807877,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004009232303120202,
      "loss": 4.7037,
      "mean_token_accuracy": 0.24938159734010695,
      "num_tokens": 75792751.0,
      "step": 33055
    },
    {
      "entropy": 5.388373374938965,
      "epoch": 3.175792507204611,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040089481727494,
      "loss": 4.6168,
      "mean_token_accuracy": 0.2589566558599472,
      "num_tokens": 75804882.0,
      "step": 33060
    },
    {
      "entropy": 5.35329155921936,
      "epoch": 3.176272814601345,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000400866401314956,
      "loss": 4.6593,
      "mean_token_accuracy": 0.251689475774765,
      "num_tokens": 75816207.0,
      "step": 33065
    },
    {
      "entropy": 5.44726939201355,
      "epoch": 3.176753121998079,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040083798243272797,
      "loss": 4.8185,
      "mean_token_accuracy": 0.25065303295850755,
      "num_tokens": 75827957.0,
      "step": 33070
    },
    {
      "entropy": 5.500483751296997,
      "epoch": 3.177233429394813,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040080956062891554,
      "loss": 4.7362,
      "mean_token_accuracy": 0.2500710278749466,
      "num_tokens": 75839227.0,
      "step": 33075
    },
    {
      "entropy": 5.475994205474853,
      "epoch": 3.1777137367915467,
      "grad_norm": 1.125,
      "learning_rate": 0.00040078113590417887,
      "loss": 4.7307,
      "mean_token_accuracy": 0.24923896044492722,
      "num_tokens": 75849831.0,
      "step": 33080
    },
    {
      "entropy": 5.343835496902466,
      "epoch": 3.1781940441882806,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040075270825917753,
      "loss": 4.685,
      "mean_token_accuracy": 0.25396904200315473,
      "num_tokens": 75860617.0,
      "step": 33085
    },
    {
      "entropy": 5.456933259963989,
      "epoch": 3.1786743515850144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004007242776945718,
      "loss": 4.7574,
      "mean_token_accuracy": 0.24276282340288163,
      "num_tokens": 75872193.0,
      "step": 33090
    },
    {
      "entropy": 5.443118095397949,
      "epoch": 3.1791546589817483,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040069584421102174,
      "loss": 4.7339,
      "mean_token_accuracy": 0.25449587404727936,
      "num_tokens": 75885536.0,
      "step": 33095
    },
    {
      "entropy": 5.518156909942627,
      "epoch": 3.179634966378482,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040066740780918725,
      "loss": 4.7558,
      "mean_token_accuracy": 0.2464570015668869,
      "num_tokens": 75897325.0,
      "step": 33100
    },
    {
      "entropy": 5.437181568145752,
      "epoch": 3.180115273775216,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004006389684897288,
      "loss": 4.7489,
      "mean_token_accuracy": 0.24540738761425018,
      "num_tokens": 75908353.0,
      "step": 33105
    },
    {
      "entropy": 5.420773458480835,
      "epoch": 3.18059558117195,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004006105262533066,
      "loss": 4.7497,
      "mean_token_accuracy": 0.2462108924984932,
      "num_tokens": 75919862.0,
      "step": 33110
    },
    {
      "entropy": 5.405204629898071,
      "epoch": 3.181075888568684,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004005820811005809,
      "loss": 4.6474,
      "mean_token_accuracy": 0.259388367831707,
      "num_tokens": 75930971.0,
      "step": 33115
    },
    {
      "entropy": 5.450860118865966,
      "epoch": 3.1815561959654177,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040055363303221226,
      "loss": 4.6219,
      "mean_token_accuracy": 0.2640230879187584,
      "num_tokens": 75941124.0,
      "step": 33120
    },
    {
      "entropy": 5.46060528755188,
      "epoch": 3.1820365033621516,
      "grad_norm": 1.25,
      "learning_rate": 0.000400525182048861,
      "loss": 4.7305,
      "mean_token_accuracy": 0.2532235771417618,
      "num_tokens": 75953040.0,
      "step": 33125
    },
    {
      "entropy": 5.4333830833435055,
      "epoch": 3.1825168107588855,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040049672815118775,
      "loss": 4.77,
      "mean_token_accuracy": 0.23977696299552917,
      "num_tokens": 75965122.0,
      "step": 33130
    },
    {
      "entropy": 5.484256553649902,
      "epoch": 3.18299711815562,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040046827133985316,
      "loss": 4.6607,
      "mean_token_accuracy": 0.25312172770500185,
      "num_tokens": 75975360.0,
      "step": 33135
    },
    {
      "entropy": 5.430519533157349,
      "epoch": 3.1834774255523537,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040043981161551784,
      "loss": 4.6301,
      "mean_token_accuracy": 0.25127819180488586,
      "num_tokens": 75985592.0,
      "step": 33140
    },
    {
      "entropy": 5.350561475753784,
      "epoch": 3.1839577329490876,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004004113489788426,
      "loss": 4.6871,
      "mean_token_accuracy": 0.24947884827852249,
      "num_tokens": 75997429.0,
      "step": 33145
    },
    {
      "entropy": 5.499626398086548,
      "epoch": 3.1844380403458215,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040038288343048823,
      "loss": 4.8547,
      "mean_token_accuracy": 0.24001419693231582,
      "num_tokens": 76008785.0,
      "step": 33150
    },
    {
      "entropy": 5.561444473266602,
      "epoch": 3.1849183477425553,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040035441497111564,
      "loss": 4.8191,
      "mean_token_accuracy": 0.24011840522289277,
      "num_tokens": 76020773.0,
      "step": 33155
    },
    {
      "entropy": 5.444511556625367,
      "epoch": 3.185398655139289,
      "grad_norm": 1.125,
      "learning_rate": 0.00040032594360138576,
      "loss": 4.7051,
      "mean_token_accuracy": 0.2541560634970665,
      "num_tokens": 76033300.0,
      "step": 33160
    },
    {
      "entropy": 5.406310510635376,
      "epoch": 3.185878962536023,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004002974693219595,
      "loss": 4.6067,
      "mean_token_accuracy": 0.25166787058115003,
      "num_tokens": 76043443.0,
      "step": 33165
    },
    {
      "entropy": 5.455263566970825,
      "epoch": 3.186359269932757,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040026899213349814,
      "loss": 4.6832,
      "mean_token_accuracy": 0.2528302654623985,
      "num_tokens": 76055417.0,
      "step": 33170
    },
    {
      "entropy": 5.447287464141846,
      "epoch": 3.186839577329491,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004002405120366628,
      "loss": 4.7026,
      "mean_token_accuracy": 0.250362491607666,
      "num_tokens": 76067259.0,
      "step": 33175
    },
    {
      "entropy": 5.453786039352417,
      "epoch": 3.1873198847262247,
      "grad_norm": 1.125,
      "learning_rate": 0.00040021202903211454,
      "loss": 4.6782,
      "mean_token_accuracy": 0.2534155920147896,
      "num_tokens": 76079827.0,
      "step": 33180
    },
    {
      "entropy": 5.437813758850098,
      "epoch": 3.1878001921229586,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004001835431205149,
      "loss": 4.7035,
      "mean_token_accuracy": 0.24580058455467224,
      "num_tokens": 76092068.0,
      "step": 33185
    },
    {
      "entropy": 5.467273235321045,
      "epoch": 3.1882804995196925,
      "grad_norm": 1.421875,
      "learning_rate": 0.00040015505430252506,
      "loss": 4.7375,
      "mean_token_accuracy": 0.24348847270011903,
      "num_tokens": 76102046.0,
      "step": 33190
    },
    {
      "entropy": 5.408592796325683,
      "epoch": 3.1887608069164264,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040012656257880645,
      "loss": 4.6704,
      "mean_token_accuracy": 0.25318306386470796,
      "num_tokens": 76113629.0,
      "step": 33195
    },
    {
      "entropy": 5.383147954940796,
      "epoch": 3.1892411143131603,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00040009806795002076,
      "loss": 4.6259,
      "mean_token_accuracy": 0.26255634129047395,
      "num_tokens": 76124593.0,
      "step": 33200
    },
    {
      "entropy": 5.552494716644287,
      "epoch": 3.189721421709894,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004000695704168292,
      "loss": 4.8813,
      "mean_token_accuracy": 0.23532852083444594,
      "num_tokens": 76135177.0,
      "step": 33205
    },
    {
      "entropy": 5.425709009170532,
      "epoch": 3.1902017291066285,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004000410699798937,
      "loss": 4.6747,
      "mean_token_accuracy": 0.25184424221515656,
      "num_tokens": 76148998.0,
      "step": 33210
    },
    {
      "entropy": 5.3515181064605715,
      "epoch": 3.1906820365033624,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040001256663987585,
      "loss": 4.7093,
      "mean_token_accuracy": 0.25747924000024797,
      "num_tokens": 76161224.0,
      "step": 33215
    },
    {
      "entropy": 5.381874752044678,
      "epoch": 3.1911623439000962,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039998406039743736,
      "loss": 4.704,
      "mean_token_accuracy": 0.252979040145874,
      "num_tokens": 76172577.0,
      "step": 33220
    },
    {
      "entropy": 5.37125883102417,
      "epoch": 3.19164265129683,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003999555512532401,
      "loss": 4.6632,
      "mean_token_accuracy": 0.2553064048290253,
      "num_tokens": 76184135.0,
      "step": 33225
    },
    {
      "entropy": 5.460771369934082,
      "epoch": 3.192122958693564,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003999270392079461,
      "loss": 4.7467,
      "mean_token_accuracy": 0.24991582632064818,
      "num_tokens": 76195831.0,
      "step": 33230
    },
    {
      "entropy": 5.497240877151489,
      "epoch": 3.192603266090298,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003998985242622171,
      "loss": 4.7559,
      "mean_token_accuracy": 0.24508977085351943,
      "num_tokens": 76208914.0,
      "step": 33235
    },
    {
      "entropy": 5.435340452194214,
      "epoch": 3.1930835734870318,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003998700064167153,
      "loss": 4.7572,
      "mean_token_accuracy": 0.24728191196918486,
      "num_tokens": 76220599.0,
      "step": 33240
    },
    {
      "entropy": 5.4143757820129395,
      "epoch": 3.1935638808837656,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003998414856721027,
      "loss": 4.6968,
      "mean_token_accuracy": 0.25106672793626783,
      "num_tokens": 76232897.0,
      "step": 33245
    },
    {
      "entropy": 5.485614061355591,
      "epoch": 3.1940441882804995,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003998129620290415,
      "loss": 4.7414,
      "mean_token_accuracy": 0.2540698364377022,
      "num_tokens": 76245948.0,
      "step": 33250
    },
    {
      "entropy": 5.449227237701416,
      "epoch": 3.1945244956772334,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00039978443548819393,
      "loss": 4.7163,
      "mean_token_accuracy": 0.2501614421606064,
      "num_tokens": 76257223.0,
      "step": 33255
    },
    {
      "entropy": 5.38938341140747,
      "epoch": 3.1950048030739673,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003997559060502224,
      "loss": 4.6358,
      "mean_token_accuracy": 0.25402323603630067,
      "num_tokens": 76267606.0,
      "step": 33260
    },
    {
      "entropy": 5.423636865615845,
      "epoch": 3.195485110470701,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003997273737157891,
      "loss": 4.6907,
      "mean_token_accuracy": 0.25215519219636917,
      "num_tokens": 76280430.0,
      "step": 33265
    },
    {
      "entropy": 5.511788463592529,
      "epoch": 3.195965417867435,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039969883848555647,
      "loss": 4.7513,
      "mean_token_accuracy": 0.24705886244773864,
      "num_tokens": 76293089.0,
      "step": 33270
    },
    {
      "entropy": 5.423926258087159,
      "epoch": 3.196445725264169,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003996703003601872,
      "loss": 4.6915,
      "mean_token_accuracy": 0.24590768069028854,
      "num_tokens": 76304936.0,
      "step": 33275
    },
    {
      "entropy": 5.3774425983428955,
      "epoch": 3.196926032660903,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039964175934034375,
      "loss": 4.5294,
      "mean_token_accuracy": 0.2690952718257904,
      "num_tokens": 76316195.0,
      "step": 33280
    },
    {
      "entropy": 5.404885149002075,
      "epoch": 3.1974063400576367,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003996132154266887,
      "loss": 4.6555,
      "mean_token_accuracy": 0.2540562480688095,
      "num_tokens": 76328128.0,
      "step": 33285
    },
    {
      "entropy": 5.360824251174927,
      "epoch": 3.1978866474543706,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003995846686198849,
      "loss": 4.6447,
      "mean_token_accuracy": 0.25378641933202745,
      "num_tokens": 76338596.0,
      "step": 33290
    },
    {
      "entropy": 5.421521186828613,
      "epoch": 3.198366954851105,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003995561189205949,
      "loss": 4.719,
      "mean_token_accuracy": 0.24626323580741882,
      "num_tokens": 76350479.0,
      "step": 33295
    },
    {
      "entropy": 5.4611763000488285,
      "epoch": 3.1988472622478388,
      "grad_norm": 1.125,
      "learning_rate": 0.0003995275663294818,
      "loss": 4.7333,
      "mean_token_accuracy": 0.24647459387779236,
      "num_tokens": 76363433.0,
      "step": 33300
    },
    {
      "entropy": 5.332909727096558,
      "epoch": 3.1993275696445727,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003994990108472084,
      "loss": 4.6105,
      "mean_token_accuracy": 0.2572768867015839,
      "num_tokens": 76374000.0,
      "step": 33305
    },
    {
      "entropy": 5.364466714859009,
      "epoch": 3.1998078770413065,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039947045247443755,
      "loss": 4.6492,
      "mean_token_accuracy": 0.25099267810583115,
      "num_tokens": 76384192.0,
      "step": 33310
    },
    {
      "entropy": 5.489482116699219,
      "epoch": 3.2002881844380404,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039944189121183247,
      "loss": 4.8147,
      "mean_token_accuracy": 0.24267087429761885,
      "num_tokens": 76396641.0,
      "step": 33315
    },
    {
      "entropy": 5.375642347335815,
      "epoch": 3.2007684918347743,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039941332706005617,
      "loss": 4.5814,
      "mean_token_accuracy": 0.261711785197258,
      "num_tokens": 76407954.0,
      "step": 33320
    },
    {
      "entropy": 5.4482122421264645,
      "epoch": 3.201248799231508,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039938476001977175,
      "loss": 4.6917,
      "mean_token_accuracy": 0.2547926530241966,
      "num_tokens": 76419024.0,
      "step": 33325
    },
    {
      "entropy": 5.464232540130615,
      "epoch": 3.201729106628242,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039935619009164264,
      "loss": 4.7277,
      "mean_token_accuracy": 0.24213629215955734,
      "num_tokens": 76429786.0,
      "step": 33330
    },
    {
      "entropy": 5.352515602111817,
      "epoch": 3.202209414024976,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003993276172763321,
      "loss": 4.5741,
      "mean_token_accuracy": 0.2603064015507698,
      "num_tokens": 76440439.0,
      "step": 33335
    },
    {
      "entropy": 5.420913171768189,
      "epoch": 3.20268972142171,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039929904157450343,
      "loss": 4.7314,
      "mean_token_accuracy": 0.24846592396497727,
      "num_tokens": 76451855.0,
      "step": 33340
    },
    {
      "entropy": 5.387923622131348,
      "epoch": 3.2031700288184437,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039927046298682007,
      "loss": 4.639,
      "mean_token_accuracy": 0.2551468268036842,
      "num_tokens": 76463662.0,
      "step": 33345
    },
    {
      "entropy": 5.426482629776001,
      "epoch": 3.2036503362151776,
      "grad_norm": 1.125,
      "learning_rate": 0.0003992418815139456,
      "loss": 4.6771,
      "mean_token_accuracy": 0.24976954758167266,
      "num_tokens": 76476273.0,
      "step": 33350
    },
    {
      "entropy": 5.504788208007812,
      "epoch": 3.2041306436119115,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039921329715654355,
      "loss": 4.7515,
      "mean_token_accuracy": 0.24488118141889573,
      "num_tokens": 76488958.0,
      "step": 33355
    },
    {
      "entropy": 5.37974362373352,
      "epoch": 3.2046109510086453,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003991847099152775,
      "loss": 4.639,
      "mean_token_accuracy": 0.25515382885932925,
      "num_tokens": 76499606.0,
      "step": 33360
    },
    {
      "entropy": 5.525476598739624,
      "epoch": 3.2050912584053792,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003991561197908114,
      "loss": 4.8095,
      "mean_token_accuracy": 0.23829956948757172,
      "num_tokens": 76510953.0,
      "step": 33365
    },
    {
      "entropy": 5.454489660263062,
      "epoch": 3.2055715658021136,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003991275267838088,
      "loss": 4.682,
      "mean_token_accuracy": 0.24955350458621978,
      "num_tokens": 76522420.0,
      "step": 33370
    },
    {
      "entropy": 5.395744562149048,
      "epoch": 3.2060518731988474,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039909893089493353,
      "loss": 4.6939,
      "mean_token_accuracy": 0.2576077550649643,
      "num_tokens": 76533748.0,
      "step": 33375
    },
    {
      "entropy": 5.432915067672729,
      "epoch": 3.2065321805955813,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039907033212484966,
      "loss": 4.6766,
      "mean_token_accuracy": 0.2539109021425247,
      "num_tokens": 76544831.0,
      "step": 33380
    },
    {
      "entropy": 5.427533197402954,
      "epoch": 3.207012487992315,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003990417304742211,
      "loss": 4.7378,
      "mean_token_accuracy": 0.24957716166973115,
      "num_tokens": 76558048.0,
      "step": 33385
    },
    {
      "entropy": 5.44864444732666,
      "epoch": 3.207492795389049,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003990131259437119,
      "loss": 4.7277,
      "mean_token_accuracy": 0.25338455885648725,
      "num_tokens": 76568461.0,
      "step": 33390
    },
    {
      "entropy": 5.415866899490356,
      "epoch": 3.207973102785783,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003989845185339861,
      "loss": 4.6907,
      "mean_token_accuracy": 0.24751657694578172,
      "num_tokens": 76580191.0,
      "step": 33395
    },
    {
      "entropy": 5.435626459121704,
      "epoch": 3.208453410182517,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003989559082457079,
      "loss": 4.6558,
      "mean_token_accuracy": 0.2547546401619911,
      "num_tokens": 76591846.0,
      "step": 33400
    },
    {
      "entropy": 5.414660978317261,
      "epoch": 3.2089337175792507,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039892729507954173,
      "loss": 4.6317,
      "mean_token_accuracy": 0.2564762070775032,
      "num_tokens": 76602301.0,
      "step": 33405
    },
    {
      "entropy": 5.399563980102539,
      "epoch": 3.2094140249759846,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039889867903615165,
      "loss": 4.6859,
      "mean_token_accuracy": 0.25172879695892336,
      "num_tokens": 76613299.0,
      "step": 33410
    },
    {
      "entropy": 5.498921728134155,
      "epoch": 3.2098943323727185,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039887006011620217,
      "loss": 4.7395,
      "mean_token_accuracy": 0.24502476900815964,
      "num_tokens": 76625022.0,
      "step": 33415
    },
    {
      "entropy": 5.392694044113159,
      "epoch": 3.2103746397694524,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039884143832035775,
      "loss": 4.6706,
      "mean_token_accuracy": 0.25429461151361465,
      "num_tokens": 76636393.0,
      "step": 33420
    },
    {
      "entropy": 5.424818229675293,
      "epoch": 3.2108549471661862,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003988128136492828,
      "loss": 4.6948,
      "mean_token_accuracy": 0.24783320277929305,
      "num_tokens": 76648953.0,
      "step": 33425
    },
    {
      "entropy": 5.42250165939331,
      "epoch": 3.21133525456292,
      "grad_norm": 1.125,
      "learning_rate": 0.000398784186103642,
      "loss": 4.6954,
      "mean_token_accuracy": 0.2506685346364975,
      "num_tokens": 76660843.0,
      "step": 33430
    },
    {
      "entropy": 5.429387855529785,
      "epoch": 3.211815561959654,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039875555568409996,
      "loss": 4.6966,
      "mean_token_accuracy": 0.2560866430401802,
      "num_tokens": 76671745.0,
      "step": 33435
    },
    {
      "entropy": 5.405222082138062,
      "epoch": 3.212295869356388,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003987269223913214,
      "loss": 4.6489,
      "mean_token_accuracy": 0.2504561021924019,
      "num_tokens": 76684386.0,
      "step": 33440
    },
    {
      "entropy": 5.4331011295318605,
      "epoch": 3.212776176753122,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039869828622597105,
      "loss": 4.6625,
      "mean_token_accuracy": 0.2557015120983124,
      "num_tokens": 76694670.0,
      "step": 33445
    },
    {
      "entropy": 5.387274599075317,
      "epoch": 3.213256484149856,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039866964718871385,
      "loss": 4.6868,
      "mean_token_accuracy": 0.25418657809495926,
      "num_tokens": 76705567.0,
      "step": 33450
    },
    {
      "entropy": 5.389308023452759,
      "epoch": 3.21373679154659,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003986410052802146,
      "loss": 4.6578,
      "mean_token_accuracy": 0.25231530219316484,
      "num_tokens": 76717211.0,
      "step": 33455
    },
    {
      "entropy": 5.477072048187256,
      "epoch": 3.214217098943324,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039861236050113845,
      "loss": 4.7564,
      "mean_token_accuracy": 0.24337191134691238,
      "num_tokens": 76728414.0,
      "step": 33460
    },
    {
      "entropy": 5.3558159351348875,
      "epoch": 3.2146974063400577,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003985837128521503,
      "loss": 4.6329,
      "mean_token_accuracy": 0.2612708568572998,
      "num_tokens": 76739790.0,
      "step": 33465
    },
    {
      "entropy": 5.459284973144531,
      "epoch": 3.2151777137367916,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003985550623339153,
      "loss": 4.7108,
      "mean_token_accuracy": 0.2521683156490326,
      "num_tokens": 76751558.0,
      "step": 33470
    },
    {
      "entropy": 5.4925768852233885,
      "epoch": 3.2156580211335255,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003985264089470987,
      "loss": 4.8416,
      "mean_token_accuracy": 0.2394635945558548,
      "num_tokens": 76763004.0,
      "step": 33475
    },
    {
      "entropy": 5.437912559509277,
      "epoch": 3.2161383285302594,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039849775269236556,
      "loss": 4.7232,
      "mean_token_accuracy": 0.25398400574922564,
      "num_tokens": 76773625.0,
      "step": 33480
    },
    {
      "entropy": 5.465202331542969,
      "epoch": 3.2166186359269933,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039846909357038135,
      "loss": 4.7606,
      "mean_token_accuracy": 0.24300415366888045,
      "num_tokens": 76786351.0,
      "step": 33485
    },
    {
      "entropy": 5.447390413284301,
      "epoch": 3.217098943323727,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003984404315818115,
      "loss": 4.671,
      "mean_token_accuracy": 0.24946658164262772,
      "num_tokens": 76796997.0,
      "step": 33490
    },
    {
      "entropy": 5.479136562347412,
      "epoch": 3.217579250720461,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039841176672732127,
      "loss": 4.7436,
      "mean_token_accuracy": 0.24495499581098557,
      "num_tokens": 76809690.0,
      "step": 33495
    },
    {
      "entropy": 5.42999234199524,
      "epoch": 3.218059558117195,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003983830990075763,
      "loss": 4.6359,
      "mean_token_accuracy": 0.25093443095684054,
      "num_tokens": 76820387.0,
      "step": 33500
    },
    {
      "entropy": 5.4627382278442385,
      "epoch": 3.218539865513929,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039835442842324216,
      "loss": 4.7999,
      "mean_token_accuracy": 0.2427885726094246,
      "num_tokens": 76833261.0,
      "step": 33505
    },
    {
      "entropy": 5.382210397720337,
      "epoch": 3.2190201729106627,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039832575497498454,
      "loss": 4.6315,
      "mean_token_accuracy": 0.2517231643199921,
      "num_tokens": 76844952.0,
      "step": 33510
    },
    {
      "entropy": 5.445790576934814,
      "epoch": 3.2195004803073966,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039829707866346895,
      "loss": 4.684,
      "mean_token_accuracy": 0.25006839632987976,
      "num_tokens": 76856076.0,
      "step": 33515
    },
    {
      "entropy": 5.525357723236084,
      "epoch": 3.219980787704131,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003982683994893614,
      "loss": 4.8652,
      "mean_token_accuracy": 0.2405822291970253,
      "num_tokens": 76867510.0,
      "step": 33520
    },
    {
      "entropy": 5.479172277450561,
      "epoch": 3.2204610951008648,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039823971745332764,
      "loss": 4.7828,
      "mean_token_accuracy": 0.2473609298467636,
      "num_tokens": 76879919.0,
      "step": 33525
    },
    {
      "entropy": 5.4146452903747555,
      "epoch": 3.2209414024975986,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003982110325560336,
      "loss": 4.6975,
      "mean_token_accuracy": 0.24968036115169526,
      "num_tokens": 76891402.0,
      "step": 33530
    },
    {
      "entropy": 5.494327259063721,
      "epoch": 3.2214217098943325,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003981823447981453,
      "loss": 4.7083,
      "mean_token_accuracy": 0.24860449880361557,
      "num_tokens": 76902703.0,
      "step": 33535
    },
    {
      "entropy": 5.4630872249603275,
      "epoch": 3.2219020172910664,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039815365418032855,
      "loss": 4.6568,
      "mean_token_accuracy": 0.2557025909423828,
      "num_tokens": 76913661.0,
      "step": 33540
    },
    {
      "entropy": 5.385120916366577,
      "epoch": 3.2223823246878003,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039812496070324983,
      "loss": 4.62,
      "mean_token_accuracy": 0.26058937311172486,
      "num_tokens": 76924175.0,
      "step": 33545
    },
    {
      "entropy": 5.439227437973022,
      "epoch": 3.222862632084534,
      "grad_norm": 1.140625,
      "learning_rate": 0.000398096264367575,
      "loss": 4.7289,
      "mean_token_accuracy": 0.25197282135486604,
      "num_tokens": 76935158.0,
      "step": 33550
    },
    {
      "entropy": 5.374966859817505,
      "epoch": 3.223342939481268,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003980675651739705,
      "loss": 4.5759,
      "mean_token_accuracy": 0.26160147339105605,
      "num_tokens": 76944918.0,
      "step": 33555
    },
    {
      "entropy": 5.320950889587403,
      "epoch": 3.223823246878002,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039803886312310253,
      "loss": 4.5981,
      "mean_token_accuracy": 0.26362376362085344,
      "num_tokens": 76955128.0,
      "step": 33560
    },
    {
      "entropy": 5.3682116031646725,
      "epoch": 3.224303554274736,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039801015821563755,
      "loss": 4.7275,
      "mean_token_accuracy": 0.249272720515728,
      "num_tokens": 76966227.0,
      "step": 33565
    },
    {
      "entropy": 5.441119432449341,
      "epoch": 3.2247838616714697,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003979814504522419,
      "loss": 4.6513,
      "mean_token_accuracy": 0.24955234974622725,
      "num_tokens": 76977059.0,
      "step": 33570
    },
    {
      "entropy": 5.400148582458496,
      "epoch": 3.2252641690682036,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039795273983358223,
      "loss": 4.639,
      "mean_token_accuracy": 0.2576239466667175,
      "num_tokens": 76987447.0,
      "step": 33575
    },
    {
      "entropy": 5.407995271682739,
      "epoch": 3.2257444764649374,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039792402636032497,
      "loss": 4.7436,
      "mean_token_accuracy": 0.25310891568660737,
      "num_tokens": 76998960.0,
      "step": 33580
    },
    {
      "entropy": 5.415073680877685,
      "epoch": 3.2262247838616713,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039789531003313696,
      "loss": 4.7361,
      "mean_token_accuracy": 0.24834656864404678,
      "num_tokens": 77011255.0,
      "step": 33585
    },
    {
      "entropy": 5.453651332855225,
      "epoch": 3.226705091258405,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003978665908526846,
      "loss": 4.6911,
      "mean_token_accuracy": 0.2518963202834129,
      "num_tokens": 77023061.0,
      "step": 33590
    },
    {
      "entropy": 5.415170812606812,
      "epoch": 3.227185398655139,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003978378688196349,
      "loss": 4.6749,
      "mean_token_accuracy": 0.25083225816488264,
      "num_tokens": 77033949.0,
      "step": 33595
    },
    {
      "entropy": 5.403607368469238,
      "epoch": 3.227665706051873,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003978091439346546,
      "loss": 4.6627,
      "mean_token_accuracy": 0.25209677815437315,
      "num_tokens": 77044990.0,
      "step": 33600
    },
    {
      "entropy": 5.45111870765686,
      "epoch": 3.2281460134486073,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003977804161984108,
      "loss": 4.7478,
      "mean_token_accuracy": 0.2434329494833946,
      "num_tokens": 77055099.0,
      "step": 33605
    },
    {
      "entropy": 5.459322881698609,
      "epoch": 3.228626320845341,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003977516856115702,
      "loss": 4.6809,
      "mean_token_accuracy": 0.25835915356874467,
      "num_tokens": 77065854.0,
      "step": 33610
    },
    {
      "entropy": 5.427380466461182,
      "epoch": 3.229106628242075,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039772295217479993,
      "loss": 4.7127,
      "mean_token_accuracy": 0.25325003564357756,
      "num_tokens": 77077531.0,
      "step": 33615
    },
    {
      "entropy": 5.479922437667847,
      "epoch": 3.229586935638809,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003976942158887671,
      "loss": 4.6746,
      "mean_token_accuracy": 0.25279997289180756,
      "num_tokens": 77088659.0,
      "step": 33620
    },
    {
      "entropy": 5.461447525024414,
      "epoch": 3.230067243035543,
      "grad_norm": 1.234375,
      "learning_rate": 0.000397665476754139,
      "loss": 4.7878,
      "mean_token_accuracy": 0.24266346991062165,
      "num_tokens": 77099868.0,
      "step": 33625
    },
    {
      "entropy": 5.433854341506958,
      "epoch": 3.2305475504322767,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003976367347715828,
      "loss": 4.6836,
      "mean_token_accuracy": 0.257380373775959,
      "num_tokens": 77112415.0,
      "step": 33630
    },
    {
      "entropy": 5.4521870613098145,
      "epoch": 3.2310278578290106,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003976079899417657,
      "loss": 4.8022,
      "mean_token_accuracy": 0.24390265643596648,
      "num_tokens": 77124978.0,
      "step": 33635
    },
    {
      "entropy": 5.479333400726318,
      "epoch": 3.2315081652257445,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003975792422653552,
      "loss": 4.7525,
      "mean_token_accuracy": 0.25077047795057295,
      "num_tokens": 77135800.0,
      "step": 33640
    },
    {
      "entropy": 5.457885599136352,
      "epoch": 3.2319884726224783,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003975504917430186,
      "loss": 4.7527,
      "mean_token_accuracy": 0.24382736086845397,
      "num_tokens": 77147269.0,
      "step": 33645
    },
    {
      "entropy": 5.428481101989746,
      "epoch": 3.2324687800192122,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003975217383754235,
      "loss": 4.7113,
      "mean_token_accuracy": 0.24747423231601715,
      "num_tokens": 77159522.0,
      "step": 33650
    },
    {
      "entropy": 5.441060638427734,
      "epoch": 3.232949087415946,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0003974929821632375,
      "loss": 4.7233,
      "mean_token_accuracy": 0.24868059605360032,
      "num_tokens": 77170884.0,
      "step": 33655
    },
    {
      "entropy": 5.364257717132569,
      "epoch": 3.23342939481268,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039746422310712814,
      "loss": 4.5747,
      "mean_token_accuracy": 0.2554344519972801,
      "num_tokens": 77181554.0,
      "step": 33660
    },
    {
      "entropy": 5.471784210205078,
      "epoch": 3.233909702209414,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003974354612077632,
      "loss": 4.7719,
      "mean_token_accuracy": 0.24584585130214692,
      "num_tokens": 77192988.0,
      "step": 33665
    },
    {
      "entropy": 5.445204019546509,
      "epoch": 3.2343900096061478,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003974066964658104,
      "loss": 4.6511,
      "mean_token_accuracy": 0.25633180886507034,
      "num_tokens": 77204107.0,
      "step": 33670
    },
    {
      "entropy": 5.528897523880005,
      "epoch": 3.2348703170028816,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039737792888193754,
      "loss": 4.8264,
      "mean_token_accuracy": 0.243452724814415,
      "num_tokens": 77217008.0,
      "step": 33675
    },
    {
      "entropy": 5.440745782852173,
      "epoch": 3.235350624399616,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003973491584568126,
      "loss": 4.7436,
      "mean_token_accuracy": 0.24723830372095107,
      "num_tokens": 77228651.0,
      "step": 33680
    },
    {
      "entropy": 5.368432521820068,
      "epoch": 3.23583093179635,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003973203851911035,
      "loss": 4.6111,
      "mean_token_accuracy": 0.26239985078573225,
      "num_tokens": 77240251.0,
      "step": 33685
    },
    {
      "entropy": 5.453677845001221,
      "epoch": 3.2363112391930837,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003972916090854782,
      "loss": 4.7105,
      "mean_token_accuracy": 0.26067701131105425,
      "num_tokens": 77252160.0,
      "step": 33690
    },
    {
      "entropy": 5.41200213432312,
      "epoch": 3.2367915465898176,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039726283014060497,
      "loss": 4.6601,
      "mean_token_accuracy": 0.2551521956920624,
      "num_tokens": 77263678.0,
      "step": 33695
    },
    {
      "entropy": 5.424459743499756,
      "epoch": 3.2372718539865515,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003972340483571519,
      "loss": 4.706,
      "mean_token_accuracy": 0.2525022953748703,
      "num_tokens": 77275180.0,
      "step": 33700
    },
    {
      "entropy": 5.431145572662354,
      "epoch": 3.2377521613832854,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039720526373578704,
      "loss": 4.7304,
      "mean_token_accuracy": 0.24507690221071243,
      "num_tokens": 77286450.0,
      "step": 33705
    },
    {
      "entropy": 5.449428844451904,
      "epoch": 3.2382324687800192,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039717647627717894,
      "loss": 4.6652,
      "mean_token_accuracy": 0.2538145810365677,
      "num_tokens": 77297011.0,
      "step": 33710
    },
    {
      "entropy": 5.380990409851075,
      "epoch": 3.238712776176753,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003971476859819958,
      "loss": 4.5807,
      "mean_token_accuracy": 0.26832272857427597,
      "num_tokens": 77307408.0,
      "step": 33715
    },
    {
      "entropy": 5.443876314163208,
      "epoch": 3.239193083573487,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003971188928509062,
      "loss": 4.7455,
      "mean_token_accuracy": 0.24821969866752625,
      "num_tokens": 77316762.0,
      "step": 33720
    },
    {
      "entropy": 5.447432136535644,
      "epoch": 3.239673390970221,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003970900968845784,
      "loss": 4.7263,
      "mean_token_accuracy": 0.24966494739055634,
      "num_tokens": 77327233.0,
      "step": 33725
    },
    {
      "entropy": 5.5075035572052,
      "epoch": 3.2401536983669548,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039706129808368115,
      "loss": 4.7697,
      "mean_token_accuracy": 0.2454906404018402,
      "num_tokens": 77338001.0,
      "step": 33730
    },
    {
      "entropy": 5.3819300651550295,
      "epoch": 3.2406340057636887,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003970324964488829,
      "loss": 4.7231,
      "mean_token_accuracy": 0.2488697201013565,
      "num_tokens": 77350243.0,
      "step": 33735
    },
    {
      "entropy": 5.416616153717041,
      "epoch": 3.2411143131604225,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039700369198085255,
      "loss": 4.7362,
      "mean_token_accuracy": 0.251139934360981,
      "num_tokens": 77360712.0,
      "step": 33740
    },
    {
      "entropy": 5.422603559494019,
      "epoch": 3.2415946205571564,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039697488468025876,
      "loss": 4.6508,
      "mean_token_accuracy": 0.2518091484904289,
      "num_tokens": 77371399.0,
      "step": 33745
    },
    {
      "entropy": 5.4743774890899655,
      "epoch": 3.2420749279538903,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003969460745477703,
      "loss": 4.772,
      "mean_token_accuracy": 0.24697636365890502,
      "num_tokens": 77381652.0,
      "step": 33750
    },
    {
      "entropy": 5.450820827484131,
      "epoch": 3.2425552353506246,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039691726158405606,
      "loss": 4.6914,
      "mean_token_accuracy": 0.2511848762631416,
      "num_tokens": 77393335.0,
      "step": 33755
    },
    {
      "entropy": 5.412785530090332,
      "epoch": 3.2430355427473585,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039688844578978516,
      "loss": 4.6969,
      "mean_token_accuracy": 0.2530561923980713,
      "num_tokens": 77404311.0,
      "step": 33760
    },
    {
      "entropy": 5.469546937942505,
      "epoch": 3.2435158501440924,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003968596271656263,
      "loss": 4.7261,
      "mean_token_accuracy": 0.2511828765273094,
      "num_tokens": 77415886.0,
      "step": 33765
    },
    {
      "entropy": 5.467490863800049,
      "epoch": 3.2439961575408263,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039683080571224885,
      "loss": 4.7206,
      "mean_token_accuracy": 0.2489307627081871,
      "num_tokens": 77426300.0,
      "step": 33770
    },
    {
      "entropy": 5.5002082824707035,
      "epoch": 3.24447646493756,
      "grad_norm": 1.125,
      "learning_rate": 0.0003968019814303219,
      "loss": 4.7291,
      "mean_token_accuracy": 0.24659742861986161,
      "num_tokens": 77437803.0,
      "step": 33775
    },
    {
      "entropy": 5.4963791847229,
      "epoch": 3.244956772334294,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003967731543205145,
      "loss": 4.7284,
      "mean_token_accuracy": 0.2560193449258804,
      "num_tokens": 77448850.0,
      "step": 33780
    },
    {
      "entropy": 5.457386255264282,
      "epoch": 3.245437079731028,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039674432438349607,
      "loss": 4.6929,
      "mean_token_accuracy": 0.2515192121267319,
      "num_tokens": 77458933.0,
      "step": 33785
    },
    {
      "entropy": 5.449470376968383,
      "epoch": 3.245917387127762,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000396715491619936,
      "loss": 4.7085,
      "mean_token_accuracy": 0.24836157262325287,
      "num_tokens": 77469916.0,
      "step": 33790
    },
    {
      "entropy": 5.512402105331421,
      "epoch": 3.2463976945244957,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003966866560305036,
      "loss": 4.7004,
      "mean_token_accuracy": 0.2522930935025215,
      "num_tokens": 77480548.0,
      "step": 33795
    },
    {
      "entropy": 5.423204612731934,
      "epoch": 3.2468780019212296,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039665781761586837,
      "loss": 4.7239,
      "mean_token_accuracy": 0.24071845263242722,
      "num_tokens": 77490664.0,
      "step": 33800
    },
    {
      "entropy": 5.445045089721679,
      "epoch": 3.2473583093179634,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003966289763766999,
      "loss": 4.6774,
      "mean_token_accuracy": 0.25720722377300265,
      "num_tokens": 77502010.0,
      "step": 33805
    },
    {
      "entropy": 5.410501289367676,
      "epoch": 3.2478386167146973,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003966001323136678,
      "loss": 4.674,
      "mean_token_accuracy": 0.25674364566802976,
      "num_tokens": 77513453.0,
      "step": 33810
    },
    {
      "entropy": 5.4680122375488285,
      "epoch": 3.248318924111431,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003965712854274416,
      "loss": 4.7984,
      "mean_token_accuracy": 0.24426113814115524,
      "num_tokens": 77526294.0,
      "step": 33815
    },
    {
      "entropy": 5.513613843917847,
      "epoch": 3.248799231508165,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003965424357186912,
      "loss": 4.7574,
      "mean_token_accuracy": 0.24706168621778488,
      "num_tokens": 77537753.0,
      "step": 33820
    },
    {
      "entropy": 5.401170635223389,
      "epoch": 3.249279538904899,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003965135831880864,
      "loss": 4.6438,
      "mean_token_accuracy": 0.2579970195889473,
      "num_tokens": 77548931.0,
      "step": 33825
    },
    {
      "entropy": 5.377698278427124,
      "epoch": 3.249759846301633,
      "grad_norm": 1.125,
      "learning_rate": 0.000396484727836297,
      "loss": 4.6671,
      "mean_token_accuracy": 0.25759522914886473,
      "num_tokens": 77560259.0,
      "step": 33830
    },
    {
      "entropy": 5.4181013107299805,
      "epoch": 3.2502401536983667,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003964558696639928,
      "loss": 4.7182,
      "mean_token_accuracy": 0.25558450669050214,
      "num_tokens": 77571688.0,
      "step": 33835
    },
    {
      "entropy": 5.475750398635864,
      "epoch": 3.250720461095101,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003964270086718441,
      "loss": 4.7697,
      "mean_token_accuracy": 0.2512074589729309,
      "num_tokens": 77583754.0,
      "step": 33840
    },
    {
      "entropy": 5.447564888000488,
      "epoch": 3.251200768491835,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039639814486052083,
      "loss": 4.6459,
      "mean_token_accuracy": 0.2646897718310356,
      "num_tokens": 77596013.0,
      "step": 33845
    },
    {
      "entropy": 5.469159460067749,
      "epoch": 3.251681075888569,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003963692782306931,
      "loss": 4.7675,
      "mean_token_accuracy": 0.24708865135908126,
      "num_tokens": 77607349.0,
      "step": 33850
    },
    {
      "entropy": 5.431390380859375,
      "epoch": 3.2521613832853027,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003963404087830311,
      "loss": 4.7054,
      "mean_token_accuracy": 0.2524047076702118,
      "num_tokens": 77618325.0,
      "step": 33855
    },
    {
      "entropy": 5.546149110794067,
      "epoch": 3.2526416906820366,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003963115365182051,
      "loss": 4.8084,
      "mean_token_accuracy": 0.237033449113369,
      "num_tokens": 77629794.0,
      "step": 33860
    },
    {
      "entropy": 5.357737970352173,
      "epoch": 3.2531219980787704,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039628266143688554,
      "loss": 4.6598,
      "mean_token_accuracy": 0.25220225751399994,
      "num_tokens": 77641800.0,
      "step": 33865
    },
    {
      "entropy": 5.389326858520508,
      "epoch": 3.2536023054755043,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003962537835397426,
      "loss": 4.6642,
      "mean_token_accuracy": 0.24882374852895736,
      "num_tokens": 77653533.0,
      "step": 33870
    },
    {
      "entropy": 5.431952381134034,
      "epoch": 3.254082612872238,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039622490282744684,
      "loss": 4.6966,
      "mean_token_accuracy": 0.2500015839934349,
      "num_tokens": 77663508.0,
      "step": 33875
    },
    {
      "entropy": 5.452558374404907,
      "epoch": 3.254562920268972,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003961960193006689,
      "loss": 4.7364,
      "mean_token_accuracy": 0.2447968080639839,
      "num_tokens": 77674956.0,
      "step": 33880
    },
    {
      "entropy": 5.480708408355713,
      "epoch": 3.255043227665706,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003961671329600792,
      "loss": 4.7242,
      "mean_token_accuracy": 0.2505678921937943,
      "num_tokens": 77686000.0,
      "step": 33885
    },
    {
      "entropy": 5.350408983230591,
      "epoch": 3.25552353506244,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003961382438063485,
      "loss": 4.6307,
      "mean_token_accuracy": 0.2503298565745354,
      "num_tokens": 77698979.0,
      "step": 33890
    },
    {
      "entropy": 5.396661233901978,
      "epoch": 3.2560038424591737,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003961093518401475,
      "loss": 4.5855,
      "mean_token_accuracy": 0.26050916165113447,
      "num_tokens": 77710397.0,
      "step": 33895
    },
    {
      "entropy": 5.41218752861023,
      "epoch": 3.2564841498559076,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039608045706214696,
      "loss": 4.6895,
      "mean_token_accuracy": 0.24470377564430237,
      "num_tokens": 77721665.0,
      "step": 33900
    },
    {
      "entropy": 5.4060399532318115,
      "epoch": 3.2569644572526415,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003960515594730177,
      "loss": 4.6957,
      "mean_token_accuracy": 0.25276950299739837,
      "num_tokens": 77733328.0,
      "step": 33905
    },
    {
      "entropy": 5.43626823425293,
      "epoch": 3.2574447646493754,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003960226590734307,
      "loss": 4.7244,
      "mean_token_accuracy": 0.24554295390844344,
      "num_tokens": 77745812.0,
      "step": 33910
    },
    {
      "entropy": 5.348901891708374,
      "epoch": 3.2579250720461097,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000395993755864057,
      "loss": 4.592,
      "mean_token_accuracy": 0.25951275080442426,
      "num_tokens": 77757473.0,
      "step": 33915
    },
    {
      "entropy": 5.474650764465332,
      "epoch": 3.2584053794428436,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003959648498455674,
      "loss": 4.7136,
      "mean_token_accuracy": 0.24815493524074556,
      "num_tokens": 77769924.0,
      "step": 33920
    },
    {
      "entropy": 5.454634809494019,
      "epoch": 3.2588856868395775,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039593594101863333,
      "loss": 4.7038,
      "mean_token_accuracy": 0.2510687544941902,
      "num_tokens": 77782048.0,
      "step": 33925
    },
    {
      "entropy": 5.345828056335449,
      "epoch": 3.2593659942363113,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039590702938392577,
      "loss": 4.5832,
      "mean_token_accuracy": 0.260757303237915,
      "num_tokens": 77792371.0,
      "step": 33930
    },
    {
      "entropy": 5.44304141998291,
      "epoch": 3.2598463016330452,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039587811494211594,
      "loss": 4.6994,
      "mean_token_accuracy": 0.2502101868391037,
      "num_tokens": 77804089.0,
      "step": 33935
    },
    {
      "entropy": 5.395117473602295,
      "epoch": 3.260326609029779,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039584919769387536,
      "loss": 4.6782,
      "mean_token_accuracy": 0.2548355981707573,
      "num_tokens": 77815962.0,
      "step": 33940
    },
    {
      "entropy": 5.479871129989624,
      "epoch": 3.260806916426513,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003958202776398751,
      "loss": 4.7873,
      "mean_token_accuracy": 0.24287697523832322,
      "num_tokens": 77827396.0,
      "step": 33945
    },
    {
      "entropy": 5.419396924972534,
      "epoch": 3.261287223823247,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003957913547807868,
      "loss": 4.7242,
      "mean_token_accuracy": 0.2560410752892494,
      "num_tokens": 77838916.0,
      "step": 33950
    },
    {
      "entropy": 5.558325958251953,
      "epoch": 3.2617675312199808,
      "grad_norm": 1.171875,
      "learning_rate": 0.000395762429117282,
      "loss": 4.8549,
      "mean_token_accuracy": 0.24638267159461974,
      "num_tokens": 77851393.0,
      "step": 33955
    },
    {
      "entropy": 5.495118188858032,
      "epoch": 3.2622478386167146,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003957335006500321,
      "loss": 4.7355,
      "mean_token_accuracy": 0.24603354781866074,
      "num_tokens": 77863124.0,
      "step": 33960
    },
    {
      "entropy": 5.398018836975098,
      "epoch": 3.2627281460134485,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039570456937970883,
      "loss": 4.675,
      "mean_token_accuracy": 0.2588006466627121,
      "num_tokens": 77875568.0,
      "step": 33965
    },
    {
      "entropy": 5.48918833732605,
      "epoch": 3.2632084534101824,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003956756353069839,
      "loss": 4.7671,
      "mean_token_accuracy": 0.25178506672382356,
      "num_tokens": 77885747.0,
      "step": 33970
    },
    {
      "entropy": 5.365253496170044,
      "epoch": 3.2636887608069163,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003956466984325291,
      "loss": 4.5935,
      "mean_token_accuracy": 0.2575103849172592,
      "num_tokens": 77897074.0,
      "step": 33975
    },
    {
      "entropy": 5.459847354888916,
      "epoch": 3.26416906820365,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039561775875701616,
      "loss": 4.717,
      "mean_token_accuracy": 0.25226900428533555,
      "num_tokens": 77907718.0,
      "step": 33980
    },
    {
      "entropy": 5.425333547592163,
      "epoch": 3.264649375600384,
      "grad_norm": 1.0625,
      "learning_rate": 0.000395588816281117,
      "loss": 4.7055,
      "mean_token_accuracy": 0.2437622994184494,
      "num_tokens": 77919768.0,
      "step": 33985
    },
    {
      "entropy": 5.481195116043091,
      "epoch": 3.2651296829971184,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003955598710055036,
      "loss": 4.7346,
      "mean_token_accuracy": 0.24786432087421417,
      "num_tokens": 77930927.0,
      "step": 33990
    },
    {
      "entropy": 5.4768575668334964,
      "epoch": 3.2656099903938522,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000395530922930848,
      "loss": 4.7037,
      "mean_token_accuracy": 0.25009425133466723,
      "num_tokens": 77941907.0,
      "step": 33995
    },
    {
      "entropy": 5.414022827148438,
      "epoch": 3.266090297790586,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003955019720578223,
      "loss": 4.6382,
      "mean_token_accuracy": 0.25246876031160354,
      "num_tokens": 77952800.0,
      "step": 34000
    },
    {
      "entropy": 5.491213178634643,
      "epoch": 3.26657060518732,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003954730183870987,
      "loss": 4.7677,
      "mean_token_accuracy": 0.2457813709974289,
      "num_tokens": 77963874.0,
      "step": 34005
    },
    {
      "entropy": 5.425838422775269,
      "epoch": 3.267050912584054,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003954440619193491,
      "loss": 4.578,
      "mean_token_accuracy": 0.2602605685591698,
      "num_tokens": 77975282.0,
      "step": 34010
    },
    {
      "entropy": 5.397683954238891,
      "epoch": 3.2675312199807878,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039541510265524626,
      "loss": 4.6235,
      "mean_token_accuracy": 0.24995785653591157,
      "num_tokens": 77987511.0,
      "step": 34015
    },
    {
      "entropy": 5.444394636154175,
      "epoch": 3.2680115273775217,
      "grad_norm": 1.25,
      "learning_rate": 0.0003953861405954622,
      "loss": 4.6759,
      "mean_token_accuracy": 0.2459824338555336,
      "num_tokens": 77997805.0,
      "step": 34020
    },
    {
      "entropy": 5.452789306640625,
      "epoch": 3.2684918347742555,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003953571757406694,
      "loss": 4.7421,
      "mean_token_accuracy": 0.24750946313142777,
      "num_tokens": 78010312.0,
      "step": 34025
    },
    {
      "entropy": 5.430243396759034,
      "epoch": 3.2689721421709894,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039532820809154044,
      "loss": 4.6517,
      "mean_token_accuracy": 0.25578898638486863,
      "num_tokens": 78019934.0,
      "step": 34030
    },
    {
      "entropy": 5.415246438980103,
      "epoch": 3.2694524495677233,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039529923764874774,
      "loss": 4.6365,
      "mean_token_accuracy": 0.2656590446829796,
      "num_tokens": 78031263.0,
      "step": 34035
    },
    {
      "entropy": 5.399154710769653,
      "epoch": 3.269932756964457,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000395270264412964,
      "loss": 4.6959,
      "mean_token_accuracy": 0.25262038558721545,
      "num_tokens": 78043829.0,
      "step": 34040
    },
    {
      "entropy": 5.545526218414307,
      "epoch": 3.270413064361191,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039524128838486184,
      "loss": 4.8314,
      "mean_token_accuracy": 0.24291644841432572,
      "num_tokens": 78055292.0,
      "step": 34045
    },
    {
      "entropy": 5.463749361038208,
      "epoch": 3.270893371757925,
      "grad_norm": 1.1875,
      "learning_rate": 0.000395212309565114,
      "loss": 4.7414,
      "mean_token_accuracy": 0.2458766683936119,
      "num_tokens": 78066969.0,
      "step": 34050
    },
    {
      "entropy": 5.530800247192383,
      "epoch": 3.271373679154659,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039518332795439326,
      "loss": 4.7954,
      "mean_token_accuracy": 0.2427621975541115,
      "num_tokens": 78077994.0,
      "step": 34055
    },
    {
      "entropy": 5.386464309692383,
      "epoch": 3.2718539865513927,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039515434355337254,
      "loss": 4.664,
      "mean_token_accuracy": 0.24558027386665343,
      "num_tokens": 78089730.0,
      "step": 34060
    },
    {
      "entropy": 5.450805521011352,
      "epoch": 3.272334293948127,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003951253563627248,
      "loss": 4.7287,
      "mean_token_accuracy": 0.2472706690430641,
      "num_tokens": 78100125.0,
      "step": 34065
    },
    {
      "entropy": 5.5077684879302975,
      "epoch": 3.2728146013448605,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003950963663831229,
      "loss": 4.7683,
      "mean_token_accuracy": 0.24255042821168898,
      "num_tokens": 78112170.0,
      "step": 34070
    },
    {
      "entropy": 5.493219518661499,
      "epoch": 3.273294908741595,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039506737361524007,
      "loss": 4.7644,
      "mean_token_accuracy": 0.2461818724870682,
      "num_tokens": 78123796.0,
      "step": 34075
    },
    {
      "entropy": 5.428887939453125,
      "epoch": 3.2737752161383287,
      "grad_norm": 1.125,
      "learning_rate": 0.00039503837805974926,
      "loss": 4.6412,
      "mean_token_accuracy": 0.2589148372411728,
      "num_tokens": 78134170.0,
      "step": 34080
    },
    {
      "entropy": 5.395914602279663,
      "epoch": 3.2742555235350626,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039500937971732376,
      "loss": 4.692,
      "mean_token_accuracy": 0.253236448764801,
      "num_tokens": 78145140.0,
      "step": 34085
    },
    {
      "entropy": 5.491656064987183,
      "epoch": 3.2747358309317964,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003949803785886369,
      "loss": 4.8202,
      "mean_token_accuracy": 0.25061438530683516,
      "num_tokens": 78157627.0,
      "step": 34090
    },
    {
      "entropy": 5.46493501663208,
      "epoch": 3.2752161383285303,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039495137467436184,
      "loss": 4.7688,
      "mean_token_accuracy": 0.24558406174182892,
      "num_tokens": 78170343.0,
      "step": 34095
    },
    {
      "entropy": 5.456604671478272,
      "epoch": 3.275696445725264,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039492236797517206,
      "loss": 4.6826,
      "mean_token_accuracy": 0.25154115408658984,
      "num_tokens": 78181393.0,
      "step": 34100
    },
    {
      "entropy": 5.447706174850464,
      "epoch": 3.276176753121998,
      "grad_norm": 1.109375,
      "learning_rate": 0.000394893358491741,
      "loss": 4.7064,
      "mean_token_accuracy": 0.24686822444200515,
      "num_tokens": 78193048.0,
      "step": 34105
    },
    {
      "entropy": 5.3727561950683596,
      "epoch": 3.276657060518732,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039486434622474216,
      "loss": 4.6385,
      "mean_token_accuracy": 0.2525596931576729,
      "num_tokens": 78203738.0,
      "step": 34110
    },
    {
      "entropy": 5.452292776107788,
      "epoch": 3.277137367915466,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039483533117484916,
      "loss": 4.7444,
      "mean_token_accuracy": 0.24564095139503478,
      "num_tokens": 78215171.0,
      "step": 34115
    },
    {
      "entropy": 5.4254233837127686,
      "epoch": 3.2776176753121997,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003948063133427356,
      "loss": 4.6527,
      "mean_token_accuracy": 0.25085035115480425,
      "num_tokens": 78225901.0,
      "step": 34120
    },
    {
      "entropy": 5.41339955329895,
      "epoch": 3.2780979827089336,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003947772927290751,
      "loss": 4.6655,
      "mean_token_accuracy": 0.25139930546283723,
      "num_tokens": 78237327.0,
      "step": 34125
    },
    {
      "entropy": 5.429017972946167,
      "epoch": 3.2785782901056675,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003947482693345416,
      "loss": 4.7067,
      "mean_token_accuracy": 0.24723327159881592,
      "num_tokens": 78248030.0,
      "step": 34130
    },
    {
      "entropy": 5.385182237625122,
      "epoch": 3.2790585975024014,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039471924315980894,
      "loss": 4.5991,
      "mean_token_accuracy": 0.2593761309981346,
      "num_tokens": 78259303.0,
      "step": 34135
    },
    {
      "entropy": 5.440959215164185,
      "epoch": 3.2795389048991357,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039469021420555087,
      "loss": 4.71,
      "mean_token_accuracy": 0.24831900447607042,
      "num_tokens": 78271284.0,
      "step": 34140
    },
    {
      "entropy": 5.380855131149292,
      "epoch": 3.280019212295869,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039466118247244143,
      "loss": 4.6441,
      "mean_token_accuracy": 0.2566983595490456,
      "num_tokens": 78282718.0,
      "step": 34145
    },
    {
      "entropy": 5.410724878311157,
      "epoch": 3.2804995196926034,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003946321479611547,
      "loss": 4.6619,
      "mean_token_accuracy": 0.2579051688313484,
      "num_tokens": 78294292.0,
      "step": 34150
    },
    {
      "entropy": 5.3962602615356445,
      "epoch": 3.2809798270893373,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039460311067236465,
      "loss": 4.7187,
      "mean_token_accuracy": 0.24541675299406052,
      "num_tokens": 78306834.0,
      "step": 34155
    },
    {
      "entropy": 5.485984134674072,
      "epoch": 3.281460134486071,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039457407060674557,
      "loss": 4.7461,
      "mean_token_accuracy": 0.2427325278520584,
      "num_tokens": 78319866.0,
      "step": 34160
    },
    {
      "entropy": 5.429823780059815,
      "epoch": 3.281940441882805,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039454502776497163,
      "loss": 4.6982,
      "mean_token_accuracy": 0.24426991641521453,
      "num_tokens": 78330933.0,
      "step": 34165
    },
    {
      "entropy": 5.427136993408203,
      "epoch": 3.282420749279539,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039451598214771706,
      "loss": 4.6769,
      "mean_token_accuracy": 0.25308824330568314,
      "num_tokens": 78342323.0,
      "step": 34170
    },
    {
      "entropy": 5.460287046432495,
      "epoch": 3.282901056676273,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003944869337556563,
      "loss": 4.6679,
      "mean_token_accuracy": 0.2628712758421898,
      "num_tokens": 78353090.0,
      "step": 34175
    },
    {
      "entropy": 5.4651618003845215,
      "epoch": 3.2833813640730067,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003944578825894639,
      "loss": 4.7358,
      "mean_token_accuracy": 0.2510448083281517,
      "num_tokens": 78362679.0,
      "step": 34180
    },
    {
      "entropy": 5.4237621307373045,
      "epoch": 3.2838616714697406,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039442882864981397,
      "loss": 4.7233,
      "mean_token_accuracy": 0.24250997453927994,
      "num_tokens": 78373592.0,
      "step": 34185
    },
    {
      "entropy": 5.390952920913696,
      "epoch": 3.2843419788664745,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039439977193738134,
      "loss": 4.6082,
      "mean_token_accuracy": 0.26498291939496993,
      "num_tokens": 78384617.0,
      "step": 34190
    },
    {
      "entropy": 5.396334266662597,
      "epoch": 3.2848222862632084,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039437071245284055,
      "loss": 4.7431,
      "mean_token_accuracy": 0.24323808550834655,
      "num_tokens": 78396253.0,
      "step": 34195
    },
    {
      "entropy": 5.469293737411499,
      "epoch": 3.2853025936599423,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003943416501968663,
      "loss": 4.7961,
      "mean_token_accuracy": 0.24020418673753738,
      "num_tokens": 78407852.0,
      "step": 34200
    },
    {
      "entropy": 5.3996106624603275,
      "epoch": 3.285782901056676,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039431258517013323,
      "loss": 4.6308,
      "mean_token_accuracy": 0.2594335973262787,
      "num_tokens": 78419434.0,
      "step": 34205
    },
    {
      "entropy": 5.509778356552124,
      "epoch": 3.28626320845341,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003942835173733163,
      "loss": 4.7189,
      "mean_token_accuracy": 0.24651930779218673,
      "num_tokens": 78431820.0,
      "step": 34210
    },
    {
      "entropy": 5.425617790222168,
      "epoch": 3.286743515850144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003942544468070903,
      "loss": 4.7245,
      "mean_token_accuracy": 0.2512019038200378,
      "num_tokens": 78443062.0,
      "step": 34215
    },
    {
      "entropy": 5.504342365264892,
      "epoch": 3.287223823246878,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003942253734721301,
      "loss": 4.7255,
      "mean_token_accuracy": 0.25646368265151975,
      "num_tokens": 78454091.0,
      "step": 34220
    },
    {
      "entropy": 5.524246549606323,
      "epoch": 3.287704130643612,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039419629736911076,
      "loss": 4.7931,
      "mean_token_accuracy": 0.2486431673169136,
      "num_tokens": 78466108.0,
      "step": 34225
    },
    {
      "entropy": 5.404341840744019,
      "epoch": 3.288184438040346,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039416721849870736,
      "loss": 4.6138,
      "mean_token_accuracy": 0.25343158692121504,
      "num_tokens": 78476690.0,
      "step": 34230
    },
    {
      "entropy": 5.522705507278443,
      "epoch": 3.28866474543708,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003941381368615951,
      "loss": 4.8339,
      "mean_token_accuracy": 0.23683058619499206,
      "num_tokens": 78487157.0,
      "step": 34235
    },
    {
      "entropy": 5.479615545272827,
      "epoch": 3.2891450528338138,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003941090524584489,
      "loss": 4.7339,
      "mean_token_accuracy": 0.24558626115322113,
      "num_tokens": 78499034.0,
      "step": 34240
    },
    {
      "entropy": 5.411904239654541,
      "epoch": 3.2896253602305476,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003940799652899442,
      "loss": 4.6279,
      "mean_token_accuracy": 0.26100166887044907,
      "num_tokens": 78509720.0,
      "step": 34245
    },
    {
      "entropy": 5.446908473968506,
      "epoch": 3.2901056676272815,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003940508753567564,
      "loss": 4.6968,
      "mean_token_accuracy": 0.25637869983911515,
      "num_tokens": 78522126.0,
      "step": 34250
    },
    {
      "entropy": 5.491124248504638,
      "epoch": 3.2905859750240154,
      "grad_norm": 1.125,
      "learning_rate": 0.00039402178265956074,
      "loss": 4.7805,
      "mean_token_accuracy": 0.24319436103105546,
      "num_tokens": 78533348.0,
      "step": 34255
    },
    {
      "entropy": 5.458863162994385,
      "epoch": 3.2910662824207493,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003939926871990328,
      "loss": 4.6895,
      "mean_token_accuracy": 0.2524368852376938,
      "num_tokens": 78545889.0,
      "step": 34260
    },
    {
      "entropy": 5.478237056732178,
      "epoch": 3.291546589817483,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003939635889758478,
      "loss": 4.7547,
      "mean_token_accuracy": 0.24658737927675248,
      "num_tokens": 78557429.0,
      "step": 34265
    },
    {
      "entropy": 5.4000804901123045,
      "epoch": 3.292026897214217,
      "grad_norm": 1.125,
      "learning_rate": 0.00039393448799068164,
      "loss": 4.6309,
      "mean_token_accuracy": 0.2621041342616081,
      "num_tokens": 78567824.0,
      "step": 34270
    },
    {
      "entropy": 5.4987852573394775,
      "epoch": 3.292507204610951,
      "grad_norm": 1.25,
      "learning_rate": 0.0003939053842442098,
      "loss": 4.7576,
      "mean_token_accuracy": 0.2500441253185272,
      "num_tokens": 78578229.0,
      "step": 34275
    },
    {
      "entropy": 5.466667890548706,
      "epoch": 3.292987512007685,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039387627773710803,
      "loss": 4.728,
      "mean_token_accuracy": 0.2515834540128708,
      "num_tokens": 78591619.0,
      "step": 34280
    },
    {
      "entropy": 5.469660806655884,
      "epoch": 3.2934678194044187,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039384716847005216,
      "loss": 4.6854,
      "mean_token_accuracy": 0.25149508416652677,
      "num_tokens": 78602576.0,
      "step": 34285
    },
    {
      "entropy": 5.394841194152832,
      "epoch": 3.2939481268011526,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039381805644371774,
      "loss": 4.6777,
      "mean_token_accuracy": 0.26274228394031524,
      "num_tokens": 78613532.0,
      "step": 34290
    },
    {
      "entropy": 5.451415252685547,
      "epoch": 3.2944284341978864,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000393788941658781,
      "loss": 4.7377,
      "mean_token_accuracy": 0.24960678219795226,
      "num_tokens": 78626120.0,
      "step": 34295
    },
    {
      "entropy": 5.441372442245483,
      "epoch": 3.2949087415946208,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039375982411591774,
      "loss": 4.7073,
      "mean_token_accuracy": 0.25311149507761,
      "num_tokens": 78636175.0,
      "step": 34300
    },
    {
      "entropy": 5.519496440887451,
      "epoch": 3.2953890489913547,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039373070381580404,
      "loss": 4.7331,
      "mean_token_accuracy": 0.25420559495687484,
      "num_tokens": 78647429.0,
      "step": 34305
    },
    {
      "entropy": 5.458052730560302,
      "epoch": 3.2958693563880885,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003937015807591159,
      "loss": 4.7672,
      "mean_token_accuracy": 0.24511714279651642,
      "num_tokens": 78659792.0,
      "step": 34310
    },
    {
      "entropy": 5.434053373336792,
      "epoch": 3.2963496637848224,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039367245494652963,
      "loss": 4.6508,
      "mean_token_accuracy": 0.25855632722377775,
      "num_tokens": 78670614.0,
      "step": 34315
    },
    {
      "entropy": 5.452944374084472,
      "epoch": 3.2968299711815563,
      "grad_norm": 1.25,
      "learning_rate": 0.00039364332637872125,
      "loss": 4.6574,
      "mean_token_accuracy": 0.2551427572965622,
      "num_tokens": 78681475.0,
      "step": 34320
    },
    {
      "entropy": 5.412503480911255,
      "epoch": 3.29731027857829,
      "grad_norm": 1.125,
      "learning_rate": 0.00039361419505636714,
      "loss": 4.6769,
      "mean_token_accuracy": 0.25362746268510816,
      "num_tokens": 78692251.0,
      "step": 34325
    },
    {
      "entropy": 5.43065128326416,
      "epoch": 3.297790585975024,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039358506098014363,
      "loss": 4.694,
      "mean_token_accuracy": 0.2519388094544411,
      "num_tokens": 78704584.0,
      "step": 34330
    },
    {
      "entropy": 5.447978401184082,
      "epoch": 3.298270893371758,
      "grad_norm": 1.125,
      "learning_rate": 0.00039355592415072716,
      "loss": 4.6477,
      "mean_token_accuracy": 0.2506603091955185,
      "num_tokens": 78715467.0,
      "step": 34335
    },
    {
      "entropy": 5.412213754653931,
      "epoch": 3.298751200768492,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039352678456879415,
      "loss": 4.6578,
      "mean_token_accuracy": 0.2594792157411575,
      "num_tokens": 78725835.0,
      "step": 34340
    },
    {
      "entropy": 5.393380546569825,
      "epoch": 3.2992315081652257,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003934976422350212,
      "loss": 4.6352,
      "mean_token_accuracy": 0.25850801914930344,
      "num_tokens": 78737143.0,
      "step": 34345
    },
    {
      "entropy": 5.569334030151367,
      "epoch": 3.2997118155619596,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003934684971500848,
      "loss": 4.8599,
      "mean_token_accuracy": 0.2421200841665268,
      "num_tokens": 78748641.0,
      "step": 34350
    },
    {
      "entropy": 5.4546655178070065,
      "epoch": 3.3001921229586935,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039343934931466165,
      "loss": 4.6821,
      "mean_token_accuracy": 0.24861938506364822,
      "num_tokens": 78760200.0,
      "step": 34355
    },
    {
      "entropy": 5.389844274520874,
      "epoch": 3.3006724303554273,
      "grad_norm": 1.125,
      "learning_rate": 0.00039341019872942855,
      "loss": 4.6756,
      "mean_token_accuracy": 0.25381863564252855,
      "num_tokens": 78771347.0,
      "step": 34360
    },
    {
      "entropy": 5.350956773757934,
      "epoch": 3.3011527377521612,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039338104539506227,
      "loss": 4.6011,
      "mean_token_accuracy": 0.25799970626831054,
      "num_tokens": 78783514.0,
      "step": 34365
    },
    {
      "entropy": 5.478760147094727,
      "epoch": 3.301633045148895,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003933518893122396,
      "loss": 4.7554,
      "mean_token_accuracy": 0.24441684335470198,
      "num_tokens": 78794621.0,
      "step": 34370
    },
    {
      "entropy": 5.4779926300048825,
      "epoch": 3.3021133525456294,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003933227304816375,
      "loss": 4.7054,
      "mean_token_accuracy": 0.25459485203027726,
      "num_tokens": 78806293.0,
      "step": 34375
    },
    {
      "entropy": 5.513264989852905,
      "epoch": 3.302593659942363,
      "grad_norm": 1.109375,
      "learning_rate": 0.000393293568903933,
      "loss": 4.7469,
      "mean_token_accuracy": 0.24231422543525696,
      "num_tokens": 78816930.0,
      "step": 34380
    },
    {
      "entropy": 5.453431081771851,
      "epoch": 3.303073967339097,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003932644045798029,
      "loss": 4.7714,
      "mean_token_accuracy": 0.2421731486916542,
      "num_tokens": 78828321.0,
      "step": 34385
    },
    {
      "entropy": 5.460171031951904,
      "epoch": 3.303554274735831,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003932352375099247,
      "loss": 4.7787,
      "mean_token_accuracy": 0.24337219893932344,
      "num_tokens": 78840967.0,
      "step": 34390
    },
    {
      "entropy": 5.423111343383789,
      "epoch": 3.304034582132565,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003932060676949753,
      "loss": 4.6949,
      "mean_token_accuracy": 0.24898483455181122,
      "num_tokens": 78853037.0,
      "step": 34395
    },
    {
      "entropy": 5.500659418106079,
      "epoch": 3.304514889529299,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003931768951356319,
      "loss": 4.7426,
      "mean_token_accuracy": 0.24745261520147324,
      "num_tokens": 78863441.0,
      "step": 34400
    },
    {
      "entropy": 5.356772804260254,
      "epoch": 3.3049951969260327,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003931477198325721,
      "loss": 4.622,
      "mean_token_accuracy": 0.2601510226726532,
      "num_tokens": 78875770.0,
      "step": 34405
    },
    {
      "entropy": 5.458569478988648,
      "epoch": 3.3054755043227666,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000393118541786473,
      "loss": 4.6992,
      "mean_token_accuracy": 0.2563567027449608,
      "num_tokens": 78886239.0,
      "step": 34410
    },
    {
      "entropy": 5.445851039886475,
      "epoch": 3.3059558117195005,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039308936099801203,
      "loss": 4.7335,
      "mean_token_accuracy": 0.24258170276880264,
      "num_tokens": 78897497.0,
      "step": 34415
    },
    {
      "entropy": 5.48602819442749,
      "epoch": 3.3064361191162344,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003930601774678669,
      "loss": 4.7325,
      "mean_token_accuracy": 0.24394038021564485,
      "num_tokens": 78908643.0,
      "step": 34420
    },
    {
      "entropy": 5.482816553115844,
      "epoch": 3.3069164265129682,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039303099119671487,
      "loss": 4.7935,
      "mean_token_accuracy": 0.2410024657845497,
      "num_tokens": 78921240.0,
      "step": 34425
    },
    {
      "entropy": 5.439303064346314,
      "epoch": 3.307396733909702,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039300180218523374,
      "loss": 4.7183,
      "mean_token_accuracy": 0.2556604340672493,
      "num_tokens": 78933803.0,
      "step": 34430
    },
    {
      "entropy": 5.540589714050293,
      "epoch": 3.307877041306436,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003929726104341013,
      "loss": 4.8122,
      "mean_token_accuracy": 0.24526645690202714,
      "num_tokens": 78944876.0,
      "step": 34435
    },
    {
      "entropy": 5.42168288230896,
      "epoch": 3.30835734870317,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039294341594399494,
      "loss": 4.6498,
      "mean_token_accuracy": 0.25822554975748063,
      "num_tokens": 78957044.0,
      "step": 34440
    },
    {
      "entropy": 5.373810434341431,
      "epoch": 3.3088376560999038,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039291421871559274,
      "loss": 4.6606,
      "mean_token_accuracy": 0.2579470112919807,
      "num_tokens": 78968654.0,
      "step": 34445
    },
    {
      "entropy": 5.420226812362671,
      "epoch": 3.309317963496638,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039288501874957263,
      "loss": 4.7227,
      "mean_token_accuracy": 0.2519470855593681,
      "num_tokens": 78980201.0,
      "step": 34450
    },
    {
      "entropy": 5.491055345535278,
      "epoch": 3.3097982708933715,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003928558160466123,
      "loss": 4.7603,
      "mean_token_accuracy": 0.24161802977323532,
      "num_tokens": 78991172.0,
      "step": 34455
    },
    {
      "entropy": 5.4881298542022705,
      "epoch": 3.310278578290106,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003928266106073899,
      "loss": 4.6992,
      "mean_token_accuracy": 0.24989808797836305,
      "num_tokens": 79001944.0,
      "step": 34460
    },
    {
      "entropy": 5.453688716888427,
      "epoch": 3.3107588856868397,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039279740243258353,
      "loss": 4.7279,
      "mean_token_accuracy": 0.2506378084421158,
      "num_tokens": 79014715.0,
      "step": 34465
    },
    {
      "entropy": 5.410131120681763,
      "epoch": 3.3112391930835736,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003927681915228712,
      "loss": 4.649,
      "mean_token_accuracy": 0.25669342577457427,
      "num_tokens": 79025499.0,
      "step": 34470
    },
    {
      "entropy": 5.398714685440064,
      "epoch": 3.3117195004803075,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003927389778789312,
      "loss": 4.739,
      "mean_token_accuracy": 0.24904851913452147,
      "num_tokens": 79037400.0,
      "step": 34475
    },
    {
      "entropy": 5.452772855758667,
      "epoch": 3.3121998078770414,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003927097615014418,
      "loss": 4.7209,
      "mean_token_accuracy": 0.2515050247311592,
      "num_tokens": 79048450.0,
      "step": 34480
    },
    {
      "entropy": 5.470528793334961,
      "epoch": 3.3126801152737753,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003926805423910811,
      "loss": 4.7484,
      "mean_token_accuracy": 0.254166966676712,
      "num_tokens": 79061579.0,
      "step": 34485
    },
    {
      "entropy": 5.42203049659729,
      "epoch": 3.313160422670509,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003926513205485278,
      "loss": 4.6755,
      "mean_token_accuracy": 0.24527239352464675,
      "num_tokens": 79073106.0,
      "step": 34490
    },
    {
      "entropy": 5.457114362716675,
      "epoch": 3.313640730067243,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003926220959744602,
      "loss": 4.7658,
      "mean_token_accuracy": 0.24519893229007722,
      "num_tokens": 79085528.0,
      "step": 34495
    },
    {
      "entropy": 5.400688314437867,
      "epoch": 3.314121037463977,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003925928686695567,
      "loss": 4.6133,
      "mean_token_accuracy": 0.26209972351789473,
      "num_tokens": 79096429.0,
      "step": 34500
    },
    {
      "entropy": 5.477209997177124,
      "epoch": 3.314601344860711,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000392563638634496,
      "loss": 4.7585,
      "mean_token_accuracy": 0.24704079180955887,
      "num_tokens": 79107688.0,
      "step": 34505
    },
    {
      "entropy": 5.447778129577637,
      "epoch": 3.3150816522574447,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003925344058699567,
      "loss": 4.6873,
      "mean_token_accuracy": 0.25391780138015746,
      "num_tokens": 79119549.0,
      "step": 34510
    },
    {
      "entropy": 5.391026830673217,
      "epoch": 3.3155619596541785,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003925051703766175,
      "loss": 4.6658,
      "mean_token_accuracy": 0.25586340129375457,
      "num_tokens": 79131193.0,
      "step": 34515
    },
    {
      "entropy": 5.437295150756836,
      "epoch": 3.3160422670509124,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003924759321551572,
      "loss": 4.7444,
      "mean_token_accuracy": 0.2428869366645813,
      "num_tokens": 79142117.0,
      "step": 34520
    },
    {
      "entropy": 5.469495058059692,
      "epoch": 3.3165225744476463,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003924466912062546,
      "loss": 4.7126,
      "mean_token_accuracy": 0.24899600446224213,
      "num_tokens": 79154734.0,
      "step": 34525
    },
    {
      "entropy": 5.476527166366577,
      "epoch": 3.31700288184438,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003924174475305885,
      "loss": 4.7466,
      "mean_token_accuracy": 0.25035484731197355,
      "num_tokens": 79166077.0,
      "step": 34530
    },
    {
      "entropy": 5.489509057998657,
      "epoch": 3.3174831892411145,
      "grad_norm": 1.25,
      "learning_rate": 0.00039238820112883795,
      "loss": 4.729,
      "mean_token_accuracy": 0.24258051365613936,
      "num_tokens": 79177544.0,
      "step": 34535
    },
    {
      "entropy": 5.422629165649414,
      "epoch": 3.3179634966378484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003923589520016819,
      "loss": 4.6924,
      "mean_token_accuracy": 0.25530115216970445,
      "num_tokens": 79189606.0,
      "step": 34540
    },
    {
      "entropy": 5.371073007583618,
      "epoch": 3.3184438040345823,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039232970014979965,
      "loss": 4.6251,
      "mean_token_accuracy": 0.2560166552662849,
      "num_tokens": 79201385.0,
      "step": 34545
    },
    {
      "entropy": 5.458832454681397,
      "epoch": 3.318924111431316,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003923004455738699,
      "loss": 4.7742,
      "mean_token_accuracy": 0.24098608195781707,
      "num_tokens": 79213481.0,
      "step": 34550
    },
    {
      "entropy": 5.4243370532989506,
      "epoch": 3.31940441882805,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039227118827457234,
      "loss": 4.6667,
      "mean_token_accuracy": 0.24903956353664397,
      "num_tokens": 79226369.0,
      "step": 34555
    },
    {
      "entropy": 5.488967657089233,
      "epoch": 3.319884726224784,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039224192825258584,
      "loss": 4.778,
      "mean_token_accuracy": 0.2453838810324669,
      "num_tokens": 79238236.0,
      "step": 34560
    },
    {
      "entropy": 5.527775716781616,
      "epoch": 3.320365033621518,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039221266550859004,
      "loss": 4.7288,
      "mean_token_accuracy": 0.24717819839715957,
      "num_tokens": 79248536.0,
      "step": 34565
    },
    {
      "entropy": 5.436396169662475,
      "epoch": 3.3208453410182517,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039218340004326414,
      "loss": 4.6904,
      "mean_token_accuracy": 0.25087960809469223,
      "num_tokens": 79259273.0,
      "step": 34570
    },
    {
      "entropy": 5.304122686386108,
      "epoch": 3.3213256484149856,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003921541318572876,
      "loss": 4.6026,
      "mean_token_accuracy": 0.25680462419986727,
      "num_tokens": 79271986.0,
      "step": 34575
    },
    {
      "entropy": 5.417111158370972,
      "epoch": 3.3218059558117194,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039212486095134005,
      "loss": 4.6996,
      "mean_token_accuracy": 0.2479342982172966,
      "num_tokens": 79282602.0,
      "step": 34580
    },
    {
      "entropy": 5.443144655227661,
      "epoch": 3.3222862632084533,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003920955873261011,
      "loss": 4.7052,
      "mean_token_accuracy": 0.2555766418576241,
      "num_tokens": 79295034.0,
      "step": 34585
    },
    {
      "entropy": 5.424291610717773,
      "epoch": 3.322766570605187,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003920663109822502,
      "loss": 4.6987,
      "mean_token_accuracy": 0.25074367970228195,
      "num_tokens": 79307434.0,
      "step": 34590
    },
    {
      "entropy": 5.499506139755249,
      "epoch": 3.323246878001921,
      "grad_norm": 1.125,
      "learning_rate": 0.00039203703192046717,
      "loss": 4.7585,
      "mean_token_accuracy": 0.24713200479745864,
      "num_tokens": 79318185.0,
      "step": 34595
    },
    {
      "entropy": 5.346632289886474,
      "epoch": 3.323727185398655,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003920077501414318,
      "loss": 4.6375,
      "mean_token_accuracy": 0.2604114443063736,
      "num_tokens": 79329800.0,
      "step": 34600
    },
    {
      "entropy": 5.3849855899810795,
      "epoch": 3.324207492795389,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039197846564582395,
      "loss": 4.7254,
      "mean_token_accuracy": 0.24393080025911332,
      "num_tokens": 79342537.0,
      "step": 34605
    },
    {
      "entropy": 5.442325210571289,
      "epoch": 3.324687800192123,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039194917843432347,
      "loss": 4.7157,
      "mean_token_accuracy": 0.24928591698408126,
      "num_tokens": 79353727.0,
      "step": 34610
    },
    {
      "entropy": 5.412034225463867,
      "epoch": 3.325168107588857,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003919198885076103,
      "loss": 4.6481,
      "mean_token_accuracy": 0.26344154626131056,
      "num_tokens": 79364337.0,
      "step": 34615
    },
    {
      "entropy": 5.467574262619019,
      "epoch": 3.325648414985591,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039189059586636465,
      "loss": 4.7143,
      "mean_token_accuracy": 0.24707631319761275,
      "num_tokens": 79375890.0,
      "step": 34620
    },
    {
      "entropy": 5.463693571090698,
      "epoch": 3.326128722382325,
      "grad_norm": 1.25,
      "learning_rate": 0.0003918613005112663,
      "loss": 4.7173,
      "mean_token_accuracy": 0.24542176723480225,
      "num_tokens": 79386686.0,
      "step": 34625
    },
    {
      "entropy": 5.4296331882476805,
      "epoch": 3.3266090297790587,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003918320024429956,
      "loss": 4.7308,
      "mean_token_accuracy": 0.25033366978168486,
      "num_tokens": 79397852.0,
      "step": 34630
    },
    {
      "entropy": 5.418598365783692,
      "epoch": 3.3270893371757926,
      "grad_norm": 1.125,
      "learning_rate": 0.0003918027016622328,
      "loss": 4.6958,
      "mean_token_accuracy": 0.2552506148815155,
      "num_tokens": 79409547.0,
      "step": 34635
    },
    {
      "entropy": 5.339698219299317,
      "epoch": 3.3275696445725265,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003917733981696579,
      "loss": 4.6831,
      "mean_token_accuracy": 0.25935749858617785,
      "num_tokens": 79421906.0,
      "step": 34640
    },
    {
      "entropy": 5.421996307373047,
      "epoch": 3.3280499519692603,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003917440919659516,
      "loss": 4.6755,
      "mean_token_accuracy": 0.25561088770627977,
      "num_tokens": 79433767.0,
      "step": 34645
    },
    {
      "entropy": 5.501058340072632,
      "epoch": 3.3285302593659942,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003917147830517941,
      "loss": 4.7235,
      "mean_token_accuracy": 0.24425368756055832,
      "num_tokens": 79445488.0,
      "step": 34650
    },
    {
      "entropy": 5.47869462966919,
      "epoch": 3.329010566762728,
      "grad_norm": 1.125,
      "learning_rate": 0.0003916854714278659,
      "loss": 4.7044,
      "mean_token_accuracy": 0.25322427600622177,
      "num_tokens": 79457908.0,
      "step": 34655
    },
    {
      "entropy": 5.470390796661377,
      "epoch": 3.329490874159462,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003916561570948476,
      "loss": 4.7446,
      "mean_token_accuracy": 0.24571965783834457,
      "num_tokens": 79469876.0,
      "step": 34660
    },
    {
      "entropy": 5.407791328430176,
      "epoch": 3.329971181556196,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003916268400534197,
      "loss": 4.6364,
      "mean_token_accuracy": 0.2555285394191742,
      "num_tokens": 79480532.0,
      "step": 34665
    },
    {
      "entropy": 5.416766262054443,
      "epoch": 3.3304514889529298,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003915975203042628,
      "loss": 4.7101,
      "mean_token_accuracy": 0.2553961932659149,
      "num_tokens": 79491298.0,
      "step": 34670
    },
    {
      "entropy": 5.420481204986572,
      "epoch": 3.3309317963496636,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039156819784805783,
      "loss": 4.6907,
      "mean_token_accuracy": 0.24287602305412292,
      "num_tokens": 79502794.0,
      "step": 34675
    },
    {
      "entropy": 5.4109962463378904,
      "epoch": 3.3314121037463975,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003915388726854854,
      "loss": 4.6998,
      "mean_token_accuracy": 0.24830862730741501,
      "num_tokens": 79513104.0,
      "step": 34680
    },
    {
      "entropy": 5.5001733779907225,
      "epoch": 3.331892411143132,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039150954481722634,
      "loss": 4.7882,
      "mean_token_accuracy": 0.24315405637025833,
      "num_tokens": 79524660.0,
      "step": 34685
    },
    {
      "entropy": 5.496211576461792,
      "epoch": 3.3323727185398653,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003914802142439617,
      "loss": 4.7157,
      "mean_token_accuracy": 0.24753042310476303,
      "num_tokens": 79538027.0,
      "step": 34690
    },
    {
      "entropy": 5.424083709716797,
      "epoch": 3.3328530259365996,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003914508809663723,
      "loss": 4.6478,
      "mean_token_accuracy": 0.26089958101511,
      "num_tokens": 79548894.0,
      "step": 34695
    },
    {
      "entropy": 5.4437602996826175,
      "epoch": 3.3333333333333335,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039142154498513913,
      "loss": 4.7707,
      "mean_token_accuracy": 0.24599283039569855,
      "num_tokens": 79560520.0,
      "step": 34700
    },
    {
      "entropy": 5.507673740386963,
      "epoch": 3.3338136407300674,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039139220630094357,
      "loss": 4.7182,
      "mean_token_accuracy": 0.2535771012306213,
      "num_tokens": 79571440.0,
      "step": 34705
    },
    {
      "entropy": 5.477480411529541,
      "epoch": 3.3342939481268012,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039136286491446657,
      "loss": 4.7507,
      "mean_token_accuracy": 0.24317355155944825,
      "num_tokens": 79582046.0,
      "step": 34710
    },
    {
      "entropy": 5.421837043762207,
      "epoch": 3.334774255523535,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039133352082638923,
      "loss": 4.6898,
      "mean_token_accuracy": 0.24912893325090407,
      "num_tokens": 79593704.0,
      "step": 34715
    },
    {
      "entropy": 5.477133178710938,
      "epoch": 3.335254562920269,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039130417403739315,
      "loss": 4.7601,
      "mean_token_accuracy": 0.24939435571432114,
      "num_tokens": 79605783.0,
      "step": 34720
    },
    {
      "entropy": 5.437319135665893,
      "epoch": 3.335734870317003,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003912748245481594,
      "loss": 4.6377,
      "mean_token_accuracy": 0.2537251189351082,
      "num_tokens": 79617140.0,
      "step": 34725
    },
    {
      "entropy": 5.409595060348511,
      "epoch": 3.3362151777137368,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039124547235936947,
      "loss": 4.6985,
      "mean_token_accuracy": 0.24817233085632323,
      "num_tokens": 79627763.0,
      "step": 34730
    },
    {
      "entropy": 5.484588718414306,
      "epoch": 3.3366954851104706,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039121611747170495,
      "loss": 4.8107,
      "mean_token_accuracy": 0.24115159064531327,
      "num_tokens": 79640056.0,
      "step": 34735
    },
    {
      "entropy": 5.482151889801026,
      "epoch": 3.3371757925072045,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039118675988584724,
      "loss": 4.7288,
      "mean_token_accuracy": 0.24685075283050537,
      "num_tokens": 79652795.0,
      "step": 34740
    },
    {
      "entropy": 5.573814868927002,
      "epoch": 3.3376560999039384,
      "grad_norm": 1.125,
      "learning_rate": 0.0003911573996024779,
      "loss": 4.8632,
      "mean_token_accuracy": 0.23959876596927643,
      "num_tokens": 79663734.0,
      "step": 34745
    },
    {
      "entropy": 5.447584438323974,
      "epoch": 3.3381364073006723,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003911280366222787,
      "loss": 4.6453,
      "mean_token_accuracy": 0.25663392543792723,
      "num_tokens": 79675092.0,
      "step": 34750
    },
    {
      "entropy": 5.491541576385498,
      "epoch": 3.338616714697406,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039109867094593134,
      "loss": 4.7961,
      "mean_token_accuracy": 0.24251189827919006,
      "num_tokens": 79687126.0,
      "step": 34755
    },
    {
      "entropy": 5.359338855743408,
      "epoch": 3.3390970220941405,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003910693025741175,
      "loss": 4.6512,
      "mean_token_accuracy": 0.25626715570688247,
      "num_tokens": 79698511.0,
      "step": 34760
    },
    {
      "entropy": 5.467512750625611,
      "epoch": 3.339577329490874,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039103993150751916,
      "loss": 4.7319,
      "mean_token_accuracy": 0.25120809078216555,
      "num_tokens": 79709407.0,
      "step": 34765
    },
    {
      "entropy": 5.486828136444092,
      "epoch": 3.3400576368876083,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039101055774681825,
      "loss": 4.6958,
      "mean_token_accuracy": 0.25202953070402145,
      "num_tokens": 79720562.0,
      "step": 34770
    },
    {
      "entropy": 5.475890159606934,
      "epoch": 3.340537944284342,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003909811812926966,
      "loss": 4.7781,
      "mean_token_accuracy": 0.24072598665952682,
      "num_tokens": 79731687.0,
      "step": 34775
    },
    {
      "entropy": 5.460471487045288,
      "epoch": 3.341018251681076,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003909518021458363,
      "loss": 4.7417,
      "mean_token_accuracy": 0.25359519720077517,
      "num_tokens": 79742954.0,
      "step": 34780
    },
    {
      "entropy": 5.482081699371338,
      "epoch": 3.34149855907781,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003909224203069195,
      "loss": 4.7799,
      "mean_token_accuracy": 0.24031828492879867,
      "num_tokens": 79754233.0,
      "step": 34785
    },
    {
      "entropy": 5.446634101867676,
      "epoch": 3.341978866474544,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003908930357766283,
      "loss": 4.612,
      "mean_token_accuracy": 0.26319279819726943,
      "num_tokens": 79765348.0,
      "step": 34790
    },
    {
      "entropy": 5.4337952613830565,
      "epoch": 3.3424591738712777,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003908636485556449,
      "loss": 4.6929,
      "mean_token_accuracy": 0.2549081936478615,
      "num_tokens": 79776993.0,
      "step": 34795
    },
    {
      "entropy": 5.511778545379639,
      "epoch": 3.3429394812680115,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039083425864465165,
      "loss": 4.8248,
      "mean_token_accuracy": 0.24691056758165358,
      "num_tokens": 79787321.0,
      "step": 34800
    },
    {
      "entropy": 5.433262062072754,
      "epoch": 3.3434197886647454,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003908048660443309,
      "loss": 4.6864,
      "mean_token_accuracy": 0.2502284958958626,
      "num_tokens": 79798130.0,
      "step": 34805
    },
    {
      "entropy": 5.4443612575531,
      "epoch": 3.3439000960614793,
      "grad_norm": 1.125,
      "learning_rate": 0.0003907754707553651,
      "loss": 4.6504,
      "mean_token_accuracy": 0.2513242706656456,
      "num_tokens": 79809616.0,
      "step": 34810
    },
    {
      "entropy": 5.390579700469971,
      "epoch": 3.344380403458213,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003907460727784365,
      "loss": 4.6777,
      "mean_token_accuracy": 0.25259238481521606,
      "num_tokens": 79820355.0,
      "step": 34815
    },
    {
      "entropy": 5.4724681854248045,
      "epoch": 3.344860710854947,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039071667211422787,
      "loss": 4.7081,
      "mean_token_accuracy": 0.24902935028076173,
      "num_tokens": 79831733.0,
      "step": 34820
    },
    {
      "entropy": 5.47444863319397,
      "epoch": 3.345341018251681,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003906872687634217,
      "loss": 4.7483,
      "mean_token_accuracy": 0.2466981053352356,
      "num_tokens": 79842961.0,
      "step": 34825
    },
    {
      "entropy": 5.4285722255706785,
      "epoch": 3.345821325648415,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039065786272670066,
      "loss": 4.7433,
      "mean_token_accuracy": 0.2522162064909935,
      "num_tokens": 79855142.0,
      "step": 34830
    },
    {
      "entropy": 5.405714988708496,
      "epoch": 3.3463016330451487,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003906284540047475,
      "loss": 4.6451,
      "mean_token_accuracy": 0.25556202232837677,
      "num_tokens": 79865973.0,
      "step": 34835
    },
    {
      "entropy": 5.530129766464233,
      "epoch": 3.3467819404418826,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039059904259824507,
      "loss": 4.7249,
      "mean_token_accuracy": 0.24524388909339906,
      "num_tokens": 79877424.0,
      "step": 34840
    },
    {
      "entropy": 5.519464445114136,
      "epoch": 3.347262247838617,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000390569628507876,
      "loss": 4.7517,
      "mean_token_accuracy": 0.24286214560270308,
      "num_tokens": 79889956.0,
      "step": 34845
    },
    {
      "entropy": 5.423182392120362,
      "epoch": 3.347742555235351,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039054021173432336,
      "loss": 4.7021,
      "mean_token_accuracy": 0.2522093325853348,
      "num_tokens": 79902454.0,
      "step": 34850
    },
    {
      "entropy": 5.441210556030273,
      "epoch": 3.3482228626320847,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003905107922782701,
      "loss": 4.7507,
      "mean_token_accuracy": 0.24828798919916154,
      "num_tokens": 79914425.0,
      "step": 34855
    },
    {
      "entropy": 5.394536733627319,
      "epoch": 3.3487031700288186,
      "grad_norm": 1.125,
      "learning_rate": 0.0003904813701403993,
      "loss": 4.6456,
      "mean_token_accuracy": 0.26126213520765307,
      "num_tokens": 79926225.0,
      "step": 34860
    },
    {
      "entropy": 5.3941840648651125,
      "epoch": 3.3491834774255524,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039045194532139396,
      "loss": 4.6574,
      "mean_token_accuracy": 0.25668366253376007,
      "num_tokens": 79937804.0,
      "step": 34865
    },
    {
      "entropy": 5.47864670753479,
      "epoch": 3.3496637848222863,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003904225178219372,
      "loss": 4.6989,
      "mean_token_accuracy": 0.252862748503685,
      "num_tokens": 79949588.0,
      "step": 34870
    },
    {
      "entropy": 5.42878794670105,
      "epoch": 3.35014409221902,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039039308764271237,
      "loss": 4.6704,
      "mean_token_accuracy": 0.25188447535037994,
      "num_tokens": 79960243.0,
      "step": 34875
    },
    {
      "entropy": 5.385711669921875,
      "epoch": 3.350624399615754,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003903636547844026,
      "loss": 4.7021,
      "mean_token_accuracy": 0.2539800554513931,
      "num_tokens": 79971780.0,
      "step": 34880
    },
    {
      "entropy": 5.423802757263184,
      "epoch": 3.351104707012488,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039033421924769145,
      "loss": 4.7126,
      "mean_token_accuracy": 0.248578043282032,
      "num_tokens": 79982105.0,
      "step": 34885
    },
    {
      "entropy": 5.44408392906189,
      "epoch": 3.351585014409222,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039030478103326216,
      "loss": 4.6733,
      "mean_token_accuracy": 0.25946062207221987,
      "num_tokens": 79993456.0,
      "step": 34890
    },
    {
      "entropy": 5.529705762863159,
      "epoch": 3.3520653218059557,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039027534014179823,
      "loss": 4.7725,
      "mean_token_accuracy": 0.24057641625404358,
      "num_tokens": 80006335.0,
      "step": 34895
    },
    {
      "entropy": 5.420498275756836,
      "epoch": 3.3525456292026896,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003902458965739832,
      "loss": 4.7016,
      "mean_token_accuracy": 0.2540232926607132,
      "num_tokens": 80017574.0,
      "step": 34900
    },
    {
      "entropy": 5.435838413238526,
      "epoch": 3.3530259365994235,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003902164503305006,
      "loss": 4.712,
      "mean_token_accuracy": 0.2362843707203865,
      "num_tokens": 80029470.0,
      "step": 34905
    },
    {
      "entropy": 5.426451110839844,
      "epoch": 3.3535062439961574,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003901870014120343,
      "loss": 4.6816,
      "mean_token_accuracy": 0.2552762612700462,
      "num_tokens": 80040599.0,
      "step": 34910
    },
    {
      "entropy": 5.475981903076172,
      "epoch": 3.3539865513928913,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003901575498192678,
      "loss": 4.7214,
      "mean_token_accuracy": 0.25050544142723086,
      "num_tokens": 80052597.0,
      "step": 34915
    },
    {
      "entropy": 5.482787418365478,
      "epoch": 3.3544668587896256,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003901280955528849,
      "loss": 4.7454,
      "mean_token_accuracy": 0.23988714963197708,
      "num_tokens": 80063208.0,
      "step": 34920
    },
    {
      "entropy": 5.455665063858032,
      "epoch": 3.3549471661863595,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003900986386135695,
      "loss": 4.7324,
      "mean_token_accuracy": 0.25143360942602155,
      "num_tokens": 80074046.0,
      "step": 34925
    },
    {
      "entropy": 5.4629011154174805,
      "epoch": 3.3554274735830933,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039006917900200543,
      "loss": 4.7784,
      "mean_token_accuracy": 0.24790282398462296,
      "num_tokens": 80086054.0,
      "step": 34930
    },
    {
      "entropy": 5.526428985595703,
      "epoch": 3.3559077809798272,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039003971671887675,
      "loss": 4.6794,
      "mean_token_accuracy": 0.2514375612139702,
      "num_tokens": 80096030.0,
      "step": 34935
    },
    {
      "entropy": 5.359608125686646,
      "epoch": 3.356388088376561,
      "grad_norm": 1.125,
      "learning_rate": 0.0003900102517648674,
      "loss": 4.5665,
      "mean_token_accuracy": 0.260147912800312,
      "num_tokens": 80106412.0,
      "step": 34940
    },
    {
      "entropy": 5.45989761352539,
      "epoch": 3.356868395773295,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003899807841406617,
      "loss": 4.7248,
      "mean_token_accuracy": 0.2517839834094048,
      "num_tokens": 80116580.0,
      "step": 34945
    },
    {
      "entropy": 5.4579047679901125,
      "epoch": 3.357348703170029,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003899513138469434,
      "loss": 4.7287,
      "mean_token_accuracy": 0.24973418265581132,
      "num_tokens": 80129055.0,
      "step": 34950
    },
    {
      "entropy": 5.460215139389038,
      "epoch": 3.3578290105667628,
      "grad_norm": 1.09375,
      "learning_rate": 0.000389921840884397,
      "loss": 4.7839,
      "mean_token_accuracy": 0.24403431862592698,
      "num_tokens": 80140626.0,
      "step": 34955
    },
    {
      "entropy": 5.52929458618164,
      "epoch": 3.3583093179634966,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038989236525370676,
      "loss": 4.7612,
      "mean_token_accuracy": 0.2490891233086586,
      "num_tokens": 80151896.0,
      "step": 34960
    },
    {
      "entropy": 5.4729715347290036,
      "epoch": 3.3587896253602305,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003898628869555569,
      "loss": 4.683,
      "mean_token_accuracy": 0.2578965499997139,
      "num_tokens": 80162801.0,
      "step": 34965
    },
    {
      "entropy": 5.454912185668945,
      "epoch": 3.3592699327569644,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038983340599063187,
      "loss": 4.6748,
      "mean_token_accuracy": 0.253310626745224,
      "num_tokens": 80174998.0,
      "step": 34970
    },
    {
      "entropy": 5.546263837814331,
      "epoch": 3.3597502401536983,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003898039223596162,
      "loss": 4.8368,
      "mean_token_accuracy": 0.24416219294071198,
      "num_tokens": 80188092.0,
      "step": 34975
    },
    {
      "entropy": 5.4933130741119385,
      "epoch": 3.360230547550432,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003897744360631943,
      "loss": 4.7194,
      "mean_token_accuracy": 0.24917811155319214,
      "num_tokens": 80199749.0,
      "step": 34980
    },
    {
      "entropy": 5.401525545120239,
      "epoch": 3.360710854947166,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038974494710205084,
      "loss": 4.6763,
      "mean_token_accuracy": 0.25481424331665037,
      "num_tokens": 80211714.0,
      "step": 34985
    },
    {
      "entropy": 5.418367195129394,
      "epoch": 3.3611911623439,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038971545547687036,
      "loss": 4.7068,
      "mean_token_accuracy": 0.2541360169649124,
      "num_tokens": 80222258.0,
      "step": 34990
    },
    {
      "entropy": 5.3327117443084715,
      "epoch": 3.3616714697406342,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038968596118833766,
      "loss": 4.6749,
      "mean_token_accuracy": 0.25465797632932663,
      "num_tokens": 80234532.0,
      "step": 34995
    },
    {
      "entropy": 5.458303546905517,
      "epoch": 3.3621517771373677,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038965646423713744,
      "loss": 4.6516,
      "mean_token_accuracy": 0.2511284977197647,
      "num_tokens": 80245826.0,
      "step": 35000
    },
    {
      "entropy": 5.453481149673462,
      "epoch": 3.362632084534102,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038962696462395473,
      "loss": 4.6488,
      "mean_token_accuracy": 0.2509426474571228,
      "num_tokens": 80256715.0,
      "step": 35005
    },
    {
      "entropy": 5.4430468559265135,
      "epoch": 3.363112391930836,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003895974623494742,
      "loss": 4.7221,
      "mean_token_accuracy": 0.2538894176483154,
      "num_tokens": 80267713.0,
      "step": 35010
    },
    {
      "entropy": 5.4650195121765135,
      "epoch": 3.3635926993275698,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038956795741438085,
      "loss": 4.7486,
      "mean_token_accuracy": 0.24982526153326035,
      "num_tokens": 80278935.0,
      "step": 35015
    },
    {
      "entropy": 5.394759654998779,
      "epoch": 3.3640730067243036,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038953844981935975,
      "loss": 4.6062,
      "mean_token_accuracy": 0.25909767746925355,
      "num_tokens": 80289735.0,
      "step": 35020
    },
    {
      "entropy": 5.421020936965943,
      "epoch": 3.3645533141210375,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038950893956509597,
      "loss": 4.7611,
      "mean_token_accuracy": 0.2508338883519173,
      "num_tokens": 80301617.0,
      "step": 35025
    },
    {
      "entropy": 5.358046436309815,
      "epoch": 3.3650336215177714,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003894794266522746,
      "loss": 4.5992,
      "mean_token_accuracy": 0.2614386886358261,
      "num_tokens": 80312996.0,
      "step": 35030
    },
    {
      "entropy": 5.434061574935913,
      "epoch": 3.3655139289145053,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038944991108158094,
      "loss": 4.6707,
      "mean_token_accuracy": 0.25243222564458845,
      "num_tokens": 80324435.0,
      "step": 35035
    },
    {
      "entropy": 5.423876476287842,
      "epoch": 3.365994236311239,
      "grad_norm": 1.125,
      "learning_rate": 0.0003894203928537001,
      "loss": 4.7249,
      "mean_token_accuracy": 0.2534633755683899,
      "num_tokens": 80335922.0,
      "step": 35040
    },
    {
      "entropy": 5.430365371704101,
      "epoch": 3.366474543707973,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038939087196931754,
      "loss": 4.7202,
      "mean_token_accuracy": 0.25512915700674055,
      "num_tokens": 80348405.0,
      "step": 35045
    },
    {
      "entropy": 5.438399028778076,
      "epoch": 3.366954851104707,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038936134842911863,
      "loss": 4.698,
      "mean_token_accuracy": 0.25191017985343933,
      "num_tokens": 80361387.0,
      "step": 35050
    },
    {
      "entropy": 5.4067878246307375,
      "epoch": 3.367435158501441,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003893318222337888,
      "loss": 4.7371,
      "mean_token_accuracy": 0.25058156698942186,
      "num_tokens": 80372500.0,
      "step": 35055
    },
    {
      "entropy": 5.53137059211731,
      "epoch": 3.3679154658981747,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038930229338401354,
      "loss": 4.8154,
      "mean_token_accuracy": 0.24741076827049255,
      "num_tokens": 80384311.0,
      "step": 35060
    },
    {
      "entropy": 5.4676916122436525,
      "epoch": 3.3683957732949086,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003892727618804783,
      "loss": 4.6889,
      "mean_token_accuracy": 0.24851370900869368,
      "num_tokens": 80396259.0,
      "step": 35065
    },
    {
      "entropy": 5.4317771911621096,
      "epoch": 3.3688760806916425,
      "grad_norm": 1.28125,
      "learning_rate": 0.000389243227723869,
      "loss": 4.7088,
      "mean_token_accuracy": 0.2503512978553772,
      "num_tokens": 80407420.0,
      "step": 35070
    },
    {
      "entropy": 5.381503486633301,
      "epoch": 3.3693563880883763,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003892136909148711,
      "loss": 4.6477,
      "mean_token_accuracy": 0.2512021154165268,
      "num_tokens": 80417130.0,
      "step": 35075
    },
    {
      "entropy": 5.47384352684021,
      "epoch": 3.3698366954851107,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003891841514541706,
      "loss": 4.6992,
      "mean_token_accuracy": 0.2564546033740044,
      "num_tokens": 80429405.0,
      "step": 35080
    },
    {
      "entropy": 5.4365949630737305,
      "epoch": 3.3703170028818445,
      "grad_norm": 1.328125,
      "learning_rate": 0.000389154609342453,
      "loss": 4.7045,
      "mean_token_accuracy": 0.24882621318101883,
      "num_tokens": 80439895.0,
      "step": 35085
    },
    {
      "entropy": 5.368251132965088,
      "epoch": 3.3707973102785784,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003891250645804044,
      "loss": 4.6671,
      "mean_token_accuracy": 0.2569687247276306,
      "num_tokens": 80451884.0,
      "step": 35090
    },
    {
      "entropy": 5.488925981521606,
      "epoch": 3.3712776176753123,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038909551716871074,
      "loss": 4.72,
      "mean_token_accuracy": 0.25182813704013823,
      "num_tokens": 80462629.0,
      "step": 35095
    },
    {
      "entropy": 5.483624076843261,
      "epoch": 3.371757925072046,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003890659671080579,
      "loss": 4.738,
      "mean_token_accuracy": 0.246837118268013,
      "num_tokens": 80472066.0,
      "step": 35100
    },
    {
      "entropy": 5.4299157619476315,
      "epoch": 3.37223823246878,
      "grad_norm": 1.078125,
      "learning_rate": 0.000389036414399132,
      "loss": 4.6686,
      "mean_token_accuracy": 0.25806080400943754,
      "num_tokens": 80483687.0,
      "step": 35105
    },
    {
      "entropy": 5.426649379730224,
      "epoch": 3.372718539865514,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003890068590426191,
      "loss": 4.6996,
      "mean_token_accuracy": 0.25084578096866605,
      "num_tokens": 80493323.0,
      "step": 35110
    },
    {
      "entropy": 5.454758262634277,
      "epoch": 3.373198847262248,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003889773010392056,
      "loss": 4.6979,
      "mean_token_accuracy": 0.2568440869450569,
      "num_tokens": 80504454.0,
      "step": 35115
    },
    {
      "entropy": 5.412213373184204,
      "epoch": 3.3736791546589817,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038894774038957756,
      "loss": 4.6148,
      "mean_token_accuracy": 0.26067000180482863,
      "num_tokens": 80516201.0,
      "step": 35120
    },
    {
      "entropy": 5.382900333404541,
      "epoch": 3.3741594620557156,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038891817709442135,
      "loss": 4.723,
      "mean_token_accuracy": 0.24845882058143615,
      "num_tokens": 80528319.0,
      "step": 35125
    },
    {
      "entropy": 5.400237512588501,
      "epoch": 3.3746397694524495,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038888861115442334,
      "loss": 4.6482,
      "mean_token_accuracy": 0.2601949214935303,
      "num_tokens": 80539684.0,
      "step": 35130
    },
    {
      "entropy": 5.406467151641846,
      "epoch": 3.3751200768491834,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003888590425702699,
      "loss": 4.6591,
      "mean_token_accuracy": 0.25962394326925275,
      "num_tokens": 80550971.0,
      "step": 35135
    },
    {
      "entropy": 5.415207147598267,
      "epoch": 3.3756003842459172,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003888294713426477,
      "loss": 4.6889,
      "mean_token_accuracy": 0.2548618331551552,
      "num_tokens": 80562931.0,
      "step": 35140
    },
    {
      "entropy": 5.400937557220459,
      "epoch": 3.376080691642651,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038879989747224317,
      "loss": 4.6703,
      "mean_token_accuracy": 0.2633875533938408,
      "num_tokens": 80573899.0,
      "step": 35145
    },
    {
      "entropy": 5.4318304538726805,
      "epoch": 3.376560999039385,
      "grad_norm": 1.125,
      "learning_rate": 0.0003887703209597428,
      "loss": 4.6773,
      "mean_token_accuracy": 0.249067060649395,
      "num_tokens": 80585453.0,
      "step": 35150
    },
    {
      "entropy": 5.424689292907715,
      "epoch": 3.3770413064361193,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003887407418058335,
      "loss": 4.7204,
      "mean_token_accuracy": 0.2540655881166458,
      "num_tokens": 80597911.0,
      "step": 35155
    },
    {
      "entropy": 5.464784383773804,
      "epoch": 3.377521613832853,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038871116001120196,
      "loss": 4.6929,
      "mean_token_accuracy": 0.2513432398438454,
      "num_tokens": 80609350.0,
      "step": 35160
    },
    {
      "entropy": 5.435538196563721,
      "epoch": 3.378001921229587,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003886815755765348,
      "loss": 4.7483,
      "mean_token_accuracy": 0.2469846338033676,
      "num_tokens": 80620710.0,
      "step": 35165
    },
    {
      "entropy": 5.4656895160675045,
      "epoch": 3.378482228626321,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003886519885025191,
      "loss": 4.8208,
      "mean_token_accuracy": 0.23917608857154846,
      "num_tokens": 80633031.0,
      "step": 35170
    },
    {
      "entropy": 5.399071264266968,
      "epoch": 3.378962536023055,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038862239878984173,
      "loss": 4.614,
      "mean_token_accuracy": 0.25785605013370516,
      "num_tokens": 80644047.0,
      "step": 35175
    },
    {
      "entropy": 5.411724519729614,
      "epoch": 3.3794428434197887,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003885928064391897,
      "loss": 4.7022,
      "mean_token_accuracy": 0.25269855707883837,
      "num_tokens": 80655769.0,
      "step": 35180
    },
    {
      "entropy": 5.464472389221191,
      "epoch": 3.3799231508165226,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003885632114512499,
      "loss": 4.7557,
      "mean_token_accuracy": 0.2443406879901886,
      "num_tokens": 80667881.0,
      "step": 35185
    },
    {
      "entropy": 5.461730575561523,
      "epoch": 3.3804034582132565,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038853361382670956,
      "loss": 4.7722,
      "mean_token_accuracy": 0.24470954537391662,
      "num_tokens": 80679021.0,
      "step": 35190
    },
    {
      "entropy": 5.411584806442261,
      "epoch": 3.3808837656099904,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038850401356625583,
      "loss": 4.7352,
      "mean_token_accuracy": 0.24643190801143647,
      "num_tokens": 80689366.0,
      "step": 35195
    },
    {
      "entropy": 5.357751321792603,
      "epoch": 3.3813640730067243,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003884744106705759,
      "loss": 4.6368,
      "mean_token_accuracy": 0.2490009143948555,
      "num_tokens": 80700755.0,
      "step": 35200
    },
    {
      "entropy": 5.46356348991394,
      "epoch": 3.381844380403458,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038844480514035727,
      "loss": 4.6913,
      "mean_token_accuracy": 0.2501776233315468,
      "num_tokens": 80712186.0,
      "step": 35205
    },
    {
      "entropy": 5.403799152374267,
      "epoch": 3.382324687800192,
      "grad_norm": 1.296875,
      "learning_rate": 0.000388415196976287,
      "loss": 4.6687,
      "mean_token_accuracy": 0.2590024724602699,
      "num_tokens": 80723865.0,
      "step": 35210
    },
    {
      "entropy": 5.411276912689209,
      "epoch": 3.382804995196926,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003883855861790526,
      "loss": 4.7126,
      "mean_token_accuracy": 0.24666176587343216,
      "num_tokens": 80735817.0,
      "step": 35215
    },
    {
      "entropy": 5.436880016326905,
      "epoch": 3.38328530259366,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003883559727493417,
      "loss": 4.7688,
      "mean_token_accuracy": 0.24823495000600815,
      "num_tokens": 80747782.0,
      "step": 35220
    },
    {
      "entropy": 5.450526571273803,
      "epoch": 3.3837656099903937,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003883263566878416,
      "loss": 4.7214,
      "mean_token_accuracy": 0.23424150347709655,
      "num_tokens": 80759119.0,
      "step": 35225
    },
    {
      "entropy": 5.399685287475586,
      "epoch": 3.384245917387128,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038829673799524006,
      "loss": 4.6187,
      "mean_token_accuracy": 0.26171707808971406,
      "num_tokens": 80769792.0,
      "step": 35230
    },
    {
      "entropy": 5.390163993835449,
      "epoch": 3.3847262247838614,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038826711667222464,
      "loss": 4.6569,
      "mean_token_accuracy": 0.25189554393291474,
      "num_tokens": 80781685.0,
      "step": 35235
    },
    {
      "entropy": 5.352747297286987,
      "epoch": 3.3852065321805958,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038823749271948315,
      "loss": 4.6139,
      "mean_token_accuracy": 0.2608355075120926,
      "num_tokens": 80792499.0,
      "step": 35240
    },
    {
      "entropy": 5.463081026077271,
      "epoch": 3.3856868395773296,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038820786613770334,
      "loss": 4.6954,
      "mean_token_accuracy": 0.25536195039749143,
      "num_tokens": 80803020.0,
      "step": 35245
    },
    {
      "entropy": 5.472555446624756,
      "epoch": 3.3861671469740635,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038817823692757303,
      "loss": 4.7231,
      "mean_token_accuracy": 0.2550787955522537,
      "num_tokens": 80814543.0,
      "step": 35250
    },
    {
      "entropy": 5.376118564605713,
      "epoch": 3.3866474543707974,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038814860508978004,
      "loss": 4.6553,
      "mean_token_accuracy": 0.25672752410173416,
      "num_tokens": 80826686.0,
      "step": 35255
    },
    {
      "entropy": 5.435895204544067,
      "epoch": 3.3871277617675313,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003881189706250125,
      "loss": 4.7659,
      "mean_token_accuracy": 0.2446002721786499,
      "num_tokens": 80838372.0,
      "step": 35260
    },
    {
      "entropy": 5.496024036407471,
      "epoch": 3.387608069164265,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038808933353395836,
      "loss": 4.7289,
      "mean_token_accuracy": 0.2491311937570572,
      "num_tokens": 80849334.0,
      "step": 35265
    },
    {
      "entropy": 5.369501113891602,
      "epoch": 3.388088376560999,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038805969381730564,
      "loss": 4.5895,
      "mean_token_accuracy": 0.25531365275382994,
      "num_tokens": 80860682.0,
      "step": 35270
    },
    {
      "entropy": 5.4299560546875,
      "epoch": 3.388568683957733,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038803005147574265,
      "loss": 4.7364,
      "mean_token_accuracy": 0.2492813214659691,
      "num_tokens": 80872026.0,
      "step": 35275
    },
    {
      "entropy": 5.482859182357788,
      "epoch": 3.389048991354467,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003880004065099575,
      "loss": 4.7454,
      "mean_token_accuracy": 0.24573743343353271,
      "num_tokens": 80884146.0,
      "step": 35280
    },
    {
      "entropy": 5.4335041522979735,
      "epoch": 3.3895292987512007,
      "grad_norm": 1.125,
      "learning_rate": 0.0003879707589206383,
      "loss": 4.7234,
      "mean_token_accuracy": 0.2498758092522621,
      "num_tokens": 80895975.0,
      "step": 35285
    },
    {
      "entropy": 5.5157122135162355,
      "epoch": 3.3900096061479346,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003879411087084736,
      "loss": 4.788,
      "mean_token_accuracy": 0.24256788045167924,
      "num_tokens": 80908144.0,
      "step": 35290
    },
    {
      "entropy": 5.435681009292603,
      "epoch": 3.3904899135446684,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038791145587415186,
      "loss": 4.6885,
      "mean_token_accuracy": 0.25282672643661497,
      "num_tokens": 80918616.0,
      "step": 35295
    },
    {
      "entropy": 5.486489534378052,
      "epoch": 3.3909702209414023,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038788180041836117,
      "loss": 4.7912,
      "mean_token_accuracy": 0.2465907961130142,
      "num_tokens": 80929690.0,
      "step": 35300
    },
    {
      "entropy": 5.468409299850464,
      "epoch": 3.3914505283381366,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038785214234179037,
      "loss": 4.6763,
      "mean_token_accuracy": 0.2500907227396965,
      "num_tokens": 80940817.0,
      "step": 35305
    },
    {
      "entropy": 5.451190328598022,
      "epoch": 3.39193083573487,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038782248164512804,
      "loss": 4.6648,
      "mean_token_accuracy": 0.25516152679920195,
      "num_tokens": 80950951.0,
      "step": 35310
    },
    {
      "entropy": 5.425511407852173,
      "epoch": 3.3924111431316044,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038779281832906253,
      "loss": 4.703,
      "mean_token_accuracy": 0.2520365685224533,
      "num_tokens": 80963511.0,
      "step": 35315
    },
    {
      "entropy": 5.521159267425537,
      "epoch": 3.3928914505283383,
      "grad_norm": 1.125,
      "learning_rate": 0.00038776315239428275,
      "loss": 4.7933,
      "mean_token_accuracy": 0.2460867390036583,
      "num_tokens": 80975905.0,
      "step": 35320
    },
    {
      "entropy": 5.483807325363159,
      "epoch": 3.393371757925072,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038773348384147743,
      "loss": 4.7332,
      "mean_token_accuracy": 0.25070747882127764,
      "num_tokens": 80986850.0,
      "step": 35325
    },
    {
      "entropy": 5.349221992492676,
      "epoch": 3.393852065321806,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003877038126713354,
      "loss": 4.5815,
      "mean_token_accuracy": 0.26785062402486803,
      "num_tokens": 80998190.0,
      "step": 35330
    },
    {
      "entropy": 5.367937278747559,
      "epoch": 3.39433237271854,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038767413888454537,
      "loss": 4.6704,
      "mean_token_accuracy": 0.24956746250391007,
      "num_tokens": 81011086.0,
      "step": 35335
    },
    {
      "entropy": 5.552329397201538,
      "epoch": 3.394812680115274,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038764446248179665,
      "loss": 4.7802,
      "mean_token_accuracy": 0.2514008194208145,
      "num_tokens": 81022074.0,
      "step": 35340
    },
    {
      "entropy": 5.444555187225342,
      "epoch": 3.3952929875120077,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003876147834637778,
      "loss": 4.6884,
      "mean_token_accuracy": 0.24930763244628906,
      "num_tokens": 81033019.0,
      "step": 35345
    },
    {
      "entropy": 5.475594234466553,
      "epoch": 3.3957732949087416,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038758510183117806,
      "loss": 4.7721,
      "mean_token_accuracy": 0.24808503836393356,
      "num_tokens": 81044955.0,
      "step": 35350
    },
    {
      "entropy": 5.4080610275268555,
      "epoch": 3.3962536023054755,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003875554175846866,
      "loss": 4.6112,
      "mean_token_accuracy": 0.260129976272583,
      "num_tokens": 81055980.0,
      "step": 35355
    },
    {
      "entropy": 5.42901930809021,
      "epoch": 3.3967339097022093,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038752573072499267,
      "loss": 4.6455,
      "mean_token_accuracy": 0.2581823989748955,
      "num_tokens": 81067960.0,
      "step": 35360
    },
    {
      "entropy": 5.459894752502441,
      "epoch": 3.3972142170989432,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038749604125278524,
      "loss": 4.7422,
      "mean_token_accuracy": 0.24741540402173995,
      "num_tokens": 81078084.0,
      "step": 35365
    },
    {
      "entropy": 5.459806060791015,
      "epoch": 3.397694524495677,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003874663491687539,
      "loss": 4.6527,
      "mean_token_accuracy": 0.2548586890101433,
      "num_tokens": 81089376.0,
      "step": 35370
    },
    {
      "entropy": 5.516777229309082,
      "epoch": 3.398174831892411,
      "grad_norm": 1.125,
      "learning_rate": 0.00038743665447358785,
      "loss": 4.7603,
      "mean_token_accuracy": 0.24361282140016555,
      "num_tokens": 81100906.0,
      "step": 35375
    },
    {
      "entropy": 5.480842638015747,
      "epoch": 3.398655139289145,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003874069571679766,
      "loss": 4.7359,
      "mean_token_accuracy": 0.251839841902256,
      "num_tokens": 81112839.0,
      "step": 35380
    },
    {
      "entropy": 5.412104701995849,
      "epoch": 3.3991354466858787,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038737725725260946,
      "loss": 4.6449,
      "mean_token_accuracy": 0.2558803975582123,
      "num_tokens": 81126015.0,
      "step": 35385
    },
    {
      "entropy": 5.4641529560089115,
      "epoch": 3.399615754082613,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038734755472817617,
      "loss": 4.7515,
      "mean_token_accuracy": 0.25005447417497634,
      "num_tokens": 81139526.0,
      "step": 35390
    },
    {
      "entropy": 5.431187295913697,
      "epoch": 3.400096061479347,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038731784959536626,
      "loss": 4.6244,
      "mean_token_accuracy": 0.2587812960147858,
      "num_tokens": 81152340.0,
      "step": 35395
    },
    {
      "entropy": 5.543612098693847,
      "epoch": 3.400576368876081,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038728814185486944,
      "loss": 4.7489,
      "mean_token_accuracy": 0.24386952370405196,
      "num_tokens": 81162724.0,
      "step": 35400
    },
    {
      "entropy": 5.4537413120269775,
      "epoch": 3.4010566762728147,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003872584315073753,
      "loss": 4.7141,
      "mean_token_accuracy": 0.24765170365571976,
      "num_tokens": 81172933.0,
      "step": 35405
    },
    {
      "entropy": 5.419608306884766,
      "epoch": 3.4015369836695486,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003872287185535738,
      "loss": 4.7236,
      "mean_token_accuracy": 0.2470464825630188,
      "num_tokens": 81183774.0,
      "step": 35410
    },
    {
      "entropy": 5.346129703521728,
      "epoch": 3.4020172910662825,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038719900299415475,
      "loss": 4.5621,
      "mean_token_accuracy": 0.26794341802597044,
      "num_tokens": 81194896.0,
      "step": 35415
    },
    {
      "entropy": 5.461100959777832,
      "epoch": 3.4024975984630164,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003871692848298079,
      "loss": 4.6888,
      "mean_token_accuracy": 0.24966205656528473,
      "num_tokens": 81204889.0,
      "step": 35420
    },
    {
      "entropy": 5.482469129562378,
      "epoch": 3.4029779058597502,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038713956406122334,
      "loss": 4.761,
      "mean_token_accuracy": 0.24237653762102127,
      "num_tokens": 81217177.0,
      "step": 35425
    },
    {
      "entropy": 5.509892845153809,
      "epoch": 3.403458213256484,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038710984068909116,
      "loss": 4.781,
      "mean_token_accuracy": 0.24309907406568526,
      "num_tokens": 81228767.0,
      "step": 35430
    },
    {
      "entropy": 5.430819416046143,
      "epoch": 3.403938520653218,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003870801147141014,
      "loss": 4.651,
      "mean_token_accuracy": 0.24610039740800857,
      "num_tokens": 81239599.0,
      "step": 35435
    },
    {
      "entropy": 5.440488338470459,
      "epoch": 3.404418828049952,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003870503861369441,
      "loss": 4.7189,
      "mean_token_accuracy": 0.2514369830489159,
      "num_tokens": 81251051.0,
      "step": 35440
    },
    {
      "entropy": 5.462915754318237,
      "epoch": 3.4048991354466858,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038702065495830956,
      "loss": 4.7536,
      "mean_token_accuracy": 0.24939581453800203,
      "num_tokens": 81262484.0,
      "step": 35445
    },
    {
      "entropy": 5.424311876296997,
      "epoch": 3.4053794428434196,
      "grad_norm": 1.25,
      "learning_rate": 0.0003869909211788881,
      "loss": 4.6712,
      "mean_token_accuracy": 0.2593751817941666,
      "num_tokens": 81273068.0,
      "step": 35450
    },
    {
      "entropy": 5.448486375808716,
      "epoch": 3.4058597502401535,
      "grad_norm": 1.25,
      "learning_rate": 0.00038696118479936994,
      "loss": 4.7303,
      "mean_token_accuracy": 0.2469109833240509,
      "num_tokens": 81285239.0,
      "step": 35455
    },
    {
      "entropy": 5.499677658081055,
      "epoch": 3.4063400576368874,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038693144582044553,
      "loss": 4.7114,
      "mean_token_accuracy": 0.2477413982152939,
      "num_tokens": 81295998.0,
      "step": 35460
    },
    {
      "entropy": 5.380808115005493,
      "epoch": 3.4068203650336217,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038690170424280534,
      "loss": 4.6156,
      "mean_token_accuracy": 0.2517089322209358,
      "num_tokens": 81307569.0,
      "step": 35465
    },
    {
      "entropy": 5.402000522613525,
      "epoch": 3.4073006724303556,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003868719600671399,
      "loss": 4.666,
      "mean_token_accuracy": 0.24996984452009202,
      "num_tokens": 81319023.0,
      "step": 35470
    },
    {
      "entropy": 5.349354410171509,
      "epoch": 3.4077809798270895,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038684221329413965,
      "loss": 4.647,
      "mean_token_accuracy": 0.253397274017334,
      "num_tokens": 81330087.0,
      "step": 35475
    },
    {
      "entropy": 5.408992624282837,
      "epoch": 3.4082612872238234,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003868124639244954,
      "loss": 4.5979,
      "mean_token_accuracy": 0.26265921592712405,
      "num_tokens": 81341533.0,
      "step": 35480
    },
    {
      "entropy": 5.490687561035156,
      "epoch": 3.4087415946205573,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038678271195889766,
      "loss": 4.7576,
      "mean_token_accuracy": 0.24031210094690322,
      "num_tokens": 81353505.0,
      "step": 35485
    },
    {
      "entropy": 5.4626445293426515,
      "epoch": 3.409221902017291,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038675295739803734,
      "loss": 4.7334,
      "mean_token_accuracy": 0.2505513235926628,
      "num_tokens": 81365246.0,
      "step": 35490
    },
    {
      "entropy": 5.559177303314209,
      "epoch": 3.409702209414025,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038672320024260516,
      "loss": 4.8076,
      "mean_token_accuracy": 0.24149993658065796,
      "num_tokens": 81376403.0,
      "step": 35495
    },
    {
      "entropy": 5.417534112930298,
      "epoch": 3.410182516810759,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038669344049329204,
      "loss": 4.69,
      "mean_token_accuracy": 0.2552834928035736,
      "num_tokens": 81386865.0,
      "step": 35500
    },
    {
      "entropy": 5.3302570343017575,
      "epoch": 3.410662824207493,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038666367815078887,
      "loss": 4.6705,
      "mean_token_accuracy": 0.2557420641183853,
      "num_tokens": 81398191.0,
      "step": 35505
    },
    {
      "entropy": 5.433197355270385,
      "epoch": 3.4111431316042267,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003866339132157867,
      "loss": 4.7441,
      "mean_token_accuracy": 0.24978163689374924,
      "num_tokens": 81410694.0,
      "step": 35510
    },
    {
      "entropy": 5.467283058166504,
      "epoch": 3.4116234390009605,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003866041456889764,
      "loss": 4.6797,
      "mean_token_accuracy": 0.24898177832365037,
      "num_tokens": 81422195.0,
      "step": 35515
    },
    {
      "entropy": 5.457956171035766,
      "epoch": 3.4121037463976944,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038657437557104946,
      "loss": 4.7118,
      "mean_token_accuracy": 0.24962015002965926,
      "num_tokens": 81432979.0,
      "step": 35520
    },
    {
      "entropy": 5.385507535934448,
      "epoch": 3.4125840537944283,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003865446028626967,
      "loss": 4.6457,
      "mean_token_accuracy": 0.2555881440639496,
      "num_tokens": 81444328.0,
      "step": 35525
    },
    {
      "entropy": 5.531324100494385,
      "epoch": 3.413064361191162,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038651482756460947,
      "loss": 4.8957,
      "mean_token_accuracy": 0.24308789521455765,
      "num_tokens": 81455530.0,
      "step": 35530
    },
    {
      "entropy": 5.438281917572022,
      "epoch": 3.413544668587896,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038648504967747914,
      "loss": 4.6825,
      "mean_token_accuracy": 0.25229328870773315,
      "num_tokens": 81466462.0,
      "step": 35535
    },
    {
      "entropy": 5.515604305267334,
      "epoch": 3.4140249759846304,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038645526920199697,
      "loss": 4.7878,
      "mean_token_accuracy": 0.24304087460041046,
      "num_tokens": 81479459.0,
      "step": 35540
    },
    {
      "entropy": 5.470956087112427,
      "epoch": 3.414505283381364,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003864254861388544,
      "loss": 4.8282,
      "mean_token_accuracy": 0.2456494003534317,
      "num_tokens": 81491026.0,
      "step": 35545
    },
    {
      "entropy": 5.488528060913086,
      "epoch": 3.414985590778098,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038639570048874295,
      "loss": 4.7213,
      "mean_token_accuracy": 0.25411768108606336,
      "num_tokens": 81504469.0,
      "step": 35550
    },
    {
      "entropy": 5.44148097038269,
      "epoch": 3.415465898174832,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038636591225235407,
      "loss": 4.6531,
      "mean_token_accuracy": 0.252341990172863,
      "num_tokens": 81515873.0,
      "step": 35555
    },
    {
      "entropy": 5.424675559997558,
      "epoch": 3.415946205571566,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003863361214303794,
      "loss": 4.6903,
      "mean_token_accuracy": 0.2514757364988327,
      "num_tokens": 81528169.0,
      "step": 35560
    },
    {
      "entropy": 5.475280714035034,
      "epoch": 3.4164265129683,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003863063280235106,
      "loss": 4.6823,
      "mean_token_accuracy": 0.251334910094738,
      "num_tokens": 81538831.0,
      "step": 35565
    },
    {
      "entropy": 5.528482675552368,
      "epoch": 3.4169068203650337,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038627653203243933,
      "loss": 4.8153,
      "mean_token_accuracy": 0.24372090995311738,
      "num_tokens": 81549809.0,
      "step": 35570
    },
    {
      "entropy": 5.368585777282715,
      "epoch": 3.4173871277617676,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003862467334578574,
      "loss": 4.6334,
      "mean_token_accuracy": 0.2550181090831757,
      "num_tokens": 81560537.0,
      "step": 35575
    },
    {
      "entropy": 5.387383699417114,
      "epoch": 3.4178674351585014,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038621693230045677,
      "loss": 4.679,
      "mean_token_accuracy": 0.2555195868015289,
      "num_tokens": 81571983.0,
      "step": 35580
    },
    {
      "entropy": 5.3786547660827635,
      "epoch": 3.4183477425552353,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003861871285609291,
      "loss": 4.652,
      "mean_token_accuracy": 0.25013935565948486,
      "num_tokens": 81584783.0,
      "step": 35585
    },
    {
      "entropy": 5.413335561752319,
      "epoch": 3.418828049951969,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003861573222399665,
      "loss": 4.6131,
      "mean_token_accuracy": 0.2556250587105751,
      "num_tokens": 81596619.0,
      "step": 35590
    },
    {
      "entropy": 5.558559894561768,
      "epoch": 3.419308357348703,
      "grad_norm": 1.21875,
      "learning_rate": 0.000386127513338261,
      "loss": 4.8142,
      "mean_token_accuracy": 0.23957280367612838,
      "num_tokens": 81607981.0,
      "step": 35595
    },
    {
      "entropy": 5.482106161117554,
      "epoch": 3.419788664745437,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003860977018565046,
      "loss": 4.794,
      "mean_token_accuracy": 0.24504088163375853,
      "num_tokens": 81620444.0,
      "step": 35600
    },
    {
      "entropy": 5.4369285106658936,
      "epoch": 3.420268972142171,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003860678877953894,
      "loss": 4.6835,
      "mean_token_accuracy": 0.2475135326385498,
      "num_tokens": 81631668.0,
      "step": 35605
    },
    {
      "entropy": 5.461303758621216,
      "epoch": 3.4207492795389047,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003860380711556077,
      "loss": 4.8027,
      "mean_token_accuracy": 0.2467698022723198,
      "num_tokens": 81643230.0,
      "step": 35610
    },
    {
      "entropy": 5.480565547943115,
      "epoch": 3.421229586935639,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038600825193785173,
      "loss": 4.7302,
      "mean_token_accuracy": 0.2488080605864525,
      "num_tokens": 81654217.0,
      "step": 35615
    },
    {
      "entropy": 5.449506855010986,
      "epoch": 3.4217098943323725,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003859784301428137,
      "loss": 4.7724,
      "mean_token_accuracy": 0.24280257225036622,
      "num_tokens": 81666025.0,
      "step": 35620
    },
    {
      "entropy": 5.463738584518433,
      "epoch": 3.422190201729107,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003859486057711861,
      "loss": 4.7087,
      "mean_token_accuracy": 0.24777246713638307,
      "num_tokens": 81677417.0,
      "step": 35625
    },
    {
      "entropy": 5.4760682582855225,
      "epoch": 3.4226705091258407,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003859187788236613,
      "loss": 4.7488,
      "mean_token_accuracy": 0.25132103115320203,
      "num_tokens": 81687898.0,
      "step": 35630
    },
    {
      "entropy": 5.362353801727295,
      "epoch": 3.4231508165225746,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038588894930093184,
      "loss": 4.6475,
      "mean_token_accuracy": 0.25639188587665557,
      "num_tokens": 81699532.0,
      "step": 35635
    },
    {
      "entropy": 5.4849974632263185,
      "epoch": 3.4236311239193085,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038585911720369023,
      "loss": 4.7553,
      "mean_token_accuracy": 0.24141464680433272,
      "num_tokens": 81709164.0,
      "step": 35640
    },
    {
      "entropy": 5.488157510757446,
      "epoch": 3.4241114313160423,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003858292825326291,
      "loss": 4.7074,
      "mean_token_accuracy": 0.25369535982608793,
      "num_tokens": 81721354.0,
      "step": 35645
    },
    {
      "entropy": 5.397724723815918,
      "epoch": 3.4245917387127762,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003857994452884412,
      "loss": 4.6158,
      "mean_token_accuracy": 0.2613354593515396,
      "num_tokens": 81731793.0,
      "step": 35650
    },
    {
      "entropy": 5.479383754730224,
      "epoch": 3.42507204610951,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003857696054718191,
      "loss": 4.7324,
      "mean_token_accuracy": 0.2553535312414169,
      "num_tokens": 81743254.0,
      "step": 35655
    },
    {
      "entropy": 5.5165081977844235,
      "epoch": 3.425552353506244,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003857397630834557,
      "loss": 4.7868,
      "mean_token_accuracy": 0.24327779263257981,
      "num_tokens": 81755080.0,
      "step": 35660
    },
    {
      "entropy": 5.453504991531372,
      "epoch": 3.426032660902978,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038570991812404384,
      "loss": 4.7302,
      "mean_token_accuracy": 0.25163761228322984,
      "num_tokens": 81767367.0,
      "step": 35665
    },
    {
      "entropy": 5.472439241409302,
      "epoch": 3.4265129682997117,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003856800705942764,
      "loss": 4.7881,
      "mean_token_accuracy": 0.23987525701522827,
      "num_tokens": 81778513.0,
      "step": 35670
    },
    {
      "entropy": 5.443566513061524,
      "epoch": 3.4269932756964456,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038565022049484636,
      "loss": 4.6613,
      "mean_token_accuracy": 0.25930006057024,
      "num_tokens": 81789845.0,
      "step": 35675
    },
    {
      "entropy": 5.41775803565979,
      "epoch": 3.4274735830931795,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038562036782644675,
      "loss": 4.7429,
      "mean_token_accuracy": 0.24853651970624924,
      "num_tokens": 81801287.0,
      "step": 35680
    },
    {
      "entropy": 5.417262935638428,
      "epoch": 3.4279538904899134,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003855905125897708,
      "loss": 4.6754,
      "mean_token_accuracy": 0.25566571950912476,
      "num_tokens": 81811462.0,
      "step": 35685
    },
    {
      "entropy": 5.381610012054443,
      "epoch": 3.4284341978866473,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038556065478551147,
      "loss": 4.6395,
      "mean_token_accuracy": 0.24988506138324737,
      "num_tokens": 81822590.0,
      "step": 35690
    },
    {
      "entropy": 5.474045324325561,
      "epoch": 3.428914505283381,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000385530794414362,
      "loss": 4.7014,
      "mean_token_accuracy": 0.25802018493413925,
      "num_tokens": 81834755.0,
      "step": 35695
    },
    {
      "entropy": 5.422270393371582,
      "epoch": 3.4293948126801155,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003855009314770157,
      "loss": 4.5987,
      "mean_token_accuracy": 0.25736205130815504,
      "num_tokens": 81846259.0,
      "step": 35700
    },
    {
      "entropy": 5.3920446872711185,
      "epoch": 3.4298751200768494,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038547106597416593,
      "loss": 4.6443,
      "mean_token_accuracy": 0.25840216428041457,
      "num_tokens": 81856262.0,
      "step": 35705
    },
    {
      "entropy": 5.4418230056762695,
      "epoch": 3.4303554274735832,
      "grad_norm": 1.140625,
      "learning_rate": 0.000385441197906506,
      "loss": 4.7294,
      "mean_token_accuracy": 0.2471067041158676,
      "num_tokens": 81867201.0,
      "step": 35710
    },
    {
      "entropy": 5.542460680007935,
      "epoch": 3.430835734870317,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038541132727472945,
      "loss": 4.9007,
      "mean_token_accuracy": 0.244399031996727,
      "num_tokens": 81878654.0,
      "step": 35715
    },
    {
      "entropy": 5.471721649169922,
      "epoch": 3.431316042267051,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038538145407952964,
      "loss": 4.6555,
      "mean_token_accuracy": 0.258257058262825,
      "num_tokens": 81889672.0,
      "step": 35720
    },
    {
      "entropy": 5.453236436843872,
      "epoch": 3.431796349663785,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003853515783216003,
      "loss": 4.6883,
      "mean_token_accuracy": 0.2467265620827675,
      "num_tokens": 81901072.0,
      "step": 35725
    },
    {
      "entropy": 5.420616579055786,
      "epoch": 3.4322766570605188,
      "grad_norm": 1.203125,
      "learning_rate": 0.000385321700001635,
      "loss": 4.6465,
      "mean_token_accuracy": 0.2574083849787712,
      "num_tokens": 81913146.0,
      "step": 35730
    },
    {
      "entropy": 5.480524110794067,
      "epoch": 3.4327569644572526,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003852918191203274,
      "loss": 4.7611,
      "mean_token_accuracy": 0.2430671289563179,
      "num_tokens": 81924648.0,
      "step": 35735
    },
    {
      "entropy": 5.4793946743011475,
      "epoch": 3.4332372718539865,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003852619356783712,
      "loss": 4.7684,
      "mean_token_accuracy": 0.2400476023554802,
      "num_tokens": 81935744.0,
      "step": 35740
    },
    {
      "entropy": 5.514432573318482,
      "epoch": 3.4337175792507204,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003852320496764603,
      "loss": 4.7065,
      "mean_token_accuracy": 0.24774158447980882,
      "num_tokens": 81946931.0,
      "step": 35745
    },
    {
      "entropy": 5.438637971878052,
      "epoch": 3.4341978866474543,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038520216111528855,
      "loss": 4.6858,
      "mean_token_accuracy": 0.25942299962043763,
      "num_tokens": 81958009.0,
      "step": 35750
    },
    {
      "entropy": 5.4275459289550785,
      "epoch": 3.434678194044188,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003851722699955499,
      "loss": 4.732,
      "mean_token_accuracy": 0.24873974919319153,
      "num_tokens": 81970648.0,
      "step": 35755
    },
    {
      "entropy": 5.4551760196685795,
      "epoch": 3.435158501440922,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003851423763179382,
      "loss": 4.673,
      "mean_token_accuracy": 0.2614157050848007,
      "num_tokens": 81982293.0,
      "step": 35760
    },
    {
      "entropy": 5.385718965530396,
      "epoch": 3.435638808837656,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038511248008314756,
      "loss": 4.6748,
      "mean_token_accuracy": 0.255496147274971,
      "num_tokens": 81993856.0,
      "step": 35765
    },
    {
      "entropy": 5.483035087585449,
      "epoch": 3.43611911623439,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003850825812918722,
      "loss": 4.7657,
      "mean_token_accuracy": 0.2477887362241745,
      "num_tokens": 82005359.0,
      "step": 35770
    },
    {
      "entropy": 5.493106842041016,
      "epoch": 3.436599423631124,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003850526799448061,
      "loss": 4.7101,
      "mean_token_accuracy": 0.25207181721925737,
      "num_tokens": 82017789.0,
      "step": 35775
    },
    {
      "entropy": 5.4666468620300295,
      "epoch": 3.437079731027858,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003850227760426436,
      "loss": 4.6945,
      "mean_token_accuracy": 0.2516950398683548,
      "num_tokens": 82028147.0,
      "step": 35780
    },
    {
      "entropy": 5.416484069824219,
      "epoch": 3.437560038424592,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038499286958607894,
      "loss": 4.6916,
      "mean_token_accuracy": 0.24891269207000732,
      "num_tokens": 82039485.0,
      "step": 35785
    },
    {
      "entropy": 5.393529081344605,
      "epoch": 3.438040345821326,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003849629605758065,
      "loss": 4.6201,
      "mean_token_accuracy": 0.26460592448711395,
      "num_tokens": 82050004.0,
      "step": 35790
    },
    {
      "entropy": 5.435658073425293,
      "epoch": 3.4385206532180597,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003849330490125205,
      "loss": 4.6396,
      "mean_token_accuracy": 0.2567748069763184,
      "num_tokens": 82060621.0,
      "step": 35795
    },
    {
      "entropy": 5.43260645866394,
      "epoch": 3.4390009606147935,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038490313489691566,
      "loss": 4.7002,
      "mean_token_accuracy": 0.2486884981393814,
      "num_tokens": 82070955.0,
      "step": 35800
    },
    {
      "entropy": 5.51964316368103,
      "epoch": 3.4394812680115274,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003848732182296863,
      "loss": 4.8992,
      "mean_token_accuracy": 0.23345067203044892,
      "num_tokens": 82082623.0,
      "step": 35805
    },
    {
      "entropy": 5.46042423248291,
      "epoch": 3.4399615754082613,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038484329901152713,
      "loss": 4.6517,
      "mean_token_accuracy": 0.2551464274525642,
      "num_tokens": 82092707.0,
      "step": 35810
    },
    {
      "entropy": 5.430969190597534,
      "epoch": 3.440441882804995,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038481337724313264,
      "loss": 4.6704,
      "mean_token_accuracy": 0.2563789516687393,
      "num_tokens": 82105114.0,
      "step": 35815
    },
    {
      "entropy": 5.364910650253296,
      "epoch": 3.440922190201729,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003847834529251977,
      "loss": 4.6478,
      "mean_token_accuracy": 0.2626077145338058,
      "num_tokens": 82117638.0,
      "step": 35820
    },
    {
      "entropy": 5.419521522521973,
      "epoch": 3.441402497598463,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038475352605841693,
      "loss": 4.7307,
      "mean_token_accuracy": 0.2461439475417137,
      "num_tokens": 82129100.0,
      "step": 35825
    },
    {
      "entropy": 5.491789531707764,
      "epoch": 3.441882804995197,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003847235966434852,
      "loss": 4.7584,
      "mean_token_accuracy": 0.2511852473020554,
      "num_tokens": 82140776.0,
      "step": 35830
    },
    {
      "entropy": 5.505823850631714,
      "epoch": 3.4423631123919307,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003846936646810974,
      "loss": 4.766,
      "mean_token_accuracy": 0.244469752907753,
      "num_tokens": 82151569.0,
      "step": 35835
    },
    {
      "entropy": 5.515319395065307,
      "epoch": 3.4428434197886646,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038466373017194834,
      "loss": 4.7944,
      "mean_token_accuracy": 0.24727065861225128,
      "num_tokens": 82162696.0,
      "step": 35840
    },
    {
      "entropy": 5.369102478027344,
      "epoch": 3.4433237271853985,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003846337931167332,
      "loss": 4.6034,
      "mean_token_accuracy": 0.2594327092170715,
      "num_tokens": 82173580.0,
      "step": 35845
    },
    {
      "entropy": 5.354727935791016,
      "epoch": 3.443804034582133,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038460385351614683,
      "loss": 4.6987,
      "mean_token_accuracy": 0.25836297422647475,
      "num_tokens": 82185839.0,
      "step": 35850
    },
    {
      "entropy": 5.430714225769043,
      "epoch": 3.4442843419788662,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038457391137088455,
      "loss": 4.6982,
      "mean_token_accuracy": 0.24720740914344788,
      "num_tokens": 82196682.0,
      "step": 35855
    },
    {
      "entropy": 5.459149122238159,
      "epoch": 3.4447646493756006,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038454396668164136,
      "loss": 4.6961,
      "mean_token_accuracy": 0.2503824681043625,
      "num_tokens": 82207349.0,
      "step": 35860
    },
    {
      "entropy": 5.443182277679443,
      "epoch": 3.4452449567723344,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003845140194491126,
      "loss": 4.6681,
      "mean_token_accuracy": 0.25365370213985444,
      "num_tokens": 82218547.0,
      "step": 35865
    },
    {
      "entropy": 5.450742530822754,
      "epoch": 3.4457252641690683,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038448406967399334,
      "loss": 4.6983,
      "mean_token_accuracy": 0.252657613158226,
      "num_tokens": 82230246.0,
      "step": 35870
    },
    {
      "entropy": 5.393023157119751,
      "epoch": 3.446205571565802,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038445411735697917,
      "loss": 4.6602,
      "mean_token_accuracy": 0.253817643225193,
      "num_tokens": 82240924.0,
      "step": 35875
    },
    {
      "entropy": 5.4055084705352785,
      "epoch": 3.446685878962536,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003844241624987655,
      "loss": 4.7214,
      "mean_token_accuracy": 0.2490170955657959,
      "num_tokens": 82252951.0,
      "step": 35880
    },
    {
      "entropy": 5.4994011402130125,
      "epoch": 3.44716618635927,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003843942051000476,
      "loss": 4.7778,
      "mean_token_accuracy": 0.24058375656604766,
      "num_tokens": 82264377.0,
      "step": 35885
    },
    {
      "entropy": 5.538672208786011,
      "epoch": 3.447646493756004,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000384364245161521,
      "loss": 4.8134,
      "mean_token_accuracy": 0.23650054186582564,
      "num_tokens": 82276736.0,
      "step": 35890
    },
    {
      "entropy": 5.504140186309814,
      "epoch": 3.4481268011527377,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003843342826838815,
      "loss": 4.7373,
      "mean_token_accuracy": 0.24804002791643143,
      "num_tokens": 82288574.0,
      "step": 35895
    },
    {
      "entropy": 5.412194490432739,
      "epoch": 3.4486071085494716,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038430431766782463,
      "loss": 4.6333,
      "mean_token_accuracy": 0.25923354625701905,
      "num_tokens": 82299570.0,
      "step": 35900
    },
    {
      "entropy": 5.482267189025879,
      "epoch": 3.4490874159462055,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000384274350114046,
      "loss": 4.7716,
      "mean_token_accuracy": 0.2406499832868576,
      "num_tokens": 82310449.0,
      "step": 35905
    },
    {
      "entropy": 5.465459108352661,
      "epoch": 3.4495677233429394,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038424438002324145,
      "loss": 4.7473,
      "mean_token_accuracy": 0.2465823158621788,
      "num_tokens": 82321517.0,
      "step": 35910
    },
    {
      "entropy": 5.401159334182739,
      "epoch": 3.4500480307396733,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038421440739610683,
      "loss": 4.7143,
      "mean_token_accuracy": 0.24782880544662475,
      "num_tokens": 82334555.0,
      "step": 35915
    },
    {
      "entropy": 5.3567129611969,
      "epoch": 3.450528338136407,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038418443223333797,
      "loss": 4.5761,
      "mean_token_accuracy": 0.2594401925802231,
      "num_tokens": 82347166.0,
      "step": 35920
    },
    {
      "entropy": 5.373970890045166,
      "epoch": 3.451008645533141,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003841544545356308,
      "loss": 4.629,
      "mean_token_accuracy": 0.2534667491912842,
      "num_tokens": 82358220.0,
      "step": 35925
    },
    {
      "entropy": 5.447188901901245,
      "epoch": 3.451488952929875,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038412447430368125,
      "loss": 4.7268,
      "mean_token_accuracy": 0.24769855588674544,
      "num_tokens": 82369154.0,
      "step": 35930
    },
    {
      "entropy": 5.446628093719482,
      "epoch": 3.4519692603266092,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038409449153818556,
      "loss": 4.658,
      "mean_token_accuracy": 0.2563991218805313,
      "num_tokens": 82380311.0,
      "step": 35935
    },
    {
      "entropy": 5.419747018814087,
      "epoch": 3.452449567723343,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038406450623983964,
      "loss": 4.6803,
      "mean_token_accuracy": 0.2544480711221695,
      "num_tokens": 82390895.0,
      "step": 35940
    },
    {
      "entropy": 5.4287079811096195,
      "epoch": 3.452929875120077,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038403451840933966,
      "loss": 4.6419,
      "mean_token_accuracy": 0.2539542391896248,
      "num_tokens": 82400880.0,
      "step": 35945
    },
    {
      "entropy": 5.4110846519470215,
      "epoch": 3.453410182516811,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038400452804738204,
      "loss": 4.7056,
      "mean_token_accuracy": 0.25832283943891526,
      "num_tokens": 82413566.0,
      "step": 35950
    },
    {
      "entropy": 5.401994132995606,
      "epoch": 3.4538904899135447,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038397453515466297,
      "loss": 4.6376,
      "mean_token_accuracy": 0.256507508456707,
      "num_tokens": 82424434.0,
      "step": 35955
    },
    {
      "entropy": 5.490840673446655,
      "epoch": 3.4543707973102786,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003839445397318787,
      "loss": 4.7705,
      "mean_token_accuracy": 0.24574447125196458,
      "num_tokens": 82436310.0,
      "step": 35960
    },
    {
      "entropy": 5.406431436538696,
      "epoch": 3.4548511047070125,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003839145417797258,
      "loss": 4.6284,
      "mean_token_accuracy": 0.25457228273153304,
      "num_tokens": 82446484.0,
      "step": 35965
    },
    {
      "entropy": 5.49995436668396,
      "epoch": 3.4553314121037464,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003838845412989006,
      "loss": 4.7922,
      "mean_token_accuracy": 0.24083980172872543,
      "num_tokens": 82458169.0,
      "step": 35970
    },
    {
      "entropy": 5.404331159591675,
      "epoch": 3.4558117195004803,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003838545382900997,
      "loss": 4.6829,
      "mean_token_accuracy": 0.2497251182794571,
      "num_tokens": 82469407.0,
      "step": 35975
    },
    {
      "entropy": 5.378988027572632,
      "epoch": 3.456292026897214,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003838245327540196,
      "loss": 4.6335,
      "mean_token_accuracy": 0.26226264238357544,
      "num_tokens": 82480757.0,
      "step": 35980
    },
    {
      "entropy": 5.40749888420105,
      "epoch": 3.456772334293948,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038379452469135706,
      "loss": 4.7153,
      "mean_token_accuracy": 0.2506769821047783,
      "num_tokens": 82491073.0,
      "step": 35985
    },
    {
      "entropy": 5.444006013870239,
      "epoch": 3.457252641690682,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038376451410280864,
      "loss": 4.7225,
      "mean_token_accuracy": 0.25751113295555117,
      "num_tokens": 82502243.0,
      "step": 35990
    },
    {
      "entropy": 5.449823570251465,
      "epoch": 3.457732949087416,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038373450098907124,
      "loss": 4.684,
      "mean_token_accuracy": 0.24550838023424149,
      "num_tokens": 82514831.0,
      "step": 35995
    },
    {
      "entropy": 5.4286524772644045,
      "epoch": 3.4582132564841497,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038370448535084156,
      "loss": 4.6978,
      "mean_token_accuracy": 0.2553088992834091,
      "num_tokens": 82525907.0,
      "step": 36000
    },
    {
      "epoch": 3.4582132564841497,
      "eval_entropy": 5.1584456023909535,
      "eval_loss": 4.800304889678955,
      "eval_mean_token_accuracy": 0.2550250307626401,
      "eval_num_tokens": 82525907.0,
      "eval_runtime": 20.7114,
      "eval_samples_per_second": 1584.396,
      "eval_steps_per_second": 198.056,
      "step": 36000
    },
    {
      "entropy": 5.4647856712341305,
      "epoch": 3.4586935638808836,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003836744671888165,
      "loss": 4.7308,
      "mean_token_accuracy": 0.24819555431604384,
      "num_tokens": 82537130.0,
      "step": 36005
    },
    {
      "entropy": 5.436455345153808,
      "epoch": 3.459173871277618,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038364444650369306,
      "loss": 4.6772,
      "mean_token_accuracy": 0.24661630243062974,
      "num_tokens": 82548244.0,
      "step": 36010
    },
    {
      "entropy": 5.4360747814178465,
      "epoch": 3.4596541786743518,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038361442329616814,
      "loss": 4.651,
      "mean_token_accuracy": 0.2544337660074234,
      "num_tokens": 82559291.0,
      "step": 36015
    },
    {
      "entropy": 5.419740772247314,
      "epoch": 3.4601344860710856,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038358439756693886,
      "loss": 4.6939,
      "mean_token_accuracy": 0.2557618722319603,
      "num_tokens": 82570929.0,
      "step": 36020
    },
    {
      "entropy": 5.404842948913574,
      "epoch": 3.4606147934678195,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038355436931670225,
      "loss": 4.7011,
      "mean_token_accuracy": 0.2525505155324936,
      "num_tokens": 82583016.0,
      "step": 36025
    },
    {
      "entropy": 5.47595624923706,
      "epoch": 3.4610951008645534,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038352433854615557,
      "loss": 4.7339,
      "mean_token_accuracy": 0.2487241134047508,
      "num_tokens": 82593950.0,
      "step": 36030
    },
    {
      "entropy": 5.3942361831665036,
      "epoch": 3.4615754082612873,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038349430525599606,
      "loss": 4.6223,
      "mean_token_accuracy": 0.2542311534285545,
      "num_tokens": 82605764.0,
      "step": 36035
    },
    {
      "entropy": 5.448212337493897,
      "epoch": 3.462055715658021,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038346426944692075,
      "loss": 4.723,
      "mean_token_accuracy": 0.25119868814945223,
      "num_tokens": 82617336.0,
      "step": 36040
    },
    {
      "entropy": 5.419525575637818,
      "epoch": 3.462536023054755,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003834342311196274,
      "loss": 4.6012,
      "mean_token_accuracy": 0.26600709855556487,
      "num_tokens": 82628999.0,
      "step": 36045
    },
    {
      "entropy": 5.372236347198486,
      "epoch": 3.463016330451489,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038340419027481305,
      "loss": 4.7144,
      "mean_token_accuracy": 0.2555257171392441,
      "num_tokens": 82640141.0,
      "step": 36050
    },
    {
      "entropy": 5.42683219909668,
      "epoch": 3.463496637848223,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003833741469131753,
      "loss": 4.7168,
      "mean_token_accuracy": 0.24762749671936035,
      "num_tokens": 82651320.0,
      "step": 36055
    },
    {
      "entropy": 5.429924535751343,
      "epoch": 3.4639769452449567,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003833441010354117,
      "loss": 4.6962,
      "mean_token_accuracy": 0.2531484544277191,
      "num_tokens": 82662126.0,
      "step": 36060
    },
    {
      "entropy": 5.38309326171875,
      "epoch": 3.4644572526416906,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038331405264221974,
      "loss": 4.6382,
      "mean_token_accuracy": 0.2580956667661667,
      "num_tokens": 82673266.0,
      "step": 36065
    },
    {
      "entropy": 5.488017463684082,
      "epoch": 3.4649375600384245,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003832840017342972,
      "loss": 4.7341,
      "mean_token_accuracy": 0.24522817730903626,
      "num_tokens": 82684797.0,
      "step": 36070
    },
    {
      "entropy": 5.394346427917481,
      "epoch": 3.4654178674351583,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038325394831234164,
      "loss": 4.6467,
      "mean_token_accuracy": 0.2536184385418892,
      "num_tokens": 82697531.0,
      "step": 36075
    },
    {
      "entropy": 5.497323894500733,
      "epoch": 3.465898174831892,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003832238923770508,
      "loss": 4.7224,
      "mean_token_accuracy": 0.25393322855234146,
      "num_tokens": 82708481.0,
      "step": 36080
    },
    {
      "entropy": 5.401921892166138,
      "epoch": 3.4663784822286265,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003831938339291225,
      "loss": 4.6884,
      "mean_token_accuracy": 0.25188650786876676,
      "num_tokens": 82719569.0,
      "step": 36085
    },
    {
      "entropy": 5.513607215881348,
      "epoch": 3.46685878962536,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038316377296925474,
      "loss": 4.7899,
      "mean_token_accuracy": 0.24494290649890899,
      "num_tokens": 82730219.0,
      "step": 36090
    },
    {
      "entropy": 5.479703283309936,
      "epoch": 3.4673390970220943,
      "grad_norm": 1.375,
      "learning_rate": 0.00038313370949814537,
      "loss": 4.7565,
      "mean_token_accuracy": 0.24564829766750335,
      "num_tokens": 82741865.0,
      "step": 36095
    },
    {
      "entropy": 5.417892169952393,
      "epoch": 3.467819404418828,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003831036435164923,
      "loss": 4.6705,
      "mean_token_accuracy": 0.25596632808446884,
      "num_tokens": 82752751.0,
      "step": 36100
    },
    {
      "entropy": 5.455991649627686,
      "epoch": 3.468299711815562,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003830735750249936,
      "loss": 4.7733,
      "mean_token_accuracy": 0.24601333886384963,
      "num_tokens": 82763946.0,
      "step": 36105
    },
    {
      "entropy": 5.419434404373169,
      "epoch": 3.468780019212296,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003830435040243475,
      "loss": 4.6948,
      "mean_token_accuracy": 0.2520451217889786,
      "num_tokens": 82775659.0,
      "step": 36110
    },
    {
      "entropy": 5.453275299072265,
      "epoch": 3.46926032660903,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000383013430515252,
      "loss": 4.7125,
      "mean_token_accuracy": 0.2412989169359207,
      "num_tokens": 82787633.0,
      "step": 36115
    },
    {
      "entropy": 5.4894026756286625,
      "epoch": 3.4697406340057637,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003829833544984054,
      "loss": 4.7457,
      "mean_token_accuracy": 0.25063948035240174,
      "num_tokens": 82799983.0,
      "step": 36120
    },
    {
      "entropy": 5.506343126296997,
      "epoch": 3.4702209414024976,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003829532759745059,
      "loss": 4.7379,
      "mean_token_accuracy": 0.24954410940408706,
      "num_tokens": 82811733.0,
      "step": 36125
    },
    {
      "entropy": 5.518206167221069,
      "epoch": 3.4707012487992315,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038292319494425195,
      "loss": 4.7286,
      "mean_token_accuracy": 0.2537417128682137,
      "num_tokens": 82823555.0,
      "step": 36130
    },
    {
      "entropy": 5.453533601760864,
      "epoch": 3.4711815561959654,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003828931114083419,
      "loss": 4.7059,
      "mean_token_accuracy": 0.25786224007606506,
      "num_tokens": 82835288.0,
      "step": 36135
    },
    {
      "entropy": 5.458036851882935,
      "epoch": 3.4716618635926992,
      "grad_norm": 1.25,
      "learning_rate": 0.000382863025367474,
      "loss": 4.6838,
      "mean_token_accuracy": 0.26428106278181074,
      "num_tokens": 82846298.0,
      "step": 36140
    },
    {
      "entropy": 5.453117418289184,
      "epoch": 3.472142170989433,
      "grad_norm": 1.125,
      "learning_rate": 0.0003828329368223471,
      "loss": 4.7448,
      "mean_token_accuracy": 0.24490631967782975,
      "num_tokens": 82859265.0,
      "step": 36145
    },
    {
      "entropy": 5.406797266006469,
      "epoch": 3.472622478386167,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003828028457736596,
      "loss": 4.7278,
      "mean_token_accuracy": 0.2461111754179001,
      "num_tokens": 82870167.0,
      "step": 36150
    },
    {
      "entropy": 5.488490724563599,
      "epoch": 3.473102785782901,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003827727522221101,
      "loss": 4.7532,
      "mean_token_accuracy": 0.24884352684020997,
      "num_tokens": 82883034.0,
      "step": 36155
    },
    {
      "entropy": 5.449736928939819,
      "epoch": 3.473583093179635,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038274265616839733,
      "loss": 4.629,
      "mean_token_accuracy": 0.2605826437473297,
      "num_tokens": 82894080.0,
      "step": 36160
    },
    {
      "entropy": 5.429360580444336,
      "epoch": 3.4740634005763686,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038271255761322,
      "loss": 4.7084,
      "mean_token_accuracy": 0.25290615409612655,
      "num_tokens": 82905158.0,
      "step": 36165
    },
    {
      "entropy": 5.3868566989898685,
      "epoch": 3.474543707973103,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003826824565572769,
      "loss": 4.6589,
      "mean_token_accuracy": 0.24952100962400436,
      "num_tokens": 82917246.0,
      "step": 36170
    },
    {
      "entropy": 5.372503137588501,
      "epoch": 3.475024015369837,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003826523530012669,
      "loss": 4.6396,
      "mean_token_accuracy": 0.2532766178250313,
      "num_tokens": 82928895.0,
      "step": 36175
    },
    {
      "entropy": 5.435168313980102,
      "epoch": 3.4755043227665707,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000382622246945889,
      "loss": 4.7035,
      "mean_token_accuracy": 0.2546683639287949,
      "num_tokens": 82939239.0,
      "step": 36180
    },
    {
      "entropy": 5.42557282447815,
      "epoch": 3.4759846301633046,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038259213839184205,
      "loss": 4.678,
      "mean_token_accuracy": 0.25251253992319106,
      "num_tokens": 82950457.0,
      "step": 36185
    },
    {
      "entropy": 5.4442707061767575,
      "epoch": 3.4764649375600385,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038256202733982515,
      "loss": 4.6578,
      "mean_token_accuracy": 0.25268775671720506,
      "num_tokens": 82961589.0,
      "step": 36190
    },
    {
      "entropy": 5.401222467422485,
      "epoch": 3.4769452449567724,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003825319137905373,
      "loss": 4.6045,
      "mean_token_accuracy": 0.2612211212515831,
      "num_tokens": 82972397.0,
      "step": 36195
    },
    {
      "entropy": 5.416146516799927,
      "epoch": 3.4774255523535063,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003825017977446778,
      "loss": 4.6869,
      "mean_token_accuracy": 0.24991412758827208,
      "num_tokens": 82984239.0,
      "step": 36200
    },
    {
      "entropy": 5.413333368301392,
      "epoch": 3.47790585975024,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038247167920294574,
      "loss": 4.7398,
      "mean_token_accuracy": 0.2571500241756439,
      "num_tokens": 82996544.0,
      "step": 36205
    },
    {
      "entropy": 5.57362060546875,
      "epoch": 3.478386167146974,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038244155816604037,
      "loss": 4.8525,
      "mean_token_accuracy": 0.24130430370569228,
      "num_tokens": 83009002.0,
      "step": 36210
    },
    {
      "entropy": 5.4199052333831785,
      "epoch": 3.478866474543708,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003824114346346611,
      "loss": 4.6519,
      "mean_token_accuracy": 0.24797742515802385,
      "num_tokens": 83021225.0,
      "step": 36215
    },
    {
      "entropy": 5.439920091629029,
      "epoch": 3.479346781940442,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003823813086095073,
      "loss": 4.7588,
      "mean_token_accuracy": 0.25253650099039077,
      "num_tokens": 83032682.0,
      "step": 36220
    },
    {
      "entropy": 5.434450244903564,
      "epoch": 3.4798270893371757,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038235118009127833,
      "loss": 4.6561,
      "mean_token_accuracy": 0.25776110142469405,
      "num_tokens": 83045035.0,
      "step": 36225
    },
    {
      "entropy": 5.488147211074829,
      "epoch": 3.4803073967339095,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003823210490806737,
      "loss": 4.8526,
      "mean_token_accuracy": 0.23997996002435684,
      "num_tokens": 83056724.0,
      "step": 36230
    },
    {
      "entropy": 5.4747861385345455,
      "epoch": 3.4807877041306434,
      "grad_norm": 1.15625,
      "learning_rate": 0.000382290915578393,
      "loss": 4.7527,
      "mean_token_accuracy": 0.2512222409248352,
      "num_tokens": 83067736.0,
      "step": 36235
    },
    {
      "entropy": 5.437187814712525,
      "epoch": 3.4812680115273773,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003822607795851359,
      "loss": 4.6633,
      "mean_token_accuracy": 0.2626345783472061,
      "num_tokens": 83079367.0,
      "step": 36240
    },
    {
      "entropy": 5.529883241653442,
      "epoch": 3.4817483189241116,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003822306411016019,
      "loss": 4.7558,
      "mean_token_accuracy": 0.24739763438701629,
      "num_tokens": 83091138.0,
      "step": 36245
    },
    {
      "entropy": 5.390830993652344,
      "epoch": 3.4822286263208455,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003822005001284909,
      "loss": 4.6199,
      "mean_token_accuracy": 0.2639778897166252,
      "num_tokens": 83101627.0,
      "step": 36250
    },
    {
      "entropy": 5.431215429306031,
      "epoch": 3.4827089337175794,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003821703566665026,
      "loss": 4.7063,
      "mean_token_accuracy": 0.2558892294764519,
      "num_tokens": 83112569.0,
      "step": 36255
    },
    {
      "entropy": 5.395221757888794,
      "epoch": 3.4831892411143133,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003821402107163368,
      "loss": 4.6063,
      "mean_token_accuracy": 0.25979232490062715,
      "num_tokens": 83124019.0,
      "step": 36260
    },
    {
      "entropy": 5.438871812820435,
      "epoch": 3.483669548511047,
      "grad_norm": 1.125,
      "learning_rate": 0.0003821100622786935,
      "loss": 4.6696,
      "mean_token_accuracy": 0.25903777182102206,
      "num_tokens": 83135032.0,
      "step": 36265
    },
    {
      "entropy": 5.367235994338989,
      "epoch": 3.484149855907781,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038207991135427255,
      "loss": 4.5741,
      "mean_token_accuracy": 0.25253976583480836,
      "num_tokens": 83147362.0,
      "step": 36270
    },
    {
      "entropy": 5.437580299377442,
      "epoch": 3.484630163304515,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000382049757943774,
      "loss": 4.6927,
      "mean_token_accuracy": 0.25320435911417005,
      "num_tokens": 83158531.0,
      "step": 36275
    },
    {
      "entropy": 5.490598392486572,
      "epoch": 3.485110470701249,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038201960204789796,
      "loss": 4.7993,
      "mean_token_accuracy": 0.24569968432188033,
      "num_tokens": 83169565.0,
      "step": 36280
    },
    {
      "entropy": 5.457918930053711,
      "epoch": 3.4855907780979827,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003819894436673446,
      "loss": 4.7494,
      "mean_token_accuracy": 0.2503797024488449,
      "num_tokens": 83181084.0,
      "step": 36285
    },
    {
      "entropy": 5.474738025665284,
      "epoch": 3.4860710854947166,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038195928280281406,
      "loss": 4.7432,
      "mean_token_accuracy": 0.23850466012954713,
      "num_tokens": 83193477.0,
      "step": 36290
    },
    {
      "entropy": 5.497291851043701,
      "epoch": 3.4865513928914504,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003819291194550066,
      "loss": 4.7328,
      "mean_token_accuracy": 0.2520383149385452,
      "num_tokens": 83204408.0,
      "step": 36295
    },
    {
      "entropy": 5.511803913116455,
      "epoch": 3.4870317002881843,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003818989536246224,
      "loss": 4.7886,
      "mean_token_accuracy": 0.2518546089529991,
      "num_tokens": 83215152.0,
      "step": 36300
    },
    {
      "entropy": 5.36034574508667,
      "epoch": 3.487512007684918,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038186878531236196,
      "loss": 4.6562,
      "mean_token_accuracy": 0.2554333284497261,
      "num_tokens": 83227094.0,
      "step": 36305
    },
    {
      "entropy": 5.4376849174499515,
      "epoch": 3.487992315081652,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038183861451892567,
      "loss": 4.7183,
      "mean_token_accuracy": 0.2532540559768677,
      "num_tokens": 83238149.0,
      "step": 36310
    },
    {
      "entropy": 5.444032192230225,
      "epoch": 3.488472622478386,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000381808441245014,
      "loss": 4.7483,
      "mean_token_accuracy": 0.2471893236041069,
      "num_tokens": 83249991.0,
      "step": 36315
    },
    {
      "entropy": 5.427116775512696,
      "epoch": 3.4889529298751203,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003817782654913274,
      "loss": 4.7051,
      "mean_token_accuracy": 0.25286481380462644,
      "num_tokens": 83262987.0,
      "step": 36320
    },
    {
      "entropy": 5.48630166053772,
      "epoch": 3.489433237271854,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003817480872585667,
      "loss": 4.7328,
      "mean_token_accuracy": 0.24932989329099656,
      "num_tokens": 83274155.0,
      "step": 36325
    },
    {
      "entropy": 5.5203040599822994,
      "epoch": 3.489913544668588,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038171790654743234,
      "loss": 4.7156,
      "mean_token_accuracy": 0.2498101532459259,
      "num_tokens": 83286314.0,
      "step": 36330
    },
    {
      "entropy": 5.440716981887817,
      "epoch": 3.490393852065322,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000381687723358625,
      "loss": 4.7374,
      "mean_token_accuracy": 0.25142858922481537,
      "num_tokens": 83297435.0,
      "step": 36335
    },
    {
      "entropy": 5.390204620361328,
      "epoch": 3.490874159462056,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003816575376928456,
      "loss": 4.6001,
      "mean_token_accuracy": 0.25878116935491563,
      "num_tokens": 83308248.0,
      "step": 36340
    },
    {
      "entropy": 5.436470556259155,
      "epoch": 3.4913544668587897,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038162734955079483,
      "loss": 4.6742,
      "mean_token_accuracy": 0.2516919568181038,
      "num_tokens": 83319446.0,
      "step": 36345
    },
    {
      "entropy": 5.542528820037842,
      "epoch": 3.4918347742555236,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003815971589331736,
      "loss": 4.7798,
      "mean_token_accuracy": 0.24855855852365494,
      "num_tokens": 83330340.0,
      "step": 36350
    },
    {
      "entropy": 5.380068635940551,
      "epoch": 3.4923150816522575,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003815669658406829,
      "loss": 4.6415,
      "mean_token_accuracy": 0.2548734411597252,
      "num_tokens": 83342338.0,
      "step": 36355
    },
    {
      "entropy": 5.423019981384277,
      "epoch": 3.4927953890489913,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038153677027402374,
      "loss": 4.672,
      "mean_token_accuracy": 0.25645369440317156,
      "num_tokens": 83353272.0,
      "step": 36360
    },
    {
      "entropy": 5.478996610641479,
      "epoch": 3.493275696445725,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038150657223389703,
      "loss": 4.7537,
      "mean_token_accuracy": 0.25019470155239104,
      "num_tokens": 83364691.0,
      "step": 36365
    },
    {
      "entropy": 5.554847526550293,
      "epoch": 3.493756003842459,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038147637172100397,
      "loss": 4.6987,
      "mean_token_accuracy": 0.25069051384925845,
      "num_tokens": 83375110.0,
      "step": 36370
    },
    {
      "entropy": 5.383763265609741,
      "epoch": 3.494236311239193,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003814461687360459,
      "loss": 4.6917,
      "mean_token_accuracy": 0.24474346935749053,
      "num_tokens": 83387540.0,
      "step": 36375
    },
    {
      "entropy": 5.488369369506836,
      "epoch": 3.494716618635927,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038141596327972375,
      "loss": 4.7674,
      "mean_token_accuracy": 0.2447591543197632,
      "num_tokens": 83398895.0,
      "step": 36380
    },
    {
      "entropy": 5.430051946640015,
      "epoch": 3.4951969260326607,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003813857553527389,
      "loss": 4.6693,
      "mean_token_accuracy": 0.25104349106550217,
      "num_tokens": 83410712.0,
      "step": 36385
    },
    {
      "entropy": 5.483949756622314,
      "epoch": 3.4956772334293946,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038135554495579273,
      "loss": 4.7485,
      "mean_token_accuracy": 0.24416958838701247,
      "num_tokens": 83421168.0,
      "step": 36390
    },
    {
      "entropy": 5.4750292778015135,
      "epoch": 3.496157540826129,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003813253320895866,
      "loss": 4.7569,
      "mean_token_accuracy": 0.24521174132823945,
      "num_tokens": 83432634.0,
      "step": 36395
    },
    {
      "entropy": 5.448449563980103,
      "epoch": 3.4966378482228624,
      "grad_norm": 1.25,
      "learning_rate": 0.000381295116754822,
      "loss": 4.7144,
      "mean_token_accuracy": 0.247922120988369,
      "num_tokens": 83444024.0,
      "step": 36400
    },
    {
      "entropy": 5.502452993392945,
      "epoch": 3.4971181556195967,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038126489895220047,
      "loss": 4.762,
      "mean_token_accuracy": 0.2426598086953163,
      "num_tokens": 83455223.0,
      "step": 36405
    },
    {
      "entropy": 5.56572995185852,
      "epoch": 3.4975984630163306,
      "grad_norm": 1.609375,
      "learning_rate": 0.00038123467868242345,
      "loss": 4.7842,
      "mean_token_accuracy": 0.2446329414844513,
      "num_tokens": 83467295.0,
      "step": 36410
    },
    {
      "entropy": 5.528389739990234,
      "epoch": 3.4980787704130645,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038120445594619264,
      "loss": 4.8261,
      "mean_token_accuracy": 0.2445413276553154,
      "num_tokens": 83478445.0,
      "step": 36415
    },
    {
      "entropy": 5.456834077835083,
      "epoch": 3.4985590778097984,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003811742307442098,
      "loss": 4.7383,
      "mean_token_accuracy": 0.2470931127667427,
      "num_tokens": 83490257.0,
      "step": 36420
    },
    {
      "entropy": 5.438460159301758,
      "epoch": 3.4990393852065322,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038114400307717657,
      "loss": 4.6752,
      "mean_token_accuracy": 0.2544380620121956,
      "num_tokens": 83501773.0,
      "step": 36425
    },
    {
      "entropy": 5.459212160110473,
      "epoch": 3.499519692603266,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003811137729457947,
      "loss": 4.6893,
      "mean_token_accuracy": 0.2588293254375458,
      "num_tokens": 83513654.0,
      "step": 36430
    },
    {
      "entropy": 5.430193138122559,
      "epoch": 3.5,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003810835403507662,
      "loss": 4.7043,
      "mean_token_accuracy": 0.24802484214305878,
      "num_tokens": 83525251.0,
      "step": 36435
    },
    {
      "entropy": 5.46633677482605,
      "epoch": 3.500480307396734,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038105330529279285,
      "loss": 4.7619,
      "mean_token_accuracy": 0.2514957919716835,
      "num_tokens": 83536937.0,
      "step": 36440
    },
    {
      "entropy": 5.463500165939331,
      "epoch": 3.5009606147934678,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003810230677725767,
      "loss": 4.708,
      "mean_token_accuracy": 0.24798151850700378,
      "num_tokens": 83549686.0,
      "step": 36445
    },
    {
      "entropy": 5.500936603546142,
      "epoch": 3.5014409221902016,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038099282779081975,
      "loss": 4.7265,
      "mean_token_accuracy": 0.24922872930765153,
      "num_tokens": 83561774.0,
      "step": 36450
    },
    {
      "entropy": 5.471020317077636,
      "epoch": 3.5019212295869355,
      "grad_norm": 1.25,
      "learning_rate": 0.00038096258534822404,
      "loss": 4.7279,
      "mean_token_accuracy": 0.24963091611862182,
      "num_tokens": 83573236.0,
      "step": 36455
    },
    {
      "entropy": 5.432536220550537,
      "epoch": 3.5024015369836694,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038093234044549174,
      "loss": 4.685,
      "mean_token_accuracy": 0.2498992547392845,
      "num_tokens": 83584899.0,
      "step": 36460
    },
    {
      "entropy": 5.459561920166015,
      "epoch": 3.5028818443804033,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038090209308332503,
      "loss": 4.7271,
      "mean_token_accuracy": 0.2534923151135445,
      "num_tokens": 83596880.0,
      "step": 36465
    },
    {
      "entropy": 5.434033918380737,
      "epoch": 3.5033621517771376,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038087184326242627,
      "loss": 4.7288,
      "mean_token_accuracy": 0.25024732351303103,
      "num_tokens": 83608747.0,
      "step": 36470
    },
    {
      "entropy": 5.436532497406006,
      "epoch": 3.503842459173871,
      "grad_norm": 1.25,
      "learning_rate": 0.0003808415909834976,
      "loss": 4.7233,
      "mean_token_accuracy": 0.2545244634151459,
      "num_tokens": 83619971.0,
      "step": 36475
    },
    {
      "entropy": 5.523594045639038,
      "epoch": 3.5043227665706054,
      "grad_norm": 1.25,
      "learning_rate": 0.00038081133624724147,
      "loss": 4.8114,
      "mean_token_accuracy": 0.24909809678792955,
      "num_tokens": 83631030.0,
      "step": 36480
    },
    {
      "entropy": 5.460321712493896,
      "epoch": 3.5048030739673393,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038078107905436035,
      "loss": 4.6693,
      "mean_token_accuracy": 0.2537594273686409,
      "num_tokens": 83642854.0,
      "step": 36485
    },
    {
      "entropy": 5.4888267993927,
      "epoch": 3.505283381364073,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038075081940555666,
      "loss": 4.6845,
      "mean_token_accuracy": 0.253227935731411,
      "num_tokens": 83653669.0,
      "step": 36490
    },
    {
      "entropy": 5.461727142333984,
      "epoch": 3.505763688760807,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038072055730153297,
      "loss": 4.7026,
      "mean_token_accuracy": 0.2580073356628418,
      "num_tokens": 83665098.0,
      "step": 36495
    },
    {
      "entropy": 5.400018262863159,
      "epoch": 3.506243996157541,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038069029274299184,
      "loss": 4.7151,
      "mean_token_accuracy": 0.2523759454488754,
      "num_tokens": 83677494.0,
      "step": 36500
    },
    {
      "entropy": 5.4288147449493405,
      "epoch": 3.506724303554275,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003806600257306359,
      "loss": 4.6469,
      "mean_token_accuracy": 0.26018828302621844,
      "num_tokens": 83689128.0,
      "step": 36505
    },
    {
      "entropy": 5.368082761764526,
      "epoch": 3.5072046109510087,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038062975626516806,
      "loss": 4.6396,
      "mean_token_accuracy": 0.25730445981025696,
      "num_tokens": 83700113.0,
      "step": 36510
    },
    {
      "entropy": 5.481581115722657,
      "epoch": 3.5076849183477425,
      "grad_norm": 1.4375,
      "learning_rate": 0.00038059948434729086,
      "loss": 4.6797,
      "mean_token_accuracy": 0.24951327592134476,
      "num_tokens": 83709549.0,
      "step": 36515
    },
    {
      "entropy": 5.453924655914307,
      "epoch": 3.5081652257444764,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038056920997770713,
      "loss": 4.7042,
      "mean_token_accuracy": 0.251926551759243,
      "num_tokens": 83720912.0,
      "step": 36520
    },
    {
      "entropy": 5.42933840751648,
      "epoch": 3.5086455331412103,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038053893315711997,
      "loss": 4.7324,
      "mean_token_accuracy": 0.24318698048591614,
      "num_tokens": 83732885.0,
      "step": 36525
    },
    {
      "entropy": 5.491782855987549,
      "epoch": 3.509125840537944,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000380508653886232,
      "loss": 4.7433,
      "mean_token_accuracy": 0.2467532768845558,
      "num_tokens": 83744521.0,
      "step": 36530
    },
    {
      "entropy": 5.524216890335083,
      "epoch": 3.509606147934678,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003804783721657465,
      "loss": 4.7401,
      "mean_token_accuracy": 0.2579554483294487,
      "num_tokens": 83756184.0,
      "step": 36535
    },
    {
      "entropy": 5.4300150871276855,
      "epoch": 3.510086455331412,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038044808799636635,
      "loss": 4.6933,
      "mean_token_accuracy": 0.2595406100153923,
      "num_tokens": 83767643.0,
      "step": 36540
    },
    {
      "entropy": 5.380952835083008,
      "epoch": 3.5105667627281463,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003804178013787947,
      "loss": 4.6196,
      "mean_token_accuracy": 0.2616360321640968,
      "num_tokens": 83778326.0,
      "step": 36545
    },
    {
      "entropy": 5.444450426101684,
      "epoch": 3.5110470701248797,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038038751231373477,
      "loss": 4.7942,
      "mean_token_accuracy": 0.2494828775525093,
      "num_tokens": 83790396.0,
      "step": 36550
    },
    {
      "entropy": 5.4545072555542,
      "epoch": 3.511527377521614,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038035722080188974,
      "loss": 4.721,
      "mean_token_accuracy": 0.25568192452192307,
      "num_tokens": 83803199.0,
      "step": 36555
    },
    {
      "entropy": 5.448311424255371,
      "epoch": 3.5120076849183475,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038032692684396286,
      "loss": 4.701,
      "mean_token_accuracy": 0.25126462876796724,
      "num_tokens": 83815184.0,
      "step": 36560
    },
    {
      "entropy": 5.590494680404663,
      "epoch": 3.512487992315082,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003802966304406575,
      "loss": 4.8037,
      "mean_token_accuracy": 0.24605314433574677,
      "num_tokens": 83827627.0,
      "step": 36565
    },
    {
      "entropy": 5.45779709815979,
      "epoch": 3.5129682997118157,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038026633159267707,
      "loss": 4.6965,
      "mean_token_accuracy": 0.25480577200651167,
      "num_tokens": 83840771.0,
      "step": 36570
    },
    {
      "entropy": 5.439200687408447,
      "epoch": 3.5134486071085496,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038023603030072503,
      "loss": 4.6857,
      "mean_token_accuracy": 0.2564528092741966,
      "num_tokens": 83851311.0,
      "step": 36575
    },
    {
      "entropy": 5.431161165237427,
      "epoch": 3.5139289145052834,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003802057265655048,
      "loss": 4.7372,
      "mean_token_accuracy": 0.2535404905676842,
      "num_tokens": 83863887.0,
      "step": 36580
    },
    {
      "entropy": 5.514787197113037,
      "epoch": 3.5144092219020173,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038017542038772003,
      "loss": 4.7071,
      "mean_token_accuracy": 0.24963390380144118,
      "num_tokens": 83874183.0,
      "step": 36585
    },
    {
      "entropy": 5.471960830688476,
      "epoch": 3.514889529298751,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038014511176807426,
      "loss": 4.7304,
      "mean_token_accuracy": 0.25249582827091216,
      "num_tokens": 83885973.0,
      "step": 36590
    },
    {
      "entropy": 5.398511600494385,
      "epoch": 3.515369836695485,
      "grad_norm": 1.125,
      "learning_rate": 0.00038011480070727124,
      "loss": 4.6432,
      "mean_token_accuracy": 0.2518604189157486,
      "num_tokens": 83897808.0,
      "step": 36595
    },
    {
      "entropy": 5.466485643386841,
      "epoch": 3.515850144092219,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003800844872060147,
      "loss": 4.719,
      "mean_token_accuracy": 0.25101846307516096,
      "num_tokens": 83909876.0,
      "step": 36600
    },
    {
      "entropy": 5.400876426696778,
      "epoch": 3.516330451488953,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038005417126500836,
      "loss": 4.6853,
      "mean_token_accuracy": 0.25138611942529676,
      "num_tokens": 83921181.0,
      "step": 36605
    },
    {
      "entropy": 5.511036443710327,
      "epoch": 3.5168107588856867,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038002385288495613,
      "loss": 4.7125,
      "mean_token_accuracy": 0.24740612655878066,
      "num_tokens": 83932946.0,
      "step": 36610
    },
    {
      "entropy": 5.420216274261475,
      "epoch": 3.5172910662824206,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003799935320665619,
      "loss": 4.6794,
      "mean_token_accuracy": 0.2541394293308258,
      "num_tokens": 83945439.0,
      "step": 36615
    },
    {
      "entropy": 5.467505884170532,
      "epoch": 3.517771373679155,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037996320881052956,
      "loss": 4.7431,
      "mean_token_accuracy": 0.2482215642929077,
      "num_tokens": 83956979.0,
      "step": 36620
    },
    {
      "entropy": 5.55766863822937,
      "epoch": 3.5182516810758884,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037993288311756327,
      "loss": 4.7902,
      "mean_token_accuracy": 0.2412840723991394,
      "num_tokens": 83968371.0,
      "step": 36625
    },
    {
      "entropy": 5.516800785064698,
      "epoch": 3.5187319884726227,
      "grad_norm": 1.203125,
      "learning_rate": 0.000379902554988367,
      "loss": 4.704,
      "mean_token_accuracy": 0.25260202437639234,
      "num_tokens": 83980815.0,
      "step": 36630
    },
    {
      "entropy": 5.397502756118774,
      "epoch": 3.519212295869356,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003798722244236449,
      "loss": 4.6818,
      "mean_token_accuracy": 0.2531205967068672,
      "num_tokens": 83991764.0,
      "step": 36635
    },
    {
      "entropy": 5.550442790985107,
      "epoch": 3.5196926032660905,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003798418914241011,
      "loss": 4.8023,
      "mean_token_accuracy": 0.2455929473042488,
      "num_tokens": 84003423.0,
      "step": 36640
    },
    {
      "entropy": 5.443557357788086,
      "epoch": 3.5201729106628243,
      "grad_norm": 1.25,
      "learning_rate": 0.00037981155599043997,
      "loss": 4.6993,
      "mean_token_accuracy": 0.25185528844594957,
      "num_tokens": 84014431.0,
      "step": 36645
    },
    {
      "entropy": 5.470828199386597,
      "epoch": 3.520653218059558,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003797812181233657,
      "loss": 4.6987,
      "mean_token_accuracy": 0.2500268265604973,
      "num_tokens": 84026024.0,
      "step": 36650
    },
    {
      "entropy": 5.494072961807251,
      "epoch": 3.521133525456292,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037975087782358264,
      "loss": 4.7286,
      "mean_token_accuracy": 0.25210876762866974,
      "num_tokens": 84037533.0,
      "step": 36655
    },
    {
      "entropy": 5.547641324996948,
      "epoch": 3.521613832853026,
      "grad_norm": 1.25,
      "learning_rate": 0.0003797205350917953,
      "loss": 4.7537,
      "mean_token_accuracy": 0.25201797634363177,
      "num_tokens": 84049065.0,
      "step": 36660
    },
    {
      "entropy": 5.467063474655151,
      "epoch": 3.52209414024976,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037969018992870814,
      "loss": 4.677,
      "mean_token_accuracy": 0.25756129920482634,
      "num_tokens": 84060339.0,
      "step": 36665
    },
    {
      "entropy": 5.4384137153625485,
      "epoch": 3.5225744476464937,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037965984233502553,
      "loss": 4.732,
      "mean_token_accuracy": 0.24951465725898742,
      "num_tokens": 84071470.0,
      "step": 36670
    },
    {
      "entropy": 5.51066484451294,
      "epoch": 3.5230547550432276,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003796294923114522,
      "loss": 4.7715,
      "mean_token_accuracy": 0.23964198231697081,
      "num_tokens": 84082831.0,
      "step": 36675
    },
    {
      "entropy": 5.463075113296509,
      "epoch": 3.5235350624399615,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037959913985869276,
      "loss": 4.6948,
      "mean_token_accuracy": 0.25141708105802535,
      "num_tokens": 84094547.0,
      "step": 36680
    },
    {
      "entropy": 5.476634311676025,
      "epoch": 3.5240153698366954,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037956878497745185,
      "loss": 4.6563,
      "mean_token_accuracy": 0.2535689607262611,
      "num_tokens": 84106117.0,
      "step": 36685
    },
    {
      "entropy": 5.412801027297974,
      "epoch": 3.5244956772334293,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003795384276684343,
      "loss": 4.6238,
      "mean_token_accuracy": 0.25972455739974976,
      "num_tokens": 84116654.0,
      "step": 36690
    },
    {
      "entropy": 5.44591383934021,
      "epoch": 3.524975984630163,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003795080679323449,
      "loss": 4.7394,
      "mean_token_accuracy": 0.2509776219725609,
      "num_tokens": 84128217.0,
      "step": 36695
    },
    {
      "entropy": 5.487013721466065,
      "epoch": 3.525456292026897,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003794777057698884,
      "loss": 4.718,
      "mean_token_accuracy": 0.2460033804178238,
      "num_tokens": 84139638.0,
      "step": 36700
    },
    {
      "entropy": 5.551719522476196,
      "epoch": 3.5259365994236314,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003794473411817699,
      "loss": 4.7808,
      "mean_token_accuracy": 0.2477921202778816,
      "num_tokens": 84150945.0,
      "step": 36705
    },
    {
      "entropy": 5.444630813598633,
      "epoch": 3.526416906820365,
      "grad_norm": 1.25,
      "learning_rate": 0.0003794169741686942,
      "loss": 4.6255,
      "mean_token_accuracy": 0.25777965486049653,
      "num_tokens": 84163443.0,
      "step": 36710
    },
    {
      "entropy": 5.518018341064453,
      "epoch": 3.526897214217099,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003793866047313665,
      "loss": 4.8074,
      "mean_token_accuracy": 0.24586497247219086,
      "num_tokens": 84174291.0,
      "step": 36715
    },
    {
      "entropy": 5.422544765472412,
      "epoch": 3.527377521613833,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037935623287049173,
      "loss": 4.6458,
      "mean_token_accuracy": 0.25914565324783323,
      "num_tokens": 84185331.0,
      "step": 36720
    },
    {
      "entropy": 5.355278635025025,
      "epoch": 3.527857829010567,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037932585858677515,
      "loss": 4.6583,
      "mean_token_accuracy": 0.25656032711267474,
      "num_tokens": 84196930.0,
      "step": 36725
    },
    {
      "entropy": 5.385110139846802,
      "epoch": 3.5283381364073008,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037929548188092186,
      "loss": 4.6534,
      "mean_token_accuracy": 0.26053203642368317,
      "num_tokens": 84208635.0,
      "step": 36730
    },
    {
      "entropy": 5.47473177909851,
      "epoch": 3.5288184438040346,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037926510275363725,
      "loss": 4.6952,
      "mean_token_accuracy": 0.2554880201816559,
      "num_tokens": 84219392.0,
      "step": 36735
    },
    {
      "entropy": 5.399322748184204,
      "epoch": 3.5292987512007685,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003792347212056265,
      "loss": 4.6362,
      "mean_token_accuracy": 0.2595602676272392,
      "num_tokens": 84230540.0,
      "step": 36740
    },
    {
      "entropy": 5.5062366962432865,
      "epoch": 3.5297790585975024,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037920433723759506,
      "loss": 4.7027,
      "mean_token_accuracy": 0.2470795214176178,
      "num_tokens": 84241339.0,
      "step": 36745
    },
    {
      "entropy": 5.483756256103516,
      "epoch": 3.5302593659942363,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003791739508502483,
      "loss": 4.7128,
      "mean_token_accuracy": 0.25058208853006364,
      "num_tokens": 84252658.0,
      "step": 36750
    },
    {
      "entropy": 5.391508913040161,
      "epoch": 3.53073967339097,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037914356204429183,
      "loss": 4.7023,
      "mean_token_accuracy": 0.24640007615089415,
      "num_tokens": 84264522.0,
      "step": 36755
    },
    {
      "entropy": 5.491046953201294,
      "epoch": 3.531219980787704,
      "grad_norm": 1.25,
      "learning_rate": 0.00037911317082043103,
      "loss": 4.7561,
      "mean_token_accuracy": 0.25346573293209074,
      "num_tokens": 84276330.0,
      "step": 36760
    },
    {
      "entropy": 5.541143369674683,
      "epoch": 3.531700288184438,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003790827771793715,
      "loss": 4.7964,
      "mean_token_accuracy": 0.24483138769865037,
      "num_tokens": 84287288.0,
      "step": 36765
    },
    {
      "entropy": 5.431784582138062,
      "epoch": 3.532180595581172,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000379052381121819,
      "loss": 4.6602,
      "mean_token_accuracy": 0.2526552274823189,
      "num_tokens": 84297472.0,
      "step": 36770
    },
    {
      "entropy": 5.374803352355957,
      "epoch": 3.5326609029779057,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003790219826484792,
      "loss": 4.6479,
      "mean_token_accuracy": 0.25969946533441546,
      "num_tokens": 84308726.0,
      "step": 36775
    },
    {
      "entropy": 5.493534135818481,
      "epoch": 3.53314121037464,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037899158176005784,
      "loss": 4.7975,
      "mean_token_accuracy": 0.24403763711452484,
      "num_tokens": 84320439.0,
      "step": 36780
    },
    {
      "entropy": 5.557023906707764,
      "epoch": 3.5336215177713735,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037896117845726067,
      "loss": 4.7927,
      "mean_token_accuracy": 0.2507328882813454,
      "num_tokens": 84331321.0,
      "step": 36785
    },
    {
      "entropy": 5.557481908798218,
      "epoch": 3.534101825168108,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037893077274079375,
      "loss": 4.8343,
      "mean_token_accuracy": 0.24213165938854217,
      "num_tokens": 84344095.0,
      "step": 36790
    },
    {
      "entropy": 5.474826240539551,
      "epoch": 3.5345821325648417,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037890036461136285,
      "loss": 4.765,
      "mean_token_accuracy": 0.2435059890151024,
      "num_tokens": 84355254.0,
      "step": 36795
    },
    {
      "entropy": 5.4023706912994385,
      "epoch": 3.5350624399615755,
      "grad_norm": 1.140625,
      "learning_rate": 0.000378869954069674,
      "loss": 4.5992,
      "mean_token_accuracy": 0.26474927067756654,
      "num_tokens": 84365739.0,
      "step": 36800
    },
    {
      "entropy": 5.438440561294556,
      "epoch": 3.5355427473583094,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003788395411164332,
      "loss": 4.7674,
      "mean_token_accuracy": 0.24816315919160842,
      "num_tokens": 84377881.0,
      "step": 36805
    },
    {
      "entropy": 5.471712875366211,
      "epoch": 3.5360230547550433,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037880912575234666,
      "loss": 4.7053,
      "mean_token_accuracy": 0.2519576892256737,
      "num_tokens": 84388748.0,
      "step": 36810
    },
    {
      "entropy": 5.424633598327636,
      "epoch": 3.536503362151777,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003787787079781205,
      "loss": 4.6915,
      "mean_token_accuracy": 0.253153894841671,
      "num_tokens": 84399056.0,
      "step": 36815
    },
    {
      "entropy": 5.492865943908692,
      "epoch": 3.536983669548511,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003787482877944608,
      "loss": 4.741,
      "mean_token_accuracy": 0.250135101377964,
      "num_tokens": 84411441.0,
      "step": 36820
    },
    {
      "entropy": 5.508409214019776,
      "epoch": 3.537463976945245,
      "grad_norm": 1.21875,
      "learning_rate": 0.000378717865202074,
      "loss": 4.7935,
      "mean_token_accuracy": 0.24111467003822326,
      "num_tokens": 84423430.0,
      "step": 36825
    },
    {
      "entropy": 5.568131017684936,
      "epoch": 3.537944284341979,
      "grad_norm": 1.375,
      "learning_rate": 0.00037868744020166635,
      "loss": 4.8238,
      "mean_token_accuracy": 0.24189769327640534,
      "num_tokens": 84435002.0,
      "step": 36830
    },
    {
      "entropy": 5.470731019973755,
      "epoch": 3.5384245917387127,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037865701279394414,
      "loss": 4.6949,
      "mean_token_accuracy": 0.25386230647563934,
      "num_tokens": 84446858.0,
      "step": 36835
    },
    {
      "entropy": 5.443322944641113,
      "epoch": 3.5389048991354466,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037862658297961393,
      "loss": 4.6929,
      "mean_token_accuracy": 0.25952042937278746,
      "num_tokens": 84457603.0,
      "step": 36840
    },
    {
      "entropy": 5.487119960784912,
      "epoch": 3.5393852065321805,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003785961507593823,
      "loss": 4.7465,
      "mean_token_accuracy": 0.24943623095750808,
      "num_tokens": 84468755.0,
      "step": 36845
    },
    {
      "entropy": 5.439061832427979,
      "epoch": 3.5398655139289144,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003785657161339555,
      "loss": 4.6919,
      "mean_token_accuracy": 0.256575471162796,
      "num_tokens": 84480248.0,
      "step": 36850
    },
    {
      "entropy": 5.472072124481201,
      "epoch": 3.5403458213256487,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037853527910404033,
      "loss": 4.7843,
      "mean_token_accuracy": 0.24227406233549117,
      "num_tokens": 84493290.0,
      "step": 36855
    },
    {
      "entropy": 5.448037958145141,
      "epoch": 3.540826128722382,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003785048396703435,
      "loss": 4.7218,
      "mean_token_accuracy": 0.24319998174905777,
      "num_tokens": 84505353.0,
      "step": 36860
    },
    {
      "entropy": 5.565755128860474,
      "epoch": 3.5413064361191164,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037847439783357165,
      "loss": 4.8504,
      "mean_token_accuracy": 0.2467345491051674,
      "num_tokens": 84517871.0,
      "step": 36865
    },
    {
      "entropy": 5.412446165084839,
      "epoch": 3.54178674351585,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003784439535944314,
      "loss": 4.6055,
      "mean_token_accuracy": 0.2604482039809227,
      "num_tokens": 84528513.0,
      "step": 36870
    },
    {
      "entropy": 5.594400119781494,
      "epoch": 3.542267050912584,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003784135069536298,
      "loss": 4.8229,
      "mean_token_accuracy": 0.23678039014339447,
      "num_tokens": 84539990.0,
      "step": 36875
    },
    {
      "entropy": 5.463593292236328,
      "epoch": 3.542747358309318,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037838305791187363,
      "loss": 4.6537,
      "mean_token_accuracy": 0.2557010903954506,
      "num_tokens": 84551218.0,
      "step": 36880
    },
    {
      "entropy": 5.4172944068908695,
      "epoch": 3.543227665706052,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037835260646986986,
      "loss": 4.6134,
      "mean_token_accuracy": 0.2576006084680557,
      "num_tokens": 84561672.0,
      "step": 36885
    },
    {
      "entropy": 5.369071960449219,
      "epoch": 3.543707973102786,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037832215262832545,
      "loss": 4.6957,
      "mean_token_accuracy": 0.2550359070301056,
      "num_tokens": 84574115.0,
      "step": 36890
    },
    {
      "entropy": 5.473688125610352,
      "epoch": 3.5441882804995197,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003782916963879474,
      "loss": 4.7338,
      "mean_token_accuracy": 0.244407719373703,
      "num_tokens": 84586014.0,
      "step": 36895
    },
    {
      "entropy": 5.448237180709839,
      "epoch": 3.5446685878962536,
      "grad_norm": 1.1875,
      "learning_rate": 0.000378261237749443,
      "loss": 4.6796,
      "mean_token_accuracy": 0.2564145907759666,
      "num_tokens": 84597720.0,
      "step": 36900
    },
    {
      "entropy": 5.3816718578338625,
      "epoch": 3.5451488952929875,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003782307767135192,
      "loss": 4.659,
      "mean_token_accuracy": 0.26075489073991776,
      "num_tokens": 84609610.0,
      "step": 36905
    },
    {
      "entropy": 5.45015778541565,
      "epoch": 3.5456292026897214,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037820031328088325,
      "loss": 4.6864,
      "mean_token_accuracy": 0.25659467428922655,
      "num_tokens": 84619806.0,
      "step": 36910
    },
    {
      "entropy": 5.3795360088348385,
      "epoch": 3.5461095100864553,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003781698474522425,
      "loss": 4.6434,
      "mean_token_accuracy": 0.2573245480656624,
      "num_tokens": 84631030.0,
      "step": 36915
    },
    {
      "entropy": 5.410390615463257,
      "epoch": 3.546589817483189,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037813937922830435,
      "loss": 4.7333,
      "mean_token_accuracy": 0.24768309146165848,
      "num_tokens": 84643914.0,
      "step": 36920
    },
    {
      "entropy": 5.457704210281372,
      "epoch": 3.547070124879923,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037810890860977595,
      "loss": 4.6658,
      "mean_token_accuracy": 0.25092208534479143,
      "num_tokens": 84655055.0,
      "step": 36925
    },
    {
      "entropy": 5.442407417297363,
      "epoch": 3.547550432276657,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037807843559736494,
      "loss": 4.7146,
      "mean_token_accuracy": 0.24996191263198853,
      "num_tokens": 84667840.0,
      "step": 36930
    },
    {
      "entropy": 5.4222253322601315,
      "epoch": 3.5480307396733908,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037804796019177874,
      "loss": 4.6846,
      "mean_token_accuracy": 0.25155219435691833,
      "num_tokens": 84678412.0,
      "step": 36935
    },
    {
      "entropy": 5.396407556533814,
      "epoch": 3.548511047070125,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003780174823937248,
      "loss": 4.6395,
      "mean_token_accuracy": 0.25999880135059356,
      "num_tokens": 84688725.0,
      "step": 36940
    },
    {
      "entropy": 5.501947116851807,
      "epoch": 3.5489913544668585,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037798700220391087,
      "loss": 4.8015,
      "mean_token_accuracy": 0.2505126252770424,
      "num_tokens": 84702489.0,
      "step": 36945
    },
    {
      "entropy": 5.404710006713867,
      "epoch": 3.549471661863593,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037795651962304456,
      "loss": 4.592,
      "mean_token_accuracy": 0.2633151814341545,
      "num_tokens": 84712933.0,
      "step": 36950
    },
    {
      "entropy": 5.392220067977905,
      "epoch": 3.5499519692603267,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037792603465183354,
      "loss": 4.718,
      "mean_token_accuracy": 0.24885118156671523,
      "num_tokens": 84723540.0,
      "step": 36955
    },
    {
      "entropy": 5.466289758682251,
      "epoch": 3.5504322766570606,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003778955472909857,
      "loss": 4.7758,
      "mean_token_accuracy": 0.24764496237039565,
      "num_tokens": 84735671.0,
      "step": 36960
    },
    {
      "entropy": 5.592088222503662,
      "epoch": 3.5509125840537945,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037786505754120874,
      "loss": 4.8611,
      "mean_token_accuracy": 0.2349093645811081,
      "num_tokens": 84746719.0,
      "step": 36965
    },
    {
      "entropy": 5.5165962219238285,
      "epoch": 3.5513928914505284,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037783456540321065,
      "loss": 4.7975,
      "mean_token_accuracy": 0.24087553173303605,
      "num_tokens": 84758774.0,
      "step": 36970
    },
    {
      "entropy": 5.454417562484741,
      "epoch": 3.5518731988472623,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037780407087769925,
      "loss": 4.7395,
      "mean_token_accuracy": 0.24590039551258086,
      "num_tokens": 84771047.0,
      "step": 36975
    },
    {
      "entropy": 5.413994979858399,
      "epoch": 3.552353506243996,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003777735739653826,
      "loss": 4.6715,
      "mean_token_accuracy": 0.2591075971722603,
      "num_tokens": 84781249.0,
      "step": 36980
    },
    {
      "entropy": 5.356760740280151,
      "epoch": 3.55283381364073,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003777430746669687,
      "loss": 4.599,
      "mean_token_accuracy": 0.25461663901805875,
      "num_tokens": 84791722.0,
      "step": 36985
    },
    {
      "entropy": 5.4702489376068115,
      "epoch": 3.553314121037464,
      "grad_norm": 1.125,
      "learning_rate": 0.00037771257298316573,
      "loss": 4.7408,
      "mean_token_accuracy": 0.24863242208957673,
      "num_tokens": 84804327.0,
      "step": 36990
    },
    {
      "entropy": 5.418333530426025,
      "epoch": 3.553794428434198,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003776820689146818,
      "loss": 4.709,
      "mean_token_accuracy": 0.2528533563017845,
      "num_tokens": 84814381.0,
      "step": 36995
    },
    {
      "entropy": 5.4383687496185305,
      "epoch": 3.5542747358309317,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037765156246222513,
      "loss": 4.6856,
      "mean_token_accuracy": 0.25532124638557435,
      "num_tokens": 84826274.0,
      "step": 37000
    },
    {
      "entropy": 5.452550840377808,
      "epoch": 3.5547550432276656,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037762105362650403,
      "loss": 4.6827,
      "mean_token_accuracy": 0.2510853707790375,
      "num_tokens": 84838946.0,
      "step": 37005
    },
    {
      "entropy": 5.570109939575195,
      "epoch": 3.5552353506243994,
      "grad_norm": 1.0,
      "learning_rate": 0.00037759054240822675,
      "loss": 4.789,
      "mean_token_accuracy": 0.2427710250020027,
      "num_tokens": 84851767.0,
      "step": 37010
    },
    {
      "entropy": 5.434012365341187,
      "epoch": 3.5557156580211338,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003775600288081018,
      "loss": 4.688,
      "mean_token_accuracy": 0.25075376629829405,
      "num_tokens": 84863712.0,
      "step": 37015
    },
    {
      "entropy": 5.362864923477173,
      "epoch": 3.556195965417867,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003775295128268374,
      "loss": 4.6425,
      "mean_token_accuracy": 0.262833996117115,
      "num_tokens": 84875699.0,
      "step": 37020
    },
    {
      "entropy": 5.471260356903076,
      "epoch": 3.5566762728146015,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037749899446514216,
      "loss": 4.6829,
      "mean_token_accuracy": 0.2520051583647728,
      "num_tokens": 84886772.0,
      "step": 37025
    },
    {
      "entropy": 5.441158771514893,
      "epoch": 3.5571565802113354,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003774684737237247,
      "loss": 4.7206,
      "mean_token_accuracy": 0.24816972762346268,
      "num_tokens": 84898501.0,
      "step": 37030
    },
    {
      "entropy": 5.510035657882691,
      "epoch": 3.5576368876080693,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003774379506032935,
      "loss": 4.7923,
      "mean_token_accuracy": 0.24063135534524918,
      "num_tokens": 84908418.0,
      "step": 37035
    },
    {
      "entropy": 5.507362651824951,
      "epoch": 3.558117195004803,
      "grad_norm": 1.40625,
      "learning_rate": 0.00037740742510455726,
      "loss": 4.7781,
      "mean_token_accuracy": 0.24827453196048738,
      "num_tokens": 84919792.0,
      "step": 37040
    },
    {
      "entropy": 5.4145612716674805,
      "epoch": 3.558597502401537,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003773768972282248,
      "loss": 4.7243,
      "mean_token_accuracy": 0.25520930290222166,
      "num_tokens": 84931606.0,
      "step": 37045
    },
    {
      "entropy": 5.378095436096191,
      "epoch": 3.559077809798271,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003773463669750046,
      "loss": 4.6838,
      "mean_token_accuracy": 0.25869992226362226,
      "num_tokens": 84943099.0,
      "step": 37050
    },
    {
      "entropy": 5.5576924800872805,
      "epoch": 3.559558117195005,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003773158343456058,
      "loss": 4.828,
      "mean_token_accuracy": 0.2357689082622528,
      "num_tokens": 84954444.0,
      "step": 37055
    },
    {
      "entropy": 5.4774292469024655,
      "epoch": 3.5600384245917387,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037728529934073714,
      "loss": 4.6559,
      "mean_token_accuracy": 0.2509851410984993,
      "num_tokens": 84965342.0,
      "step": 37060
    },
    {
      "entropy": 5.439002227783203,
      "epoch": 3.5605187319884726,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003772547619611075,
      "loss": 4.722,
      "mean_token_accuracy": 0.24919249713420868,
      "num_tokens": 84977685.0,
      "step": 37065
    },
    {
      "entropy": 5.516214513778687,
      "epoch": 3.5609990393852065,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003772242222074259,
      "loss": 4.7812,
      "mean_token_accuracy": 0.23962417095899582,
      "num_tokens": 84990148.0,
      "step": 37070
    },
    {
      "entropy": 5.5796546459198,
      "epoch": 3.5614793467819403,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037719368008040145,
      "loss": 4.817,
      "mean_token_accuracy": 0.24594101756811143,
      "num_tokens": 85001611.0,
      "step": 37075
    },
    {
      "entropy": 5.59600043296814,
      "epoch": 3.561959654178674,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037716313558074327,
      "loss": 4.8168,
      "mean_token_accuracy": 0.24252541810274125,
      "num_tokens": 85012558.0,
      "step": 37080
    },
    {
      "entropy": 5.447354316711426,
      "epoch": 3.562439961575408,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003771325887091603,
      "loss": 4.6879,
      "mean_token_accuracy": 0.2546050578355789,
      "num_tokens": 85022102.0,
      "step": 37085
    },
    {
      "entropy": 5.417292165756225,
      "epoch": 3.5629202689721424,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000377102039466362,
      "loss": 4.6498,
      "mean_token_accuracy": 0.2565863639116287,
      "num_tokens": 85033497.0,
      "step": 37090
    },
    {
      "entropy": 5.414370965957642,
      "epoch": 3.563400576368876,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003770714878530575,
      "loss": 4.7104,
      "mean_token_accuracy": 0.2516645982861519,
      "num_tokens": 85044938.0,
      "step": 37095
    },
    {
      "entropy": 5.500911855697632,
      "epoch": 3.56388088376561,
      "grad_norm": 2.5,
      "learning_rate": 0.00037704093386995607,
      "loss": 4.7185,
      "mean_token_accuracy": 0.2552750647068024,
      "num_tokens": 85056545.0,
      "step": 37100
    },
    {
      "entropy": 5.396948289871216,
      "epoch": 3.5643611911623436,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003770103775177673,
      "loss": 4.6294,
      "mean_token_accuracy": 0.26339801400899887,
      "num_tokens": 85067660.0,
      "step": 37105
    },
    {
      "entropy": 5.404934644699097,
      "epoch": 3.564841498559078,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037697981879720034,
      "loss": 4.6818,
      "mean_token_accuracy": 0.25580503344535827,
      "num_tokens": 85078813.0,
      "step": 37110
    },
    {
      "entropy": 5.533222436904907,
      "epoch": 3.565321805955812,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037694925770896486,
      "loss": 4.7428,
      "mean_token_accuracy": 0.23973799496889114,
      "num_tokens": 85089473.0,
      "step": 37115
    },
    {
      "entropy": 5.558396625518799,
      "epoch": 3.5658021133525457,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037691869425377037,
      "loss": 4.8407,
      "mean_token_accuracy": 0.24237702935934066,
      "num_tokens": 85101325.0,
      "step": 37120
    },
    {
      "entropy": 5.5426530838012695,
      "epoch": 3.5662824207492796,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003768881284323263,
      "loss": 4.7825,
      "mean_token_accuracy": 0.24280323684215546,
      "num_tokens": 85112601.0,
      "step": 37125
    },
    {
      "entropy": 5.4234254360198975,
      "epoch": 3.5667627281460135,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003768575602453426,
      "loss": 4.6904,
      "mean_token_accuracy": 0.26005417555570604,
      "num_tokens": 85123189.0,
      "step": 37130
    },
    {
      "entropy": 5.471990442276001,
      "epoch": 3.5672430355427474,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037682698969352876,
      "loss": 4.7017,
      "mean_token_accuracy": 0.2526450291275978,
      "num_tokens": 85133575.0,
      "step": 37135
    },
    {
      "entropy": 5.524107980728149,
      "epoch": 3.5677233429394812,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003767964167775946,
      "loss": 4.7064,
      "mean_token_accuracy": 0.254293255507946,
      "num_tokens": 85144549.0,
      "step": 37140
    },
    {
      "entropy": 5.481521987915039,
      "epoch": 3.568203650336215,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003767658414982498,
      "loss": 4.7384,
      "mean_token_accuracy": 0.25821213275194166,
      "num_tokens": 85155715.0,
      "step": 37145
    },
    {
      "entropy": 5.486953783035278,
      "epoch": 3.568683957732949,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003767352638562044,
      "loss": 4.7474,
      "mean_token_accuracy": 0.2463550463318825,
      "num_tokens": 85168268.0,
      "step": 37150
    },
    {
      "entropy": 5.469719552993775,
      "epoch": 3.569164265129683,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003767046838521683,
      "loss": 4.7356,
      "mean_token_accuracy": 0.24904571026563643,
      "num_tokens": 85181326.0,
      "step": 37155
    },
    {
      "entropy": 5.456112337112427,
      "epoch": 3.5696445725264168,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003766741014868515,
      "loss": 4.6347,
      "mean_token_accuracy": 0.2571116402745247,
      "num_tokens": 85191640.0,
      "step": 37160
    },
    {
      "entropy": 5.4544775009155275,
      "epoch": 3.570124879923151,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037664351676096385,
      "loss": 4.6961,
      "mean_token_accuracy": 0.25767491310834884,
      "num_tokens": 85202801.0,
      "step": 37165
    },
    {
      "entropy": 5.428199195861817,
      "epoch": 3.5706051873198845,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037661292967521563,
      "loss": 4.6557,
      "mean_token_accuracy": 0.24801378548145295,
      "num_tokens": 85213833.0,
      "step": 37170
    },
    {
      "entropy": 5.469465684890747,
      "epoch": 3.571085494716619,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037658234023031694,
      "loss": 4.6859,
      "mean_token_accuracy": 0.2610522612929344,
      "num_tokens": 85223856.0,
      "step": 37175
    },
    {
      "entropy": 5.4622045993804935,
      "epoch": 3.5715658021133523,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003765517484269779,
      "loss": 4.6953,
      "mean_token_accuracy": 0.2573678344488144,
      "num_tokens": 85236282.0,
      "step": 37180
    },
    {
      "entropy": 5.4229858875274655,
      "epoch": 3.5720461095100866,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037652115426590874,
      "loss": 4.71,
      "mean_token_accuracy": 0.2492672622203827,
      "num_tokens": 85247430.0,
      "step": 37185
    },
    {
      "entropy": 5.57228307723999,
      "epoch": 3.5725264169068205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037649055774782,
      "loss": 4.7594,
      "mean_token_accuracy": 0.24832040071487427,
      "num_tokens": 85259091.0,
      "step": 37190
    },
    {
      "entropy": 5.449297142028809,
      "epoch": 3.5730067243035544,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003764599588734218,
      "loss": 4.6337,
      "mean_token_accuracy": 0.26269001364707945,
      "num_tokens": 85270686.0,
      "step": 37195
    },
    {
      "entropy": 5.398440742492676,
      "epoch": 3.5734870317002883,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037642935764342454,
      "loss": 4.7034,
      "mean_token_accuracy": 0.2501357063651085,
      "num_tokens": 85281419.0,
      "step": 37200
    },
    {
      "entropy": 5.455207252502442,
      "epoch": 3.573967339097022,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003763987540585389,
      "loss": 4.7282,
      "mean_token_accuracy": 0.2481430247426033,
      "num_tokens": 85293116.0,
      "step": 37205
    },
    {
      "entropy": 5.55713906288147,
      "epoch": 3.574447646493756,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003763681481194752,
      "loss": 4.7678,
      "mean_token_accuracy": 0.2507124975323677,
      "num_tokens": 85304567.0,
      "step": 37210
    },
    {
      "entropy": 5.434828424453736,
      "epoch": 3.57492795389049,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037633753982694413,
      "loss": 4.7365,
      "mean_token_accuracy": 0.2485759288072586,
      "num_tokens": 85315947.0,
      "step": 37215
    },
    {
      "entropy": 5.445933771133423,
      "epoch": 3.5754082612872238,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037630692918165634,
      "loss": 4.7478,
      "mean_token_accuracy": 0.24712882339954376,
      "num_tokens": 85327309.0,
      "step": 37220
    },
    {
      "entropy": 5.479194593429566,
      "epoch": 3.5758885686839577,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003762763161843224,
      "loss": 4.7832,
      "mean_token_accuracy": 0.24957028329372405,
      "num_tokens": 85339034.0,
      "step": 37225
    },
    {
      "entropy": 5.489444303512573,
      "epoch": 3.5763688760806915,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003762457008356531,
      "loss": 4.7108,
      "mean_token_accuracy": 0.25528854727745054,
      "num_tokens": 85351764.0,
      "step": 37230
    },
    {
      "entropy": 5.385315895080566,
      "epoch": 3.5768491834774254,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037621508313635945,
      "loss": 4.6417,
      "mean_token_accuracy": 0.25789857357740403,
      "num_tokens": 85363234.0,
      "step": 37235
    },
    {
      "entropy": 5.424396181106568,
      "epoch": 3.5773294908741593,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003761844630871519,
      "loss": 4.6472,
      "mean_token_accuracy": 0.25863605588674543,
      "num_tokens": 85374148.0,
      "step": 37240
    },
    {
      "entropy": 5.4250133514404295,
      "epoch": 3.577809798270893,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003761538406887417,
      "loss": 4.648,
      "mean_token_accuracy": 0.25836222171783446,
      "num_tokens": 85385952.0,
      "step": 37245
    },
    {
      "entropy": 5.5126060962677,
      "epoch": 3.5782901056676275,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003761232159418397,
      "loss": 4.7361,
      "mean_token_accuracy": 0.24678527265787126,
      "num_tokens": 85396182.0,
      "step": 37250
    },
    {
      "entropy": 5.451197147369385,
      "epoch": 3.578770413064361,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003760925888471568,
      "loss": 4.6893,
      "mean_token_accuracy": 0.2554732084274292,
      "num_tokens": 85408109.0,
      "step": 37255
    },
    {
      "entropy": 5.518185377120972,
      "epoch": 3.5792507204610953,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003760619594054042,
      "loss": 4.8514,
      "mean_token_accuracy": 0.24443228244781495,
      "num_tokens": 85420622.0,
      "step": 37260
    },
    {
      "entropy": 5.419106101989746,
      "epoch": 3.579731027857829,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037603132761729303,
      "loss": 4.6598,
      "mean_token_accuracy": 0.25412497073411944,
      "num_tokens": 85431890.0,
      "step": 37265
    },
    {
      "entropy": 5.403204154968262,
      "epoch": 3.580211335254563,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003760006934835344,
      "loss": 4.7068,
      "mean_token_accuracy": 0.24514196962118148,
      "num_tokens": 85442953.0,
      "step": 37270
    },
    {
      "entropy": 5.491116380691528,
      "epoch": 3.580691642651297,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003759700570048396,
      "loss": 4.7605,
      "mean_token_accuracy": 0.25219975262880323,
      "num_tokens": 85454125.0,
      "step": 37275
    },
    {
      "entropy": 5.568990707397461,
      "epoch": 3.581171950048031,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003759394181819199,
      "loss": 4.8406,
      "mean_token_accuracy": 0.2388811230659485,
      "num_tokens": 85465162.0,
      "step": 37280
    },
    {
      "entropy": 5.4360494136810305,
      "epoch": 3.5816522574447647,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037590877701548667,
      "loss": 4.6306,
      "mean_token_accuracy": 0.2545792758464813,
      "num_tokens": 85475316.0,
      "step": 37285
    },
    {
      "entropy": 5.501903390884399,
      "epoch": 3.5821325648414986,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003758781335062512,
      "loss": 4.7492,
      "mean_token_accuracy": 0.2503475174307823,
      "num_tokens": 85487138.0,
      "step": 37290
    },
    {
      "entropy": 5.484175014495849,
      "epoch": 3.5826128722382324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003758474876549251,
      "loss": 4.6651,
      "mean_token_accuracy": 0.255440054833889,
      "num_tokens": 85498141.0,
      "step": 37295
    },
    {
      "entropy": 5.369169759750366,
      "epoch": 3.5830931796349663,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037581683946221974,
      "loss": 4.6947,
      "mean_token_accuracy": 0.2532894104719162,
      "num_tokens": 85510123.0,
      "step": 37300
    },
    {
      "entropy": 5.4933445930480955,
      "epoch": 3.5835734870317,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003757861889288467,
      "loss": 4.7896,
      "mean_token_accuracy": 0.23330225944519042,
      "num_tokens": 85522003.0,
      "step": 37305
    },
    {
      "entropy": 5.518648958206176,
      "epoch": 3.584053794428434,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037575553605551765,
      "loss": 4.7398,
      "mean_token_accuracy": 0.24140042662620545,
      "num_tokens": 85533400.0,
      "step": 37310
    },
    {
      "entropy": 5.405669164657593,
      "epoch": 3.584534101825168,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037572488084294423,
      "loss": 4.6349,
      "mean_token_accuracy": 0.2600130051374435,
      "num_tokens": 85543393.0,
      "step": 37315
    },
    {
      "entropy": 5.359893941879273,
      "epoch": 3.585014409221902,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037569422329183824,
      "loss": 4.6976,
      "mean_token_accuracy": 0.2592314973473549,
      "num_tokens": 85554596.0,
      "step": 37320
    },
    {
      "entropy": 5.491022157669067,
      "epoch": 3.585494716618636,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003756635634029112,
      "loss": 4.7712,
      "mean_token_accuracy": 0.24470193088054656,
      "num_tokens": 85566144.0,
      "step": 37325
    },
    {
      "entropy": 5.553358173370361,
      "epoch": 3.5859750240153696,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037563290117687535,
      "loss": 4.7688,
      "mean_token_accuracy": 0.2500207781791687,
      "num_tokens": 85576824.0,
      "step": 37330
    },
    {
      "entropy": 5.520667839050293,
      "epoch": 3.586455331412104,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003756022366144423,
      "loss": 4.7679,
      "mean_token_accuracy": 0.24482745826244354,
      "num_tokens": 85588719.0,
      "step": 37335
    },
    {
      "entropy": 5.393313550949097,
      "epoch": 3.586935638808838,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003755715697163239,
      "loss": 4.6147,
      "mean_token_accuracy": 0.25342566668987276,
      "num_tokens": 85598783.0,
      "step": 37340
    },
    {
      "entropy": 5.47696590423584,
      "epoch": 3.5874159462055717,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037554090048323247,
      "loss": 4.8003,
      "mean_token_accuracy": 0.24251838773488998,
      "num_tokens": 85610222.0,
      "step": 37345
    },
    {
      "entropy": 5.479221630096435,
      "epoch": 3.5878962536023056,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037551022891587976,
      "loss": 4.7106,
      "mean_token_accuracy": 0.25088177621364594,
      "num_tokens": 85620660.0,
      "step": 37350
    },
    {
      "entropy": 5.586337995529175,
      "epoch": 3.5883765609990395,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037547955501497806,
      "loss": 4.839,
      "mean_token_accuracy": 0.2394196569919586,
      "num_tokens": 85632466.0,
      "step": 37355
    },
    {
      "entropy": 5.593289041519165,
      "epoch": 3.5888568683957733,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003754488787812394,
      "loss": 4.8538,
      "mean_token_accuracy": 0.24349947422742843,
      "num_tokens": 85643343.0,
      "step": 37360
    },
    {
      "entropy": 5.3972392082214355,
      "epoch": 3.589337175792507,
      "grad_norm": 1.125,
      "learning_rate": 0.0003754182002153761,
      "loss": 4.6846,
      "mean_token_accuracy": 0.25362654477357865,
      "num_tokens": 85655864.0,
      "step": 37365
    },
    {
      "entropy": 5.480119848251343,
      "epoch": 3.589817483189241,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003753875193181003,
      "loss": 4.7919,
      "mean_token_accuracy": 0.2522342726588249,
      "num_tokens": 85667729.0,
      "step": 37370
    },
    {
      "entropy": 5.514600467681885,
      "epoch": 3.590297790585975,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037535683609012446,
      "loss": 4.7907,
      "mean_token_accuracy": 0.24308239072561263,
      "num_tokens": 85679252.0,
      "step": 37375
    },
    {
      "entropy": 5.46419997215271,
      "epoch": 3.590778097982709,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037532615053216084,
      "loss": 4.7234,
      "mean_token_accuracy": 0.25842811465263366,
      "num_tokens": 85690598.0,
      "step": 37380
    },
    {
      "entropy": 5.495190906524658,
      "epoch": 3.5912584053794427,
      "grad_norm": 1.28125,
      "learning_rate": 0.000375295462644922,
      "loss": 4.7235,
      "mean_token_accuracy": 0.25608272552490235,
      "num_tokens": 85703537.0,
      "step": 37385
    },
    {
      "entropy": 5.631138467788697,
      "epoch": 3.5917387127761766,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003752647724291202,
      "loss": 4.9219,
      "mean_token_accuracy": 0.23826923966407776,
      "num_tokens": 85717163.0,
      "step": 37390
    },
    {
      "entropy": 5.536502075195313,
      "epoch": 3.5922190201729105,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037523407988546815,
      "loss": 4.7827,
      "mean_token_accuracy": 0.24201785176992416,
      "num_tokens": 85728749.0,
      "step": 37395
    },
    {
      "entropy": 5.527349996566772,
      "epoch": 3.592699327569645,
      "grad_norm": 1.25,
      "learning_rate": 0.0003752033850146785,
      "loss": 4.8068,
      "mean_token_accuracy": 0.2452990174293518,
      "num_tokens": 85738653.0,
      "step": 37400
    },
    {
      "entropy": 5.421929740905762,
      "epoch": 3.5931796349663783,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003751726878174637,
      "loss": 4.6631,
      "mean_token_accuracy": 0.25856322795152664,
      "num_tokens": 85749244.0,
      "step": 37405
    },
    {
      "entropy": 5.476386642456054,
      "epoch": 3.5936599423631126,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003751419882945365,
      "loss": 4.6808,
      "mean_token_accuracy": 0.25879946202039716,
      "num_tokens": 85760314.0,
      "step": 37410
    },
    {
      "entropy": 5.451622581481933,
      "epoch": 3.594140249759846,
      "grad_norm": 1.25,
      "learning_rate": 0.0003751112864466097,
      "loss": 4.6401,
      "mean_token_accuracy": 0.2598521277308464,
      "num_tokens": 85771319.0,
      "step": 37415
    },
    {
      "entropy": 5.484802007675171,
      "epoch": 3.5946205571565804,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003750805822743962,
      "loss": 4.7355,
      "mean_token_accuracy": 0.2567942813038826,
      "num_tokens": 85781687.0,
      "step": 37420
    },
    {
      "entropy": 5.454227161407471,
      "epoch": 3.5951008645533142,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003750498757786087,
      "loss": 4.697,
      "mean_token_accuracy": 0.24871593713760376,
      "num_tokens": 85793775.0,
      "step": 37425
    },
    {
      "entropy": 5.507655382156372,
      "epoch": 3.595581171950048,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003750191669599602,
      "loss": 4.7923,
      "mean_token_accuracy": 0.2482352375984192,
      "num_tokens": 85805404.0,
      "step": 37430
    },
    {
      "entropy": 5.516337919235229,
      "epoch": 3.596061479346782,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003749884558191636,
      "loss": 4.7606,
      "mean_token_accuracy": 0.24544950425624848,
      "num_tokens": 85818182.0,
      "step": 37435
    },
    {
      "entropy": 5.520508193969727,
      "epoch": 3.596541786743516,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037495774235693193,
      "loss": 4.7379,
      "mean_token_accuracy": 0.2454267367720604,
      "num_tokens": 85828747.0,
      "step": 37440
    },
    {
      "entropy": 5.500737094879151,
      "epoch": 3.5970220941402498,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037492702657397837,
      "loss": 4.6897,
      "mean_token_accuracy": 0.25242246985435485,
      "num_tokens": 85839914.0,
      "step": 37445
    },
    {
      "entropy": 5.4806359767913815,
      "epoch": 3.5975024015369836,
      "grad_norm": 1.25,
      "learning_rate": 0.000374896308471016,
      "loss": 4.6555,
      "mean_token_accuracy": 0.25756998658180236,
      "num_tokens": 85851252.0,
      "step": 37450
    },
    {
      "entropy": 5.419042253494263,
      "epoch": 3.5979827089337175,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037486558804875785,
      "loss": 4.6361,
      "mean_token_accuracy": 0.256014358997345,
      "num_tokens": 85863287.0,
      "step": 37455
    },
    {
      "entropy": 5.527282857894898,
      "epoch": 3.5984630163304514,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003748348653079174,
      "loss": 4.8533,
      "mean_token_accuracy": 0.23997599631547928,
      "num_tokens": 85874575.0,
      "step": 37460
    },
    {
      "entropy": 5.471496105194092,
      "epoch": 3.5989433237271853,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037480414024920777,
      "loss": 4.6825,
      "mean_token_accuracy": 0.25211990922689437,
      "num_tokens": 85886267.0,
      "step": 37465
    },
    {
      "entropy": 5.437094211578369,
      "epoch": 3.599423631123919,
      "grad_norm": 1.375,
      "learning_rate": 0.0003747734128733424,
      "loss": 4.7466,
      "mean_token_accuracy": 0.24758899956941605,
      "num_tokens": 85897149.0,
      "step": 37470
    },
    {
      "entropy": 5.316774559020996,
      "epoch": 3.5999039385206535,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003747426831810346,
      "loss": 4.5614,
      "mean_token_accuracy": 0.2669482335448265,
      "num_tokens": 85909045.0,
      "step": 37475
    },
    {
      "entropy": 5.416320323944092,
      "epoch": 3.600384245917387,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037471195117299796,
      "loss": 4.7095,
      "mean_token_accuracy": 0.25623527765274046,
      "num_tokens": 85919795.0,
      "step": 37480
    },
    {
      "entropy": 5.494394207000733,
      "epoch": 3.6008645533141213,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003746812168499458,
      "loss": 4.7328,
      "mean_token_accuracy": 0.25328807681798937,
      "num_tokens": 85930348.0,
      "step": 37485
    },
    {
      "entropy": 5.560242462158203,
      "epoch": 3.6013448607108547,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003746504802125918,
      "loss": 4.8271,
      "mean_token_accuracy": 0.24162171483039857,
      "num_tokens": 85941956.0,
      "step": 37490
    },
    {
      "entropy": 5.471708917617798,
      "epoch": 3.601825168107589,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037461974126164953,
      "loss": 4.6706,
      "mean_token_accuracy": 0.2549857750535011,
      "num_tokens": 85953208.0,
      "step": 37495
    },
    {
      "entropy": 5.480360460281372,
      "epoch": 3.602305475504323,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037458899999783267,
      "loss": 4.7477,
      "mean_token_accuracy": 0.2454298809170723,
      "num_tokens": 85967527.0,
      "step": 37500
    },
    {
      "entropy": 5.5060296058654785,
      "epoch": 3.6027857829010568,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037455825642185504,
      "loss": 4.7396,
      "mean_token_accuracy": 0.2498505175113678,
      "num_tokens": 85979013.0,
      "step": 37505
    },
    {
      "entropy": 5.501752090454102,
      "epoch": 3.6032660902977907,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003745275105344303,
      "loss": 4.7198,
      "mean_token_accuracy": 0.25052082538604736,
      "num_tokens": 85990331.0,
      "step": 37510
    },
    {
      "entropy": 5.501833581924439,
      "epoch": 3.6037463976945245,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003744967623362723,
      "loss": 4.7506,
      "mean_token_accuracy": 0.2500546768307686,
      "num_tokens": 86001883.0,
      "step": 37515
    },
    {
      "entropy": 5.46913595199585,
      "epoch": 3.6042267050912584,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003744660118280949,
      "loss": 4.7334,
      "mean_token_accuracy": 0.24878235012292862,
      "num_tokens": 86013069.0,
      "step": 37520
    },
    {
      "entropy": 5.530709457397461,
      "epoch": 3.6047070124879923,
      "grad_norm": 1.078125,
      "learning_rate": 0.000374435259010612,
      "loss": 4.7781,
      "mean_token_accuracy": 0.2458219662308693,
      "num_tokens": 86026107.0,
      "step": 37525
    },
    {
      "entropy": 5.46111831665039,
      "epoch": 3.605187319884726,
      "grad_norm": 1.375,
      "learning_rate": 0.00037440450388453767,
      "loss": 4.646,
      "mean_token_accuracy": 0.2629260718822479,
      "num_tokens": 86036914.0,
      "step": 37530
    },
    {
      "entropy": 5.46657428741455,
      "epoch": 3.60566762728146,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000374373746450586,
      "loss": 4.7618,
      "mean_token_accuracy": 0.24630853533744812,
      "num_tokens": 86047922.0,
      "step": 37535
    },
    {
      "entropy": 5.43938512802124,
      "epoch": 3.606147934678194,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037434298670947097,
      "loss": 4.7377,
      "mean_token_accuracy": 0.2496066212654114,
      "num_tokens": 86058843.0,
      "step": 37540
    },
    {
      "entropy": 5.479861497879028,
      "epoch": 3.606628242074928,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037431222466190677,
      "loss": 4.6414,
      "mean_token_accuracy": 0.2550934195518494,
      "num_tokens": 86069212.0,
      "step": 37545
    },
    {
      "entropy": 5.453286266326904,
      "epoch": 3.6071085494716617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037428146030860765,
      "loss": 4.676,
      "mean_token_accuracy": 0.2520549684762955,
      "num_tokens": 86080952.0,
      "step": 37550
    },
    {
      "entropy": 5.416334438323974,
      "epoch": 3.6075888568683956,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003742506936502878,
      "loss": 4.6933,
      "mean_token_accuracy": 0.25733979791402817,
      "num_tokens": 86091976.0,
      "step": 37555
    },
    {
      "entropy": 5.4051107406616214,
      "epoch": 3.60806916426513,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003742199246876616,
      "loss": 4.6332,
      "mean_token_accuracy": 0.2566567897796631,
      "num_tokens": 86103135.0,
      "step": 37560
    },
    {
      "entropy": 5.576596355438232,
      "epoch": 3.6085494716618634,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037418915342144333,
      "loss": 4.7254,
      "mean_token_accuracy": 0.251104636490345,
      "num_tokens": 86113495.0,
      "step": 37565
    },
    {
      "entropy": 5.499048948287964,
      "epoch": 3.6090297790585977,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003741583798523474,
      "loss": 4.7162,
      "mean_token_accuracy": 0.24477304071187972,
      "num_tokens": 86125489.0,
      "step": 37570
    },
    {
      "entropy": 5.504081583023071,
      "epoch": 3.6095100864553316,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003741276039810884,
      "loss": 4.7083,
      "mean_token_accuracy": 0.25074398666620257,
      "num_tokens": 86136109.0,
      "step": 37575
    },
    {
      "entropy": 5.472125053405762,
      "epoch": 3.6099903938520654,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003740968258083808,
      "loss": 4.7562,
      "mean_token_accuracy": 0.246445532143116,
      "num_tokens": 86147375.0,
      "step": 37580
    },
    {
      "entropy": 5.4586687088012695,
      "epoch": 3.6104707012487993,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003740660453349391,
      "loss": 4.7069,
      "mean_token_accuracy": 0.24855585396289825,
      "num_tokens": 86158277.0,
      "step": 37585
    },
    {
      "entropy": 5.503960609436035,
      "epoch": 3.610951008645533,
      "grad_norm": 1.28125,
      "learning_rate": 0.000374035262561478,
      "loss": 4.7564,
      "mean_token_accuracy": 0.24259666800498964,
      "num_tokens": 86169280.0,
      "step": 37590
    },
    {
      "entropy": 5.500898170471191,
      "epoch": 3.611431316042267,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037400447748871224,
      "loss": 4.7043,
      "mean_token_accuracy": 0.25495010763406756,
      "num_tokens": 86179918.0,
      "step": 37595
    },
    {
      "entropy": 5.350613737106324,
      "epoch": 3.611911623439001,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037397369011735643,
      "loss": 4.5808,
      "mean_token_accuracy": 0.263748599588871,
      "num_tokens": 86192616.0,
      "step": 37600
    },
    {
      "entropy": 5.444384622573852,
      "epoch": 3.612391930835735,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003739429004481254,
      "loss": 4.7029,
      "mean_token_accuracy": 0.24713116884231567,
      "num_tokens": 86203797.0,
      "step": 37605
    },
    {
      "entropy": 5.420926094055176,
      "epoch": 3.6128722382324687,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037391210848173413,
      "loss": 4.6559,
      "mean_token_accuracy": 0.25671450793743134,
      "num_tokens": 86214408.0,
      "step": 37610
    },
    {
      "entropy": 5.44876537322998,
      "epoch": 3.6133525456292026,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037388131421889733,
      "loss": 4.6939,
      "mean_token_accuracy": 0.25350883305072786,
      "num_tokens": 86226382.0,
      "step": 37615
    },
    {
      "entropy": 5.493449401855469,
      "epoch": 3.6138328530259365,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003738505176603301,
      "loss": 4.7298,
      "mean_token_accuracy": 0.2510909244418144,
      "num_tokens": 86236962.0,
      "step": 37620
    },
    {
      "entropy": 5.425980281829834,
      "epoch": 3.6143131604226704,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037381971880674726,
      "loss": 4.643,
      "mean_token_accuracy": 0.26170885413885114,
      "num_tokens": 86248867.0,
      "step": 37625
    },
    {
      "entropy": 5.516648960113526,
      "epoch": 3.6147934678194042,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037378891765886405,
      "loss": 4.7419,
      "mean_token_accuracy": 0.24824082255363464,
      "num_tokens": 86260264.0,
      "step": 37630
    },
    {
      "entropy": 5.484749889373779,
      "epoch": 3.6152737752161386,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037375811421739555,
      "loss": 4.7593,
      "mean_token_accuracy": 0.24273585230112077,
      "num_tokens": 86271707.0,
      "step": 37635
    },
    {
      "entropy": 5.428352165222168,
      "epoch": 3.615754082612872,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037372730848305683,
      "loss": 4.6658,
      "mean_token_accuracy": 0.25984631329774854,
      "num_tokens": 86282532.0,
      "step": 37640
    },
    {
      "entropy": 5.51300630569458,
      "epoch": 3.6162343900096063,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037369650045656315,
      "loss": 4.7747,
      "mean_token_accuracy": 0.24692470729351043,
      "num_tokens": 86295256.0,
      "step": 37645
    },
    {
      "entropy": 5.479423189163208,
      "epoch": 3.61671469740634,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037366569013862986,
      "loss": 4.7156,
      "mean_token_accuracy": 0.2524994879961014,
      "num_tokens": 86306326.0,
      "step": 37650
    },
    {
      "entropy": 5.401057767868042,
      "epoch": 3.617195004803074,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003736348775299722,
      "loss": 4.6223,
      "mean_token_accuracy": 0.26352909207344055,
      "num_tokens": 86317313.0,
      "step": 37655
    },
    {
      "entropy": 5.4642692565917965,
      "epoch": 3.617675312199808,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037360406263130556,
      "loss": 4.6781,
      "mean_token_accuracy": 0.25660122036933897,
      "num_tokens": 86327139.0,
      "step": 37660
    },
    {
      "entropy": 5.423729133605957,
      "epoch": 3.618155619596542,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037357324544334537,
      "loss": 4.7058,
      "mean_token_accuracy": 0.24919614046812058,
      "num_tokens": 86338743.0,
      "step": 37665
    },
    {
      "entropy": 5.47803144454956,
      "epoch": 3.6186359269932757,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003735424259668072,
      "loss": 4.7773,
      "mean_token_accuracy": 0.24376185536384581,
      "num_tokens": 86349790.0,
      "step": 37670
    },
    {
      "entropy": 5.50017237663269,
      "epoch": 3.6191162343900096,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003735116042024064,
      "loss": 4.7235,
      "mean_token_accuracy": 0.2533008947968483,
      "num_tokens": 86361689.0,
      "step": 37675
    },
    {
      "entropy": 5.522277212142944,
      "epoch": 3.6195965417867435,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003734807801508587,
      "loss": 4.7272,
      "mean_token_accuracy": 0.2500563129782677,
      "num_tokens": 86372005.0,
      "step": 37680
    },
    {
      "entropy": 5.485615968704224,
      "epoch": 3.6200768491834774,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037344995381287984,
      "loss": 4.816,
      "mean_token_accuracy": 0.24284866452217102,
      "num_tokens": 86383789.0,
      "step": 37685
    },
    {
      "entropy": 5.420523548126221,
      "epoch": 3.6205571565802113,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037341912518918524,
      "loss": 4.6953,
      "mean_token_accuracy": 0.2491873100399971,
      "num_tokens": 86395831.0,
      "step": 37690
    },
    {
      "entropy": 5.477121448516845,
      "epoch": 3.621037463976945,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037338829428049087,
      "loss": 4.6669,
      "mean_token_accuracy": 0.25675160735845565,
      "num_tokens": 86406807.0,
      "step": 37695
    },
    {
      "entropy": 5.498269701004029,
      "epoch": 3.621517771373679,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037335746108751246,
      "loss": 4.7506,
      "mean_token_accuracy": 0.24916112571954727,
      "num_tokens": 86418025.0,
      "step": 37700
    },
    {
      "entropy": 5.514704704284668,
      "epoch": 3.621998078770413,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003733266256109658,
      "loss": 4.738,
      "mean_token_accuracy": 0.24326128512620926,
      "num_tokens": 86429019.0,
      "step": 37705
    },
    {
      "entropy": 5.56147313117981,
      "epoch": 3.6224783861671472,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037329578785156686,
      "loss": 4.8178,
      "mean_token_accuracy": 0.24109542965888978,
      "num_tokens": 86439471.0,
      "step": 37710
    },
    {
      "entropy": 5.5193675518035885,
      "epoch": 3.6229586935638807,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003732649478100317,
      "loss": 4.7683,
      "mean_token_accuracy": 0.2443425938487053,
      "num_tokens": 86450966.0,
      "step": 37715
    },
    {
      "entropy": 5.356217956542968,
      "epoch": 3.623439000960615,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037323410548707617,
      "loss": 4.5835,
      "mean_token_accuracy": 0.25899662375450133,
      "num_tokens": 86461467.0,
      "step": 37720
    },
    {
      "entropy": 5.446416568756104,
      "epoch": 3.6239193083573484,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003732032608834164,
      "loss": 4.7652,
      "mean_token_accuracy": 0.24830191880464553,
      "num_tokens": 86472743.0,
      "step": 37725
    },
    {
      "entropy": 5.475582075119019,
      "epoch": 3.6243996157540828,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037317241399976856,
      "loss": 4.7297,
      "mean_token_accuracy": 0.25301122963428496,
      "num_tokens": 86484101.0,
      "step": 37730
    },
    {
      "entropy": 5.471561431884766,
      "epoch": 3.6248799231508166,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037314156483684875,
      "loss": 4.718,
      "mean_token_accuracy": 0.24752050191164016,
      "num_tokens": 86496017.0,
      "step": 37735
    },
    {
      "entropy": 5.496461343765259,
      "epoch": 3.6253602305475505,
      "grad_norm": 1.125,
      "learning_rate": 0.0003731107133953732,
      "loss": 4.7323,
      "mean_token_accuracy": 0.24142497181892394,
      "num_tokens": 86508198.0,
      "step": 37740
    },
    {
      "entropy": 5.621002197265625,
      "epoch": 3.6258405379442844,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037307985967605827,
      "loss": 4.8775,
      "mean_token_accuracy": 0.24180345237255096,
      "num_tokens": 86518712.0,
      "step": 37745
    },
    {
      "entropy": 5.440371799468994,
      "epoch": 3.6263208453410183,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037304900367962017,
      "loss": 4.6772,
      "mean_token_accuracy": 0.25067765861749647,
      "num_tokens": 86531340.0,
      "step": 37750
    },
    {
      "entropy": 5.428462314605713,
      "epoch": 3.626801152737752,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003730181454067754,
      "loss": 4.6503,
      "mean_token_accuracy": 0.2551438584923744,
      "num_tokens": 86541548.0,
      "step": 37755
    },
    {
      "entropy": 5.452396965026855,
      "epoch": 3.627281460134486,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037298728485824043,
      "loss": 4.7539,
      "mean_token_accuracy": 0.25072367638349535,
      "num_tokens": 86553259.0,
      "step": 37760
    },
    {
      "entropy": 5.420817232131958,
      "epoch": 3.62776176753122,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003729564220347315,
      "loss": 4.6788,
      "mean_token_accuracy": 0.24594148844480515,
      "num_tokens": 86565199.0,
      "step": 37765
    },
    {
      "entropy": 5.392025184631348,
      "epoch": 3.628242074927954,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003729255569369654,
      "loss": 4.578,
      "mean_token_accuracy": 0.26236034780740736,
      "num_tokens": 86574949.0,
      "step": 37770
    },
    {
      "entropy": 5.444803142547608,
      "epoch": 3.6287223823246877,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003728946895656586,
      "loss": 4.6718,
      "mean_token_accuracy": 0.2526092201471329,
      "num_tokens": 86586039.0,
      "step": 37775
    },
    {
      "entropy": 5.519192218780518,
      "epoch": 3.6292026897214216,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003728638199215279,
      "loss": 4.7478,
      "mean_token_accuracy": 0.24560384303331376,
      "num_tokens": 86598365.0,
      "step": 37780
    },
    {
      "entropy": 5.410275173187256,
      "epoch": 3.629682997118156,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037283294800528985,
      "loss": 4.6154,
      "mean_token_accuracy": 0.2612846538424492,
      "num_tokens": 86609460.0,
      "step": 37785
    },
    {
      "entropy": 5.41770305633545,
      "epoch": 3.6301633045148893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003728020738176612,
      "loss": 4.6882,
      "mean_token_accuracy": 0.25548186153173447,
      "num_tokens": 86621195.0,
      "step": 37790
    },
    {
      "entropy": 5.483548212051391,
      "epoch": 3.6306436119116237,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003727711973593589,
      "loss": 4.7328,
      "mean_token_accuracy": 0.2487204611301422,
      "num_tokens": 86632671.0,
      "step": 37795
    },
    {
      "entropy": 5.481105661392212,
      "epoch": 3.631123919308357,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003727403186310996,
      "loss": 4.7187,
      "mean_token_accuracy": 0.24748198837041854,
      "num_tokens": 86644224.0,
      "step": 37800
    },
    {
      "entropy": 5.569194889068603,
      "epoch": 3.6316042267050914,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037270943763360046,
      "loss": 4.8212,
      "mean_token_accuracy": 0.24751319736242294,
      "num_tokens": 86655302.0,
      "step": 37805
    },
    {
      "entropy": 5.5555023670196535,
      "epoch": 3.6320845341018253,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003726785543675782,
      "loss": 4.7937,
      "mean_token_accuracy": 0.24569666832685472,
      "num_tokens": 86665907.0,
      "step": 37810
    },
    {
      "entropy": 5.386057186126709,
      "epoch": 3.632564841498559,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037264766883375003,
      "loss": 4.6359,
      "mean_token_accuracy": 0.26034427881240846,
      "num_tokens": 86676746.0,
      "step": 37815
    },
    {
      "entropy": 5.402319669723511,
      "epoch": 3.633045148895293,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003726167810328329,
      "loss": 4.7146,
      "mean_token_accuracy": 0.24880185276269912,
      "num_tokens": 86689274.0,
      "step": 37820
    },
    {
      "entropy": 5.456144523620606,
      "epoch": 3.633525456292027,
      "grad_norm": 1.390625,
      "learning_rate": 0.000372585890965544,
      "loss": 4.712,
      "mean_token_accuracy": 0.25044218748807906,
      "num_tokens": 86700427.0,
      "step": 37825
    },
    {
      "entropy": 5.451106834411621,
      "epoch": 3.634005763688761,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003725549986326004,
      "loss": 4.6809,
      "mean_token_accuracy": 0.25733633041381837,
      "num_tokens": 86712780.0,
      "step": 37830
    },
    {
      "entropy": 5.474854183197022,
      "epoch": 3.6344860710854947,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003725241040347196,
      "loss": 4.7184,
      "mean_token_accuracy": 0.2514024540781975,
      "num_tokens": 86724359.0,
      "step": 37835
    },
    {
      "entropy": 5.468010663986206,
      "epoch": 3.6349663784822286,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003724932071726186,
      "loss": 4.7735,
      "mean_token_accuracy": 0.24754831492900847,
      "num_tokens": 86735540.0,
      "step": 37840
    },
    {
      "entropy": 5.4330486297607425,
      "epoch": 3.6354466858789625,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037246230804701476,
      "loss": 4.7149,
      "mean_token_accuracy": 0.2532291144132614,
      "num_tokens": 86746879.0,
      "step": 37845
    },
    {
      "entropy": 5.473910236358643,
      "epoch": 3.6359269932756964,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037243140665862556,
      "loss": 4.7115,
      "mean_token_accuracy": 0.25441018044948577,
      "num_tokens": 86756783.0,
      "step": 37850
    },
    {
      "entropy": 5.505873298645019,
      "epoch": 3.6364073006724302,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037240050300816843,
      "loss": 4.7194,
      "mean_token_accuracy": 0.25907407104969027,
      "num_tokens": 86768722.0,
      "step": 37855
    },
    {
      "entropy": 5.477000999450683,
      "epoch": 3.636887608069164,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037236959709636085,
      "loss": 4.7792,
      "mean_token_accuracy": 0.25092304199934007,
      "num_tokens": 86780386.0,
      "step": 37860
    },
    {
      "entropy": 5.485908460617066,
      "epoch": 3.637367915465898,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037233868892392035,
      "loss": 4.7179,
      "mean_token_accuracy": 0.24923425167798996,
      "num_tokens": 86793008.0,
      "step": 37865
    },
    {
      "entropy": 5.532158613204956,
      "epoch": 3.6378482228626323,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003723077784915645,
      "loss": 4.6978,
      "mean_token_accuracy": 0.25102945417165756,
      "num_tokens": 86804451.0,
      "step": 37870
    },
    {
      "entropy": 5.526382732391357,
      "epoch": 3.6383285302593658,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037227686580001095,
      "loss": 4.7679,
      "mean_token_accuracy": 0.2504990190267563,
      "num_tokens": 86816391.0,
      "step": 37875
    },
    {
      "entropy": 5.552110481262207,
      "epoch": 3.6388088376561,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003722459508499775,
      "loss": 4.809,
      "mean_token_accuracy": 0.2464088723063469,
      "num_tokens": 86828163.0,
      "step": 37880
    },
    {
      "entropy": 5.532535600662231,
      "epoch": 3.639289145052834,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003722150336421818,
      "loss": 4.7206,
      "mean_token_accuracy": 0.24648042917251586,
      "num_tokens": 86839311.0,
      "step": 37885
    },
    {
      "entropy": 5.563254356384277,
      "epoch": 3.639769452449568,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003721841141773417,
      "loss": 4.8131,
      "mean_token_accuracy": 0.23855464458465575,
      "num_tokens": 86849921.0,
      "step": 37890
    },
    {
      "entropy": 5.371118259429932,
      "epoch": 3.6402497598463017,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037215319245617503,
      "loss": 4.6226,
      "mean_token_accuracy": 0.25099276155233385,
      "num_tokens": 86862858.0,
      "step": 37895
    },
    {
      "entropy": 5.446948146820068,
      "epoch": 3.6407300672430356,
      "grad_norm": 1.25,
      "learning_rate": 0.00037212226847939975,
      "loss": 4.6457,
      "mean_token_accuracy": 0.25914138555526733,
      "num_tokens": 86874152.0,
      "step": 37900
    },
    {
      "entropy": 5.4868096828460695,
      "epoch": 3.6412103746397695,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003720913422477338,
      "loss": 4.6793,
      "mean_token_accuracy": 0.25628081411123277,
      "num_tokens": 86886126.0,
      "step": 37905
    },
    {
      "entropy": 5.620455312728882,
      "epoch": 3.6416906820365034,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037206041376189515,
      "loss": 4.869,
      "mean_token_accuracy": 0.24005090892314912,
      "num_tokens": 86896046.0,
      "step": 37910
    },
    {
      "entropy": 5.400245237350464,
      "epoch": 3.6421709894332372,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003720294830226018,
      "loss": 4.6516,
      "mean_token_accuracy": 0.25322025567293166,
      "num_tokens": 86907946.0,
      "step": 37915
    },
    {
      "entropy": 5.38089394569397,
      "epoch": 3.642651296829971,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003719985500305721,
      "loss": 4.6396,
      "mean_token_accuracy": 0.2580250158905983,
      "num_tokens": 86919849.0,
      "step": 37920
    },
    {
      "entropy": 5.539474821090698,
      "epoch": 3.643131604226705,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003719676147865241,
      "loss": 4.7894,
      "mean_token_accuracy": 0.24884163439273835,
      "num_tokens": 86930325.0,
      "step": 37925
    },
    {
      "entropy": 5.400912618637085,
      "epoch": 3.643611911623439,
      "grad_norm": 1.15625,
      "learning_rate": 0.000371936677291176,
      "loss": 4.6219,
      "mean_token_accuracy": 0.25573321133852006,
      "num_tokens": 86942076.0,
      "step": 37930
    },
    {
      "entropy": 5.373451995849609,
      "epoch": 3.6440922190201728,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003719057375452461,
      "loss": 4.5943,
      "mean_token_accuracy": 0.2689600795507431,
      "num_tokens": 86953226.0,
      "step": 37935
    },
    {
      "entropy": 5.494539451599121,
      "epoch": 3.6445725264169067,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003718747955494527,
      "loss": 4.7862,
      "mean_token_accuracy": 0.24687643945217133,
      "num_tokens": 86964606.0,
      "step": 37940
    },
    {
      "entropy": 5.4188157081604,
      "epoch": 3.645052833813641,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037184385130451425,
      "loss": 4.6369,
      "mean_token_accuracy": 0.2572964310646057,
      "num_tokens": 86976738.0,
      "step": 37945
    },
    {
      "entropy": 5.495331430435181,
      "epoch": 3.6455331412103744,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037181290481114903,
      "loss": 4.8257,
      "mean_token_accuracy": 0.24890100806951523,
      "num_tokens": 86988445.0,
      "step": 37950
    },
    {
      "entropy": 5.441614389419556,
      "epoch": 3.6460134486071087,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003717819560700758,
      "loss": 4.6485,
      "mean_token_accuracy": 0.2622308641672134,
      "num_tokens": 87000818.0,
      "step": 37955
    },
    {
      "entropy": 5.47670578956604,
      "epoch": 3.6464937560038426,
      "grad_norm": 1.125,
      "learning_rate": 0.0003717510050820128,
      "loss": 4.7468,
      "mean_token_accuracy": 0.25206083059310913,
      "num_tokens": 87012398.0,
      "step": 37960
    },
    {
      "entropy": 5.547105932235718,
      "epoch": 3.6469740634005765,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037172005184767885,
      "loss": 4.7899,
      "mean_token_accuracy": 0.24522292166948317,
      "num_tokens": 87022885.0,
      "step": 37965
    },
    {
      "entropy": 5.504031229019165,
      "epoch": 3.6474543707973104,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003716890963677924,
      "loss": 4.7269,
      "mean_token_accuracy": 0.2548235669732094,
      "num_tokens": 87033973.0,
      "step": 37970
    },
    {
      "entropy": 5.485598373413086,
      "epoch": 3.6479346781940443,
      "grad_norm": 1.125,
      "learning_rate": 0.0003716581386430723,
      "loss": 4.703,
      "mean_token_accuracy": 0.25136757642030716,
      "num_tokens": 87046391.0,
      "step": 37975
    },
    {
      "entropy": 5.505124044418335,
      "epoch": 3.648414985590778,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003716271786742372,
      "loss": 4.7451,
      "mean_token_accuracy": 0.25007263123989104,
      "num_tokens": 87056913.0,
      "step": 37980
    },
    {
      "entropy": 5.5228846073150635,
      "epoch": 3.648895292987512,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037159621646200595,
      "loss": 4.7108,
      "mean_token_accuracy": 0.2532086044549942,
      "num_tokens": 87068905.0,
      "step": 37985
    },
    {
      "entropy": 5.497816753387451,
      "epoch": 3.649375600384246,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003715652520070974,
      "loss": 4.7963,
      "mean_token_accuracy": 0.24695555865764618,
      "num_tokens": 87080186.0,
      "step": 37990
    },
    {
      "entropy": 5.438949251174927,
      "epoch": 3.64985590778098,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003715342853102305,
      "loss": 4.6937,
      "mean_token_accuracy": 0.2543922752141953,
      "num_tokens": 87091130.0,
      "step": 37995
    },
    {
      "entropy": 5.57559642791748,
      "epoch": 3.6503362151777137,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003715033163721241,
      "loss": 4.7867,
      "mean_token_accuracy": 0.24391983151435853,
      "num_tokens": 87103482.0,
      "step": 38000
    },
    {
      "entropy": 5.495943307876587,
      "epoch": 3.6508165225744476,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003714723451934972,
      "loss": 4.759,
      "mean_token_accuracy": 0.24323913007974624,
      "num_tokens": 87114850.0,
      "step": 38005
    },
    {
      "entropy": 5.476016044616699,
      "epoch": 3.6512968299711814,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037144137177506887,
      "loss": 4.7569,
      "mean_token_accuracy": 0.24965815395116805,
      "num_tokens": 87125840.0,
      "step": 38010
    },
    {
      "entropy": 5.505307960510254,
      "epoch": 3.6517771373679153,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003714103961175584,
      "loss": 4.7499,
      "mean_token_accuracy": 0.24790554046630858,
      "num_tokens": 87138216.0,
      "step": 38015
    },
    {
      "entropy": 5.487857389450073,
      "epoch": 3.6522574447646496,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003713794182216847,
      "loss": 4.6937,
      "mean_token_accuracy": 0.2554118737578392,
      "num_tokens": 87149071.0,
      "step": 38020
    },
    {
      "entropy": 5.429833698272705,
      "epoch": 3.652737752161383,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037134843808816717,
      "loss": 4.6956,
      "mean_token_accuracy": 0.25844560712575915,
      "num_tokens": 87160564.0,
      "step": 38025
    },
    {
      "entropy": 5.353420448303223,
      "epoch": 3.6532180595581174,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037131745571772497,
      "loss": 4.6127,
      "mean_token_accuracy": 0.26833585649728775,
      "num_tokens": 87171536.0,
      "step": 38030
    },
    {
      "entropy": 5.5676689624786375,
      "epoch": 3.653698366954851,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037128647111107736,
      "loss": 4.809,
      "mean_token_accuracy": 0.2465489149093628,
      "num_tokens": 87183064.0,
      "step": 38035
    },
    {
      "entropy": 5.537548398971557,
      "epoch": 3.654178674351585,
      "grad_norm": 1.21875,
      "learning_rate": 0.000371255484268944,
      "loss": 4.7101,
      "mean_token_accuracy": 0.24942291378974915,
      "num_tokens": 87194034.0,
      "step": 38040
    },
    {
      "entropy": 5.488871002197266,
      "epoch": 3.654658981748319,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037122449519204397,
      "loss": 4.7477,
      "mean_token_accuracy": 0.2509173363447189,
      "num_tokens": 87205083.0,
      "step": 38045
    },
    {
      "entropy": 5.51708197593689,
      "epoch": 3.655139289145053,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037119350388109697,
      "loss": 4.7305,
      "mean_token_accuracy": 0.24972273111343385,
      "num_tokens": 87216034.0,
      "step": 38050
    },
    {
      "entropy": 5.455567646026611,
      "epoch": 3.655619596541787,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037116251033682236,
      "loss": 4.6989,
      "mean_token_accuracy": 0.2531959980726242,
      "num_tokens": 87227231.0,
      "step": 38055
    },
    {
      "entropy": 5.515823316574097,
      "epoch": 3.6560999039385207,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003711315145599398,
      "loss": 4.7376,
      "mean_token_accuracy": 0.25224713534116744,
      "num_tokens": 87239630.0,
      "step": 38060
    },
    {
      "entropy": 5.501846265792847,
      "epoch": 3.6565802113352546,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000371100516551169,
      "loss": 4.7098,
      "mean_token_accuracy": 0.25202121734619143,
      "num_tokens": 87250008.0,
      "step": 38065
    },
    {
      "entropy": 5.474589538574219,
      "epoch": 3.6570605187319885,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003710695163112295,
      "loss": 4.704,
      "mean_token_accuracy": 0.25481015294790266,
      "num_tokens": 87260802.0,
      "step": 38070
    },
    {
      "entropy": 5.508035135269165,
      "epoch": 3.6575408261287223,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003710385138408412,
      "loss": 4.8197,
      "mean_token_accuracy": 0.24998641908168792,
      "num_tokens": 87271960.0,
      "step": 38075
    },
    {
      "entropy": 5.398660182952881,
      "epoch": 3.658021133525456,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003710075091407237,
      "loss": 4.6548,
      "mean_token_accuracy": 0.2513419210910797,
      "num_tokens": 87285138.0,
      "step": 38080
    },
    {
      "entropy": 5.468804025650025,
      "epoch": 3.65850144092219,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003709765022115969,
      "loss": 4.7065,
      "mean_token_accuracy": 0.24952240884304047,
      "num_tokens": 87296650.0,
      "step": 38085
    },
    {
      "entropy": 5.45039119720459,
      "epoch": 3.658981748318924,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003709454930541808,
      "loss": 4.706,
      "mean_token_accuracy": 0.2549412250518799,
      "num_tokens": 87307315.0,
      "step": 38090
    },
    {
      "entropy": 5.403020811080933,
      "epoch": 3.659462055715658,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037091448166919516,
      "loss": 4.5745,
      "mean_token_accuracy": 0.2604144215583801,
      "num_tokens": 87318842.0,
      "step": 38095
    },
    {
      "entropy": 5.443148612976074,
      "epoch": 3.6599423631123917,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037088346805736014,
      "loss": 4.6276,
      "mean_token_accuracy": 0.26228197067976,
      "num_tokens": 87329308.0,
      "step": 38100
    },
    {
      "entropy": 5.507600164413452,
      "epoch": 3.660422670509126,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003708524522193957,
      "loss": 4.7876,
      "mean_token_accuracy": 0.24187227785587312,
      "num_tokens": 87340186.0,
      "step": 38105
    },
    {
      "entropy": 5.530103349685669,
      "epoch": 3.6609029779058595,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037082143415602186,
      "loss": 4.8011,
      "mean_token_accuracy": 0.23927779644727706,
      "num_tokens": 87352547.0,
      "step": 38110
    },
    {
      "entropy": 5.476965522766113,
      "epoch": 3.661383285302594,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003707904138679589,
      "loss": 4.7493,
      "mean_token_accuracy": 0.24897976964712143,
      "num_tokens": 87363602.0,
      "step": 38115
    },
    {
      "entropy": 5.487685489654541,
      "epoch": 3.6618635926993277,
      "grad_norm": 1.171875,
      "learning_rate": 0.000370759391355927,
      "loss": 4.7418,
      "mean_token_accuracy": 0.24877915680408477,
      "num_tokens": 87375127.0,
      "step": 38120
    },
    {
      "entropy": 5.513380718231201,
      "epoch": 3.6623439000960616,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003707283666206464,
      "loss": 4.7566,
      "mean_token_accuracy": 0.2492217868566513,
      "num_tokens": 87386436.0,
      "step": 38125
    },
    {
      "entropy": 5.4438581466674805,
      "epoch": 3.6628242074927955,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003706973396628373,
      "loss": 4.7083,
      "mean_token_accuracy": 0.25373556911945344,
      "num_tokens": 87399584.0,
      "step": 38130
    },
    {
      "entropy": 5.523203897476196,
      "epoch": 3.6633045148895294,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003706663104832202,
      "loss": 4.7585,
      "mean_token_accuracy": 0.24550649970769883,
      "num_tokens": 87410244.0,
      "step": 38135
    },
    {
      "entropy": 5.456813335418701,
      "epoch": 3.6637848222862632,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037063527908251536,
      "loss": 4.664,
      "mean_token_accuracy": 0.25745151489973067,
      "num_tokens": 87421907.0,
      "step": 38140
    },
    {
      "entropy": 5.404771614074707,
      "epoch": 3.664265129682997,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003706042454614434,
      "loss": 4.5825,
      "mean_token_accuracy": 0.2626073777675629,
      "num_tokens": 87432202.0,
      "step": 38145
    },
    {
      "entropy": 5.39242525100708,
      "epoch": 3.664745437079731,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003705732096207247,
      "loss": 4.6084,
      "mean_token_accuracy": 0.25889242738485335,
      "num_tokens": 87443460.0,
      "step": 38150
    },
    {
      "entropy": 5.3990613460540775,
      "epoch": 3.665225744476465,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037054217156107987,
      "loss": 4.6611,
      "mean_token_accuracy": 0.25907652378082274,
      "num_tokens": 87455900.0,
      "step": 38155
    },
    {
      "entropy": 5.4942010879516605,
      "epoch": 3.6657060518731988,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037051113128322953,
      "loss": 4.7171,
      "mean_token_accuracy": 0.2501246467232704,
      "num_tokens": 87467593.0,
      "step": 38160
    },
    {
      "entropy": 5.473310136795044,
      "epoch": 3.6661863592699326,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037048008878789437,
      "loss": 4.7068,
      "mean_token_accuracy": 0.24786543399095534,
      "num_tokens": 87479230.0,
      "step": 38165
    },
    {
      "entropy": 5.427902030944824,
      "epoch": 3.6666666666666665,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037044904407579506,
      "loss": 4.6678,
      "mean_token_accuracy": 0.2641862541437149,
      "num_tokens": 87489777.0,
      "step": 38170
    },
    {
      "entropy": 5.464749002456665,
      "epoch": 3.6671469740634004,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037041799714765226,
      "loss": 4.789,
      "mean_token_accuracy": 0.2400815263390541,
      "num_tokens": 87501202.0,
      "step": 38175
    },
    {
      "entropy": 5.47803750038147,
      "epoch": 3.6676272814601347,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000370386948004187,
      "loss": 4.7151,
      "mean_token_accuracy": 0.25264199525117875,
      "num_tokens": 87512684.0,
      "step": 38180
    },
    {
      "entropy": 5.533303260803223,
      "epoch": 3.668107588856868,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037035589664612004,
      "loss": 4.7113,
      "mean_token_accuracy": 0.25584533959627154,
      "num_tokens": 87525905.0,
      "step": 38185
    },
    {
      "entropy": 5.419366502761841,
      "epoch": 3.6685878962536025,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003703248430741723,
      "loss": 4.662,
      "mean_token_accuracy": 0.25366371870040894,
      "num_tokens": 87536764.0,
      "step": 38190
    },
    {
      "entropy": 5.510813617706299,
      "epoch": 3.6690682036503364,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003702937872890647,
      "loss": 4.7512,
      "mean_token_accuracy": 0.248629193007946,
      "num_tokens": 87548807.0,
      "step": 38195
    },
    {
      "entropy": 5.3477167129516605,
      "epoch": 3.6695485110470702,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003702627292915184,
      "loss": 4.5824,
      "mean_token_accuracy": 0.260172875225544,
      "num_tokens": 87559570.0,
      "step": 38200
    },
    {
      "entropy": 5.503118228912354,
      "epoch": 3.670028818443804,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003702316690822544,
      "loss": 4.7561,
      "mean_token_accuracy": 0.24869336187839508,
      "num_tokens": 87570758.0,
      "step": 38205
    },
    {
      "entropy": 5.421717548370362,
      "epoch": 3.670509125840538,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037020060666199386,
      "loss": 4.5958,
      "mean_token_accuracy": 0.25436814725399015,
      "num_tokens": 87581985.0,
      "step": 38210
    },
    {
      "entropy": 5.435144758224487,
      "epoch": 3.670989433237272,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003701695420314578,
      "loss": 4.702,
      "mean_token_accuracy": 0.24760954976081848,
      "num_tokens": 87595130.0,
      "step": 38215
    },
    {
      "entropy": 5.486284923553467,
      "epoch": 3.6714697406340058,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037013847519136765,
      "loss": 4.7338,
      "mean_token_accuracy": 0.24651833474636078,
      "num_tokens": 87608073.0,
      "step": 38220
    },
    {
      "entropy": 5.381429052352905,
      "epoch": 3.6719500480307397,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003701074061424446,
      "loss": 4.6016,
      "mean_token_accuracy": 0.2669172868132591,
      "num_tokens": 87618889.0,
      "step": 38225
    },
    {
      "entropy": 5.4366944313049315,
      "epoch": 3.6724303554274735,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037007633488541,
      "loss": 4.6144,
      "mean_token_accuracy": 0.26342170536518095,
      "num_tokens": 87630386.0,
      "step": 38230
    },
    {
      "entropy": 5.511225318908691,
      "epoch": 3.6729106628242074,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037004526142098517,
      "loss": 4.7699,
      "mean_token_accuracy": 0.24544866681098937,
      "num_tokens": 87642761.0,
      "step": 38235
    },
    {
      "entropy": 5.492319917678833,
      "epoch": 3.6733909702209413,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037001418574989165,
      "loss": 4.7236,
      "mean_token_accuracy": 0.2483659192919731,
      "num_tokens": 87653661.0,
      "step": 38240
    },
    {
      "entropy": 5.398644161224365,
      "epoch": 3.673871277617675,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036998310787285084,
      "loss": 4.682,
      "mean_token_accuracy": 0.25857133269309995,
      "num_tokens": 87665025.0,
      "step": 38245
    },
    {
      "entropy": 5.489118576049805,
      "epoch": 3.674351585014409,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003699520277905843,
      "loss": 4.7132,
      "mean_token_accuracy": 0.2523573160171509,
      "num_tokens": 87676349.0,
      "step": 38250
    },
    {
      "entropy": 5.397767877578735,
      "epoch": 3.6748318924111434,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003699209455038137,
      "loss": 4.5746,
      "mean_token_accuracy": 0.2638733103871346,
      "num_tokens": 87688149.0,
      "step": 38255
    },
    {
      "entropy": 5.4717369556427,
      "epoch": 3.675312199807877,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036988986101326056,
      "loss": 4.6364,
      "mean_token_accuracy": 0.25743257999420166,
      "num_tokens": 87699323.0,
      "step": 38260
    },
    {
      "entropy": 5.504780864715576,
      "epoch": 3.675792507204611,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003698587743196465,
      "loss": 4.7036,
      "mean_token_accuracy": 0.2558631539344788,
      "num_tokens": 87709994.0,
      "step": 38265
    },
    {
      "entropy": 5.411701297760009,
      "epoch": 3.6762728146013446,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036982768542369353,
      "loss": 4.7178,
      "mean_token_accuracy": 0.254097743332386,
      "num_tokens": 87721979.0,
      "step": 38270
    },
    {
      "entropy": 5.4818370819091795,
      "epoch": 3.676753121998079,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036979659432612324,
      "loss": 4.6862,
      "mean_token_accuracy": 0.25241653621196747,
      "num_tokens": 87732194.0,
      "step": 38275
    },
    {
      "entropy": 5.382015037536621,
      "epoch": 3.677233429394813,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003697655010276575,
      "loss": 4.5773,
      "mean_token_accuracy": 0.26404929459095,
      "num_tokens": 87742736.0,
      "step": 38280
    },
    {
      "entropy": 5.445033597946167,
      "epoch": 3.6777137367915467,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003697344055290182,
      "loss": 4.6797,
      "mean_token_accuracy": 0.25588560849428177,
      "num_tokens": 87753557.0,
      "step": 38285
    },
    {
      "entropy": 5.4767251968383786,
      "epoch": 3.6781940441882806,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003697033078309274,
      "loss": 4.7483,
      "mean_token_accuracy": 0.2482931062579155,
      "num_tokens": 87766585.0,
      "step": 38290
    },
    {
      "entropy": 5.46252293586731,
      "epoch": 3.6786743515850144,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036967220793410687,
      "loss": 4.7552,
      "mean_token_accuracy": 0.24849268794059753,
      "num_tokens": 87777679.0,
      "step": 38295
    },
    {
      "entropy": 5.4304310321807865,
      "epoch": 3.6791546589817483,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003696411058392789,
      "loss": 4.6363,
      "mean_token_accuracy": 0.25999634265899657,
      "num_tokens": 87790083.0,
      "step": 38300
    },
    {
      "entropy": 5.430845022201538,
      "epoch": 3.679634966378482,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003696100015471654,
      "loss": 4.6366,
      "mean_token_accuracy": 0.2609268814325333,
      "num_tokens": 87802293.0,
      "step": 38305
    },
    {
      "entropy": 5.393534851074219,
      "epoch": 3.680115273775216,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003695788950584886,
      "loss": 4.6499,
      "mean_token_accuracy": 0.255362406373024,
      "num_tokens": 87813367.0,
      "step": 38310
    },
    {
      "entropy": 5.4174353122711185,
      "epoch": 3.68059558117195,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036954778637397084,
      "loss": 4.6727,
      "mean_token_accuracy": 0.2536328822374344,
      "num_tokens": 87824669.0,
      "step": 38315
    },
    {
      "entropy": 5.476462507247925,
      "epoch": 3.681075888568684,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036951667549433404,
      "loss": 4.82,
      "mean_token_accuracy": 0.25047376900911333,
      "num_tokens": 87836728.0,
      "step": 38320
    },
    {
      "entropy": 5.51965446472168,
      "epoch": 3.6815561959654177,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036948556242030083,
      "loss": 4.7897,
      "mean_token_accuracy": 0.24105577766895295,
      "num_tokens": 87848391.0,
      "step": 38325
    },
    {
      "entropy": 5.540651845932007,
      "epoch": 3.682036503362152,
      "grad_norm": 1.125,
      "learning_rate": 0.00036945444715259334,
      "loss": 4.7451,
      "mean_token_accuracy": 0.2568321079015732,
      "num_tokens": 87860837.0,
      "step": 38330
    },
    {
      "entropy": 5.474909830093384,
      "epoch": 3.6825168107588855,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003694233296919341,
      "loss": 4.7432,
      "mean_token_accuracy": 0.24913543611764907,
      "num_tokens": 87873766.0,
      "step": 38335
    },
    {
      "entropy": 5.471793746948242,
      "epoch": 3.68299711815562,
      "grad_norm": 1.125,
      "learning_rate": 0.00036939221003904547,
      "loss": 4.712,
      "mean_token_accuracy": 0.25766322314739226,
      "num_tokens": 87885151.0,
      "step": 38340
    },
    {
      "entropy": 5.358002996444702,
      "epoch": 3.6834774255523532,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036936108819465006,
      "loss": 4.58,
      "mean_token_accuracy": 0.2671238571405411,
      "num_tokens": 87896967.0,
      "step": 38345
    },
    {
      "entropy": 5.446291780471801,
      "epoch": 3.6839577329490876,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036932996415947047,
      "loss": 4.6897,
      "mean_token_accuracy": 0.24891044497489928,
      "num_tokens": 87908392.0,
      "step": 38350
    },
    {
      "entropy": 5.5227879047393795,
      "epoch": 3.6844380403458215,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036929883793422906,
      "loss": 4.8113,
      "mean_token_accuracy": 0.24961352795362474,
      "num_tokens": 87920409.0,
      "step": 38355
    },
    {
      "entropy": 5.467597484588623,
      "epoch": 3.6849183477425553,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003692677095196487,
      "loss": 4.7275,
      "mean_token_accuracy": 0.25097474455833435,
      "num_tokens": 87931609.0,
      "step": 38360
    },
    {
      "entropy": 5.528201246261597,
      "epoch": 3.685398655139289,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000369236578916452,
      "loss": 4.8065,
      "mean_token_accuracy": 0.23090119361877443,
      "num_tokens": 87943599.0,
      "step": 38365
    },
    {
      "entropy": 5.503591632843017,
      "epoch": 3.685878962536023,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036920544612536186,
      "loss": 4.7116,
      "mean_token_accuracy": 0.24814435690641404,
      "num_tokens": 87953659.0,
      "step": 38370
    },
    {
      "entropy": 5.5202313423156735,
      "epoch": 3.686359269932757,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003691743111471009,
      "loss": 4.8087,
      "mean_token_accuracy": 0.23560016304254533,
      "num_tokens": 87966174.0,
      "step": 38375
    },
    {
      "entropy": 5.467265462875366,
      "epoch": 3.686839577329491,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003691431739823922,
      "loss": 4.6979,
      "mean_token_accuracy": 0.2502520874142647,
      "num_tokens": 87978804.0,
      "step": 38380
    },
    {
      "entropy": 5.523390436172486,
      "epoch": 3.6873198847262247,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036911203463195843,
      "loss": 4.7119,
      "mean_token_accuracy": 0.2562424331903458,
      "num_tokens": 87989807.0,
      "step": 38385
    },
    {
      "entropy": 5.430996131896973,
      "epoch": 3.6878001921229586,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036908089309652266,
      "loss": 4.6791,
      "mean_token_accuracy": 0.2567839428782463,
      "num_tokens": 88000978.0,
      "step": 38390
    },
    {
      "entropy": 5.4663331508636475,
      "epoch": 3.6882804995196925,
      "grad_norm": 1.21875,
      "learning_rate": 0.000369049749376808,
      "loss": 4.7456,
      "mean_token_accuracy": 0.253655044734478,
      "num_tokens": 88012421.0,
      "step": 38395
    },
    {
      "entropy": 5.518753147125244,
      "epoch": 3.6887608069164264,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036901860347353735,
      "loss": 4.7497,
      "mean_token_accuracy": 0.25511270612478254,
      "num_tokens": 88023315.0,
      "step": 38400
    },
    {
      "entropy": 5.453296422958374,
      "epoch": 3.6892411143131603,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000368987455387434,
      "loss": 4.7095,
      "mean_token_accuracy": 0.25909202843904494,
      "num_tokens": 88035368.0,
      "step": 38405
    },
    {
      "entropy": 5.396697950363159,
      "epoch": 3.689721421709894,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003689563051192209,
      "loss": 4.6915,
      "mean_token_accuracy": 0.2557314783334732,
      "num_tokens": 88045809.0,
      "step": 38410
    },
    {
      "entropy": 5.510910367965698,
      "epoch": 3.6902017291066285,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036892515266962145,
      "loss": 4.7278,
      "mean_token_accuracy": 0.2547199338674545,
      "num_tokens": 88056456.0,
      "step": 38415
    },
    {
      "entropy": 5.518176889419555,
      "epoch": 3.690682036503362,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003688939980393588,
      "loss": 4.7199,
      "mean_token_accuracy": 0.25056171268224714,
      "num_tokens": 88069387.0,
      "step": 38420
    },
    {
      "entropy": 5.5294147491455075,
      "epoch": 3.6911623439000962,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003688628412291564,
      "loss": 4.7146,
      "mean_token_accuracy": 0.2486061543226242,
      "num_tokens": 88079393.0,
      "step": 38425
    },
    {
      "entropy": 5.54340353012085,
      "epoch": 3.69164265129683,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036883168223973754,
      "loss": 4.7459,
      "mean_token_accuracy": 0.2420138329267502,
      "num_tokens": 88089151.0,
      "step": 38430
    },
    {
      "entropy": 5.5008831977844235,
      "epoch": 3.692122958693564,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003688005210718257,
      "loss": 4.6948,
      "mean_token_accuracy": 0.24936534762382506,
      "num_tokens": 88100497.0,
      "step": 38435
    },
    {
      "entropy": 5.551054906845093,
      "epoch": 3.692603266090298,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003687693577261442,
      "loss": 4.7968,
      "mean_token_accuracy": 0.2444600060582161,
      "num_tokens": 88112818.0,
      "step": 38440
    },
    {
      "entropy": 5.557637453079224,
      "epoch": 3.6930835734870318,
      "grad_norm": 1.25,
      "learning_rate": 0.00036873819220341667,
      "loss": 4.746,
      "mean_token_accuracy": 0.2515460342168808,
      "num_tokens": 88123128.0,
      "step": 38445
    },
    {
      "entropy": 5.4180501937866214,
      "epoch": 3.6935638808837656,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036870702450436655,
      "loss": 4.6934,
      "mean_token_accuracy": 0.2558251515030861,
      "num_tokens": 88136062.0,
      "step": 38450
    },
    {
      "entropy": 5.5300075054168705,
      "epoch": 3.6940441882804995,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036867585462971775,
      "loss": 4.777,
      "mean_token_accuracy": 0.24212975800037384,
      "num_tokens": 88147234.0,
      "step": 38455
    },
    {
      "entropy": 5.4155683517456055,
      "epoch": 3.6945244956772334,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003686446825801937,
      "loss": 4.6778,
      "mean_token_accuracy": 0.25313362032175063,
      "num_tokens": 88158537.0,
      "step": 38460
    },
    {
      "entropy": 5.472520256042481,
      "epoch": 3.6950048030739673,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003686135083565181,
      "loss": 4.6744,
      "mean_token_accuracy": 0.25624648928642274,
      "num_tokens": 88169652.0,
      "step": 38465
    },
    {
      "entropy": 5.527430725097656,
      "epoch": 3.695485110470701,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036858233195941493,
      "loss": 4.7761,
      "mean_token_accuracy": 0.23938588351011275,
      "num_tokens": 88181919.0,
      "step": 38470
    },
    {
      "entropy": 5.407576990127564,
      "epoch": 3.695965417867435,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003685511533896079,
      "loss": 4.65,
      "mean_token_accuracy": 0.254591204226017,
      "num_tokens": 88194070.0,
      "step": 38475
    },
    {
      "entropy": 5.474805402755737,
      "epoch": 3.696445725264169,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036851997264782083,
      "loss": 4.7046,
      "mean_token_accuracy": 0.24697917848825454,
      "num_tokens": 88206410.0,
      "step": 38480
    },
    {
      "entropy": 5.434221887588501,
      "epoch": 3.696926032660903,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036848878973477767,
      "loss": 4.7419,
      "mean_token_accuracy": 0.2550902470946312,
      "num_tokens": 88218921.0,
      "step": 38485
    },
    {
      "entropy": 5.506269884109497,
      "epoch": 3.697406340057637,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036845760465120247,
      "loss": 4.6978,
      "mean_token_accuracy": 0.2538252428174019,
      "num_tokens": 88231026.0,
      "step": 38490
    },
    {
      "entropy": 5.602132558822632,
      "epoch": 3.6978866474543706,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003684264173978192,
      "loss": 4.8097,
      "mean_token_accuracy": 0.24987199306488037,
      "num_tokens": 88243731.0,
      "step": 38495
    },
    {
      "entropy": 5.4390558242797855,
      "epoch": 3.698366954851105,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003683952279753519,
      "loss": 4.6373,
      "mean_token_accuracy": 0.26003620624542234,
      "num_tokens": 88254890.0,
      "step": 38500
    },
    {
      "entropy": 5.460673666000366,
      "epoch": 3.6988472622478388,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036836403638452477,
      "loss": 4.7181,
      "mean_token_accuracy": 0.2438870683312416,
      "num_tokens": 88267169.0,
      "step": 38505
    },
    {
      "entropy": 5.555178642272949,
      "epoch": 3.6993275696445727,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003683328426260619,
      "loss": 4.8423,
      "mean_token_accuracy": 0.24317368268966674,
      "num_tokens": 88278635.0,
      "step": 38510
    },
    {
      "entropy": 5.472474765777588,
      "epoch": 3.6998078770413065,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036830164670068754,
      "loss": 4.6673,
      "mean_token_accuracy": 0.2616072237491608,
      "num_tokens": 88290155.0,
      "step": 38515
    },
    {
      "entropy": 5.524127388000489,
      "epoch": 3.7002881844380404,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036827044860912605,
      "loss": 4.73,
      "mean_token_accuracy": 0.24552934467792512,
      "num_tokens": 88301250.0,
      "step": 38520
    },
    {
      "entropy": 5.472802972793579,
      "epoch": 3.7007684918347743,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003682392483521017,
      "loss": 4.7344,
      "mean_token_accuracy": 0.25309175103902815,
      "num_tokens": 88312214.0,
      "step": 38525
    },
    {
      "entropy": 5.458769845962524,
      "epoch": 3.701248799231508,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036820804593033893,
      "loss": 4.7372,
      "mean_token_accuracy": 0.24743417650461197,
      "num_tokens": 88322793.0,
      "step": 38530
    },
    {
      "entropy": 5.46553144454956,
      "epoch": 3.701729106628242,
      "grad_norm": 1.21875,
      "learning_rate": 0.000368176841344562,
      "loss": 4.7046,
      "mean_token_accuracy": 0.2485368460416794,
      "num_tokens": 88335032.0,
      "step": 38535
    },
    {
      "entropy": 5.478254604339599,
      "epoch": 3.702209414024976,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036814563459549557,
      "loss": 4.6821,
      "mean_token_accuracy": 0.25288943350315096,
      "num_tokens": 88348186.0,
      "step": 38540
    },
    {
      "entropy": 5.5660624504089355,
      "epoch": 3.70268972142171,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036811442568386403,
      "loss": 4.7606,
      "mean_token_accuracy": 0.24690347164869308,
      "num_tokens": 88358117.0,
      "step": 38545
    },
    {
      "entropy": 5.578690719604492,
      "epoch": 3.7031700288184437,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036808321461039207,
      "loss": 4.8264,
      "mean_token_accuracy": 0.2383086159825325,
      "num_tokens": 88368606.0,
      "step": 38550
    },
    {
      "entropy": 5.499757242202759,
      "epoch": 3.7036503362151776,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003680520013758042,
      "loss": 4.7205,
      "mean_token_accuracy": 0.24842409193515777,
      "num_tokens": 88379676.0,
      "step": 38555
    },
    {
      "entropy": 5.514234018325806,
      "epoch": 3.7041306436119115,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036802078598082523,
      "loss": 4.762,
      "mean_token_accuracy": 0.25002527981996536,
      "num_tokens": 88390824.0,
      "step": 38560
    },
    {
      "entropy": 5.492612552642822,
      "epoch": 3.704610951008646,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003679895684261798,
      "loss": 4.7482,
      "mean_token_accuracy": 0.2452264755964279,
      "num_tokens": 88401783.0,
      "step": 38565
    },
    {
      "entropy": 5.569248533248901,
      "epoch": 3.7050912584053792,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003679583487125927,
      "loss": 4.7947,
      "mean_token_accuracy": 0.24277548342943192,
      "num_tokens": 88414366.0,
      "step": 38570
    },
    {
      "entropy": 5.511212253570557,
      "epoch": 3.7055715658021136,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003679271268407888,
      "loss": 4.7984,
      "mean_token_accuracy": 0.2396158203482628,
      "num_tokens": 88425458.0,
      "step": 38575
    },
    {
      "entropy": 5.520334911346436,
      "epoch": 3.706051873198847,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003678959028114929,
      "loss": 4.7978,
      "mean_token_accuracy": 0.24296584874391555,
      "num_tokens": 88435766.0,
      "step": 38580
    },
    {
      "entropy": 5.5195465087890625,
      "epoch": 3.7065321805955813,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036786467662543,
      "loss": 4.7192,
      "mean_token_accuracy": 0.24991735816001892,
      "num_tokens": 88447807.0,
      "step": 38585
    },
    {
      "entropy": 5.466838026046753,
      "epoch": 3.707012487992315,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003678334482833251,
      "loss": 4.7258,
      "mean_token_accuracy": 0.25328120589256287,
      "num_tokens": 88459248.0,
      "step": 38590
    },
    {
      "entropy": 5.461731815338135,
      "epoch": 3.707492795389049,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036780221778590316,
      "loss": 4.655,
      "mean_token_accuracy": 0.26124205738306044,
      "num_tokens": 88471094.0,
      "step": 38595
    },
    {
      "entropy": 5.527285099029541,
      "epoch": 3.707973102785783,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003677709851338893,
      "loss": 4.7773,
      "mean_token_accuracy": 0.2508664220571518,
      "num_tokens": 88482154.0,
      "step": 38600
    },
    {
      "entropy": 5.498313093185425,
      "epoch": 3.708453410182517,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036773975032800867,
      "loss": 4.6947,
      "mean_token_accuracy": 0.2554201826453209,
      "num_tokens": 88493267.0,
      "step": 38605
    },
    {
      "entropy": 5.450923919677734,
      "epoch": 3.7089337175792507,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003677085133689864,
      "loss": 4.7404,
      "mean_token_accuracy": 0.25418524593114855,
      "num_tokens": 88505510.0,
      "step": 38610
    },
    {
      "entropy": 5.437158298492432,
      "epoch": 3.7094140249759846,
      "grad_norm": 1.125,
      "learning_rate": 0.00036767727425754767,
      "loss": 4.6761,
      "mean_token_accuracy": 0.25304255187511443,
      "num_tokens": 88516657.0,
      "step": 38615
    },
    {
      "entropy": 5.519437599182129,
      "epoch": 3.7098943323727185,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003676460329944179,
      "loss": 4.7433,
      "mean_token_accuracy": 0.24908487796783446,
      "num_tokens": 88527470.0,
      "step": 38620
    },
    {
      "entropy": 5.583340978622436,
      "epoch": 3.7103746397694524,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003676147895803223,
      "loss": 4.8224,
      "mean_token_accuracy": 0.24797419160604478,
      "num_tokens": 88539267.0,
      "step": 38625
    },
    {
      "entropy": 5.386899518966675,
      "epoch": 3.7108549471661862,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003675835440159863,
      "loss": 4.6029,
      "mean_token_accuracy": 0.26475532352924347,
      "num_tokens": 88551466.0,
      "step": 38630
    },
    {
      "entropy": 5.412972736358642,
      "epoch": 3.71133525456292,
      "grad_norm": 1.25,
      "learning_rate": 0.0003675522963021354,
      "loss": 4.6491,
      "mean_token_accuracy": 0.25852490663528443,
      "num_tokens": 88563991.0,
      "step": 38635
    },
    {
      "entropy": 5.5567010879516605,
      "epoch": 3.7118155619596545,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036752104643949493,
      "loss": 4.8333,
      "mean_token_accuracy": 0.24269133806228638,
      "num_tokens": 88574308.0,
      "step": 38640
    },
    {
      "entropy": 5.546703624725342,
      "epoch": 3.712295869356388,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003674897944287904,
      "loss": 4.7734,
      "mean_token_accuracy": 0.24347683489322663,
      "num_tokens": 88585510.0,
      "step": 38645
    },
    {
      "entropy": 5.425597906112671,
      "epoch": 3.712776176753122,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003674585402707477,
      "loss": 4.6173,
      "mean_token_accuracy": 0.26182337254285815,
      "num_tokens": 88596575.0,
      "step": 38650
    },
    {
      "entropy": 5.545491409301758,
      "epoch": 3.7132564841498557,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003674272839660921,
      "loss": 4.838,
      "mean_token_accuracy": 0.24222707897424697,
      "num_tokens": 88608549.0,
      "step": 38655
    },
    {
      "entropy": 5.3976929664611815,
      "epoch": 3.71373679154659,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036739602551554945,
      "loss": 4.673,
      "mean_token_accuracy": 0.25746863782405854,
      "num_tokens": 88620611.0,
      "step": 38660
    },
    {
      "entropy": 5.544831705093384,
      "epoch": 3.714217098943324,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003673647649198454,
      "loss": 4.8001,
      "mean_token_accuracy": 0.24469462186098098,
      "num_tokens": 88630600.0,
      "step": 38665
    },
    {
      "entropy": 5.452911806106568,
      "epoch": 3.7146974063400577,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036733350217970593,
      "loss": 4.6456,
      "mean_token_accuracy": 0.25616577416658404,
      "num_tokens": 88642846.0,
      "step": 38670
    },
    {
      "entropy": 5.44217438697815,
      "epoch": 3.7151777137367916,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036730223729585654,
      "loss": 4.716,
      "mean_token_accuracy": 0.2530002251267433,
      "num_tokens": 88654981.0,
      "step": 38675
    },
    {
      "entropy": 5.583939886093139,
      "epoch": 3.7156580211335255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003672709702690234,
      "loss": 4.7952,
      "mean_token_accuracy": 0.24524906277656555,
      "num_tokens": 88666737.0,
      "step": 38680
    },
    {
      "entropy": 5.510693693161011,
      "epoch": 3.7161383285302594,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036723970109993223,
      "loss": 4.6892,
      "mean_token_accuracy": 0.25506328493356706,
      "num_tokens": 88678037.0,
      "step": 38685
    },
    {
      "entropy": 5.514085149765014,
      "epoch": 3.7166186359269933,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003672084297893091,
      "loss": 4.6805,
      "mean_token_accuracy": 0.25483577847480776,
      "num_tokens": 88689215.0,
      "step": 38690
    },
    {
      "entropy": 5.371277332305908,
      "epoch": 3.717098943323727,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036717715633788014,
      "loss": 4.6838,
      "mean_token_accuracy": 0.25476180464029313,
      "num_tokens": 88701848.0,
      "step": 38695
    },
    {
      "entropy": 5.422053956985474,
      "epoch": 3.717579250720461,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036714588074637116,
      "loss": 4.6268,
      "mean_token_accuracy": 0.2569938898086548,
      "num_tokens": 88712745.0,
      "step": 38700
    },
    {
      "entropy": 5.404322910308838,
      "epoch": 3.718059558117195,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003671146030155086,
      "loss": 4.6156,
      "mean_token_accuracy": 0.26125945448875426,
      "num_tokens": 88725048.0,
      "step": 38705
    },
    {
      "entropy": 5.427138948440552,
      "epoch": 3.718539865513929,
      "grad_norm": 1.125,
      "learning_rate": 0.0003670833231460184,
      "loss": 4.6648,
      "mean_token_accuracy": 0.2553469017148018,
      "num_tokens": 88736218.0,
      "step": 38710
    },
    {
      "entropy": 5.554109048843384,
      "epoch": 3.7190201729106627,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036705204113862683,
      "loss": 4.8091,
      "mean_token_accuracy": 0.24781275689601898,
      "num_tokens": 88747967.0,
      "step": 38715
    },
    {
      "entropy": 5.529304122924804,
      "epoch": 3.7195004803073966,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003670207569940603,
      "loss": 4.6981,
      "mean_token_accuracy": 0.2526485949754715,
      "num_tokens": 88759565.0,
      "step": 38720
    },
    {
      "entropy": 5.385045337677002,
      "epoch": 3.719980787704131,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036698947071304495,
      "loss": 4.5717,
      "mean_token_accuracy": 0.26471147537231443,
      "num_tokens": 88771153.0,
      "step": 38725
    },
    {
      "entropy": 5.501430177688599,
      "epoch": 3.7204610951008643,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036695818229630735,
      "loss": 4.7866,
      "mean_token_accuracy": 0.24974988251924515,
      "num_tokens": 88782615.0,
      "step": 38730
    },
    {
      "entropy": 5.494764995574951,
      "epoch": 3.7209414024975986,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003669268917445737,
      "loss": 4.7404,
      "mean_token_accuracy": 0.25167073756456376,
      "num_tokens": 88794054.0,
      "step": 38735
    },
    {
      "entropy": 5.464716815948487,
      "epoch": 3.7214217098943325,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003668955990585706,
      "loss": 4.7071,
      "mean_token_accuracy": 0.25608427077531815,
      "num_tokens": 88804094.0,
      "step": 38740
    },
    {
      "entropy": 5.466986894607544,
      "epoch": 3.7219020172910664,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036686430423902466,
      "loss": 4.7182,
      "mean_token_accuracy": 0.2462858572602272,
      "num_tokens": 88814661.0,
      "step": 38745
    },
    {
      "entropy": 5.446202945709229,
      "epoch": 3.7223823246878003,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003668330072866623,
      "loss": 4.7388,
      "mean_token_accuracy": 0.245011343061924,
      "num_tokens": 88825256.0,
      "step": 38750
    },
    {
      "entropy": 5.477546167373657,
      "epoch": 3.722862632084534,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036680170820221013,
      "loss": 4.7208,
      "mean_token_accuracy": 0.2524253711104393,
      "num_tokens": 88835914.0,
      "step": 38755
    },
    {
      "entropy": 5.439700031280518,
      "epoch": 3.723342939481268,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000366770406986395,
      "loss": 4.6568,
      "mean_token_accuracy": 0.2560461640357971,
      "num_tokens": 88846847.0,
      "step": 38760
    },
    {
      "entropy": 5.540718078613281,
      "epoch": 3.723823246878002,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003667391036399435,
      "loss": 4.764,
      "mean_token_accuracy": 0.2507845088839531,
      "num_tokens": 88857899.0,
      "step": 38765
    },
    {
      "entropy": 5.459467172622681,
      "epoch": 3.724303554274736,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003667077981635824,
      "loss": 4.6745,
      "mean_token_accuracy": 0.2592965379357338,
      "num_tokens": 88867972.0,
      "step": 38770
    },
    {
      "entropy": 5.465718936920166,
      "epoch": 3.7247838616714697,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036667649055803855,
      "loss": 4.7635,
      "mean_token_accuracy": 0.24648467898368837,
      "num_tokens": 88880241.0,
      "step": 38775
    },
    {
      "entropy": 5.55481071472168,
      "epoch": 3.7252641690682036,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036664518082403876,
      "loss": 4.7539,
      "mean_token_accuracy": 0.25383228808641434,
      "num_tokens": 88891833.0,
      "step": 38780
    },
    {
      "entropy": 5.442574548721313,
      "epoch": 3.7257444764649374,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003666138689623101,
      "loss": 4.6405,
      "mean_token_accuracy": 0.2593009933829308,
      "num_tokens": 88902537.0,
      "step": 38785
    },
    {
      "entropy": 5.413526296615601,
      "epoch": 3.7262247838616713,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036658255497357936,
      "loss": 4.7073,
      "mean_token_accuracy": 0.25460578203201295,
      "num_tokens": 88914499.0,
      "step": 38790
    },
    {
      "entropy": 5.490294933319092,
      "epoch": 3.726705091258405,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036655123885857366,
      "loss": 4.773,
      "mean_token_accuracy": 0.25093507021665573,
      "num_tokens": 88927240.0,
      "step": 38795
    },
    {
      "entropy": 5.4537865161895756,
      "epoch": 3.7271853986551395,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036651992061802007,
      "loss": 4.6771,
      "mean_token_accuracy": 0.25546124279499055,
      "num_tokens": 88939877.0,
      "step": 38800
    },
    {
      "entropy": 5.418507814407349,
      "epoch": 3.727665706051873,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003664886002526457,
      "loss": 4.6649,
      "mean_token_accuracy": 0.25949409455060957,
      "num_tokens": 88952144.0,
      "step": 38805
    },
    {
      "entropy": 5.519764184951782,
      "epoch": 3.7281460134486073,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036645727776317763,
      "loss": 4.7892,
      "mean_token_accuracy": 0.24387079179286958,
      "num_tokens": 88963700.0,
      "step": 38810
    },
    {
      "entropy": 5.573657035827637,
      "epoch": 3.728626320845341,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003664259531503432,
      "loss": 4.7961,
      "mean_token_accuracy": 0.24817942082881927,
      "num_tokens": 88973744.0,
      "step": 38815
    },
    {
      "entropy": 5.478315019607544,
      "epoch": 3.729106628242075,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003663946264148696,
      "loss": 4.6876,
      "mean_token_accuracy": 0.2554256200790405,
      "num_tokens": 88984506.0,
      "step": 38820
    },
    {
      "entropy": 5.539937257766724,
      "epoch": 3.729586935638809,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003663632975574841,
      "loss": 4.7678,
      "mean_token_accuracy": 0.2522327616810799,
      "num_tokens": 88995703.0,
      "step": 38825
    },
    {
      "entropy": 5.443521928787232,
      "epoch": 3.730067243035543,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003663319665789142,
      "loss": 4.6428,
      "mean_token_accuracy": 0.26063731461763384,
      "num_tokens": 89006529.0,
      "step": 38830
    },
    {
      "entropy": 5.462705802917481,
      "epoch": 3.7305475504322767,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003663006334798872,
      "loss": 4.7662,
      "mean_token_accuracy": 0.2447535514831543,
      "num_tokens": 89018555.0,
      "step": 38835
    },
    {
      "entropy": 5.449809312820435,
      "epoch": 3.7310278578290106,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003662692982611306,
      "loss": 4.657,
      "mean_token_accuracy": 0.25916541963815687,
      "num_tokens": 89029603.0,
      "step": 38840
    },
    {
      "entropy": 5.471740293502807,
      "epoch": 3.7315081652257445,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036623796092337196,
      "loss": 4.7121,
      "mean_token_accuracy": 0.25260649174451827,
      "num_tokens": 89041560.0,
      "step": 38845
    },
    {
      "entropy": 5.525808811187744,
      "epoch": 3.7319884726224783,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003662066214673388,
      "loss": 4.7982,
      "mean_token_accuracy": 0.2489354819059372,
      "num_tokens": 89052133.0,
      "step": 38850
    },
    {
      "entropy": 5.458691120147705,
      "epoch": 3.7324687800192122,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003661752798937586,
      "loss": 4.6381,
      "mean_token_accuracy": 0.2559501141309738,
      "num_tokens": 89064829.0,
      "step": 38855
    },
    {
      "entropy": 5.570803833007813,
      "epoch": 3.732949087415946,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036614393620335923,
      "loss": 4.8379,
      "mean_token_accuracy": 0.23987323343753814,
      "num_tokens": 89076310.0,
      "step": 38860
    },
    {
      "entropy": 5.488734531402588,
      "epoch": 3.73342939481268,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003661125903968683,
      "loss": 4.7425,
      "mean_token_accuracy": 0.2439393073320389,
      "num_tokens": 89088556.0,
      "step": 38865
    },
    {
      "entropy": 5.435473442077637,
      "epoch": 3.733909702209414,
      "grad_norm": 1.125,
      "learning_rate": 0.0003660812424750135,
      "loss": 4.597,
      "mean_token_accuracy": 0.262081541121006,
      "num_tokens": 89100020.0,
      "step": 38870
    },
    {
      "entropy": 5.427335214614868,
      "epoch": 3.734390009606148,
      "grad_norm": 1.25,
      "learning_rate": 0.0003660498924385228,
      "loss": 4.5947,
      "mean_token_accuracy": 0.2666704714298248,
      "num_tokens": 89111538.0,
      "step": 38875
    },
    {
      "entropy": 5.4645452976226805,
      "epoch": 3.7348703170028816,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036601854028812394,
      "loss": 4.7166,
      "mean_token_accuracy": 0.2513332635164261,
      "num_tokens": 89122674.0,
      "step": 38880
    },
    {
      "entropy": 5.518290615081787,
      "epoch": 3.735350624399616,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036598718602454474,
      "loss": 4.7464,
      "mean_token_accuracy": 0.2528709605336189,
      "num_tokens": 89135165.0,
      "step": 38885
    },
    {
      "entropy": 5.471973848342896,
      "epoch": 3.7358309317963494,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003659558296485133,
      "loss": 4.6886,
      "mean_token_accuracy": 0.2554155960679054,
      "num_tokens": 89146353.0,
      "step": 38890
    },
    {
      "entropy": 5.454482793807983,
      "epoch": 3.7363112391930837,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036592447116075765,
      "loss": 4.7139,
      "mean_token_accuracy": 0.24524301886558533,
      "num_tokens": 89157589.0,
      "step": 38895
    },
    {
      "entropy": 5.451625680923462,
      "epoch": 3.7367915465898176,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036589311056200573,
      "loss": 4.6556,
      "mean_token_accuracy": 0.25837873071432116,
      "num_tokens": 89169384.0,
      "step": 38900
    },
    {
      "entropy": 5.479080533981323,
      "epoch": 3.7372718539865515,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003658617478529857,
      "loss": 4.6973,
      "mean_token_accuracy": 0.2463983878493309,
      "num_tokens": 89180443.0,
      "step": 38905
    },
    {
      "entropy": 5.414032363891602,
      "epoch": 3.7377521613832854,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036583038303442567,
      "loss": 4.6794,
      "mean_token_accuracy": 0.24818435162305832,
      "num_tokens": 89192434.0,
      "step": 38910
    },
    {
      "entropy": 5.495210933685303,
      "epoch": 3.7382324687800192,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036579901610705377,
      "loss": 4.6944,
      "mean_token_accuracy": 0.2519203692674637,
      "num_tokens": 89203965.0,
      "step": 38915
    },
    {
      "entropy": 5.4787672996521,
      "epoch": 3.738712776176753,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036576764707159853,
      "loss": 4.6513,
      "mean_token_accuracy": 0.2519580528140068,
      "num_tokens": 89214754.0,
      "step": 38920
    },
    {
      "entropy": 5.369419384002685,
      "epoch": 3.739193083573487,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003657362759287879,
      "loss": 4.6224,
      "mean_token_accuracy": 0.2599217787384987,
      "num_tokens": 89226831.0,
      "step": 38925
    },
    {
      "entropy": 5.52705569267273,
      "epoch": 3.739673390970221,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036570490267935044,
      "loss": 4.8149,
      "mean_token_accuracy": 0.24979456067085265,
      "num_tokens": 89237526.0,
      "step": 38930
    },
    {
      "entropy": 5.4272246837615965,
      "epoch": 3.7401536983669548,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036567352732401443,
      "loss": 4.5789,
      "mean_token_accuracy": 0.2628600850701332,
      "num_tokens": 89248841.0,
      "step": 38935
    },
    {
      "entropy": 5.437876129150391,
      "epoch": 3.7406340057636887,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036564214986350844,
      "loss": 4.6794,
      "mean_token_accuracy": 0.2509500414133072,
      "num_tokens": 89260208.0,
      "step": 38940
    },
    {
      "entropy": 5.43917031288147,
      "epoch": 3.7411143131604225,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036561077029856077,
      "loss": 4.7078,
      "mean_token_accuracy": 0.2623023480176926,
      "num_tokens": 89270844.0,
      "step": 38945
    },
    {
      "entropy": 5.491954040527344,
      "epoch": 3.7415946205571564,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036557938862990017,
      "loss": 4.7704,
      "mean_token_accuracy": 0.2536311596632004,
      "num_tokens": 89284301.0,
      "step": 38950
    },
    {
      "entropy": 5.476388835906983,
      "epoch": 3.7420749279538903,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036554800485825506,
      "loss": 4.7066,
      "mean_token_accuracy": 0.2536375179886818,
      "num_tokens": 89296753.0,
      "step": 38955
    },
    {
      "entropy": 5.5293371200561525,
      "epoch": 3.7425552353506246,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036551661898435414,
      "loss": 4.7863,
      "mean_token_accuracy": 0.24399580508470536,
      "num_tokens": 89308745.0,
      "step": 38960
    },
    {
      "entropy": 5.481930351257324,
      "epoch": 3.743035542747358,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003654852310089262,
      "loss": 4.7329,
      "mean_token_accuracy": 0.2540664941072464,
      "num_tokens": 89320273.0,
      "step": 38965
    },
    {
      "entropy": 5.588536357879638,
      "epoch": 3.7435158501440924,
      "grad_norm": 1.25,
      "learning_rate": 0.00036545384093269984,
      "loss": 4.844,
      "mean_token_accuracy": 0.23887089490890503,
      "num_tokens": 89331657.0,
      "step": 38970
    },
    {
      "entropy": 5.432617044448852,
      "epoch": 3.7439961575408263,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036542244875640385,
      "loss": 4.6796,
      "mean_token_accuracy": 0.25480982661247253,
      "num_tokens": 89342679.0,
      "step": 38975
    },
    {
      "entropy": 5.435176801681519,
      "epoch": 3.74447646493756,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036539105448076713,
      "loss": 4.6143,
      "mean_token_accuracy": 0.26403372585773466,
      "num_tokens": 89353830.0,
      "step": 38980
    },
    {
      "entropy": 5.382896852493286,
      "epoch": 3.744956772334294,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003653596581065185,
      "loss": 4.6291,
      "mean_token_accuracy": 0.2595733255147934,
      "num_tokens": 89365397.0,
      "step": 38985
    },
    {
      "entropy": 5.436226177215576,
      "epoch": 3.745437079731028,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003653282596343869,
      "loss": 4.6596,
      "mean_token_accuracy": 0.25469553768634795,
      "num_tokens": 89377255.0,
      "step": 38990
    },
    {
      "entropy": 5.3860677719116214,
      "epoch": 3.745917387127762,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003652968590651014,
      "loss": 4.6405,
      "mean_token_accuracy": 0.2630636379122734,
      "num_tokens": 89388603.0,
      "step": 38995
    },
    {
      "entropy": 5.560388088226318,
      "epoch": 3.7463976945244957,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036526545639939097,
      "loss": 4.709,
      "mean_token_accuracy": 0.252469876408577,
      "num_tokens": 89400510.0,
      "step": 39000
    },
    {
      "epoch": 3.7463976945244957,
      "eval_entropy": 5.227918525372175,
      "eval_loss": 4.7917160987854,
      "eval_mean_token_accuracy": 0.25726336557082086,
      "eval_num_tokens": 89400510.0,
      "eval_runtime": 20.4615,
      "eval_samples_per_second": 1603.741,
      "eval_steps_per_second": 200.474,
      "step": 39000
    },
    {
      "entropy": 5.590927696228027,
      "epoch": 3.7468780019212296,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003652340516379846,
      "loss": 4.8306,
      "mean_token_accuracy": 0.23742213249206542,
      "num_tokens": 89411512.0,
      "step": 39005
    },
    {
      "entropy": 5.46753830909729,
      "epoch": 3.7473583093179634,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036520264478161164,
      "loss": 4.6856,
      "mean_token_accuracy": 0.25560342222452165,
      "num_tokens": 89422394.0,
      "step": 39010
    },
    {
      "entropy": 5.428706455230713,
      "epoch": 3.7478386167146973,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003651712358310009,
      "loss": 4.6885,
      "mean_token_accuracy": 0.2495415359735489,
      "num_tokens": 89433420.0,
      "step": 39015
    },
    {
      "entropy": 5.475094556808472,
      "epoch": 3.748318924111431,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036513982478688197,
      "loss": 4.6905,
      "mean_token_accuracy": 0.2570991352200508,
      "num_tokens": 89444722.0,
      "step": 39020
    },
    {
      "entropy": 5.526190185546875,
      "epoch": 3.748799231508165,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036510841164998385,
      "loss": 4.7143,
      "mean_token_accuracy": 0.25435220897197724,
      "num_tokens": 89456601.0,
      "step": 39025
    },
    {
      "entropy": 5.427340269088745,
      "epoch": 3.749279538904899,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003650769964210361,
      "loss": 4.6935,
      "mean_token_accuracy": 0.2493784546852112,
      "num_tokens": 89467753.0,
      "step": 39030
    },
    {
      "entropy": 5.526816749572754,
      "epoch": 3.7497598463016333,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003650455791007679,
      "loss": 4.8028,
      "mean_token_accuracy": 0.2507305070757866,
      "num_tokens": 89479303.0,
      "step": 39035
    },
    {
      "entropy": 5.569991111755371,
      "epoch": 3.7502401536983667,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036501415968990874,
      "loss": 4.752,
      "mean_token_accuracy": 0.2438623920083046,
      "num_tokens": 89491334.0,
      "step": 39040
    },
    {
      "entropy": 5.413542413711548,
      "epoch": 3.750720461095101,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003649827381891881,
      "loss": 4.5987,
      "mean_token_accuracy": 0.2620018571615219,
      "num_tokens": 89502788.0,
      "step": 39045
    },
    {
      "entropy": 5.421383190155029,
      "epoch": 3.751200768491835,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003649513145993355,
      "loss": 4.6776,
      "mean_token_accuracy": 0.2477606400847435,
      "num_tokens": 89514479.0,
      "step": 39050
    },
    {
      "entropy": 5.45076994895935,
      "epoch": 3.751681075888569,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003649198889210804,
      "loss": 4.725,
      "mean_token_accuracy": 0.25653997510671617,
      "num_tokens": 89526534.0,
      "step": 39055
    },
    {
      "entropy": 5.533136606216431,
      "epoch": 3.7521613832853027,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003648884611551525,
      "loss": 4.7259,
      "mean_token_accuracy": 0.2527101367712021,
      "num_tokens": 89537231.0,
      "step": 39060
    },
    {
      "entropy": 5.501824283599854,
      "epoch": 3.7526416906820366,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036485703130228147,
      "loss": 4.7158,
      "mean_token_accuracy": 0.24319298714399337,
      "num_tokens": 89547374.0,
      "step": 39065
    },
    {
      "entropy": 5.392236328125,
      "epoch": 3.7531219980787704,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036482559936319703,
      "loss": 4.6199,
      "mean_token_accuracy": 0.25472029894590376,
      "num_tokens": 89557776.0,
      "step": 39070
    },
    {
      "entropy": 5.427426862716675,
      "epoch": 3.7536023054755043,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036479416533862885,
      "loss": 4.6846,
      "mean_token_accuracy": 0.26100080013275145,
      "num_tokens": 89569181.0,
      "step": 39075
    },
    {
      "entropy": 5.502686977386475,
      "epoch": 3.754082612872238,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036476272922930674,
      "loss": 4.7115,
      "mean_token_accuracy": 0.2466324284672737,
      "num_tokens": 89580375.0,
      "step": 39080
    },
    {
      "entropy": 5.604035949707031,
      "epoch": 3.754562920268972,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036473129103596077,
      "loss": 4.7827,
      "mean_token_accuracy": 0.2451749712228775,
      "num_tokens": 89591787.0,
      "step": 39085
    },
    {
      "entropy": 5.460624837875367,
      "epoch": 3.755043227665706,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003646998507593205,
      "loss": 4.6862,
      "mean_token_accuracy": 0.2498982474207878,
      "num_tokens": 89604389.0,
      "step": 39090
    },
    {
      "entropy": 5.359866285324097,
      "epoch": 3.75552353506244,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003646684084001161,
      "loss": 4.5861,
      "mean_token_accuracy": 0.26685369610786436,
      "num_tokens": 89615900.0,
      "step": 39095
    },
    {
      "entropy": 5.394830513000488,
      "epoch": 3.7560038424591737,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003646369639590776,
      "loss": 4.6351,
      "mean_token_accuracy": 0.2531131595373154,
      "num_tokens": 89627056.0,
      "step": 39100
    },
    {
      "entropy": 5.4478950023651125,
      "epoch": 3.7564841498559076,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003646055174369348,
      "loss": 4.654,
      "mean_token_accuracy": 0.25947661101818087,
      "num_tokens": 89637194.0,
      "step": 39105
    },
    {
      "entropy": 5.501287984848022,
      "epoch": 3.756964457252642,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003645740688344181,
      "loss": 4.6889,
      "mean_token_accuracy": 0.25691403448581696,
      "num_tokens": 89648132.0,
      "step": 39110
    },
    {
      "entropy": 5.441798877716065,
      "epoch": 3.7574447646493754,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003645426181522575,
      "loss": 4.6167,
      "mean_token_accuracy": 0.2561358168721199,
      "num_tokens": 89658236.0,
      "step": 39115
    },
    {
      "entropy": 5.493663024902344,
      "epoch": 3.7579250720461097,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036451116539118304,
      "loss": 4.7136,
      "mean_token_accuracy": 0.2472725808620453,
      "num_tokens": 89669660.0,
      "step": 39120
    },
    {
      "entropy": 5.475358390808106,
      "epoch": 3.7584053794428436,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036447971055192526,
      "loss": 4.6933,
      "mean_token_accuracy": 0.2507383957505226,
      "num_tokens": 89681667.0,
      "step": 39125
    },
    {
      "entropy": 5.3472514152526855,
      "epoch": 3.7588856868395775,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003644482536352142,
      "loss": 4.5672,
      "mean_token_accuracy": 0.26284876763820647,
      "num_tokens": 89694924.0,
      "step": 39130
    },
    {
      "entropy": 5.484437084197998,
      "epoch": 3.7593659942363113,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036441679464178044,
      "loss": 4.7039,
      "mean_token_accuracy": 0.2502909153699875,
      "num_tokens": 89705855.0,
      "step": 39135
    },
    {
      "entropy": 5.460918235778808,
      "epoch": 3.7598463016330452,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003643853335723541,
      "loss": 4.7009,
      "mean_token_accuracy": 0.24722906500101088,
      "num_tokens": 89716766.0,
      "step": 39140
    },
    {
      "entropy": 5.500425529479981,
      "epoch": 3.760326609029779,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036435387042766575,
      "loss": 4.7085,
      "mean_token_accuracy": 0.2523765072226524,
      "num_tokens": 89729595.0,
      "step": 39145
    },
    {
      "entropy": 5.480041933059693,
      "epoch": 3.760806916426513,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003643224052084459,
      "loss": 4.6788,
      "mean_token_accuracy": 0.2544781193137169,
      "num_tokens": 89740972.0,
      "step": 39150
    },
    {
      "entropy": 5.432016134262085,
      "epoch": 3.761287223823247,
      "grad_norm": 1.21875,
      "learning_rate": 0.000364290937915425,
      "loss": 4.6914,
      "mean_token_accuracy": 0.25093468725681306,
      "num_tokens": 89754660.0,
      "step": 39155
    },
    {
      "entropy": 5.555732727050781,
      "epoch": 3.7617675312199808,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036425946854933374,
      "loss": 4.7746,
      "mean_token_accuracy": 0.24797807335853578,
      "num_tokens": 89766298.0,
      "step": 39160
    },
    {
      "entropy": 5.5191480159759525,
      "epoch": 3.7622478386167146,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036422799711090253,
      "loss": 4.7695,
      "mean_token_accuracy": 0.2444112181663513,
      "num_tokens": 89778133.0,
      "step": 39165
    },
    {
      "entropy": 5.4534063816070555,
      "epoch": 3.7627281460134485,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003641965236008623,
      "loss": 4.7201,
      "mean_token_accuracy": 0.2529525771737099,
      "num_tokens": 89789408.0,
      "step": 39170
    },
    {
      "entropy": 5.4718883514404295,
      "epoch": 3.7632084534101824,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003641650480199436,
      "loss": 4.6845,
      "mean_token_accuracy": 0.2540855869650841,
      "num_tokens": 89800287.0,
      "step": 39175
    },
    {
      "entropy": 5.463309717178345,
      "epoch": 3.7636887608069163,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036413357036887723,
      "loss": 4.6835,
      "mean_token_accuracy": 0.24752452969551086,
      "num_tokens": 89811639.0,
      "step": 39180
    },
    {
      "entropy": 5.476664113998413,
      "epoch": 3.7641690682036506,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000364102090648394,
      "loss": 4.7185,
      "mean_token_accuracy": 0.2519086256623268,
      "num_tokens": 89823075.0,
      "step": 39185
    },
    {
      "entropy": 5.411709499359131,
      "epoch": 3.764649375600384,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003640706088592249,
      "loss": 4.6671,
      "mean_token_accuracy": 0.25259134769439695,
      "num_tokens": 89835268.0,
      "step": 39190
    },
    {
      "entropy": 5.5447978496551515,
      "epoch": 3.7651296829971184,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036403912500210066,
      "loss": 4.7416,
      "mean_token_accuracy": 0.25143427699804305,
      "num_tokens": 89846983.0,
      "step": 39195
    },
    {
      "entropy": 5.489446783065796,
      "epoch": 3.765609990393852,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003640076390777524,
      "loss": 4.6655,
      "mean_token_accuracy": 0.25904382914304735,
      "num_tokens": 89857372.0,
      "step": 39200
    },
    {
      "entropy": 5.4394842147827145,
      "epoch": 3.766090297790586,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000363976151086911,
      "loss": 4.691,
      "mean_token_accuracy": 0.2546086773276329,
      "num_tokens": 89869286.0,
      "step": 39205
    },
    {
      "entropy": 5.470434904098511,
      "epoch": 3.76657060518732,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003639446610303076,
      "loss": 4.7269,
      "mean_token_accuracy": 0.24716968834400177,
      "num_tokens": 89879646.0,
      "step": 39210
    },
    {
      "entropy": 5.457228183746338,
      "epoch": 3.767050912584054,
      "grad_norm": 1.125,
      "learning_rate": 0.00036391316890867334,
      "loss": 4.6823,
      "mean_token_accuracy": 0.25454663932323457,
      "num_tokens": 89891870.0,
      "step": 39215
    },
    {
      "entropy": 5.523255205154419,
      "epoch": 3.7675312199807878,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036388167472273925,
      "loss": 4.7046,
      "mean_token_accuracy": 0.2531433880329132,
      "num_tokens": 89903383.0,
      "step": 39220
    },
    {
      "entropy": 5.632803916931152,
      "epoch": 3.7680115273775217,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003638501784732366,
      "loss": 4.9116,
      "mean_token_accuracy": 0.23996198177337646,
      "num_tokens": 89913707.0,
      "step": 39225
    },
    {
      "entropy": 5.447080421447754,
      "epoch": 3.7684918347742555,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003638186801608966,
      "loss": 4.6447,
      "mean_token_accuracy": 0.2607200115919113,
      "num_tokens": 89924983.0,
      "step": 39230
    },
    {
      "entropy": 5.437960004806518,
      "epoch": 3.7689721421709894,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036378717978645067,
      "loss": 4.7175,
      "mean_token_accuracy": 0.25345922708511354,
      "num_tokens": 89936404.0,
      "step": 39235
    },
    {
      "entropy": 5.539598369598389,
      "epoch": 3.7694524495677233,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036375567735063,
      "loss": 4.8063,
      "mean_token_accuracy": 0.24925985038280488,
      "num_tokens": 89947916.0,
      "step": 39240
    },
    {
      "entropy": 5.559914445877075,
      "epoch": 3.769932756964457,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003637241728541661,
      "loss": 4.766,
      "mean_token_accuracy": 0.2542991876602173,
      "num_tokens": 89959436.0,
      "step": 39245
    },
    {
      "entropy": 5.521519327163697,
      "epoch": 3.770413064361191,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036369266629779037,
      "loss": 4.7291,
      "mean_token_accuracy": 0.24881502985954285,
      "num_tokens": 89969977.0,
      "step": 39250
    },
    {
      "entropy": 5.494840478897094,
      "epoch": 3.770893371757925,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003636611576822342,
      "loss": 4.7458,
      "mean_token_accuracy": 0.24381092488765715,
      "num_tokens": 89981189.0,
      "step": 39255
    },
    {
      "entropy": 5.446835613250732,
      "epoch": 3.771373679154659,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003636296470082293,
      "loss": 4.7508,
      "mean_token_accuracy": 0.2481696769595146,
      "num_tokens": 89994120.0,
      "step": 39260
    },
    {
      "entropy": 5.3807940006256105,
      "epoch": 3.7718539865513927,
      "grad_norm": 1.125,
      "learning_rate": 0.00036359813427650726,
      "loss": 4.587,
      "mean_token_accuracy": 0.261432945728302,
      "num_tokens": 90005953.0,
      "step": 39265
    },
    {
      "entropy": 5.458310461044311,
      "epoch": 3.772334293948127,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036356661948779945,
      "loss": 4.6373,
      "mean_token_accuracy": 0.2528861314058304,
      "num_tokens": 90017875.0,
      "step": 39270
    },
    {
      "entropy": 5.52795991897583,
      "epoch": 3.7728146013448605,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036353510264283776,
      "loss": 4.7044,
      "mean_token_accuracy": 0.250826732814312,
      "num_tokens": 90028096.0,
      "step": 39275
    },
    {
      "entropy": 5.538585758209228,
      "epoch": 3.773294908741595,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003635035837423539,
      "loss": 4.7434,
      "mean_token_accuracy": 0.2472092628479004,
      "num_tokens": 90038834.0,
      "step": 39280
    },
    {
      "entropy": 5.515896368026733,
      "epoch": 3.7737752161383287,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003634720627870797,
      "loss": 4.7818,
      "mean_token_accuracy": 0.24688680917024614,
      "num_tokens": 90051429.0,
      "step": 39285
    },
    {
      "entropy": 5.487292766571045,
      "epoch": 3.7742555235350626,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036344053977774684,
      "loss": 4.6806,
      "mean_token_accuracy": 0.2539188593626022,
      "num_tokens": 90062269.0,
      "step": 39290
    },
    {
      "entropy": 5.426702499389648,
      "epoch": 3.7747358309317964,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003634090147150873,
      "loss": 4.6693,
      "mean_token_accuracy": 0.2582255005836487,
      "num_tokens": 90073573.0,
      "step": 39295
    },
    {
      "entropy": 5.416153478622436,
      "epoch": 3.7752161383285303,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003633774875998329,
      "loss": 4.6052,
      "mean_token_accuracy": 0.2657013475894928,
      "num_tokens": 90083543.0,
      "step": 39300
    },
    {
      "entropy": 5.48156099319458,
      "epoch": 3.775696445725264,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003633459584327157,
      "loss": 4.7469,
      "mean_token_accuracy": 0.2472907930612564,
      "num_tokens": 90095108.0,
      "step": 39305
    },
    {
      "entropy": 5.4967796325683596,
      "epoch": 3.776176753121998,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036331442721446763,
      "loss": 4.8221,
      "mean_token_accuracy": 0.24715470373630524,
      "num_tokens": 90107272.0,
      "step": 39310
    },
    {
      "entropy": 5.547845363616943,
      "epoch": 3.776657060518732,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003632828939458209,
      "loss": 4.7714,
      "mean_token_accuracy": 0.24456723630428315,
      "num_tokens": 90118167.0,
      "step": 39315
    },
    {
      "entropy": 5.528298711776733,
      "epoch": 3.777137367915466,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036325135862750745,
      "loss": 4.7155,
      "mean_token_accuracy": 0.2464755654335022,
      "num_tokens": 90129790.0,
      "step": 39320
    },
    {
      "entropy": 5.494391584396363,
      "epoch": 3.7776176753121997,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036321982126025947,
      "loss": 4.7181,
      "mean_token_accuracy": 0.2522956684231758,
      "num_tokens": 90141367.0,
      "step": 39325
    },
    {
      "entropy": 5.516352939605713,
      "epoch": 3.7780979827089336,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003631882818448093,
      "loss": 4.6794,
      "mean_token_accuracy": 0.2552202925086021,
      "num_tokens": 90151755.0,
      "step": 39330
    },
    {
      "entropy": 5.479482936859131,
      "epoch": 3.7785782901056675,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036315674038188905,
      "loss": 4.6952,
      "mean_token_accuracy": 0.25240288078784945,
      "num_tokens": 90163262.0,
      "step": 39335
    },
    {
      "entropy": 5.6302406787872314,
      "epoch": 3.7790585975024014,
      "grad_norm": 1.234375,
      "learning_rate": 0.000363125196872231,
      "loss": 4.8706,
      "mean_token_accuracy": 0.2370237171649933,
      "num_tokens": 90174704.0,
      "step": 39340
    },
    {
      "entropy": 5.447763872146607,
      "epoch": 3.7795389048991357,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003630936513165676,
      "loss": 4.6213,
      "mean_token_accuracy": 0.25585859566926955,
      "num_tokens": 90186885.0,
      "step": 39345
    },
    {
      "entropy": 5.473380851745605,
      "epoch": 3.780019212295869,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003630621037156312,
      "loss": 4.6828,
      "mean_token_accuracy": 0.25623034983873366,
      "num_tokens": 90198213.0,
      "step": 39350
    },
    {
      "entropy": 5.517498874664307,
      "epoch": 3.7804995196926034,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036303055407015437,
      "loss": 4.7558,
      "mean_token_accuracy": 0.24842083007097243,
      "num_tokens": 90209623.0,
      "step": 39355
    },
    {
      "entropy": 5.504548025131226,
      "epoch": 3.7809798270893373,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036299900238086924,
      "loss": 4.7027,
      "mean_token_accuracy": 0.2521663561463356,
      "num_tokens": 90221808.0,
      "step": 39360
    },
    {
      "entropy": 5.507753705978393,
      "epoch": 3.781460134486071,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003629674486485087,
      "loss": 4.7252,
      "mean_token_accuracy": 0.24658691585063935,
      "num_tokens": 90232894.0,
      "step": 39365
    },
    {
      "entropy": 5.395488834381103,
      "epoch": 3.781940441882805,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003629358928738052,
      "loss": 4.6255,
      "mean_token_accuracy": 0.2599883571267128,
      "num_tokens": 90243240.0,
      "step": 39370
    },
    {
      "entropy": 5.409471130371093,
      "epoch": 3.782420749279539,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036290433505749135,
      "loss": 4.6336,
      "mean_token_accuracy": 0.2577241688966751,
      "num_tokens": 90255082.0,
      "step": 39375
    },
    {
      "entropy": 5.55864782333374,
      "epoch": 3.782901056676273,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003628727752002999,
      "loss": 4.7689,
      "mean_token_accuracy": 0.24711925983428956,
      "num_tokens": 90266521.0,
      "step": 39380
    },
    {
      "entropy": 5.472974872589111,
      "epoch": 3.7833813640730067,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003628412133029636,
      "loss": 4.6297,
      "mean_token_accuracy": 0.2581799104809761,
      "num_tokens": 90277145.0,
      "step": 39385
    },
    {
      "entropy": 5.399630165100097,
      "epoch": 3.7838616714697406,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003628096493662151,
      "loss": 4.6033,
      "mean_token_accuracy": 0.25865921229124067,
      "num_tokens": 90289290.0,
      "step": 39390
    },
    {
      "entropy": 5.499638843536377,
      "epoch": 3.7843419788664745,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003627780833907873,
      "loss": 4.7013,
      "mean_token_accuracy": 0.249581478536129,
      "num_tokens": 90299863.0,
      "step": 39395
    },
    {
      "entropy": 5.455315113067627,
      "epoch": 3.7848222862632084,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036274651537741305,
      "loss": 4.7291,
      "mean_token_accuracy": 0.25199161767959594,
      "num_tokens": 90310905.0,
      "step": 39400
    },
    {
      "entropy": 5.504467725753784,
      "epoch": 3.7853025936599423,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003627149453268252,
      "loss": 4.7915,
      "mean_token_accuracy": 0.24920207262039185,
      "num_tokens": 90322971.0,
      "step": 39405
    },
    {
      "entropy": 5.533896636962891,
      "epoch": 3.785782901056676,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036268337323975685,
      "loss": 4.6987,
      "mean_token_accuracy": 0.25060611963272095,
      "num_tokens": 90333782.0,
      "step": 39410
    },
    {
      "entropy": 5.5907268047332765,
      "epoch": 3.78626320845341,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00036265179911694094,
      "loss": 4.818,
      "mean_token_accuracy": 0.23819293975830078,
      "num_tokens": 90343695.0,
      "step": 39415
    },
    {
      "entropy": 5.440638780593872,
      "epoch": 3.7867435158501443,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003626202229591105,
      "loss": 4.6748,
      "mean_token_accuracy": 0.2541544422507286,
      "num_tokens": 90354932.0,
      "step": 39420
    },
    {
      "entropy": 5.490943098068238,
      "epoch": 3.787223823246878,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036258864476699864,
      "loss": 4.7014,
      "mean_token_accuracy": 0.2554903268814087,
      "num_tokens": 90367449.0,
      "step": 39425
    },
    {
      "entropy": 5.417992496490479,
      "epoch": 3.787704130643612,
      "grad_norm": 1.25,
      "learning_rate": 0.0003625570645413386,
      "loss": 4.6029,
      "mean_token_accuracy": 0.26528569757938386,
      "num_tokens": 90378915.0,
      "step": 39430
    },
    {
      "entropy": 5.451073884963989,
      "epoch": 3.7881844380403455,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003625254822828635,
      "loss": 4.6702,
      "mean_token_accuracy": 0.25344584733247755,
      "num_tokens": 90391868.0,
      "step": 39435
    },
    {
      "entropy": 5.49691686630249,
      "epoch": 3.78866474543708,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003624938979923065,
      "loss": 4.7388,
      "mean_token_accuracy": 0.24893763959407805,
      "num_tokens": 90402520.0,
      "step": 39440
    },
    {
      "entropy": 5.61725754737854,
      "epoch": 3.7891450528338138,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036246231167040105,
      "loss": 4.8023,
      "mean_token_accuracy": 0.2452710136771202,
      "num_tokens": 90413425.0,
      "step": 39445
    },
    {
      "entropy": 5.59978756904602,
      "epoch": 3.7896253602305476,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003624307233178805,
      "loss": 4.8448,
      "mean_token_accuracy": 0.2421219229698181,
      "num_tokens": 90424976.0,
      "step": 39450
    },
    {
      "entropy": 5.43693323135376,
      "epoch": 3.7901056676272815,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036239913293547807,
      "loss": 4.6398,
      "mean_token_accuracy": 0.2576769918203354,
      "num_tokens": 90436177.0,
      "step": 39455
    },
    {
      "entropy": 5.4916712760925295,
      "epoch": 3.7905859750240154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036236754052392737,
      "loss": 4.7512,
      "mean_token_accuracy": 0.25435228943824767,
      "num_tokens": 90448066.0,
      "step": 39460
    },
    {
      "entropy": 5.469911718368531,
      "epoch": 3.7910662824207493,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003623359460839618,
      "loss": 4.6697,
      "mean_token_accuracy": 0.2551996484398842,
      "num_tokens": 90457913.0,
      "step": 39465
    },
    {
      "entropy": 5.3914247989654545,
      "epoch": 3.791546589817483,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003623043496163149,
      "loss": 4.6467,
      "mean_token_accuracy": 0.26029030829668043,
      "num_tokens": 90469637.0,
      "step": 39470
    },
    {
      "entropy": 5.44749493598938,
      "epoch": 3.792026897214217,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003622727511217202,
      "loss": 4.719,
      "mean_token_accuracy": 0.24962675869464873,
      "num_tokens": 90481222.0,
      "step": 39475
    },
    {
      "entropy": 5.486592245101929,
      "epoch": 3.792507204610951,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003622411506009113,
      "loss": 4.6516,
      "mean_token_accuracy": 0.2535059526562691,
      "num_tokens": 90492911.0,
      "step": 39480
    },
    {
      "entropy": 5.463081121444702,
      "epoch": 3.792987512007685,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003622095480546221,
      "loss": 4.723,
      "mean_token_accuracy": 0.2509325578808784,
      "num_tokens": 90504583.0,
      "step": 39485
    },
    {
      "entropy": 5.464928770065308,
      "epoch": 3.7934678194044187,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036217794348358606,
      "loss": 4.7313,
      "mean_token_accuracy": 0.2499052032828331,
      "num_tokens": 90516006.0,
      "step": 39490
    },
    {
      "entropy": 5.389527225494385,
      "epoch": 3.793948126801153,
      "grad_norm": 1.3125,
      "learning_rate": 0.000362146336888537,
      "loss": 4.5935,
      "mean_token_accuracy": 0.2583906948566437,
      "num_tokens": 90527362.0,
      "step": 39495
    },
    {
      "entropy": 5.4370039939880375,
      "epoch": 3.7944284341978864,
      "grad_norm": 1.25,
      "learning_rate": 0.00036211472827020886,
      "loss": 4.6658,
      "mean_token_accuracy": 0.2573280781507492,
      "num_tokens": 90538635.0,
      "step": 39500
    },
    {
      "entropy": 5.486967658996582,
      "epoch": 3.7949087415946208,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003620831176293354,
      "loss": 4.6755,
      "mean_token_accuracy": 0.2564308598637581,
      "num_tokens": 90549996.0,
      "step": 39505
    },
    {
      "entropy": 5.529063892364502,
      "epoch": 3.795389048991354,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036205150496665053,
      "loss": 4.7819,
      "mean_token_accuracy": 0.2437766522169113,
      "num_tokens": 90563360.0,
      "step": 39510
    },
    {
      "entropy": 5.532676839828492,
      "epoch": 3.7958693563880885,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036201989028288815,
      "loss": 4.7876,
      "mean_token_accuracy": 0.24393652379512787,
      "num_tokens": 90574617.0,
      "step": 39515
    },
    {
      "entropy": 5.524231863021851,
      "epoch": 3.7963496637848224,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036198827357878234,
      "loss": 4.7738,
      "mean_token_accuracy": 0.24272819608449936,
      "num_tokens": 90586209.0,
      "step": 39520
    },
    {
      "entropy": 5.580117416381836,
      "epoch": 3.7968299711815563,
      "grad_norm": 1.125,
      "learning_rate": 0.0003619566548550671,
      "loss": 4.7953,
      "mean_token_accuracy": 0.2431970328092575,
      "num_tokens": 90598409.0,
      "step": 39525
    },
    {
      "entropy": 5.517055463790894,
      "epoch": 3.79731027857829,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003619250341124765,
      "loss": 4.6998,
      "mean_token_accuracy": 0.25577917546033857,
      "num_tokens": 90610073.0,
      "step": 39530
    },
    {
      "entropy": 5.460142183303833,
      "epoch": 3.797790585975024,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003618934113517448,
      "loss": 4.7513,
      "mean_token_accuracy": 0.2534287080168724,
      "num_tokens": 90621782.0,
      "step": 39535
    },
    {
      "entropy": 5.4465858936309814,
      "epoch": 3.798270893371758,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003618617865736061,
      "loss": 4.7112,
      "mean_token_accuracy": 0.2472532108426094,
      "num_tokens": 90633771.0,
      "step": 39540
    },
    {
      "entropy": 5.547481203079224,
      "epoch": 3.798751200768492,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036183015977879463,
      "loss": 4.82,
      "mean_token_accuracy": 0.24109215289354324,
      "num_tokens": 90644965.0,
      "step": 39545
    },
    {
      "entropy": 5.454978322982788,
      "epoch": 3.7992315081652257,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036179853096804474,
      "loss": 4.6737,
      "mean_token_accuracy": 0.25241821855306623,
      "num_tokens": 90656691.0,
      "step": 39550
    },
    {
      "entropy": 5.531814193725586,
      "epoch": 3.7997118155619596,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036176690014209055,
      "loss": 4.7385,
      "mean_token_accuracy": 0.24784428328275682,
      "num_tokens": 90667760.0,
      "step": 39555
    },
    {
      "entropy": 5.463553857803345,
      "epoch": 3.8001921229586935,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003617352673016667,
      "loss": 4.7111,
      "mean_token_accuracy": 0.25163390636444094,
      "num_tokens": 90679640.0,
      "step": 39560
    },
    {
      "entropy": 5.440100955963135,
      "epoch": 3.8006724303554273,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036170363244750756,
      "loss": 4.632,
      "mean_token_accuracy": 0.26551122814416883,
      "num_tokens": 90693187.0,
      "step": 39565
    },
    {
      "entropy": 5.549955701828003,
      "epoch": 3.8011527377521612,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003616719955803474,
      "loss": 4.8332,
      "mean_token_accuracy": 0.24352775812149047,
      "num_tokens": 90704540.0,
      "step": 39570
    },
    {
      "entropy": 5.487769889831543,
      "epoch": 3.801633045148895,
      "grad_norm": 1.25,
      "learning_rate": 0.0003616403567009209,
      "loss": 4.7016,
      "mean_token_accuracy": 0.26105557531118395,
      "num_tokens": 90716577.0,
      "step": 39575
    },
    {
      "entropy": 5.41668291091919,
      "epoch": 3.8021133525456294,
      "grad_norm": 1.125,
      "learning_rate": 0.0003616087158099626,
      "loss": 4.5687,
      "mean_token_accuracy": 0.2671796604990959,
      "num_tokens": 90727134.0,
      "step": 39580
    },
    {
      "entropy": 5.470618200302124,
      "epoch": 3.802593659942363,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003615770729082071,
      "loss": 4.697,
      "mean_token_accuracy": 0.25269024521112443,
      "num_tokens": 90738844.0,
      "step": 39585
    },
    {
      "entropy": 5.39982795715332,
      "epoch": 3.803073967339097,
      "grad_norm": 1.21875,
      "learning_rate": 0.000361545427996389,
      "loss": 4.6583,
      "mean_token_accuracy": 0.25608165711164477,
      "num_tokens": 90749882.0,
      "step": 39590
    },
    {
      "entropy": 5.473925924301147,
      "epoch": 3.803554274735831,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003615137810752431,
      "loss": 4.7179,
      "mean_token_accuracy": 0.25350992679595946,
      "num_tokens": 90762244.0,
      "step": 39595
    },
    {
      "entropy": 5.587137031555176,
      "epoch": 3.804034582132565,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036148213214550406,
      "loss": 4.7901,
      "mean_token_accuracy": 0.2528738588094711,
      "num_tokens": 90774452.0,
      "step": 39600
    },
    {
      "entropy": 5.506032180786133,
      "epoch": 3.804514889529299,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003614504812079068,
      "loss": 4.7026,
      "mean_token_accuracy": 0.2552053689956665,
      "num_tokens": 90785506.0,
      "step": 39605
    },
    {
      "entropy": 5.428396987915039,
      "epoch": 3.8049951969260327,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036141882826318597,
      "loss": 4.6039,
      "mean_token_accuracy": 0.26205611526966094,
      "num_tokens": 90796606.0,
      "step": 39610
    },
    {
      "entropy": 5.488014888763428,
      "epoch": 3.8054755043227666,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003613871733120766,
      "loss": 4.745,
      "mean_token_accuracy": 0.2517171695828438,
      "num_tokens": 90807949.0,
      "step": 39615
    },
    {
      "entropy": 5.489482831954956,
      "epoch": 3.8059558117195005,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036135551635531354,
      "loss": 4.7829,
      "mean_token_accuracy": 0.24808284193277358,
      "num_tokens": 90819206.0,
      "step": 39620
    },
    {
      "entropy": 5.615074682235718,
      "epoch": 3.8064361191162344,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003613238573936319,
      "loss": 4.8565,
      "mean_token_accuracy": 0.2404411941766739,
      "num_tokens": 90830335.0,
      "step": 39625
    },
    {
      "entropy": 5.483059978485107,
      "epoch": 3.8069164265129682,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003612921964277665,
      "loss": 4.6672,
      "mean_token_accuracy": 0.25816395729780195,
      "num_tokens": 90841236.0,
      "step": 39630
    },
    {
      "entropy": 5.43152928352356,
      "epoch": 3.807396733909702,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003612605334584526,
      "loss": 4.7177,
      "mean_token_accuracy": 0.24765414744615555,
      "num_tokens": 90852772.0,
      "step": 39635
    },
    {
      "entropy": 5.41912612915039,
      "epoch": 3.807877041306436,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036122886848642536,
      "loss": 4.6011,
      "mean_token_accuracy": 0.2584525108337402,
      "num_tokens": 90864692.0,
      "step": 39640
    },
    {
      "entropy": 5.51203556060791,
      "epoch": 3.80835734870317,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003611972015124197,
      "loss": 4.7476,
      "mean_token_accuracy": 0.2524311810731888,
      "num_tokens": 90875398.0,
      "step": 39645
    },
    {
      "entropy": 5.420019626617432,
      "epoch": 3.8088376560999038,
      "grad_norm": 1.140625,
      "learning_rate": 0.000361165532537171,
      "loss": 4.5873,
      "mean_token_accuracy": 0.2597927048802376,
      "num_tokens": 90886997.0,
      "step": 39650
    },
    {
      "entropy": 5.549698543548584,
      "epoch": 3.809317963496638,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003611338615614145,
      "loss": 4.7579,
      "mean_token_accuracy": 0.2470494419336319,
      "num_tokens": 90898674.0,
      "step": 39655
    },
    {
      "entropy": 5.518749809265136,
      "epoch": 3.8097982708933715,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036110218858588554,
      "loss": 4.7543,
      "mean_token_accuracy": 0.24508826285600663,
      "num_tokens": 90910165.0,
      "step": 39660
    },
    {
      "entropy": 5.474596357345581,
      "epoch": 3.810278578290106,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036107051361131934,
      "loss": 4.7291,
      "mean_token_accuracy": 0.2520622193813324,
      "num_tokens": 90922380.0,
      "step": 39665
    },
    {
      "entropy": 5.5078026294708256,
      "epoch": 3.8107588856868397,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036103883663845155,
      "loss": 4.6842,
      "mean_token_accuracy": 0.25365769416093825,
      "num_tokens": 90932913.0,
      "step": 39670
    },
    {
      "entropy": 5.4667699337005615,
      "epoch": 3.8112391930835736,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003610071576680173,
      "loss": 4.7194,
      "mean_token_accuracy": 0.24996305555105208,
      "num_tokens": 90945247.0,
      "step": 39675
    },
    {
      "entropy": 5.549861431121826,
      "epoch": 3.8117195004803075,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036097547670075226,
      "loss": 4.7871,
      "mean_token_accuracy": 0.24392125457525254,
      "num_tokens": 90956064.0,
      "step": 39680
    },
    {
      "entropy": 5.545874881744385,
      "epoch": 3.8121998078770414,
      "grad_norm": 1.34375,
      "learning_rate": 0.000360943793737392,
      "loss": 4.8684,
      "mean_token_accuracy": 0.23507051765918732,
      "num_tokens": 90967636.0,
      "step": 39685
    },
    {
      "entropy": 5.516132593154907,
      "epoch": 3.8126801152737753,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000360912108778672,
      "loss": 4.7305,
      "mean_token_accuracy": 0.2545801028609276,
      "num_tokens": 90979128.0,
      "step": 39690
    },
    {
      "entropy": 5.472958278656006,
      "epoch": 3.813160422670509,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000360880421825328,
      "loss": 4.72,
      "mean_token_accuracy": 0.24912549406290055,
      "num_tokens": 90990743.0,
      "step": 39695
    },
    {
      "entropy": 5.532824611663818,
      "epoch": 3.813640730067243,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003608487328780956,
      "loss": 4.7052,
      "mean_token_accuracy": 0.24870754033327103,
      "num_tokens": 91001160.0,
      "step": 39700
    },
    {
      "entropy": 5.568990325927734,
      "epoch": 3.814121037463977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036081704193771056,
      "loss": 4.7271,
      "mean_token_accuracy": 0.257616662979126,
      "num_tokens": 91012125.0,
      "step": 39705
    },
    {
      "entropy": 5.486910581588745,
      "epoch": 3.814601344860711,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003607853490049085,
      "loss": 4.7184,
      "mean_token_accuracy": 0.25573179572820665,
      "num_tokens": 91023557.0,
      "step": 39710
    },
    {
      "entropy": 5.480859899520874,
      "epoch": 3.8150816522574447,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003607536540804255,
      "loss": 4.6973,
      "mean_token_accuracy": 0.2489120304584503,
      "num_tokens": 91035274.0,
      "step": 39715
    },
    {
      "entropy": 5.5695099353790285,
      "epoch": 3.8155619596541785,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036072195716499726,
      "loss": 4.7086,
      "mean_token_accuracy": 0.25520243048667907,
      "num_tokens": 91047516.0,
      "step": 39720
    },
    {
      "entropy": 5.364712047576904,
      "epoch": 3.8160422670509124,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036069025825935974,
      "loss": 4.5742,
      "mean_token_accuracy": 0.2577007159590721,
      "num_tokens": 91058529.0,
      "step": 39725
    },
    {
      "entropy": 5.486225605010986,
      "epoch": 3.8165225744476468,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036065855736424886,
      "loss": 4.6912,
      "mean_token_accuracy": 0.256699401140213,
      "num_tokens": 91071277.0,
      "step": 39730
    },
    {
      "entropy": 5.419186019897461,
      "epoch": 3.81700288184438,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003606268544804006,
      "loss": 4.6441,
      "mean_token_accuracy": 0.2610947862267494,
      "num_tokens": 91083157.0,
      "step": 39735
    },
    {
      "entropy": 5.534114074707031,
      "epoch": 3.8174831892411145,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036059514960855104,
      "loss": 4.7788,
      "mean_token_accuracy": 0.24647645503282548,
      "num_tokens": 91095945.0,
      "step": 39740
    },
    {
      "entropy": 5.544696426391601,
      "epoch": 3.817963496637848,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003605634427494363,
      "loss": 4.7344,
      "mean_token_accuracy": 0.24719117134809493,
      "num_tokens": 91107210.0,
      "step": 39745
    },
    {
      "entropy": 5.497523260116577,
      "epoch": 3.8184438040345823,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003605317339037925,
      "loss": 4.7233,
      "mean_token_accuracy": 0.24563139528036118,
      "num_tokens": 91120000.0,
      "step": 39750
    },
    {
      "entropy": 5.452379322052002,
      "epoch": 3.818924111431316,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036050002307235575,
      "loss": 4.753,
      "mean_token_accuracy": 0.25193591713905333,
      "num_tokens": 91131794.0,
      "step": 39755
    },
    {
      "entropy": 5.5389011859893795,
      "epoch": 3.81940441882805,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003604683102558624,
      "loss": 4.7807,
      "mean_token_accuracy": 0.24960489124059676,
      "num_tokens": 91143644.0,
      "step": 39760
    },
    {
      "entropy": 5.528754615783692,
      "epoch": 3.819884726224784,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036043659545504866,
      "loss": 4.6793,
      "mean_token_accuracy": 0.25727511793375013,
      "num_tokens": 91153735.0,
      "step": 39765
    },
    {
      "entropy": 5.45685453414917,
      "epoch": 3.820365033621518,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003604048786706509,
      "loss": 4.6655,
      "mean_token_accuracy": 0.25718460530042647,
      "num_tokens": 91165436.0,
      "step": 39770
    },
    {
      "entropy": 5.47251672744751,
      "epoch": 3.8208453410182517,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003603731599034054,
      "loss": 4.7391,
      "mean_token_accuracy": 0.25179578065872193,
      "num_tokens": 91177203.0,
      "step": 39775
    },
    {
      "entropy": 5.4349555492401125,
      "epoch": 3.8213256484149856,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036034143915404865,
      "loss": 4.6706,
      "mean_token_accuracy": 0.25842871367931364,
      "num_tokens": 91190486.0,
      "step": 39780
    },
    {
      "entropy": 5.534273862838745,
      "epoch": 3.8218059558117194,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003603097164233171,
      "loss": 4.7385,
      "mean_token_accuracy": 0.24686895608901976,
      "num_tokens": 91201859.0,
      "step": 39785
    },
    {
      "entropy": 5.566938591003418,
      "epoch": 3.8222862632084533,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003602779917119473,
      "loss": 4.7938,
      "mean_token_accuracy": 0.24299954622983932,
      "num_tokens": 91212743.0,
      "step": 39790
    },
    {
      "entropy": 5.5255646228790285,
      "epoch": 3.822766570605187,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036024626502067565,
      "loss": 4.7266,
      "mean_token_accuracy": 0.24696884751319886,
      "num_tokens": 91223890.0,
      "step": 39795
    },
    {
      "entropy": 5.427174186706543,
      "epoch": 3.823246878001921,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036021453635023894,
      "loss": 4.6153,
      "mean_token_accuracy": 0.26346398442983626,
      "num_tokens": 91235821.0,
      "step": 39800
    },
    {
      "entropy": 5.460089588165284,
      "epoch": 3.8237271853986554,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036018280570137374,
      "loss": 4.7332,
      "mean_token_accuracy": 0.25181041955947875,
      "num_tokens": 91246951.0,
      "step": 39805
    },
    {
      "entropy": 5.460354328155518,
      "epoch": 3.824207492795389,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003601510730748167,
      "loss": 4.6394,
      "mean_token_accuracy": 0.259240859746933,
      "num_tokens": 91257093.0,
      "step": 39810
    },
    {
      "entropy": 5.540520477294922,
      "epoch": 3.824687800192123,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003601193384713046,
      "loss": 4.7816,
      "mean_token_accuracy": 0.25157881528139114,
      "num_tokens": 91269390.0,
      "step": 39815
    },
    {
      "entropy": 5.4836962699890135,
      "epoch": 3.8251681075888566,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003600876018915742,
      "loss": 4.6797,
      "mean_token_accuracy": 0.24652302712202073,
      "num_tokens": 91280427.0,
      "step": 39820
    },
    {
      "entropy": 5.543156337738037,
      "epoch": 3.825648414985591,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003600558633363624,
      "loss": 4.776,
      "mean_token_accuracy": 0.24543946832418442,
      "num_tokens": 91292415.0,
      "step": 39825
    },
    {
      "entropy": 5.461116456985474,
      "epoch": 3.826128722382325,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003600241228064059,
      "loss": 4.6821,
      "mean_token_accuracy": 0.25770277827978133,
      "num_tokens": 91302750.0,
      "step": 39830
    },
    {
      "entropy": 5.469386434555053,
      "epoch": 3.8266090297790587,
      "grad_norm": 1.25,
      "learning_rate": 0.00035999238030244183,
      "loss": 4.6365,
      "mean_token_accuracy": 0.25567193925380705,
      "num_tokens": 91313889.0,
      "step": 39835
    },
    {
      "entropy": 5.536604595184326,
      "epoch": 3.8270893371757926,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035996063582520704,
      "loss": 4.7453,
      "mean_token_accuracy": 0.24149753898382187,
      "num_tokens": 91324306.0,
      "step": 39840
    },
    {
      "entropy": 5.503219127655029,
      "epoch": 3.8275696445725265,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035992888937543856,
      "loss": 4.7793,
      "mean_token_accuracy": 0.2490900471806526,
      "num_tokens": 91335317.0,
      "step": 39845
    },
    {
      "entropy": 5.543606281280518,
      "epoch": 3.8280499519692603,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035989714095387344,
      "loss": 4.772,
      "mean_token_accuracy": 0.2489593133330345,
      "num_tokens": 91346582.0,
      "step": 39850
    },
    {
      "entropy": 5.438648223876953,
      "epoch": 3.8285302593659942,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035986539056124877,
      "loss": 4.6429,
      "mean_token_accuracy": 0.25796595960855484,
      "num_tokens": 91359135.0,
      "step": 39855
    },
    {
      "entropy": 5.577924489974976,
      "epoch": 3.829010566762728,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003598336381983018,
      "loss": 4.8016,
      "mean_token_accuracy": 0.25004742592573165,
      "num_tokens": 91370309.0,
      "step": 39860
    },
    {
      "entropy": 5.472116327285766,
      "epoch": 3.829490874159462,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035980188386576967,
      "loss": 4.6766,
      "mean_token_accuracy": 0.25063818842172625,
      "num_tokens": 91381322.0,
      "step": 39865
    },
    {
      "entropy": 5.498747062683106,
      "epoch": 3.829971181556196,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003597701275643896,
      "loss": 4.7321,
      "mean_token_accuracy": 0.2544034108519554,
      "num_tokens": 91393466.0,
      "step": 39870
    },
    {
      "entropy": 5.506223964691162,
      "epoch": 3.8304514889529298,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003597383692948988,
      "loss": 4.665,
      "mean_token_accuracy": 0.25725397914648057,
      "num_tokens": 91403530.0,
      "step": 39875
    },
    {
      "entropy": 5.485867214202881,
      "epoch": 3.8309317963496636,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003597066090580347,
      "loss": 4.6688,
      "mean_token_accuracy": 0.25245262384414674,
      "num_tokens": 91414563.0,
      "step": 39880
    },
    {
      "entropy": 5.51817831993103,
      "epoch": 3.8314121037463975,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003596748468545348,
      "loss": 4.7883,
      "mean_token_accuracy": 0.2450470209121704,
      "num_tokens": 91427125.0,
      "step": 39885
    },
    {
      "entropy": 5.513775205612182,
      "epoch": 3.831892411143132,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035964308268513623,
      "loss": 4.7301,
      "mean_token_accuracy": 0.25456334054470064,
      "num_tokens": 91438987.0,
      "step": 39890
    },
    {
      "entropy": 5.447529125213623,
      "epoch": 3.8323727185398653,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035961131655057674,
      "loss": 4.6081,
      "mean_token_accuracy": 0.2640045925974846,
      "num_tokens": 91450986.0,
      "step": 39895
    },
    {
      "entropy": 5.457462167739868,
      "epoch": 3.8328530259365996,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003595795484515936,
      "loss": 4.6968,
      "mean_token_accuracy": 0.25541130304336546,
      "num_tokens": 91462590.0,
      "step": 39900
    },
    {
      "entropy": 5.545726871490478,
      "epoch": 3.8333333333333335,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035954777838892454,
      "loss": 4.7505,
      "mean_token_accuracy": 0.24720518887043,
      "num_tokens": 91472630.0,
      "step": 39905
    },
    {
      "entropy": 5.520501279830933,
      "epoch": 3.8338136407300674,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003595160063633072,
      "loss": 4.7615,
      "mean_token_accuracy": 0.25371151715517043,
      "num_tokens": 91483335.0,
      "step": 39910
    },
    {
      "entropy": 5.522676038742065,
      "epoch": 3.8342939481268012,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003594842323754791,
      "loss": 4.808,
      "mean_token_accuracy": 0.24989186376333236,
      "num_tokens": 91494395.0,
      "step": 39915
    },
    {
      "entropy": 5.46566858291626,
      "epoch": 3.834774255523535,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035945245642617795,
      "loss": 4.7032,
      "mean_token_accuracy": 0.2586660489439964,
      "num_tokens": 91506412.0,
      "step": 39920
    },
    {
      "entropy": 5.492944145202637,
      "epoch": 3.835254562920269,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035942067851614165,
      "loss": 4.6911,
      "mean_token_accuracy": 0.26302239000797273,
      "num_tokens": 91517624.0,
      "step": 39925
    },
    {
      "entropy": 5.412717485427857,
      "epoch": 3.835734870317003,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035938889864610776,
      "loss": 4.6879,
      "mean_token_accuracy": 0.26049199253320693,
      "num_tokens": 91529497.0,
      "step": 39930
    },
    {
      "entropy": 5.596102237701416,
      "epoch": 3.8362151777137368,
      "grad_norm": 1.25,
      "learning_rate": 0.00035935711681681423,
      "loss": 4.7761,
      "mean_token_accuracy": 0.24630105644464492,
      "num_tokens": 91540454.0,
      "step": 39935
    },
    {
      "entropy": 5.472919559478759,
      "epoch": 3.8366954851104706,
      "grad_norm": 1.21875,
      "learning_rate": 0.000359325333028999,
      "loss": 4.6704,
      "mean_token_accuracy": 0.25813925862312315,
      "num_tokens": 91551153.0,
      "step": 39940
    },
    {
      "entropy": 5.482219266891479,
      "epoch": 3.8371757925072045,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035929354728339986,
      "loss": 4.7302,
      "mean_token_accuracy": 0.2546871855854988,
      "num_tokens": 91562108.0,
      "step": 39945
    },
    {
      "entropy": 5.4527058601379395,
      "epoch": 3.8376560999039384,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003592617595807549,
      "loss": 4.6181,
      "mean_token_accuracy": 0.2563063889741898,
      "num_tokens": 91572803.0,
      "step": 39950
    },
    {
      "entropy": 5.5192121982574465,
      "epoch": 3.8381364073006723,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035922996992180213,
      "loss": 4.7497,
      "mean_token_accuracy": 0.2507021114230156,
      "num_tokens": 91584292.0,
      "step": 39955
    },
    {
      "entropy": 5.5313334465026855,
      "epoch": 3.838616714697406,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035919817830727945,
      "loss": 4.7628,
      "mean_token_accuracy": 0.25070351660251616,
      "num_tokens": 91596396.0,
      "step": 39960
    },
    {
      "entropy": 5.54119758605957,
      "epoch": 3.8390970220941405,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003591663847379252,
      "loss": 4.742,
      "mean_token_accuracy": 0.25487800538539884,
      "num_tokens": 91608278.0,
      "step": 39965
    },
    {
      "entropy": 5.437678575515747,
      "epoch": 3.839577329490874,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003591345892144774,
      "loss": 4.6508,
      "mean_token_accuracy": 0.2686997473239899,
      "num_tokens": 91619579.0,
      "step": 39970
    },
    {
      "entropy": 5.44267315864563,
      "epoch": 3.8400576368876083,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035910279173767426,
      "loss": 4.718,
      "mean_token_accuracy": 0.26055595874786375,
      "num_tokens": 91631586.0,
      "step": 39975
    },
    {
      "entropy": 5.427404546737671,
      "epoch": 3.840537944284342,
      "grad_norm": 1.203125,
      "learning_rate": 0.000359070992308254,
      "loss": 4.6614,
      "mean_token_accuracy": 0.26081218123435973,
      "num_tokens": 91643164.0,
      "step": 39980
    },
    {
      "entropy": 5.504410076141357,
      "epoch": 3.841018251681076,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000359039190926955,
      "loss": 4.6882,
      "mean_token_accuracy": 0.2565660610795021,
      "num_tokens": 91654111.0,
      "step": 39985
    },
    {
      "entropy": 5.361556816101074,
      "epoch": 3.84149855907781,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003590073875945154,
      "loss": 4.5136,
      "mean_token_accuracy": 0.2652531906962395,
      "num_tokens": 91665474.0,
      "step": 39990
    },
    {
      "entropy": 5.47675199508667,
      "epoch": 3.841978866474544,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035897558231167375,
      "loss": 4.7574,
      "mean_token_accuracy": 0.24453403949737548,
      "num_tokens": 91676609.0,
      "step": 39995
    },
    {
      "entropy": 5.432303953170776,
      "epoch": 3.8424591738712777,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003589437750791685,
      "loss": 4.6485,
      "mean_token_accuracy": 0.25436092019081114,
      "num_tokens": 91687213.0,
      "step": 40000
    },
    {
      "entropy": 5.441453456878662,
      "epoch": 3.8429394812680115,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035891196589773805,
      "loss": 4.7079,
      "mean_token_accuracy": 0.25416945815086367,
      "num_tokens": 91699220.0,
      "step": 40005
    },
    {
      "entropy": 5.413247108459473,
      "epoch": 3.8434197886647454,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035888015476812085,
      "loss": 4.6086,
      "mean_token_accuracy": 0.2588384822010994,
      "num_tokens": 91710020.0,
      "step": 40010
    },
    {
      "entropy": 5.459276485443115,
      "epoch": 3.8439000960614793,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003588483416910557,
      "loss": 4.6906,
      "mean_token_accuracy": 0.25607648193836213,
      "num_tokens": 91720708.0,
      "step": 40015
    },
    {
      "entropy": 5.5007476806640625,
      "epoch": 3.844380403458213,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003588165266672809,
      "loss": 4.7272,
      "mean_token_accuracy": 0.2478409931063652,
      "num_tokens": 91731545.0,
      "step": 40020
    },
    {
      "entropy": 5.463721179962159,
      "epoch": 3.844860710854947,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035878470969753524,
      "loss": 4.6862,
      "mean_token_accuracy": 0.2528249308466911,
      "num_tokens": 91743848.0,
      "step": 40025
    },
    {
      "entropy": 5.491224002838135,
      "epoch": 3.845341018251681,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035875289078255747,
      "loss": 4.6979,
      "mean_token_accuracy": 0.2573836833238602,
      "num_tokens": 91754000.0,
      "step": 40030
    },
    {
      "entropy": 5.466449308395386,
      "epoch": 3.845821325648415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035872106992308614,
      "loss": 4.6935,
      "mean_token_accuracy": 0.2543403670191765,
      "num_tokens": 91765563.0,
      "step": 40035
    },
    {
      "entropy": 5.498899316787719,
      "epoch": 3.846301633045149,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003586892471198603,
      "loss": 4.6851,
      "mean_token_accuracy": 0.25722140073776245,
      "num_tokens": 91777394.0,
      "step": 40040
    },
    {
      "entropy": 5.469364166259766,
      "epoch": 3.8467819404418826,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003586574223736186,
      "loss": 4.6446,
      "mean_token_accuracy": 0.25978811681270597,
      "num_tokens": 91787161.0,
      "step": 40045
    },
    {
      "entropy": 5.5065062046051025,
      "epoch": 3.847262247838617,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035862559568509994,
      "loss": 4.7479,
      "mean_token_accuracy": 0.2509462907910347,
      "num_tokens": 91797001.0,
      "step": 40050
    },
    {
      "entropy": 5.535959243774414,
      "epoch": 3.8477425552353504,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003585937670550433,
      "loss": 4.7466,
      "mean_token_accuracy": 0.24167226403951644,
      "num_tokens": 91808195.0,
      "step": 40055
    },
    {
      "entropy": 5.546880531311035,
      "epoch": 3.8482228626320847,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035856193648418753,
      "loss": 4.7101,
      "mean_token_accuracy": 0.2519577473402023,
      "num_tokens": 91819907.0,
      "step": 40060
    },
    {
      "entropy": 5.538257884979248,
      "epoch": 3.8487031700288186,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035853010397327186,
      "loss": 4.7578,
      "mean_token_accuracy": 0.24582972675561904,
      "num_tokens": 91831960.0,
      "step": 40065
    },
    {
      "entropy": 5.526252174377442,
      "epoch": 3.8491834774255524,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003584982695230351,
      "loss": 4.783,
      "mean_token_accuracy": 0.24745328724384308,
      "num_tokens": 91844339.0,
      "step": 40070
    },
    {
      "entropy": 5.424618291854858,
      "epoch": 3.8496637848222863,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003584664331342166,
      "loss": 4.7061,
      "mean_token_accuracy": 0.25554744601249696,
      "num_tokens": 91856207.0,
      "step": 40075
    },
    {
      "entropy": 5.534861135482788,
      "epoch": 3.85014409221902,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003584345948075552,
      "loss": 4.7359,
      "mean_token_accuracy": 0.2553664743900299,
      "num_tokens": 91868307.0,
      "step": 40080
    },
    {
      "entropy": 5.514063549041748,
      "epoch": 3.850624399615754,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003584027545437903,
      "loss": 4.7074,
      "mean_token_accuracy": 0.2541347607970238,
      "num_tokens": 91881164.0,
      "step": 40085
    },
    {
      "entropy": 5.493542766571045,
      "epoch": 3.851104707012488,
      "grad_norm": 1.25,
      "learning_rate": 0.00035837091234366107,
      "loss": 4.748,
      "mean_token_accuracy": 0.255477711558342,
      "num_tokens": 91893725.0,
      "step": 40090
    },
    {
      "entropy": 5.3770692348480225,
      "epoch": 3.851585014409222,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000358339068207907,
      "loss": 4.5863,
      "mean_token_accuracy": 0.2574327692389488,
      "num_tokens": 91904646.0,
      "step": 40095
    },
    {
      "entropy": 5.488199043273926,
      "epoch": 3.8520653218059557,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035830722213726695,
      "loss": 4.6707,
      "mean_token_accuracy": 0.261575935781002,
      "num_tokens": 91916001.0,
      "step": 40100
    },
    {
      "entropy": 5.505285120010376,
      "epoch": 3.8525456292026896,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035827537413248077,
      "loss": 4.7645,
      "mean_token_accuracy": 0.2442758396267891,
      "num_tokens": 91927247.0,
      "step": 40105
    },
    {
      "entropy": 5.510593700408935,
      "epoch": 3.8530259365994235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003582435241942877,
      "loss": 4.7289,
      "mean_token_accuracy": 0.2503656417131424,
      "num_tokens": 91938081.0,
      "step": 40110
    },
    {
      "entropy": 5.533720254898071,
      "epoch": 3.8535062439961574,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035821167232342706,
      "loss": 4.7596,
      "mean_token_accuracy": 0.25065523833036424,
      "num_tokens": 91949497.0,
      "step": 40115
    },
    {
      "entropy": 5.442519044876098,
      "epoch": 3.8539865513928913,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003581798185206385,
      "loss": 4.7024,
      "mean_token_accuracy": 0.256391766667366,
      "num_tokens": 91960134.0,
      "step": 40120
    },
    {
      "entropy": 5.454991674423217,
      "epoch": 3.8544668587896256,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035814796278666166,
      "loss": 4.6555,
      "mean_token_accuracy": 0.2613529846072197,
      "num_tokens": 91970402.0,
      "step": 40125
    },
    {
      "entropy": 5.383323526382446,
      "epoch": 3.854947166186359,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035811610512223585,
      "loss": 4.6602,
      "mean_token_accuracy": 0.258158415555954,
      "num_tokens": 91980542.0,
      "step": 40130
    },
    {
      "entropy": 5.4126396656036375,
      "epoch": 3.8554274735830933,
      "grad_norm": 1.171875,
      "learning_rate": 0.000358084245528101,
      "loss": 4.5754,
      "mean_token_accuracy": 0.26610650271177294,
      "num_tokens": 91992078.0,
      "step": 40135
    },
    {
      "entropy": 5.460590553283692,
      "epoch": 3.8559077809798272,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035805238400499667,
      "loss": 4.6555,
      "mean_token_accuracy": 0.2536802276968956,
      "num_tokens": 92002940.0,
      "step": 40140
    },
    {
      "entropy": 5.491058683395385,
      "epoch": 3.856388088376561,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003580205205536626,
      "loss": 4.677,
      "mean_token_accuracy": 0.25986259430646896,
      "num_tokens": 92015911.0,
      "step": 40145
    },
    {
      "entropy": 5.488272666931152,
      "epoch": 3.856868395773295,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003579886551748385,
      "loss": 4.7839,
      "mean_token_accuracy": 0.24396724104881287,
      "num_tokens": 92028008.0,
      "step": 40150
    },
    {
      "entropy": 5.608595991134644,
      "epoch": 3.857348703170029,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035795678786926434,
      "loss": 4.8653,
      "mean_token_accuracy": 0.24399546682834625,
      "num_tokens": 92039833.0,
      "step": 40155
    },
    {
      "entropy": 5.436152029037475,
      "epoch": 3.8578290105667628,
      "grad_norm": 1.25,
      "learning_rate": 0.00035792491863767984,
      "loss": 4.5924,
      "mean_token_accuracy": 0.25942930281162263,
      "num_tokens": 92051929.0,
      "step": 40160
    },
    {
      "entropy": 5.572142457962036,
      "epoch": 3.8583093179634966,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035789304748082496,
      "loss": 4.7253,
      "mean_token_accuracy": 0.2531443268060684,
      "num_tokens": 92062521.0,
      "step": 40165
    },
    {
      "entropy": 5.420902729034424,
      "epoch": 3.8587896253602305,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035786117439943967,
      "loss": 4.6704,
      "mean_token_accuracy": 0.2559236854314804,
      "num_tokens": 92074120.0,
      "step": 40170
    },
    {
      "entropy": 5.4998644351959225,
      "epoch": 3.8592699327569644,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035782929939426393,
      "loss": 4.6961,
      "mean_token_accuracy": 0.25394646674394605,
      "num_tokens": 92085687.0,
      "step": 40175
    },
    {
      "entropy": 5.460897302627563,
      "epoch": 3.8597502401536983,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003577974224660379,
      "loss": 4.6535,
      "mean_token_accuracy": 0.2629792481660843,
      "num_tokens": 92098034.0,
      "step": 40180
    },
    {
      "entropy": 5.544803428649902,
      "epoch": 3.860230547550432,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003577655436155015,
      "loss": 4.7925,
      "mean_token_accuracy": 0.24530757665634156,
      "num_tokens": 92108220.0,
      "step": 40185
    },
    {
      "entropy": 5.498795080184936,
      "epoch": 3.860710854947166,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003577336628433949,
      "loss": 4.6519,
      "mean_token_accuracy": 0.258937793970108,
      "num_tokens": 92120061.0,
      "step": 40190
    },
    {
      "entropy": 5.456291532516479,
      "epoch": 3.8611911623439,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035770178015045843,
      "loss": 4.6837,
      "mean_token_accuracy": 0.25112791955471037,
      "num_tokens": 92131199.0,
      "step": 40195
    },
    {
      "entropy": 5.4358930587768555,
      "epoch": 3.8616714697406342,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003576698955374321,
      "loss": 4.7014,
      "mean_token_accuracy": 0.2528317242860794,
      "num_tokens": 92142738.0,
      "step": 40200
    },
    {
      "entropy": 5.483589506149292,
      "epoch": 3.8621517771373677,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003576380090050564,
      "loss": 4.7088,
      "mean_token_accuracy": 0.2532301604747772,
      "num_tokens": 92154090.0,
      "step": 40205
    },
    {
      "entropy": 5.522116422653198,
      "epoch": 3.862632084534102,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003576061205540714,
      "loss": 4.7105,
      "mean_token_accuracy": 0.24969658255577087,
      "num_tokens": 92166100.0,
      "step": 40210
    },
    {
      "entropy": 5.456019592285156,
      "epoch": 3.863112391930836,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003575742301852176,
      "loss": 4.734,
      "mean_token_accuracy": 0.2491842433810234,
      "num_tokens": 92178336.0,
      "step": 40215
    },
    {
      "entropy": 5.543620586395264,
      "epoch": 3.8635926993275698,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003575423378992354,
      "loss": 4.691,
      "mean_token_accuracy": 0.24875947833061218,
      "num_tokens": 92189211.0,
      "step": 40220
    },
    {
      "entropy": 5.346169137954712,
      "epoch": 3.8640730067243036,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003575104436968652,
      "loss": 4.6259,
      "mean_token_accuracy": 0.2626384660601616,
      "num_tokens": 92200699.0,
      "step": 40225
    },
    {
      "entropy": 5.407923936843872,
      "epoch": 3.8645533141210375,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035747854757884744,
      "loss": 4.6097,
      "mean_token_accuracy": 0.2560175731778145,
      "num_tokens": 92212640.0,
      "step": 40230
    },
    {
      "entropy": 5.588851881027222,
      "epoch": 3.8650336215177714,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035744664954592283,
      "loss": 4.7914,
      "mean_token_accuracy": 0.2521863952279091,
      "num_tokens": 92225888.0,
      "step": 40235
    },
    {
      "entropy": 5.476335430145264,
      "epoch": 3.8655139289145053,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035741474959883175,
      "loss": 4.6777,
      "mean_token_accuracy": 0.25126454383134844,
      "num_tokens": 92237842.0,
      "step": 40240
    },
    {
      "entropy": 5.497070407867431,
      "epoch": 3.865994236311239,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000357382847738315,
      "loss": 4.7713,
      "mean_token_accuracy": 0.24821067601442337,
      "num_tokens": 92249986.0,
      "step": 40245
    },
    {
      "entropy": 5.491580009460449,
      "epoch": 3.866474543707973,
      "grad_norm": 1.25,
      "learning_rate": 0.00035735094396511304,
      "loss": 4.7379,
      "mean_token_accuracy": 0.24387039095163346,
      "num_tokens": 92262682.0,
      "step": 40250
    },
    {
      "entropy": 5.519636583328247,
      "epoch": 3.866954851104707,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003573190382799667,
      "loss": 4.7229,
      "mean_token_accuracy": 0.2542654678225517,
      "num_tokens": 92274890.0,
      "step": 40255
    },
    {
      "entropy": 5.494300031661988,
      "epoch": 3.867435158501441,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035728713068361684,
      "loss": 4.7534,
      "mean_token_accuracy": 0.2531323805451393,
      "num_tokens": 92284790.0,
      "step": 40260
    },
    {
      "entropy": 5.4754303932189945,
      "epoch": 3.8679154658981747,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035725522117680404,
      "loss": 4.7479,
      "mean_token_accuracy": 0.2491675928235054,
      "num_tokens": 92296108.0,
      "step": 40265
    },
    {
      "entropy": 5.540681457519531,
      "epoch": 3.8683957732949086,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003572233097602693,
      "loss": 4.751,
      "mean_token_accuracy": 0.2440392330288887,
      "num_tokens": 92307648.0,
      "step": 40270
    },
    {
      "entropy": 5.472910308837891,
      "epoch": 3.868876080691643,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035719139643475344,
      "loss": 4.7589,
      "mean_token_accuracy": 0.24763625115156174,
      "num_tokens": 92320405.0,
      "step": 40275
    },
    {
      "entropy": 5.497980785369873,
      "epoch": 3.8693563880883763,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035715948120099747,
      "loss": 4.7129,
      "mean_token_accuracy": 0.25611880868673326,
      "num_tokens": 92331903.0,
      "step": 40280
    },
    {
      "entropy": 5.543249845504761,
      "epoch": 3.8698366954851107,
      "grad_norm": 1.25,
      "learning_rate": 0.00035712756405974216,
      "loss": 4.7702,
      "mean_token_accuracy": 0.24904498904943467,
      "num_tokens": 92343336.0,
      "step": 40285
    },
    {
      "entropy": 5.476924324035645,
      "epoch": 3.870317002881844,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003570956450117288,
      "loss": 4.743,
      "mean_token_accuracy": 0.24904401004314422,
      "num_tokens": 92356283.0,
      "step": 40290
    },
    {
      "entropy": 5.487257385253907,
      "epoch": 3.8707973102785784,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003570637240576983,
      "loss": 4.7315,
      "mean_token_accuracy": 0.25146324187517166,
      "num_tokens": 92367893.0,
      "step": 40295
    },
    {
      "entropy": 5.487522888183594,
      "epoch": 3.8712776176753123,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035703180119839187,
      "loss": 4.6388,
      "mean_token_accuracy": 0.2528821796178818,
      "num_tokens": 92379019.0,
      "step": 40300
    },
    {
      "entropy": 5.425402593612671,
      "epoch": 3.871757925072046,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003569998764345506,
      "loss": 4.6357,
      "mean_token_accuracy": 0.26301144510507585,
      "num_tokens": 92390082.0,
      "step": 40305
    },
    {
      "entropy": 5.445456600189209,
      "epoch": 3.87223823246878,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035696794976691564,
      "loss": 4.6884,
      "mean_token_accuracy": 0.2562153473496437,
      "num_tokens": 92401323.0,
      "step": 40310
    },
    {
      "entropy": 5.466891241073609,
      "epoch": 3.872718539865514,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035693602119622823,
      "loss": 4.6604,
      "mean_token_accuracy": 0.2512710362672806,
      "num_tokens": 92412329.0,
      "step": 40315
    },
    {
      "entropy": 5.62332854270935,
      "epoch": 3.873198847262248,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003569040907232298,
      "loss": 4.8386,
      "mean_token_accuracy": 0.24497848600149155,
      "num_tokens": 92423267.0,
      "step": 40320
    },
    {
      "entropy": 5.400581169128418,
      "epoch": 3.8736791546589817,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035687215834866154,
      "loss": 4.6462,
      "mean_token_accuracy": 0.25937180668115617,
      "num_tokens": 92434519.0,
      "step": 40325
    },
    {
      "entropy": 5.4409482955932615,
      "epoch": 3.8741594620557156,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035684022407326486,
      "loss": 4.6246,
      "mean_token_accuracy": 0.2559319093823433,
      "num_tokens": 92445834.0,
      "step": 40330
    },
    {
      "entropy": 5.432407331466675,
      "epoch": 3.8746397694524495,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035680828789778126,
      "loss": 4.7334,
      "mean_token_accuracy": 0.2501261979341507,
      "num_tokens": 92457648.0,
      "step": 40335
    },
    {
      "entropy": 5.481592655181885,
      "epoch": 3.8751200768491834,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003567763498229521,
      "loss": 4.7273,
      "mean_token_accuracy": 0.25851349979639054,
      "num_tokens": 92469080.0,
      "step": 40340
    },
    {
      "entropy": 5.5206304550170895,
      "epoch": 3.8756003842459172,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003567444098495189,
      "loss": 4.7428,
      "mean_token_accuracy": 0.2468013733625412,
      "num_tokens": 92481679.0,
      "step": 40345
    },
    {
      "entropy": 5.494037532806397,
      "epoch": 3.8760806916426516,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035671246797822325,
      "loss": 4.7012,
      "mean_token_accuracy": 0.255867101252079,
      "num_tokens": 92493143.0,
      "step": 40350
    },
    {
      "entropy": 5.469069528579712,
      "epoch": 3.876560999039385,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003566805242098067,
      "loss": 4.654,
      "mean_token_accuracy": 0.25358297526836393,
      "num_tokens": 92504302.0,
      "step": 40355
    },
    {
      "entropy": 5.52483868598938,
      "epoch": 3.8770413064361193,
      "grad_norm": 1.203125,
      "learning_rate": 0.000356648578545011,
      "loss": 4.7121,
      "mean_token_accuracy": 0.2499252587556839,
      "num_tokens": 92514797.0,
      "step": 40360
    },
    {
      "entropy": 5.555805397033692,
      "epoch": 3.8775216138328528,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003566166309845777,
      "loss": 4.7949,
      "mean_token_accuracy": 0.24363376945257187,
      "num_tokens": 92526492.0,
      "step": 40365
    },
    {
      "entropy": 5.462129163742065,
      "epoch": 3.878001921229587,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003565846815292486,
      "loss": 4.7251,
      "mean_token_accuracy": 0.25032509714365003,
      "num_tokens": 92536980.0,
      "step": 40370
    },
    {
      "entropy": 5.435908412933349,
      "epoch": 3.878482228626321,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003565527301797654,
      "loss": 4.6848,
      "mean_token_accuracy": 0.2518529802560806,
      "num_tokens": 92548384.0,
      "step": 40375
    },
    {
      "entropy": 5.415638017654419,
      "epoch": 3.878962536023055,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035652077693687,
      "loss": 4.6907,
      "mean_token_accuracy": 0.2524256855249405,
      "num_tokens": 92559853.0,
      "step": 40380
    },
    {
      "entropy": 5.514239645004272,
      "epoch": 3.8794428434197887,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003564888218013042,
      "loss": 4.7859,
      "mean_token_accuracy": 0.24062766432762145,
      "num_tokens": 92570740.0,
      "step": 40385
    },
    {
      "entropy": 5.378829669952393,
      "epoch": 3.8799231508165226,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035645686477380996,
      "loss": 4.5813,
      "mean_token_accuracy": 0.2633738473057747,
      "num_tokens": 92583246.0,
      "step": 40390
    },
    {
      "entropy": 5.5912047863006595,
      "epoch": 3.8804034582132565,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003564249058551293,
      "loss": 4.8012,
      "mean_token_accuracy": 0.246279539167881,
      "num_tokens": 92595551.0,
      "step": 40395
    },
    {
      "entropy": 5.444998168945313,
      "epoch": 3.8808837656099904,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003563929450460039,
      "loss": 4.6063,
      "mean_token_accuracy": 0.26408320367336274,
      "num_tokens": 92608620.0,
      "step": 40400
    },
    {
      "entropy": 5.424344491958618,
      "epoch": 3.8813640730067243,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003563609823471761,
      "loss": 4.6829,
      "mean_token_accuracy": 0.2501148745417595,
      "num_tokens": 92620342.0,
      "step": 40405
    },
    {
      "entropy": 5.387424468994141,
      "epoch": 3.881844380403458,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035632901775938794,
      "loss": 4.6601,
      "mean_token_accuracy": 0.26059499233961103,
      "num_tokens": 92630633.0,
      "step": 40410
    },
    {
      "entropy": 5.487371349334717,
      "epoch": 3.882324687800192,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003562970512833814,
      "loss": 4.8665,
      "mean_token_accuracy": 0.23385860174894332,
      "num_tokens": 92643134.0,
      "step": 40415
    },
    {
      "entropy": 5.500598049163818,
      "epoch": 3.882804995196926,
      "grad_norm": 1.125,
      "learning_rate": 0.0003562650829198988,
      "loss": 4.6339,
      "mean_token_accuracy": 0.25875622779130936,
      "num_tokens": 92654765.0,
      "step": 40420
    },
    {
      "entropy": 5.538136434555054,
      "epoch": 3.88328530259366,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035623311266968225,
      "loss": 4.7633,
      "mean_token_accuracy": 0.25320742577314376,
      "num_tokens": 92666321.0,
      "step": 40425
    },
    {
      "entropy": 5.4459497928619385,
      "epoch": 3.8837656099903937,
      "grad_norm": 1.109375,
      "learning_rate": 0.000356201140533474,
      "loss": 4.6989,
      "mean_token_accuracy": 0.25215385407209395,
      "num_tokens": 92679214.0,
      "step": 40430
    },
    {
      "entropy": 5.383370876312256,
      "epoch": 3.884245917387128,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003561691665120164,
      "loss": 4.6869,
      "mean_token_accuracy": 0.2555983603000641,
      "num_tokens": 92692064.0,
      "step": 40435
    },
    {
      "entropy": 5.520705795288086,
      "epoch": 3.8847262247838614,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003561371906060518,
      "loss": 4.7199,
      "mean_token_accuracy": 0.25554081350564956,
      "num_tokens": 92704732.0,
      "step": 40440
    },
    {
      "entropy": 5.520569562911987,
      "epoch": 3.8852065321805958,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035610521281632257,
      "loss": 4.7118,
      "mean_token_accuracy": 0.2563439905643463,
      "num_tokens": 92714176.0,
      "step": 40445
    },
    {
      "entropy": 5.46630892753601,
      "epoch": 3.8856868395773296,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003560732331435711,
      "loss": 4.6778,
      "mean_token_accuracy": 0.2507488563656807,
      "num_tokens": 92725874.0,
      "step": 40450
    },
    {
      "entropy": 5.363512277603149,
      "epoch": 3.8861671469740635,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035604125158853996,
      "loss": 4.5972,
      "mean_token_accuracy": 0.2632381200790405,
      "num_tokens": 92738040.0,
      "step": 40455
    },
    {
      "entropy": 5.527319526672363,
      "epoch": 3.8866474543707974,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003560092681519715,
      "loss": 4.7031,
      "mean_token_accuracy": 0.2527434751391411,
      "num_tokens": 92749818.0,
      "step": 40460
    },
    {
      "entropy": 5.496093893051148,
      "epoch": 3.8871277617675313,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035597728283460843,
      "loss": 4.7003,
      "mean_token_accuracy": 0.25649800449609755,
      "num_tokens": 92761209.0,
      "step": 40465
    },
    {
      "entropy": 5.419841623306274,
      "epoch": 3.887608069164265,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035594529563719336,
      "loss": 4.62,
      "mean_token_accuracy": 0.2567827597260475,
      "num_tokens": 92771698.0,
      "step": 40470
    },
    {
      "entropy": 5.480290126800537,
      "epoch": 3.888088376560999,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003559133065604688,
      "loss": 4.7361,
      "mean_token_accuracy": 0.25420363396406176,
      "num_tokens": 92783753.0,
      "step": 40475
    },
    {
      "entropy": 5.469707059860229,
      "epoch": 3.888568683957733,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003558813156051776,
      "loss": 4.6664,
      "mean_token_accuracy": 0.25528246760368345,
      "num_tokens": 92796640.0,
      "step": 40480
    },
    {
      "entropy": 5.527066659927368,
      "epoch": 3.889048991354467,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003558493227720624,
      "loss": 4.7465,
      "mean_token_accuracy": 0.25088978111743926,
      "num_tokens": 92808163.0,
      "step": 40485
    },
    {
      "entropy": 5.450945329666138,
      "epoch": 3.8895292987512007,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035581732806186595,
      "loss": 4.6528,
      "mean_token_accuracy": 0.2623646005988121,
      "num_tokens": 92818953.0,
      "step": 40490
    },
    {
      "entropy": 5.457440757751465,
      "epoch": 3.8900096061479346,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003557853314753311,
      "loss": 4.7058,
      "mean_token_accuracy": 0.24933891892433166,
      "num_tokens": 92830555.0,
      "step": 40495
    },
    {
      "entropy": 5.560046052932739,
      "epoch": 3.8904899135446684,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035575333301320083,
      "loss": 4.7496,
      "mean_token_accuracy": 0.24887769371271135,
      "num_tokens": 92841491.0,
      "step": 40500
    },
    {
      "entropy": 5.517280912399292,
      "epoch": 3.8909702209414023,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035572133267621793,
      "loss": 4.7298,
      "mean_token_accuracy": 0.25752563774585724,
      "num_tokens": 92853102.0,
      "step": 40505
    },
    {
      "entropy": 5.462132406234741,
      "epoch": 3.8914505283381366,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003556893304651253,
      "loss": 4.6918,
      "mean_token_accuracy": 0.25059599429368973,
      "num_tokens": 92863034.0,
      "step": 40510
    },
    {
      "entropy": 5.463887071609497,
      "epoch": 3.89193083573487,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003556573263806661,
      "loss": 4.7175,
      "mean_token_accuracy": 0.25086410641670226,
      "num_tokens": 92875076.0,
      "step": 40515
    },
    {
      "entropy": 5.456852769851684,
      "epoch": 3.8924111431316044,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003556253204235833,
      "loss": 4.7006,
      "mean_token_accuracy": 0.25358116924762725,
      "num_tokens": 92886725.0,
      "step": 40520
    },
    {
      "entropy": 5.530939197540283,
      "epoch": 3.8928914505283383,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003555933125946199,
      "loss": 4.7228,
      "mean_token_accuracy": 0.24409096091985702,
      "num_tokens": 92897244.0,
      "step": 40525
    },
    {
      "entropy": 5.578475046157837,
      "epoch": 3.893371757925072,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003555613028945191,
      "loss": 4.784,
      "mean_token_accuracy": 0.2484847515821457,
      "num_tokens": 92908161.0,
      "step": 40530
    },
    {
      "entropy": 5.454455423355102,
      "epoch": 3.893852065321806,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035552929132402414,
      "loss": 4.6555,
      "mean_token_accuracy": 0.2530456006526947,
      "num_tokens": 92919065.0,
      "step": 40535
    },
    {
      "entropy": 5.546929836273193,
      "epoch": 3.89433237271854,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035549727788387805,
      "loss": 4.7902,
      "mean_token_accuracy": 0.24524426758289336,
      "num_tokens": 92930224.0,
      "step": 40540
    },
    {
      "entropy": 5.50043215751648,
      "epoch": 3.894812680115274,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035546526257482426,
      "loss": 4.7244,
      "mean_token_accuracy": 0.2603514015674591,
      "num_tokens": 92942109.0,
      "step": 40545
    },
    {
      "entropy": 5.471337366104126,
      "epoch": 3.8952929875120077,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035543324539760604,
      "loss": 4.6563,
      "mean_token_accuracy": 0.25815410166978836,
      "num_tokens": 92954984.0,
      "step": 40550
    },
    {
      "entropy": 5.475752067565918,
      "epoch": 3.8957732949087416,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003554012263529666,
      "loss": 4.7105,
      "mean_token_accuracy": 0.24808614403009416,
      "num_tokens": 92965946.0,
      "step": 40555
    },
    {
      "entropy": 5.505849885940552,
      "epoch": 3.8962536023054755,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003553692054416495,
      "loss": 4.6959,
      "mean_token_accuracy": 0.2547337025403976,
      "num_tokens": 92978035.0,
      "step": 40560
    },
    {
      "entropy": 5.479494619369507,
      "epoch": 3.8967339097022093,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003553371826643981,
      "loss": 4.7253,
      "mean_token_accuracy": 0.25114298015832903,
      "num_tokens": 92989647.0,
      "step": 40565
    },
    {
      "entropy": 5.410842990875244,
      "epoch": 3.8972142170989432,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003553051580219558,
      "loss": 4.6694,
      "mean_token_accuracy": 0.2566041752696037,
      "num_tokens": 93000699.0,
      "step": 40570
    },
    {
      "entropy": 5.394049596786499,
      "epoch": 3.897694524495677,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003552731315150662,
      "loss": 4.6035,
      "mean_token_accuracy": 0.260003587603569,
      "num_tokens": 93011463.0,
      "step": 40575
    },
    {
      "entropy": 5.527338933944702,
      "epoch": 3.898174831892411,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035524110314447295,
      "loss": 4.7277,
      "mean_token_accuracy": 0.2526299610733986,
      "num_tokens": 93022147.0,
      "step": 40580
    },
    {
      "entropy": 5.4013103485107425,
      "epoch": 3.8986551392891453,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003552090729109195,
      "loss": 4.5964,
      "mean_token_accuracy": 0.2616342231631279,
      "num_tokens": 93035155.0,
      "step": 40585
    },
    {
      "entropy": 5.548247146606445,
      "epoch": 3.8991354466858787,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003551770408151496,
      "loss": 4.7087,
      "mean_token_accuracy": 0.257383269071579,
      "num_tokens": 93045724.0,
      "step": 40590
    },
    {
      "entropy": 5.497446298599243,
      "epoch": 3.899615754082613,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035514500685790673,
      "loss": 4.7602,
      "mean_token_accuracy": 0.25260433852672576,
      "num_tokens": 93057629.0,
      "step": 40595
    },
    {
      "entropy": 5.455983304977417,
      "epoch": 3.9000960614793465,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035511297103993486,
      "loss": 4.7212,
      "mean_token_accuracy": 0.2530237421393394,
      "num_tokens": 93068426.0,
      "step": 40600
    },
    {
      "entropy": 5.474006605148316,
      "epoch": 3.900576368876081,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003550809333619778,
      "loss": 4.7132,
      "mean_token_accuracy": 0.2543066531419754,
      "num_tokens": 93079344.0,
      "step": 40605
    },
    {
      "entropy": 5.4450935363769535,
      "epoch": 3.9010566762728147,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003550488938247791,
      "loss": 4.7021,
      "mean_token_accuracy": 0.2497902914881706,
      "num_tokens": 93091111.0,
      "step": 40610
    },
    {
      "entropy": 5.510510683059692,
      "epoch": 3.9015369836695486,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003550168524290828,
      "loss": 4.7391,
      "mean_token_accuracy": 0.2497907668352127,
      "num_tokens": 93102202.0,
      "step": 40615
    },
    {
      "entropy": 5.417462778091431,
      "epoch": 3.9020172910662825,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035498480917563294,
      "loss": 4.652,
      "mean_token_accuracy": 0.2614865466952324,
      "num_tokens": 93113575.0,
      "step": 40620
    },
    {
      "entropy": 5.466286993026733,
      "epoch": 3.9024975984630164,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003549527640651732,
      "loss": 4.7107,
      "mean_token_accuracy": 0.2519951343536377,
      "num_tokens": 93125402.0,
      "step": 40625
    },
    {
      "entropy": 5.469207715988159,
      "epoch": 3.9029779058597502,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003549207170984476,
      "loss": 4.662,
      "mean_token_accuracy": 0.26707591116428375,
      "num_tokens": 93138147.0,
      "step": 40630
    },
    {
      "entropy": 5.53174991607666,
      "epoch": 3.903458213256484,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035488866827620047,
      "loss": 4.6812,
      "mean_token_accuracy": 0.25745621621608733,
      "num_tokens": 93150299.0,
      "step": 40635
    },
    {
      "entropy": 5.404725074768066,
      "epoch": 3.903938520653218,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003548566175991756,
      "loss": 4.579,
      "mean_token_accuracy": 0.26690110862255095,
      "num_tokens": 93162224.0,
      "step": 40640
    },
    {
      "entropy": 5.474523448944092,
      "epoch": 3.904418828049952,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035482456506811714,
      "loss": 4.7238,
      "mean_token_accuracy": 0.2569157987833023,
      "num_tokens": 93173144.0,
      "step": 40645
    },
    {
      "entropy": 5.612341690063476,
      "epoch": 3.9048991354466858,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035479251068376935,
      "loss": 4.8617,
      "mean_token_accuracy": 0.23579824417829515,
      "num_tokens": 93183515.0,
      "step": 40650
    },
    {
      "entropy": 5.393214416503906,
      "epoch": 3.9053794428434196,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035476045444687637,
      "loss": 4.6101,
      "mean_token_accuracy": 0.26359681636095045,
      "num_tokens": 93194184.0,
      "step": 40655
    },
    {
      "entropy": 5.550841760635376,
      "epoch": 3.905859750240154,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003547283963581825,
      "loss": 4.8734,
      "mean_token_accuracy": 0.24649241864681243,
      "num_tokens": 93205088.0,
      "step": 40660
    },
    {
      "entropy": 5.473793077468872,
      "epoch": 3.9063400576368874,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000354696336418432,
      "loss": 4.6779,
      "mean_token_accuracy": 0.25819325894117356,
      "num_tokens": 93216160.0,
      "step": 40665
    },
    {
      "entropy": 5.440382862091065,
      "epoch": 3.9068203650336217,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035466427462836923,
      "loss": 4.625,
      "mean_token_accuracy": 0.26170748472213745,
      "num_tokens": 93227658.0,
      "step": 40670
    },
    {
      "entropy": 5.475371789932251,
      "epoch": 3.907300672430355,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035463221098873854,
      "loss": 4.7098,
      "mean_token_accuracy": 0.25360559225082396,
      "num_tokens": 93238684.0,
      "step": 40675
    },
    {
      "entropy": 5.530123472213745,
      "epoch": 3.9077809798270895,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003546001455002844,
      "loss": 4.8439,
      "mean_token_accuracy": 0.23734061568975448,
      "num_tokens": 93249772.0,
      "step": 40680
    },
    {
      "entropy": 5.509752035140991,
      "epoch": 3.9082612872238234,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003545680781637512,
      "loss": 4.7032,
      "mean_token_accuracy": 0.25314990133047105,
      "num_tokens": 93261642.0,
      "step": 40685
    },
    {
      "entropy": 5.492084598541259,
      "epoch": 3.9087415946205573,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003545360089798836,
      "loss": 4.6719,
      "mean_token_accuracy": 0.25453430563211443,
      "num_tokens": 93273704.0,
      "step": 40690
    },
    {
      "entropy": 5.4664044857025145,
      "epoch": 3.909221902017291,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003545039379494259,
      "loss": 4.7053,
      "mean_token_accuracy": 0.2537610948085785,
      "num_tokens": 93285396.0,
      "step": 40695
    },
    {
      "entropy": 5.52794771194458,
      "epoch": 3.909702209414025,
      "grad_norm": 1.1875,
      "learning_rate": 0.000354471865073123,
      "loss": 4.7417,
      "mean_token_accuracy": 0.24906659871339798,
      "num_tokens": 93297411.0,
      "step": 40700
    },
    {
      "entropy": 5.487019824981689,
      "epoch": 3.910182516810759,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003544397903517192,
      "loss": 4.6522,
      "mean_token_accuracy": 0.2583379209041595,
      "num_tokens": 93310545.0,
      "step": 40705
    },
    {
      "entropy": 5.448687887191772,
      "epoch": 3.910662824207493,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003544077137859595,
      "loss": 4.6568,
      "mean_token_accuracy": 0.2579941675066948,
      "num_tokens": 93321851.0,
      "step": 40710
    },
    {
      "entropy": 5.494442129135132,
      "epoch": 3.9111431316042267,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035437563537658835,
      "loss": 4.6782,
      "mean_token_accuracy": 0.255921496450901,
      "num_tokens": 93332662.0,
      "step": 40715
    },
    {
      "entropy": 5.40415997505188,
      "epoch": 3.9116234390009605,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035434355512435075,
      "loss": 4.6174,
      "mean_token_accuracy": 0.25912383645772935,
      "num_tokens": 93344979.0,
      "step": 40720
    },
    {
      "entropy": 5.480033826828003,
      "epoch": 3.9121037463976944,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003543114730299913,
      "loss": 4.6628,
      "mean_token_accuracy": 0.262871190905571,
      "num_tokens": 93356427.0,
      "step": 40725
    },
    {
      "entropy": 5.517547988891602,
      "epoch": 3.9125840537944283,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003542793890942551,
      "loss": 4.7141,
      "mean_token_accuracy": 0.2513091966509819,
      "num_tokens": 93368377.0,
      "step": 40730
    },
    {
      "entropy": 5.484983634948731,
      "epoch": 3.913064361191162,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003542473033178868,
      "loss": 4.717,
      "mean_token_accuracy": 0.25161408483982084,
      "num_tokens": 93378203.0,
      "step": 40735
    },
    {
      "entropy": 5.383976030349731,
      "epoch": 3.913544668587896,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003542152157016315,
      "loss": 4.6098,
      "mean_token_accuracy": 0.26396198719739916,
      "num_tokens": 93390326.0,
      "step": 40740
    },
    {
      "entropy": 5.444323205947876,
      "epoch": 3.9140249759846304,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000354183126246234,
      "loss": 4.6839,
      "mean_token_accuracy": 0.2572936609387398,
      "num_tokens": 93401364.0,
      "step": 40745
    },
    {
      "entropy": 5.431288862228394,
      "epoch": 3.914505283381364,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003541510349524395,
      "loss": 4.6095,
      "mean_token_accuracy": 0.26549686342477796,
      "num_tokens": 93412468.0,
      "step": 40750
    },
    {
      "entropy": 5.4944079399108885,
      "epoch": 3.914985590778098,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000354118941820993,
      "loss": 4.7813,
      "mean_token_accuracy": 0.24600871354341508,
      "num_tokens": 93424605.0,
      "step": 40755
    },
    {
      "entropy": 5.433273506164551,
      "epoch": 3.915465898174832,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035408684685263954,
      "loss": 4.5901,
      "mean_token_accuracy": 0.25419677793979645,
      "num_tokens": 93435624.0,
      "step": 40760
    },
    {
      "entropy": 5.583983469009399,
      "epoch": 3.915946205571566,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003540547500481244,
      "loss": 4.8346,
      "mean_token_accuracy": 0.23547417372465135,
      "num_tokens": 93447455.0,
      "step": 40765
    },
    {
      "entropy": 5.488719034194946,
      "epoch": 3.9164265129683,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003540226514081927,
      "loss": 4.6842,
      "mean_token_accuracy": 0.2541194036602974,
      "num_tokens": 93459928.0,
      "step": 40770
    },
    {
      "entropy": 5.499241542816162,
      "epoch": 3.9169068203650337,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035399055093358965,
      "loss": 4.7448,
      "mean_token_accuracy": 0.2525720432400703,
      "num_tokens": 93471277.0,
      "step": 40775
    },
    {
      "entropy": 5.540633916854858,
      "epoch": 3.9173871277617676,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035395844862506063,
      "loss": 4.7052,
      "mean_token_accuracy": 0.25147412419319154,
      "num_tokens": 93483085.0,
      "step": 40780
    },
    {
      "entropy": 5.439450597763061,
      "epoch": 3.9178674351585014,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003539263444833509,
      "loss": 4.6959,
      "mean_token_accuracy": 0.25819700956344604,
      "num_tokens": 93494769.0,
      "step": 40785
    },
    {
      "entropy": 5.398166227340698,
      "epoch": 3.9183477425552353,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035389423850920576,
      "loss": 4.6635,
      "mean_token_accuracy": 0.25370920151472093,
      "num_tokens": 93506860.0,
      "step": 40790
    },
    {
      "entropy": 5.392376327514649,
      "epoch": 3.918828049951969,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035386213070337063,
      "loss": 4.6487,
      "mean_token_accuracy": 0.254948790371418,
      "num_tokens": 93519260.0,
      "step": 40795
    },
    {
      "entropy": 5.500998640060425,
      "epoch": 3.919308357348703,
      "grad_norm": 1.125,
      "learning_rate": 0.00035383002106659114,
      "loss": 4.6687,
      "mean_token_accuracy": 0.24985571950674057,
      "num_tokens": 93530261.0,
      "step": 40800
    },
    {
      "entropy": 5.516722345352173,
      "epoch": 3.919788664745437,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035379790959961247,
      "loss": 4.7453,
      "mean_token_accuracy": 0.25325244963169097,
      "num_tokens": 93542317.0,
      "step": 40805
    },
    {
      "entropy": 5.461720895767212,
      "epoch": 3.920268972142171,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003537657963031804,
      "loss": 4.6767,
      "mean_token_accuracy": 0.25789555758237837,
      "num_tokens": 93554038.0,
      "step": 40810
    },
    {
      "entropy": 5.482616662979126,
      "epoch": 3.9207492795389047,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035373368117804047,
      "loss": 4.686,
      "mean_token_accuracy": 0.25276737064123156,
      "num_tokens": 93565263.0,
      "step": 40815
    },
    {
      "entropy": 5.579107761383057,
      "epoch": 3.921229586935639,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035370156422493814,
      "loss": 4.7954,
      "mean_token_accuracy": 0.24843704849481582,
      "num_tokens": 93575724.0,
      "step": 40820
    },
    {
      "entropy": 5.506059646606445,
      "epoch": 3.9217098943323725,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035366944544461926,
      "loss": 4.6794,
      "mean_token_accuracy": 0.25528088212013245,
      "num_tokens": 93587289.0,
      "step": 40825
    },
    {
      "entropy": 5.522081613540649,
      "epoch": 3.922190201729107,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003536373248378295,
      "loss": 4.7799,
      "mean_token_accuracy": 0.25133405327796937,
      "num_tokens": 93599993.0,
      "step": 40830
    },
    {
      "entropy": 5.4688352108001705,
      "epoch": 3.9226705091258407,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035360520240531446,
      "loss": 4.7038,
      "mean_token_accuracy": 0.25477886348962786,
      "num_tokens": 93611542.0,
      "step": 40835
    },
    {
      "entropy": 5.425338506698608,
      "epoch": 3.9231508165225746,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035357307814782007,
      "loss": 4.6293,
      "mean_token_accuracy": 0.25597025603055956,
      "num_tokens": 93622650.0,
      "step": 40840
    },
    {
      "entropy": 5.476326513290405,
      "epoch": 3.9236311239193085,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003535409520660922,
      "loss": 4.6242,
      "mean_token_accuracy": 0.259578050673008,
      "num_tokens": 93634481.0,
      "step": 40845
    },
    {
      "entropy": 5.42905559539795,
      "epoch": 3.9241114313160423,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003535088241608765,
      "loss": 4.7132,
      "mean_token_accuracy": 0.2511111244559288,
      "num_tokens": 93647225.0,
      "step": 40850
    },
    {
      "entropy": 5.455616188049317,
      "epoch": 3.9245917387127762,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003534766944329191,
      "loss": 4.659,
      "mean_token_accuracy": 0.2636229872703552,
      "num_tokens": 93657984.0,
      "step": 40855
    },
    {
      "entropy": 5.503946828842163,
      "epoch": 3.92507204610951,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035344456288296595,
      "loss": 4.6982,
      "mean_token_accuracy": 0.24757488071918488,
      "num_tokens": 93669420.0,
      "step": 40860
    },
    {
      "entropy": 5.479670000076294,
      "epoch": 3.925552353506244,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003534124295117629,
      "loss": 4.6388,
      "mean_token_accuracy": 0.2648668706417084,
      "num_tokens": 93680649.0,
      "step": 40865
    },
    {
      "entropy": 5.481721591949463,
      "epoch": 3.926032660902978,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003533802943200561,
      "loss": 4.7612,
      "mean_token_accuracy": 0.24293488413095474,
      "num_tokens": 93693314.0,
      "step": 40870
    },
    {
      "entropy": 5.5228899955749515,
      "epoch": 3.9265129682997117,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003533481573085916,
      "loss": 4.6536,
      "mean_token_accuracy": 0.25507027506828306,
      "num_tokens": 93704271.0,
      "step": 40875
    },
    {
      "entropy": 5.461873340606689,
      "epoch": 3.9269932756964456,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003533160184781155,
      "loss": 4.6604,
      "mean_token_accuracy": 0.26373025476932527,
      "num_tokens": 93715561.0,
      "step": 40880
    },
    {
      "entropy": 5.447552251815796,
      "epoch": 3.9274735830931795,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000353283877829374,
      "loss": 4.7207,
      "mean_token_accuracy": 0.25096276551485064,
      "num_tokens": 93727115.0,
      "step": 40885
    },
    {
      "entropy": 5.501708936691284,
      "epoch": 3.9279538904899134,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003532517353631134,
      "loss": 4.6735,
      "mean_token_accuracy": 0.25376079231500626,
      "num_tokens": 93738439.0,
      "step": 40890
    },
    {
      "entropy": 5.513786840438843,
      "epoch": 3.9284341978866477,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003532195910800798,
      "loss": 4.6743,
      "mean_token_accuracy": 0.2564789280295372,
      "num_tokens": 93749722.0,
      "step": 40895
    },
    {
      "entropy": 5.508312082290649,
      "epoch": 3.928914505283381,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003531874449810195,
      "loss": 4.7025,
      "mean_token_accuracy": 0.2497492641210556,
      "num_tokens": 93761122.0,
      "step": 40900
    },
    {
      "entropy": 5.527310132980347,
      "epoch": 3.9293948126801155,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035315529706667905,
      "loss": 4.7241,
      "mean_token_accuracy": 0.24996845573186874,
      "num_tokens": 93771369.0,
      "step": 40905
    },
    {
      "entropy": 5.531266355514527,
      "epoch": 3.929875120076849,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003531231473378046,
      "loss": 4.7426,
      "mean_token_accuracy": 0.24091127067804335,
      "num_tokens": 93782380.0,
      "step": 40910
    },
    {
      "entropy": 5.488978242874145,
      "epoch": 3.9303554274735832,
      "grad_norm": 1.25,
      "learning_rate": 0.00035309099579514255,
      "loss": 4.6844,
      "mean_token_accuracy": 0.24977275282144545,
      "num_tokens": 93793162.0,
      "step": 40915
    },
    {
      "entropy": 5.55873064994812,
      "epoch": 3.930835734870317,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003530588424394396,
      "loss": 4.7561,
      "mean_token_accuracy": 0.2504671931266785,
      "num_tokens": 93804441.0,
      "step": 40920
    },
    {
      "entropy": 5.521816539764404,
      "epoch": 3.931316042267051,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035302668727144206,
      "loss": 4.7231,
      "mean_token_accuracy": 0.250603212416172,
      "num_tokens": 93816080.0,
      "step": 40925
    },
    {
      "entropy": 5.492085933685303,
      "epoch": 3.931796349663785,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035299453029189647,
      "loss": 4.7352,
      "mean_token_accuracy": 0.248833267390728,
      "num_tokens": 93828494.0,
      "step": 40930
    },
    {
      "entropy": 5.519677352905274,
      "epoch": 3.9322766570605188,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035296237150154964,
      "loss": 4.7448,
      "mean_token_accuracy": 0.24560018479824067,
      "num_tokens": 93839323.0,
      "step": 40935
    },
    {
      "entropy": 5.553723001480103,
      "epoch": 3.9327569644572526,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003529302109011479,
      "loss": 4.7464,
      "mean_token_accuracy": 0.2538365706801414,
      "num_tokens": 93851474.0,
      "step": 40940
    },
    {
      "entropy": 5.535925579071045,
      "epoch": 3.9332372718539865,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035289804849143806,
      "loss": 4.8098,
      "mean_token_accuracy": 0.24475409090518951,
      "num_tokens": 93863484.0,
      "step": 40945
    },
    {
      "entropy": 5.45222544670105,
      "epoch": 3.9337175792507204,
      "grad_norm": 1.109375,
      "learning_rate": 0.000352865884273167,
      "loss": 4.6761,
      "mean_token_accuracy": 0.26238584220409394,
      "num_tokens": 93874124.0,
      "step": 40950
    },
    {
      "entropy": 5.4992493152618405,
      "epoch": 3.9341978866474543,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003528337182470812,
      "loss": 4.7186,
      "mean_token_accuracy": 0.25079509168863295,
      "num_tokens": 93884880.0,
      "step": 40955
    },
    {
      "entropy": 5.401667547225952,
      "epoch": 3.934678194044188,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035280155041392757,
      "loss": 4.6375,
      "mean_token_accuracy": 0.26116674542427065,
      "num_tokens": 93895737.0,
      "step": 40960
    },
    {
      "entropy": 5.521459531784058,
      "epoch": 3.935158501440922,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003527693807744531,
      "loss": 4.704,
      "mean_token_accuracy": 0.25368867814540863,
      "num_tokens": 93906916.0,
      "step": 40965
    },
    {
      "entropy": 5.505608940124512,
      "epoch": 3.9356388088376564,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003527372093294044,
      "loss": 4.7154,
      "mean_token_accuracy": 0.24684349596500396,
      "num_tokens": 93918471.0,
      "step": 40970
    },
    {
      "entropy": 5.420804262161255,
      "epoch": 3.93611911623439,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003527050360795285,
      "loss": 4.6048,
      "mean_token_accuracy": 0.26088410168886184,
      "num_tokens": 93929874.0,
      "step": 40975
    },
    {
      "entropy": 5.499696207046509,
      "epoch": 3.936599423631124,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003526728610255725,
      "loss": 4.7454,
      "mean_token_accuracy": 0.24481874853372573,
      "num_tokens": 93940785.0,
      "step": 40980
    },
    {
      "entropy": 5.527274894714355,
      "epoch": 3.9370797310278576,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035264068416828326,
      "loss": 4.7109,
      "mean_token_accuracy": 0.24619605988264084,
      "num_tokens": 93953255.0,
      "step": 40985
    },
    {
      "entropy": 5.4343420505523685,
      "epoch": 3.937560038424592,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003526085055084079,
      "loss": 4.6331,
      "mean_token_accuracy": 0.255363667011261,
      "num_tokens": 93964581.0,
      "step": 40990
    },
    {
      "entropy": 5.472287940979004,
      "epoch": 3.938040345821326,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003525763250466935,
      "loss": 4.7531,
      "mean_token_accuracy": 0.24579035937786103,
      "num_tokens": 93976222.0,
      "step": 40995
    },
    {
      "entropy": 5.512660121917724,
      "epoch": 3.9385206532180597,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003525441427838872,
      "loss": 4.7769,
      "mean_token_accuracy": 0.2474414125084877,
      "num_tokens": 93986936.0,
      "step": 41000
    },
    {
      "entropy": 5.523522233963012,
      "epoch": 3.9390009606147935,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003525119587207361,
      "loss": 4.7153,
      "mean_token_accuracy": 0.2521307498216629,
      "num_tokens": 93999518.0,
      "step": 41005
    },
    {
      "entropy": 5.561968183517456,
      "epoch": 3.9394812680115274,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003524797728579875,
      "loss": 4.717,
      "mean_token_accuracy": 0.25578452199697493,
      "num_tokens": 94010500.0,
      "step": 41010
    },
    {
      "entropy": 5.358495616912842,
      "epoch": 3.9399615754082613,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003524475851963887,
      "loss": 4.6498,
      "mean_token_accuracy": 0.25632222294807433,
      "num_tokens": 94020848.0,
      "step": 41015
    },
    {
      "entropy": 5.5377342224121096,
      "epoch": 3.940441882804995,
      "grad_norm": 1.125,
      "learning_rate": 0.00035241539573668693,
      "loss": 4.7653,
      "mean_token_accuracy": 0.2464689016342163,
      "num_tokens": 94033154.0,
      "step": 41020
    },
    {
      "entropy": 5.551317405700684,
      "epoch": 3.940922190201729,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035238320447962947,
      "loss": 4.698,
      "mean_token_accuracy": 0.25080652087926864,
      "num_tokens": 94044131.0,
      "step": 41025
    },
    {
      "entropy": 5.533302259445191,
      "epoch": 3.941402497598463,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003523510114259638,
      "loss": 4.6782,
      "mean_token_accuracy": 0.25143401324748993,
      "num_tokens": 94055877.0,
      "step": 41030
    },
    {
      "entropy": 5.406872653961182,
      "epoch": 3.941882804995197,
      "grad_norm": 1.25,
      "learning_rate": 0.0003523188165764374,
      "loss": 4.6405,
      "mean_token_accuracy": 0.26190524101257323,
      "num_tokens": 94067341.0,
      "step": 41035
    },
    {
      "entropy": 5.447862339019776,
      "epoch": 3.9423631123919307,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035228661993179757,
      "loss": 4.6308,
      "mean_token_accuracy": 0.2555582135915756,
      "num_tokens": 94079618.0,
      "step": 41040
    },
    {
      "entropy": 5.544518089294433,
      "epoch": 3.9428434197886646,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000352254421492792,
      "loss": 4.7539,
      "mean_token_accuracy": 0.24601770639419557,
      "num_tokens": 94090607.0,
      "step": 41045
    },
    {
      "entropy": 5.509518146514893,
      "epoch": 3.9433237271853985,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035222222126016814,
      "loss": 4.7344,
      "mean_token_accuracy": 0.24548502564430236,
      "num_tokens": 94102078.0,
      "step": 41050
    },
    {
      "entropy": 5.440370845794678,
      "epoch": 3.943804034582133,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003521900192346736,
      "loss": 4.5699,
      "mean_token_accuracy": 0.26460869163274764,
      "num_tokens": 94113524.0,
      "step": 41055
    },
    {
      "entropy": 5.461919212341309,
      "epoch": 3.9442843419788662,
      "grad_norm": 1.28125,
      "learning_rate": 0.000352157815417056,
      "loss": 4.7112,
      "mean_token_accuracy": 0.2594931647181511,
      "num_tokens": 94125003.0,
      "step": 41060
    },
    {
      "entropy": 5.548984527587891,
      "epoch": 3.9447646493756006,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035212560980806305,
      "loss": 4.8663,
      "mean_token_accuracy": 0.24094715416431428,
      "num_tokens": 94136670.0,
      "step": 41065
    },
    {
      "entropy": 5.5390111923217775,
      "epoch": 3.9452449567723344,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003520934024084425,
      "loss": 4.7099,
      "mean_token_accuracy": 0.25325572341680525,
      "num_tokens": 94148541.0,
      "step": 41070
    },
    {
      "entropy": 5.545483112335205,
      "epoch": 3.9457252641690683,
      "grad_norm": 1.171875,
      "learning_rate": 0.000352061193218942,
      "loss": 4.7741,
      "mean_token_accuracy": 0.2540639042854309,
      "num_tokens": 94159936.0,
      "step": 41075
    },
    {
      "entropy": 5.438854789733886,
      "epoch": 3.946205571565802,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003520289822403095,
      "loss": 4.7041,
      "mean_token_accuracy": 0.2522335097193718,
      "num_tokens": 94171042.0,
      "step": 41080
    },
    {
      "entropy": 5.5194179058074955,
      "epoch": 3.946685878962536,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035199676947329274,
      "loss": 4.7767,
      "mean_token_accuracy": 0.2529826179146767,
      "num_tokens": 94181610.0,
      "step": 41085
    },
    {
      "entropy": 5.585697507858276,
      "epoch": 3.94716618635927,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035196455491863954,
      "loss": 4.7804,
      "mean_token_accuracy": 0.24593095928430558,
      "num_tokens": 94192684.0,
      "step": 41090
    },
    {
      "entropy": 5.553096342086792,
      "epoch": 3.947646493756004,
      "grad_norm": 1.140625,
      "learning_rate": 0.000351932338577098,
      "loss": 4.8021,
      "mean_token_accuracy": 0.24243473261594772,
      "num_tokens": 94204211.0,
      "step": 41095
    },
    {
      "entropy": 5.443163537979126,
      "epoch": 3.9481268011527377,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035190012044941605,
      "loss": 4.6661,
      "mean_token_accuracy": 0.25807429403066634,
      "num_tokens": 94216175.0,
      "step": 41100
    },
    {
      "entropy": 5.496305227279663,
      "epoch": 3.9486071085494716,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003518679005363415,
      "loss": 4.6548,
      "mean_token_accuracy": 0.25692663341760635,
      "num_tokens": 94227735.0,
      "step": 41105
    },
    {
      "entropy": 5.504522800445557,
      "epoch": 3.9490874159462055,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003518356788386226,
      "loss": 4.6706,
      "mean_token_accuracy": 0.26016792058944704,
      "num_tokens": 94238247.0,
      "step": 41110
    },
    {
      "entropy": 5.519880104064941,
      "epoch": 3.9495677233429394,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035180345535700745,
      "loss": 4.7683,
      "mean_token_accuracy": 0.24709718376398088,
      "num_tokens": 94249839.0,
      "step": 41115
    },
    {
      "entropy": 5.5559248447418215,
      "epoch": 3.9500480307396733,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003517712300922441,
      "loss": 4.7742,
      "mean_token_accuracy": 0.25452250987291336,
      "num_tokens": 94261794.0,
      "step": 41120
    },
    {
      "entropy": 5.515202045440674,
      "epoch": 3.950528338136407,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035173900304508074,
      "loss": 4.7582,
      "mean_token_accuracy": 0.24309491515159606,
      "num_tokens": 94276197.0,
      "step": 41125
    },
    {
      "entropy": 5.544722223281861,
      "epoch": 3.9510086455331415,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003517067742162656,
      "loss": 4.7436,
      "mean_token_accuracy": 0.2434785023331642,
      "num_tokens": 94288419.0,
      "step": 41130
    },
    {
      "entropy": 5.528930044174194,
      "epoch": 3.951488952929875,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000351674543606547,
      "loss": 4.7674,
      "mean_token_accuracy": 0.24719220250844956,
      "num_tokens": 94300521.0,
      "step": 41135
    },
    {
      "entropy": 5.391448450088501,
      "epoch": 3.9519692603266092,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003516423112166731,
      "loss": 4.5756,
      "mean_token_accuracy": 0.26304915100336074,
      "num_tokens": 94311349.0,
      "step": 41140
    },
    {
      "entropy": 5.4477903842926025,
      "epoch": 3.952449567723343,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035161007704739226,
      "loss": 4.629,
      "mean_token_accuracy": 0.25871192663908005,
      "num_tokens": 94322383.0,
      "step": 41145
    },
    {
      "entropy": 5.466660499572754,
      "epoch": 3.952929875120077,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000351577841099453,
      "loss": 4.7182,
      "mean_token_accuracy": 0.24501297026872634,
      "num_tokens": 94334807.0,
      "step": 41150
    },
    {
      "entropy": 5.467305612564087,
      "epoch": 3.953410182516811,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003515456033736036,
      "loss": 4.6274,
      "mean_token_accuracy": 0.25920387655496596,
      "num_tokens": 94345444.0,
      "step": 41155
    },
    {
      "entropy": 5.494179725646973,
      "epoch": 3.9538904899135447,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035151336387059275,
      "loss": 4.758,
      "mean_token_accuracy": 0.24698775857686997,
      "num_tokens": 94356802.0,
      "step": 41160
    },
    {
      "entropy": 5.480099678039551,
      "epoch": 3.9543707973102786,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003514811225911686,
      "loss": 4.7187,
      "mean_token_accuracy": 0.2590271562337875,
      "num_tokens": 94367694.0,
      "step": 41165
    },
    {
      "entropy": 5.494482374191284,
      "epoch": 3.9548511047070125,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035144887953607995,
      "loss": 4.7076,
      "mean_token_accuracy": 0.2597197204828262,
      "num_tokens": 94379714.0,
      "step": 41170
    },
    {
      "entropy": 5.559393644332886,
      "epoch": 3.9553314121037464,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003514166347060753,
      "loss": 4.7653,
      "mean_token_accuracy": 0.25578671544790266,
      "num_tokens": 94391405.0,
      "step": 41175
    },
    {
      "entropy": 5.432549285888672,
      "epoch": 3.9558117195004803,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003513843881019033,
      "loss": 4.5883,
      "mean_token_accuracy": 0.2664534240961075,
      "num_tokens": 94402785.0,
      "step": 41180
    },
    {
      "entropy": 5.461072063446045,
      "epoch": 3.956292026897214,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035135213972431267,
      "loss": 4.6923,
      "mean_token_accuracy": 0.2557574287056923,
      "num_tokens": 94414373.0,
      "step": 41185
    },
    {
      "entropy": 5.434358072280884,
      "epoch": 3.956772334293948,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035131988957405206,
      "loss": 4.6334,
      "mean_token_accuracy": 0.2553956791758537,
      "num_tokens": 94427285.0,
      "step": 41190
    },
    {
      "entropy": 5.506437206268311,
      "epoch": 3.957252641690682,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035128763765187026,
      "loss": 4.7213,
      "mean_token_accuracy": 0.25124952346086504,
      "num_tokens": 94439520.0,
      "step": 41195
    },
    {
      "entropy": 5.474166202545166,
      "epoch": 3.957732949087416,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000351255383958516,
      "loss": 4.6622,
      "mean_token_accuracy": 0.2578286424279213,
      "num_tokens": 94450707.0,
      "step": 41200
    },
    {
      "entropy": 5.479242563247681,
      "epoch": 3.95821325648415,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003512231284947382,
      "loss": 4.7056,
      "mean_token_accuracy": 0.25326761454343794,
      "num_tokens": 94461941.0,
      "step": 41205
    },
    {
      "entropy": 5.558971500396728,
      "epoch": 3.9586935638808836,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003511908712612857,
      "loss": 4.7396,
      "mean_token_accuracy": 0.25445913672447207,
      "num_tokens": 94471931.0,
      "step": 41210
    },
    {
      "entropy": 5.515032434463501,
      "epoch": 3.959173871277618,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003511586122589073,
      "loss": 4.7146,
      "mean_token_accuracy": 0.25163989067077636,
      "num_tokens": 94482692.0,
      "step": 41215
    },
    {
      "entropy": 5.4277578830719,
      "epoch": 3.9596541786743513,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003511263514883522,
      "loss": 4.6324,
      "mean_token_accuracy": 0.2597229197621346,
      "num_tokens": 94493650.0,
      "step": 41220
    },
    {
      "entropy": 5.409870195388794,
      "epoch": 3.9601344860710856,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003510940889503692,
      "loss": 4.5474,
      "mean_token_accuracy": 0.2705882415175438,
      "num_tokens": 94503878.0,
      "step": 41225
    },
    {
      "entropy": 5.4445545196533205,
      "epoch": 3.9606147934678195,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035106182464570736,
      "loss": 4.7051,
      "mean_token_accuracy": 0.25478485226631165,
      "num_tokens": 94516930.0,
      "step": 41230
    },
    {
      "entropy": 5.475064325332641,
      "epoch": 3.9610951008645534,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003510295585751159,
      "loss": 4.6922,
      "mean_token_accuracy": 0.25040285289287567,
      "num_tokens": 94527523.0,
      "step": 41235
    },
    {
      "entropy": 5.526189565658569,
      "epoch": 3.9615754082612873,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035099729073934376,
      "loss": 4.7098,
      "mean_token_accuracy": 0.2576648831367493,
      "num_tokens": 94537594.0,
      "step": 41240
    },
    {
      "entropy": 5.556706953048706,
      "epoch": 3.962055715658021,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003509650211391403,
      "loss": 4.7816,
      "mean_token_accuracy": 0.24524501860141754,
      "num_tokens": 94548293.0,
      "step": 41245
    },
    {
      "entropy": 5.494301509857178,
      "epoch": 3.962536023054755,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035093274977525456,
      "loss": 4.7227,
      "mean_token_accuracy": 0.25461623072624207,
      "num_tokens": 94559116.0,
      "step": 41250
    },
    {
      "entropy": 5.49155478477478,
      "epoch": 3.963016330451489,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003509004766484359,
      "loss": 4.7768,
      "mean_token_accuracy": 0.24442456513643265,
      "num_tokens": 94570208.0,
      "step": 41255
    },
    {
      "entropy": 5.466205739974976,
      "epoch": 3.963496637848223,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035086820175943344,
      "loss": 4.6536,
      "mean_token_accuracy": 0.26021797955036163,
      "num_tokens": 94580982.0,
      "step": 41260
    },
    {
      "entropy": 5.475544214248657,
      "epoch": 3.9639769452449567,
      "grad_norm": 1.25,
      "learning_rate": 0.0003508359251089967,
      "loss": 4.7154,
      "mean_token_accuracy": 0.2533408984541893,
      "num_tokens": 94593364.0,
      "step": 41265
    },
    {
      "entropy": 5.522725534439087,
      "epoch": 3.9644572526416906,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035080364669787495,
      "loss": 4.6748,
      "mean_token_accuracy": 0.2555069521069527,
      "num_tokens": 94605072.0,
      "step": 41270
    },
    {
      "entropy": 5.453592967987061,
      "epoch": 3.9649375600384245,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035077136652681757,
      "loss": 4.7892,
      "mean_token_accuracy": 0.24817353636026382,
      "num_tokens": 94617226.0,
      "step": 41275
    },
    {
      "entropy": 5.488528728485107,
      "epoch": 3.9654178674351583,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035073908459657413,
      "loss": 4.6817,
      "mean_token_accuracy": 0.2523385226726532,
      "num_tokens": 94629807.0,
      "step": 41280
    },
    {
      "entropy": 5.50472731590271,
      "epoch": 3.965898174831892,
      "grad_norm": 1.25,
      "learning_rate": 0.00035070680090789404,
      "loss": 4.6567,
      "mean_token_accuracy": 0.25468950122594836,
      "num_tokens": 94640866.0,
      "step": 41285
    },
    {
      "entropy": 5.461478900909424,
      "epoch": 3.9663784822286265,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003506745154615268,
      "loss": 4.6816,
      "mean_token_accuracy": 0.24654823839664458,
      "num_tokens": 94651872.0,
      "step": 41290
    },
    {
      "entropy": 5.500684499740601,
      "epoch": 3.96685878962536,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035064222825822197,
      "loss": 4.7627,
      "mean_token_accuracy": 0.2496602490544319,
      "num_tokens": 94664890.0,
      "step": 41295
    },
    {
      "entropy": 5.380647993087768,
      "epoch": 3.9673390970220943,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035060993929872923,
      "loss": 4.5333,
      "mean_token_accuracy": 0.2643838420510292,
      "num_tokens": 94677447.0,
      "step": 41300
    },
    {
      "entropy": 5.517150783538819,
      "epoch": 3.967819404418828,
      "grad_norm": 1.125,
      "learning_rate": 0.0003505776485837983,
      "loss": 4.7603,
      "mean_token_accuracy": 0.2489765003323555,
      "num_tokens": 94690432.0,
      "step": 41305
    },
    {
      "entropy": 5.4529320240020756,
      "epoch": 3.968299711815562,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003505453561141787,
      "loss": 4.649,
      "mean_token_accuracy": 0.25323302000761033,
      "num_tokens": 94703357.0,
      "step": 41310
    },
    {
      "entropy": 5.4593874454498295,
      "epoch": 3.968780019212296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003505130618906203,
      "loss": 4.6455,
      "mean_token_accuracy": 0.25993207544088365,
      "num_tokens": 94715276.0,
      "step": 41315
    },
    {
      "entropy": 5.509416675567627,
      "epoch": 3.96926032660903,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003504807659138728,
      "loss": 4.6881,
      "mean_token_accuracy": 0.25428789108991623,
      "num_tokens": 94725692.0,
      "step": 41320
    },
    {
      "entropy": 5.566633176803589,
      "epoch": 3.9697406340057637,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003504484681846861,
      "loss": 4.7992,
      "mean_token_accuracy": 0.2506256952881813,
      "num_tokens": 94737211.0,
      "step": 41325
    },
    {
      "entropy": 5.515867376327515,
      "epoch": 3.9702209414024976,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035041616870380997,
      "loss": 4.7582,
      "mean_token_accuracy": 0.24803271144628525,
      "num_tokens": 94749140.0,
      "step": 41330
    },
    {
      "entropy": 5.559043216705322,
      "epoch": 3.9707012487992315,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035038386747199437,
      "loss": 4.6777,
      "mean_token_accuracy": 0.2508071959018707,
      "num_tokens": 94759409.0,
      "step": 41335
    },
    {
      "entropy": 5.4981451511383055,
      "epoch": 3.9711815561959654,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003503515644899892,
      "loss": 4.6959,
      "mean_token_accuracy": 0.24782017022371292,
      "num_tokens": 94770267.0,
      "step": 41340
    },
    {
      "entropy": 5.492672967910766,
      "epoch": 3.9716618635926992,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003503192597585444,
      "loss": 4.6946,
      "mean_token_accuracy": 0.25269081741571425,
      "num_tokens": 94780479.0,
      "step": 41345
    },
    {
      "entropy": 5.422207975387574,
      "epoch": 3.972142170989433,
      "grad_norm": 1.25,
      "learning_rate": 0.0003502869532784101,
      "loss": 4.6257,
      "mean_token_accuracy": 0.2606149062514305,
      "num_tokens": 94791192.0,
      "step": 41350
    },
    {
      "entropy": 5.498775243759155,
      "epoch": 3.972622478386167,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035025464505033626,
      "loss": 4.6777,
      "mean_token_accuracy": 0.2501717388629913,
      "num_tokens": 94802090.0,
      "step": 41355
    },
    {
      "entropy": 5.3977799892425535,
      "epoch": 3.973102785782901,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000350222335075073,
      "loss": 4.6246,
      "mean_token_accuracy": 0.2567461311817169,
      "num_tokens": 94814246.0,
      "step": 41360
    },
    {
      "entropy": 5.497043657302856,
      "epoch": 3.973583093179635,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035019002335337057,
      "loss": 4.6938,
      "mean_token_accuracy": 0.25703668892383574,
      "num_tokens": 94825754.0,
      "step": 41365
    },
    {
      "entropy": 5.498170518875122,
      "epoch": 3.9740634005763686,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003501577098859791,
      "loss": 4.7896,
      "mean_token_accuracy": 0.2468671455979347,
      "num_tokens": 94837790.0,
      "step": 41370
    },
    {
      "entropy": 5.4627762794494625,
      "epoch": 3.974543707973103,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035012539467364865,
      "loss": 4.6982,
      "mean_token_accuracy": 0.24557386934757233,
      "num_tokens": 94848694.0,
      "step": 41375
    },
    {
      "entropy": 5.456984901428223,
      "epoch": 3.975024015369837,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003500930777171296,
      "loss": 4.7142,
      "mean_token_accuracy": 0.2528990522027016,
      "num_tokens": 94861602.0,
      "step": 41380
    },
    {
      "entropy": 5.5325182437896725,
      "epoch": 3.9755043227665707,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003500607590171724,
      "loss": 4.7151,
      "mean_token_accuracy": 0.254108390212059,
      "num_tokens": 94872857.0,
      "step": 41385
    },
    {
      "entropy": 5.604924631118775,
      "epoch": 3.9759846301633046,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003500284385745272,
      "loss": 4.8188,
      "mean_token_accuracy": 0.24908133149147033,
      "num_tokens": 94884716.0,
      "step": 41390
    },
    {
      "entropy": 5.4834857940673825,
      "epoch": 3.9764649375600385,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003499961163899444,
      "loss": 4.6838,
      "mean_token_accuracy": 0.257437589764595,
      "num_tokens": 94897721.0,
      "step": 41395
    },
    {
      "entropy": 5.557518672943115,
      "epoch": 3.9769452449567724,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034996379246417453,
      "loss": 4.7226,
      "mean_token_accuracy": 0.2523579359054565,
      "num_tokens": 94908099.0,
      "step": 41400
    },
    {
      "entropy": 5.5267900943756105,
      "epoch": 3.9774255523535063,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034993146679796795,
      "loss": 4.7126,
      "mean_token_accuracy": 0.25212245583534243,
      "num_tokens": 94919944.0,
      "step": 41405
    },
    {
      "entropy": 5.506744432449341,
      "epoch": 3.97790585975024,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003498991393920752,
      "loss": 4.7236,
      "mean_token_accuracy": 0.2561176627874374,
      "num_tokens": 94932104.0,
      "step": 41410
    },
    {
      "entropy": 5.502374839782715,
      "epoch": 3.978386167146974,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034986681024724696,
      "loss": 4.7816,
      "mean_token_accuracy": 0.24877255409955978,
      "num_tokens": 94943869.0,
      "step": 41415
    },
    {
      "entropy": 5.548053646087647,
      "epoch": 3.978866474543708,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034983447936423357,
      "loss": 4.7678,
      "mean_token_accuracy": 0.24750000685453416,
      "num_tokens": 94955812.0,
      "step": 41420
    },
    {
      "entropy": 5.443542242050171,
      "epoch": 3.979346781940442,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003498021467437858,
      "loss": 4.6537,
      "mean_token_accuracy": 0.2545794203877449,
      "num_tokens": 94967320.0,
      "step": 41425
    },
    {
      "entropy": 5.509357881546021,
      "epoch": 3.9798270893371757,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003497698123866544,
      "loss": 4.7121,
      "mean_token_accuracy": 0.25804625302553175,
      "num_tokens": 94978249.0,
      "step": 41430
    },
    {
      "entropy": 5.516163921356201,
      "epoch": 3.9803073967339095,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034973747629358983,
      "loss": 4.7351,
      "mean_token_accuracy": 0.2563894703984261,
      "num_tokens": 94990776.0,
      "step": 41435
    },
    {
      "entropy": 5.548998069763184,
      "epoch": 3.980787704130644,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034970513846534305,
      "loss": 4.7367,
      "mean_token_accuracy": 0.25457852333784103,
      "num_tokens": 95001361.0,
      "step": 41440
    },
    {
      "entropy": 5.523902082443238,
      "epoch": 3.9812680115273773,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003496727989026648,
      "loss": 4.7708,
      "mean_token_accuracy": 0.2513301014900208,
      "num_tokens": 95013748.0,
      "step": 41445
    },
    {
      "entropy": 5.428549671173096,
      "epoch": 3.9817483189241116,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034964045760630584,
      "loss": 4.6567,
      "mean_token_accuracy": 0.25513907670974734,
      "num_tokens": 95025700.0,
      "step": 41450
    },
    {
      "entropy": 5.558842182159424,
      "epoch": 3.982228626320845,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034960811457701715,
      "loss": 4.7646,
      "mean_token_accuracy": 0.24897675216197968,
      "num_tokens": 95037653.0,
      "step": 41455
    },
    {
      "entropy": 5.437463474273682,
      "epoch": 3.9827089337175794,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034957576981554955,
      "loss": 4.6645,
      "mean_token_accuracy": 0.25923409312963486,
      "num_tokens": 95049859.0,
      "step": 41460
    },
    {
      "entropy": 5.38766393661499,
      "epoch": 3.9831892411143133,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034954342332265407,
      "loss": 4.64,
      "mean_token_accuracy": 0.254102298617363,
      "num_tokens": 95062862.0,
      "step": 41465
    },
    {
      "entropy": 5.486344528198242,
      "epoch": 3.983669548511047,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034951107509908157,
      "loss": 4.7166,
      "mean_token_accuracy": 0.24639478474855422,
      "num_tokens": 95073555.0,
      "step": 41470
    },
    {
      "entropy": 5.506278085708618,
      "epoch": 3.984149855907781,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003494787251455832,
      "loss": 4.6586,
      "mean_token_accuracy": 0.25958363264799117,
      "num_tokens": 95086165.0,
      "step": 41475
    },
    {
      "entropy": 5.5201374053955075,
      "epoch": 3.984630163304515,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034944637346291,
      "loss": 4.7565,
      "mean_token_accuracy": 0.25018196552991867,
      "num_tokens": 95098475.0,
      "step": 41480
    },
    {
      "entropy": 5.514064645767212,
      "epoch": 3.985110470701249,
      "grad_norm": 1.140625,
      "learning_rate": 0.000349414020051813,
      "loss": 4.7608,
      "mean_token_accuracy": 0.245589879155159,
      "num_tokens": 95109972.0,
      "step": 41485
    },
    {
      "entropy": 5.503448534011841,
      "epoch": 3.9855907780979827,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003493816649130435,
      "loss": 4.746,
      "mean_token_accuracy": 0.25326373875141145,
      "num_tokens": 95122043.0,
      "step": 41490
    },
    {
      "entropy": 5.484879446029663,
      "epoch": 3.9860710854947166,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003493493080473526,
      "loss": 4.6291,
      "mean_token_accuracy": 0.2591151297092438,
      "num_tokens": 95133685.0,
      "step": 41495
    },
    {
      "entropy": 5.492282915115356,
      "epoch": 3.9865513928914504,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034931694945549146,
      "loss": 4.7,
      "mean_token_accuracy": 0.25468470305204394,
      "num_tokens": 95146018.0,
      "step": 41500
    },
    {
      "entropy": 5.484355926513672,
      "epoch": 3.9870317002881843,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003492845891382115,
      "loss": 4.7439,
      "mean_token_accuracy": 0.24532463252544404,
      "num_tokens": 95158262.0,
      "step": 41505
    },
    {
      "entropy": 5.485703468322754,
      "epoch": 3.987512007684918,
      "grad_norm": 1.265625,
      "learning_rate": 0.000349252227096264,
      "loss": 4.7047,
      "mean_token_accuracy": 0.24982773810625075,
      "num_tokens": 95170162.0,
      "step": 41510
    },
    {
      "entropy": 5.465477657318115,
      "epoch": 3.9879923150816525,
      "grad_norm": 1.125,
      "learning_rate": 0.0003492198633304002,
      "loss": 4.6997,
      "mean_token_accuracy": 0.258037631213665,
      "num_tokens": 95182574.0,
      "step": 41515
    },
    {
      "entropy": 5.55047197341919,
      "epoch": 3.988472622478386,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003491874978413716,
      "loss": 4.771,
      "mean_token_accuracy": 0.24546557515859604,
      "num_tokens": 95194527.0,
      "step": 41520
    },
    {
      "entropy": 5.455594205856324,
      "epoch": 3.9889529298751203,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003491551306299296,
      "loss": 4.66,
      "mean_token_accuracy": 0.2527119666337967,
      "num_tokens": 95207842.0,
      "step": 41525
    },
    {
      "entropy": 5.531377458572388,
      "epoch": 3.9894332372718537,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034912276169682556,
      "loss": 4.7467,
      "mean_token_accuracy": 0.2557212769985199,
      "num_tokens": 95219694.0,
      "step": 41530
    },
    {
      "entropy": 5.43764476776123,
      "epoch": 3.989913544668588,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034909039104281123,
      "loss": 4.6625,
      "mean_token_accuracy": 0.2618744492530823,
      "num_tokens": 95231397.0,
      "step": 41535
    },
    {
      "entropy": 5.435226249694824,
      "epoch": 3.990393852065322,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003490580186686379,
      "loss": 4.6658,
      "mean_token_accuracy": 0.259309259057045,
      "num_tokens": 95243575.0,
      "step": 41540
    },
    {
      "entropy": 5.551360082626343,
      "epoch": 3.990874159462056,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003490256445750573,
      "loss": 4.8162,
      "mean_token_accuracy": 0.2469842478632927,
      "num_tokens": 95255866.0,
      "step": 41545
    },
    {
      "entropy": 5.544286012649536,
      "epoch": 3.9913544668587897,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003489932687628211,
      "loss": 4.7162,
      "mean_token_accuracy": 0.25218199342489245,
      "num_tokens": 95266812.0,
      "step": 41550
    },
    {
      "entropy": 5.484330129623413,
      "epoch": 3.9918347742555236,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034896089123268096,
      "loss": 4.7352,
      "mean_token_accuracy": 0.2506725713610649,
      "num_tokens": 95278482.0,
      "step": 41555
    },
    {
      "entropy": 5.426892805099487,
      "epoch": 3.9923150816522575,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003489285119853884,
      "loss": 4.7458,
      "mean_token_accuracy": 0.2546478509902954,
      "num_tokens": 95289374.0,
      "step": 41560
    },
    {
      "entropy": 5.482873868942261,
      "epoch": 3.9927953890489913,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003488961310216954,
      "loss": 4.6659,
      "mean_token_accuracy": 0.25531949251890185,
      "num_tokens": 95301913.0,
      "step": 41565
    },
    {
      "entropy": 5.60161600112915,
      "epoch": 3.993275696445725,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003488637483423537,
      "loss": 4.8308,
      "mean_token_accuracy": 0.2436303436756134,
      "num_tokens": 95314560.0,
      "step": 41570
    },
    {
      "entropy": 5.565365219116211,
      "epoch": 3.993756003842459,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034883136394811505,
      "loss": 4.7537,
      "mean_token_accuracy": 0.24786891490221025,
      "num_tokens": 95326956.0,
      "step": 41575
    },
    {
      "entropy": 5.56496262550354,
      "epoch": 3.994236311239193,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034879897783973143,
      "loss": 4.7086,
      "mean_token_accuracy": 0.25369215160608294,
      "num_tokens": 95338405.0,
      "step": 41580
    },
    {
      "entropy": 5.519147109985352,
      "epoch": 3.994716618635927,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003487665900179545,
      "loss": 4.7498,
      "mean_token_accuracy": 0.24504537731409073,
      "num_tokens": 95349269.0,
      "step": 41585
    },
    {
      "entropy": 5.595450019836425,
      "epoch": 3.9951969260326607,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003487342004835366,
      "loss": 4.7762,
      "mean_token_accuracy": 0.24560199826955795,
      "num_tokens": 95359940.0,
      "step": 41590
    },
    {
      "entropy": 5.529043865203858,
      "epoch": 3.9956772334293946,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003487018092372294,
      "loss": 4.7786,
      "mean_token_accuracy": 0.25018419325351715,
      "num_tokens": 95370830.0,
      "step": 41595
    },
    {
      "entropy": 5.548371315002441,
      "epoch": 3.996157540826129,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000348669416279785,
      "loss": 4.7973,
      "mean_token_accuracy": 0.2513919621706009,
      "num_tokens": 95382805.0,
      "step": 41600
    },
    {
      "entropy": 5.428062438964844,
      "epoch": 3.9966378482228624,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034863702161195566,
      "loss": 4.5941,
      "mean_token_accuracy": 0.2660010129213333,
      "num_tokens": 95394380.0,
      "step": 41605
    },
    {
      "entropy": 5.40377779006958,
      "epoch": 3.9971181556195967,
      "grad_norm": 1.125,
      "learning_rate": 0.0003486046252344933,
      "loss": 4.5434,
      "mean_token_accuracy": 0.26172801703214643,
      "num_tokens": 95406240.0,
      "step": 41610
    },
    {
      "entropy": 5.568137741088867,
      "epoch": 3.9975984630163306,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034857222714814996,
      "loss": 4.7638,
      "mean_token_accuracy": 0.24481877237558364,
      "num_tokens": 95417021.0,
      "step": 41615
    },
    {
      "entropy": 5.515288496017456,
      "epoch": 3.9980787704130645,
      "grad_norm": 1.203125,
      "learning_rate": 0.000348539827353678,
      "loss": 4.808,
      "mean_token_accuracy": 0.24185216277837754,
      "num_tokens": 95429315.0,
      "step": 41620
    },
    {
      "entropy": 5.523983478546143,
      "epoch": 3.9985590778097984,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003485074258518298,
      "loss": 4.7302,
      "mean_token_accuracy": 0.2585476666688919,
      "num_tokens": 95441149.0,
      "step": 41625
    },
    {
      "entropy": 5.552683019638062,
      "epoch": 3.9990393852065322,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003484750226433573,
      "loss": 4.7967,
      "mean_token_accuracy": 0.24947046786546706,
      "num_tokens": 95452753.0,
      "step": 41630
    },
    {
      "entropy": 5.363332080841064,
      "epoch": 3.999519692603266,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034844261772901293,
      "loss": 4.6339,
      "mean_token_accuracy": 0.26227917075157164,
      "num_tokens": 95464901.0,
      "step": 41635
    },
    {
      "entropy": 5.561138725280761,
      "epoch": 4.0,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003484102111095492,
      "loss": 4.7164,
      "mean_token_accuracy": 0.25602803826332093,
      "num_tokens": 95474144.0,
      "step": 41640
    },
    {
      "entropy": 5.547472715377808,
      "epoch": 4.000480307396734,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034837780278571814,
      "loss": 4.7129,
      "mean_token_accuracy": 0.25026958137750627,
      "num_tokens": 95483973.0,
      "step": 41645
    },
    {
      "entropy": 5.508199024200439,
      "epoch": 4.000960614793468,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003483453927582725,
      "loss": 4.7347,
      "mean_token_accuracy": 0.25161905586719513,
      "num_tokens": 95495989.0,
      "step": 41650
    },
    {
      "entropy": 5.434222745895386,
      "epoch": 4.001440922190202,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003483129810279647,
      "loss": 4.6092,
      "mean_token_accuracy": 0.259767447412014,
      "num_tokens": 95507286.0,
      "step": 41655
    },
    {
      "entropy": 5.543717002868652,
      "epoch": 4.0019212295869355,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000348280567595547,
      "loss": 4.7337,
      "mean_token_accuracy": 0.25097402930259705,
      "num_tokens": 95519499.0,
      "step": 41660
    },
    {
      "entropy": 5.48430552482605,
      "epoch": 4.00240153698367,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034824815246177227,
      "loss": 4.6541,
      "mean_token_accuracy": 0.2543994188308716,
      "num_tokens": 95532134.0,
      "step": 41665
    },
    {
      "entropy": 5.493575620651245,
      "epoch": 4.002881844380403,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003482157356273929,
      "loss": 4.694,
      "mean_token_accuracy": 0.25688207894563675,
      "num_tokens": 95542157.0,
      "step": 41670
    },
    {
      "entropy": 5.510220003128052,
      "epoch": 4.003362151777138,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003481833170931615,
      "loss": 4.63,
      "mean_token_accuracy": 0.26091318577528,
      "num_tokens": 95553345.0,
      "step": 41675
    },
    {
      "entropy": 5.455819225311279,
      "epoch": 4.003842459173871,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034815089685983085,
      "loss": 4.6276,
      "mean_token_accuracy": 0.2571519106626511,
      "num_tokens": 95564795.0,
      "step": 41680
    },
    {
      "entropy": 5.4980553150177,
      "epoch": 4.004322766570605,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003481184749281536,
      "loss": 4.7286,
      "mean_token_accuracy": 0.2538881704211235,
      "num_tokens": 95576777.0,
      "step": 41685
    },
    {
      "entropy": 5.500484275817871,
      "epoch": 4.004803073967339,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034808605129888246,
      "loss": 4.712,
      "mean_token_accuracy": 0.25739518254995347,
      "num_tokens": 95588483.0,
      "step": 41690
    },
    {
      "entropy": 5.465500211715698,
      "epoch": 4.005283381364073,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034805362597277017,
      "loss": 4.5665,
      "mean_token_accuracy": 0.2696992203593254,
      "num_tokens": 95600300.0,
      "step": 41695
    },
    {
      "entropy": 5.4691853523254395,
      "epoch": 4.005763688760807,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034802119895056973,
      "loss": 4.6625,
      "mean_token_accuracy": 0.24954085648059846,
      "num_tokens": 95611596.0,
      "step": 41700
    },
    {
      "entropy": 5.4556553840637205,
      "epoch": 4.006243996157541,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034798877023303385,
      "loss": 4.6671,
      "mean_token_accuracy": 0.2543606385588646,
      "num_tokens": 95622813.0,
      "step": 41705
    },
    {
      "entropy": 5.4095738410949705,
      "epoch": 4.006724303554274,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034795633982091534,
      "loss": 4.6017,
      "mean_token_accuracy": 0.2615666136145592,
      "num_tokens": 95635862.0,
      "step": 41710
    },
    {
      "entropy": 5.589762020111084,
      "epoch": 4.007204610951009,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034792390771496735,
      "loss": 4.7763,
      "mean_token_accuracy": 0.24974295496940613,
      "num_tokens": 95647396.0,
      "step": 41715
    },
    {
      "entropy": 5.4964500904083256,
      "epoch": 4.007684918347743,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034789147391594275,
      "loss": 4.6943,
      "mean_token_accuracy": 0.257252162694931,
      "num_tokens": 95657882.0,
      "step": 41720
    },
    {
      "entropy": 5.467151737213134,
      "epoch": 4.008165225744476,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003478590384245945,
      "loss": 4.5822,
      "mean_token_accuracy": 0.26394361853599546,
      "num_tokens": 95669915.0,
      "step": 41725
    },
    {
      "entropy": 5.4489734172821045,
      "epoch": 4.008645533141211,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034782660124167583,
      "loss": 4.5676,
      "mean_token_accuracy": 0.259989969432354,
      "num_tokens": 95680660.0,
      "step": 41730
    },
    {
      "entropy": 5.482440853118897,
      "epoch": 4.009125840537944,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003477941623679397,
      "loss": 4.7127,
      "mean_token_accuracy": 0.24588490277528763,
      "num_tokens": 95692604.0,
      "step": 41735
    },
    {
      "entropy": 5.440522241592407,
      "epoch": 4.0096061479346785,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003477617218041393,
      "loss": 4.6276,
      "mean_token_accuracy": 0.25803819596767424,
      "num_tokens": 95703761.0,
      "step": 41740
    },
    {
      "entropy": 5.3917155265808105,
      "epoch": 4.010086455331412,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003477292795510277,
      "loss": 4.5414,
      "mean_token_accuracy": 0.2688129603862762,
      "num_tokens": 95715776.0,
      "step": 41745
    },
    {
      "entropy": 5.502365779876709,
      "epoch": 4.010566762728146,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003476968356093582,
      "loss": 4.7548,
      "mean_token_accuracy": 0.25010496377944946,
      "num_tokens": 95727812.0,
      "step": 41750
    },
    {
      "entropy": 5.4864836692810055,
      "epoch": 4.01104707012488,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003476643899798841,
      "loss": 4.6622,
      "mean_token_accuracy": 0.25816797465085983,
      "num_tokens": 95740453.0,
      "step": 41755
    },
    {
      "entropy": 5.4328008651733395,
      "epoch": 4.011527377521614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034763194266335855,
      "loss": 4.5693,
      "mean_token_accuracy": 0.272362557053566,
      "num_tokens": 95753436.0,
      "step": 41760
    },
    {
      "entropy": 5.53506908416748,
      "epoch": 4.0120076849183475,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003475994936605351,
      "loss": 4.7278,
      "mean_token_accuracy": 0.24991032481193542,
      "num_tokens": 95763936.0,
      "step": 41765
    },
    {
      "entropy": 5.498421764373779,
      "epoch": 4.012487992315082,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034756704297216686,
      "loss": 4.696,
      "mean_token_accuracy": 0.25391274243593215,
      "num_tokens": 95775535.0,
      "step": 41770
    },
    {
      "entropy": 5.48714804649353,
      "epoch": 4.012968299711815,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034753459059900745,
      "loss": 4.7302,
      "mean_token_accuracy": 0.2535557091236115,
      "num_tokens": 95786814.0,
      "step": 41775
    },
    {
      "entropy": 5.559858417510986,
      "epoch": 4.01344860710855,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034750213654181016,
      "loss": 4.7259,
      "mean_token_accuracy": 0.261352975666523,
      "num_tokens": 95798141.0,
      "step": 41780
    },
    {
      "entropy": 5.500548791885376,
      "epoch": 4.013928914505283,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034746968080132855,
      "loss": 4.6996,
      "mean_token_accuracy": 0.2611154466867447,
      "num_tokens": 95811244.0,
      "step": 41785
    },
    {
      "entropy": 5.484286689758301,
      "epoch": 4.014409221902017,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003474372233783162,
      "loss": 4.6023,
      "mean_token_accuracy": 0.26459618508815763,
      "num_tokens": 95821824.0,
      "step": 41790
    },
    {
      "entropy": 5.427321529388427,
      "epoch": 4.014889529298751,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034740476427352657,
      "loss": 4.563,
      "mean_token_accuracy": 0.2657317191362381,
      "num_tokens": 95833122.0,
      "step": 41795
    },
    {
      "entropy": 5.385173273086548,
      "epoch": 4.015369836695485,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003473723034877133,
      "loss": 4.6172,
      "mean_token_accuracy": 0.26238625645637514,
      "num_tokens": 95844325.0,
      "step": 41800
    },
    {
      "entropy": 5.48657603263855,
      "epoch": 4.015850144092219,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034733984102163005,
      "loss": 4.7265,
      "mean_token_accuracy": 0.2481127440929413,
      "num_tokens": 95855788.0,
      "step": 41805
    },
    {
      "entropy": 5.600787925720215,
      "epoch": 4.016330451488953,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003473073768760306,
      "loss": 4.7089,
      "mean_token_accuracy": 0.2524610623717308,
      "num_tokens": 95867387.0,
      "step": 41810
    },
    {
      "entropy": 5.577070331573486,
      "epoch": 4.016810758885687,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034727491105166847,
      "loss": 4.7186,
      "mean_token_accuracy": 0.2522706910967827,
      "num_tokens": 95878258.0,
      "step": 41815
    },
    {
      "entropy": 5.5447791576385494,
      "epoch": 4.017291066282421,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003472424435492975,
      "loss": 4.677,
      "mean_token_accuracy": 0.25619468092918396,
      "num_tokens": 95889312.0,
      "step": 41820
    },
    {
      "entropy": 5.5722503662109375,
      "epoch": 4.017771373679155,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003472099743696716,
      "loss": 4.8076,
      "mean_token_accuracy": 0.24843946546316148,
      "num_tokens": 95901930.0,
      "step": 41825
    },
    {
      "entropy": 5.430638837814331,
      "epoch": 4.018251681075888,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003471775035135444,
      "loss": 4.5889,
      "mean_token_accuracy": 0.25722510665655135,
      "num_tokens": 95913877.0,
      "step": 41830
    },
    {
      "entropy": 5.460920143127441,
      "epoch": 4.018731988472623,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034714503098166996,
      "loss": 4.6859,
      "mean_token_accuracy": 0.25670529454946517,
      "num_tokens": 95925259.0,
      "step": 41835
    },
    {
      "entropy": 5.49409327507019,
      "epoch": 4.019212295869356,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034711255677480216,
      "loss": 4.6172,
      "mean_token_accuracy": 0.26218634843826294,
      "num_tokens": 95936612.0,
      "step": 41840
    },
    {
      "entropy": 5.4348245620727536,
      "epoch": 4.0196926032660905,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003470800808936948,
      "loss": 4.6625,
      "mean_token_accuracy": 0.25477389097213743,
      "num_tokens": 95948505.0,
      "step": 41845
    },
    {
      "entropy": 5.402983665466309,
      "epoch": 4.020172910662824,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034704760333910205,
      "loss": 4.5681,
      "mean_token_accuracy": 0.26907903254032134,
      "num_tokens": 95959767.0,
      "step": 41850
    },
    {
      "entropy": 5.58946270942688,
      "epoch": 4.020653218059558,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003470151241117779,
      "loss": 4.8171,
      "mean_token_accuracy": 0.25058099776506426,
      "num_tokens": 95971470.0,
      "step": 41855
    },
    {
      "entropy": 5.508378696441651,
      "epoch": 4.021133525456292,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003469826432124764,
      "loss": 4.6297,
      "mean_token_accuracy": 0.2589709609746933,
      "num_tokens": 95983610.0,
      "step": 41860
    },
    {
      "entropy": 5.497771167755127,
      "epoch": 4.021613832853026,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003469501606419517,
      "loss": 4.6667,
      "mean_token_accuracy": 0.2640172645449638,
      "num_tokens": 95995199.0,
      "step": 41865
    },
    {
      "entropy": 5.546493625640869,
      "epoch": 4.022094140249759,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003469176764009579,
      "loss": 4.7055,
      "mean_token_accuracy": 0.2511676952242851,
      "num_tokens": 96005881.0,
      "step": 41870
    },
    {
      "entropy": 5.52286319732666,
      "epoch": 4.022574447646494,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034688519049024924,
      "loss": 4.6212,
      "mean_token_accuracy": 0.2578404664993286,
      "num_tokens": 96016973.0,
      "step": 41875
    },
    {
      "entropy": 5.508524799346924,
      "epoch": 4.023054755043228,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034685270291057987,
      "loss": 4.739,
      "mean_token_accuracy": 0.23848807364702224,
      "num_tokens": 96029500.0,
      "step": 41880
    },
    {
      "entropy": 5.449189281463623,
      "epoch": 4.0235350624399615,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034682021366270403,
      "loss": 4.5841,
      "mean_token_accuracy": 0.2653824657201767,
      "num_tokens": 96042239.0,
      "step": 41885
    },
    {
      "entropy": 5.627670812606811,
      "epoch": 4.024015369836696,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034678772274737626,
      "loss": 4.7646,
      "mean_token_accuracy": 0.24189835041761398,
      "num_tokens": 96053385.0,
      "step": 41890
    },
    {
      "entropy": 5.479885482788086,
      "epoch": 4.024495677233429,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034675523016535067,
      "loss": 4.6613,
      "mean_token_accuracy": 0.2594537943601608,
      "num_tokens": 96065600.0,
      "step": 41895
    },
    {
      "entropy": 5.464064693450927,
      "epoch": 4.024975984630164,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034672273591738155,
      "loss": 4.697,
      "mean_token_accuracy": 0.25005239993333817,
      "num_tokens": 96076591.0,
      "step": 41900
    },
    {
      "entropy": 5.417671012878418,
      "epoch": 4.025456292026897,
      "grad_norm": 2.359375,
      "learning_rate": 0.0003466902400042236,
      "loss": 4.613,
      "mean_token_accuracy": 0.2691223591566086,
      "num_tokens": 96088643.0,
      "step": 41905
    },
    {
      "entropy": 5.511043405532837,
      "epoch": 4.025936599423631,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003466577424266312,
      "loss": 4.6824,
      "mean_token_accuracy": 0.2559850960969925,
      "num_tokens": 96101592.0,
      "step": 41910
    },
    {
      "entropy": 5.444753980636596,
      "epoch": 4.026416906820365,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034662524318535877,
      "loss": 4.6028,
      "mean_token_accuracy": 0.2617674067616463,
      "num_tokens": 96113434.0,
      "step": 41915
    },
    {
      "entropy": 5.51462459564209,
      "epoch": 4.026897214217099,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034659274228116086,
      "loss": 4.685,
      "mean_token_accuracy": 0.25746078938245776,
      "num_tokens": 96123576.0,
      "step": 41920
    },
    {
      "entropy": 5.492161893844605,
      "epoch": 4.027377521613833,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034656023971479206,
      "loss": 4.7781,
      "mean_token_accuracy": 0.2459753468632698,
      "num_tokens": 96135415.0,
      "step": 41925
    },
    {
      "entropy": 5.553878927230835,
      "epoch": 4.027857829010567,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003465277354870069,
      "loss": 4.7389,
      "mean_token_accuracy": 0.25373526513576505,
      "num_tokens": 96146740.0,
      "step": 41930
    },
    {
      "entropy": 5.412007427215576,
      "epoch": 4.0283381364073,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034649522959856027,
      "loss": 4.5709,
      "mean_token_accuracy": 0.2689517945051193,
      "num_tokens": 96158650.0,
      "step": 41935
    },
    {
      "entropy": 5.524919843673706,
      "epoch": 4.028818443804035,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034646272205020664,
      "loss": 4.725,
      "mean_token_accuracy": 0.25609623789787295,
      "num_tokens": 96169394.0,
      "step": 41940
    },
    {
      "entropy": 5.499782562255859,
      "epoch": 4.029298751200768,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003464302128427008,
      "loss": 4.7182,
      "mean_token_accuracy": 0.2583547502756119,
      "num_tokens": 96180791.0,
      "step": 41945
    },
    {
      "entropy": 5.377293872833252,
      "epoch": 4.029779058597502,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003463977019767975,
      "loss": 4.5266,
      "mean_token_accuracy": 0.27166308760643004,
      "num_tokens": 96191705.0,
      "step": 41950
    },
    {
      "entropy": 5.526702642440796,
      "epoch": 4.030259365994237,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003463651894532516,
      "loss": 4.8123,
      "mean_token_accuracy": 0.24675203263759612,
      "num_tokens": 96203742.0,
      "step": 41955
    },
    {
      "entropy": 5.482724285125732,
      "epoch": 4.03073967339097,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000346332675272818,
      "loss": 4.6557,
      "mean_token_accuracy": 0.26603133976459503,
      "num_tokens": 96214991.0,
      "step": 41960
    },
    {
      "entropy": 5.431916761398315,
      "epoch": 4.0312199807877045,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003463001594362513,
      "loss": 4.6165,
      "mean_token_accuracy": 0.26204258054494856,
      "num_tokens": 96227122.0,
      "step": 41965
    },
    {
      "entropy": 5.464972019195557,
      "epoch": 4.031700288184438,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003462676419443067,
      "loss": 4.6029,
      "mean_token_accuracy": 0.26340333223342893,
      "num_tokens": 96237508.0,
      "step": 41970
    },
    {
      "entropy": 5.4406311988830565,
      "epoch": 4.032180595581172,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003462351227977391,
      "loss": 4.7183,
      "mean_token_accuracy": 0.2551876872777939,
      "num_tokens": 96250375.0,
      "step": 41975
    },
    {
      "entropy": 5.50390853881836,
      "epoch": 4.032660902977906,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034620260199730345,
      "loss": 4.6925,
      "mean_token_accuracy": 0.25703396648168564,
      "num_tokens": 96261199.0,
      "step": 41980
    },
    {
      "entropy": 5.529475450515747,
      "epoch": 4.03314121037464,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034617007954375486,
      "loss": 4.7099,
      "mean_token_accuracy": 0.251586776971817,
      "num_tokens": 96272492.0,
      "step": 41985
    },
    {
      "entropy": 5.456419467926025,
      "epoch": 4.0336215177713735,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003461375554378483,
      "loss": 4.5904,
      "mean_token_accuracy": 0.2567483648657799,
      "num_tokens": 96284657.0,
      "step": 41990
    },
    {
      "entropy": 5.588797760009766,
      "epoch": 4.034101825168108,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034610502968033895,
      "loss": 4.7853,
      "mean_token_accuracy": 0.24248917549848556,
      "num_tokens": 96296958.0,
      "step": 41995
    },
    {
      "entropy": 5.4682825088500975,
      "epoch": 4.034582132564841,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000346072502271982,
      "loss": 4.6302,
      "mean_token_accuracy": 0.26397144198417666,
      "num_tokens": 96307923.0,
      "step": 42000
    },
    {
      "epoch": 4.034582132564841,
      "eval_entropy": 5.229828721311021,
      "eval_loss": 4.784153461456299,
      "eval_mean_token_accuracy": 0.2587579297885364,
      "eval_num_tokens": 96307923.0,
      "eval_runtime": 20.6505,
      "eval_samples_per_second": 1589.068,
      "eval_steps_per_second": 198.64,
      "step": 42000
    },
    {
      "entropy": 5.495956659317017,
      "epoch": 4.0350624399615755,
      "grad_norm": 1.125,
      "learning_rate": 0.0003460399732135326,
      "loss": 4.6873,
      "mean_token_accuracy": 0.2539865165948868,
      "num_tokens": 96320177.0,
      "step": 42005
    },
    {
      "entropy": 5.5131641864776615,
      "epoch": 4.035542747358309,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003460074425057458,
      "loss": 4.7091,
      "mean_token_accuracy": 0.25096220076084136,
      "num_tokens": 96331860.0,
      "step": 42010
    },
    {
      "entropy": 5.348714113235474,
      "epoch": 4.036023054755043,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034597491014937723,
      "loss": 4.518,
      "mean_token_accuracy": 0.26746706366539,
      "num_tokens": 96342111.0,
      "step": 42015
    },
    {
      "entropy": 5.447357845306397,
      "epoch": 4.036503362151777,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003459423761451819,
      "loss": 4.6957,
      "mean_token_accuracy": 0.253644572198391,
      "num_tokens": 96353930.0,
      "step": 42020
    },
    {
      "entropy": 5.481619024276734,
      "epoch": 4.036983669548511,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003459098404939153,
      "loss": 4.6995,
      "mean_token_accuracy": 0.2568279430270195,
      "num_tokens": 96364924.0,
      "step": 42025
    },
    {
      "entropy": 5.553538084030151,
      "epoch": 4.037463976945245,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034587730319633265,
      "loss": 4.7058,
      "mean_token_accuracy": 0.2551100492477417,
      "num_tokens": 96377043.0,
      "step": 42030
    },
    {
      "entropy": 5.485311651229859,
      "epoch": 4.037944284341979,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034584476425318964,
      "loss": 4.629,
      "mean_token_accuracy": 0.26472594290971757,
      "num_tokens": 96388122.0,
      "step": 42035
    },
    {
      "entropy": 5.434488916397095,
      "epoch": 4.038424591738713,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034581222366524147,
      "loss": 4.6421,
      "mean_token_accuracy": 0.254853592813015,
      "num_tokens": 96399347.0,
      "step": 42040
    },
    {
      "entropy": 5.448543214797974,
      "epoch": 4.038904899135447,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034577968143324365,
      "loss": 4.6129,
      "mean_token_accuracy": 0.26251412779092786,
      "num_tokens": 96411249.0,
      "step": 42045
    },
    {
      "entropy": 5.537498474121094,
      "epoch": 4.039385206532181,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000345747137557952,
      "loss": 4.761,
      "mean_token_accuracy": 0.2526927664875984,
      "num_tokens": 96423274.0,
      "step": 42050
    },
    {
      "entropy": 5.550829076766968,
      "epoch": 4.039865513928914,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003457145920401218,
      "loss": 4.7248,
      "mean_token_accuracy": 0.25434657782316206,
      "num_tokens": 96434822.0,
      "step": 42055
    },
    {
      "entropy": 5.481413555145264,
      "epoch": 4.040345821325649,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034568204488050875,
      "loss": 4.6496,
      "mean_token_accuracy": 0.2573492243885994,
      "num_tokens": 96446313.0,
      "step": 42060
    },
    {
      "entropy": 5.569108533859253,
      "epoch": 4.040826128722382,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003456494960798685,
      "loss": 4.7197,
      "mean_token_accuracy": 0.24860191345214844,
      "num_tokens": 96457669.0,
      "step": 42065
    },
    {
      "entropy": 5.491832494735718,
      "epoch": 4.041306436119116,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003456169456389567,
      "loss": 4.6693,
      "mean_token_accuracy": 0.260526505112648,
      "num_tokens": 96469056.0,
      "step": 42070
    },
    {
      "entropy": 5.383673810958863,
      "epoch": 4.04178674351585,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034558439355852915,
      "loss": 4.5328,
      "mean_token_accuracy": 0.26611380875110624,
      "num_tokens": 96479783.0,
      "step": 42075
    },
    {
      "entropy": 5.432693147659302,
      "epoch": 4.042267050912584,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003455518398393416,
      "loss": 4.6588,
      "mean_token_accuracy": 0.25946155190467834,
      "num_tokens": 96491473.0,
      "step": 42080
    },
    {
      "entropy": 5.511789083480835,
      "epoch": 4.042747358309318,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034551928448214983,
      "loss": 4.7211,
      "mean_token_accuracy": 0.25815297216176986,
      "num_tokens": 96502380.0,
      "step": 42085
    },
    {
      "entropy": 5.636602830886841,
      "epoch": 4.043227665706052,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003454867274877096,
      "loss": 4.8094,
      "mean_token_accuracy": 0.2482210159301758,
      "num_tokens": 96514157.0,
      "step": 42090
    },
    {
      "entropy": 5.600018930435181,
      "epoch": 4.043707973102785,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034545416885677695,
      "loss": 4.8055,
      "mean_token_accuracy": 0.24459494650363922,
      "num_tokens": 96526549.0,
      "step": 42095
    },
    {
      "entropy": 5.472967958450317,
      "epoch": 4.04418828049952,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003454216085901076,
      "loss": 4.6972,
      "mean_token_accuracy": 0.2533607974648476,
      "num_tokens": 96539158.0,
      "step": 42100
    },
    {
      "entropy": 5.3986876010894775,
      "epoch": 4.044668587896253,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034538904668845766,
      "loss": 4.6018,
      "mean_token_accuracy": 0.2671900540590286,
      "num_tokens": 96552604.0,
      "step": 42105
    },
    {
      "entropy": 5.420937442779541,
      "epoch": 4.0451488952929875,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003453564831525831,
      "loss": 4.6043,
      "mean_token_accuracy": 0.2609040305018425,
      "num_tokens": 96565457.0,
      "step": 42110
    },
    {
      "entropy": 5.50702052116394,
      "epoch": 4.045629202689722,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003453239179832399,
      "loss": 4.6867,
      "mean_token_accuracy": 0.2570617452263832,
      "num_tokens": 96575605.0,
      "step": 42115
    },
    {
      "entropy": 5.488935470581055,
      "epoch": 4.046109510086455,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003452913511811841,
      "loss": 4.6935,
      "mean_token_accuracy": 0.25457257926464083,
      "num_tokens": 96586426.0,
      "step": 42120
    },
    {
      "entropy": 5.493999385833741,
      "epoch": 4.04658981748319,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003452587827471719,
      "loss": 4.66,
      "mean_token_accuracy": 0.25197656005620955,
      "num_tokens": 96597955.0,
      "step": 42125
    },
    {
      "entropy": 5.447873258590699,
      "epoch": 4.047070124879923,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034522621268195936,
      "loss": 4.5944,
      "mean_token_accuracy": 0.2628923922777176,
      "num_tokens": 96609148.0,
      "step": 42130
    },
    {
      "entropy": 5.433773708343506,
      "epoch": 4.047550432276657,
      "grad_norm": 1.25,
      "learning_rate": 0.0003451936409863027,
      "loss": 4.6003,
      "mean_token_accuracy": 0.26537771970033647,
      "num_tokens": 96621221.0,
      "step": 42135
    },
    {
      "entropy": 5.497773265838623,
      "epoch": 4.048030739673391,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034516106766095813,
      "loss": 4.7098,
      "mean_token_accuracy": 0.25761554539203646,
      "num_tokens": 96632567.0,
      "step": 42140
    },
    {
      "entropy": 5.504492473602295,
      "epoch": 4.048511047070125,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003451284927066819,
      "loss": 4.5953,
      "mean_token_accuracy": 0.2626398906111717,
      "num_tokens": 96643501.0,
      "step": 42145
    },
    {
      "entropy": 5.437385320663452,
      "epoch": 4.0489913544668585,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003450959161242303,
      "loss": 4.6345,
      "mean_token_accuracy": 0.2616502195596695,
      "num_tokens": 96654395.0,
      "step": 42150
    },
    {
      "entropy": 5.525277233123779,
      "epoch": 4.049471661863593,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003450633379143597,
      "loss": 4.7808,
      "mean_token_accuracy": 0.24430831223726274,
      "num_tokens": 96666135.0,
      "step": 42155
    },
    {
      "entropy": 5.464745903015137,
      "epoch": 4.049951969260326,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034503075807782634,
      "loss": 4.6067,
      "mean_token_accuracy": 0.2633389472961426,
      "num_tokens": 96678149.0,
      "step": 42160
    },
    {
      "entropy": 5.462284278869629,
      "epoch": 4.050432276657061,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003449981766153869,
      "loss": 4.6213,
      "mean_token_accuracy": 0.2719177544116974,
      "num_tokens": 96688557.0,
      "step": 42165
    },
    {
      "entropy": 5.479218816757202,
      "epoch": 4.050912584053794,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003449655935277976,
      "loss": 4.6672,
      "mean_token_accuracy": 0.259587237238884,
      "num_tokens": 96699404.0,
      "step": 42170
    },
    {
      "entropy": 5.438632011413574,
      "epoch": 4.051392891450528,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034493300881581484,
      "loss": 4.6108,
      "mean_token_accuracy": 0.2583493009209633,
      "num_tokens": 96711673.0,
      "step": 42175
    },
    {
      "entropy": 5.451039743423462,
      "epoch": 4.051873198847262,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003449004224801954,
      "loss": 4.6362,
      "mean_token_accuracy": 0.2681694656610489,
      "num_tokens": 96723113.0,
      "step": 42180
    },
    {
      "entropy": 5.549369144439697,
      "epoch": 4.052353506243996,
      "grad_norm": 2.109375,
      "learning_rate": 0.0003448678345216957,
      "loss": 4.8044,
      "mean_token_accuracy": 0.24937510639429092,
      "num_tokens": 96733323.0,
      "step": 42185
    },
    {
      "entropy": 5.513789653778076,
      "epoch": 4.0528338136407305,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034483524494107235,
      "loss": 4.6722,
      "mean_token_accuracy": 0.25546194314956666,
      "num_tokens": 96743524.0,
      "step": 42190
    },
    {
      "entropy": 5.600406789779663,
      "epoch": 4.053314121037464,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034480265373908187,
      "loss": 4.8016,
      "mean_token_accuracy": 0.24711752384901048,
      "num_tokens": 96756565.0,
      "step": 42195
    },
    {
      "entropy": 5.477747058868408,
      "epoch": 4.053794428434198,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034477006091648116,
      "loss": 4.6163,
      "mean_token_accuracy": 0.26164445728063584,
      "num_tokens": 96767391.0,
      "step": 42200
    },
    {
      "entropy": 5.4294291019439695,
      "epoch": 4.054274735830932,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034473746647402674,
      "loss": 4.6258,
      "mean_token_accuracy": 0.26534524112939833,
      "num_tokens": 96778074.0,
      "step": 42205
    },
    {
      "entropy": 5.389470529556275,
      "epoch": 4.054755043227666,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003447048704124754,
      "loss": 4.5332,
      "mean_token_accuracy": 0.27466691583395003,
      "num_tokens": 96789669.0,
      "step": 42210
    },
    {
      "entropy": 5.536287260055542,
      "epoch": 4.055235350624399,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003446722727325841,
      "loss": 4.7251,
      "mean_token_accuracy": 0.2566806897521019,
      "num_tokens": 96800089.0,
      "step": 42215
    },
    {
      "entropy": 5.555914115905762,
      "epoch": 4.055715658021134,
      "grad_norm": 1.25,
      "learning_rate": 0.0003446396734351094,
      "loss": 4.7764,
      "mean_token_accuracy": 0.24473524540662767,
      "num_tokens": 96812248.0,
      "step": 42220
    },
    {
      "entropy": 5.585545396804809,
      "epoch": 4.056195965417867,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003446070725208083,
      "loss": 4.7973,
      "mean_token_accuracy": 0.23909653276205062,
      "num_tokens": 96824016.0,
      "step": 42225
    },
    {
      "entropy": 5.482309484481812,
      "epoch": 4.0566762728146015,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003445744699904377,
      "loss": 4.6365,
      "mean_token_accuracy": 0.257051782310009,
      "num_tokens": 96835128.0,
      "step": 42230
    },
    {
      "entropy": 5.438259077072144,
      "epoch": 4.057156580211335,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003445418658447544,
      "loss": 4.6277,
      "mean_token_accuracy": 0.25690867453813554,
      "num_tokens": 96846755.0,
      "step": 42235
    },
    {
      "entropy": 5.472283506393433,
      "epoch": 4.057636887608069,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034450926008451556,
      "loss": 4.679,
      "mean_token_accuracy": 0.2580919861793518,
      "num_tokens": 96858262.0,
      "step": 42240
    },
    {
      "entropy": 5.4701074123382565,
      "epoch": 4.058117195004803,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003444766527104781,
      "loss": 4.6836,
      "mean_token_accuracy": 0.25073648542165755,
      "num_tokens": 96871533.0,
      "step": 42245
    },
    {
      "entropy": 5.481945896148682,
      "epoch": 4.058597502401537,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000344444043723399,
      "loss": 4.6677,
      "mean_token_accuracy": 0.2609832614660263,
      "num_tokens": 96883668.0,
      "step": 42250
    },
    {
      "entropy": 5.517292547225952,
      "epoch": 4.0590778097982705,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003444114331240355,
      "loss": 4.7239,
      "mean_token_accuracy": 0.2512525036931038,
      "num_tokens": 96894196.0,
      "step": 42255
    },
    {
      "entropy": 5.573529291152954,
      "epoch": 4.059558117195005,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003443788209131446,
      "loss": 4.7111,
      "mean_token_accuracy": 0.25915242582559583,
      "num_tokens": 96905570.0,
      "step": 42260
    },
    {
      "entropy": 5.518734169006348,
      "epoch": 4.060038424591739,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034434620709148343,
      "loss": 4.67,
      "mean_token_accuracy": 0.2659013465046883,
      "num_tokens": 96916852.0,
      "step": 42265
    },
    {
      "entropy": 5.474788618087769,
      "epoch": 4.060518731988473,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003443135916598093,
      "loss": 4.6514,
      "mean_token_accuracy": 0.25174660235643387,
      "num_tokens": 96927851.0,
      "step": 42270
    },
    {
      "entropy": 5.455215692520142,
      "epoch": 4.060999039385207,
      "grad_norm": 1.125,
      "learning_rate": 0.00034428097461887934,
      "loss": 4.6881,
      "mean_token_accuracy": 0.25293929874897003,
      "num_tokens": 96939976.0,
      "step": 42275
    },
    {
      "entropy": 5.463224840164185,
      "epoch": 4.06147934678194,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003442483559694509,
      "loss": 4.7089,
      "mean_token_accuracy": 0.25125221759080885,
      "num_tokens": 96951920.0,
      "step": 42280
    },
    {
      "entropy": 5.50343017578125,
      "epoch": 4.061959654178675,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034421573571228134,
      "loss": 4.6951,
      "mean_token_accuracy": 0.25697204768657683,
      "num_tokens": 96963389.0,
      "step": 42285
    },
    {
      "entropy": 5.461284017562866,
      "epoch": 4.062439961575408,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003441831138481279,
      "loss": 4.6627,
      "mean_token_accuracy": 0.26713597625494,
      "num_tokens": 96975250.0,
      "step": 42290
    },
    {
      "entropy": 5.419812536239624,
      "epoch": 4.062920268972142,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003441504903777479,
      "loss": 4.5974,
      "mean_token_accuracy": 0.2629875883460045,
      "num_tokens": 96986116.0,
      "step": 42295
    },
    {
      "entropy": 5.471432065963745,
      "epoch": 4.063400576368876,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034411786530189894,
      "loss": 4.6661,
      "mean_token_accuracy": 0.25610075891017914,
      "num_tokens": 96998370.0,
      "step": 42300
    },
    {
      "entropy": 5.427270221710205,
      "epoch": 4.06388088376561,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034408523862133834,
      "loss": 4.5956,
      "mean_token_accuracy": 0.2541428878903389,
      "num_tokens": 97010866.0,
      "step": 42305
    },
    {
      "entropy": 5.5291155815124515,
      "epoch": 4.064361191162344,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034405261033682363,
      "loss": 4.6908,
      "mean_token_accuracy": 0.25422141551971433,
      "num_tokens": 97023127.0,
      "step": 42310
    },
    {
      "entropy": 5.517888879776001,
      "epoch": 4.064841498559078,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003440199804491123,
      "loss": 4.7374,
      "mean_token_accuracy": 0.2515604704618454,
      "num_tokens": 97035117.0,
      "step": 42315
    },
    {
      "entropy": 5.5439671039581295,
      "epoch": 4.065321805955811,
      "grad_norm": 1.25,
      "learning_rate": 0.0003439873489589621,
      "loss": 4.7289,
      "mean_token_accuracy": 0.2477584719657898,
      "num_tokens": 97046711.0,
      "step": 42320
    },
    {
      "entropy": 5.511455011367798,
      "epoch": 4.065802113352546,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003439547158671304,
      "loss": 4.6519,
      "mean_token_accuracy": 0.2529826298356056,
      "num_tokens": 97059120.0,
      "step": 42325
    },
    {
      "entropy": 5.444996690750122,
      "epoch": 4.066282420749279,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034392208117437503,
      "loss": 4.6482,
      "mean_token_accuracy": 0.2537728428840637,
      "num_tokens": 97071879.0,
      "step": 42330
    },
    {
      "entropy": 5.467194128036499,
      "epoch": 4.0667627281460135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034388944488145356,
      "loss": 4.6824,
      "mean_token_accuracy": 0.2609917983412743,
      "num_tokens": 97082883.0,
      "step": 42335
    },
    {
      "entropy": 5.456636381149292,
      "epoch": 4.067243035542747,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034385680698912377,
      "loss": 4.6631,
      "mean_token_accuracy": 0.25637172907590866,
      "num_tokens": 97094174.0,
      "step": 42340
    },
    {
      "entropy": 5.569712114334107,
      "epoch": 4.067723342939481,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034382416749814335,
      "loss": 4.7218,
      "mean_token_accuracy": 0.24793127179145813,
      "num_tokens": 97104328.0,
      "step": 42345
    },
    {
      "entropy": 5.554775190353394,
      "epoch": 4.068203650336216,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003437915264092701,
      "loss": 4.6837,
      "mean_token_accuracy": 0.2568752110004425,
      "num_tokens": 97117533.0,
      "step": 42350
    },
    {
      "entropy": 5.528671455383301,
      "epoch": 4.068683957732949,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003437588837232619,
      "loss": 4.7205,
      "mean_token_accuracy": 0.25234463810920715,
      "num_tokens": 97129355.0,
      "step": 42355
    },
    {
      "entropy": 5.460422229766846,
      "epoch": 4.069164265129683,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034372623944087656,
      "loss": 4.6474,
      "mean_token_accuracy": 0.260855370759964,
      "num_tokens": 97141424.0,
      "step": 42360
    },
    {
      "entropy": 5.474782848358155,
      "epoch": 4.069644572526417,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003436935935628721,
      "loss": 4.678,
      "mean_token_accuracy": 0.2604090079665184,
      "num_tokens": 97152900.0,
      "step": 42365
    },
    {
      "entropy": 5.516853284835816,
      "epoch": 4.070124879923151,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034366094609000633,
      "loss": 4.7253,
      "mean_token_accuracy": 0.25317352861166,
      "num_tokens": 97163894.0,
      "step": 42370
    },
    {
      "entropy": 5.494958734512329,
      "epoch": 4.0706051873198845,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003436282970230372,
      "loss": 4.7043,
      "mean_token_accuracy": 0.2517710417509079,
      "num_tokens": 97175385.0,
      "step": 42375
    },
    {
      "entropy": 5.524476671218872,
      "epoch": 4.071085494716619,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034359564636272287,
      "loss": 4.6569,
      "mean_token_accuracy": 0.26241626441478727,
      "num_tokens": 97186830.0,
      "step": 42380
    },
    {
      "entropy": 5.560749578475952,
      "epoch": 4.071565802113352,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003435629941098213,
      "loss": 4.7369,
      "mean_token_accuracy": 0.2540200769901276,
      "num_tokens": 97197816.0,
      "step": 42385
    },
    {
      "entropy": 5.523098611831665,
      "epoch": 4.072046109510087,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003435303402650906,
      "loss": 4.6611,
      "mean_token_accuracy": 0.2602622225880623,
      "num_tokens": 97208721.0,
      "step": 42390
    },
    {
      "entropy": 5.5389657497406,
      "epoch": 4.07252641690682,
      "grad_norm": 1.234375,
      "learning_rate": 0.000343497684829289,
      "loss": 4.7773,
      "mean_token_accuracy": 0.24761004000902176,
      "num_tokens": 97220487.0,
      "step": 42395
    },
    {
      "entropy": 5.466276216506958,
      "epoch": 4.073006724303554,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003434650278031744,
      "loss": 4.6288,
      "mean_token_accuracy": 0.2647788882255554,
      "num_tokens": 97231995.0,
      "step": 42400
    },
    {
      "entropy": 5.498526000976563,
      "epoch": 4.073487031700288,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034343236918750523,
      "loss": 4.5807,
      "mean_token_accuracy": 0.27066588699817656,
      "num_tokens": 97243168.0,
      "step": 42405
    },
    {
      "entropy": 5.499664068222046,
      "epoch": 4.073967339097022,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034339970898303963,
      "loss": 4.6429,
      "mean_token_accuracy": 0.267056243121624,
      "num_tokens": 97254361.0,
      "step": 42410
    },
    {
      "entropy": 5.46985650062561,
      "epoch": 4.074447646493756,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034336704719053584,
      "loss": 4.629,
      "mean_token_accuracy": 0.2599684566259384,
      "num_tokens": 97265535.0,
      "step": 42415
    },
    {
      "entropy": 5.391230201721191,
      "epoch": 4.07492795389049,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003433343838107523,
      "loss": 4.6556,
      "mean_token_accuracy": 0.2633891269564629,
      "num_tokens": 97277426.0,
      "step": 42420
    },
    {
      "entropy": 5.437358236312866,
      "epoch": 4.075408261287224,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034330171884444725,
      "loss": 4.6238,
      "mean_token_accuracy": 0.2561158090829849,
      "num_tokens": 97288897.0,
      "step": 42425
    },
    {
      "entropy": 5.454502391815185,
      "epoch": 4.075888568683958,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034326905229237914,
      "loss": 4.6574,
      "mean_token_accuracy": 0.25457168072462083,
      "num_tokens": 97299288.0,
      "step": 42430
    },
    {
      "entropy": 5.575408935546875,
      "epoch": 4.076368876080692,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034323638415530627,
      "loss": 4.7001,
      "mean_token_accuracy": 0.2578587606549263,
      "num_tokens": 97309122.0,
      "step": 42435
    },
    {
      "entropy": 5.407442712783814,
      "epoch": 4.076849183477425,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003432037144339872,
      "loss": 4.6024,
      "mean_token_accuracy": 0.2654115706682205,
      "num_tokens": 97320390.0,
      "step": 42440
    },
    {
      "entropy": 5.430894422531128,
      "epoch": 4.07732949087416,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034317104312918054,
      "loss": 4.7038,
      "mean_token_accuracy": 0.25004568994045256,
      "num_tokens": 97332062.0,
      "step": 42445
    },
    {
      "entropy": 5.483252239227295,
      "epoch": 4.077809798270893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003431383702416445,
      "loss": 4.6129,
      "mean_token_accuracy": 0.2611764043569565,
      "num_tokens": 97343197.0,
      "step": 42450
    },
    {
      "entropy": 5.528693819046021,
      "epoch": 4.0782901056676275,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034310569577213795,
      "loss": 4.6674,
      "mean_token_accuracy": 0.25614446550607683,
      "num_tokens": 97354667.0,
      "step": 42455
    },
    {
      "entropy": 5.58443603515625,
      "epoch": 4.078770413064361,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003430730197214193,
      "loss": 4.8037,
      "mean_token_accuracy": 0.24740854650735855,
      "num_tokens": 97365011.0,
      "step": 42460
    },
    {
      "entropy": 5.4460304260253904,
      "epoch": 4.079250720461095,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003430403420902473,
      "loss": 4.6959,
      "mean_token_accuracy": 0.251497757434845,
      "num_tokens": 97376285.0,
      "step": 42465
    },
    {
      "entropy": 5.447171449661255,
      "epoch": 4.079731027857829,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034300766287938066,
      "loss": 4.7015,
      "mean_token_accuracy": 0.2555937647819519,
      "num_tokens": 97387901.0,
      "step": 42470
    },
    {
      "entropy": 5.486668539047241,
      "epoch": 4.080211335254563,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034297498208957797,
      "loss": 4.6265,
      "mean_token_accuracy": 0.2660987854003906,
      "num_tokens": 97399854.0,
      "step": 42475
    },
    {
      "entropy": 5.456594800949096,
      "epoch": 4.0806916426512965,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000342942299721598,
      "loss": 4.6259,
      "mean_token_accuracy": 0.26028558760881426,
      "num_tokens": 97410867.0,
      "step": 42480
    },
    {
      "entropy": 5.474694108963012,
      "epoch": 4.081171950048031,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003429096157761997,
      "loss": 4.7058,
      "mean_token_accuracy": 0.2518029123544693,
      "num_tokens": 97422890.0,
      "step": 42485
    },
    {
      "entropy": 5.432509326934815,
      "epoch": 4.081652257444764,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003428769302541416,
      "loss": 4.6186,
      "mean_token_accuracy": 0.25718304663896563,
      "num_tokens": 97434465.0,
      "step": 42490
    },
    {
      "entropy": 5.442902326583862,
      "epoch": 4.082132564841499,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003428442431561828,
      "loss": 4.5973,
      "mean_token_accuracy": 0.26498415172100065,
      "num_tokens": 97445358.0,
      "step": 42495
    },
    {
      "entropy": 5.457403612136841,
      "epoch": 4.082612872238233,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003428115544830823,
      "loss": 4.6236,
      "mean_token_accuracy": 0.2556000486016273,
      "num_tokens": 97456120.0,
      "step": 42500
    },
    {
      "entropy": 5.47347936630249,
      "epoch": 4.083093179634966,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034277886423559863,
      "loss": 4.6827,
      "mean_token_accuracy": 0.25540692508220675,
      "num_tokens": 97467377.0,
      "step": 42505
    },
    {
      "entropy": 5.452052927017212,
      "epoch": 4.083573487031701,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034274617241449105,
      "loss": 4.6992,
      "mean_token_accuracy": 0.25108240991830827,
      "num_tokens": 97479250.0,
      "step": 42510
    },
    {
      "entropy": 5.553588771820069,
      "epoch": 4.084053794428434,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034271347902051864,
      "loss": 4.7325,
      "mean_token_accuracy": 0.25506940186023713,
      "num_tokens": 97491157.0,
      "step": 42515
    },
    {
      "entropy": 5.426689863204956,
      "epoch": 4.084534101825168,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003426807840544401,
      "loss": 4.5565,
      "mean_token_accuracy": 0.2615838810801506,
      "num_tokens": 97503075.0,
      "step": 42520
    },
    {
      "entropy": 5.509971952438354,
      "epoch": 4.085014409221902,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003426480875170149,
      "loss": 4.763,
      "mean_token_accuracy": 0.2469555988907814,
      "num_tokens": 97514026.0,
      "step": 42525
    },
    {
      "entropy": 5.486274194717407,
      "epoch": 4.085494716618636,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034261538940900195,
      "loss": 4.6479,
      "mean_token_accuracy": 0.25809302777051923,
      "num_tokens": 97526866.0,
      "step": 42530
    },
    {
      "entropy": 5.539806127548218,
      "epoch": 4.08597502401537,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003425826897311604,
      "loss": 4.7474,
      "mean_token_accuracy": 0.25054112672805784,
      "num_tokens": 97538632.0,
      "step": 42535
    },
    {
      "entropy": 5.482585430145264,
      "epoch": 4.086455331412104,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034254998848424947,
      "loss": 4.6478,
      "mean_token_accuracy": 0.2513435587286949,
      "num_tokens": 97549288.0,
      "step": 42540
    },
    {
      "entropy": 5.520724773406982,
      "epoch": 4.086935638808837,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003425172856690284,
      "loss": 4.747,
      "mean_token_accuracy": 0.2511808440089226,
      "num_tokens": 97560582.0,
      "step": 42545
    },
    {
      "entropy": 5.457403945922851,
      "epoch": 4.087415946205572,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034248458128625643,
      "loss": 4.5972,
      "mean_token_accuracy": 0.25831068605184554,
      "num_tokens": 97571974.0,
      "step": 42550
    },
    {
      "entropy": 5.470881128311158,
      "epoch": 4.087896253602305,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034245187533669285,
      "loss": 4.6483,
      "mean_token_accuracy": 0.2584894299507141,
      "num_tokens": 97582456.0,
      "step": 42555
    },
    {
      "entropy": 5.513875722885132,
      "epoch": 4.0883765609990395,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034241916782109714,
      "loss": 4.7373,
      "mean_token_accuracy": 0.252739417552948,
      "num_tokens": 97593246.0,
      "step": 42560
    },
    {
      "entropy": 5.413125228881836,
      "epoch": 4.088856868395773,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034238645874022845,
      "loss": 4.5372,
      "mean_token_accuracy": 0.2677224099636078,
      "num_tokens": 97604004.0,
      "step": 42565
    },
    {
      "entropy": 5.5298995018005375,
      "epoch": 4.089337175792507,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034235374809484626,
      "loss": 4.7289,
      "mean_token_accuracy": 0.2519468620419502,
      "num_tokens": 97616122.0,
      "step": 42570
    },
    {
      "entropy": 5.4438403129577635,
      "epoch": 4.0898174831892415,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003423210358857101,
      "loss": 4.6199,
      "mean_token_accuracy": 0.25683494508266447,
      "num_tokens": 97627550.0,
      "step": 42575
    },
    {
      "entropy": 5.524833869934082,
      "epoch": 4.090297790585975,
      "grad_norm": 1.125,
      "learning_rate": 0.00034228832211357935,
      "loss": 4.7539,
      "mean_token_accuracy": 0.24522749334573746,
      "num_tokens": 97639952.0,
      "step": 42580
    },
    {
      "entropy": 5.431515407562256,
      "epoch": 4.090778097982709,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034225560677921354,
      "loss": 4.5922,
      "mean_token_accuracy": 0.2634821593761444,
      "num_tokens": 97653158.0,
      "step": 42585
    },
    {
      "entropy": 5.5099437713623045,
      "epoch": 4.091258405379443,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034222288988337233,
      "loss": 4.7466,
      "mean_token_accuracy": 0.2555833637714386,
      "num_tokens": 97666768.0,
      "step": 42590
    },
    {
      "entropy": 5.514379978179932,
      "epoch": 4.091738712776177,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003421901714268152,
      "loss": 4.6685,
      "mean_token_accuracy": 0.260109905898571,
      "num_tokens": 97678128.0,
      "step": 42595
    },
    {
      "entropy": 5.505461692810059,
      "epoch": 4.0922190201729105,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034215745141030177,
      "loss": 4.7129,
      "mean_token_accuracy": 0.254145510494709,
      "num_tokens": 97689565.0,
      "step": 42600
    },
    {
      "entropy": 5.460262203216553,
      "epoch": 4.092699327569645,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034212472983459176,
      "loss": 4.6367,
      "mean_token_accuracy": 0.26155361235141755,
      "num_tokens": 97701315.0,
      "step": 42605
    },
    {
      "entropy": 5.5240113735198975,
      "epoch": 4.093179634966378,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003420920067004448,
      "loss": 4.7134,
      "mean_token_accuracy": 0.25764271318912507,
      "num_tokens": 97712404.0,
      "step": 42610
    },
    {
      "entropy": 5.5000988960266115,
      "epoch": 4.093659942363113,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003420592820086207,
      "loss": 4.6224,
      "mean_token_accuracy": 0.25599916726350785,
      "num_tokens": 97723547.0,
      "step": 42615
    },
    {
      "entropy": 5.45414924621582,
      "epoch": 4.094140249759846,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034202655575987925,
      "loss": 4.6444,
      "mean_token_accuracy": 0.2699910461902618,
      "num_tokens": 97734133.0,
      "step": 42620
    },
    {
      "entropy": 5.5187397480010985,
      "epoch": 4.09462055715658,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034199382795498015,
      "loss": 4.7382,
      "mean_token_accuracy": 0.24838437736034394,
      "num_tokens": 97746479.0,
      "step": 42625
    },
    {
      "entropy": 5.530471658706665,
      "epoch": 4.095100864553314,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034196109859468325,
      "loss": 4.6723,
      "mean_token_accuracy": 0.260334113240242,
      "num_tokens": 97758042.0,
      "step": 42630
    },
    {
      "entropy": 5.463621139526367,
      "epoch": 4.095581171950048,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003419283676797485,
      "loss": 4.669,
      "mean_token_accuracy": 0.26399700790643693,
      "num_tokens": 97770338.0,
      "step": 42635
    },
    {
      "entropy": 5.474463176727295,
      "epoch": 4.0960614793467816,
      "grad_norm": 1.375,
      "learning_rate": 0.00034189563521093587,
      "loss": 4.6617,
      "mean_token_accuracy": 0.2557863175868988,
      "num_tokens": 97781274.0,
      "step": 42640
    },
    {
      "entropy": 5.543714237213135,
      "epoch": 4.096541786743516,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003418629011890051,
      "loss": 4.7629,
      "mean_token_accuracy": 0.2506802827119827,
      "num_tokens": 97792241.0,
      "step": 42645
    },
    {
      "entropy": 5.519195652008056,
      "epoch": 4.09702209414025,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034183016561471644,
      "loss": 4.7227,
      "mean_token_accuracy": 0.2530124679207802,
      "num_tokens": 97803315.0,
      "step": 42650
    },
    {
      "entropy": 5.637829923629761,
      "epoch": 4.097502401536984,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034179742848882967,
      "loss": 4.8267,
      "mean_token_accuracy": 0.24220484346151352,
      "num_tokens": 97815138.0,
      "step": 42655
    },
    {
      "entropy": 5.507266426086426,
      "epoch": 4.097982708933718,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034176468981210494,
      "loss": 4.6356,
      "mean_token_accuracy": 0.26421907395124433,
      "num_tokens": 97826846.0,
      "step": 42660
    },
    {
      "entropy": 5.503274154663086,
      "epoch": 4.098463016330451,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003417319495853025,
      "loss": 4.701,
      "mean_token_accuracy": 0.253885018825531,
      "num_tokens": 97838561.0,
      "step": 42665
    },
    {
      "entropy": 5.5226243019104,
      "epoch": 4.098943323727186,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003416992078091823,
      "loss": 4.6808,
      "mean_token_accuracy": 0.2544016644358635,
      "num_tokens": 97850306.0,
      "step": 42670
    },
    {
      "entropy": 5.515234756469726,
      "epoch": 4.099423631123919,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003416664644845044,
      "loss": 4.6801,
      "mean_token_accuracy": 0.25506350100040437,
      "num_tokens": 97862193.0,
      "step": 42675
    },
    {
      "entropy": 5.521270084381103,
      "epoch": 4.0999039385206535,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034163371961202935,
      "loss": 4.7276,
      "mean_token_accuracy": 0.24445707798004152,
      "num_tokens": 97874557.0,
      "step": 42680
    },
    {
      "entropy": 5.53608283996582,
      "epoch": 4.100384245917387,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034160097319251714,
      "loss": 4.7292,
      "mean_token_accuracy": 0.25125124752521516,
      "num_tokens": 97885334.0,
      "step": 42685
    },
    {
      "entropy": 5.562976980209351,
      "epoch": 4.100864553314121,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034156822522672805,
      "loss": 4.7721,
      "mean_token_accuracy": 0.2445255234837532,
      "num_tokens": 97896857.0,
      "step": 42690
    },
    {
      "entropy": 5.425723695755005,
      "epoch": 4.101344860710855,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003415354757154225,
      "loss": 4.6064,
      "mean_token_accuracy": 0.2588916689157486,
      "num_tokens": 97908172.0,
      "step": 42695
    },
    {
      "entropy": 5.455512619018554,
      "epoch": 4.101825168107589,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034150272465936066,
      "loss": 4.618,
      "mean_token_accuracy": 0.2626450449228287,
      "num_tokens": 97920176.0,
      "step": 42700
    },
    {
      "entropy": 5.452777671813965,
      "epoch": 4.1023054755043225,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034146997205930307,
      "loss": 4.6502,
      "mean_token_accuracy": 0.2547474503517151,
      "num_tokens": 97932225.0,
      "step": 42705
    },
    {
      "entropy": 5.486677789688111,
      "epoch": 4.102785782901057,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003414372179160101,
      "loss": 4.6353,
      "mean_token_accuracy": 0.2599442288279533,
      "num_tokens": 97942457.0,
      "step": 42710
    },
    {
      "entropy": 5.5726786136627195,
      "epoch": 4.10326609029779,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003414044622302424,
      "loss": 4.7642,
      "mean_token_accuracy": 0.25318109393119814,
      "num_tokens": 97954278.0,
      "step": 42715
    },
    {
      "entropy": 5.539187908172607,
      "epoch": 4.1037463976945245,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034137170500276004,
      "loss": 4.6922,
      "mean_token_accuracy": 0.24845859706401824,
      "num_tokens": 97966441.0,
      "step": 42720
    },
    {
      "entropy": 5.539091968536377,
      "epoch": 4.104226705091258,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003413389462343238,
      "loss": 4.6591,
      "mean_token_accuracy": 0.2587722137570381,
      "num_tokens": 97976810.0,
      "step": 42725
    },
    {
      "entropy": 5.490980339050293,
      "epoch": 4.104707012487992,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003413061859256943,
      "loss": 4.6525,
      "mean_token_accuracy": 0.2575728878378868,
      "num_tokens": 97989012.0,
      "step": 42730
    },
    {
      "entropy": 5.605491304397583,
      "epoch": 4.105187319884727,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034127342407763196,
      "loss": 4.8173,
      "mean_token_accuracy": 0.23798614144325256,
      "num_tokens": 98000280.0,
      "step": 42735
    },
    {
      "entropy": 5.5638752460479735,
      "epoch": 4.10566762728146,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003412406606908976,
      "loss": 4.7425,
      "mean_token_accuracy": 0.24998596161603928,
      "num_tokens": 98012097.0,
      "step": 42740
    },
    {
      "entropy": 5.550551891326904,
      "epoch": 4.106147934678194,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003412078957662517,
      "loss": 4.7854,
      "mean_token_accuracy": 0.2443050041794777,
      "num_tokens": 98022602.0,
      "step": 42745
    },
    {
      "entropy": 5.537406921386719,
      "epoch": 4.106628242074928,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034117512930445504,
      "loss": 4.7222,
      "mean_token_accuracy": 0.24959359019994737,
      "num_tokens": 98034535.0,
      "step": 42750
    },
    {
      "entropy": 5.4846641540527346,
      "epoch": 4.107108549471662,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034114236130626855,
      "loss": 4.6401,
      "mean_token_accuracy": 0.25843715518712995,
      "num_tokens": 98046525.0,
      "step": 42755
    },
    {
      "entropy": 5.525589418411255,
      "epoch": 4.107588856868396,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003411095917724527,
      "loss": 4.7177,
      "mean_token_accuracy": 0.26045708209276197,
      "num_tokens": 98057540.0,
      "step": 42760
    },
    {
      "entropy": 5.51448073387146,
      "epoch": 4.10806916426513,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034107682070376845,
      "loss": 4.7,
      "mean_token_accuracy": 0.26219588816165923,
      "num_tokens": 98068578.0,
      "step": 42765
    },
    {
      "entropy": 5.465487623214722,
      "epoch": 4.108549471661863,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034104404810097663,
      "loss": 4.6491,
      "mean_token_accuracy": 0.2534323543310165,
      "num_tokens": 98079981.0,
      "step": 42770
    },
    {
      "entropy": 5.494054698944092,
      "epoch": 4.109029779058598,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003410112739648381,
      "loss": 4.6747,
      "mean_token_accuracy": 0.26122272163629534,
      "num_tokens": 98091579.0,
      "step": 42775
    },
    {
      "entropy": 5.531874465942383,
      "epoch": 4.109510086455331,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034097849829611387,
      "loss": 4.7457,
      "mean_token_accuracy": 0.2461780235171318,
      "num_tokens": 98103239.0,
      "step": 42780
    },
    {
      "entropy": 5.5029284954071045,
      "epoch": 4.109990393852065,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034094572109556477,
      "loss": 4.6478,
      "mean_token_accuracy": 0.260518516600132,
      "num_tokens": 98114865.0,
      "step": 42785
    },
    {
      "entropy": 5.4606608867645265,
      "epoch": 4.110470701248799,
      "grad_norm": 1.484375,
      "learning_rate": 0.00034091294236395186,
      "loss": 4.6284,
      "mean_token_accuracy": 0.2539664670825005,
      "num_tokens": 98125831.0,
      "step": 42790
    },
    {
      "entropy": 5.454886245727539,
      "epoch": 4.110951008645533,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034088016210203617,
      "loss": 4.6425,
      "mean_token_accuracy": 0.25233357548713686,
      "num_tokens": 98136129.0,
      "step": 42795
    },
    {
      "entropy": 5.53878231048584,
      "epoch": 4.111431316042267,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003408473803105788,
      "loss": 4.7532,
      "mean_token_accuracy": 0.24865238219499589,
      "num_tokens": 98148022.0,
      "step": 42800
    },
    {
      "entropy": 5.48397045135498,
      "epoch": 4.111911623439001,
      "grad_norm": 1.25,
      "learning_rate": 0.0003408145969903406,
      "loss": 4.6184,
      "mean_token_accuracy": 0.25537666827440264,
      "num_tokens": 98159567.0,
      "step": 42805
    },
    {
      "entropy": 5.549623298645019,
      "epoch": 4.112391930835735,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034078181214208304,
      "loss": 4.7619,
      "mean_token_accuracy": 0.2548246577382088,
      "num_tokens": 98171957.0,
      "step": 42810
    },
    {
      "entropy": 5.41923189163208,
      "epoch": 4.112872238232469,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003407490257665671,
      "loss": 4.6086,
      "mean_token_accuracy": 0.26333325207233427,
      "num_tokens": 98184553.0,
      "step": 42815
    },
    {
      "entropy": 5.559159851074218,
      "epoch": 4.113352545629203,
      "grad_norm": 1.15625,
      "learning_rate": 0.000340716237864554,
      "loss": 4.7666,
      "mean_token_accuracy": 0.2413361355662346,
      "num_tokens": 98196510.0,
      "step": 42820
    },
    {
      "entropy": 5.543231725692749,
      "epoch": 4.1138328530259365,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034068344843680493,
      "loss": 4.7961,
      "mean_token_accuracy": 0.23852113932371138,
      "num_tokens": 98207234.0,
      "step": 42825
    },
    {
      "entropy": 5.484008932113648,
      "epoch": 4.114313160422671,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034065065748408135,
      "loss": 4.6265,
      "mean_token_accuracy": 0.2566272497177124,
      "num_tokens": 98218853.0,
      "step": 42830
    },
    {
      "entropy": 5.5188652038574215,
      "epoch": 4.114793467819404,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034061786500714434,
      "loss": 4.6951,
      "mean_token_accuracy": 0.2580799087882042,
      "num_tokens": 98230668.0,
      "step": 42835
    },
    {
      "entropy": 5.445721578598023,
      "epoch": 4.115273775216139,
      "grad_norm": 1.375,
      "learning_rate": 0.00034058507100675545,
      "loss": 4.6088,
      "mean_token_accuracy": 0.25986621975898744,
      "num_tokens": 98241982.0,
      "step": 42840
    },
    {
      "entropy": 5.527672910690308,
      "epoch": 4.115754082612872,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034055227548367595,
      "loss": 4.7366,
      "mean_token_accuracy": 0.252566197514534,
      "num_tokens": 98253726.0,
      "step": 42845
    },
    {
      "entropy": 5.585765695571899,
      "epoch": 4.116234390009606,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003405194784386672,
      "loss": 4.6692,
      "mean_token_accuracy": 0.25420495718717573,
      "num_tokens": 98264290.0,
      "step": 42850
    },
    {
      "entropy": 5.4866255760192875,
      "epoch": 4.11671469740634,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003404866798724907,
      "loss": 4.6503,
      "mean_token_accuracy": 0.25912766754627226,
      "num_tokens": 98276175.0,
      "step": 42855
    },
    {
      "entropy": 5.380297613143921,
      "epoch": 4.117195004803074,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000340453879785908,
      "loss": 4.6099,
      "mean_token_accuracy": 0.25874359905719757,
      "num_tokens": 98287023.0,
      "step": 42860
    },
    {
      "entropy": 5.48995532989502,
      "epoch": 4.1176753121998075,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034042107817968054,
      "loss": 4.6427,
      "mean_token_accuracy": 0.2641780212521553,
      "num_tokens": 98297576.0,
      "step": 42865
    },
    {
      "entropy": 5.47585825920105,
      "epoch": 4.118155619596542,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034038827505456995,
      "loss": 4.6759,
      "mean_token_accuracy": 0.25060486048460007,
      "num_tokens": 98308951.0,
      "step": 42870
    },
    {
      "entropy": 5.487484931945801,
      "epoch": 4.118635926993275,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003403554704113378,
      "loss": 4.6905,
      "mean_token_accuracy": 0.25849495977163317,
      "num_tokens": 98320696.0,
      "step": 42875
    },
    {
      "entropy": 5.523585653305053,
      "epoch": 4.11911623439001,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003403226642507456,
      "loss": 4.6918,
      "mean_token_accuracy": 0.25542200803756715,
      "num_tokens": 98331778.0,
      "step": 42880
    },
    {
      "entropy": 5.559002208709717,
      "epoch": 4.119596541786744,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034028985657355516,
      "loss": 4.684,
      "mean_token_accuracy": 0.2582764163613319,
      "num_tokens": 98343142.0,
      "step": 42885
    },
    {
      "entropy": 5.474043130874634,
      "epoch": 4.120076849183477,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003402570473805281,
      "loss": 4.7733,
      "mean_token_accuracy": 0.24784097224473953,
      "num_tokens": 98354907.0,
      "step": 42890
    },
    {
      "entropy": 5.54673752784729,
      "epoch": 4.120557156580212,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003402242366724262,
      "loss": 4.8166,
      "mean_token_accuracy": 0.2450270026922226,
      "num_tokens": 98366846.0,
      "step": 42895
    },
    {
      "entropy": 5.529747247695923,
      "epoch": 4.121037463976945,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034019142445001116,
      "loss": 4.6869,
      "mean_token_accuracy": 0.2566752225160599,
      "num_tokens": 98379572.0,
      "step": 42900
    },
    {
      "entropy": 5.512189292907715,
      "epoch": 4.1215177713736795,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034015861071404484,
      "loss": 4.7005,
      "mean_token_accuracy": 0.24982386827468872,
      "num_tokens": 98391151.0,
      "step": 42905
    },
    {
      "entropy": 5.472418642044067,
      "epoch": 4.121998078770413,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003401257954652891,
      "loss": 4.6394,
      "mean_token_accuracy": 0.26286680698394777,
      "num_tokens": 98402012.0,
      "step": 42910
    },
    {
      "entropy": 5.433299541473389,
      "epoch": 4.122478386167147,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034009297870450574,
      "loss": 4.6035,
      "mean_token_accuracy": 0.2621582433581352,
      "num_tokens": 98413533.0,
      "step": 42915
    },
    {
      "entropy": 5.428326559066773,
      "epoch": 4.122958693563881,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003400601604324567,
      "loss": 4.5857,
      "mean_token_accuracy": 0.26660341322422026,
      "num_tokens": 98425056.0,
      "step": 42920
    },
    {
      "entropy": 5.5615557670593265,
      "epoch": 4.123439000960615,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034002734064990386,
      "loss": 4.7238,
      "mean_token_accuracy": 0.2580337062478065,
      "num_tokens": 98436768.0,
      "step": 42925
    },
    {
      "entropy": 5.577818965911865,
      "epoch": 4.123919308357348,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003399945193576093,
      "loss": 4.758,
      "mean_token_accuracy": 0.2477502390742302,
      "num_tokens": 98449360.0,
      "step": 42930
    },
    {
      "entropy": 5.484556436538696,
      "epoch": 4.124399615754083,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003399616965563351,
      "loss": 4.6433,
      "mean_token_accuracy": 0.25716733932495117,
      "num_tokens": 98461717.0,
      "step": 42935
    },
    {
      "entropy": 5.509142351150513,
      "epoch": 4.124879923150816,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003399288722468431,
      "loss": 4.6677,
      "mean_token_accuracy": 0.25710240602493284,
      "num_tokens": 98472186.0,
      "step": 42940
    },
    {
      "entropy": 5.5612757205963135,
      "epoch": 4.1253602305475505,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003398960464298955,
      "loss": 4.6961,
      "mean_token_accuracy": 0.25546908378601074,
      "num_tokens": 98483266.0,
      "step": 42945
    },
    {
      "entropy": 5.444724988937378,
      "epoch": 4.125840537944284,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003398632191062544,
      "loss": 4.5864,
      "mean_token_accuracy": 0.2647333204746246,
      "num_tokens": 98494984.0,
      "step": 42950
    },
    {
      "entropy": 5.4821343421936035,
      "epoch": 4.126320845341018,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000339830390276682,
      "loss": 4.6354,
      "mean_token_accuracy": 0.2619659170508385,
      "num_tokens": 98506564.0,
      "step": 42955
    },
    {
      "entropy": 5.43461799621582,
      "epoch": 4.126801152737752,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033979755994194033,
      "loss": 4.6386,
      "mean_token_accuracy": 0.2622184917330742,
      "num_tokens": 98519732.0,
      "step": 42960
    },
    {
      "entropy": 5.552219390869141,
      "epoch": 4.127281460134486,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003397647281027919,
      "loss": 4.7492,
      "mean_token_accuracy": 0.2488288924098015,
      "num_tokens": 98530271.0,
      "step": 42965
    },
    {
      "entropy": 5.479260396957398,
      "epoch": 4.12776176753122,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003397318947599987,
      "loss": 4.6079,
      "mean_token_accuracy": 0.2626535475254059,
      "num_tokens": 98540178.0,
      "step": 42970
    },
    {
      "entropy": 5.48417649269104,
      "epoch": 4.128242074927954,
      "grad_norm": 1.25,
      "learning_rate": 0.00033969905991432307,
      "loss": 4.7098,
      "mean_token_accuracy": 0.24931986182928084,
      "num_tokens": 98551513.0,
      "step": 42975
    },
    {
      "entropy": 5.548981761932373,
      "epoch": 4.128722382324688,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033966622356652746,
      "loss": 4.7641,
      "mean_token_accuracy": 0.25223190784454347,
      "num_tokens": 98563971.0,
      "step": 42980
    },
    {
      "entropy": 5.587743854522705,
      "epoch": 4.129202689721422,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003396333857173741,
      "loss": 4.7029,
      "mean_token_accuracy": 0.25290267765522,
      "num_tokens": 98575348.0,
      "step": 42985
    },
    {
      "entropy": 5.514242315292359,
      "epoch": 4.129682997118156,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003396005463676255,
      "loss": 4.6744,
      "mean_token_accuracy": 0.25381863713264463,
      "num_tokens": 98586238.0,
      "step": 42990
    },
    {
      "entropy": 5.451392793655396,
      "epoch": 4.130163304514889,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000339567705518044,
      "loss": 4.6582,
      "mean_token_accuracy": 0.25725641995668413,
      "num_tokens": 98599051.0,
      "step": 42995
    },
    {
      "entropy": 5.455086040496826,
      "epoch": 4.130643611911624,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033953486316939214,
      "loss": 4.6324,
      "mean_token_accuracy": 0.25839643478393554,
      "num_tokens": 98610526.0,
      "step": 43000
    },
    {
      "entropy": 5.434266662597656,
      "epoch": 4.131123919308357,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033950201932243237,
      "loss": 4.5147,
      "mean_token_accuracy": 0.26549834161996844,
      "num_tokens": 98621720.0,
      "step": 43005
    },
    {
      "entropy": 5.488291454315186,
      "epoch": 4.131604226705091,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033946917397792724,
      "loss": 4.6657,
      "mean_token_accuracy": 0.26154539436101915,
      "num_tokens": 98632248.0,
      "step": 43010
    },
    {
      "entropy": 5.424181270599365,
      "epoch": 4.132084534101825,
      "grad_norm": 1.25,
      "learning_rate": 0.0003394363271366393,
      "loss": 4.6819,
      "mean_token_accuracy": 0.2529924318194389,
      "num_tokens": 98643097.0,
      "step": 43015
    },
    {
      "entropy": 5.445079326629639,
      "epoch": 4.132564841498559,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003394034787993312,
      "loss": 4.5755,
      "mean_token_accuracy": 0.2683174654841423,
      "num_tokens": 98653429.0,
      "step": 43020
    },
    {
      "entropy": 5.4680811882019045,
      "epoch": 4.133045148895293,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033937062896676563,
      "loss": 4.6225,
      "mean_token_accuracy": 0.25969298779964445,
      "num_tokens": 98664972.0,
      "step": 43025
    },
    {
      "entropy": 5.481714582443237,
      "epoch": 4.133525456292027,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00033933777763970506,
      "loss": 4.721,
      "mean_token_accuracy": 0.2532467141747475,
      "num_tokens": 98675075.0,
      "step": 43030
    },
    {
      "entropy": 5.561950302124023,
      "epoch": 4.13400576368876,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003393049248189124,
      "loss": 4.7464,
      "mean_token_accuracy": 0.2501841619610786,
      "num_tokens": 98686749.0,
      "step": 43035
    },
    {
      "entropy": 5.541613960266114,
      "epoch": 4.134486071085495,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003392720705051504,
      "loss": 4.7559,
      "mean_token_accuracy": 0.2473876416683197,
      "num_tokens": 98698229.0,
      "step": 43040
    },
    {
      "entropy": 5.466405296325684,
      "epoch": 4.134966378482229,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033923921469918174,
      "loss": 4.7003,
      "mean_token_accuracy": 0.25393896400928495,
      "num_tokens": 98708302.0,
      "step": 43045
    },
    {
      "entropy": 5.484797143936158,
      "epoch": 4.1354466858789625,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003392063574017692,
      "loss": 4.6775,
      "mean_token_accuracy": 0.25860474109649656,
      "num_tokens": 98719882.0,
      "step": 43050
    },
    {
      "entropy": 5.5040850162506105,
      "epoch": 4.135926993275697,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033917349861367575,
      "loss": 4.6625,
      "mean_token_accuracy": 0.260753308236599,
      "num_tokens": 98730123.0,
      "step": 43055
    },
    {
      "entropy": 5.4734358310699465,
      "epoch": 4.13640730067243,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003391406383356642,
      "loss": 4.66,
      "mean_token_accuracy": 0.2617115259170532,
      "num_tokens": 98739605.0,
      "step": 43060
    },
    {
      "entropy": 5.47500433921814,
      "epoch": 4.136887608069165,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033910777656849745,
      "loss": 4.6936,
      "mean_token_accuracy": 0.2583306357264519,
      "num_tokens": 98752474.0,
      "step": 43065
    },
    {
      "entropy": 5.586598205566406,
      "epoch": 4.137367915465898,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003390749133129386,
      "loss": 4.7596,
      "mean_token_accuracy": 0.25142938643693924,
      "num_tokens": 98764982.0,
      "step": 43070
    },
    {
      "entropy": 5.4864692211151125,
      "epoch": 4.137848222862632,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003390420485697504,
      "loss": 4.662,
      "mean_token_accuracy": 0.2588947400450706,
      "num_tokens": 98777464.0,
      "step": 43075
    },
    {
      "entropy": 5.538907194137574,
      "epoch": 4.138328530259366,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033900918233969606,
      "loss": 4.6809,
      "mean_token_accuracy": 0.2544194430112839,
      "num_tokens": 98788672.0,
      "step": 43080
    },
    {
      "entropy": 5.630599069595337,
      "epoch": 4.1388088376561,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033897631462353854,
      "loss": 4.8547,
      "mean_token_accuracy": 0.2412879705429077,
      "num_tokens": 98801531.0,
      "step": 43085
    },
    {
      "entropy": 5.492377281188965,
      "epoch": 4.1392891450528335,
      "grad_norm": 1.3125,
      "learning_rate": 0.000338943445422041,
      "loss": 4.654,
      "mean_token_accuracy": 0.2579874262213707,
      "num_tokens": 98811534.0,
      "step": 43090
    },
    {
      "entropy": 5.569633531570434,
      "epoch": 4.139769452449568,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033891057473596655,
      "loss": 4.7251,
      "mean_token_accuracy": 0.25743876695632933,
      "num_tokens": 98823220.0,
      "step": 43095
    },
    {
      "entropy": 5.460314607620239,
      "epoch": 4.140249759846301,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003388777025660783,
      "loss": 4.6123,
      "mean_token_accuracy": 0.25761880427598954,
      "num_tokens": 98834389.0,
      "step": 43100
    },
    {
      "entropy": 5.519323635101318,
      "epoch": 4.140730067243036,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0003388448289131395,
      "loss": 4.7135,
      "mean_token_accuracy": 0.2522661581635475,
      "num_tokens": 98846399.0,
      "step": 43105
    },
    {
      "entropy": 5.588692140579224,
      "epoch": 4.141210374639769,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033881195377791337,
      "loss": 4.7903,
      "mean_token_accuracy": 0.24622575491666793,
      "num_tokens": 98858448.0,
      "step": 43110
    },
    {
      "entropy": 5.409054327011108,
      "epoch": 4.141690682036503,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033877907716116314,
      "loss": 4.5344,
      "mean_token_accuracy": 0.26499540358781815,
      "num_tokens": 98871113.0,
      "step": 43115
    },
    {
      "entropy": 5.5210460186004635,
      "epoch": 4.142170989433238,
      "grad_norm": 1.25,
      "learning_rate": 0.000338746199063652,
      "loss": 4.6799,
      "mean_token_accuracy": 0.25308232605457304,
      "num_tokens": 98881779.0,
      "step": 43120
    },
    {
      "entropy": 5.477627372741699,
      "epoch": 4.142651296829971,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003387133194861435,
      "loss": 4.6583,
      "mean_token_accuracy": 0.25559635907411576,
      "num_tokens": 98893490.0,
      "step": 43125
    },
    {
      "entropy": 5.5806640625,
      "epoch": 4.1431316042267055,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033868043842940097,
      "loss": 4.752,
      "mean_token_accuracy": 0.24833920001983642,
      "num_tokens": 98904485.0,
      "step": 43130
    },
    {
      "entropy": 5.51014461517334,
      "epoch": 4.143611911623439,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033864755589418764,
      "loss": 4.6713,
      "mean_token_accuracy": 0.25411323755979537,
      "num_tokens": 98916276.0,
      "step": 43135
    },
    {
      "entropy": 5.45262336730957,
      "epoch": 4.144092219020173,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033861467188126714,
      "loss": 4.6944,
      "mean_token_accuracy": 0.2603691205382347,
      "num_tokens": 98927891.0,
      "step": 43140
    },
    {
      "entropy": 5.519540786743164,
      "epoch": 4.144572526416907,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033858178639140286,
      "loss": 4.6663,
      "mean_token_accuracy": 0.2609233185648918,
      "num_tokens": 98939396.0,
      "step": 43145
    },
    {
      "entropy": 5.488612174987793,
      "epoch": 4.145052833813641,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003385488994253583,
      "loss": 4.631,
      "mean_token_accuracy": 0.2649221420288086,
      "num_tokens": 98950327.0,
      "step": 43150
    },
    {
      "entropy": 5.53660945892334,
      "epoch": 4.145533141210374,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000338516010983897,
      "loss": 4.7332,
      "mean_token_accuracy": 0.25753948241472246,
      "num_tokens": 98961867.0,
      "step": 43155
    },
    {
      "entropy": 5.513646221160888,
      "epoch": 4.146013448607109,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033848312106778247,
      "loss": 4.6887,
      "mean_token_accuracy": 0.2505826517939568,
      "num_tokens": 98972998.0,
      "step": 43160
    },
    {
      "entropy": 5.505939483642578,
      "epoch": 4.146493756003842,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033845022967777833,
      "loss": 4.7623,
      "mean_token_accuracy": 0.24711156338453294,
      "num_tokens": 98986898.0,
      "step": 43165
    },
    {
      "entropy": 5.555199241638183,
      "epoch": 4.1469740634005765,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033841733681464837,
      "loss": 4.6923,
      "mean_token_accuracy": 0.2603279545903206,
      "num_tokens": 98997881.0,
      "step": 43170
    },
    {
      "entropy": 5.438412523269653,
      "epoch": 4.14745437079731,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033838444247915613,
      "loss": 4.6289,
      "mean_token_accuracy": 0.2609734803438187,
      "num_tokens": 99012328.0,
      "step": 43175
    },
    {
      "entropy": 5.528254842758178,
      "epoch": 4.147934678194044,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003383515466720653,
      "loss": 4.7373,
      "mean_token_accuracy": 0.24781260788440704,
      "num_tokens": 99023383.0,
      "step": 43180
    },
    {
      "entropy": 5.514659357070923,
      "epoch": 4.148414985590778,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033831864939413975,
      "loss": 4.6691,
      "mean_token_accuracy": 0.25689933598041537,
      "num_tokens": 99034264.0,
      "step": 43185
    },
    {
      "entropy": 5.506821012496948,
      "epoch": 4.148895292987512,
      "grad_norm": 1.25,
      "learning_rate": 0.0003382857506461432,
      "loss": 4.6642,
      "mean_token_accuracy": 0.25968800485134125,
      "num_tokens": 99046593.0,
      "step": 43190
    },
    {
      "entropy": 5.451734447479248,
      "epoch": 4.149375600384246,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033825285042883937,
      "loss": 4.6853,
      "mean_token_accuracy": 0.2589380621910095,
      "num_tokens": 99059068.0,
      "step": 43195
    },
    {
      "entropy": 5.558466529846191,
      "epoch": 4.14985590778098,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033821994874299216,
      "loss": 4.7988,
      "mean_token_accuracy": 0.24539484977722167,
      "num_tokens": 99070547.0,
      "step": 43200
    },
    {
      "entropy": 5.536341428756714,
      "epoch": 4.150336215177714,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003381870455893655,
      "loss": 4.7027,
      "mean_token_accuracy": 0.2632414221763611,
      "num_tokens": 99082118.0,
      "step": 43205
    },
    {
      "entropy": 5.477193069458008,
      "epoch": 4.1508165225744476,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033815414096872326,
      "loss": 4.6148,
      "mean_token_accuracy": 0.2545272409915924,
      "num_tokens": 99093514.0,
      "step": 43210
    },
    {
      "entropy": 5.479601907730102,
      "epoch": 4.151296829971182,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003381212348818294,
      "loss": 4.6597,
      "mean_token_accuracy": 0.2583903193473816,
      "num_tokens": 99104309.0,
      "step": 43215
    },
    {
      "entropy": 5.451570796966553,
      "epoch": 4.151777137367915,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003380883273294479,
      "loss": 4.6861,
      "mean_token_accuracy": 0.25401201397180556,
      "num_tokens": 99116503.0,
      "step": 43220
    },
    {
      "entropy": 5.49790325164795,
      "epoch": 4.15225744476465,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033805541831234275,
      "loss": 4.648,
      "mean_token_accuracy": 0.2567359820008278,
      "num_tokens": 99127212.0,
      "step": 43225
    },
    {
      "entropy": 5.5020514011383055,
      "epoch": 4.152737752161383,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000338022507831278,
      "loss": 4.6516,
      "mean_token_accuracy": 0.25775266587734225,
      "num_tokens": 99137947.0,
      "step": 43230
    },
    {
      "entropy": 5.538798856735229,
      "epoch": 4.153218059558117,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033798959588701786,
      "loss": 4.7022,
      "mean_token_accuracy": 0.25258989781141283,
      "num_tokens": 99150084.0,
      "step": 43235
    },
    {
      "entropy": 5.495609283447266,
      "epoch": 4.153698366954851,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003379566824803263,
      "loss": 4.6871,
      "mean_token_accuracy": 0.2537431433796883,
      "num_tokens": 99160126.0,
      "step": 43240
    },
    {
      "entropy": 5.464425802230835,
      "epoch": 4.154178674351585,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003379237676119675,
      "loss": 4.7106,
      "mean_token_accuracy": 0.24690164625644684,
      "num_tokens": 99172349.0,
      "step": 43245
    },
    {
      "entropy": 5.45983853340149,
      "epoch": 4.154658981748319,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003378908512827057,
      "loss": 4.6093,
      "mean_token_accuracy": 0.256497161090374,
      "num_tokens": 99183826.0,
      "step": 43250
    },
    {
      "entropy": 5.487994623184204,
      "epoch": 4.155139289145053,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033785793349330495,
      "loss": 4.6515,
      "mean_token_accuracy": 0.25864339619874954,
      "num_tokens": 99195713.0,
      "step": 43255
    },
    {
      "entropy": 5.537251424789429,
      "epoch": 4.155619596541786,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033782501424452975,
      "loss": 4.7337,
      "mean_token_accuracy": 0.2543799117207527,
      "num_tokens": 99206544.0,
      "step": 43260
    },
    {
      "entropy": 5.396960496902466,
      "epoch": 4.156099903938521,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033779209353714424,
      "loss": 4.5471,
      "mean_token_accuracy": 0.27446779906749724,
      "num_tokens": 99218073.0,
      "step": 43265
    },
    {
      "entropy": 5.475465965270996,
      "epoch": 4.156580211335255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003377591713719128,
      "loss": 4.6637,
      "mean_token_accuracy": 0.2525331437587738,
      "num_tokens": 99228627.0,
      "step": 43270
    },
    {
      "entropy": 5.506232166290284,
      "epoch": 4.1570605187319885,
      "grad_norm": 1.25,
      "learning_rate": 0.0003377262477495997,
      "loss": 4.6613,
      "mean_token_accuracy": 0.26358873546123507,
      "num_tokens": 99240863.0,
      "step": 43275
    },
    {
      "entropy": 5.537413167953491,
      "epoch": 4.157540826128723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003376933226709694,
      "loss": 4.7145,
      "mean_token_accuracy": 0.25228308588266374,
      "num_tokens": 99253255.0,
      "step": 43280
    },
    {
      "entropy": 5.53718409538269,
      "epoch": 4.158021133525456,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033766039613678634,
      "loss": 4.6948,
      "mean_token_accuracy": 0.25893714427948,
      "num_tokens": 99264850.0,
      "step": 43285
    },
    {
      "entropy": 5.456399869918823,
      "epoch": 4.1585014409221905,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033762746814781496,
      "loss": 4.5842,
      "mean_token_accuracy": 0.2676131188869476,
      "num_tokens": 99275389.0,
      "step": 43290
    },
    {
      "entropy": 5.436407804489136,
      "epoch": 4.158981748318924,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003375945387048197,
      "loss": 4.665,
      "mean_token_accuracy": 0.25336011201143266,
      "num_tokens": 99286265.0,
      "step": 43295
    },
    {
      "entropy": 5.458422517776489,
      "epoch": 4.159462055715658,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003375616078085651,
      "loss": 4.7041,
      "mean_token_accuracy": 0.2515016943216324,
      "num_tokens": 99298382.0,
      "step": 43300
    },
    {
      "entropy": 5.52871675491333,
      "epoch": 4.159942363112392,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033752867545981576,
      "loss": 4.6896,
      "mean_token_accuracy": 0.25385682284832,
      "num_tokens": 99308706.0,
      "step": 43305
    },
    {
      "entropy": 5.498509550094605,
      "epoch": 4.160422670509126,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033749574165933626,
      "loss": 4.6331,
      "mean_token_accuracy": 0.25693808048963546,
      "num_tokens": 99318952.0,
      "step": 43310
    },
    {
      "entropy": 5.490941524505615,
      "epoch": 4.1609029779058595,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003374628064078912,
      "loss": 4.7071,
      "mean_token_accuracy": 0.25066235810518267,
      "num_tokens": 99330857.0,
      "step": 43315
    },
    {
      "entropy": 5.42292332649231,
      "epoch": 4.161383285302594,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033742986970624526,
      "loss": 4.5171,
      "mean_token_accuracy": 0.2713790759444237,
      "num_tokens": 99341214.0,
      "step": 43320
    },
    {
      "entropy": 5.483273267745972,
      "epoch": 4.161863592699327,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033739693155516315,
      "loss": 4.6527,
      "mean_token_accuracy": 0.2572027429938316,
      "num_tokens": 99353312.0,
      "step": 43325
    },
    {
      "entropy": 5.541375780105591,
      "epoch": 4.162343900096062,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033736399195540955,
      "loss": 4.6875,
      "mean_token_accuracy": 0.25509467124938967,
      "num_tokens": 99363965.0,
      "step": 43330
    },
    {
      "entropy": 5.525278997421265,
      "epoch": 4.162824207492795,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033733105090774917,
      "loss": 4.6338,
      "mean_token_accuracy": 0.2664958521723747,
      "num_tokens": 99375027.0,
      "step": 43335
    },
    {
      "entropy": 5.593852233886719,
      "epoch": 4.163304514889529,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000337298108412947,
      "loss": 4.8088,
      "mean_token_accuracy": 0.24160257279872893,
      "num_tokens": 99385998.0,
      "step": 43340
    },
    {
      "entropy": 5.434388303756714,
      "epoch": 4.163784822286263,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033726516447176764,
      "loss": 4.597,
      "mean_token_accuracy": 0.26296195685863494,
      "num_tokens": 99396358.0,
      "step": 43345
    },
    {
      "entropy": 5.523823070526123,
      "epoch": 4.164265129682997,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003372322190849761,
      "loss": 4.751,
      "mean_token_accuracy": 0.25105679482221605,
      "num_tokens": 99407289.0,
      "step": 43350
    },
    {
      "entropy": 5.517546224594116,
      "epoch": 4.164745437079731,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003371992722533372,
      "loss": 4.7313,
      "mean_token_accuracy": 0.25423239916563034,
      "num_tokens": 99419036.0,
      "step": 43355
    },
    {
      "entropy": 5.5534743309021,
      "epoch": 4.165225744476465,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003371663239776159,
      "loss": 4.649,
      "mean_token_accuracy": 0.24892188161611556,
      "num_tokens": 99429589.0,
      "step": 43360
    },
    {
      "entropy": 5.489415693283081,
      "epoch": 4.165706051873199,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033713337425857713,
      "loss": 4.6246,
      "mean_token_accuracy": 0.2556702718138695,
      "num_tokens": 99440398.0,
      "step": 43365
    },
    {
      "entropy": 5.484642553329468,
      "epoch": 4.166186359269933,
      "grad_norm": 1.203125,
      "learning_rate": 0.000337100423096986,
      "loss": 4.6289,
      "mean_token_accuracy": 0.25964813977479934,
      "num_tokens": 99452000.0,
      "step": 43370
    },
    {
      "entropy": 5.559202766418457,
      "epoch": 4.166666666666667,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003370674704936074,
      "loss": 4.6815,
      "mean_token_accuracy": 0.25492479652166367,
      "num_tokens": 99462403.0,
      "step": 43375
    },
    {
      "entropy": 5.4348366260528564,
      "epoch": 4.1671469740634,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033703451644920637,
      "loss": 4.5677,
      "mean_token_accuracy": 0.2633687868714333,
      "num_tokens": 99474284.0,
      "step": 43380
    },
    {
      "entropy": 5.48563404083252,
      "epoch": 4.167627281460135,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033700156096454816,
      "loss": 4.733,
      "mean_token_accuracy": 0.2532726556062698,
      "num_tokens": 99486189.0,
      "step": 43385
    },
    {
      "entropy": 5.532251453399658,
      "epoch": 4.168107588856868,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033696860404039777,
      "loss": 4.7588,
      "mean_token_accuracy": 0.2482762634754181,
      "num_tokens": 99498047.0,
      "step": 43390
    },
    {
      "entropy": 5.444976758956909,
      "epoch": 4.1685878962536025,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033693564567752036,
      "loss": 4.629,
      "mean_token_accuracy": 0.2557430773973465,
      "num_tokens": 99509523.0,
      "step": 43395
    },
    {
      "entropy": 5.488572454452514,
      "epoch": 4.169068203650336,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003369026858766813,
      "loss": 4.6061,
      "mean_token_accuracy": 0.25546724945306776,
      "num_tokens": 99521357.0,
      "step": 43400
    },
    {
      "entropy": 5.493000078201294,
      "epoch": 4.16954851104707,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033686972463864553,
      "loss": 4.7133,
      "mean_token_accuracy": 0.252421560883522,
      "num_tokens": 99532484.0,
      "step": 43405
    },
    {
      "entropy": 5.556303787231445,
      "epoch": 4.170028818443804,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003368367619641785,
      "loss": 4.6572,
      "mean_token_accuracy": 0.2651986196637154,
      "num_tokens": 99543082.0,
      "step": 43410
    },
    {
      "entropy": 5.4952672004699705,
      "epoch": 4.170509125840538,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003368037978540455,
      "loss": 4.7236,
      "mean_token_accuracy": 0.24802924692630768,
      "num_tokens": 99555380.0,
      "step": 43415
    },
    {
      "entropy": 5.52055082321167,
      "epoch": 4.1709894332372714,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003367708323090119,
      "loss": 4.7417,
      "mean_token_accuracy": 0.2546717807650566,
      "num_tokens": 99566394.0,
      "step": 43420
    },
    {
      "entropy": 5.454279613494873,
      "epoch": 4.171469740634006,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000336737865329843,
      "loss": 4.665,
      "mean_token_accuracy": 0.2532691940665245,
      "num_tokens": 99579156.0,
      "step": 43425
    },
    {
      "entropy": 5.501789855957031,
      "epoch": 4.17195004803074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033670489691730405,
      "loss": 4.716,
      "mean_token_accuracy": 0.252107672393322,
      "num_tokens": 99589899.0,
      "step": 43430
    },
    {
      "entropy": 5.543061780929565,
      "epoch": 4.1724303554274735,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003366719270721607,
      "loss": 4.7309,
      "mean_token_accuracy": 0.2494884416460991,
      "num_tokens": 99602337.0,
      "step": 43435
    },
    {
      "entropy": 5.550251770019531,
      "epoch": 4.172910662824208,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033663895579517835,
      "loss": 4.7628,
      "mean_token_accuracy": 0.2460479184985161,
      "num_tokens": 99613375.0,
      "step": 43440
    },
    {
      "entropy": 5.485026931762695,
      "epoch": 4.173390970220941,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033660598308712247,
      "loss": 4.6198,
      "mean_token_accuracy": 0.2666208744049072,
      "num_tokens": 99624156.0,
      "step": 43445
    },
    {
      "entropy": 5.4403345584869385,
      "epoch": 4.173871277617676,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033657300894875857,
      "loss": 4.5914,
      "mean_token_accuracy": 0.2615476220846176,
      "num_tokens": 99635360.0,
      "step": 43450
    },
    {
      "entropy": 5.519686222076416,
      "epoch": 4.174351585014409,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033654003338085225,
      "loss": 4.6108,
      "mean_token_accuracy": 0.26603085845708846,
      "num_tokens": 99646479.0,
      "step": 43455
    },
    {
      "entropy": 5.461757040023803,
      "epoch": 4.174831892411143,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033650705638416914,
      "loss": 4.6669,
      "mean_token_accuracy": 0.2556250154972076,
      "num_tokens": 99656688.0,
      "step": 43460
    },
    {
      "entropy": 5.438313293457031,
      "epoch": 4.175312199807877,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003364740779594747,
      "loss": 4.6426,
      "mean_token_accuracy": 0.2605939909815788,
      "num_tokens": 99667859.0,
      "step": 43465
    },
    {
      "entropy": 5.514748191833496,
      "epoch": 4.175792507204611,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033644109810753483,
      "loss": 4.675,
      "mean_token_accuracy": 0.2556849762797356,
      "num_tokens": 99679012.0,
      "step": 43470
    },
    {
      "entropy": 5.488751029968261,
      "epoch": 4.176272814601345,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000336408116829115,
      "loss": 4.6536,
      "mean_token_accuracy": 0.2516706004738808,
      "num_tokens": 99689734.0,
      "step": 43475
    },
    {
      "entropy": 5.444997787475586,
      "epoch": 4.176753121998079,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033637513412498117,
      "loss": 4.649,
      "mean_token_accuracy": 0.2611166566610336,
      "num_tokens": 99700445.0,
      "step": 43480
    },
    {
      "entropy": 5.5047403335571286,
      "epoch": 4.177233429394812,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003363421499958989,
      "loss": 4.7264,
      "mean_token_accuracy": 0.2515763819217682,
      "num_tokens": 99711317.0,
      "step": 43485
    },
    {
      "entropy": 5.463526916503906,
      "epoch": 4.177713736791547,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033630916444263407,
      "loss": 4.5969,
      "mean_token_accuracy": 0.2638493850827217,
      "num_tokens": 99722355.0,
      "step": 43490
    },
    {
      "entropy": 5.508136224746704,
      "epoch": 4.17819404418828,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003362761774659525,
      "loss": 4.6694,
      "mean_token_accuracy": 0.26058436632156373,
      "num_tokens": 99733010.0,
      "step": 43495
    },
    {
      "entropy": 5.520628976821899,
      "epoch": 4.178674351585014,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033624318906662,
      "loss": 4.7239,
      "mean_token_accuracy": 0.25135488659143446,
      "num_tokens": 99744724.0,
      "step": 43500
    },
    {
      "entropy": 5.558080625534058,
      "epoch": 4.179154658981748,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033621019924540255,
      "loss": 4.761,
      "mean_token_accuracy": 0.24501983374357222,
      "num_tokens": 99756772.0,
      "step": 43505
    },
    {
      "entropy": 5.5426891326904295,
      "epoch": 4.179634966378482,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000336177208003066,
      "loss": 4.734,
      "mean_token_accuracy": 0.25285973846912385,
      "num_tokens": 99768398.0,
      "step": 43510
    },
    {
      "entropy": 5.568021154403686,
      "epoch": 4.1801152737752165,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003361442153403763,
      "loss": 4.8338,
      "mean_token_accuracy": 0.2434015080332756,
      "num_tokens": 99780110.0,
      "step": 43515
    },
    {
      "entropy": 5.519728517532348,
      "epoch": 4.18059558117195,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003361112212580995,
      "loss": 4.6285,
      "mean_token_accuracy": 0.26332264542579653,
      "num_tokens": 99791329.0,
      "step": 43520
    },
    {
      "entropy": 5.508914613723755,
      "epoch": 4.181075888568684,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003360782257570017,
      "loss": 4.7392,
      "mean_token_accuracy": 0.24570897668600084,
      "num_tokens": 99802890.0,
      "step": 43525
    },
    {
      "entropy": 5.508522891998291,
      "epoch": 4.181556195965418,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003360452288378488,
      "loss": 4.6569,
      "mean_token_accuracy": 0.2580455124378204,
      "num_tokens": 99814076.0,
      "step": 43530
    },
    {
      "entropy": 5.517406463623047,
      "epoch": 4.182036503362152,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033601223050140687,
      "loss": 4.6571,
      "mean_token_accuracy": 0.2577746227383614,
      "num_tokens": 99825894.0,
      "step": 43535
    },
    {
      "entropy": 5.533355808258056,
      "epoch": 4.1825168107588855,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003359792307484423,
      "loss": 4.6602,
      "mean_token_accuracy": 0.25946532338857653,
      "num_tokens": 99837217.0,
      "step": 43540
    },
    {
      "entropy": 5.6051819801330565,
      "epoch": 4.18299711815562,
      "grad_norm": 1.375,
      "learning_rate": 0.00033594622957972096,
      "loss": 4.8203,
      "mean_token_accuracy": 0.24639834016561507,
      "num_tokens": 99848728.0,
      "step": 43545
    },
    {
      "entropy": 5.450102424621582,
      "epoch": 4.183477425552353,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003359132269960091,
      "loss": 4.6332,
      "mean_token_accuracy": 0.2685096234083176,
      "num_tokens": 99860116.0,
      "step": 43550
    },
    {
      "entropy": 5.52916145324707,
      "epoch": 4.183957732949088,
      "grad_norm": 1.28125,
      "learning_rate": 0.000335880222998073,
      "loss": 4.6461,
      "mean_token_accuracy": 0.2580014273524284,
      "num_tokens": 99870720.0,
      "step": 43555
    },
    {
      "entropy": 5.474533462524414,
      "epoch": 4.184438040345821,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003358472175866789,
      "loss": 4.6426,
      "mean_token_accuracy": 0.2583266064524651,
      "num_tokens": 99881882.0,
      "step": 43560
    },
    {
      "entropy": 5.611022472381592,
      "epoch": 4.184918347742555,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003358142107625931,
      "loss": 4.7355,
      "mean_token_accuracy": 0.2540289252996445,
      "num_tokens": 99892845.0,
      "step": 43565
    },
    {
      "entropy": 5.467004013061524,
      "epoch": 4.185398655139289,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003357812025265819,
      "loss": 4.6601,
      "mean_token_accuracy": 0.2572259560227394,
      "num_tokens": 99905024.0,
      "step": 43570
    },
    {
      "entropy": 5.543931198120117,
      "epoch": 4.185878962536023,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033574819287941174,
      "loss": 4.7197,
      "mean_token_accuracy": 0.2538453355431557,
      "num_tokens": 99917899.0,
      "step": 43575
    },
    {
      "entropy": 5.5160938739776615,
      "epoch": 4.1863592699327565,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033571518182184886,
      "loss": 4.6854,
      "mean_token_accuracy": 0.25182845145463945,
      "num_tokens": 99929245.0,
      "step": 43580
    },
    {
      "entropy": 5.54601263999939,
      "epoch": 4.186839577329491,
      "grad_norm": 1.25,
      "learning_rate": 0.00033568216935465976,
      "loss": 4.7622,
      "mean_token_accuracy": 0.2515923991799355,
      "num_tokens": 99940806.0,
      "step": 43585
    },
    {
      "entropy": 5.517718982696533,
      "epoch": 4.187319884726225,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003356491554786109,
      "loss": 4.6707,
      "mean_token_accuracy": 0.2559166863560677,
      "num_tokens": 99953369.0,
      "step": 43590
    },
    {
      "entropy": 5.500888347625732,
      "epoch": 4.187800192122959,
      "grad_norm": 1.25,
      "learning_rate": 0.00033561614019446867,
      "loss": 4.6898,
      "mean_token_accuracy": 0.25542817264795303,
      "num_tokens": 99964563.0,
      "step": 43595
    },
    {
      "entropy": 5.521173143386841,
      "epoch": 4.188280499519693,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033558312350299963,
      "loss": 4.7414,
      "mean_token_accuracy": 0.2532419592142105,
      "num_tokens": 99975143.0,
      "step": 43600
    },
    {
      "entropy": 5.645272922515869,
      "epoch": 4.188760806916426,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033555010540497045,
      "loss": 4.768,
      "mean_token_accuracy": 0.24832975566387178,
      "num_tokens": 99987430.0,
      "step": 43605
    },
    {
      "entropy": 5.4194801330566404,
      "epoch": 4.189241114313161,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003355170859011476,
      "loss": 4.615,
      "mean_token_accuracy": 0.2703607067465782,
      "num_tokens": 99998638.0,
      "step": 43610
    },
    {
      "entropy": 5.510542726516723,
      "epoch": 4.189721421709894,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003354840649922977,
      "loss": 4.7432,
      "mean_token_accuracy": 0.24767607003450393,
      "num_tokens": 100011054.0,
      "step": 43615
    },
    {
      "entropy": 5.610476970672607,
      "epoch": 4.1902017291066285,
      "grad_norm": 1.25,
      "learning_rate": 0.0003354510426791874,
      "loss": 4.8,
      "mean_token_accuracy": 0.24953981190919877,
      "num_tokens": 100022695.0,
      "step": 43620
    },
    {
      "entropy": 5.533863735198975,
      "epoch": 4.190682036503362,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033541801896258344,
      "loss": 4.6923,
      "mean_token_accuracy": 0.25160864144563677,
      "num_tokens": 100034402.0,
      "step": 43625
    },
    {
      "entropy": 5.554067659378052,
      "epoch": 4.191162343900096,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003353849938432524,
      "loss": 4.7811,
      "mean_token_accuracy": 0.2500628262758255,
      "num_tokens": 100046429.0,
      "step": 43630
    },
    {
      "entropy": 5.42811336517334,
      "epoch": 4.19164265129683,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003353519673219612,
      "loss": 4.6127,
      "mean_token_accuracy": 0.26224776804447175,
      "num_tokens": 100057019.0,
      "step": 43635
    },
    {
      "entropy": 5.550305414199829,
      "epoch": 4.192122958693564,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003353189393994765,
      "loss": 4.7724,
      "mean_token_accuracy": 0.2531927078962326,
      "num_tokens": 100069763.0,
      "step": 43640
    },
    {
      "entropy": 5.4554407596588135,
      "epoch": 4.192603266090297,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033528591007656516,
      "loss": 4.6088,
      "mean_token_accuracy": 0.26081993728876113,
      "num_tokens": 100082838.0,
      "step": 43645
    },
    {
      "entropy": 5.513876867294312,
      "epoch": 4.193083573487032,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003352528793539939,
      "loss": 4.7118,
      "mean_token_accuracy": 0.25527664870023725,
      "num_tokens": 100095422.0,
      "step": 43650
    },
    {
      "entropy": 5.4633095264434814,
      "epoch": 4.193563880883765,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033521984723252973,
      "loss": 4.656,
      "mean_token_accuracy": 0.26042686998844145,
      "num_tokens": 100107000.0,
      "step": 43655
    },
    {
      "entropy": 5.49920802116394,
      "epoch": 4.1940441882804995,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033518681371293965,
      "loss": 4.7119,
      "mean_token_accuracy": 0.2562492981553078,
      "num_tokens": 100121086.0,
      "step": 43660
    },
    {
      "entropy": 5.445305204391479,
      "epoch": 4.194524495677234,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033515377879599035,
      "loss": 4.5957,
      "mean_token_accuracy": 0.2614863947033882,
      "num_tokens": 100131476.0,
      "step": 43665
    },
    {
      "entropy": 5.53065071105957,
      "epoch": 4.195004803073967,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033512074248244895,
      "loss": 4.6587,
      "mean_token_accuracy": 0.2545874297618866,
      "num_tokens": 100142461.0,
      "step": 43670
    },
    {
      "entropy": 5.445517730712891,
      "epoch": 4.195485110470702,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003350877047730825,
      "loss": 4.5981,
      "mean_token_accuracy": 0.26303073167800906,
      "num_tokens": 100154317.0,
      "step": 43675
    },
    {
      "entropy": 5.522095966339111,
      "epoch": 4.195965417867435,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033505466566865783,
      "loss": 4.6354,
      "mean_token_accuracy": 0.26425004452466966,
      "num_tokens": 100165197.0,
      "step": 43680
    },
    {
      "entropy": 5.463684606552124,
      "epoch": 4.196445725264169,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003350216251699422,
      "loss": 4.6422,
      "mean_token_accuracy": 0.2577901318669319,
      "num_tokens": 100176745.0,
      "step": 43685
    },
    {
      "entropy": 5.451168203353882,
      "epoch": 4.196926032660903,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033498858327770276,
      "loss": 4.6535,
      "mean_token_accuracy": 0.25664073079824445,
      "num_tokens": 100188202.0,
      "step": 43690
    },
    {
      "entropy": 5.51883282661438,
      "epoch": 4.197406340057637,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033495553999270643,
      "loss": 4.7167,
      "mean_token_accuracy": 0.25556024014949796,
      "num_tokens": 100199044.0,
      "step": 43695
    },
    {
      "entropy": 5.459451484680176,
      "epoch": 4.197886647454371,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033492249531572054,
      "loss": 4.6462,
      "mean_token_accuracy": 0.25705484598875045,
      "num_tokens": 100210272.0,
      "step": 43700
    },
    {
      "entropy": 5.44035758972168,
      "epoch": 4.198366954851105,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003348894492475123,
      "loss": 4.5973,
      "mean_token_accuracy": 0.26651277393102646,
      "num_tokens": 100220737.0,
      "step": 43705
    },
    {
      "entropy": 5.590808296203614,
      "epoch": 4.198847262247838,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003348564017888488,
      "loss": 4.779,
      "mean_token_accuracy": 0.25458365082740786,
      "num_tokens": 100232330.0,
      "step": 43710
    },
    {
      "entropy": 5.481125259399414,
      "epoch": 4.199327569644573,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033482335294049735,
      "loss": 4.6176,
      "mean_token_accuracy": 0.2655268281698227,
      "num_tokens": 100243092.0,
      "step": 43715
    },
    {
      "entropy": 5.559219169616699,
      "epoch": 4.199807877041306,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003347903027032254,
      "loss": 4.7722,
      "mean_token_accuracy": 0.2503727152943611,
      "num_tokens": 100254820.0,
      "step": 43720
    },
    {
      "entropy": 5.567133140563965,
      "epoch": 4.20028818443804,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003347572510778001,
      "loss": 4.748,
      "mean_token_accuracy": 0.2488605797290802,
      "num_tokens": 100267272.0,
      "step": 43725
    },
    {
      "entropy": 5.4776623249053955,
      "epoch": 4.200768491834774,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003347241980649889,
      "loss": 4.6297,
      "mean_token_accuracy": 0.26487537920475007,
      "num_tokens": 100279813.0,
      "step": 43730
    },
    {
      "entropy": 5.495695018768311,
      "epoch": 4.201248799231508,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033469114366555913,
      "loss": 4.6816,
      "mean_token_accuracy": 0.2549319893121719,
      "num_tokens": 100292062.0,
      "step": 43735
    },
    {
      "entropy": 5.519081115722656,
      "epoch": 4.2017291066282425,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033465808788027824,
      "loss": 4.7083,
      "mean_token_accuracy": 0.2590801864862442,
      "num_tokens": 100303595.0,
      "step": 43740
    },
    {
      "entropy": 5.423225927352905,
      "epoch": 4.202209414024976,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033462503070991374,
      "loss": 4.6418,
      "mean_token_accuracy": 0.25993315875530243,
      "num_tokens": 100315496.0,
      "step": 43745
    },
    {
      "entropy": 5.470452117919922,
      "epoch": 4.20268972142171,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003345919721552331,
      "loss": 4.6397,
      "mean_token_accuracy": 0.2598977491259575,
      "num_tokens": 100326999.0,
      "step": 43750
    },
    {
      "entropy": 5.473793363571167,
      "epoch": 4.203170028818444,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033455891221700375,
      "loss": 4.6832,
      "mean_token_accuracy": 0.2580163523554802,
      "num_tokens": 100339168.0,
      "step": 43755
    },
    {
      "entropy": 5.521692562103271,
      "epoch": 4.203650336215178,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003345258508959933,
      "loss": 4.6315,
      "mean_token_accuracy": 0.25631310492753984,
      "num_tokens": 100350522.0,
      "step": 43760
    },
    {
      "entropy": 5.441352081298828,
      "epoch": 4.2041306436119115,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003344927881929695,
      "loss": 4.6297,
      "mean_token_accuracy": 0.2587842956185341,
      "num_tokens": 100361852.0,
      "step": 43765
    },
    {
      "entropy": 5.514843940734863,
      "epoch": 4.204610951008646,
      "grad_norm": 1.375,
      "learning_rate": 0.00033445972410869955,
      "loss": 4.7208,
      "mean_token_accuracy": 0.2513153046369553,
      "num_tokens": 100372276.0,
      "step": 43770
    },
    {
      "entropy": 5.446982479095459,
      "epoch": 4.205091258405379,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003344266586439516,
      "loss": 4.6457,
      "mean_token_accuracy": 0.26421415358781813,
      "num_tokens": 100383363.0,
      "step": 43775
    },
    {
      "entropy": 5.540467929840088,
      "epoch": 4.2055715658021136,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033439359179949303,
      "loss": 4.72,
      "mean_token_accuracy": 0.25321348905563357,
      "num_tokens": 100395912.0,
      "step": 43780
    },
    {
      "entropy": 5.470946407318115,
      "epoch": 4.206051873198847,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033436052357609157,
      "loss": 4.6311,
      "mean_token_accuracy": 0.2574260398745537,
      "num_tokens": 100405798.0,
      "step": 43785
    },
    {
      "entropy": 5.489752197265625,
      "epoch": 4.206532180595581,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003343274539745151,
      "loss": 4.6638,
      "mean_token_accuracy": 0.26077390611171725,
      "num_tokens": 100416457.0,
      "step": 43790
    },
    {
      "entropy": 5.419922971725464,
      "epoch": 4.207012487992315,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003342943829955313,
      "loss": 4.7412,
      "mean_token_accuracy": 0.247335085272789,
      "num_tokens": 100427799.0,
      "step": 43795
    },
    {
      "entropy": 5.433965587615967,
      "epoch": 4.207492795389049,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033426131063990787,
      "loss": 4.6507,
      "mean_token_accuracy": 0.2564327746629715,
      "num_tokens": 100439035.0,
      "step": 43800
    },
    {
      "entropy": 5.471464729309082,
      "epoch": 4.2079731027857825,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033422823690841286,
      "loss": 4.6404,
      "mean_token_accuracy": 0.2611569672822952,
      "num_tokens": 100450490.0,
      "step": 43805
    },
    {
      "entropy": 5.456157493591308,
      "epoch": 4.208453410182517,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003341951618018141,
      "loss": 4.6299,
      "mean_token_accuracy": 0.26058926433324814,
      "num_tokens": 100461426.0,
      "step": 43810
    },
    {
      "entropy": 5.5324469089508055,
      "epoch": 4.208933717579251,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033416208532087937,
      "loss": 4.6977,
      "mean_token_accuracy": 0.2549543440341949,
      "num_tokens": 100473051.0,
      "step": 43815
    },
    {
      "entropy": 5.452414703369141,
      "epoch": 4.209414024975985,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003341290074663768,
      "loss": 4.648,
      "mean_token_accuracy": 0.25892663598060606,
      "num_tokens": 100484269.0,
      "step": 43820
    },
    {
      "entropy": 5.477554082870483,
      "epoch": 4.209894332372719,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003340959282390742,
      "loss": 4.7185,
      "mean_token_accuracy": 0.25513082295656203,
      "num_tokens": 100495928.0,
      "step": 43825
    },
    {
      "entropy": 5.477842664718628,
      "epoch": 4.210374639769452,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033406284763973964,
      "loss": 4.6458,
      "mean_token_accuracy": 0.26348865032196045,
      "num_tokens": 100507132.0,
      "step": 43830
    },
    {
      "entropy": 5.523759555816651,
      "epoch": 4.210854947166187,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003340297656691411,
      "loss": 4.7419,
      "mean_token_accuracy": 0.2583146959543228,
      "num_tokens": 100518414.0,
      "step": 43835
    },
    {
      "entropy": 5.488240098953247,
      "epoch": 4.21133525456292,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003339966823280467,
      "loss": 4.6489,
      "mean_token_accuracy": 0.25958153754472735,
      "num_tokens": 100528809.0,
      "step": 43840
    },
    {
      "entropy": 5.543964910507202,
      "epoch": 4.2118155619596545,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033396359761722453,
      "loss": 4.6883,
      "mean_token_accuracy": 0.2471017673611641,
      "num_tokens": 100539224.0,
      "step": 43845
    },
    {
      "entropy": 5.580245876312256,
      "epoch": 4.212295869356388,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033393051153744274,
      "loss": 4.7993,
      "mean_token_accuracy": 0.24539947956800462,
      "num_tokens": 100550984.0,
      "step": 43850
    },
    {
      "entropy": 5.574827527999878,
      "epoch": 4.212776176753122,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033389742408946937,
      "loss": 4.7382,
      "mean_token_accuracy": 0.24940597862005234,
      "num_tokens": 100562786.0,
      "step": 43855
    },
    {
      "entropy": 5.485958623886108,
      "epoch": 4.213256484149856,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003338643352740728,
      "loss": 4.6298,
      "mean_token_accuracy": 0.2580674812197685,
      "num_tokens": 100573966.0,
      "step": 43860
    },
    {
      "entropy": 5.57563042640686,
      "epoch": 4.21373679154659,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003338312450920211,
      "loss": 4.8035,
      "mean_token_accuracy": 0.24788830280303956,
      "num_tokens": 100586451.0,
      "step": 43865
    },
    {
      "entropy": 5.575334310531616,
      "epoch": 4.214217098943323,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003337981535440826,
      "loss": 4.7855,
      "mean_token_accuracy": 0.24810133427381514,
      "num_tokens": 100597820.0,
      "step": 43870
    },
    {
      "entropy": 5.5162207126617435,
      "epoch": 4.214697406340058,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033376506063102556,
      "loss": 4.752,
      "mean_token_accuracy": 0.2520545542240143,
      "num_tokens": 100611423.0,
      "step": 43875
    },
    {
      "entropy": 5.511425638198853,
      "epoch": 4.215177713736791,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003337319663536183,
      "loss": 4.7572,
      "mean_token_accuracy": 0.25242583453655243,
      "num_tokens": 100623220.0,
      "step": 43880
    },
    {
      "entropy": 5.506828880310058,
      "epoch": 4.2156580211335255,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003336988707126293,
      "loss": 4.6427,
      "mean_token_accuracy": 0.2653315007686615,
      "num_tokens": 100634317.0,
      "step": 43885
    },
    {
      "entropy": 5.592591571807861,
      "epoch": 4.216138328530259,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003336657737088267,
      "loss": 4.7597,
      "mean_token_accuracy": 0.2539572313427925,
      "num_tokens": 100645743.0,
      "step": 43890
    },
    {
      "entropy": 5.468446683883667,
      "epoch": 4.216618635926993,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033363267534297896,
      "loss": 4.7229,
      "mean_token_accuracy": 0.2569320097565651,
      "num_tokens": 100658240.0,
      "step": 43895
    },
    {
      "entropy": 5.555034971237182,
      "epoch": 4.217098943323728,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003335995756158548,
      "loss": 4.7518,
      "mean_token_accuracy": 0.2555656567215919,
      "num_tokens": 100669096.0,
      "step": 43900
    },
    {
      "entropy": 5.500002431869507,
      "epoch": 4.217579250720461,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033356647452822236,
      "loss": 4.6134,
      "mean_token_accuracy": 0.2620124951004982,
      "num_tokens": 100682389.0,
      "step": 43905
    },
    {
      "entropy": 5.5209314823150635,
      "epoch": 4.218059558117195,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033353337208085035,
      "loss": 4.6537,
      "mean_token_accuracy": 0.2559983253479004,
      "num_tokens": 100693896.0,
      "step": 43910
    },
    {
      "entropy": 5.481320714950561,
      "epoch": 4.218539865513929,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003335002682745072,
      "loss": 4.71,
      "mean_token_accuracy": 0.2518001437187195,
      "num_tokens": 100704997.0,
      "step": 43915
    },
    {
      "entropy": 5.508323764801025,
      "epoch": 4.219020172910663,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003334671631099615,
      "loss": 4.6831,
      "mean_token_accuracy": 0.25368653833866117,
      "num_tokens": 100716340.0,
      "step": 43920
    },
    {
      "entropy": 5.513459587097168,
      "epoch": 4.2195004803073966,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000333434056587982,
      "loss": 4.6764,
      "mean_token_accuracy": 0.2579905465245247,
      "num_tokens": 100727897.0,
      "step": 43925
    },
    {
      "entropy": 5.528485250473023,
      "epoch": 4.219980787704131,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033340094870933714,
      "loss": 4.7247,
      "mean_token_accuracy": 0.25644742846488955,
      "num_tokens": 100738848.0,
      "step": 43930
    },
    {
      "entropy": 5.465096950531006,
      "epoch": 4.220461095100864,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033336783947479566,
      "loss": 4.6697,
      "mean_token_accuracy": 0.2657225474715233,
      "num_tokens": 100750029.0,
      "step": 43935
    },
    {
      "entropy": 5.532225704193115,
      "epoch": 4.220941402497599,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033333472888512626,
      "loss": 4.7155,
      "mean_token_accuracy": 0.2516299977898598,
      "num_tokens": 100762037.0,
      "step": 43940
    },
    {
      "entropy": 5.628166580200196,
      "epoch": 4.221421709894332,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003333016169410978,
      "loss": 4.7998,
      "mean_token_accuracy": 0.2477319657802582,
      "num_tokens": 100774269.0,
      "step": 43945
    },
    {
      "entropy": 5.479964780807495,
      "epoch": 4.221902017291066,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033326850364347876,
      "loss": 4.6589,
      "mean_token_accuracy": 0.2582293048501015,
      "num_tokens": 100785940.0,
      "step": 43950
    },
    {
      "entropy": 5.473235177993774,
      "epoch": 4.2223823246878,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003332353889930381,
      "loss": 4.6988,
      "mean_token_accuracy": 0.25374770909547806,
      "num_tokens": 100797281.0,
      "step": 43955
    },
    {
      "entropy": 5.492359972000122,
      "epoch": 4.222862632084534,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003332022729905448,
      "loss": 4.7426,
      "mean_token_accuracy": 0.25205540359020234,
      "num_tokens": 100807585.0,
      "step": 43960
    },
    {
      "entropy": 5.457491493225097,
      "epoch": 4.223342939481268,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003331691556367674,
      "loss": 4.6187,
      "mean_token_accuracy": 0.26379910856485367,
      "num_tokens": 100818235.0,
      "step": 43965
    },
    {
      "entropy": 5.506855297088623,
      "epoch": 4.223823246878002,
      "grad_norm": 1.25,
      "learning_rate": 0.0003331360369324749,
      "loss": 4.7072,
      "mean_token_accuracy": 0.2546180605888367,
      "num_tokens": 100829844.0,
      "step": 43970
    },
    {
      "entropy": 5.444299602508545,
      "epoch": 4.224303554274736,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003331029168784363,
      "loss": 4.6344,
      "mean_token_accuracy": 0.2594666212797165,
      "num_tokens": 100841983.0,
      "step": 43975
    },
    {
      "entropy": 5.513641595840454,
      "epoch": 4.22478386167147,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033306979547542047,
      "loss": 4.6842,
      "mean_token_accuracy": 0.2565369427204132,
      "num_tokens": 100853683.0,
      "step": 43980
    },
    {
      "entropy": 5.488329029083252,
      "epoch": 4.225264169068204,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003330366727241965,
      "loss": 4.6567,
      "mean_token_accuracy": 0.2538486123085022,
      "num_tokens": 100865583.0,
      "step": 43985
    },
    {
      "entropy": 5.5721923351287845,
      "epoch": 4.2257444764649374,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003330035486255333,
      "loss": 4.7229,
      "mean_token_accuracy": 0.2543579488992691,
      "num_tokens": 100876486.0,
      "step": 43990
    },
    {
      "entropy": 5.464834356307984,
      "epoch": 4.226224783861672,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033297042318019995,
      "loss": 4.5956,
      "mean_token_accuracy": 0.25671605616807935,
      "num_tokens": 100887630.0,
      "step": 43995
    },
    {
      "entropy": 5.508366918563842,
      "epoch": 4.226705091258405,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003329372963889654,
      "loss": 4.6691,
      "mean_token_accuracy": 0.25626787543296814,
      "num_tokens": 100899666.0,
      "step": 44000
    },
    {
      "entropy": 5.44833812713623,
      "epoch": 4.2271853986551395,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000332904168252599,
      "loss": 4.6187,
      "mean_token_accuracy": 0.2611506193876266,
      "num_tokens": 100911066.0,
      "step": 44005
    },
    {
      "entropy": 5.417370176315307,
      "epoch": 4.227665706051873,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033287103877186966,
      "loss": 4.6417,
      "mean_token_accuracy": 0.2634035810828209,
      "num_tokens": 100923241.0,
      "step": 44010
    },
    {
      "entropy": 5.553122520446777,
      "epoch": 4.228146013448607,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003328379079475466,
      "loss": 4.7387,
      "mean_token_accuracy": 0.2560730427503586,
      "num_tokens": 100935203.0,
      "step": 44015
    },
    {
      "entropy": 5.439465141296386,
      "epoch": 4.228626320845341,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003328047757803992,
      "loss": 4.6446,
      "mean_token_accuracy": 0.2588789716362953,
      "num_tokens": 100948133.0,
      "step": 44020
    },
    {
      "entropy": 5.633664274215699,
      "epoch": 4.229106628242075,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003327716422711965,
      "loss": 4.7957,
      "mean_token_accuracy": 0.2465050458908081,
      "num_tokens": 100960448.0,
      "step": 44025
    },
    {
      "entropy": 5.562252855300903,
      "epoch": 4.2295869356388085,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033273850742070784,
      "loss": 4.7144,
      "mean_token_accuracy": 0.2588727056980133,
      "num_tokens": 100971372.0,
      "step": 44030
    },
    {
      "entropy": 5.542672634124756,
      "epoch": 4.230067243035543,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003327053712297024,
      "loss": 4.7394,
      "mean_token_accuracy": 0.24839710742235183,
      "num_tokens": 100982656.0,
      "step": 44035
    },
    {
      "entropy": 5.531379890441895,
      "epoch": 4.230547550432276,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003326722336989497,
      "loss": 4.7164,
      "mean_token_accuracy": 0.24778151214122773,
      "num_tokens": 100994410.0,
      "step": 44040
    },
    {
      "entropy": 5.527891778945923,
      "epoch": 4.231027857829011,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003326390948292189,
      "loss": 4.6929,
      "mean_token_accuracy": 0.2503281027078629,
      "num_tokens": 101007345.0,
      "step": 44045
    },
    {
      "entropy": 5.614350128173828,
      "epoch": 4.231508165225744,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003326059546212795,
      "loss": 4.8267,
      "mean_token_accuracy": 0.24763794094324112,
      "num_tokens": 101018780.0,
      "step": 44050
    },
    {
      "entropy": 5.55621075630188,
      "epoch": 4.231988472622478,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003325728130759009,
      "loss": 4.696,
      "mean_token_accuracy": 0.25887004733085633,
      "num_tokens": 101030221.0,
      "step": 44055
    },
    {
      "entropy": 5.40625033378601,
      "epoch": 4.232468780019213,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003325396701938525,
      "loss": 4.6177,
      "mean_token_accuracy": 0.26696089208126067,
      "num_tokens": 101041877.0,
      "step": 44060
    },
    {
      "entropy": 5.491464042663575,
      "epoch": 4.232949087415946,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003325065259759039,
      "loss": 4.6403,
      "mean_token_accuracy": 0.25922152400016785,
      "num_tokens": 101052331.0,
      "step": 44065
    },
    {
      "entropy": 5.456812858581543,
      "epoch": 4.23342939481268,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003324733804228244,
      "loss": 4.6135,
      "mean_token_accuracy": 0.2580410972237587,
      "num_tokens": 101063211.0,
      "step": 44070
    },
    {
      "entropy": 5.457903051376343,
      "epoch": 4.233909702209414,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033244023353538367,
      "loss": 4.6976,
      "mean_token_accuracy": 0.2603307723999023,
      "num_tokens": 101074245.0,
      "step": 44075
    },
    {
      "entropy": 5.457386541366577,
      "epoch": 4.234390009606148,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003324070853143514,
      "loss": 4.6007,
      "mean_token_accuracy": 0.26209567338228223,
      "num_tokens": 101084232.0,
      "step": 44080
    },
    {
      "entropy": 5.610564851760865,
      "epoch": 4.234870317002882,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033237393576049704,
      "loss": 4.826,
      "mean_token_accuracy": 0.23919305354356765,
      "num_tokens": 101094433.0,
      "step": 44085
    },
    {
      "entropy": 5.548173809051514,
      "epoch": 4.235350624399616,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033234078487459016,
      "loss": 4.6902,
      "mean_token_accuracy": 0.26171218007802965,
      "num_tokens": 101106895.0,
      "step": 44090
    },
    {
      "entropy": 5.546690559387207,
      "epoch": 4.235830931796349,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003323076326574006,
      "loss": 4.803,
      "mean_token_accuracy": 0.2525580570101738,
      "num_tokens": 101118572.0,
      "step": 44095
    },
    {
      "entropy": 5.559020376205444,
      "epoch": 4.236311239193084,
      "grad_norm": 1.34375,
      "learning_rate": 0.000332274479109698,
      "loss": 4.7462,
      "mean_token_accuracy": 0.24413999766111374,
      "num_tokens": 101129257.0,
      "step": 44100
    },
    {
      "entropy": 5.571775341033936,
      "epoch": 4.236791546589817,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000332241324232252,
      "loss": 4.7682,
      "mean_token_accuracy": 0.25776603519916536,
      "num_tokens": 101140025.0,
      "step": 44105
    },
    {
      "entropy": 5.508843183517456,
      "epoch": 4.2372718539865515,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033220816802583247,
      "loss": 4.6744,
      "mean_token_accuracy": 0.25979476869106294,
      "num_tokens": 101151432.0,
      "step": 44110
    },
    {
      "entropy": 5.4819458484649655,
      "epoch": 4.237752161383285,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003321750104912092,
      "loss": 4.6828,
      "mean_token_accuracy": 0.25225200206041337,
      "num_tokens": 101162624.0,
      "step": 44115
    },
    {
      "entropy": 5.507258558273316,
      "epoch": 4.238232468780019,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033214185162915185,
      "loss": 4.7413,
      "mean_token_accuracy": 0.25330639332532884,
      "num_tokens": 101175055.0,
      "step": 44120
    },
    {
      "entropy": 5.577723598480224,
      "epoch": 4.238712776176753,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033210869144043037,
      "loss": 4.7565,
      "mean_token_accuracy": 0.25035512894392015,
      "num_tokens": 101186051.0,
      "step": 44125
    },
    {
      "entropy": 5.530425500869751,
      "epoch": 4.239193083573487,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033207552992581474,
      "loss": 4.7243,
      "mean_token_accuracy": 0.25671758204698564,
      "num_tokens": 101198009.0,
      "step": 44130
    },
    {
      "entropy": 5.435327625274658,
      "epoch": 4.239673390970221,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033204236708607476,
      "loss": 4.6518,
      "mean_token_accuracy": 0.266046741604805,
      "num_tokens": 101208950.0,
      "step": 44135
    },
    {
      "entropy": 5.503330183029175,
      "epoch": 4.240153698366955,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003320092029219804,
      "loss": 4.7143,
      "mean_token_accuracy": 0.25580689311027527,
      "num_tokens": 101220673.0,
      "step": 44140
    },
    {
      "entropy": 5.520607423782349,
      "epoch": 4.240634005763689,
      "grad_norm": 1.25,
      "learning_rate": 0.0003319760374343016,
      "loss": 4.7071,
      "mean_token_accuracy": 0.2536234423518181,
      "num_tokens": 101233629.0,
      "step": 44145
    },
    {
      "entropy": 5.43081693649292,
      "epoch": 4.2411143131604225,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003319428706238085,
      "loss": 4.6067,
      "mean_token_accuracy": 0.2721107557415962,
      "num_tokens": 101245395.0,
      "step": 44150
    },
    {
      "entropy": 5.476445245742798,
      "epoch": 4.241594620557157,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003319097024912709,
      "loss": 4.6877,
      "mean_token_accuracy": 0.25334481298923495,
      "num_tokens": 101257504.0,
      "step": 44155
    },
    {
      "entropy": 5.5306854248046875,
      "epoch": 4.24207492795389,
      "grad_norm": 1.1875,
      "learning_rate": 0.000331876533037459,
      "loss": 4.7649,
      "mean_token_accuracy": 0.24223595410585402,
      "num_tokens": 101268418.0,
      "step": 44160
    },
    {
      "entropy": 5.473517036437988,
      "epoch": 4.242555235350625,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000331843362263143,
      "loss": 4.6291,
      "mean_token_accuracy": 0.2653240874409676,
      "num_tokens": 101280509.0,
      "step": 44165
    },
    {
      "entropy": 5.577295017242432,
      "epoch": 4.243035542747358,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033181019016909286,
      "loss": 4.7313,
      "mean_token_accuracy": 0.25148926079273226,
      "num_tokens": 101292869.0,
      "step": 44170
    },
    {
      "entropy": 5.546703338623047,
      "epoch": 4.243515850144092,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003317770167560788,
      "loss": 4.6491,
      "mean_token_accuracy": 0.26165668219327926,
      "num_tokens": 101304772.0,
      "step": 44175
    },
    {
      "entropy": 5.500024318695068,
      "epoch": 4.243996157540826,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033174384202487096,
      "loss": 4.6997,
      "mean_token_accuracy": 0.2509234845638275,
      "num_tokens": 101315445.0,
      "step": 44180
    },
    {
      "entropy": 5.449222373962402,
      "epoch": 4.24447646493756,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003317106659762397,
      "loss": 4.6572,
      "mean_token_accuracy": 0.26012925505638124,
      "num_tokens": 101326102.0,
      "step": 44185
    },
    {
      "entropy": 5.441334056854248,
      "epoch": 4.244956772334294,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033167748861095503,
      "loss": 4.6437,
      "mean_token_accuracy": 0.2536982700228691,
      "num_tokens": 101337419.0,
      "step": 44190
    },
    {
      "entropy": 5.581371545791626,
      "epoch": 4.245437079731028,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003316443099297874,
      "loss": 4.6881,
      "mean_token_accuracy": 0.2589432328939438,
      "num_tokens": 101348630.0,
      "step": 44195
    },
    {
      "entropy": 5.4678404331207275,
      "epoch": 4.245917387127761,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033161112993350725,
      "loss": 4.6215,
      "mean_token_accuracy": 0.2552360758185387,
      "num_tokens": 101361357.0,
      "step": 44200
    },
    {
      "entropy": 5.531602621078491,
      "epoch": 4.246397694524496,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033157794862288467,
      "loss": 4.6479,
      "mean_token_accuracy": 0.2566849336028099,
      "num_tokens": 101372155.0,
      "step": 44205
    },
    {
      "entropy": 5.413741111755371,
      "epoch": 4.24687800192123,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033154476599869006,
      "loss": 4.5622,
      "mean_token_accuracy": 0.2685359105467796,
      "num_tokens": 101383011.0,
      "step": 44210
    },
    {
      "entropy": 5.458719205856323,
      "epoch": 4.247358309317963,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000331511582061694,
      "loss": 4.6302,
      "mean_token_accuracy": 0.25655865371227266,
      "num_tokens": 101394741.0,
      "step": 44215
    },
    {
      "entropy": 5.55318169593811,
      "epoch": 4.247838616714698,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003314783968126667,
      "loss": 4.77,
      "mean_token_accuracy": 0.25234091430902483,
      "num_tokens": 101405719.0,
      "step": 44220
    },
    {
      "entropy": 5.450058126449585,
      "epoch": 4.248318924111431,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003314452102523788,
      "loss": 4.6389,
      "mean_token_accuracy": 0.256949582695961,
      "num_tokens": 101417073.0,
      "step": 44225
    },
    {
      "entropy": 5.533262825012207,
      "epoch": 4.2487992315081655,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033141202238160075,
      "loss": 4.7398,
      "mean_token_accuracy": 0.25323628187179564,
      "num_tokens": 101429674.0,
      "step": 44230
    },
    {
      "entropy": 5.522257518768311,
      "epoch": 4.249279538904899,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033137883320110296,
      "loss": 4.6761,
      "mean_token_accuracy": 0.2557082921266556,
      "num_tokens": 101441841.0,
      "step": 44235
    },
    {
      "entropy": 5.537293815612793,
      "epoch": 4.249759846301633,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003313456427116561,
      "loss": 4.7097,
      "mean_token_accuracy": 0.2522064924240112,
      "num_tokens": 101452946.0,
      "step": 44240
    },
    {
      "entropy": 5.566970443725586,
      "epoch": 4.250240153698367,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003313124509140308,
      "loss": 4.7202,
      "mean_token_accuracy": 0.2513511896133423,
      "num_tokens": 101462919.0,
      "step": 44245
    },
    {
      "entropy": 5.421511268615722,
      "epoch": 4.250720461095101,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033127925780899754,
      "loss": 4.5843,
      "mean_token_accuracy": 0.26537591964006424,
      "num_tokens": 101474718.0,
      "step": 44250
    },
    {
      "entropy": 5.443533039093017,
      "epoch": 4.2512007684918345,
      "grad_norm": 1.21875,
      "learning_rate": 0.000331246063397327,
      "loss": 4.6434,
      "mean_token_accuracy": 0.2578091472387314,
      "num_tokens": 101487179.0,
      "step": 44255
    },
    {
      "entropy": 5.45994815826416,
      "epoch": 4.251681075888569,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033121286767979,
      "loss": 4.6697,
      "mean_token_accuracy": 0.2608169078826904,
      "num_tokens": 101498816.0,
      "step": 44260
    },
    {
      "entropy": 5.4946861743927,
      "epoch": 4.252161383285302,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000331179670657157,
      "loss": 4.7028,
      "mean_token_accuracy": 0.255339777469635,
      "num_tokens": 101510945.0,
      "step": 44265
    },
    {
      "entropy": 5.497782516479492,
      "epoch": 4.252641690682037,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003311464723301989,
      "loss": 4.6483,
      "mean_token_accuracy": 0.25645890831947327,
      "num_tokens": 101522344.0,
      "step": 44270
    },
    {
      "entropy": 5.498247718811035,
      "epoch": 4.25312199807877,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003311132726996866,
      "loss": 4.6729,
      "mean_token_accuracy": 0.2579328641295433,
      "num_tokens": 101534071.0,
      "step": 44275
    },
    {
      "entropy": 5.461125326156616,
      "epoch": 4.253602305475504,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003310800717663905,
      "loss": 4.654,
      "mean_token_accuracy": 0.25893698036670687,
      "num_tokens": 101546418.0,
      "step": 44280
    },
    {
      "entropy": 5.475929927825928,
      "epoch": 4.254082612872239,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033104686953108173,
      "loss": 4.6724,
      "mean_token_accuracy": 0.25925796031951903,
      "num_tokens": 101559423.0,
      "step": 44285
    },
    {
      "entropy": 5.540446472167969,
      "epoch": 4.254562920268972,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003310136659945311,
      "loss": 4.6892,
      "mean_token_accuracy": 0.2602899566292763,
      "num_tokens": 101570308.0,
      "step": 44290
    },
    {
      "entropy": 5.464824199676514,
      "epoch": 4.255043227665706,
      "grad_norm": 1.25,
      "learning_rate": 0.00033098046115750946,
      "loss": 4.6364,
      "mean_token_accuracy": 0.2577469289302826,
      "num_tokens": 101582689.0,
      "step": 44295
    },
    {
      "entropy": 5.384541273117065,
      "epoch": 4.25552353506244,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003309472550207877,
      "loss": 4.5669,
      "mean_token_accuracy": 0.2743761256337166,
      "num_tokens": 101593675.0,
      "step": 44300
    },
    {
      "entropy": 5.460159969329834,
      "epoch": 4.256003842459174,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033091404758513686,
      "loss": 4.5934,
      "mean_token_accuracy": 0.2728377342224121,
      "num_tokens": 101604057.0,
      "step": 44305
    },
    {
      "entropy": 5.590318012237549,
      "epoch": 4.256484149855908,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033088083885132784,
      "loss": 4.6571,
      "mean_token_accuracy": 0.260494464635849,
      "num_tokens": 101615063.0,
      "step": 44310
    },
    {
      "entropy": 5.4573187828063965,
      "epoch": 4.256964457252642,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033084762882013176,
      "loss": 4.6008,
      "mean_token_accuracy": 0.2607095554471016,
      "num_tokens": 101626059.0,
      "step": 44315
    },
    {
      "entropy": 5.479970216751099,
      "epoch": 4.257444764649375,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033081441749231944,
      "loss": 4.712,
      "mean_token_accuracy": 0.24853493720293046,
      "num_tokens": 101637799.0,
      "step": 44320
    },
    {
      "entropy": 5.42863826751709,
      "epoch": 4.25792507204611,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003307812048686621,
      "loss": 4.6277,
      "mean_token_accuracy": 0.2669756978750229,
      "num_tokens": 101648863.0,
      "step": 44325
    },
    {
      "entropy": 5.47434287071228,
      "epoch": 4.258405379442843,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003307479909499309,
      "loss": 4.611,
      "mean_token_accuracy": 0.2619078129529953,
      "num_tokens": 101660144.0,
      "step": 44330
    },
    {
      "entropy": 5.536876678466797,
      "epoch": 4.2588856868395775,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003307147757368969,
      "loss": 4.6907,
      "mean_token_accuracy": 0.2542210340499878,
      "num_tokens": 101671388.0,
      "step": 44335
    },
    {
      "entropy": 5.459747219085694,
      "epoch": 4.259365994236311,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033068155923033114,
      "loss": 4.6678,
      "mean_token_accuracy": 0.26380970478057864,
      "num_tokens": 101682542.0,
      "step": 44340
    },
    {
      "entropy": 5.485137701034546,
      "epoch": 4.259846301633045,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033064834143100495,
      "loss": 4.6595,
      "mean_token_accuracy": 0.263799324631691,
      "num_tokens": 101692683.0,
      "step": 44345
    },
    {
      "entropy": 5.49869213104248,
      "epoch": 4.260326609029779,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003306151223396896,
      "loss": 4.7169,
      "mean_token_accuracy": 0.25054513812065127,
      "num_tokens": 101704566.0,
      "step": 44350
    },
    {
      "entropy": 5.535801792144776,
      "epoch": 4.260806916426513,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003305819019571562,
      "loss": 4.7208,
      "mean_token_accuracy": 0.24859009683132172,
      "num_tokens": 101716751.0,
      "step": 44355
    },
    {
      "entropy": 5.602010107040405,
      "epoch": 4.261287223823247,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033054868028417606,
      "loss": 4.755,
      "mean_token_accuracy": 0.2552390918135643,
      "num_tokens": 101728614.0,
      "step": 44360
    },
    {
      "entropy": 5.5710248947143555,
      "epoch": 4.261767531219981,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033051545732152047,
      "loss": 4.7444,
      "mean_token_accuracy": 0.24946054965257644,
      "num_tokens": 101739740.0,
      "step": 44365
    },
    {
      "entropy": 5.460970258712768,
      "epoch": 4.262247838616715,
      "grad_norm": 1.25,
      "learning_rate": 0.0003304822330699609,
      "loss": 4.6059,
      "mean_token_accuracy": 0.2692405581474304,
      "num_tokens": 101750971.0,
      "step": 44370
    },
    {
      "entropy": 5.483867597579956,
      "epoch": 4.2627281460134485,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003304490075302686,
      "loss": 4.6273,
      "mean_token_accuracy": 0.2563737347722054,
      "num_tokens": 101762955.0,
      "step": 44375
    },
    {
      "entropy": 5.438728094100952,
      "epoch": 4.263208453410183,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033041578070321494,
      "loss": 4.5937,
      "mean_token_accuracy": 0.26653056442737577,
      "num_tokens": 101773314.0,
      "step": 44380
    },
    {
      "entropy": 5.422705268859863,
      "epoch": 4.263688760806916,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003303825525895714,
      "loss": 4.6506,
      "mean_token_accuracy": 0.2521201312541962,
      "num_tokens": 101784006.0,
      "step": 44385
    },
    {
      "entropy": 5.496889305114746,
      "epoch": 4.264169068203651,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033034932319010946,
      "loss": 4.6362,
      "mean_token_accuracy": 0.26194821298122406,
      "num_tokens": 101795585.0,
      "step": 44390
    },
    {
      "entropy": 5.55492262840271,
      "epoch": 4.264649375600384,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033031609250560066,
      "loss": 4.7483,
      "mean_token_accuracy": 0.24895693063735963,
      "num_tokens": 101806672.0,
      "step": 44395
    },
    {
      "entropy": 5.577178573608398,
      "epoch": 4.265129682997118,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003302828605368164,
      "loss": 4.7689,
      "mean_token_accuracy": 0.24846653789281845,
      "num_tokens": 101819353.0,
      "step": 44400
    },
    {
      "entropy": 5.563149070739746,
      "epoch": 4.265609990393852,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033024962728452826,
      "loss": 4.7047,
      "mean_token_accuracy": 0.2552025094628334,
      "num_tokens": 101830739.0,
      "step": 44405
    },
    {
      "entropy": 5.454914093017578,
      "epoch": 4.266090297790586,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003302163927495078,
      "loss": 4.7252,
      "mean_token_accuracy": 0.2497829794883728,
      "num_tokens": 101842484.0,
      "step": 44410
    },
    {
      "entropy": 5.496386337280273,
      "epoch": 4.26657060518732,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003301831569325267,
      "loss": 4.7102,
      "mean_token_accuracy": 0.2503887593746185,
      "num_tokens": 101854393.0,
      "step": 44415
    },
    {
      "entropy": 5.487008762359619,
      "epoch": 4.267050912584054,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033014991983435654,
      "loss": 4.5763,
      "mean_token_accuracy": 0.26216208934783936,
      "num_tokens": 101866266.0,
      "step": 44420
    },
    {
      "entropy": 5.4905468940734865,
      "epoch": 4.267531219980787,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003301166814557691,
      "loss": 4.5953,
      "mean_token_accuracy": 0.26770241260528566,
      "num_tokens": 101876727.0,
      "step": 44425
    },
    {
      "entropy": 5.530528974533081,
      "epoch": 4.268011527377522,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033008344179753584,
      "loss": 4.7282,
      "mean_token_accuracy": 0.25292412340641024,
      "num_tokens": 101887996.0,
      "step": 44430
    },
    {
      "entropy": 5.505264711380005,
      "epoch": 4.268491834774256,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033005020086042874,
      "loss": 4.6887,
      "mean_token_accuracy": 0.2561803162097931,
      "num_tokens": 101899676.0,
      "step": 44435
    },
    {
      "entropy": 5.495248031616211,
      "epoch": 4.268972142170989,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033001695864521934,
      "loss": 4.6819,
      "mean_token_accuracy": 0.2574470922350883,
      "num_tokens": 101910709.0,
      "step": 44440
    },
    {
      "entropy": 5.516668510437012,
      "epoch": 4.269452449567724,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032998371515267954,
      "loss": 4.7117,
      "mean_token_accuracy": 0.25544205904006956,
      "num_tokens": 101921375.0,
      "step": 44445
    },
    {
      "entropy": 5.505091571807862,
      "epoch": 4.269932756964457,
      "grad_norm": 1.25,
      "learning_rate": 0.0003299504703835812,
      "loss": 4.6521,
      "mean_token_accuracy": 0.25517240166664124,
      "num_tokens": 101932899.0,
      "step": 44450
    },
    {
      "entropy": 5.550200891494751,
      "epoch": 4.2704130643611915,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003299172243386961,
      "loss": 4.7263,
      "mean_token_accuracy": 0.251295730471611,
      "num_tokens": 101943749.0,
      "step": 44455
    },
    {
      "entropy": 5.500307607650757,
      "epoch": 4.270893371757925,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003298839770187961,
      "loss": 4.6667,
      "mean_token_accuracy": 0.25709623396396636,
      "num_tokens": 101955166.0,
      "step": 44460
    },
    {
      "entropy": 5.516109895706177,
      "epoch": 4.271373679154659,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032985072842465314,
      "loss": 4.7044,
      "mean_token_accuracy": 0.25631502717733384,
      "num_tokens": 101966875.0,
      "step": 44465
    },
    {
      "entropy": 5.553261041641235,
      "epoch": 4.271853986551393,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032981747855703915,
      "loss": 4.733,
      "mean_token_accuracy": 0.25036687701940535,
      "num_tokens": 101978137.0,
      "step": 44470
    },
    {
      "entropy": 5.459897756576538,
      "epoch": 4.272334293948127,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032978422741672603,
      "loss": 4.6171,
      "mean_token_accuracy": 0.2580574586987495,
      "num_tokens": 101988953.0,
      "step": 44475
    },
    {
      "entropy": 5.567010021209716,
      "epoch": 4.2728146013448605,
      "grad_norm": 1.25,
      "learning_rate": 0.0003297509750044858,
      "loss": 4.7744,
      "mean_token_accuracy": 0.24903898388147355,
      "num_tokens": 102001298.0,
      "step": 44480
    },
    {
      "entropy": 5.521309852600098,
      "epoch": 4.273294908741595,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003297177213210907,
      "loss": 4.6395,
      "mean_token_accuracy": 0.2595987990498543,
      "num_tokens": 102012934.0,
      "step": 44485
    },
    {
      "entropy": 5.453932619094848,
      "epoch": 4.273775216138328,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003296844663673124,
      "loss": 4.6136,
      "mean_token_accuracy": 0.26631012856960296,
      "num_tokens": 102023863.0,
      "step": 44490
    },
    {
      "entropy": 5.436812019348144,
      "epoch": 4.2742555235350626,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003296512101439232,
      "loss": 4.557,
      "mean_token_accuracy": 0.26789123862981795,
      "num_tokens": 102034586.0,
      "step": 44495
    },
    {
      "entropy": 5.409916639328003,
      "epoch": 4.274735830931796,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032961795265169525,
      "loss": 4.6184,
      "mean_token_accuracy": 0.2638685256242752,
      "num_tokens": 102044933.0,
      "step": 44500
    },
    {
      "entropy": 5.5561073303222654,
      "epoch": 4.27521613832853,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003295846938914005,
      "loss": 4.7913,
      "mean_token_accuracy": 0.25508708953857423,
      "num_tokens": 102056370.0,
      "step": 44505
    },
    {
      "entropy": 5.502010917663574,
      "epoch": 4.275696445725265,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003295514338638114,
      "loss": 4.7105,
      "mean_token_accuracy": 0.2517759621143341,
      "num_tokens": 102067428.0,
      "step": 44510
    },
    {
      "entropy": 5.465698623657227,
      "epoch": 4.276176753121998,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003295181725696999,
      "loss": 4.5948,
      "mean_token_accuracy": 0.2696466654539108,
      "num_tokens": 102079262.0,
      "step": 44515
    },
    {
      "entropy": 5.464466285705567,
      "epoch": 4.276657060518732,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003294849100098383,
      "loss": 4.6491,
      "mean_token_accuracy": 0.2644658759236336,
      "num_tokens": 102088835.0,
      "step": 44520
    },
    {
      "entropy": 5.590981340408325,
      "epoch": 4.277137367915466,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003294516461849988,
      "loss": 4.7669,
      "mean_token_accuracy": 0.25061325877904894,
      "num_tokens": 102100119.0,
      "step": 44525
    },
    {
      "entropy": 5.568872308731079,
      "epoch": 4.2776176753122,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000329418381095954,
      "loss": 4.7436,
      "mean_token_accuracy": 0.2512673631310463,
      "num_tokens": 102111279.0,
      "step": 44530
    },
    {
      "entropy": 5.443983888626098,
      "epoch": 4.278097982708934,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003293851147434758,
      "loss": 4.6545,
      "mean_token_accuracy": 0.25950236469507215,
      "num_tokens": 102122977.0,
      "step": 44535
    },
    {
      "entropy": 5.535566091537476,
      "epoch": 4.278578290105668,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032935184712833674,
      "loss": 4.7678,
      "mean_token_accuracy": 0.256123423576355,
      "num_tokens": 102135343.0,
      "step": 44540
    },
    {
      "entropy": 5.532382392883301,
      "epoch": 4.279058597502401,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003293185782513093,
      "loss": 4.6756,
      "mean_token_accuracy": 0.2554986298084259,
      "num_tokens": 102148331.0,
      "step": 44545
    },
    {
      "entropy": 5.50025200843811,
      "epoch": 4.279538904899136,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032928530811316564,
      "loss": 4.6297,
      "mean_token_accuracy": 0.2646732643246651,
      "num_tokens": 102159720.0,
      "step": 44550
    },
    {
      "entropy": 5.546621513366699,
      "epoch": 4.280019212295869,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003292520367146784,
      "loss": 4.7567,
      "mean_token_accuracy": 0.24603029191493989,
      "num_tokens": 102172323.0,
      "step": 44555
    },
    {
      "entropy": 5.489952278137207,
      "epoch": 4.2804995196926034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003292187640566199,
      "loss": 4.6273,
      "mean_token_accuracy": 0.2593034774065018,
      "num_tokens": 102183998.0,
      "step": 44560
    },
    {
      "entropy": 5.511522388458252,
      "epoch": 4.280979827089337,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032918549013976267,
      "loss": 4.6436,
      "mean_token_accuracy": 0.2576747953891754,
      "num_tokens": 102196193.0,
      "step": 44565
    },
    {
      "entropy": 5.478483009338379,
      "epoch": 4.281460134486071,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003291522149648794,
      "loss": 4.6072,
      "mean_token_accuracy": 0.26265241503715514,
      "num_tokens": 102208042.0,
      "step": 44570
    },
    {
      "entropy": 5.521187162399292,
      "epoch": 4.281940441882805,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032911893853274234,
      "loss": 4.6951,
      "mean_token_accuracy": 0.25891953855752947,
      "num_tokens": 102219006.0,
      "step": 44575
    },
    {
      "entropy": 5.520261907577515,
      "epoch": 4.282420749279539,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003290856608441243,
      "loss": 4.6699,
      "mean_token_accuracy": 0.26305961459875105,
      "num_tokens": 102230733.0,
      "step": 44580
    },
    {
      "entropy": 5.516479206085205,
      "epoch": 4.282901056676272,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003290523818997978,
      "loss": 4.7101,
      "mean_token_accuracy": 0.256427638232708,
      "num_tokens": 102240871.0,
      "step": 44585
    },
    {
      "entropy": 5.457573318481446,
      "epoch": 4.283381364073007,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032901910170053553,
      "loss": 4.5896,
      "mean_token_accuracy": 0.262369441986084,
      "num_tokens": 102252427.0,
      "step": 44590
    },
    {
      "entropy": 5.533185720443726,
      "epoch": 4.28386167146974,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003289858202471101,
      "loss": 4.7461,
      "mean_token_accuracy": 0.25627746284008024,
      "num_tokens": 102263576.0,
      "step": 44595
    },
    {
      "entropy": 5.565053653717041,
      "epoch": 4.2843419788664745,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032895253754029415,
      "loss": 4.6496,
      "mean_token_accuracy": 0.2613336831331253,
      "num_tokens": 102275475.0,
      "step": 44600
    },
    {
      "entropy": 5.527697896957397,
      "epoch": 4.284822286263209,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032891925358086063,
      "loss": 4.6989,
      "mean_token_accuracy": 0.2500453144311905,
      "num_tokens": 102287757.0,
      "step": 44605
    },
    {
      "entropy": 5.507824420928955,
      "epoch": 4.285302593659942,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032888596836958206,
      "loss": 4.6228,
      "mean_token_accuracy": 0.25856528282165525,
      "num_tokens": 102298732.0,
      "step": 44610
    },
    {
      "entropy": 5.551400136947632,
      "epoch": 4.285782901056677,
      "grad_norm": 1.4375,
      "learning_rate": 0.00032885268190723134,
      "loss": 4.7164,
      "mean_token_accuracy": 0.25702660232782365,
      "num_tokens": 102310061.0,
      "step": 44615
    },
    {
      "entropy": 5.517699146270752,
      "epoch": 4.28626320845341,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003288193941945812,
      "loss": 4.7087,
      "mean_token_accuracy": 0.2517948761582375,
      "num_tokens": 102320937.0,
      "step": 44620
    },
    {
      "entropy": 5.559198188781738,
      "epoch": 4.286743515850144,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003287861052324047,
      "loss": 4.7963,
      "mean_token_accuracy": 0.24399843961000442,
      "num_tokens": 102331849.0,
      "step": 44625
    },
    {
      "entropy": 5.580558824539184,
      "epoch": 4.287223823246878,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003287528150214744,
      "loss": 4.716,
      "mean_token_accuracy": 0.2539939031004906,
      "num_tokens": 102343504.0,
      "step": 44630
    },
    {
      "entropy": 5.555150842666626,
      "epoch": 4.287704130643612,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032871952356256336,
      "loss": 4.7226,
      "mean_token_accuracy": 0.25152686685323716,
      "num_tokens": 102354476.0,
      "step": 44635
    },
    {
      "entropy": 5.451582527160644,
      "epoch": 4.2881844380403455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003286862308564445,
      "loss": 4.5892,
      "mean_token_accuracy": 0.2664786636829376,
      "num_tokens": 102365733.0,
      "step": 44640
    },
    {
      "entropy": 5.583823871612549,
      "epoch": 4.28866474543708,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032865293690389076,
      "loss": 4.7346,
      "mean_token_accuracy": 0.24812645763158797,
      "num_tokens": 102377954.0,
      "step": 44645
    },
    {
      "entropy": 5.61136908531189,
      "epoch": 4.289145052833813,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003286196417056752,
      "loss": 4.7344,
      "mean_token_accuracy": 0.25312345176935197,
      "num_tokens": 102388378.0,
      "step": 44650
    },
    {
      "entropy": 5.446613597869873,
      "epoch": 4.289625360230548,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032858634526257077,
      "loss": 4.667,
      "mean_token_accuracy": 0.25753533840179443,
      "num_tokens": 102399702.0,
      "step": 44655
    },
    {
      "entropy": 5.513202381134033,
      "epoch": 4.290105667627281,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003285530475753505,
      "loss": 4.6786,
      "mean_token_accuracy": 0.2552290290594101,
      "num_tokens": 102410393.0,
      "step": 44660
    },
    {
      "entropy": 5.509484434127808,
      "epoch": 4.290585975024015,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032851974864478743,
      "loss": 4.7485,
      "mean_token_accuracy": 0.2627446487545967,
      "num_tokens": 102423463.0,
      "step": 44665
    },
    {
      "entropy": 5.523440027236939,
      "epoch": 4.291066282420749,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032848644847165484,
      "loss": 4.6409,
      "mean_token_accuracy": 0.2638435661792755,
      "num_tokens": 102434822.0,
      "step": 44670
    },
    {
      "entropy": 5.485598230361939,
      "epoch": 4.291546589817483,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003284531470567256,
      "loss": 4.6307,
      "mean_token_accuracy": 0.25734863728284835,
      "num_tokens": 102446302.0,
      "step": 44675
    },
    {
      "entropy": 5.4787684917449955,
      "epoch": 4.2920268972142175,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003284198444007731,
      "loss": 4.6594,
      "mean_token_accuracy": 0.25548793077468873,
      "num_tokens": 102456866.0,
      "step": 44680
    },
    {
      "entropy": 5.506260204315185,
      "epoch": 4.292507204610951,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032838654050457047,
      "loss": 4.7581,
      "mean_token_accuracy": 0.25031192451715467,
      "num_tokens": 102468498.0,
      "step": 44685
    },
    {
      "entropy": 5.477948427200317,
      "epoch": 4.292987512007685,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003283532353688908,
      "loss": 4.6677,
      "mean_token_accuracy": 0.2577355563640594,
      "num_tokens": 102481723.0,
      "step": 44690
    },
    {
      "entropy": 5.537821960449219,
      "epoch": 4.293467819404419,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032831992899450745,
      "loss": 4.6734,
      "mean_token_accuracy": 0.2578752875328064,
      "num_tokens": 102492828.0,
      "step": 44695
    },
    {
      "entropy": 5.469836235046387,
      "epoch": 4.293948126801153,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032828662138219375,
      "loss": 4.6537,
      "mean_token_accuracy": 0.26160493195056916,
      "num_tokens": 102503261.0,
      "step": 44700
    },
    {
      "entropy": 5.420002412796021,
      "epoch": 4.2944284341978864,
      "grad_norm": 1.25,
      "learning_rate": 0.0003282533125327228,
      "loss": 4.5589,
      "mean_token_accuracy": 0.26131436675786973,
      "num_tokens": 102514383.0,
      "step": 44705
    },
    {
      "entropy": 5.449392509460449,
      "epoch": 4.294908741594621,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003282200024468682,
      "loss": 4.7047,
      "mean_token_accuracy": 0.2546020969748497,
      "num_tokens": 102524441.0,
      "step": 44710
    },
    {
      "entropy": 5.528486156463623,
      "epoch": 4.295389048991354,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003281866911254031,
      "loss": 4.6585,
      "mean_token_accuracy": 0.26178061962127686,
      "num_tokens": 102534785.0,
      "step": 44715
    },
    {
      "entropy": 5.475444316864014,
      "epoch": 4.2958693563880885,
      "grad_norm": 1.3125,
      "learning_rate": 0.000328153378569101,
      "loss": 4.6092,
      "mean_token_accuracy": 0.26343763768672945,
      "num_tokens": 102546596.0,
      "step": 44720
    },
    {
      "entropy": 5.455709981918335,
      "epoch": 4.296349663784822,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003281200647787352,
      "loss": 4.6392,
      "mean_token_accuracy": 0.2612660452723503,
      "num_tokens": 102557035.0,
      "step": 44725
    },
    {
      "entropy": 5.518209218978882,
      "epoch": 4.296829971181556,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032808674975507937,
      "loss": 4.7194,
      "mean_token_accuracy": 0.2551042765378952,
      "num_tokens": 102568580.0,
      "step": 44730
    },
    {
      "entropy": 5.647634124755859,
      "epoch": 4.29731027857829,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003280534334989067,
      "loss": 4.756,
      "mean_token_accuracy": 0.25441714078187944,
      "num_tokens": 102578858.0,
      "step": 44735
    },
    {
      "entropy": 5.519574308395386,
      "epoch": 4.297790585975024,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032802011601099095,
      "loss": 4.6648,
      "mean_token_accuracy": 0.25683062374591825,
      "num_tokens": 102590855.0,
      "step": 44740
    },
    {
      "entropy": 5.459544849395752,
      "epoch": 4.2982708933717575,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032798679729210555,
      "loss": 4.6496,
      "mean_token_accuracy": 0.2601732164621353,
      "num_tokens": 102602548.0,
      "step": 44745
    },
    {
      "entropy": 5.503080940246582,
      "epoch": 4.298751200768492,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000327953477343024,
      "loss": 4.7384,
      "mean_token_accuracy": 0.24911360144615174,
      "num_tokens": 102613635.0,
      "step": 44750
    },
    {
      "entropy": 5.542904806137085,
      "epoch": 4.299231508165226,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032792015616452,
      "loss": 4.7088,
      "mean_token_accuracy": 0.24628829061985016,
      "num_tokens": 102624958.0,
      "step": 44755
    },
    {
      "entropy": 5.529013395309448,
      "epoch": 4.29971181556196,
      "grad_norm": 1.15625,
      "learning_rate": 0.000327886833757367,
      "loss": 4.6282,
      "mean_token_accuracy": 0.25579986721277237,
      "num_tokens": 102635616.0,
      "step": 44760
    },
    {
      "entropy": 5.544986867904663,
      "epoch": 4.300192122958694,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003278535101223388,
      "loss": 4.6882,
      "mean_token_accuracy": 0.2558743417263031,
      "num_tokens": 102646654.0,
      "step": 44765
    },
    {
      "entropy": 5.553482580184936,
      "epoch": 4.300672430355427,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003278201852602091,
      "loss": 4.6556,
      "mean_token_accuracy": 0.2665554046630859,
      "num_tokens": 102657624.0,
      "step": 44770
    },
    {
      "entropy": 5.478507471084595,
      "epoch": 4.301152737752162,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032778685917175153,
      "loss": 4.645,
      "mean_token_accuracy": 0.2605857715010643,
      "num_tokens": 102668931.0,
      "step": 44775
    },
    {
      "entropy": 5.517552661895752,
      "epoch": 4.301633045148895,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032775353185773983,
      "loss": 4.6501,
      "mean_token_accuracy": 0.255835685133934,
      "num_tokens": 102680680.0,
      "step": 44780
    },
    {
      "entropy": 5.500795459747314,
      "epoch": 4.302113352545629,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032772020331894776,
      "loss": 4.696,
      "mean_token_accuracy": 0.2557999029755592,
      "num_tokens": 102692972.0,
      "step": 44785
    },
    {
      "entropy": 5.535572814941406,
      "epoch": 4.302593659942363,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003276868735561491,
      "loss": 4.6562,
      "mean_token_accuracy": 0.2514923378825188,
      "num_tokens": 102705739.0,
      "step": 44790
    },
    {
      "entropy": 5.516630697250366,
      "epoch": 4.303073967339097,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00032765354257011765,
      "loss": 4.6439,
      "mean_token_accuracy": 0.2620949521660805,
      "num_tokens": 102716368.0,
      "step": 44795
    },
    {
      "entropy": 5.5031599521636965,
      "epoch": 4.303554274735831,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032762021036162735,
      "loss": 4.7336,
      "mean_token_accuracy": 0.2493380308151245,
      "num_tokens": 102729920.0,
      "step": 44800
    },
    {
      "entropy": 5.5421788692474365,
      "epoch": 4.304034582132565,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003275868769314521,
      "loss": 4.7203,
      "mean_token_accuracy": 0.251312281191349,
      "num_tokens": 102740374.0,
      "step": 44805
    },
    {
      "entropy": 5.5508264064788815,
      "epoch": 4.304514889529298,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032755354228036554,
      "loss": 4.7395,
      "mean_token_accuracy": 0.24523482471704483,
      "num_tokens": 102750820.0,
      "step": 44810
    },
    {
      "entropy": 5.547393131256103,
      "epoch": 4.304995196926033,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003275202064091419,
      "loss": 4.7026,
      "mean_token_accuracy": 0.25574740320444106,
      "num_tokens": 102761312.0,
      "step": 44815
    },
    {
      "entropy": 5.521642780303955,
      "epoch": 4.305475504322766,
      "grad_norm": 1.375,
      "learning_rate": 0.00032748686931855495,
      "loss": 4.6938,
      "mean_token_accuracy": 0.2518611267209053,
      "num_tokens": 102772288.0,
      "step": 44820
    },
    {
      "entropy": 5.51058554649353,
      "epoch": 4.3059558117195005,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003274535310093788,
      "loss": 4.6139,
      "mean_token_accuracy": 0.2614827141165733,
      "num_tokens": 102783533.0,
      "step": 44825
    },
    {
      "entropy": 5.511208534240723,
      "epoch": 4.306436119116235,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003274201914823874,
      "loss": 4.6868,
      "mean_token_accuracy": 0.25917146652936934,
      "num_tokens": 102795940.0,
      "step": 44830
    },
    {
      "entropy": 5.470823907852173,
      "epoch": 4.306916426512968,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032738685073835475,
      "loss": 4.684,
      "mean_token_accuracy": 0.2630195617675781,
      "num_tokens": 102807507.0,
      "step": 44835
    },
    {
      "entropy": 5.487682294845581,
      "epoch": 4.307396733909703,
      "grad_norm": 1.21875,
      "learning_rate": 0.000327353508778055,
      "loss": 4.6035,
      "mean_token_accuracy": 0.2614815950393677,
      "num_tokens": 102817895.0,
      "step": 44840
    },
    {
      "entropy": 5.533781576156616,
      "epoch": 4.307877041306436,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032732016560226235,
      "loss": 4.6778,
      "mean_token_accuracy": 0.2540199518203735,
      "num_tokens": 102829141.0,
      "step": 44845
    },
    {
      "entropy": 5.4988316059112545,
      "epoch": 4.30835734870317,
      "grad_norm": 1.25,
      "learning_rate": 0.00032728682121175077,
      "loss": 4.6784,
      "mean_token_accuracy": 0.26064886152744293,
      "num_tokens": 102841488.0,
      "step": 44850
    },
    {
      "entropy": 5.487931776046753,
      "epoch": 4.308837656099904,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032725347560729433,
      "loss": 4.7094,
      "mean_token_accuracy": 0.2499549850821495,
      "num_tokens": 102853276.0,
      "step": 44855
    },
    {
      "entropy": 5.505508041381836,
      "epoch": 4.309317963496638,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032722012878966744,
      "loss": 4.6779,
      "mean_token_accuracy": 0.2549630805850029,
      "num_tokens": 102864850.0,
      "step": 44860
    },
    {
      "entropy": 5.530338430404663,
      "epoch": 4.3097982708933715,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032718678075964425,
      "loss": 4.7006,
      "mean_token_accuracy": 0.2497489035129547,
      "num_tokens": 102876692.0,
      "step": 44865
    },
    {
      "entropy": 5.484028053283692,
      "epoch": 4.310278578290106,
      "grad_norm": 1.125,
      "learning_rate": 0.0003271534315179989,
      "loss": 4.6649,
      "mean_token_accuracy": 0.25464330762624743,
      "num_tokens": 102887693.0,
      "step": 44870
    },
    {
      "entropy": 5.491369390487671,
      "epoch": 4.310758885686839,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032712008106550584,
      "loss": 4.6957,
      "mean_token_accuracy": 0.2559821605682373,
      "num_tokens": 102900493.0,
      "step": 44875
    },
    {
      "entropy": 5.554059743881226,
      "epoch": 4.311239193083574,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032708672940293923,
      "loss": 4.6969,
      "mean_token_accuracy": 0.25011534094810484,
      "num_tokens": 102911983.0,
      "step": 44880
    },
    {
      "entropy": 5.500639343261719,
      "epoch": 4.311719500480307,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032705337653107336,
      "loss": 4.6282,
      "mean_token_accuracy": 0.265678970515728,
      "num_tokens": 102923204.0,
      "step": 44885
    },
    {
      "entropy": 5.5340780258178714,
      "epoch": 4.312199807877041,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032702002245068267,
      "loss": 4.7304,
      "mean_token_accuracy": 0.24877720177173615,
      "num_tokens": 102934091.0,
      "step": 44890
    },
    {
      "entropy": 5.469866514205933,
      "epoch": 4.312680115273775,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032698666716254155,
      "loss": 4.5995,
      "mean_token_accuracy": 0.26554068326950075,
      "num_tokens": 102946044.0,
      "step": 44895
    },
    {
      "entropy": 5.6002685546875,
      "epoch": 4.313160422670509,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003269533106674244,
      "loss": 4.8208,
      "mean_token_accuracy": 0.24726193249225617,
      "num_tokens": 102957399.0,
      "step": 44900
    },
    {
      "entropy": 5.5103844165802,
      "epoch": 4.3136407300672435,
      "grad_norm": 1.25,
      "learning_rate": 0.00032691995296610566,
      "loss": 4.6467,
      "mean_token_accuracy": 0.2623570397496223,
      "num_tokens": 102968757.0,
      "step": 44905
    },
    {
      "entropy": 5.467669868469239,
      "epoch": 4.314121037463977,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032688659405935973,
      "loss": 4.6185,
      "mean_token_accuracy": 0.2624053254723549,
      "num_tokens": 102980128.0,
      "step": 44910
    },
    {
      "entropy": 5.525893974304199,
      "epoch": 4.314601344860711,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003268532339479612,
      "loss": 4.7303,
      "mean_token_accuracy": 0.25026995986700057,
      "num_tokens": 102990951.0,
      "step": 44915
    },
    {
      "entropy": 5.462504196166992,
      "epoch": 4.315081652257445,
      "grad_norm": 1.25,
      "learning_rate": 0.0003268198726326846,
      "loss": 4.6689,
      "mean_token_accuracy": 0.25799338668584826,
      "num_tokens": 103003911.0,
      "step": 44920
    },
    {
      "entropy": 5.539969635009766,
      "epoch": 4.315561959654179,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003267865101143043,
      "loss": 4.6842,
      "mean_token_accuracy": 0.25835048109292985,
      "num_tokens": 103015584.0,
      "step": 44925
    },
    {
      "entropy": 5.541767978668213,
      "epoch": 4.316042267050912,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032675314639359505,
      "loss": 4.7021,
      "mean_token_accuracy": 0.2649006187915802,
      "num_tokens": 103027632.0,
      "step": 44930
    },
    {
      "entropy": 5.5256187438964846,
      "epoch": 4.316522574447647,
      "grad_norm": 1.375,
      "learning_rate": 0.0003267197814713315,
      "loss": 4.6629,
      "mean_token_accuracy": 0.25452738255262375,
      "num_tokens": 103037784.0,
      "step": 44935
    },
    {
      "entropy": 5.4722565650939945,
      "epoch": 4.31700288184438,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032668641534828807,
      "loss": 4.6377,
      "mean_token_accuracy": 0.25448388457298277,
      "num_tokens": 103048753.0,
      "step": 44940
    },
    {
      "entropy": 5.503896808624267,
      "epoch": 4.3174831892411145,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032665304802523965,
      "loss": 4.714,
      "mean_token_accuracy": 0.24981288164854049,
      "num_tokens": 103061372.0,
      "step": 44945
    },
    {
      "entropy": 5.527155971527099,
      "epoch": 4.317963496637848,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032661967950296084,
      "loss": 4.6799,
      "mean_token_accuracy": 0.2609217017889023,
      "num_tokens": 103074791.0,
      "step": 44950
    },
    {
      "entropy": 5.531623601913452,
      "epoch": 4.318443804034582,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003265863097822263,
      "loss": 4.692,
      "mean_token_accuracy": 0.25946722477674483,
      "num_tokens": 103086790.0,
      "step": 44955
    },
    {
      "entropy": 5.540306568145752,
      "epoch": 4.318924111431316,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032655293886381083,
      "loss": 4.6727,
      "mean_token_accuracy": 0.2550214886665344,
      "num_tokens": 103097819.0,
      "step": 44960
    },
    {
      "entropy": 5.534500694274902,
      "epoch": 4.31940441882805,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003265195667484892,
      "loss": 4.7308,
      "mean_token_accuracy": 0.2497068703174591,
      "num_tokens": 103110071.0,
      "step": 44965
    },
    {
      "entropy": 5.4823317527771,
      "epoch": 4.3198847262247835,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032648619343703626,
      "loss": 4.6744,
      "mean_token_accuracy": 0.2605892315506935,
      "num_tokens": 103122586.0,
      "step": 44970
    },
    {
      "entropy": 5.529036569595337,
      "epoch": 4.320365033621518,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003264528189302267,
      "loss": 4.6555,
      "mean_token_accuracy": 0.2576582029461861,
      "num_tokens": 103133531.0,
      "step": 44975
    },
    {
      "entropy": 5.464519834518432,
      "epoch": 4.320845341018252,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032641944322883557,
      "loss": 4.5974,
      "mean_token_accuracy": 0.2673592746257782,
      "num_tokens": 103144099.0,
      "step": 44980
    },
    {
      "entropy": 5.442235326766967,
      "epoch": 4.321325648414986,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003263860663336376,
      "loss": 4.6217,
      "mean_token_accuracy": 0.25632911920547485,
      "num_tokens": 103156221.0,
      "step": 44985
    },
    {
      "entropy": 5.495186281204224,
      "epoch": 4.32180595581172,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032635268824540774,
      "loss": 4.6077,
      "mean_token_accuracy": 0.2685782790184021,
      "num_tokens": 103167121.0,
      "step": 44990
    },
    {
      "entropy": 5.65065746307373,
      "epoch": 4.322286263208453,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032631930896492097,
      "loss": 4.7836,
      "mean_token_accuracy": 0.2468547433614731,
      "num_tokens": 103178169.0,
      "step": 44995
    },
    {
      "entropy": 5.562743902206421,
      "epoch": 4.322766570605188,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003262859284929522,
      "loss": 4.7268,
      "mean_token_accuracy": 0.25498434603214265,
      "num_tokens": 103190790.0,
      "step": 45000
    },
    {
      "epoch": 4.322766570605188,
      "eval_entropy": 5.247945466197334,
      "eval_loss": 4.787253379821777,
      "eval_mean_token_accuracy": 0.2588939967895356,
      "eval_num_tokens": 103190790.0,
      "eval_runtime": 20.7092,
      "eval_samples_per_second": 1584.56,
      "eval_steps_per_second": 198.076,
      "step": 45000
    },
    {
      "entropy": 5.497464179992676,
      "epoch": 4.323246878001921,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003262525468302765,
      "loss": 4.6486,
      "mean_token_accuracy": 0.25804266333580017,
      "num_tokens": 103201993.0,
      "step": 45005
    },
    {
      "entropy": 5.511553812026977,
      "epoch": 4.323727185398655,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003262191639776689,
      "loss": 4.6698,
      "mean_token_accuracy": 0.2585488423705101,
      "num_tokens": 103213938.0,
      "step": 45010
    },
    {
      "entropy": 5.48817629814148,
      "epoch": 4.324207492795389,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003261857799359043,
      "loss": 4.6548,
      "mean_token_accuracy": 0.25204968452453613,
      "num_tokens": 103227098.0,
      "step": 45015
    },
    {
      "entropy": 5.518828582763672,
      "epoch": 4.324687800192123,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003261523947057579,
      "loss": 4.6784,
      "mean_token_accuracy": 0.25786491483449936,
      "num_tokens": 103239811.0,
      "step": 45020
    },
    {
      "entropy": 5.619755125045776,
      "epoch": 4.325168107588857,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003261190082880048,
      "loss": 4.8149,
      "mean_token_accuracy": 0.2482788547873497,
      "num_tokens": 103252160.0,
      "step": 45025
    },
    {
      "entropy": 5.598443698883057,
      "epoch": 4.325648414985591,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003260856206834201,
      "loss": 4.7958,
      "mean_token_accuracy": 0.24559347629547118,
      "num_tokens": 103262752.0,
      "step": 45030
    },
    {
      "entropy": 5.525638914108276,
      "epoch": 4.326128722382324,
      "grad_norm": 1.25,
      "learning_rate": 0.00032605223189277903,
      "loss": 4.7256,
      "mean_token_accuracy": 0.2532047048211098,
      "num_tokens": 103274539.0,
      "step": 45035
    },
    {
      "entropy": 5.628654289245605,
      "epoch": 4.326609029779059,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003260188419168568,
      "loss": 4.7614,
      "mean_token_accuracy": 0.2488217145204544,
      "num_tokens": 103285929.0,
      "step": 45040
    },
    {
      "entropy": 5.560004234313965,
      "epoch": 4.327089337175792,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032598545075642843,
      "loss": 4.7057,
      "mean_token_accuracy": 0.25087147802114484,
      "num_tokens": 103297710.0,
      "step": 45045
    },
    {
      "entropy": 5.521257114410401,
      "epoch": 4.3275696445725265,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003259520584122693,
      "loss": 4.725,
      "mean_token_accuracy": 0.2555710792541504,
      "num_tokens": 103310080.0,
      "step": 45050
    },
    {
      "entropy": 5.511566066741944,
      "epoch": 4.328049951969261,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003259186648851547,
      "loss": 4.6946,
      "mean_token_accuracy": 0.25995665937662127,
      "num_tokens": 103322660.0,
      "step": 45055
    },
    {
      "entropy": 5.516691398620606,
      "epoch": 4.328530259365994,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003258852701758599,
      "loss": 4.6755,
      "mean_token_accuracy": 0.25605475157499313,
      "num_tokens": 103335302.0,
      "step": 45060
    },
    {
      "entropy": 5.5091523170471195,
      "epoch": 4.3290105667627286,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003258518742851602,
      "loss": 4.662,
      "mean_token_accuracy": 0.26123411357402804,
      "num_tokens": 103346957.0,
      "step": 45065
    },
    {
      "entropy": 5.537324857711792,
      "epoch": 4.329490874159462,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000325818477213831,
      "loss": 4.6813,
      "mean_token_accuracy": 0.252600160241127,
      "num_tokens": 103359003.0,
      "step": 45070
    },
    {
      "entropy": 5.531919765472412,
      "epoch": 4.329971181556196,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003257850789626477,
      "loss": 4.7171,
      "mean_token_accuracy": 0.2561795100569725,
      "num_tokens": 103370407.0,
      "step": 45075
    },
    {
      "entropy": 5.606730031967163,
      "epoch": 4.33045148895293,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003257516795323856,
      "loss": 4.8314,
      "mean_token_accuracy": 0.24294845759868622,
      "num_tokens": 103382524.0,
      "step": 45080
    },
    {
      "entropy": 5.630810213088989,
      "epoch": 4.330931796349664,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032571827892382014,
      "loss": 4.8021,
      "mean_token_accuracy": 0.24824876785278321,
      "num_tokens": 103393685.0,
      "step": 45085
    },
    {
      "entropy": 5.481455230712891,
      "epoch": 4.3314121037463975,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000325684877137727,
      "loss": 4.5776,
      "mean_token_accuracy": 0.25790923088788986,
      "num_tokens": 103404105.0,
      "step": 45090
    },
    {
      "entropy": 5.583245038986206,
      "epoch": 4.331892411143132,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003256514741748814,
      "loss": 4.791,
      "mean_token_accuracy": 0.24718202352523805,
      "num_tokens": 103415765.0,
      "step": 45095
    },
    {
      "entropy": 5.472154378890991,
      "epoch": 4.332372718539865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003256180700360591,
      "loss": 4.682,
      "mean_token_accuracy": 0.2537522718310356,
      "num_tokens": 103426490.0,
      "step": 45100
    },
    {
      "entropy": 5.502405166625977,
      "epoch": 4.3328530259366,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003255846647220354,
      "loss": 4.6257,
      "mean_token_accuracy": 0.2602687567472458,
      "num_tokens": 103437864.0,
      "step": 45105
    },
    {
      "entropy": 5.502850818634033,
      "epoch": 4.333333333333333,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032555125823358603,
      "loss": 4.6661,
      "mean_token_accuracy": 0.2578860655426979,
      "num_tokens": 103450119.0,
      "step": 45110
    },
    {
      "entropy": 5.441209316253662,
      "epoch": 4.333813640730067,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003255178505714866,
      "loss": 4.676,
      "mean_token_accuracy": 0.25870270282030106,
      "num_tokens": 103461220.0,
      "step": 45115
    },
    {
      "entropy": 5.580495929718017,
      "epoch": 4.334293948126801,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003254844417365126,
      "loss": 4.7267,
      "mean_token_accuracy": 0.2586792320013046,
      "num_tokens": 103473375.0,
      "step": 45120
    },
    {
      "entropy": 5.511885690689087,
      "epoch": 4.334774255523535,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003254510317294398,
      "loss": 4.6862,
      "mean_token_accuracy": 0.25522768348455427,
      "num_tokens": 103484575.0,
      "step": 45125
    },
    {
      "entropy": 5.522336196899414,
      "epoch": 4.335254562920269,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032541762055104386,
      "loss": 4.674,
      "mean_token_accuracy": 0.25866190493106844,
      "num_tokens": 103496005.0,
      "step": 45130
    },
    {
      "entropy": 5.486805057525634,
      "epoch": 4.335734870317003,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032538420820210054,
      "loss": 4.6458,
      "mean_token_accuracy": 0.26664622873067856,
      "num_tokens": 103506795.0,
      "step": 45135
    },
    {
      "entropy": 5.503609323501587,
      "epoch": 4.336215177713736,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003253507946833854,
      "loss": 4.6512,
      "mean_token_accuracy": 0.25571922659873964,
      "num_tokens": 103518698.0,
      "step": 45140
    },
    {
      "entropy": 5.5577620506286625,
      "epoch": 4.336695485110471,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003253173799956744,
      "loss": 4.6667,
      "mean_token_accuracy": 0.2588439881801605,
      "num_tokens": 103529261.0,
      "step": 45145
    },
    {
      "entropy": 5.516618728637695,
      "epoch": 4.337175792507205,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003252839641397432,
      "loss": 4.6706,
      "mean_token_accuracy": 0.26147380769252776,
      "num_tokens": 103540823.0,
      "step": 45150
    },
    {
      "entropy": 5.466664552688599,
      "epoch": 4.337656099903938,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003252505471163677,
      "loss": 4.65,
      "mean_token_accuracy": 0.26092431545257566,
      "num_tokens": 103552097.0,
      "step": 45155
    },
    {
      "entropy": 5.545445680618286,
      "epoch": 4.338136407300673,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032521712892632365,
      "loss": 4.6908,
      "mean_token_accuracy": 0.25138054937124255,
      "num_tokens": 103563836.0,
      "step": 45160
    },
    {
      "entropy": 5.452725267410278,
      "epoch": 4.338616714697406,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003251837095703871,
      "loss": 4.5702,
      "mean_token_accuracy": 0.2672850385308266,
      "num_tokens": 103574925.0,
      "step": 45165
    },
    {
      "entropy": 5.638764572143555,
      "epoch": 4.3390970220941405,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032515028904933375,
      "loss": 4.7889,
      "mean_token_accuracy": 0.24279546439647676,
      "num_tokens": 103586304.0,
      "step": 45170
    },
    {
      "entropy": 5.563543272018433,
      "epoch": 4.339577329490874,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003251168673639396,
      "loss": 4.6638,
      "mean_token_accuracy": 0.2553952232003212,
      "num_tokens": 103597326.0,
      "step": 45175
    },
    {
      "entropy": 5.500528764724732,
      "epoch": 4.340057636887608,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032508344451498064,
      "loss": 4.743,
      "mean_token_accuracy": 0.25295754075050353,
      "num_tokens": 103608688.0,
      "step": 45180
    },
    {
      "entropy": 5.491665029525757,
      "epoch": 4.340537944284342,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003250500205032328,
      "loss": 4.6591,
      "mean_token_accuracy": 0.25825362652540207,
      "num_tokens": 103619521.0,
      "step": 45185
    },
    {
      "entropy": 5.533912229537964,
      "epoch": 4.341018251681076,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032501659532947205,
      "loss": 4.6612,
      "mean_token_accuracy": 0.25558904409408567,
      "num_tokens": 103631829.0,
      "step": 45190
    },
    {
      "entropy": 5.547182607650757,
      "epoch": 4.3414985590778095,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032498316899447454,
      "loss": 4.772,
      "mean_token_accuracy": 0.24825990945100784,
      "num_tokens": 103642417.0,
      "step": 45195
    },
    {
      "entropy": 5.459686946868897,
      "epoch": 4.341978866474544,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032494974149901626,
      "loss": 4.6055,
      "mean_token_accuracy": 0.26289051920175555,
      "num_tokens": 103653921.0,
      "step": 45200
    },
    {
      "entropy": 5.415164470672607,
      "epoch": 4.342459173871277,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003249163128438732,
      "loss": 4.6201,
      "mean_token_accuracy": 0.26166763752698896,
      "num_tokens": 103665917.0,
      "step": 45205
    },
    {
      "entropy": 5.497642898559571,
      "epoch": 4.3429394812680115,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032488288302982176,
      "loss": 4.7043,
      "mean_token_accuracy": 0.25741944164037706,
      "num_tokens": 103676353.0,
      "step": 45210
    },
    {
      "entropy": 5.601546382904052,
      "epoch": 4.343419788664745,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003248494520576378,
      "loss": 4.7476,
      "mean_token_accuracy": 0.25151998549699783,
      "num_tokens": 103688334.0,
      "step": 45215
    },
    {
      "entropy": 5.516785955429077,
      "epoch": 4.343900096061479,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032481601992809754,
      "loss": 4.6433,
      "mean_token_accuracy": 0.2564839914441109,
      "num_tokens": 103699853.0,
      "step": 45220
    },
    {
      "entropy": 5.488981437683106,
      "epoch": 4.344380403458214,
      "grad_norm": 1.25,
      "learning_rate": 0.0003247825866419773,
      "loss": 4.6289,
      "mean_token_accuracy": 0.26027390658855437,
      "num_tokens": 103709891.0,
      "step": 45225
    },
    {
      "entropy": 5.4333240509033205,
      "epoch": 4.344860710854947,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032474915220005317,
      "loss": 4.6401,
      "mean_token_accuracy": 0.26104417592287066,
      "num_tokens": 103721154.0,
      "step": 45230
    },
    {
      "entropy": 5.565346813201904,
      "epoch": 4.345341018251681,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032471571660310145,
      "loss": 4.741,
      "mean_token_accuracy": 0.25310502052307127,
      "num_tokens": 103732831.0,
      "step": 45235
    },
    {
      "entropy": 5.520807266235352,
      "epoch": 4.345821325648415,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003246822798518985,
      "loss": 4.6223,
      "mean_token_accuracy": 0.26117744892835615,
      "num_tokens": 103745392.0,
      "step": 45240
    },
    {
      "entropy": 5.537290811538696,
      "epoch": 4.346301633045149,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032464884194722046,
      "loss": 4.6561,
      "mean_token_accuracy": 0.2552200898528099,
      "num_tokens": 103757191.0,
      "step": 45245
    },
    {
      "entropy": 5.51285719871521,
      "epoch": 4.346781940441883,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032461540288984377,
      "loss": 4.7053,
      "mean_token_accuracy": 0.25106072425842285,
      "num_tokens": 103769434.0,
      "step": 45250
    },
    {
      "entropy": 5.611399078369141,
      "epoch": 4.347262247838617,
      "grad_norm": 1.25,
      "learning_rate": 0.0003245819626805447,
      "loss": 4.7954,
      "mean_token_accuracy": 0.24846526235342026,
      "num_tokens": 103780906.0,
      "step": 45255
    },
    {
      "entropy": 5.545927286148071,
      "epoch": 4.34774255523535,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032454852132009974,
      "loss": 4.6904,
      "mean_token_accuracy": 0.2547703891992569,
      "num_tokens": 103791998.0,
      "step": 45260
    },
    {
      "entropy": 5.490305614471436,
      "epoch": 4.348222862632085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003245150788092852,
      "loss": 4.6519,
      "mean_token_accuracy": 0.26250392496585845,
      "num_tokens": 103804401.0,
      "step": 45265
    },
    {
      "entropy": 5.48433141708374,
      "epoch": 4.348703170028818,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032448163514887756,
      "loss": 4.6615,
      "mean_token_accuracy": 0.25520412176847457,
      "num_tokens": 103815774.0,
      "step": 45270
    },
    {
      "entropy": 5.550991153717041,
      "epoch": 4.3491834774255524,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003244481903396533,
      "loss": 4.6763,
      "mean_token_accuracy": 0.2545759126543999,
      "num_tokens": 103827370.0,
      "step": 45275
    },
    {
      "entropy": 5.537365674972534,
      "epoch": 4.349663784822286,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032441474438238893,
      "loss": 4.65,
      "mean_token_accuracy": 0.25918678790330885,
      "num_tokens": 103839102.0,
      "step": 45280
    },
    {
      "entropy": 5.518983602523804,
      "epoch": 4.35014409221902,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003243812972778609,
      "loss": 4.7065,
      "mean_token_accuracy": 0.25214729756116866,
      "num_tokens": 103852014.0,
      "step": 45285
    },
    {
      "entropy": 5.488629961013794,
      "epoch": 4.350624399615754,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032434784902684574,
      "loss": 4.6438,
      "mean_token_accuracy": 0.26328829526901243,
      "num_tokens": 103865308.0,
      "step": 45290
    },
    {
      "entropy": 5.544042539596558,
      "epoch": 4.351104707012488,
      "grad_norm": 1.25,
      "learning_rate": 0.00032431439963012004,
      "loss": 4.6846,
      "mean_token_accuracy": 0.25607795268297195,
      "num_tokens": 103877961.0,
      "step": 45295
    },
    {
      "entropy": 5.5107337474823,
      "epoch": 4.351585014409222,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003242809490884605,
      "loss": 4.6448,
      "mean_token_accuracy": 0.257288458943367,
      "num_tokens": 103888625.0,
      "step": 45300
    },
    {
      "entropy": 5.417885398864746,
      "epoch": 4.352065321805956,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003242474974026436,
      "loss": 4.589,
      "mean_token_accuracy": 0.26810288727283477,
      "num_tokens": 103899177.0,
      "step": 45305
    },
    {
      "entropy": 5.491866874694824,
      "epoch": 4.35254562920269,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003242140445734461,
      "loss": 4.6588,
      "mean_token_accuracy": 0.2608944490551949,
      "num_tokens": 103909687.0,
      "step": 45310
    },
    {
      "entropy": 5.468952751159668,
      "epoch": 4.3530259365994235,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032418059060164455,
      "loss": 4.6917,
      "mean_token_accuracy": 0.2584999963641167,
      "num_tokens": 103921137.0,
      "step": 45315
    },
    {
      "entropy": 5.537650156021118,
      "epoch": 4.353506243996158,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032414713548801577,
      "loss": 4.7593,
      "mean_token_accuracy": 0.2485166847705841,
      "num_tokens": 103934014.0,
      "step": 45320
    },
    {
      "entropy": 5.577564859390259,
      "epoch": 4.353986551392891,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032411367923333646,
      "loss": 4.6526,
      "mean_token_accuracy": 0.2644514039158821,
      "num_tokens": 103945824.0,
      "step": 45325
    },
    {
      "entropy": 5.563113164901734,
      "epoch": 4.354466858789626,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032408022183838336,
      "loss": 4.5737,
      "mean_token_accuracy": 0.2656689822673798,
      "num_tokens": 103956880.0,
      "step": 45330
    },
    {
      "entropy": 5.476344156265259,
      "epoch": 4.354947166186359,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032404676330393317,
      "loss": 4.6544,
      "mean_token_accuracy": 0.2617462635040283,
      "num_tokens": 103967663.0,
      "step": 45335
    },
    {
      "entropy": 5.532754278182983,
      "epoch": 4.355427473583093,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003240133036307628,
      "loss": 4.78,
      "mean_token_accuracy": 0.2426105961203575,
      "num_tokens": 103978736.0,
      "step": 45340
    },
    {
      "entropy": 5.459945201873779,
      "epoch": 4.355907780979827,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003239798428196491,
      "loss": 4.6285,
      "mean_token_accuracy": 0.26190603226423265,
      "num_tokens": 103990777.0,
      "step": 45345
    },
    {
      "entropy": 5.491395950317383,
      "epoch": 4.356388088376561,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032394638087136887,
      "loss": 4.6769,
      "mean_token_accuracy": 0.2572349101305008,
      "num_tokens": 104002563.0,
      "step": 45350
    },
    {
      "entropy": 5.504003858566284,
      "epoch": 4.3568683957732945,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000323912917786699,
      "loss": 4.6656,
      "mean_token_accuracy": 0.2531269058585167,
      "num_tokens": 104013955.0,
      "step": 45355
    },
    {
      "entropy": 5.563384342193603,
      "epoch": 4.357348703170029,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032387945356641653,
      "loss": 4.704,
      "mean_token_accuracy": 0.2502195328474045,
      "num_tokens": 104026086.0,
      "step": 45360
    },
    {
      "entropy": 5.59903130531311,
      "epoch": 4.357829010566762,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003238459882112981,
      "loss": 4.7885,
      "mean_token_accuracy": 0.25163508355617525,
      "num_tokens": 104038235.0,
      "step": 45365
    },
    {
      "entropy": 5.420061922073364,
      "epoch": 4.358309317963497,
      "grad_norm": 1.171875,
      "learning_rate": 0.000323812521722121,
      "loss": 4.5934,
      "mean_token_accuracy": 0.26166157275438306,
      "num_tokens": 104049574.0,
      "step": 45370
    },
    {
      "entropy": 5.440373516082763,
      "epoch": 4.358789625360231,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003237790540996621,
      "loss": 4.6555,
      "mean_token_accuracy": 0.2580373391509056,
      "num_tokens": 104060715.0,
      "step": 45375
    },
    {
      "entropy": 5.492627239227295,
      "epoch": 4.359269932756964,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032374558534469834,
      "loss": 4.6541,
      "mean_token_accuracy": 0.25810271948575975,
      "num_tokens": 104074234.0,
      "step": 45380
    },
    {
      "entropy": 5.548341894149781,
      "epoch": 4.359750240153699,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003237121154580068,
      "loss": 4.6677,
      "mean_token_accuracy": 0.26356969475746156,
      "num_tokens": 104084725.0,
      "step": 45385
    },
    {
      "entropy": 5.444989538192749,
      "epoch": 4.360230547550432,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032367864444036467,
      "loss": 4.655,
      "mean_token_accuracy": 0.26242454499006274,
      "num_tokens": 104096816.0,
      "step": 45390
    },
    {
      "entropy": 5.644735336303711,
      "epoch": 4.3607108549471665,
      "grad_norm": 1.25,
      "learning_rate": 0.00032364517229254894,
      "loss": 4.8107,
      "mean_token_accuracy": 0.23768325746059418,
      "num_tokens": 104108869.0,
      "step": 45395
    },
    {
      "entropy": 5.6090521812438965,
      "epoch": 4.3611911623439,
      "grad_norm": 1.25,
      "learning_rate": 0.00032361169901533673,
      "loss": 4.7429,
      "mean_token_accuracy": 0.25344652533531187,
      "num_tokens": 104120556.0,
      "step": 45400
    },
    {
      "entropy": 5.497588396072388,
      "epoch": 4.361671469740634,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032357822460950524,
      "loss": 4.6441,
      "mean_token_accuracy": 0.2558200716972351,
      "num_tokens": 104132253.0,
      "step": 45405
    },
    {
      "entropy": 5.5107409954071045,
      "epoch": 4.362151777137368,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003235447490758316,
      "loss": 4.6885,
      "mean_token_accuracy": 0.25961272418498993,
      "num_tokens": 104143211.0,
      "step": 45410
    },
    {
      "entropy": 5.457952690124512,
      "epoch": 4.362632084534102,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032351127241509305,
      "loss": 4.5601,
      "mean_token_accuracy": 0.26251655369997023,
      "num_tokens": 104154482.0,
      "step": 45415
    },
    {
      "entropy": 5.572168254852295,
      "epoch": 4.363112391930835,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003234777946280668,
      "loss": 4.7508,
      "mean_token_accuracy": 0.2496996268630028,
      "num_tokens": 104165681.0,
      "step": 45420
    },
    {
      "entropy": 5.5123673439025875,
      "epoch": 4.36359269932757,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032344431571553004,
      "loss": 4.6254,
      "mean_token_accuracy": 0.2632372960448265,
      "num_tokens": 104177363.0,
      "step": 45425
    },
    {
      "entropy": 5.4743083953857425,
      "epoch": 4.364073006724303,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032341083567826026,
      "loss": 4.6788,
      "mean_token_accuracy": 0.25238995552062987,
      "num_tokens": 104188521.0,
      "step": 45430
    },
    {
      "entropy": 5.51814112663269,
      "epoch": 4.3645533141210375,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032337735451703456,
      "loss": 4.7483,
      "mean_token_accuracy": 0.24738895297050476,
      "num_tokens": 104200597.0,
      "step": 45435
    },
    {
      "entropy": 5.608283519744873,
      "epoch": 4.365033621517771,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032334387223263033,
      "loss": 4.7849,
      "mean_token_accuracy": 0.24413712620735167,
      "num_tokens": 104213528.0,
      "step": 45440
    },
    {
      "entropy": 5.547825622558594,
      "epoch": 4.365513928914505,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032331038882582494,
      "loss": 4.7315,
      "mean_token_accuracy": 0.252690526843071,
      "num_tokens": 104224314.0,
      "step": 45445
    },
    {
      "entropy": 5.457800006866455,
      "epoch": 4.36599423631124,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003232769042973958,
      "loss": 4.6259,
      "mean_token_accuracy": 0.2582730546593666,
      "num_tokens": 104235369.0,
      "step": 45450
    },
    {
      "entropy": 5.447577905654907,
      "epoch": 4.366474543707973,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003232434186481203,
      "loss": 4.6509,
      "mean_token_accuracy": 0.2618774399161339,
      "num_tokens": 104246430.0,
      "step": 45455
    },
    {
      "entropy": 5.6107241153717045,
      "epoch": 4.366954851104707,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003232099318787758,
      "loss": 4.825,
      "mean_token_accuracy": 0.24445984810590743,
      "num_tokens": 104257904.0,
      "step": 45460
    },
    {
      "entropy": 5.5459291458129885,
      "epoch": 4.367435158501441,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003231764439901399,
      "loss": 4.6584,
      "mean_token_accuracy": 0.25771324932575224,
      "num_tokens": 104268546.0,
      "step": 45465
    },
    {
      "entropy": 5.562265634536743,
      "epoch": 4.367915465898175,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032314295498298993,
      "loss": 4.6935,
      "mean_token_accuracy": 0.2570853665471077,
      "num_tokens": 104279630.0,
      "step": 45470
    },
    {
      "entropy": 5.435853815078735,
      "epoch": 4.368395773294909,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032310946485810364,
      "loss": 4.5669,
      "mean_token_accuracy": 0.2659861221909523,
      "num_tokens": 104291700.0,
      "step": 45475
    },
    {
      "entropy": 5.503277826309204,
      "epoch": 4.368876080691643,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032307597361625834,
      "loss": 4.6972,
      "mean_token_accuracy": 0.2556163176894188,
      "num_tokens": 104304458.0,
      "step": 45480
    },
    {
      "entropy": 5.501146221160889,
      "epoch": 4.369356388088376,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003230424812582318,
      "loss": 4.7028,
      "mean_token_accuracy": 0.2603914588689804,
      "num_tokens": 104315276.0,
      "step": 45485
    },
    {
      "entropy": 5.508283233642578,
      "epoch": 4.369836695485111,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032300898778480134,
      "loss": 4.5695,
      "mean_token_accuracy": 0.26874043643474577,
      "num_tokens": 104327261.0,
      "step": 45490
    },
    {
      "entropy": 5.518549919128418,
      "epoch": 4.370317002881844,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032297549319674487,
      "loss": 4.6961,
      "mean_token_accuracy": 0.2509321630001068,
      "num_tokens": 104338756.0,
      "step": 45495
    },
    {
      "entropy": 5.5343897342681885,
      "epoch": 4.370797310278578,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032294199749483986,
      "loss": 4.6382,
      "mean_token_accuracy": 0.2590306654572487,
      "num_tokens": 104350041.0,
      "step": 45500
    },
    {
      "entropy": 5.6526929378509525,
      "epoch": 4.371277617675312,
      "grad_norm": 1.3125,
      "learning_rate": 0.000322908500679864,
      "loss": 4.863,
      "mean_token_accuracy": 0.2436918944120407,
      "num_tokens": 104360418.0,
      "step": 45505
    },
    {
      "entropy": 5.531517839431762,
      "epoch": 4.371757925072046,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032287500275259506,
      "loss": 4.6648,
      "mean_token_accuracy": 0.2583818927407265,
      "num_tokens": 104372099.0,
      "step": 45510
    },
    {
      "entropy": 5.451370429992676,
      "epoch": 4.37223823246878,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032284150371381066,
      "loss": 4.6374,
      "mean_token_accuracy": 0.2614053875207901,
      "num_tokens": 104382223.0,
      "step": 45515
    },
    {
      "entropy": 5.56297287940979,
      "epoch": 4.372718539865514,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003228080035642886,
      "loss": 4.7624,
      "mean_token_accuracy": 0.25410415083169935,
      "num_tokens": 104393377.0,
      "step": 45520
    },
    {
      "entropy": 5.522823810577393,
      "epoch": 4.373198847262248,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003227745023048067,
      "loss": 4.7009,
      "mean_token_accuracy": 0.2613919675350189,
      "num_tokens": 104405474.0,
      "step": 45525
    },
    {
      "entropy": 5.547591733932495,
      "epoch": 4.373679154658982,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032274099993614275,
      "loss": 4.7715,
      "mean_token_accuracy": 0.25535869002342226,
      "num_tokens": 104416381.0,
      "step": 45530
    },
    {
      "entropy": 5.453718042373657,
      "epoch": 4.374159462055716,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032270749645907447,
      "loss": 4.6399,
      "mean_token_accuracy": 0.2598991826176643,
      "num_tokens": 104426410.0,
      "step": 45535
    },
    {
      "entropy": 5.5648997783660885,
      "epoch": 4.3746397694524495,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032267399187437977,
      "loss": 4.7726,
      "mean_token_accuracy": 0.24686568677425386,
      "num_tokens": 104437800.0,
      "step": 45540
    },
    {
      "entropy": 5.520082759857178,
      "epoch": 4.375120076849184,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003226404861828366,
      "loss": 4.6446,
      "mean_token_accuracy": 0.26885059773921965,
      "num_tokens": 104449467.0,
      "step": 45545
    },
    {
      "entropy": 5.465857410430909,
      "epoch": 4.375600384245917,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003226069793852229,
      "loss": 4.6013,
      "mean_token_accuracy": 0.26117583066225053,
      "num_tokens": 104460057.0,
      "step": 45550
    },
    {
      "entropy": 5.53790807723999,
      "epoch": 4.376080691642652,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003225734714823163,
      "loss": 4.7666,
      "mean_token_accuracy": 0.2522888869047165,
      "num_tokens": 104471251.0,
      "step": 45555
    },
    {
      "entropy": 5.5623894214630125,
      "epoch": 4.376560999039385,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003225399624748951,
      "loss": 4.6927,
      "mean_token_accuracy": 0.25465400964021684,
      "num_tokens": 104482340.0,
      "step": 45560
    },
    {
      "entropy": 5.52806978225708,
      "epoch": 4.377041306436119,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003225064523637371,
      "loss": 4.7029,
      "mean_token_accuracy": 0.2566553711891174,
      "num_tokens": 104493163.0,
      "step": 45565
    },
    {
      "entropy": 5.465260124206543,
      "epoch": 4.377521613832853,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003224729411496203,
      "loss": 4.6579,
      "mean_token_accuracy": 0.25439535528421403,
      "num_tokens": 104505240.0,
      "step": 45570
    },
    {
      "entropy": 5.440442037582398,
      "epoch": 4.378001921229587,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003224394288333228,
      "loss": 4.6032,
      "mean_token_accuracy": 0.2585084542632103,
      "num_tokens": 104516594.0,
      "step": 45575
    },
    {
      "entropy": 5.478280448913575,
      "epoch": 4.3784822286263205,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003224059154156226,
      "loss": 4.5916,
      "mean_token_accuracy": 0.26753092408180235,
      "num_tokens": 104529335.0,
      "step": 45580
    },
    {
      "entropy": 5.456008195877075,
      "epoch": 4.378962536023055,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032237240089729786,
      "loss": 4.6132,
      "mean_token_accuracy": 0.267363241314888,
      "num_tokens": 104540795.0,
      "step": 45585
    },
    {
      "entropy": 5.506799983978271,
      "epoch": 4.379442843419788,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032233888527912657,
      "loss": 4.6815,
      "mean_token_accuracy": 0.25539650171995165,
      "num_tokens": 104551834.0,
      "step": 45590
    },
    {
      "entropy": 5.475774526596069,
      "epoch": 4.379923150816523,
      "grad_norm": 1.296875,
      "learning_rate": 0.000322305368561887,
      "loss": 4.6876,
      "mean_token_accuracy": 0.2504399254918098,
      "num_tokens": 104564931.0,
      "step": 45595
    },
    {
      "entropy": 5.563869190216065,
      "epoch": 4.380403458213257,
      "grad_norm": 1.65625,
      "learning_rate": 0.0003222718507463572,
      "loss": 4.6542,
      "mean_token_accuracy": 0.26550161838531494,
      "num_tokens": 104576865.0,
      "step": 45600
    },
    {
      "entropy": 5.503190612792968,
      "epoch": 4.38088376560999,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032223833183331546,
      "loss": 4.6454,
      "mean_token_accuracy": 0.25908231139183047,
      "num_tokens": 104587831.0,
      "step": 45605
    },
    {
      "entropy": 5.5579283237457275,
      "epoch": 4.381364073006725,
      "grad_norm": 1.25,
      "learning_rate": 0.00032220481182353985,
      "loss": 4.7232,
      "mean_token_accuracy": 0.24967961013317108,
      "num_tokens": 104598722.0,
      "step": 45610
    },
    {
      "entropy": 5.5552599906921385,
      "epoch": 4.381844380403458,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032217129071780877,
      "loss": 4.7368,
      "mean_token_accuracy": 0.2458847165107727,
      "num_tokens": 104609709.0,
      "step": 45615
    },
    {
      "entropy": 5.5145892143249515,
      "epoch": 4.3823246878001925,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003221377685169003,
      "loss": 4.6988,
      "mean_token_accuracy": 0.2543454736471176,
      "num_tokens": 104620893.0,
      "step": 45620
    },
    {
      "entropy": 5.517578887939453,
      "epoch": 4.382804995196926,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003221042452215928,
      "loss": 4.7016,
      "mean_token_accuracy": 0.24492756724357606,
      "num_tokens": 104633273.0,
      "step": 45625
    },
    {
      "entropy": 5.513828563690185,
      "epoch": 4.38328530259366,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003220707208326647,
      "loss": 4.6957,
      "mean_token_accuracy": 0.24845919162034988,
      "num_tokens": 104643741.0,
      "step": 45630
    },
    {
      "entropy": 5.500738096237183,
      "epoch": 4.383765609990394,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032203719535089416,
      "loss": 4.648,
      "mean_token_accuracy": 0.2603710934519768,
      "num_tokens": 104655515.0,
      "step": 45635
    },
    {
      "entropy": 5.619020700454712,
      "epoch": 4.384245917387128,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003220036687770596,
      "loss": 4.8148,
      "mean_token_accuracy": 0.2539605498313904,
      "num_tokens": 104666234.0,
      "step": 45640
    },
    {
      "entropy": 5.5838103771209715,
      "epoch": 4.384726224783861,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003219701411119395,
      "loss": 4.7439,
      "mean_token_accuracy": 0.2539485231041908,
      "num_tokens": 104677610.0,
      "step": 45645
    },
    {
      "entropy": 5.5480721950531,
      "epoch": 4.385206532180596,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032193661235631213,
      "loss": 4.6192,
      "mean_token_accuracy": 0.25824839621782303,
      "num_tokens": 104688453.0,
      "step": 45650
    },
    {
      "entropy": 5.489593935012818,
      "epoch": 4.385686839577329,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032190308251095603,
      "loss": 4.6082,
      "mean_token_accuracy": 0.26367749869823454,
      "num_tokens": 104700014.0,
      "step": 45655
    },
    {
      "entropy": 5.496459484100342,
      "epoch": 4.3861671469740635,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003218695515766496,
      "loss": 4.6998,
      "mean_token_accuracy": 0.25279911458492277,
      "num_tokens": 104711320.0,
      "step": 45660
    },
    {
      "entropy": 5.510981225967408,
      "epoch": 4.386647454370797,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00032183601955417137,
      "loss": 4.6641,
      "mean_token_accuracy": 0.259125842154026,
      "num_tokens": 104721971.0,
      "step": 45665
    },
    {
      "entropy": 5.585281848907471,
      "epoch": 4.387127761767531,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003218024864442998,
      "loss": 4.757,
      "mean_token_accuracy": 0.252734038233757,
      "num_tokens": 104733875.0,
      "step": 45670
    },
    {
      "entropy": 5.496800231933594,
      "epoch": 4.387608069164266,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032176895224781353,
      "loss": 4.6041,
      "mean_token_accuracy": 0.2648328572511673,
      "num_tokens": 104744575.0,
      "step": 45675
    },
    {
      "entropy": 5.396614265441895,
      "epoch": 4.388088376560999,
      "grad_norm": 1.25,
      "learning_rate": 0.0003217354169654911,
      "loss": 4.5307,
      "mean_token_accuracy": 0.26399931460618975,
      "num_tokens": 104755703.0,
      "step": 45680
    },
    {
      "entropy": 5.535372161865235,
      "epoch": 4.388568683957733,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032170188059811095,
      "loss": 4.6947,
      "mean_token_accuracy": 0.2576228305697441,
      "num_tokens": 104766334.0,
      "step": 45685
    },
    {
      "entropy": 5.504290008544922,
      "epoch": 4.389048991354467,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032166834314645184,
      "loss": 4.6718,
      "mean_token_accuracy": 0.26022865921258925,
      "num_tokens": 104778326.0,
      "step": 45690
    },
    {
      "entropy": 5.4746746063232425,
      "epoch": 4.389529298751201,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003216348046112924,
      "loss": 4.6777,
      "mean_token_accuracy": 0.2539794921875,
      "num_tokens": 104789936.0,
      "step": 45695
    },
    {
      "entropy": 5.577811431884766,
      "epoch": 4.390009606147935,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032160126499341125,
      "loss": 4.7414,
      "mean_token_accuracy": 0.24884448498487471,
      "num_tokens": 104801539.0,
      "step": 45700
    },
    {
      "entropy": 5.508926105499268,
      "epoch": 4.390489913544669,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032156772429358716,
      "loss": 4.6575,
      "mean_token_accuracy": 0.25891243368387223,
      "num_tokens": 104813889.0,
      "step": 45705
    },
    {
      "entropy": 5.619964456558227,
      "epoch": 4.390970220941402,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032153418251259873,
      "loss": 4.823,
      "mean_token_accuracy": 0.24014477133750917,
      "num_tokens": 104826116.0,
      "step": 45710
    },
    {
      "entropy": 5.5395101547241214,
      "epoch": 4.391450528338137,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032150063965122485,
      "loss": 4.6772,
      "mean_token_accuracy": 0.2477065160870552,
      "num_tokens": 104837350.0,
      "step": 45715
    },
    {
      "entropy": 5.443986606597901,
      "epoch": 4.39193083573487,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032146709571024407,
      "loss": 4.6103,
      "mean_token_accuracy": 0.2653942614793777,
      "num_tokens": 104848382.0,
      "step": 45720
    },
    {
      "entropy": 5.4681778907775875,
      "epoch": 4.392411143131604,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003214335506904354,
      "loss": 4.6749,
      "mean_token_accuracy": 0.2560461267828941,
      "num_tokens": 104859598.0,
      "step": 45725
    },
    {
      "entropy": 5.401128053665161,
      "epoch": 4.392891450528338,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032140000459257757,
      "loss": 4.5681,
      "mean_token_accuracy": 0.26250134855508805,
      "num_tokens": 104871070.0,
      "step": 45730
    },
    {
      "entropy": 5.407061576843262,
      "epoch": 4.393371757925072,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032136645741744936,
      "loss": 4.6335,
      "mean_token_accuracy": 0.26224365532398225,
      "num_tokens": 104883811.0,
      "step": 45735
    },
    {
      "entropy": 5.6192567348480225,
      "epoch": 4.393852065321806,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032133290916582966,
      "loss": 4.7994,
      "mean_token_accuracy": 0.24703235030174256,
      "num_tokens": 104894100.0,
      "step": 45740
    },
    {
      "entropy": 5.588043308258056,
      "epoch": 4.39433237271854,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003212993598384975,
      "loss": 4.654,
      "mean_token_accuracy": 0.2606150969862938,
      "num_tokens": 104906353.0,
      "step": 45745
    },
    {
      "entropy": 5.469870567321777,
      "epoch": 4.394812680115273,
      "grad_norm": 1.375,
      "learning_rate": 0.0003212658094362316,
      "loss": 4.5608,
      "mean_token_accuracy": 0.26295554339885713,
      "num_tokens": 104916810.0,
      "step": 45750
    },
    {
      "entropy": 5.475241851806641,
      "epoch": 4.395292987512008,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032123225795981097,
      "loss": 4.6421,
      "mean_token_accuracy": 0.25963430255651476,
      "num_tokens": 104927904.0,
      "step": 45755
    },
    {
      "entropy": 5.504133701324463,
      "epoch": 4.395773294908741,
      "grad_norm": 1.25,
      "learning_rate": 0.00032119870541001466,
      "loss": 4.7321,
      "mean_token_accuracy": 0.25331525802612304,
      "num_tokens": 104939993.0,
      "step": 45760
    },
    {
      "entropy": 5.545083618164062,
      "epoch": 4.3962536023054755,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003211651517876216,
      "loss": 4.7268,
      "mean_token_accuracy": 0.24988382160663605,
      "num_tokens": 104951004.0,
      "step": 45765
    },
    {
      "entropy": 5.542053604125977,
      "epoch": 4.39673390970221,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032113159709341074,
      "loss": 4.703,
      "mean_token_accuracy": 0.25798029750585555,
      "num_tokens": 104963684.0,
      "step": 45770
    },
    {
      "entropy": 5.533334636688233,
      "epoch": 4.397214217098943,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032109804132816113,
      "loss": 4.6845,
      "mean_token_accuracy": 0.2509001612663269,
      "num_tokens": 104975943.0,
      "step": 45775
    },
    {
      "entropy": 5.523946905136109,
      "epoch": 4.3976945244956775,
      "grad_norm": 1.265625,
      "learning_rate": 0.000321064484492652,
      "loss": 4.6488,
      "mean_token_accuracy": 0.26451202780008315,
      "num_tokens": 104987775.0,
      "step": 45780
    },
    {
      "entropy": 5.47424807548523,
      "epoch": 4.398174831892411,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032103092658766226,
      "loss": 4.5903,
      "mean_token_accuracy": 0.26453826725482943,
      "num_tokens": 104999420.0,
      "step": 45785
    },
    {
      "entropy": 5.5709761619567875,
      "epoch": 4.398655139289145,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032099736761397106,
      "loss": 4.7366,
      "mean_token_accuracy": 0.25650458633899687,
      "num_tokens": 105010759.0,
      "step": 45790
    },
    {
      "entropy": 5.495577239990235,
      "epoch": 4.399135446685879,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032096380757235774,
      "loss": 4.6744,
      "mean_token_accuracy": 0.255560377240181,
      "num_tokens": 105021386.0,
      "step": 45795
    },
    {
      "entropy": 5.480046129226684,
      "epoch": 4.399615754082613,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003209302464636011,
      "loss": 4.6498,
      "mean_token_accuracy": 0.26319126337766646,
      "num_tokens": 105031917.0,
      "step": 45800
    },
    {
      "entropy": 5.451712131500244,
      "epoch": 4.4000960614793465,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003208966842884806,
      "loss": 4.6489,
      "mean_token_accuracy": 0.25888400971889497,
      "num_tokens": 105043020.0,
      "step": 45805
    },
    {
      "entropy": 5.5474607944488525,
      "epoch": 4.400576368876081,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003208631210477754,
      "loss": 4.7617,
      "mean_token_accuracy": 0.2498544469475746,
      "num_tokens": 105053453.0,
      "step": 45810
    },
    {
      "entropy": 5.5656355857849125,
      "epoch": 4.401056676272814,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003208295567422646,
      "loss": 4.7421,
      "mean_token_accuracy": 0.25564656406641006,
      "num_tokens": 105064268.0,
      "step": 45815
    },
    {
      "entropy": 5.547365522384643,
      "epoch": 4.401536983669549,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003207959913727277,
      "loss": 4.6456,
      "mean_token_accuracy": 0.2656495898962021,
      "num_tokens": 105075797.0,
      "step": 45820
    },
    {
      "entropy": 5.464559078216553,
      "epoch": 4.402017291066282,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032076242493994374,
      "loss": 4.6542,
      "mean_token_accuracy": 0.26518007516860964,
      "num_tokens": 105088310.0,
      "step": 45825
    },
    {
      "entropy": 5.509503412246704,
      "epoch": 4.402497598463016,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003207288574446923,
      "loss": 4.635,
      "mean_token_accuracy": 0.25566650927066803,
      "num_tokens": 105099240.0,
      "step": 45830
    },
    {
      "entropy": 5.500174283981323,
      "epoch": 4.40297790585975,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032069528888775244,
      "loss": 4.6541,
      "mean_token_accuracy": 0.25719973891973497,
      "num_tokens": 105109433.0,
      "step": 45835
    },
    {
      "entropy": 5.585388898849487,
      "epoch": 4.403458213256484,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003206617192699038,
      "loss": 4.7612,
      "mean_token_accuracy": 0.25464988052845,
      "num_tokens": 105121098.0,
      "step": 45840
    },
    {
      "entropy": 5.5365190505981445,
      "epoch": 4.4039385206532184,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032062814859192554,
      "loss": 4.7436,
      "mean_token_accuracy": 0.24891144633293152,
      "num_tokens": 105132978.0,
      "step": 45845
    },
    {
      "entropy": 5.451737356185913,
      "epoch": 4.404418828049952,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003205945768545972,
      "loss": 4.6211,
      "mean_token_accuracy": 0.26073912978172303,
      "num_tokens": 105145917.0,
      "step": 45850
    },
    {
      "entropy": 5.484078407287598,
      "epoch": 4.404899135446686,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003205610040586981,
      "loss": 4.6009,
      "mean_token_accuracy": 0.26883557736873626,
      "num_tokens": 105157444.0,
      "step": 45855
    },
    {
      "entropy": 5.565348768234253,
      "epoch": 4.40537944284342,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003205274302050078,
      "loss": 4.7071,
      "mean_token_accuracy": 0.2553897276520729,
      "num_tokens": 105168784.0,
      "step": 45860
    },
    {
      "entropy": 5.442002296447754,
      "epoch": 4.405859750240154,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003204938552943058,
      "loss": 4.6445,
      "mean_token_accuracy": 0.2593810424208641,
      "num_tokens": 105178964.0,
      "step": 45865
    },
    {
      "entropy": 5.485612392425537,
      "epoch": 4.406340057636887,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003204602793273716,
      "loss": 4.5978,
      "mean_token_accuracy": 0.26933740079402924,
      "num_tokens": 105189057.0,
      "step": 45870
    },
    {
      "entropy": 5.517908382415771,
      "epoch": 4.406820365033622,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032042670230498463,
      "loss": 4.7043,
      "mean_token_accuracy": 0.25367715358734133,
      "num_tokens": 105199477.0,
      "step": 45875
    },
    {
      "entropy": 5.426127672195435,
      "epoch": 4.407300672430355,
      "grad_norm": 1.125,
      "learning_rate": 0.0003203931242279245,
      "loss": 4.5651,
      "mean_token_accuracy": 0.2664554014801979,
      "num_tokens": 105212348.0,
      "step": 45880
    },
    {
      "entropy": 5.469936227798462,
      "epoch": 4.4077809798270895,
      "grad_norm": 1.25,
      "learning_rate": 0.0003203595450969709,
      "loss": 4.6821,
      "mean_token_accuracy": 0.252135357260704,
      "num_tokens": 105223956.0,
      "step": 45885
    },
    {
      "entropy": 5.56318302154541,
      "epoch": 4.408261287223823,
      "grad_norm": 1.25,
      "learning_rate": 0.0003203259649129033,
      "loss": 4.71,
      "mean_token_accuracy": 0.24875291734933852,
      "num_tokens": 105234848.0,
      "step": 45890
    },
    {
      "entropy": 5.548501253128052,
      "epoch": 4.408741594620557,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032029238367650145,
      "loss": 4.7362,
      "mean_token_accuracy": 0.2522242873907089,
      "num_tokens": 105246442.0,
      "step": 45895
    },
    {
      "entropy": 5.588698768615723,
      "epoch": 4.409221902017291,
      "grad_norm": 1.25,
      "learning_rate": 0.0003202588013885449,
      "loss": 4.7292,
      "mean_token_accuracy": 0.2554861634969711,
      "num_tokens": 105258183.0,
      "step": 45900
    },
    {
      "entropy": 5.5053263187408445,
      "epoch": 4.409702209414025,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032022521804981333,
      "loss": 4.654,
      "mean_token_accuracy": 0.25530389696359634,
      "num_tokens": 105269964.0,
      "step": 45905
    },
    {
      "entropy": 5.492571496963501,
      "epoch": 4.4101825168107585,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032019163366108654,
      "loss": 4.6504,
      "mean_token_accuracy": 0.25988986194133756,
      "num_tokens": 105282238.0,
      "step": 45910
    },
    {
      "entropy": 5.487559604644775,
      "epoch": 4.410662824207493,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003201580482231442,
      "loss": 4.6428,
      "mean_token_accuracy": 0.2590697929263115,
      "num_tokens": 105294841.0,
      "step": 45915
    },
    {
      "entropy": 5.510750246047974,
      "epoch": 4.411143131604227,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003201244617367661,
      "loss": 4.6341,
      "mean_token_accuracy": 0.2595221266150475,
      "num_tokens": 105305397.0,
      "step": 45920
    },
    {
      "entropy": 5.521962833404541,
      "epoch": 4.4116234390009605,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003200908742027319,
      "loss": 4.67,
      "mean_token_accuracy": 0.2533691629767418,
      "num_tokens": 105318601.0,
      "step": 45925
    },
    {
      "entropy": 5.511748123168945,
      "epoch": 4.412103746397695,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003200572856218216,
      "loss": 4.7015,
      "mean_token_accuracy": 0.25284676253795624,
      "num_tokens": 105329798.0,
      "step": 45930
    },
    {
      "entropy": 5.5125096321105955,
      "epoch": 4.412584053794428,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032002369599481485,
      "loss": 4.6561,
      "mean_token_accuracy": 0.2587107390165329,
      "num_tokens": 105340646.0,
      "step": 45935
    },
    {
      "entropy": 5.518161392211914,
      "epoch": 4.413064361191163,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003199901053224916,
      "loss": 4.7066,
      "mean_token_accuracy": 0.2511584132909775,
      "num_tokens": 105353026.0,
      "step": 45940
    },
    {
      "entropy": 5.468749856948852,
      "epoch": 4.413544668587896,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031995651360563164,
      "loss": 4.7024,
      "mean_token_accuracy": 0.25279681086540223,
      "num_tokens": 105364394.0,
      "step": 45945
    },
    {
      "entropy": 5.473783493041992,
      "epoch": 4.41402497598463,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031992292084501503,
      "loss": 4.6421,
      "mean_token_accuracy": 0.2641745075583458,
      "num_tokens": 105375630.0,
      "step": 45950
    },
    {
      "entropy": 5.510735607147216,
      "epoch": 4.414505283381364,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031988932704142146,
      "loss": 4.7289,
      "mean_token_accuracy": 0.25668237209320066,
      "num_tokens": 105387036.0,
      "step": 45955
    },
    {
      "entropy": 5.46125898361206,
      "epoch": 4.414985590778098,
      "grad_norm": 1.25,
      "learning_rate": 0.0003198557321956312,
      "loss": 4.6273,
      "mean_token_accuracy": 0.2563016712665558,
      "num_tokens": 105398752.0,
      "step": 45960
    },
    {
      "entropy": 5.469834518432617,
      "epoch": 4.415465898174832,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003198221363084239,
      "loss": 4.6444,
      "mean_token_accuracy": 0.2662768989801407,
      "num_tokens": 105410843.0,
      "step": 45965
    },
    {
      "entropy": 5.596012306213379,
      "epoch": 4.415946205571566,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003197885393805797,
      "loss": 4.7634,
      "mean_token_accuracy": 0.2561506018042564,
      "num_tokens": 105421999.0,
      "step": 45970
    },
    {
      "entropy": 5.574089241027832,
      "epoch": 4.416426512968299,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031975494141287864,
      "loss": 4.6936,
      "mean_token_accuracy": 0.2542667791247368,
      "num_tokens": 105433864.0,
      "step": 45975
    },
    {
      "entropy": 5.538476133346558,
      "epoch": 4.416906820365034,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003197213424061007,
      "loss": 4.7016,
      "mean_token_accuracy": 0.25803841203451156,
      "num_tokens": 105444924.0,
      "step": 45980
    },
    {
      "entropy": 5.499686813354492,
      "epoch": 4.417387127761767,
      "grad_norm": 1.3125,
      "learning_rate": 0.000319687742361026,
      "loss": 4.6774,
      "mean_token_accuracy": 0.25158836841583254,
      "num_tokens": 105455922.0,
      "step": 45985
    },
    {
      "entropy": 5.507193565368652,
      "epoch": 4.417867435158501,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031965414127843464,
      "loss": 4.7075,
      "mean_token_accuracy": 0.2557115748524666,
      "num_tokens": 105468352.0,
      "step": 45990
    },
    {
      "entropy": 5.54549503326416,
      "epoch": 4.418347742555236,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031962053915910664,
      "loss": 4.7219,
      "mean_token_accuracy": 0.25068873465061187,
      "num_tokens": 105479453.0,
      "step": 45995
    },
    {
      "entropy": 5.550672721862793,
      "epoch": 4.418828049951969,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031958693600382226,
      "loss": 4.6681,
      "mean_token_accuracy": 0.25468706637620925,
      "num_tokens": 105491512.0,
      "step": 46000
    },
    {
      "entropy": 5.552539396286011,
      "epoch": 4.4193083573487035,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003195533318133617,
      "loss": 4.7461,
      "mean_token_accuracy": 0.25374791771173477,
      "num_tokens": 105503159.0,
      "step": 46005
    },
    {
      "entropy": 5.518359851837158,
      "epoch": 4.419788664745437,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031951972658850495,
      "loss": 4.7036,
      "mean_token_accuracy": 0.25482640564441683,
      "num_tokens": 105514748.0,
      "step": 46010
    },
    {
      "entropy": 5.524551630020142,
      "epoch": 4.420268972142171,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031948612033003246,
      "loss": 4.6557,
      "mean_token_accuracy": 0.26061936765909194,
      "num_tokens": 105526046.0,
      "step": 46015
    },
    {
      "entropy": 5.509602642059326,
      "epoch": 4.420749279538905,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031945251303872426,
      "loss": 4.7459,
      "mean_token_accuracy": 0.24842083752155303,
      "num_tokens": 105539565.0,
      "step": 46020
    },
    {
      "entropy": 5.588917589187622,
      "epoch": 4.421229586935639,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031941890471536076,
      "loss": 4.7292,
      "mean_token_accuracy": 0.2483423501253128,
      "num_tokens": 105550562.0,
      "step": 46025
    },
    {
      "entropy": 5.581390762329102,
      "epoch": 4.4217098943323725,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031938529536072215,
      "loss": 4.7046,
      "mean_token_accuracy": 0.25627081245183947,
      "num_tokens": 105561715.0,
      "step": 46030
    },
    {
      "entropy": 5.52883505821228,
      "epoch": 4.422190201729107,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003193516849755888,
      "loss": 4.6561,
      "mean_token_accuracy": 0.25479593873023987,
      "num_tokens": 105572694.0,
      "step": 46035
    },
    {
      "entropy": 5.575862741470337,
      "epoch": 4.42267050912584,
      "grad_norm": 1.21875,
      "learning_rate": 0.000319318073560741,
      "loss": 4.7371,
      "mean_token_accuracy": 0.25167997032403944,
      "num_tokens": 105584586.0,
      "step": 46040
    },
    {
      "entropy": 5.449387788772583,
      "epoch": 4.423150816522575,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031928446111695915,
      "loss": 4.5921,
      "mean_token_accuracy": 0.2669928029179573,
      "num_tokens": 105595778.0,
      "step": 46045
    },
    {
      "entropy": 5.412598466873169,
      "epoch": 4.423631123919308,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003192508476450236,
      "loss": 4.5405,
      "mean_token_accuracy": 0.26675224155187605,
      "num_tokens": 105607986.0,
      "step": 46050
    },
    {
      "entropy": 5.497135877609253,
      "epoch": 4.424111431316042,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031921723314571476,
      "loss": 4.6619,
      "mean_token_accuracy": 0.25940761566162107,
      "num_tokens": 105620933.0,
      "step": 46055
    },
    {
      "entropy": 5.590590190887451,
      "epoch": 4.424591738712776,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031918361761981305,
      "loss": 4.8066,
      "mean_token_accuracy": 0.24447111785411835,
      "num_tokens": 105632113.0,
      "step": 46060
    },
    {
      "entropy": 5.599877500534058,
      "epoch": 4.42507204610951,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00031915000106809904,
      "loss": 4.724,
      "mean_token_accuracy": 0.2581471249461174,
      "num_tokens": 105643978.0,
      "step": 46065
    },
    {
      "entropy": 5.431103897094727,
      "epoch": 4.425552353506244,
      "grad_norm": 1.140625,
      "learning_rate": 0.000319116383491353,
      "loss": 4.5628,
      "mean_token_accuracy": 0.2682539612054825,
      "num_tokens": 105655193.0,
      "step": 46070
    },
    {
      "entropy": 5.4635590553283695,
      "epoch": 4.426032660902978,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003190827648903555,
      "loss": 4.7009,
      "mean_token_accuracy": 0.2505642995238304,
      "num_tokens": 105667144.0,
      "step": 46075
    },
    {
      "entropy": 5.560956573486328,
      "epoch": 4.426512968299712,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003190491452658872,
      "loss": 4.7225,
      "mean_token_accuracy": 0.25444667488336564,
      "num_tokens": 105679836.0,
      "step": 46080
    },
    {
      "entropy": 5.574412822723389,
      "epoch": 4.426993275696446,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003190155246187286,
      "loss": 4.7022,
      "mean_token_accuracy": 0.25319476127624513,
      "num_tokens": 105690437.0,
      "step": 46085
    },
    {
      "entropy": 5.526764297485352,
      "epoch": 4.42747358309318,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031898190294966006,
      "loss": 4.7389,
      "mean_token_accuracy": 0.24958496391773224,
      "num_tokens": 105701614.0,
      "step": 46090
    },
    {
      "entropy": 5.598005962371826,
      "epoch": 4.427953890489913,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003189482802594624,
      "loss": 4.8465,
      "mean_token_accuracy": 0.2478555828332901,
      "num_tokens": 105713725.0,
      "step": 46095
    },
    {
      "entropy": 5.545028924942017,
      "epoch": 4.428434197886648,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031891465654891615,
      "loss": 4.7179,
      "mean_token_accuracy": 0.2501699835062027,
      "num_tokens": 105726397.0,
      "step": 46100
    },
    {
      "entropy": 5.445765447616577,
      "epoch": 4.428914505283381,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000318881031818802,
      "loss": 4.5661,
      "mean_token_accuracy": 0.2681219756603241,
      "num_tokens": 105738401.0,
      "step": 46105
    },
    {
      "entropy": 5.489689445495605,
      "epoch": 4.4293948126801155,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031884740606990056,
      "loss": 4.5876,
      "mean_token_accuracy": 0.270773409307003,
      "num_tokens": 105749529.0,
      "step": 46110
    },
    {
      "entropy": 5.459286499023437,
      "epoch": 4.429875120076849,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031881377930299256,
      "loss": 4.6253,
      "mean_token_accuracy": 0.2674023285508156,
      "num_tokens": 105761965.0,
      "step": 46115
    },
    {
      "entropy": 5.4677684783935545,
      "epoch": 4.430355427473583,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003187801515188588,
      "loss": 4.6544,
      "mean_token_accuracy": 0.2623530238866806,
      "num_tokens": 105773358.0,
      "step": 46120
    },
    {
      "entropy": 5.499896669387818,
      "epoch": 4.430835734870317,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003187465227182798,
      "loss": 4.7146,
      "mean_token_accuracy": 0.257863937318325,
      "num_tokens": 105784815.0,
      "step": 46125
    },
    {
      "entropy": 5.479617500305176,
      "epoch": 4.431316042267051,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031871289290203646,
      "loss": 4.626,
      "mean_token_accuracy": 0.25562343299388884,
      "num_tokens": 105796383.0,
      "step": 46130
    },
    {
      "entropy": 5.511359071731567,
      "epoch": 4.431796349663784,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031867926207090956,
      "loss": 4.6923,
      "mean_token_accuracy": 0.2593562290072441,
      "num_tokens": 105808287.0,
      "step": 46135
    },
    {
      "entropy": 5.533601999282837,
      "epoch": 4.432276657060519,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003186456302256799,
      "loss": 4.6787,
      "mean_token_accuracy": 0.25316769629716873,
      "num_tokens": 105819111.0,
      "step": 46140
    },
    {
      "entropy": 5.496621370315552,
      "epoch": 4.432756964457253,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003186119973671283,
      "loss": 4.6821,
      "mean_token_accuracy": 0.25955477058887483,
      "num_tokens": 105830656.0,
      "step": 46145
    },
    {
      "entropy": 5.46826138496399,
      "epoch": 4.4332372718539865,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031857836349603565,
      "loss": 4.6167,
      "mean_token_accuracy": 0.2642111986875534,
      "num_tokens": 105843773.0,
      "step": 46150
    },
    {
      "entropy": 5.481546020507812,
      "epoch": 4.433717579250721,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031854472861318284,
      "loss": 4.642,
      "mean_token_accuracy": 0.25779951363801956,
      "num_tokens": 105855031.0,
      "step": 46155
    },
    {
      "entropy": 5.5053798198699955,
      "epoch": 4.434197886647454,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031851109271935064,
      "loss": 4.6375,
      "mean_token_accuracy": 0.2593184277415276,
      "num_tokens": 105865604.0,
      "step": 46160
    },
    {
      "entropy": 5.515771579742432,
      "epoch": 4.434678194044189,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003184774558153201,
      "loss": 4.7124,
      "mean_token_accuracy": 0.2561858415603638,
      "num_tokens": 105877406.0,
      "step": 46165
    },
    {
      "entropy": 5.463824987411499,
      "epoch": 4.435158501440922,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031844381790187217,
      "loss": 4.6669,
      "mean_token_accuracy": 0.26072334200143815,
      "num_tokens": 105889379.0,
      "step": 46170
    },
    {
      "entropy": 5.544010972976684,
      "epoch": 4.435638808837656,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031841017897978773,
      "loss": 4.6799,
      "mean_token_accuracy": 0.25318965017795564,
      "num_tokens": 105900712.0,
      "step": 46175
    },
    {
      "entropy": 5.521235084533691,
      "epoch": 4.43611911623439,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031837653904984793,
      "loss": 4.7142,
      "mean_token_accuracy": 0.25882029682397845,
      "num_tokens": 105912388.0,
      "step": 46180
    },
    {
      "entropy": 5.560369443893433,
      "epoch": 4.436599423631124,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003183428981128336,
      "loss": 4.7343,
      "mean_token_accuracy": 0.2511944055557251,
      "num_tokens": 105922391.0,
      "step": 46185
    },
    {
      "entropy": 5.48270468711853,
      "epoch": 4.437079731027858,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031830925616952593,
      "loss": 4.5624,
      "mean_token_accuracy": 0.2618590757250786,
      "num_tokens": 105932856.0,
      "step": 46190
    },
    {
      "entropy": 5.558771705627441,
      "epoch": 4.437560038424592,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003182756132207059,
      "loss": 4.7806,
      "mean_token_accuracy": 0.24908700585365295,
      "num_tokens": 105944703.0,
      "step": 46195
    },
    {
      "entropy": 5.4874646186828615,
      "epoch": 4.438040345821325,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003182419692671547,
      "loss": 4.5971,
      "mean_token_accuracy": 0.2683944493532181,
      "num_tokens": 105956734.0,
      "step": 46200
    },
    {
      "entropy": 5.462072944641113,
      "epoch": 4.43852065321806,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031820832430965335,
      "loss": 4.6563,
      "mean_token_accuracy": 0.26092215031385424,
      "num_tokens": 105967937.0,
      "step": 46205
    },
    {
      "entropy": 5.514260911941529,
      "epoch": 4.439000960614793,
      "grad_norm": 1.375,
      "learning_rate": 0.0003181746783489829,
      "loss": 4.7116,
      "mean_token_accuracy": 0.2580775499343872,
      "num_tokens": 105978505.0,
      "step": 46210
    },
    {
      "entropy": 5.517890405654907,
      "epoch": 4.439481268011527,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031814103138592476,
      "loss": 4.7186,
      "mean_token_accuracy": 0.257938452064991,
      "num_tokens": 105990267.0,
      "step": 46215
    },
    {
      "entropy": 5.573607015609741,
      "epoch": 4.439961575408262,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031810738342125995,
      "loss": 4.7236,
      "mean_token_accuracy": 0.2538390174508095,
      "num_tokens": 106002001.0,
      "step": 46220
    },
    {
      "entropy": 5.582072353363037,
      "epoch": 4.440441882804995,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003180737344557696,
      "loss": 4.7471,
      "mean_token_accuracy": 0.24854371398687364,
      "num_tokens": 106012769.0,
      "step": 46225
    },
    {
      "entropy": 5.481355142593384,
      "epoch": 4.4409221902017295,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003180400844902352,
      "loss": 4.6929,
      "mean_token_accuracy": 0.2552500620484352,
      "num_tokens": 106024317.0,
      "step": 46230
    },
    {
      "entropy": 5.493901491165161,
      "epoch": 4.441402497598463,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031800643352543773,
      "loss": 4.6887,
      "mean_token_accuracy": 0.2555053517222404,
      "num_tokens": 106035589.0,
      "step": 46235
    },
    {
      "entropy": 5.5712587356567385,
      "epoch": 4.441882804995197,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003179727815621586,
      "loss": 4.778,
      "mean_token_accuracy": 0.24933671653270723,
      "num_tokens": 106047897.0,
      "step": 46240
    },
    {
      "entropy": 5.459808397293091,
      "epoch": 4.442363112391931,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031793912860117907,
      "loss": 4.6101,
      "mean_token_accuracy": 0.26003405451774597,
      "num_tokens": 106059162.0,
      "step": 46245
    },
    {
      "entropy": 5.547631120681762,
      "epoch": 4.442843419788665,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003179054746432805,
      "loss": 4.7881,
      "mean_token_accuracy": 0.23962197750806807,
      "num_tokens": 106071442.0,
      "step": 46250
    },
    {
      "entropy": 5.546511507034301,
      "epoch": 4.4433237271853985,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031787181968924417,
      "loss": 4.6915,
      "mean_token_accuracy": 0.25347217470407485,
      "num_tokens": 106081696.0,
      "step": 46255
    },
    {
      "entropy": 5.530494165420532,
      "epoch": 4.443804034582133,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003178381637398515,
      "loss": 4.7039,
      "mean_token_accuracy": 0.2503719389438629,
      "num_tokens": 106093100.0,
      "step": 46260
    },
    {
      "entropy": 5.509852695465088,
      "epoch": 4.444284341978866,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031780450679588394,
      "loss": 4.6737,
      "mean_token_accuracy": 0.254303178191185,
      "num_tokens": 106104700.0,
      "step": 46265
    },
    {
      "entropy": 5.565346193313599,
      "epoch": 4.444764649375601,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003177708488581227,
      "loss": 4.7103,
      "mean_token_accuracy": 0.25683118104934693,
      "num_tokens": 106116593.0,
      "step": 46270
    },
    {
      "entropy": 5.564843416213989,
      "epoch": 4.445244956772334,
      "grad_norm": 1.25,
      "learning_rate": 0.0003177371899273496,
      "loss": 4.7455,
      "mean_token_accuracy": 0.254109825193882,
      "num_tokens": 106128558.0,
      "step": 46275
    },
    {
      "entropy": 5.666146564483642,
      "epoch": 4.445725264169068,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031770353000434566,
      "loss": 4.8252,
      "mean_token_accuracy": 0.24088504165410995,
      "num_tokens": 106139309.0,
      "step": 46280
    },
    {
      "entropy": 5.462948513031006,
      "epoch": 4.446205571565802,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003176698690898925,
      "loss": 4.662,
      "mean_token_accuracy": 0.25597624480724335,
      "num_tokens": 106150812.0,
      "step": 46285
    },
    {
      "entropy": 5.53485426902771,
      "epoch": 4.446685878962536,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003176362071847719,
      "loss": 4.6819,
      "mean_token_accuracy": 0.2593931004405022,
      "num_tokens": 106160986.0,
      "step": 46290
    },
    {
      "entropy": 5.423777294158936,
      "epoch": 4.4471661863592695,
      "grad_norm": 1.265625,
      "learning_rate": 0.000317602544289765,
      "loss": 4.6616,
      "mean_token_accuracy": 0.26063325107097624,
      "num_tokens": 106172125.0,
      "step": 46295
    },
    {
      "entropy": 5.5316712856292725,
      "epoch": 4.447646493756004,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031756888040565355,
      "loss": 4.7222,
      "mean_token_accuracy": 0.2539876863360405,
      "num_tokens": 106185405.0,
      "step": 46300
    },
    {
      "entropy": 5.565937805175781,
      "epoch": 4.448126801152737,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003175352155332192,
      "loss": 4.7115,
      "mean_token_accuracy": 0.2589643314480782,
      "num_tokens": 106196144.0,
      "step": 46305
    },
    {
      "entropy": 5.460125684738159,
      "epoch": 4.448607108549472,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031750154967324337,
      "loss": 4.6469,
      "mean_token_accuracy": 0.2603305011987686,
      "num_tokens": 106207708.0,
      "step": 46310
    },
    {
      "entropy": 5.585235214233398,
      "epoch": 4.449087415946206,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031746788282650774,
      "loss": 4.7777,
      "mean_token_accuracy": 0.25499268174171447,
      "num_tokens": 106218568.0,
      "step": 46315
    },
    {
      "entropy": 5.518364429473877,
      "epoch": 4.449567723342939,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000317434214993794,
      "loss": 4.6676,
      "mean_token_accuracy": 0.2664498150348663,
      "num_tokens": 106231188.0,
      "step": 46320
    },
    {
      "entropy": 5.524969673156738,
      "epoch": 4.450048030739674,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003174005461758838,
      "loss": 4.7059,
      "mean_token_accuracy": 0.2511515155434608,
      "num_tokens": 106242158.0,
      "step": 46325
    },
    {
      "entropy": 5.538444948196411,
      "epoch": 4.450528338136407,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003173668763735588,
      "loss": 4.6609,
      "mean_token_accuracy": 0.26061938256025313,
      "num_tokens": 106252382.0,
      "step": 46330
    },
    {
      "entropy": 5.4890978813171385,
      "epoch": 4.4510086455331415,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031733320558760076,
      "loss": 4.7402,
      "mean_token_accuracy": 0.25179931372404096,
      "num_tokens": 106265033.0,
      "step": 46335
    },
    {
      "entropy": 5.4794281959533695,
      "epoch": 4.451488952929875,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031729953381879136,
      "loss": 4.7149,
      "mean_token_accuracy": 0.25377084910869596,
      "num_tokens": 106277776.0,
      "step": 46340
    },
    {
      "entropy": 5.430533599853516,
      "epoch": 4.451969260326609,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003172658610679124,
      "loss": 4.6269,
      "mean_token_accuracy": 0.25687346756458285,
      "num_tokens": 106290031.0,
      "step": 46345
    },
    {
      "entropy": 5.5516785144805905,
      "epoch": 4.452449567723343,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003172321873357457,
      "loss": 4.7418,
      "mean_token_accuracy": 0.2526622638106346,
      "num_tokens": 106300889.0,
      "step": 46350
    },
    {
      "entropy": 5.540915632247925,
      "epoch": 4.452929875120077,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031719851262307293,
      "loss": 4.6973,
      "mean_token_accuracy": 0.25465422719717024,
      "num_tokens": 106314321.0,
      "step": 46355
    },
    {
      "entropy": 5.509042739868164,
      "epoch": 4.45341018251681,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000317164836930676,
      "loss": 4.7214,
      "mean_token_accuracy": 0.2537279173731804,
      "num_tokens": 106325930.0,
      "step": 46360
    },
    {
      "entropy": 5.568079948425293,
      "epoch": 4.453890489913545,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031713116025933684,
      "loss": 4.7326,
      "mean_token_accuracy": 0.25453432500362394,
      "num_tokens": 106338295.0,
      "step": 46365
    },
    {
      "entropy": 5.4030379295349125,
      "epoch": 4.454370797310278,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031709748260983726,
      "loss": 4.5921,
      "mean_token_accuracy": 0.2614446997642517,
      "num_tokens": 106350140.0,
      "step": 46370
    },
    {
      "entropy": 5.477293825149536,
      "epoch": 4.4548511047070125,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031706380398295895,
      "loss": 4.6586,
      "mean_token_accuracy": 0.25913745909929276,
      "num_tokens": 106362923.0,
      "step": 46375
    },
    {
      "entropy": 5.5062473773956295,
      "epoch": 4.455331412103746,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003170301243794842,
      "loss": 4.7198,
      "mean_token_accuracy": 0.25454009175300596,
      "num_tokens": 106374447.0,
      "step": 46380
    },
    {
      "entropy": 5.474219036102295,
      "epoch": 4.45581171950048,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003169964438001948,
      "loss": 4.6498,
      "mean_token_accuracy": 0.2598590895533562,
      "num_tokens": 106386543.0,
      "step": 46385
    },
    {
      "entropy": 5.530458259582519,
      "epoch": 4.456292026897215,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003169627622458725,
      "loss": 4.7441,
      "mean_token_accuracy": 0.25129126012325287,
      "num_tokens": 106397805.0,
      "step": 46390
    },
    {
      "entropy": 5.566376256942749,
      "epoch": 4.456772334293948,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003169290797172996,
      "loss": 4.7767,
      "mean_token_accuracy": 0.24934292286634446,
      "num_tokens": 106409208.0,
      "step": 46395
    },
    {
      "entropy": 5.505934333801269,
      "epoch": 4.457252641690682,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003168953962152578,
      "loss": 4.6596,
      "mean_token_accuracy": 0.2592508628964424,
      "num_tokens": 106421159.0,
      "step": 46400
    },
    {
      "entropy": 5.5242870330810545,
      "epoch": 4.457732949087416,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003168617117405295,
      "loss": 4.6988,
      "mean_token_accuracy": 0.2541607514023781,
      "num_tokens": 106432826.0,
      "step": 46405
    },
    {
      "entropy": 5.495956993103027,
      "epoch": 4.45821325648415,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003168280262938965,
      "loss": 4.6277,
      "mean_token_accuracy": 0.2617301791906357,
      "num_tokens": 106443959.0,
      "step": 46410
    },
    {
      "entropy": 5.497550058364868,
      "epoch": 4.458693563880884,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003167943398761409,
      "loss": 4.6467,
      "mean_token_accuracy": 0.26026311665773394,
      "num_tokens": 106455087.0,
      "step": 46415
    },
    {
      "entropy": 5.492155122756958,
      "epoch": 4.459173871277618,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031676065248804485,
      "loss": 4.6153,
      "mean_token_accuracy": 0.2572315663099289,
      "num_tokens": 106466296.0,
      "step": 46420
    },
    {
      "entropy": 5.5404948711395265,
      "epoch": 4.459654178674351,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031672696413039046,
      "loss": 4.7235,
      "mean_token_accuracy": 0.24963639974594115,
      "num_tokens": 106476072.0,
      "step": 46425
    },
    {
      "entropy": 5.547529411315918,
      "epoch": 4.460134486071086,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031669327480395986,
      "loss": 4.7366,
      "mean_token_accuracy": 0.24745362102985383,
      "num_tokens": 106487309.0,
      "step": 46430
    },
    {
      "entropy": 5.55568380355835,
      "epoch": 4.460614793467819,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003166595845095352,
      "loss": 4.7499,
      "mean_token_accuracy": 0.24725572019815445,
      "num_tokens": 106498661.0,
      "step": 46435
    },
    {
      "entropy": 5.580624198913574,
      "epoch": 4.461095100864553,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031662589324789873,
      "loss": 4.7022,
      "mean_token_accuracy": 0.24920917749404908,
      "num_tokens": 106509809.0,
      "step": 46440
    },
    {
      "entropy": 5.454278564453125,
      "epoch": 4.461575408261287,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003165922010198326,
      "loss": 4.6484,
      "mean_token_accuracy": 0.26244388371706007,
      "num_tokens": 106522662.0,
      "step": 46445
    },
    {
      "entropy": 5.521432828903198,
      "epoch": 4.462055715658021,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031655850782611903,
      "loss": 4.6743,
      "mean_token_accuracy": 0.2518733814358711,
      "num_tokens": 106534176.0,
      "step": 46450
    },
    {
      "entropy": 5.51481032371521,
      "epoch": 4.462536023054755,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031652481366754046,
      "loss": 4.6488,
      "mean_token_accuracy": 0.2586863860487938,
      "num_tokens": 106544626.0,
      "step": 46455
    },
    {
      "entropy": 5.537321281433106,
      "epoch": 4.463016330451489,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003164911185448789,
      "loss": 4.8065,
      "mean_token_accuracy": 0.24360592812299728,
      "num_tokens": 106557112.0,
      "step": 46460
    },
    {
      "entropy": 5.577640104293823,
      "epoch": 4.463496637848223,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003164574224589168,
      "loss": 4.7908,
      "mean_token_accuracy": 0.2497845247387886,
      "num_tokens": 106568226.0,
      "step": 46465
    },
    {
      "entropy": 5.5811761856079105,
      "epoch": 4.463976945244957,
      "grad_norm": 1.125,
      "learning_rate": 0.0003164237254104364,
      "loss": 4.7321,
      "mean_token_accuracy": 0.24850240349769592,
      "num_tokens": 106580592.0,
      "step": 46470
    },
    {
      "entropy": 5.542504739761353,
      "epoch": 4.464457252641691,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031639002740022017,
      "loss": 4.7817,
      "mean_token_accuracy": 0.2465704783797264,
      "num_tokens": 106591655.0,
      "step": 46475
    },
    {
      "entropy": 5.550870895385742,
      "epoch": 4.4649375600384245,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003163563284290504,
      "loss": 4.6908,
      "mean_token_accuracy": 0.25551681220531464,
      "num_tokens": 106603852.0,
      "step": 46480
    },
    {
      "entropy": 5.541644620895386,
      "epoch": 4.465417867435159,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031632262849770945,
      "loss": 4.7266,
      "mean_token_accuracy": 0.25002380311489103,
      "num_tokens": 106616237.0,
      "step": 46485
    },
    {
      "entropy": 5.532409429550171,
      "epoch": 4.465898174831892,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031628892760697974,
      "loss": 4.7176,
      "mean_token_accuracy": 0.2547330528497696,
      "num_tokens": 106626812.0,
      "step": 46490
    },
    {
      "entropy": 5.492036771774292,
      "epoch": 4.4663784822286265,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031625522575764374,
      "loss": 4.6623,
      "mean_token_accuracy": 0.25778899639844893,
      "num_tokens": 106636791.0,
      "step": 46495
    },
    {
      "entropy": 5.568886137008667,
      "epoch": 4.46685878962536,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000316221522950484,
      "loss": 4.7686,
      "mean_token_accuracy": 0.25235469043254855,
      "num_tokens": 106647838.0,
      "step": 46500
    },
    {
      "entropy": 5.504886817932129,
      "epoch": 4.467339097022094,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003161878191862829,
      "loss": 4.6352,
      "mean_token_accuracy": 0.25521883815526963,
      "num_tokens": 106659128.0,
      "step": 46505
    },
    {
      "entropy": 5.599607849121094,
      "epoch": 4.467819404418828,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003161541144658228,
      "loss": 4.7385,
      "mean_token_accuracy": 0.2579991206526756,
      "num_tokens": 106670276.0,
      "step": 46510
    },
    {
      "entropy": 5.529261684417724,
      "epoch": 4.468299711815562,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031612040878988647,
      "loss": 4.6929,
      "mean_token_accuracy": 0.25087714195251465,
      "num_tokens": 106681199.0,
      "step": 46515
    },
    {
      "entropy": 5.4749351978302006,
      "epoch": 4.4687800192122955,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031608670215925634,
      "loss": 4.7116,
      "mean_token_accuracy": 0.25546321272850037,
      "num_tokens": 106693544.0,
      "step": 46520
    },
    {
      "entropy": 5.407833433151245,
      "epoch": 4.46926032660903,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031605299457471495,
      "loss": 4.6174,
      "mean_token_accuracy": 0.26443699896335604,
      "num_tokens": 106704390.0,
      "step": 46525
    },
    {
      "entropy": 5.580186748504639,
      "epoch": 4.469740634005763,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003160192860370449,
      "loss": 4.7204,
      "mean_token_accuracy": 0.2542656555771828,
      "num_tokens": 106715620.0,
      "step": 46530
    },
    {
      "entropy": 5.612737417221069,
      "epoch": 4.470220941402498,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031598557654702894,
      "loss": 4.7929,
      "mean_token_accuracy": 0.24792052209377288,
      "num_tokens": 106726541.0,
      "step": 46535
    },
    {
      "entropy": 5.517569732666016,
      "epoch": 4.470701248799232,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031595186610544954,
      "loss": 4.7194,
      "mean_token_accuracy": 0.2480940118432045,
      "num_tokens": 106737317.0,
      "step": 46540
    },
    {
      "entropy": 5.466408824920654,
      "epoch": 4.471181556195965,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031591815471308945,
      "loss": 4.6058,
      "mean_token_accuracy": 0.2625848054885864,
      "num_tokens": 106749634.0,
      "step": 46545
    },
    {
      "entropy": 5.5172513961792,
      "epoch": 4.4716618635927,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003158844423707312,
      "loss": 4.7024,
      "mean_token_accuracy": 0.25954050570726395,
      "num_tokens": 106760674.0,
      "step": 46550
    },
    {
      "entropy": 5.666163063049316,
      "epoch": 4.472142170989433,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003158507290791577,
      "loss": 4.8816,
      "mean_token_accuracy": 0.24612315446138383,
      "num_tokens": 106772783.0,
      "step": 46555
    },
    {
      "entropy": 5.510955286026001,
      "epoch": 4.472622478386167,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031581701483915153,
      "loss": 4.5979,
      "mean_token_accuracy": 0.2743449002504349,
      "num_tokens": 106785379.0,
      "step": 46560
    },
    {
      "entropy": 5.457203722000122,
      "epoch": 4.473102785782901,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031578329965149544,
      "loss": 4.6793,
      "mean_token_accuracy": 0.26008242517709734,
      "num_tokens": 106796185.0,
      "step": 46565
    },
    {
      "entropy": 5.536980152130127,
      "epoch": 4.473583093179635,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003157495835169723,
      "loss": 4.7532,
      "mean_token_accuracy": 0.25249561965465545,
      "num_tokens": 106807880.0,
      "step": 46570
    },
    {
      "entropy": 5.520671224594116,
      "epoch": 4.474063400576369,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003157158664363648,
      "loss": 4.5824,
      "mean_token_accuracy": 0.2655043229460716,
      "num_tokens": 106818837.0,
      "step": 46575
    },
    {
      "entropy": 5.503313827514648,
      "epoch": 4.474543707973103,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031568214841045576,
      "loss": 4.6656,
      "mean_token_accuracy": 0.2606605261564255,
      "num_tokens": 106829606.0,
      "step": 46580
    },
    {
      "entropy": 5.445046138763428,
      "epoch": 4.475024015369836,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003156484294400281,
      "loss": 4.6243,
      "mean_token_accuracy": 0.2583214953541756,
      "num_tokens": 106840218.0,
      "step": 46585
    },
    {
      "entropy": 5.471058034896851,
      "epoch": 4.475504322766571,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003156147095258646,
      "loss": 4.6555,
      "mean_token_accuracy": 0.2633750453591347,
      "num_tokens": 106852020.0,
      "step": 46590
    },
    {
      "entropy": 5.502137231826782,
      "epoch": 4.475984630163304,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031558098866874814,
      "loss": 4.7047,
      "mean_token_accuracy": 0.25537863224744795,
      "num_tokens": 106863068.0,
      "step": 46595
    },
    {
      "entropy": 5.571085119247437,
      "epoch": 4.4764649375600385,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031554726686946154,
      "loss": 4.733,
      "mean_token_accuracy": 0.255980122089386,
      "num_tokens": 106874319.0,
      "step": 46600
    },
    {
      "entropy": 5.565481328964234,
      "epoch": 4.476945244956772,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000315513544128788,
      "loss": 4.6868,
      "mean_token_accuracy": 0.2583144187927246,
      "num_tokens": 106886332.0,
      "step": 46605
    },
    {
      "entropy": 5.519355154037475,
      "epoch": 4.477425552353506,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003154798204475101,
      "loss": 4.7418,
      "mean_token_accuracy": 0.2552322790026665,
      "num_tokens": 106898976.0,
      "step": 46610
    },
    {
      "entropy": 5.564067459106445,
      "epoch": 4.477905859750241,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003154460958264109,
      "loss": 4.6755,
      "mean_token_accuracy": 0.2591277003288269,
      "num_tokens": 106911078.0,
      "step": 46615
    },
    {
      "entropy": 5.580243444442749,
      "epoch": 4.478386167146974,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003154123702662737,
      "loss": 4.7305,
      "mean_token_accuracy": 0.2623673155903816,
      "num_tokens": 106922980.0,
      "step": 46620
    },
    {
      "entropy": 5.573634099960327,
      "epoch": 4.478866474543708,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003153786437678811,
      "loss": 4.7694,
      "mean_token_accuracy": 0.2491643488407135,
      "num_tokens": 106935361.0,
      "step": 46625
    },
    {
      "entropy": 5.477863931655884,
      "epoch": 4.479346781940442,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003153449163320163,
      "loss": 4.562,
      "mean_token_accuracy": 0.26534217447042463,
      "num_tokens": 106947104.0,
      "step": 46630
    },
    {
      "entropy": 5.558799791336059,
      "epoch": 4.479827089337176,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003153111879594623,
      "loss": 4.7547,
      "mean_token_accuracy": 0.24568957090377808,
      "num_tokens": 106958886.0,
      "step": 46635
    },
    {
      "entropy": 5.448854303359985,
      "epoch": 4.4803073967339095,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003152774586510023,
      "loss": 4.5685,
      "mean_token_accuracy": 0.2656298041343689,
      "num_tokens": 106970156.0,
      "step": 46640
    },
    {
      "entropy": 5.550650501251221,
      "epoch": 4.480787704130644,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003152437284074193,
      "loss": 4.7211,
      "mean_token_accuracy": 0.25305787324905393,
      "num_tokens": 106980426.0,
      "step": 46645
    },
    {
      "entropy": 5.456212091445923,
      "epoch": 4.481268011527377,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003152099972294964,
      "loss": 4.6222,
      "mean_token_accuracy": 0.2694146290421486,
      "num_tokens": 106991348.0,
      "step": 46650
    },
    {
      "entropy": 5.510574722290039,
      "epoch": 4.481748318924112,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003151762651180168,
      "loss": 4.6221,
      "mean_token_accuracy": 0.26680674105882646,
      "num_tokens": 107002414.0,
      "step": 46655
    },
    {
      "entropy": 5.480461692810058,
      "epoch": 4.482228626320845,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031514253207376355,
      "loss": 4.6379,
      "mean_token_accuracy": 0.2596712291240692,
      "num_tokens": 107014729.0,
      "step": 46660
    },
    {
      "entropy": 5.5183069705963135,
      "epoch": 4.482708933717579,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031510879809751996,
      "loss": 4.7132,
      "mean_token_accuracy": 0.2523802578449249,
      "num_tokens": 107026699.0,
      "step": 46665
    },
    {
      "entropy": 5.501795291900635,
      "epoch": 4.483189241114313,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031507506319006917,
      "loss": 4.6603,
      "mean_token_accuracy": 0.2534980311989784,
      "num_tokens": 107038474.0,
      "step": 46670
    },
    {
      "entropy": 5.543220520019531,
      "epoch": 4.483669548511047,
      "grad_norm": 1.375,
      "learning_rate": 0.00031504132735219435,
      "loss": 4.7361,
      "mean_token_accuracy": 0.25508278161287307,
      "num_tokens": 107049459.0,
      "step": 46675
    },
    {
      "entropy": 5.458370494842529,
      "epoch": 4.484149855907781,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031500759058467894,
      "loss": 4.6194,
      "mean_token_accuracy": 0.2659431308507919,
      "num_tokens": 107060162.0,
      "step": 46680
    },
    {
      "entropy": 5.533887767791748,
      "epoch": 4.484630163304515,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000314973852888306,
      "loss": 4.7236,
      "mean_token_accuracy": 0.25451084822416303,
      "num_tokens": 107072532.0,
      "step": 46685
    },
    {
      "entropy": 5.500760269165039,
      "epoch": 4.485110470701249,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003149401142638589,
      "loss": 4.6367,
      "mean_token_accuracy": 0.2543952360749245,
      "num_tokens": 107084070.0,
      "step": 46690
    },
    {
      "entropy": 5.544087553024292,
      "epoch": 4.485590778097983,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031490637471212087,
      "loss": 4.7339,
      "mean_token_accuracy": 0.25978267192840576,
      "num_tokens": 107096089.0,
      "step": 46695
    },
    {
      "entropy": 5.599828672409058,
      "epoch": 4.486071085494717,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031487263423387544,
      "loss": 4.7615,
      "mean_token_accuracy": 0.2472938284277916,
      "num_tokens": 107107596.0,
      "step": 46700
    },
    {
      "entropy": 5.563036298751831,
      "epoch": 4.48655139289145,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031483889282990567,
      "loss": 4.7089,
      "mean_token_accuracy": 0.25663903951644895,
      "num_tokens": 107119205.0,
      "step": 46705
    },
    {
      "entropy": 5.5355041980743405,
      "epoch": 4.487031700288185,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031480515050099527,
      "loss": 4.7176,
      "mean_token_accuracy": 0.24806324541568756,
      "num_tokens": 107131341.0,
      "step": 46710
    },
    {
      "entropy": 5.515993452072143,
      "epoch": 4.487512007684918,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003147714072479275,
      "loss": 4.6953,
      "mean_token_accuracy": 0.25100730955600736,
      "num_tokens": 107142846.0,
      "step": 46715
    },
    {
      "entropy": 5.578174114227295,
      "epoch": 4.4879923150816525,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003147376630714856,
      "loss": 4.818,
      "mean_token_accuracy": 0.24234989136457444,
      "num_tokens": 107154300.0,
      "step": 46720
    },
    {
      "entropy": 5.5929769515991214,
      "epoch": 4.488472622478386,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003147039179724533,
      "loss": 4.7388,
      "mean_token_accuracy": 0.2593110501766205,
      "num_tokens": 107166387.0,
      "step": 46725
    },
    {
      "entropy": 5.654472827911377,
      "epoch": 4.48895292987512,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031467017195161383,
      "loss": 4.8648,
      "mean_token_accuracy": 0.2452406257390976,
      "num_tokens": 107179848.0,
      "step": 46730
    },
    {
      "entropy": 5.539542961120605,
      "epoch": 4.489433237271854,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003146364250097508,
      "loss": 4.7355,
      "mean_token_accuracy": 0.26116210520267485,
      "num_tokens": 107192837.0,
      "step": 46735
    },
    {
      "entropy": 5.502597761154175,
      "epoch": 4.489913544668588,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003146026771476476,
      "loss": 4.6519,
      "mean_token_accuracy": 0.2530822977423668,
      "num_tokens": 107204055.0,
      "step": 46740
    },
    {
      "entropy": 5.5088848114013675,
      "epoch": 4.4903938520653215,
      "grad_norm": 1.5,
      "learning_rate": 0.00031456892836608793,
      "loss": 4.6838,
      "mean_token_accuracy": 0.25956064760684966,
      "num_tokens": 107216144.0,
      "step": 46745
    },
    {
      "entropy": 5.520947551727295,
      "epoch": 4.490874159462056,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031453517866585523,
      "loss": 4.6275,
      "mean_token_accuracy": 0.25629578679800036,
      "num_tokens": 107227970.0,
      "step": 46750
    },
    {
      "entropy": 5.482166385650634,
      "epoch": 4.491354466858789,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000314501428047733,
      "loss": 4.6879,
      "mean_token_accuracy": 0.25661240220069886,
      "num_tokens": 107237765.0,
      "step": 46755
    },
    {
      "entropy": 5.537572574615479,
      "epoch": 4.491834774255524,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031446767651250495,
      "loss": 4.7297,
      "mean_token_accuracy": 0.2527179032564163,
      "num_tokens": 107249298.0,
      "step": 46760
    },
    {
      "entropy": 5.649592542648316,
      "epoch": 4.492315081652258,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031443392406095474,
      "loss": 4.8153,
      "mean_token_accuracy": 0.24991151988506316,
      "num_tokens": 107260856.0,
      "step": 46765
    },
    {
      "entropy": 5.541652822494507,
      "epoch": 4.492795389048991,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003144001706938658,
      "loss": 4.7357,
      "mean_token_accuracy": 0.2541315987706184,
      "num_tokens": 107272027.0,
      "step": 46770
    },
    {
      "entropy": 5.475083303451538,
      "epoch": 4.493275696445726,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003143664164120219,
      "loss": 4.6692,
      "mean_token_accuracy": 0.2627368852496147,
      "num_tokens": 107284918.0,
      "step": 46775
    },
    {
      "entropy": 5.515578699111939,
      "epoch": 4.493756003842459,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031433266121620675,
      "loss": 4.6673,
      "mean_token_accuracy": 0.25907507389783857,
      "num_tokens": 107297270.0,
      "step": 46780
    },
    {
      "entropy": 5.581371116638183,
      "epoch": 4.494236311239193,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031429890510720394,
      "loss": 4.6926,
      "mean_token_accuracy": 0.2504153698682785,
      "num_tokens": 107308846.0,
      "step": 46785
    },
    {
      "entropy": 5.506519222259522,
      "epoch": 4.494716618635927,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003142651480857973,
      "loss": 4.6575,
      "mean_token_accuracy": 0.25633908063173294,
      "num_tokens": 107319958.0,
      "step": 46790
    },
    {
      "entropy": 5.465163612365723,
      "epoch": 4.495196926032661,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003142313901527705,
      "loss": 4.6549,
      "mean_token_accuracy": 0.25952501893043517,
      "num_tokens": 107329451.0,
      "step": 46795
    },
    {
      "entropy": 5.5339274406433105,
      "epoch": 4.495677233429395,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003141976313089073,
      "loss": 4.6538,
      "mean_token_accuracy": 0.2604288995265961,
      "num_tokens": 107340362.0,
      "step": 46800
    },
    {
      "entropy": 5.543996047973633,
      "epoch": 4.496157540826129,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003141638715549915,
      "loss": 4.7099,
      "mean_token_accuracy": 0.25271949768066404,
      "num_tokens": 107351765.0,
      "step": 46805
    },
    {
      "entropy": 5.539916324615478,
      "epoch": 4.496637848222862,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003141301108918069,
      "loss": 4.7102,
      "mean_token_accuracy": 0.24739460051059722,
      "num_tokens": 107362506.0,
      "step": 46810
    },
    {
      "entropy": 5.537537908554077,
      "epoch": 4.497118155619597,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003140963493201373,
      "loss": 4.7381,
      "mean_token_accuracy": 0.2553750991821289,
      "num_tokens": 107374671.0,
      "step": 46815
    },
    {
      "entropy": 5.424888181686401,
      "epoch": 4.49759846301633,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003140625868407666,
      "loss": 4.624,
      "mean_token_accuracy": 0.2581144079566002,
      "num_tokens": 107387183.0,
      "step": 46820
    },
    {
      "entropy": 5.540734195709229,
      "epoch": 4.4980787704130645,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003140288234544786,
      "loss": 4.7525,
      "mean_token_accuracy": 0.256303046643734,
      "num_tokens": 107399194.0,
      "step": 46825
    },
    {
      "entropy": 5.588509273529053,
      "epoch": 4.498559077809798,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031399505916205716,
      "loss": 4.7915,
      "mean_token_accuracy": 0.24412254393100738,
      "num_tokens": 107411359.0,
      "step": 46830
    },
    {
      "entropy": 5.463913059234619,
      "epoch": 4.499039385206532,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031396129396428637,
      "loss": 4.6096,
      "mean_token_accuracy": 0.26695099622011187,
      "num_tokens": 107422175.0,
      "step": 46835
    },
    {
      "entropy": 5.523390102386474,
      "epoch": 4.499519692603266,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003139275278619499,
      "loss": 4.697,
      "mean_token_accuracy": 0.2539649188518524,
      "num_tokens": 107433774.0,
      "step": 46840
    },
    {
      "entropy": 5.472951173782349,
      "epoch": 4.5,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031389376085583183,
      "loss": 4.6102,
      "mean_token_accuracy": 0.2690969958901405,
      "num_tokens": 107444913.0,
      "step": 46845
    },
    {
      "entropy": 5.429235172271729,
      "epoch": 4.500480307396733,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031385999294671627,
      "loss": 4.5925,
      "mean_token_accuracy": 0.26877438873052595,
      "num_tokens": 107456049.0,
      "step": 46850
    },
    {
      "entropy": 5.500035190582276,
      "epoch": 4.500960614793468,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031382622413538686,
      "loss": 4.6038,
      "mean_token_accuracy": 0.2570596277713776,
      "num_tokens": 107466889.0,
      "step": 46855
    },
    {
      "entropy": 5.430062007904053,
      "epoch": 4.501440922190202,
      "grad_norm": 1.375,
      "learning_rate": 0.0003137924544226279,
      "loss": 4.6107,
      "mean_token_accuracy": 0.2587294191122055,
      "num_tokens": 107477470.0,
      "step": 46860
    },
    {
      "entropy": 5.609763050079346,
      "epoch": 4.5019212295869355,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003137586838092234,
      "loss": 4.7664,
      "mean_token_accuracy": 0.24959103763103485,
      "num_tokens": 107488277.0,
      "step": 46865
    },
    {
      "entropy": 5.503981685638427,
      "epoch": 4.50240153698367,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003137249122959573,
      "loss": 4.6268,
      "mean_token_accuracy": 0.2595207914710045,
      "num_tokens": 107499986.0,
      "step": 46870
    },
    {
      "entropy": 5.5298765182495115,
      "epoch": 4.502881844380403,
      "grad_norm": 1.25,
      "learning_rate": 0.0003136911398836137,
      "loss": 4.7357,
      "mean_token_accuracy": 0.2539049655199051,
      "num_tokens": 107511645.0,
      "step": 46875
    },
    {
      "entropy": 5.441519641876221,
      "epoch": 4.503362151777138,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003136573665729768,
      "loss": 4.5753,
      "mean_token_accuracy": 0.26811682879924775,
      "num_tokens": 107523128.0,
      "step": 46880
    },
    {
      "entropy": 5.57589898109436,
      "epoch": 4.503842459173871,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031362359236483054,
      "loss": 4.7076,
      "mean_token_accuracy": 0.2524935156106949,
      "num_tokens": 107533403.0,
      "step": 46885
    },
    {
      "entropy": 5.541082382202148,
      "epoch": 4.504322766570605,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003135898172599592,
      "loss": 4.7741,
      "mean_token_accuracy": 0.24784417301416398,
      "num_tokens": 107545351.0,
      "step": 46890
    },
    {
      "entropy": 5.50999436378479,
      "epoch": 4.504803073967339,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003135560412591469,
      "loss": 4.7223,
      "mean_token_accuracy": 0.256531722843647,
      "num_tokens": 107557373.0,
      "step": 46895
    },
    {
      "entropy": 5.546692609786987,
      "epoch": 4.505283381364073,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031352226436317783,
      "loss": 4.7329,
      "mean_token_accuracy": 0.25827307999134064,
      "num_tokens": 107567797.0,
      "step": 46900
    },
    {
      "entropy": 5.523524045944214,
      "epoch": 4.505763688760807,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003134884865728361,
      "loss": 4.7176,
      "mean_token_accuracy": 0.2600399672985077,
      "num_tokens": 107580180.0,
      "step": 46905
    },
    {
      "entropy": 5.541374444961548,
      "epoch": 4.506243996157541,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003134547078889061,
      "loss": 4.6616,
      "mean_token_accuracy": 0.257000757753849,
      "num_tokens": 107591774.0,
      "step": 46910
    },
    {
      "entropy": 5.524510478973388,
      "epoch": 4.506724303554275,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003134209283121719,
      "loss": 4.6602,
      "mean_token_accuracy": 0.25777578800916673,
      "num_tokens": 107602666.0,
      "step": 46915
    },
    {
      "entropy": 5.516753005981445,
      "epoch": 4.507204610951009,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031338714784341783,
      "loss": 4.7093,
      "mean_token_accuracy": 0.25818620473146436,
      "num_tokens": 107614178.0,
      "step": 46920
    },
    {
      "entropy": 5.532459402084351,
      "epoch": 4.507684918347742,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003133533664834282,
      "loss": 4.71,
      "mean_token_accuracy": 0.257749380171299,
      "num_tokens": 107625317.0,
      "step": 46925
    },
    {
      "entropy": 5.510802602767944,
      "epoch": 4.508165225744476,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031331958423298724,
      "loss": 4.6489,
      "mean_token_accuracy": 0.26544099003076554,
      "num_tokens": 107635541.0,
      "step": 46930
    },
    {
      "entropy": 5.480500316619873,
      "epoch": 4.508645533141211,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003132858010928794,
      "loss": 4.6733,
      "mean_token_accuracy": 0.25392867922782897,
      "num_tokens": 107646236.0,
      "step": 46935
    },
    {
      "entropy": 5.574073028564453,
      "epoch": 4.509125840537944,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031325201706388885,
      "loss": 4.6965,
      "mean_token_accuracy": 0.2629899069666862,
      "num_tokens": 107656488.0,
      "step": 46940
    },
    {
      "entropy": 5.563017082214356,
      "epoch": 4.5096061479346785,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031321823214680014,
      "loss": 4.7615,
      "mean_token_accuracy": 0.24775015115737914,
      "num_tokens": 107668909.0,
      "step": 46945
    },
    {
      "entropy": 5.65773754119873,
      "epoch": 4.510086455331412,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031318444634239744,
      "loss": 4.8011,
      "mean_token_accuracy": 0.24959700852632521,
      "num_tokens": 107680777.0,
      "step": 46950
    },
    {
      "entropy": 5.488824892044067,
      "epoch": 4.510566762728146,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003131506596514654,
      "loss": 4.6622,
      "mean_token_accuracy": 0.2617749959230423,
      "num_tokens": 107692004.0,
      "step": 46955
    },
    {
      "entropy": 5.475271797180175,
      "epoch": 4.51104707012488,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003131168720747883,
      "loss": 4.611,
      "mean_token_accuracy": 0.264707912504673,
      "num_tokens": 107703743.0,
      "step": 46960
    },
    {
      "entropy": 5.567829847335815,
      "epoch": 4.511527377521614,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003130830836131505,
      "loss": 4.7646,
      "mean_token_accuracy": 0.2438416451215744,
      "num_tokens": 107715703.0,
      "step": 46965
    },
    {
      "entropy": 5.544291591644287,
      "epoch": 4.5120076849183475,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003130492942673367,
      "loss": 4.6911,
      "mean_token_accuracy": 0.2534520924091339,
      "num_tokens": 107726548.0,
      "step": 46970
    },
    {
      "entropy": 5.536421012878418,
      "epoch": 4.512487992315082,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031301550403813126,
      "loss": 4.7212,
      "mean_token_accuracy": 0.25991000682115556,
      "num_tokens": 107737863.0,
      "step": 46975
    },
    {
      "entropy": 5.554172658920288,
      "epoch": 4.512968299711815,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003129817129263188,
      "loss": 4.7406,
      "mean_token_accuracy": 0.25616328567266466,
      "num_tokens": 107750049.0,
      "step": 46980
    },
    {
      "entropy": 5.538708114624024,
      "epoch": 4.51344860710855,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003129479209326835,
      "loss": 4.6736,
      "mean_token_accuracy": 0.2595045730471611,
      "num_tokens": 107761970.0,
      "step": 46985
    },
    {
      "entropy": 5.579148674011231,
      "epoch": 4.513928914505283,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031291412805801034,
      "loss": 4.7587,
      "mean_token_accuracy": 0.2541925460100174,
      "num_tokens": 107773391.0,
      "step": 46990
    },
    {
      "entropy": 5.500624322891236,
      "epoch": 4.514409221902017,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003128803343030836,
      "loss": 4.6873,
      "mean_token_accuracy": 0.25110419690608976,
      "num_tokens": 107785077.0,
      "step": 46995
    },
    {
      "entropy": 5.524694967269897,
      "epoch": 4.514889529298751,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031284653966868813,
      "loss": 4.7141,
      "mean_token_accuracy": 0.25356399416923525,
      "num_tokens": 107795969.0,
      "step": 47000
    },
    {
      "entropy": 5.50043625831604,
      "epoch": 4.515369836695485,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031281274415560827,
      "loss": 4.5698,
      "mean_token_accuracy": 0.27407377064228056,
      "num_tokens": 107806547.0,
      "step": 47005
    },
    {
      "entropy": 5.581594467163086,
      "epoch": 4.515850144092219,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003127789477646288,
      "loss": 4.7586,
      "mean_token_accuracy": 0.2484728515148163,
      "num_tokens": 107818131.0,
      "step": 47010
    },
    {
      "entropy": 5.519807624816894,
      "epoch": 4.516330451488953,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003127451504965344,
      "loss": 4.6876,
      "mean_token_accuracy": 0.25792024433612826,
      "num_tokens": 107830346.0,
      "step": 47015
    },
    {
      "entropy": 5.497030115127563,
      "epoch": 4.516810758885687,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003127113523521096,
      "loss": 4.7122,
      "mean_token_accuracy": 0.24859710782766342,
      "num_tokens": 107841094.0,
      "step": 47020
    },
    {
      "entropy": 5.567450428009034,
      "epoch": 4.517291066282421,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003126775533321392,
      "loss": 4.7306,
      "mean_token_accuracy": 0.25045571476221085,
      "num_tokens": 107852951.0,
      "step": 47025
    },
    {
      "entropy": 5.507938861846924,
      "epoch": 4.517771373679155,
      "grad_norm": 1.1875,
      "learning_rate": 0.000312643753437408,
      "loss": 4.6337,
      "mean_token_accuracy": 0.25879035890102386,
      "num_tokens": 107864008.0,
      "step": 47030
    },
    {
      "entropy": 5.416028451919556,
      "epoch": 4.518251681075888,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003126099526687005,
      "loss": 4.5778,
      "mean_token_accuracy": 0.26751805394887923,
      "num_tokens": 107874522.0,
      "step": 47035
    },
    {
      "entropy": 5.478145599365234,
      "epoch": 4.518731988472623,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031257615102680154,
      "loss": 4.6894,
      "mean_token_accuracy": 0.2575576052069664,
      "num_tokens": 107886832.0,
      "step": 47040
    },
    {
      "entropy": 5.562099409103394,
      "epoch": 4.519212295869356,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031254234851249606,
      "loss": 4.7326,
      "mean_token_accuracy": 0.2551735818386078,
      "num_tokens": 107897361.0,
      "step": 47045
    },
    {
      "entropy": 5.540876579284668,
      "epoch": 4.5196926032660905,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031250854512656864,
      "loss": 4.6531,
      "mean_token_accuracy": 0.2641635939478874,
      "num_tokens": 107908097.0,
      "step": 47050
    },
    {
      "entropy": 5.563678789138794,
      "epoch": 4.520172910662824,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003124747408698042,
      "loss": 4.7148,
      "mean_token_accuracy": 0.2564733847975731,
      "num_tokens": 107919317.0,
      "step": 47055
    },
    {
      "entropy": 5.6344053745269775,
      "epoch": 4.520653218059558,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003124409357429876,
      "loss": 4.7797,
      "mean_token_accuracy": 0.25252267718315125,
      "num_tokens": 107930459.0,
      "step": 47060
    },
    {
      "entropy": 5.57196536064148,
      "epoch": 4.521133525456292,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031240712974690355,
      "loss": 4.7481,
      "mean_token_accuracy": 0.2527761161327362,
      "num_tokens": 107942292.0,
      "step": 47065
    },
    {
      "entropy": 5.526624298095703,
      "epoch": 4.521613832853026,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031237332288233704,
      "loss": 4.6809,
      "mean_token_accuracy": 0.24988166093826295,
      "num_tokens": 107954348.0,
      "step": 47070
    },
    {
      "entropy": 5.493578720092773,
      "epoch": 4.522094140249759,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031233951515007306,
      "loss": 4.7141,
      "mean_token_accuracy": 0.2495623916387558,
      "num_tokens": 107966068.0,
      "step": 47075
    },
    {
      "entropy": 5.530089616775513,
      "epoch": 4.522574447646494,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003123057065508964,
      "loss": 4.6846,
      "mean_token_accuracy": 0.2573012337088585,
      "num_tokens": 107977001.0,
      "step": 47080
    },
    {
      "entropy": 5.517795562744141,
      "epoch": 4.523054755043228,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003122718970855918,
      "loss": 4.6317,
      "mean_token_accuracy": 0.25947786718606947,
      "num_tokens": 107989016.0,
      "step": 47085
    },
    {
      "entropy": 5.5700945377349855,
      "epoch": 4.5235350624399615,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003122380867549447,
      "loss": 4.7309,
      "mean_token_accuracy": 0.25184461772441863,
      "num_tokens": 107998828.0,
      "step": 47090
    },
    {
      "entropy": 5.500339698791504,
      "epoch": 4.524015369836696,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031220427555973964,
      "loss": 4.6744,
      "mean_token_accuracy": 0.2567019581794739,
      "num_tokens": 108010038.0,
      "step": 47095
    },
    {
      "entropy": 5.503466606140137,
      "epoch": 4.524495677233429,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031217046350076177,
      "loss": 4.6826,
      "mean_token_accuracy": 0.2593303948640823,
      "num_tokens": 108022036.0,
      "step": 47100
    },
    {
      "entropy": 5.515724802017212,
      "epoch": 4.524975984630164,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031213665057879615,
      "loss": 4.6844,
      "mean_token_accuracy": 0.2531577110290527,
      "num_tokens": 108033956.0,
      "step": 47105
    },
    {
      "entropy": 5.5944336414337155,
      "epoch": 4.525456292026897,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003121028367946278,
      "loss": 4.7039,
      "mean_token_accuracy": 0.2578944280743599,
      "num_tokens": 108045706.0,
      "step": 47110
    },
    {
      "entropy": 5.530661773681641,
      "epoch": 4.525936599423631,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003120690221490417,
      "loss": 4.6934,
      "mean_token_accuracy": 0.2511250048875809,
      "num_tokens": 108057780.0,
      "step": 47115
    },
    {
      "entropy": 5.53219084739685,
      "epoch": 4.526416906820365,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003120352066428231,
      "loss": 4.6676,
      "mean_token_accuracy": 0.2533661678433418,
      "num_tokens": 108068782.0,
      "step": 47120
    },
    {
      "entropy": 5.445348787307739,
      "epoch": 4.526897214217099,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031200139027675683,
      "loss": 4.5756,
      "mean_token_accuracy": 0.25743907541036604,
      "num_tokens": 108080639.0,
      "step": 47125
    },
    {
      "entropy": 5.509738636016846,
      "epoch": 4.527377521613833,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031196757305162814,
      "loss": 4.6725,
      "mean_token_accuracy": 0.2572475135326385,
      "num_tokens": 108092513.0,
      "step": 47130
    },
    {
      "entropy": 5.482576608657837,
      "epoch": 4.527857829010567,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003119337549682223,
      "loss": 4.6002,
      "mean_token_accuracy": 0.2613320767879486,
      "num_tokens": 108104590.0,
      "step": 47135
    },
    {
      "entropy": 5.4643501281738285,
      "epoch": 4.5283381364073,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003118999360273243,
      "loss": 4.5876,
      "mean_token_accuracy": 0.2634564384818077,
      "num_tokens": 108116781.0,
      "step": 47140
    },
    {
      "entropy": 5.442094469070435,
      "epoch": 4.528818443804035,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031186611622971926,
      "loss": 4.6422,
      "mean_token_accuracy": 0.2603076308965683,
      "num_tokens": 108127823.0,
      "step": 47145
    },
    {
      "entropy": 5.489366865158081,
      "epoch": 4.529298751200768,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031183229557619265,
      "loss": 4.6295,
      "mean_token_accuracy": 0.25503932386636735,
      "num_tokens": 108138546.0,
      "step": 47150
    },
    {
      "entropy": 5.436348485946655,
      "epoch": 4.529779058597502,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003117984740675294,
      "loss": 4.5523,
      "mean_token_accuracy": 0.26178523898124695,
      "num_tokens": 108148962.0,
      "step": 47155
    },
    {
      "entropy": 5.529645729064941,
      "epoch": 4.530259365994237,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003117646517045148,
      "loss": 4.6711,
      "mean_token_accuracy": 0.26660366356372833,
      "num_tokens": 108160997.0,
      "step": 47160
    },
    {
      "entropy": 5.500250244140625,
      "epoch": 4.53073967339097,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003117308284879342,
      "loss": 4.6837,
      "mean_token_accuracy": 0.25615732967853544,
      "num_tokens": 108172503.0,
      "step": 47165
    },
    {
      "entropy": 5.47862491607666,
      "epoch": 4.5312199807877045,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003116970044185728,
      "loss": 4.6601,
      "mean_token_accuracy": 0.25659599155187607,
      "num_tokens": 108184515.0,
      "step": 47170
    },
    {
      "entropy": 5.539178943634033,
      "epoch": 4.531700288184438,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000311663179497216,
      "loss": 4.6311,
      "mean_token_accuracy": 0.26309180706739427,
      "num_tokens": 108196003.0,
      "step": 47175
    },
    {
      "entropy": 5.616916275024414,
      "epoch": 4.532180595581172,
      "grad_norm": 1.203125,
      "learning_rate": 0.000311629353724649,
      "loss": 4.7552,
      "mean_token_accuracy": 0.25294009745121004,
      "num_tokens": 108207467.0,
      "step": 47180
    },
    {
      "entropy": 5.539917039871216,
      "epoch": 4.532660902977906,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031159552710165715,
      "loss": 4.7476,
      "mean_token_accuracy": 0.2567729026079178,
      "num_tokens": 108218608.0,
      "step": 47185
    },
    {
      "entropy": 5.529085445404053,
      "epoch": 4.53314121037464,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031156169962902585,
      "loss": 4.7519,
      "mean_token_accuracy": 0.2557568535208702,
      "num_tokens": 108230684.0,
      "step": 47190
    },
    {
      "entropy": 5.518197345733642,
      "epoch": 4.5336215177713735,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003115278713075405,
      "loss": 4.7306,
      "mean_token_accuracy": 0.25719657987356187,
      "num_tokens": 108242127.0,
      "step": 47195
    },
    {
      "entropy": 5.439710330963135,
      "epoch": 4.534101825168108,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003114940421379864,
      "loss": 4.6138,
      "mean_token_accuracy": 0.26327614635229113,
      "num_tokens": 108253491.0,
      "step": 47200
    },
    {
      "entropy": 5.502122974395752,
      "epoch": 4.534582132564841,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000311460212121149,
      "loss": 4.6823,
      "mean_token_accuracy": 0.25853836089372634,
      "num_tokens": 108267164.0,
      "step": 47205
    },
    {
      "entropy": 5.50104775428772,
      "epoch": 4.5350624399615755,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031142638125781374,
      "loss": 4.633,
      "mean_token_accuracy": 0.26403038054704664,
      "num_tokens": 108278506.0,
      "step": 47210
    },
    {
      "entropy": 5.5761713027954105,
      "epoch": 4.535542747358309,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031139254954876605,
      "loss": 4.7385,
      "mean_token_accuracy": 0.25017344802618025,
      "num_tokens": 108288766.0,
      "step": 47215
    },
    {
      "entropy": 5.513028001785278,
      "epoch": 4.536023054755043,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003113587169947914,
      "loss": 4.6871,
      "mean_token_accuracy": 0.2512996941804886,
      "num_tokens": 108300632.0,
      "step": 47220
    },
    {
      "entropy": 5.557114791870117,
      "epoch": 4.536503362151777,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031132488359667546,
      "loss": 4.7065,
      "mean_token_accuracy": 0.2513019621372223,
      "num_tokens": 108312168.0,
      "step": 47225
    },
    {
      "entropy": 5.528599596023559,
      "epoch": 4.536983669548511,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003112910493552034,
      "loss": 4.6724,
      "mean_token_accuracy": 0.25993437319993973,
      "num_tokens": 108323214.0,
      "step": 47230
    },
    {
      "entropy": 5.471493148803711,
      "epoch": 4.537463976945245,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031125721427116104,
      "loss": 4.6169,
      "mean_token_accuracy": 0.2625260591506958,
      "num_tokens": 108334329.0,
      "step": 47235
    },
    {
      "entropy": 5.51688175201416,
      "epoch": 4.537944284341979,
      "grad_norm": 1.375,
      "learning_rate": 0.00031122337834533375,
      "loss": 4.7193,
      "mean_token_accuracy": 0.2543919235467911,
      "num_tokens": 108345402.0,
      "step": 47240
    },
    {
      "entropy": 5.504413795471192,
      "epoch": 4.538424591738713,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003111895415785073,
      "loss": 4.6461,
      "mean_token_accuracy": 0.25171881467103957,
      "num_tokens": 108356807.0,
      "step": 47245
    },
    {
      "entropy": 5.625108242034912,
      "epoch": 4.538904899135447,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003111557039714671,
      "loss": 4.8301,
      "mean_token_accuracy": 0.24416645467281342,
      "num_tokens": 108367804.0,
      "step": 47250
    },
    {
      "entropy": 5.5224738121032715,
      "epoch": 4.539385206532181,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031112186552499876,
      "loss": 4.6447,
      "mean_token_accuracy": 0.2536953791975975,
      "num_tokens": 108380201.0,
      "step": 47255
    },
    {
      "entropy": 5.435463857650757,
      "epoch": 4.539865513928914,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000311088026239888,
      "loss": 4.5132,
      "mean_token_accuracy": 0.2733698830008507,
      "num_tokens": 108391767.0,
      "step": 47260
    },
    {
      "entropy": 5.515857267379761,
      "epoch": 4.540345821325649,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031105418611692044,
      "loss": 4.6703,
      "mean_token_accuracy": 0.2533727988600731,
      "num_tokens": 108402532.0,
      "step": 47265
    },
    {
      "entropy": 5.5863312721252445,
      "epoch": 4.540826128722382,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003110203451568817,
      "loss": 4.7411,
      "mean_token_accuracy": 0.2495850533246994,
      "num_tokens": 108413136.0,
      "step": 47270
    },
    {
      "entropy": 5.494755744934082,
      "epoch": 4.541306436119116,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003109865033605575,
      "loss": 4.6143,
      "mean_token_accuracy": 0.2649141579866409,
      "num_tokens": 108425101.0,
      "step": 47275
    },
    {
      "entropy": 5.479862260818481,
      "epoch": 4.54178674351585,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003109526607287336,
      "loss": 4.6093,
      "mean_token_accuracy": 0.26576751470565796,
      "num_tokens": 108436258.0,
      "step": 47280
    },
    {
      "entropy": 5.50996880531311,
      "epoch": 4.542267050912584,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031091881726219567,
      "loss": 4.6533,
      "mean_token_accuracy": 0.2600066542625427,
      "num_tokens": 108447697.0,
      "step": 47285
    },
    {
      "entropy": 5.4908812046051025,
      "epoch": 4.542747358309318,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003108849729617294,
      "loss": 4.6655,
      "mean_token_accuracy": 0.2571572855114937,
      "num_tokens": 108459414.0,
      "step": 47290
    },
    {
      "entropy": 5.561719989776611,
      "epoch": 4.543227665706052,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031085112782812066,
      "loss": 4.7576,
      "mean_token_accuracy": 0.24506974816322327,
      "num_tokens": 108469182.0,
      "step": 47295
    },
    {
      "entropy": 5.505053758621216,
      "epoch": 4.543707973102785,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003108172818621552,
      "loss": 4.7085,
      "mean_token_accuracy": 0.25480308681726455,
      "num_tokens": 108481824.0,
      "step": 47300
    },
    {
      "entropy": 5.493735408782959,
      "epoch": 4.54418828049952,
      "grad_norm": 1.375,
      "learning_rate": 0.0003107834350646188,
      "loss": 4.6314,
      "mean_token_accuracy": 0.26502475887537,
      "num_tokens": 108491954.0,
      "step": 47305
    },
    {
      "entropy": 5.520175743103027,
      "epoch": 4.544668587896254,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003107495874362973,
      "loss": 4.7076,
      "mean_token_accuracy": 0.2511263877153397,
      "num_tokens": 108502828.0,
      "step": 47310
    },
    {
      "entropy": 5.482628440856933,
      "epoch": 4.5451488952929875,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003107157389779765,
      "loss": 4.6594,
      "mean_token_accuracy": 0.26029803454875944,
      "num_tokens": 108513781.0,
      "step": 47315
    },
    {
      "entropy": 5.516935777664185,
      "epoch": 4.545629202689721,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003106818896904423,
      "loss": 4.6369,
      "mean_token_accuracy": 0.25906695425510406,
      "num_tokens": 108524315.0,
      "step": 47320
    },
    {
      "entropy": 5.480649042129516,
      "epoch": 4.546109510086455,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031064803957448064,
      "loss": 4.6614,
      "mean_token_accuracy": 0.2576272666454315,
      "num_tokens": 108534972.0,
      "step": 47325
    },
    {
      "entropy": 5.6430878162384035,
      "epoch": 4.54658981748319,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003106141886308773,
      "loss": 4.8082,
      "mean_token_accuracy": 0.2466110870242119,
      "num_tokens": 108545571.0,
      "step": 47330
    },
    {
      "entropy": 5.5534882068634035,
      "epoch": 4.547070124879923,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003105803368604183,
      "loss": 4.7538,
      "mean_token_accuracy": 0.24767766743898392,
      "num_tokens": 108557643.0,
      "step": 47335
    },
    {
      "entropy": 5.543441724777222,
      "epoch": 4.547550432276657,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031054648426388944,
      "loss": 4.6983,
      "mean_token_accuracy": 0.24857674539089203,
      "num_tokens": 108569872.0,
      "step": 47340
    },
    {
      "entropy": 5.56609001159668,
      "epoch": 4.548030739673391,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031051263084207686,
      "loss": 4.6781,
      "mean_token_accuracy": 0.2572523608803749,
      "num_tokens": 108583387.0,
      "step": 47345
    },
    {
      "entropy": 5.438651847839355,
      "epoch": 4.548511047070125,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031047877659576636,
      "loss": 4.6186,
      "mean_token_accuracy": 0.2612829148769379,
      "num_tokens": 108595474.0,
      "step": 47350
    },
    {
      "entropy": 5.46137375831604,
      "epoch": 4.5489913544668585,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031044492152574406,
      "loss": 4.6054,
      "mean_token_accuracy": 0.26428184360265733,
      "num_tokens": 108606386.0,
      "step": 47355
    },
    {
      "entropy": 5.531386041641236,
      "epoch": 4.549471661863593,
      "grad_norm": 1.25,
      "learning_rate": 0.000310411065632796,
      "loss": 4.6864,
      "mean_token_accuracy": 0.25351606905460355,
      "num_tokens": 108617492.0,
      "step": 47360
    },
    {
      "entropy": 5.474215936660767,
      "epoch": 4.549951969260326,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031037720891770803,
      "loss": 4.6171,
      "mean_token_accuracy": 0.2628219500184059,
      "num_tokens": 108628956.0,
      "step": 47365
    },
    {
      "entropy": 5.439387702941895,
      "epoch": 4.550432276657061,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003103433513812663,
      "loss": 4.6019,
      "mean_token_accuracy": 0.2592271685600281,
      "num_tokens": 108639238.0,
      "step": 47370
    },
    {
      "entropy": 5.499400424957275,
      "epoch": 4.550912584053794,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031030949302425704,
      "loss": 4.6944,
      "mean_token_accuracy": 0.2513200372457504,
      "num_tokens": 108650323.0,
      "step": 47375
    },
    {
      "entropy": 5.501909351348877,
      "epoch": 4.551392891450528,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003102756338474661,
      "loss": 4.6045,
      "mean_token_accuracy": 0.2645012751221657,
      "num_tokens": 108662231.0,
      "step": 47380
    },
    {
      "entropy": 5.497310066223145,
      "epoch": 4.551873198847263,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003102417738516797,
      "loss": 4.6478,
      "mean_token_accuracy": 0.2586538061499596,
      "num_tokens": 108672571.0,
      "step": 47385
    },
    {
      "entropy": 5.51612811088562,
      "epoch": 4.552353506243996,
      "grad_norm": 1.515625,
      "learning_rate": 0.000310207913037684,
      "loss": 4.7419,
      "mean_token_accuracy": 0.24826167672872543,
      "num_tokens": 108683631.0,
      "step": 47390
    },
    {
      "entropy": 5.530269765853882,
      "epoch": 4.55283381364073,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031017405140626507,
      "loss": 4.7157,
      "mean_token_accuracy": 0.25372287780046465,
      "num_tokens": 108693986.0,
      "step": 47395
    },
    {
      "entropy": 5.605749845504761,
      "epoch": 4.553314121037464,
      "grad_norm": 1.25,
      "learning_rate": 0.00031014018895820903,
      "loss": 4.7444,
      "mean_token_accuracy": 0.24835253208875657,
      "num_tokens": 108705987.0,
      "step": 47400
    },
    {
      "entropy": 5.553306579589844,
      "epoch": 4.553794428434198,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003101063256943023,
      "loss": 4.7296,
      "mean_token_accuracy": 0.24900679737329484,
      "num_tokens": 108717154.0,
      "step": 47405
    },
    {
      "entropy": 5.550335884094238,
      "epoch": 4.554274735830932,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031007246161533086,
      "loss": 4.734,
      "mean_token_accuracy": 0.26352435499429705,
      "num_tokens": 108728225.0,
      "step": 47410
    },
    {
      "entropy": 5.505600929260254,
      "epoch": 4.554755043227666,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000310038596722081,
      "loss": 4.6746,
      "mean_token_accuracy": 0.26056499779224396,
      "num_tokens": 108739616.0,
      "step": 47415
    },
    {
      "entropy": 5.547707176208496,
      "epoch": 4.555235350624399,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000310004731015339,
      "loss": 4.6909,
      "mean_token_accuracy": 0.25535380095243454,
      "num_tokens": 108751409.0,
      "step": 47420
    },
    {
      "entropy": 5.5668360710144045,
      "epoch": 4.555715658021134,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00030997086449589107,
      "loss": 4.7013,
      "mean_token_accuracy": 0.2497764840722084,
      "num_tokens": 108762260.0,
      "step": 47425
    },
    {
      "entropy": 5.555566072463989,
      "epoch": 4.556195965417867,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030993699716452345,
      "loss": 4.7118,
      "mean_token_accuracy": 0.26107655465602875,
      "num_tokens": 108774283.0,
      "step": 47430
    },
    {
      "entropy": 5.490749263763428,
      "epoch": 4.5566762728146015,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003099031290220226,
      "loss": 4.6553,
      "mean_token_accuracy": 0.26220354735851287,
      "num_tokens": 108786081.0,
      "step": 47435
    },
    {
      "entropy": 5.518607759475708,
      "epoch": 4.557156580211335,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030986926006917463,
      "loss": 4.7169,
      "mean_token_accuracy": 0.2600632295012474,
      "num_tokens": 108798409.0,
      "step": 47440
    },
    {
      "entropy": 5.577763891220092,
      "epoch": 4.557636887608069,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003098353903067661,
      "loss": 4.6836,
      "mean_token_accuracy": 0.24651682376861572,
      "num_tokens": 108808807.0,
      "step": 47445
    },
    {
      "entropy": 5.543392086029053,
      "epoch": 4.558117195004803,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003098015197355832,
      "loss": 4.6779,
      "mean_token_accuracy": 0.2527883604168892,
      "num_tokens": 108820308.0,
      "step": 47450
    },
    {
      "entropy": 5.488467121124268,
      "epoch": 4.558597502401537,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003097676483564123,
      "loss": 4.6237,
      "mean_token_accuracy": 0.25944290310144424,
      "num_tokens": 108832242.0,
      "step": 47455
    },
    {
      "entropy": 5.545747232437134,
      "epoch": 4.559077809798271,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003097337761700399,
      "loss": 4.7141,
      "mean_token_accuracy": 0.2557047441601753,
      "num_tokens": 108843044.0,
      "step": 47460
    },
    {
      "entropy": 5.565375900268554,
      "epoch": 4.559558117195005,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003096999031772523,
      "loss": 4.775,
      "mean_token_accuracy": 0.24516549855470657,
      "num_tokens": 108855245.0,
      "step": 47465
    },
    {
      "entropy": 5.568190193176269,
      "epoch": 4.560038424591738,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030966602937883606,
      "loss": 4.7095,
      "mean_token_accuracy": 0.2530239954590797,
      "num_tokens": 108866991.0,
      "step": 47470
    },
    {
      "entropy": 5.4862922668457035,
      "epoch": 4.560518731988473,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003096321547755775,
      "loss": 4.6174,
      "mean_token_accuracy": 0.2606368660926819,
      "num_tokens": 108878341.0,
      "step": 47475
    },
    {
      "entropy": 5.551780128479004,
      "epoch": 4.560999039385207,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003095982793682632,
      "loss": 4.6866,
      "mean_token_accuracy": 0.25960084348917006,
      "num_tokens": 108889595.0,
      "step": 47480
    },
    {
      "entropy": 5.448693370819091,
      "epoch": 4.56147934678194,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003095644031576795,
      "loss": 4.6143,
      "mean_token_accuracy": 0.2589838609099388,
      "num_tokens": 108901336.0,
      "step": 47485
    },
    {
      "entropy": 5.526521348953247,
      "epoch": 4.561959654178675,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030953052614461304,
      "loss": 4.6788,
      "mean_token_accuracy": 0.2531534552574158,
      "num_tokens": 108912163.0,
      "step": 47490
    },
    {
      "entropy": 5.518086767196655,
      "epoch": 4.562439961575408,
      "grad_norm": 1.25,
      "learning_rate": 0.0003094966483298503,
      "loss": 4.6895,
      "mean_token_accuracy": 0.25721909999847414,
      "num_tokens": 108923876.0,
      "step": 47495
    },
    {
      "entropy": 5.540822505950928,
      "epoch": 4.562920268972142,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030946276971417785,
      "loss": 4.6883,
      "mean_token_accuracy": 0.259449265897274,
      "num_tokens": 108935529.0,
      "step": 47500
    },
    {
      "entropy": 5.613669109344483,
      "epoch": 4.563400576368876,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030942889029838216,
      "loss": 4.8206,
      "mean_token_accuracy": 0.24694718718528746,
      "num_tokens": 108947524.0,
      "step": 47505
    },
    {
      "entropy": 5.5590253353118895,
      "epoch": 4.56388088376561,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030939501008324995,
      "loss": 4.6992,
      "mean_token_accuracy": 0.2535183012485504,
      "num_tokens": 108958890.0,
      "step": 47510
    },
    {
      "entropy": 5.497640228271484,
      "epoch": 4.564361191162344,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030936112906956766,
      "loss": 4.669,
      "mean_token_accuracy": 0.25495033860206606,
      "num_tokens": 108971483.0,
      "step": 47515
    },
    {
      "entropy": 5.60307731628418,
      "epoch": 4.564841498559078,
      "grad_norm": 1.328125,
      "learning_rate": 0.000309327247258122,
      "loss": 4.7721,
      "mean_token_accuracy": 0.2455498605966568,
      "num_tokens": 108983127.0,
      "step": 47520
    },
    {
      "entropy": 5.529120063781738,
      "epoch": 4.565321805955811,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003092933646496996,
      "loss": 4.747,
      "mean_token_accuracy": 0.2523654654622078,
      "num_tokens": 108994764.0,
      "step": 47525
    },
    {
      "entropy": 5.571789932250977,
      "epoch": 4.565802113352546,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003092594812450871,
      "loss": 4.6898,
      "mean_token_accuracy": 0.2594235688447952,
      "num_tokens": 109005806.0,
      "step": 47530
    },
    {
      "entropy": 5.526821136474609,
      "epoch": 4.56628242074928,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003092255970450712,
      "loss": 4.7116,
      "mean_token_accuracy": 0.2527335822582245,
      "num_tokens": 109017819.0,
      "step": 47535
    },
    {
      "entropy": 5.4747549533844,
      "epoch": 4.5667627281460135,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003091917120504385,
      "loss": 4.5892,
      "mean_token_accuracy": 0.2620863109827042,
      "num_tokens": 109031049.0,
      "step": 47540
    },
    {
      "entropy": 5.497329568862915,
      "epoch": 4.567243035542747,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003091578262619758,
      "loss": 4.6751,
      "mean_token_accuracy": 0.2546574339270592,
      "num_tokens": 109042206.0,
      "step": 47545
    },
    {
      "entropy": 5.502302455902099,
      "epoch": 4.567723342939481,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030912393968046984,
      "loss": 4.6889,
      "mean_token_accuracy": 0.25278534591197965,
      "num_tokens": 109054340.0,
      "step": 47550
    },
    {
      "entropy": 5.457642698287964,
      "epoch": 4.568203650336216,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030909005230670734,
      "loss": 4.6395,
      "mean_token_accuracy": 0.2663937881588936,
      "num_tokens": 109065450.0,
      "step": 47555
    },
    {
      "entropy": 5.5127593040466305,
      "epoch": 4.568683957732949,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030905616414147503,
      "loss": 4.6778,
      "mean_token_accuracy": 0.2507965207099915,
      "num_tokens": 109075649.0,
      "step": 47560
    },
    {
      "entropy": 5.437182283401489,
      "epoch": 4.569164265129683,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030902227518555967,
      "loss": 4.6512,
      "mean_token_accuracy": 0.2644876778125763,
      "num_tokens": 109085403.0,
      "step": 47565
    },
    {
      "entropy": 5.487937688827515,
      "epoch": 4.569644572526417,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030898838543974816,
      "loss": 4.6887,
      "mean_token_accuracy": 0.25814932882785796,
      "num_tokens": 109096047.0,
      "step": 47570
    },
    {
      "entropy": 5.558182334899902,
      "epoch": 4.570124879923151,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030895449490482725,
      "loss": 4.6652,
      "mean_token_accuracy": 0.2647016361355782,
      "num_tokens": 109107543.0,
      "step": 47575
    },
    {
      "entropy": 5.449149560928345,
      "epoch": 4.5706051873198845,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030892060358158376,
      "loss": 4.5928,
      "mean_token_accuracy": 0.26470627784729006,
      "num_tokens": 109119099.0,
      "step": 47580
    },
    {
      "entropy": 5.42692551612854,
      "epoch": 4.571085494716619,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030888671147080466,
      "loss": 4.6268,
      "mean_token_accuracy": 0.256746469438076,
      "num_tokens": 109130362.0,
      "step": 47585
    },
    {
      "entropy": 5.523314476013184,
      "epoch": 4.571565802113352,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003088528185732767,
      "loss": 4.7315,
      "mean_token_accuracy": 0.2513928860425949,
      "num_tokens": 109141632.0,
      "step": 47590
    },
    {
      "entropy": 5.547695207595825,
      "epoch": 4.572046109510087,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00030881892488978676,
      "loss": 4.7275,
      "mean_token_accuracy": 0.2537601009011269,
      "num_tokens": 109153422.0,
      "step": 47595
    },
    {
      "entropy": 5.576894235610962,
      "epoch": 4.57252641690682,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030878503042112184,
      "loss": 4.7691,
      "mean_token_accuracy": 0.24899111688137054,
      "num_tokens": 109165412.0,
      "step": 47600
    },
    {
      "entropy": 5.491971969604492,
      "epoch": 4.573006724303554,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030875113516806886,
      "loss": 4.6417,
      "mean_token_accuracy": 0.26153901815414426,
      "num_tokens": 109176419.0,
      "step": 47605
    },
    {
      "entropy": 5.453462409973144,
      "epoch": 4.573487031700288,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003087172391314147,
      "loss": 4.6653,
      "mean_token_accuracy": 0.25924576818943024,
      "num_tokens": 109187456.0,
      "step": 47610
    },
    {
      "entropy": 5.541285657882691,
      "epoch": 4.573967339097022,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030868334231194645,
      "loss": 4.7261,
      "mean_token_accuracy": 0.2564445823431015,
      "num_tokens": 109198687.0,
      "step": 47615
    },
    {
      "entropy": 5.571504259109497,
      "epoch": 4.574447646493756,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003086494447104509,
      "loss": 4.677,
      "mean_token_accuracy": 0.25889245569705965,
      "num_tokens": 109211091.0,
      "step": 47620
    },
    {
      "entropy": 5.48736252784729,
      "epoch": 4.57492795389049,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003086155463277152,
      "loss": 4.6844,
      "mean_token_accuracy": 0.2611022859811783,
      "num_tokens": 109222767.0,
      "step": 47625
    },
    {
      "entropy": 5.483443880081177,
      "epoch": 4.575408261287224,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003085816471645264,
      "loss": 4.6853,
      "mean_token_accuracy": 0.2578812301158905,
      "num_tokens": 109234030.0,
      "step": 47630
    },
    {
      "entropy": 5.500910329818725,
      "epoch": 4.575888568683958,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003085477472216714,
      "loss": 4.7062,
      "mean_token_accuracy": 0.2568618804216385,
      "num_tokens": 109245453.0,
      "step": 47635
    },
    {
      "entropy": 5.488492345809936,
      "epoch": 4.576368876080692,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003085138464999373,
      "loss": 4.5875,
      "mean_token_accuracy": 0.26095940321683886,
      "num_tokens": 109257322.0,
      "step": 47640
    },
    {
      "entropy": 5.501231002807617,
      "epoch": 4.576849183477425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003084799450001113,
      "loss": 4.6949,
      "mean_token_accuracy": 0.24649676978588103,
      "num_tokens": 109270696.0,
      "step": 47645
    },
    {
      "entropy": 5.462875413894653,
      "epoch": 4.57732949087416,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003084460427229803,
      "loss": 4.6381,
      "mean_token_accuracy": 0.25616317242383957,
      "num_tokens": 109282505.0,
      "step": 47650
    },
    {
      "entropy": 5.467487430572509,
      "epoch": 4.577809798270893,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003084121396693316,
      "loss": 4.6517,
      "mean_token_accuracy": 0.2660590007901192,
      "num_tokens": 109292984.0,
      "step": 47655
    },
    {
      "entropy": 5.4950666427612305,
      "epoch": 4.5782901056676275,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030837823583995216,
      "loss": 4.6985,
      "mean_token_accuracy": 0.25920102149248125,
      "num_tokens": 109303797.0,
      "step": 47660
    },
    {
      "entropy": 5.557495641708374,
      "epoch": 4.578770413064361,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003083443312356292,
      "loss": 4.6983,
      "mean_token_accuracy": 0.2619380339980125,
      "num_tokens": 109315535.0,
      "step": 47665
    },
    {
      "entropy": 5.515939664840698,
      "epoch": 4.579250720461095,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030831042585714984,
      "loss": 4.6005,
      "mean_token_accuracy": 0.2572850897908211,
      "num_tokens": 109326773.0,
      "step": 47670
    },
    {
      "entropy": 5.539491510391235,
      "epoch": 4.579731027857829,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030827651970530143,
      "loss": 4.6881,
      "mean_token_accuracy": 0.2536906287074089,
      "num_tokens": 109337153.0,
      "step": 47675
    },
    {
      "entropy": 5.488747930526733,
      "epoch": 4.580211335254563,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003082426127808709,
      "loss": 4.7052,
      "mean_token_accuracy": 0.25314697027206423,
      "num_tokens": 109348854.0,
      "step": 47680
    },
    {
      "entropy": 5.457170867919922,
      "epoch": 4.5806916426512965,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003082087050846457,
      "loss": 4.6139,
      "mean_token_accuracy": 0.26403619050979615,
      "num_tokens": 109360902.0,
      "step": 47685
    },
    {
      "entropy": 5.588356971740723,
      "epoch": 4.581171950048031,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003081747966174129,
      "loss": 4.8187,
      "mean_token_accuracy": 0.2535311967134476,
      "num_tokens": 109372372.0,
      "step": 47690
    },
    {
      "entropy": 5.536856317520142,
      "epoch": 4.581652257444764,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003081408873799599,
      "loss": 4.6482,
      "mean_token_accuracy": 0.25868945866823195,
      "num_tokens": 109383807.0,
      "step": 47695
    },
    {
      "entropy": 5.468714046478271,
      "epoch": 4.582132564841499,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003081069773730739,
      "loss": 4.601,
      "mean_token_accuracy": 0.2662356957793236,
      "num_tokens": 109396344.0,
      "step": 47700
    },
    {
      "entropy": 5.499070262908935,
      "epoch": 4.582612872238233,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003080730665975423,
      "loss": 4.6829,
      "mean_token_accuracy": 0.26542383432388306,
      "num_tokens": 109408102.0,
      "step": 47705
    },
    {
      "entropy": 5.539494037628174,
      "epoch": 4.583093179634966,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003080391550541521,
      "loss": 4.7393,
      "mean_token_accuracy": 0.2480764478445053,
      "num_tokens": 109418352.0,
      "step": 47710
    },
    {
      "entropy": 5.642609691619873,
      "epoch": 4.583573487031701,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003080052427436909,
      "loss": 4.7397,
      "mean_token_accuracy": 0.2552469834685326,
      "num_tokens": 109429176.0,
      "step": 47715
    },
    {
      "entropy": 5.544283723831176,
      "epoch": 4.584053794428434,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000307971329666946,
      "loss": 4.6341,
      "mean_token_accuracy": 0.2645201817154884,
      "num_tokens": 109440370.0,
      "step": 47720
    },
    {
      "entropy": 5.615363693237304,
      "epoch": 4.584534101825168,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003079374158247047,
      "loss": 4.787,
      "mean_token_accuracy": 0.24806243032217026,
      "num_tokens": 109452268.0,
      "step": 47725
    },
    {
      "entropy": 5.5467452049255375,
      "epoch": 4.585014409221902,
      "grad_norm": 1.25,
      "learning_rate": 0.0003079035012177544,
      "loss": 4.6928,
      "mean_token_accuracy": 0.262269951403141,
      "num_tokens": 109463398.0,
      "step": 47730
    },
    {
      "entropy": 5.593846607208252,
      "epoch": 4.585494716618636,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003078695858468825,
      "loss": 4.7478,
      "mean_token_accuracy": 0.24805080890655518,
      "num_tokens": 109475860.0,
      "step": 47735
    },
    {
      "entropy": 5.478199481964111,
      "epoch": 4.58597502401537,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003078356697128765,
      "loss": 4.6268,
      "mean_token_accuracy": 0.261628820002079,
      "num_tokens": 109487355.0,
      "step": 47740
    },
    {
      "entropy": 5.431950378417969,
      "epoch": 4.586455331412104,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003078017528165236,
      "loss": 4.6047,
      "mean_token_accuracy": 0.267733097076416,
      "num_tokens": 109498477.0,
      "step": 47745
    },
    {
      "entropy": 5.608608436584473,
      "epoch": 4.586935638808837,
      "grad_norm": 1.46875,
      "learning_rate": 0.00030776783515861156,
      "loss": 4.7677,
      "mean_token_accuracy": 0.2513077184557915,
      "num_tokens": 109508997.0,
      "step": 47750
    },
    {
      "entropy": 5.459484720230103,
      "epoch": 4.587415946205572,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030773391673992754,
      "loss": 4.6791,
      "mean_token_accuracy": 0.25531307607889175,
      "num_tokens": 109520271.0,
      "step": 47755
    },
    {
      "entropy": 5.524590015411377,
      "epoch": 4.587896253602305,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003076999975612593,
      "loss": 4.7136,
      "mean_token_accuracy": 0.24560218304395676,
      "num_tokens": 109532118.0,
      "step": 47760
    },
    {
      "entropy": 5.55395131111145,
      "epoch": 4.5883765609990395,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003076660776233942,
      "loss": 4.7157,
      "mean_token_accuracy": 0.2536682218313217,
      "num_tokens": 109542936.0,
      "step": 47765
    },
    {
      "entropy": 5.538242483139038,
      "epoch": 4.588856868395773,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003076321569271197,
      "loss": 4.686,
      "mean_token_accuracy": 0.2574518233537674,
      "num_tokens": 109555542.0,
      "step": 47770
    },
    {
      "entropy": 5.497792959213257,
      "epoch": 4.589337175792507,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030759823547322346,
      "loss": 4.7059,
      "mean_token_accuracy": 0.2588791519403458,
      "num_tokens": 109566347.0,
      "step": 47775
    },
    {
      "entropy": 5.483173656463623,
      "epoch": 4.5898174831892415,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000307564313262493,
      "loss": 4.591,
      "mean_token_accuracy": 0.2622149035334587,
      "num_tokens": 109577263.0,
      "step": 47780
    },
    {
      "entropy": 5.543571662902832,
      "epoch": 4.590297790585975,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030753039029571586,
      "loss": 4.7425,
      "mean_token_accuracy": 0.25336888879537584,
      "num_tokens": 109587998.0,
      "step": 47785
    },
    {
      "entropy": 5.5112488746643065,
      "epoch": 4.590778097982709,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030749646657367963,
      "loss": 4.6538,
      "mean_token_accuracy": 0.26500523239374163,
      "num_tokens": 109600419.0,
      "step": 47790
    },
    {
      "entropy": 5.4944188594818115,
      "epoch": 4.591258405379443,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000307462542097172,
      "loss": 4.649,
      "mean_token_accuracy": 0.2683283567428589,
      "num_tokens": 109612100.0,
      "step": 47795
    },
    {
      "entropy": 5.516100311279297,
      "epoch": 4.591738712776177,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030742861686698056,
      "loss": 4.6397,
      "mean_token_accuracy": 0.25588996410369874,
      "num_tokens": 109624163.0,
      "step": 47800
    },
    {
      "entropy": 5.613224983215332,
      "epoch": 4.5922190201729105,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003073946908838928,
      "loss": 4.7471,
      "mean_token_accuracy": 0.2548310399055481,
      "num_tokens": 109634751.0,
      "step": 47805
    },
    {
      "entropy": 5.5200902938842775,
      "epoch": 4.592699327569645,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003073607641486966,
      "loss": 4.6856,
      "mean_token_accuracy": 0.26005166321992873,
      "num_tokens": 109645681.0,
      "step": 47810
    },
    {
      "entropy": 5.530058240890503,
      "epoch": 4.593179634966378,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030732683666217955,
      "loss": 4.7325,
      "mean_token_accuracy": 0.2590674474835396,
      "num_tokens": 109657817.0,
      "step": 47815
    },
    {
      "entropy": 5.554967164993286,
      "epoch": 4.593659942363113,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030729290842512936,
      "loss": 4.6351,
      "mean_token_accuracy": 0.26257012486457826,
      "num_tokens": 109668520.0,
      "step": 47820
    },
    {
      "entropy": 5.557923269271851,
      "epoch": 4.594140249759846,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030725897943833364,
      "loss": 4.7343,
      "mean_token_accuracy": 0.2527798727154732,
      "num_tokens": 109678856.0,
      "step": 47825
    },
    {
      "entropy": 5.53437032699585,
      "epoch": 4.59462055715658,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030722504970258027,
      "loss": 4.6785,
      "mean_token_accuracy": 0.2651423722505569,
      "num_tokens": 109690610.0,
      "step": 47830
    },
    {
      "entropy": 5.472569847106934,
      "epoch": 4.595100864553314,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030719111921865687,
      "loss": 4.6494,
      "mean_token_accuracy": 0.25848219990730287,
      "num_tokens": 109702473.0,
      "step": 47835
    },
    {
      "entropy": 5.497023105621338,
      "epoch": 4.595581171950048,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003071571879873514,
      "loss": 4.6254,
      "mean_token_accuracy": 0.2603593558073044,
      "num_tokens": 109713802.0,
      "step": 47840
    },
    {
      "entropy": 5.433384370803833,
      "epoch": 4.5960614793467816,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003071232560094514,
      "loss": 4.6474,
      "mean_token_accuracy": 0.25725975185632705,
      "num_tokens": 109725801.0,
      "step": 47845
    },
    {
      "entropy": 5.549112701416016,
      "epoch": 4.596541786743516,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030708932328574477,
      "loss": 4.7451,
      "mean_token_accuracy": 0.25105162113904955,
      "num_tokens": 109737570.0,
      "step": 47850
    },
    {
      "entropy": 5.508540678024292,
      "epoch": 4.59702209414025,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030705538981701946,
      "loss": 4.617,
      "mean_token_accuracy": 0.26246498227119447,
      "num_tokens": 109749535.0,
      "step": 47855
    },
    {
      "entropy": 5.465529346466065,
      "epoch": 4.597502401536984,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030702145560406314,
      "loss": 4.6126,
      "mean_token_accuracy": 0.264761021733284,
      "num_tokens": 109761020.0,
      "step": 47860
    },
    {
      "entropy": 5.519585752487183,
      "epoch": 4.597982708933718,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030698752064766365,
      "loss": 4.7324,
      "mean_token_accuracy": 0.2514249846339226,
      "num_tokens": 109773669.0,
      "step": 47865
    },
    {
      "entropy": 5.57053771018982,
      "epoch": 4.598463016330451,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000306953584948609,
      "loss": 4.78,
      "mean_token_accuracy": 0.23886046558618546,
      "num_tokens": 109784306.0,
      "step": 47870
    },
    {
      "entropy": 5.546192836761475,
      "epoch": 4.598943323727186,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003069196485076869,
      "loss": 4.7193,
      "mean_token_accuracy": 0.251239138841629,
      "num_tokens": 109796794.0,
      "step": 47875
    },
    {
      "entropy": 5.476920652389526,
      "epoch": 4.599423631123919,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003068857113256854,
      "loss": 4.5859,
      "mean_token_accuracy": 0.26685684472322463,
      "num_tokens": 109809139.0,
      "step": 47880
    },
    {
      "entropy": 5.522867679595947,
      "epoch": 4.5999039385206535,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003068517734033923,
      "loss": 4.7128,
      "mean_token_accuracy": 0.25245907306671145,
      "num_tokens": 109821791.0,
      "step": 47885
    },
    {
      "entropy": 5.53520655632019,
      "epoch": 4.600384245917387,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030681783474159564,
      "loss": 4.5854,
      "mean_token_accuracy": 0.2640076279640198,
      "num_tokens": 109832600.0,
      "step": 47890
    },
    {
      "entropy": 5.53027753829956,
      "epoch": 4.600864553314121,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003067838953410833,
      "loss": 4.6533,
      "mean_token_accuracy": 0.2571261703968048,
      "num_tokens": 109844035.0,
      "step": 47895
    },
    {
      "entropy": 5.452734518051147,
      "epoch": 4.601344860710855,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030674995520264337,
      "loss": 4.6771,
      "mean_token_accuracy": 0.2619456693530083,
      "num_tokens": 109856278.0,
      "step": 47900
    },
    {
      "entropy": 5.516651439666748,
      "epoch": 4.601825168107589,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030671601432706374,
      "loss": 4.6563,
      "mean_token_accuracy": 0.25702384412288665,
      "num_tokens": 109867093.0,
      "step": 47905
    },
    {
      "entropy": 5.521115112304687,
      "epoch": 4.6023054755043225,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003066820727151324,
      "loss": 4.7174,
      "mean_token_accuracy": 0.2532689034938812,
      "num_tokens": 109878304.0,
      "step": 47910
    },
    {
      "entropy": 5.555454063415527,
      "epoch": 4.602785782901057,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030664813036763744,
      "loss": 4.7046,
      "mean_token_accuracy": 0.2568758472800255,
      "num_tokens": 109889067.0,
      "step": 47915
    },
    {
      "entropy": 5.493565320968628,
      "epoch": 4.60326609029779,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003066141872853668,
      "loss": 4.6043,
      "mean_token_accuracy": 0.26482044756412504,
      "num_tokens": 109899802.0,
      "step": 47920
    },
    {
      "entropy": 5.57615327835083,
      "epoch": 4.6037463976945245,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003065802434691087,
      "loss": 4.7407,
      "mean_token_accuracy": 0.2520313963294029,
      "num_tokens": 109910510.0,
      "step": 47925
    },
    {
      "entropy": 5.526368713378906,
      "epoch": 4.604226705091259,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030654629891965106,
      "loss": 4.7265,
      "mean_token_accuracy": 0.25723725110292434,
      "num_tokens": 109922493.0,
      "step": 47930
    },
    {
      "entropy": 5.488086366653443,
      "epoch": 4.604707012487992,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003065123536377821,
      "loss": 4.7074,
      "mean_token_accuracy": 0.25875732153654096,
      "num_tokens": 109934256.0,
      "step": 47935
    },
    {
      "entropy": 5.534248399734497,
      "epoch": 4.605187319884726,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003064784076242897,
      "loss": 4.6489,
      "mean_token_accuracy": 0.26358423084020616,
      "num_tokens": 109945982.0,
      "step": 47940
    },
    {
      "entropy": 5.552996110916138,
      "epoch": 4.60566762728146,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003064444608799623,
      "loss": 4.7271,
      "mean_token_accuracy": 0.2559385225176811,
      "num_tokens": 109956838.0,
      "step": 47945
    },
    {
      "entropy": 5.5286510467529295,
      "epoch": 4.606147934678194,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030641051340558786,
      "loss": 4.7379,
      "mean_token_accuracy": 0.25099442452192305,
      "num_tokens": 109968750.0,
      "step": 47950
    },
    {
      "entropy": 5.515021419525146,
      "epoch": 4.606628242074928,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030637656520195454,
      "loss": 4.6983,
      "mean_token_accuracy": 0.25710817873477937,
      "num_tokens": 109979146.0,
      "step": 47955
    },
    {
      "entropy": 5.536544942855835,
      "epoch": 4.607108549471662,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003063426162698506,
      "loss": 4.7541,
      "mean_token_accuracy": 0.25192147940397264,
      "num_tokens": 109991331.0,
      "step": 47960
    },
    {
      "entropy": 5.54798321723938,
      "epoch": 4.607588856868396,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003063086666100641,
      "loss": 4.7047,
      "mean_token_accuracy": 0.25957546979188917,
      "num_tokens": 110002819.0,
      "step": 47965
    },
    {
      "entropy": 5.429078006744385,
      "epoch": 4.60806916426513,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003062747162233834,
      "loss": 4.6079,
      "mean_token_accuracy": 0.261640565097332,
      "num_tokens": 110013888.0,
      "step": 47970
    },
    {
      "entropy": 5.523548269271851,
      "epoch": 4.608549471661863,
      "grad_norm": 1.25,
      "learning_rate": 0.0003062407651105966,
      "loss": 4.6761,
      "mean_token_accuracy": 0.24984775334596634,
      "num_tokens": 110026536.0,
      "step": 47975
    },
    {
      "entropy": 5.5532652854919435,
      "epoch": 4.609029779058598,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000306206813272492,
      "loss": 4.6984,
      "mean_token_accuracy": 0.25639779716730116,
      "num_tokens": 110038470.0,
      "step": 47980
    },
    {
      "entropy": 5.493975734710693,
      "epoch": 4.609510086455331,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003061728607098579,
      "loss": 4.6257,
      "mean_token_accuracy": 0.2575662061572075,
      "num_tokens": 110049087.0,
      "step": 47985
    },
    {
      "entropy": 5.523640489578247,
      "epoch": 4.609990393852065,
      "grad_norm": 1.25,
      "learning_rate": 0.0003061389074234825,
      "loss": 4.6769,
      "mean_token_accuracy": 0.2650493085384369,
      "num_tokens": 110060515.0,
      "step": 47990
    },
    {
      "entropy": 5.544516086578369,
      "epoch": 4.610470701248799,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003061049534141542,
      "loss": 4.7351,
      "mean_token_accuracy": 0.25318553596735,
      "num_tokens": 110071180.0,
      "step": 47995
    },
    {
      "entropy": 5.529023218154907,
      "epoch": 4.610951008645533,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003060709986826612,
      "loss": 4.6997,
      "mean_token_accuracy": 0.25110477954149246,
      "num_tokens": 110083288.0,
      "step": 48000
    },
    {
      "epoch": 4.610951008645533,
      "eval_entropy": 5.268608357602942,
      "eval_loss": 4.7801361083984375,
      "eval_mean_token_accuracy": 0.25996012354078,
      "eval_num_tokens": 110083288.0,
      "eval_runtime": 20.693,
      "eval_samples_per_second": 1585.8,
      "eval_steps_per_second": 198.231,
      "step": 48000
    },
    {
      "entropy": 5.482731008529663,
      "epoch": 4.6114313160422675,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030603704322979187,
      "loss": 4.6412,
      "mean_token_accuracy": 0.2612991020083427,
      "num_tokens": 110094492.0,
      "step": 48005
    },
    {
      "entropy": 5.475678110122681,
      "epoch": 4.611911623439001,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030600308705633453,
      "loss": 4.5806,
      "mean_token_accuracy": 0.2645551860332489,
      "num_tokens": 110105671.0,
      "step": 48010
    },
    {
      "entropy": 5.594863653182983,
      "epoch": 4.612391930835734,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003059691301630776,
      "loss": 4.815,
      "mean_token_accuracy": 0.24393034875392913,
      "num_tokens": 110117866.0,
      "step": 48015
    },
    {
      "entropy": 5.543740463256836,
      "epoch": 4.612872238232469,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003059351725508095,
      "loss": 4.6543,
      "mean_token_accuracy": 0.26270572394132613,
      "num_tokens": 110128619.0,
      "step": 48020
    },
    {
      "entropy": 5.483699131011963,
      "epoch": 4.613352545629203,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030590121422031843,
      "loss": 4.6304,
      "mean_token_accuracy": 0.2607370391488075,
      "num_tokens": 110140366.0,
      "step": 48025
    },
    {
      "entropy": 5.518457841873169,
      "epoch": 4.6138328530259365,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030586725517239296,
      "loss": 4.6723,
      "mean_token_accuracy": 0.265580427646637,
      "num_tokens": 110151187.0,
      "step": 48030
    },
    {
      "entropy": 5.581230640411377,
      "epoch": 4.614313160422671,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003058332954078215,
      "loss": 4.7087,
      "mean_token_accuracy": 0.25984827876091005,
      "num_tokens": 110162243.0,
      "step": 48035
    },
    {
      "entropy": 5.555036354064941,
      "epoch": 4.614793467819404,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003057993349273925,
      "loss": 4.7283,
      "mean_token_accuracy": 0.2525343492627144,
      "num_tokens": 110174551.0,
      "step": 48040
    },
    {
      "entropy": 5.557419681549073,
      "epoch": 4.615273775216139,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003057653737318943,
      "loss": 4.7464,
      "mean_token_accuracy": 0.2641154617071152,
      "num_tokens": 110184889.0,
      "step": 48045
    },
    {
      "entropy": 5.556359767913818,
      "epoch": 4.615754082612872,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030573141182211564,
      "loss": 4.6961,
      "mean_token_accuracy": 0.24642374366521835,
      "num_tokens": 110195749.0,
      "step": 48050
    },
    {
      "entropy": 5.4820620059967045,
      "epoch": 4.616234390009606,
      "grad_norm": 1.25,
      "learning_rate": 0.0003056974491988447,
      "loss": 4.6319,
      "mean_token_accuracy": 0.25696843564510347,
      "num_tokens": 110206703.0,
      "step": 48055
    },
    {
      "entropy": 5.508238887786865,
      "epoch": 4.61671469740634,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003056634858628702,
      "loss": 4.7371,
      "mean_token_accuracy": 0.2517372310161591,
      "num_tokens": 110218610.0,
      "step": 48060
    },
    {
      "entropy": 5.542927932739258,
      "epoch": 4.617195004803074,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003056295218149806,
      "loss": 4.6586,
      "mean_token_accuracy": 0.25777973234653473,
      "num_tokens": 110229562.0,
      "step": 48065
    },
    {
      "entropy": 5.591064071655273,
      "epoch": 4.6176753121998075,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003055955570559645,
      "loss": 4.7659,
      "mean_token_accuracy": 0.24922859370708467,
      "num_tokens": 110242328.0,
      "step": 48070
    },
    {
      "entropy": 5.440161657333374,
      "epoch": 4.618155619596542,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030556159158661036,
      "loss": 4.5846,
      "mean_token_accuracy": 0.2645569175481796,
      "num_tokens": 110254649.0,
      "step": 48075
    },
    {
      "entropy": 5.531926059722901,
      "epoch": 4.618635926993276,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003055276254077068,
      "loss": 4.7581,
      "mean_token_accuracy": 0.25029812902212145,
      "num_tokens": 110265649.0,
      "step": 48080
    },
    {
      "entropy": 5.620811605453492,
      "epoch": 4.61911623439001,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003054936585200424,
      "loss": 4.6893,
      "mean_token_accuracy": 0.2555052191019058,
      "num_tokens": 110275994.0,
      "step": 48085
    },
    {
      "entropy": 5.533994722366333,
      "epoch": 4.619596541786743,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003054596909244059,
      "loss": 4.6715,
      "mean_token_accuracy": 0.2604081019759178,
      "num_tokens": 110287422.0,
      "step": 48090
    },
    {
      "entropy": 5.462889671325684,
      "epoch": 4.620076849183477,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030542572262158574,
      "loss": 4.6371,
      "mean_token_accuracy": 0.26055665612220763,
      "num_tokens": 110298238.0,
      "step": 48095
    },
    {
      "entropy": 5.497009372711181,
      "epoch": 4.620557156580212,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030539175361237066,
      "loss": 4.6469,
      "mean_token_accuracy": 0.257544519007206,
      "num_tokens": 110309517.0,
      "step": 48100
    },
    {
      "entropy": 5.617881917953492,
      "epoch": 4.621037463976945,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003053577838975494,
      "loss": 4.7482,
      "mean_token_accuracy": 0.25470310151576997,
      "num_tokens": 110320174.0,
      "step": 48105
    },
    {
      "entropy": 5.483324193954468,
      "epoch": 4.6215177713736795,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030532381347791044,
      "loss": 4.6154,
      "mean_token_accuracy": 0.2664415031671524,
      "num_tokens": 110331403.0,
      "step": 48110
    },
    {
      "entropy": 5.492835426330567,
      "epoch": 4.621998078770413,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030528984235424255,
      "loss": 4.6196,
      "mean_token_accuracy": 0.2610191717743874,
      "num_tokens": 110341661.0,
      "step": 48115
    },
    {
      "entropy": 5.531013059616089,
      "epoch": 4.622478386167147,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030525587052733456,
      "loss": 4.7368,
      "mean_token_accuracy": 0.2545017793774605,
      "num_tokens": 110353712.0,
      "step": 48120
    },
    {
      "entropy": 5.521227645874023,
      "epoch": 4.622958693563881,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003052218979979751,
      "loss": 4.6815,
      "mean_token_accuracy": 0.26020281165838244,
      "num_tokens": 110366541.0,
      "step": 48125
    },
    {
      "entropy": 5.542605352401734,
      "epoch": 4.623439000960615,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003051879247669528,
      "loss": 4.7014,
      "mean_token_accuracy": 0.25271919518709185,
      "num_tokens": 110377284.0,
      "step": 48130
    },
    {
      "entropy": 5.528365707397461,
      "epoch": 4.623919308357348,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003051539508350566,
      "loss": 4.6963,
      "mean_token_accuracy": 0.25517299175262453,
      "num_tokens": 110388317.0,
      "step": 48135
    },
    {
      "entropy": 5.494279241561889,
      "epoch": 4.624399615754083,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003051199762030752,
      "loss": 4.7081,
      "mean_token_accuracy": 0.2554492443799973,
      "num_tokens": 110399993.0,
      "step": 48140
    },
    {
      "entropy": 5.509506797790527,
      "epoch": 4.624879923150816,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003050860008717974,
      "loss": 4.6766,
      "mean_token_accuracy": 0.26060790568590164,
      "num_tokens": 110411033.0,
      "step": 48145
    },
    {
      "entropy": 5.442971420288086,
      "epoch": 4.6253602305475505,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030505202484201196,
      "loss": 4.6035,
      "mean_token_accuracy": 0.264740988612175,
      "num_tokens": 110422860.0,
      "step": 48150
    },
    {
      "entropy": 5.499479198455811,
      "epoch": 4.625840537944284,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003050180481145077,
      "loss": 4.6799,
      "mean_token_accuracy": 0.2640187978744507,
      "num_tokens": 110434632.0,
      "step": 48155
    },
    {
      "entropy": 5.507255697250367,
      "epoch": 4.626320845341018,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003049840706900736,
      "loss": 4.6562,
      "mean_token_accuracy": 0.26794320791959764,
      "num_tokens": 110445177.0,
      "step": 48160
    },
    {
      "entropy": 5.570025110244751,
      "epoch": 4.626801152737752,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030495009256949836,
      "loss": 4.7517,
      "mean_token_accuracy": 0.2490591362118721,
      "num_tokens": 110456877.0,
      "step": 48165
    },
    {
      "entropy": 5.549390840530395,
      "epoch": 4.627281460134486,
      "grad_norm": 1.234375,
      "learning_rate": 0.000304916113753571,
      "loss": 4.6891,
      "mean_token_accuracy": 0.2548264801502228,
      "num_tokens": 110468958.0,
      "step": 48170
    },
    {
      "entropy": 5.490346002578735,
      "epoch": 4.62776176753122,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003048821342430802,
      "loss": 4.68,
      "mean_token_accuracy": 0.25934825986623766,
      "num_tokens": 110480673.0,
      "step": 48175
    },
    {
      "entropy": 5.580412530899048,
      "epoch": 4.628242074927954,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030484815403881494,
      "loss": 4.7214,
      "mean_token_accuracy": 0.251126629114151,
      "num_tokens": 110491137.0,
      "step": 48180
    },
    {
      "entropy": 5.510111570358276,
      "epoch": 4.628722382324688,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003048141731415643,
      "loss": 4.6729,
      "mean_token_accuracy": 0.2610503226518631,
      "num_tokens": 110502812.0,
      "step": 48185
    },
    {
      "entropy": 5.435949516296387,
      "epoch": 4.629202689721422,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030478019155211693,
      "loss": 4.5651,
      "mean_token_accuracy": 0.2703158140182495,
      "num_tokens": 110515058.0,
      "step": 48190
    },
    {
      "entropy": 5.473851060867309,
      "epoch": 4.629682997118156,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030474620927126205,
      "loss": 4.6766,
      "mean_token_accuracy": 0.26350001245737076,
      "num_tokens": 110527024.0,
      "step": 48195
    },
    {
      "entropy": 5.580460739135742,
      "epoch": 4.630163304514889,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003047122262997885,
      "loss": 4.761,
      "mean_token_accuracy": 0.2544008165597916,
      "num_tokens": 110537481.0,
      "step": 48200
    },
    {
      "entropy": 5.6109046459198,
      "epoch": 4.630643611911624,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003046782426384853,
      "loss": 4.778,
      "mean_token_accuracy": 0.25145078748464583,
      "num_tokens": 110548882.0,
      "step": 48205
    },
    {
      "entropy": 5.514479446411133,
      "epoch": 4.631123919308357,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030464425828814144,
      "loss": 4.6741,
      "mean_token_accuracy": 0.2556160777807236,
      "num_tokens": 110562298.0,
      "step": 48210
    },
    {
      "entropy": 5.497320032119751,
      "epoch": 4.631604226705091,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030461027324954595,
      "loss": 4.6971,
      "mean_token_accuracy": 0.25715157091617585,
      "num_tokens": 110574756.0,
      "step": 48215
    },
    {
      "entropy": 5.513008546829224,
      "epoch": 4.632084534101825,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030457628752348774,
      "loss": 4.6684,
      "mean_token_accuracy": 0.25990469604730604,
      "num_tokens": 110586632.0,
      "step": 48220
    },
    {
      "entropy": 5.537447309494018,
      "epoch": 4.632564841498559,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030454230111075605,
      "loss": 4.7171,
      "mean_token_accuracy": 0.25381266325712204,
      "num_tokens": 110598506.0,
      "step": 48225
    },
    {
      "entropy": 5.553474569320679,
      "epoch": 4.633045148895293,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003045083140121398,
      "loss": 4.6243,
      "mean_token_accuracy": 0.26574052721261976,
      "num_tokens": 110610130.0,
      "step": 48230
    },
    {
      "entropy": 5.480491733551025,
      "epoch": 4.633525456292027,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003044743262284281,
      "loss": 4.6911,
      "mean_token_accuracy": 0.253385117650032,
      "num_tokens": 110622017.0,
      "step": 48235
    },
    {
      "entropy": 5.559721517562866,
      "epoch": 4.63400576368876,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030444033776041014,
      "loss": 4.7956,
      "mean_token_accuracy": 0.244127157330513,
      "num_tokens": 110633993.0,
      "step": 48240
    },
    {
      "entropy": 5.547993612289429,
      "epoch": 4.634486071085495,
      "grad_norm": 1.25,
      "learning_rate": 0.0003044063486088749,
      "loss": 4.6756,
      "mean_token_accuracy": 0.2544633254408836,
      "num_tokens": 110645791.0,
      "step": 48245
    },
    {
      "entropy": 5.452391672134399,
      "epoch": 4.634966378482229,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030437235877461155,
      "loss": 4.5569,
      "mean_token_accuracy": 0.27090358585119245,
      "num_tokens": 110656254.0,
      "step": 48250
    },
    {
      "entropy": 5.551088094711304,
      "epoch": 4.6354466858789625,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030433836825840925,
      "loss": 4.6916,
      "mean_token_accuracy": 0.2554528921842575,
      "num_tokens": 110668757.0,
      "step": 48255
    },
    {
      "entropy": 5.544843769073486,
      "epoch": 4.635926993275697,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003043043770610572,
      "loss": 4.7203,
      "mean_token_accuracy": 0.2547572419047356,
      "num_tokens": 110679152.0,
      "step": 48260
    },
    {
      "entropy": 5.512409019470215,
      "epoch": 4.63640730067243,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030427038518334457,
      "loss": 4.7001,
      "mean_token_accuracy": 0.25431473553180695,
      "num_tokens": 110691683.0,
      "step": 48265
    },
    {
      "entropy": 5.508939027786255,
      "epoch": 4.636887608069165,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003042363926260604,
      "loss": 4.64,
      "mean_token_accuracy": 0.263793882727623,
      "num_tokens": 110702203.0,
      "step": 48270
    },
    {
      "entropy": 5.436768817901611,
      "epoch": 4.637367915465898,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003042023993899941,
      "loss": 4.5841,
      "mean_token_accuracy": 0.26239892542362214,
      "num_tokens": 110712864.0,
      "step": 48275
    },
    {
      "entropy": 5.43401198387146,
      "epoch": 4.637848222862632,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003041684054759348,
      "loss": 4.6133,
      "mean_token_accuracy": 0.2609002083539963,
      "num_tokens": 110726186.0,
      "step": 48280
    },
    {
      "entropy": 5.476765298843384,
      "epoch": 4.638328530259366,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003041344108846716,
      "loss": 4.6325,
      "mean_token_accuracy": 0.2610874608159065,
      "num_tokens": 110738125.0,
      "step": 48285
    },
    {
      "entropy": 5.48120436668396,
      "epoch": 4.6388088376561,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003041004156169941,
      "loss": 4.6123,
      "mean_token_accuracy": 0.25772055238485336,
      "num_tokens": 110750537.0,
      "step": 48290
    },
    {
      "entropy": 5.50983510017395,
      "epoch": 4.6392891450528335,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003040664196736912,
      "loss": 4.6567,
      "mean_token_accuracy": 0.2550644800066948,
      "num_tokens": 110762525.0,
      "step": 48295
    },
    {
      "entropy": 5.587147235870361,
      "epoch": 4.639769452449568,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003040324230555524,
      "loss": 4.7113,
      "mean_token_accuracy": 0.24679161012172698,
      "num_tokens": 110774337.0,
      "step": 48300
    },
    {
      "entropy": 5.613292503356933,
      "epoch": 4.640249759846301,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030399842576336693,
      "loss": 4.7525,
      "mean_token_accuracy": 0.24908613413572311,
      "num_tokens": 110784438.0,
      "step": 48305
    },
    {
      "entropy": 5.558576822280884,
      "epoch": 4.640730067243036,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030396442779792414,
      "loss": 4.7503,
      "mean_token_accuracy": 0.25030112117528913,
      "num_tokens": 110795409.0,
      "step": 48310
    },
    {
      "entropy": 5.463618230819702,
      "epoch": 4.641210374639769,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003039304291600133,
      "loss": 4.6407,
      "mean_token_accuracy": 0.25883212834596636,
      "num_tokens": 110805923.0,
      "step": 48315
    },
    {
      "entropy": 5.4460248947143555,
      "epoch": 4.641690682036503,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030389642985042387,
      "loss": 4.6537,
      "mean_token_accuracy": 0.26455303877592085,
      "num_tokens": 110817301.0,
      "step": 48320
    },
    {
      "entropy": 5.532053852081299,
      "epoch": 4.642170989433238,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003038624298699451,
      "loss": 4.6893,
      "mean_token_accuracy": 0.2576109245419502,
      "num_tokens": 110828736.0,
      "step": 48325
    },
    {
      "entropy": 5.482851600646972,
      "epoch": 4.642651296829971,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030382842921936645,
      "loss": 4.6748,
      "mean_token_accuracy": 0.2578031688928604,
      "num_tokens": 110841611.0,
      "step": 48330
    },
    {
      "entropy": 5.584382009506226,
      "epoch": 4.6431316042267055,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030379442789947727,
      "loss": 4.7089,
      "mean_token_accuracy": 0.2618792325258255,
      "num_tokens": 110854050.0,
      "step": 48335
    },
    {
      "entropy": 5.543910932540894,
      "epoch": 4.643611911623439,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030376042591106697,
      "loss": 4.7084,
      "mean_token_accuracy": 0.2532899439334869,
      "num_tokens": 110865165.0,
      "step": 48340
    },
    {
      "entropy": 5.579921102523803,
      "epoch": 4.644092219020173,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003037264232549249,
      "loss": 4.7643,
      "mean_token_accuracy": 0.2501338109374046,
      "num_tokens": 110876538.0,
      "step": 48345
    },
    {
      "entropy": 5.5218945980072025,
      "epoch": 4.644572526416907,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030369241993184065,
      "loss": 4.6957,
      "mean_token_accuracy": 0.25474375784397124,
      "num_tokens": 110888465.0,
      "step": 48350
    },
    {
      "entropy": 5.619806337356567,
      "epoch": 4.645052833813641,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003036584159426036,
      "loss": 4.7718,
      "mean_token_accuracy": 0.24849555492401124,
      "num_tokens": 110899710.0,
      "step": 48355
    },
    {
      "entropy": 5.5712508201599125,
      "epoch": 4.645533141210374,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003036244112880032,
      "loss": 4.7216,
      "mean_token_accuracy": 0.2525214239954948,
      "num_tokens": 110910267.0,
      "step": 48360
    },
    {
      "entropy": 5.453323602676392,
      "epoch": 4.646013448607109,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030359040596882904,
      "loss": 4.6079,
      "mean_token_accuracy": 0.25901674926280976,
      "num_tokens": 110920798.0,
      "step": 48365
    },
    {
      "entropy": 5.5097911834716795,
      "epoch": 4.646493756003842,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030355639998587045,
      "loss": 4.6909,
      "mean_token_accuracy": 0.261538128554821,
      "num_tokens": 110932741.0,
      "step": 48370
    },
    {
      "entropy": 5.579377174377441,
      "epoch": 4.6469740634005765,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030352239333991714,
      "loss": 4.759,
      "mean_token_accuracy": 0.26306499987840654,
      "num_tokens": 110944511.0,
      "step": 48375
    },
    {
      "entropy": 5.525500202178955,
      "epoch": 4.64745437079731,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030348838603175857,
      "loss": 4.699,
      "mean_token_accuracy": 0.2519214525818825,
      "num_tokens": 110955760.0,
      "step": 48380
    },
    {
      "entropy": 5.715081596374512,
      "epoch": 4.647934678194044,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003034543780621842,
      "loss": 4.8603,
      "mean_token_accuracy": 0.2428887367248535,
      "num_tokens": 110967397.0,
      "step": 48385
    },
    {
      "entropy": 5.600068426132202,
      "epoch": 4.648414985590778,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003034203694319837,
      "loss": 4.7223,
      "mean_token_accuracy": 0.25541370213031767,
      "num_tokens": 110980851.0,
      "step": 48390
    },
    {
      "entropy": 5.546074390411377,
      "epoch": 4.648895292987512,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003033863601419466,
      "loss": 4.6688,
      "mean_token_accuracy": 0.2567230865359306,
      "num_tokens": 110991431.0,
      "step": 48395
    },
    {
      "entropy": 5.500859642028809,
      "epoch": 4.649375600384246,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030335235019286245,
      "loss": 4.6824,
      "mean_token_accuracy": 0.255844047665596,
      "num_tokens": 111003541.0,
      "step": 48400
    },
    {
      "entropy": 5.47667384147644,
      "epoch": 4.64985590778098,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030331833958552096,
      "loss": 4.6431,
      "mean_token_accuracy": 0.26452057957649233,
      "num_tokens": 111014661.0,
      "step": 48405
    },
    {
      "entropy": 5.558774089813232,
      "epoch": 4.650336215177714,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003032843283207118,
      "loss": 4.7557,
      "mean_token_accuracy": 0.25208091735839844,
      "num_tokens": 111026270.0,
      "step": 48410
    },
    {
      "entropy": 5.520854473114014,
      "epoch": 4.6508165225744476,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003032503163992244,
      "loss": 4.6859,
      "mean_token_accuracy": 0.25695353001356125,
      "num_tokens": 111037283.0,
      "step": 48415
    },
    {
      "entropy": 5.54622893333435,
      "epoch": 4.651296829971182,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003032163038218485,
      "loss": 4.6525,
      "mean_token_accuracy": 0.2612086832523346,
      "num_tokens": 111049187.0,
      "step": 48420
    },
    {
      "entropy": 5.583186388015747,
      "epoch": 4.651777137367915,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030318229058937395,
      "loss": 4.8195,
      "mean_token_accuracy": 0.2516620740294456,
      "num_tokens": 111061433.0,
      "step": 48425
    },
    {
      "entropy": 5.567944192886353,
      "epoch": 4.65225744476465,
      "grad_norm": 1.375,
      "learning_rate": 0.00030314827670259025,
      "loss": 4.6884,
      "mean_token_accuracy": 0.24604581594467162,
      "num_tokens": 111072262.0,
      "step": 48430
    },
    {
      "entropy": 5.515257978439331,
      "epoch": 4.652737752161383,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003031142621622871,
      "loss": 4.6168,
      "mean_token_accuracy": 0.2657399088144302,
      "num_tokens": 111084208.0,
      "step": 48435
    },
    {
      "entropy": 5.653163480758667,
      "epoch": 4.653218059558117,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003030802469692543,
      "loss": 4.9205,
      "mean_token_accuracy": 0.23456165045499802,
      "num_tokens": 111095605.0,
      "step": 48440
    },
    {
      "entropy": 5.555815505981445,
      "epoch": 4.653698366954851,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003030462311242816,
      "loss": 4.6645,
      "mean_token_accuracy": 0.2600269794464111,
      "num_tokens": 111106260.0,
      "step": 48445
    },
    {
      "entropy": 5.4351911544799805,
      "epoch": 4.654178674351585,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030301221462815863,
      "loss": 4.5972,
      "mean_token_accuracy": 0.25920582711696627,
      "num_tokens": 111118317.0,
      "step": 48450
    },
    {
      "entropy": 5.475290012359619,
      "epoch": 4.654658981748319,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003029781974816752,
      "loss": 4.6639,
      "mean_token_accuracy": 0.25647643506526946,
      "num_tokens": 111129575.0,
      "step": 48455
    },
    {
      "entropy": 5.521022939682007,
      "epoch": 4.655139289145053,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030294417968562123,
      "loss": 4.5989,
      "mean_token_accuracy": 0.2589129939675331,
      "num_tokens": 111141436.0,
      "step": 48460
    },
    {
      "entropy": 5.451436996459961,
      "epoch": 4.655619596541786,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003029101612407862,
      "loss": 4.626,
      "mean_token_accuracy": 0.2648378312587738,
      "num_tokens": 111151869.0,
      "step": 48465
    },
    {
      "entropy": 5.502292251586914,
      "epoch": 4.656099903938521,
      "grad_norm": 1.25,
      "learning_rate": 0.0003028761421479602,
      "loss": 4.7094,
      "mean_token_accuracy": 0.2568670302629471,
      "num_tokens": 111163226.0,
      "step": 48470
    },
    {
      "entropy": 5.488179492950439,
      "epoch": 4.656580211335255,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030284212240793296,
      "loss": 4.6939,
      "mean_token_accuracy": 0.2600498542189598,
      "num_tokens": 111175593.0,
      "step": 48475
    },
    {
      "entropy": 5.60559196472168,
      "epoch": 4.6570605187319885,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030280810202149434,
      "loss": 4.71,
      "mean_token_accuracy": 0.2575052618980408,
      "num_tokens": 111187690.0,
      "step": 48480
    },
    {
      "entropy": 5.557754898071289,
      "epoch": 4.657540826128722,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030277408098943403,
      "loss": 4.7053,
      "mean_token_accuracy": 0.25035312622785566,
      "num_tokens": 111200319.0,
      "step": 48485
    },
    {
      "entropy": 5.549062585830688,
      "epoch": 4.658021133525456,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003027400593125422,
      "loss": 4.7536,
      "mean_token_accuracy": 0.2531613394618034,
      "num_tokens": 111211883.0,
      "step": 48490
    },
    {
      "entropy": 5.497089910507202,
      "epoch": 4.6585014409221905,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003027060369916085,
      "loss": 4.6158,
      "mean_token_accuracy": 0.25904891192913054,
      "num_tokens": 111222192.0,
      "step": 48495
    },
    {
      "entropy": 5.507549524307251,
      "epoch": 4.658981748318924,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030267201402742286,
      "loss": 4.5995,
      "mean_token_accuracy": 0.26853623390197756,
      "num_tokens": 111233210.0,
      "step": 48500
    },
    {
      "entropy": 5.551734828948975,
      "epoch": 4.659462055715658,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003026379904207752,
      "loss": 4.6301,
      "mean_token_accuracy": 0.25542787462472916,
      "num_tokens": 111243605.0,
      "step": 48505
    },
    {
      "entropy": 5.515162134170533,
      "epoch": 4.659942363112392,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030260396617245555,
      "loss": 4.6794,
      "mean_token_accuracy": 0.2628790393471718,
      "num_tokens": 111253208.0,
      "step": 48510
    },
    {
      "entropy": 5.456355571746826,
      "epoch": 4.660422670509126,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003025699412832537,
      "loss": 4.6611,
      "mean_token_accuracy": 0.25552108883857727,
      "num_tokens": 111265129.0,
      "step": 48515
    },
    {
      "entropy": 5.511182117462158,
      "epoch": 4.6609029779058595,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003025359157539598,
      "loss": 4.6863,
      "mean_token_accuracy": 0.26012490391731263,
      "num_tokens": 111275864.0,
      "step": 48520
    },
    {
      "entropy": 5.551564645767212,
      "epoch": 4.661383285302594,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030250188958536354,
      "loss": 4.714,
      "mean_token_accuracy": 0.26039248108863833,
      "num_tokens": 111288364.0,
      "step": 48525
    },
    {
      "entropy": 5.48758454322815,
      "epoch": 4.661863592699327,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030246786277825515,
      "loss": 4.6297,
      "mean_token_accuracy": 0.2582154154777527,
      "num_tokens": 111299692.0,
      "step": 48530
    },
    {
      "entropy": 5.574607563018799,
      "epoch": 4.662343900096062,
      "grad_norm": 1.25,
      "learning_rate": 0.0003024338353334246,
      "loss": 4.7259,
      "mean_token_accuracy": 0.2593665897846222,
      "num_tokens": 111311024.0,
      "step": 48535
    },
    {
      "entropy": 5.552976608276367,
      "epoch": 4.662824207492795,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003023998072516618,
      "loss": 4.6455,
      "mean_token_accuracy": 0.2555416777729988,
      "num_tokens": 111322099.0,
      "step": 48540
    },
    {
      "entropy": 5.48212776184082,
      "epoch": 4.663304514889529,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003023657785337568,
      "loss": 4.6459,
      "mean_token_accuracy": 0.2644735500216484,
      "num_tokens": 111332940.0,
      "step": 48545
    },
    {
      "entropy": 5.443404579162598,
      "epoch": 4.663784822286264,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030233174918049973,
      "loss": 4.6065,
      "mean_token_accuracy": 0.2634139969944954,
      "num_tokens": 111343991.0,
      "step": 48550
    },
    {
      "entropy": 5.571722412109375,
      "epoch": 4.664265129682997,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030229771919268054,
      "loss": 4.7918,
      "mean_token_accuracy": 0.24870240241289138,
      "num_tokens": 111355367.0,
      "step": 48555
    },
    {
      "entropy": 5.513676404953003,
      "epoch": 4.6647454370797306,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030226368857108946,
      "loss": 4.7881,
      "mean_token_accuracy": 0.25050302445888517,
      "num_tokens": 111368232.0,
      "step": 48560
    },
    {
      "entropy": 5.514728212356568,
      "epoch": 4.665225744476465,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003022296573165165,
      "loss": 4.6233,
      "mean_token_accuracy": 0.25996540784835814,
      "num_tokens": 111380490.0,
      "step": 48565
    },
    {
      "entropy": 5.559771585464477,
      "epoch": 4.665706051873199,
      "grad_norm": 1.40625,
      "learning_rate": 0.00030219562542975167,
      "loss": 4.7816,
      "mean_token_accuracy": 0.24342858791351318,
      "num_tokens": 111392380.0,
      "step": 48570
    },
    {
      "entropy": 5.54527063369751,
      "epoch": 4.666186359269933,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003021615929115852,
      "loss": 4.6875,
      "mean_token_accuracy": 0.25831636488437654,
      "num_tokens": 111404407.0,
      "step": 48575
    },
    {
      "entropy": 5.508069849014282,
      "epoch": 4.666666666666667,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030212755976280714,
      "loss": 4.6675,
      "mean_token_accuracy": 0.26248078644275663,
      "num_tokens": 111414435.0,
      "step": 48580
    },
    {
      "entropy": 5.523973226547241,
      "epoch": 4.6671469740634,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030209352598420776,
      "loss": 4.682,
      "mean_token_accuracy": 0.2544418811798096,
      "num_tokens": 111425352.0,
      "step": 48585
    },
    {
      "entropy": 5.504978036880493,
      "epoch": 4.667627281460135,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030205949157657716,
      "loss": 4.7363,
      "mean_token_accuracy": 0.24681894928216935,
      "num_tokens": 111437316.0,
      "step": 48590
    },
    {
      "entropy": 5.443516397476197,
      "epoch": 4.668107588856868,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003020254565407055,
      "loss": 4.607,
      "mean_token_accuracy": 0.2609421476721764,
      "num_tokens": 111450032.0,
      "step": 48595
    },
    {
      "entropy": 5.470867013931274,
      "epoch": 4.6685878962536025,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000301991420877383,
      "loss": 4.6452,
      "mean_token_accuracy": 0.2594276949763298,
      "num_tokens": 111461332.0,
      "step": 48600
    },
    {
      "entropy": 5.535652160644531,
      "epoch": 4.669068203650336,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003019573845873998,
      "loss": 4.7643,
      "mean_token_accuracy": 0.255573208630085,
      "num_tokens": 111473359.0,
      "step": 48605
    },
    {
      "entropy": 5.563342905044555,
      "epoch": 4.66954851104707,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003019233476715463,
      "loss": 4.6991,
      "mean_token_accuracy": 0.25608727931976316,
      "num_tokens": 111484912.0,
      "step": 48610
    },
    {
      "entropy": 5.472263622283935,
      "epoch": 4.670028818443804,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003018893101306125,
      "loss": 4.6371,
      "mean_token_accuracy": 0.25938103944063184,
      "num_tokens": 111496985.0,
      "step": 48615
    },
    {
      "entropy": 5.521542692184449,
      "epoch": 4.670509125840538,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003018552719653888,
      "loss": 4.69,
      "mean_token_accuracy": 0.2540324732661247,
      "num_tokens": 111508329.0,
      "step": 48620
    },
    {
      "entropy": 5.602862358093262,
      "epoch": 4.670989433237272,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030182123317666544,
      "loss": 4.8038,
      "mean_token_accuracy": 0.2426360383629799,
      "num_tokens": 111520281.0,
      "step": 48625
    },
    {
      "entropy": 5.547756862640381,
      "epoch": 4.671469740634006,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003017871937652327,
      "loss": 4.6189,
      "mean_token_accuracy": 0.25958493500947954,
      "num_tokens": 111531602.0,
      "step": 48630
    },
    {
      "entropy": 5.572002220153808,
      "epoch": 4.671950048030739,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003017531537318808,
      "loss": 4.784,
      "mean_token_accuracy": 0.24815921038389205,
      "num_tokens": 111543792.0,
      "step": 48635
    },
    {
      "entropy": 5.459470415115357,
      "epoch": 4.6724303554274735,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003017191130774002,
      "loss": 4.6354,
      "mean_token_accuracy": 0.25827121138572695,
      "num_tokens": 111555167.0,
      "step": 48640
    },
    {
      "entropy": 5.624302721023559,
      "epoch": 4.672910662824208,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000301685071802581,
      "loss": 4.8085,
      "mean_token_accuracy": 0.2521528095006943,
      "num_tokens": 111565819.0,
      "step": 48645
    },
    {
      "entropy": 5.594880867004394,
      "epoch": 4.673390970220941,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003016510299082138,
      "loss": 4.774,
      "mean_token_accuracy": 0.25240061432123184,
      "num_tokens": 111577921.0,
      "step": 48650
    },
    {
      "entropy": 5.5675488948822025,
      "epoch": 4.673871277617676,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030161698739508875,
      "loss": 4.7337,
      "mean_token_accuracy": 0.25573980808258057,
      "num_tokens": 111588860.0,
      "step": 48655
    },
    {
      "entropy": 5.585457992553711,
      "epoch": 4.674351585014409,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030158294426399637,
      "loss": 4.7459,
      "mean_token_accuracy": 0.2551688998937607,
      "num_tokens": 111600775.0,
      "step": 48660
    },
    {
      "entropy": 5.517655658721924,
      "epoch": 4.674831892411143,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003015489005157269,
      "loss": 4.6607,
      "mean_token_accuracy": 0.25463277846574783,
      "num_tokens": 111611953.0,
      "step": 48665
    },
    {
      "entropy": 5.526106309890747,
      "epoch": 4.675312199807877,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030151485615107073,
      "loss": 4.6088,
      "mean_token_accuracy": 0.2690506249666214,
      "num_tokens": 111624083.0,
      "step": 48670
    },
    {
      "entropy": 5.488324213027954,
      "epoch": 4.675792507204611,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030148081117081836,
      "loss": 4.6588,
      "mean_token_accuracy": 0.26316955387592317,
      "num_tokens": 111634939.0,
      "step": 48675
    },
    {
      "entropy": 5.574198770523071,
      "epoch": 4.676272814601345,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030144676557576026,
      "loss": 4.7576,
      "mean_token_accuracy": 0.2504470646381378,
      "num_tokens": 111646434.0,
      "step": 48680
    },
    {
      "entropy": 5.631341361999512,
      "epoch": 4.676753121998079,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003014127193666868,
      "loss": 4.8357,
      "mean_token_accuracy": 0.24507858157157897,
      "num_tokens": 111658266.0,
      "step": 48685
    },
    {
      "entropy": 5.4889263153076175,
      "epoch": 4.677233429394812,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003013786725443884,
      "loss": 4.6194,
      "mean_token_accuracy": 0.2630831807851791,
      "num_tokens": 111669261.0,
      "step": 48690
    },
    {
      "entropy": 5.551577568054199,
      "epoch": 4.677713736791547,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003013446251096555,
      "loss": 4.7771,
      "mean_token_accuracy": 0.2547946870326996,
      "num_tokens": 111681444.0,
      "step": 48695
    },
    {
      "entropy": 5.482543420791626,
      "epoch": 4.678194044188281,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030131057706327867,
      "loss": 4.6096,
      "mean_token_accuracy": 0.2608513742685318,
      "num_tokens": 111693461.0,
      "step": 48700
    },
    {
      "entropy": 5.483924722671508,
      "epoch": 4.678674351585014,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003012765284060484,
      "loss": 4.7391,
      "mean_token_accuracy": 0.2562659800052643,
      "num_tokens": 111705977.0,
      "step": 48705
    },
    {
      "entropy": 5.556692695617675,
      "epoch": 4.679154658981748,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003012424791387551,
      "loss": 4.749,
      "mean_token_accuracy": 0.2548192232847214,
      "num_tokens": 111717534.0,
      "step": 48710
    },
    {
      "entropy": 5.564305925369263,
      "epoch": 4.679634966378482,
      "grad_norm": 1.25,
      "learning_rate": 0.00030120842926218945,
      "loss": 4.712,
      "mean_token_accuracy": 0.25111387819051745,
      "num_tokens": 111728552.0,
      "step": 48715
    },
    {
      "entropy": 5.512849283218384,
      "epoch": 4.6801152737752165,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030117437877714184,
      "loss": 4.6949,
      "mean_token_accuracy": 0.25393199920654297,
      "num_tokens": 111740759.0,
      "step": 48720
    },
    {
      "entropy": 5.564038467407227,
      "epoch": 4.68059558117195,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030114032768440287,
      "loss": 4.7301,
      "mean_token_accuracy": 0.2549808919429779,
      "num_tokens": 111752045.0,
      "step": 48725
    },
    {
      "entropy": 5.60195894241333,
      "epoch": 4.681075888568684,
      "grad_norm": 1.375,
      "learning_rate": 0.0003011062759847631,
      "loss": 4.745,
      "mean_token_accuracy": 0.25244194120168684,
      "num_tokens": 111762624.0,
      "step": 48730
    },
    {
      "entropy": 5.511817216873169,
      "epoch": 4.681556195965418,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030107222367901313,
      "loss": 4.6769,
      "mean_token_accuracy": 0.2604965537786484,
      "num_tokens": 111774253.0,
      "step": 48735
    },
    {
      "entropy": 5.512941694259643,
      "epoch": 4.682036503362152,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030103817076794357,
      "loss": 4.7096,
      "mean_token_accuracy": 0.25500820726156237,
      "num_tokens": 111785528.0,
      "step": 48740
    },
    {
      "entropy": 5.482176303863525,
      "epoch": 4.6825168107588855,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030100411725234505,
      "loss": 4.6939,
      "mean_token_accuracy": 0.2544483780860901,
      "num_tokens": 111795692.0,
      "step": 48745
    },
    {
      "entropy": 5.486674785614014,
      "epoch": 4.68299711815562,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000300970063133008,
      "loss": 4.6174,
      "mean_token_accuracy": 0.26800805181264875,
      "num_tokens": 111807056.0,
      "step": 48750
    },
    {
      "entropy": 5.528381109237671,
      "epoch": 4.683477425552353,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030093600841072326,
      "loss": 4.6686,
      "mean_token_accuracy": 0.25898256301879885,
      "num_tokens": 111817976.0,
      "step": 48755
    },
    {
      "entropy": 5.565084981918335,
      "epoch": 4.683957732949088,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030090195308628157,
      "loss": 4.6703,
      "mean_token_accuracy": 0.25994457602500914,
      "num_tokens": 111829693.0,
      "step": 48760
    },
    {
      "entropy": 5.495015287399292,
      "epoch": 4.684438040345821,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003008678971604732,
      "loss": 4.734,
      "mean_token_accuracy": 0.24894391000270844,
      "num_tokens": 111842162.0,
      "step": 48765
    },
    {
      "entropy": 5.496126699447632,
      "epoch": 4.684918347742555,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003008338406340892,
      "loss": 4.6702,
      "mean_token_accuracy": 0.2621521756052971,
      "num_tokens": 111853176.0,
      "step": 48770
    },
    {
      "entropy": 5.508040428161621,
      "epoch": 4.685398655139289,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030079978350792014,
      "loss": 4.6806,
      "mean_token_accuracy": 0.2595526874065399,
      "num_tokens": 111865154.0,
      "step": 48775
    },
    {
      "entropy": 5.577926778793335,
      "epoch": 4.685878962536023,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003007657257827567,
      "loss": 4.8029,
      "mean_token_accuracy": 0.24502019137144088,
      "num_tokens": 111876429.0,
      "step": 48780
    },
    {
      "entropy": 5.514642810821533,
      "epoch": 4.6863592699327565,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030073166745938956,
      "loss": 4.6626,
      "mean_token_accuracy": 0.26526148021221163,
      "num_tokens": 111886623.0,
      "step": 48785
    },
    {
      "entropy": 5.480263090133667,
      "epoch": 4.686839577329491,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030069760853860964,
      "loss": 4.6623,
      "mean_token_accuracy": 0.260053439438343,
      "num_tokens": 111897137.0,
      "step": 48790
    },
    {
      "entropy": 5.6201399803161625,
      "epoch": 4.687319884726225,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003006635490212074,
      "loss": 4.7818,
      "mean_token_accuracy": 0.2550304889678955,
      "num_tokens": 111907535.0,
      "step": 48795
    },
    {
      "entropy": 5.507296323776245,
      "epoch": 4.687800192122959,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030062948890797385,
      "loss": 4.6142,
      "mean_token_accuracy": 0.26123653501272204,
      "num_tokens": 111919519.0,
      "step": 48800
    },
    {
      "entropy": 5.552355718612671,
      "epoch": 4.688280499519693,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003005954281996997,
      "loss": 4.7408,
      "mean_token_accuracy": 0.24931202083826065,
      "num_tokens": 111931308.0,
      "step": 48805
    },
    {
      "entropy": 5.536308813095093,
      "epoch": 4.688760806916426,
      "grad_norm": 1.375,
      "learning_rate": 0.00030056136689717553,
      "loss": 4.6948,
      "mean_token_accuracy": 0.2501939028501511,
      "num_tokens": 111942804.0,
      "step": 48810
    },
    {
      "entropy": 5.550281000137329,
      "epoch": 4.689241114313161,
      "grad_norm": 1.25,
      "learning_rate": 0.0003005273050011925,
      "loss": 4.7188,
      "mean_token_accuracy": 0.2574050098657608,
      "num_tokens": 111953725.0,
      "step": 48815
    },
    {
      "entropy": 5.622869968414307,
      "epoch": 4.689721421709894,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030049324251254123,
      "loss": 4.7619,
      "mean_token_accuracy": 0.25205001831054685,
      "num_tokens": 111966342.0,
      "step": 48820
    },
    {
      "entropy": 5.539585876464844,
      "epoch": 4.6902017291066285,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030045917943201253,
      "loss": 4.6745,
      "mean_token_accuracy": 0.260216598212719,
      "num_tokens": 111978120.0,
      "step": 48825
    },
    {
      "entropy": 5.506312847137451,
      "epoch": 4.690682036503362,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030042511576039725,
      "loss": 4.6351,
      "mean_token_accuracy": 0.2624510571360588,
      "num_tokens": 111989168.0,
      "step": 48830
    },
    {
      "entropy": 5.49687442779541,
      "epoch": 4.691162343900096,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030039105149848636,
      "loss": 4.6446,
      "mean_token_accuracy": 0.26483620405197145,
      "num_tokens": 112000205.0,
      "step": 48835
    },
    {
      "entropy": 5.576359558105469,
      "epoch": 4.69164265129683,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003003569866470706,
      "loss": 4.6891,
      "mean_token_accuracy": 0.25691736936569215,
      "num_tokens": 112012065.0,
      "step": 48840
    },
    {
      "entropy": 5.61665358543396,
      "epoch": 4.692122958693564,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030032292120694086,
      "loss": 4.7452,
      "mean_token_accuracy": 0.2557525023818016,
      "num_tokens": 112023809.0,
      "step": 48845
    },
    {
      "entropy": 5.526021432876587,
      "epoch": 4.692603266090297,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003002888551788882,
      "loss": 4.7004,
      "mean_token_accuracy": 0.25031147450208663,
      "num_tokens": 112035683.0,
      "step": 48850
    },
    {
      "entropy": 5.540627479553223,
      "epoch": 4.693083573487032,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030025478856370334,
      "loss": 4.6669,
      "mean_token_accuracy": 0.25090662837028505,
      "num_tokens": 112047015.0,
      "step": 48855
    },
    {
      "entropy": 5.529377603530884,
      "epoch": 4.693563880883765,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003002207213621773,
      "loss": 4.6581,
      "mean_token_accuracy": 0.25676982998847964,
      "num_tokens": 112057315.0,
      "step": 48860
    },
    {
      "entropy": 5.410574626922608,
      "epoch": 4.6940441882804995,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000300186653575101,
      "loss": 4.5522,
      "mean_token_accuracy": 0.26578310430049895,
      "num_tokens": 112068580.0,
      "step": 48865
    },
    {
      "entropy": 5.497357368469238,
      "epoch": 4.694524495677234,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003001525852032654,
      "loss": 4.654,
      "mean_token_accuracy": 0.26446128636598587,
      "num_tokens": 112079104.0,
      "step": 48870
    },
    {
      "entropy": 5.462286710739136,
      "epoch": 4.695004803073967,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003001185162474614,
      "loss": 4.5664,
      "mean_token_accuracy": 0.2689906179904938,
      "num_tokens": 112090228.0,
      "step": 48875
    },
    {
      "entropy": 5.526598405838013,
      "epoch": 4.695485110470702,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003000844467084801,
      "loss": 4.6726,
      "mean_token_accuracy": 0.2634465605020523,
      "num_tokens": 112101540.0,
      "step": 48880
    },
    {
      "entropy": 5.571476459503174,
      "epoch": 4.695965417867435,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003000503765871124,
      "loss": 4.7332,
      "mean_token_accuracy": 0.25890509635210035,
      "num_tokens": 112112998.0,
      "step": 48885
    },
    {
      "entropy": 5.554394960403442,
      "epoch": 4.696445725264169,
      "grad_norm": 1.40625,
      "learning_rate": 0.00030001630588414934,
      "loss": 4.7435,
      "mean_token_accuracy": 0.252851502597332,
      "num_tokens": 112125032.0,
      "step": 48890
    },
    {
      "entropy": 5.5861584663391115,
      "epoch": 4.696926032660903,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029998223460038205,
      "loss": 4.7292,
      "mean_token_accuracy": 0.2512446895241737,
      "num_tokens": 112137364.0,
      "step": 48895
    },
    {
      "entropy": 5.517335224151611,
      "epoch": 4.697406340057637,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002999481627366013,
      "loss": 4.6049,
      "mean_token_accuracy": 0.26435559391975405,
      "num_tokens": 112148103.0,
      "step": 48900
    },
    {
      "entropy": 5.528702592849731,
      "epoch": 4.697886647454371,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029991409029359837,
      "loss": 4.6992,
      "mean_token_accuracy": 0.2596530348062515,
      "num_tokens": 112159976.0,
      "step": 48905
    },
    {
      "entropy": 5.464732599258423,
      "epoch": 4.698366954851105,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029988001727216426,
      "loss": 4.6234,
      "mean_token_accuracy": 0.2623106986284256,
      "num_tokens": 112170839.0,
      "step": 48910
    },
    {
      "entropy": 5.441685771942138,
      "epoch": 4.698847262247838,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029984594367309,
      "loss": 4.5702,
      "mean_token_accuracy": 0.2659035712480545,
      "num_tokens": 112182195.0,
      "step": 48915
    },
    {
      "entropy": 5.538768863677978,
      "epoch": 4.699327569644573,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029981186949716663,
      "loss": 4.7479,
      "mean_token_accuracy": 0.2538313552737236,
      "num_tokens": 112193457.0,
      "step": 48920
    },
    {
      "entropy": 5.496483659744262,
      "epoch": 4.699807877041306,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029977779474518545,
      "loss": 4.675,
      "mean_token_accuracy": 0.258239883184433,
      "num_tokens": 112204665.0,
      "step": 48925
    },
    {
      "entropy": 5.620136499404907,
      "epoch": 4.70028818443804,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029974371941793743,
      "loss": 4.8064,
      "mean_token_accuracy": 0.253908708691597,
      "num_tokens": 112217381.0,
      "step": 48930
    },
    {
      "entropy": 5.554287147521973,
      "epoch": 4.700768491834774,
      "grad_norm": 1.125,
      "learning_rate": 0.0002997096435162137,
      "loss": 4.6463,
      "mean_token_accuracy": 0.26224340349435804,
      "num_tokens": 112228096.0,
      "step": 48935
    },
    {
      "entropy": 5.462896680831909,
      "epoch": 4.701248799231508,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002996755670408053,
      "loss": 4.6755,
      "mean_token_accuracy": 0.24748289138078688,
      "num_tokens": 112238974.0,
      "step": 48940
    },
    {
      "entropy": 5.526189708709717,
      "epoch": 4.7017291066282425,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029964148999250367,
      "loss": 4.7416,
      "mean_token_accuracy": 0.24741180390119552,
      "num_tokens": 112250022.0,
      "step": 48945
    },
    {
      "entropy": 5.509835910797119,
      "epoch": 4.702209414024976,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029960741237209976,
      "loss": 4.6963,
      "mean_token_accuracy": 0.2478325143456459,
      "num_tokens": 112260654.0,
      "step": 48950
    },
    {
      "entropy": 5.551683473587036,
      "epoch": 4.70268972142171,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002995733341803848,
      "loss": 4.7164,
      "mean_token_accuracy": 0.2528392493724823,
      "num_tokens": 112271864.0,
      "step": 48955
    },
    {
      "entropy": 5.552858352661133,
      "epoch": 4.703170028818444,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00029953925541814993,
      "loss": 4.7196,
      "mean_token_accuracy": 0.25396152436733244,
      "num_tokens": 112282907.0,
      "step": 48960
    },
    {
      "entropy": 5.578418779373169,
      "epoch": 4.703650336215178,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029950517608618647,
      "loss": 4.7608,
      "mean_token_accuracy": 0.2401827946305275,
      "num_tokens": 112294887.0,
      "step": 48965
    },
    {
      "entropy": 5.485395431518555,
      "epoch": 4.7041306436119115,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029947109618528564,
      "loss": 4.6977,
      "mean_token_accuracy": 0.2499213397502899,
      "num_tokens": 112305401.0,
      "step": 48970
    },
    {
      "entropy": 5.484477186203003,
      "epoch": 4.704610951008646,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029943701571623853,
      "loss": 4.6645,
      "mean_token_accuracy": 0.26305621713399885,
      "num_tokens": 112316692.0,
      "step": 48975
    },
    {
      "entropy": 5.606156349182129,
      "epoch": 4.705091258405379,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002994029346798365,
      "loss": 4.7867,
      "mean_token_accuracy": 0.24035360366106034,
      "num_tokens": 112328864.0,
      "step": 48980
    },
    {
      "entropy": 5.587063884735107,
      "epoch": 4.7055715658021136,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029936885307687086,
      "loss": 4.7471,
      "mean_token_accuracy": 0.24984240233898164,
      "num_tokens": 112340308.0,
      "step": 48985
    },
    {
      "entropy": 5.570561361312866,
      "epoch": 4.706051873198847,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002993347709081327,
      "loss": 4.6992,
      "mean_token_accuracy": 0.2519601210951805,
      "num_tokens": 112352044.0,
      "step": 48990
    },
    {
      "entropy": 5.503772354125976,
      "epoch": 4.706532180595581,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029930068817441346,
      "loss": 4.6914,
      "mean_token_accuracy": 0.25846275240182875,
      "num_tokens": 112363928.0,
      "step": 48995
    },
    {
      "entropy": 5.5338115215301515,
      "epoch": 4.707012487992315,
      "grad_norm": 1.25,
      "learning_rate": 0.0002992666048765045,
      "loss": 4.6781,
      "mean_token_accuracy": 0.2554185092449188,
      "num_tokens": 112375034.0,
      "step": 49000
    },
    {
      "entropy": 5.537279367446899,
      "epoch": 4.707492795389049,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029923252101519694,
      "loss": 4.7254,
      "mean_token_accuracy": 0.2595970258116722,
      "num_tokens": 112386984.0,
      "step": 49005
    },
    {
      "entropy": 5.585250759124756,
      "epoch": 4.7079731027857825,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029919843659128215,
      "loss": 4.6937,
      "mean_token_accuracy": 0.2546818494796753,
      "num_tokens": 112398207.0,
      "step": 49010
    },
    {
      "entropy": 5.595265197753906,
      "epoch": 4.708453410182517,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002991643516055516,
      "loss": 4.7649,
      "mean_token_accuracy": 0.2517458856105804,
      "num_tokens": 112408793.0,
      "step": 49015
    },
    {
      "entropy": 5.548152065277099,
      "epoch": 4.708933717579251,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002991302660587965,
      "loss": 4.7336,
      "mean_token_accuracy": 0.2499803841114044,
      "num_tokens": 112420738.0,
      "step": 49020
    },
    {
      "entropy": 5.562401628494262,
      "epoch": 4.709414024975985,
      "grad_norm": 1.375,
      "learning_rate": 0.00029909617995180833,
      "loss": 4.6453,
      "mean_token_accuracy": 0.25695914775133133,
      "num_tokens": 112432279.0,
      "step": 49025
    },
    {
      "entropy": 5.536653852462768,
      "epoch": 4.709894332372719,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029906209328537845,
      "loss": 4.7948,
      "mean_token_accuracy": 0.2453932136297226,
      "num_tokens": 112444826.0,
      "step": 49030
    },
    {
      "entropy": 5.501629447937011,
      "epoch": 4.710374639769452,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002990280060602981,
      "loss": 4.6723,
      "mean_token_accuracy": 0.2608791649341583,
      "num_tokens": 112457250.0,
      "step": 49035
    },
    {
      "entropy": 5.537286424636841,
      "epoch": 4.710854947166187,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029899391827735883,
      "loss": 4.7243,
      "mean_token_accuracy": 0.2580835849046707,
      "num_tokens": 112468657.0,
      "step": 49040
    },
    {
      "entropy": 5.633216857910156,
      "epoch": 4.71133525456292,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002989598299373521,
      "loss": 4.7776,
      "mean_token_accuracy": 0.25025725960731504,
      "num_tokens": 112481641.0,
      "step": 49045
    },
    {
      "entropy": 5.558393049240112,
      "epoch": 4.7118155619596545,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002989257410410692,
      "loss": 4.6914,
      "mean_token_accuracy": 0.2536150857806206,
      "num_tokens": 112492939.0,
      "step": 49050
    },
    {
      "entropy": 5.622831392288208,
      "epoch": 4.712295869356388,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002988916515893016,
      "loss": 4.795,
      "mean_token_accuracy": 0.2475050151348114,
      "num_tokens": 112504809.0,
      "step": 49055
    },
    {
      "entropy": 5.477626943588257,
      "epoch": 4.712776176753122,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029885756158284086,
      "loss": 4.5886,
      "mean_token_accuracy": 0.2722963184118271,
      "num_tokens": 112515955.0,
      "step": 49060
    },
    {
      "entropy": 5.510362672805786,
      "epoch": 4.713256484149856,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002988234710224784,
      "loss": 4.6145,
      "mean_token_accuracy": 0.2655546277761459,
      "num_tokens": 112527441.0,
      "step": 49065
    },
    {
      "entropy": 5.647089099884033,
      "epoch": 4.71373679154659,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00029878937990900556,
      "loss": 4.776,
      "mean_token_accuracy": 0.24968857020139695,
      "num_tokens": 112536688.0,
      "step": 49070
    },
    {
      "entropy": 5.536451196670532,
      "epoch": 4.714217098943323,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002987552882432141,
      "loss": 4.6738,
      "mean_token_accuracy": 0.25716388672590257,
      "num_tokens": 112548281.0,
      "step": 49075
    },
    {
      "entropy": 5.467496538162232,
      "epoch": 4.714697406340058,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002987211960258954,
      "loss": 4.6106,
      "mean_token_accuracy": 0.25962152928113935,
      "num_tokens": 112560284.0,
      "step": 49080
    },
    {
      "entropy": 5.437899351119995,
      "epoch": 4.715177713736791,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002986871032578408,
      "loss": 4.6713,
      "mean_token_accuracy": 0.25958692133426664,
      "num_tokens": 112570805.0,
      "step": 49085
    },
    {
      "entropy": 5.486164999008179,
      "epoch": 4.7156580211335255,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029865300993984215,
      "loss": 4.6815,
      "mean_token_accuracy": 0.2566055357456207,
      "num_tokens": 112582983.0,
      "step": 49090
    },
    {
      "entropy": 5.527470636367798,
      "epoch": 4.71613832853026,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029861891607269075,
      "loss": 4.6634,
      "mean_token_accuracy": 0.25184299051761627,
      "num_tokens": 112592801.0,
      "step": 49095
    },
    {
      "entropy": 5.568046426773071,
      "epoch": 4.716618635926993,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029858482165717824,
      "loss": 4.743,
      "mean_token_accuracy": 0.24802331924438475,
      "num_tokens": 112603133.0,
      "step": 49100
    },
    {
      "entropy": 5.543206739425659,
      "epoch": 4.717098943323727,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029855072669409636,
      "loss": 4.6576,
      "mean_token_accuracy": 0.25365490168333055,
      "num_tokens": 112614408.0,
      "step": 49105
    },
    {
      "entropy": 5.5376263618469235,
      "epoch": 4.717579250720461,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029851663118423647,
      "loss": 4.6652,
      "mean_token_accuracy": 0.25586207807064054,
      "num_tokens": 112627112.0,
      "step": 49110
    },
    {
      "entropy": 5.603524208068848,
      "epoch": 4.718059558117195,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029848253512839015,
      "loss": 4.7226,
      "mean_token_accuracy": 0.2557846412062645,
      "num_tokens": 112638579.0,
      "step": 49115
    },
    {
      "entropy": 5.522995138168335,
      "epoch": 4.718539865513929,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029844843852734913,
      "loss": 4.6829,
      "mean_token_accuracy": 0.2611123412847519,
      "num_tokens": 112650032.0,
      "step": 49120
    },
    {
      "entropy": 5.577251768112182,
      "epoch": 4.719020172910663,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000298414341381905,
      "loss": 4.7189,
      "mean_token_accuracy": 0.24955229014158248,
      "num_tokens": 112661287.0,
      "step": 49125
    },
    {
      "entropy": 5.534765625,
      "epoch": 4.7195004803073966,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002983802436928493,
      "loss": 4.6774,
      "mean_token_accuracy": 0.2626756697893143,
      "num_tokens": 112671533.0,
      "step": 49130
    },
    {
      "entropy": 5.551130151748657,
      "epoch": 4.719980787704131,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029834614546097395,
      "loss": 4.6303,
      "mean_token_accuracy": 0.2579751372337341,
      "num_tokens": 112682684.0,
      "step": 49135
    },
    {
      "entropy": 5.570973539352417,
      "epoch": 4.720461095100864,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029831204668707035,
      "loss": 4.7093,
      "mean_token_accuracy": 0.2602310344576836,
      "num_tokens": 112693308.0,
      "step": 49140
    },
    {
      "entropy": 5.576630926132202,
      "epoch": 4.720941402497599,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029827794737193026,
      "loss": 4.781,
      "mean_token_accuracy": 0.24589239060878754,
      "num_tokens": 112704757.0,
      "step": 49145
    },
    {
      "entropy": 5.554714679718018,
      "epoch": 4.721421709894332,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002982438475163453,
      "loss": 4.7013,
      "mean_token_accuracy": 0.25798326581716535,
      "num_tokens": 112715347.0,
      "step": 49150
    },
    {
      "entropy": 5.475700235366821,
      "epoch": 4.721902017291066,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029820974712110724,
      "loss": 4.5829,
      "mean_token_accuracy": 0.264636792242527,
      "num_tokens": 112727590.0,
      "step": 49155
    },
    {
      "entropy": 5.533917474746704,
      "epoch": 4.7223823246878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002981756461870078,
      "loss": 4.6761,
      "mean_token_accuracy": 0.26362662613391874,
      "num_tokens": 112738730.0,
      "step": 49160
    },
    {
      "entropy": 5.6045334815979,
      "epoch": 4.722862632084534,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002981415447148387,
      "loss": 4.7586,
      "mean_token_accuracy": 0.25815286934375764,
      "num_tokens": 112750177.0,
      "step": 49165
    },
    {
      "entropy": 5.488722467422486,
      "epoch": 4.7233429394812685,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002981074427053916,
      "loss": 4.6531,
      "mean_token_accuracy": 0.26353961527347564,
      "num_tokens": 112762436.0,
      "step": 49170
    },
    {
      "entropy": 5.538221263885498,
      "epoch": 4.723823246878002,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029807334015945826,
      "loss": 4.663,
      "mean_token_accuracy": 0.2541380599141121,
      "num_tokens": 112774360.0,
      "step": 49175
    },
    {
      "entropy": 5.441414260864258,
      "epoch": 4.724303554274735,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029803923707783056,
      "loss": 4.5731,
      "mean_token_accuracy": 0.26665245890617373,
      "num_tokens": 112786196.0,
      "step": 49180
    },
    {
      "entropy": 5.496576118469238,
      "epoch": 4.72478386167147,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029800513346130007,
      "loss": 4.7208,
      "mean_token_accuracy": 0.25646141171455383,
      "num_tokens": 112797080.0,
      "step": 49185
    },
    {
      "entropy": 5.555210018157959,
      "epoch": 4.725264169068204,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002979710293106588,
      "loss": 4.7073,
      "mean_token_accuracy": 0.26625740379095075,
      "num_tokens": 112809012.0,
      "step": 49190
    },
    {
      "entropy": 5.530941915512085,
      "epoch": 4.7257444764649374,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002979369246266984,
      "loss": 4.6831,
      "mean_token_accuracy": 0.25339420288801195,
      "num_tokens": 112821037.0,
      "step": 49195
    },
    {
      "entropy": 5.5187300682067875,
      "epoch": 4.726224783861672,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002979028194102107,
      "loss": 4.687,
      "mean_token_accuracy": 0.255971172451973,
      "num_tokens": 112832802.0,
      "step": 49200
    },
    {
      "entropy": 5.554801511764526,
      "epoch": 4.726705091258405,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029786871366198754,
      "loss": 4.6851,
      "mean_token_accuracy": 0.2596752405166626,
      "num_tokens": 112843235.0,
      "step": 49205
    },
    {
      "entropy": 5.552907419204712,
      "epoch": 4.7271853986551395,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002978346073828208,
      "loss": 4.7295,
      "mean_token_accuracy": 0.2583544045686722,
      "num_tokens": 112853751.0,
      "step": 49210
    },
    {
      "entropy": 5.591862487792969,
      "epoch": 4.727665706051873,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002978005005735023,
      "loss": 4.8088,
      "mean_token_accuracy": 0.2428588718175888,
      "num_tokens": 112864247.0,
      "step": 49215
    },
    {
      "entropy": 5.54439148902893,
      "epoch": 4.728146013448607,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002977663932348238,
      "loss": 4.7059,
      "mean_token_accuracy": 0.2510977819561958,
      "num_tokens": 112875667.0,
      "step": 49220
    },
    {
      "entropy": 5.577328491210937,
      "epoch": 4.728626320845341,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029773228536757727,
      "loss": 4.7135,
      "mean_token_accuracy": 0.26075640618801116,
      "num_tokens": 112887805.0,
      "step": 49225
    },
    {
      "entropy": 5.597927761077881,
      "epoch": 4.729106628242075,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029769817697255455,
      "loss": 4.7799,
      "mean_token_accuracy": 0.2504475265741348,
      "num_tokens": 112899515.0,
      "step": 49230
    },
    {
      "entropy": 5.527606201171875,
      "epoch": 4.7295869356388085,
      "grad_norm": 1.25,
      "learning_rate": 0.0002976640680505476,
      "loss": 4.6562,
      "mean_token_accuracy": 0.26085434556007386,
      "num_tokens": 112910701.0,
      "step": 49235
    },
    {
      "entropy": 5.452523899078369,
      "epoch": 4.730067243035543,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002976299586023483,
      "loss": 4.5356,
      "mean_token_accuracy": 0.27346600592136383,
      "num_tokens": 112921681.0,
      "step": 49240
    },
    {
      "entropy": 5.490186023712158,
      "epoch": 4.730547550432277,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029759584862874854,
      "loss": 4.6365,
      "mean_token_accuracy": 0.26371408849954603,
      "num_tokens": 112933199.0,
      "step": 49245
    },
    {
      "entropy": 5.497173738479614,
      "epoch": 4.731027857829011,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002975617381305403,
      "loss": 4.7125,
      "mean_token_accuracy": 0.2557569816708565,
      "num_tokens": 112946087.0,
      "step": 49250
    },
    {
      "entropy": 5.4554678440094,
      "epoch": 4.731508165225744,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002975276271085156,
      "loss": 4.5971,
      "mean_token_accuracy": 0.2699870079755783,
      "num_tokens": 112958347.0,
      "step": 49255
    },
    {
      "entropy": 5.5558733463287355,
      "epoch": 4.731988472622478,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029749351556346607,
      "loss": 4.6045,
      "mean_token_accuracy": 0.2586759805679321,
      "num_tokens": 112969105.0,
      "step": 49260
    },
    {
      "entropy": 5.4568315029144285,
      "epoch": 4.732468780019213,
      "grad_norm": 1.21875,
      "learning_rate": 0.000297459403496184,
      "loss": 4.6112,
      "mean_token_accuracy": 0.2624738872051239,
      "num_tokens": 112981031.0,
      "step": 49265
    },
    {
      "entropy": 5.477939558029175,
      "epoch": 4.732949087415946,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002974252909074613,
      "loss": 4.601,
      "mean_token_accuracy": 0.26660146117210387,
      "num_tokens": 112993418.0,
      "step": 49270
    },
    {
      "entropy": 5.48498306274414,
      "epoch": 4.73342939481268,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002973911777980899,
      "loss": 4.6426,
      "mean_token_accuracy": 0.26274425983428956,
      "num_tokens": 113004858.0,
      "step": 49275
    },
    {
      "entropy": 5.5645617008209225,
      "epoch": 4.733909702209414,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002973570641688618,
      "loss": 4.7557,
      "mean_token_accuracy": 0.2546677738428116,
      "num_tokens": 113017086.0,
      "step": 49280
    },
    {
      "entropy": 5.589614009857177,
      "epoch": 4.734390009606148,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029732295002056914,
      "loss": 4.7237,
      "mean_token_accuracy": 0.25346164405345917,
      "num_tokens": 113028802.0,
      "step": 49285
    },
    {
      "entropy": 5.609011936187744,
      "epoch": 4.734870317002882,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002972888353540038,
      "loss": 4.7522,
      "mean_token_accuracy": 0.2519893363118172,
      "num_tokens": 113042352.0,
      "step": 49290
    },
    {
      "entropy": 5.575958204269409,
      "epoch": 4.735350624399616,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002972547201699579,
      "loss": 4.7111,
      "mean_token_accuracy": 0.2618120491504669,
      "num_tokens": 113053262.0,
      "step": 49295
    },
    {
      "entropy": 5.479461574554444,
      "epoch": 4.735830931796349,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002972206044692235,
      "loss": 4.6262,
      "mean_token_accuracy": 0.25774873048067093,
      "num_tokens": 113064618.0,
      "step": 49300
    },
    {
      "entropy": 5.456086874008179,
      "epoch": 4.736311239193084,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029718648825259266,
      "loss": 4.5585,
      "mean_token_accuracy": 0.2673366144299507,
      "num_tokens": 113076358.0,
      "step": 49305
    },
    {
      "entropy": 5.5939866065979,
      "epoch": 4.736791546589817,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029715237152085744,
      "loss": 4.7391,
      "mean_token_accuracy": 0.25206594467163085,
      "num_tokens": 113086307.0,
      "step": 49310
    },
    {
      "entropy": 5.533039808273315,
      "epoch": 4.7372718539865515,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029711825427480994,
      "loss": 4.664,
      "mean_token_accuracy": 0.2624142363667488,
      "num_tokens": 113097240.0,
      "step": 49315
    },
    {
      "entropy": 5.496601200103759,
      "epoch": 4.737752161383285,
      "grad_norm": 1.375,
      "learning_rate": 0.00029708413651524223,
      "loss": 4.7018,
      "mean_token_accuracy": 0.2564418762922287,
      "num_tokens": 113107745.0,
      "step": 49320
    },
    {
      "entropy": 5.525255870819092,
      "epoch": 4.738232468780019,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029705001824294636,
      "loss": 4.6952,
      "mean_token_accuracy": 0.2541597917675972,
      "num_tokens": 113118346.0,
      "step": 49325
    },
    {
      "entropy": 5.50911865234375,
      "epoch": 4.738712776176753,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029701589945871473,
      "loss": 4.6587,
      "mean_token_accuracy": 0.2668694883584976,
      "num_tokens": 113130610.0,
      "step": 49330
    },
    {
      "entropy": 5.5350377559661865,
      "epoch": 4.739193083573487,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029698178016333913,
      "loss": 4.6778,
      "mean_token_accuracy": 0.2554922342300415,
      "num_tokens": 113142926.0,
      "step": 49335
    },
    {
      "entropy": 5.589195489883423,
      "epoch": 4.739673390970221,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002969476603576119,
      "loss": 4.7185,
      "mean_token_accuracy": 0.24883398562669753,
      "num_tokens": 113155754.0,
      "step": 49340
    },
    {
      "entropy": 5.542986297607422,
      "epoch": 4.740153698366955,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029691354004232525,
      "loss": 4.7396,
      "mean_token_accuracy": 0.2511470064520836,
      "num_tokens": 113167733.0,
      "step": 49345
    },
    {
      "entropy": 5.54359188079834,
      "epoch": 4.740634005763689,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002968794192182712,
      "loss": 4.727,
      "mean_token_accuracy": 0.2541450932621956,
      "num_tokens": 113180469.0,
      "step": 49350
    },
    {
      "entropy": 5.661234188079834,
      "epoch": 4.7411143131604225,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029684529788624205,
      "loss": 4.7701,
      "mean_token_accuracy": 0.24897526502609252,
      "num_tokens": 113191238.0,
      "step": 49355
    },
    {
      "entropy": 5.5442211627960205,
      "epoch": 4.741594620557157,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002968111760470299,
      "loss": 4.7008,
      "mean_token_accuracy": 0.25241912305355074,
      "num_tokens": 113202912.0,
      "step": 49360
    },
    {
      "entropy": 5.557088422775268,
      "epoch": 4.74207492795389,
      "grad_norm": 1.3125,
      "learning_rate": 0.000296777053701427,
      "loss": 4.7307,
      "mean_token_accuracy": 0.25303651094436647,
      "num_tokens": 113215062.0,
      "step": 49365
    },
    {
      "entropy": 5.573876571655274,
      "epoch": 4.742555235350625,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002967429308502256,
      "loss": 4.7482,
      "mean_token_accuracy": 0.2479296386241913,
      "num_tokens": 113225422.0,
      "step": 49370
    },
    {
      "entropy": 5.607876300811768,
      "epoch": 4.743035542747358,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002967088074942179,
      "loss": 4.7922,
      "mean_token_accuracy": 0.2586304023861885,
      "num_tokens": 113237197.0,
      "step": 49375
    },
    {
      "entropy": 5.523729419708252,
      "epoch": 4.743515850144092,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002966746836341962,
      "loss": 4.6593,
      "mean_token_accuracy": 0.2581161022186279,
      "num_tokens": 113248015.0,
      "step": 49380
    },
    {
      "entropy": 5.49337854385376,
      "epoch": 4.743996157540826,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002966405592709526,
      "loss": 4.6915,
      "mean_token_accuracy": 0.2505333170294762,
      "num_tokens": 113259760.0,
      "step": 49385
    },
    {
      "entropy": 5.5498298645019535,
      "epoch": 4.74447646493756,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029660643440527964,
      "loss": 4.6722,
      "mean_token_accuracy": 0.2518792673945427,
      "num_tokens": 113270826.0,
      "step": 49390
    },
    {
      "entropy": 5.533018636703491,
      "epoch": 4.744956772334294,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029657230903796925,
      "loss": 4.6861,
      "mean_token_accuracy": 0.2640738278627396,
      "num_tokens": 113283078.0,
      "step": 49395
    },
    {
      "entropy": 5.535366725921631,
      "epoch": 4.745437079731028,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000296538183169814,
      "loss": 4.7407,
      "mean_token_accuracy": 0.25631731301546096,
      "num_tokens": 113294893.0,
      "step": 49400
    },
    {
      "entropy": 5.536608362197876,
      "epoch": 4.745917387127761,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002965040568016061,
      "loss": 4.6498,
      "mean_token_accuracy": 0.26297094225883483,
      "num_tokens": 113306613.0,
      "step": 49405
    },
    {
      "entropy": 5.4759876251220705,
      "epoch": 4.746397694524496,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002964699299341378,
      "loss": 4.6662,
      "mean_token_accuracy": 0.25841007232666013,
      "num_tokens": 113318192.0,
      "step": 49410
    },
    {
      "entropy": 5.589792919158936,
      "epoch": 4.74687800192123,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029643580256820157,
      "loss": 4.7847,
      "mean_token_accuracy": 0.24553309381008148,
      "num_tokens": 113329105.0,
      "step": 49415
    },
    {
      "entropy": 5.580944967269898,
      "epoch": 4.747358309317963,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029640167470458957,
      "loss": 4.7788,
      "mean_token_accuracy": 0.2507141798734665,
      "num_tokens": 113340938.0,
      "step": 49420
    },
    {
      "entropy": 5.584104585647583,
      "epoch": 4.747838616714698,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002963675463440943,
      "loss": 4.6067,
      "mean_token_accuracy": 0.26158059686422347,
      "num_tokens": 113352566.0,
      "step": 49425
    },
    {
      "entropy": 5.6110388278961185,
      "epoch": 4.748318924111431,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029633341748750797,
      "loss": 4.7825,
      "mean_token_accuracy": 0.25348674356937406,
      "num_tokens": 113364246.0,
      "step": 49430
    },
    {
      "entropy": 5.574675750732422,
      "epoch": 4.7487992315081655,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029629928813562314,
      "loss": 4.7449,
      "mean_token_accuracy": 0.25426546186208726,
      "num_tokens": 113376182.0,
      "step": 49435
    },
    {
      "entropy": 5.434208345413208,
      "epoch": 4.749279538904899,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029626515828923197,
      "loss": 4.6327,
      "mean_token_accuracy": 0.26506689190864563,
      "num_tokens": 113388304.0,
      "step": 49440
    },
    {
      "entropy": 5.567615795135498,
      "epoch": 4.749759846301633,
      "grad_norm": 1.34375,
      "learning_rate": 0.000296231027949127,
      "loss": 4.7449,
      "mean_token_accuracy": 0.26279556900262835,
      "num_tokens": 113399660.0,
      "step": 49445
    },
    {
      "entropy": 5.607358837127686,
      "epoch": 4.750240153698367,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002961968971161006,
      "loss": 4.7504,
      "mean_token_accuracy": 0.25975899696350097,
      "num_tokens": 113410979.0,
      "step": 49450
    },
    {
      "entropy": 5.5146702289581295,
      "epoch": 4.750720461095101,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029616276579094524,
      "loss": 4.6215,
      "mean_token_accuracy": 0.2639973729848862,
      "num_tokens": 113422403.0,
      "step": 49455
    },
    {
      "entropy": 5.522399663925171,
      "epoch": 4.7512007684918345,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002961286339744533,
      "loss": 4.6254,
      "mean_token_accuracy": 0.2643934562802315,
      "num_tokens": 113434609.0,
      "step": 49460
    },
    {
      "entropy": 5.526288414001465,
      "epoch": 4.751681075888569,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029609450166741724,
      "loss": 4.6985,
      "mean_token_accuracy": 0.26115372478961946,
      "num_tokens": 113446518.0,
      "step": 49465
    },
    {
      "entropy": 5.499017953872681,
      "epoch": 4.752161383285302,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029606036887062936,
      "loss": 4.6808,
      "mean_token_accuracy": 0.2571819305419922,
      "num_tokens": 113457400.0,
      "step": 49470
    },
    {
      "entropy": 5.587454557418823,
      "epoch": 4.752641690682037,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029602623558488235,
      "loss": 4.7753,
      "mean_token_accuracy": 0.24927662760019303,
      "num_tokens": 113468510.0,
      "step": 49475
    },
    {
      "entropy": 5.5415833473205565,
      "epoch": 4.75312199807877,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002959921018109686,
      "loss": 4.6455,
      "mean_token_accuracy": 0.2662084326148033,
      "num_tokens": 113479767.0,
      "step": 49480
    },
    {
      "entropy": 5.45601224899292,
      "epoch": 4.753602305475504,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002959579675496805,
      "loss": 4.5732,
      "mean_token_accuracy": 0.26155356913805006,
      "num_tokens": 113490844.0,
      "step": 49485
    },
    {
      "entropy": 5.459970712661743,
      "epoch": 4.754082612872239,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002959238328018107,
      "loss": 4.6449,
      "mean_token_accuracy": 0.2596724972128868,
      "num_tokens": 113502131.0,
      "step": 49490
    },
    {
      "entropy": 5.5255921363830565,
      "epoch": 4.754562920268972,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029588969756815154,
      "loss": 4.6636,
      "mean_token_accuracy": 0.25711351335048677,
      "num_tokens": 113515178.0,
      "step": 49495
    },
    {
      "entropy": 5.520210933685303,
      "epoch": 4.755043227665706,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002958555618494958,
      "loss": 4.6558,
      "mean_token_accuracy": 0.25803935527801514,
      "num_tokens": 113526571.0,
      "step": 49500
    },
    {
      "entropy": 5.533684301376343,
      "epoch": 4.75552353506244,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029582142564663565,
      "loss": 4.6633,
      "mean_token_accuracy": 0.2575841695070267,
      "num_tokens": 113537601.0,
      "step": 49505
    },
    {
      "entropy": 5.586101913452149,
      "epoch": 4.756003842459174,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029578728896036393,
      "loss": 4.778,
      "mean_token_accuracy": 0.24205198884010315,
      "num_tokens": 113550007.0,
      "step": 49510
    },
    {
      "entropy": 5.480841159820557,
      "epoch": 4.756484149855908,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029575315179147305,
      "loss": 4.6147,
      "mean_token_accuracy": 0.26247892528772354,
      "num_tokens": 113562467.0,
      "step": 49515
    },
    {
      "entropy": 5.521850728988648,
      "epoch": 4.756964457252642,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002957190141407557,
      "loss": 4.6284,
      "mean_token_accuracy": 0.26132559925317767,
      "num_tokens": 113572580.0,
      "step": 49520
    },
    {
      "entropy": 5.580967664718628,
      "epoch": 4.757444764649375,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029568487600900426,
      "loss": 4.757,
      "mean_token_accuracy": 0.2577021285891533,
      "num_tokens": 113584577.0,
      "step": 49525
    },
    {
      "entropy": 5.523101568222046,
      "epoch": 4.75792507204611,
      "grad_norm": 1.25,
      "learning_rate": 0.00029565073739701146,
      "loss": 4.6971,
      "mean_token_accuracy": 0.25609846115112306,
      "num_tokens": 113596895.0,
      "step": 49530
    },
    {
      "entropy": 5.511553573608398,
      "epoch": 4.758405379442843,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002956165983055698,
      "loss": 4.673,
      "mean_token_accuracy": 0.25816661715507505,
      "num_tokens": 113607915.0,
      "step": 49535
    },
    {
      "entropy": 5.563381099700928,
      "epoch": 4.7588856868395775,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000295582458735472,
      "loss": 4.6798,
      "mean_token_accuracy": 0.25635210424661636,
      "num_tokens": 113620000.0,
      "step": 49540
    },
    {
      "entropy": 5.593824434280395,
      "epoch": 4.759365994236311,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029554831868751065,
      "loss": 4.7193,
      "mean_token_accuracy": 0.2630499333143234,
      "num_tokens": 113630026.0,
      "step": 49545
    },
    {
      "entropy": 5.512263059616089,
      "epoch": 4.759846301633045,
      "grad_norm": 1.25,
      "learning_rate": 0.00029551417816247834,
      "loss": 4.7302,
      "mean_token_accuracy": 0.2500907063484192,
      "num_tokens": 113642403.0,
      "step": 49550
    },
    {
      "entropy": 5.447823190689087,
      "epoch": 4.760326609029779,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029548003716116774,
      "loss": 4.5198,
      "mean_token_accuracy": 0.27486561983823776,
      "num_tokens": 113652733.0,
      "step": 49555
    },
    {
      "entropy": 5.549574899673462,
      "epoch": 4.760806916426513,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002954458956843714,
      "loss": 4.7212,
      "mean_token_accuracy": 0.25417918860912325,
      "num_tokens": 113663417.0,
      "step": 49560
    },
    {
      "entropy": 5.459422969818116,
      "epoch": 4.761287223823247,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029541175373288213,
      "loss": 4.6268,
      "mean_token_accuracy": 0.2605663165450096,
      "num_tokens": 113673852.0,
      "step": 49565
    },
    {
      "entropy": 5.588094043731689,
      "epoch": 4.761767531219981,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002953776113074926,
      "loss": 4.8257,
      "mean_token_accuracy": 0.24661738723516463,
      "num_tokens": 113686032.0,
      "step": 49570
    },
    {
      "entropy": 5.5563029766082765,
      "epoch": 4.762247838616715,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029534346840899545,
      "loss": 4.6589,
      "mean_token_accuracy": 0.26090497374534605,
      "num_tokens": 113696979.0,
      "step": 49575
    },
    {
      "entropy": 5.512016773223877,
      "epoch": 4.7627281460134485,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029530932503818333,
      "loss": 4.6917,
      "mean_token_accuracy": 0.2645789533853531,
      "num_tokens": 113709101.0,
      "step": 49580
    },
    {
      "entropy": 5.457629632949829,
      "epoch": 4.763208453410183,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029527518119584897,
      "loss": 4.6431,
      "mean_token_accuracy": 0.2676907777786255,
      "num_tokens": 113721022.0,
      "step": 49585
    },
    {
      "entropy": 5.480819797515869,
      "epoch": 4.763688760806916,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002952410368827851,
      "loss": 4.7301,
      "mean_token_accuracy": 0.25309356600046157,
      "num_tokens": 113733252.0,
      "step": 49590
    },
    {
      "entropy": 5.5798454761505125,
      "epoch": 4.764169068203651,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029520689209978446,
      "loss": 4.7467,
      "mean_token_accuracy": 0.25815181732177733,
      "num_tokens": 113744827.0,
      "step": 49595
    },
    {
      "entropy": 5.579656457901001,
      "epoch": 4.764649375600384,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029517274684763984,
      "loss": 4.79,
      "mean_token_accuracy": 0.2465256318449974,
      "num_tokens": 113756214.0,
      "step": 49600
    },
    {
      "entropy": 5.548507118225098,
      "epoch": 4.765129682997118,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002951386011271439,
      "loss": 4.7656,
      "mean_token_accuracy": 0.24630593806505202,
      "num_tokens": 113767174.0,
      "step": 49605
    },
    {
      "entropy": 5.537016534805298,
      "epoch": 4.765609990393852,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029510445493908946,
      "loss": 4.6726,
      "mean_token_accuracy": 0.2571592047810555,
      "num_tokens": 113779249.0,
      "step": 49610
    },
    {
      "entropy": 5.570648860931397,
      "epoch": 4.766090297790586,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002950703082842693,
      "loss": 4.6898,
      "mean_token_accuracy": 0.2562208130955696,
      "num_tokens": 113790737.0,
      "step": 49615
    },
    {
      "entropy": 5.537118291854858,
      "epoch": 4.76657060518732,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002950361611634761,
      "loss": 4.6819,
      "mean_token_accuracy": 0.2566343009471893,
      "num_tokens": 113802456.0,
      "step": 49620
    },
    {
      "entropy": 5.624145030975342,
      "epoch": 4.767050912584054,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002950020135775027,
      "loss": 4.7721,
      "mean_token_accuracy": 0.2529041290283203,
      "num_tokens": 113813702.0,
      "step": 49625
    },
    {
      "entropy": 5.587974596023559,
      "epoch": 4.767531219980787,
      "grad_norm": 1.515625,
      "learning_rate": 0.00029496786552714213,
      "loss": 4.713,
      "mean_token_accuracy": 0.2536953926086426,
      "num_tokens": 113823688.0,
      "step": 49630
    },
    {
      "entropy": 5.583051919937134,
      "epoch": 4.768011527377522,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002949337170131868,
      "loss": 4.7228,
      "mean_token_accuracy": 0.2529280290007591,
      "num_tokens": 113835794.0,
      "step": 49635
    },
    {
      "entropy": 5.599518585205078,
      "epoch": 4.768491834774256,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029489956803642985,
      "loss": 4.791,
      "mean_token_accuracy": 0.24717210084199906,
      "num_tokens": 113847140.0,
      "step": 49640
    },
    {
      "entropy": 5.497231388092041,
      "epoch": 4.768972142170989,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002948654185976641,
      "loss": 4.6389,
      "mean_token_accuracy": 0.2556879907846451,
      "num_tokens": 113859043.0,
      "step": 49645
    },
    {
      "entropy": 5.566942739486694,
      "epoch": 4.769452449567723,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029483126869768217,
      "loss": 4.7702,
      "mean_token_accuracy": 0.24996569752693176,
      "num_tokens": 113871098.0,
      "step": 49650
    },
    {
      "entropy": 5.496382236480713,
      "epoch": 4.769932756964457,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002947971183372771,
      "loss": 4.7078,
      "mean_token_accuracy": 0.26090003848075866,
      "num_tokens": 113882865.0,
      "step": 49655
    },
    {
      "entropy": 5.466874647140503,
      "epoch": 4.7704130643611915,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029476296751724184,
      "loss": 4.6293,
      "mean_token_accuracy": 0.2599530279636383,
      "num_tokens": 113894926.0,
      "step": 49660
    },
    {
      "entropy": 5.594895505905152,
      "epoch": 4.770893371757925,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029472881623836906,
      "loss": 4.717,
      "mean_token_accuracy": 0.25706481784582136,
      "num_tokens": 113906035.0,
      "step": 49665
    },
    {
      "entropy": 5.57821569442749,
      "epoch": 4.771373679154659,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029469466450145177,
      "loss": 4.7142,
      "mean_token_accuracy": 0.2555663540959358,
      "num_tokens": 113917196.0,
      "step": 49670
    },
    {
      "entropy": 5.583044195175171,
      "epoch": 4.771853986551393,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00029466051230728283,
      "loss": 4.7638,
      "mean_token_accuracy": 0.25747443586587904,
      "num_tokens": 113928556.0,
      "step": 49675
    },
    {
      "entropy": 5.51786322593689,
      "epoch": 4.772334293948127,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002946263596566552,
      "loss": 4.6541,
      "mean_token_accuracy": 0.2597781792283058,
      "num_tokens": 113940123.0,
      "step": 49680
    },
    {
      "entropy": 5.50214147567749,
      "epoch": 4.7728146013448605,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029459220655036176,
      "loss": 4.6201,
      "mean_token_accuracy": 0.26924831569194796,
      "num_tokens": 113951913.0,
      "step": 49685
    },
    {
      "entropy": 5.598922824859619,
      "epoch": 4.773294908741595,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029455805298919555,
      "loss": 4.7875,
      "mean_token_accuracy": 0.25388905107975007,
      "num_tokens": 113963726.0,
      "step": 49690
    },
    {
      "entropy": 5.523237895965576,
      "epoch": 4.773775216138328,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002945238989739493,
      "loss": 4.674,
      "mean_token_accuracy": 0.25816998928785323,
      "num_tokens": 113974180.0,
      "step": 49695
    },
    {
      "entropy": 5.582477140426636,
      "epoch": 4.7742555235350626,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002944897445054161,
      "loss": 4.7261,
      "mean_token_accuracy": 0.25319081395864484,
      "num_tokens": 113984950.0,
      "step": 49700
    },
    {
      "entropy": 5.586855077743531,
      "epoch": 4.774735830931796,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000294455589584389,
      "loss": 4.7972,
      "mean_token_accuracy": 0.24847812801599503,
      "num_tokens": 113996034.0,
      "step": 49705
    },
    {
      "entropy": 5.531797313690186,
      "epoch": 4.77521613832853,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029442143421166083,
      "loss": 4.6656,
      "mean_token_accuracy": 0.26202153265476225,
      "num_tokens": 114007497.0,
      "step": 49710
    },
    {
      "entropy": 5.533063745498657,
      "epoch": 4.775696445725265,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002943872783880246,
      "loss": 4.7132,
      "mean_token_accuracy": 0.25767094641923904,
      "num_tokens": 114018285.0,
      "step": 49715
    },
    {
      "entropy": 5.559014987945557,
      "epoch": 4.776176753121998,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002943531221142734,
      "loss": 4.7495,
      "mean_token_accuracy": 0.2490769773721695,
      "num_tokens": 114029568.0,
      "step": 49720
    },
    {
      "entropy": 5.558981323242188,
      "epoch": 4.7766570605187315,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002943189653912001,
      "loss": 4.6735,
      "mean_token_accuracy": 0.26377323269844055,
      "num_tokens": 114040711.0,
      "step": 49725
    },
    {
      "entropy": 5.5560869693756105,
      "epoch": 4.777137367915466,
      "grad_norm": 1.125,
      "learning_rate": 0.0002942848082195978,
      "loss": 4.7192,
      "mean_token_accuracy": 0.2564304038882256,
      "num_tokens": 114053245.0,
      "step": 49730
    },
    {
      "entropy": 5.48664116859436,
      "epoch": 4.7776176753122,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002942506506002595,
      "loss": 4.6089,
      "mean_token_accuracy": 0.264312182366848,
      "num_tokens": 114065076.0,
      "step": 49735
    },
    {
      "entropy": 5.534460020065308,
      "epoch": 4.778097982708934,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002942164925339783,
      "loss": 4.7084,
      "mean_token_accuracy": 0.25789757668972013,
      "num_tokens": 114076180.0,
      "step": 49740
    },
    {
      "entropy": 5.580488920211792,
      "epoch": 4.778578290105668,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002941823340215472,
      "loss": 4.7123,
      "mean_token_accuracy": 0.2615086942911148,
      "num_tokens": 114087829.0,
      "step": 49745
    },
    {
      "entropy": 5.624134922027588,
      "epoch": 4.779058597502401,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029414817506375927,
      "loss": 4.8257,
      "mean_token_accuracy": 0.24053840339183807,
      "num_tokens": 114100139.0,
      "step": 49750
    },
    {
      "entropy": 5.533710241317749,
      "epoch": 4.779538904899136,
      "grad_norm": 1.25,
      "learning_rate": 0.00029411401566140755,
      "loss": 4.6416,
      "mean_token_accuracy": 0.257844415307045,
      "num_tokens": 114111877.0,
      "step": 49755
    },
    {
      "entropy": 5.58784236907959,
      "epoch": 4.780019212295869,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002940798558152851,
      "loss": 4.7342,
      "mean_token_accuracy": 0.25136772692203524,
      "num_tokens": 114122604.0,
      "step": 49760
    },
    {
      "entropy": 5.5559040069580075,
      "epoch": 4.7804995196926034,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029404569552618514,
      "loss": 4.7642,
      "mean_token_accuracy": 0.24990426898002624,
      "num_tokens": 114134718.0,
      "step": 49765
    },
    {
      "entropy": 5.572395372390747,
      "epoch": 4.780979827089337,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029401153479490075,
      "loss": 4.75,
      "mean_token_accuracy": 0.24648224413394929,
      "num_tokens": 114146905.0,
      "step": 49770
    },
    {
      "entropy": 5.6129437446594235,
      "epoch": 4.781460134486071,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002939773736222248,
      "loss": 4.7317,
      "mean_token_accuracy": 0.24930551946163176,
      "num_tokens": 114158004.0,
      "step": 49775
    },
    {
      "entropy": 5.543867778778076,
      "epoch": 4.781940441882805,
      "grad_norm": 1.25,
      "learning_rate": 0.0002939432120089506,
      "loss": 4.6912,
      "mean_token_accuracy": 0.2547830954194069,
      "num_tokens": 114169206.0,
      "step": 49780
    },
    {
      "entropy": 5.593410301208496,
      "epoch": 4.782420749279539,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029390904995587144,
      "loss": 4.7834,
      "mean_token_accuracy": 0.24712566137313843,
      "num_tokens": 114180988.0,
      "step": 49785
    },
    {
      "entropy": 5.497448968887329,
      "epoch": 4.782901056676273,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002938748874637801,
      "loss": 4.6806,
      "mean_token_accuracy": 0.25708419233560564,
      "num_tokens": 114192601.0,
      "step": 49790
    },
    {
      "entropy": 5.538060712814331,
      "epoch": 4.783381364073007,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029384072453346997,
      "loss": 4.7454,
      "mean_token_accuracy": 0.2560406669974327,
      "num_tokens": 114203938.0,
      "step": 49795
    },
    {
      "entropy": 5.529000854492187,
      "epoch": 4.78386167146974,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029380656116573424,
      "loss": 4.7051,
      "mean_token_accuracy": 0.25576271414756774,
      "num_tokens": 114215918.0,
      "step": 49800
    },
    {
      "entropy": 5.561012172698975,
      "epoch": 4.7843419788664745,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029377239736136596,
      "loss": 4.6655,
      "mean_token_accuracy": 0.26845733970403673,
      "num_tokens": 114227625.0,
      "step": 49805
    },
    {
      "entropy": 5.568856191635132,
      "epoch": 4.784822286263209,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002937382331211583,
      "loss": 4.6807,
      "mean_token_accuracy": 0.25524991005659103,
      "num_tokens": 114239578.0,
      "step": 49810
    },
    {
      "entropy": 5.5460371494293215,
      "epoch": 4.785302593659942,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002937040684459046,
      "loss": 4.7234,
      "mean_token_accuracy": 0.2612452432513237,
      "num_tokens": 114252082.0,
      "step": 49815
    },
    {
      "entropy": 5.548304319381714,
      "epoch": 4.785782901056677,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002936699033363979,
      "loss": 4.7187,
      "mean_token_accuracy": 0.256961789727211,
      "num_tokens": 114263501.0,
      "step": 49820
    },
    {
      "entropy": 5.534068822860718,
      "epoch": 4.78626320845341,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029363573779343154,
      "loss": 4.6352,
      "mean_token_accuracy": 0.2651775494217873,
      "num_tokens": 114274624.0,
      "step": 49825
    },
    {
      "entropy": 5.457083702087402,
      "epoch": 4.786743515850144,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002936015718177986,
      "loss": 4.5813,
      "mean_token_accuracy": 0.2664045557379723,
      "num_tokens": 114286354.0,
      "step": 49830
    },
    {
      "entropy": 5.565068054199219,
      "epoch": 4.787223823246878,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029356740541029247,
      "loss": 4.7598,
      "mean_token_accuracy": 0.25671923756599424,
      "num_tokens": 114298241.0,
      "step": 49835
    },
    {
      "entropy": 5.516598272323608,
      "epoch": 4.787704130643612,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002935332385717064,
      "loss": 4.5973,
      "mean_token_accuracy": 0.26334256678819656,
      "num_tokens": 114309728.0,
      "step": 49840
    },
    {
      "entropy": 5.477047348022461,
      "epoch": 4.7881844380403455,
      "grad_norm": 1.25,
      "learning_rate": 0.0002934990713028334,
      "loss": 4.5948,
      "mean_token_accuracy": 0.2634457632899284,
      "num_tokens": 114320816.0,
      "step": 49845
    },
    {
      "entropy": 5.498006772994995,
      "epoch": 4.78866474543708,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002934649036044669,
      "loss": 4.6534,
      "mean_token_accuracy": 0.26340007185935976,
      "num_tokens": 114331518.0,
      "step": 49850
    },
    {
      "entropy": 5.516376066207886,
      "epoch": 4.789145052833813,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002934307354774003,
      "loss": 4.6659,
      "mean_token_accuracy": 0.2591585427522659,
      "num_tokens": 114343280.0,
      "step": 49855
    },
    {
      "entropy": 5.542487144470215,
      "epoch": 4.789625360230548,
      "grad_norm": 1.375,
      "learning_rate": 0.00029339656692242667,
      "loss": 4.6397,
      "mean_token_accuracy": 0.25911676436662673,
      "num_tokens": 114353833.0,
      "step": 49860
    },
    {
      "entropy": 5.516024208068847,
      "epoch": 4.790105667627281,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029336239794033935,
      "loss": 4.6692,
      "mean_token_accuracy": 0.25274134874343873,
      "num_tokens": 114366012.0,
      "step": 49865
    },
    {
      "entropy": 5.41039891242981,
      "epoch": 4.790585975024015,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029332822853193175,
      "loss": 4.611,
      "mean_token_accuracy": 0.2621765360236168,
      "num_tokens": 114377374.0,
      "step": 49870
    },
    {
      "entropy": 5.535114336013794,
      "epoch": 4.791066282420749,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002932940586979971,
      "loss": 4.7026,
      "mean_token_accuracy": 0.2532339051365852,
      "num_tokens": 114388894.0,
      "step": 49875
    },
    {
      "entropy": 5.4877543449401855,
      "epoch": 4.791546589817483,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029325988843932865,
      "loss": 4.617,
      "mean_token_accuracy": 0.2680918872356415,
      "num_tokens": 114401204.0,
      "step": 49880
    },
    {
      "entropy": 5.527503490447998,
      "epoch": 4.7920268972142175,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029322571775671993,
      "loss": 4.6548,
      "mean_token_accuracy": 0.2611271530389786,
      "num_tokens": 114413248.0,
      "step": 49885
    },
    {
      "entropy": 5.514663934707642,
      "epoch": 4.792507204610951,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029319154665096407,
      "loss": 4.6948,
      "mean_token_accuracy": 0.25309518426656724,
      "num_tokens": 114426135.0,
      "step": 49890
    },
    {
      "entropy": 5.546713876724243,
      "epoch": 4.792987512007685,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029315737512285454,
      "loss": 4.7033,
      "mean_token_accuracy": 0.25190480798482895,
      "num_tokens": 114438386.0,
      "step": 49895
    },
    {
      "entropy": 5.504230213165283,
      "epoch": 4.793467819404419,
      "grad_norm": 1.25,
      "learning_rate": 0.0002931232031731847,
      "loss": 4.6623,
      "mean_token_accuracy": 0.2636158004403114,
      "num_tokens": 114449436.0,
      "step": 49900
    },
    {
      "entropy": 5.496984481811523,
      "epoch": 4.793948126801153,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029308903080274784,
      "loss": 4.6809,
      "mean_token_accuracy": 0.25716357976198195,
      "num_tokens": 114461010.0,
      "step": 49905
    },
    {
      "entropy": 5.524355888366699,
      "epoch": 4.7944284341978864,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002930548580123374,
      "loss": 4.6728,
      "mean_token_accuracy": 0.2619256108999252,
      "num_tokens": 114471712.0,
      "step": 49910
    },
    {
      "entropy": 5.5330945491790775,
      "epoch": 4.794908741594621,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029302068480274685,
      "loss": 4.6135,
      "mean_token_accuracy": 0.2612399265170097,
      "num_tokens": 114483109.0,
      "step": 49915
    },
    {
      "entropy": 5.463362073898315,
      "epoch": 4.795389048991354,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002929865111747694,
      "loss": 4.6386,
      "mean_token_accuracy": 0.2693366974592209,
      "num_tokens": 114494088.0,
      "step": 49920
    },
    {
      "entropy": 5.546717691421509,
      "epoch": 4.7958693563880885,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00029295233712919866,
      "loss": 4.7182,
      "mean_token_accuracy": 0.256349878013134,
      "num_tokens": 114505090.0,
      "step": 49925
    },
    {
      "entropy": 5.551745319366455,
      "epoch": 4.796349663784822,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000292918162666828,
      "loss": 4.6294,
      "mean_token_accuracy": 0.254243890941143,
      "num_tokens": 114516640.0,
      "step": 49930
    },
    {
      "entropy": 5.519516134262085,
      "epoch": 4.796829971181556,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002928839877884507,
      "loss": 4.7016,
      "mean_token_accuracy": 0.2507118433713913,
      "num_tokens": 114526514.0,
      "step": 49935
    },
    {
      "entropy": 5.536012411117554,
      "epoch": 4.79731027857829,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002928498124948603,
      "loss": 4.687,
      "mean_token_accuracy": 0.2609941393136978,
      "num_tokens": 114537206.0,
      "step": 49940
    },
    {
      "entropy": 5.627393293380737,
      "epoch": 4.797790585975024,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029281563678685034,
      "loss": 4.8295,
      "mean_token_accuracy": 0.2492863953113556,
      "num_tokens": 114548830.0,
      "step": 49945
    },
    {
      "entropy": 5.5051332950592045,
      "epoch": 4.7982708933717575,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002927814606652142,
      "loss": 4.6642,
      "mean_token_accuracy": 0.25883574336767196,
      "num_tokens": 114559777.0,
      "step": 49950
    },
    {
      "entropy": 5.5179657459259035,
      "epoch": 4.798751200768492,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029274728413074524,
      "loss": 4.6417,
      "mean_token_accuracy": 0.2595978558063507,
      "num_tokens": 114570299.0,
      "step": 49955
    },
    {
      "entropy": 5.524763202667236,
      "epoch": 4.799231508165226,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002927131071842372,
      "loss": 4.6928,
      "mean_token_accuracy": 0.2534759908914566,
      "num_tokens": 114582062.0,
      "step": 49960
    },
    {
      "entropy": 5.622017335891724,
      "epoch": 4.79971181556196,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029267892982648327,
      "loss": 4.7729,
      "mean_token_accuracy": 0.24696791023015977,
      "num_tokens": 114594279.0,
      "step": 49965
    },
    {
      "entropy": 5.551771831512451,
      "epoch": 4.800192122958694,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029264475205827714,
      "loss": 4.6915,
      "mean_token_accuracy": 0.2613398939371109,
      "num_tokens": 114606889.0,
      "step": 49970
    },
    {
      "entropy": 5.522386503219605,
      "epoch": 4.800672430355427,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029261057388041227,
      "loss": 4.6778,
      "mean_token_accuracy": 0.26410857737064364,
      "num_tokens": 114617447.0,
      "step": 49975
    },
    {
      "entropy": 5.572467470169068,
      "epoch": 4.801152737752162,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002925763952936821,
      "loss": 4.7381,
      "mean_token_accuracy": 0.25321913212537767,
      "num_tokens": 114628730.0,
      "step": 49980
    },
    {
      "entropy": 5.530237197875977,
      "epoch": 4.801633045148895,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029254221629888025,
      "loss": 4.6663,
      "mean_token_accuracy": 0.2605368956923485,
      "num_tokens": 114639971.0,
      "step": 49985
    },
    {
      "entropy": 5.578056812286377,
      "epoch": 4.802113352545629,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029250803689680026,
      "loss": 4.735,
      "mean_token_accuracy": 0.2500262185931206,
      "num_tokens": 114652596.0,
      "step": 49990
    },
    {
      "entropy": 5.449384355545044,
      "epoch": 4.802593659942363,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002924738570882356,
      "loss": 4.6025,
      "mean_token_accuracy": 0.25668410062789915,
      "num_tokens": 114663965.0,
      "step": 49995
    },
    {
      "entropy": 5.462963724136353,
      "epoch": 4.803073967339097,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029243967687397996,
      "loss": 4.6287,
      "mean_token_accuracy": 0.2623051702976227,
      "num_tokens": 114676279.0,
      "step": 50000
    },
    {
      "entropy": 5.484331798553467,
      "epoch": 4.803554274735831,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002924054962548267,
      "loss": 4.6148,
      "mean_token_accuracy": 0.26603019386529925,
      "num_tokens": 114687910.0,
      "step": 50005
    },
    {
      "entropy": 5.478596925735474,
      "epoch": 4.804034582132565,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029237131523156956,
      "loss": 4.5863,
      "mean_token_accuracy": 0.2624948427081108,
      "num_tokens": 114700006.0,
      "step": 50010
    },
    {
      "entropy": 5.52850308418274,
      "epoch": 4.804514889529298,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000292337133805002,
      "loss": 4.6598,
      "mean_token_accuracy": 0.2600063070654869,
      "num_tokens": 114711221.0,
      "step": 50015
    },
    {
      "entropy": 5.5020304203033445,
      "epoch": 4.804995196926033,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002923029519759178,
      "loss": 4.6414,
      "mean_token_accuracy": 0.26458075940608977,
      "num_tokens": 114723091.0,
      "step": 50020
    },
    {
      "entropy": 5.549348020553589,
      "epoch": 4.805475504322766,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002922687697451103,
      "loss": 4.6401,
      "mean_token_accuracy": 0.25872829258441926,
      "num_tokens": 114734347.0,
      "step": 50025
    },
    {
      "entropy": 5.516128444671631,
      "epoch": 4.8059558117195005,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029223458711337327,
      "loss": 4.7631,
      "mean_token_accuracy": 0.26016910523176195,
      "num_tokens": 114746191.0,
      "step": 50030
    },
    {
      "entropy": 5.55009093284607,
      "epoch": 4.806436119116235,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002922004040815004,
      "loss": 4.6848,
      "mean_token_accuracy": 0.2551736429333687,
      "num_tokens": 114758686.0,
      "step": 50035
    },
    {
      "entropy": 5.565339374542236,
      "epoch": 4.806916426512968,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002921662206502851,
      "loss": 4.7515,
      "mean_token_accuracy": 0.2545465797185898,
      "num_tokens": 114771616.0,
      "step": 50040
    },
    {
      "entropy": 5.561860513687134,
      "epoch": 4.807396733909703,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002921320368205212,
      "loss": 4.6451,
      "mean_token_accuracy": 0.26260995119810104,
      "num_tokens": 114781850.0,
      "step": 50045
    },
    {
      "entropy": 5.528982925415039,
      "epoch": 4.807877041306436,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002920978525930023,
      "loss": 4.6499,
      "mean_token_accuracy": 0.25334090143442156,
      "num_tokens": 114791997.0,
      "step": 50050
    },
    {
      "entropy": 5.577595520019531,
      "epoch": 4.80835734870317,
      "grad_norm": 1.25,
      "learning_rate": 0.0002920636679685219,
      "loss": 4.6896,
      "mean_token_accuracy": 0.2630211591720581,
      "num_tokens": 114802650.0,
      "step": 50055
    },
    {
      "entropy": 5.56360821723938,
      "epoch": 4.808837656099904,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002920294829478739,
      "loss": 4.7312,
      "mean_token_accuracy": 0.25488759726285937,
      "num_tokens": 114813792.0,
      "step": 50060
    },
    {
      "entropy": 5.512821435928345,
      "epoch": 4.809317963496638,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029199529753185183,
      "loss": 4.6839,
      "mean_token_accuracy": 0.2544720396399498,
      "num_tokens": 114824957.0,
      "step": 50065
    },
    {
      "entropy": 5.553933477401733,
      "epoch": 4.8097982708933715,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002919611117212495,
      "loss": 4.6423,
      "mean_token_accuracy": 0.263177689909935,
      "num_tokens": 114836091.0,
      "step": 50070
    },
    {
      "entropy": 5.517169332504272,
      "epoch": 4.810278578290106,
      "grad_norm": 1.375,
      "learning_rate": 0.00029192692551686044,
      "loss": 4.6044,
      "mean_token_accuracy": 0.2667677029967308,
      "num_tokens": 114846967.0,
      "step": 50075
    },
    {
      "entropy": 5.577963495254517,
      "epoch": 4.810758885686839,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002918927389194785,
      "loss": 4.7178,
      "mean_token_accuracy": 0.2569907858967781,
      "num_tokens": 114858035.0,
      "step": 50080
    },
    {
      "entropy": 5.526707744598388,
      "epoch": 4.811239193083574,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002918585519298972,
      "loss": 4.6679,
      "mean_token_accuracy": 0.2581997662782669,
      "num_tokens": 114870176.0,
      "step": 50085
    },
    {
      "entropy": 5.537578821182251,
      "epoch": 4.811719500480307,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002918243645489105,
      "loss": 4.6683,
      "mean_token_accuracy": 0.2629324868321419,
      "num_tokens": 114882453.0,
      "step": 50090
    },
    {
      "entropy": 5.494445323944092,
      "epoch": 4.812199807877041,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000291790176777312,
      "loss": 4.6833,
      "mean_token_accuracy": 0.2588220611214638,
      "num_tokens": 114894582.0,
      "step": 50095
    },
    {
      "entropy": 5.562508821487427,
      "epoch": 4.812680115273775,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002917559886158954,
      "loss": 4.6579,
      "mean_token_accuracy": 0.2655416503548622,
      "num_tokens": 114905901.0,
      "step": 50100
    },
    {
      "entropy": 5.478827476501465,
      "epoch": 4.813160422670509,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029172180006545453,
      "loss": 4.5703,
      "mean_token_accuracy": 0.26794796884059907,
      "num_tokens": 114917625.0,
      "step": 50105
    },
    {
      "entropy": 5.436439800262451,
      "epoch": 4.8136407300672435,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002916876111267832,
      "loss": 4.5716,
      "mean_token_accuracy": 0.2677062273025513,
      "num_tokens": 114929817.0,
      "step": 50110
    },
    {
      "entropy": 5.5615276336669925,
      "epoch": 4.814121037463977,
      "grad_norm": 1.25,
      "learning_rate": 0.00029165342180067496,
      "loss": 4.7338,
      "mean_token_accuracy": 0.2516311138868332,
      "num_tokens": 114941763.0,
      "step": 50115
    },
    {
      "entropy": 5.588158226013183,
      "epoch": 4.814601344860711,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002916192320879237,
      "loss": 4.6666,
      "mean_token_accuracy": 0.2681097090244293,
      "num_tokens": 114952909.0,
      "step": 50120
    },
    {
      "entropy": 5.556729459762574,
      "epoch": 4.815081652257445,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029158504198932334,
      "loss": 4.6873,
      "mean_token_accuracy": 0.2585921615362167,
      "num_tokens": 114964937.0,
      "step": 50125
    },
    {
      "entropy": 5.641573524475097,
      "epoch": 4.815561959654179,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029155085150566753,
      "loss": 4.8203,
      "mean_token_accuracy": 0.2480808362364769,
      "num_tokens": 114977171.0,
      "step": 50130
    },
    {
      "entropy": 5.555619716644287,
      "epoch": 4.816042267050912,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029151666063775,
      "loss": 4.6849,
      "mean_token_accuracy": 0.2553645059466362,
      "num_tokens": 114987160.0,
      "step": 50135
    },
    {
      "entropy": 5.485742807388306,
      "epoch": 4.816522574447647,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00029148246938636476,
      "loss": 4.6232,
      "mean_token_accuracy": 0.2648033186793327,
      "num_tokens": 114997737.0,
      "step": 50140
    },
    {
      "entropy": 5.573244094848633,
      "epoch": 4.81700288184438,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002914482777523055,
      "loss": 4.7299,
      "mean_token_accuracy": 0.25218756347894666,
      "num_tokens": 115008403.0,
      "step": 50145
    },
    {
      "entropy": 5.509153509140015,
      "epoch": 4.8174831892411145,
      "grad_norm": 1.34375,
      "learning_rate": 0.000291414085736366,
      "loss": 4.6531,
      "mean_token_accuracy": 0.2592646166682243,
      "num_tokens": 115018806.0,
      "step": 50150
    },
    {
      "entropy": 5.4991504669189455,
      "epoch": 4.817963496637848,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029137989333934025,
      "loss": 4.6395,
      "mean_token_accuracy": 0.26021278500556944,
      "num_tokens": 115030297.0,
      "step": 50155
    },
    {
      "entropy": 5.541189908981323,
      "epoch": 4.818443804034582,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00029134570056202195,
      "loss": 4.7111,
      "mean_token_accuracy": 0.25974312871694566,
      "num_tokens": 115040767.0,
      "step": 50160
    },
    {
      "entropy": 5.52358570098877,
      "epoch": 4.818924111431316,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029131150740520514,
      "loss": 4.6256,
      "mean_token_accuracy": 0.2555144801735878,
      "num_tokens": 115051441.0,
      "step": 50165
    },
    {
      "entropy": 5.53644585609436,
      "epoch": 4.81940441882805,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029127731386968344,
      "loss": 4.7308,
      "mean_token_accuracy": 0.2536251828074455,
      "num_tokens": 115062343.0,
      "step": 50170
    },
    {
      "entropy": 5.540202713012695,
      "epoch": 4.8198847262247835,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029124311995625094,
      "loss": 4.6772,
      "mean_token_accuracy": 0.2564846321940422,
      "num_tokens": 115072555.0,
      "step": 50175
    },
    {
      "entropy": 5.54416069984436,
      "epoch": 4.820365033621518,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029120892566570137,
      "loss": 4.6934,
      "mean_token_accuracy": 0.26331354677677155,
      "num_tokens": 115083123.0,
      "step": 50180
    },
    {
      "entropy": 5.562497329711914,
      "epoch": 4.820845341018252,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029117473099882864,
      "loss": 4.6896,
      "mean_token_accuracy": 0.2560726508498192,
      "num_tokens": 115095471.0,
      "step": 50185
    },
    {
      "entropy": 5.513739347457886,
      "epoch": 4.821325648414986,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002911405359564267,
      "loss": 4.6337,
      "mean_token_accuracy": 0.25969005525112154,
      "num_tokens": 115106649.0,
      "step": 50190
    },
    {
      "entropy": 5.528295803070068,
      "epoch": 4.821805955811719,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002911063405392895,
      "loss": 4.6416,
      "mean_token_accuracy": 0.26398885399103167,
      "num_tokens": 115117740.0,
      "step": 50195
    },
    {
      "entropy": 5.584143352508545,
      "epoch": 4.822286263208453,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029107214474821083,
      "loss": 4.7758,
      "mean_token_accuracy": 0.2528025537729263,
      "num_tokens": 115127920.0,
      "step": 50200
    },
    {
      "entropy": 5.50728588104248,
      "epoch": 4.822766570605188,
      "grad_norm": 1.25,
      "learning_rate": 0.00029103794858398476,
      "loss": 4.6436,
      "mean_token_accuracy": 0.25949900448322294,
      "num_tokens": 115140279.0,
      "step": 50205
    },
    {
      "entropy": 5.557386350631714,
      "epoch": 4.823246878001921,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002910037520474051,
      "loss": 4.5878,
      "mean_token_accuracy": 0.27139981240034106,
      "num_tokens": 115150350.0,
      "step": 50210
    },
    {
      "entropy": 5.5966551303863525,
      "epoch": 4.823727185398655,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002909695551392658,
      "loss": 4.7954,
      "mean_token_accuracy": 0.24566807299852372,
      "num_tokens": 115161563.0,
      "step": 50215
    },
    {
      "entropy": 5.558107852935791,
      "epoch": 4.824207492795389,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029093535786036095,
      "loss": 4.7015,
      "mean_token_accuracy": 0.2544710785150528,
      "num_tokens": 115172125.0,
      "step": 50220
    },
    {
      "entropy": 5.509967422485351,
      "epoch": 4.824687800192123,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029090116021148426,
      "loss": 4.6728,
      "mean_token_accuracy": 0.26122442781925204,
      "num_tokens": 115183509.0,
      "step": 50225
    },
    {
      "entropy": 5.460070276260376,
      "epoch": 4.825168107588857,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029086696219342994,
      "loss": 4.6284,
      "mean_token_accuracy": 0.2625263899564743,
      "num_tokens": 115195111.0,
      "step": 50230
    },
    {
      "entropy": 5.558806848526001,
      "epoch": 4.825648414985591,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000290832763806992,
      "loss": 4.6865,
      "mean_token_accuracy": 0.2535636439919472,
      "num_tokens": 115206965.0,
      "step": 50235
    },
    {
      "entropy": 5.530310726165771,
      "epoch": 4.826128722382324,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002907985650529641,
      "loss": 4.6556,
      "mean_token_accuracy": 0.25781970620155337,
      "num_tokens": 115218826.0,
      "step": 50240
    },
    {
      "entropy": 5.593046712875366,
      "epoch": 4.826609029779059,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002907643659321405,
      "loss": 4.7105,
      "mean_token_accuracy": 0.25243105441331865,
      "num_tokens": 115228652.0,
      "step": 50245
    },
    {
      "entropy": 5.586111831665039,
      "epoch": 4.827089337175792,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002907301664453152,
      "loss": 4.7331,
      "mean_token_accuracy": 0.2545425847172737,
      "num_tokens": 115240235.0,
      "step": 50250
    },
    {
      "entropy": 5.63349027633667,
      "epoch": 4.8275696445725265,
      "grad_norm": 1.265625,
      "learning_rate": 0.000290695966593282,
      "loss": 4.8226,
      "mean_token_accuracy": 0.2507226914167404,
      "num_tokens": 115251929.0,
      "step": 50255
    },
    {
      "entropy": 5.585833549499512,
      "epoch": 4.828049951969261,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002906617663768351,
      "loss": 4.6724,
      "mean_token_accuracy": 0.2584625408053398,
      "num_tokens": 115263042.0,
      "step": 50260
    },
    {
      "entropy": 5.547235822677612,
      "epoch": 4.828530259365994,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029062756579676866,
      "loss": 4.6636,
      "mean_token_accuracy": 0.2543299153447151,
      "num_tokens": 115273157.0,
      "step": 50265
    },
    {
      "entropy": 5.542889356613159,
      "epoch": 4.829010566762728,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002905933648538764,
      "loss": 4.7226,
      "mean_token_accuracy": 0.25562981218099595,
      "num_tokens": 115285354.0,
      "step": 50270
    },
    {
      "entropy": 5.512237501144409,
      "epoch": 4.829490874159462,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002905591635489525,
      "loss": 4.724,
      "mean_token_accuracy": 0.24702924937009813,
      "num_tokens": 115297719.0,
      "step": 50275
    },
    {
      "entropy": 5.577894830703736,
      "epoch": 4.829971181556196,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029052496188279104,
      "loss": 4.7033,
      "mean_token_accuracy": 0.2559619516134262,
      "num_tokens": 115310130.0,
      "step": 50280
    },
    {
      "entropy": 5.487441539764404,
      "epoch": 4.83045148895293,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029049075985618613,
      "loss": 4.6863,
      "mean_token_accuracy": 0.26008655279874804,
      "num_tokens": 115321065.0,
      "step": 50285
    },
    {
      "entropy": 5.430618143081665,
      "epoch": 4.830931796349664,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002904565574699317,
      "loss": 4.6018,
      "mean_token_accuracy": 0.2654637679457664,
      "num_tokens": 115332270.0,
      "step": 50290
    },
    {
      "entropy": 5.484411573410034,
      "epoch": 4.8314121037463975,
      "grad_norm": 1.125,
      "learning_rate": 0.0002904223547248219,
      "loss": 4.6157,
      "mean_token_accuracy": 0.26415336281061175,
      "num_tokens": 115344621.0,
      "step": 50295
    },
    {
      "entropy": 5.543502235412598,
      "epoch": 4.831892411143132,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002903881516216508,
      "loss": 4.5872,
      "mean_token_accuracy": 0.2671678900718689,
      "num_tokens": 115355429.0,
      "step": 50300
    },
    {
      "entropy": 5.581674289703369,
      "epoch": 4.832372718539865,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029035394816121266,
      "loss": 4.7385,
      "mean_token_accuracy": 0.25365095734596255,
      "num_tokens": 115367345.0,
      "step": 50305
    },
    {
      "entropy": 5.574899816513062,
      "epoch": 4.8328530259366,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002903197443443013,
      "loss": 4.7247,
      "mean_token_accuracy": 0.25790069699287416,
      "num_tokens": 115380111.0,
      "step": 50310
    },
    {
      "entropy": 5.519667625427246,
      "epoch": 4.833333333333333,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000290285540171711,
      "loss": 4.6761,
      "mean_token_accuracy": 0.25364880412817004,
      "num_tokens": 115393156.0,
      "step": 50315
    },
    {
      "entropy": 5.643899202346802,
      "epoch": 4.833813640730067,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002902513356442358,
      "loss": 4.7747,
      "mean_token_accuracy": 0.2479974627494812,
      "num_tokens": 115404675.0,
      "step": 50320
    },
    {
      "entropy": 5.576779127120972,
      "epoch": 4.834293948126801,
      "grad_norm": 1.25,
      "learning_rate": 0.00029021713076266996,
      "loss": 4.7341,
      "mean_token_accuracy": 0.25305938720703125,
      "num_tokens": 115416187.0,
      "step": 50325
    },
    {
      "entropy": 5.567765378952027,
      "epoch": 4.834774255523535,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002901829255278075,
      "loss": 4.6418,
      "mean_token_accuracy": 0.2607333272695541,
      "num_tokens": 115427375.0,
      "step": 50330
    },
    {
      "entropy": 5.5950539112091064,
      "epoch": 4.8352545629202694,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029014871994044255,
      "loss": 4.7312,
      "mean_token_accuracy": 0.2510968506336212,
      "num_tokens": 115438795.0,
      "step": 50335
    },
    {
      "entropy": 5.52950382232666,
      "epoch": 4.835734870317003,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002901145140013694,
      "loss": 4.6706,
      "mean_token_accuracy": 0.2637393340468407,
      "num_tokens": 115449944.0,
      "step": 50340
    },
    {
      "entropy": 5.463509035110474,
      "epoch": 4.836215177713736,
      "grad_norm": 1.234375,
      "learning_rate": 0.000290080307711382,
      "loss": 4.5661,
      "mean_token_accuracy": 0.2626357689499855,
      "num_tokens": 115461700.0,
      "step": 50345
    },
    {
      "entropy": 5.547169876098633,
      "epoch": 4.836695485110471,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029004610107127463,
      "loss": 4.6828,
      "mean_token_accuracy": 0.26037521958351134,
      "num_tokens": 115474173.0,
      "step": 50350
    },
    {
      "entropy": 5.55211763381958,
      "epoch": 4.837175792507205,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002900118940818415,
      "loss": 4.6561,
      "mean_token_accuracy": 0.257410129904747,
      "num_tokens": 115485275.0,
      "step": 50355
    },
    {
      "entropy": 5.509793376922607,
      "epoch": 4.837656099903938,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028997768674387685,
      "loss": 4.6542,
      "mean_token_accuracy": 0.26170649379491806,
      "num_tokens": 115496109.0,
      "step": 50360
    },
    {
      "entropy": 5.5103524684906,
      "epoch": 4.838136407300673,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002899434790581748,
      "loss": 4.6727,
      "mean_token_accuracy": 0.25184550136327744,
      "num_tokens": 115507212.0,
      "step": 50365
    },
    {
      "entropy": 5.581096363067627,
      "epoch": 4.838616714697406,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028990927102552936,
      "loss": 4.6939,
      "mean_token_accuracy": 0.25266123265028,
      "num_tokens": 115518401.0,
      "step": 50370
    },
    {
      "entropy": 5.558498334884644,
      "epoch": 4.8390970220941405,
      "grad_norm": 1.3125,
      "learning_rate": 0.000289875062646735,
      "loss": 4.6617,
      "mean_token_accuracy": 0.2548498764634132,
      "num_tokens": 115529480.0,
      "step": 50375
    },
    {
      "entropy": 5.497745132446289,
      "epoch": 4.839577329490874,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000289840853922586,
      "loss": 4.5527,
      "mean_token_accuracy": 0.26993132531642916,
      "num_tokens": 115539760.0,
      "step": 50380
    },
    {
      "entropy": 5.49000391960144,
      "epoch": 4.840057636887608,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028980664485387633,
      "loss": 4.6007,
      "mean_token_accuracy": 0.26378265470266343,
      "num_tokens": 115550054.0,
      "step": 50385
    },
    {
      "entropy": 5.48883957862854,
      "epoch": 4.840537944284342,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028977243544140027,
      "loss": 4.6473,
      "mean_token_accuracy": 0.26780229806900024,
      "num_tokens": 115561304.0,
      "step": 50390
    },
    {
      "entropy": 5.540748739242554,
      "epoch": 4.841018251681076,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028973822568595213,
      "loss": 4.748,
      "mean_token_accuracy": 0.2532925844192505,
      "num_tokens": 115573432.0,
      "step": 50395
    },
    {
      "entropy": 5.47693281173706,
      "epoch": 4.8414985590778095,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028970401558832626,
      "loss": 4.6231,
      "mean_token_accuracy": 0.26157341301441195,
      "num_tokens": 115584834.0,
      "step": 50400
    },
    {
      "entropy": 5.560884046554565,
      "epoch": 4.841978866474544,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028966980514931673,
      "loss": 4.7048,
      "mean_token_accuracy": 0.2641489177942276,
      "num_tokens": 115595977.0,
      "step": 50405
    },
    {
      "entropy": 5.588549375534058,
      "epoch": 4.842459173871278,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028963559436971794,
      "loss": 4.7429,
      "mean_token_accuracy": 0.2514436185359955,
      "num_tokens": 115607724.0,
      "step": 50410
    },
    {
      "entropy": 5.652096128463745,
      "epoch": 4.8429394812680115,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002896013832503241,
      "loss": 4.7426,
      "mean_token_accuracy": 0.24860565662384032,
      "num_tokens": 115618114.0,
      "step": 50415
    },
    {
      "entropy": 5.560735273361206,
      "epoch": 4.843419788664745,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028956717179192944,
      "loss": 4.7354,
      "mean_token_accuracy": 0.2544237270951271,
      "num_tokens": 115628723.0,
      "step": 50420
    },
    {
      "entropy": 5.579500913619995,
      "epoch": 4.843900096061479,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002895329599953283,
      "loss": 4.7931,
      "mean_token_accuracy": 0.247293384373188,
      "num_tokens": 115640816.0,
      "step": 50425
    },
    {
      "entropy": 5.679240465164185,
      "epoch": 4.844380403458214,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002894987478613151,
      "loss": 4.8495,
      "mean_token_accuracy": 0.24472607672214508,
      "num_tokens": 115651679.0,
      "step": 50430
    },
    {
      "entropy": 5.488152933120728,
      "epoch": 4.844860710854947,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028946453539068395,
      "loss": 4.6621,
      "mean_token_accuracy": 0.26404655128717425,
      "num_tokens": 115663507.0,
      "step": 50435
    },
    {
      "entropy": 5.586029148101806,
      "epoch": 4.845341018251681,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002894303225842292,
      "loss": 4.7247,
      "mean_token_accuracy": 0.25594974160194395,
      "num_tokens": 115674194.0,
      "step": 50440
    },
    {
      "entropy": 5.530804872512817,
      "epoch": 4.845821325648415,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028939610944274526,
      "loss": 4.6709,
      "mean_token_accuracy": 0.25887687504291534,
      "num_tokens": 115684853.0,
      "step": 50445
    },
    {
      "entropy": 5.519256925582885,
      "epoch": 4.846301633045149,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002893618959670264,
      "loss": 4.6513,
      "mean_token_accuracy": 0.26007645279169084,
      "num_tokens": 115694704.0,
      "step": 50450
    },
    {
      "entropy": 5.60690712928772,
      "epoch": 4.846781940441883,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002893276821578669,
      "loss": 4.8266,
      "mean_token_accuracy": 0.25121413022279737,
      "num_tokens": 115706178.0,
      "step": 50455
    },
    {
      "entropy": 5.630817651748657,
      "epoch": 4.847262247838617,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028929346801606127,
      "loss": 4.6972,
      "mean_token_accuracy": 0.24984704852104186,
      "num_tokens": 115716887.0,
      "step": 50460
    },
    {
      "entropy": 5.460723829269409,
      "epoch": 4.84774255523535,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028925925354240375,
      "loss": 4.6264,
      "mean_token_accuracy": 0.2609607055783272,
      "num_tokens": 115728142.0,
      "step": 50465
    },
    {
      "entropy": 5.518139743804932,
      "epoch": 4.848222862632085,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002892250387376886,
      "loss": 4.6107,
      "mean_token_accuracy": 0.25865538269281385,
      "num_tokens": 115737754.0,
      "step": 50470
    },
    {
      "entropy": 5.49315299987793,
      "epoch": 4.848703170028818,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002891908236027104,
      "loss": 4.6857,
      "mean_token_accuracy": 0.25954806357622145,
      "num_tokens": 115749263.0,
      "step": 50475
    },
    {
      "entropy": 5.426728343963623,
      "epoch": 4.8491834774255524,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002891566081382634,
      "loss": 4.5281,
      "mean_token_accuracy": 0.27004934251308443,
      "num_tokens": 115760630.0,
      "step": 50480
    },
    {
      "entropy": 5.5574303150177,
      "epoch": 4.849663784822286,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002891223923451419,
      "loss": 4.7236,
      "mean_token_accuracy": 0.25209314078092576,
      "num_tokens": 115773259.0,
      "step": 50485
    },
    {
      "entropy": 5.5230615615844725,
      "epoch": 4.85014409221902,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002890881762241405,
      "loss": 4.6231,
      "mean_token_accuracy": 0.2669981047511101,
      "num_tokens": 115785419.0,
      "step": 50490
    },
    {
      "entropy": 5.409414958953858,
      "epoch": 4.850624399615754,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002890539597760534,
      "loss": 4.5511,
      "mean_token_accuracy": 0.27560293972492217,
      "num_tokens": 115796421.0,
      "step": 50495
    },
    {
      "entropy": 5.512583208084107,
      "epoch": 4.851104707012488,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002890197430016751,
      "loss": 4.6469,
      "mean_token_accuracy": 0.2655674308538437,
      "num_tokens": 115807300.0,
      "step": 50500
    },
    {
      "entropy": 5.439594316482544,
      "epoch": 4.851585014409222,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002889855259018,
      "loss": 4.57,
      "mean_token_accuracy": 0.2648017853498459,
      "num_tokens": 115818003.0,
      "step": 50505
    },
    {
      "entropy": 5.514274549484253,
      "epoch": 4.852065321805956,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028895130847722257,
      "loss": 4.6565,
      "mean_token_accuracy": 0.25102643966674804,
      "num_tokens": 115829781.0,
      "step": 50510
    },
    {
      "entropy": 5.594828748703003,
      "epoch": 4.85254562920269,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028891709072873705,
      "loss": 4.6968,
      "mean_token_accuracy": 0.2577891245484352,
      "num_tokens": 115840503.0,
      "step": 50515
    },
    {
      "entropy": 5.4531268119812015,
      "epoch": 4.8530259365994235,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002888828726571382,
      "loss": 4.6104,
      "mean_token_accuracy": 0.2734969437122345,
      "num_tokens": 115851890.0,
      "step": 50520
    },
    {
      "entropy": 5.512522125244141,
      "epoch": 4.853506243996158,
      "grad_norm": 1.375,
      "learning_rate": 0.0002888486542632202,
      "loss": 4.6465,
      "mean_token_accuracy": 0.258397713303566,
      "num_tokens": 115863584.0,
      "step": 50525
    },
    {
      "entropy": 5.470711517333984,
      "epoch": 4.853986551392891,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028881443554777745,
      "loss": 4.6107,
      "mean_token_accuracy": 0.2620575428009033,
      "num_tokens": 115874961.0,
      "step": 50530
    },
    {
      "entropy": 5.601215744018555,
      "epoch": 4.854466858789626,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028878021651160463,
      "loss": 4.7034,
      "mean_token_accuracy": 0.258908112347126,
      "num_tokens": 115886123.0,
      "step": 50535
    },
    {
      "entropy": 5.5389491558074955,
      "epoch": 4.854947166186359,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002887459971554961,
      "loss": 4.6659,
      "mean_token_accuracy": 0.25403787940740585,
      "num_tokens": 115897566.0,
      "step": 50540
    },
    {
      "entropy": 5.572155380249024,
      "epoch": 4.855427473583093,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028871177748024634,
      "loss": 4.757,
      "mean_token_accuracy": 0.2530954793095589,
      "num_tokens": 115909475.0,
      "step": 50545
    },
    {
      "entropy": 5.571916484832764,
      "epoch": 4.855907780979827,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002886775574866498,
      "loss": 4.7047,
      "mean_token_accuracy": 0.25870698392391206,
      "num_tokens": 115920922.0,
      "step": 50550
    },
    {
      "entropy": 5.5136631488800045,
      "epoch": 4.856388088376561,
      "grad_norm": 1.25,
      "learning_rate": 0.000288643337175501,
      "loss": 4.699,
      "mean_token_accuracy": 0.25161079615354537,
      "num_tokens": 115932952.0,
      "step": 50555
    },
    {
      "entropy": 5.596230411529541,
      "epoch": 4.8568683957732945,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028860911654759444,
      "loss": 4.7362,
      "mean_token_accuracy": 0.25323085486888885,
      "num_tokens": 115943730.0,
      "step": 50560
    },
    {
      "entropy": 5.620119953155518,
      "epoch": 4.857348703170029,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002885748956037246,
      "loss": 4.6918,
      "mean_token_accuracy": 0.26110541075468063,
      "num_tokens": 115954602.0,
      "step": 50565
    },
    {
      "entropy": 5.522569561004639,
      "epoch": 4.857829010566762,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028854067434468594,
      "loss": 4.6977,
      "mean_token_accuracy": 0.26328202784061433,
      "num_tokens": 115966128.0,
      "step": 50570
    },
    {
      "entropy": 5.4988133907318115,
      "epoch": 4.858309317963497,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000288506452771273,
      "loss": 4.7102,
      "mean_token_accuracy": 0.2518626496195793,
      "num_tokens": 115976613.0,
      "step": 50575
    },
    {
      "entropy": 5.509871530532837,
      "epoch": 4.858789625360231,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002884722308842805,
      "loss": 4.6358,
      "mean_token_accuracy": 0.2633798986673355,
      "num_tokens": 115987484.0,
      "step": 50580
    },
    {
      "entropy": 5.606991195678711,
      "epoch": 4.859269932756964,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002884380086845027,
      "loss": 4.7299,
      "mean_token_accuracy": 0.25040450841188433,
      "num_tokens": 115999062.0,
      "step": 50585
    },
    {
      "entropy": 5.520765733718872,
      "epoch": 4.859750240153699,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028840378617273424,
      "loss": 4.6504,
      "mean_token_accuracy": 0.25930708199739455,
      "num_tokens": 116010514.0,
      "step": 50590
    },
    {
      "entropy": 5.623202276229859,
      "epoch": 4.860230547550432,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002883695633497697,
      "loss": 4.7619,
      "mean_token_accuracy": 0.24905590415000917,
      "num_tokens": 116022641.0,
      "step": 50595
    },
    {
      "entropy": 5.610523700714111,
      "epoch": 4.8607108549471665,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002883353402164035,
      "loss": 4.7371,
      "mean_token_accuracy": 0.24768817722797393,
      "num_tokens": 116033663.0,
      "step": 50600
    },
    {
      "entropy": 5.5336244106292725,
      "epoch": 4.8611911623439,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028830111677343037,
      "loss": 4.6931,
      "mean_token_accuracy": 0.2553511530160904,
      "num_tokens": 116045384.0,
      "step": 50605
    },
    {
      "entropy": 5.564023780822754,
      "epoch": 4.861671469740634,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002882668930216448,
      "loss": 4.663,
      "mean_token_accuracy": 0.2562651991844177,
      "num_tokens": 116056850.0,
      "step": 50610
    },
    {
      "entropy": 5.515863084793091,
      "epoch": 4.862151777137368,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002882326689618414,
      "loss": 4.638,
      "mean_token_accuracy": 0.2605817288160324,
      "num_tokens": 116068235.0,
      "step": 50615
    },
    {
      "entropy": 5.44949893951416,
      "epoch": 4.862632084534102,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028819844459481467,
      "loss": 4.6021,
      "mean_token_accuracy": 0.26112758815288545,
      "num_tokens": 116080556.0,
      "step": 50620
    },
    {
      "entropy": 5.498001432418823,
      "epoch": 4.863112391930835,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028816421992135927,
      "loss": 4.5925,
      "mean_token_accuracy": 0.27513571083545685,
      "num_tokens": 116091583.0,
      "step": 50625
    },
    {
      "entropy": 5.559062194824219,
      "epoch": 4.86359269932757,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028812999494226977,
      "loss": 4.63,
      "mean_token_accuracy": 0.26352627724409106,
      "num_tokens": 116101942.0,
      "step": 50630
    },
    {
      "entropy": 5.538974142074585,
      "epoch": 4.864073006724303,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028809576965834073,
      "loss": 4.6675,
      "mean_token_accuracy": 0.2589496701955795,
      "num_tokens": 116112097.0,
      "step": 50635
    },
    {
      "entropy": 5.431685066223144,
      "epoch": 4.8645533141210375,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002880615440703668,
      "loss": 4.6094,
      "mean_token_accuracy": 0.26506890803575517,
      "num_tokens": 116124716.0,
      "step": 50640
    },
    {
      "entropy": 5.492398309707641,
      "epoch": 4.865033621517771,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002880273181791427,
      "loss": 4.6496,
      "mean_token_accuracy": 0.2606605663895607,
      "num_tokens": 116135795.0,
      "step": 50645
    },
    {
      "entropy": 5.58632459640503,
      "epoch": 4.865513928914505,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002879930919854629,
      "loss": 4.7101,
      "mean_token_accuracy": 0.26342547684907913,
      "num_tokens": 116147581.0,
      "step": 50650
    },
    {
      "entropy": 5.509091567993164,
      "epoch": 4.86599423631124,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028795886549012206,
      "loss": 4.5873,
      "mean_token_accuracy": 0.26923919171094896,
      "num_tokens": 116157998.0,
      "step": 50655
    },
    {
      "entropy": 5.491602373123169,
      "epoch": 4.866474543707973,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028792463869391487,
      "loss": 4.6323,
      "mean_token_accuracy": 0.2660223960876465,
      "num_tokens": 116168904.0,
      "step": 50660
    },
    {
      "entropy": 5.553592348098755,
      "epoch": 4.866954851104707,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028789041159763603,
      "loss": 4.7616,
      "mean_token_accuracy": 0.25571623593568804,
      "num_tokens": 116179305.0,
      "step": 50665
    },
    {
      "entropy": 5.581087684631347,
      "epoch": 4.867435158501441,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028785618420208,
      "loss": 4.7105,
      "mean_token_accuracy": 0.24800153374671935,
      "num_tokens": 116189299.0,
      "step": 50670
    },
    {
      "entropy": 5.545233297348022,
      "epoch": 4.867915465898175,
      "grad_norm": 1.25,
      "learning_rate": 0.00028782195650804156,
      "loss": 4.6738,
      "mean_token_accuracy": 0.25453554689884184,
      "num_tokens": 116199877.0,
      "step": 50675
    },
    {
      "entropy": 5.502768468856812,
      "epoch": 4.868395773294909,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002877877285163154,
      "loss": 4.6698,
      "mean_token_accuracy": 0.2590806812047958,
      "num_tokens": 116210480.0,
      "step": 50680
    },
    {
      "entropy": 5.573952722549438,
      "epoch": 4.868876080691643,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028775350022769615,
      "loss": 4.7973,
      "mean_token_accuracy": 0.24447559118270873,
      "num_tokens": 116221780.0,
      "step": 50685
    },
    {
      "entropy": 5.565600633621216,
      "epoch": 4.869356388088376,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028771927164297843,
      "loss": 4.6392,
      "mean_token_accuracy": 0.2603668600320816,
      "num_tokens": 116234414.0,
      "step": 50690
    },
    {
      "entropy": 5.589218044281006,
      "epoch": 4.869836695485111,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028768504276295706,
      "loss": 4.6975,
      "mean_token_accuracy": 0.25620718002319337,
      "num_tokens": 116245562.0,
      "step": 50695
    },
    {
      "entropy": 5.541265201568604,
      "epoch": 4.870317002881844,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002876508135884267,
      "loss": 4.6949,
      "mean_token_accuracy": 0.25972579419612885,
      "num_tokens": 116256565.0,
      "step": 50700
    },
    {
      "entropy": 5.561945581436158,
      "epoch": 4.870797310278578,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002876165841201819,
      "loss": 4.7642,
      "mean_token_accuracy": 0.2567469820380211,
      "num_tokens": 116268077.0,
      "step": 50705
    },
    {
      "entropy": 5.524457883834839,
      "epoch": 4.871277617675312,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028758235435901754,
      "loss": 4.6152,
      "mean_token_accuracy": 0.2623887926340103,
      "num_tokens": 116278338.0,
      "step": 50710
    },
    {
      "entropy": 5.490522766113282,
      "epoch": 4.871757925072046,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002875481243057283,
      "loss": 4.6548,
      "mean_token_accuracy": 0.263328792154789,
      "num_tokens": 116290892.0,
      "step": 50715
    },
    {
      "entropy": 5.412061500549316,
      "epoch": 4.87223823246878,
      "grad_norm": 1.375,
      "learning_rate": 0.00028751389396110884,
      "loss": 4.5344,
      "mean_token_accuracy": 0.2689974203705788,
      "num_tokens": 116301072.0,
      "step": 50720
    },
    {
      "entropy": 5.527578592300415,
      "epoch": 4.872718539865514,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028747966332595394,
      "loss": 4.7228,
      "mean_token_accuracy": 0.25718191266059875,
      "num_tokens": 116313263.0,
      "step": 50725
    },
    {
      "entropy": 5.68074746131897,
      "epoch": 4.873198847262248,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002874454324010583,
      "loss": 4.7511,
      "mean_token_accuracy": 0.25006173700094225,
      "num_tokens": 116324076.0,
      "step": 50730
    },
    {
      "entropy": 5.52501335144043,
      "epoch": 4.873679154658982,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002874112011872167,
      "loss": 4.6612,
      "mean_token_accuracy": 0.2605654910206795,
      "num_tokens": 116335513.0,
      "step": 50735
    },
    {
      "entropy": 5.5434952735900875,
      "epoch": 4.874159462055716,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002873769696852238,
      "loss": 4.738,
      "mean_token_accuracy": 0.2500367105007172,
      "num_tokens": 116346750.0,
      "step": 50740
    },
    {
      "entropy": 5.490592002868652,
      "epoch": 4.8746397694524495,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002873427378958745,
      "loss": 4.6877,
      "mean_token_accuracy": 0.2558460786938667,
      "num_tokens": 116358359.0,
      "step": 50745
    },
    {
      "entropy": 5.496728181838989,
      "epoch": 4.875120076849184,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028730850581996345,
      "loss": 4.5936,
      "mean_token_accuracy": 0.26802582293748856,
      "num_tokens": 116369346.0,
      "step": 50750
    },
    {
      "entropy": 5.621200704574585,
      "epoch": 4.875600384245917,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002872742734582854,
      "loss": 4.762,
      "mean_token_accuracy": 0.24805329591035843,
      "num_tokens": 116380846.0,
      "step": 50755
    },
    {
      "entropy": 5.522297143936157,
      "epoch": 4.876080691642652,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028724004081163523,
      "loss": 4.6199,
      "mean_token_accuracy": 0.26039344966411593,
      "num_tokens": 116392492.0,
      "step": 50760
    },
    {
      "entropy": 5.567150115966797,
      "epoch": 4.876560999039385,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028720580788080757,
      "loss": 4.6472,
      "mean_token_accuracy": 0.2635364755988121,
      "num_tokens": 116403788.0,
      "step": 50765
    },
    {
      "entropy": 5.4437782764434814,
      "epoch": 4.877041306436119,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028717157466659737,
      "loss": 4.5642,
      "mean_token_accuracy": 0.27679601311683655,
      "num_tokens": 116415208.0,
      "step": 50770
    },
    {
      "entropy": 5.502903032302856,
      "epoch": 4.877521613832853,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028713734116979936,
      "loss": 4.6017,
      "mean_token_accuracy": 0.2649358808994293,
      "num_tokens": 116426338.0,
      "step": 50775
    },
    {
      "entropy": 5.648551511764526,
      "epoch": 4.878001921229587,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028710310739120826,
      "loss": 4.7699,
      "mean_token_accuracy": 0.25092683285474776,
      "num_tokens": 116436253.0,
      "step": 50780
    },
    {
      "entropy": 5.626560258865356,
      "epoch": 4.8784822286263205,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028706887333161896,
      "loss": 4.8096,
      "mean_token_accuracy": 0.24788620918989182,
      "num_tokens": 116447070.0,
      "step": 50785
    },
    {
      "entropy": 5.583043479919434,
      "epoch": 4.878962536023055,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002870346389918263,
      "loss": 4.7789,
      "mean_token_accuracy": 0.24888496845960617,
      "num_tokens": 116459482.0,
      "step": 50790
    },
    {
      "entropy": 5.587989044189453,
      "epoch": 4.879442843419788,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028700040437262505,
      "loss": 4.7469,
      "mean_token_accuracy": 0.2496942862868309,
      "num_tokens": 116471100.0,
      "step": 50795
    },
    {
      "entropy": 5.461436605453491,
      "epoch": 4.879923150816523,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028696616947481004,
      "loss": 4.6327,
      "mean_token_accuracy": 0.2603986859321594,
      "num_tokens": 116484004.0,
      "step": 50800
    },
    {
      "entropy": 5.571089792251587,
      "epoch": 4.880403458213257,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028693193429917615,
      "loss": 4.7332,
      "mean_token_accuracy": 0.24298716932535172,
      "num_tokens": 116495451.0,
      "step": 50805
    },
    {
      "entropy": 5.624762964248657,
      "epoch": 4.88088376560999,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002868976988465181,
      "loss": 4.7685,
      "mean_token_accuracy": 0.2456412509083748,
      "num_tokens": 116506551.0,
      "step": 50810
    },
    {
      "entropy": 5.505647087097168,
      "epoch": 4.881364073006724,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002868634631176308,
      "loss": 4.5727,
      "mean_token_accuracy": 0.2606074005365372,
      "num_tokens": 116518310.0,
      "step": 50815
    },
    {
      "entropy": 5.619717025756836,
      "epoch": 4.881844380403458,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028682922711330917,
      "loss": 4.8183,
      "mean_token_accuracy": 0.25027257949113846,
      "num_tokens": 116528835.0,
      "step": 50820
    },
    {
      "entropy": 5.50112247467041,
      "epoch": 4.8823246878001925,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028679499083434794,
      "loss": 4.6716,
      "mean_token_accuracy": 0.2580739170312881,
      "num_tokens": 116540189.0,
      "step": 50825
    },
    {
      "entropy": 5.533471775054932,
      "epoch": 4.882804995196926,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028676075428154206,
      "loss": 4.6266,
      "mean_token_accuracy": 0.2721930220723152,
      "num_tokens": 116551527.0,
      "step": 50830
    },
    {
      "entropy": 5.603919649124146,
      "epoch": 4.88328530259366,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002867265174556865,
      "loss": 4.7968,
      "mean_token_accuracy": 0.2555998980998993,
      "num_tokens": 116563058.0,
      "step": 50835
    },
    {
      "entropy": 5.540848207473755,
      "epoch": 4.883765609990394,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002866922803575759,
      "loss": 4.6458,
      "mean_token_accuracy": 0.2593246266245842,
      "num_tokens": 116574399.0,
      "step": 50840
    },
    {
      "entropy": 5.518783760070801,
      "epoch": 4.884245917387128,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028665804298800537,
      "loss": 4.6506,
      "mean_token_accuracy": 0.25347021371126177,
      "num_tokens": 116586451.0,
      "step": 50845
    },
    {
      "entropy": 5.4972021102905275,
      "epoch": 4.884726224783861,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028662380534776966,
      "loss": 4.615,
      "mean_token_accuracy": 0.2647184744477272,
      "num_tokens": 116597423.0,
      "step": 50850
    },
    {
      "entropy": 5.54445514678955,
      "epoch": 4.885206532180596,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028658956743766366,
      "loss": 4.6925,
      "mean_token_accuracy": 0.2574873462319374,
      "num_tokens": 116608756.0,
      "step": 50855
    },
    {
      "entropy": 5.553913545608521,
      "epoch": 4.885686839577329,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002865553292584824,
      "loss": 4.6504,
      "mean_token_accuracy": 0.2602318525314331,
      "num_tokens": 116619516.0,
      "step": 50860
    },
    {
      "entropy": 5.561381387710571,
      "epoch": 4.8861671469740635,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002865210908110207,
      "loss": 4.7522,
      "mean_token_accuracy": 0.25957666635513305,
      "num_tokens": 116631482.0,
      "step": 50865
    },
    {
      "entropy": 5.528856420516968,
      "epoch": 4.886647454370797,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002864868520960734,
      "loss": 4.6521,
      "mean_token_accuracy": 0.26196133345365524,
      "num_tokens": 116643193.0,
      "step": 50870
    },
    {
      "entropy": 5.566158103942871,
      "epoch": 4.887127761767531,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028645261311443555,
      "loss": 4.6975,
      "mean_token_accuracy": 0.25461900383234026,
      "num_tokens": 116654154.0,
      "step": 50875
    },
    {
      "entropy": 5.585364723205567,
      "epoch": 4.887608069164266,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000286418373866902,
      "loss": 4.687,
      "mean_token_accuracy": 0.2564746499061584,
      "num_tokens": 116666200.0,
      "step": 50880
    },
    {
      "entropy": 5.583368349075317,
      "epoch": 4.888088376560999,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028638413435426777,
      "loss": 4.7487,
      "mean_token_accuracy": 0.24483132660388945,
      "num_tokens": 116678409.0,
      "step": 50885
    },
    {
      "entropy": 5.563484239578247,
      "epoch": 4.8885686839577325,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028634989457732767,
      "loss": 4.7302,
      "mean_token_accuracy": 0.26344092190265656,
      "num_tokens": 116689448.0,
      "step": 50890
    },
    {
      "entropy": 5.5547106742858885,
      "epoch": 4.889048991354467,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002863156545368768,
      "loss": 4.7,
      "mean_token_accuracy": 0.25284708887338636,
      "num_tokens": 116700976.0,
      "step": 50895
    },
    {
      "entropy": 5.575588893890381,
      "epoch": 4.889529298751201,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028628141423370993,
      "loss": 4.6807,
      "mean_token_accuracy": 0.2606223836541176,
      "num_tokens": 116713117.0,
      "step": 50900
    },
    {
      "entropy": 5.635686540603638,
      "epoch": 4.890009606147935,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028624717366862223,
      "loss": 4.8168,
      "mean_token_accuracy": 0.24600475430488586,
      "num_tokens": 116725260.0,
      "step": 50905
    },
    {
      "entropy": 5.5029548645019535,
      "epoch": 4.890489913544669,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028621293284240856,
      "loss": 4.6987,
      "mean_token_accuracy": 0.2581894502043724,
      "num_tokens": 116737508.0,
      "step": 50910
    },
    {
      "entropy": 5.455356740951538,
      "epoch": 4.890970220941402,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002861786917558638,
      "loss": 4.5731,
      "mean_token_accuracy": 0.26095047146081923,
      "num_tokens": 116748841.0,
      "step": 50915
    },
    {
      "entropy": 5.493468475341797,
      "epoch": 4.891450528338137,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028614445040978304,
      "loss": 4.5906,
      "mean_token_accuracy": 0.25476653724908827,
      "num_tokens": 116761053.0,
      "step": 50920
    },
    {
      "entropy": 5.631753396987915,
      "epoch": 4.89193083573487,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028611020880496123,
      "loss": 4.8301,
      "mean_token_accuracy": 0.24466420263051986,
      "num_tokens": 116772521.0,
      "step": 50925
    },
    {
      "entropy": 5.618339204788208,
      "epoch": 4.892411143131604,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002860759669421934,
      "loss": 4.7244,
      "mean_token_accuracy": 0.2544622927904129,
      "num_tokens": 116784564.0,
      "step": 50930
    },
    {
      "entropy": 5.550128841400147,
      "epoch": 4.892891450528338,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002860417248222745,
      "loss": 4.679,
      "mean_token_accuracy": 0.2546504497528076,
      "num_tokens": 116796100.0,
      "step": 50935
    },
    {
      "entropy": 5.6206317901611325,
      "epoch": 4.893371757925072,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002860074824459995,
      "loss": 4.7385,
      "mean_token_accuracy": 0.2643949195742607,
      "num_tokens": 116807644.0,
      "step": 50940
    },
    {
      "entropy": 5.506115484237671,
      "epoch": 4.893852065321806,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002859732398141634,
      "loss": 4.6766,
      "mean_token_accuracy": 0.25188429951667785,
      "num_tokens": 116818913.0,
      "step": 50945
    },
    {
      "entropy": 5.495435619354248,
      "epoch": 4.89433237271854,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002859389969275613,
      "loss": 4.6413,
      "mean_token_accuracy": 0.2529128804802895,
      "num_tokens": 116829418.0,
      "step": 50950
    },
    {
      "entropy": 5.490136337280274,
      "epoch": 4.894812680115274,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002859047537869882,
      "loss": 4.6459,
      "mean_token_accuracy": 0.258088093996048,
      "num_tokens": 116841186.0,
      "step": 50955
    },
    {
      "entropy": 5.481443405151367,
      "epoch": 4.895292987512008,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002858705103932391,
      "loss": 4.6277,
      "mean_token_accuracy": 0.25477508902549745,
      "num_tokens": 116851309.0,
      "step": 50960
    },
    {
      "entropy": 5.563046503067016,
      "epoch": 4.895773294908741,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028583626674710897,
      "loss": 4.7102,
      "mean_token_accuracy": 0.2524739742279053,
      "num_tokens": 116863779.0,
      "step": 50965
    },
    {
      "entropy": 5.617795133590699,
      "epoch": 4.8962536023054755,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000285802022849393,
      "loss": 4.779,
      "mean_token_accuracy": 0.25409391820430755,
      "num_tokens": 116876673.0,
      "step": 50970
    },
    {
      "entropy": 5.566997146606445,
      "epoch": 4.89673390970221,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002857677787008861,
      "loss": 4.702,
      "mean_token_accuracy": 0.2517378330230713,
      "num_tokens": 116887470.0,
      "step": 50975
    },
    {
      "entropy": 5.464937925338745,
      "epoch": 4.897214217098943,
      "grad_norm": 1.25,
      "learning_rate": 0.00028573353430238324,
      "loss": 4.581,
      "mean_token_accuracy": 0.2665947139263153,
      "num_tokens": 116899203.0,
      "step": 50980
    },
    {
      "entropy": 5.489330768585205,
      "epoch": 4.8976945244956775,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028569928965467973,
      "loss": 4.692,
      "mean_token_accuracy": 0.2578034117817879,
      "num_tokens": 116911460.0,
      "step": 50985
    },
    {
      "entropy": 5.52569808959961,
      "epoch": 4.898174831892411,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028566504475857044,
      "loss": 4.7079,
      "mean_token_accuracy": 0.25523747205734254,
      "num_tokens": 116922411.0,
      "step": 50990
    },
    {
      "entropy": 5.494235801696777,
      "epoch": 4.898655139289145,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002856307996148505,
      "loss": 4.5827,
      "mean_token_accuracy": 0.26547201722860336,
      "num_tokens": 116934398.0,
      "step": 50995
    },
    {
      "entropy": 5.51976842880249,
      "epoch": 4.899135446685879,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028559655422431493,
      "loss": 4.6952,
      "mean_token_accuracy": 0.25828570425510405,
      "num_tokens": 116945170.0,
      "step": 51000
    },
    {
      "epoch": 4.899135446685879,
      "eval_entropy": 5.213690584081373,
      "eval_loss": 4.777991771697998,
      "eval_mean_token_accuracy": 0.2617244982882164,
      "eval_num_tokens": 116945170.0,
      "eval_runtime": 20.624,
      "eval_samples_per_second": 1591.106,
      "eval_steps_per_second": 198.894,
      "step": 51000
    },
    {
      "entropy": 5.5024114608764645,
      "epoch": 4.899615754082613,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002855623085877588,
      "loss": 4.6707,
      "mean_token_accuracy": 0.25882659405469893,
      "num_tokens": 116955711.0,
      "step": 51005
    },
    {
      "entropy": 5.43154354095459,
      "epoch": 4.9000960614793465,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002855280627059773,
      "loss": 4.574,
      "mean_token_accuracy": 0.267194963991642,
      "num_tokens": 116966589.0,
      "step": 51010
    },
    {
      "entropy": 5.484604167938232,
      "epoch": 4.900576368876081,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002854938165797654,
      "loss": 4.6032,
      "mean_token_accuracy": 0.26520181596279147,
      "num_tokens": 116978529.0,
      "step": 51015
    },
    {
      "entropy": 5.535462045669556,
      "epoch": 4.901056676272814,
      "grad_norm": 1.25,
      "learning_rate": 0.00028545957020991824,
      "loss": 4.735,
      "mean_token_accuracy": 0.2533364400267601,
      "num_tokens": 116990146.0,
      "step": 51020
    },
    {
      "entropy": 5.600887966156006,
      "epoch": 4.901536983669549,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000285425323597231,
      "loss": 4.733,
      "mean_token_accuracy": 0.25515916049480436,
      "num_tokens": 117001520.0,
      "step": 51025
    },
    {
      "entropy": 5.482072639465332,
      "epoch": 4.902017291066282,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002853910767424986,
      "loss": 4.6225,
      "mean_token_accuracy": 0.2596193298697472,
      "num_tokens": 117013505.0,
      "step": 51030
    },
    {
      "entropy": 5.5848602771759035,
      "epoch": 4.902497598463016,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002853568296465163,
      "loss": 4.7356,
      "mean_token_accuracy": 0.2494861364364624,
      "num_tokens": 117024025.0,
      "step": 51035
    },
    {
      "entropy": 5.5342652797698975,
      "epoch": 4.90297790585975,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028532258231007913,
      "loss": 4.6393,
      "mean_token_accuracy": 0.26067875176668165,
      "num_tokens": 117035451.0,
      "step": 51040
    },
    {
      "entropy": 5.6254926204681395,
      "epoch": 4.903458213256484,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002852883347339823,
      "loss": 4.7852,
      "mean_token_accuracy": 0.24933143854141235,
      "num_tokens": 117047068.0,
      "step": 51045
    },
    {
      "entropy": 5.56856255531311,
      "epoch": 4.9039385206532184,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002852540869190209,
      "loss": 4.714,
      "mean_token_accuracy": 0.25108253210783005,
      "num_tokens": 117059029.0,
      "step": 51050
    },
    {
      "entropy": 5.5184314250946045,
      "epoch": 4.904418828049952,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028521983886599,
      "loss": 4.6023,
      "mean_token_accuracy": 0.26659607887268066,
      "num_tokens": 117071066.0,
      "step": 51055
    },
    {
      "entropy": 5.577201986312867,
      "epoch": 4.904899135446686,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002851855905756848,
      "loss": 4.7091,
      "mean_token_accuracy": 0.2520953640341759,
      "num_tokens": 117084099.0,
      "step": 51060
    },
    {
      "entropy": 5.51625018119812,
      "epoch": 4.90537944284342,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002851513420489004,
      "loss": 4.6079,
      "mean_token_accuracy": 0.2668176174163818,
      "num_tokens": 117094944.0,
      "step": 51065
    },
    {
      "entropy": 5.664257287979126,
      "epoch": 4.905859750240154,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028511709328643206,
      "loss": 4.8229,
      "mean_token_accuracy": 0.24096553772687912,
      "num_tokens": 117106219.0,
      "step": 51070
    },
    {
      "entropy": 5.526396417617798,
      "epoch": 4.906340057636887,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028508284428907483,
      "loss": 4.6177,
      "mean_token_accuracy": 0.26499418914318085,
      "num_tokens": 117117634.0,
      "step": 51075
    },
    {
      "entropy": 5.483735418319702,
      "epoch": 4.906820365033622,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002850485950576238,
      "loss": 4.6133,
      "mean_token_accuracy": 0.26682575792074203,
      "num_tokens": 117128655.0,
      "step": 51080
    },
    {
      "entropy": 5.5215881824493405,
      "epoch": 4.907300672430355,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002850143455928743,
      "loss": 4.6628,
      "mean_token_accuracy": 0.25844532400369646,
      "num_tokens": 117139048.0,
      "step": 51085
    },
    {
      "entropy": 5.597785139083863,
      "epoch": 4.9077809798270895,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002849800958956215,
      "loss": 4.7755,
      "mean_token_accuracy": 0.24873762130737304,
      "num_tokens": 117150960.0,
      "step": 51090
    },
    {
      "entropy": 5.554533529281616,
      "epoch": 4.908261287223823,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028494584596666043,
      "loss": 4.6643,
      "mean_token_accuracy": 0.2539591148495674,
      "num_tokens": 117162525.0,
      "step": 51095
    },
    {
      "entropy": 5.560188865661621,
      "epoch": 4.908741594620557,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028491159580678634,
      "loss": 4.6884,
      "mean_token_accuracy": 0.25919677764177323,
      "num_tokens": 117173601.0,
      "step": 51100
    },
    {
      "entropy": 5.575367403030396,
      "epoch": 4.909221902017291,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002848773454167945,
      "loss": 4.7374,
      "mean_token_accuracy": 0.25152466595172884,
      "num_tokens": 117183878.0,
      "step": 51105
    },
    {
      "entropy": 5.530104541778565,
      "epoch": 4.909702209414025,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028484309479747993,
      "loss": 4.7389,
      "mean_token_accuracy": 0.24433691799640656,
      "num_tokens": 117196021.0,
      "step": 51110
    },
    {
      "entropy": 5.533364582061767,
      "epoch": 4.9101825168107585,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028480884394963804,
      "loss": 4.686,
      "mean_token_accuracy": 0.2631233260035515,
      "num_tokens": 117207663.0,
      "step": 51115
    },
    {
      "entropy": 5.57168493270874,
      "epoch": 4.910662824207493,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002847745928740638,
      "loss": 4.7514,
      "mean_token_accuracy": 0.25046988427639005,
      "num_tokens": 117218353.0,
      "step": 51120
    },
    {
      "entropy": 5.570907354354858,
      "epoch": 4.911143131604227,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028474034157155263,
      "loss": 4.7015,
      "mean_token_accuracy": 0.2564591497182846,
      "num_tokens": 117230941.0,
      "step": 51125
    },
    {
      "entropy": 5.5881084442138675,
      "epoch": 4.9116234390009605,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028470609004289964,
      "loss": 4.6451,
      "mean_token_accuracy": 0.2647249087691307,
      "num_tokens": 117242592.0,
      "step": 51130
    },
    {
      "entropy": 5.494238471984863,
      "epoch": 4.912103746397695,
      "grad_norm": 2.046875,
      "learning_rate": 0.0002846718382889001,
      "loss": 4.6347,
      "mean_token_accuracy": 0.2698680341243744,
      "num_tokens": 117254343.0,
      "step": 51135
    },
    {
      "entropy": 5.453977108001709,
      "epoch": 4.912584053794428,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002846375863103491,
      "loss": 4.5946,
      "mean_token_accuracy": 0.2632354602217674,
      "num_tokens": 117266227.0,
      "step": 51140
    },
    {
      "entropy": 5.4391014099121096,
      "epoch": 4.913064361191163,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028460333410804193,
      "loss": 4.574,
      "mean_token_accuracy": 0.2599513649940491,
      "num_tokens": 117277838.0,
      "step": 51145
    },
    {
      "entropy": 5.46767110824585,
      "epoch": 4.913544668587896,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000284569081682774,
      "loss": 4.558,
      "mean_token_accuracy": 0.2710151866078377,
      "num_tokens": 117288751.0,
      "step": 51150
    },
    {
      "entropy": 5.482272052764893,
      "epoch": 4.91402497598463,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002845348290353403,
      "loss": 4.6434,
      "mean_token_accuracy": 0.26111134737730024,
      "num_tokens": 117300699.0,
      "step": 51155
    },
    {
      "entropy": 5.607781600952149,
      "epoch": 4.914505283381364,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028450057616653617,
      "loss": 4.7442,
      "mean_token_accuracy": 0.2473454013466835,
      "num_tokens": 117312274.0,
      "step": 51160
    },
    {
      "entropy": 5.5843315601348875,
      "epoch": 4.914985590778098,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000284466323077157,
      "loss": 4.7197,
      "mean_token_accuracy": 0.2549989655613899,
      "num_tokens": 117324162.0,
      "step": 51165
    },
    {
      "entropy": 5.531850957870484,
      "epoch": 4.915465898174832,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00028443206976799785,
      "loss": 4.6632,
      "mean_token_accuracy": 0.2611521080136299,
      "num_tokens": 117335271.0,
      "step": 51170
    },
    {
      "entropy": 5.535458326339722,
      "epoch": 4.915946205571566,
      "grad_norm": 1.4375,
      "learning_rate": 0.000284397816239854,
      "loss": 4.6996,
      "mean_token_accuracy": 0.2641322359442711,
      "num_tokens": 117345861.0,
      "step": 51175
    },
    {
      "entropy": 5.612810754776001,
      "epoch": 4.916426512968299,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002843635624935208,
      "loss": 4.7605,
      "mean_token_accuracy": 0.2518650308251381,
      "num_tokens": 117358050.0,
      "step": 51180
    },
    {
      "entropy": 5.51639952659607,
      "epoch": 4.916906820365034,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002843293085297935,
      "loss": 4.6616,
      "mean_token_accuracy": 0.2628836020827293,
      "num_tokens": 117369899.0,
      "step": 51185
    },
    {
      "entropy": 5.586766576766967,
      "epoch": 4.917387127761767,
      "grad_norm": 1.375,
      "learning_rate": 0.0002842950543494673,
      "loss": 4.7691,
      "mean_token_accuracy": 0.2514227405190468,
      "num_tokens": 117381887.0,
      "step": 51190
    },
    {
      "entropy": 5.572305631637573,
      "epoch": 4.917867435158501,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028426079995333756,
      "loss": 4.7211,
      "mean_token_accuracy": 0.25633364021778104,
      "num_tokens": 117392641.0,
      "step": 51195
    },
    {
      "entropy": 5.5276508808135985,
      "epoch": 4.918347742555236,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002842265453421996,
      "loss": 4.6703,
      "mean_token_accuracy": 0.2551921963691711,
      "num_tokens": 117403437.0,
      "step": 51200
    },
    {
      "entropy": 5.540388107299805,
      "epoch": 4.918828049951969,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002841922905168486,
      "loss": 4.7009,
      "mean_token_accuracy": 0.2583330810070038,
      "num_tokens": 117415388.0,
      "step": 51205
    },
    {
      "entropy": 5.5181872844696045,
      "epoch": 4.9193083573487035,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002841580354780799,
      "loss": 4.6032,
      "mean_token_accuracy": 0.26094191819429396,
      "num_tokens": 117426456.0,
      "step": 51210
    },
    {
      "entropy": 5.623219776153564,
      "epoch": 4.919788664745437,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028412378022668876,
      "loss": 4.7723,
      "mean_token_accuracy": 0.25183276981115343,
      "num_tokens": 117438919.0,
      "step": 51215
    },
    {
      "entropy": 5.552955055236817,
      "epoch": 4.920268972142171,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028408952476347064,
      "loss": 4.6796,
      "mean_token_accuracy": 0.26188296228647234,
      "num_tokens": 117449536.0,
      "step": 51220
    },
    {
      "entropy": 5.539761638641357,
      "epoch": 4.920749279538905,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028405526908922076,
      "loss": 4.693,
      "mean_token_accuracy": 0.2616008251905441,
      "num_tokens": 117461174.0,
      "step": 51225
    },
    {
      "entropy": 5.531009006500244,
      "epoch": 4.921229586935639,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028402101320473437,
      "loss": 4.6392,
      "mean_token_accuracy": 0.26709650456905365,
      "num_tokens": 117472645.0,
      "step": 51230
    },
    {
      "entropy": 5.517780494689942,
      "epoch": 4.9217098943323725,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028398675711080684,
      "loss": 4.6243,
      "mean_token_accuracy": 0.2534431263804436,
      "num_tokens": 117483187.0,
      "step": 51235
    },
    {
      "entropy": 5.487917470932007,
      "epoch": 4.922190201729107,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002839525008082335,
      "loss": 4.6365,
      "mean_token_accuracy": 0.2655090928077698,
      "num_tokens": 117493860.0,
      "step": 51240
    },
    {
      "entropy": 5.514786911010742,
      "epoch": 4.92267050912584,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028391824429780966,
      "loss": 4.6846,
      "mean_token_accuracy": 0.24897494465112685,
      "num_tokens": 117505892.0,
      "step": 51245
    },
    {
      "entropy": 5.445349407196045,
      "epoch": 4.923150816522575,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002838839875803307,
      "loss": 4.5966,
      "mean_token_accuracy": 0.26159733682870867,
      "num_tokens": 117517161.0,
      "step": 51250
    },
    {
      "entropy": 5.488727188110351,
      "epoch": 4.923631123919308,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002838497306565919,
      "loss": 4.6595,
      "mean_token_accuracy": 0.2586659610271454,
      "num_tokens": 117528828.0,
      "step": 51255
    },
    {
      "entropy": 5.539328002929688,
      "epoch": 4.924111431316042,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028381547352738867,
      "loss": 4.6695,
      "mean_token_accuracy": 0.2560058206319809,
      "num_tokens": 117540319.0,
      "step": 51260
    },
    {
      "entropy": 5.535774898529053,
      "epoch": 4.924591738712776,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028378121619351625,
      "loss": 4.6698,
      "mean_token_accuracy": 0.2583089619874954,
      "num_tokens": 117550375.0,
      "step": 51265
    },
    {
      "entropy": 5.462867498397827,
      "epoch": 4.92507204610951,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028374695865577014,
      "loss": 4.6328,
      "mean_token_accuracy": 0.26197291165590286,
      "num_tokens": 117561383.0,
      "step": 51270
    },
    {
      "entropy": 5.565555763244629,
      "epoch": 4.925552353506244,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002837127009149456,
      "loss": 4.6803,
      "mean_token_accuracy": 0.25302226692438123,
      "num_tokens": 117572280.0,
      "step": 51275
    },
    {
      "entropy": 5.554919576644897,
      "epoch": 4.926032660902978,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028367844297183797,
      "loss": 4.7309,
      "mean_token_accuracy": 0.25317842364311216,
      "num_tokens": 117584497.0,
      "step": 51280
    },
    {
      "entropy": 5.520206546783447,
      "epoch": 4.926512968299712,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028364418482724275,
      "loss": 4.5744,
      "mean_token_accuracy": 0.2657239347696304,
      "num_tokens": 117595199.0,
      "step": 51285
    },
    {
      "entropy": 5.599620199203491,
      "epoch": 4.926993275696446,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028360992648195516,
      "loss": 4.7935,
      "mean_token_accuracy": 0.24663035124540328,
      "num_tokens": 117606746.0,
      "step": 51290
    },
    {
      "entropy": 5.520365810394287,
      "epoch": 4.92747358309318,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002835756679367707,
      "loss": 4.7141,
      "mean_token_accuracy": 0.2630465865135193,
      "num_tokens": 117619059.0,
      "step": 51295
    },
    {
      "entropy": 5.501797103881836,
      "epoch": 4.927953890489913,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002835414091924847,
      "loss": 4.6066,
      "mean_token_accuracy": 0.26156901121139525,
      "num_tokens": 117633006.0,
      "step": 51300
    },
    {
      "entropy": 5.578409481048584,
      "epoch": 4.928434197886648,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028350715024989245,
      "loss": 4.7288,
      "mean_token_accuracy": 0.2569585725665092,
      "num_tokens": 117643987.0,
      "step": 51305
    },
    {
      "entropy": 5.571826457977295,
      "epoch": 4.928914505283381,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002834728911097895,
      "loss": 4.6792,
      "mean_token_accuracy": 0.2597357288002968,
      "num_tokens": 117655251.0,
      "step": 51310
    },
    {
      "entropy": 5.551329326629639,
      "epoch": 4.9293948126801155,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028343863177297113,
      "loss": 4.7338,
      "mean_token_accuracy": 0.26101367324590685,
      "num_tokens": 117666970.0,
      "step": 51315
    },
    {
      "entropy": 5.606207466125488,
      "epoch": 4.929875120076849,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002834043722402328,
      "loss": 4.7721,
      "mean_token_accuracy": 0.24822360128164292,
      "num_tokens": 117678855.0,
      "step": 51320
    },
    {
      "entropy": 5.5212780952453615,
      "epoch": 4.930355427473583,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028337011251236994,
      "loss": 4.6255,
      "mean_token_accuracy": 0.26105307042598724,
      "num_tokens": 117689289.0,
      "step": 51325
    },
    {
      "entropy": 5.479053020477295,
      "epoch": 4.930835734870317,
      "grad_norm": 1.25,
      "learning_rate": 0.00028333585259017794,
      "loss": 4.6308,
      "mean_token_accuracy": 0.25932188928127287,
      "num_tokens": 117699872.0,
      "step": 51330
    },
    {
      "entropy": 5.546660423278809,
      "epoch": 4.931316042267051,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028330159247445217,
      "loss": 4.7034,
      "mean_token_accuracy": 0.26212421208620074,
      "num_tokens": 117711569.0,
      "step": 51335
    },
    {
      "entropy": 5.544947385787964,
      "epoch": 4.931796349663784,
      "grad_norm": 1.390625,
      "learning_rate": 0.000283267332165988,
      "loss": 4.7489,
      "mean_token_accuracy": 0.2534447446465492,
      "num_tokens": 117722654.0,
      "step": 51340
    },
    {
      "entropy": 5.667176866531372,
      "epoch": 4.932276657060519,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000283233071665581,
      "loss": 4.7658,
      "mean_token_accuracy": 0.2515133395791054,
      "num_tokens": 117733658.0,
      "step": 51345
    },
    {
      "entropy": 5.599289798736573,
      "epoch": 4.932756964457253,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002831988109740265,
      "loss": 4.6983,
      "mean_token_accuracy": 0.2599704757332802,
      "num_tokens": 117744911.0,
      "step": 51350
    },
    {
      "entropy": 5.549695253372192,
      "epoch": 4.9332372718539865,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028316455009211994,
      "loss": 4.6876,
      "mean_token_accuracy": 0.2581245735287666,
      "num_tokens": 117756759.0,
      "step": 51355
    },
    {
      "entropy": 5.434660387039185,
      "epoch": 4.93371757925072,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028313028902065684,
      "loss": 4.6254,
      "mean_token_accuracy": 0.2713098779320717,
      "num_tokens": 117767791.0,
      "step": 51360
    },
    {
      "entropy": 5.509320306777954,
      "epoch": 4.934197886647454,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002830960277604325,
      "loss": 4.6102,
      "mean_token_accuracy": 0.2564079463481903,
      "num_tokens": 117778384.0,
      "step": 51365
    },
    {
      "entropy": 5.520854902267456,
      "epoch": 4.934678194044189,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028306176631224243,
      "loss": 4.6387,
      "mean_token_accuracy": 0.26084153801202775,
      "num_tokens": 117789570.0,
      "step": 51370
    },
    {
      "entropy": 5.488220167160034,
      "epoch": 4.935158501440922,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028302750467688214,
      "loss": 4.5864,
      "mean_token_accuracy": 0.2694110631942749,
      "num_tokens": 117799829.0,
      "step": 51375
    },
    {
      "entropy": 5.511489677429199,
      "epoch": 4.935638808837656,
      "grad_norm": 1.328125,
      "learning_rate": 0.000282993242855147,
      "loss": 4.6787,
      "mean_token_accuracy": 0.25530096143484116,
      "num_tokens": 117812345.0,
      "step": 51380
    },
    {
      "entropy": 5.528390979766845,
      "epoch": 4.93611911623439,
      "grad_norm": 1.25,
      "learning_rate": 0.00028295898084783257,
      "loss": 4.6651,
      "mean_token_accuracy": 0.25470546931028365,
      "num_tokens": 117823621.0,
      "step": 51385
    },
    {
      "entropy": 5.482465839385986,
      "epoch": 4.936599423631124,
      "grad_norm": 1.25,
      "learning_rate": 0.00028292471865573414,
      "loss": 4.6642,
      "mean_token_accuracy": 0.2619722053408623,
      "num_tokens": 117835476.0,
      "step": 51390
    },
    {
      "entropy": 5.597140741348267,
      "epoch": 4.937079731027858,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002828904562796473,
      "loss": 4.7461,
      "mean_token_accuracy": 0.2565469443798065,
      "num_tokens": 117847361.0,
      "step": 51395
    },
    {
      "entropy": 5.5708287239074705,
      "epoch": 4.937560038424592,
      "grad_norm": 1.375,
      "learning_rate": 0.0002828561937203675,
      "loss": 4.6666,
      "mean_token_accuracy": 0.26173228174448016,
      "num_tokens": 117858579.0,
      "step": 51400
    },
    {
      "entropy": 5.5987907409667965,
      "epoch": 4.938040345821325,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028282193097869024,
      "loss": 4.7868,
      "mean_token_accuracy": 0.2503933683037758,
      "num_tokens": 117870156.0,
      "step": 51405
    },
    {
      "entropy": 5.604306888580322,
      "epoch": 4.93852065321806,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002827876680554109,
      "loss": 4.7378,
      "mean_token_accuracy": 0.25139549523592,
      "num_tokens": 117882198.0,
      "step": 51410
    },
    {
      "entropy": 5.504951286315918,
      "epoch": 4.939000960614793,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002827534049513251,
      "loss": 4.6074,
      "mean_token_accuracy": 0.26768241375684737,
      "num_tokens": 117892127.0,
      "step": 51415
    },
    {
      "entropy": 5.56313042640686,
      "epoch": 4.939481268011527,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028271914166722826,
      "loss": 4.7109,
      "mean_token_accuracy": 0.2538334161043167,
      "num_tokens": 117903662.0,
      "step": 51420
    },
    {
      "entropy": 5.52234787940979,
      "epoch": 4.939961575408262,
      "grad_norm": 1.25,
      "learning_rate": 0.0002826848782039158,
      "loss": 4.6469,
      "mean_token_accuracy": 0.26016461104154587,
      "num_tokens": 117915163.0,
      "step": 51425
    },
    {
      "entropy": 5.539756774902344,
      "epoch": 4.940441882804995,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028265061456218333,
      "loss": 4.72,
      "mean_token_accuracy": 0.2527052819728851,
      "num_tokens": 117927321.0,
      "step": 51430
    },
    {
      "entropy": 5.589084911346435,
      "epoch": 4.940922190201729,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028261635074282635,
      "loss": 4.7249,
      "mean_token_accuracy": 0.2494872733950615,
      "num_tokens": 117938528.0,
      "step": 51435
    },
    {
      "entropy": 5.568204307556153,
      "epoch": 4.941402497598463,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002825820867466403,
      "loss": 4.7386,
      "mean_token_accuracy": 0.2493496999144554,
      "num_tokens": 117949832.0,
      "step": 51440
    },
    {
      "entropy": 5.489717674255371,
      "epoch": 4.941882804995197,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002825478225744206,
      "loss": 4.6875,
      "mean_token_accuracy": 0.25657597482204436,
      "num_tokens": 117960606.0,
      "step": 51445
    },
    {
      "entropy": 5.553892755508423,
      "epoch": 4.942363112391931,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028251355822696294,
      "loss": 4.6375,
      "mean_token_accuracy": 0.2620597779750824,
      "num_tokens": 117972729.0,
      "step": 51450
    },
    {
      "entropy": 5.554510402679443,
      "epoch": 4.942843419788665,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002824792937050628,
      "loss": 4.5875,
      "mean_token_accuracy": 0.27297150492668154,
      "num_tokens": 117984723.0,
      "step": 51455
    },
    {
      "entropy": 5.575796508789063,
      "epoch": 4.9433237271853985,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002824450290095157,
      "loss": 4.7562,
      "mean_token_accuracy": 0.25235749036073685,
      "num_tokens": 117995483.0,
      "step": 51460
    },
    {
      "entropy": 5.461840629577637,
      "epoch": 4.943804034582133,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002824107641411171,
      "loss": 4.6816,
      "mean_token_accuracy": 0.2533165365457535,
      "num_tokens": 118007126.0,
      "step": 51465
    },
    {
      "entropy": 5.555578231811523,
      "epoch": 4.944284341978866,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002823764991006625,
      "loss": 4.7869,
      "mean_token_accuracy": 0.2469429552555084,
      "num_tokens": 118019582.0,
      "step": 51470
    },
    {
      "entropy": 5.571907615661621,
      "epoch": 4.944764649375601,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002823422338889476,
      "loss": 4.7469,
      "mean_token_accuracy": 0.2529526188969612,
      "num_tokens": 118031228.0,
      "step": 51475
    },
    {
      "entropy": 5.585746383666992,
      "epoch": 4.945244956772334,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002823079685067677,
      "loss": 4.6946,
      "mean_token_accuracy": 0.2552746504545212,
      "num_tokens": 118041904.0,
      "step": 51480
    },
    {
      "entropy": 5.515159273147583,
      "epoch": 4.945725264169068,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002822737029549186,
      "loss": 4.6601,
      "mean_token_accuracy": 0.26399490237236023,
      "num_tokens": 118053424.0,
      "step": 51485
    },
    {
      "entropy": 5.485728120803833,
      "epoch": 4.946205571565802,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002822394372341957,
      "loss": 4.6901,
      "mean_token_accuracy": 0.2645538836717606,
      "num_tokens": 118064736.0,
      "step": 51490
    },
    {
      "entropy": 5.563724517822266,
      "epoch": 4.946685878962536,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028220517134539447,
      "loss": 4.6466,
      "mean_token_accuracy": 0.25376574993133544,
      "num_tokens": 118075445.0,
      "step": 51495
    },
    {
      "entropy": 5.506791353225708,
      "epoch": 4.94716618635927,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028217090528931063,
      "loss": 4.6745,
      "mean_token_accuracy": 0.2612826868891716,
      "num_tokens": 118087668.0,
      "step": 51500
    },
    {
      "entropy": 5.518533563613891,
      "epoch": 4.947646493756004,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002821366390667397,
      "loss": 4.667,
      "mean_token_accuracy": 0.2573585197329521,
      "num_tokens": 118099623.0,
      "step": 51505
    },
    {
      "entropy": 5.5387043952941895,
      "epoch": 4.948126801152737,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002821023726784771,
      "loss": 4.8083,
      "mean_token_accuracy": 0.24955426752567292,
      "num_tokens": 118111729.0,
      "step": 51510
    },
    {
      "entropy": 5.565611410140991,
      "epoch": 4.948607108549472,
      "grad_norm": 1.375,
      "learning_rate": 0.0002820681061253185,
      "loss": 4.6905,
      "mean_token_accuracy": 0.26347003430128096,
      "num_tokens": 118122352.0,
      "step": 51515
    },
    {
      "entropy": 5.551986360549927,
      "epoch": 4.949087415946206,
      "grad_norm": 1.375,
      "learning_rate": 0.00028203383940805956,
      "loss": 4.7488,
      "mean_token_accuracy": 0.24439332634210587,
      "num_tokens": 118133328.0,
      "step": 51520
    },
    {
      "entropy": 5.565909147262573,
      "epoch": 4.949567723342939,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002819995725274957,
      "loss": 4.6909,
      "mean_token_accuracy": 0.2573073446750641,
      "num_tokens": 118143854.0,
      "step": 51525
    },
    {
      "entropy": 5.479266500473022,
      "epoch": 4.950048030739674,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028196530548442255,
      "loss": 4.6122,
      "mean_token_accuracy": 0.2657322704792023,
      "num_tokens": 118153966.0,
      "step": 51530
    },
    {
      "entropy": 5.563882780075073,
      "epoch": 4.950528338136407,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028193103827963567,
      "loss": 4.6557,
      "mean_token_accuracy": 0.25617633014917374,
      "num_tokens": 118166325.0,
      "step": 51535
    },
    {
      "entropy": 5.430317831039429,
      "epoch": 4.9510086455331415,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028189677091393074,
      "loss": 4.584,
      "mean_token_accuracy": 0.2651230603456497,
      "num_tokens": 118177838.0,
      "step": 51540
    },
    {
      "entropy": 5.503604221343994,
      "epoch": 4.951488952929875,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002818625033881032,
      "loss": 4.6272,
      "mean_token_accuracy": 0.26626219004392626,
      "num_tokens": 118188802.0,
      "step": 51545
    },
    {
      "entropy": 5.542930030822754,
      "epoch": 4.951969260326609,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028182823570294867,
      "loss": 4.7042,
      "mean_token_accuracy": 0.2523477151989937,
      "num_tokens": 118198912.0,
      "step": 51550
    },
    {
      "entropy": 5.54972243309021,
      "epoch": 4.952449567723343,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028179396785926286,
      "loss": 4.6596,
      "mean_token_accuracy": 0.25798899829387667,
      "num_tokens": 118209491.0,
      "step": 51555
    },
    {
      "entropy": 5.514846134185791,
      "epoch": 4.952929875120077,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028175969985784135,
      "loss": 4.6593,
      "mean_token_accuracy": 0.25963545590639114,
      "num_tokens": 118223126.0,
      "step": 51560
    },
    {
      "entropy": 5.481546115875244,
      "epoch": 4.95341018251681,
      "grad_norm": 1.25,
      "learning_rate": 0.0002817254316994795,
      "loss": 4.6579,
      "mean_token_accuracy": 0.25876267403364184,
      "num_tokens": 118234225.0,
      "step": 51565
    },
    {
      "entropy": 5.564443969726563,
      "epoch": 4.953890489913545,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028169116338497323,
      "loss": 4.6971,
      "mean_token_accuracy": 0.2566273510456085,
      "num_tokens": 118245456.0,
      "step": 51570
    },
    {
      "entropy": 5.559312248229981,
      "epoch": 4.954370797310279,
      "grad_norm": 1.25,
      "learning_rate": 0.00028165689491511806,
      "loss": 4.672,
      "mean_token_accuracy": 0.2601869359612465,
      "num_tokens": 118257235.0,
      "step": 51575
    },
    {
      "entropy": 5.541658878326416,
      "epoch": 4.9548511047070125,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002816226262907094,
      "loss": 4.5967,
      "mean_token_accuracy": 0.26444386690855026,
      "num_tokens": 118266435.0,
      "step": 51580
    },
    {
      "entropy": 5.528103590011597,
      "epoch": 4.955331412103746,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028158835751254316,
      "loss": 4.6904,
      "mean_token_accuracy": 0.25494868606328963,
      "num_tokens": 118277384.0,
      "step": 51585
    },
    {
      "entropy": 5.563806247711182,
      "epoch": 4.95581171950048,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002815540885814147,
      "loss": 4.7586,
      "mean_token_accuracy": 0.2570146694779396,
      "num_tokens": 118289144.0,
      "step": 51590
    },
    {
      "entropy": 5.460924339294434,
      "epoch": 4.956292026897215,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028151981949811985,
      "loss": 4.5871,
      "mean_token_accuracy": 0.26174538731575014,
      "num_tokens": 118301469.0,
      "step": 51595
    },
    {
      "entropy": 5.524900197982788,
      "epoch": 4.956772334293948,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028148555026345415,
      "loss": 4.617,
      "mean_token_accuracy": 0.2560672998428345,
      "num_tokens": 118313883.0,
      "step": 51600
    },
    {
      "entropy": 5.5286657333374025,
      "epoch": 4.957252641690682,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002814512808782132,
      "loss": 4.6979,
      "mean_token_accuracy": 0.2596608877182007,
      "num_tokens": 118325263.0,
      "step": 51605
    },
    {
      "entropy": 5.529976654052734,
      "epoch": 4.957732949087416,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028141701134319267,
      "loss": 4.6739,
      "mean_token_accuracy": 0.2582085967063904,
      "num_tokens": 118336000.0,
      "step": 51610
    },
    {
      "entropy": 5.544608545303345,
      "epoch": 4.95821325648415,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002813827416591882,
      "loss": 4.6677,
      "mean_token_accuracy": 0.2592195048928261,
      "num_tokens": 118347982.0,
      "step": 51615
    },
    {
      "entropy": 5.544896841049194,
      "epoch": 4.958693563880884,
      "grad_norm": 1.375,
      "learning_rate": 0.0002813484718269954,
      "loss": 4.6822,
      "mean_token_accuracy": 0.25938306003808975,
      "num_tokens": 118358595.0,
      "step": 51620
    },
    {
      "entropy": 5.604404020309448,
      "epoch": 4.959173871277618,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028131420184740995,
      "loss": 4.7459,
      "mean_token_accuracy": 0.2552385166287422,
      "num_tokens": 118370764.0,
      "step": 51625
    },
    {
      "entropy": 5.6147643566131595,
      "epoch": 4.959654178674351,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002812799317212275,
      "loss": 4.7001,
      "mean_token_accuracy": 0.25465650856494904,
      "num_tokens": 118382160.0,
      "step": 51630
    },
    {
      "entropy": 5.512167453765869,
      "epoch": 4.960134486071086,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028124566144924364,
      "loss": 4.6542,
      "mean_token_accuracy": 0.25913476645946504,
      "num_tokens": 118393663.0,
      "step": 51635
    },
    {
      "entropy": 5.425873231887818,
      "epoch": 4.960614793467819,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002812113910322541,
      "loss": 4.6119,
      "mean_token_accuracy": 0.2580916821956635,
      "num_tokens": 118404377.0,
      "step": 51640
    },
    {
      "entropy": 5.469603967666626,
      "epoch": 4.961095100864553,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028117712047105454,
      "loss": 4.6228,
      "mean_token_accuracy": 0.2668896734714508,
      "num_tokens": 118416034.0,
      "step": 51645
    },
    {
      "entropy": 5.600849437713623,
      "epoch": 4.961575408261287,
      "grad_norm": 1.25,
      "learning_rate": 0.0002811428497664405,
      "loss": 4.6936,
      "mean_token_accuracy": 0.25920410454273224,
      "num_tokens": 118426779.0,
      "step": 51650
    },
    {
      "entropy": 5.487762498855591,
      "epoch": 4.962055715658021,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002811085789192078,
      "loss": 4.5749,
      "mean_token_accuracy": 0.268405544757843,
      "num_tokens": 118438789.0,
      "step": 51655
    },
    {
      "entropy": 5.5540525913238525,
      "epoch": 4.962536023054755,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028107430793015205,
      "loss": 4.7591,
      "mean_token_accuracy": 0.25037663280963895,
      "num_tokens": 118450168.0,
      "step": 51660
    },
    {
      "entropy": 5.548331928253174,
      "epoch": 4.963016330451489,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002810400368000688,
      "loss": 4.7498,
      "mean_token_accuracy": 0.24609206169843673,
      "num_tokens": 118462223.0,
      "step": 51665
    },
    {
      "entropy": 5.603349828720093,
      "epoch": 4.963496637848223,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002810057655297539,
      "loss": 4.7224,
      "mean_token_accuracy": 0.26264663487672807,
      "num_tokens": 118472412.0,
      "step": 51670
    },
    {
      "entropy": 5.510373926162719,
      "epoch": 4.963976945244957,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000280971494120003,
      "loss": 4.6186,
      "mean_token_accuracy": 0.2668510004878044,
      "num_tokens": 118483809.0,
      "step": 51675
    },
    {
      "entropy": 5.539427471160889,
      "epoch": 4.964457252641691,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002809372225716117,
      "loss": 4.7464,
      "mean_token_accuracy": 0.24477775692939757,
      "num_tokens": 118495665.0,
      "step": 51680
    },
    {
      "entropy": 5.535234642028809,
      "epoch": 4.9649375600384245,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002809029508853757,
      "loss": 4.6863,
      "mean_token_accuracy": 0.2561086490750313,
      "num_tokens": 118507976.0,
      "step": 51685
    },
    {
      "entropy": 5.549560689926148,
      "epoch": 4.965417867435159,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002808686790620907,
      "loss": 4.7317,
      "mean_token_accuracy": 0.25855873972177507,
      "num_tokens": 118518844.0,
      "step": 51690
    },
    {
      "entropy": 5.549389743804932,
      "epoch": 4.965898174831892,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00028083440710255245,
      "loss": 4.6894,
      "mean_token_accuracy": 0.25390584468841554,
      "num_tokens": 118529856.0,
      "step": 51695
    },
    {
      "entropy": 5.4617517471313475,
      "epoch": 4.9663784822286265,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002808001350075565,
      "loss": 4.5817,
      "mean_token_accuracy": 0.2638910785317421,
      "num_tokens": 118540544.0,
      "step": 51700
    },
    {
      "entropy": 5.550138473510742,
      "epoch": 4.96685878962536,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028076586277789875,
      "loss": 4.7113,
      "mean_token_accuracy": 0.2583997964859009,
      "num_tokens": 118551233.0,
      "step": 51705
    },
    {
      "entropy": 5.545741558074951,
      "epoch": 4.967339097022094,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002807315904143747,
      "loss": 4.6481,
      "mean_token_accuracy": 0.2585466459393501,
      "num_tokens": 118562241.0,
      "step": 51710
    },
    {
      "entropy": 5.58536171913147,
      "epoch": 4.967819404418828,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002806973179177802,
      "loss": 4.725,
      "mean_token_accuracy": 0.25938562154769895,
      "num_tokens": 118573961.0,
      "step": 51715
    },
    {
      "entropy": 5.531906986236573,
      "epoch": 4.968299711815562,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028066304528891087,
      "loss": 4.7131,
      "mean_token_accuracy": 0.2569832786917686,
      "num_tokens": 118585443.0,
      "step": 51720
    },
    {
      "entropy": 5.544739627838135,
      "epoch": 4.9687800192122955,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002806287725285625,
      "loss": 4.7499,
      "mean_token_accuracy": 0.25496858805418016,
      "num_tokens": 118596702.0,
      "step": 51725
    },
    {
      "entropy": 5.539208555221558,
      "epoch": 4.96926032660903,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002805944996375306,
      "loss": 4.7179,
      "mean_token_accuracy": 0.259472993016243,
      "num_tokens": 118608117.0,
      "step": 51730
    },
    {
      "entropy": 5.6288337230682375,
      "epoch": 4.969740634005763,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002805602266166111,
      "loss": 4.7373,
      "mean_token_accuracy": 0.2573870480060577,
      "num_tokens": 118619103.0,
      "step": 51735
    },
    {
      "entropy": 5.525048494338989,
      "epoch": 4.970220941402498,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002805259534665998,
      "loss": 4.686,
      "mean_token_accuracy": 0.2528285324573517,
      "num_tokens": 118631060.0,
      "step": 51740
    },
    {
      "entropy": 5.4538702964782715,
      "epoch": 4.970701248799232,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028049168018829215,
      "loss": 4.6004,
      "mean_token_accuracy": 0.264948745071888,
      "num_tokens": 118642447.0,
      "step": 51745
    },
    {
      "entropy": 5.550220060348511,
      "epoch": 4.971181556195965,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002804574067824839,
      "loss": 4.6921,
      "mean_token_accuracy": 0.25359936654567716,
      "num_tokens": 118654614.0,
      "step": 51750
    },
    {
      "entropy": 5.641423082351684,
      "epoch": 4.9716618635927,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000280423133249971,
      "loss": 4.7672,
      "mean_token_accuracy": 0.24738623052835465,
      "num_tokens": 118666630.0,
      "step": 51755
    },
    {
      "entropy": 5.514713764190674,
      "epoch": 4.972142170989433,
      "grad_norm": 1.296875,
      "learning_rate": 0.000280388859591549,
      "loss": 4.6566,
      "mean_token_accuracy": 0.26751860082149503,
      "num_tokens": 118677263.0,
      "step": 51760
    },
    {
      "entropy": 5.443077898025512,
      "epoch": 4.972622478386167,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028035458580801366,
      "loss": 4.5874,
      "mean_token_accuracy": 0.2669768914580345,
      "num_tokens": 118687639.0,
      "step": 51765
    },
    {
      "entropy": 5.5635992050170895,
      "epoch": 4.973102785782901,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002803203119001608,
      "loss": 4.66,
      "mean_token_accuracy": 0.25658543556928637,
      "num_tokens": 118699349.0,
      "step": 51770
    },
    {
      "entropy": 5.602519559860229,
      "epoch": 4.973583093179635,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028028603786878605,
      "loss": 4.7432,
      "mean_token_accuracy": 0.2518546223640442,
      "num_tokens": 118709948.0,
      "step": 51775
    },
    {
      "entropy": 5.504171419143677,
      "epoch": 4.974063400576369,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002802517637146852,
      "loss": 4.6444,
      "mean_token_accuracy": 0.2602444216609001,
      "num_tokens": 118721671.0,
      "step": 51780
    },
    {
      "entropy": 5.483778047561645,
      "epoch": 4.974543707973103,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028021748943865397,
      "loss": 4.5873,
      "mean_token_accuracy": 0.2653959825634956,
      "num_tokens": 118733459.0,
      "step": 51785
    },
    {
      "entropy": 5.482273578643799,
      "epoch": 4.975024015369836,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002801832150414881,
      "loss": 4.5689,
      "mean_token_accuracy": 0.2650371327996254,
      "num_tokens": 118743871.0,
      "step": 51790
    },
    {
      "entropy": 5.431676197052002,
      "epoch": 4.975504322766571,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002801489405239835,
      "loss": 4.6044,
      "mean_token_accuracy": 0.25979074090719223,
      "num_tokens": 118755199.0,
      "step": 51795
    },
    {
      "entropy": 5.499019479751587,
      "epoch": 4.975984630163304,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002801146658869357,
      "loss": 4.6356,
      "mean_token_accuracy": 0.2596592873334885,
      "num_tokens": 118766286.0,
      "step": 51800
    },
    {
      "entropy": 5.5290834426879885,
      "epoch": 4.9764649375600385,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028008039113114054,
      "loss": 4.6291,
      "mean_token_accuracy": 0.2619482487440109,
      "num_tokens": 118777289.0,
      "step": 51805
    },
    {
      "entropy": 5.542686653137207,
      "epoch": 4.976945244956772,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002800461162573938,
      "loss": 4.6723,
      "mean_token_accuracy": 0.26175862550735474,
      "num_tokens": 118789621.0,
      "step": 51810
    },
    {
      "entropy": 5.474313068389892,
      "epoch": 4.977425552353506,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028001184126649124,
      "loss": 4.674,
      "mean_token_accuracy": 0.2576261341571808,
      "num_tokens": 118800716.0,
      "step": 51815
    },
    {
      "entropy": 5.554514741897583,
      "epoch": 4.977905859750241,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002799775661592286,
      "loss": 4.7377,
      "mean_token_accuracy": 0.2501810610294342,
      "num_tokens": 118810523.0,
      "step": 51820
    },
    {
      "entropy": 5.564634466171265,
      "epoch": 4.978386167146974,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002799432909364016,
      "loss": 4.7459,
      "mean_token_accuracy": 0.25247587263584137,
      "num_tokens": 118822502.0,
      "step": 51825
    },
    {
      "entropy": 5.478321170806884,
      "epoch": 4.978866474543708,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027990901559880616,
      "loss": 4.592,
      "mean_token_accuracy": 0.26180538684129717,
      "num_tokens": 118834288.0,
      "step": 51830
    },
    {
      "entropy": 5.606925773620605,
      "epoch": 4.979346781940442,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002798747401472379,
      "loss": 4.7969,
      "mean_token_accuracy": 0.24929216206073762,
      "num_tokens": 118846365.0,
      "step": 51835
    },
    {
      "entropy": 5.520565748214722,
      "epoch": 4.979827089337176,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027984046458249265,
      "loss": 4.706,
      "mean_token_accuracy": 0.2559014305472374,
      "num_tokens": 118858047.0,
      "step": 51840
    },
    {
      "entropy": 5.540775537490845,
      "epoch": 4.9803073967339095,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027980618890536624,
      "loss": 4.6131,
      "mean_token_accuracy": 0.2666496574878693,
      "num_tokens": 118868105.0,
      "step": 51845
    },
    {
      "entropy": 5.542187833786011,
      "epoch": 4.980787704130644,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002797719131166543,
      "loss": 4.6413,
      "mean_token_accuracy": 0.2578593701124191,
      "num_tokens": 118878505.0,
      "step": 51850
    },
    {
      "entropy": 5.501705169677734,
      "epoch": 4.981268011527377,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027973763721715275,
      "loss": 4.6492,
      "mean_token_accuracy": 0.2640196859836578,
      "num_tokens": 118890801.0,
      "step": 51855
    },
    {
      "entropy": 5.5488362312316895,
      "epoch": 4.981748318924112,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027970336120765733,
      "loss": 4.7192,
      "mean_token_accuracy": 0.2569694250822067,
      "num_tokens": 118901980.0,
      "step": 51860
    },
    {
      "entropy": 5.4904381275177006,
      "epoch": 4.982228626320845,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027966908508896384,
      "loss": 4.6666,
      "mean_token_accuracy": 0.2619649514555931,
      "num_tokens": 118913271.0,
      "step": 51865
    },
    {
      "entropy": 5.488394117355346,
      "epoch": 4.982708933717579,
      "grad_norm": 1.296875,
      "learning_rate": 0.000279634808861868,
      "loss": 4.6354,
      "mean_token_accuracy": 0.26284522712230685,
      "num_tokens": 118924682.0,
      "step": 51870
    },
    {
      "entropy": 5.576658630371094,
      "epoch": 4.983189241114313,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002796005325271658,
      "loss": 4.7732,
      "mean_token_accuracy": 0.2418246954679489,
      "num_tokens": 118937498.0,
      "step": 51875
    },
    {
      "entropy": 5.568092012405396,
      "epoch": 4.983669548511047,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027956625608565276,
      "loss": 4.7184,
      "mean_token_accuracy": 0.258978958427906,
      "num_tokens": 118949523.0,
      "step": 51880
    },
    {
      "entropy": 5.451725339889526,
      "epoch": 4.984149855907781,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027953197953812484,
      "loss": 4.6069,
      "mean_token_accuracy": 0.2601226791739464,
      "num_tokens": 118961895.0,
      "step": 51885
    },
    {
      "entropy": 5.510698413848877,
      "epoch": 4.984630163304515,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002794977028853779,
      "loss": 4.6281,
      "mean_token_accuracy": 0.2666772440075874,
      "num_tokens": 118973065.0,
      "step": 51890
    },
    {
      "entropy": 5.597929811477661,
      "epoch": 4.985110470701249,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002794634261282075,
      "loss": 4.7178,
      "mean_token_accuracy": 0.25668822079896925,
      "num_tokens": 118984231.0,
      "step": 51895
    },
    {
      "entropy": 5.5185691833496096,
      "epoch": 4.985590778097983,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002794291492674098,
      "loss": 4.6731,
      "mean_token_accuracy": 0.25278338342905043,
      "num_tokens": 118995401.0,
      "step": 51900
    },
    {
      "entropy": 5.568142938613891,
      "epoch": 4.986071085494717,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002793948723037803,
      "loss": 4.7772,
      "mean_token_accuracy": 0.2521271139383316,
      "num_tokens": 119008641.0,
      "step": 51905
    },
    {
      "entropy": 5.618983745574951,
      "epoch": 4.98655139289145,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027936059523811486,
      "loss": 4.7896,
      "mean_token_accuracy": 0.2461071565747261,
      "num_tokens": 119020049.0,
      "step": 51910
    },
    {
      "entropy": 5.630252456665039,
      "epoch": 4.987031700288185,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027932631807120933,
      "loss": 4.7407,
      "mean_token_accuracy": 0.2548306420445442,
      "num_tokens": 119031066.0,
      "step": 51915
    },
    {
      "entropy": 5.489589595794678,
      "epoch": 4.987512007684918,
      "grad_norm": 1.25,
      "learning_rate": 0.0002792920408038597,
      "loss": 4.5386,
      "mean_token_accuracy": 0.265263968706131,
      "num_tokens": 119042608.0,
      "step": 51920
    },
    {
      "entropy": 5.4772663593292235,
      "epoch": 4.9879923150816525,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002792577634368615,
      "loss": 4.6536,
      "mean_token_accuracy": 0.2593495845794678,
      "num_tokens": 119054606.0,
      "step": 51925
    },
    {
      "entropy": 5.516843795776367,
      "epoch": 4.988472622478386,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027922348597101066,
      "loss": 4.6336,
      "mean_token_accuracy": 0.2605273976922035,
      "num_tokens": 119065124.0,
      "step": 51930
    },
    {
      "entropy": 5.502050733566284,
      "epoch": 4.98895292987512,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027918920840710315,
      "loss": 4.6243,
      "mean_token_accuracy": 0.26170111894607545,
      "num_tokens": 119075179.0,
      "step": 51935
    },
    {
      "entropy": 5.516996574401856,
      "epoch": 4.989433237271854,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002791549307459345,
      "loss": 4.6349,
      "mean_token_accuracy": 0.2672989174723625,
      "num_tokens": 119087102.0,
      "step": 51940
    },
    {
      "entropy": 5.574186897277832,
      "epoch": 4.989913544668588,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002791206529883008,
      "loss": 4.7481,
      "mean_token_accuracy": 0.2552924692630768,
      "num_tokens": 119099396.0,
      "step": 51945
    },
    {
      "entropy": 5.63450870513916,
      "epoch": 4.9903938520653215,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002790863751349978,
      "loss": 4.7561,
      "mean_token_accuracy": 0.2565490588545799,
      "num_tokens": 119111214.0,
      "step": 51950
    },
    {
      "entropy": 5.5312072277069095,
      "epoch": 4.990874159462056,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002790520971868212,
      "loss": 4.718,
      "mean_token_accuracy": 0.25569494664669035,
      "num_tokens": 119122041.0,
      "step": 51955
    },
    {
      "entropy": 5.569270467758178,
      "epoch": 4.991354466858789,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027901781914456696,
      "loss": 4.6885,
      "mean_token_accuracy": 0.256528876721859,
      "num_tokens": 119133806.0,
      "step": 51960
    },
    {
      "entropy": 5.649181509017945,
      "epoch": 4.991834774255524,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027898354100903093,
      "loss": 4.7948,
      "mean_token_accuracy": 0.2557347357273102,
      "num_tokens": 119147183.0,
      "step": 51965
    },
    {
      "entropy": 5.534633541107178,
      "epoch": 4.992315081652258,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002789492627810089,
      "loss": 4.6463,
      "mean_token_accuracy": 0.2640285938978195,
      "num_tokens": 119159258.0,
      "step": 51970
    },
    {
      "entropy": 5.4646100997924805,
      "epoch": 4.992795389048991,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002789149844612967,
      "loss": 4.5963,
      "mean_token_accuracy": 0.267288638651371,
      "num_tokens": 119169991.0,
      "step": 51975
    },
    {
      "entropy": 5.536606645584106,
      "epoch": 4.993275696445725,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002788807060506902,
      "loss": 4.706,
      "mean_token_accuracy": 0.2565942332148552,
      "num_tokens": 119181448.0,
      "step": 51980
    },
    {
      "entropy": 5.542834758758545,
      "epoch": 4.993756003842459,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027884642754998524,
      "loss": 4.7644,
      "mean_token_accuracy": 0.2568905219435692,
      "num_tokens": 119192921.0,
      "step": 51985
    },
    {
      "entropy": 5.43321681022644,
      "epoch": 4.994236311239193,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027881214895997757,
      "loss": 4.5714,
      "mean_token_accuracy": 0.2660097986459732,
      "num_tokens": 119205311.0,
      "step": 51990
    },
    {
      "entropy": 5.513586473464966,
      "epoch": 4.994716618635927,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002787778702814633,
      "loss": 4.6677,
      "mean_token_accuracy": 0.25775708109140394,
      "num_tokens": 119216742.0,
      "step": 51995
    },
    {
      "entropy": 5.546374368667602,
      "epoch": 4.995196926032661,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002787435915152379,
      "loss": 4.7234,
      "mean_token_accuracy": 0.26075543761253356,
      "num_tokens": 119228785.0,
      "step": 52000
    },
    {
      "entropy": 5.553421211242676,
      "epoch": 4.995677233429395,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002787093126620975,
      "loss": 4.702,
      "mean_token_accuracy": 0.25553335547447203,
      "num_tokens": 119240279.0,
      "step": 52005
    },
    {
      "entropy": 5.544262886047363,
      "epoch": 4.996157540826129,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027867503372283793,
      "loss": 4.6574,
      "mean_token_accuracy": 0.25307015627622603,
      "num_tokens": 119251865.0,
      "step": 52010
    },
    {
      "entropy": 5.555152368545532,
      "epoch": 4.996637848222862,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002786407546982549,
      "loss": 4.6875,
      "mean_token_accuracy": 0.2604197099804878,
      "num_tokens": 119264050.0,
      "step": 52015
    },
    {
      "entropy": 5.5496964931488035,
      "epoch": 4.997118155619597,
      "grad_norm": 1.375,
      "learning_rate": 0.00027860647558914444,
      "loss": 4.7554,
      "mean_token_accuracy": 0.25983370989561083,
      "num_tokens": 119275215.0,
      "step": 52020
    },
    {
      "entropy": 5.47252631187439,
      "epoch": 4.99759846301633,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002785721963963023,
      "loss": 4.5498,
      "mean_token_accuracy": 0.2731470733880997,
      "num_tokens": 119286879.0,
      "step": 52025
    },
    {
      "entropy": 5.517118310928344,
      "epoch": 4.9980787704130645,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027853791712052436,
      "loss": 4.6395,
      "mean_token_accuracy": 0.26311379969120025,
      "num_tokens": 119297238.0,
      "step": 52030
    },
    {
      "entropy": 5.469548225402832,
      "epoch": 4.998559077809798,
      "grad_norm": 1.25,
      "learning_rate": 0.0002785036377626064,
      "loss": 4.6368,
      "mean_token_accuracy": 0.2651268407702446,
      "num_tokens": 119308430.0,
      "step": 52035
    },
    {
      "entropy": 5.475651454925537,
      "epoch": 4.999039385206532,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002784693583233445,
      "loss": 4.7122,
      "mean_token_accuracy": 0.24937602430582045,
      "num_tokens": 119319693.0,
      "step": 52040
    },
    {
      "entropy": 5.608645820617676,
      "epoch": 4.999519692603267,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027843507880353434,
      "loss": 4.6949,
      "mean_token_accuracy": 0.26031134128570554,
      "num_tokens": 119331966.0,
      "step": 52045
    },
    {
      "entropy": 5.512818717956543,
      "epoch": 5.0,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027840079920397187,
      "loss": 4.6282,
      "mean_token_accuracy": 0.27048894613981245,
      "num_tokens": 119342680.0,
      "step": 52050
    },
    {
      "entropy": 5.5216131687164305,
      "epoch": 5.000480307396734,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027836651952545293,
      "loss": 4.6566,
      "mean_token_accuracy": 0.2607802331447601,
      "num_tokens": 119355096.0,
      "step": 52055
    },
    {
      "entropy": 5.538910913467407,
      "epoch": 5.000960614793468,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002783322397687734,
      "loss": 4.6943,
      "mean_token_accuracy": 0.2529942512512207,
      "num_tokens": 119367710.0,
      "step": 52060
    },
    {
      "entropy": 5.503893089294434,
      "epoch": 5.001440922190202,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002782979599347292,
      "loss": 4.6367,
      "mean_token_accuracy": 0.26240693479776384,
      "num_tokens": 119378266.0,
      "step": 52065
    },
    {
      "entropy": 5.553658580780029,
      "epoch": 5.0019212295869355,
      "grad_norm": 1.25,
      "learning_rate": 0.0002782636800241161,
      "loss": 4.6882,
      "mean_token_accuracy": 0.2572544112801552,
      "num_tokens": 119390606.0,
      "step": 52070
    },
    {
      "entropy": 5.552737855911255,
      "epoch": 5.00240153698367,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027822940003773006,
      "loss": 4.672,
      "mean_token_accuracy": 0.2534304216504097,
      "num_tokens": 119401544.0,
      "step": 52075
    },
    {
      "entropy": 5.57387146949768,
      "epoch": 5.002881844380403,
      "grad_norm": 1.375,
      "learning_rate": 0.00027819511997636696,
      "loss": 4.7073,
      "mean_token_accuracy": 0.25502718836069105,
      "num_tokens": 119412040.0,
      "step": 52080
    },
    {
      "entropy": 5.527398777008057,
      "epoch": 5.003362151777138,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002781608398408227,
      "loss": 4.6585,
      "mean_token_accuracy": 0.26742509454488755,
      "num_tokens": 119423733.0,
      "step": 52085
    },
    {
      "entropy": 5.5740992546081545,
      "epoch": 5.003842459173871,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027812655963189303,
      "loss": 4.6474,
      "mean_token_accuracy": 0.2626393258571625,
      "num_tokens": 119436029.0,
      "step": 52090
    },
    {
      "entropy": 5.483333539962769,
      "epoch": 5.004322766570605,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027809227935037396,
      "loss": 4.6198,
      "mean_token_accuracy": 0.2638164207339287,
      "num_tokens": 119447574.0,
      "step": 52095
    },
    {
      "entropy": 5.434429454803467,
      "epoch": 5.004803073967339,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027805799899706136,
      "loss": 4.609,
      "mean_token_accuracy": 0.2711408168077469,
      "num_tokens": 119459683.0,
      "step": 52100
    },
    {
      "entropy": 5.4971997261047365,
      "epoch": 5.005283381364073,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002780237185727512,
      "loss": 4.6698,
      "mean_token_accuracy": 0.26026288866996766,
      "num_tokens": 119472515.0,
      "step": 52105
    },
    {
      "entropy": 5.501448726654052,
      "epoch": 5.005763688760807,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027798943807823913,
      "loss": 4.6096,
      "mean_token_accuracy": 0.2678440406918526,
      "num_tokens": 119483121.0,
      "step": 52110
    },
    {
      "entropy": 5.524995613098144,
      "epoch": 5.006243996157541,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002779551575143212,
      "loss": 4.6566,
      "mean_token_accuracy": 0.2553727254271507,
      "num_tokens": 119494053.0,
      "step": 52115
    },
    {
      "entropy": 5.468675994873047,
      "epoch": 5.006724303554274,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027792087688179333,
      "loss": 4.5528,
      "mean_token_accuracy": 0.27176657170057295,
      "num_tokens": 119505244.0,
      "step": 52120
    },
    {
      "entropy": 5.616493988037109,
      "epoch": 5.007204610951009,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027788659618145135,
      "loss": 4.7654,
      "mean_token_accuracy": 0.25455480217933657,
      "num_tokens": 119516915.0,
      "step": 52125
    },
    {
      "entropy": 5.5489133358001705,
      "epoch": 5.007684918347743,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027785231541409126,
      "loss": 4.6405,
      "mean_token_accuracy": 0.26532520204782484,
      "num_tokens": 119528575.0,
      "step": 52130
    },
    {
      "entropy": 5.512305212020874,
      "epoch": 5.008165225744476,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002778180345805088,
      "loss": 4.6678,
      "mean_token_accuracy": 0.2613469809293747,
      "num_tokens": 119540272.0,
      "step": 52135
    },
    {
      "entropy": 5.596897315979004,
      "epoch": 5.008645533141211,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027778375368149995,
      "loss": 4.6825,
      "mean_token_accuracy": 0.252946674823761,
      "num_tokens": 119552891.0,
      "step": 52140
    },
    {
      "entropy": 5.571494007110596,
      "epoch": 5.009125840537944,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002777494727178606,
      "loss": 4.6002,
      "mean_token_accuracy": 0.2651546701788902,
      "num_tokens": 119563295.0,
      "step": 52145
    },
    {
      "entropy": 5.535918664932251,
      "epoch": 5.0096061479346785,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027771519169038665,
      "loss": 4.7273,
      "mean_token_accuracy": 0.25082520842552186,
      "num_tokens": 119575045.0,
      "step": 52150
    },
    {
      "entropy": 5.515399026870727,
      "epoch": 5.010086455331412,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000277680910599874,
      "loss": 4.7173,
      "mean_token_accuracy": 0.24970537424087524,
      "num_tokens": 119586729.0,
      "step": 52155
    },
    {
      "entropy": 5.576267576217651,
      "epoch": 5.010566762728146,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027764662944711866,
      "loss": 4.6397,
      "mean_token_accuracy": 0.2614287868142128,
      "num_tokens": 119598602.0,
      "step": 52160
    },
    {
      "entropy": 5.566324377059937,
      "epoch": 5.01104707012488,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027761234823291635,
      "loss": 4.6407,
      "mean_token_accuracy": 0.26882209032773974,
      "num_tokens": 119609144.0,
      "step": 52165
    },
    {
      "entropy": 5.497970199584961,
      "epoch": 5.011527377521614,
      "grad_norm": 1.328125,
      "learning_rate": 0.000277578066958063,
      "loss": 4.5768,
      "mean_token_accuracy": 0.2670840352773666,
      "num_tokens": 119621567.0,
      "step": 52170
    },
    {
      "entropy": 5.43648362159729,
      "epoch": 5.0120076849183475,
      "grad_norm": 1.375,
      "learning_rate": 0.00027754378562335483,
      "loss": 4.5867,
      "mean_token_accuracy": 0.2607375979423523,
      "num_tokens": 119631370.0,
      "step": 52175
    },
    {
      "entropy": 5.46486463546753,
      "epoch": 5.012487992315082,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002775095042295874,
      "loss": 4.6083,
      "mean_token_accuracy": 0.25972203761339185,
      "num_tokens": 119643499.0,
      "step": 52180
    },
    {
      "entropy": 5.49013500213623,
      "epoch": 5.012968299711815,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027747522277755666,
      "loss": 4.6405,
      "mean_token_accuracy": 0.25851071923971175,
      "num_tokens": 119656209.0,
      "step": 52185
    },
    {
      "entropy": 5.568468046188355,
      "epoch": 5.01344860710855,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002774409412680586,
      "loss": 4.6744,
      "mean_token_accuracy": 0.261203370988369,
      "num_tokens": 119667379.0,
      "step": 52190
    },
    {
      "entropy": 5.504886722564697,
      "epoch": 5.013928914505283,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002774066597018892,
      "loss": 4.6581,
      "mean_token_accuracy": 0.2636328756809235,
      "num_tokens": 119678394.0,
      "step": 52195
    },
    {
      "entropy": 5.478828811645508,
      "epoch": 5.014409221902017,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027737237807984425,
      "loss": 4.5884,
      "mean_token_accuracy": 0.2642806902527809,
      "num_tokens": 119689818.0,
      "step": 52200
    },
    {
      "entropy": 5.445979261398316,
      "epoch": 5.014889529298751,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002773380964027198,
      "loss": 4.5868,
      "mean_token_accuracy": 0.2750174641609192,
      "num_tokens": 119700768.0,
      "step": 52205
    },
    {
      "entropy": 5.47836127281189,
      "epoch": 5.015369836695485,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002773038146713116,
      "loss": 4.6794,
      "mean_token_accuracy": 0.2621482267975807,
      "num_tokens": 119712097.0,
      "step": 52210
    },
    {
      "entropy": 5.510090303421021,
      "epoch": 5.015850144092219,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027726953288641573,
      "loss": 4.6458,
      "mean_token_accuracy": 0.2644563093781471,
      "num_tokens": 119722308.0,
      "step": 52215
    },
    {
      "entropy": 5.583770751953125,
      "epoch": 5.016330451488953,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027723525104882807,
      "loss": 4.7211,
      "mean_token_accuracy": 0.25023149102926256,
      "num_tokens": 119734014.0,
      "step": 52220
    },
    {
      "entropy": 5.466789436340332,
      "epoch": 5.016810758885687,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027720096915934443,
      "loss": 4.5599,
      "mean_token_accuracy": 0.27057508528232577,
      "num_tokens": 119745552.0,
      "step": 52225
    },
    {
      "entropy": 5.486229515075683,
      "epoch": 5.017291066282421,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002771666872187609,
      "loss": 4.6212,
      "mean_token_accuracy": 0.26505357921123507,
      "num_tokens": 119756235.0,
      "step": 52230
    },
    {
      "entropy": 5.5400598526000975,
      "epoch": 5.017771373679155,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002771324052278733,
      "loss": 4.6755,
      "mean_token_accuracy": 0.26198966801166534,
      "num_tokens": 119766911.0,
      "step": 52235
    },
    {
      "entropy": 5.517800378799438,
      "epoch": 5.018251681075888,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027709812318747756,
      "loss": 4.623,
      "mean_token_accuracy": 0.2578218400478363,
      "num_tokens": 119778395.0,
      "step": 52240
    },
    {
      "entropy": 5.5787207126617435,
      "epoch": 5.018731988472623,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027706384109836964,
      "loss": 4.7568,
      "mean_token_accuracy": 0.2523998185992241,
      "num_tokens": 119790698.0,
      "step": 52245
    },
    {
      "entropy": 5.627835083007812,
      "epoch": 5.019212295869356,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002770295589613455,
      "loss": 4.7781,
      "mean_token_accuracy": 0.2459750145673752,
      "num_tokens": 119803935.0,
      "step": 52250
    },
    {
      "entropy": 5.561001491546631,
      "epoch": 5.0196926032660905,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027699527677720097,
      "loss": 4.62,
      "mean_token_accuracy": 0.25754912495613097,
      "num_tokens": 119815423.0,
      "step": 52255
    },
    {
      "entropy": 5.53338623046875,
      "epoch": 5.020172910662824,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027696099454673206,
      "loss": 4.6,
      "mean_token_accuracy": 0.26782400757074354,
      "num_tokens": 119826837.0,
      "step": 52260
    },
    {
      "entropy": 5.536273002624512,
      "epoch": 5.020653218059558,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027692671227073473,
      "loss": 4.6817,
      "mean_token_accuracy": 0.2617028817534447,
      "num_tokens": 119837214.0,
      "step": 52265
    },
    {
      "entropy": 5.562199497222901,
      "epoch": 5.021133525456292,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002768924299500048,
      "loss": 4.7163,
      "mean_token_accuracy": 0.24573484063148499,
      "num_tokens": 119846845.0,
      "step": 52270
    },
    {
      "entropy": 5.622506475448608,
      "epoch": 5.021613832853026,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002768581475853382,
      "loss": 4.7143,
      "mean_token_accuracy": 0.2589005663990974,
      "num_tokens": 119857006.0,
      "step": 52275
    },
    {
      "entropy": 5.58756799697876,
      "epoch": 5.022094140249759,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002768238651775311,
      "loss": 4.7004,
      "mean_token_accuracy": 0.2517814800143242,
      "num_tokens": 119868379.0,
      "step": 52280
    },
    {
      "entropy": 5.58413462638855,
      "epoch": 5.022574447646494,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002767895827273792,
      "loss": 4.7016,
      "mean_token_accuracy": 0.2578504025936127,
      "num_tokens": 119880474.0,
      "step": 52285
    },
    {
      "entropy": 5.526469373703003,
      "epoch": 5.023054755043228,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027675530023567844,
      "loss": 4.6002,
      "mean_token_accuracy": 0.26344412565231323,
      "num_tokens": 119890388.0,
      "step": 52290
    },
    {
      "entropy": 5.570560169219971,
      "epoch": 5.0235350624399615,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002767210177032249,
      "loss": 4.7294,
      "mean_token_accuracy": 0.26335313618183137,
      "num_tokens": 119903375.0,
      "step": 52295
    },
    {
      "entropy": 5.500209093093872,
      "epoch": 5.024015369836696,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002766867351308144,
      "loss": 4.6463,
      "mean_token_accuracy": 0.2620485842227936,
      "num_tokens": 119914458.0,
      "step": 52300
    },
    {
      "entropy": 5.53210940361023,
      "epoch": 5.024495677233429,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002766524525192429,
      "loss": 4.6513,
      "mean_token_accuracy": 0.26140987128019333,
      "num_tokens": 119925801.0,
      "step": 52305
    },
    {
      "entropy": 5.525508165359497,
      "epoch": 5.024975984630164,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002766181698693064,
      "loss": 4.6293,
      "mean_token_accuracy": 0.26602145135402677,
      "num_tokens": 119937030.0,
      "step": 52310
    },
    {
      "entropy": 5.579918956756591,
      "epoch": 5.025456292026897,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002765838871818008,
      "loss": 4.6754,
      "mean_token_accuracy": 0.26110813319683074,
      "num_tokens": 119948339.0,
      "step": 52315
    },
    {
      "entropy": 5.527939224243164,
      "epoch": 5.025936599423631,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027654960445752197,
      "loss": 4.6312,
      "mean_token_accuracy": 0.26079574376344683,
      "num_tokens": 119960157.0,
      "step": 52320
    },
    {
      "entropy": 5.546169996261597,
      "epoch": 5.026416906820365,
      "grad_norm": 1.484375,
      "learning_rate": 0.000276515321697266,
      "loss": 4.5964,
      "mean_token_accuracy": 0.25978343188762665,
      "num_tokens": 119970201.0,
      "step": 52325
    },
    {
      "entropy": 5.5873565673828125,
      "epoch": 5.026897214217099,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002764810389018287,
      "loss": 4.7214,
      "mean_token_accuracy": 0.2544189304113388,
      "num_tokens": 119981385.0,
      "step": 52330
    },
    {
      "entropy": 5.606989526748658,
      "epoch": 5.027377521613833,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002764467560720061,
      "loss": 4.7288,
      "mean_token_accuracy": 0.2589354932308197,
      "num_tokens": 119992240.0,
      "step": 52335
    },
    {
      "entropy": 5.625872278213501,
      "epoch": 5.027857829010567,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027641247320859413,
      "loss": 4.7237,
      "mean_token_accuracy": 0.25958980321884156,
      "num_tokens": 120002900.0,
      "step": 52340
    },
    {
      "entropy": 5.515445852279663,
      "epoch": 5.0283381364073,
      "grad_norm": 1.375,
      "learning_rate": 0.0002763781903123888,
      "loss": 4.6355,
      "mean_token_accuracy": 0.26399927884340285,
      "num_tokens": 120015019.0,
      "step": 52345
    },
    {
      "entropy": 5.555207252502441,
      "epoch": 5.028818443804035,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002763439073841859,
      "loss": 4.6626,
      "mean_token_accuracy": 0.25598437041044236,
      "num_tokens": 120026078.0,
      "step": 52350
    },
    {
      "entropy": 5.4439949035644535,
      "epoch": 5.029298751200768,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027630962442478144,
      "loss": 4.5589,
      "mean_token_accuracy": 0.2729287028312683,
      "num_tokens": 120039411.0,
      "step": 52355
    },
    {
      "entropy": 5.492659330368042,
      "epoch": 5.029779058597502,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002762753414349714,
      "loss": 4.6098,
      "mean_token_accuracy": 0.26824352890253067,
      "num_tokens": 120050539.0,
      "step": 52360
    },
    {
      "entropy": 5.574702215194702,
      "epoch": 5.030259365994237,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002762410584155518,
      "loss": 4.6352,
      "mean_token_accuracy": 0.2639838680624962,
      "num_tokens": 120061656.0,
      "step": 52365
    },
    {
      "entropy": 5.543430519104004,
      "epoch": 5.03073967339097,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002762067753673184,
      "loss": 4.6084,
      "mean_token_accuracy": 0.2705629736185074,
      "num_tokens": 120073475.0,
      "step": 52370
    },
    {
      "entropy": 5.512114953994751,
      "epoch": 5.0312199807877045,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002761724922910673,
      "loss": 4.567,
      "mean_token_accuracy": 0.2617191463708878,
      "num_tokens": 120083947.0,
      "step": 52375
    },
    {
      "entropy": 5.517523097991943,
      "epoch": 5.031700288184438,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027613820918759443,
      "loss": 4.6176,
      "mean_token_accuracy": 0.2606909781694412,
      "num_tokens": 120095990.0,
      "step": 52380
    },
    {
      "entropy": 5.541116952896118,
      "epoch": 5.032180595581172,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027610392605769573,
      "loss": 4.6009,
      "mean_token_accuracy": 0.26576084047555926,
      "num_tokens": 120106809.0,
      "step": 52385
    },
    {
      "entropy": 5.4543379783630375,
      "epoch": 5.032660902977906,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027606964290216715,
      "loss": 4.582,
      "mean_token_accuracy": 0.26951454430818556,
      "num_tokens": 120117908.0,
      "step": 52390
    },
    {
      "entropy": 5.515008068084716,
      "epoch": 5.03314121037464,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027603535972180453,
      "loss": 4.6095,
      "mean_token_accuracy": 0.2605175495147705,
      "num_tokens": 120129412.0,
      "step": 52395
    },
    {
      "entropy": 5.537090063095093,
      "epoch": 5.0336215177713735,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027600107651740407,
      "loss": 4.6758,
      "mean_token_accuracy": 0.2600330889225006,
      "num_tokens": 120140063.0,
      "step": 52400
    },
    {
      "entropy": 5.510074329376221,
      "epoch": 5.034101825168108,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027596679328976156,
      "loss": 4.665,
      "mean_token_accuracy": 0.257499623298645,
      "num_tokens": 120151624.0,
      "step": 52405
    },
    {
      "entropy": 5.53288049697876,
      "epoch": 5.034582132564841,
      "grad_norm": 1.25,
      "learning_rate": 0.0002759325100396729,
      "loss": 4.6688,
      "mean_token_accuracy": 0.2588732525706291,
      "num_tokens": 120164139.0,
      "step": 52410
    },
    {
      "entropy": 5.500956630706787,
      "epoch": 5.0350624399615755,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027589822676793424,
      "loss": 4.6259,
      "mean_token_accuracy": 0.26334625780582427,
      "num_tokens": 120175696.0,
      "step": 52415
    },
    {
      "entropy": 5.574534463882446,
      "epoch": 5.035542747358309,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002758639434753413,
      "loss": 4.6927,
      "mean_token_accuracy": 0.2550934001803398,
      "num_tokens": 120186867.0,
      "step": 52420
    },
    {
      "entropy": 5.519118165969848,
      "epoch": 5.036023054755043,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027582966016269025,
      "loss": 4.6457,
      "mean_token_accuracy": 0.25617814660072324,
      "num_tokens": 120198508.0,
      "step": 52425
    },
    {
      "entropy": 5.526713800430298,
      "epoch": 5.036503362151777,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027579537683077693,
      "loss": 4.6031,
      "mean_token_accuracy": 0.2666749686002731,
      "num_tokens": 120209542.0,
      "step": 52430
    },
    {
      "entropy": 5.477036046981811,
      "epoch": 5.036983669548511,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002757610934803974,
      "loss": 4.6612,
      "mean_token_accuracy": 0.2563027501106262,
      "num_tokens": 120221757.0,
      "step": 52435
    },
    {
      "entropy": 5.391278505325317,
      "epoch": 5.037463976945245,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027572681011234733,
      "loss": 4.5311,
      "mean_token_accuracy": 0.2784095585346222,
      "num_tokens": 120233727.0,
      "step": 52440
    },
    {
      "entropy": 5.549258899688721,
      "epoch": 5.037944284341979,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002756925267274231,
      "loss": 4.667,
      "mean_token_accuracy": 0.2651970058679581,
      "num_tokens": 120244926.0,
      "step": 52445
    },
    {
      "entropy": 5.5474920749664305,
      "epoch": 5.038424591738713,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002756582433264203,
      "loss": 4.6616,
      "mean_token_accuracy": 0.2627565681934357,
      "num_tokens": 120256134.0,
      "step": 52450
    },
    {
      "entropy": 5.537676382064819,
      "epoch": 5.038904899135447,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002756239599101351,
      "loss": 4.6084,
      "mean_token_accuracy": 0.2639744535088539,
      "num_tokens": 120266331.0,
      "step": 52455
    },
    {
      "entropy": 5.522514533996582,
      "epoch": 5.039385206532181,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027558967647936343,
      "loss": 4.6418,
      "mean_token_accuracy": 0.25538658648729323,
      "num_tokens": 120277626.0,
      "step": 52460
    },
    {
      "entropy": 5.447792100906372,
      "epoch": 5.039865513928914,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027555539303490126,
      "loss": 4.5992,
      "mean_token_accuracy": 0.26052683442831037,
      "num_tokens": 120288017.0,
      "step": 52465
    },
    {
      "entropy": 5.523684883117676,
      "epoch": 5.040345821325649,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002755211095775444,
      "loss": 4.6179,
      "mean_token_accuracy": 0.26761474311351774,
      "num_tokens": 120297936.0,
      "step": 52470
    },
    {
      "entropy": 5.514960050582886,
      "epoch": 5.040826128722382,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002754868261080891,
      "loss": 4.6189,
      "mean_token_accuracy": 0.2683106318116188,
      "num_tokens": 120309560.0,
      "step": 52475
    },
    {
      "entropy": 5.512055921554565,
      "epoch": 5.041306436119116,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027545254262733096,
      "loss": 4.6307,
      "mean_token_accuracy": 0.26013223081827164,
      "num_tokens": 120320820.0,
      "step": 52480
    },
    {
      "entropy": 5.566263246536255,
      "epoch": 5.04178674351585,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027541825913606616,
      "loss": 4.6926,
      "mean_token_accuracy": 0.26040478348731994,
      "num_tokens": 120331884.0,
      "step": 52485
    },
    {
      "entropy": 5.56316647529602,
      "epoch": 5.042267050912584,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002753839756350907,
      "loss": 4.7025,
      "mean_token_accuracy": 0.24859307259321212,
      "num_tokens": 120344745.0,
      "step": 52490
    },
    {
      "entropy": 5.575461530685425,
      "epoch": 5.042747358309318,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002753496921252004,
      "loss": 4.6354,
      "mean_token_accuracy": 0.2637710109353065,
      "num_tokens": 120356022.0,
      "step": 52495
    },
    {
      "entropy": 5.551992511749267,
      "epoch": 5.043227665706052,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002753154086071913,
      "loss": 4.6811,
      "mean_token_accuracy": 0.26889817267656324,
      "num_tokens": 120367879.0,
      "step": 52500
    },
    {
      "entropy": 5.479800176620484,
      "epoch": 5.043707973102785,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027528112508185936,
      "loss": 4.6311,
      "mean_token_accuracy": 0.26513236463069917,
      "num_tokens": 120380030.0,
      "step": 52505
    },
    {
      "entropy": 5.549280500411987,
      "epoch": 5.04418828049952,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027524684155000057,
      "loss": 4.6367,
      "mean_token_accuracy": 0.259766286611557,
      "num_tokens": 120391993.0,
      "step": 52510
    },
    {
      "entropy": 5.594520854949951,
      "epoch": 5.044668587896253,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002752125580124108,
      "loss": 4.7054,
      "mean_token_accuracy": 0.2543171584606171,
      "num_tokens": 120404468.0,
      "step": 52515
    },
    {
      "entropy": 5.47774715423584,
      "epoch": 5.0451488952929875,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027517827446988606,
      "loss": 4.5868,
      "mean_token_accuracy": 0.26448792964220047,
      "num_tokens": 120416145.0,
      "step": 52520
    },
    {
      "entropy": 5.531938648223877,
      "epoch": 5.045629202689722,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002751439909232223,
      "loss": 4.6224,
      "mean_token_accuracy": 0.2662616729736328,
      "num_tokens": 120427485.0,
      "step": 52525
    },
    {
      "entropy": 5.536355590820312,
      "epoch": 5.046109510086455,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027510970737321556,
      "loss": 4.6351,
      "mean_token_accuracy": 0.2646491214632988,
      "num_tokens": 120438380.0,
      "step": 52530
    },
    {
      "entropy": 5.485915851593018,
      "epoch": 5.04658981748319,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002750754238206617,
      "loss": 4.6704,
      "mean_token_accuracy": 0.2577394604682922,
      "num_tokens": 120450242.0,
      "step": 52535
    },
    {
      "entropy": 5.522487831115723,
      "epoch": 5.047070124879923,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027504114026635676,
      "loss": 4.6547,
      "mean_token_accuracy": 0.2656126216053963,
      "num_tokens": 120461071.0,
      "step": 52540
    },
    {
      "entropy": 5.63253960609436,
      "epoch": 5.047550432276657,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002750068567110966,
      "loss": 4.7526,
      "mean_token_accuracy": 0.25615263283252715,
      "num_tokens": 120472375.0,
      "step": 52545
    },
    {
      "entropy": 5.617427110671997,
      "epoch": 5.048030739673391,
      "grad_norm": 1.375,
      "learning_rate": 0.00027497257315567733,
      "loss": 4.7083,
      "mean_token_accuracy": 0.260556623339653,
      "num_tokens": 120483381.0,
      "step": 52550
    },
    {
      "entropy": 5.545552587509155,
      "epoch": 5.048511047070125,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002749382896008947,
      "loss": 4.6581,
      "mean_token_accuracy": 0.26240517348051073,
      "num_tokens": 120494275.0,
      "step": 52555
    },
    {
      "entropy": 5.51878547668457,
      "epoch": 5.0489913544668585,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027490400604754503,
      "loss": 4.7019,
      "mean_token_accuracy": 0.2546693429350853,
      "num_tokens": 120505452.0,
      "step": 52560
    },
    {
      "entropy": 5.4854237079620365,
      "epoch": 5.049471661863593,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027486972249642387,
      "loss": 4.6668,
      "mean_token_accuracy": 0.26462713032960894,
      "num_tokens": 120516795.0,
      "step": 52565
    },
    {
      "entropy": 5.629294967651367,
      "epoch": 5.049951969260326,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002748354389483274,
      "loss": 4.7667,
      "mean_token_accuracy": 0.25320799499750135,
      "num_tokens": 120528972.0,
      "step": 52570
    },
    {
      "entropy": 5.534354877471924,
      "epoch": 5.050432276657061,
      "grad_norm": 1.375,
      "learning_rate": 0.0002748011554040515,
      "loss": 4.6175,
      "mean_token_accuracy": 0.2675777167081833,
      "num_tokens": 120539280.0,
      "step": 52575
    },
    {
      "entropy": 5.585940790176392,
      "epoch": 5.050912584053794,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002747668718643923,
      "loss": 4.6988,
      "mean_token_accuracy": 0.25977672189474105,
      "num_tokens": 120550339.0,
      "step": 52580
    },
    {
      "entropy": 5.499801731109619,
      "epoch": 5.051392891450528,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027473258833014553,
      "loss": 4.689,
      "mean_token_accuracy": 0.25979952663183215,
      "num_tokens": 120563334.0,
      "step": 52585
    },
    {
      "entropy": 5.543687868118286,
      "epoch": 5.051873198847262,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027469830480210736,
      "loss": 4.6195,
      "mean_token_accuracy": 0.2684008300304413,
      "num_tokens": 120573707.0,
      "step": 52590
    },
    {
      "entropy": 5.529169273376465,
      "epoch": 5.052353506243996,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002746640212810737,
      "loss": 4.6048,
      "mean_token_accuracy": 0.2616755455732346,
      "num_tokens": 120584986.0,
      "step": 52595
    },
    {
      "entropy": 5.566828441619873,
      "epoch": 5.0528338136407305,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002746297377678403,
      "loss": 4.7306,
      "mean_token_accuracy": 0.2502318546175957,
      "num_tokens": 120596258.0,
      "step": 52600
    },
    {
      "entropy": 5.536663389205932,
      "epoch": 5.053314121037464,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027459545426320337,
      "loss": 4.6802,
      "mean_token_accuracy": 0.2612490221858025,
      "num_tokens": 120609036.0,
      "step": 52605
    },
    {
      "entropy": 5.60493016242981,
      "epoch": 5.053794428434198,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002745611707679588,
      "loss": 4.7823,
      "mean_token_accuracy": 0.2529679909348488,
      "num_tokens": 120620648.0,
      "step": 52610
    },
    {
      "entropy": 5.5585826396942135,
      "epoch": 5.054274735830932,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027452688728290246,
      "loss": 4.6368,
      "mean_token_accuracy": 0.26699173301458357,
      "num_tokens": 120632463.0,
      "step": 52615
    },
    {
      "entropy": 5.457834005355835,
      "epoch": 5.054755043227666,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027449260380883056,
      "loss": 4.5914,
      "mean_token_accuracy": 0.2648575633764267,
      "num_tokens": 120643681.0,
      "step": 52620
    },
    {
      "entropy": 5.472521209716797,
      "epoch": 5.055235350624399,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002744583203465388,
      "loss": 4.5742,
      "mean_token_accuracy": 0.2742185816168785,
      "num_tokens": 120654600.0,
      "step": 52625
    },
    {
      "entropy": 5.519409275054931,
      "epoch": 5.055715658021134,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027442403689682316,
      "loss": 4.5846,
      "mean_token_accuracy": 0.2657577946782112,
      "num_tokens": 120665560.0,
      "step": 52630
    },
    {
      "entropy": 5.561632585525513,
      "epoch": 5.056195965417867,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027438975346047976,
      "loss": 4.7343,
      "mean_token_accuracy": 0.2530520439147949,
      "num_tokens": 120677070.0,
      "step": 52635
    },
    {
      "entropy": 5.550606918334961,
      "epoch": 5.0566762728146015,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027435547003830445,
      "loss": 4.6268,
      "mean_token_accuracy": 0.2554906338453293,
      "num_tokens": 120689470.0,
      "step": 52640
    },
    {
      "entropy": 5.5208899021148685,
      "epoch": 5.057156580211335,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027432118663109323,
      "loss": 4.6842,
      "mean_token_accuracy": 0.25979979783296586,
      "num_tokens": 120701693.0,
      "step": 52645
    },
    {
      "entropy": 5.513149929046631,
      "epoch": 5.057636887608069,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027428690323964204,
      "loss": 4.6113,
      "mean_token_accuracy": 0.2612448692321777,
      "num_tokens": 120712597.0,
      "step": 52650
    },
    {
      "entropy": 5.533696842193604,
      "epoch": 5.058117195004803,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002742526198647469,
      "loss": 4.6882,
      "mean_token_accuracy": 0.25174703299999235,
      "num_tokens": 120724993.0,
      "step": 52655
    },
    {
      "entropy": 5.543549299240112,
      "epoch": 5.058597502401537,
      "grad_norm": 1.5,
      "learning_rate": 0.00027421833650720357,
      "loss": 4.6312,
      "mean_token_accuracy": 0.2618659555912018,
      "num_tokens": 120735333.0,
      "step": 52660
    },
    {
      "entropy": 5.53416337966919,
      "epoch": 5.0590778097982705,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002741840531678082,
      "loss": 4.705,
      "mean_token_accuracy": 0.25978888720273974,
      "num_tokens": 120747865.0,
      "step": 52665
    },
    {
      "entropy": 5.524687433242798,
      "epoch": 5.059558117195005,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027414976984735666,
      "loss": 4.6701,
      "mean_token_accuracy": 0.2609471529722214,
      "num_tokens": 120758811.0,
      "step": 52670
    },
    {
      "entropy": 5.622972917556763,
      "epoch": 5.060038424591739,
      "grad_norm": 1.25,
      "learning_rate": 0.00027411548654664506,
      "loss": 4.7707,
      "mean_token_accuracy": 0.2501683205366135,
      "num_tokens": 120770938.0,
      "step": 52675
    },
    {
      "entropy": 5.533102369308471,
      "epoch": 5.060518731988473,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027408120326646916,
      "loss": 4.613,
      "mean_token_accuracy": 0.2660989612340927,
      "num_tokens": 120781794.0,
      "step": 52680
    },
    {
      "entropy": 5.626890850067139,
      "epoch": 5.060999039385207,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000274046920007625,
      "loss": 4.7284,
      "mean_token_accuracy": 0.2537282437086105,
      "num_tokens": 120792753.0,
      "step": 52685
    },
    {
      "entropy": 5.5414612770080565,
      "epoch": 5.06147934678194,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002740126367709085,
      "loss": 4.654,
      "mean_token_accuracy": 0.25814990103244784,
      "num_tokens": 120804845.0,
      "step": 52690
    },
    {
      "entropy": 5.531492757797241,
      "epoch": 5.061959654178675,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027397835355711563,
      "loss": 4.6295,
      "mean_token_accuracy": 0.2710217982530594,
      "num_tokens": 120816359.0,
      "step": 52695
    },
    {
      "entropy": 5.436241865158081,
      "epoch": 5.062439961575408,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002739440703670424,
      "loss": 4.5483,
      "mean_token_accuracy": 0.2662169337272644,
      "num_tokens": 120828251.0,
      "step": 52700
    },
    {
      "entropy": 5.587940883636475,
      "epoch": 5.062920268972142,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002739097872014847,
      "loss": 4.6926,
      "mean_token_accuracy": 0.2523655503988266,
      "num_tokens": 120839190.0,
      "step": 52705
    },
    {
      "entropy": 5.496385765075684,
      "epoch": 5.063400576368876,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002738755040612386,
      "loss": 4.5567,
      "mean_token_accuracy": 0.27212749421596527,
      "num_tokens": 120850826.0,
      "step": 52710
    },
    {
      "entropy": 5.593372249603272,
      "epoch": 5.06388088376561,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027384122094709985,
      "loss": 4.6852,
      "mean_token_accuracy": 0.25689181685447693,
      "num_tokens": 120862360.0,
      "step": 52715
    },
    {
      "entropy": 5.491254186630249,
      "epoch": 5.064361191162344,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002738069378598645,
      "loss": 4.6655,
      "mean_token_accuracy": 0.26066823303699493,
      "num_tokens": 120874796.0,
      "step": 52720
    },
    {
      "entropy": 5.559783792495727,
      "epoch": 5.064841498559078,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002737726548003285,
      "loss": 4.6466,
      "mean_token_accuracy": 0.2568985790014267,
      "num_tokens": 120885894.0,
      "step": 52725
    },
    {
      "entropy": 5.58178391456604,
      "epoch": 5.065321805955811,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027373837176928787,
      "loss": 4.7101,
      "mean_token_accuracy": 0.25739361345767975,
      "num_tokens": 120896869.0,
      "step": 52730
    },
    {
      "entropy": 5.452416658401489,
      "epoch": 5.065802113352546,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027370408876753845,
      "loss": 4.5649,
      "mean_token_accuracy": 0.26553806513547895,
      "num_tokens": 120909331.0,
      "step": 52735
    },
    {
      "entropy": 5.5398108959198,
      "epoch": 5.066282420749279,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027366980579587635,
      "loss": 4.7229,
      "mean_token_accuracy": 0.2594629883766174,
      "num_tokens": 120920192.0,
      "step": 52740
    },
    {
      "entropy": 5.560996007919312,
      "epoch": 5.0667627281460135,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027363552285509737,
      "loss": 4.6162,
      "mean_token_accuracy": 0.25971888154745104,
      "num_tokens": 120930767.0,
      "step": 52745
    },
    {
      "entropy": 5.49798583984375,
      "epoch": 5.067243035542747,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027360123994599737,
      "loss": 4.677,
      "mean_token_accuracy": 0.2578368648886681,
      "num_tokens": 120941731.0,
      "step": 52750
    },
    {
      "entropy": 5.623564720153809,
      "epoch": 5.067723342939481,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027356695706937253,
      "loss": 4.735,
      "mean_token_accuracy": 0.24779539704322814,
      "num_tokens": 120955707.0,
      "step": 52755
    },
    {
      "entropy": 5.555979585647583,
      "epoch": 5.068203650336216,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002735326742260187,
      "loss": 4.641,
      "mean_token_accuracy": 0.2585094481706619,
      "num_tokens": 120967408.0,
      "step": 52760
    },
    {
      "entropy": 5.596336936950683,
      "epoch": 5.068683957732949,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002734983914167317,
      "loss": 4.6951,
      "mean_token_accuracy": 0.26413038223981855,
      "num_tokens": 120979853.0,
      "step": 52765
    },
    {
      "entropy": 5.526946067810059,
      "epoch": 5.069164265129683,
      "grad_norm": 1.25,
      "learning_rate": 0.00027346410864230783,
      "loss": 4.6999,
      "mean_token_accuracy": 0.25343251377344134,
      "num_tokens": 120991948.0,
      "step": 52770
    },
    {
      "entropy": 5.5075380325317385,
      "epoch": 5.069644572526417,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027342982590354266,
      "loss": 4.5819,
      "mean_token_accuracy": 0.2620419755578041,
      "num_tokens": 121003804.0,
      "step": 52775
    },
    {
      "entropy": 5.521994876861572,
      "epoch": 5.070124879923151,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002733955432012322,
      "loss": 4.6551,
      "mean_token_accuracy": 0.261733116209507,
      "num_tokens": 121014619.0,
      "step": 52780
    },
    {
      "entropy": 5.560255002975464,
      "epoch": 5.0706051873198845,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027336126053617257,
      "loss": 4.6552,
      "mean_token_accuracy": 0.257463338971138,
      "num_tokens": 121025272.0,
      "step": 52785
    },
    {
      "entropy": 5.491431808471679,
      "epoch": 5.071085494716619,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002733269779091595,
      "loss": 4.6078,
      "mean_token_accuracy": 0.26251736730337144,
      "num_tokens": 121037017.0,
      "step": 52790
    },
    {
      "entropy": 5.6108757019042965,
      "epoch": 5.071565802113352,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027329269532098906,
      "loss": 4.7041,
      "mean_token_accuracy": 0.259819483757019,
      "num_tokens": 121047933.0,
      "step": 52795
    },
    {
      "entropy": 5.5595386028289795,
      "epoch": 5.072046109510087,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027325841277245735,
      "loss": 4.6795,
      "mean_token_accuracy": 0.26316856145858764,
      "num_tokens": 121061318.0,
      "step": 52800
    },
    {
      "entropy": 5.592967844009399,
      "epoch": 5.07252641690682,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002732241302643599,
      "loss": 4.6766,
      "mean_token_accuracy": 0.25736442357301714,
      "num_tokens": 121073191.0,
      "step": 52805
    },
    {
      "entropy": 5.553551435470581,
      "epoch": 5.073006724303554,
      "grad_norm": 1.3125,
      "learning_rate": 0.000273189847797493,
      "loss": 4.7388,
      "mean_token_accuracy": 0.24726446121931075,
      "num_tokens": 121084318.0,
      "step": 52810
    },
    {
      "entropy": 5.530810832977295,
      "epoch": 5.073487031700288,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027315556537265236,
      "loss": 4.6718,
      "mean_token_accuracy": 0.2559325724840164,
      "num_tokens": 121094590.0,
      "step": 52815
    },
    {
      "entropy": 5.474588680267334,
      "epoch": 5.073967339097022,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027312128299063405,
      "loss": 4.6185,
      "mean_token_accuracy": 0.2588049456477165,
      "num_tokens": 121105268.0,
      "step": 52820
    },
    {
      "entropy": 5.504098749160766,
      "epoch": 5.074447646493756,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027308700065223395,
      "loss": 4.6353,
      "mean_token_accuracy": 0.25917191803455353,
      "num_tokens": 121116467.0,
      "step": 52825
    },
    {
      "entropy": 5.605092144012451,
      "epoch": 5.07492795389049,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027305271835824813,
      "loss": 4.6728,
      "mean_token_accuracy": 0.2688612103462219,
      "num_tokens": 121128247.0,
      "step": 52830
    },
    {
      "entropy": 5.542609786987304,
      "epoch": 5.075408261287224,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002730184361094723,
      "loss": 4.7018,
      "mean_token_accuracy": 0.2614315003156662,
      "num_tokens": 121138293.0,
      "step": 52835
    },
    {
      "entropy": 5.503521585464478,
      "epoch": 5.075888568683958,
      "grad_norm": 1.25,
      "learning_rate": 0.00027298415390670253,
      "loss": 4.6359,
      "mean_token_accuracy": 0.26842528879642485,
      "num_tokens": 121150470.0,
      "step": 52840
    },
    {
      "entropy": 5.520740747451782,
      "epoch": 5.076368876080692,
      "grad_norm": 1.25,
      "learning_rate": 0.0002729498717507347,
      "loss": 4.6086,
      "mean_token_accuracy": 0.2668144702911377,
      "num_tokens": 121162053.0,
      "step": 52845
    },
    {
      "entropy": 5.556865882873535,
      "epoch": 5.076849183477425,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027291558964236476,
      "loss": 4.6623,
      "mean_token_accuracy": 0.26225738525390624,
      "num_tokens": 121173497.0,
      "step": 52850
    },
    {
      "entropy": 5.538127708435058,
      "epoch": 5.07732949087416,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002728813075823886,
      "loss": 4.654,
      "mean_token_accuracy": 0.2658750593662262,
      "num_tokens": 121184123.0,
      "step": 52855
    },
    {
      "entropy": 5.515976238250732,
      "epoch": 5.077809798270893,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027284702557160234,
      "loss": 4.6584,
      "mean_token_accuracy": 0.2578831806778908,
      "num_tokens": 121196845.0,
      "step": 52860
    },
    {
      "entropy": 5.494178581237793,
      "epoch": 5.0782901056676275,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002728127436108016,
      "loss": 4.6466,
      "mean_token_accuracy": 0.26650662571191785,
      "num_tokens": 121209424.0,
      "step": 52865
    },
    {
      "entropy": 5.477860164642334,
      "epoch": 5.078770413064361,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027277846170078253,
      "loss": 4.6466,
      "mean_token_accuracy": 0.2631461203098297,
      "num_tokens": 121222170.0,
      "step": 52870
    },
    {
      "entropy": 5.4837016582489015,
      "epoch": 5.079250720461095,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027274417984234095,
      "loss": 4.6191,
      "mean_token_accuracy": 0.2612843617796898,
      "num_tokens": 121234182.0,
      "step": 52875
    },
    {
      "entropy": 5.503811359405518,
      "epoch": 5.079731027857829,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002727098980362728,
      "loss": 4.6109,
      "mean_token_accuracy": 0.2680299818515778,
      "num_tokens": 121246122.0,
      "step": 52880
    },
    {
      "entropy": 5.537132930755615,
      "epoch": 5.080211335254563,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000272675616283374,
      "loss": 4.7104,
      "mean_token_accuracy": 0.25040834248065946,
      "num_tokens": 121258181.0,
      "step": 52885
    },
    {
      "entropy": 5.540223121643066,
      "epoch": 5.0806916426512965,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002726413345844406,
      "loss": 4.5869,
      "mean_token_accuracy": 0.27290032207965853,
      "num_tokens": 121270167.0,
      "step": 52890
    },
    {
      "entropy": 5.456027507781982,
      "epoch": 5.081171950048031,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027260705294026834,
      "loss": 4.6515,
      "mean_token_accuracy": 0.2594391480088234,
      "num_tokens": 121280563.0,
      "step": 52895
    },
    {
      "entropy": 5.543133306503296,
      "epoch": 5.081652257444764,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002725727713516532,
      "loss": 4.6606,
      "mean_token_accuracy": 0.259412944316864,
      "num_tokens": 121292924.0,
      "step": 52900
    },
    {
      "entropy": 5.565569114685059,
      "epoch": 5.082132564841499,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002725384898193911,
      "loss": 4.6675,
      "mean_token_accuracy": 0.2561723828315735,
      "num_tokens": 121303884.0,
      "step": 52905
    },
    {
      "entropy": 5.47353229522705,
      "epoch": 5.082612872238233,
      "grad_norm": 1.3125,
      "learning_rate": 0.000272504208344278,
      "loss": 4.5907,
      "mean_token_accuracy": 0.26457871198654176,
      "num_tokens": 121314562.0,
      "step": 52910
    },
    {
      "entropy": 5.40302004814148,
      "epoch": 5.083093179634966,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002724699269271097,
      "loss": 4.5995,
      "mean_token_accuracy": 0.2585220232605934,
      "num_tokens": 121326130.0,
      "step": 52915
    },
    {
      "entropy": 5.524237108230591,
      "epoch": 5.083573487031701,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027243564556868224,
      "loss": 4.6338,
      "mean_token_accuracy": 0.2601589739322662,
      "num_tokens": 121338691.0,
      "step": 52920
    },
    {
      "entropy": 5.5513756275177,
      "epoch": 5.084053794428434,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027240136426979144,
      "loss": 4.6227,
      "mean_token_accuracy": 0.2640117347240448,
      "num_tokens": 121350338.0,
      "step": 52925
    },
    {
      "entropy": 5.508493661880493,
      "epoch": 5.084534101825168,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027236708303123323,
      "loss": 4.6626,
      "mean_token_accuracy": 0.2603862091898918,
      "num_tokens": 121363450.0,
      "step": 52930
    },
    {
      "entropy": 5.557983875274658,
      "epoch": 5.085014409221902,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002723328018538036,
      "loss": 4.6754,
      "mean_token_accuracy": 0.257698230445385,
      "num_tokens": 121374560.0,
      "step": 52935
    },
    {
      "entropy": 5.570313453674316,
      "epoch": 5.085494716618636,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002722985207382984,
      "loss": 4.6742,
      "mean_token_accuracy": 0.2599464252591133,
      "num_tokens": 121385097.0,
      "step": 52940
    },
    {
      "entropy": 5.5814783573150635,
      "epoch": 5.08597502401537,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027226423968551344,
      "loss": 4.6834,
      "mean_token_accuracy": 0.2610358461737633,
      "num_tokens": 121396670.0,
      "step": 52945
    },
    {
      "entropy": 5.6222881317138675,
      "epoch": 5.086455331412104,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027222995869624476,
      "loss": 4.7275,
      "mean_token_accuracy": 0.2512193277478218,
      "num_tokens": 121409903.0,
      "step": 52950
    },
    {
      "entropy": 5.514986371994018,
      "epoch": 5.086935638808837,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027219567777128824,
      "loss": 4.5783,
      "mean_token_accuracy": 0.26169556230306623,
      "num_tokens": 121420336.0,
      "step": 52955
    },
    {
      "entropy": 5.476877880096436,
      "epoch": 5.087415946205572,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002721613969114396,
      "loss": 4.6189,
      "mean_token_accuracy": 0.26682259887456894,
      "num_tokens": 121431617.0,
      "step": 52960
    },
    {
      "entropy": 5.431775522232056,
      "epoch": 5.087896253602305,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027212711611749506,
      "loss": 4.6698,
      "mean_token_accuracy": 0.26125864386558534,
      "num_tokens": 121444889.0,
      "step": 52965
    },
    {
      "entropy": 5.511652135848999,
      "epoch": 5.0883765609990395,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027209283539025034,
      "loss": 4.6152,
      "mean_token_accuracy": 0.2642847657203674,
      "num_tokens": 121456776.0,
      "step": 52970
    },
    {
      "entropy": 5.580621576309204,
      "epoch": 5.088856868395773,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027205855473050127,
      "loss": 4.7543,
      "mean_token_accuracy": 0.2529293656349182,
      "num_tokens": 121470039.0,
      "step": 52975
    },
    {
      "entropy": 5.605852842330933,
      "epoch": 5.089337175792507,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000272024274139044,
      "loss": 4.7002,
      "mean_token_accuracy": 0.25682538449764253,
      "num_tokens": 121482349.0,
      "step": 52980
    },
    {
      "entropy": 5.557800769805908,
      "epoch": 5.0898174831892415,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002719899936166741,
      "loss": 4.6433,
      "mean_token_accuracy": 0.2618229627609253,
      "num_tokens": 121494283.0,
      "step": 52985
    },
    {
      "entropy": 5.491266250610352,
      "epoch": 5.090297790585975,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002719557131641876,
      "loss": 4.6497,
      "mean_token_accuracy": 0.2618023604154587,
      "num_tokens": 121504654.0,
      "step": 52990
    },
    {
      "entropy": 5.476589202880859,
      "epoch": 5.090778097982709,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002719214327823804,
      "loss": 4.6006,
      "mean_token_accuracy": 0.2625830054283142,
      "num_tokens": 121515580.0,
      "step": 52995
    },
    {
      "entropy": 5.607587671279907,
      "epoch": 5.091258405379443,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027188715247204845,
      "loss": 4.7813,
      "mean_token_accuracy": 0.2506200224161148,
      "num_tokens": 121527079.0,
      "step": 53000
    },
    {
      "entropy": 5.573591136932373,
      "epoch": 5.091738712776177,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002718528722339875,
      "loss": 4.7286,
      "mean_token_accuracy": 0.2571015328168869,
      "num_tokens": 121538958.0,
      "step": 53005
    },
    {
      "entropy": 5.577002429962159,
      "epoch": 5.0922190201729105,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027181859206899366,
      "loss": 4.6125,
      "mean_token_accuracy": 0.26750885546207426,
      "num_tokens": 121550378.0,
      "step": 53010
    },
    {
      "entropy": 5.489740991592408,
      "epoch": 5.092699327569645,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002717843119778625,
      "loss": 4.612,
      "mean_token_accuracy": 0.2672847792506218,
      "num_tokens": 121561621.0,
      "step": 53015
    },
    {
      "entropy": 5.531423902511596,
      "epoch": 5.093179634966378,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027175003196139015,
      "loss": 4.6892,
      "mean_token_accuracy": 0.26253926157951357,
      "num_tokens": 121572581.0,
      "step": 53020
    },
    {
      "entropy": 5.506611347198486,
      "epoch": 5.093659942363113,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002717157520203724,
      "loss": 4.6368,
      "mean_token_accuracy": 0.25811494290828707,
      "num_tokens": 121583886.0,
      "step": 53025
    },
    {
      "entropy": 5.568693399429321,
      "epoch": 5.094140249759846,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027168147215560505,
      "loss": 4.6397,
      "mean_token_accuracy": 0.2634089946746826,
      "num_tokens": 121594935.0,
      "step": 53030
    },
    {
      "entropy": 5.548572254180908,
      "epoch": 5.09462055715658,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002716471923678842,
      "loss": 4.6692,
      "mean_token_accuracy": 0.2637352541089058,
      "num_tokens": 121605810.0,
      "step": 53035
    },
    {
      "entropy": 5.477960729598999,
      "epoch": 5.095100864553314,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002716129126580056,
      "loss": 4.5745,
      "mean_token_accuracy": 0.260130213201046,
      "num_tokens": 121616863.0,
      "step": 53040
    },
    {
      "entropy": 5.576207160949707,
      "epoch": 5.095581171950048,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000271578633026765,
      "loss": 4.714,
      "mean_token_accuracy": 0.2592612564563751,
      "num_tokens": 121629868.0,
      "step": 53045
    },
    {
      "entropy": 5.5619079113006595,
      "epoch": 5.0960614793467816,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002715443534749584,
      "loss": 4.7201,
      "mean_token_accuracy": 0.2577244058251381,
      "num_tokens": 121642093.0,
      "step": 53050
    },
    {
      "entropy": 5.599740839004516,
      "epoch": 5.096541786743516,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002715100740033817,
      "loss": 4.7091,
      "mean_token_accuracy": 0.2541333794593811,
      "num_tokens": 121653200.0,
      "step": 53055
    },
    {
      "entropy": 5.535562992095947,
      "epoch": 5.09702209414025,
      "grad_norm": 1.25,
      "learning_rate": 0.00027147579461283064,
      "loss": 4.6269,
      "mean_token_accuracy": 0.26101663112640383,
      "num_tokens": 121665184.0,
      "step": 53060
    },
    {
      "entropy": 5.5719482421875,
      "epoch": 5.097502401536984,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027144151530410123,
      "loss": 4.7222,
      "mean_token_accuracy": 0.25822199136018753,
      "num_tokens": 121676092.0,
      "step": 53065
    },
    {
      "entropy": 5.577082920074463,
      "epoch": 5.097982708933718,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027140723607798934,
      "loss": 4.6963,
      "mean_token_accuracy": 0.25299872159957887,
      "num_tokens": 121687996.0,
      "step": 53070
    },
    {
      "entropy": 5.576039266586304,
      "epoch": 5.098463016330451,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027137295693529064,
      "loss": 4.6672,
      "mean_token_accuracy": 0.2557502418756485,
      "num_tokens": 121699208.0,
      "step": 53075
    },
    {
      "entropy": 5.505693483352661,
      "epoch": 5.098943323727186,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002713386778768011,
      "loss": 4.6666,
      "mean_token_accuracy": 0.2578994229435921,
      "num_tokens": 121711065.0,
      "step": 53080
    },
    {
      "entropy": 5.582651853561401,
      "epoch": 5.099423631123919,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027130439890331667,
      "loss": 4.7256,
      "mean_token_accuracy": 0.258173955976963,
      "num_tokens": 121722573.0,
      "step": 53085
    },
    {
      "entropy": 5.637018442153931,
      "epoch": 5.0999039385206535,
      "grad_norm": 1.265625,
      "learning_rate": 0.000271270120015633,
      "loss": 4.7871,
      "mean_token_accuracy": 0.25361995697021483,
      "num_tokens": 121734407.0,
      "step": 53090
    },
    {
      "entropy": 5.486701631546021,
      "epoch": 5.100384245917387,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027123584121454614,
      "loss": 4.5986,
      "mean_token_accuracy": 0.26571850180625917,
      "num_tokens": 121745832.0,
      "step": 53095
    },
    {
      "entropy": 5.501519012451172,
      "epoch": 5.100864553314121,
      "grad_norm": 1.25,
      "learning_rate": 0.000271201562500852,
      "loss": 4.6099,
      "mean_token_accuracy": 0.2618995293974876,
      "num_tokens": 121758145.0,
      "step": 53100
    },
    {
      "entropy": 5.5189581394195555,
      "epoch": 5.101344860710855,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002711672838753462,
      "loss": 4.6081,
      "mean_token_accuracy": 0.2661005347967148,
      "num_tokens": 121769583.0,
      "step": 53105
    },
    {
      "entropy": 5.53486475944519,
      "epoch": 5.101825168107589,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002711330053388246,
      "loss": 4.6444,
      "mean_token_accuracy": 0.2622625470161438,
      "num_tokens": 121781398.0,
      "step": 53110
    },
    {
      "entropy": 5.477579832077026,
      "epoch": 5.1023054755043225,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002710987268920832,
      "loss": 4.5595,
      "mean_token_accuracy": 0.26889011412858965,
      "num_tokens": 121792650.0,
      "step": 53115
    },
    {
      "entropy": 5.506315422058106,
      "epoch": 5.102785782901057,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002710644485359178,
      "loss": 4.6951,
      "mean_token_accuracy": 0.2632329374551773,
      "num_tokens": 121803771.0,
      "step": 53120
    },
    {
      "entropy": 5.481941366195679,
      "epoch": 5.10326609029779,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027103017027112414,
      "loss": 4.6191,
      "mean_token_accuracy": 0.25902451425790785,
      "num_tokens": 121815085.0,
      "step": 53125
    },
    {
      "entropy": 5.508751630783081,
      "epoch": 5.1037463976945245,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027099589209849827,
      "loss": 4.6855,
      "mean_token_accuracy": 0.26272686272859574,
      "num_tokens": 121827297.0,
      "step": 53130
    },
    {
      "entropy": 5.526847219467163,
      "epoch": 5.104226705091258,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002709616140188358,
      "loss": 4.5779,
      "mean_token_accuracy": 0.2710634171962738,
      "num_tokens": 121838805.0,
      "step": 53135
    },
    {
      "entropy": 5.601501941680908,
      "epoch": 5.104707012487992,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027092733603293267,
      "loss": 4.6846,
      "mean_token_accuracy": 0.2570012301206589,
      "num_tokens": 121850069.0,
      "step": 53140
    },
    {
      "entropy": 5.5165986061096195,
      "epoch": 5.105187319884727,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002708930581415847,
      "loss": 4.6539,
      "mean_token_accuracy": 0.2675357684493065,
      "num_tokens": 121860985.0,
      "step": 53145
    },
    {
      "entropy": 5.578860473632813,
      "epoch": 5.10566762728146,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027085878034558775,
      "loss": 4.7235,
      "mean_token_accuracy": 0.259079672396183,
      "num_tokens": 121872105.0,
      "step": 53150
    },
    {
      "entropy": 5.475114727020264,
      "epoch": 5.106147934678194,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002708245026457376,
      "loss": 4.5528,
      "mean_token_accuracy": 0.2674885272979736,
      "num_tokens": 121883199.0,
      "step": 53155
    },
    {
      "entropy": 5.487955856323242,
      "epoch": 5.106628242074928,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027079022504283016,
      "loss": 4.5598,
      "mean_token_accuracy": 0.268089659512043,
      "num_tokens": 121894134.0,
      "step": 53160
    },
    {
      "entropy": 5.56188006401062,
      "epoch": 5.107108549471662,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00027075594753766115,
      "loss": 4.6926,
      "mean_token_accuracy": 0.26408462077379224,
      "num_tokens": 121904624.0,
      "step": 53165
    },
    {
      "entropy": 5.561642265319824,
      "epoch": 5.107588856868396,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027072167013102643,
      "loss": 4.7193,
      "mean_token_accuracy": 0.25318487286567687,
      "num_tokens": 121916220.0,
      "step": 53170
    },
    {
      "entropy": 5.479261970520019,
      "epoch": 5.10806916426513,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027068739282372177,
      "loss": 4.6196,
      "mean_token_accuracy": 0.26344057023525236,
      "num_tokens": 121927736.0,
      "step": 53175
    },
    {
      "entropy": 5.505460929870606,
      "epoch": 5.108549471661863,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002706531156165431,
      "loss": 4.7037,
      "mean_token_accuracy": 0.2600427567958832,
      "num_tokens": 121939499.0,
      "step": 53180
    },
    {
      "entropy": 5.494511985778809,
      "epoch": 5.109029779058598,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002706188385102862,
      "loss": 4.6065,
      "mean_token_accuracy": 0.26400420665740965,
      "num_tokens": 121950856.0,
      "step": 53185
    },
    {
      "entropy": 5.516808748245239,
      "epoch": 5.109510086455331,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002705845615057469,
      "loss": 4.6299,
      "mean_token_accuracy": 0.2628717586398125,
      "num_tokens": 121960942.0,
      "step": 53190
    },
    {
      "entropy": 5.528564929962158,
      "epoch": 5.109990393852065,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002705502846037209,
      "loss": 4.6407,
      "mean_token_accuracy": 0.2649009317159653,
      "num_tokens": 121972832.0,
      "step": 53195
    },
    {
      "entropy": 5.512624979019165,
      "epoch": 5.110470701248799,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002705160078050041,
      "loss": 4.6458,
      "mean_token_accuracy": 0.261512453854084,
      "num_tokens": 121984550.0,
      "step": 53200
    },
    {
      "entropy": 5.530774450302124,
      "epoch": 5.110951008645533,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027048173111039227,
      "loss": 4.7188,
      "mean_token_accuracy": 0.25089156329631807,
      "num_tokens": 121997418.0,
      "step": 53205
    },
    {
      "entropy": 5.530567932128906,
      "epoch": 5.111431316042267,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002704474545206812,
      "loss": 4.5971,
      "mean_token_accuracy": 0.2659377887845039,
      "num_tokens": 122008771.0,
      "step": 53210
    },
    {
      "entropy": 5.48203182220459,
      "epoch": 5.111911623439001,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027041317803666674,
      "loss": 4.5725,
      "mean_token_accuracy": 0.2632775828242302,
      "num_tokens": 122020801.0,
      "step": 53215
    },
    {
      "entropy": 5.529645681381226,
      "epoch": 5.112391930835735,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002703789016591447,
      "loss": 4.6678,
      "mean_token_accuracy": 0.2671017274260521,
      "num_tokens": 122032499.0,
      "step": 53220
    },
    {
      "entropy": 5.5205543518066404,
      "epoch": 5.112872238232469,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002703446253889108,
      "loss": 4.6721,
      "mean_token_accuracy": 0.25777481198310853,
      "num_tokens": 122044492.0,
      "step": 53225
    },
    {
      "entropy": 5.551367616653442,
      "epoch": 5.113352545629203,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002703103492267609,
      "loss": 4.664,
      "mean_token_accuracy": 0.2646737307310104,
      "num_tokens": 122055647.0,
      "step": 53230
    },
    {
      "entropy": 5.514669513702392,
      "epoch": 5.1138328530259365,
      "grad_norm": 1.375,
      "learning_rate": 0.0002702760731734907,
      "loss": 4.641,
      "mean_token_accuracy": 0.2615489557385445,
      "num_tokens": 122066844.0,
      "step": 53235
    },
    {
      "entropy": 5.485998821258545,
      "epoch": 5.114313160422671,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002702417972298961,
      "loss": 4.6065,
      "mean_token_accuracy": 0.26750562638044356,
      "num_tokens": 122078838.0,
      "step": 53240
    },
    {
      "entropy": 5.469153642654419,
      "epoch": 5.114793467819404,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002702075213967729,
      "loss": 4.656,
      "mean_token_accuracy": 0.2608339011669159,
      "num_tokens": 122091260.0,
      "step": 53245
    },
    {
      "entropy": 5.503183889389038,
      "epoch": 5.115273775216139,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002701732456749168,
      "loss": 4.6357,
      "mean_token_accuracy": 0.26911372095346453,
      "num_tokens": 122101735.0,
      "step": 53250
    },
    {
      "entropy": 5.5584859371185305,
      "epoch": 5.115754082612872,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027013897006512356,
      "loss": 4.6898,
      "mean_token_accuracy": 0.26206494122743607,
      "num_tokens": 122112617.0,
      "step": 53255
    },
    {
      "entropy": 5.553332233428955,
      "epoch": 5.116234390009606,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027010469456818906,
      "loss": 4.6702,
      "mean_token_accuracy": 0.26098917573690417,
      "num_tokens": 122124623.0,
      "step": 53260
    },
    {
      "entropy": 5.594034433364868,
      "epoch": 5.11671469740634,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002700704191849089,
      "loss": 4.7361,
      "mean_token_accuracy": 0.2544580057263374,
      "num_tokens": 122135826.0,
      "step": 53265
    },
    {
      "entropy": 5.498718166351319,
      "epoch": 5.117195004803074,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027003614391607906,
      "loss": 4.5866,
      "mean_token_accuracy": 0.2674205735325813,
      "num_tokens": 122146480.0,
      "step": 53270
    },
    {
      "entropy": 5.572028017044067,
      "epoch": 5.1176753121998075,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002700018687624952,
      "loss": 4.7105,
      "mean_token_accuracy": 0.2549558192491531,
      "num_tokens": 122158331.0,
      "step": 53275
    },
    {
      "entropy": 5.53701605796814,
      "epoch": 5.118155619596542,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002699675937249532,
      "loss": 4.681,
      "mean_token_accuracy": 0.25821129977703094,
      "num_tokens": 122170549.0,
      "step": 53280
    },
    {
      "entropy": 5.563247394561768,
      "epoch": 5.118635926993275,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026993331880424855,
      "loss": 4.7223,
      "mean_token_accuracy": 0.2533079028129578,
      "num_tokens": 122181894.0,
      "step": 53285
    },
    {
      "entropy": 5.546642255783081,
      "epoch": 5.11911623439001,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002698990440011773,
      "loss": 4.6865,
      "mean_token_accuracy": 0.2592563912272453,
      "num_tokens": 122193649.0,
      "step": 53290
    },
    {
      "entropy": 5.451125574111939,
      "epoch": 5.119596541786744,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002698647693165351,
      "loss": 4.6069,
      "mean_token_accuracy": 0.26519315391778947,
      "num_tokens": 122206616.0,
      "step": 53295
    },
    {
      "entropy": 5.535603952407837,
      "epoch": 5.120076849183477,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026983049475111766,
      "loss": 4.6557,
      "mean_token_accuracy": 0.25226091891527175,
      "num_tokens": 122218578.0,
      "step": 53300
    },
    {
      "entropy": 5.544290590286255,
      "epoch": 5.120557156580212,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026979622030572076,
      "loss": 4.596,
      "mean_token_accuracy": 0.2672847345471382,
      "num_tokens": 122230533.0,
      "step": 53305
    },
    {
      "entropy": 5.5069413661956785,
      "epoch": 5.121037463976945,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002697619459811403,
      "loss": 4.6164,
      "mean_token_accuracy": 0.26193786412477493,
      "num_tokens": 122241595.0,
      "step": 53310
    },
    {
      "entropy": 5.510664081573486,
      "epoch": 5.1215177713736795,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002697276717781718,
      "loss": 4.6281,
      "mean_token_accuracy": 0.2666506037116051,
      "num_tokens": 122253073.0,
      "step": 53315
    },
    {
      "entropy": 5.517863607406616,
      "epoch": 5.121998078770413,
      "grad_norm": 1.375,
      "learning_rate": 0.00026969339769761117,
      "loss": 4.6062,
      "mean_token_accuracy": 0.2658717393875122,
      "num_tokens": 122264469.0,
      "step": 53320
    },
    {
      "entropy": 5.54456934928894,
      "epoch": 5.122478386167147,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000269659123740254,
      "loss": 4.6489,
      "mean_token_accuracy": 0.26327773183584213,
      "num_tokens": 122276130.0,
      "step": 53325
    },
    {
      "entropy": 5.594854164123535,
      "epoch": 5.122958693563881,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026962484990689616,
      "loss": 4.6675,
      "mean_token_accuracy": 0.26238672286272047,
      "num_tokens": 122288225.0,
      "step": 53330
    },
    {
      "entropy": 5.566983699798584,
      "epoch": 5.123439000960615,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026959057619833335,
      "loss": 4.6669,
      "mean_token_accuracy": 0.25525187104940417,
      "num_tokens": 122299536.0,
      "step": 53335
    },
    {
      "entropy": 5.475725841522217,
      "epoch": 5.123919308357348,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002695563026153613,
      "loss": 4.5668,
      "mean_token_accuracy": 0.2645181447267532,
      "num_tokens": 122309899.0,
      "step": 53340
    },
    {
      "entropy": 5.580260944366455,
      "epoch": 5.124399615754083,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002695220291587758,
      "loss": 4.7647,
      "mean_token_accuracy": 0.2605865105986595,
      "num_tokens": 122322261.0,
      "step": 53345
    },
    {
      "entropy": 5.484611892700196,
      "epoch": 5.124879923150816,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002694877558293724,
      "loss": 4.5769,
      "mean_token_accuracy": 0.27113310992717743,
      "num_tokens": 122334557.0,
      "step": 53350
    },
    {
      "entropy": 5.553295087814331,
      "epoch": 5.1253602305475505,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000269453482627947,
      "loss": 4.6817,
      "mean_token_accuracy": 0.25226841419935225,
      "num_tokens": 122346503.0,
      "step": 53355
    },
    {
      "entropy": 5.498297166824341,
      "epoch": 5.125840537944284,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002694192095552953,
      "loss": 4.6191,
      "mean_token_accuracy": 0.2678842216730118,
      "num_tokens": 122357355.0,
      "step": 53360
    },
    {
      "entropy": 5.446576690673828,
      "epoch": 5.126320845341018,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026938493661221287,
      "loss": 4.6131,
      "mean_token_accuracy": 0.2729318022727966,
      "num_tokens": 122369324.0,
      "step": 53365
    },
    {
      "entropy": 5.547396230697632,
      "epoch": 5.126801152737752,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002693506637994957,
      "loss": 4.6816,
      "mean_token_accuracy": 0.25662108659744265,
      "num_tokens": 122381434.0,
      "step": 53370
    },
    {
      "entropy": 5.52668571472168,
      "epoch": 5.127281460134486,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002693163911179392,
      "loss": 4.5937,
      "mean_token_accuracy": 0.2701208308339119,
      "num_tokens": 122393231.0,
      "step": 53375
    },
    {
      "entropy": 5.508250188827515,
      "epoch": 5.12776176753122,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002692821185683393,
      "loss": 4.6349,
      "mean_token_accuracy": 0.2590149357914925,
      "num_tokens": 122404983.0,
      "step": 53380
    },
    {
      "entropy": 5.484838581085205,
      "epoch": 5.128242074927954,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026924784615149156,
      "loss": 4.5874,
      "mean_token_accuracy": 0.2614042475819588,
      "num_tokens": 122415390.0,
      "step": 53385
    },
    {
      "entropy": 5.5098466873168945,
      "epoch": 5.128722382324688,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026921357386819176,
      "loss": 4.6776,
      "mean_token_accuracy": 0.2551340743899345,
      "num_tokens": 122427287.0,
      "step": 53390
    },
    {
      "entropy": 5.518045473098755,
      "epoch": 5.129202689721422,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002691793017192356,
      "loss": 4.6093,
      "mean_token_accuracy": 0.26869903951883317,
      "num_tokens": 122438952.0,
      "step": 53395
    },
    {
      "entropy": 5.53027868270874,
      "epoch": 5.129682997118156,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002691450297054189,
      "loss": 4.6997,
      "mean_token_accuracy": 0.2615835666656494,
      "num_tokens": 122450857.0,
      "step": 53400
    },
    {
      "entropy": 5.543721055984497,
      "epoch": 5.130163304514889,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002691107578275371,
      "loss": 4.6734,
      "mean_token_accuracy": 0.260125894844532,
      "num_tokens": 122462685.0,
      "step": 53405
    },
    {
      "entropy": 5.5551488399505615,
      "epoch": 5.130643611911624,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000269076486086386,
      "loss": 4.6346,
      "mean_token_accuracy": 0.26915704309940336,
      "num_tokens": 122473556.0,
      "step": 53410
    },
    {
      "entropy": 5.564883708953857,
      "epoch": 5.131123919308357,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026904221448276135,
      "loss": 4.7185,
      "mean_token_accuracy": 0.26279559135437014,
      "num_tokens": 122485258.0,
      "step": 53415
    },
    {
      "entropy": 5.537863302230835,
      "epoch": 5.131604226705091,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026900794301745875,
      "loss": 4.6091,
      "mean_token_accuracy": 0.2737228378653526,
      "num_tokens": 122496044.0,
      "step": 53420
    },
    {
      "entropy": 5.479403448104859,
      "epoch": 5.132084534101825,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002689736716912739,
      "loss": 4.6646,
      "mean_token_accuracy": 0.2607581987977028,
      "num_tokens": 122507140.0,
      "step": 53425
    },
    {
      "entropy": 5.5298449993133545,
      "epoch": 5.132564841498559,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002689394005050026,
      "loss": 4.7272,
      "mean_token_accuracy": 0.2569567546248436,
      "num_tokens": 122519402.0,
      "step": 53430
    },
    {
      "entropy": 5.5870977401733395,
      "epoch": 5.133045148895293,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002689051294594404,
      "loss": 4.7128,
      "mean_token_accuracy": 0.2603044852614403,
      "num_tokens": 122529735.0,
      "step": 53435
    },
    {
      "entropy": 5.491140556335449,
      "epoch": 5.133525456292027,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000268870858555383,
      "loss": 4.5689,
      "mean_token_accuracy": 0.2627863109111786,
      "num_tokens": 122541466.0,
      "step": 53440
    },
    {
      "entropy": 5.512656497955322,
      "epoch": 5.13400576368876,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000268836587793626,
      "loss": 4.6252,
      "mean_token_accuracy": 0.2633108049631119,
      "num_tokens": 122553044.0,
      "step": 53445
    },
    {
      "entropy": 5.444668197631836,
      "epoch": 5.134486071085495,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002688023171749651,
      "loss": 4.6351,
      "mean_token_accuracy": 0.2612456053495407,
      "num_tokens": 122564536.0,
      "step": 53450
    },
    {
      "entropy": 5.4944579124450685,
      "epoch": 5.134966378482229,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002687680467001961,
      "loss": 4.5948,
      "mean_token_accuracy": 0.27198070138692854,
      "num_tokens": 122575940.0,
      "step": 53455
    },
    {
      "entropy": 5.5924256324768065,
      "epoch": 5.1354466858789625,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002687337763701145,
      "loss": 4.7395,
      "mean_token_accuracy": 0.2591514140367508,
      "num_tokens": 122586349.0,
      "step": 53460
    },
    {
      "entropy": 5.571359300613404,
      "epoch": 5.135926993275697,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026869950618551604,
      "loss": 4.7043,
      "mean_token_accuracy": 0.257324181497097,
      "num_tokens": 122598747.0,
      "step": 53465
    },
    {
      "entropy": 5.448312854766845,
      "epoch": 5.13640730067243,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002686652361471962,
      "loss": 4.598,
      "mean_token_accuracy": 0.2669940873980522,
      "num_tokens": 122610515.0,
      "step": 53470
    },
    {
      "entropy": 5.486784934997559,
      "epoch": 5.136887608069165,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026863096625595096,
      "loss": 4.617,
      "mean_token_accuracy": 0.2642500728368759,
      "num_tokens": 122622578.0,
      "step": 53475
    },
    {
      "entropy": 5.441051340103149,
      "epoch": 5.137367915465898,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026859669651257566,
      "loss": 4.6203,
      "mean_token_accuracy": 0.26699415147304534,
      "num_tokens": 122635059.0,
      "step": 53480
    },
    {
      "entropy": 5.562892150878906,
      "epoch": 5.137848222862632,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002685624269178661,
      "loss": 4.6423,
      "mean_token_accuracy": 0.2639608770608902,
      "num_tokens": 122645723.0,
      "step": 53485
    },
    {
      "entropy": 5.520143842697143,
      "epoch": 5.138328530259366,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002685281574726178,
      "loss": 4.6395,
      "mean_token_accuracy": 0.2648613005876541,
      "num_tokens": 122657114.0,
      "step": 53490
    },
    {
      "entropy": 5.549754810333252,
      "epoch": 5.1388088376561,
      "grad_norm": 1.375,
      "learning_rate": 0.0002684938881776265,
      "loss": 4.7389,
      "mean_token_accuracy": 0.2581649884581566,
      "num_tokens": 122668863.0,
      "step": 53495
    },
    {
      "entropy": 5.517233610153198,
      "epoch": 5.1392891450528335,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002684596190336879,
      "loss": 4.5827,
      "mean_token_accuracy": 0.26539460122585296,
      "num_tokens": 122681091.0,
      "step": 53500
    },
    {
      "entropy": 5.4868663311004635,
      "epoch": 5.139769452449568,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002684253500415974,
      "loss": 4.623,
      "mean_token_accuracy": 0.26610043048858645,
      "num_tokens": 122692004.0,
      "step": 53505
    },
    {
      "entropy": 5.573522901535034,
      "epoch": 5.140249759846301,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026839108120215073,
      "loss": 4.7877,
      "mean_token_accuracy": 0.24903466552495956,
      "num_tokens": 122704250.0,
      "step": 53510
    },
    {
      "entropy": 5.487596273422241,
      "epoch": 5.140730067243036,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026835681251614364,
      "loss": 4.6653,
      "mean_token_accuracy": 0.2640439882874489,
      "num_tokens": 122716601.0,
      "step": 53515
    },
    {
      "entropy": 5.565268468856812,
      "epoch": 5.141210374639769,
      "grad_norm": 1.25,
      "learning_rate": 0.00026832254398437164,
      "loss": 4.6919,
      "mean_token_accuracy": 0.2561373606324196,
      "num_tokens": 122727565.0,
      "step": 53520
    },
    {
      "entropy": 5.605062961578369,
      "epoch": 5.141690682036503,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026828827560763024,
      "loss": 4.7207,
      "mean_token_accuracy": 0.2607172608375549,
      "num_tokens": 122738764.0,
      "step": 53525
    },
    {
      "entropy": 5.532729196548462,
      "epoch": 5.142170989433238,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002682540073867152,
      "loss": 4.6579,
      "mean_token_accuracy": 0.2616593539714813,
      "num_tokens": 122749763.0,
      "step": 53530
    },
    {
      "entropy": 5.470004606246948,
      "epoch": 5.142651296829971,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026821973932242206,
      "loss": 4.6211,
      "mean_token_accuracy": 0.26184715181589124,
      "num_tokens": 122762041.0,
      "step": 53535
    },
    {
      "entropy": 5.471022367477417,
      "epoch": 5.1431316042267055,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026818547141554646,
      "loss": 4.6055,
      "mean_token_accuracy": 0.2671834468841553,
      "num_tokens": 122774856.0,
      "step": 53540
    },
    {
      "entropy": 5.458657073974609,
      "epoch": 5.143611911623439,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026815120366688387,
      "loss": 4.6242,
      "mean_token_accuracy": 0.2664401769638062,
      "num_tokens": 122785821.0,
      "step": 53545
    },
    {
      "entropy": 5.512715864181518,
      "epoch": 5.144092219020173,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026811693607723026,
      "loss": 4.6471,
      "mean_token_accuracy": 0.2678177744150162,
      "num_tokens": 122796820.0,
      "step": 53550
    },
    {
      "entropy": 5.503797912597657,
      "epoch": 5.144572526416907,
      "grad_norm": 1.5,
      "learning_rate": 0.00026808266864738074,
      "loss": 4.6824,
      "mean_token_accuracy": 0.25835914313793185,
      "num_tokens": 122808110.0,
      "step": 53555
    },
    {
      "entropy": 5.4905060768127445,
      "epoch": 5.145052833813641,
      "grad_norm": 1.375,
      "learning_rate": 0.00026804840137813115,
      "loss": 4.6405,
      "mean_token_accuracy": 0.26190929412841796,
      "num_tokens": 122819426.0,
      "step": 53560
    },
    {
      "entropy": 5.517619991302491,
      "epoch": 5.145533141210374,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026801413427027704,
      "loss": 4.7067,
      "mean_token_accuracy": 0.25127777457237244,
      "num_tokens": 122830326.0,
      "step": 53565
    },
    {
      "entropy": 5.501135587692261,
      "epoch": 5.146013448607109,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000267979867324614,
      "loss": 4.6419,
      "mean_token_accuracy": 0.25443431437015535,
      "num_tokens": 122841628.0,
      "step": 53570
    },
    {
      "entropy": 5.503137016296387,
      "epoch": 5.146493756003842,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002679456005419376,
      "loss": 4.5695,
      "mean_token_accuracy": 0.2688832268118858,
      "num_tokens": 122851908.0,
      "step": 53575
    },
    {
      "entropy": 5.515015125274658,
      "epoch": 5.1469740634005765,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002679113339230435,
      "loss": 4.6819,
      "mean_token_accuracy": 0.2590707615017891,
      "num_tokens": 122861928.0,
      "step": 53580
    },
    {
      "entropy": 5.482940721511841,
      "epoch": 5.14745437079731,
      "grad_norm": 1.453125,
      "learning_rate": 0.000267877067468727,
      "loss": 4.6092,
      "mean_token_accuracy": 0.2652646407485008,
      "num_tokens": 122872935.0,
      "step": 53585
    },
    {
      "entropy": 5.6168242454528805,
      "epoch": 5.147934678194044,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026784280117978393,
      "loss": 4.7253,
      "mean_token_accuracy": 0.25462296307086946,
      "num_tokens": 122884144.0,
      "step": 53590
    },
    {
      "entropy": 5.48942084312439,
      "epoch": 5.148414985590778,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026780853505700973,
      "loss": 4.5706,
      "mean_token_accuracy": 0.2658432960510254,
      "num_tokens": 122895103.0,
      "step": 53595
    },
    {
      "entropy": 5.497958087921143,
      "epoch": 5.148895292987512,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002677742691012,
      "loss": 4.6337,
      "mean_token_accuracy": 0.26926074326038363,
      "num_tokens": 122907438.0,
      "step": 53600
    },
    {
      "entropy": 5.527880764007568,
      "epoch": 5.149375600384246,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002677400033131503,
      "loss": 4.7135,
      "mean_token_accuracy": 0.25740388929843905,
      "num_tokens": 122920614.0,
      "step": 53605
    },
    {
      "entropy": 5.481552934646606,
      "epoch": 5.14985590778098,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026770573769365624,
      "loss": 4.6178,
      "mean_token_accuracy": 0.2615706294775009,
      "num_tokens": 122932526.0,
      "step": 53610
    },
    {
      "entropy": 5.601031064987183,
      "epoch": 5.150336215177714,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026767147224351326,
      "loss": 4.69,
      "mean_token_accuracy": 0.2595282718539238,
      "num_tokens": 122944708.0,
      "step": 53615
    },
    {
      "entropy": 5.525435543060302,
      "epoch": 5.1508165225744476,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026763720696351684,
      "loss": 4.6031,
      "mean_token_accuracy": 0.2644816726446152,
      "num_tokens": 122955276.0,
      "step": 53620
    },
    {
      "entropy": 5.517215967178345,
      "epoch": 5.151296829971182,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026760294185446267,
      "loss": 4.6561,
      "mean_token_accuracy": 0.26104129701852796,
      "num_tokens": 122966717.0,
      "step": 53625
    },
    {
      "entropy": 5.539074230194092,
      "epoch": 5.151777137367915,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002675686769171462,
      "loss": 4.6767,
      "mean_token_accuracy": 0.2645763874053955,
      "num_tokens": 122977039.0,
      "step": 53630
    },
    {
      "entropy": 5.5140218257904055,
      "epoch": 5.15225744476465,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026753441215236296,
      "loss": 4.6884,
      "mean_token_accuracy": 0.26003326028585433,
      "num_tokens": 122988053.0,
      "step": 53635
    },
    {
      "entropy": 5.485171747207642,
      "epoch": 5.152737752161383,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002675001475609087,
      "loss": 4.5787,
      "mean_token_accuracy": 0.26853003054857255,
      "num_tokens": 122998814.0,
      "step": 53640
    },
    {
      "entropy": 5.525277423858642,
      "epoch": 5.153218059558117,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002674658831435785,
      "loss": 4.643,
      "mean_token_accuracy": 0.2660157859325409,
      "num_tokens": 123009540.0,
      "step": 53645
    },
    {
      "entropy": 5.552337408065796,
      "epoch": 5.153698366954851,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026743161890116826,
      "loss": 4.6686,
      "mean_token_accuracy": 0.25290638357400896,
      "num_tokens": 123021942.0,
      "step": 53650
    },
    {
      "entropy": 5.507959508895874,
      "epoch": 5.154178674351585,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002673973548344733,
      "loss": 4.5463,
      "mean_token_accuracy": 0.27508341073989867,
      "num_tokens": 123033821.0,
      "step": 53655
    },
    {
      "entropy": 5.47700777053833,
      "epoch": 5.154658981748319,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002673630909442891,
      "loss": 4.6191,
      "mean_token_accuracy": 0.2636307135224342,
      "num_tokens": 123044411.0,
      "step": 53660
    },
    {
      "entropy": 5.429988193511963,
      "epoch": 5.155139289145053,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002673288272314114,
      "loss": 4.6144,
      "mean_token_accuracy": 0.26252923607826234,
      "num_tokens": 123055605.0,
      "step": 53665
    },
    {
      "entropy": 5.506734132766724,
      "epoch": 5.155619596541786,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002672945636966356,
      "loss": 4.6876,
      "mean_token_accuracy": 0.259563672542572,
      "num_tokens": 123066842.0,
      "step": 53670
    },
    {
      "entropy": 5.530355310440063,
      "epoch": 5.156099903938521,
      "grad_norm": 1.234375,
      "learning_rate": 0.000267260300340757,
      "loss": 4.6887,
      "mean_token_accuracy": 0.26061061322689055,
      "num_tokens": 123079118.0,
      "step": 53675
    },
    {
      "entropy": 5.569586086273193,
      "epoch": 5.156580211335255,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002672260371645713,
      "loss": 4.6806,
      "mean_token_accuracy": 0.2621635600924492,
      "num_tokens": 123090176.0,
      "step": 53680
    },
    {
      "entropy": 5.47922534942627,
      "epoch": 5.1570605187319885,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026719177416887393,
      "loss": 4.6443,
      "mean_token_accuracy": 0.26614016741514207,
      "num_tokens": 123101861.0,
      "step": 53685
    },
    {
      "entropy": 5.498856878280639,
      "epoch": 5.157540826128723,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002671575113544603,
      "loss": 4.6429,
      "mean_token_accuracy": 0.2645736336708069,
      "num_tokens": 123113392.0,
      "step": 53690
    },
    {
      "entropy": 5.5243431568145756,
      "epoch": 5.158021133525456,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002671232487221261,
      "loss": 4.6249,
      "mean_token_accuracy": 0.2611073568463326,
      "num_tokens": 123123778.0,
      "step": 53695
    },
    {
      "entropy": 5.542735147476196,
      "epoch": 5.1585014409221905,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026708898627266663,
      "loss": 4.6638,
      "mean_token_accuracy": 0.26169671565294267,
      "num_tokens": 123135337.0,
      "step": 53700
    },
    {
      "entropy": 5.552350378036499,
      "epoch": 5.158981748318924,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002670547240068774,
      "loss": 4.6415,
      "mean_token_accuracy": 0.25851753950119016,
      "num_tokens": 123145903.0,
      "step": 53705
    },
    {
      "entropy": 5.500127696990967,
      "epoch": 5.159462055715658,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002670204619255538,
      "loss": 4.6276,
      "mean_token_accuracy": 0.2634359821677208,
      "num_tokens": 123158006.0,
      "step": 53710
    },
    {
      "entropy": 5.558793640136718,
      "epoch": 5.159942363112392,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026698620002949154,
      "loss": 4.6692,
      "mean_token_accuracy": 0.2605114936828613,
      "num_tokens": 123170473.0,
      "step": 53715
    },
    {
      "entropy": 5.5524406909942625,
      "epoch": 5.160422670509126,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002669519383194858,
      "loss": 4.714,
      "mean_token_accuracy": 0.25903272479772566,
      "num_tokens": 123183231.0,
      "step": 53720
    },
    {
      "entropy": 5.450572109222412,
      "epoch": 5.1609029779058595,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002669176767963322,
      "loss": 4.542,
      "mean_token_accuracy": 0.27125568091869356,
      "num_tokens": 123194726.0,
      "step": 53725
    },
    {
      "entropy": 5.501053094863892,
      "epoch": 5.161383285302594,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026688341546082616,
      "loss": 4.6097,
      "mean_token_accuracy": 0.2668407201766968,
      "num_tokens": 123206205.0,
      "step": 53730
    },
    {
      "entropy": 5.601476335525513,
      "epoch": 5.161863592699327,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002668491543137631,
      "loss": 4.7053,
      "mean_token_accuracy": 0.25517233163118364,
      "num_tokens": 123217465.0,
      "step": 53735
    },
    {
      "entropy": 5.556103754043579,
      "epoch": 5.162343900096062,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002668148933559385,
      "loss": 4.6847,
      "mean_token_accuracy": 0.2595504730939865,
      "num_tokens": 123229655.0,
      "step": 53740
    },
    {
      "entropy": 5.587949275970459,
      "epoch": 5.162824207492795,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002667806325881477,
      "loss": 4.6898,
      "mean_token_accuracy": 0.26147216707468035,
      "num_tokens": 123240313.0,
      "step": 53745
    },
    {
      "entropy": 5.409068012237549,
      "epoch": 5.163304514889529,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026674637201118623,
      "loss": 4.5427,
      "mean_token_accuracy": 0.2741131976246834,
      "num_tokens": 123252663.0,
      "step": 53750
    },
    {
      "entropy": 5.641554641723633,
      "epoch": 5.163784822286263,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002667121116258495,
      "loss": 4.7473,
      "mean_token_accuracy": 0.2493695229291916,
      "num_tokens": 123263678.0,
      "step": 53755
    },
    {
      "entropy": 5.496404504776001,
      "epoch": 5.164265129682997,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000266677851432933,
      "loss": 4.5789,
      "mean_token_accuracy": 0.26684722751379014,
      "num_tokens": 123273899.0,
      "step": 53760
    },
    {
      "entropy": 5.475841951370239,
      "epoch": 5.164745437079731,
      "grad_norm": 1.4375,
      "learning_rate": 0.000266643591433232,
      "loss": 4.6237,
      "mean_token_accuracy": 0.26242931336164477,
      "num_tokens": 123285380.0,
      "step": 53765
    },
    {
      "entropy": 5.493885564804077,
      "epoch": 5.165225744476465,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026660933162754195,
      "loss": 4.6247,
      "mean_token_accuracy": 0.2657759919762611,
      "num_tokens": 123296350.0,
      "step": 53770
    },
    {
      "entropy": 5.51646785736084,
      "epoch": 5.165706051873199,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002665750720166584,
      "loss": 4.6198,
      "mean_token_accuracy": 0.26666283458471296,
      "num_tokens": 123307050.0,
      "step": 53775
    },
    {
      "entropy": 5.594228458404541,
      "epoch": 5.166186359269933,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026654081260137666,
      "loss": 4.7134,
      "mean_token_accuracy": 0.25647643953561783,
      "num_tokens": 123318333.0,
      "step": 53780
    },
    {
      "entropy": 5.579801845550537,
      "epoch": 5.166666666666667,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002665065533824921,
      "loss": 4.7693,
      "mean_token_accuracy": 0.24967241585254668,
      "num_tokens": 123330145.0,
      "step": 53785
    },
    {
      "entropy": 5.520553636550903,
      "epoch": 5.1671469740634,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026647229436080013,
      "loss": 4.6613,
      "mean_token_accuracy": 0.26072648763656614,
      "num_tokens": 123341409.0,
      "step": 53790
    },
    {
      "entropy": 5.488593196868896,
      "epoch": 5.167627281460135,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002664380355370962,
      "loss": 4.5915,
      "mean_token_accuracy": 0.26715020686388014,
      "num_tokens": 123352929.0,
      "step": 53795
    },
    {
      "entropy": 5.469095849990845,
      "epoch": 5.168107588856868,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002664037769121756,
      "loss": 4.5756,
      "mean_token_accuracy": 0.2734148263931274,
      "num_tokens": 123363200.0,
      "step": 53800
    },
    {
      "entropy": 5.5155477046966555,
      "epoch": 5.1685878962536025,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002663695184868338,
      "loss": 4.6417,
      "mean_token_accuracy": 0.26304286271333693,
      "num_tokens": 123374889.0,
      "step": 53805
    },
    {
      "entropy": 5.543001699447632,
      "epoch": 5.169068203650336,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002663352602618661,
      "loss": 4.7042,
      "mean_token_accuracy": 0.26115291863679885,
      "num_tokens": 123385866.0,
      "step": 53810
    },
    {
      "entropy": 5.610185337066651,
      "epoch": 5.16954851104707,
      "grad_norm": 1.328125,
      "learning_rate": 0.000266301002238068,
      "loss": 4.7756,
      "mean_token_accuracy": 0.2471502333879471,
      "num_tokens": 123396607.0,
      "step": 53815
    },
    {
      "entropy": 5.511433458328247,
      "epoch": 5.170028818443804,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002662667444162349,
      "loss": 4.6756,
      "mean_token_accuracy": 0.2606548428535461,
      "num_tokens": 123407847.0,
      "step": 53820
    },
    {
      "entropy": 5.593584299087524,
      "epoch": 5.170509125840538,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026623248679716197,
      "loss": 4.7154,
      "mean_token_accuracy": 0.2581436976790428,
      "num_tokens": 123419354.0,
      "step": 53825
    },
    {
      "entropy": 5.483731937408447,
      "epoch": 5.1709894332372714,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002661982293816446,
      "loss": 4.6104,
      "mean_token_accuracy": 0.25874750316143036,
      "num_tokens": 123431271.0,
      "step": 53830
    },
    {
      "entropy": 5.541299104690552,
      "epoch": 5.171469740634006,
      "grad_norm": 1.25,
      "learning_rate": 0.0002661639721704782,
      "loss": 4.6927,
      "mean_token_accuracy": 0.2605564698576927,
      "num_tokens": 123444046.0,
      "step": 53835
    },
    {
      "entropy": 5.552159547805786,
      "epoch": 5.17195004803074,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026612971516445824,
      "loss": 4.6762,
      "mean_token_accuracy": 0.2562086895108223,
      "num_tokens": 123454999.0,
      "step": 53840
    },
    {
      "entropy": 5.513676309585572,
      "epoch": 5.1724303554274735,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002660954583643798,
      "loss": 4.6519,
      "mean_token_accuracy": 0.25909584760665894,
      "num_tokens": 123466568.0,
      "step": 53845
    },
    {
      "entropy": 5.516601324081421,
      "epoch": 5.172910662824208,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026606120177103857,
      "loss": 4.7218,
      "mean_token_accuracy": 0.2569027543067932,
      "num_tokens": 123478406.0,
      "step": 53850
    },
    {
      "entropy": 5.546694898605347,
      "epoch": 5.173390970220941,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002660269453852296,
      "loss": 4.6195,
      "mean_token_accuracy": 0.263571472465992,
      "num_tokens": 123489479.0,
      "step": 53855
    },
    {
      "entropy": 5.587923622131347,
      "epoch": 5.173871277617676,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026599268920774827,
      "loss": 4.7424,
      "mean_token_accuracy": 0.2579196885228157,
      "num_tokens": 123502899.0,
      "step": 53860
    },
    {
      "entropy": 5.524159383773804,
      "epoch": 5.174351585014409,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026595843323939,
      "loss": 4.6084,
      "mean_token_accuracy": 0.2598324462771416,
      "num_tokens": 123514333.0,
      "step": 53865
    },
    {
      "entropy": 5.49234619140625,
      "epoch": 5.174831892411143,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026592417748095003,
      "loss": 4.5918,
      "mean_token_accuracy": 0.27378859370946884,
      "num_tokens": 123525889.0,
      "step": 53870
    },
    {
      "entropy": 5.54571123123169,
      "epoch": 5.175312199807877,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002658899219332238,
      "loss": 4.637,
      "mean_token_accuracy": 0.2655519753694534,
      "num_tokens": 123537195.0,
      "step": 53875
    },
    {
      "entropy": 5.566466331481934,
      "epoch": 5.175792507204611,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002658556665970065,
      "loss": 4.7256,
      "mean_token_accuracy": 0.2535940229892731,
      "num_tokens": 123548939.0,
      "step": 53880
    },
    {
      "entropy": 5.676243638992309,
      "epoch": 5.176272814601345,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026582141147309335,
      "loss": 4.7777,
      "mean_token_accuracy": 0.2506107956171036,
      "num_tokens": 123561024.0,
      "step": 53885
    },
    {
      "entropy": 5.597729063034057,
      "epoch": 5.176753121998079,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026578715656227987,
      "loss": 4.7237,
      "mean_token_accuracy": 0.25263654440641403,
      "num_tokens": 123574016.0,
      "step": 53890
    },
    {
      "entropy": 5.4600177764892575,
      "epoch": 5.177233429394812,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002657529018653612,
      "loss": 4.584,
      "mean_token_accuracy": 0.2661802038550377,
      "num_tokens": 123586645.0,
      "step": 53895
    },
    {
      "entropy": 5.546150016784668,
      "epoch": 5.177713736791547,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002657186473831328,
      "loss": 4.6401,
      "mean_token_accuracy": 0.26834252327680586,
      "num_tokens": 123597830.0,
      "step": 53900
    },
    {
      "entropy": 5.5711017608642575,
      "epoch": 5.17819404418828,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002656843931163897,
      "loss": 4.6773,
      "mean_token_accuracy": 0.2545118182897568,
      "num_tokens": 123608711.0,
      "step": 53905
    },
    {
      "entropy": 5.560781574249267,
      "epoch": 5.178674351585014,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002656501390659275,
      "loss": 4.6943,
      "mean_token_accuracy": 0.26482574790716173,
      "num_tokens": 123619414.0,
      "step": 53910
    },
    {
      "entropy": 5.583746242523193,
      "epoch": 5.179154658981748,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002656158852325412,
      "loss": 4.694,
      "mean_token_accuracy": 0.2617642402648926,
      "num_tokens": 123632557.0,
      "step": 53915
    },
    {
      "entropy": 5.520421504974365,
      "epoch": 5.179634966378482,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002655816316170262,
      "loss": 4.5949,
      "mean_token_accuracy": 0.26308528184890745,
      "num_tokens": 123645182.0,
      "step": 53920
    },
    {
      "entropy": 5.47586727142334,
      "epoch": 5.1801152737752165,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002655473782201777,
      "loss": 4.5833,
      "mean_token_accuracy": 0.25977045148611067,
      "num_tokens": 123656861.0,
      "step": 53925
    },
    {
      "entropy": 5.529742479324341,
      "epoch": 5.18059558117195,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026551312504279105,
      "loss": 4.7094,
      "mean_token_accuracy": 0.2548145160079002,
      "num_tokens": 123667586.0,
      "step": 53930
    },
    {
      "entropy": 5.437369108200073,
      "epoch": 5.181075888568684,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002654788720856615,
      "loss": 4.5172,
      "mean_token_accuracy": 0.27495688796043394,
      "num_tokens": 123678303.0,
      "step": 53935
    },
    {
      "entropy": 5.5040748596191404,
      "epoch": 5.181556195965418,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002654446193495843,
      "loss": 4.6702,
      "mean_token_accuracy": 0.2607876271009445,
      "num_tokens": 123689847.0,
      "step": 53940
    },
    {
      "entropy": 5.457675933837891,
      "epoch": 5.182036503362152,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026541036683535457,
      "loss": 4.6324,
      "mean_token_accuracy": 0.2627925112843513,
      "num_tokens": 123700234.0,
      "step": 53945
    },
    {
      "entropy": 5.5880903720855715,
      "epoch": 5.1825168107588855,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026537611454376765,
      "loss": 4.7419,
      "mean_token_accuracy": 0.25264094322919844,
      "num_tokens": 123712340.0,
      "step": 53950
    },
    {
      "entropy": 5.549473905563355,
      "epoch": 5.18299711815562,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026534186247561875,
      "loss": 4.7141,
      "mean_token_accuracy": 0.2540805354714394,
      "num_tokens": 123723725.0,
      "step": 53955
    },
    {
      "entropy": 5.574302673339844,
      "epoch": 5.183477425552353,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026530761063170324,
      "loss": 4.6648,
      "mean_token_accuracy": 0.26784915030002593,
      "num_tokens": 123735849.0,
      "step": 53960
    },
    {
      "entropy": 5.55084457397461,
      "epoch": 5.183957732949088,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002652733590128161,
      "loss": 4.6351,
      "mean_token_accuracy": 0.25724791586399076,
      "num_tokens": 123747333.0,
      "step": 53965
    },
    {
      "entropy": 5.529787158966064,
      "epoch": 5.184438040345821,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026523910761975286,
      "loss": 4.6786,
      "mean_token_accuracy": 0.26405244171619413,
      "num_tokens": 123760391.0,
      "step": 53970
    },
    {
      "entropy": 5.553308391571045,
      "epoch": 5.184918347742555,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002652048564533084,
      "loss": 4.7049,
      "mean_token_accuracy": 0.2557730808854103,
      "num_tokens": 123771710.0,
      "step": 53975
    },
    {
      "entropy": 5.532438564300537,
      "epoch": 5.185398655139289,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002651706055142781,
      "loss": 4.6672,
      "mean_token_accuracy": 0.2661603316664696,
      "num_tokens": 123783316.0,
      "step": 53980
    },
    {
      "entropy": 5.611898565292359,
      "epoch": 5.185878962536023,
      "grad_norm": 1.375,
      "learning_rate": 0.0002651363548034572,
      "loss": 4.748,
      "mean_token_accuracy": 0.2468015283346176,
      "num_tokens": 123793958.0,
      "step": 53985
    },
    {
      "entropy": 5.593298101425171,
      "epoch": 5.1863592699327565,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002651021043216408,
      "loss": 4.7107,
      "mean_token_accuracy": 0.25669427365064623,
      "num_tokens": 123806370.0,
      "step": 53990
    },
    {
      "entropy": 5.62945966720581,
      "epoch": 5.186839577329491,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002650678540696242,
      "loss": 4.8025,
      "mean_token_accuracy": 0.24492372423410416,
      "num_tokens": 123818059.0,
      "step": 53995
    },
    {
      "entropy": 5.57509593963623,
      "epoch": 5.187319884726225,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026503360404820254,
      "loss": 4.6689,
      "mean_token_accuracy": 0.26476056724786756,
      "num_tokens": 123829338.0,
      "step": 54000
    },
    {
      "epoch": 5.187319884726225,
      "eval_entropy": 5.237379206732271,
      "eval_loss": 4.7721123695373535,
      "eval_mean_token_accuracy": 0.2626099801285856,
      "eval_num_tokens": 123829338.0,
      "eval_runtime": 20.6553,
      "eval_samples_per_second": 1588.693,
      "eval_steps_per_second": 198.593,
      "step": 54000
    },
    {
      "entropy": 5.466054487228393,
      "epoch": 5.187800192122959,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026499935425817106,
      "loss": 4.5834,
      "mean_token_accuracy": 0.2699845194816589,
      "num_tokens": 123841470.0,
      "step": 54005
    },
    {
      "entropy": 5.5014256000518795,
      "epoch": 5.188280499519693,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002649651047003248,
      "loss": 4.6923,
      "mean_token_accuracy": 0.264163176715374,
      "num_tokens": 123854236.0,
      "step": 54010
    },
    {
      "entropy": 5.547169733047485,
      "epoch": 5.188760806916426,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026493085537545897,
      "loss": 4.6635,
      "mean_token_accuracy": 0.2583999067544937,
      "num_tokens": 123865509.0,
      "step": 54015
    },
    {
      "entropy": 5.583293771743774,
      "epoch": 5.189241114313161,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026489660628436877,
      "loss": 4.6715,
      "mean_token_accuracy": 0.25985675752162934,
      "num_tokens": 123876250.0,
      "step": 54020
    },
    {
      "entropy": 5.542149782180786,
      "epoch": 5.189721421709894,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002648623574278494,
      "loss": 4.7382,
      "mean_token_accuracy": 0.25791009813547133,
      "num_tokens": 123888154.0,
      "step": 54025
    },
    {
      "entropy": 5.567514038085937,
      "epoch": 5.1902017291066285,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026482810880669607,
      "loss": 4.6648,
      "mean_token_accuracy": 0.2615774840116501,
      "num_tokens": 123899622.0,
      "step": 54030
    },
    {
      "entropy": 5.62809648513794,
      "epoch": 5.190682036503362,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026479386042170376,
      "loss": 4.723,
      "mean_token_accuracy": 0.25886420011520384,
      "num_tokens": 123911975.0,
      "step": 54035
    },
    {
      "entropy": 5.50596866607666,
      "epoch": 5.191162343900096,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026475961227366774,
      "loss": 4.6028,
      "mean_token_accuracy": 0.26992115676403045,
      "num_tokens": 123923186.0,
      "step": 54040
    },
    {
      "entropy": 5.50607476234436,
      "epoch": 5.19164265129683,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002647253643633831,
      "loss": 4.6174,
      "mean_token_accuracy": 0.25983927249908445,
      "num_tokens": 123934002.0,
      "step": 54045
    },
    {
      "entropy": 5.486478233337403,
      "epoch": 5.192122958693564,
      "grad_norm": 1.390625,
      "learning_rate": 0.000264691116691645,
      "loss": 4.6392,
      "mean_token_accuracy": 0.26186815947294234,
      "num_tokens": 123945630.0,
      "step": 54050
    },
    {
      "entropy": 5.561257791519165,
      "epoch": 5.192603266090297,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026465686925924856,
      "loss": 4.6634,
      "mean_token_accuracy": 0.2666387766599655,
      "num_tokens": 123957313.0,
      "step": 54055
    },
    {
      "entropy": 5.547973155975342,
      "epoch": 5.193083573487032,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026462262206698896,
      "loss": 4.6829,
      "mean_token_accuracy": 0.26213539093732835,
      "num_tokens": 123968094.0,
      "step": 54060
    },
    {
      "entropy": 5.6215098857879635,
      "epoch": 5.193563880883765,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026458837511566114,
      "loss": 4.6778,
      "mean_token_accuracy": 0.25547039806842803,
      "num_tokens": 123980758.0,
      "step": 54065
    },
    {
      "entropy": 5.578080940246582,
      "epoch": 5.1940441882804995,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002645541284060604,
      "loss": 4.6865,
      "mean_token_accuracy": 0.25903760343790055,
      "num_tokens": 123992601.0,
      "step": 54070
    },
    {
      "entropy": 5.482993459701538,
      "epoch": 5.194524495677234,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026451988193898176,
      "loss": 4.59,
      "mean_token_accuracy": 0.26345366388559344,
      "num_tokens": 124003522.0,
      "step": 54075
    },
    {
      "entropy": 5.520755338668823,
      "epoch": 5.195004803073967,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026448563571522033,
      "loss": 4.6577,
      "mean_token_accuracy": 0.2712667569518089,
      "num_tokens": 124014992.0,
      "step": 54080
    },
    {
      "entropy": 5.562494707107544,
      "epoch": 5.195485110470702,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002644513897355713,
      "loss": 4.6287,
      "mean_token_accuracy": 0.2672502636909485,
      "num_tokens": 124026079.0,
      "step": 54085
    },
    {
      "entropy": 5.546261501312256,
      "epoch": 5.195965417867435,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002644171440008296,
      "loss": 4.6356,
      "mean_token_accuracy": 0.2663283497095108,
      "num_tokens": 124037921.0,
      "step": 54090
    },
    {
      "entropy": 5.544683742523193,
      "epoch": 5.196445725264169,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026438289851179043,
      "loss": 4.7629,
      "mean_token_accuracy": 0.25337263345718386,
      "num_tokens": 124049291.0,
      "step": 54095
    },
    {
      "entropy": 5.542958164215088,
      "epoch": 5.196926032660903,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026434865326924883,
      "loss": 4.6728,
      "mean_token_accuracy": 0.26376787573099136,
      "num_tokens": 124059644.0,
      "step": 54100
    },
    {
      "entropy": 5.597841072082519,
      "epoch": 5.197406340057637,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002643144082739998,
      "loss": 4.6977,
      "mean_token_accuracy": 0.26227881759405136,
      "num_tokens": 124069387.0,
      "step": 54105
    },
    {
      "entropy": 5.644629621505738,
      "epoch": 5.197886647454371,
      "grad_norm": 1.5,
      "learning_rate": 0.00026428016352683854,
      "loss": 4.7518,
      "mean_token_accuracy": 0.25120740234851835,
      "num_tokens": 124079855.0,
      "step": 54110
    },
    {
      "entropy": 5.540892934799194,
      "epoch": 5.198366954851105,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026424591902856004,
      "loss": 4.6122,
      "mean_token_accuracy": 0.2676249906420708,
      "num_tokens": 124091464.0,
      "step": 54115
    },
    {
      "entropy": 5.589572477340698,
      "epoch": 5.198847262247838,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026421167477995943,
      "loss": 4.7656,
      "mean_token_accuracy": 0.24648280143737794,
      "num_tokens": 124102668.0,
      "step": 54120
    },
    {
      "entropy": 5.443126010894775,
      "epoch": 5.199327569644573,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026417743078183164,
      "loss": 4.5669,
      "mean_token_accuracy": 0.2630400896072388,
      "num_tokens": 124114900.0,
      "step": 54125
    },
    {
      "entropy": 5.542607021331787,
      "epoch": 5.199807877041306,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002641431870349718,
      "loss": 4.6363,
      "mean_token_accuracy": 0.25117470175027845,
      "num_tokens": 124126259.0,
      "step": 54130
    },
    {
      "entropy": 5.547062063217163,
      "epoch": 5.20028818443804,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026410894354017485,
      "loss": 4.6555,
      "mean_token_accuracy": 0.26289336532354357,
      "num_tokens": 124138097.0,
      "step": 54135
    },
    {
      "entropy": 5.479561138153076,
      "epoch": 5.200768491834774,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00026407470029823595,
      "loss": 4.6117,
      "mean_token_accuracy": 0.2709352284669876,
      "num_tokens": 124149165.0,
      "step": 54140
    },
    {
      "entropy": 5.586152696609497,
      "epoch": 5.201248799231508,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026404045730995,
      "loss": 4.8322,
      "mean_token_accuracy": 0.24607330858707427,
      "num_tokens": 124160473.0,
      "step": 54145
    },
    {
      "entropy": 5.52005615234375,
      "epoch": 5.2017291066282425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026400621457611226,
      "loss": 4.627,
      "mean_token_accuracy": 0.26155584305524826,
      "num_tokens": 124171673.0,
      "step": 54150
    },
    {
      "entropy": 5.553641891479492,
      "epoch": 5.202209414024976,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002639719720975174,
      "loss": 4.6202,
      "mean_token_accuracy": 0.2640159383416176,
      "num_tokens": 124183116.0,
      "step": 54155
    },
    {
      "entropy": 5.549924421310425,
      "epoch": 5.20268972142171,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026393772987496066,
      "loss": 4.6944,
      "mean_token_accuracy": 0.2573907092213631,
      "num_tokens": 124194268.0,
      "step": 54160
    },
    {
      "entropy": 5.5889452457427975,
      "epoch": 5.203170028818444,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026390348790923695,
      "loss": 4.789,
      "mean_token_accuracy": 0.25140427947044375,
      "num_tokens": 124205484.0,
      "step": 54165
    },
    {
      "entropy": 5.549752044677734,
      "epoch": 5.203650336215178,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026386924620114136,
      "loss": 4.5714,
      "mean_token_accuracy": 0.2709269031882286,
      "num_tokens": 124215631.0,
      "step": 54170
    },
    {
      "entropy": 5.571962547302246,
      "epoch": 5.2041306436119115,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002638350047514687,
      "loss": 4.6709,
      "mean_token_accuracy": 0.2560152396559715,
      "num_tokens": 124227243.0,
      "step": 54175
    },
    {
      "entropy": 5.572308874130249,
      "epoch": 5.204610951008646,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002638007635610143,
      "loss": 4.6692,
      "mean_token_accuracy": 0.256303483247757,
      "num_tokens": 124238956.0,
      "step": 54180
    },
    {
      "entropy": 5.521768617630005,
      "epoch": 5.205091258405379,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002637665226305728,
      "loss": 4.6202,
      "mean_token_accuracy": 0.2680636510252953,
      "num_tokens": 124249770.0,
      "step": 54185
    },
    {
      "entropy": 5.485435676574707,
      "epoch": 5.2055715658021136,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026373228196093916,
      "loss": 4.5818,
      "mean_token_accuracy": 0.2696264207363129,
      "num_tokens": 124260976.0,
      "step": 54190
    },
    {
      "entropy": 5.503664207458496,
      "epoch": 5.206051873198847,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002636980415529086,
      "loss": 4.5833,
      "mean_token_accuracy": 0.2685582637786865,
      "num_tokens": 124271967.0,
      "step": 54195
    },
    {
      "entropy": 5.585066509246826,
      "epoch": 5.206532180595581,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026366380140727596,
      "loss": 4.6373,
      "mean_token_accuracy": 0.2651407092809677,
      "num_tokens": 124283732.0,
      "step": 54200
    },
    {
      "entropy": 5.606437158584595,
      "epoch": 5.207012487992315,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026362956152483616,
      "loss": 4.7655,
      "mean_token_accuracy": 0.24525614380836486,
      "num_tokens": 124296732.0,
      "step": 54205
    },
    {
      "entropy": 5.5085954666137695,
      "epoch": 5.207492795389049,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026359532190638413,
      "loss": 4.6187,
      "mean_token_accuracy": 0.26462615728378297,
      "num_tokens": 124307755.0,
      "step": 54210
    },
    {
      "entropy": 5.500861644744873,
      "epoch": 5.2079731027857825,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026356108255271504,
      "loss": 4.6137,
      "mean_token_accuracy": 0.2571627736091614,
      "num_tokens": 124319204.0,
      "step": 54215
    },
    {
      "entropy": 5.566493129730224,
      "epoch": 5.208453410182517,
      "grad_norm": 1.5,
      "learning_rate": 0.00026352684346462354,
      "loss": 4.6505,
      "mean_token_accuracy": 0.2651854410767555,
      "num_tokens": 124330430.0,
      "step": 54220
    },
    {
      "entropy": 5.4860883235931395,
      "epoch": 5.208933717579251,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002634926046429046,
      "loss": 4.6244,
      "mean_token_accuracy": 0.2604998707771301,
      "num_tokens": 124342025.0,
      "step": 54225
    },
    {
      "entropy": 5.533004570007324,
      "epoch": 5.209414024975985,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002634583660883532,
      "loss": 4.6992,
      "mean_token_accuracy": 0.25164759904146194,
      "num_tokens": 124352548.0,
      "step": 54230
    },
    {
      "entropy": 5.598302412033081,
      "epoch": 5.209894332372719,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026342412780176435,
      "loss": 4.7106,
      "mean_token_accuracy": 0.2526181012392044,
      "num_tokens": 124364208.0,
      "step": 54235
    },
    {
      "entropy": 5.621190881729126,
      "epoch": 5.210374639769452,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026338988978393284,
      "loss": 4.6878,
      "mean_token_accuracy": 0.25685794800519945,
      "num_tokens": 124375462.0,
      "step": 54240
    },
    {
      "entropy": 5.52484016418457,
      "epoch": 5.210854947166187,
      "grad_norm": 1.375,
      "learning_rate": 0.00026335565203565377,
      "loss": 4.5928,
      "mean_token_accuracy": 0.2705108135938644,
      "num_tokens": 124387070.0,
      "step": 54245
    },
    {
      "entropy": 5.547539949417114,
      "epoch": 5.21133525456292,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002633214145577217,
      "loss": 4.6987,
      "mean_token_accuracy": 0.2555496975779533,
      "num_tokens": 124398429.0,
      "step": 54250
    },
    {
      "entropy": 5.586432647705078,
      "epoch": 5.2118155619596545,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002632871773509317,
      "loss": 4.6867,
      "mean_token_accuracy": 0.25660969167947767,
      "num_tokens": 124409212.0,
      "step": 54255
    },
    {
      "entropy": 5.538433170318603,
      "epoch": 5.212295869356388,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026325294041607873,
      "loss": 4.6333,
      "mean_token_accuracy": 0.2546467140316963,
      "num_tokens": 124421302.0,
      "step": 54260
    },
    {
      "entropy": 5.514083909988403,
      "epoch": 5.212776176753122,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002632187037539576,
      "loss": 4.671,
      "mean_token_accuracy": 0.2611333057284355,
      "num_tokens": 124434593.0,
      "step": 54265
    },
    {
      "entropy": 5.635720586776733,
      "epoch": 5.213256484149856,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002631844673653631,
      "loss": 4.7925,
      "mean_token_accuracy": 0.24852387607097626,
      "num_tokens": 124446223.0,
      "step": 54270
    },
    {
      "entropy": 5.651836729049682,
      "epoch": 5.21373679154659,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002631502312510904,
      "loss": 4.773,
      "mean_token_accuracy": 0.24980027973651886,
      "num_tokens": 124457978.0,
      "step": 54275
    },
    {
      "entropy": 5.578315734863281,
      "epoch": 5.214217098943323,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026311599541193394,
      "loss": 4.7151,
      "mean_token_accuracy": 0.25400244146585466,
      "num_tokens": 124468565.0,
      "step": 54280
    },
    {
      "entropy": 5.529309940338135,
      "epoch": 5.214697406340058,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002630817598486888,
      "loss": 4.6246,
      "mean_token_accuracy": 0.263103586435318,
      "num_tokens": 124480010.0,
      "step": 54285
    },
    {
      "entropy": 5.547715520858764,
      "epoch": 5.215177713736791,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002630475245621499,
      "loss": 4.6996,
      "mean_token_accuracy": 0.2590800210833549,
      "num_tokens": 124491539.0,
      "step": 54290
    },
    {
      "entropy": 5.53080005645752,
      "epoch": 5.2156580211335255,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002630132895531119,
      "loss": 4.692,
      "mean_token_accuracy": 0.25923787951469424,
      "num_tokens": 124503605.0,
      "step": 54295
    },
    {
      "entropy": 5.474802255630493,
      "epoch": 5.216138328530259,
      "grad_norm": 1.25,
      "learning_rate": 0.0002629790548223697,
      "loss": 4.5781,
      "mean_token_accuracy": 0.2682120889425278,
      "num_tokens": 124514867.0,
      "step": 54300
    },
    {
      "entropy": 5.519566822052002,
      "epoch": 5.216618635926993,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002629448203707183,
      "loss": 4.5971,
      "mean_token_accuracy": 0.25817507058382033,
      "num_tokens": 124526382.0,
      "step": 54305
    },
    {
      "entropy": 5.492760086059571,
      "epoch": 5.217098943323728,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026291058619895227,
      "loss": 4.5619,
      "mean_token_accuracy": 0.2672959163784981,
      "num_tokens": 124538379.0,
      "step": 54310
    },
    {
      "entropy": 5.5772558689117435,
      "epoch": 5.217579250720461,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002628763523078666,
      "loss": 4.732,
      "mean_token_accuracy": 0.25197682082653045,
      "num_tokens": 124548804.0,
      "step": 54315
    },
    {
      "entropy": 5.553146600723267,
      "epoch": 5.218059558117195,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026284211869825587,
      "loss": 4.689,
      "mean_token_accuracy": 0.26213688403367996,
      "num_tokens": 124560198.0,
      "step": 54320
    },
    {
      "entropy": 5.5535805225372314,
      "epoch": 5.218539865513929,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002628078853709151,
      "loss": 4.6552,
      "mean_token_accuracy": 0.2640894755721092,
      "num_tokens": 124572299.0,
      "step": 54325
    },
    {
      "entropy": 5.664479684829712,
      "epoch": 5.219020172910663,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002627736523266391,
      "loss": 4.8057,
      "mean_token_accuracy": 0.25363700836896896,
      "num_tokens": 124583590.0,
      "step": 54330
    },
    {
      "entropy": 5.55840311050415,
      "epoch": 5.2195004803073966,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002627394195662226,
      "loss": 4.5961,
      "mean_token_accuracy": 0.26573398560285566,
      "num_tokens": 124594709.0,
      "step": 54335
    },
    {
      "entropy": 5.481765031814575,
      "epoch": 5.219980787704131,
      "grad_norm": 1.375,
      "learning_rate": 0.0002627051870904603,
      "loss": 4.6484,
      "mean_token_accuracy": 0.2632294297218323,
      "num_tokens": 124606245.0,
      "step": 54340
    },
    {
      "entropy": 5.5267805576324465,
      "epoch": 5.220461095100864,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000262670954900147,
      "loss": 4.6922,
      "mean_token_accuracy": 0.2597979918122292,
      "num_tokens": 124618124.0,
      "step": 54345
    },
    {
      "entropy": 5.524998903274536,
      "epoch": 5.220941402497599,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002626367229960777,
      "loss": 4.6617,
      "mean_token_accuracy": 0.2541933223605156,
      "num_tokens": 124630793.0,
      "step": 54350
    },
    {
      "entropy": 5.65198860168457,
      "epoch": 5.221421709894332,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026260249137904684,
      "loss": 4.7571,
      "mean_token_accuracy": 0.2526246041059494,
      "num_tokens": 124642183.0,
      "step": 54355
    },
    {
      "entropy": 5.539201641082764,
      "epoch": 5.221902017291066,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026256826004984935,
      "loss": 4.5984,
      "mean_token_accuracy": 0.2649953380227089,
      "num_tokens": 124653582.0,
      "step": 54360
    },
    {
      "entropy": 5.5751667499542235,
      "epoch": 5.2223823246878,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026253402900928,
      "loss": 4.7739,
      "mean_token_accuracy": 0.2566149398684502,
      "num_tokens": 124663975.0,
      "step": 54365
    },
    {
      "entropy": 5.485727405548095,
      "epoch": 5.222862632084534,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002624997982581334,
      "loss": 4.5669,
      "mean_token_accuracy": 0.2702478677034378,
      "num_tokens": 124675099.0,
      "step": 54370
    },
    {
      "entropy": 5.469759559631347,
      "epoch": 5.223342939481268,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026246556779720436,
      "loss": 4.6271,
      "mean_token_accuracy": 0.2710998013615608,
      "num_tokens": 124687663.0,
      "step": 54375
    },
    {
      "entropy": 5.4362749576568605,
      "epoch": 5.223823246878002,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026243133762728765,
      "loss": 4.6177,
      "mean_token_accuracy": 0.27108197957277297,
      "num_tokens": 124698502.0,
      "step": 54380
    },
    {
      "entropy": 5.525581169128418,
      "epoch": 5.224303554274736,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026239710774917795,
      "loss": 4.6563,
      "mean_token_accuracy": 0.26065611094236374,
      "num_tokens": 124709728.0,
      "step": 54385
    },
    {
      "entropy": 5.56702446937561,
      "epoch": 5.22478386167147,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026236287816367,
      "loss": 4.717,
      "mean_token_accuracy": 0.2574765309691429,
      "num_tokens": 124723329.0,
      "step": 54390
    },
    {
      "entropy": 5.502901554107666,
      "epoch": 5.225264169068204,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002623286488715586,
      "loss": 4.647,
      "mean_token_accuracy": 0.2616195797920227,
      "num_tokens": 124733410.0,
      "step": 54395
    },
    {
      "entropy": 5.57887978553772,
      "epoch": 5.2257444764649374,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002622944198736382,
      "loss": 4.7117,
      "mean_token_accuracy": 0.25656848549842837,
      "num_tokens": 124745030.0,
      "step": 54400
    },
    {
      "entropy": 5.532849454879761,
      "epoch": 5.226224783861672,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026226019117070367,
      "loss": 4.645,
      "mean_token_accuracy": 0.2617356672883034,
      "num_tokens": 124756825.0,
      "step": 54405
    },
    {
      "entropy": 5.576071166992188,
      "epoch": 5.226705091258405,
      "grad_norm": 1.25,
      "learning_rate": 0.00026222596276354965,
      "loss": 4.6585,
      "mean_token_accuracy": 0.26417728662490847,
      "num_tokens": 124768658.0,
      "step": 54410
    },
    {
      "entropy": 5.5132224559783936,
      "epoch": 5.2271853986551395,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026219173465297083,
      "loss": 4.6583,
      "mean_token_accuracy": 0.25720139443874357,
      "num_tokens": 124780058.0,
      "step": 54415
    },
    {
      "entropy": 5.5636475563049315,
      "epoch": 5.227665706051873,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026215750683976195,
      "loss": 4.7779,
      "mean_token_accuracy": 0.25773694813251496,
      "num_tokens": 124791898.0,
      "step": 54420
    },
    {
      "entropy": 5.513664197921753,
      "epoch": 5.228146013448607,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026212327932471777,
      "loss": 4.6234,
      "mean_token_accuracy": 0.26536839455366135,
      "num_tokens": 124803363.0,
      "step": 54425
    },
    {
      "entropy": 5.539771032333374,
      "epoch": 5.228626320845341,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002620890521086326,
      "loss": 4.627,
      "mean_token_accuracy": 0.2648194894194603,
      "num_tokens": 124814625.0,
      "step": 54430
    },
    {
      "entropy": 5.622366952896118,
      "epoch": 5.229106628242075,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002620548251923013,
      "loss": 4.7427,
      "mean_token_accuracy": 0.2591641113162041,
      "num_tokens": 124825934.0,
      "step": 54435
    },
    {
      "entropy": 5.4572266101837155,
      "epoch": 5.2295869356388085,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002620205985765186,
      "loss": 4.5968,
      "mean_token_accuracy": 0.2636733502149582,
      "num_tokens": 124837051.0,
      "step": 54440
    },
    {
      "entropy": 5.534987545013427,
      "epoch": 5.230067243035543,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026198637226207904,
      "loss": 4.7056,
      "mean_token_accuracy": 0.2542790174484253,
      "num_tokens": 124849199.0,
      "step": 54445
    },
    {
      "entropy": 5.5308348655700685,
      "epoch": 5.230547550432276,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026195214624977724,
      "loss": 4.7156,
      "mean_token_accuracy": 0.259436297416687,
      "num_tokens": 124861906.0,
      "step": 54450
    },
    {
      "entropy": 5.50099368095398,
      "epoch": 5.231027857829011,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000261917920540408,
      "loss": 4.6307,
      "mean_token_accuracy": 0.2671395614743233,
      "num_tokens": 124873035.0,
      "step": 54455
    },
    {
      "entropy": 5.6354154586792,
      "epoch": 5.231508165225744,
      "grad_norm": 1.375,
      "learning_rate": 0.0002618836951347657,
      "loss": 4.7403,
      "mean_token_accuracy": 0.25390574932098386,
      "num_tokens": 124884986.0,
      "step": 54460
    },
    {
      "entropy": 5.5933867454528805,
      "epoch": 5.231988472622478,
      "grad_norm": 1.390625,
      "learning_rate": 0.000261849470033645,
      "loss": 4.6363,
      "mean_token_accuracy": 0.26063687205314634,
      "num_tokens": 124895858.0,
      "step": 54465
    },
    {
      "entropy": 5.530251598358154,
      "epoch": 5.232468780019213,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002618152452378406,
      "loss": 4.5959,
      "mean_token_accuracy": 0.2615845263004303,
      "num_tokens": 124906770.0,
      "step": 54470
    },
    {
      "entropy": 5.556694602966308,
      "epoch": 5.232949087415946,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000261781020748147,
      "loss": 4.7169,
      "mean_token_accuracy": 0.26501874476671217,
      "num_tokens": 124918022.0,
      "step": 54475
    },
    {
      "entropy": 5.587048149108886,
      "epoch": 5.23342939481268,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026174679656535896,
      "loss": 4.6749,
      "mean_token_accuracy": 0.2638152062892914,
      "num_tokens": 124929260.0,
      "step": 54480
    },
    {
      "entropy": 5.498611736297607,
      "epoch": 5.233909702209414,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000261712572690271,
      "loss": 4.6163,
      "mean_token_accuracy": 0.25608127415180204,
      "num_tokens": 124941260.0,
      "step": 54485
    },
    {
      "entropy": 5.501444101333618,
      "epoch": 5.234390009606148,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002616783491236775,
      "loss": 4.6378,
      "mean_token_accuracy": 0.26256779581308365,
      "num_tokens": 124951801.0,
      "step": 54490
    },
    {
      "entropy": 5.520589303970337,
      "epoch": 5.234870317002882,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002616441258663732,
      "loss": 4.6726,
      "mean_token_accuracy": 0.2592776983976364,
      "num_tokens": 124964034.0,
      "step": 54495
    },
    {
      "entropy": 5.591238927841187,
      "epoch": 5.235350624399616,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026160990291915264,
      "loss": 4.7478,
      "mean_token_accuracy": 0.25723532140254973,
      "num_tokens": 124975737.0,
      "step": 54500
    },
    {
      "entropy": 5.5502965450286865,
      "epoch": 5.235830931796349,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002615756802828104,
      "loss": 4.6204,
      "mean_token_accuracy": 0.2610964491963387,
      "num_tokens": 124986814.0,
      "step": 54505
    },
    {
      "entropy": 5.569236755371094,
      "epoch": 5.236311239193084,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026154145795814093,
      "loss": 4.6751,
      "mean_token_accuracy": 0.2623548835515976,
      "num_tokens": 124998220.0,
      "step": 54510
    },
    {
      "entropy": 5.583431243896484,
      "epoch": 5.236791546589817,
      "grad_norm": 1.296875,
      "learning_rate": 0.000261507235945939,
      "loss": 4.6252,
      "mean_token_accuracy": 0.2663289487361908,
      "num_tokens": 125010213.0,
      "step": 54515
    },
    {
      "entropy": 5.6103253841400145,
      "epoch": 5.2372718539865515,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002614730142469989,
      "loss": 4.7119,
      "mean_token_accuracy": 0.26201075613498687,
      "num_tokens": 125022490.0,
      "step": 54520
    },
    {
      "entropy": 5.566197204589844,
      "epoch": 5.237752161383285,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026143879286211517,
      "loss": 4.6476,
      "mean_token_accuracy": 0.26790006160736085,
      "num_tokens": 125033416.0,
      "step": 54525
    },
    {
      "entropy": 5.499303817749023,
      "epoch": 5.238232468780019,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002614045717920825,
      "loss": 4.6056,
      "mean_token_accuracy": 0.265033595263958,
      "num_tokens": 125044948.0,
      "step": 54530
    },
    {
      "entropy": 5.56029372215271,
      "epoch": 5.238712776176753,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026137035103769525,
      "loss": 4.6574,
      "mean_token_accuracy": 0.2595566764473915,
      "num_tokens": 125055651.0,
      "step": 54535
    },
    {
      "entropy": 5.536634397506714,
      "epoch": 5.239193083573487,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002613361305997479,
      "loss": 4.6175,
      "mean_token_accuracy": 0.2643937334418297,
      "num_tokens": 125067276.0,
      "step": 54540
    },
    {
      "entropy": 5.593018817901611,
      "epoch": 5.239673390970221,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002613019104790353,
      "loss": 4.7205,
      "mean_token_accuracy": 0.2513465255498886,
      "num_tokens": 125077804.0,
      "step": 54545
    },
    {
      "entropy": 5.58446626663208,
      "epoch": 5.240153698366955,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026126769067635143,
      "loss": 4.7313,
      "mean_token_accuracy": 0.25657123029232026,
      "num_tokens": 125089356.0,
      "step": 54550
    },
    {
      "entropy": 5.476530981063843,
      "epoch": 5.240634005763689,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026123347119249104,
      "loss": 4.5719,
      "mean_token_accuracy": 0.26985283493995665,
      "num_tokens": 125101320.0,
      "step": 54555
    },
    {
      "entropy": 5.502368927001953,
      "epoch": 5.2411143131604225,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002611992520282486,
      "loss": 4.6333,
      "mean_token_accuracy": 0.26234945505857465,
      "num_tokens": 125112063.0,
      "step": 54560
    },
    {
      "entropy": 5.460709047317505,
      "epoch": 5.241594620557157,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002611650331844185,
      "loss": 4.5808,
      "mean_token_accuracy": 0.2680113360285759,
      "num_tokens": 125123064.0,
      "step": 54565
    },
    {
      "entropy": 5.515719747543335,
      "epoch": 5.24207492795389,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002611308146617952,
      "loss": 4.6586,
      "mean_token_accuracy": 0.2673598945140839,
      "num_tokens": 125133966.0,
      "step": 54570
    },
    {
      "entropy": 5.57940354347229,
      "epoch": 5.242555235350625,
      "grad_norm": 1.375,
      "learning_rate": 0.0002610965964611734,
      "loss": 4.6903,
      "mean_token_accuracy": 0.25699369311332704,
      "num_tokens": 125144401.0,
      "step": 54575
    },
    {
      "entropy": 5.505997943878174,
      "epoch": 5.243035542747358,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026106237858334724,
      "loss": 4.684,
      "mean_token_accuracy": 0.26325090527534484,
      "num_tokens": 125156103.0,
      "step": 54580
    },
    {
      "entropy": 5.462887620925903,
      "epoch": 5.243515850144092,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002610281610291113,
      "loss": 4.5711,
      "mean_token_accuracy": 0.27369050979614257,
      "num_tokens": 125167110.0,
      "step": 54585
    },
    {
      "entropy": 5.450797891616821,
      "epoch": 5.243996157540826,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026099394379925983,
      "loss": 4.5679,
      "mean_token_accuracy": 0.27362782061100005,
      "num_tokens": 125177480.0,
      "step": 54590
    },
    {
      "entropy": 5.493690013885498,
      "epoch": 5.24447646493756,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002609597268945875,
      "loss": 4.603,
      "mean_token_accuracy": 0.2698627144098282,
      "num_tokens": 125188597.0,
      "step": 54595
    },
    {
      "entropy": 5.585047578811645,
      "epoch": 5.244956772334294,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002609255103158886,
      "loss": 4.7297,
      "mean_token_accuracy": 0.2537602424621582,
      "num_tokens": 125199811.0,
      "step": 54600
    },
    {
      "entropy": 5.546578073501587,
      "epoch": 5.245437079731028,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002608912940639576,
      "loss": 4.7192,
      "mean_token_accuracy": 0.25901479423046114,
      "num_tokens": 125210791.0,
      "step": 54605
    },
    {
      "entropy": 5.571379470825195,
      "epoch": 5.245917387127761,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002608570781395888,
      "loss": 4.6411,
      "mean_token_accuracy": 0.26229827255010607,
      "num_tokens": 125223973.0,
      "step": 54610
    },
    {
      "entropy": 5.58584942817688,
      "epoch": 5.246397694524496,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002608228625435767,
      "loss": 4.7217,
      "mean_token_accuracy": 0.25617428719997404,
      "num_tokens": 125235549.0,
      "step": 54615
    },
    {
      "entropy": 5.441174840927124,
      "epoch": 5.24687800192123,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026078864727671565,
      "loss": 4.5346,
      "mean_token_accuracy": 0.27407369911670687,
      "num_tokens": 125246020.0,
      "step": 54620
    },
    {
      "entropy": 5.53413577079773,
      "epoch": 5.247358309317963,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002607544323397999,
      "loss": 4.6983,
      "mean_token_accuracy": 0.25926165133714674,
      "num_tokens": 125258098.0,
      "step": 54625
    },
    {
      "entropy": 5.510282421112061,
      "epoch": 5.247838616714698,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026072021773362403,
      "loss": 4.6286,
      "mean_token_accuracy": 0.26129295825958254,
      "num_tokens": 125269894.0,
      "step": 54630
    },
    {
      "entropy": 5.62501277923584,
      "epoch": 5.248318924111431,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002606860034589823,
      "loss": 4.7168,
      "mean_token_accuracy": 0.2543525189161301,
      "num_tokens": 125280588.0,
      "step": 54635
    },
    {
      "entropy": 5.519901609420776,
      "epoch": 5.2487992315081655,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026065178951666904,
      "loss": 4.6501,
      "mean_token_accuracy": 0.2524042218923569,
      "num_tokens": 125292554.0,
      "step": 54640
    },
    {
      "entropy": 5.516995525360107,
      "epoch": 5.249279538904899,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002606175759074786,
      "loss": 4.6655,
      "mean_token_accuracy": 0.2640373080968857,
      "num_tokens": 125303522.0,
      "step": 54645
    },
    {
      "entropy": 5.584539556503296,
      "epoch": 5.249759846301633,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026058336263220535,
      "loss": 4.7347,
      "mean_token_accuracy": 0.25229599326848984,
      "num_tokens": 125315609.0,
      "step": 54650
    },
    {
      "entropy": 5.517589902877807,
      "epoch": 5.250240153698367,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026054914969164363,
      "loss": 4.6578,
      "mean_token_accuracy": 0.25698635429143907,
      "num_tokens": 125327337.0,
      "step": 54655
    },
    {
      "entropy": 5.529704952239991,
      "epoch": 5.250720461095101,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026051493708658774,
      "loss": 4.7211,
      "mean_token_accuracy": 0.2601387411355972,
      "num_tokens": 125339042.0,
      "step": 54660
    },
    {
      "entropy": 5.502070951461792,
      "epoch": 5.2512007684918345,
      "grad_norm": 1.359375,
      "learning_rate": 0.000260480724817832,
      "loss": 4.6485,
      "mean_token_accuracy": 0.26065910905599593,
      "num_tokens": 125349811.0,
      "step": 54665
    },
    {
      "entropy": 5.538998937606811,
      "epoch": 5.251681075888569,
      "grad_norm": 1.25,
      "learning_rate": 0.00026044651288617077,
      "loss": 4.6826,
      "mean_token_accuracy": 0.2586281284689903,
      "num_tokens": 125361265.0,
      "step": 54670
    },
    {
      "entropy": 5.496763849258423,
      "epoch": 5.252161383285302,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026041230129239816,
      "loss": 4.6291,
      "mean_token_accuracy": 0.25477585196495056,
      "num_tokens": 125372235.0,
      "step": 54675
    },
    {
      "entropy": 5.564211750030518,
      "epoch": 5.252641690682037,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026037809003730874,
      "loss": 4.7693,
      "mean_token_accuracy": 0.25299178510904313,
      "num_tokens": 125384659.0,
      "step": 54680
    },
    {
      "entropy": 5.52730016708374,
      "epoch": 5.25312199807877,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026034387912169657,
      "loss": 4.5929,
      "mean_token_accuracy": 0.2653956338763237,
      "num_tokens": 125395737.0,
      "step": 54685
    },
    {
      "entropy": 5.479855251312256,
      "epoch": 5.253602305475504,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026030966854635593,
      "loss": 4.555,
      "mean_token_accuracy": 0.2696737825870514,
      "num_tokens": 125406466.0,
      "step": 54690
    },
    {
      "entropy": 5.529741525650024,
      "epoch": 5.254082612872239,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002602754583120814,
      "loss": 4.6933,
      "mean_token_accuracy": 0.2570851922035217,
      "num_tokens": 125418308.0,
      "step": 54695
    },
    {
      "entropy": 5.521808052062989,
      "epoch": 5.254562920268972,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026024124841966687,
      "loss": 4.6333,
      "mean_token_accuracy": 0.2625695884227753,
      "num_tokens": 125430265.0,
      "step": 54700
    },
    {
      "entropy": 5.486533451080322,
      "epoch": 5.255043227665706,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002602070388699067,
      "loss": 4.5999,
      "mean_token_accuracy": 0.26587690114974977,
      "num_tokens": 125441585.0,
      "step": 54705
    },
    {
      "entropy": 5.569755887985229,
      "epoch": 5.25552353506244,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002601728296635951,
      "loss": 4.6778,
      "mean_token_accuracy": 0.2562676563858986,
      "num_tokens": 125451733.0,
      "step": 54710
    },
    {
      "entropy": 5.563566827774048,
      "epoch": 5.256003842459174,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002601386208015265,
      "loss": 4.7369,
      "mean_token_accuracy": 0.251931531727314,
      "num_tokens": 125462765.0,
      "step": 54715
    },
    {
      "entropy": 5.577304840087891,
      "epoch": 5.256484149855908,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026010441228449494,
      "loss": 4.6432,
      "mean_token_accuracy": 0.2671768993139267,
      "num_tokens": 125474511.0,
      "step": 54720
    },
    {
      "entropy": 5.501955127716064,
      "epoch": 5.256964457252642,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026007020411329473,
      "loss": 4.6337,
      "mean_token_accuracy": 0.2609743535518646,
      "num_tokens": 125485300.0,
      "step": 54725
    },
    {
      "entropy": 5.499500370025634,
      "epoch": 5.257444764649375,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026003599628872,
      "loss": 4.5547,
      "mean_token_accuracy": 0.27032450437545774,
      "num_tokens": 125497097.0,
      "step": 54730
    },
    {
      "entropy": 5.504213476181031,
      "epoch": 5.25792507204611,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026000178881156505,
      "loss": 4.6407,
      "mean_token_accuracy": 0.26705477833747865,
      "num_tokens": 125509708.0,
      "step": 54735
    },
    {
      "entropy": 5.454208660125732,
      "epoch": 5.258405379442843,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025996758168262394,
      "loss": 4.5804,
      "mean_token_accuracy": 0.2685962438583374,
      "num_tokens": 125521393.0,
      "step": 54740
    },
    {
      "entropy": 5.559672117233276,
      "epoch": 5.2588856868395775,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025993337490269104,
      "loss": 4.7396,
      "mean_token_accuracy": 0.2503510877490044,
      "num_tokens": 125532724.0,
      "step": 54745
    },
    {
      "entropy": 5.562157487869262,
      "epoch": 5.259365994236311,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002598991684725604,
      "loss": 4.6857,
      "mean_token_accuracy": 0.25991769433021544,
      "num_tokens": 125545082.0,
      "step": 54750
    },
    {
      "entropy": 5.544813871383667,
      "epoch": 5.259846301633045,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025986496239302633,
      "loss": 4.6577,
      "mean_token_accuracy": 0.26186855882406235,
      "num_tokens": 125556505.0,
      "step": 54755
    },
    {
      "entropy": 5.5092381000518795,
      "epoch": 5.260326609029779,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002598307566648828,
      "loss": 4.5444,
      "mean_token_accuracy": 0.272791762650013,
      "num_tokens": 125568639.0,
      "step": 54760
    },
    {
      "entropy": 5.61170711517334,
      "epoch": 5.260806916426513,
      "grad_norm": 1.46875,
      "learning_rate": 0.000259796551288924,
      "loss": 4.6994,
      "mean_token_accuracy": 0.25958520323038103,
      "num_tokens": 125579599.0,
      "step": 54765
    },
    {
      "entropy": 5.546229457855224,
      "epoch": 5.261287223823247,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025976234626594416,
      "loss": 4.7347,
      "mean_token_accuracy": 0.25414823889732363,
      "num_tokens": 125592548.0,
      "step": 54770
    },
    {
      "entropy": 5.525974893569947,
      "epoch": 5.261767531219981,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002597281415967374,
      "loss": 4.6308,
      "mean_token_accuracy": 0.25714067220687864,
      "num_tokens": 125602983.0,
      "step": 54775
    },
    {
      "entropy": 5.502378511428833,
      "epoch": 5.262247838616715,
      "grad_norm": 1.25,
      "learning_rate": 0.0002596939372820978,
      "loss": 4.6272,
      "mean_token_accuracy": 0.2629909157752991,
      "num_tokens": 125613225.0,
      "step": 54780
    },
    {
      "entropy": 5.611821222305298,
      "epoch": 5.2627281460134485,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002596597333228197,
      "loss": 4.7759,
      "mean_token_accuracy": 0.25233552902936934,
      "num_tokens": 125625067.0,
      "step": 54785
    },
    {
      "entropy": 5.524601078033447,
      "epoch": 5.263208453410183,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025962552971969696,
      "loss": 4.625,
      "mean_token_accuracy": 0.26304842084646224,
      "num_tokens": 125636518.0,
      "step": 54790
    },
    {
      "entropy": 5.508805751800537,
      "epoch": 5.263688760806916,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025959132647352366,
      "loss": 4.6426,
      "mean_token_accuracy": 0.2577420845627785,
      "num_tokens": 125647862.0,
      "step": 54795
    },
    {
      "entropy": 5.461550760269165,
      "epoch": 5.264169068203651,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002595571235850941,
      "loss": 4.6142,
      "mean_token_accuracy": 0.26440279483795165,
      "num_tokens": 125659767.0,
      "step": 54800
    },
    {
      "entropy": 5.576639604568482,
      "epoch": 5.264649375600384,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025952292105520225,
      "loss": 4.7165,
      "mean_token_accuracy": 0.2576194405555725,
      "num_tokens": 125672420.0,
      "step": 54805
    },
    {
      "entropy": 5.588123750686646,
      "epoch": 5.265129682997118,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002594887188846422,
      "loss": 4.703,
      "mean_token_accuracy": 0.2550281584262848,
      "num_tokens": 125682775.0,
      "step": 54810
    },
    {
      "entropy": 5.526512527465821,
      "epoch": 5.265609990393852,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025945451707420816,
      "loss": 4.6569,
      "mean_token_accuracy": 0.26226186454296113,
      "num_tokens": 125694765.0,
      "step": 54815
    },
    {
      "entropy": 5.539035654067993,
      "epoch": 5.266090297790586,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002594203156246939,
      "loss": 4.7804,
      "mean_token_accuracy": 0.2527936607599258,
      "num_tokens": 125706181.0,
      "step": 54820
    },
    {
      "entropy": 5.616540098190308,
      "epoch": 5.26657060518732,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025938611453689376,
      "loss": 4.7779,
      "mean_token_accuracy": 0.24690356850624084,
      "num_tokens": 125718338.0,
      "step": 54825
    },
    {
      "entropy": 5.547737073898316,
      "epoch": 5.267050912584054,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002593519138116016,
      "loss": 4.6275,
      "mean_token_accuracy": 0.2667463839054108,
      "num_tokens": 125730067.0,
      "step": 54830
    },
    {
      "entropy": 5.602959299087525,
      "epoch": 5.267531219980787,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002593177134496116,
      "loss": 4.7639,
      "mean_token_accuracy": 0.25290298610925677,
      "num_tokens": 125741701.0,
      "step": 54835
    },
    {
      "entropy": 5.514694309234619,
      "epoch": 5.268011527377522,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025928351345171763,
      "loss": 4.6098,
      "mean_token_accuracy": 0.2621592625975609,
      "num_tokens": 125752743.0,
      "step": 54840
    },
    {
      "entropy": 5.583968591690064,
      "epoch": 5.268491834774256,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025924931381871404,
      "loss": 4.741,
      "mean_token_accuracy": 0.25440294593572615,
      "num_tokens": 125764588.0,
      "step": 54845
    },
    {
      "entropy": 5.605746603012085,
      "epoch": 5.268972142170989,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002592151145513944,
      "loss": 4.752,
      "mean_token_accuracy": 0.25741364508867265,
      "num_tokens": 125776958.0,
      "step": 54850
    },
    {
      "entropy": 5.553819274902343,
      "epoch": 5.269452449567724,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025918091565055296,
      "loss": 4.7212,
      "mean_token_accuracy": 0.2535534903407097,
      "num_tokens": 125788859.0,
      "step": 54855
    },
    {
      "entropy": 5.499166202545166,
      "epoch": 5.269932756964457,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025914671711698366,
      "loss": 4.6489,
      "mean_token_accuracy": 0.26311215609312055,
      "num_tokens": 125801075.0,
      "step": 54860
    },
    {
      "entropy": 5.534758281707764,
      "epoch": 5.2704130643611915,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025911251895148057,
      "loss": 4.6955,
      "mean_token_accuracy": 0.25671363472938535,
      "num_tokens": 125812639.0,
      "step": 54865
    },
    {
      "entropy": 5.5101580142974855,
      "epoch": 5.270893371757925,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002590783211548376,
      "loss": 4.5795,
      "mean_token_accuracy": 0.2705597370862961,
      "num_tokens": 125823923.0,
      "step": 54870
    },
    {
      "entropy": 5.48015570640564,
      "epoch": 5.271373679154659,
      "grad_norm": 1.25,
      "learning_rate": 0.0002590441237278488,
      "loss": 4.6011,
      "mean_token_accuracy": 0.26548234522342684,
      "num_tokens": 125834471.0,
      "step": 54875
    },
    {
      "entropy": 5.512644100189209,
      "epoch": 5.271853986551393,
      "grad_norm": 1.34375,
      "learning_rate": 0.000259009926671308,
      "loss": 4.7228,
      "mean_token_accuracy": 0.2569010227918625,
      "num_tokens": 125846281.0,
      "step": 54880
    },
    {
      "entropy": 5.647366905212403,
      "epoch": 5.272334293948127,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002589757299860092,
      "loss": 4.7865,
      "mean_token_accuracy": 0.25097724944353106,
      "num_tokens": 125858902.0,
      "step": 54885
    },
    {
      "entropy": 5.626152753829956,
      "epoch": 5.2728146013448605,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002589415336727464,
      "loss": 4.7387,
      "mean_token_accuracy": 0.258055080473423,
      "num_tokens": 125869778.0,
      "step": 54890
    },
    {
      "entropy": 5.558892869949341,
      "epoch": 5.273294908741595,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025890733773231353,
      "loss": 4.7163,
      "mean_token_accuracy": 0.25628057271242144,
      "num_tokens": 125880642.0,
      "step": 54895
    },
    {
      "entropy": 5.503886938095093,
      "epoch": 5.273775216138328,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002588731421655044,
      "loss": 4.6352,
      "mean_token_accuracy": 0.26197507083415983,
      "num_tokens": 125893363.0,
      "step": 54900
    },
    {
      "entropy": 5.545036125183105,
      "epoch": 5.2742555235350626,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025883894697311316,
      "loss": 4.6382,
      "mean_token_accuracy": 0.2585666134953499,
      "num_tokens": 125904927.0,
      "step": 54905
    },
    {
      "entropy": 5.575498151779175,
      "epoch": 5.274735830931796,
      "grad_norm": 1.5,
      "learning_rate": 0.00025880475215593347,
      "loss": 4.8058,
      "mean_token_accuracy": 0.24697300642728806,
      "num_tokens": 125915816.0,
      "step": 54910
    },
    {
      "entropy": 5.474265718460083,
      "epoch": 5.27521613832853,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002587705577147594,
      "loss": 4.6204,
      "mean_token_accuracy": 0.26376415342092513,
      "num_tokens": 125927980.0,
      "step": 54915
    },
    {
      "entropy": 5.594429922103882,
      "epoch": 5.275696445725265,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002587363636503848,
      "loss": 4.7129,
      "mean_token_accuracy": 0.251473543047905,
      "num_tokens": 125938899.0,
      "step": 54920
    },
    {
      "entropy": 5.466967582702637,
      "epoch": 5.276176753121998,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002587021699636035,
      "loss": 4.5339,
      "mean_token_accuracy": 0.27096815705299376,
      "num_tokens": 125950358.0,
      "step": 54925
    },
    {
      "entropy": 5.529627990722656,
      "epoch": 5.276657060518732,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002586679766552094,
      "loss": 4.7065,
      "mean_token_accuracy": 0.25522648692131045,
      "num_tokens": 125961344.0,
      "step": 54930
    },
    {
      "entropy": 5.633451318740844,
      "epoch": 5.277137367915466,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002586337837259965,
      "loss": 4.7527,
      "mean_token_accuracy": 0.2515799254179001,
      "num_tokens": 125972975.0,
      "step": 54935
    },
    {
      "entropy": 5.543687725067139,
      "epoch": 5.2776176753122,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002585995911767584,
      "loss": 4.6341,
      "mean_token_accuracy": 0.26119251996278764,
      "num_tokens": 125984836.0,
      "step": 54940
    },
    {
      "entropy": 5.703379201889038,
      "epoch": 5.278097982708934,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002585653990082892,
      "loss": 4.8941,
      "mean_token_accuracy": 0.23891335874795913,
      "num_tokens": 125998800.0,
      "step": 54945
    },
    {
      "entropy": 5.606506586074829,
      "epoch": 5.278578290105668,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002585312072213825,
      "loss": 4.7261,
      "mean_token_accuracy": 0.25095461010932923,
      "num_tokens": 126010219.0,
      "step": 54950
    },
    {
      "entropy": 5.572795295715332,
      "epoch": 5.279058597502401,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002584970158168324,
      "loss": 4.7023,
      "mean_token_accuracy": 0.2600542798638344,
      "num_tokens": 126020953.0,
      "step": 54955
    },
    {
      "entropy": 5.602072095870971,
      "epoch": 5.279538904899136,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002584628247954325,
      "loss": 4.7384,
      "mean_token_accuracy": 0.2527218908071518,
      "num_tokens": 126032626.0,
      "step": 54960
    },
    {
      "entropy": 5.4979606628417965,
      "epoch": 5.280019212295869,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025842863415797684,
      "loss": 4.6195,
      "mean_token_accuracy": 0.25473962277173995,
      "num_tokens": 126044028.0,
      "step": 54965
    },
    {
      "entropy": 5.51640362739563,
      "epoch": 5.2804995196926034,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000258394443905259,
      "loss": 4.6278,
      "mean_token_accuracy": 0.25769226998090744,
      "num_tokens": 126054513.0,
      "step": 54970
    },
    {
      "entropy": 5.527943325042725,
      "epoch": 5.280979827089337,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002583602540380728,
      "loss": 4.6367,
      "mean_token_accuracy": 0.2597272485494614,
      "num_tokens": 126067367.0,
      "step": 54975
    },
    {
      "entropy": 5.575689458847046,
      "epoch": 5.281460134486071,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002583260645572122,
      "loss": 4.7165,
      "mean_token_accuracy": 0.253537580370903,
      "num_tokens": 126079078.0,
      "step": 54980
    },
    {
      "entropy": 5.593950700759888,
      "epoch": 5.281940441882805,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002582918754634708,
      "loss": 4.6698,
      "mean_token_accuracy": 0.25793920159339906,
      "num_tokens": 126090393.0,
      "step": 54985
    },
    {
      "entropy": 5.637573289871216,
      "epoch": 5.282420749279539,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002582576867576424,
      "loss": 4.7168,
      "mean_token_accuracy": 0.2597266063094139,
      "num_tokens": 126102899.0,
      "step": 54990
    },
    {
      "entropy": 5.499338150024414,
      "epoch": 5.282901056676272,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025822349844052103,
      "loss": 4.6492,
      "mean_token_accuracy": 0.26135557889938354,
      "num_tokens": 126115196.0,
      "step": 54995
    },
    {
      "entropy": 5.515150880813598,
      "epoch": 5.283381364073007,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025818931051290004,
      "loss": 4.6682,
      "mean_token_accuracy": 0.2577248141169548,
      "num_tokens": 126125284.0,
      "step": 55000
    },
    {
      "entropy": 5.531428623199463,
      "epoch": 5.28386167146974,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002581551229755733,
      "loss": 4.6503,
      "mean_token_accuracy": 0.2589609906077385,
      "num_tokens": 126136502.0,
      "step": 55005
    },
    {
      "entropy": 5.481320524215699,
      "epoch": 5.2843419788664745,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025812093582933463,
      "loss": 4.675,
      "mean_token_accuracy": 0.25746806114912035,
      "num_tokens": 126148043.0,
      "step": 55010
    },
    {
      "entropy": 5.554512214660645,
      "epoch": 5.284822286263209,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025808674907497765,
      "loss": 4.6985,
      "mean_token_accuracy": 0.25524491220712664,
      "num_tokens": 126158636.0,
      "step": 55015
    },
    {
      "entropy": 5.576088762283325,
      "epoch": 5.285302593659942,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002580525627132962,
      "loss": 4.7048,
      "mean_token_accuracy": 0.2545991212129593,
      "num_tokens": 126169588.0,
      "step": 55020
    },
    {
      "entropy": 5.675016641616821,
      "epoch": 5.285782901056677,
      "grad_norm": 1.296875,
      "learning_rate": 0.000258018376745084,
      "loss": 4.7604,
      "mean_token_accuracy": 0.25504216402769087,
      "num_tokens": 126180935.0,
      "step": 55025
    },
    {
      "entropy": 5.499374485015869,
      "epoch": 5.28626320845341,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002579841911711346,
      "loss": 4.5668,
      "mean_token_accuracy": 0.26688591092824937,
      "num_tokens": 126192601.0,
      "step": 55030
    },
    {
      "entropy": 5.466223049163818,
      "epoch": 5.286743515850144,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025795000599224165,
      "loss": 4.6243,
      "mean_token_accuracy": 0.26266719251871107,
      "num_tokens": 126204961.0,
      "step": 55035
    },
    {
      "entropy": 5.534532165527343,
      "epoch": 5.287223823246878,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025791582120919906,
      "loss": 4.6489,
      "mean_token_accuracy": 0.2596264570951462,
      "num_tokens": 126216233.0,
      "step": 55040
    },
    {
      "entropy": 5.499772834777832,
      "epoch": 5.287704130643612,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002578816368228003,
      "loss": 4.5609,
      "mean_token_accuracy": 0.2651773884892464,
      "num_tokens": 126226291.0,
      "step": 55045
    },
    {
      "entropy": 5.564332008361816,
      "epoch": 5.2881844380403455,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025784745283383907,
      "loss": 4.6547,
      "mean_token_accuracy": 0.2659172579646111,
      "num_tokens": 126237802.0,
      "step": 55050
    },
    {
      "entropy": 5.542859983444214,
      "epoch": 5.28866474543708,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002578132692431092,
      "loss": 4.6898,
      "mean_token_accuracy": 0.2542552173137665,
      "num_tokens": 126248332.0,
      "step": 55055
    },
    {
      "entropy": 5.507566833496094,
      "epoch": 5.289145052833813,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002577790860514041,
      "loss": 4.7203,
      "mean_token_accuracy": 0.25619492381811143,
      "num_tokens": 126260303.0,
      "step": 55060
    },
    {
      "entropy": 5.431608009338379,
      "epoch": 5.289625360230548,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002577449032595175,
      "loss": 4.579,
      "mean_token_accuracy": 0.27020958364009856,
      "num_tokens": 126271494.0,
      "step": 55065
    },
    {
      "entropy": 5.5183165073394775,
      "epoch": 5.290105667627281,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000257710720868243,
      "loss": 4.6359,
      "mean_token_accuracy": 0.2584421783685684,
      "num_tokens": 126283236.0,
      "step": 55070
    },
    {
      "entropy": 5.5172045707702635,
      "epoch": 5.290585975024015,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002576765388783742,
      "loss": 4.6957,
      "mean_token_accuracy": 0.26251471787691116,
      "num_tokens": 126294388.0,
      "step": 55075
    },
    {
      "entropy": 5.591417980194092,
      "epoch": 5.291066282420749,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025764235729070466,
      "loss": 4.763,
      "mean_token_accuracy": 0.2601486638188362,
      "num_tokens": 126306142.0,
      "step": 55080
    },
    {
      "entropy": 5.558472537994385,
      "epoch": 5.291546589817483,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025760817610602824,
      "loss": 4.6993,
      "mean_token_accuracy": 0.25604364573955535,
      "num_tokens": 126318078.0,
      "step": 55085
    },
    {
      "entropy": 5.524991989135742,
      "epoch": 5.2920268972142175,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002575739953251382,
      "loss": 4.5883,
      "mean_token_accuracy": 0.26104727387428284,
      "num_tokens": 126328176.0,
      "step": 55090
    },
    {
      "entropy": 5.515807247161865,
      "epoch": 5.292507204610951,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002575398149488283,
      "loss": 4.5985,
      "mean_token_accuracy": 0.26832794696092604,
      "num_tokens": 126337639.0,
      "step": 55095
    },
    {
      "entropy": 5.5726817607879635,
      "epoch": 5.292987512007685,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025750563497789193,
      "loss": 4.6838,
      "mean_token_accuracy": 0.25376760512590407,
      "num_tokens": 126348205.0,
      "step": 55100
    },
    {
      "entropy": 5.540373086929321,
      "epoch": 5.293467819404419,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025747145541312285,
      "loss": 4.6944,
      "mean_token_accuracy": 0.2578039586544037,
      "num_tokens": 126359212.0,
      "step": 55105
    },
    {
      "entropy": 5.514739036560059,
      "epoch": 5.293948126801153,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002574372762553145,
      "loss": 4.6166,
      "mean_token_accuracy": 0.26295205056667326,
      "num_tokens": 126370153.0,
      "step": 55110
    },
    {
      "entropy": 5.488339757919311,
      "epoch": 5.2944284341978864,
      "grad_norm": 1.375,
      "learning_rate": 0.00025740309750526055,
      "loss": 4.648,
      "mean_token_accuracy": 0.26011880934238435,
      "num_tokens": 126381769.0,
      "step": 55115
    },
    {
      "entropy": 5.5282364845275875,
      "epoch": 5.294908741594621,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025736891916375434,
      "loss": 4.6378,
      "mean_token_accuracy": 0.2643355131149292,
      "num_tokens": 126393276.0,
      "step": 55120
    },
    {
      "entropy": 5.555696296691894,
      "epoch": 5.295389048991354,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025733474123158947,
      "loss": 4.672,
      "mean_token_accuracy": 0.2614723280072212,
      "num_tokens": 126405327.0,
      "step": 55125
    },
    {
      "entropy": 5.602077436447144,
      "epoch": 5.2958693563880885,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025730056370955947,
      "loss": 4.7181,
      "mean_token_accuracy": 0.2489883705973625,
      "num_tokens": 126416999.0,
      "step": 55130
    },
    {
      "entropy": 5.5248411178588865,
      "epoch": 5.296349663784822,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025726638659845775,
      "loss": 4.6124,
      "mean_token_accuracy": 0.26868176460266113,
      "num_tokens": 126428639.0,
      "step": 55135
    },
    {
      "entropy": 5.492456912994385,
      "epoch": 5.296829971181556,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002572322098990779,
      "loss": 4.638,
      "mean_token_accuracy": 0.26261280179023744,
      "num_tokens": 126440158.0,
      "step": 55140
    },
    {
      "entropy": 5.548699522018433,
      "epoch": 5.29731027857829,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002571980336122135,
      "loss": 4.621,
      "mean_token_accuracy": 0.2600473716855049,
      "num_tokens": 126451239.0,
      "step": 55145
    },
    {
      "entropy": 5.496588897705078,
      "epoch": 5.297790585975024,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002571638577386578,
      "loss": 4.6042,
      "mean_token_accuracy": 0.2580656141042709,
      "num_tokens": 126464503.0,
      "step": 55150
    },
    {
      "entropy": 5.5624330043792725,
      "epoch": 5.2982708933717575,
      "grad_norm": 1.375,
      "learning_rate": 0.0002571296822792044,
      "loss": 4.7431,
      "mean_token_accuracy": 0.2583991914987564,
      "num_tokens": 126475607.0,
      "step": 55155
    },
    {
      "entropy": 5.501075649261475,
      "epoch": 5.298751200768492,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025709550723464664,
      "loss": 4.5992,
      "mean_token_accuracy": 0.2653396472334862,
      "num_tokens": 126487642.0,
      "step": 55160
    },
    {
      "entropy": 5.508317756652832,
      "epoch": 5.299231508165226,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002570613326057781,
      "loss": 4.6209,
      "mean_token_accuracy": 0.26735909283161163,
      "num_tokens": 126498556.0,
      "step": 55165
    },
    {
      "entropy": 5.590753507614136,
      "epoch": 5.29971181556196,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025702715839339216,
      "loss": 4.6759,
      "mean_token_accuracy": 0.2608557865023613,
      "num_tokens": 126509069.0,
      "step": 55170
    },
    {
      "entropy": 5.550948667526245,
      "epoch": 5.300192122958694,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002569929845982822,
      "loss": 4.7076,
      "mean_token_accuracy": 0.256986066699028,
      "num_tokens": 126520453.0,
      "step": 55175
    },
    {
      "entropy": 5.578773736953735,
      "epoch": 5.300672430355427,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025695881122124166,
      "loss": 4.6689,
      "mean_token_accuracy": 0.26047507524490354,
      "num_tokens": 126531831.0,
      "step": 55180
    },
    {
      "entropy": 5.450032806396484,
      "epoch": 5.301152737752162,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025692463826306394,
      "loss": 4.5784,
      "mean_token_accuracy": 0.27275600284338,
      "num_tokens": 126543312.0,
      "step": 55185
    },
    {
      "entropy": 5.560554265975952,
      "epoch": 5.301633045148895,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002568904657245425,
      "loss": 4.7374,
      "mean_token_accuracy": 0.25896051079034804,
      "num_tokens": 126555513.0,
      "step": 55190
    },
    {
      "entropy": 5.532449054718017,
      "epoch": 5.302113352545629,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025685629360647065,
      "loss": 4.6864,
      "mean_token_accuracy": 0.26452773809432983,
      "num_tokens": 126566497.0,
      "step": 55195
    },
    {
      "entropy": 5.538296222686768,
      "epoch": 5.302593659942363,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025682212190964175,
      "loss": 4.699,
      "mean_token_accuracy": 0.25442043840885165,
      "num_tokens": 126578879.0,
      "step": 55200
    },
    {
      "entropy": 5.527514123916626,
      "epoch": 5.303073967339097,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002567879506348493,
      "loss": 4.6309,
      "mean_token_accuracy": 0.26010398417711256,
      "num_tokens": 126590971.0,
      "step": 55205
    },
    {
      "entropy": 5.673728799819946,
      "epoch": 5.303554274735831,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025675377978288646,
      "loss": 4.81,
      "mean_token_accuracy": 0.2507008582353592,
      "num_tokens": 126603298.0,
      "step": 55210
    },
    {
      "entropy": 5.507661628723144,
      "epoch": 5.304034582132565,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025671960935454667,
      "loss": 4.6235,
      "mean_token_accuracy": 0.2693760484457016,
      "num_tokens": 126615216.0,
      "step": 55215
    },
    {
      "entropy": 5.590365600585938,
      "epoch": 5.304514889529298,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002566854393506233,
      "loss": 4.6409,
      "mean_token_accuracy": 0.2660472646355629,
      "num_tokens": 126626719.0,
      "step": 55220
    },
    {
      "entropy": 5.545523262023925,
      "epoch": 5.304995196926033,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025665126977190967,
      "loss": 4.6971,
      "mean_token_accuracy": 0.25283497869968413,
      "num_tokens": 126639258.0,
      "step": 55225
    },
    {
      "entropy": 5.494320440292358,
      "epoch": 5.305475504322766,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000256617100619199,
      "loss": 4.6909,
      "mean_token_accuracy": 0.25970378816127776,
      "num_tokens": 126650373.0,
      "step": 55230
    },
    {
      "entropy": 5.513554620742798,
      "epoch": 5.3059558117195005,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002565829318932848,
      "loss": 4.7306,
      "mean_token_accuracy": 0.25303491950035095,
      "num_tokens": 126661695.0,
      "step": 55235
    },
    {
      "entropy": 5.461932134628296,
      "epoch": 5.306436119116235,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025654876359496015,
      "loss": 4.6069,
      "mean_token_accuracy": 0.2649464800953865,
      "num_tokens": 126674351.0,
      "step": 55240
    },
    {
      "entropy": 5.507643985748291,
      "epoch": 5.306916426512968,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002565145957250185,
      "loss": 4.6034,
      "mean_token_accuracy": 0.26212728172540667,
      "num_tokens": 126686305.0,
      "step": 55245
    },
    {
      "entropy": 5.613053703308106,
      "epoch": 5.307396733909703,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002564804282842529,
      "loss": 4.7564,
      "mean_token_accuracy": 0.25575061738491056,
      "num_tokens": 126697051.0,
      "step": 55250
    },
    {
      "entropy": 5.4637798309326175,
      "epoch": 5.307877041306436,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002564462612734568,
      "loss": 4.5572,
      "mean_token_accuracy": 0.27127224802970884,
      "num_tokens": 126708925.0,
      "step": 55255
    },
    {
      "entropy": 5.531184530258178,
      "epoch": 5.30835734870317,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002564120946934235,
      "loss": 4.7046,
      "mean_token_accuracy": 0.258019357919693,
      "num_tokens": 126720886.0,
      "step": 55260
    },
    {
      "entropy": 5.589685249328613,
      "epoch": 5.308837656099904,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025637792854494613,
      "loss": 4.6771,
      "mean_token_accuracy": 0.26434046775102615,
      "num_tokens": 126732592.0,
      "step": 55265
    },
    {
      "entropy": 5.568224763870239,
      "epoch": 5.309317963496638,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025634376282881793,
      "loss": 4.6675,
      "mean_token_accuracy": 0.2582146093249321,
      "num_tokens": 126744064.0,
      "step": 55270
    },
    {
      "entropy": 5.521299552917481,
      "epoch": 5.3097982708933715,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002563095975458322,
      "loss": 4.6209,
      "mean_token_accuracy": 0.25388316959142687,
      "num_tokens": 126755310.0,
      "step": 55275
    },
    {
      "entropy": 5.559928369522095,
      "epoch": 5.310278578290106,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025627543269678203,
      "loss": 4.6907,
      "mean_token_accuracy": 0.26000425666570665,
      "num_tokens": 126766165.0,
      "step": 55280
    },
    {
      "entropy": 5.536421585083008,
      "epoch": 5.310758885686839,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002562412682824608,
      "loss": 4.6698,
      "mean_token_accuracy": 0.2566259950399399,
      "num_tokens": 126778048.0,
      "step": 55285
    },
    {
      "entropy": 5.561265182495117,
      "epoch": 5.311239193083574,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002562071043036616,
      "loss": 4.718,
      "mean_token_accuracy": 0.2562347322702408,
      "num_tokens": 126788590.0,
      "step": 55290
    },
    {
      "entropy": 5.463772487640381,
      "epoch": 5.311719500480307,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025617294076117777,
      "loss": 4.6102,
      "mean_token_accuracy": 0.2658187061548233,
      "num_tokens": 126799984.0,
      "step": 55295
    },
    {
      "entropy": 5.485464811325073,
      "epoch": 5.312199807877041,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002561387776558021,
      "loss": 4.5987,
      "mean_token_accuracy": 0.2616944655776024,
      "num_tokens": 126810367.0,
      "step": 55300
    },
    {
      "entropy": 5.592571496963501,
      "epoch": 5.312680115273775,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025610461498832817,
      "loss": 4.7453,
      "mean_token_accuracy": 0.2551272436976433,
      "num_tokens": 126823287.0,
      "step": 55305
    },
    {
      "entropy": 5.516830110549927,
      "epoch": 5.313160422670509,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002560704527595489,
      "loss": 4.5892,
      "mean_token_accuracy": 0.2657239243388176,
      "num_tokens": 126833864.0,
      "step": 55310
    },
    {
      "entropy": 5.5713389873504635,
      "epoch": 5.3136407300672435,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002560362909702575,
      "loss": 4.6863,
      "mean_token_accuracy": 0.26107514798641207,
      "num_tokens": 126847237.0,
      "step": 55315
    },
    {
      "entropy": 5.531566953659057,
      "epoch": 5.314121037463977,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025600212962124713,
      "loss": 4.7299,
      "mean_token_accuracy": 0.2541442781686783,
      "num_tokens": 126858325.0,
      "step": 55320
    },
    {
      "entropy": 5.504649257659912,
      "epoch": 5.314601344860711,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000255967968713311,
      "loss": 4.6625,
      "mean_token_accuracy": 0.260300974547863,
      "num_tokens": 126870948.0,
      "step": 55325
    },
    {
      "entropy": 5.574971389770508,
      "epoch": 5.315081652257445,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025593380824724205,
      "loss": 4.6421,
      "mean_token_accuracy": 0.2578972727060318,
      "num_tokens": 126882079.0,
      "step": 55330
    },
    {
      "entropy": 5.672845792770386,
      "epoch": 5.315561959654179,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002558996482238334,
      "loss": 4.7508,
      "mean_token_accuracy": 0.2630935937166214,
      "num_tokens": 126892397.0,
      "step": 55335
    },
    {
      "entropy": 5.567334318161011,
      "epoch": 5.316042267050912,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025586548864387827,
      "loss": 4.6765,
      "mean_token_accuracy": 0.2564263328909874,
      "num_tokens": 126904417.0,
      "step": 55340
    },
    {
      "entropy": 5.504290151596069,
      "epoch": 5.316522574447647,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002558313295081696,
      "loss": 4.6275,
      "mean_token_accuracy": 0.26796463429927825,
      "num_tokens": 126915564.0,
      "step": 55345
    },
    {
      "entropy": 5.4395716190338135,
      "epoch": 5.31700288184438,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002557971708175006,
      "loss": 4.6239,
      "mean_token_accuracy": 0.2684971377253532,
      "num_tokens": 126926663.0,
      "step": 55350
    },
    {
      "entropy": 5.557532691955567,
      "epoch": 5.3174831892411145,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002557630125726643,
      "loss": 4.6749,
      "mean_token_accuracy": 0.25652710646390914,
      "num_tokens": 126937853.0,
      "step": 55355
    },
    {
      "entropy": 5.56812891960144,
      "epoch": 5.317963496637848,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002557288547744537,
      "loss": 4.7577,
      "mean_token_accuracy": 0.25734245777130127,
      "num_tokens": 126948296.0,
      "step": 55360
    },
    {
      "entropy": 5.485380935668945,
      "epoch": 5.318443804034582,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025569469742366184,
      "loss": 4.5572,
      "mean_token_accuracy": 0.2659107238054276,
      "num_tokens": 126959903.0,
      "step": 55365
    },
    {
      "entropy": 5.5652756690979,
      "epoch": 5.318924111431316,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002556605405210818,
      "loss": 4.7264,
      "mean_token_accuracy": 0.2526276782155037,
      "num_tokens": 126971912.0,
      "step": 55370
    },
    {
      "entropy": 5.482588148117065,
      "epoch": 5.31940441882805,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025562638406750654,
      "loss": 4.5929,
      "mean_token_accuracy": 0.26915431916713717,
      "num_tokens": 126982372.0,
      "step": 55375
    },
    {
      "entropy": 5.509996175765991,
      "epoch": 5.3198847262247835,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025559222806372916,
      "loss": 4.6582,
      "mean_token_accuracy": 0.2580577194690704,
      "num_tokens": 126994097.0,
      "step": 55380
    },
    {
      "entropy": 5.451895713806152,
      "epoch": 5.320365033621518,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002555580725105427,
      "loss": 4.6022,
      "mean_token_accuracy": 0.26729162633419035,
      "num_tokens": 127005527.0,
      "step": 55385
    },
    {
      "entropy": 5.538306331634521,
      "epoch": 5.320845341018252,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025552391740873997,
      "loss": 4.662,
      "mean_token_accuracy": 0.25585713386535647,
      "num_tokens": 127016957.0,
      "step": 55390
    },
    {
      "entropy": 5.529235458374023,
      "epoch": 5.321325648414986,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025548976275911403,
      "loss": 4.663,
      "mean_token_accuracy": 0.2624725133180618,
      "num_tokens": 127026688.0,
      "step": 55395
    },
    {
      "entropy": 5.542287397384643,
      "epoch": 5.32180595581172,
      "grad_norm": 1.375,
      "learning_rate": 0.0002554556085624579,
      "loss": 4.6941,
      "mean_token_accuracy": 0.25551012754440305,
      "num_tokens": 127038178.0,
      "step": 55400
    },
    {
      "entropy": 5.5577466011047365,
      "epoch": 5.322286263208453,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002554214548195645,
      "loss": 4.6823,
      "mean_token_accuracy": 0.2547640159726143,
      "num_tokens": 127049744.0,
      "step": 55405
    },
    {
      "entropy": 5.6074316024780275,
      "epoch": 5.322766570605188,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002553873015312267,
      "loss": 4.7139,
      "mean_token_accuracy": 0.2549991950392723,
      "num_tokens": 127060126.0,
      "step": 55410
    },
    {
      "entropy": 5.557883024215698,
      "epoch": 5.323246878001921,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025535314869823764,
      "loss": 4.7047,
      "mean_token_accuracy": 0.2536875709891319,
      "num_tokens": 127071015.0,
      "step": 55415
    },
    {
      "entropy": 5.5626317977905275,
      "epoch": 5.323727185398655,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002553189963213901,
      "loss": 4.7184,
      "mean_token_accuracy": 0.2548591509461403,
      "num_tokens": 127083054.0,
      "step": 55420
    },
    {
      "entropy": 5.6180188179016115,
      "epoch": 5.324207492795389,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000255284844401477,
      "loss": 4.7424,
      "mean_token_accuracy": 0.2577680990099907,
      "num_tokens": 127094600.0,
      "step": 55425
    },
    {
      "entropy": 5.543489789962768,
      "epoch": 5.324687800192123,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002552506929392913,
      "loss": 4.6614,
      "mean_token_accuracy": 0.26501006036996844,
      "num_tokens": 127107514.0,
      "step": 55430
    },
    {
      "entropy": 5.590196895599365,
      "epoch": 5.325168107588857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025521654193562583,
      "loss": 4.7237,
      "mean_token_accuracy": 0.2564946115016937,
      "num_tokens": 127118246.0,
      "step": 55435
    },
    {
      "entropy": 5.53484787940979,
      "epoch": 5.325648414985591,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025518239139127346,
      "loss": 4.6409,
      "mean_token_accuracy": 0.2663682371377945,
      "num_tokens": 127129188.0,
      "step": 55440
    },
    {
      "entropy": 5.499359464645385,
      "epoch": 5.326128722382324,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002551482413070272,
      "loss": 4.679,
      "mean_token_accuracy": 0.26022323966026306,
      "num_tokens": 127140465.0,
      "step": 55445
    },
    {
      "entropy": 5.519657802581787,
      "epoch": 5.326609029779059,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002551140916836798,
      "loss": 4.69,
      "mean_token_accuracy": 0.2531002938747406,
      "num_tokens": 127152724.0,
      "step": 55450
    },
    {
      "entropy": 5.534122180938721,
      "epoch": 5.327089337175792,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025507994252202415,
      "loss": 4.6604,
      "mean_token_accuracy": 0.2633105367422104,
      "num_tokens": 127163270.0,
      "step": 55455
    },
    {
      "entropy": 5.548373699188232,
      "epoch": 5.3275696445725265,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002550457938228531,
      "loss": 4.6667,
      "mean_token_accuracy": 0.262422077357769,
      "num_tokens": 127174247.0,
      "step": 55460
    },
    {
      "entropy": 5.572892856597901,
      "epoch": 5.328049951969261,
      "grad_norm": 1.375,
      "learning_rate": 0.0002550116455869595,
      "loss": 4.6757,
      "mean_token_accuracy": 0.2582643151283264,
      "num_tokens": 127184689.0,
      "step": 55465
    },
    {
      "entropy": 5.527895736694336,
      "epoch": 5.328530259365994,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025497749781513605,
      "loss": 4.6621,
      "mean_token_accuracy": 0.26737871319055556,
      "num_tokens": 127197374.0,
      "step": 55470
    },
    {
      "entropy": 5.501508092880249,
      "epoch": 5.3290105667627286,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025494335050817575,
      "loss": 4.5612,
      "mean_token_accuracy": 0.2748560279607773,
      "num_tokens": 127208273.0,
      "step": 55475
    },
    {
      "entropy": 5.542698955535888,
      "epoch": 5.329490874159462,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002549092036668712,
      "loss": 4.7079,
      "mean_token_accuracy": 0.26032260060310364,
      "num_tokens": 127218564.0,
      "step": 55480
    },
    {
      "entropy": 5.542709493637085,
      "epoch": 5.329971181556196,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002548750572920153,
      "loss": 4.6669,
      "mean_token_accuracy": 0.2636725053191185,
      "num_tokens": 127230026.0,
      "step": 55485
    },
    {
      "entropy": 5.610084962844849,
      "epoch": 5.33045148895293,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002548409113844008,
      "loss": 4.717,
      "mean_token_accuracy": 0.2604688614606857,
      "num_tokens": 127241343.0,
      "step": 55490
    },
    {
      "entropy": 5.504941654205322,
      "epoch": 5.330931796349664,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002548067659448205,
      "loss": 4.5723,
      "mean_token_accuracy": 0.2709062650799751,
      "num_tokens": 127253695.0,
      "step": 55495
    },
    {
      "entropy": 5.527880573272705,
      "epoch": 5.3314121037463975,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025477262097406715,
      "loss": 4.5978,
      "mean_token_accuracy": 0.2661074593663216,
      "num_tokens": 127264138.0,
      "step": 55500
    },
    {
      "entropy": 5.611672449111938,
      "epoch": 5.331892411143132,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002547384764729335,
      "loss": 4.7729,
      "mean_token_accuracy": 0.24507382959127427,
      "num_tokens": 127275802.0,
      "step": 55505
    },
    {
      "entropy": 5.549834680557251,
      "epoch": 5.332372718539865,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025470433244221223,
      "loss": 4.666,
      "mean_token_accuracy": 0.2657864883542061,
      "num_tokens": 127287506.0,
      "step": 55510
    },
    {
      "entropy": 5.543027019500732,
      "epoch": 5.3328530259366,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002546701888826961,
      "loss": 4.6672,
      "mean_token_accuracy": 0.2626886650919914,
      "num_tokens": 127298214.0,
      "step": 55515
    },
    {
      "entropy": 5.553904485702515,
      "epoch": 5.333333333333333,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002546360457951778,
      "loss": 4.6445,
      "mean_token_accuracy": 0.25915652215480806,
      "num_tokens": 127310042.0,
      "step": 55520
    },
    {
      "entropy": 5.518614625930786,
      "epoch": 5.333813640730067,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025460190318045,
      "loss": 4.6736,
      "mean_token_accuracy": 0.26168023198843005,
      "num_tokens": 127323281.0,
      "step": 55525
    },
    {
      "entropy": 5.5707783699035645,
      "epoch": 5.334293948126801,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002545677610393055,
      "loss": 4.6432,
      "mean_token_accuracy": 0.25793426632881167,
      "num_tokens": 127335626.0,
      "step": 55530
    },
    {
      "entropy": 5.5211889266967775,
      "epoch": 5.334774255523535,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000254533619372537,
      "loss": 4.6363,
      "mean_token_accuracy": 0.2571772634983063,
      "num_tokens": 127347559.0,
      "step": 55535
    },
    {
      "entropy": 5.537135124206543,
      "epoch": 5.335254562920269,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000254499478180937,
      "loss": 4.6937,
      "mean_token_accuracy": 0.2538318172097206,
      "num_tokens": 127358512.0,
      "step": 55540
    },
    {
      "entropy": 5.55135588645935,
      "epoch": 5.335734870317003,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002544653374652982,
      "loss": 4.6313,
      "mean_token_accuracy": 0.2707674726843834,
      "num_tokens": 127369144.0,
      "step": 55545
    },
    {
      "entropy": 5.54573655128479,
      "epoch": 5.336215177713736,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025443119722641326,
      "loss": 4.7295,
      "mean_token_accuracy": 0.2548900872468948,
      "num_tokens": 127380910.0,
      "step": 55550
    },
    {
      "entropy": 5.524030876159668,
      "epoch": 5.336695485110471,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025439705746507497,
      "loss": 4.6564,
      "mean_token_accuracy": 0.25768280029296875,
      "num_tokens": 127391373.0,
      "step": 55555
    },
    {
      "entropy": 5.623467111587525,
      "epoch": 5.337175792507205,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025436291818207573,
      "loss": 4.6797,
      "mean_token_accuracy": 0.25983692705631256,
      "num_tokens": 127401987.0,
      "step": 55560
    },
    {
      "entropy": 5.5384934425354,
      "epoch": 5.337656099903938,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025432877937820836,
      "loss": 4.6124,
      "mean_token_accuracy": 0.26704676300287244,
      "num_tokens": 127414427.0,
      "step": 55565
    },
    {
      "entropy": 5.514112091064453,
      "epoch": 5.338136407300673,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025429464105426524,
      "loss": 4.6361,
      "mean_token_accuracy": 0.25847320854663847,
      "num_tokens": 127425739.0,
      "step": 55570
    },
    {
      "entropy": 5.5101423263549805,
      "epoch": 5.338616714697406,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025426050321103916,
      "loss": 4.7089,
      "mean_token_accuracy": 0.2615717530250549,
      "num_tokens": 127439237.0,
      "step": 55575
    },
    {
      "entropy": 5.538356018066406,
      "epoch": 5.3390970220941405,
      "grad_norm": 1.25,
      "learning_rate": 0.00025422636584932255,
      "loss": 4.6694,
      "mean_token_accuracy": 0.2628179430961609,
      "num_tokens": 127450634.0,
      "step": 55580
    },
    {
      "entropy": 5.569144582748413,
      "epoch": 5.339577329490874,
      "grad_norm": 1.265625,
      "learning_rate": 0.000254192228969908,
      "loss": 4.663,
      "mean_token_accuracy": 0.26336103975772857,
      "num_tokens": 127462198.0,
      "step": 55585
    },
    {
      "entropy": 5.534789419174194,
      "epoch": 5.340057636887608,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025415809257358816,
      "loss": 4.6362,
      "mean_token_accuracy": 0.2632430657744408,
      "num_tokens": 127474573.0,
      "step": 55590
    },
    {
      "entropy": 5.532739305496216,
      "epoch": 5.340537944284342,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002541239566611556,
      "loss": 4.6332,
      "mean_token_accuracy": 0.26551695466041564,
      "num_tokens": 127486332.0,
      "step": 55595
    },
    {
      "entropy": 5.570854997634887,
      "epoch": 5.341018251681076,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002540898212334027,
      "loss": 4.6816,
      "mean_token_accuracy": 0.26189617812633514,
      "num_tokens": 127497196.0,
      "step": 55600
    },
    {
      "entropy": 5.516167449951172,
      "epoch": 5.3414985590778095,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025405568629112207,
      "loss": 4.6497,
      "mean_token_accuracy": 0.2628775715827942,
      "num_tokens": 127508218.0,
      "step": 55605
    },
    {
      "entropy": 5.484581851959229,
      "epoch": 5.341978866474544,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025402155183510615,
      "loss": 4.5592,
      "mean_token_accuracy": 0.2670773446559906,
      "num_tokens": 127520196.0,
      "step": 55610
    },
    {
      "entropy": 5.451721572875977,
      "epoch": 5.342459173871277,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002539874178661476,
      "loss": 4.5096,
      "mean_token_accuracy": 0.27492485493421553,
      "num_tokens": 127532171.0,
      "step": 55615
    },
    {
      "entropy": 5.509429979324341,
      "epoch": 5.3429394812680115,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025395328438503875,
      "loss": 4.6451,
      "mean_token_accuracy": 0.2649844765663147,
      "num_tokens": 127543963.0,
      "step": 55620
    },
    {
      "entropy": 5.527005243301391,
      "epoch": 5.343419788664745,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002539191513925722,
      "loss": 4.6383,
      "mean_token_accuracy": 0.2639534339308739,
      "num_tokens": 127554066.0,
      "step": 55625
    },
    {
      "entropy": 5.5265381813049315,
      "epoch": 5.343900096061479,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025388501888954035,
      "loss": 4.674,
      "mean_token_accuracy": 0.263622023165226,
      "num_tokens": 127566133.0,
      "step": 55630
    },
    {
      "entropy": 5.527291679382325,
      "epoch": 5.344380403458214,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002538508868767356,
      "loss": 4.6419,
      "mean_token_accuracy": 0.26393268555402755,
      "num_tokens": 127577598.0,
      "step": 55635
    },
    {
      "entropy": 5.518230247497558,
      "epoch": 5.344860710854947,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002538167553549505,
      "loss": 4.649,
      "mean_token_accuracy": 0.2573309257626534,
      "num_tokens": 127589528.0,
      "step": 55640
    },
    {
      "entropy": 5.510964059829712,
      "epoch": 5.345341018251681,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002537826243249774,
      "loss": 4.5776,
      "mean_token_accuracy": 0.2699788510799408,
      "num_tokens": 127600410.0,
      "step": 55645
    },
    {
      "entropy": 5.595870542526245,
      "epoch": 5.345821325648415,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00025374849378760877,
      "loss": 4.7083,
      "mean_token_accuracy": 0.25696443766355515,
      "num_tokens": 127612748.0,
      "step": 55650
    },
    {
      "entropy": 5.5060888767242435,
      "epoch": 5.346301633045149,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002537143637436371,
      "loss": 4.6943,
      "mean_token_accuracy": 0.2572061255574226,
      "num_tokens": 127624617.0,
      "step": 55655
    },
    {
      "entropy": 5.561973714828492,
      "epoch": 5.346781940441883,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002536802341938547,
      "loss": 4.6611,
      "mean_token_accuracy": 0.2570936486124992,
      "num_tokens": 127637032.0,
      "step": 55660
    },
    {
      "entropy": 5.5713037014007565,
      "epoch": 5.347262247838617,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025364610513905377,
      "loss": 4.7013,
      "mean_token_accuracy": 0.25053686648607254,
      "num_tokens": 127648247.0,
      "step": 55665
    },
    {
      "entropy": 5.533555936813355,
      "epoch": 5.34774255523535,
      "grad_norm": 1.328125,
      "learning_rate": 0.000253611976580027,
      "loss": 4.612,
      "mean_token_accuracy": 0.26428918838500975,
      "num_tokens": 127660268.0,
      "step": 55670
    },
    {
      "entropy": 5.47346682548523,
      "epoch": 5.348222862632085,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002535778485175666,
      "loss": 4.6459,
      "mean_token_accuracy": 0.25906799286603927,
      "num_tokens": 127672489.0,
      "step": 55675
    },
    {
      "entropy": 5.5597617626190186,
      "epoch": 5.348703170028818,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002535437209524649,
      "loss": 4.6142,
      "mean_token_accuracy": 0.26810908764600755,
      "num_tokens": 127683654.0,
      "step": 55680
    },
    {
      "entropy": 5.598185110092163,
      "epoch": 5.3491834774255524,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002535095938855144,
      "loss": 4.7878,
      "mean_token_accuracy": 0.2520979791879654,
      "num_tokens": 127695314.0,
      "step": 55685
    },
    {
      "entropy": 5.60713415145874,
      "epoch": 5.349663784822286,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002534754673175073,
      "loss": 4.732,
      "mean_token_accuracy": 0.24681747406721116,
      "num_tokens": 127706732.0,
      "step": 55690
    },
    {
      "entropy": 5.488000297546387,
      "epoch": 5.35014409221902,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002534413412492359,
      "loss": 4.6242,
      "mean_token_accuracy": 0.27018008232116697,
      "num_tokens": 127719689.0,
      "step": 55695
    },
    {
      "entropy": 5.526687717437744,
      "epoch": 5.350624399615754,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002534072156814925,
      "loss": 4.6766,
      "mean_token_accuracy": 0.26118214428424835,
      "num_tokens": 127732113.0,
      "step": 55700
    },
    {
      "entropy": 5.464137077331543,
      "epoch": 5.351104707012488,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025337309061506947,
      "loss": 4.5745,
      "mean_token_accuracy": 0.2678020536899567,
      "num_tokens": 127743997.0,
      "step": 55705
    },
    {
      "entropy": 5.616693305969238,
      "epoch": 5.351585014409222,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000253338966050759,
      "loss": 4.7424,
      "mean_token_accuracy": 0.2465811237692833,
      "num_tokens": 127755967.0,
      "step": 55710
    },
    {
      "entropy": 5.523854303359985,
      "epoch": 5.352065321805956,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025330484198935355,
      "loss": 4.636,
      "mean_token_accuracy": 0.26098983734846115,
      "num_tokens": 127767322.0,
      "step": 55715
    },
    {
      "entropy": 5.510820627212524,
      "epoch": 5.35254562920269,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002532707184316451,
      "loss": 4.5887,
      "mean_token_accuracy": 0.26919025182724,
      "num_tokens": 127778205.0,
      "step": 55720
    },
    {
      "entropy": 5.57858943939209,
      "epoch": 5.3530259365994235,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002532365953784262,
      "loss": 4.6691,
      "mean_token_accuracy": 0.2594780772924423,
      "num_tokens": 127788559.0,
      "step": 55725
    },
    {
      "entropy": 5.5419114112854,
      "epoch": 5.353506243996158,
      "grad_norm": 1.25,
      "learning_rate": 0.0002532024728304888,
      "loss": 4.7313,
      "mean_token_accuracy": 0.26141108721494677,
      "num_tokens": 127800729.0,
      "step": 55730
    },
    {
      "entropy": 5.562019157409668,
      "epoch": 5.353986551392891,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002531683507886253,
      "loss": 4.7074,
      "mean_token_accuracy": 0.2550425291061401,
      "num_tokens": 127813128.0,
      "step": 55735
    },
    {
      "entropy": 5.587524938583374,
      "epoch": 5.354466858789626,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002531342292536279,
      "loss": 4.6444,
      "mean_token_accuracy": 0.26106803864240646,
      "num_tokens": 127824087.0,
      "step": 55740
    },
    {
      "entropy": 5.471872615814209,
      "epoch": 5.354947166186359,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002531001082262887,
      "loss": 4.5761,
      "mean_token_accuracy": 0.2698852926492691,
      "num_tokens": 127835938.0,
      "step": 55745
    },
    {
      "entropy": 5.521651649475098,
      "epoch": 5.355427473583093,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025306598770740004,
      "loss": 4.6547,
      "mean_token_accuracy": 0.25974135398864745,
      "num_tokens": 127848552.0,
      "step": 55750
    },
    {
      "entropy": 5.518854761123658,
      "epoch": 5.355907780979827,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025303186769775395,
      "loss": 4.6137,
      "mean_token_accuracy": 0.26839018166065215,
      "num_tokens": 127862165.0,
      "step": 55755
    },
    {
      "entropy": 5.540425968170166,
      "epoch": 5.356388088376561,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025299774819814263,
      "loss": 4.6499,
      "mean_token_accuracy": 0.2605888843536377,
      "num_tokens": 127874070.0,
      "step": 55760
    },
    {
      "entropy": 5.544436454772949,
      "epoch": 5.3568683957732945,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002529636292093584,
      "loss": 4.68,
      "mean_token_accuracy": 0.2584880694746971,
      "num_tokens": 127886197.0,
      "step": 55765
    },
    {
      "entropy": 5.48245210647583,
      "epoch": 5.357348703170029,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002529295107321932,
      "loss": 4.5959,
      "mean_token_accuracy": 0.2678855299949646,
      "num_tokens": 127897073.0,
      "step": 55770
    },
    {
      "entropy": 5.5277259349823,
      "epoch": 5.357829010566762,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002528953927674392,
      "loss": 4.6684,
      "mean_token_accuracy": 0.2533862069249153,
      "num_tokens": 127907718.0,
      "step": 55775
    },
    {
      "entropy": 5.546145105361939,
      "epoch": 5.358309317963497,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002528612753158885,
      "loss": 4.6094,
      "mean_token_accuracy": 0.26188151985406877,
      "num_tokens": 127918910.0,
      "step": 55780
    },
    {
      "entropy": 5.472902584075928,
      "epoch": 5.358789625360231,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002528271583783333,
      "loss": 4.6315,
      "mean_token_accuracy": 0.2635385513305664,
      "num_tokens": 127929459.0,
      "step": 55785
    },
    {
      "entropy": 5.4764872074127195,
      "epoch": 5.359269932756964,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002527930419555656,
      "loss": 4.649,
      "mean_token_accuracy": 0.26336565911769866,
      "num_tokens": 127941541.0,
      "step": 55790
    },
    {
      "entropy": 5.630435514450073,
      "epoch": 5.359750240153699,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002527589260483776,
      "loss": 4.7336,
      "mean_token_accuracy": 0.26156100779771807,
      "num_tokens": 127954000.0,
      "step": 55795
    },
    {
      "entropy": 5.567986488342285,
      "epoch": 5.360230547550432,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002527248106575613,
      "loss": 4.6435,
      "mean_token_accuracy": 0.26384624689817426,
      "num_tokens": 127964717.0,
      "step": 55800
    },
    {
      "entropy": 5.513719367980957,
      "epoch": 5.3607108549471665,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025269069578390875,
      "loss": 4.6479,
      "mean_token_accuracy": 0.2570356473326683,
      "num_tokens": 127976284.0,
      "step": 55805
    },
    {
      "entropy": 5.53089165687561,
      "epoch": 5.3611911623439,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002526565814282119,
      "loss": 4.6453,
      "mean_token_accuracy": 0.27024729251861573,
      "num_tokens": 127987854.0,
      "step": 55810
    },
    {
      "entropy": 5.506406927108765,
      "epoch": 5.361671469740634,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025262246759126295,
      "loss": 4.6523,
      "mean_token_accuracy": 0.2605032831430435,
      "num_tokens": 127999490.0,
      "step": 55815
    },
    {
      "entropy": 5.53217887878418,
      "epoch": 5.362151777137368,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025258835427385384,
      "loss": 4.6495,
      "mean_token_accuracy": 0.26278277933597566,
      "num_tokens": 128009623.0,
      "step": 55820
    },
    {
      "entropy": 5.528398132324218,
      "epoch": 5.362632084534102,
      "grad_norm": 1.375,
      "learning_rate": 0.00025255424147677656,
      "loss": 4.6144,
      "mean_token_accuracy": 0.26032963693141936,
      "num_tokens": 128021751.0,
      "step": 55825
    },
    {
      "entropy": 5.580239629745483,
      "epoch": 5.363112391930835,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025252012920082316,
      "loss": 4.7234,
      "mean_token_accuracy": 0.25459985733032225,
      "num_tokens": 128034248.0,
      "step": 55830
    },
    {
      "entropy": 5.585423851013184,
      "epoch": 5.36359269932757,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002524860174467858,
      "loss": 4.6876,
      "mean_token_accuracy": 0.26354406028985977,
      "num_tokens": 128046719.0,
      "step": 55835
    },
    {
      "entropy": 5.505116415023804,
      "epoch": 5.364073006724303,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025245190621545603,
      "loss": 4.6449,
      "mean_token_accuracy": 0.2677885442972183,
      "num_tokens": 128058695.0,
      "step": 55840
    },
    {
      "entropy": 5.53577184677124,
      "epoch": 5.3645533141210375,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002524177955076261,
      "loss": 4.6883,
      "mean_token_accuracy": 0.2546923667192459,
      "num_tokens": 128070593.0,
      "step": 55845
    },
    {
      "entropy": 5.5029693126678465,
      "epoch": 5.365033621517771,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002523836853240879,
      "loss": 4.6118,
      "mean_token_accuracy": 0.2662314802408218,
      "num_tokens": 128081180.0,
      "step": 55850
    },
    {
      "entropy": 5.562829923629761,
      "epoch": 5.365513928914505,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002523495756656334,
      "loss": 4.6825,
      "mean_token_accuracy": 0.26062356531620023,
      "num_tokens": 128093986.0,
      "step": 55855
    },
    {
      "entropy": 5.519537115097046,
      "epoch": 5.36599423631124,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002523154665330544,
      "loss": 4.6818,
      "mean_token_accuracy": 0.2654334381222725,
      "num_tokens": 128105788.0,
      "step": 55860
    },
    {
      "entropy": 5.517454767227173,
      "epoch": 5.366474543707973,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025228135792714314,
      "loss": 4.6425,
      "mean_token_accuracy": 0.2597613468766212,
      "num_tokens": 128116336.0,
      "step": 55865
    },
    {
      "entropy": 5.542007064819336,
      "epoch": 5.366954851104707,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002522472498486911,
      "loss": 4.6921,
      "mean_token_accuracy": 0.25545276403427125,
      "num_tokens": 128128030.0,
      "step": 55870
    },
    {
      "entropy": 5.575599908828735,
      "epoch": 5.367435158501441,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025221314229849037,
      "loss": 4.7107,
      "mean_token_accuracy": 0.25785051584243773,
      "num_tokens": 128139113.0,
      "step": 55875
    },
    {
      "entropy": 5.588823223114014,
      "epoch": 5.367915465898175,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002521790352773328,
      "loss": 4.7277,
      "mean_token_accuracy": 0.2578424334526062,
      "num_tokens": 128151912.0,
      "step": 55880
    },
    {
      "entropy": 5.4865625381469725,
      "epoch": 5.368395773294909,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025214492878601033,
      "loss": 4.6891,
      "mean_token_accuracy": 0.2580589532852173,
      "num_tokens": 128164623.0,
      "step": 55885
    },
    {
      "entropy": 5.556678342819214,
      "epoch": 5.368876080691643,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025211082282531466,
      "loss": 4.6895,
      "mean_token_accuracy": 0.2531604304909706,
      "num_tokens": 128176354.0,
      "step": 55890
    },
    {
      "entropy": 5.5483973026275635,
      "epoch": 5.369356388088376,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025207671739603776,
      "loss": 4.6814,
      "mean_token_accuracy": 0.258969384431839,
      "num_tokens": 128189464.0,
      "step": 55895
    },
    {
      "entropy": 5.479499292373657,
      "epoch": 5.369836695485111,
      "grad_norm": 1.375,
      "learning_rate": 0.00025204261249897153,
      "loss": 4.5932,
      "mean_token_accuracy": 0.2689276680350304,
      "num_tokens": 128200939.0,
      "step": 55900
    },
    {
      "entropy": 5.480694341659546,
      "epoch": 5.370317002881844,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025200850813490754,
      "loss": 4.5885,
      "mean_token_accuracy": 0.2620487928390503,
      "num_tokens": 128213059.0,
      "step": 55905
    },
    {
      "entropy": 5.549109840393067,
      "epoch": 5.370797310278578,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025197440430463775,
      "loss": 4.6934,
      "mean_token_accuracy": 0.26136996895074843,
      "num_tokens": 128224824.0,
      "step": 55910
    },
    {
      "entropy": 5.492754983901977,
      "epoch": 5.371277617675312,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025194030100895396,
      "loss": 4.6286,
      "mean_token_accuracy": 0.2609574794769287,
      "num_tokens": 128235621.0,
      "step": 55915
    },
    {
      "entropy": 5.5819591045379635,
      "epoch": 5.371757925072046,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002519061982486478,
      "loss": 4.7618,
      "mean_token_accuracy": 0.25833494365215304,
      "num_tokens": 128246209.0,
      "step": 55920
    },
    {
      "entropy": 5.615984773635864,
      "epoch": 5.37223823246878,
      "grad_norm": 1.375,
      "learning_rate": 0.00025187209602451114,
      "loss": 4.7712,
      "mean_token_accuracy": 0.2573565408587456,
      "num_tokens": 128257445.0,
      "step": 55925
    },
    {
      "entropy": 5.603793144226074,
      "epoch": 5.372718539865514,
      "grad_norm": 1.421875,
      "learning_rate": 0.000251837994337336,
      "loss": 4.7799,
      "mean_token_accuracy": 0.2481815606355667,
      "num_tokens": 128269075.0,
      "step": 55930
    },
    {
      "entropy": 5.534134864807129,
      "epoch": 5.373198847262248,
      "grad_norm": 1.25,
      "learning_rate": 0.0002518038931879136,
      "loss": 4.6018,
      "mean_token_accuracy": 0.2646864905953407,
      "num_tokens": 128279728.0,
      "step": 55935
    },
    {
      "entropy": 5.50441255569458,
      "epoch": 5.373679154658982,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025176979257703595,
      "loss": 4.593,
      "mean_token_accuracy": 0.26898993998765947,
      "num_tokens": 128290197.0,
      "step": 55940
    },
    {
      "entropy": 5.510973739624023,
      "epoch": 5.374159462055716,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002517356925054948,
      "loss": 4.6589,
      "mean_token_accuracy": 0.2611307367682457,
      "num_tokens": 128300897.0,
      "step": 55945
    },
    {
      "entropy": 5.582307529449463,
      "epoch": 5.3746397694524495,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002517015929740817,
      "loss": 4.6879,
      "mean_token_accuracy": 0.26200558692216874,
      "num_tokens": 128311651.0,
      "step": 55950
    },
    {
      "entropy": 5.552903032302856,
      "epoch": 5.375120076849184,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002516674939835885,
      "loss": 4.6936,
      "mean_token_accuracy": 0.2554493024945259,
      "num_tokens": 128324003.0,
      "step": 55955
    },
    {
      "entropy": 5.578551578521728,
      "epoch": 5.375600384245917,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025163339553480686,
      "loss": 4.7603,
      "mean_token_accuracy": 0.25198402255773544,
      "num_tokens": 128335112.0,
      "step": 55960
    },
    {
      "entropy": 5.547757196426391,
      "epoch": 5.376080691642652,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002515992976285283,
      "loss": 4.6784,
      "mean_token_accuracy": 0.2609168916940689,
      "num_tokens": 128347591.0,
      "step": 55965
    },
    {
      "entropy": 5.539571237564087,
      "epoch": 5.376560999039385,
      "grad_norm": 1.375,
      "learning_rate": 0.0002515652002655446,
      "loss": 4.616,
      "mean_token_accuracy": 0.2635528385639191,
      "num_tokens": 128359340.0,
      "step": 55970
    },
    {
      "entropy": 5.50718731880188,
      "epoch": 5.377041306436119,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025153110344664734,
      "loss": 4.6147,
      "mean_token_accuracy": 0.26138823181390763,
      "num_tokens": 128371065.0,
      "step": 55975
    },
    {
      "entropy": 5.550933218002319,
      "epoch": 5.377521613832853,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025149700717262815,
      "loss": 4.6594,
      "mean_token_accuracy": 0.2595532566308975,
      "num_tokens": 128382729.0,
      "step": 55980
    },
    {
      "entropy": 5.537700557708741,
      "epoch": 5.378001921229587,
      "grad_norm": 1.25,
      "learning_rate": 0.00025146291144427865,
      "loss": 4.6844,
      "mean_token_accuracy": 0.2576386481523514,
      "num_tokens": 128394763.0,
      "step": 55985
    },
    {
      "entropy": 5.594408988952637,
      "epoch": 5.3784822286263205,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002514288162623905,
      "loss": 4.7557,
      "mean_token_accuracy": 0.24921808242797852,
      "num_tokens": 128406331.0,
      "step": 55990
    },
    {
      "entropy": 5.48071608543396,
      "epoch": 5.378962536023055,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025139472162775527,
      "loss": 4.5607,
      "mean_token_accuracy": 0.26461431533098223,
      "num_tokens": 128416732.0,
      "step": 55995
    },
    {
      "entropy": 5.626568222045899,
      "epoch": 5.379442843419788,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002513606275411645,
      "loss": 4.7461,
      "mean_token_accuracy": 0.2595515847206116,
      "num_tokens": 128427331.0,
      "step": 56000
    },
    {
      "entropy": 5.510546112060547,
      "epoch": 5.379923150816523,
      "grad_norm": 1.375,
      "learning_rate": 0.0002513265340034097,
      "loss": 4.6362,
      "mean_token_accuracy": 0.26129461973905566,
      "num_tokens": 128439206.0,
      "step": 56005
    },
    {
      "entropy": 5.545032072067261,
      "epoch": 5.380403458213257,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002512924410152825,
      "loss": 4.7087,
      "mean_token_accuracy": 0.2556914329528809,
      "num_tokens": 128451883.0,
      "step": 56010
    },
    {
      "entropy": 5.586467313766479,
      "epoch": 5.38088376560999,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025125834857757444,
      "loss": 4.6758,
      "mean_token_accuracy": 0.2588024392724037,
      "num_tokens": 128463004.0,
      "step": 56015
    },
    {
      "entropy": 5.564021062850952,
      "epoch": 5.381364073006725,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002512242566910771,
      "loss": 4.6234,
      "mean_token_accuracy": 0.26593668311834334,
      "num_tokens": 128475613.0,
      "step": 56020
    },
    {
      "entropy": 5.6006872177124025,
      "epoch": 5.381844380403458,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025119016535658193,
      "loss": 4.7354,
      "mean_token_accuracy": 0.25518318861722944,
      "num_tokens": 128487483.0,
      "step": 56025
    },
    {
      "entropy": 5.588147163391113,
      "epoch": 5.3823246878001925,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002511560745748803,
      "loss": 4.6874,
      "mean_token_accuracy": 0.2538062885403633,
      "num_tokens": 128498897.0,
      "step": 56030
    },
    {
      "entropy": 5.524995994567871,
      "epoch": 5.382804995196926,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025112198434676386,
      "loss": 4.6599,
      "mean_token_accuracy": 0.2606758281588554,
      "num_tokens": 128510357.0,
      "step": 56035
    },
    {
      "entropy": 5.565561723709107,
      "epoch": 5.38328530259366,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002510878946730241,
      "loss": 4.6777,
      "mean_token_accuracy": 0.26287419646978377,
      "num_tokens": 128521683.0,
      "step": 56040
    },
    {
      "entropy": 5.557748556137085,
      "epoch": 5.383765609990394,
      "grad_norm": 1.25,
      "learning_rate": 0.0002510538055544523,
      "loss": 4.6416,
      "mean_token_accuracy": 0.26877193748950956,
      "num_tokens": 128532190.0,
      "step": 56045
    },
    {
      "entropy": 5.5279974937438965,
      "epoch": 5.384245917387128,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025101971699184014,
      "loss": 4.6488,
      "mean_token_accuracy": 0.25554143637418747,
      "num_tokens": 128543591.0,
      "step": 56050
    },
    {
      "entropy": 5.532762956619263,
      "epoch": 5.384726224783861,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025098562898597887,
      "loss": 4.6956,
      "mean_token_accuracy": 0.2584079146385193,
      "num_tokens": 128555592.0,
      "step": 56055
    },
    {
      "entropy": 5.5605720520019535,
      "epoch": 5.385206532180596,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025095154153766006,
      "loss": 4.7277,
      "mean_token_accuracy": 0.2543741285800934,
      "num_tokens": 128568773.0,
      "step": 56060
    },
    {
      "entropy": 5.531645965576172,
      "epoch": 5.385686839577329,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025091745464767495,
      "loss": 4.6216,
      "mean_token_accuracy": 0.261121129989624,
      "num_tokens": 128580346.0,
      "step": 56065
    },
    {
      "entropy": 5.574544382095337,
      "epoch": 5.3861671469740635,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025088336831681507,
      "loss": 4.6871,
      "mean_token_accuracy": 0.25895233899354936,
      "num_tokens": 128591478.0,
      "step": 56070
    },
    {
      "entropy": 5.541067409515381,
      "epoch": 5.386647454370797,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002508492825458719,
      "loss": 4.6723,
      "mean_token_accuracy": 0.2583088383078575,
      "num_tokens": 128603320.0,
      "step": 56075
    },
    {
      "entropy": 5.663867092132568,
      "epoch": 5.387127761767531,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025081519733563654,
      "loss": 4.7661,
      "mean_token_accuracy": 0.2499040499329567,
      "num_tokens": 128614478.0,
      "step": 56080
    },
    {
      "entropy": 5.549566984176636,
      "epoch": 5.387608069164266,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025078111268690044,
      "loss": 4.6505,
      "mean_token_accuracy": 0.2570642501115799,
      "num_tokens": 128626198.0,
      "step": 56085
    },
    {
      "entropy": 5.512462377548218,
      "epoch": 5.388088376560999,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025074702860045507,
      "loss": 4.639,
      "mean_token_accuracy": 0.2650735929608345,
      "num_tokens": 128639042.0,
      "step": 56090
    },
    {
      "entropy": 5.530058670043945,
      "epoch": 5.388568683957733,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002507129450770916,
      "loss": 4.6696,
      "mean_token_accuracy": 0.2605062186717987,
      "num_tokens": 128649738.0,
      "step": 56095
    },
    {
      "entropy": 5.539209699630737,
      "epoch": 5.389048991354467,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025067886211760154,
      "loss": 4.6274,
      "mean_token_accuracy": 0.26733191311359406,
      "num_tokens": 128660420.0,
      "step": 56100
    },
    {
      "entropy": 5.532786798477173,
      "epoch": 5.389529298751201,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000250644779722776,
      "loss": 4.7155,
      "mean_token_accuracy": 0.258499276638031,
      "num_tokens": 128671861.0,
      "step": 56105
    },
    {
      "entropy": 5.484158039093018,
      "epoch": 5.390009606147935,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002506106978934065,
      "loss": 4.669,
      "mean_token_accuracy": 0.25916404128074644,
      "num_tokens": 128683220.0,
      "step": 56110
    },
    {
      "entropy": 5.6400000095367435,
      "epoch": 5.390489913544669,
      "grad_norm": 1.328125,
      "learning_rate": 0.000250576616630284,
      "loss": 4.772,
      "mean_token_accuracy": 0.24975321739912032,
      "num_tokens": 128695056.0,
      "step": 56115
    },
    {
      "entropy": 5.594123077392578,
      "epoch": 5.390970220941402,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002505425359342,
      "loss": 4.7487,
      "mean_token_accuracy": 0.2513478174805641,
      "num_tokens": 128705696.0,
      "step": 56120
    },
    {
      "entropy": 5.5771777629852295,
      "epoch": 5.391450528338137,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025050845580594565,
      "loss": 4.6569,
      "mean_token_accuracy": 0.2631892442703247,
      "num_tokens": 128717557.0,
      "step": 56125
    },
    {
      "entropy": 5.599954843521118,
      "epoch": 5.39193083573487,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002504743762463123,
      "loss": 4.7308,
      "mean_token_accuracy": 0.2502649426460266,
      "num_tokens": 128729784.0,
      "step": 56130
    },
    {
      "entropy": 5.49848952293396,
      "epoch": 5.392411143131604,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002504402972560911,
      "loss": 4.5595,
      "mean_token_accuracy": 0.26736327856779096,
      "num_tokens": 128741225.0,
      "step": 56135
    },
    {
      "entropy": 5.554753732681275,
      "epoch": 5.392891450528338,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002504062188360733,
      "loss": 4.6881,
      "mean_token_accuracy": 0.2547506779432297,
      "num_tokens": 128753302.0,
      "step": 56140
    },
    {
      "entropy": 5.5722143173217775,
      "epoch": 5.393371757925072,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025037214098705,
      "loss": 4.6471,
      "mean_token_accuracy": 0.2672333285212517,
      "num_tokens": 128764402.0,
      "step": 56145
    },
    {
      "entropy": 5.5690874576568605,
      "epoch": 5.393852065321806,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002503380637098125,
      "loss": 4.7357,
      "mean_token_accuracy": 0.2583767846226692,
      "num_tokens": 128776314.0,
      "step": 56150
    },
    {
      "entropy": 5.560726308822632,
      "epoch": 5.39433237271854,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002503039870051519,
      "loss": 4.7644,
      "mean_token_accuracy": 0.2536254584789276,
      "num_tokens": 128789099.0,
      "step": 56155
    },
    {
      "entropy": 5.492218494415283,
      "epoch": 5.394812680115273,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025026991087385937,
      "loss": 4.624,
      "mean_token_accuracy": 0.2641188845038414,
      "num_tokens": 128800492.0,
      "step": 56160
    },
    {
      "entropy": 5.498374176025391,
      "epoch": 5.395292987512008,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025023583531672607,
      "loss": 4.6066,
      "mean_token_accuracy": 0.2638247013092041,
      "num_tokens": 128811524.0,
      "step": 56165
    },
    {
      "entropy": 5.520421886444092,
      "epoch": 5.395773294908741,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002502017603345433,
      "loss": 4.6353,
      "mean_token_accuracy": 0.26780261397361754,
      "num_tokens": 128822710.0,
      "step": 56170
    },
    {
      "entropy": 5.557804536819458,
      "epoch": 5.3962536023054755,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025016768592810184,
      "loss": 4.7066,
      "mean_token_accuracy": 0.2632907286286354,
      "num_tokens": 128833871.0,
      "step": 56175
    },
    {
      "entropy": 5.518630456924439,
      "epoch": 5.39673390970221,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000250133612098193,
      "loss": 4.6585,
      "mean_token_accuracy": 0.2611532911658287,
      "num_tokens": 128846386.0,
      "step": 56180
    },
    {
      "entropy": 5.5275664806365965,
      "epoch": 5.397214217098943,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002500995388456079,
      "loss": 4.6913,
      "mean_token_accuracy": 0.2589512914419174,
      "num_tokens": 128857195.0,
      "step": 56185
    },
    {
      "entropy": 5.470988845825195,
      "epoch": 5.3976945244956775,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002500654661711375,
      "loss": 4.5504,
      "mean_token_accuracy": 0.27044502049684527,
      "num_tokens": 128868807.0,
      "step": 56190
    },
    {
      "entropy": 5.572797679901123,
      "epoch": 5.398174831892411,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025003139407557296,
      "loss": 4.7126,
      "mean_token_accuracy": 0.2584269270300865,
      "num_tokens": 128881210.0,
      "step": 56195
    },
    {
      "entropy": 5.541218090057373,
      "epoch": 5.398655139289145,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002499973225597054,
      "loss": 4.6706,
      "mean_token_accuracy": 0.2628727912902832,
      "num_tokens": 128892238.0,
      "step": 56200
    },
    {
      "entropy": 5.491672849655151,
      "epoch": 5.399135446685879,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002499632516243257,
      "loss": 4.5842,
      "mean_token_accuracy": 0.263763827085495,
      "num_tokens": 128903498.0,
      "step": 56205
    },
    {
      "entropy": 5.5034754276275635,
      "epoch": 5.399615754082613,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000249929181270225,
      "loss": 4.6252,
      "mean_token_accuracy": 0.25650223195552824,
      "num_tokens": 128914828.0,
      "step": 56210
    },
    {
      "entropy": 5.610933017730713,
      "epoch": 5.4000960614793465,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002498951114981942,
      "loss": 4.7467,
      "mean_token_accuracy": 0.2584263250231743,
      "num_tokens": 128925815.0,
      "step": 56215
    },
    {
      "entropy": 5.488397073745728,
      "epoch": 5.400576368876081,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002498610423090243,
      "loss": 4.5809,
      "mean_token_accuracy": 0.2680394098162651,
      "num_tokens": 128936531.0,
      "step": 56220
    },
    {
      "entropy": 5.5213793277740475,
      "epoch": 5.401056676272814,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002498269737035064,
      "loss": 4.632,
      "mean_token_accuracy": 0.2629841357469559,
      "num_tokens": 128948189.0,
      "step": 56225
    },
    {
      "entropy": 5.4579753398895265,
      "epoch": 5.401536983669549,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002497929056824316,
      "loss": 4.5727,
      "mean_token_accuracy": 0.2721137449145317,
      "num_tokens": 128959171.0,
      "step": 56230
    },
    {
      "entropy": 5.399478387832642,
      "epoch": 5.402017291066282,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002497588382465905,
      "loss": 4.5133,
      "mean_token_accuracy": 0.2679280757904053,
      "num_tokens": 128970972.0,
      "step": 56235
    },
    {
      "entropy": 5.425375080108642,
      "epoch": 5.402497598463016,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024972477139677425,
      "loss": 4.5411,
      "mean_token_accuracy": 0.26835915446281433,
      "num_tokens": 128983154.0,
      "step": 56240
    },
    {
      "entropy": 5.472449922561646,
      "epoch": 5.40297790585975,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024969070513377375,
      "loss": 4.6374,
      "mean_token_accuracy": 0.2596636489033699,
      "num_tokens": 128995430.0,
      "step": 56245
    },
    {
      "entropy": 5.561673021316528,
      "epoch": 5.403458213256484,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002496566394583799,
      "loss": 4.6752,
      "mean_token_accuracy": 0.2578496292233467,
      "num_tokens": 129007236.0,
      "step": 56250
    },
    {
      "entropy": 5.547601318359375,
      "epoch": 5.4039385206532184,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002496225743713837,
      "loss": 4.6243,
      "mean_token_accuracy": 0.2660650432109833,
      "num_tokens": 129019372.0,
      "step": 56255
    },
    {
      "entropy": 5.48132357597351,
      "epoch": 5.404418828049952,
      "grad_norm": 1.28125,
      "learning_rate": 0.000249588509873576,
      "loss": 4.6223,
      "mean_token_accuracy": 0.26945777982473373,
      "num_tokens": 129031146.0,
      "step": 56260
    },
    {
      "entropy": 5.501860952377319,
      "epoch": 5.404899135446686,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024955444596574754,
      "loss": 4.7175,
      "mean_token_accuracy": 0.2540468260645866,
      "num_tokens": 129042727.0,
      "step": 56265
    },
    {
      "entropy": 5.570419883728027,
      "epoch": 5.40537944284342,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002495203826486893,
      "loss": 4.6802,
      "mean_token_accuracy": 0.2703128129243851,
      "num_tokens": 129054562.0,
      "step": 56270
    },
    {
      "entropy": 5.5517627716064455,
      "epoch": 5.405859750240154,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024948631992319215,
      "loss": 4.6,
      "mean_token_accuracy": 0.2682794749736786,
      "num_tokens": 129065558.0,
      "step": 56275
    },
    {
      "entropy": 5.588101816177368,
      "epoch": 5.406340057636887,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002494522577900469,
      "loss": 4.6945,
      "mean_token_accuracy": 0.2625554069876671,
      "num_tokens": 129076726.0,
      "step": 56280
    },
    {
      "entropy": 5.511349248886108,
      "epoch": 5.406820365033622,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002494181962500443,
      "loss": 4.6884,
      "mean_token_accuracy": 0.25776808708906174,
      "num_tokens": 129087502.0,
      "step": 56285
    },
    {
      "entropy": 5.508719158172608,
      "epoch": 5.407300672430355,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002493841353039753,
      "loss": 4.7082,
      "mean_token_accuracy": 0.25593016892671583,
      "num_tokens": 129098816.0,
      "step": 56290
    },
    {
      "entropy": 5.618701839447022,
      "epoch": 5.4077809798270895,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024935007495263054,
      "loss": 4.6967,
      "mean_token_accuracy": 0.25622748136520385,
      "num_tokens": 129109466.0,
      "step": 56295
    },
    {
      "entropy": 5.487134504318237,
      "epoch": 5.408261287223823,
      "grad_norm": 1.375,
      "learning_rate": 0.000249316015196801,
      "loss": 4.5796,
      "mean_token_accuracy": 0.2705535888671875,
      "num_tokens": 129121706.0,
      "step": 56300
    },
    {
      "entropy": 5.485473155975342,
      "epoch": 5.408741594620557,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002492819560372772,
      "loss": 4.6633,
      "mean_token_accuracy": 0.2574503421783447,
      "num_tokens": 129133687.0,
      "step": 56305
    },
    {
      "entropy": 5.547125816345215,
      "epoch": 5.409221902017291,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024924789747485005,
      "loss": 4.6606,
      "mean_token_accuracy": 0.26446772664785384,
      "num_tokens": 129145259.0,
      "step": 56310
    },
    {
      "entropy": 5.511079168319702,
      "epoch": 5.409702209414025,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024921383951031025,
      "loss": 4.639,
      "mean_token_accuracy": 0.2640919744968414,
      "num_tokens": 129158415.0,
      "step": 56315
    },
    {
      "entropy": 5.565230989456177,
      "epoch": 5.4101825168107585,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002491797821444485,
      "loss": 4.6866,
      "mean_token_accuracy": 0.2574472546577454,
      "num_tokens": 129169593.0,
      "step": 56320
    },
    {
      "entropy": 5.531215333938599,
      "epoch": 5.410662824207493,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024914572537805557,
      "loss": 4.7029,
      "mean_token_accuracy": 0.26091964542865753,
      "num_tokens": 129181394.0,
      "step": 56325
    },
    {
      "entropy": 5.481481266021729,
      "epoch": 5.411143131604227,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024911166921192215,
      "loss": 4.5859,
      "mean_token_accuracy": 0.2658102631568909,
      "num_tokens": 129193761.0,
      "step": 56330
    },
    {
      "entropy": 5.521560621261597,
      "epoch": 5.4116234390009605,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024907761364683886,
      "loss": 4.6279,
      "mean_token_accuracy": 0.2570706993341446,
      "num_tokens": 129206034.0,
      "step": 56335
    },
    {
      "entropy": 5.586119556427002,
      "epoch": 5.412103746397695,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024904355868359646,
      "loss": 4.7331,
      "mean_token_accuracy": 0.25184202641248704,
      "num_tokens": 129218787.0,
      "step": 56340
    },
    {
      "entropy": 5.633624315261841,
      "epoch": 5.412584053794428,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024900950432298546,
      "loss": 4.6822,
      "mean_token_accuracy": 0.26271059066057206,
      "num_tokens": 129230586.0,
      "step": 56345
    },
    {
      "entropy": 5.532357215881348,
      "epoch": 5.413064361191163,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024897545056579666,
      "loss": 4.7156,
      "mean_token_accuracy": 0.25674937218427657,
      "num_tokens": 129241262.0,
      "step": 56350
    },
    {
      "entropy": 5.509925127029419,
      "epoch": 5.413544668587896,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002489413974128206,
      "loss": 4.6348,
      "mean_token_accuracy": 0.26712643504142763,
      "num_tokens": 129251781.0,
      "step": 56355
    },
    {
      "entropy": 5.525384426116943,
      "epoch": 5.41402497598463,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000248907344864848,
      "loss": 4.7062,
      "mean_token_accuracy": 0.2609278500080109,
      "num_tokens": 129263494.0,
      "step": 56360
    },
    {
      "entropy": 5.621101665496826,
      "epoch": 5.414505283381364,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024887329292266923,
      "loss": 4.7221,
      "mean_token_accuracy": 0.26306972503662107,
      "num_tokens": 129275946.0,
      "step": 56365
    },
    {
      "entropy": 5.559212732315063,
      "epoch": 5.414985590778098,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024883924158707505,
      "loss": 4.7018,
      "mean_token_accuracy": 0.2628184348344803,
      "num_tokens": 129287270.0,
      "step": 56370
    },
    {
      "entropy": 5.527575922012329,
      "epoch": 5.415465898174832,
      "grad_norm": 1.375,
      "learning_rate": 0.00024880519085885606,
      "loss": 4.6708,
      "mean_token_accuracy": 0.2560626372694969,
      "num_tokens": 129298857.0,
      "step": 56375
    },
    {
      "entropy": 5.533285760879517,
      "epoch": 5.415946205571566,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002487711407388028,
      "loss": 4.7395,
      "mean_token_accuracy": 0.26252520382404326,
      "num_tokens": 129311340.0,
      "step": 56380
    },
    {
      "entropy": 5.485003566741943,
      "epoch": 5.416426512968299,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024873709122770563,
      "loss": 4.5809,
      "mean_token_accuracy": 0.267927759885788,
      "num_tokens": 129322445.0,
      "step": 56385
    },
    {
      "entropy": 5.551542329788208,
      "epoch": 5.416906820365034,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002487030423263553,
      "loss": 4.6669,
      "mean_token_accuracy": 0.25974806398153305,
      "num_tokens": 129334970.0,
      "step": 56390
    },
    {
      "entropy": 5.520151615142822,
      "epoch": 5.417387127761767,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002486689940355422,
      "loss": 4.6314,
      "mean_token_accuracy": 0.2620521351695061,
      "num_tokens": 129346151.0,
      "step": 56395
    },
    {
      "entropy": 5.555585432052612,
      "epoch": 5.417867435158501,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002486349463560569,
      "loss": 4.6975,
      "mean_token_accuracy": 0.2636315867304802,
      "num_tokens": 129356961.0,
      "step": 56400
    },
    {
      "entropy": 5.550709772109985,
      "epoch": 5.418347742555236,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002486008992886898,
      "loss": 4.6746,
      "mean_token_accuracy": 0.2608560279011726,
      "num_tokens": 129368363.0,
      "step": 56405
    },
    {
      "entropy": 5.555657529830933,
      "epoch": 5.418828049951969,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002485668528342316,
      "loss": 4.6617,
      "mean_token_accuracy": 0.25485379695892335,
      "num_tokens": 129378841.0,
      "step": 56410
    },
    {
      "entropy": 5.582209491729737,
      "epoch": 5.4193083573487035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024853280699347247,
      "loss": 4.7182,
      "mean_token_accuracy": 0.25456133037805556,
      "num_tokens": 129390075.0,
      "step": 56415
    },
    {
      "entropy": 5.505593204498291,
      "epoch": 5.419788664745437,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002484987617672029,
      "loss": 4.614,
      "mean_token_accuracy": 0.2574278101325035,
      "num_tokens": 129401593.0,
      "step": 56420
    },
    {
      "entropy": 5.4515069961547855,
      "epoch": 5.420268972142171,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002484647171562135,
      "loss": 4.6337,
      "mean_token_accuracy": 0.26569215655326844,
      "num_tokens": 129413004.0,
      "step": 56425
    },
    {
      "entropy": 5.527483034133911,
      "epoch": 5.420749279538905,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024843067316129445,
      "loss": 4.6389,
      "mean_token_accuracy": 0.2576039180159569,
      "num_tokens": 129423878.0,
      "step": 56430
    },
    {
      "entropy": 5.54182276725769,
      "epoch": 5.421229586935639,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002483966297832364,
      "loss": 4.6615,
      "mean_token_accuracy": 0.2577246353030205,
      "num_tokens": 129434272.0,
      "step": 56435
    },
    {
      "entropy": 5.576495218276977,
      "epoch": 5.4217098943323725,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024836258702282967,
      "loss": 4.709,
      "mean_token_accuracy": 0.25237604677677156,
      "num_tokens": 129445525.0,
      "step": 56440
    },
    {
      "entropy": 5.573019981384277,
      "epoch": 5.422190201729107,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002483285448808644,
      "loss": 4.7167,
      "mean_token_accuracy": 0.2546921044588089,
      "num_tokens": 129458145.0,
      "step": 56445
    },
    {
      "entropy": 5.576039218902588,
      "epoch": 5.42267050912584,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002482945033581312,
      "loss": 4.6464,
      "mean_token_accuracy": 0.2551166623830795,
      "num_tokens": 129469061.0,
      "step": 56450
    },
    {
      "entropy": 5.498339223861694,
      "epoch": 5.423150816522575,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002482604624554203,
      "loss": 4.6387,
      "mean_token_accuracy": 0.26811501681804656,
      "num_tokens": 129479696.0,
      "step": 56455
    },
    {
      "entropy": 5.482086801528931,
      "epoch": 5.423631123919308,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024822642217352204,
      "loss": 4.6464,
      "mean_token_accuracy": 0.26193323731422424,
      "num_tokens": 129492094.0,
      "step": 56460
    },
    {
      "entropy": 5.516125154495239,
      "epoch": 5.424111431316042,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024819238251322677,
      "loss": 4.6411,
      "mean_token_accuracy": 0.2539446547627449,
      "num_tokens": 129502484.0,
      "step": 56465
    },
    {
      "entropy": 5.509613037109375,
      "epoch": 5.424591738712776,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024815834347532496,
      "loss": 4.6176,
      "mean_token_accuracy": 0.26600011587142947,
      "num_tokens": 129514251.0,
      "step": 56470
    },
    {
      "entropy": 5.515117025375366,
      "epoch": 5.42507204610951,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002481243050606065,
      "loss": 4.6102,
      "mean_token_accuracy": 0.27263174653053285,
      "num_tokens": 129526283.0,
      "step": 56475
    },
    {
      "entropy": 5.509347200393677,
      "epoch": 5.425552353506244,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024809026726986197,
      "loss": 4.6823,
      "mean_token_accuracy": 0.2584213107824326,
      "num_tokens": 129539440.0,
      "step": 56480
    },
    {
      "entropy": 5.541688776016235,
      "epoch": 5.426032660902978,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024805623010388145,
      "loss": 4.6596,
      "mean_token_accuracy": 0.2673891007900238,
      "num_tokens": 129551244.0,
      "step": 56485
    },
    {
      "entropy": 5.633539342880249,
      "epoch": 5.426512968299712,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002480221935634553,
      "loss": 4.7618,
      "mean_token_accuracy": 0.25096030384302137,
      "num_tokens": 129563269.0,
      "step": 56490
    },
    {
      "entropy": 5.5140989303588865,
      "epoch": 5.426993275696446,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002479881576493737,
      "loss": 4.6326,
      "mean_token_accuracy": 0.2605672538280487,
      "num_tokens": 129574905.0,
      "step": 56495
    },
    {
      "entropy": 5.551241207122803,
      "epoch": 5.42747358309318,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024795412236242693,
      "loss": 4.6839,
      "mean_token_accuracy": 0.25998460352420805,
      "num_tokens": 129585849.0,
      "step": 56500
    },
    {
      "entropy": 5.53188271522522,
      "epoch": 5.427953890489913,
      "grad_norm": 1.5,
      "learning_rate": 0.0002479200877034051,
      "loss": 4.687,
      "mean_token_accuracy": 0.26355256587266923,
      "num_tokens": 129596668.0,
      "step": 56505
    },
    {
      "entropy": 5.552738618850708,
      "epoch": 5.428434197886648,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024788605367309836,
      "loss": 4.6829,
      "mean_token_accuracy": 0.26246472746133803,
      "num_tokens": 129607508.0,
      "step": 56510
    },
    {
      "entropy": 5.4738280296325685,
      "epoch": 5.428914505283381,
      "grad_norm": 1.40625,
      "learning_rate": 0.000247852020272297,
      "loss": 4.5645,
      "mean_token_accuracy": 0.2696376979351044,
      "num_tokens": 129618946.0,
      "step": 56515
    },
    {
      "entropy": 5.564498329162598,
      "epoch": 5.4293948126801155,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000247817987501791,
      "loss": 4.6261,
      "mean_token_accuracy": 0.26321070045232775,
      "num_tokens": 129628770.0,
      "step": 56520
    },
    {
      "entropy": 5.631966400146484,
      "epoch": 5.429875120076849,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024778395536237077,
      "loss": 4.7602,
      "mean_token_accuracy": 0.2508438125252724,
      "num_tokens": 129639617.0,
      "step": 56525
    },
    {
      "entropy": 5.482368040084839,
      "epoch": 5.430355427473583,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024774992385482626,
      "loss": 4.5648,
      "mean_token_accuracy": 0.2737558990716934,
      "num_tokens": 129651626.0,
      "step": 56530
    },
    {
      "entropy": 5.615383815765381,
      "epoch": 5.430835734870317,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024771589297994755,
      "loss": 4.7407,
      "mean_token_accuracy": 0.2556650832295418,
      "num_tokens": 129662103.0,
      "step": 56535
    },
    {
      "entropy": 5.4628197193145756,
      "epoch": 5.431316042267051,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002476818627385248,
      "loss": 4.5526,
      "mean_token_accuracy": 0.26059342473745345,
      "num_tokens": 129674730.0,
      "step": 56540
    },
    {
      "entropy": 5.513824844360352,
      "epoch": 5.431796349663784,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000247647833131348,
      "loss": 4.6199,
      "mean_token_accuracy": 0.2677641972899437,
      "num_tokens": 129686218.0,
      "step": 56545
    },
    {
      "entropy": 5.549482679367065,
      "epoch": 5.432276657060519,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002476138041592074,
      "loss": 4.6995,
      "mean_token_accuracy": 0.25563000291585924,
      "num_tokens": 129697303.0,
      "step": 56550
    },
    {
      "entropy": 5.503363561630249,
      "epoch": 5.432756964457253,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024757977582289296,
      "loss": 4.5898,
      "mean_token_accuracy": 0.2635250985622406,
      "num_tokens": 129706351.0,
      "step": 56555
    },
    {
      "entropy": 5.511469841003418,
      "epoch": 5.4332372718539865,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002475457481231947,
      "loss": 4.6895,
      "mean_token_accuracy": 0.2644759863615036,
      "num_tokens": 129718949.0,
      "step": 56560
    },
    {
      "entropy": 5.591460466384888,
      "epoch": 5.433717579250721,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002475117210609026,
      "loss": 4.675,
      "mean_token_accuracy": 0.2566027492284775,
      "num_tokens": 129730710.0,
      "step": 56565
    },
    {
      "entropy": 5.584670162200927,
      "epoch": 5.434197886647454,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024747769463680675,
      "loss": 4.6822,
      "mean_token_accuracy": 0.26027259826660154,
      "num_tokens": 129743065.0,
      "step": 56570
    },
    {
      "entropy": 5.467707872390747,
      "epoch": 5.434678194044189,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024744366885169703,
      "loss": 4.6288,
      "mean_token_accuracy": 0.2663594365119934,
      "num_tokens": 129754714.0,
      "step": 56575
    },
    {
      "entropy": 5.5676435947418215,
      "epoch": 5.435158501440922,
      "grad_norm": 1.375,
      "learning_rate": 0.0002474096437063636,
      "loss": 4.7103,
      "mean_token_accuracy": 0.26029509752988816,
      "num_tokens": 129766106.0,
      "step": 56580
    },
    {
      "entropy": 5.519738578796387,
      "epoch": 5.435638808837656,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024737561920159623,
      "loss": 4.6926,
      "mean_token_accuracy": 0.25863988399505616,
      "num_tokens": 129777992.0,
      "step": 56585
    },
    {
      "entropy": 5.606692600250244,
      "epoch": 5.43611911623439,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000247341595338185,
      "loss": 4.7515,
      "mean_token_accuracy": 0.2478460431098938,
      "num_tokens": 129789568.0,
      "step": 56590
    },
    {
      "entropy": 5.521671533584595,
      "epoch": 5.436599423631124,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002473075721169198,
      "loss": 4.6434,
      "mean_token_accuracy": 0.26511354744434357,
      "num_tokens": 129801201.0,
      "step": 56595
    },
    {
      "entropy": 5.50982174873352,
      "epoch": 5.437079731027858,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002472735495385905,
      "loss": 4.5809,
      "mean_token_accuracy": 0.26335916072130205,
      "num_tokens": 129811953.0,
      "step": 56600
    },
    {
      "entropy": 5.5868841171264645,
      "epoch": 5.437560038424592,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024723952760398715,
      "loss": 4.7747,
      "mean_token_accuracy": 0.2512345850467682,
      "num_tokens": 129822733.0,
      "step": 56605
    },
    {
      "entropy": 5.587636280059814,
      "epoch": 5.438040345821325,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002472055063138994,
      "loss": 4.6801,
      "mean_token_accuracy": 0.2658589169383049,
      "num_tokens": 129834257.0,
      "step": 56610
    },
    {
      "entropy": 5.502604722976685,
      "epoch": 5.43852065321806,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002471714856691174,
      "loss": 4.6149,
      "mean_token_accuracy": 0.26036375761032104,
      "num_tokens": 129845217.0,
      "step": 56615
    },
    {
      "entropy": 5.513389778137207,
      "epoch": 5.439000960614793,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002471374656704308,
      "loss": 4.6008,
      "mean_token_accuracy": 0.27019681185483935,
      "num_tokens": 129857083.0,
      "step": 56620
    },
    {
      "entropy": 5.519506597518921,
      "epoch": 5.439481268011527,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024710344631862944,
      "loss": 4.6647,
      "mean_token_accuracy": 0.2602595627307892,
      "num_tokens": 129868658.0,
      "step": 56625
    },
    {
      "entropy": 5.410558128356934,
      "epoch": 5.439961575408262,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024706942761450337,
      "loss": 4.5095,
      "mean_token_accuracy": 0.27553085088729856,
      "num_tokens": 129880027.0,
      "step": 56630
    },
    {
      "entropy": 5.571093463897705,
      "epoch": 5.440441882804995,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024703540955884213,
      "loss": 4.7481,
      "mean_token_accuracy": 0.2522931307554245,
      "num_tokens": 129890375.0,
      "step": 56635
    },
    {
      "entropy": 5.5419422626495365,
      "epoch": 5.4409221902017295,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002470013921524357,
      "loss": 4.6857,
      "mean_token_accuracy": 0.2576116755604744,
      "num_tokens": 129902382.0,
      "step": 56640
    },
    {
      "entropy": 5.458096313476562,
      "epoch": 5.441402497598463,
      "grad_norm": 1.375,
      "learning_rate": 0.0002469673753960738,
      "loss": 4.5818,
      "mean_token_accuracy": 0.2695191249251366,
      "num_tokens": 129913825.0,
      "step": 56645
    },
    {
      "entropy": 5.5955267429351805,
      "epoch": 5.441882804995197,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002469333592905462,
      "loss": 4.7325,
      "mean_token_accuracy": 0.2514244943857193,
      "num_tokens": 129927427.0,
      "step": 56650
    },
    {
      "entropy": 5.642384386062622,
      "epoch": 5.442363112391931,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002468993438366427,
      "loss": 4.7232,
      "mean_token_accuracy": 0.26040330827236174,
      "num_tokens": 129938884.0,
      "step": 56655
    },
    {
      "entropy": 5.596259641647339,
      "epoch": 5.442843419788665,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002468653290351529,
      "loss": 4.6959,
      "mean_token_accuracy": 0.25573405623435974,
      "num_tokens": 129949935.0,
      "step": 56660
    },
    {
      "entropy": 5.502532482147217,
      "epoch": 5.4433237271853985,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024683131488686667,
      "loss": 4.6482,
      "mean_token_accuracy": 0.2704387068748474,
      "num_tokens": 129961816.0,
      "step": 56665
    },
    {
      "entropy": 5.462919282913208,
      "epoch": 5.443804034582133,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024679730139257365,
      "loss": 4.598,
      "mean_token_accuracy": 0.26438137888908386,
      "num_tokens": 129973953.0,
      "step": 56670
    },
    {
      "entropy": 5.51595139503479,
      "epoch": 5.444284341978866,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024676328855306355,
      "loss": 4.6818,
      "mean_token_accuracy": 0.2590455174446106,
      "num_tokens": 129985457.0,
      "step": 56675
    },
    {
      "entropy": 5.54557237625122,
      "epoch": 5.444764649375601,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002467292763691261,
      "loss": 4.5717,
      "mean_token_accuracy": 0.2694111347198486,
      "num_tokens": 129996346.0,
      "step": 56680
    },
    {
      "entropy": 5.548817920684814,
      "epoch": 5.445244956772334,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002466952648415509,
      "loss": 4.7011,
      "mean_token_accuracy": 0.25775333642959597,
      "num_tokens": 130008364.0,
      "step": 56685
    },
    {
      "entropy": 5.547415828704834,
      "epoch": 5.445725264169068,
      "grad_norm": 1.375,
      "learning_rate": 0.0002466612539711275,
      "loss": 4.6061,
      "mean_token_accuracy": 0.2646182537078857,
      "num_tokens": 130019040.0,
      "step": 56690
    },
    {
      "entropy": 5.571617603302002,
      "epoch": 5.446205571565802,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002466272437586457,
      "loss": 4.7354,
      "mean_token_accuracy": 0.25241724997758863,
      "num_tokens": 130031457.0,
      "step": 56695
    },
    {
      "entropy": 5.487679672241211,
      "epoch": 5.446685878962536,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000246593234204895,
      "loss": 4.5914,
      "mean_token_accuracy": 0.26557876020669935,
      "num_tokens": 130044049.0,
      "step": 56700
    },
    {
      "entropy": 5.580559921264649,
      "epoch": 5.4471661863592695,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002465592253106651,
      "loss": 4.6979,
      "mean_token_accuracy": 0.2602560594677925,
      "num_tokens": 130054636.0,
      "step": 56705
    },
    {
      "entropy": 5.561609268188477,
      "epoch": 5.447646493756004,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002465252170767456,
      "loss": 4.6813,
      "mean_token_accuracy": 0.26064145416021345,
      "num_tokens": 130064998.0,
      "step": 56710
    },
    {
      "entropy": 5.5325140953063965,
      "epoch": 5.448126801152737,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000246491209503926,
      "loss": 4.6502,
      "mean_token_accuracy": 0.255254927277565,
      "num_tokens": 130077073.0,
      "step": 56715
    },
    {
      "entropy": 5.555516386032105,
      "epoch": 5.448607108549472,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002464572025929957,
      "loss": 4.7145,
      "mean_token_accuracy": 0.2517714723944664,
      "num_tokens": 130087789.0,
      "step": 56720
    },
    {
      "entropy": 5.5960205078125,
      "epoch": 5.449087415946206,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002464231963447445,
      "loss": 4.7303,
      "mean_token_accuracy": 0.2594448909163475,
      "num_tokens": 130100800.0,
      "step": 56725
    },
    {
      "entropy": 5.6040455341339115,
      "epoch": 5.449567723342939,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002463891907599618,
      "loss": 4.749,
      "mean_token_accuracy": 0.25280977189540865,
      "num_tokens": 130111444.0,
      "step": 56730
    },
    {
      "entropy": 5.633148336410523,
      "epoch": 5.450048030739674,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002463551858394372,
      "loss": 4.7601,
      "mean_token_accuracy": 0.2588896930217743,
      "num_tokens": 130122695.0,
      "step": 56735
    },
    {
      "entropy": 5.490234851837158,
      "epoch": 5.450528338136407,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024632118158396014,
      "loss": 4.5584,
      "mean_token_accuracy": 0.2769703805446625,
      "num_tokens": 130135390.0,
      "step": 56740
    },
    {
      "entropy": 5.626525497436523,
      "epoch": 5.4510086455331415,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024628717799432006,
      "loss": 4.7631,
      "mean_token_accuracy": 0.25725177228450774,
      "num_tokens": 130147141.0,
      "step": 56745
    },
    {
      "entropy": 5.552938842773438,
      "epoch": 5.451488952929875,
      "grad_norm": 1.25,
      "learning_rate": 0.0002462531750713064,
      "loss": 4.6756,
      "mean_token_accuracy": 0.2589268460869789,
      "num_tokens": 130158745.0,
      "step": 56750
    },
    {
      "entropy": 5.554532289505005,
      "epoch": 5.451969260326609,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002462191728157086,
      "loss": 4.6963,
      "mean_token_accuracy": 0.25738777965307236,
      "num_tokens": 130170065.0,
      "step": 56755
    },
    {
      "entropy": 5.5509268283844,
      "epoch": 5.452449567723343,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024618517122831625,
      "loss": 4.6479,
      "mean_token_accuracy": 0.2600059360265732,
      "num_tokens": 130182278.0,
      "step": 56760
    },
    {
      "entropy": 5.520087242126465,
      "epoch": 5.452929875120077,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024615117030991863,
      "loss": 4.6201,
      "mean_token_accuracy": 0.2688250750303268,
      "num_tokens": 130192867.0,
      "step": 56765
    },
    {
      "entropy": 5.480492496490479,
      "epoch": 5.45341018251681,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002461171700613052,
      "loss": 4.6315,
      "mean_token_accuracy": 0.2637235924601555,
      "num_tokens": 130203380.0,
      "step": 56770
    },
    {
      "entropy": 5.63015079498291,
      "epoch": 5.453890489913545,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024608317048326533,
      "loss": 4.7942,
      "mean_token_accuracy": 0.253578907251358,
      "num_tokens": 130214769.0,
      "step": 56775
    },
    {
      "entropy": 5.526625108718872,
      "epoch": 5.454370797310278,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002460491715765883,
      "loss": 4.619,
      "mean_token_accuracy": 0.26503183245658873,
      "num_tokens": 130226158.0,
      "step": 56780
    },
    {
      "entropy": 5.609940052032471,
      "epoch": 5.4548511047070125,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024601517334206356,
      "loss": 4.7281,
      "mean_token_accuracy": 0.2537651792168617,
      "num_tokens": 130237233.0,
      "step": 56785
    },
    {
      "entropy": 5.521343088150024,
      "epoch": 5.455331412103746,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024598117578048044,
      "loss": 4.6392,
      "mean_token_accuracy": 0.2627836734056473,
      "num_tokens": 130248533.0,
      "step": 56790
    },
    {
      "entropy": 5.5648904800415036,
      "epoch": 5.45581171950048,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024594717889262824,
      "loss": 4.6725,
      "mean_token_accuracy": 0.26618797779083253,
      "num_tokens": 130260323.0,
      "step": 56795
    },
    {
      "entropy": 5.617459106445312,
      "epoch": 5.456292026897215,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024591318267929633,
      "loss": 4.794,
      "mean_token_accuracy": 0.25159071385860443,
      "num_tokens": 130271788.0,
      "step": 56800
    },
    {
      "entropy": 5.546799516677856,
      "epoch": 5.456772334293948,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024587918714127395,
      "loss": 4.6881,
      "mean_token_accuracy": 0.25652709752321246,
      "num_tokens": 130282877.0,
      "step": 56805
    },
    {
      "entropy": 5.566553115844727,
      "epoch": 5.457252641690682,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024584519227935033,
      "loss": 4.6808,
      "mean_token_accuracy": 0.26165701597929003,
      "num_tokens": 130295543.0,
      "step": 56810
    },
    {
      "entropy": 5.476506662368775,
      "epoch": 5.457732949087416,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024581119809431474,
      "loss": 4.6119,
      "mean_token_accuracy": 0.2711492940783501,
      "num_tokens": 130307677.0,
      "step": 56815
    },
    {
      "entropy": 5.520495271682739,
      "epoch": 5.45821325648415,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002457772045869565,
      "loss": 4.671,
      "mean_token_accuracy": 0.2634625717997551,
      "num_tokens": 130319326.0,
      "step": 56820
    },
    {
      "entropy": 5.494034481048584,
      "epoch": 5.458693563880884,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024574321175806473,
      "loss": 4.5872,
      "mean_token_accuracy": 0.2630060136318207,
      "num_tokens": 130329670.0,
      "step": 56825
    },
    {
      "entropy": 5.570886087417603,
      "epoch": 5.459173871277618,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024570921960842887,
      "loss": 4.6558,
      "mean_token_accuracy": 0.25651742368936536,
      "num_tokens": 130340481.0,
      "step": 56830
    },
    {
      "entropy": 5.518858623504639,
      "epoch": 5.459654178674351,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024567522813883784,
      "loss": 4.6649,
      "mean_token_accuracy": 0.25617583841085434,
      "num_tokens": 130352684.0,
      "step": 56835
    },
    {
      "entropy": 5.546882915496826,
      "epoch": 5.460134486071086,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024564123735008094,
      "loss": 4.6824,
      "mean_token_accuracy": 0.25774860084056855,
      "num_tokens": 130364407.0,
      "step": 56840
    },
    {
      "entropy": 5.539954090118409,
      "epoch": 5.460614793467819,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002456072472429473,
      "loss": 4.6228,
      "mean_token_accuracy": 0.2637721002101898,
      "num_tokens": 130375971.0,
      "step": 56845
    },
    {
      "entropy": 5.494595050811768,
      "epoch": 5.461095100864553,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024557325781822615,
      "loss": 4.5715,
      "mean_token_accuracy": 0.2683729037642479,
      "num_tokens": 130387794.0,
      "step": 56850
    },
    {
      "entropy": 5.559223175048828,
      "epoch": 5.461575408261287,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002455392690767065,
      "loss": 4.7428,
      "mean_token_accuracy": 0.2561682850122452,
      "num_tokens": 130398957.0,
      "step": 56855
    },
    {
      "entropy": 5.516384887695312,
      "epoch": 5.462055715658021,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002455052810191777,
      "loss": 4.6701,
      "mean_token_accuracy": 0.2532560870051384,
      "num_tokens": 130409058.0,
      "step": 56860
    },
    {
      "entropy": 5.537959098815918,
      "epoch": 5.462536023054755,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024547129364642855,
      "loss": 4.7133,
      "mean_token_accuracy": 0.2600004807114601,
      "num_tokens": 130420221.0,
      "step": 56865
    },
    {
      "entropy": 5.570669507980346,
      "epoch": 5.463016330451489,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002454373069592483,
      "loss": 4.6839,
      "mean_token_accuracy": 0.26110884100198745,
      "num_tokens": 130433191.0,
      "step": 56870
    },
    {
      "entropy": 5.450593185424805,
      "epoch": 5.463496637848223,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002454033209584261,
      "loss": 4.5917,
      "mean_token_accuracy": 0.26523403972387316,
      "num_tokens": 130445804.0,
      "step": 56875
    },
    {
      "entropy": 5.582517004013061,
      "epoch": 5.463976945244957,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002453693356447508,
      "loss": 4.7394,
      "mean_token_accuracy": 0.25770522505044935,
      "num_tokens": 130457534.0,
      "step": 56880
    },
    {
      "entropy": 5.507993745803833,
      "epoch": 5.464457252641691,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024533535101901155,
      "loss": 4.6452,
      "mean_token_accuracy": 0.2624885901808739,
      "num_tokens": 130468747.0,
      "step": 56885
    },
    {
      "entropy": 5.571642637252808,
      "epoch": 5.4649375600384245,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024530136708199746,
      "loss": 4.6722,
      "mean_token_accuracy": 0.2566432893276215,
      "num_tokens": 130480168.0,
      "step": 56890
    },
    {
      "entropy": 5.593768119812012,
      "epoch": 5.465417867435159,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002452673838344973,
      "loss": 4.6968,
      "mean_token_accuracy": 0.25993430316448213,
      "num_tokens": 130492010.0,
      "step": 56895
    },
    {
      "entropy": 5.481921052932739,
      "epoch": 5.465898174831892,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024523340127730026,
      "loss": 4.6101,
      "mean_token_accuracy": 0.2620413824915886,
      "num_tokens": 130504733.0,
      "step": 56900
    },
    {
      "entropy": 5.551282453536987,
      "epoch": 5.4663784822286265,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024519941941119526,
      "loss": 4.6639,
      "mean_token_accuracy": 0.26524169445037843,
      "num_tokens": 130516142.0,
      "step": 56905
    },
    {
      "entropy": 5.626088047027588,
      "epoch": 5.46685878962536,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002451654382369712,
      "loss": 4.7583,
      "mean_token_accuracy": 0.259819883108139,
      "num_tokens": 130527160.0,
      "step": 56910
    },
    {
      "entropy": 5.474025058746338,
      "epoch": 5.467339097022094,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024513145775541717,
      "loss": 4.5425,
      "mean_token_accuracy": 0.27762746810913086,
      "num_tokens": 130538705.0,
      "step": 56915
    },
    {
      "entropy": 5.498658800125122,
      "epoch": 5.467819404418828,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000245097477967322,
      "loss": 4.6162,
      "mean_token_accuracy": 0.26338459849357604,
      "num_tokens": 130549863.0,
      "step": 56920
    },
    {
      "entropy": 5.527609729766846,
      "epoch": 5.468299711815562,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024506349887347455,
      "loss": 4.6773,
      "mean_token_accuracy": 0.2595185309648514,
      "num_tokens": 130561237.0,
      "step": 56925
    },
    {
      "entropy": 5.5529216766357425,
      "epoch": 5.4687800192122955,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024502952047466376,
      "loss": 4.6765,
      "mean_token_accuracy": 0.264561365544796,
      "num_tokens": 130574601.0,
      "step": 56930
    },
    {
      "entropy": 5.580254173278808,
      "epoch": 5.46926032660903,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024499554277167854,
      "loss": 4.7058,
      "mean_token_accuracy": 0.259328792989254,
      "num_tokens": 130586110.0,
      "step": 56935
    },
    {
      "entropy": 5.593357229232788,
      "epoch": 5.469740634005763,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002449615657653076,
      "loss": 4.704,
      "mean_token_accuracy": 0.25529026836156843,
      "num_tokens": 130597443.0,
      "step": 56940
    },
    {
      "entropy": 5.562255859375,
      "epoch": 5.470220941402498,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024492758945634,
      "loss": 4.777,
      "mean_token_accuracy": 0.24873481690883636,
      "num_tokens": 130610322.0,
      "step": 56945
    },
    {
      "entropy": 5.58198094367981,
      "epoch": 5.470701248799232,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002448936138455646,
      "loss": 4.7323,
      "mean_token_accuracy": 0.2588675573468208,
      "num_tokens": 130621354.0,
      "step": 56950
    },
    {
      "entropy": 5.617544555664063,
      "epoch": 5.471181556195965,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024485963893376993,
      "loss": 4.683,
      "mean_token_accuracy": 0.26295955032110213,
      "num_tokens": 130631555.0,
      "step": 56955
    },
    {
      "entropy": 5.5925758361816404,
      "epoch": 5.4716618635927,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000244825664721745,
      "loss": 4.7031,
      "mean_token_accuracy": 0.253360290825367,
      "num_tokens": 130641644.0,
      "step": 56960
    },
    {
      "entropy": 5.477405738830567,
      "epoch": 5.472142170989433,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002447916912102785,
      "loss": 4.6408,
      "mean_token_accuracy": 0.26500210613012315,
      "num_tokens": 130653246.0,
      "step": 56965
    },
    {
      "entropy": 5.498541784286499,
      "epoch": 5.472622478386167,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024475771840015925,
      "loss": 4.6309,
      "mean_token_accuracy": 0.2634164109826088,
      "num_tokens": 130664211.0,
      "step": 56970
    },
    {
      "entropy": 5.550621461868286,
      "epoch": 5.473102785782901,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000244723746292176,
      "loss": 4.6509,
      "mean_token_accuracy": 0.2607397362589836,
      "num_tokens": 130676098.0,
      "step": 56975
    },
    {
      "entropy": 5.587848758697509,
      "epoch": 5.473583093179635,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002446897748871175,
      "loss": 4.6913,
      "mean_token_accuracy": 0.2558111369609833,
      "num_tokens": 130686715.0,
      "step": 56980
    },
    {
      "entropy": 5.531907415390014,
      "epoch": 5.474063400576369,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002446558041857723,
      "loss": 4.6469,
      "mean_token_accuracy": 0.2619988411664963,
      "num_tokens": 130699247.0,
      "step": 56985
    },
    {
      "entropy": 5.501251602172852,
      "epoch": 5.474543707973103,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002446218341889293,
      "loss": 4.6285,
      "mean_token_accuracy": 0.26478763669729233,
      "num_tokens": 130711105.0,
      "step": 56990
    },
    {
      "entropy": 5.523826456069946,
      "epoch": 5.475024015369836,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002445878648973771,
      "loss": 4.64,
      "mean_token_accuracy": 0.2623570591211319,
      "num_tokens": 130722261.0,
      "step": 56995
    },
    {
      "entropy": 5.4866351127624515,
      "epoch": 5.475504322766571,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002445538963119043,
      "loss": 4.5689,
      "mean_token_accuracy": 0.2692617267370224,
      "num_tokens": 130734411.0,
      "step": 57000
    },
    {
      "epoch": 5.475504322766571,
      "eval_entropy": 5.215512507604657,
      "eval_loss": 4.765960693359375,
      "eval_mean_token_accuracy": 0.26356647907870245,
      "eval_num_tokens": 130734411.0,
      "eval_runtime": 20.6395,
      "eval_samples_per_second": 1589.916,
      "eval_steps_per_second": 198.746,
      "step": 57000
    },
    {
      "entropy": 5.523036956787109,
      "epoch": 5.475984630163304,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024451992843329976,
      "loss": 4.6746,
      "mean_token_accuracy": 0.2628266721963882,
      "num_tokens": 130746564.0,
      "step": 57005
    },
    {
      "entropy": 5.6342339515686035,
      "epoch": 5.4764649375600385,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002444859612623519,
      "loss": 4.794,
      "mean_token_accuracy": 0.24799077659845353,
      "num_tokens": 130756490.0,
      "step": 57010
    },
    {
      "entropy": 5.494487857818603,
      "epoch": 5.476945244956772,
      "grad_norm": 1.375,
      "learning_rate": 0.0002444519947998494,
      "loss": 4.6436,
      "mean_token_accuracy": 0.2604633063077927,
      "num_tokens": 130768766.0,
      "step": 57015
    },
    {
      "entropy": 5.547615242004395,
      "epoch": 5.477425552353506,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002444180290465809,
      "loss": 4.6677,
      "mean_token_accuracy": 0.26478780508041383,
      "num_tokens": 130779370.0,
      "step": 57020
    },
    {
      "entropy": 5.590169239044189,
      "epoch": 5.477905859750241,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002443840640033349,
      "loss": 4.6613,
      "mean_token_accuracy": 0.2601738780736923,
      "num_tokens": 130790227.0,
      "step": 57025
    },
    {
      "entropy": 5.6144123554229735,
      "epoch": 5.478386167146974,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024435009967090006,
      "loss": 4.7327,
      "mean_token_accuracy": 0.25722897946834566,
      "num_tokens": 130800073.0,
      "step": 57030
    },
    {
      "entropy": 5.556662940979004,
      "epoch": 5.478866474543708,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002443161360500649,
      "loss": 4.7708,
      "mean_token_accuracy": 0.2546671137213707,
      "num_tokens": 130811621.0,
      "step": 57035
    },
    {
      "entropy": 5.471826028823853,
      "epoch": 5.479346781940442,
      "grad_norm": 1.359375,
      "learning_rate": 0.000244282173141618,
      "loss": 4.6089,
      "mean_token_accuracy": 0.26939199268817904,
      "num_tokens": 130823625.0,
      "step": 57040
    },
    {
      "entropy": 5.544023084640503,
      "epoch": 5.479827089337176,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024424821094634776,
      "loss": 4.6469,
      "mean_token_accuracy": 0.2559865191578865,
      "num_tokens": 130834956.0,
      "step": 57045
    },
    {
      "entropy": 5.524384212493897,
      "epoch": 5.4803073967339095,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024421424946504273,
      "loss": 4.5681,
      "mean_token_accuracy": 0.2692248895764351,
      "num_tokens": 130846955.0,
      "step": 57050
    },
    {
      "entropy": 5.585924959182739,
      "epoch": 5.480787704130644,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002441802886984915,
      "loss": 4.6871,
      "mean_token_accuracy": 0.25811772495508195,
      "num_tokens": 130858660.0,
      "step": 57055
    },
    {
      "entropy": 5.522906303405762,
      "epoch": 5.481268011527377,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002441463286474823,
      "loss": 4.6839,
      "mean_token_accuracy": 0.2613209620118141,
      "num_tokens": 130870684.0,
      "step": 57060
    },
    {
      "entropy": 5.448321676254272,
      "epoch": 5.481748318924112,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024411236931280383,
      "loss": 4.5304,
      "mean_token_accuracy": 0.27332779169082644,
      "num_tokens": 130881999.0,
      "step": 57065
    },
    {
      "entropy": 5.593876361846924,
      "epoch": 5.482228626320845,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002440784106952445,
      "loss": 4.7592,
      "mean_token_accuracy": 0.2538190811872482,
      "num_tokens": 130892942.0,
      "step": 57070
    },
    {
      "entropy": 5.5780401706695555,
      "epoch": 5.482708933717579,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024404445279559256,
      "loss": 4.7323,
      "mean_token_accuracy": 0.24856967329978943,
      "num_tokens": 130905733.0,
      "step": 57075
    },
    {
      "entropy": 5.457538223266601,
      "epoch": 5.483189241114313,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002440104956146365,
      "loss": 4.5607,
      "mean_token_accuracy": 0.26915186196565627,
      "num_tokens": 130917974.0,
      "step": 57080
    },
    {
      "entropy": 5.5099938869476315,
      "epoch": 5.483669548511047,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002439765391531647,
      "loss": 4.643,
      "mean_token_accuracy": 0.2589855507016182,
      "num_tokens": 130929210.0,
      "step": 57085
    },
    {
      "entropy": 5.635993814468383,
      "epoch": 5.484149855907781,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024394258341196558,
      "loss": 4.7267,
      "mean_token_accuracy": 0.2561116024851799,
      "num_tokens": 130939296.0,
      "step": 57090
    },
    {
      "entropy": 5.50272798538208,
      "epoch": 5.484630163304515,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024390862839182743,
      "loss": 4.6259,
      "mean_token_accuracy": 0.26199691742658615,
      "num_tokens": 130951239.0,
      "step": 57095
    },
    {
      "entropy": 5.602286529541016,
      "epoch": 5.485110470701249,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024387467409353876,
      "loss": 4.7341,
      "mean_token_accuracy": 0.2525560989975929,
      "num_tokens": 130962765.0,
      "step": 57100
    },
    {
      "entropy": 5.556296730041504,
      "epoch": 5.485590778097983,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024384072051788754,
      "loss": 4.6963,
      "mean_token_accuracy": 0.2578822076320648,
      "num_tokens": 130974481.0,
      "step": 57105
    },
    {
      "entropy": 5.524720907211304,
      "epoch": 5.486071085494717,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002438067676656623,
      "loss": 4.6252,
      "mean_token_accuracy": 0.25913084447383883,
      "num_tokens": 130986603.0,
      "step": 57110
    },
    {
      "entropy": 5.500079488754272,
      "epoch": 5.48655139289145,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024377281553765128,
      "loss": 4.6309,
      "mean_token_accuracy": 0.2670019015669823,
      "num_tokens": 130997872.0,
      "step": 57115
    },
    {
      "entropy": 5.502181148529052,
      "epoch": 5.487031700288185,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002437388641346428,
      "loss": 4.6362,
      "mean_token_accuracy": 0.26211037784814833,
      "num_tokens": 131009651.0,
      "step": 57120
    },
    {
      "entropy": 5.568862104415894,
      "epoch": 5.487512007684918,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024370491345742507,
      "loss": 4.7144,
      "mean_token_accuracy": 0.2521620526909828,
      "num_tokens": 131021304.0,
      "step": 57125
    },
    {
      "entropy": 5.516410636901855,
      "epoch": 5.4879923150816525,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002436709635067863,
      "loss": 4.6407,
      "mean_token_accuracy": 0.26186486780643464,
      "num_tokens": 131032860.0,
      "step": 57130
    },
    {
      "entropy": 5.566761207580567,
      "epoch": 5.488472622478386,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024363701428351474,
      "loss": 4.7019,
      "mean_token_accuracy": 0.2541632205247879,
      "num_tokens": 131044973.0,
      "step": 57135
    },
    {
      "entropy": 5.633077096939087,
      "epoch": 5.48895292987512,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024360306578839863,
      "loss": 4.8041,
      "mean_token_accuracy": 0.2477980598807335,
      "num_tokens": 131057796.0,
      "step": 57140
    },
    {
      "entropy": 5.520298814773559,
      "epoch": 5.489433237271854,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024356911802222604,
      "loss": 4.5922,
      "mean_token_accuracy": 0.26488053798675537,
      "num_tokens": 131069551.0,
      "step": 57145
    },
    {
      "entropy": 5.498936462402344,
      "epoch": 5.489913544668588,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002435351709857852,
      "loss": 4.7076,
      "mean_token_accuracy": 0.2608163461089134,
      "num_tokens": 131081613.0,
      "step": 57150
    },
    {
      "entropy": 5.5752012729644775,
      "epoch": 5.4903938520653215,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024350122467986423,
      "loss": 4.7086,
      "mean_token_accuracy": 0.25999532639980316,
      "num_tokens": 131094659.0,
      "step": 57155
    },
    {
      "entropy": 5.565412712097168,
      "epoch": 5.490874159462056,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002434672791052514,
      "loss": 4.5685,
      "mean_token_accuracy": 0.2670077860355377,
      "num_tokens": 131105784.0,
      "step": 57160
    },
    {
      "entropy": 5.456490707397461,
      "epoch": 5.491354466858789,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002434333342627346,
      "loss": 4.5222,
      "mean_token_accuracy": 0.2715805396437645,
      "num_tokens": 131117807.0,
      "step": 57165
    },
    {
      "entropy": 5.570270919799805,
      "epoch": 5.491834774255524,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002433993901531021,
      "loss": 4.7427,
      "mean_token_accuracy": 0.2483867421746254,
      "num_tokens": 131129887.0,
      "step": 57170
    },
    {
      "entropy": 5.503546857833863,
      "epoch": 5.492315081652258,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002433654467771419,
      "loss": 4.6235,
      "mean_token_accuracy": 0.26273743212223055,
      "num_tokens": 131141743.0,
      "step": 57175
    },
    {
      "entropy": 5.508662557601928,
      "epoch": 5.492795389048991,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024333150413564215,
      "loss": 4.629,
      "mean_token_accuracy": 0.263763801753521,
      "num_tokens": 131152976.0,
      "step": 57180
    },
    {
      "entropy": 5.50236701965332,
      "epoch": 5.493275696445726,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024329756222939077,
      "loss": 4.6527,
      "mean_token_accuracy": 0.2584283635020256,
      "num_tokens": 131164816.0,
      "step": 57185
    },
    {
      "entropy": 5.434281921386718,
      "epoch": 5.493756003842459,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024326362105917592,
      "loss": 4.5872,
      "mean_token_accuracy": 0.2699295818805695,
      "num_tokens": 131176702.0,
      "step": 57190
    },
    {
      "entropy": 5.5519325733184814,
      "epoch": 5.494236311239193,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002432296806257855,
      "loss": 4.6501,
      "mean_token_accuracy": 0.26383000910282134,
      "num_tokens": 131188006.0,
      "step": 57195
    },
    {
      "entropy": 5.555857706069946,
      "epoch": 5.494716618635927,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024319574093000757,
      "loss": 4.6583,
      "mean_token_accuracy": 0.26339243948459623,
      "num_tokens": 131200680.0,
      "step": 57200
    },
    {
      "entropy": 5.509709453582763,
      "epoch": 5.495196926032661,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024316180197263005,
      "loss": 4.6674,
      "mean_token_accuracy": 0.2640952989459038,
      "num_tokens": 131211334.0,
      "step": 57205
    },
    {
      "entropy": 5.57027850151062,
      "epoch": 5.495677233429395,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000243127863754441,
      "loss": 4.7855,
      "mean_token_accuracy": 0.25323367267847063,
      "num_tokens": 131223092.0,
      "step": 57210
    },
    {
      "entropy": 5.525978755950928,
      "epoch": 5.496157540826129,
      "grad_norm": 1.375,
      "learning_rate": 0.00024309392627622827,
      "loss": 4.5431,
      "mean_token_accuracy": 0.27190871983766557,
      "num_tokens": 131232863.0,
      "step": 57215
    },
    {
      "entropy": 5.620137023925781,
      "epoch": 5.496637848222862,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024305998953878,
      "loss": 4.7066,
      "mean_token_accuracy": 0.2525646731257439,
      "num_tokens": 131243145.0,
      "step": 57220
    },
    {
      "entropy": 5.470514154434204,
      "epoch": 5.497118155619597,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024302605354288375,
      "loss": 4.6182,
      "mean_token_accuracy": 0.2638845399022102,
      "num_tokens": 131253394.0,
      "step": 57225
    },
    {
      "entropy": 5.434082555770874,
      "epoch": 5.49759846301633,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024299211828932767,
      "loss": 4.6148,
      "mean_token_accuracy": 0.2620638906955719,
      "num_tokens": 131265652.0,
      "step": 57230
    },
    {
      "entropy": 5.527689838409424,
      "epoch": 5.4980787704130645,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002429581837788995,
      "loss": 4.6761,
      "mean_token_accuracy": 0.258992138504982,
      "num_tokens": 131277492.0,
      "step": 57235
    },
    {
      "entropy": 5.621117401123047,
      "epoch": 5.498559077809798,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024292425001238717,
      "loss": 4.7597,
      "mean_token_accuracy": 0.2516088172793388,
      "num_tokens": 131288930.0,
      "step": 57240
    },
    {
      "entropy": 5.594272613525391,
      "epoch": 5.499039385206532,
      "grad_norm": 1.375,
      "learning_rate": 0.0002428903169905785,
      "loss": 4.7004,
      "mean_token_accuracy": 0.2568515732884407,
      "num_tokens": 131300903.0,
      "step": 57245
    },
    {
      "entropy": 5.628520393371582,
      "epoch": 5.499519692603266,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024285638471426146,
      "loss": 4.7019,
      "mean_token_accuracy": 0.2584541827440262,
      "num_tokens": 131311472.0,
      "step": 57250
    },
    {
      "entropy": 5.584710121154785,
      "epoch": 5.5,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002428224531842236,
      "loss": 4.6869,
      "mean_token_accuracy": 0.25689481049776075,
      "num_tokens": 131322455.0,
      "step": 57255
    },
    {
      "entropy": 5.594171094894409,
      "epoch": 5.500480307396733,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002427885224012528,
      "loss": 4.7614,
      "mean_token_accuracy": 0.2580018758773804,
      "num_tokens": 131334083.0,
      "step": 57260
    },
    {
      "entropy": 5.612411117553711,
      "epoch": 5.500960614793468,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024275459236613688,
      "loss": 4.7186,
      "mean_token_accuracy": 0.2541418015956879,
      "num_tokens": 131346415.0,
      "step": 57265
    },
    {
      "entropy": 5.526780319213867,
      "epoch": 5.501440922190202,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002427206630796636,
      "loss": 4.6343,
      "mean_token_accuracy": 0.2620985671877861,
      "num_tokens": 131358080.0,
      "step": 57270
    },
    {
      "entropy": 5.477666902542114,
      "epoch": 5.5019212295869355,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002426867345426207,
      "loss": 4.5707,
      "mean_token_accuracy": 0.26579258888959884,
      "num_tokens": 131369248.0,
      "step": 57275
    },
    {
      "entropy": 5.621822357177734,
      "epoch": 5.50240153698367,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024265280675579592,
      "loss": 4.7508,
      "mean_token_accuracy": 0.2565549984574318,
      "num_tokens": 131380529.0,
      "step": 57280
    },
    {
      "entropy": 5.501650142669678,
      "epoch": 5.502881844380403,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024261887971997682,
      "loss": 4.567,
      "mean_token_accuracy": 0.2717057019472122,
      "num_tokens": 131391729.0,
      "step": 57285
    },
    {
      "entropy": 5.514071750640869,
      "epoch": 5.503362151777138,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002425849534359512,
      "loss": 4.6706,
      "mean_token_accuracy": 0.2553828492760658,
      "num_tokens": 131404174.0,
      "step": 57290
    },
    {
      "entropy": 5.5825306415557865,
      "epoch": 5.503842459173871,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024255102790450667,
      "loss": 4.6957,
      "mean_token_accuracy": 0.257053031027317,
      "num_tokens": 131416123.0,
      "step": 57295
    },
    {
      "entropy": 5.52015962600708,
      "epoch": 5.504322766570605,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024251710312643093,
      "loss": 4.6469,
      "mean_token_accuracy": 0.2602158084511757,
      "num_tokens": 131428729.0,
      "step": 57300
    },
    {
      "entropy": 5.576592922210693,
      "epoch": 5.504803073967339,
      "grad_norm": 1.25,
      "learning_rate": 0.00024248317910251167,
      "loss": 4.6852,
      "mean_token_accuracy": 0.2605853796005249,
      "num_tokens": 131441231.0,
      "step": 57305
    },
    {
      "entropy": 5.600662517547607,
      "epoch": 5.505283381364073,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002424492558335365,
      "loss": 4.7565,
      "mean_token_accuracy": 0.2569032937288284,
      "num_tokens": 131452045.0,
      "step": 57310
    },
    {
      "entropy": 5.500004434585572,
      "epoch": 5.505763688760807,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002424153333202928,
      "loss": 4.6461,
      "mean_token_accuracy": 0.2585559353232384,
      "num_tokens": 131462729.0,
      "step": 57315
    },
    {
      "entropy": 5.4601860523223875,
      "epoch": 5.506243996157541,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00024238141156356836,
      "loss": 4.6359,
      "mean_token_accuracy": 0.2611518740653992,
      "num_tokens": 131474711.0,
      "step": 57320
    },
    {
      "entropy": 5.521010541915894,
      "epoch": 5.506724303554275,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024234749056415068,
      "loss": 4.5777,
      "mean_token_accuracy": 0.26131015717983247,
      "num_tokens": 131485575.0,
      "step": 57325
    },
    {
      "entropy": 5.595154428482056,
      "epoch": 5.507204610951009,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024231357032282724,
      "loss": 4.7491,
      "mean_token_accuracy": 0.25063241422176363,
      "num_tokens": 131496793.0,
      "step": 57330
    },
    {
      "entropy": 5.535326814651489,
      "epoch": 5.507684918347742,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002422796508403857,
      "loss": 4.6481,
      "mean_token_accuracy": 0.2637123540043831,
      "num_tokens": 131507444.0,
      "step": 57335
    },
    {
      "entropy": 5.560691261291504,
      "epoch": 5.508165225744476,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024224573211761364,
      "loss": 4.6268,
      "mean_token_accuracy": 0.2604785621166229,
      "num_tokens": 131519257.0,
      "step": 57340
    },
    {
      "entropy": 5.478405570983886,
      "epoch": 5.508645533141211,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024221181415529826,
      "loss": 4.579,
      "mean_token_accuracy": 0.2662571907043457,
      "num_tokens": 131530241.0,
      "step": 57345
    },
    {
      "entropy": 5.501017332077026,
      "epoch": 5.509125840537944,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024217789695422725,
      "loss": 4.6045,
      "mean_token_accuracy": 0.26855501532554626,
      "num_tokens": 131540593.0,
      "step": 57350
    },
    {
      "entropy": 5.539168071746826,
      "epoch": 5.5096061479346785,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024214398051518798,
      "loss": 4.655,
      "mean_token_accuracy": 0.26141014844179156,
      "num_tokens": 131553677.0,
      "step": 57355
    },
    {
      "entropy": 5.553333711624146,
      "epoch": 5.510086455331412,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002421100648389679,
      "loss": 4.6814,
      "mean_token_accuracy": 0.2609366595745087,
      "num_tokens": 131563451.0,
      "step": 57360
    },
    {
      "entropy": 5.441383647918701,
      "epoch": 5.510566762728146,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024207614992635446,
      "loss": 4.5362,
      "mean_token_accuracy": 0.26979308426380155,
      "num_tokens": 131574929.0,
      "step": 57365
    },
    {
      "entropy": 5.490345573425293,
      "epoch": 5.51104707012488,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024204223577813517,
      "loss": 4.6166,
      "mean_token_accuracy": 0.26347485929727554,
      "num_tokens": 131585648.0,
      "step": 57370
    },
    {
      "entropy": 5.489303874969482,
      "epoch": 5.511527377521614,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024200832239509724,
      "loss": 4.601,
      "mean_token_accuracy": 0.27165926694869996,
      "num_tokens": 131596917.0,
      "step": 57375
    },
    {
      "entropy": 5.4503173828125,
      "epoch": 5.5120076849183475,
      "grad_norm": 1.25,
      "learning_rate": 0.00024197440977802805,
      "loss": 4.5154,
      "mean_token_accuracy": 0.2731546476483345,
      "num_tokens": 131608242.0,
      "step": 57380
    },
    {
      "entropy": 5.563050079345703,
      "epoch": 5.512487992315082,
      "grad_norm": 1.34375,
      "learning_rate": 0.000241940497927715,
      "loss": 4.722,
      "mean_token_accuracy": 0.2591072231531143,
      "num_tokens": 131619614.0,
      "step": 57385
    },
    {
      "entropy": 5.551971578598023,
      "epoch": 5.512968299711815,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024190658684494543,
      "loss": 4.7037,
      "mean_token_accuracy": 0.262266731262207,
      "num_tokens": 131631112.0,
      "step": 57390
    },
    {
      "entropy": 5.6070631980896,
      "epoch": 5.51344860710855,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024187267653050666,
      "loss": 4.708,
      "mean_token_accuracy": 0.25856501907110213,
      "num_tokens": 131642884.0,
      "step": 57395
    },
    {
      "entropy": 5.544540596008301,
      "epoch": 5.513928914505283,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024183876698518597,
      "loss": 4.6704,
      "mean_token_accuracy": 0.2614371940493584,
      "num_tokens": 131655067.0,
      "step": 57400
    },
    {
      "entropy": 5.5691690921783445,
      "epoch": 5.514409221902017,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024180485820977067,
      "loss": 4.6789,
      "mean_token_accuracy": 0.25876782089471817,
      "num_tokens": 131666559.0,
      "step": 57405
    },
    {
      "entropy": 5.489535236358643,
      "epoch": 5.514889529298751,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024177095020504796,
      "loss": 4.581,
      "mean_token_accuracy": 0.26723990887403487,
      "num_tokens": 131678757.0,
      "step": 57410
    },
    {
      "entropy": 5.5477441310882565,
      "epoch": 5.515369836695485,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002417370429718051,
      "loss": 4.6371,
      "mean_token_accuracy": 0.2697236731648445,
      "num_tokens": 131689053.0,
      "step": 57415
    },
    {
      "entropy": 5.561307430267334,
      "epoch": 5.515850144092219,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024170313651082937,
      "loss": 4.6174,
      "mean_token_accuracy": 0.2624927654862404,
      "num_tokens": 131699450.0,
      "step": 57420
    },
    {
      "entropy": 5.552760934829712,
      "epoch": 5.516330451488953,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024166923082290787,
      "loss": 4.758,
      "mean_token_accuracy": 0.2542558953166008,
      "num_tokens": 131711297.0,
      "step": 57425
    },
    {
      "entropy": 5.57269320487976,
      "epoch": 5.516810758885687,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024163532590882797,
      "loss": 4.7123,
      "mean_token_accuracy": 0.2544345647096634,
      "num_tokens": 131722901.0,
      "step": 57430
    },
    {
      "entropy": 5.619903135299682,
      "epoch": 5.517291066282421,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024160142176937667,
      "loss": 4.7302,
      "mean_token_accuracy": 0.2569656133651733,
      "num_tokens": 131734255.0,
      "step": 57435
    },
    {
      "entropy": 5.525383329391479,
      "epoch": 5.517771373679155,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024156751840534113,
      "loss": 4.6634,
      "mean_token_accuracy": 0.25774671882390976,
      "num_tokens": 131745997.0,
      "step": 57440
    },
    {
      "entropy": 5.495254230499268,
      "epoch": 5.518251681075888,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024153361581750866,
      "loss": 4.6397,
      "mean_token_accuracy": 0.266185462474823,
      "num_tokens": 131757475.0,
      "step": 57445
    },
    {
      "entropy": 5.558307075500489,
      "epoch": 5.518731988472623,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024149971400666616,
      "loss": 4.6832,
      "mean_token_accuracy": 0.2604598984122276,
      "num_tokens": 131768955.0,
      "step": 57450
    },
    {
      "entropy": 5.531042623519897,
      "epoch": 5.519212295869356,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024146581297360083,
      "loss": 4.6745,
      "mean_token_accuracy": 0.2594736278057098,
      "num_tokens": 131780595.0,
      "step": 57455
    },
    {
      "entropy": 5.595713901519775,
      "epoch": 5.5196926032660905,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002414319127190998,
      "loss": 4.7215,
      "mean_token_accuracy": 0.25298340767621996,
      "num_tokens": 131790236.0,
      "step": 57460
    },
    {
      "entropy": 5.556583023071289,
      "epoch": 5.520172910662824,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024139801324395005,
      "loss": 4.6861,
      "mean_token_accuracy": 0.25199582427740097,
      "num_tokens": 131801405.0,
      "step": 57465
    },
    {
      "entropy": 5.5856071472167965,
      "epoch": 5.520653218059558,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024136411454893863,
      "loss": 4.7499,
      "mean_token_accuracy": 0.2583117797970772,
      "num_tokens": 131812514.0,
      "step": 57470
    },
    {
      "entropy": 5.5523253917694095,
      "epoch": 5.521133525456292,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002413302166348526,
      "loss": 4.6589,
      "mean_token_accuracy": 0.2601150572299957,
      "num_tokens": 131825340.0,
      "step": 57475
    },
    {
      "entropy": 5.552395057678223,
      "epoch": 5.521613832853026,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024129631950247895,
      "loss": 4.6864,
      "mean_token_accuracy": 0.2586754634976387,
      "num_tokens": 131837289.0,
      "step": 57480
    },
    {
      "entropy": 5.625013303756714,
      "epoch": 5.522094140249759,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024126242315260476,
      "loss": 4.7375,
      "mean_token_accuracy": 0.2556637957692146,
      "num_tokens": 131849427.0,
      "step": 57485
    },
    {
      "entropy": 5.516476583480835,
      "epoch": 5.522574447646494,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024122852758601694,
      "loss": 4.5759,
      "mean_token_accuracy": 0.26424673199653625,
      "num_tokens": 131861684.0,
      "step": 57490
    },
    {
      "entropy": 5.513039445877075,
      "epoch": 5.523054755043228,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002411946328035023,
      "loss": 4.6621,
      "mean_token_accuracy": 0.2633254438638687,
      "num_tokens": 131873624.0,
      "step": 57495
    },
    {
      "entropy": 5.496319913864136,
      "epoch": 5.5235350624399615,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000241160738805848,
      "loss": 4.5993,
      "mean_token_accuracy": 0.26981275230646135,
      "num_tokens": 131885031.0,
      "step": 57500
    },
    {
      "entropy": 5.549086809158325,
      "epoch": 5.524015369836696,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002411268455938408,
      "loss": 4.6589,
      "mean_token_accuracy": 0.25977465212345124,
      "num_tokens": 131896119.0,
      "step": 57505
    },
    {
      "entropy": 5.488938188552856,
      "epoch": 5.524495677233429,
      "grad_norm": 1.375,
      "learning_rate": 0.0002410929531682677,
      "loss": 4.6245,
      "mean_token_accuracy": 0.2645671620965004,
      "num_tokens": 131908026.0,
      "step": 57510
    },
    {
      "entropy": 5.497917509078979,
      "epoch": 5.524975984630164,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024105906152991548,
      "loss": 4.6583,
      "mean_token_accuracy": 0.2593472182750702,
      "num_tokens": 131919349.0,
      "step": 57515
    },
    {
      "entropy": 5.5905462265014645,
      "epoch": 5.525456292026897,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024102517067957124,
      "loss": 4.7056,
      "mean_token_accuracy": 0.25701661258935926,
      "num_tokens": 131931929.0,
      "step": 57520
    },
    {
      "entropy": 5.558129835128784,
      "epoch": 5.525936599423631,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002409912806180215,
      "loss": 4.6406,
      "mean_token_accuracy": 0.2629298761487007,
      "num_tokens": 131944340.0,
      "step": 57525
    },
    {
      "entropy": 5.560352230072022,
      "epoch": 5.526416906820365,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024095739134605328,
      "loss": 4.6904,
      "mean_token_accuracy": 0.26617399603128433,
      "num_tokens": 131956348.0,
      "step": 57530
    },
    {
      "entropy": 5.513148307800293,
      "epoch": 5.526897214217099,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002409235028644533,
      "loss": 4.6117,
      "mean_token_accuracy": 0.26905166208744047,
      "num_tokens": 131967384.0,
      "step": 57535
    },
    {
      "entropy": 5.6147984027862545,
      "epoch": 5.527377521613833,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024088961517400843,
      "loss": 4.7868,
      "mean_token_accuracy": 0.24289864301681519,
      "num_tokens": 131979007.0,
      "step": 57540
    },
    {
      "entropy": 5.570032215118408,
      "epoch": 5.527857829010567,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024085572827550538,
      "loss": 4.6476,
      "mean_token_accuracy": 0.26478919088840486,
      "num_tokens": 131990060.0,
      "step": 57545
    },
    {
      "entropy": 5.502352619171143,
      "epoch": 5.5283381364073,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000240821842169731,
      "loss": 4.585,
      "mean_token_accuracy": 0.2673342823982239,
      "num_tokens": 132000479.0,
      "step": 57550
    },
    {
      "entropy": 5.454243850708008,
      "epoch": 5.528818443804035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000240787956857472,
      "loss": 4.6001,
      "mean_token_accuracy": 0.27169129997491837,
      "num_tokens": 132012639.0,
      "step": 57555
    },
    {
      "entropy": 5.459134864807129,
      "epoch": 5.529298751200768,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024075407233951497,
      "loss": 4.5626,
      "mean_token_accuracy": 0.2691186532378197,
      "num_tokens": 132023191.0,
      "step": 57560
    },
    {
      "entropy": 5.4833019256591795,
      "epoch": 5.529779058597502,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024072018861664667,
      "loss": 4.611,
      "mean_token_accuracy": 0.26840769201517106,
      "num_tokens": 132034121.0,
      "step": 57565
    },
    {
      "entropy": 5.480403900146484,
      "epoch": 5.530259365994237,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024068630568965384,
      "loss": 4.6351,
      "mean_token_accuracy": 0.26461388915777206,
      "num_tokens": 132046670.0,
      "step": 57570
    },
    {
      "entropy": 5.546130228042602,
      "epoch": 5.53073967339097,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024065242355932309,
      "loss": 4.6644,
      "mean_token_accuracy": 0.2548812285065651,
      "num_tokens": 132058121.0,
      "step": 57575
    },
    {
      "entropy": 5.56287145614624,
      "epoch": 5.5312199807877045,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024061854222644108,
      "loss": 4.6794,
      "mean_token_accuracy": 0.25830062478780746,
      "num_tokens": 132069845.0,
      "step": 57580
    },
    {
      "entropy": 5.519787883758545,
      "epoch": 5.531700288184438,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002405846616917945,
      "loss": 4.5965,
      "mean_token_accuracy": 0.2671031802892685,
      "num_tokens": 132081191.0,
      "step": 57585
    },
    {
      "entropy": 5.562394523620606,
      "epoch": 5.532180595581172,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002405507819561698,
      "loss": 4.6541,
      "mean_token_accuracy": 0.2667646124958992,
      "num_tokens": 132092014.0,
      "step": 57590
    },
    {
      "entropy": 5.4668541431427,
      "epoch": 5.532660902977906,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024051690302035361,
      "loss": 4.5882,
      "mean_token_accuracy": 0.2663127452135086,
      "num_tokens": 132103501.0,
      "step": 57595
    },
    {
      "entropy": 5.475134658813476,
      "epoch": 5.53314121037464,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024048302488513258,
      "loss": 4.6025,
      "mean_token_accuracy": 0.27331652492284775,
      "num_tokens": 132114254.0,
      "step": 57600
    },
    {
      "entropy": 5.527251529693603,
      "epoch": 5.5336215177713735,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002404491475512932,
      "loss": 4.6736,
      "mean_token_accuracy": 0.264431294798851,
      "num_tokens": 132126551.0,
      "step": 57605
    },
    {
      "entropy": 5.614603805541992,
      "epoch": 5.534101825168108,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024041527101962206,
      "loss": 4.7281,
      "mean_token_accuracy": 0.2597188174724579,
      "num_tokens": 132137637.0,
      "step": 57610
    },
    {
      "entropy": 5.473398160934448,
      "epoch": 5.534582132564841,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024038139529090568,
      "loss": 4.6049,
      "mean_token_accuracy": 0.26871782541275024,
      "num_tokens": 132148109.0,
      "step": 57615
    },
    {
      "entropy": 5.477512407302856,
      "epoch": 5.5350624399615755,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024034752036593038,
      "loss": 4.6742,
      "mean_token_accuracy": 0.26471829116344453,
      "num_tokens": 132159843.0,
      "step": 57620
    },
    {
      "entropy": 5.55811767578125,
      "epoch": 5.535542747358309,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024031364624548286,
      "loss": 4.7055,
      "mean_token_accuracy": 0.2562703534960747,
      "num_tokens": 132171628.0,
      "step": 57625
    },
    {
      "entropy": 5.610421800613404,
      "epoch": 5.536023054755043,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024027977293034943,
      "loss": 4.7216,
      "mean_token_accuracy": 0.25570945888757707,
      "num_tokens": 132182868.0,
      "step": 57630
    },
    {
      "entropy": 5.608910036087036,
      "epoch": 5.536503362151777,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024024590042131654,
      "loss": 4.6948,
      "mean_token_accuracy": 0.2548675462603569,
      "num_tokens": 132195052.0,
      "step": 57635
    },
    {
      "entropy": 5.513676118850708,
      "epoch": 5.536983669548511,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002402120287191707,
      "loss": 4.6562,
      "mean_token_accuracy": 0.262767618894577,
      "num_tokens": 132205374.0,
      "step": 57640
    },
    {
      "entropy": 5.455185317993164,
      "epoch": 5.537463976945245,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024017815782469826,
      "loss": 4.4941,
      "mean_token_accuracy": 0.2726869761943817,
      "num_tokens": 132216034.0,
      "step": 57645
    },
    {
      "entropy": 5.532267999649048,
      "epoch": 5.537944284341979,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024014428773868558,
      "loss": 4.6682,
      "mean_token_accuracy": 0.26072792857885363,
      "num_tokens": 132227608.0,
      "step": 57650
    },
    {
      "entropy": 5.449679613113403,
      "epoch": 5.538424591738713,
      "grad_norm": 1.15625,
      "learning_rate": 0.000240110418461919,
      "loss": 4.5838,
      "mean_token_accuracy": 0.2661838933825493,
      "num_tokens": 132240858.0,
      "step": 57655
    },
    {
      "entropy": 5.571811676025391,
      "epoch": 5.538904899135447,
      "grad_norm": 1.375,
      "learning_rate": 0.00024007654999518496,
      "loss": 4.7203,
      "mean_token_accuracy": 0.25150139927864074,
      "num_tokens": 132251285.0,
      "step": 57660
    },
    {
      "entropy": 5.568498182296753,
      "epoch": 5.539385206532181,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024004268233926968,
      "loss": 4.672,
      "mean_token_accuracy": 0.2578626319766045,
      "num_tokens": 132262620.0,
      "step": 57665
    },
    {
      "entropy": 5.506626319885254,
      "epoch": 5.539865513928914,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002400088154949595,
      "loss": 4.5619,
      "mean_token_accuracy": 0.270212784409523,
      "num_tokens": 132273765.0,
      "step": 57670
    },
    {
      "entropy": 5.484952688217163,
      "epoch": 5.540345821325649,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023997494946304088,
      "loss": 4.5774,
      "mean_token_accuracy": 0.27194471955299376,
      "num_tokens": 132285174.0,
      "step": 57675
    },
    {
      "entropy": 5.485638427734375,
      "epoch": 5.540826128722382,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023994108424429977,
      "loss": 4.5926,
      "mean_token_accuracy": 0.26461776793003083,
      "num_tokens": 132296294.0,
      "step": 57680
    },
    {
      "entropy": 5.521288728713989,
      "epoch": 5.541306436119116,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023990721983952262,
      "loss": 4.6278,
      "mean_token_accuracy": 0.2639862298965454,
      "num_tokens": 132307814.0,
      "step": 57685
    },
    {
      "entropy": 5.580422210693359,
      "epoch": 5.54178674351585,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023987335624949563,
      "loss": 4.6983,
      "mean_token_accuracy": 0.2575941428542137,
      "num_tokens": 132319080.0,
      "step": 57690
    },
    {
      "entropy": 5.508660078048706,
      "epoch": 5.542267050912584,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023983949347500494,
      "loss": 4.5851,
      "mean_token_accuracy": 0.26907698065042496,
      "num_tokens": 132328869.0,
      "step": 57695
    },
    {
      "entropy": 5.558215665817261,
      "epoch": 5.542747358309318,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023980563151683687,
      "loss": 4.703,
      "mean_token_accuracy": 0.2566508173942566,
      "num_tokens": 132339503.0,
      "step": 57700
    },
    {
      "entropy": 5.556027221679687,
      "epoch": 5.543227665706052,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023977177037577753,
      "loss": 4.7079,
      "mean_token_accuracy": 0.2595405951142311,
      "num_tokens": 132351475.0,
      "step": 57705
    },
    {
      "entropy": 5.573723411560058,
      "epoch": 5.543707973102785,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023973791005261303,
      "loss": 4.6897,
      "mean_token_accuracy": 0.2593743860721588,
      "num_tokens": 132362078.0,
      "step": 57710
    },
    {
      "entropy": 5.544609069824219,
      "epoch": 5.54418828049952,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023970405054812965,
      "loss": 4.6586,
      "mean_token_accuracy": 0.2648856922984123,
      "num_tokens": 132373148.0,
      "step": 57715
    },
    {
      "entropy": 5.514827585220337,
      "epoch": 5.544668587896254,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023967019186311333,
      "loss": 4.628,
      "mean_token_accuracy": 0.26466204226017,
      "num_tokens": 132382979.0,
      "step": 57720
    },
    {
      "entropy": 5.579067659378052,
      "epoch": 5.5451488952929875,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023963633399835027,
      "loss": 4.7789,
      "mean_token_accuracy": 0.25272896885871887,
      "num_tokens": 132394655.0,
      "step": 57725
    },
    {
      "entropy": 5.560921621322632,
      "epoch": 5.545629202689721,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002396024769546265,
      "loss": 4.6875,
      "mean_token_accuracy": 0.2562889292836189,
      "num_tokens": 132404767.0,
      "step": 57730
    },
    {
      "entropy": 5.515336799621582,
      "epoch": 5.546109510086455,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023956862073272817,
      "loss": 4.6407,
      "mean_token_accuracy": 0.25786233097314837,
      "num_tokens": 132415694.0,
      "step": 57735
    },
    {
      "entropy": 5.575648355484009,
      "epoch": 5.54658981748319,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023953476533344114,
      "loss": 4.706,
      "mean_token_accuracy": 0.25481009632349017,
      "num_tokens": 132426207.0,
      "step": 57740
    },
    {
      "entropy": 5.605410957336426,
      "epoch": 5.547070124879923,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023950091075755164,
      "loss": 4.7607,
      "mean_token_accuracy": 0.2571074768900871,
      "num_tokens": 132437150.0,
      "step": 57745
    },
    {
      "entropy": 5.545283031463623,
      "epoch": 5.547550432276657,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023946705700584554,
      "loss": 4.6598,
      "mean_token_accuracy": 0.26081894189119337,
      "num_tokens": 132448821.0,
      "step": 57750
    },
    {
      "entropy": 5.54492769241333,
      "epoch": 5.548030739673391,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023943320407910897,
      "loss": 4.6698,
      "mean_token_accuracy": 0.25848652273416517,
      "num_tokens": 132459452.0,
      "step": 57755
    },
    {
      "entropy": 5.5195465087890625,
      "epoch": 5.548511047070125,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002393993519781277,
      "loss": 4.6626,
      "mean_token_accuracy": 0.26547653377056124,
      "num_tokens": 132470526.0,
      "step": 57760
    },
    {
      "entropy": 5.612005043029785,
      "epoch": 5.5489913544668585,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023936550070368772,
      "loss": 4.6807,
      "mean_token_accuracy": 0.2525515541434288,
      "num_tokens": 132481170.0,
      "step": 57765
    },
    {
      "entropy": 5.515181827545166,
      "epoch": 5.549471661863593,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002393316502565751,
      "loss": 4.5961,
      "mean_token_accuracy": 0.2626043751835823,
      "num_tokens": 132491715.0,
      "step": 57770
    },
    {
      "entropy": 5.501812982559204,
      "epoch": 5.549951969260326,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023929780063757555,
      "loss": 4.6178,
      "mean_token_accuracy": 0.27309364825487137,
      "num_tokens": 132503205.0,
      "step": 57775
    },
    {
      "entropy": 5.639882278442383,
      "epoch": 5.550432276657061,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023926395184747514,
      "loss": 4.7622,
      "mean_token_accuracy": 0.252947635948658,
      "num_tokens": 132513778.0,
      "step": 57780
    },
    {
      "entropy": 5.515955686569214,
      "epoch": 5.550912584053794,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023923010388705957,
      "loss": 4.6715,
      "mean_token_accuracy": 0.2633675590157509,
      "num_tokens": 132524899.0,
      "step": 57785
    },
    {
      "entropy": 5.555864334106445,
      "epoch": 5.551392891450528,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023919625675711481,
      "loss": 4.7215,
      "mean_token_accuracy": 0.25931763350963594,
      "num_tokens": 132536179.0,
      "step": 57790
    },
    {
      "entropy": 5.527301883697509,
      "epoch": 5.551873198847263,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023916241045842673,
      "loss": 4.5897,
      "mean_token_accuracy": 0.2669229179620743,
      "num_tokens": 132547921.0,
      "step": 57795
    },
    {
      "entropy": 5.594976568222046,
      "epoch": 5.552353506243996,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023912856499178103,
      "loss": 4.6876,
      "mean_token_accuracy": 0.26462043076753616,
      "num_tokens": 132558689.0,
      "step": 57800
    },
    {
      "entropy": 5.500507783889771,
      "epoch": 5.55283381364073,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002390947203579635,
      "loss": 4.6194,
      "mean_token_accuracy": 0.2681365981698036,
      "num_tokens": 132569786.0,
      "step": 57805
    },
    {
      "entropy": 5.561190557479859,
      "epoch": 5.553314121037464,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023906087655775987,
      "loss": 4.6768,
      "mean_token_accuracy": 0.25452143102884295,
      "num_tokens": 132581619.0,
      "step": 57810
    },
    {
      "entropy": 5.490360975265503,
      "epoch": 5.553794428434198,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002390270335919561,
      "loss": 4.6502,
      "mean_token_accuracy": 0.2670275405049324,
      "num_tokens": 132592642.0,
      "step": 57815
    },
    {
      "entropy": 5.485966205596924,
      "epoch": 5.554274735830932,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023899319146133776,
      "loss": 4.6085,
      "mean_token_accuracy": 0.2645805090665817,
      "num_tokens": 132604690.0,
      "step": 57820
    },
    {
      "entropy": 5.593080663681031,
      "epoch": 5.554755043227666,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023895935016669063,
      "loss": 4.7437,
      "mean_token_accuracy": 0.2518486365675926,
      "num_tokens": 132615185.0,
      "step": 57825
    },
    {
      "entropy": 5.527442121505738,
      "epoch": 5.555235350624399,
      "grad_norm": 1.375,
      "learning_rate": 0.00023892550970880028,
      "loss": 4.544,
      "mean_token_accuracy": 0.2720296666026115,
      "num_tokens": 132626618.0,
      "step": 57830
    },
    {
      "entropy": 5.600471782684326,
      "epoch": 5.555715658021134,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023889167008845247,
      "loss": 4.6538,
      "mean_token_accuracy": 0.2610154703259468,
      "num_tokens": 132637248.0,
      "step": 57835
    },
    {
      "entropy": 5.587549638748169,
      "epoch": 5.556195965417867,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023885783130643296,
      "loss": 4.681,
      "mean_token_accuracy": 0.25622774511575697,
      "num_tokens": 132647587.0,
      "step": 57840
    },
    {
      "entropy": 5.4741839408874515,
      "epoch": 5.5566762728146015,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023882399336352716,
      "loss": 4.5678,
      "mean_token_accuracy": 0.26961965262889864,
      "num_tokens": 132659035.0,
      "step": 57845
    },
    {
      "entropy": 5.522672176361084,
      "epoch": 5.557156580211335,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023879015626052086,
      "loss": 4.6248,
      "mean_token_accuracy": 0.2633457988500595,
      "num_tokens": 132668693.0,
      "step": 57850
    },
    {
      "entropy": 5.550029563903808,
      "epoch": 5.557636887608069,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023875631999819975,
      "loss": 4.6975,
      "mean_token_accuracy": 0.25955296605825423,
      "num_tokens": 132679588.0,
      "step": 57855
    },
    {
      "entropy": 5.604858350753784,
      "epoch": 5.558117195004803,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023872248457734914,
      "loss": 4.7067,
      "mean_token_accuracy": 0.2572587922215462,
      "num_tokens": 132690508.0,
      "step": 57860
    },
    {
      "entropy": 5.554405784606933,
      "epoch": 5.558597502401537,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023868864999875472,
      "loss": 4.643,
      "mean_token_accuracy": 0.2586081773042679,
      "num_tokens": 132701414.0,
      "step": 57865
    },
    {
      "entropy": 5.551690912246704,
      "epoch": 5.559077809798271,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000238654816263202,
      "loss": 4.6719,
      "mean_token_accuracy": 0.26184273809194564,
      "num_tokens": 132713930.0,
      "step": 57870
    },
    {
      "entropy": 5.465283346176148,
      "epoch": 5.559558117195005,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023862098337147658,
      "loss": 4.6299,
      "mean_token_accuracy": 0.2674832418560982,
      "num_tokens": 132726331.0,
      "step": 57875
    },
    {
      "entropy": 5.561721467971802,
      "epoch": 5.560038424591738,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002385871513243638,
      "loss": 4.6303,
      "mean_token_accuracy": 0.25899035334587095,
      "num_tokens": 132737052.0,
      "step": 57880
    },
    {
      "entropy": 5.566389179229736,
      "epoch": 5.560518731988473,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023855332012264946,
      "loss": 4.6614,
      "mean_token_accuracy": 0.2598510071635246,
      "num_tokens": 132748669.0,
      "step": 57885
    },
    {
      "entropy": 5.5740394115448,
      "epoch": 5.560999039385207,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023851948976711862,
      "loss": 4.7045,
      "mean_token_accuracy": 0.2572329148650169,
      "num_tokens": 132760017.0,
      "step": 57890
    },
    {
      "entropy": 5.575355195999146,
      "epoch": 5.56147934678194,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023848566025855695,
      "loss": 4.7021,
      "mean_token_accuracy": 0.2598653331398964,
      "num_tokens": 132772041.0,
      "step": 57895
    },
    {
      "entropy": 5.561232471466065,
      "epoch": 5.561959654178675,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002384518315977498,
      "loss": 4.7181,
      "mean_token_accuracy": 0.24883572459220887,
      "num_tokens": 132783535.0,
      "step": 57900
    },
    {
      "entropy": 5.518224763870239,
      "epoch": 5.562439961575408,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023841800378548262,
      "loss": 4.6269,
      "mean_token_accuracy": 0.2696820929646492,
      "num_tokens": 132795431.0,
      "step": 57905
    },
    {
      "entropy": 5.54474573135376,
      "epoch": 5.562920268972142,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023838417682254075,
      "loss": 4.6565,
      "mean_token_accuracy": 0.2703996181488037,
      "num_tokens": 132807072.0,
      "step": 57910
    },
    {
      "entropy": 5.597355794906616,
      "epoch": 5.563400576368876,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002383503507097096,
      "loss": 4.7297,
      "mean_token_accuracy": 0.2543189004063606,
      "num_tokens": 132820000.0,
      "step": 57915
    },
    {
      "entropy": 5.507348918914795,
      "epoch": 5.56388088376561,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023831652544777448,
      "loss": 4.6712,
      "mean_token_accuracy": 0.26150113344192505,
      "num_tokens": 132831996.0,
      "step": 57920
    },
    {
      "entropy": 5.554322004318237,
      "epoch": 5.564361191162344,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002382827010375207,
      "loss": 4.6879,
      "mean_token_accuracy": 0.25552174299955366,
      "num_tokens": 132842743.0,
      "step": 57925
    },
    {
      "entropy": 5.560444641113281,
      "epoch": 5.564841498559078,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023824887747973355,
      "loss": 4.6957,
      "mean_token_accuracy": 0.2537055626511574,
      "num_tokens": 132854041.0,
      "step": 57930
    },
    {
      "entropy": 5.520587205886841,
      "epoch": 5.565321805955811,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023821505477519833,
      "loss": 4.6203,
      "mean_token_accuracy": 0.2664726212620735,
      "num_tokens": 132865767.0,
      "step": 57935
    },
    {
      "entropy": 5.690574264526367,
      "epoch": 5.565802113352546,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002381812329247003,
      "loss": 4.8011,
      "mean_token_accuracy": 0.24921535402536393,
      "num_tokens": 132875923.0,
      "step": 57940
    },
    {
      "entropy": 5.5121008396148685,
      "epoch": 5.56628242074928,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023814741192902485,
      "loss": 4.6663,
      "mean_token_accuracy": 0.2598172128200531,
      "num_tokens": 132887605.0,
      "step": 57945
    },
    {
      "entropy": 5.630162525177002,
      "epoch": 5.5667627281460135,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023811359178895692,
      "loss": 4.8077,
      "mean_token_accuracy": 0.24886188060045242,
      "num_tokens": 132900479.0,
      "step": 57950
    },
    {
      "entropy": 5.482885837554932,
      "epoch": 5.567243035542747,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002380797725052819,
      "loss": 4.5677,
      "mean_token_accuracy": 0.26722682267427444,
      "num_tokens": 132913040.0,
      "step": 57955
    },
    {
      "entropy": 5.55270528793335,
      "epoch": 5.567723342939481,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023804595407878488,
      "loss": 4.6772,
      "mean_token_accuracy": 0.2578494042158127,
      "num_tokens": 132923170.0,
      "step": 57960
    },
    {
      "entropy": 5.513829231262207,
      "epoch": 5.568203650336216,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002380121365102511,
      "loss": 4.6311,
      "mean_token_accuracy": 0.2666456580162048,
      "num_tokens": 132934692.0,
      "step": 57965
    },
    {
      "entropy": 5.5409831523895265,
      "epoch": 5.568683957732949,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023797831980046575,
      "loss": 4.6967,
      "mean_token_accuracy": 0.2544615313410759,
      "num_tokens": 132944849.0,
      "step": 57970
    },
    {
      "entropy": 5.65801477432251,
      "epoch": 5.569164265129683,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023794450395021383,
      "loss": 4.786,
      "mean_token_accuracy": 0.2538959369063377,
      "num_tokens": 132955521.0,
      "step": 57975
    },
    {
      "entropy": 5.482784128189087,
      "epoch": 5.569644572526417,
      "grad_norm": 1.375,
      "learning_rate": 0.00023791068896028057,
      "loss": 4.5789,
      "mean_token_accuracy": 0.2681834936141968,
      "num_tokens": 132966318.0,
      "step": 57980
    },
    {
      "entropy": 5.628254699707031,
      "epoch": 5.570124879923151,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023787687483145086,
      "loss": 4.827,
      "mean_token_accuracy": 0.247786745429039,
      "num_tokens": 132977986.0,
      "step": 57985
    },
    {
      "entropy": 5.529298210144043,
      "epoch": 5.5706051873198845,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002378430615645099,
      "loss": 4.604,
      "mean_token_accuracy": 0.2626295477151871,
      "num_tokens": 132989007.0,
      "step": 57990
    },
    {
      "entropy": 5.58384256362915,
      "epoch": 5.571085494716619,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023780924916024277,
      "loss": 4.6543,
      "mean_token_accuracy": 0.2557422578334808,
      "num_tokens": 133001704.0,
      "step": 57995
    },
    {
      "entropy": 5.564183712005615,
      "epoch": 5.571565802113352,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023777543761943444,
      "loss": 4.6924,
      "mean_token_accuracy": 0.2553525611758232,
      "num_tokens": 133012508.0,
      "step": 58000
    },
    {
      "entropy": 5.55680627822876,
      "epoch": 5.572046109510087,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023774162694286988,
      "loss": 4.6938,
      "mean_token_accuracy": 0.2556836843490601,
      "num_tokens": 133024562.0,
      "step": 58005
    },
    {
      "entropy": 5.462881565093994,
      "epoch": 5.57252641690682,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002377078171313341,
      "loss": 4.5611,
      "mean_token_accuracy": 0.2761775031685829,
      "num_tokens": 133035607.0,
      "step": 58010
    },
    {
      "entropy": 5.499260091781617,
      "epoch": 5.573006724303554,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023767400818561207,
      "loss": 4.6637,
      "mean_token_accuracy": 0.2675306677818298,
      "num_tokens": 133046671.0,
      "step": 58015
    },
    {
      "entropy": 5.5017763614654545,
      "epoch": 5.573487031700288,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023764020010648884,
      "loss": 4.6232,
      "mean_token_accuracy": 0.26812776923179626,
      "num_tokens": 133057376.0,
      "step": 58020
    },
    {
      "entropy": 5.473006343841552,
      "epoch": 5.573967339097022,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002376063928947491,
      "loss": 4.5697,
      "mean_token_accuracy": 0.27464068233966826,
      "num_tokens": 133068924.0,
      "step": 58025
    },
    {
      "entropy": 5.433808708190918,
      "epoch": 5.574447646493756,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023757258655117798,
      "loss": 4.5362,
      "mean_token_accuracy": 0.27195069193840027,
      "num_tokens": 133081394.0,
      "step": 58030
    },
    {
      "entropy": 5.575059175491333,
      "epoch": 5.57492795389049,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023753878107656018,
      "loss": 4.7659,
      "mean_token_accuracy": 0.2550211861729622,
      "num_tokens": 133093693.0,
      "step": 58035
    },
    {
      "entropy": 5.5672143459320065,
      "epoch": 5.575408261287224,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023750497647168068,
      "loss": 4.6651,
      "mean_token_accuracy": 0.2582345262169838,
      "num_tokens": 133105273.0,
      "step": 58040
    },
    {
      "entropy": 5.53424220085144,
      "epoch": 5.575888568683958,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023747117273732426,
      "loss": 4.6452,
      "mean_token_accuracy": 0.2635346338152885,
      "num_tokens": 133117574.0,
      "step": 58045
    },
    {
      "entropy": 5.544147872924805,
      "epoch": 5.576368876080692,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023743736987427578,
      "loss": 4.679,
      "mean_token_accuracy": 0.2594455361366272,
      "num_tokens": 133128092.0,
      "step": 58050
    },
    {
      "entropy": 5.532258415222168,
      "epoch": 5.576849183477425,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023740356788332008,
      "loss": 4.7189,
      "mean_token_accuracy": 0.25369453579187395,
      "num_tokens": 133140295.0,
      "step": 58055
    },
    {
      "entropy": 5.560347175598144,
      "epoch": 5.57732949087416,
      "grad_norm": 1.375,
      "learning_rate": 0.0002373697667652419,
      "loss": 4.6654,
      "mean_token_accuracy": 0.26528194844722747,
      "num_tokens": 133151935.0,
      "step": 58060
    },
    {
      "entropy": 5.542474985122681,
      "epoch": 5.577809798270893,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000237335966520826,
      "loss": 4.6409,
      "mean_token_accuracy": 0.26991783976554873,
      "num_tokens": 133162818.0,
      "step": 58065
    },
    {
      "entropy": 5.601067590713501,
      "epoch": 5.5782901056676275,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023730216715085707,
      "loss": 4.7,
      "mean_token_accuracy": 0.2584048256278038,
      "num_tokens": 133174387.0,
      "step": 58070
    },
    {
      "entropy": 5.549365472793579,
      "epoch": 5.578770413064361,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023726836865611982,
      "loss": 4.7139,
      "mean_token_accuracy": 0.2587448388338089,
      "num_tokens": 133185969.0,
      "step": 58075
    },
    {
      "entropy": 5.4573592185974125,
      "epoch": 5.579250720461095,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002372345710373991,
      "loss": 4.5407,
      "mean_token_accuracy": 0.27464292794466016,
      "num_tokens": 133197722.0,
      "step": 58080
    },
    {
      "entropy": 5.539924621582031,
      "epoch": 5.579731027857829,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023720077429547948,
      "loss": 4.6383,
      "mean_token_accuracy": 0.26506015211343764,
      "num_tokens": 133209382.0,
      "step": 58085
    },
    {
      "entropy": 5.562364864349365,
      "epoch": 5.580211335254563,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002371669784311456,
      "loss": 4.6939,
      "mean_token_accuracy": 0.2612220972776413,
      "num_tokens": 133221301.0,
      "step": 58090
    },
    {
      "entropy": 5.498718357086181,
      "epoch": 5.5806916426512965,
      "grad_norm": 1.25,
      "learning_rate": 0.0002371331834451823,
      "loss": 4.6695,
      "mean_token_accuracy": 0.26381743997335433,
      "num_tokens": 133233296.0,
      "step": 58095
    },
    {
      "entropy": 5.52414379119873,
      "epoch": 5.581171950048031,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023709938933837388,
      "loss": 4.6068,
      "mean_token_accuracy": 0.2620381250977516,
      "num_tokens": 133243689.0,
      "step": 58100
    },
    {
      "entropy": 5.548212671279908,
      "epoch": 5.581652257444764,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023706559611150514,
      "loss": 4.6676,
      "mean_token_accuracy": 0.25511214286088946,
      "num_tokens": 133255058.0,
      "step": 58105
    },
    {
      "entropy": 5.559371137619019,
      "epoch": 5.582132564841499,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002370318037653606,
      "loss": 4.6255,
      "mean_token_accuracy": 0.272538448870182,
      "num_tokens": 133266180.0,
      "step": 58110
    },
    {
      "entropy": 5.526770114898682,
      "epoch": 5.582612872238233,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023699801230072482,
      "loss": 4.7251,
      "mean_token_accuracy": 0.2597008630633354,
      "num_tokens": 133278424.0,
      "step": 58115
    },
    {
      "entropy": 5.460002470016479,
      "epoch": 5.583093179634966,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023696422171838233,
      "loss": 4.6143,
      "mean_token_accuracy": 0.2709684044122696,
      "num_tokens": 133290164.0,
      "step": 58120
    },
    {
      "entropy": 5.498451375961304,
      "epoch": 5.583573487031701,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023693043201911785,
      "loss": 4.6859,
      "mean_token_accuracy": 0.26002724170684816,
      "num_tokens": 133302341.0,
      "step": 58125
    },
    {
      "entropy": 5.575848007202149,
      "epoch": 5.584053794428434,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002368966432037155,
      "loss": 4.7164,
      "mean_token_accuracy": 0.25216303765773773,
      "num_tokens": 133313283.0,
      "step": 58130
    },
    {
      "entropy": 5.535835456848145,
      "epoch": 5.584534101825168,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023686285527296002,
      "loss": 4.6597,
      "mean_token_accuracy": 0.2593510791659355,
      "num_tokens": 133324786.0,
      "step": 58135
    },
    {
      "entropy": 5.628232717514038,
      "epoch": 5.585014409221902,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023682906822763579,
      "loss": 4.7183,
      "mean_token_accuracy": 0.25520673841238023,
      "num_tokens": 133335413.0,
      "step": 58140
    },
    {
      "entropy": 5.557036590576172,
      "epoch": 5.585494716618636,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023679528206852724,
      "loss": 4.7118,
      "mean_token_accuracy": 0.26071573942899706,
      "num_tokens": 133346903.0,
      "step": 58145
    },
    {
      "entropy": 5.586291313171387,
      "epoch": 5.58597502401537,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023676149679641877,
      "loss": 4.7118,
      "mean_token_accuracy": 0.25620412528514863,
      "num_tokens": 133358705.0,
      "step": 58150
    },
    {
      "entropy": 5.552981233596801,
      "epoch": 5.586455331412104,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023672771241209495,
      "loss": 4.7177,
      "mean_token_accuracy": 0.2629178509116173,
      "num_tokens": 133371141.0,
      "step": 58155
    },
    {
      "entropy": 5.529493951797486,
      "epoch": 5.586935638808837,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002366939289163399,
      "loss": 4.6184,
      "mean_token_accuracy": 0.2628363847732544,
      "num_tokens": 133381801.0,
      "step": 58160
    },
    {
      "entropy": 5.490673923492432,
      "epoch": 5.587415946205572,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023666014630993804,
      "loss": 4.5727,
      "mean_token_accuracy": 0.2711659148335457,
      "num_tokens": 133393061.0,
      "step": 58165
    },
    {
      "entropy": 5.547459363937378,
      "epoch": 5.587896253602305,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023662636459367375,
      "loss": 4.6713,
      "mean_token_accuracy": 0.2573596641421318,
      "num_tokens": 133404864.0,
      "step": 58170
    },
    {
      "entropy": 5.5963860034942625,
      "epoch": 5.5883765609990395,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023659258376833132,
      "loss": 4.6413,
      "mean_token_accuracy": 0.26715923845767975,
      "num_tokens": 133416504.0,
      "step": 58175
    },
    {
      "entropy": 5.581239509582519,
      "epoch": 5.588856868395773,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023655880383469504,
      "loss": 4.6812,
      "mean_token_accuracy": 0.25699014514684676,
      "num_tokens": 133427875.0,
      "step": 58180
    },
    {
      "entropy": 5.466398620605469,
      "epoch": 5.589337175792507,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002365250247935493,
      "loss": 4.6289,
      "mean_token_accuracy": 0.262110498547554,
      "num_tokens": 133440173.0,
      "step": 58185
    },
    {
      "entropy": 5.4526047706604,
      "epoch": 5.5898174831892415,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002364912466456781,
      "loss": 4.5991,
      "mean_token_accuracy": 0.26901147067546843,
      "num_tokens": 133450662.0,
      "step": 58190
    },
    {
      "entropy": 5.501416254043579,
      "epoch": 5.590297790585975,
      "grad_norm": 1.25,
      "learning_rate": 0.0002364574693918659,
      "loss": 4.6144,
      "mean_token_accuracy": 0.26477347761392594,
      "num_tokens": 133462423.0,
      "step": 58195
    },
    {
      "entropy": 5.521052074432373,
      "epoch": 5.590778097982709,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023642369303289673,
      "loss": 4.641,
      "mean_token_accuracy": 0.2621654361486435,
      "num_tokens": 133474539.0,
      "step": 58200
    },
    {
      "entropy": 5.591468334197998,
      "epoch": 5.591258405379443,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002363899175695549,
      "loss": 4.7234,
      "mean_token_accuracy": 0.2521379441022873,
      "num_tokens": 133487454.0,
      "step": 58205
    },
    {
      "entropy": 5.553951930999756,
      "epoch": 5.591738712776177,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023635614300262454,
      "loss": 4.6884,
      "mean_token_accuracy": 0.26121708005666733,
      "num_tokens": 133498337.0,
      "step": 58210
    },
    {
      "entropy": 5.553236103057861,
      "epoch": 5.5922190201729105,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023632236933288986,
      "loss": 4.6825,
      "mean_token_accuracy": 0.2582009047269821,
      "num_tokens": 133509179.0,
      "step": 58215
    },
    {
      "entropy": 5.581151676177979,
      "epoch": 5.592699327569645,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023628859656113483,
      "loss": 4.7129,
      "mean_token_accuracy": 0.2664176538586617,
      "num_tokens": 133522283.0,
      "step": 58220
    },
    {
      "entropy": 5.5949054718017575,
      "epoch": 5.593179634966378,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023625482468814363,
      "loss": 4.7582,
      "mean_token_accuracy": 0.2538511112332344,
      "num_tokens": 133533067.0,
      "step": 58225
    },
    {
      "entropy": 5.5384430408477785,
      "epoch": 5.593659942363113,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023622105371470039,
      "loss": 4.6145,
      "mean_token_accuracy": 0.2622383087873459,
      "num_tokens": 133544845.0,
      "step": 58230
    },
    {
      "entropy": 5.545537376403809,
      "epoch": 5.594140249759846,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002361872836415891,
      "loss": 4.5571,
      "mean_token_accuracy": 0.27025384157896043,
      "num_tokens": 133556696.0,
      "step": 58235
    },
    {
      "entropy": 5.6150126457214355,
      "epoch": 5.59462055715658,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002361535144695938,
      "loss": 4.7334,
      "mean_token_accuracy": 0.2503032863140106,
      "num_tokens": 133568074.0,
      "step": 58240
    },
    {
      "entropy": 5.4901782989501955,
      "epoch": 5.595100864553314,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023611974619949867,
      "loss": 4.6032,
      "mean_token_accuracy": 0.26705812066793444,
      "num_tokens": 133578863.0,
      "step": 58245
    },
    {
      "entropy": 5.427277088165283,
      "epoch": 5.595581171950048,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002360859788320875,
      "loss": 4.5298,
      "mean_token_accuracy": 0.2714715123176575,
      "num_tokens": 133591333.0,
      "step": 58250
    },
    {
      "entropy": 5.577503204345703,
      "epoch": 5.5960614793467816,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023605221236814433,
      "loss": 4.6805,
      "mean_token_accuracy": 0.26229190081357956,
      "num_tokens": 133603080.0,
      "step": 58255
    },
    {
      "entropy": 5.519261312484741,
      "epoch": 5.596541786743516,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002360184468084532,
      "loss": 4.6627,
      "mean_token_accuracy": 0.26170453876256944,
      "num_tokens": 133614244.0,
      "step": 58260
    },
    {
      "entropy": 5.5429243564605715,
      "epoch": 5.59702209414025,
      "grad_norm": 1.375,
      "learning_rate": 0.00023598468215379792,
      "loss": 4.6674,
      "mean_token_accuracy": 0.2570794567465782,
      "num_tokens": 133626894.0,
      "step": 58265
    },
    {
      "entropy": 5.583269166946411,
      "epoch": 5.597502401536984,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023595091840496247,
      "loss": 4.6729,
      "mean_token_accuracy": 0.26409668326377866,
      "num_tokens": 133638981.0,
      "step": 58270
    },
    {
      "entropy": 5.570314693450928,
      "epoch": 5.597982708933718,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002359171555627308,
      "loss": 4.7294,
      "mean_token_accuracy": 0.25959793776273726,
      "num_tokens": 133650307.0,
      "step": 58275
    },
    {
      "entropy": 5.570968532562256,
      "epoch": 5.598463016330451,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023588339362788668,
      "loss": 4.6327,
      "mean_token_accuracy": 0.26562267541885376,
      "num_tokens": 133662012.0,
      "step": 58280
    },
    {
      "entropy": 5.513527059555054,
      "epoch": 5.598943323727186,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023584963260121406,
      "loss": 4.5762,
      "mean_token_accuracy": 0.2705957368016243,
      "num_tokens": 133673297.0,
      "step": 58285
    },
    {
      "entropy": 5.453546142578125,
      "epoch": 5.599423631123919,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023581587248349667,
      "loss": 4.5999,
      "mean_token_accuracy": 0.26586751490831373,
      "num_tokens": 133685012.0,
      "step": 58290
    },
    {
      "entropy": 5.489684581756592,
      "epoch": 5.5999039385206535,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023578211327551836,
      "loss": 4.6768,
      "mean_token_accuracy": 0.2611288756132126,
      "num_tokens": 133698332.0,
      "step": 58295
    },
    {
      "entropy": 5.559364604949951,
      "epoch": 5.600384245917387,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002357483549780629,
      "loss": 4.6521,
      "mean_token_accuracy": 0.26298362463712693,
      "num_tokens": 133709379.0,
      "step": 58300
    },
    {
      "entropy": 5.565606641769409,
      "epoch": 5.600864553314121,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023571459759191423,
      "loss": 4.6392,
      "mean_token_accuracy": 0.2679352045059204,
      "num_tokens": 133721726.0,
      "step": 58305
    },
    {
      "entropy": 5.490341424942017,
      "epoch": 5.601344860710855,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002356808411178558,
      "loss": 4.6369,
      "mean_token_accuracy": 0.2682345360517502,
      "num_tokens": 133734507.0,
      "step": 58310
    },
    {
      "entropy": 5.533181381225586,
      "epoch": 5.601825168107589,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023564708555667152,
      "loss": 4.6271,
      "mean_token_accuracy": 0.26053420305252073,
      "num_tokens": 133747045.0,
      "step": 58315
    },
    {
      "entropy": 5.507663583755493,
      "epoch": 5.6023054755043225,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023561333090914505,
      "loss": 4.6168,
      "mean_token_accuracy": 0.2703883171081543,
      "num_tokens": 133758163.0,
      "step": 58320
    },
    {
      "entropy": 5.535913419723511,
      "epoch": 5.602785782901057,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023557957717606015,
      "loss": 4.5763,
      "mean_token_accuracy": 0.2666287004947662,
      "num_tokens": 133769002.0,
      "step": 58325
    },
    {
      "entropy": 5.4642377376556395,
      "epoch": 5.60326609029779,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002355458243582004,
      "loss": 4.6064,
      "mean_token_accuracy": 0.271717569231987,
      "num_tokens": 133780110.0,
      "step": 58330
    },
    {
      "entropy": 5.491817331314087,
      "epoch": 5.6037463976945245,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023551207245634944,
      "loss": 4.5747,
      "mean_token_accuracy": 0.2722327381372452,
      "num_tokens": 133791757.0,
      "step": 58335
    },
    {
      "entropy": 5.537147808074951,
      "epoch": 5.604226705091259,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023547832147129084,
      "loss": 4.6161,
      "mean_token_accuracy": 0.26498658359050753,
      "num_tokens": 133801749.0,
      "step": 58340
    },
    {
      "entropy": 5.573628711700439,
      "epoch": 5.604707012487992,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023544457140380827,
      "loss": 4.6891,
      "mean_token_accuracy": 0.2588072940707207,
      "num_tokens": 133813499.0,
      "step": 58345
    },
    {
      "entropy": 5.543308162689209,
      "epoch": 5.605187319884726,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023541082225468534,
      "loss": 4.65,
      "mean_token_accuracy": 0.2672663822770119,
      "num_tokens": 133824622.0,
      "step": 58350
    },
    {
      "entropy": 5.495896816253662,
      "epoch": 5.60566762728146,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023537707402470555,
      "loss": 4.6283,
      "mean_token_accuracy": 0.25937599390745164,
      "num_tokens": 133837534.0,
      "step": 58355
    },
    {
      "entropy": 5.546637678146363,
      "epoch": 5.606147934678194,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023534332671465243,
      "loss": 4.7049,
      "mean_token_accuracy": 0.2593609407544136,
      "num_tokens": 133849112.0,
      "step": 58360
    },
    {
      "entropy": 5.530694437026978,
      "epoch": 5.606628242074928,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023530958032530957,
      "loss": 4.586,
      "mean_token_accuracy": 0.2722094938158989,
      "num_tokens": 133859707.0,
      "step": 58365
    },
    {
      "entropy": 5.526483535766602,
      "epoch": 5.607108549471662,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023527583485746036,
      "loss": 4.6818,
      "mean_token_accuracy": 0.25413304567337036,
      "num_tokens": 133870536.0,
      "step": 58370
    },
    {
      "entropy": 5.518785715103149,
      "epoch": 5.607588856868396,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023524209031188832,
      "loss": 4.6517,
      "mean_token_accuracy": 0.25954684168100356,
      "num_tokens": 133883103.0,
      "step": 58375
    },
    {
      "entropy": 5.516021871566773,
      "epoch": 5.60806916426513,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023520834668937685,
      "loss": 4.6268,
      "mean_token_accuracy": 0.25798688679933546,
      "num_tokens": 133894312.0,
      "step": 58380
    },
    {
      "entropy": 5.544353199005127,
      "epoch": 5.608549471661863,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023517460399070938,
      "loss": 4.6453,
      "mean_token_accuracy": 0.2665943741798401,
      "num_tokens": 133903954.0,
      "step": 58385
    },
    {
      "entropy": 5.5736973762512205,
      "epoch": 5.609029779058598,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023514086221666943,
      "loss": 4.6793,
      "mean_token_accuracy": 0.25562783926725385,
      "num_tokens": 133914843.0,
      "step": 58390
    },
    {
      "entropy": 5.533066272735596,
      "epoch": 5.609510086455331,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002351071213680403,
      "loss": 4.6811,
      "mean_token_accuracy": 0.2564109191298485,
      "num_tokens": 133925720.0,
      "step": 58395
    },
    {
      "entropy": 5.572945880889892,
      "epoch": 5.609990393852065,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023507338144560536,
      "loss": 4.6969,
      "mean_token_accuracy": 0.24832436591386794,
      "num_tokens": 133937708.0,
      "step": 58400
    },
    {
      "entropy": 5.5673887729644775,
      "epoch": 5.610470701248799,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023503964245014791,
      "loss": 4.6602,
      "mean_token_accuracy": 0.2635057345032692,
      "num_tokens": 133949936.0,
      "step": 58405
    },
    {
      "entropy": 5.483131313323975,
      "epoch": 5.610951008645533,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002350059043824513,
      "loss": 4.6057,
      "mean_token_accuracy": 0.26301083117723467,
      "num_tokens": 133962470.0,
      "step": 58410
    },
    {
      "entropy": 5.532328033447266,
      "epoch": 5.6114313160422675,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023497216724329879,
      "loss": 4.6847,
      "mean_token_accuracy": 0.2636669516563416,
      "num_tokens": 133972834.0,
      "step": 58415
    },
    {
      "entropy": 5.489754915237427,
      "epoch": 5.611911623439001,
      "grad_norm": 1.25,
      "learning_rate": 0.00023493843103347374,
      "loss": 4.5914,
      "mean_token_accuracy": 0.2656494781374931,
      "num_tokens": 133984430.0,
      "step": 58420
    },
    {
      "entropy": 5.589459657669067,
      "epoch": 5.612391930835734,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023490469575375945,
      "loss": 4.7448,
      "mean_token_accuracy": 0.2540066674351692,
      "num_tokens": 133996369.0,
      "step": 58425
    },
    {
      "entropy": 5.642325496673584,
      "epoch": 5.612872238232469,
      "grad_norm": 1.5,
      "learning_rate": 0.00023487096140493896,
      "loss": 4.7642,
      "mean_token_accuracy": 0.251657734811306,
      "num_tokens": 134007109.0,
      "step": 58430
    },
    {
      "entropy": 5.558468437194824,
      "epoch": 5.613352545629203,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002348372279877956,
      "loss": 4.6368,
      "mean_token_accuracy": 0.2658550009131432,
      "num_tokens": 134018191.0,
      "step": 58435
    },
    {
      "entropy": 5.534220886230469,
      "epoch": 5.6138328530259365,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023480349550311253,
      "loss": 4.6811,
      "mean_token_accuracy": 0.25887381583452224,
      "num_tokens": 134031220.0,
      "step": 58440
    },
    {
      "entropy": 5.567817115783692,
      "epoch": 5.614313160422671,
      "grad_norm": 1.21875,
      "learning_rate": 0.000234769763951673,
      "loss": 4.6374,
      "mean_token_accuracy": 0.2642139866948128,
      "num_tokens": 134043935.0,
      "step": 58445
    },
    {
      "entropy": 5.563514947891235,
      "epoch": 5.614793467819404,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023473603333426003,
      "loss": 4.645,
      "mean_token_accuracy": 0.2629013046622276,
      "num_tokens": 134054857.0,
      "step": 58450
    },
    {
      "entropy": 5.529746770858765,
      "epoch": 5.615273775216139,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002347023036516569,
      "loss": 4.5853,
      "mean_token_accuracy": 0.2638189822435379,
      "num_tokens": 134065596.0,
      "step": 58455
    },
    {
      "entropy": 5.560053539276123,
      "epoch": 5.615754082612872,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023466857490464654,
      "loss": 4.7002,
      "mean_token_accuracy": 0.2638878211379051,
      "num_tokens": 134076983.0,
      "step": 58460
    },
    {
      "entropy": 5.626858234405518,
      "epoch": 5.616234390009606,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023463484709401213,
      "loss": 4.736,
      "mean_token_accuracy": 0.25427754670381547,
      "num_tokens": 134087301.0,
      "step": 58465
    },
    {
      "entropy": 5.556241798400879,
      "epoch": 5.61671469740634,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023460112022053665,
      "loss": 4.6504,
      "mean_token_accuracy": 0.26578222066164015,
      "num_tokens": 134099303.0,
      "step": 58470
    },
    {
      "entropy": 5.557968997955323,
      "epoch": 5.617195004803074,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023456739428500328,
      "loss": 4.6421,
      "mean_token_accuracy": 0.2648980230093002,
      "num_tokens": 134111407.0,
      "step": 58475
    },
    {
      "entropy": 5.619232320785523,
      "epoch": 5.6176753121998075,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023453366928819492,
      "loss": 4.7426,
      "mean_token_accuracy": 0.26076449602842333,
      "num_tokens": 134123414.0,
      "step": 58480
    },
    {
      "entropy": 5.584220933914184,
      "epoch": 5.618155619596542,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002344999452308947,
      "loss": 4.7014,
      "mean_token_accuracy": 0.2545753464102745,
      "num_tokens": 134135298.0,
      "step": 58485
    },
    {
      "entropy": 5.60770173072815,
      "epoch": 5.618635926993276,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023446622211388535,
      "loss": 4.7088,
      "mean_token_accuracy": 0.25698104649782183,
      "num_tokens": 134145789.0,
      "step": 58490
    },
    {
      "entropy": 5.55497374534607,
      "epoch": 5.61911623439001,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023443249993795003,
      "loss": 4.6502,
      "mean_token_accuracy": 0.26602408587932586,
      "num_tokens": 134157961.0,
      "step": 58495
    },
    {
      "entropy": 5.534629487991333,
      "epoch": 5.619596541786743,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023439877870387154,
      "loss": 4.6001,
      "mean_token_accuracy": 0.2665229752659798,
      "num_tokens": 134168732.0,
      "step": 58500
    },
    {
      "entropy": 5.48529577255249,
      "epoch": 5.620076849183477,
      "grad_norm": 1.375,
      "learning_rate": 0.00023436505841243287,
      "loss": 4.5905,
      "mean_token_accuracy": 0.2681429386138916,
      "num_tokens": 134180343.0,
      "step": 58505
    },
    {
      "entropy": 5.552632999420166,
      "epoch": 5.620557156580212,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002343313390644169,
      "loss": 4.7146,
      "mean_token_accuracy": 0.26056160032749176,
      "num_tokens": 134191650.0,
      "step": 58510
    },
    {
      "entropy": 5.556450462341308,
      "epoch": 5.621037463976945,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023429762066060655,
      "loss": 4.6669,
      "mean_token_accuracy": 0.26321672052145006,
      "num_tokens": 134204989.0,
      "step": 58515
    },
    {
      "entropy": 5.621333026885987,
      "epoch": 5.6215177713736795,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023426390320178454,
      "loss": 4.7079,
      "mean_token_accuracy": 0.2558848589658737,
      "num_tokens": 134215716.0,
      "step": 58520
    },
    {
      "entropy": 5.528059625625611,
      "epoch": 5.621998078770413,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023423018668873363,
      "loss": 4.6339,
      "mean_token_accuracy": 0.26500587314367297,
      "num_tokens": 134228855.0,
      "step": 58525
    },
    {
      "entropy": 5.4333148956298825,
      "epoch": 5.622478386167147,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002341964711222368,
      "loss": 4.5414,
      "mean_token_accuracy": 0.2683438196778297,
      "num_tokens": 134239763.0,
      "step": 58530
    },
    {
      "entropy": 5.4882715225219725,
      "epoch": 5.622958693563881,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002341627565030768,
      "loss": 4.621,
      "mean_token_accuracy": 0.26504161655902864,
      "num_tokens": 134252025.0,
      "step": 58535
    },
    {
      "entropy": 5.6407490253448485,
      "epoch": 5.623439000960615,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023412904283203622,
      "loss": 4.7045,
      "mean_token_accuracy": 0.25901315808296205,
      "num_tokens": 134263409.0,
      "step": 58540
    },
    {
      "entropy": 5.485868358612061,
      "epoch": 5.623919308357348,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023409533010989802,
      "loss": 4.6001,
      "mean_token_accuracy": 0.26429411619901655,
      "num_tokens": 134274600.0,
      "step": 58545
    },
    {
      "entropy": 5.533655071258545,
      "epoch": 5.624399615754083,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002340616183374447,
      "loss": 4.6848,
      "mean_token_accuracy": 0.25849513709545135,
      "num_tokens": 134287232.0,
      "step": 58550
    },
    {
      "entropy": 5.6026664733886715,
      "epoch": 5.624879923150816,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023402790751545912,
      "loss": 4.6711,
      "mean_token_accuracy": 0.25374543815851214,
      "num_tokens": 134297772.0,
      "step": 58555
    },
    {
      "entropy": 5.5554944515228275,
      "epoch": 5.6253602305475505,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023399419764472384,
      "loss": 4.732,
      "mean_token_accuracy": 0.252345535159111,
      "num_tokens": 134309690.0,
      "step": 58560
    },
    {
      "entropy": 5.529226875305175,
      "epoch": 5.625840537944284,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002339604887260215,
      "loss": 4.6429,
      "mean_token_accuracy": 0.2695130884647369,
      "num_tokens": 134322019.0,
      "step": 58565
    },
    {
      "entropy": 5.567668724060058,
      "epoch": 5.626320845341018,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023392678076013474,
      "loss": 4.6444,
      "mean_token_accuracy": 0.25834252387285234,
      "num_tokens": 134333386.0,
      "step": 58570
    },
    {
      "entropy": 5.621109104156494,
      "epoch": 5.626801152737752,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023389307374784625,
      "loss": 4.7056,
      "mean_token_accuracy": 0.25250357389450073,
      "num_tokens": 134344948.0,
      "step": 58575
    },
    {
      "entropy": 5.611515903472901,
      "epoch": 5.627281460134486,
      "grad_norm": 1.25,
      "learning_rate": 0.00023385936768993842,
      "loss": 4.7438,
      "mean_token_accuracy": 0.2486122354865074,
      "num_tokens": 134356528.0,
      "step": 58580
    },
    {
      "entropy": 5.578383207321167,
      "epoch": 5.62776176753122,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023382566258719397,
      "loss": 4.6409,
      "mean_token_accuracy": 0.26438616365194323,
      "num_tokens": 134368220.0,
      "step": 58585
    },
    {
      "entropy": 5.582461357116699,
      "epoch": 5.628242074927954,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002337919584403953,
      "loss": 4.7136,
      "mean_token_accuracy": 0.2556070610880852,
      "num_tokens": 134380179.0,
      "step": 58590
    },
    {
      "entropy": 5.447785186767578,
      "epoch": 5.628722382324688,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023375825525032512,
      "loss": 4.547,
      "mean_token_accuracy": 0.26878791451454165,
      "num_tokens": 134391104.0,
      "step": 58595
    },
    {
      "entropy": 5.502379989624023,
      "epoch": 5.629202689721422,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023372455301776574,
      "loss": 4.6592,
      "mean_token_accuracy": 0.2665447562932968,
      "num_tokens": 134402287.0,
      "step": 58600
    },
    {
      "entropy": 5.5691019058227536,
      "epoch": 5.629682997118156,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002336908517434997,
      "loss": 4.7249,
      "mean_token_accuracy": 0.2591984495520592,
      "num_tokens": 134414193.0,
      "step": 58605
    },
    {
      "entropy": 5.572839260101318,
      "epoch": 5.630163304514889,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023365715142830933,
      "loss": 4.657,
      "mean_token_accuracy": 0.26022653430700304,
      "num_tokens": 134425500.0,
      "step": 58610
    },
    {
      "entropy": 5.558956241607666,
      "epoch": 5.630643611911624,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002336234520729772,
      "loss": 4.67,
      "mean_token_accuracy": 0.25854495763778684,
      "num_tokens": 134436039.0,
      "step": 58615
    },
    {
      "entropy": 5.477845573425293,
      "epoch": 5.631123919308357,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002335897536782856,
      "loss": 4.5766,
      "mean_token_accuracy": 0.27065473049879074,
      "num_tokens": 134447510.0,
      "step": 58620
    },
    {
      "entropy": 5.575573492050171,
      "epoch": 5.631604226705091,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023355605624501697,
      "loss": 4.6612,
      "mean_token_accuracy": 0.26419746279716494,
      "num_tokens": 134458318.0,
      "step": 58625
    },
    {
      "entropy": 5.59671425819397,
      "epoch": 5.632084534101825,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002335223597739537,
      "loss": 4.6965,
      "mean_token_accuracy": 0.2536198228597641,
      "num_tokens": 134468762.0,
      "step": 58630
    },
    {
      "entropy": 5.503249597549439,
      "epoch": 5.632564841498559,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023348866426587813,
      "loss": 4.5895,
      "mean_token_accuracy": 0.2662723228335381,
      "num_tokens": 134480432.0,
      "step": 58635
    },
    {
      "entropy": 5.5755311965942385,
      "epoch": 5.633045148895293,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023345496972157243,
      "loss": 4.6702,
      "mean_token_accuracy": 0.258829402923584,
      "num_tokens": 134491834.0,
      "step": 58640
    },
    {
      "entropy": 5.605033826828003,
      "epoch": 5.633525456292027,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023342127614181895,
      "loss": 4.7274,
      "mean_token_accuracy": 0.25404011756181716,
      "num_tokens": 134502630.0,
      "step": 58645
    },
    {
      "entropy": 5.512335252761841,
      "epoch": 5.63400576368876,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023338758352739998,
      "loss": 4.546,
      "mean_token_accuracy": 0.27807681262493134,
      "num_tokens": 134514462.0,
      "step": 58650
    },
    {
      "entropy": 5.439809513092041,
      "epoch": 5.634486071085495,
      "grad_norm": 1.25,
      "learning_rate": 0.00023335389187909773,
      "loss": 4.5609,
      "mean_token_accuracy": 0.2652708664536476,
      "num_tokens": 134526877.0,
      "step": 58655
    },
    {
      "entropy": 5.538192796707153,
      "epoch": 5.634966378482229,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002333202011976945,
      "loss": 4.698,
      "mean_token_accuracy": 0.25244626551866534,
      "num_tokens": 134538531.0,
      "step": 58660
    },
    {
      "entropy": 5.495603322982788,
      "epoch": 5.6354466858789625,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002332865114839725,
      "loss": 4.6075,
      "mean_token_accuracy": 0.26610482335090635,
      "num_tokens": 134549647.0,
      "step": 58665
    },
    {
      "entropy": 5.600087594985962,
      "epoch": 5.635926993275697,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023325282273871373,
      "loss": 4.7397,
      "mean_token_accuracy": 0.2539355993270874,
      "num_tokens": 134560436.0,
      "step": 58670
    },
    {
      "entropy": 5.58133602142334,
      "epoch": 5.63640730067243,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023321913496270047,
      "loss": 4.6249,
      "mean_token_accuracy": 0.26516075879335405,
      "num_tokens": 134571913.0,
      "step": 58675
    },
    {
      "entropy": 5.515505456924439,
      "epoch": 5.636887608069165,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023318544815671482,
      "loss": 4.6043,
      "mean_token_accuracy": 0.2745036959648132,
      "num_tokens": 134583057.0,
      "step": 58680
    },
    {
      "entropy": 5.466983270645142,
      "epoch": 5.637367915465898,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023315176232153895,
      "loss": 4.6012,
      "mean_token_accuracy": 0.2656380131840706,
      "num_tokens": 134593822.0,
      "step": 58685
    },
    {
      "entropy": 5.566976881027221,
      "epoch": 5.637848222862632,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023311807745795489,
      "loss": 4.7567,
      "mean_token_accuracy": 0.2566147446632385,
      "num_tokens": 134605096.0,
      "step": 58690
    },
    {
      "entropy": 5.518844842910767,
      "epoch": 5.638328530259366,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023308439356674473,
      "loss": 4.591,
      "mean_token_accuracy": 0.2661324068903923,
      "num_tokens": 134616036.0,
      "step": 58695
    },
    {
      "entropy": 5.596222400665283,
      "epoch": 5.6388088376561,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002330507106486905,
      "loss": 4.6745,
      "mean_token_accuracy": 0.26239475011825564,
      "num_tokens": 134628508.0,
      "step": 58700
    },
    {
      "entropy": 5.587874984741211,
      "epoch": 5.6392891450528335,
      "grad_norm": 1.25,
      "learning_rate": 0.00023301702870457409,
      "loss": 4.707,
      "mean_token_accuracy": 0.25968085527420043,
      "num_tokens": 134640206.0,
      "step": 58705
    },
    {
      "entropy": 5.547712945938111,
      "epoch": 5.639769452449568,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023298334773517765,
      "loss": 4.6841,
      "mean_token_accuracy": 0.26543460637331007,
      "num_tokens": 134653220.0,
      "step": 58710
    },
    {
      "entropy": 5.456151056289673,
      "epoch": 5.640249759846301,
      "grad_norm": 1.375,
      "learning_rate": 0.00023294966774128312,
      "loss": 4.5704,
      "mean_token_accuracy": 0.2703174725174904,
      "num_tokens": 134663686.0,
      "step": 58715
    },
    {
      "entropy": 5.510142850875854,
      "epoch": 5.640730067243036,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023291598872367248,
      "loss": 4.6535,
      "mean_token_accuracy": 0.2611071541905403,
      "num_tokens": 134675097.0,
      "step": 58720
    },
    {
      "entropy": 5.565818738937378,
      "epoch": 5.641210374639769,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023288231068312765,
      "loss": 4.6793,
      "mean_token_accuracy": 0.2635465696454048,
      "num_tokens": 134686530.0,
      "step": 58725
    },
    {
      "entropy": 5.581372165679932,
      "epoch": 5.641690682036503,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023284863362043047,
      "loss": 4.6824,
      "mean_token_accuracy": 0.26231415420770643,
      "num_tokens": 134697941.0,
      "step": 58730
    },
    {
      "entropy": 5.485615873336792,
      "epoch": 5.642170989433238,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023281495753636278,
      "loss": 4.6046,
      "mean_token_accuracy": 0.2683085650205612,
      "num_tokens": 134709397.0,
      "step": 58735
    },
    {
      "entropy": 5.506842613220215,
      "epoch": 5.642651296829971,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023278128243170654,
      "loss": 4.6895,
      "mean_token_accuracy": 0.2606926545500755,
      "num_tokens": 134720407.0,
      "step": 58740
    },
    {
      "entropy": 5.578091669082641,
      "epoch": 5.6431316042267055,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023274760830724355,
      "loss": 4.6161,
      "mean_token_accuracy": 0.26151124984025953,
      "num_tokens": 134730301.0,
      "step": 58745
    },
    {
      "entropy": 5.508377170562744,
      "epoch": 5.643611911623439,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002327139351637557,
      "loss": 4.6361,
      "mean_token_accuracy": 0.26190863847732543,
      "num_tokens": 134741546.0,
      "step": 58750
    },
    {
      "entropy": 5.552152299880982,
      "epoch": 5.644092219020173,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023268026300202478,
      "loss": 4.7038,
      "mean_token_accuracy": 0.2564969018101692,
      "num_tokens": 134753706.0,
      "step": 58755
    },
    {
      "entropy": 5.577584218978882,
      "epoch": 5.644572526416907,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002326465918228323,
      "loss": 4.7079,
      "mean_token_accuracy": 0.25567236393690107,
      "num_tokens": 134764363.0,
      "step": 58760
    },
    {
      "entropy": 5.560195827484131,
      "epoch": 5.645052833813641,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023261292162696028,
      "loss": 4.7115,
      "mean_token_accuracy": 0.2628921553492546,
      "num_tokens": 134776025.0,
      "step": 58765
    },
    {
      "entropy": 5.47959156036377,
      "epoch": 5.645533141210374,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023257925241519035,
      "loss": 4.5721,
      "mean_token_accuracy": 0.27195852547883986,
      "num_tokens": 134787165.0,
      "step": 58770
    },
    {
      "entropy": 5.53886866569519,
      "epoch": 5.646013448607109,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023254558418830425,
      "loss": 4.6095,
      "mean_token_accuracy": 0.2701068907976151,
      "num_tokens": 134798808.0,
      "step": 58775
    },
    {
      "entropy": 5.487927579879761,
      "epoch": 5.646493756003842,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023251191694708358,
      "loss": 4.5706,
      "mean_token_accuracy": 0.2733853325247765,
      "num_tokens": 134809736.0,
      "step": 58780
    },
    {
      "entropy": 5.400809669494629,
      "epoch": 5.6469740634005765,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002324782506923101,
      "loss": 4.5448,
      "mean_token_accuracy": 0.2716837912797928,
      "num_tokens": 134820092.0,
      "step": 58785
    },
    {
      "entropy": 5.52943115234375,
      "epoch": 5.64745437079731,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023244458542476528,
      "loss": 4.6677,
      "mean_token_accuracy": 0.25581242740154264,
      "num_tokens": 134831604.0,
      "step": 58790
    },
    {
      "entropy": 5.545262622833252,
      "epoch": 5.647934678194044,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023241092114523087,
      "loss": 4.6391,
      "mean_token_accuracy": 0.26794775724411013,
      "num_tokens": 134841854.0,
      "step": 58795
    },
    {
      "entropy": 5.536431360244751,
      "epoch": 5.648414985590778,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002323772578544884,
      "loss": 4.6154,
      "mean_token_accuracy": 0.26337661892175673,
      "num_tokens": 134852035.0,
      "step": 58800
    },
    {
      "entropy": 5.578443241119385,
      "epoch": 5.648895292987512,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002323435955533194,
      "loss": 4.6713,
      "mean_token_accuracy": 0.24919258058071136,
      "num_tokens": 134863521.0,
      "step": 58805
    },
    {
      "entropy": 5.55066237449646,
      "epoch": 5.649375600384246,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023230993424250545,
      "loss": 4.6907,
      "mean_token_accuracy": 0.250778304040432,
      "num_tokens": 134875055.0,
      "step": 58810
    },
    {
      "entropy": 5.5240734100341795,
      "epoch": 5.64985590778098,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002322762739228282,
      "loss": 4.6442,
      "mean_token_accuracy": 0.26614863574504855,
      "num_tokens": 134885898.0,
      "step": 58815
    },
    {
      "entropy": 5.5712378978729244,
      "epoch": 5.650336215177714,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002322426145950689,
      "loss": 4.6834,
      "mean_token_accuracy": 0.2577270522713661,
      "num_tokens": 134897158.0,
      "step": 58820
    },
    {
      "entropy": 5.523777914047241,
      "epoch": 5.6508165225744476,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023220895626000916,
      "loss": 4.6472,
      "mean_token_accuracy": 0.2575756713747978,
      "num_tokens": 134908007.0,
      "step": 58825
    },
    {
      "entropy": 5.511372089385986,
      "epoch": 5.651296829971182,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002321752989184303,
      "loss": 4.6153,
      "mean_token_accuracy": 0.26197607070207596,
      "num_tokens": 134920193.0,
      "step": 58830
    },
    {
      "entropy": 5.487480401992798,
      "epoch": 5.651777137367915,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002321416425711139,
      "loss": 4.5977,
      "mean_token_accuracy": 0.27022740989923477,
      "num_tokens": 134932387.0,
      "step": 58835
    },
    {
      "entropy": 5.500899124145508,
      "epoch": 5.65225744476465,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002321079872188413,
      "loss": 4.6071,
      "mean_token_accuracy": 0.26637142449617385,
      "num_tokens": 134943265.0,
      "step": 58840
    },
    {
      "entropy": 5.550785112380981,
      "epoch": 5.652737752161383,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023207433286239387,
      "loss": 4.677,
      "mean_token_accuracy": 0.2653172940015793,
      "num_tokens": 134955700.0,
      "step": 58845
    },
    {
      "entropy": 5.493730258941651,
      "epoch": 5.653218059558117,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023204067950255297,
      "loss": 4.6032,
      "mean_token_accuracy": 0.2691135212779045,
      "num_tokens": 134967004.0,
      "step": 58850
    },
    {
      "entropy": 5.486024188995361,
      "epoch": 5.653698366954851,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002320070271400999,
      "loss": 4.6316,
      "mean_token_accuracy": 0.262347973883152,
      "num_tokens": 134978298.0,
      "step": 58855
    },
    {
      "entropy": 5.54178524017334,
      "epoch": 5.654178674351585,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023197337577581607,
      "loss": 4.7096,
      "mean_token_accuracy": 0.25571833550930023,
      "num_tokens": 134989827.0,
      "step": 58860
    },
    {
      "entropy": 5.594886207580567,
      "epoch": 5.654658981748319,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023193972541048265,
      "loss": 4.7521,
      "mean_token_accuracy": 0.25544709116220476,
      "num_tokens": 135000786.0,
      "step": 58865
    },
    {
      "entropy": 5.495329713821411,
      "epoch": 5.655139289145053,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002319060760448809,
      "loss": 4.6091,
      "mean_token_accuracy": 0.2640268087387085,
      "num_tokens": 135013002.0,
      "step": 58870
    },
    {
      "entropy": 5.583287477493286,
      "epoch": 5.655619596541786,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002318724276797922,
      "loss": 4.6812,
      "mean_token_accuracy": 0.2588468000292778,
      "num_tokens": 135023934.0,
      "step": 58875
    },
    {
      "entropy": 5.511196899414062,
      "epoch": 5.656099903938521,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023183878031599758,
      "loss": 4.6668,
      "mean_token_accuracy": 0.26048215925693513,
      "num_tokens": 135036442.0,
      "step": 58880
    },
    {
      "entropy": 5.596423006057739,
      "epoch": 5.656580211335255,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023180513395427833,
      "loss": 4.7528,
      "mean_token_accuracy": 0.2529721766710281,
      "num_tokens": 135046884.0,
      "step": 58885
    },
    {
      "entropy": 5.553429222106933,
      "epoch": 5.6570605187319885,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023177148859541565,
      "loss": 4.647,
      "mean_token_accuracy": 0.2596244916319847,
      "num_tokens": 135058464.0,
      "step": 58890
    },
    {
      "entropy": 5.4832850933074955,
      "epoch": 5.657540826128722,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023173784424019057,
      "loss": 4.5861,
      "mean_token_accuracy": 0.2679513871669769,
      "num_tokens": 135068957.0,
      "step": 58895
    },
    {
      "entropy": 5.609751415252686,
      "epoch": 5.658021133525456,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023170420088938437,
      "loss": 4.6837,
      "mean_token_accuracy": 0.2621353194117546,
      "num_tokens": 135079129.0,
      "step": 58900
    },
    {
      "entropy": 5.513518571853638,
      "epoch": 5.6585014409221905,
      "grad_norm": 1.375,
      "learning_rate": 0.00023167055854377805,
      "loss": 4.6382,
      "mean_token_accuracy": 0.2679383620619774,
      "num_tokens": 135090898.0,
      "step": 58905
    },
    {
      "entropy": 5.494038486480713,
      "epoch": 5.658981748318924,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002316369172041527,
      "loss": 4.6708,
      "mean_token_accuracy": 0.2618327736854553,
      "num_tokens": 135103335.0,
      "step": 58910
    },
    {
      "entropy": 5.623790740966797,
      "epoch": 5.659462055715658,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023160327687128928,
      "loss": 4.6862,
      "mean_token_accuracy": 0.25743458718061446,
      "num_tokens": 135114931.0,
      "step": 58915
    },
    {
      "entropy": 5.673884105682373,
      "epoch": 5.659942363112392,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000231569637545969,
      "loss": 4.8131,
      "mean_token_accuracy": 0.2437228173017502,
      "num_tokens": 135126626.0,
      "step": 58920
    },
    {
      "entropy": 5.496729516983033,
      "epoch": 5.660422670509126,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023153599922897273,
      "loss": 4.6786,
      "mean_token_accuracy": 0.25928735435009004,
      "num_tokens": 135138459.0,
      "step": 58925
    },
    {
      "entropy": 5.5434962749481205,
      "epoch": 5.6609029779058595,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023150236192108148,
      "loss": 4.6365,
      "mean_token_accuracy": 0.27101048827171326,
      "num_tokens": 135149318.0,
      "step": 58930
    },
    {
      "entropy": 5.518317413330078,
      "epoch": 5.661383285302594,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023146872562307636,
      "loss": 4.5833,
      "mean_token_accuracy": 0.26925148963928225,
      "num_tokens": 135160788.0,
      "step": 58935
    },
    {
      "entropy": 5.587093925476074,
      "epoch": 5.661863592699327,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023143509033573802,
      "loss": 4.7828,
      "mean_token_accuracy": 0.2534086465835571,
      "num_tokens": 135172065.0,
      "step": 58940
    },
    {
      "entropy": 5.5311534881591795,
      "epoch": 5.662343900096062,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023140145605984758,
      "loss": 4.657,
      "mean_token_accuracy": 0.26167190372943877,
      "num_tokens": 135182343.0,
      "step": 58945
    },
    {
      "entropy": 5.506945276260376,
      "epoch": 5.662824207492795,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023136782279618585,
      "loss": 4.5837,
      "mean_token_accuracy": 0.26820054203271865,
      "num_tokens": 135193333.0,
      "step": 58950
    },
    {
      "entropy": 5.554769897460938,
      "epoch": 5.663304514889529,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023133419054553367,
      "loss": 4.6449,
      "mean_token_accuracy": 0.2676737025380135,
      "num_tokens": 135204614.0,
      "step": 58955
    },
    {
      "entropy": 5.584950065612793,
      "epoch": 5.663784822286264,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000231300559308672,
      "loss": 4.7302,
      "mean_token_accuracy": 0.2543597504496574,
      "num_tokens": 135215353.0,
      "step": 58960
    },
    {
      "entropy": 5.529166841506958,
      "epoch": 5.664265129682997,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023126692908638163,
      "loss": 4.6658,
      "mean_token_accuracy": 0.2574093699455261,
      "num_tokens": 135226207.0,
      "step": 58965
    },
    {
      "entropy": 5.542881155014038,
      "epoch": 5.6647454370797306,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023123329987944326,
      "loss": 4.6993,
      "mean_token_accuracy": 0.26099939048290255,
      "num_tokens": 135239392.0,
      "step": 58970
    },
    {
      "entropy": 5.4808046340942385,
      "epoch": 5.665225744476465,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002311996716886376,
      "loss": 4.5976,
      "mean_token_accuracy": 0.2763412415981293,
      "num_tokens": 135251018.0,
      "step": 58975
    },
    {
      "entropy": 5.475394678115845,
      "epoch": 5.665706051873199,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002311660445147456,
      "loss": 4.5732,
      "mean_token_accuracy": 0.2744705885648727,
      "num_tokens": 135263479.0,
      "step": 58980
    },
    {
      "entropy": 5.515790033340454,
      "epoch": 5.666186359269933,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023113241835854786,
      "loss": 4.5954,
      "mean_token_accuracy": 0.27133776545524596,
      "num_tokens": 135273389.0,
      "step": 58985
    },
    {
      "entropy": 5.5727239608764645,
      "epoch": 5.666666666666667,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002310987932208251,
      "loss": 4.7104,
      "mean_token_accuracy": 0.25672830939292907,
      "num_tokens": 135285318.0,
      "step": 58990
    },
    {
      "entropy": 5.4857463359832765,
      "epoch": 5.6671469740634,
      "grad_norm": 1.25,
      "learning_rate": 0.0002310651691023581,
      "loss": 4.5932,
      "mean_token_accuracy": 0.27307109236717225,
      "num_tokens": 135297038.0,
      "step": 58995
    },
    {
      "entropy": 5.450390577316284,
      "epoch": 5.667627281460135,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002310315460039273,
      "loss": 4.5834,
      "mean_token_accuracy": 0.2671181261539459,
      "num_tokens": 135309190.0,
      "step": 59000
    },
    {
      "entropy": 5.540354633331299,
      "epoch": 5.668107588856868,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023099792392631348,
      "loss": 4.666,
      "mean_token_accuracy": 0.2621891856193542,
      "num_tokens": 135320351.0,
      "step": 59005
    },
    {
      "entropy": 5.620066404342651,
      "epoch": 5.6685878962536025,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023096430287029718,
      "loss": 4.7571,
      "mean_token_accuracy": 0.24925850927829743,
      "num_tokens": 135331061.0,
      "step": 59010
    },
    {
      "entropy": 5.460798883438111,
      "epoch": 5.669068203650336,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023093068283665903,
      "loss": 4.5388,
      "mean_token_accuracy": 0.27233996242284775,
      "num_tokens": 135342482.0,
      "step": 59015
    },
    {
      "entropy": 5.5199024200439455,
      "epoch": 5.66954851104707,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023089706382617954,
      "loss": 4.6707,
      "mean_token_accuracy": 0.25939868241548536,
      "num_tokens": 135354807.0,
      "step": 59020
    },
    {
      "entropy": 5.5450530529022215,
      "epoch": 5.670028818443804,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023086344583963937,
      "loss": 4.753,
      "mean_token_accuracy": 0.25413642078638077,
      "num_tokens": 135365922.0,
      "step": 59025
    },
    {
      "entropy": 5.5308233261108395,
      "epoch": 5.670509125840538,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023082982887781883,
      "loss": 4.6087,
      "mean_token_accuracy": 0.26556703001260756,
      "num_tokens": 135377393.0,
      "step": 59030
    },
    {
      "entropy": 5.480433177947998,
      "epoch": 5.670989433237272,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023079621294149854,
      "loss": 4.6229,
      "mean_token_accuracy": 0.26273987293243406,
      "num_tokens": 135390181.0,
      "step": 59035
    },
    {
      "entropy": 5.537689971923828,
      "epoch": 5.671469740634006,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023076259803145888,
      "loss": 4.6738,
      "mean_token_accuracy": 0.2632234439253807,
      "num_tokens": 135402283.0,
      "step": 59040
    },
    {
      "entropy": 5.530559158325195,
      "epoch": 5.671950048030739,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00023072898414848038,
      "loss": 4.6183,
      "mean_token_accuracy": 0.26905272752046583,
      "num_tokens": 135414330.0,
      "step": 59045
    },
    {
      "entropy": 5.5730931758880615,
      "epoch": 5.6724303554274735,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002306953712933434,
      "loss": 4.7349,
      "mean_token_accuracy": 0.25455537289381025,
      "num_tokens": 135425630.0,
      "step": 59050
    },
    {
      "entropy": 5.5922609806060795,
      "epoch": 5.672910662824208,
      "grad_norm": 1.375,
      "learning_rate": 0.00023066175946682838,
      "loss": 4.6952,
      "mean_token_accuracy": 0.2536448657512665,
      "num_tokens": 135437450.0,
      "step": 59055
    },
    {
      "entropy": 5.550420045852661,
      "epoch": 5.673390970220941,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023062814866971565,
      "loss": 4.6854,
      "mean_token_accuracy": 0.26301526427268984,
      "num_tokens": 135449711.0,
      "step": 59060
    },
    {
      "entropy": 5.528727674484253,
      "epoch": 5.673871277617676,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023059453890278547,
      "loss": 4.6068,
      "mean_token_accuracy": 0.261833643913269,
      "num_tokens": 135460632.0,
      "step": 59065
    },
    {
      "entropy": 5.518549680709839,
      "epoch": 5.674351585014409,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023056093016681829,
      "loss": 4.6632,
      "mean_token_accuracy": 0.2620416969060898,
      "num_tokens": 135472628.0,
      "step": 59070
    },
    {
      "entropy": 5.617204999923706,
      "epoch": 5.674831892411143,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002305273224625943,
      "loss": 4.6913,
      "mean_token_accuracy": 0.2575142696499825,
      "num_tokens": 135484266.0,
      "step": 59075
    },
    {
      "entropy": 5.548964595794677,
      "epoch": 5.675312199807877,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023049371579089383,
      "loss": 4.6059,
      "mean_token_accuracy": 0.2699275717139244,
      "num_tokens": 135494888.0,
      "step": 59080
    },
    {
      "entropy": 5.547208595275879,
      "epoch": 5.675792507204611,
      "grad_norm": 1.25,
      "learning_rate": 0.0002304601101524973,
      "loss": 4.6631,
      "mean_token_accuracy": 0.2639276638627052,
      "num_tokens": 135508914.0,
      "step": 59085
    },
    {
      "entropy": 5.5379101753234865,
      "epoch": 5.676272814601345,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023042650554818455,
      "loss": 4.6369,
      "mean_token_accuracy": 0.2674645408987999,
      "num_tokens": 135519694.0,
      "step": 59090
    },
    {
      "entropy": 5.535213279724121,
      "epoch": 5.676753121998079,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023039290197873608,
      "loss": 4.6272,
      "mean_token_accuracy": 0.2634248986840248,
      "num_tokens": 135530688.0,
      "step": 59095
    },
    {
      "entropy": 5.692473077774048,
      "epoch": 5.677233429394812,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023035929944493195,
      "loss": 4.8192,
      "mean_token_accuracy": 0.25126246362924576,
      "num_tokens": 135540581.0,
      "step": 59100
    },
    {
      "entropy": 5.576046276092529,
      "epoch": 5.677713736791547,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023032569794755227,
      "loss": 4.6754,
      "mean_token_accuracy": 0.2543090909719467,
      "num_tokens": 135551384.0,
      "step": 59105
    },
    {
      "entropy": 5.601438903808594,
      "epoch": 5.678194044188281,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023029209748737727,
      "loss": 4.6907,
      "mean_token_accuracy": 0.2601005434989929,
      "num_tokens": 135562940.0,
      "step": 59110
    },
    {
      "entropy": 5.561152935028076,
      "epoch": 5.678674351585014,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023025849806518705,
      "loss": 4.6424,
      "mean_token_accuracy": 0.2737951725721359,
      "num_tokens": 135573794.0,
      "step": 59115
    },
    {
      "entropy": 5.575155782699585,
      "epoch": 5.679154658981748,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023022489968176164,
      "loss": 4.6831,
      "mean_token_accuracy": 0.25617878288030627,
      "num_tokens": 135584829.0,
      "step": 59120
    },
    {
      "entropy": 5.588682079315186,
      "epoch": 5.679634966378482,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002301913023378811,
      "loss": 4.6864,
      "mean_token_accuracy": 0.26039777845144274,
      "num_tokens": 135595755.0,
      "step": 59125
    },
    {
      "entropy": 5.4799158573150635,
      "epoch": 5.6801152737752165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023015770603432544,
      "loss": 4.4792,
      "mean_token_accuracy": 0.28029502332210543,
      "num_tokens": 135607475.0,
      "step": 59130
    },
    {
      "entropy": 5.549148321151733,
      "epoch": 5.68059558117195,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023012411077187473,
      "loss": 4.7064,
      "mean_token_accuracy": 0.250676728785038,
      "num_tokens": 135619560.0,
      "step": 59135
    },
    {
      "entropy": 5.515261554718018,
      "epoch": 5.681075888568684,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023009051655130885,
      "loss": 4.6749,
      "mean_token_accuracy": 0.25921235233545303,
      "num_tokens": 135631144.0,
      "step": 59140
    },
    {
      "entropy": 5.476147222518921,
      "epoch": 5.681556195965418,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023005692337340795,
      "loss": 4.6121,
      "mean_token_accuracy": 0.26919296085834504,
      "num_tokens": 135642911.0,
      "step": 59145
    },
    {
      "entropy": 5.53043155670166,
      "epoch": 5.682036503362152,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002300233312389517,
      "loss": 4.6405,
      "mean_token_accuracy": 0.26325685530900955,
      "num_tokens": 135655455.0,
      "step": 59150
    },
    {
      "entropy": 5.5307872772216795,
      "epoch": 5.6825168107588855,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022998974014872014,
      "loss": 4.623,
      "mean_token_accuracy": 0.27052478939294816,
      "num_tokens": 135666820.0,
      "step": 59155
    },
    {
      "entropy": 5.524213743209839,
      "epoch": 5.68299711815562,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022995615010349324,
      "loss": 4.6098,
      "mean_token_accuracy": 0.2686907395720482,
      "num_tokens": 135677788.0,
      "step": 59160
    },
    {
      "entropy": 5.492296409606934,
      "epoch": 5.683477425552353,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002299225611040508,
      "loss": 4.6518,
      "mean_token_accuracy": 0.2631725251674652,
      "num_tokens": 135690940.0,
      "step": 59165
    },
    {
      "entropy": 5.51195387840271,
      "epoch": 5.683957732949088,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022988897315117258,
      "loss": 4.6639,
      "mean_token_accuracy": 0.26373572647571564,
      "num_tokens": 135703239.0,
      "step": 59170
    },
    {
      "entropy": 5.601049327850342,
      "epoch": 5.684438040345821,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002298553862456385,
      "loss": 4.7407,
      "mean_token_accuracy": 0.2539308100938797,
      "num_tokens": 135715470.0,
      "step": 59175
    },
    {
      "entropy": 5.563876485824585,
      "epoch": 5.684918347742555,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022982180038822825,
      "loss": 4.6124,
      "mean_token_accuracy": 0.26479952186346056,
      "num_tokens": 135726125.0,
      "step": 59180
    },
    {
      "entropy": 5.51803731918335,
      "epoch": 5.685398655139289,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022978821557972164,
      "loss": 4.6288,
      "mean_token_accuracy": 0.26788348853588106,
      "num_tokens": 135737468.0,
      "step": 59185
    },
    {
      "entropy": 5.519423484802246,
      "epoch": 5.685878962536023,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002297546318208984,
      "loss": 4.6411,
      "mean_token_accuracy": 0.2615630358457565,
      "num_tokens": 135749409.0,
      "step": 59190
    },
    {
      "entropy": 5.501232719421386,
      "epoch": 5.6863592699327565,
      "grad_norm": 1.375,
      "learning_rate": 0.00022972104911253826,
      "loss": 4.6006,
      "mean_token_accuracy": 0.26818478405475615,
      "num_tokens": 135760850.0,
      "step": 59195
    },
    {
      "entropy": 5.552948522567749,
      "epoch": 5.686839577329491,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022968746745542095,
      "loss": 4.647,
      "mean_token_accuracy": 0.2643791973590851,
      "num_tokens": 135773219.0,
      "step": 59200
    },
    {
      "entropy": 5.569003868103027,
      "epoch": 5.687319884726225,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022965388685032613,
      "loss": 4.7767,
      "mean_token_accuracy": 0.2535826101899147,
      "num_tokens": 135784907.0,
      "step": 59205
    },
    {
      "entropy": 5.646407222747802,
      "epoch": 5.687800192122959,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022962030729803341,
      "loss": 4.773,
      "mean_token_accuracy": 0.2551949471235275,
      "num_tokens": 135795500.0,
      "step": 59210
    },
    {
      "entropy": 5.520957517623901,
      "epoch": 5.688280499519693,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002295867287993223,
      "loss": 4.5924,
      "mean_token_accuracy": 0.26792881786823275,
      "num_tokens": 135808289.0,
      "step": 59215
    },
    {
      "entropy": 5.4416588306427,
      "epoch": 5.688760806916426,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022955315135497257,
      "loss": 4.5928,
      "mean_token_accuracy": 0.27053644955158235,
      "num_tokens": 135820551.0,
      "step": 59220
    },
    {
      "entropy": 5.546892786026001,
      "epoch": 5.689241114313161,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022951957496576364,
      "loss": 4.585,
      "mean_token_accuracy": 0.2665901407599449,
      "num_tokens": 135833358.0,
      "step": 59225
    },
    {
      "entropy": 5.528283166885376,
      "epoch": 5.689721421709894,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022948599963247524,
      "loss": 4.6188,
      "mean_token_accuracy": 0.2662872835993767,
      "num_tokens": 135845726.0,
      "step": 59230
    },
    {
      "entropy": 5.563202953338623,
      "epoch": 5.6902017291066285,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002294524253558867,
      "loss": 4.7103,
      "mean_token_accuracy": 0.2598783686757088,
      "num_tokens": 135858344.0,
      "step": 59235
    },
    {
      "entropy": 5.5598931312561035,
      "epoch": 5.690682036503362,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002294188521367777,
      "loss": 4.7068,
      "mean_token_accuracy": 0.2530687257647514,
      "num_tokens": 135869230.0,
      "step": 59240
    },
    {
      "entropy": 5.590678024291992,
      "epoch": 5.691162343900096,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002293852799759275,
      "loss": 4.7334,
      "mean_token_accuracy": 0.2548114001750946,
      "num_tokens": 135880626.0,
      "step": 59245
    },
    {
      "entropy": 5.605448532104492,
      "epoch": 5.69164265129683,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022935170887411566,
      "loss": 4.7446,
      "mean_token_accuracy": 0.25855765491724014,
      "num_tokens": 135892660.0,
      "step": 59250
    },
    {
      "entropy": 5.578502178192139,
      "epoch": 5.692122958693564,
      "grad_norm": 1.25,
      "learning_rate": 0.0002293181388321216,
      "loss": 4.6468,
      "mean_token_accuracy": 0.26131755113601685,
      "num_tokens": 135903227.0,
      "step": 59255
    },
    {
      "entropy": 5.536616945266724,
      "epoch": 5.692603266090297,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022928456985072472,
      "loss": 4.6552,
      "mean_token_accuracy": 0.260118106007576,
      "num_tokens": 135913385.0,
      "step": 59260
    },
    {
      "entropy": 5.5541479110717775,
      "epoch": 5.693083573487032,
      "grad_norm": 1.375,
      "learning_rate": 0.00022925100193070435,
      "loss": 4.7242,
      "mean_token_accuracy": 0.2598662510514259,
      "num_tokens": 135923878.0,
      "step": 59265
    },
    {
      "entropy": 5.553419589996338,
      "epoch": 5.693563880883765,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022921743507284,
      "loss": 4.7285,
      "mean_token_accuracy": 0.25826555490493774,
      "num_tokens": 135934529.0,
      "step": 59270
    },
    {
      "entropy": 5.571819543838501,
      "epoch": 5.6940441882804995,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022918386927791074,
      "loss": 4.6598,
      "mean_token_accuracy": 0.2591456577181816,
      "num_tokens": 135946243.0,
      "step": 59275
    },
    {
      "entropy": 5.529419755935669,
      "epoch": 5.694524495677234,
      "grad_norm": 1.328125,
      "learning_rate": 0.000229150304546696,
      "loss": 4.6646,
      "mean_token_accuracy": 0.2569028168916702,
      "num_tokens": 135957961.0,
      "step": 59280
    },
    {
      "entropy": 5.530768728256225,
      "epoch": 5.695004803073967,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022911674087997504,
      "loss": 4.6879,
      "mean_token_accuracy": 0.26010646373033525,
      "num_tokens": 135968640.0,
      "step": 59285
    },
    {
      "entropy": 5.592040586471557,
      "epoch": 5.695485110470702,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022908317827852716,
      "loss": 4.7122,
      "mean_token_accuracy": 0.2533531799912453,
      "num_tokens": 135979296.0,
      "step": 59290
    },
    {
      "entropy": 5.649946880340576,
      "epoch": 5.695965417867435,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022904961674313147,
      "loss": 4.8288,
      "mean_token_accuracy": 0.24376347661018372,
      "num_tokens": 135990076.0,
      "step": 59295
    },
    {
      "entropy": 5.521974658966064,
      "epoch": 5.696445725264169,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022901605627456735,
      "loss": 4.6266,
      "mean_token_accuracy": 0.2703837096691132,
      "num_tokens": 136001405.0,
      "step": 59300
    },
    {
      "entropy": 5.564229202270508,
      "epoch": 5.696926032660903,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022898249687361378,
      "loss": 4.6756,
      "mean_token_accuracy": 0.2598623365163803,
      "num_tokens": 136012891.0,
      "step": 59305
    },
    {
      "entropy": 5.602036571502685,
      "epoch": 5.697406340057637,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022894893854104992,
      "loss": 4.6244,
      "mean_token_accuracy": 0.2580180257558823,
      "num_tokens": 136023568.0,
      "step": 59310
    },
    {
      "entropy": 5.5985993385314945,
      "epoch": 5.697886647454371,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022891538127765498,
      "loss": 4.7637,
      "mean_token_accuracy": 0.2517858386039734,
      "num_tokens": 136036674.0,
      "step": 59315
    },
    {
      "entropy": 5.532587289810181,
      "epoch": 5.698366954851105,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022888182508420813,
      "loss": 4.6541,
      "mean_token_accuracy": 0.25555074363946917,
      "num_tokens": 136047323.0,
      "step": 59320
    },
    {
      "entropy": 5.512013101577759,
      "epoch": 5.698847262247838,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022884826996148828,
      "loss": 4.61,
      "mean_token_accuracy": 0.2725240051746368,
      "num_tokens": 136060182.0,
      "step": 59325
    },
    {
      "entropy": 5.494438934326172,
      "epoch": 5.699327569644573,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002288147159102747,
      "loss": 4.6206,
      "mean_token_accuracy": 0.26547207683324814,
      "num_tokens": 136071350.0,
      "step": 59330
    },
    {
      "entropy": 5.496056175231933,
      "epoch": 5.699807877041306,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022878116293134615,
      "loss": 4.6242,
      "mean_token_accuracy": 0.2636860594153404,
      "num_tokens": 136083726.0,
      "step": 59335
    },
    {
      "entropy": 5.5314970970153805,
      "epoch": 5.70028818443804,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022874761102548174,
      "loss": 4.6375,
      "mean_token_accuracy": 0.26557891219854357,
      "num_tokens": 136095630.0,
      "step": 59340
    },
    {
      "entropy": 5.535858631134033,
      "epoch": 5.700768491834774,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022871406019346052,
      "loss": 4.6641,
      "mean_token_accuracy": 0.2621223613619804,
      "num_tokens": 136107810.0,
      "step": 59345
    },
    {
      "entropy": 5.522668504714966,
      "epoch": 5.701248799231508,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022868051043606135,
      "loss": 4.5491,
      "mean_token_accuracy": 0.2729734107851982,
      "num_tokens": 136117762.0,
      "step": 59350
    },
    {
      "entropy": 5.510346221923828,
      "epoch": 5.7017291066282425,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022864696175406318,
      "loss": 4.6444,
      "mean_token_accuracy": 0.2606217786669731,
      "num_tokens": 136129196.0,
      "step": 59355
    },
    {
      "entropy": 5.585376834869384,
      "epoch": 5.702209414024976,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022861341414824504,
      "loss": 4.6948,
      "mean_token_accuracy": 0.25301364809274673,
      "num_tokens": 136140317.0,
      "step": 59360
    },
    {
      "entropy": 5.594020414352417,
      "epoch": 5.70268972142171,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022857986761938554,
      "loss": 4.753,
      "mean_token_accuracy": 0.2571012616157532,
      "num_tokens": 136153404.0,
      "step": 59365
    },
    {
      "entropy": 5.526051187515259,
      "epoch": 5.703170028818444,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002285463221682637,
      "loss": 4.6529,
      "mean_token_accuracy": 0.264057494699955,
      "num_tokens": 136167904.0,
      "step": 59370
    },
    {
      "entropy": 5.528785657882691,
      "epoch": 5.703650336215178,
      "grad_norm": 1.59375,
      "learning_rate": 0.00022851277779565833,
      "loss": 4.6679,
      "mean_token_accuracy": 0.2613625884056091,
      "num_tokens": 136179566.0,
      "step": 59375
    },
    {
      "entropy": 5.513956499099732,
      "epoch": 5.7041306436119115,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002284792345023482,
      "loss": 4.6286,
      "mean_token_accuracy": 0.27667711973190307,
      "num_tokens": 136190513.0,
      "step": 59380
    },
    {
      "entropy": 5.584860324859619,
      "epoch": 5.704610951008646,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002284456922891121,
      "loss": 4.69,
      "mean_token_accuracy": 0.2544790580868721,
      "num_tokens": 136201728.0,
      "step": 59385
    },
    {
      "entropy": 5.465823221206665,
      "epoch": 5.705091258405379,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022841215115672887,
      "loss": 4.5715,
      "mean_token_accuracy": 0.2716005861759186,
      "num_tokens": 136213170.0,
      "step": 59390
    },
    {
      "entropy": 5.527931547164917,
      "epoch": 5.7055715658021136,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022837861110597713,
      "loss": 4.6586,
      "mean_token_accuracy": 0.2601944446563721,
      "num_tokens": 136224347.0,
      "step": 59395
    },
    {
      "entropy": 5.527190113067627,
      "epoch": 5.706051873198847,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022834507213763557,
      "loss": 4.5991,
      "mean_token_accuracy": 0.2667007610201836,
      "num_tokens": 136235660.0,
      "step": 59400
    },
    {
      "entropy": 5.497847700119019,
      "epoch": 5.706532180595581,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022831153425248286,
      "loss": 4.6743,
      "mean_token_accuracy": 0.26411198228597643,
      "num_tokens": 136246680.0,
      "step": 59405
    },
    {
      "entropy": 5.6069605350494385,
      "epoch": 5.707012487992315,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022827799745129773,
      "loss": 4.7673,
      "mean_token_accuracy": 0.25072233080863954,
      "num_tokens": 136259895.0,
      "step": 59410
    },
    {
      "entropy": 5.637871170043946,
      "epoch": 5.707492795389049,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022824446173485874,
      "loss": 4.7536,
      "mean_token_accuracy": 0.26088034361600876,
      "num_tokens": 136271157.0,
      "step": 59415
    },
    {
      "entropy": 5.53734974861145,
      "epoch": 5.7079731027857825,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002282109271039446,
      "loss": 4.6242,
      "mean_token_accuracy": 0.26325763463974,
      "num_tokens": 136284309.0,
      "step": 59420
    },
    {
      "entropy": 5.612868309020996,
      "epoch": 5.708453410182517,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002281773935593337,
      "loss": 4.7342,
      "mean_token_accuracy": 0.26012481600046156,
      "num_tokens": 136295722.0,
      "step": 59425
    },
    {
      "entropy": 5.5276079177856445,
      "epoch": 5.708933717579251,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022814386110180468,
      "loss": 4.6785,
      "mean_token_accuracy": 0.261335189640522,
      "num_tokens": 136306951.0,
      "step": 59430
    },
    {
      "entropy": 5.592543125152588,
      "epoch": 5.709414024975985,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002281103297321362,
      "loss": 4.7106,
      "mean_token_accuracy": 0.2611114203929901,
      "num_tokens": 136316837.0,
      "step": 59435
    },
    {
      "entropy": 5.544697093963623,
      "epoch": 5.709894332372719,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002280767994511065,
      "loss": 4.6298,
      "mean_token_accuracy": 0.2602734714746475,
      "num_tokens": 136328320.0,
      "step": 59440
    },
    {
      "entropy": 5.486643075942993,
      "epoch": 5.710374639769452,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022804327025949424,
      "loss": 4.6254,
      "mean_token_accuracy": 0.2606617137789726,
      "num_tokens": 136341216.0,
      "step": 59445
    },
    {
      "entropy": 5.412200212478638,
      "epoch": 5.710854947166187,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002280097421580778,
      "loss": 4.5178,
      "mean_token_accuracy": 0.2815089479088783,
      "num_tokens": 136352720.0,
      "step": 59450
    },
    {
      "entropy": 5.542259693145752,
      "epoch": 5.71133525456292,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022797621514763562,
      "loss": 4.636,
      "mean_token_accuracy": 0.2610208198428154,
      "num_tokens": 136364105.0,
      "step": 59455
    },
    {
      "entropy": 5.494031810760498,
      "epoch": 5.7118155619596545,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022794268922894607,
      "loss": 4.5737,
      "mean_token_accuracy": 0.2720492839813232,
      "num_tokens": 136375582.0,
      "step": 59460
    },
    {
      "entropy": 5.5041707992553714,
      "epoch": 5.712295869356388,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022790916440278752,
      "loss": 4.6282,
      "mean_token_accuracy": 0.2534725576639175,
      "num_tokens": 136387388.0,
      "step": 59465
    },
    {
      "entropy": 5.4966799259185795,
      "epoch": 5.712776176753122,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022787564066993838,
      "loss": 4.6612,
      "mean_token_accuracy": 0.2629344806075096,
      "num_tokens": 136398692.0,
      "step": 59470
    },
    {
      "entropy": 5.445368051528931,
      "epoch": 5.713256484149856,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022784211803117695,
      "loss": 4.5695,
      "mean_token_accuracy": 0.2710867628455162,
      "num_tokens": 136410008.0,
      "step": 59475
    },
    {
      "entropy": 5.529805326461792,
      "epoch": 5.71373679154659,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022780859648728152,
      "loss": 4.6302,
      "mean_token_accuracy": 0.2660615399479866,
      "num_tokens": 136420463.0,
      "step": 59480
    },
    {
      "entropy": 5.4803338050842285,
      "epoch": 5.714217098943323,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022777507603903032,
      "loss": 4.6469,
      "mean_token_accuracy": 0.26718112379312514,
      "num_tokens": 136431526.0,
      "step": 59485
    },
    {
      "entropy": 5.6063696384429935,
      "epoch": 5.714697406340058,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022774155668720165,
      "loss": 4.7335,
      "mean_token_accuracy": 0.256295645236969,
      "num_tokens": 136442989.0,
      "step": 59490
    },
    {
      "entropy": 5.509661674499512,
      "epoch": 5.715177713736791,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022770803843257366,
      "loss": 4.6218,
      "mean_token_accuracy": 0.26056852340698244,
      "num_tokens": 136455046.0,
      "step": 59495
    },
    {
      "entropy": 5.625962734222412,
      "epoch": 5.7156580211335255,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022767452127592458,
      "loss": 4.792,
      "mean_token_accuracy": 0.25129829049110414,
      "num_tokens": 136465905.0,
      "step": 59500
    },
    {
      "entropy": 5.545570755004883,
      "epoch": 5.71613832853026,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002276410052180326,
      "loss": 4.6669,
      "mean_token_accuracy": 0.2609337612986565,
      "num_tokens": 136476562.0,
      "step": 59505
    },
    {
      "entropy": 5.536478757858276,
      "epoch": 5.716618635926993,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000227607490259676,
      "loss": 4.6036,
      "mean_token_accuracy": 0.2722732752561569,
      "num_tokens": 136487816.0,
      "step": 59510
    },
    {
      "entropy": 5.534974193572998,
      "epoch": 5.717098943323727,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022757397640163257,
      "loss": 4.6679,
      "mean_token_accuracy": 0.2606647238135338,
      "num_tokens": 136499725.0,
      "step": 59515
    },
    {
      "entropy": 5.402845573425293,
      "epoch": 5.717579250720461,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022754046364468062,
      "loss": 4.5046,
      "mean_token_accuracy": 0.2768641859292984,
      "num_tokens": 136510919.0,
      "step": 59520
    },
    {
      "entropy": 5.535604810714721,
      "epoch": 5.718059558117195,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002275069519895982,
      "loss": 4.6677,
      "mean_token_accuracy": 0.25280829817056655,
      "num_tokens": 136522904.0,
      "step": 59525
    },
    {
      "entropy": 5.564654684066772,
      "epoch": 5.718539865513929,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022747344143716325,
      "loss": 4.7381,
      "mean_token_accuracy": 0.2581572592258453,
      "num_tokens": 136533764.0,
      "step": 59530
    },
    {
      "entropy": 5.482711458206177,
      "epoch": 5.719020172910663,
      "grad_norm": 1.375,
      "learning_rate": 0.00022743993198815393,
      "loss": 4.609,
      "mean_token_accuracy": 0.2675626024603844,
      "num_tokens": 136545684.0,
      "step": 59535
    },
    {
      "entropy": 5.542083311080932,
      "epoch": 5.7195004803073966,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002274064236433483,
      "loss": 4.7056,
      "mean_token_accuracy": 0.253683365881443,
      "num_tokens": 136557257.0,
      "step": 59540
    },
    {
      "entropy": 5.60399284362793,
      "epoch": 5.719980787704131,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022737291640352402,
      "loss": 4.6787,
      "mean_token_accuracy": 0.2647817894816399,
      "num_tokens": 136568858.0,
      "step": 59545
    },
    {
      "entropy": 5.574991273880005,
      "epoch": 5.720461095100864,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022733941026945926,
      "loss": 4.6719,
      "mean_token_accuracy": 0.2598370790481567,
      "num_tokens": 136581399.0,
      "step": 59550
    },
    {
      "entropy": 5.519377136230469,
      "epoch": 5.720941402497599,
      "grad_norm": 1.375,
      "learning_rate": 0.00022730590524193184,
      "loss": 4.6274,
      "mean_token_accuracy": 0.2624533027410507,
      "num_tokens": 136591981.0,
      "step": 59555
    },
    {
      "entropy": 5.647849082946777,
      "epoch": 5.721421709894332,
      "grad_norm": 1.25,
      "learning_rate": 0.00022727240132171966,
      "loss": 4.7822,
      "mean_token_accuracy": 0.2513081967830658,
      "num_tokens": 136603481.0,
      "step": 59560
    },
    {
      "entropy": 5.566029834747314,
      "epoch": 5.721902017291066,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022723889850960062,
      "loss": 4.669,
      "mean_token_accuracy": 0.2600757107138634,
      "num_tokens": 136615492.0,
      "step": 59565
    },
    {
      "entropy": 5.505286836624146,
      "epoch": 5.7223823246878,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002272053968063526,
      "loss": 4.6023,
      "mean_token_accuracy": 0.25991097986698153,
      "num_tokens": 136626728.0,
      "step": 59570
    },
    {
      "entropy": 5.45164680480957,
      "epoch": 5.722862632084534,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022717189621275326,
      "loss": 4.6061,
      "mean_token_accuracy": 0.27013989090919494,
      "num_tokens": 136637352.0,
      "step": 59575
    },
    {
      "entropy": 5.6457915782928465,
      "epoch": 5.7233429394812685,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022713839672958047,
      "loss": 4.806,
      "mean_token_accuracy": 0.2460642486810684,
      "num_tokens": 136650505.0,
      "step": 59580
    },
    {
      "entropy": 5.595197629928589,
      "epoch": 5.723823246878002,
      "grad_norm": 1.375,
      "learning_rate": 0.000227104898357612,
      "loss": 4.6968,
      "mean_token_accuracy": 0.25070288181304934,
      "num_tokens": 136661262.0,
      "step": 59585
    },
    {
      "entropy": 5.5474800109863285,
      "epoch": 5.724303554274735,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002270714010976256,
      "loss": 4.6331,
      "mean_token_accuracy": 0.2662283331155777,
      "num_tokens": 136673820.0,
      "step": 59590
    },
    {
      "entropy": 5.512255144119263,
      "epoch": 5.72478386167147,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022703790495039888,
      "loss": 4.6436,
      "mean_token_accuracy": 0.2685066550970078,
      "num_tokens": 136684831.0,
      "step": 59595
    },
    {
      "entropy": 5.529880809783935,
      "epoch": 5.725264169068204,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022700440991670973,
      "loss": 4.7012,
      "mean_token_accuracy": 0.256246754527092,
      "num_tokens": 136696756.0,
      "step": 59600
    },
    {
      "entropy": 5.55484676361084,
      "epoch": 5.7257444764649374,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002269709159973356,
      "loss": 4.687,
      "mean_token_accuracy": 0.2612074017524719,
      "num_tokens": 136707833.0,
      "step": 59605
    },
    {
      "entropy": 5.556168746948242,
      "epoch": 5.726224783861672,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022693742319305417,
      "loss": 4.6719,
      "mean_token_accuracy": 0.25873939841985705,
      "num_tokens": 136719477.0,
      "step": 59610
    },
    {
      "entropy": 5.5168904781341555,
      "epoch": 5.726705091258405,
      "grad_norm": 1.390625,
      "learning_rate": 0.000226903931504643,
      "loss": 4.6121,
      "mean_token_accuracy": 0.2646530032157898,
      "num_tokens": 136730821.0,
      "step": 59615
    },
    {
      "entropy": 5.550039339065552,
      "epoch": 5.7271853986551395,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022687044093287978,
      "loss": 4.7165,
      "mean_token_accuracy": 0.25829011499881743,
      "num_tokens": 136741841.0,
      "step": 59620
    },
    {
      "entropy": 5.491195249557495,
      "epoch": 5.727665706051873,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022683695147854193,
      "loss": 4.6372,
      "mean_token_accuracy": 0.27197618782520294,
      "num_tokens": 136752237.0,
      "step": 59625
    },
    {
      "entropy": 5.553032779693604,
      "epoch": 5.728146013448607,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022680346314240724,
      "loss": 4.6736,
      "mean_token_accuracy": 0.2681678235530853,
      "num_tokens": 136762620.0,
      "step": 59630
    },
    {
      "entropy": 5.533388090133667,
      "epoch": 5.728626320845341,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002267699759252529,
      "loss": 4.5738,
      "mean_token_accuracy": 0.2680268809199333,
      "num_tokens": 136773372.0,
      "step": 59635
    },
    {
      "entropy": 5.517199850082397,
      "epoch": 5.729106628242075,
      "grad_norm": 1.375,
      "learning_rate": 0.00022673648982785645,
      "loss": 4.6005,
      "mean_token_accuracy": 0.27212434113025663,
      "num_tokens": 136783294.0,
      "step": 59640
    },
    {
      "entropy": 5.520990037918091,
      "epoch": 5.7295869356388085,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022670300485099547,
      "loss": 4.6431,
      "mean_token_accuracy": 0.26495826691389085,
      "num_tokens": 136794577.0,
      "step": 59645
    },
    {
      "entropy": 5.467129564285278,
      "epoch": 5.730067243035543,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022666952099544723,
      "loss": 4.6008,
      "mean_token_accuracy": 0.26826412975788116,
      "num_tokens": 136806311.0,
      "step": 59650
    },
    {
      "entropy": 5.539089584350586,
      "epoch": 5.730547550432277,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022663603826198926,
      "loss": 4.6475,
      "mean_token_accuracy": 0.2608026683330536,
      "num_tokens": 136817789.0,
      "step": 59655
    },
    {
      "entropy": 5.566218328475952,
      "epoch": 5.731027857829011,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022660255665139895,
      "loss": 4.6653,
      "mean_token_accuracy": 0.2663218230009079,
      "num_tokens": 136829827.0,
      "step": 59660
    },
    {
      "entropy": 5.567947244644165,
      "epoch": 5.731508165225744,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022656907616445344,
      "loss": 4.6805,
      "mean_token_accuracy": 0.2642814815044403,
      "num_tokens": 136841365.0,
      "step": 59665
    },
    {
      "entropy": 5.454946231842041,
      "epoch": 5.731988472622478,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002265355968019302,
      "loss": 4.6088,
      "mean_token_accuracy": 0.2647373408079147,
      "num_tokens": 136852317.0,
      "step": 59670
    },
    {
      "entropy": 5.439703512191772,
      "epoch": 5.732468780019213,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022650211856460652,
      "loss": 4.5556,
      "mean_token_accuracy": 0.2755389377474785,
      "num_tokens": 136863056.0,
      "step": 59675
    },
    {
      "entropy": 5.446885299682617,
      "epoch": 5.732949087415946,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022646864145325958,
      "loss": 4.5756,
      "mean_token_accuracy": 0.2704134970903397,
      "num_tokens": 136875130.0,
      "step": 59680
    },
    {
      "entropy": 5.585152244567871,
      "epoch": 5.73342939481268,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022643516546866672,
      "loss": 4.6692,
      "mean_token_accuracy": 0.2601618096232414,
      "num_tokens": 136886896.0,
      "step": 59685
    },
    {
      "entropy": 5.503285217285156,
      "epoch": 5.733909702209414,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022640169061160514,
      "loss": 4.5751,
      "mean_token_accuracy": 0.26484357863664626,
      "num_tokens": 136897899.0,
      "step": 59690
    },
    {
      "entropy": 5.622876977920532,
      "epoch": 5.734390009606148,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002263682168828519,
      "loss": 4.7207,
      "mean_token_accuracy": 0.24974274635314941,
      "num_tokens": 136908775.0,
      "step": 59695
    },
    {
      "entropy": 5.5204321384429935,
      "epoch": 5.734870317002882,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002263347442831844,
      "loss": 4.6539,
      "mean_token_accuracy": 0.25471890568733213,
      "num_tokens": 136920259.0,
      "step": 59700
    },
    {
      "entropy": 5.52758641242981,
      "epoch": 5.735350624399616,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022630127281337954,
      "loss": 4.6321,
      "mean_token_accuracy": 0.26825437545776365,
      "num_tokens": 136931937.0,
      "step": 59705
    },
    {
      "entropy": 5.609677505493164,
      "epoch": 5.735830931796349,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022626780247421451,
      "loss": 4.7546,
      "mean_token_accuracy": 0.2579349517822266,
      "num_tokens": 136942943.0,
      "step": 59710
    },
    {
      "entropy": 5.679599285125732,
      "epoch": 5.736311239193084,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022623433326646641,
      "loss": 4.7808,
      "mean_token_accuracy": 0.24943675547838212,
      "num_tokens": 136953866.0,
      "step": 59715
    },
    {
      "entropy": 5.550866413116455,
      "epoch": 5.736791546589817,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022620086519091232,
      "loss": 4.6689,
      "mean_token_accuracy": 0.2584585607051849,
      "num_tokens": 136966196.0,
      "step": 59720
    },
    {
      "entropy": 5.557474517822266,
      "epoch": 5.7372718539865515,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022616739824832922,
      "loss": 4.733,
      "mean_token_accuracy": 0.2539238095283508,
      "num_tokens": 136976814.0,
      "step": 59725
    },
    {
      "entropy": 5.495530796051026,
      "epoch": 5.737752161383285,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002261339324394941,
      "loss": 4.563,
      "mean_token_accuracy": 0.26438906490802766,
      "num_tokens": 136988038.0,
      "step": 59730
    },
    {
      "entropy": 5.519497966766357,
      "epoch": 5.738232468780019,
      "grad_norm": 1.375,
      "learning_rate": 0.000226100467765184,
      "loss": 4.6273,
      "mean_token_accuracy": 0.26784211546182635,
      "num_tokens": 137000378.0,
      "step": 59735
    },
    {
      "entropy": 5.504461193084717,
      "epoch": 5.738712776176753,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022606700422617582,
      "loss": 4.5896,
      "mean_token_accuracy": 0.2687113583087921,
      "num_tokens": 137012595.0,
      "step": 59740
    },
    {
      "entropy": 5.594076347351074,
      "epoch": 5.739193083573487,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022603354182324653,
      "loss": 4.6868,
      "mean_token_accuracy": 0.2568551301956177,
      "num_tokens": 137024525.0,
      "step": 59745
    },
    {
      "entropy": 5.575265693664551,
      "epoch": 5.739673390970221,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022600008055717304,
      "loss": 4.6972,
      "mean_token_accuracy": 0.26215072721242905,
      "num_tokens": 137036221.0,
      "step": 59750
    },
    {
      "entropy": 5.645642566680908,
      "epoch": 5.740153698366955,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022596662042873207,
      "loss": 4.7608,
      "mean_token_accuracy": 0.2541129678487778,
      "num_tokens": 137046801.0,
      "step": 59755
    },
    {
      "entropy": 5.585306406021118,
      "epoch": 5.740634005763689,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022593316143870062,
      "loss": 4.7294,
      "mean_token_accuracy": 0.25440781861543654,
      "num_tokens": 137057801.0,
      "step": 59760
    },
    {
      "entropy": 5.542938089370727,
      "epoch": 5.7411143131604225,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022589970358785543,
      "loss": 4.6281,
      "mean_token_accuracy": 0.2648919329047203,
      "num_tokens": 137069082.0,
      "step": 59765
    },
    {
      "entropy": 5.532667016983032,
      "epoch": 5.741594620557157,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022586624687697337,
      "loss": 4.6866,
      "mean_token_accuracy": 0.26404026299715044,
      "num_tokens": 137079597.0,
      "step": 59770
    },
    {
      "entropy": 5.590613174438476,
      "epoch": 5.74207492795389,
      "grad_norm": 1.5,
      "learning_rate": 0.0002258327913068311,
      "loss": 4.7711,
      "mean_token_accuracy": 0.24501720070838928,
      "num_tokens": 137089888.0,
      "step": 59775
    },
    {
      "entropy": 5.465330028533936,
      "epoch": 5.742555235350625,
      "grad_norm": 1.25,
      "learning_rate": 0.00022579933687820559,
      "loss": 4.5195,
      "mean_token_accuracy": 0.27310440987348555,
      "num_tokens": 137101385.0,
      "step": 59780
    },
    {
      "entropy": 5.571900701522827,
      "epoch": 5.743035542747358,
      "grad_norm": 1.25,
      "learning_rate": 0.00022576588359187323,
      "loss": 4.6845,
      "mean_token_accuracy": 0.26261693686246873,
      "num_tokens": 137113407.0,
      "step": 59785
    },
    {
      "entropy": 5.545949172973633,
      "epoch": 5.743515850144092,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002257324314486109,
      "loss": 4.6275,
      "mean_token_accuracy": 0.2636706531047821,
      "num_tokens": 137124713.0,
      "step": 59790
    },
    {
      "entropy": 5.529363965988159,
      "epoch": 5.743996157540826,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022569898044919517,
      "loss": 4.5763,
      "mean_token_accuracy": 0.2618082031607628,
      "num_tokens": 137136215.0,
      "step": 59795
    },
    {
      "entropy": 5.474983835220337,
      "epoch": 5.74447646493756,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022566553059440275,
      "loss": 4.5938,
      "mean_token_accuracy": 0.27029612809419634,
      "num_tokens": 137147682.0,
      "step": 59800
    },
    {
      "entropy": 5.591192293167114,
      "epoch": 5.744956772334294,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022563208188501022,
      "loss": 4.7151,
      "mean_token_accuracy": 0.2589975565671921,
      "num_tokens": 137159608.0,
      "step": 59805
    },
    {
      "entropy": 5.599391460418701,
      "epoch": 5.745437079731028,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022559863432179423,
      "loss": 4.7654,
      "mean_token_accuracy": 0.2528683051466942,
      "num_tokens": 137172432.0,
      "step": 59810
    },
    {
      "entropy": 5.622009801864624,
      "epoch": 5.745917387127761,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022556518790553117,
      "loss": 4.7043,
      "mean_token_accuracy": 0.2581489637494087,
      "num_tokens": 137183443.0,
      "step": 59815
    },
    {
      "entropy": 5.594105672836304,
      "epoch": 5.746397694524496,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022553174263699767,
      "loss": 4.7062,
      "mean_token_accuracy": 0.25623814612627027,
      "num_tokens": 137194894.0,
      "step": 59820
    },
    {
      "entropy": 5.589142179489135,
      "epoch": 5.74687800192123,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022549829851697018,
      "loss": 4.703,
      "mean_token_accuracy": 0.260096350312233,
      "num_tokens": 137206711.0,
      "step": 59825
    },
    {
      "entropy": 5.563353872299194,
      "epoch": 5.747358309317963,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022546485554622526,
      "loss": 4.6929,
      "mean_token_accuracy": 0.2577813223004341,
      "num_tokens": 137217537.0,
      "step": 59830
    },
    {
      "entropy": 5.509007740020752,
      "epoch": 5.747838616714698,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022543141372553926,
      "loss": 4.5776,
      "mean_token_accuracy": 0.2665846422314644,
      "num_tokens": 137228897.0,
      "step": 59835
    },
    {
      "entropy": 5.585885858535766,
      "epoch": 5.748318924111431,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022539797305568877,
      "loss": 4.7069,
      "mean_token_accuracy": 0.2535820722579956,
      "num_tokens": 137239521.0,
      "step": 59840
    },
    {
      "entropy": 5.576646041870117,
      "epoch": 5.7487992315081655,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022536453353744996,
      "loss": 4.7033,
      "mean_token_accuracy": 0.2619184359908104,
      "num_tokens": 137250457.0,
      "step": 59845
    },
    {
      "entropy": 5.51714129447937,
      "epoch": 5.749279538904899,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022533109517159928,
      "loss": 4.6038,
      "mean_token_accuracy": 0.2672759473323822,
      "num_tokens": 137261088.0,
      "step": 59850
    },
    {
      "entropy": 5.580772352218628,
      "epoch": 5.749759846301633,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002252976579589131,
      "loss": 4.7301,
      "mean_token_accuracy": 0.25287689864635465,
      "num_tokens": 137271944.0,
      "step": 59855
    },
    {
      "entropy": 5.515393781661987,
      "epoch": 5.750240153698367,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002252642219001677,
      "loss": 4.5684,
      "mean_token_accuracy": 0.26798263043165205,
      "num_tokens": 137282339.0,
      "step": 59860
    },
    {
      "entropy": 5.516173315048218,
      "epoch": 5.750720461095101,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022523078699613936,
      "loss": 4.5801,
      "mean_token_accuracy": 0.2704198703169823,
      "num_tokens": 137293592.0,
      "step": 59865
    },
    {
      "entropy": 5.525580501556396,
      "epoch": 5.7512007684918345,
      "grad_norm": 1.375,
      "learning_rate": 0.00022519735324760449,
      "loss": 4.6528,
      "mean_token_accuracy": 0.2589230969548225,
      "num_tokens": 137304592.0,
      "step": 59870
    },
    {
      "entropy": 5.649562501907349,
      "epoch": 5.751681075888569,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002251639206553391,
      "loss": 4.748,
      "mean_token_accuracy": 0.25310464203357697,
      "num_tokens": 137315954.0,
      "step": 59875
    },
    {
      "entropy": 5.542063808441162,
      "epoch": 5.752161383285302,
      "grad_norm": 1.25,
      "learning_rate": 0.00022513048922011947,
      "loss": 4.6436,
      "mean_token_accuracy": 0.26133197396993635,
      "num_tokens": 137329007.0,
      "step": 59880
    },
    {
      "entropy": 5.586709117889404,
      "epoch": 5.752641690682037,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002250970589427218,
      "loss": 4.6372,
      "mean_token_accuracy": 0.26630851030349734,
      "num_tokens": 137339392.0,
      "step": 59885
    },
    {
      "entropy": 5.551407718658448,
      "epoch": 5.75312199807877,
      "grad_norm": 1.375,
      "learning_rate": 0.00022506362982392225,
      "loss": 4.6512,
      "mean_token_accuracy": 0.26360631734132767,
      "num_tokens": 137351694.0,
      "step": 59890
    },
    {
      "entropy": 5.60185751914978,
      "epoch": 5.753602305475504,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022503020186449697,
      "loss": 4.6843,
      "mean_token_accuracy": 0.268539260327816,
      "num_tokens": 137362395.0,
      "step": 59895
    },
    {
      "entropy": 5.556674003601074,
      "epoch": 5.754082612872239,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022499677506522208,
      "loss": 4.6072,
      "mean_token_accuracy": 0.27014352530241015,
      "num_tokens": 137373821.0,
      "step": 59900
    },
    {
      "entropy": 5.480094909667969,
      "epoch": 5.754562920268972,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002249633494268735,
      "loss": 4.5993,
      "mean_token_accuracy": 0.2666168242692947,
      "num_tokens": 137384515.0,
      "step": 59905
    },
    {
      "entropy": 5.572589874267578,
      "epoch": 5.755043227665706,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002249299249502274,
      "loss": 4.7185,
      "mean_token_accuracy": 0.26178363561630247,
      "num_tokens": 137396135.0,
      "step": 59910
    },
    {
      "entropy": 5.561922931671143,
      "epoch": 5.75552353506244,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022489650163605974,
      "loss": 4.6714,
      "mean_token_accuracy": 0.2650203362107277,
      "num_tokens": 137407634.0,
      "step": 59915
    },
    {
      "entropy": 5.572980737686157,
      "epoch": 5.756003842459174,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002248630794851465,
      "loss": 4.6993,
      "mean_token_accuracy": 0.25881229639053344,
      "num_tokens": 137418723.0,
      "step": 59920
    },
    {
      "entropy": 5.534493255615234,
      "epoch": 5.756484149855908,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022482965849826377,
      "loss": 4.6327,
      "mean_token_accuracy": 0.2598041519522667,
      "num_tokens": 137431463.0,
      "step": 59925
    },
    {
      "entropy": 5.537967681884766,
      "epoch": 5.756964457252642,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022479623867618738,
      "loss": 4.7262,
      "mean_token_accuracy": 0.25628088861703874,
      "num_tokens": 137444028.0,
      "step": 59930
    },
    {
      "entropy": 5.62020525932312,
      "epoch": 5.757444764649375,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022476282001969322,
      "loss": 4.7515,
      "mean_token_accuracy": 0.24910563230514526,
      "num_tokens": 137456204.0,
      "step": 59935
    },
    {
      "entropy": 5.615010118484497,
      "epoch": 5.75792507204611,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002247294025295572,
      "loss": 4.6882,
      "mean_token_accuracy": 0.25744840055704116,
      "num_tokens": 137467549.0,
      "step": 59940
    },
    {
      "entropy": 5.51017394065857,
      "epoch": 5.758405379442843,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022469598620655518,
      "loss": 4.5206,
      "mean_token_accuracy": 0.2729744344949722,
      "num_tokens": 137478451.0,
      "step": 59945
    },
    {
      "entropy": 5.47112078666687,
      "epoch": 5.7588856868395775,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000224662571051463,
      "loss": 4.6369,
      "mean_token_accuracy": 0.26479736119508746,
      "num_tokens": 137490488.0,
      "step": 59950
    },
    {
      "entropy": 5.52380404472351,
      "epoch": 5.759365994236311,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002246291570650564,
      "loss": 4.7056,
      "mean_token_accuracy": 0.26082976460456847,
      "num_tokens": 137501339.0,
      "step": 59955
    },
    {
      "entropy": 5.572801542282105,
      "epoch": 5.759846301633045,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002245957442481113,
      "loss": 4.7234,
      "mean_token_accuracy": 0.25053015649318694,
      "num_tokens": 137513434.0,
      "step": 59960
    },
    {
      "entropy": 5.636617851257324,
      "epoch": 5.760326609029779,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022456233260140324,
      "loss": 4.6878,
      "mean_token_accuracy": 0.2525313705205917,
      "num_tokens": 137524086.0,
      "step": 59965
    },
    {
      "entropy": 5.606710052490234,
      "epoch": 5.760806916426513,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022452892212570814,
      "loss": 4.7374,
      "mean_token_accuracy": 0.25619740039110184,
      "num_tokens": 137535904.0,
      "step": 59970
    },
    {
      "entropy": 5.622787380218506,
      "epoch": 5.761287223823247,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022449551282180153,
      "loss": 4.7123,
      "mean_token_accuracy": 0.25380403995513917,
      "num_tokens": 137548383.0,
      "step": 59975
    },
    {
      "entropy": 5.528550052642823,
      "epoch": 5.761767531219981,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002244621046904592,
      "loss": 4.6119,
      "mean_token_accuracy": 0.2674755722284317,
      "num_tokens": 137560613.0,
      "step": 59980
    },
    {
      "entropy": 5.571861028671265,
      "epoch": 5.762247838616715,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022442869773245673,
      "loss": 4.7279,
      "mean_token_accuracy": 0.2525425538420677,
      "num_tokens": 137572168.0,
      "step": 59985
    },
    {
      "entropy": 5.512009239196777,
      "epoch": 5.7627281460134485,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002243952919485697,
      "loss": 4.6154,
      "mean_token_accuracy": 0.27115142792463304,
      "num_tokens": 137583680.0,
      "step": 59990
    },
    {
      "entropy": 5.577354383468628,
      "epoch": 5.763208453410183,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022436188733957374,
      "loss": 4.6388,
      "mean_token_accuracy": 0.25621008425951003,
      "num_tokens": 137593439.0,
      "step": 59995
    },
    {
      "entropy": 5.575342512130737,
      "epoch": 5.763688760806916,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022432848390624438,
      "loss": 4.7077,
      "mean_token_accuracy": 0.261722220480442,
      "num_tokens": 137604987.0,
      "step": 60000
    },
    {
      "epoch": 5.763688760806916,
      "eval_entropy": 5.24082108418108,
      "eval_loss": 4.761427879333496,
      "eval_mean_token_accuracy": 0.2644300068370355,
      "eval_num_tokens": 137604987.0,
      "eval_runtime": 20.4306,
      "eval_samples_per_second": 1606.17,
      "eval_steps_per_second": 200.777,
      "step": 60000
    },
    {
      "entropy": 5.4598595142364506,
      "epoch": 5.764169068203651,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022429508164935712,
      "loss": 4.595,
      "mean_token_accuracy": 0.26980979442596437,
      "num_tokens": 137618165.0,
      "step": 60005
    },
    {
      "entropy": 5.570543050765991,
      "epoch": 5.764649375600384,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002242616805696876,
      "loss": 4.6913,
      "mean_token_accuracy": 0.26092699766159055,
      "num_tokens": 137631255.0,
      "step": 60010
    },
    {
      "entropy": 5.5196386814117435,
      "epoch": 5.765129682997118,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022422828066801117,
      "loss": 4.6776,
      "mean_token_accuracy": 0.26342906057834625,
      "num_tokens": 137642213.0,
      "step": 60015
    },
    {
      "entropy": 5.5405622005462645,
      "epoch": 5.765609990393852,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022419488194510328,
      "loss": 4.6718,
      "mean_token_accuracy": 0.25845237225294115,
      "num_tokens": 137652906.0,
      "step": 60020
    },
    {
      "entropy": 5.576576614379883,
      "epoch": 5.766090297790586,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022416148440173938,
      "loss": 4.6914,
      "mean_token_accuracy": 0.2553845375776291,
      "num_tokens": 137663614.0,
      "step": 60025
    },
    {
      "entropy": 5.565575122833252,
      "epoch": 5.76657060518732,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022412808803869483,
      "loss": 4.6672,
      "mean_token_accuracy": 0.2579899623990059,
      "num_tokens": 137675101.0,
      "step": 60030
    },
    {
      "entropy": 5.443623065948486,
      "epoch": 5.767050912584054,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022409469285674505,
      "loss": 4.5672,
      "mean_token_accuracy": 0.27034310102462766,
      "num_tokens": 137686164.0,
      "step": 60035
    },
    {
      "entropy": 5.51698899269104,
      "epoch": 5.767531219980787,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022406129885666525,
      "loss": 4.7159,
      "mean_token_accuracy": 0.2537114545702934,
      "num_tokens": 137697230.0,
      "step": 60040
    },
    {
      "entropy": 5.527403259277344,
      "epoch": 5.768011527377522,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002240279060392309,
      "loss": 4.6844,
      "mean_token_accuracy": 0.25371563583612444,
      "num_tokens": 137708853.0,
      "step": 60045
    },
    {
      "entropy": 5.58591513633728,
      "epoch": 5.768491834774256,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022399451440521735,
      "loss": 4.7493,
      "mean_token_accuracy": 0.2483350321650505,
      "num_tokens": 137720042.0,
      "step": 60050
    },
    {
      "entropy": 5.587172174453736,
      "epoch": 5.768972142170989,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022396112395539958,
      "loss": 4.6833,
      "mean_token_accuracy": 0.25489669740200044,
      "num_tokens": 137731096.0,
      "step": 60055
    },
    {
      "entropy": 5.481684780120849,
      "epoch": 5.769452449567723,
      "grad_norm": 1.375,
      "learning_rate": 0.00022392773469055296,
      "loss": 4.5564,
      "mean_token_accuracy": 0.2743601530790329,
      "num_tokens": 137742414.0,
      "step": 60060
    },
    {
      "entropy": 5.546535634994507,
      "epoch": 5.769932756964457,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022389434661145263,
      "loss": 4.6498,
      "mean_token_accuracy": 0.26647877842187884,
      "num_tokens": 137753968.0,
      "step": 60065
    },
    {
      "entropy": 5.515543413162232,
      "epoch": 5.7704130643611915,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022386095971887394,
      "loss": 4.576,
      "mean_token_accuracy": 0.2658779963850975,
      "num_tokens": 137766491.0,
      "step": 60070
    },
    {
      "entropy": 5.473645257949829,
      "epoch": 5.770893371757925,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022382757401359182,
      "loss": 4.6048,
      "mean_token_accuracy": 0.27227957993745805,
      "num_tokens": 137777580.0,
      "step": 60075
    },
    {
      "entropy": 5.5042859554290775,
      "epoch": 5.771373679154659,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002237941894963816,
      "loss": 4.6649,
      "mean_token_accuracy": 0.26749461591243745,
      "num_tokens": 137789266.0,
      "step": 60080
    },
    {
      "entropy": 5.529291677474975,
      "epoch": 5.771853986551393,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022376080616801815,
      "loss": 4.6517,
      "mean_token_accuracy": 0.26201701164245605,
      "num_tokens": 137800132.0,
      "step": 60085
    },
    {
      "entropy": 5.551974582672119,
      "epoch": 5.772334293948127,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022372742402927665,
      "loss": 4.6726,
      "mean_token_accuracy": 0.2567804306745529,
      "num_tokens": 137812159.0,
      "step": 60090
    },
    {
      "entropy": 5.569275951385498,
      "epoch": 5.7728146013448605,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002236940430809321,
      "loss": 4.8009,
      "mean_token_accuracy": 0.25188003927469255,
      "num_tokens": 137823996.0,
      "step": 60095
    },
    {
      "entropy": 5.542619323730468,
      "epoch": 5.773294908741595,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002236606633237595,
      "loss": 4.6615,
      "mean_token_accuracy": 0.25854751765727996,
      "num_tokens": 137835598.0,
      "step": 60100
    },
    {
      "entropy": 5.516955995559693,
      "epoch": 5.773775216138328,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002236272847585339,
      "loss": 4.5931,
      "mean_token_accuracy": 0.2679128885269165,
      "num_tokens": 137847927.0,
      "step": 60105
    },
    {
      "entropy": 5.589599990844727,
      "epoch": 5.7742555235350626,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002235939073860303,
      "loss": 4.6708,
      "mean_token_accuracy": 0.26071606427431104,
      "num_tokens": 137859540.0,
      "step": 60110
    },
    {
      "entropy": 5.579064607620239,
      "epoch": 5.774735830931796,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022356053120702342,
      "loss": 4.7384,
      "mean_token_accuracy": 0.25542480796575545,
      "num_tokens": 137870957.0,
      "step": 60115
    },
    {
      "entropy": 5.629011964797973,
      "epoch": 5.77521613832853,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022352715622228826,
      "loss": 4.7612,
      "mean_token_accuracy": 0.25402874052524566,
      "num_tokens": 137882469.0,
      "step": 60120
    },
    {
      "entropy": 5.585124444961548,
      "epoch": 5.775696445725265,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022349378243259965,
      "loss": 4.6891,
      "mean_token_accuracy": 0.2557017683982849,
      "num_tokens": 137893423.0,
      "step": 60125
    },
    {
      "entropy": 5.525625085830688,
      "epoch": 5.776176753121998,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002234604098387325,
      "loss": 4.6386,
      "mean_token_accuracy": 0.26594640165567396,
      "num_tokens": 137904834.0,
      "step": 60130
    },
    {
      "entropy": 5.58522629737854,
      "epoch": 5.7766570605187315,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002234270384414616,
      "loss": 4.7157,
      "mean_token_accuracy": 0.2628664433956146,
      "num_tokens": 137916238.0,
      "step": 60135
    },
    {
      "entropy": 5.591012239456177,
      "epoch": 5.777137367915466,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022339366824156176,
      "loss": 4.7118,
      "mean_token_accuracy": 0.259591780602932,
      "num_tokens": 137928117.0,
      "step": 60140
    },
    {
      "entropy": 5.538519620895386,
      "epoch": 5.7776176753122,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002233602992398077,
      "loss": 4.6241,
      "mean_token_accuracy": 0.26840680539608003,
      "num_tokens": 137939839.0,
      "step": 60145
    },
    {
      "entropy": 5.53415322303772,
      "epoch": 5.778097982708934,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022332693143697414,
      "loss": 4.6248,
      "mean_token_accuracy": 0.2640207216143608,
      "num_tokens": 137951551.0,
      "step": 60150
    },
    {
      "entropy": 5.592848825454712,
      "epoch": 5.778578290105668,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022329356483383571,
      "loss": 4.7274,
      "mean_token_accuracy": 0.25333368480205537,
      "num_tokens": 137963787.0,
      "step": 60155
    },
    {
      "entropy": 5.595285892486572,
      "epoch": 5.779058597502401,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022326019943116727,
      "loss": 4.6869,
      "mean_token_accuracy": 0.26121118366718293,
      "num_tokens": 137975601.0,
      "step": 60160
    },
    {
      "entropy": 5.571305131912231,
      "epoch": 5.779538904899136,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022322683522974334,
      "loss": 4.6974,
      "mean_token_accuracy": 0.25713431686162946,
      "num_tokens": 137986961.0,
      "step": 60165
    },
    {
      "entropy": 5.532755422592163,
      "epoch": 5.780019212295869,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022319347223033866,
      "loss": 4.5732,
      "mean_token_accuracy": 0.2699199080467224,
      "num_tokens": 137997839.0,
      "step": 60170
    },
    {
      "entropy": 5.556400632858276,
      "epoch": 5.7804995196926034,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002231601104337276,
      "loss": 4.6079,
      "mean_token_accuracy": 0.26559394896030425,
      "num_tokens": 138009024.0,
      "step": 60175
    },
    {
      "entropy": 5.52777738571167,
      "epoch": 5.780979827089337,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022312674984068483,
      "loss": 4.579,
      "mean_token_accuracy": 0.26199119687080386,
      "num_tokens": 138020626.0,
      "step": 60180
    },
    {
      "entropy": 5.538431835174561,
      "epoch": 5.781460134486071,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002230933904519849,
      "loss": 4.684,
      "mean_token_accuracy": 0.259531706571579,
      "num_tokens": 138032360.0,
      "step": 60185
    },
    {
      "entropy": 5.5125510692596436,
      "epoch": 5.781940441882805,
      "grad_norm": 1.375,
      "learning_rate": 0.00022306003226840233,
      "loss": 4.7023,
      "mean_token_accuracy": 0.2612654849886894,
      "num_tokens": 138043738.0,
      "step": 60190
    },
    {
      "entropy": 5.517120742797852,
      "epoch": 5.782420749279539,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002230266752907116,
      "loss": 4.6637,
      "mean_token_accuracy": 0.2679380476474762,
      "num_tokens": 138055884.0,
      "step": 60195
    },
    {
      "entropy": 5.6254127502441404,
      "epoch": 5.782901056676273,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002229933195196872,
      "loss": 4.774,
      "mean_token_accuracy": 0.25058123022317885,
      "num_tokens": 138067855.0,
      "step": 60200
    },
    {
      "entropy": 5.599462985992432,
      "epoch": 5.783381364073007,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002229599649561034,
      "loss": 4.6742,
      "mean_token_accuracy": 0.2665586441755295,
      "num_tokens": 138079970.0,
      "step": 60205
    },
    {
      "entropy": 5.489103364944458,
      "epoch": 5.78386167146974,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022292661160073468,
      "loss": 4.6518,
      "mean_token_accuracy": 0.2608478680253029,
      "num_tokens": 138091798.0,
      "step": 60210
    },
    {
      "entropy": 5.511482238769531,
      "epoch": 5.7843419788664745,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002228932594543554,
      "loss": 4.6332,
      "mean_token_accuracy": 0.2627605065703392,
      "num_tokens": 138103925.0,
      "step": 60215
    },
    {
      "entropy": 5.544844627380371,
      "epoch": 5.784822286263209,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022285990851774002,
      "loss": 4.6638,
      "mean_token_accuracy": 0.2623842403292656,
      "num_tokens": 138114663.0,
      "step": 60220
    },
    {
      "entropy": 5.544019269943237,
      "epoch": 5.785302593659942,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022282655879166262,
      "loss": 4.6735,
      "mean_token_accuracy": 0.25839159786701205,
      "num_tokens": 138127329.0,
      "step": 60225
    },
    {
      "entropy": 5.572631788253784,
      "epoch": 5.785782901056677,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002227932102768977,
      "loss": 4.6998,
      "mean_token_accuracy": 0.25775990784168246,
      "num_tokens": 138138891.0,
      "step": 60230
    },
    {
      "entropy": 5.530422830581665,
      "epoch": 5.78626320845341,
      "grad_norm": 1.25,
      "learning_rate": 0.0002227598629742194,
      "loss": 4.6642,
      "mean_token_accuracy": 0.2566253960132599,
      "num_tokens": 138151081.0,
      "step": 60235
    },
    {
      "entropy": 5.5300219535827635,
      "epoch": 5.786743515850144,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022272651688440192,
      "loss": 4.5936,
      "mean_token_accuracy": 0.266358844935894,
      "num_tokens": 138161894.0,
      "step": 60240
    },
    {
      "entropy": 5.515938949584961,
      "epoch": 5.787223823246878,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022269317200821948,
      "loss": 4.6972,
      "mean_token_accuracy": 0.25871054381132125,
      "num_tokens": 138174448.0,
      "step": 60245
    },
    {
      "entropy": 5.544196891784668,
      "epoch": 5.787704130643612,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022265982834644626,
      "loss": 4.6462,
      "mean_token_accuracy": 0.2609474524855614,
      "num_tokens": 138185334.0,
      "step": 60250
    },
    {
      "entropy": 5.547135829925537,
      "epoch": 5.7881844380403455,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022262648589985645,
      "loss": 4.6486,
      "mean_token_accuracy": 0.25739008039236067,
      "num_tokens": 138195392.0,
      "step": 60255
    },
    {
      "entropy": 5.5763167381286625,
      "epoch": 5.78866474543708,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022259314466922415,
      "loss": 4.7279,
      "mean_token_accuracy": 0.2519816875457764,
      "num_tokens": 138206812.0,
      "step": 60260
    },
    {
      "entropy": 5.546357440948486,
      "epoch": 5.789145052833813,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022255980465532338,
      "loss": 4.6484,
      "mean_token_accuracy": 0.26242350339889525,
      "num_tokens": 138218347.0,
      "step": 60265
    },
    {
      "entropy": 5.591341781616211,
      "epoch": 5.789625360230548,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022252646585892822,
      "loss": 4.7208,
      "mean_token_accuracy": 0.25764355808496475,
      "num_tokens": 138230103.0,
      "step": 60270
    },
    {
      "entropy": 5.604933881759644,
      "epoch": 5.790105667627281,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022249312828081274,
      "loss": 4.7783,
      "mean_token_accuracy": 0.25333257466554643,
      "num_tokens": 138242136.0,
      "step": 60275
    },
    {
      "entropy": 5.510834407806397,
      "epoch": 5.790585975024015,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022245979192175085,
      "loss": 4.6484,
      "mean_token_accuracy": 0.2629789441823959,
      "num_tokens": 138253447.0,
      "step": 60280
    },
    {
      "entropy": 5.545446109771729,
      "epoch": 5.791066282420749,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002224264567825166,
      "loss": 4.6749,
      "mean_token_accuracy": 0.25470740646123885,
      "num_tokens": 138266197.0,
      "step": 60285
    },
    {
      "entropy": 5.546608877182007,
      "epoch": 5.791546589817483,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022239312286388398,
      "loss": 4.6166,
      "mean_token_accuracy": 0.26150874644517896,
      "num_tokens": 138277842.0,
      "step": 60290
    },
    {
      "entropy": 5.54499888420105,
      "epoch": 5.7920268972142175,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022235979016662678,
      "loss": 4.646,
      "mean_token_accuracy": 0.26636063754558564,
      "num_tokens": 138289113.0,
      "step": 60295
    },
    {
      "entropy": 5.44972186088562,
      "epoch": 5.792507204610951,
      "grad_norm": 1.375,
      "learning_rate": 0.00022232645869151892,
      "loss": 4.5964,
      "mean_token_accuracy": 0.27183871567249296,
      "num_tokens": 138300610.0,
      "step": 60300
    },
    {
      "entropy": 5.421298027038574,
      "epoch": 5.792987512007685,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022229312843933424,
      "loss": 4.5257,
      "mean_token_accuracy": 0.2828012019395828,
      "num_tokens": 138311770.0,
      "step": 60305
    },
    {
      "entropy": 5.575873279571534,
      "epoch": 5.793467819404419,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022225979941084667,
      "loss": 4.7534,
      "mean_token_accuracy": 0.2522130712866783,
      "num_tokens": 138322951.0,
      "step": 60310
    },
    {
      "entropy": 5.649752759933472,
      "epoch": 5.793948126801153,
      "grad_norm": 1.25,
      "learning_rate": 0.00022222647160683001,
      "loss": 4.7412,
      "mean_token_accuracy": 0.2539436280727386,
      "num_tokens": 138335302.0,
      "step": 60315
    },
    {
      "entropy": 5.57062258720398,
      "epoch": 5.7944284341978864,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022219314502805793,
      "loss": 4.6374,
      "mean_token_accuracy": 0.26861045360565183,
      "num_tokens": 138346185.0,
      "step": 60320
    },
    {
      "entropy": 5.593078136444092,
      "epoch": 5.794908741594621,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022215981967530423,
      "loss": 4.7218,
      "mean_token_accuracy": 0.25015713423490527,
      "num_tokens": 138357694.0,
      "step": 60325
    },
    {
      "entropy": 5.49820761680603,
      "epoch": 5.795389048991354,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022212649554934256,
      "loss": 4.6024,
      "mean_token_accuracy": 0.2684850186109543,
      "num_tokens": 138369402.0,
      "step": 60330
    },
    {
      "entropy": 5.531736564636231,
      "epoch": 5.7958693563880885,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022209317265094668,
      "loss": 4.6661,
      "mean_token_accuracy": 0.261692413687706,
      "num_tokens": 138381065.0,
      "step": 60335
    },
    {
      "entropy": 5.5280664443969725,
      "epoch": 5.796349663784822,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022205985098089026,
      "loss": 4.6394,
      "mean_token_accuracy": 0.2614660233259201,
      "num_tokens": 138393094.0,
      "step": 60340
    },
    {
      "entropy": 5.521136713027954,
      "epoch": 5.796829971181556,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022202653053994686,
      "loss": 4.618,
      "mean_token_accuracy": 0.2659000128507614,
      "num_tokens": 138404085.0,
      "step": 60345
    },
    {
      "entropy": 5.554918956756592,
      "epoch": 5.79731027857829,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022199321132889023,
      "loss": 4.7234,
      "mean_token_accuracy": 0.260410438477993,
      "num_tokens": 138417584.0,
      "step": 60350
    },
    {
      "entropy": 5.517664766311645,
      "epoch": 5.797790585975024,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022195989334849377,
      "loss": 4.7251,
      "mean_token_accuracy": 0.2573981165885925,
      "num_tokens": 138428090.0,
      "step": 60355
    },
    {
      "entropy": 5.541953897476196,
      "epoch": 5.7982708933717575,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022192657659953104,
      "loss": 4.5852,
      "mean_token_accuracy": 0.2698019176721573,
      "num_tokens": 138438209.0,
      "step": 60360
    },
    {
      "entropy": 5.58349494934082,
      "epoch": 5.798751200768492,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022189326108277562,
      "loss": 4.6911,
      "mean_token_accuracy": 0.25813078284263613,
      "num_tokens": 138450382.0,
      "step": 60365
    },
    {
      "entropy": 5.5917980670928955,
      "epoch": 5.799231508165226,
      "grad_norm": 1.375,
      "learning_rate": 0.00022185994679900102,
      "loss": 4.6963,
      "mean_token_accuracy": 0.25382431894540786,
      "num_tokens": 138461108.0,
      "step": 60370
    },
    {
      "entropy": 5.522064256668091,
      "epoch": 5.79971181556196,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002218266337489806,
      "loss": 4.6308,
      "mean_token_accuracy": 0.2584484457969666,
      "num_tokens": 138471578.0,
      "step": 60375
    },
    {
      "entropy": 5.603425073623657,
      "epoch": 5.800192122958694,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000221793321933488,
      "loss": 4.7026,
      "mean_token_accuracy": 0.25670872032642367,
      "num_tokens": 138482701.0,
      "step": 60380
    },
    {
      "entropy": 5.583744859695434,
      "epoch": 5.800672430355427,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00022176001135329633,
      "loss": 4.6307,
      "mean_token_accuracy": 0.26516329795122145,
      "num_tokens": 138493142.0,
      "step": 60385
    },
    {
      "entropy": 5.5284830093383786,
      "epoch": 5.801152737752162,
      "grad_norm": 1.5,
      "learning_rate": 0.0002217267020091791,
      "loss": 4.6564,
      "mean_token_accuracy": 0.2582906261086464,
      "num_tokens": 138504208.0,
      "step": 60390
    },
    {
      "entropy": 5.457119607925415,
      "epoch": 5.801633045148895,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022169339390190973,
      "loss": 4.5953,
      "mean_token_accuracy": 0.2652740478515625,
      "num_tokens": 138515768.0,
      "step": 60395
    },
    {
      "entropy": 5.451863718032837,
      "epoch": 5.802113352545629,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022166008703226143,
      "loss": 4.5376,
      "mean_token_accuracy": 0.2703836590051651,
      "num_tokens": 138527100.0,
      "step": 60400
    },
    {
      "entropy": 5.472833299636841,
      "epoch": 5.802593659942363,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002216267814010075,
      "loss": 4.5257,
      "mean_token_accuracy": 0.2702449634671211,
      "num_tokens": 138537484.0,
      "step": 60405
    },
    {
      "entropy": 5.5814652919769285,
      "epoch": 5.803073967339097,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022159347700892134,
      "loss": 4.7096,
      "mean_token_accuracy": 0.2554249092936516,
      "num_tokens": 138548488.0,
      "step": 60410
    },
    {
      "entropy": 5.447199916839599,
      "epoch": 5.803554274735831,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022156017385677597,
      "loss": 4.4745,
      "mean_token_accuracy": 0.2685124695301056,
      "num_tokens": 138558827.0,
      "step": 60415
    },
    {
      "entropy": 5.576008892059326,
      "epoch": 5.804034582132565,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022152687194534464,
      "loss": 4.7078,
      "mean_token_accuracy": 0.25719823092222216,
      "num_tokens": 138570750.0,
      "step": 60420
    },
    {
      "entropy": 5.503030824661255,
      "epoch": 5.804514889529298,
      "grad_norm": 1.375,
      "learning_rate": 0.0002214935712754006,
      "loss": 4.5566,
      "mean_token_accuracy": 0.268795382976532,
      "num_tokens": 138582715.0,
      "step": 60425
    },
    {
      "entropy": 5.535984849929809,
      "epoch": 5.804995196926033,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022146027184771696,
      "loss": 4.7195,
      "mean_token_accuracy": 0.25589770525693895,
      "num_tokens": 138594360.0,
      "step": 60430
    },
    {
      "entropy": 5.544092607498169,
      "epoch": 5.805475504322766,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002214269736630668,
      "loss": 4.6612,
      "mean_token_accuracy": 0.2656631901860237,
      "num_tokens": 138605156.0,
      "step": 60435
    },
    {
      "entropy": 5.544046115875244,
      "epoch": 5.8059558117195005,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022139367672222333,
      "loss": 4.6735,
      "mean_token_accuracy": 0.26429240852594377,
      "num_tokens": 138616644.0,
      "step": 60440
    },
    {
      "entropy": 5.580761480331421,
      "epoch": 5.806436119116235,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002213603810259594,
      "loss": 4.6841,
      "mean_token_accuracy": 0.258050636947155,
      "num_tokens": 138626666.0,
      "step": 60445
    },
    {
      "entropy": 5.472038745880127,
      "epoch": 5.806916426512968,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002213270865750482,
      "loss": 4.6313,
      "mean_token_accuracy": 0.2651954352855682,
      "num_tokens": 138637953.0,
      "step": 60450
    },
    {
      "entropy": 5.484606122970581,
      "epoch": 5.807396733909703,
      "grad_norm": 1.375,
      "learning_rate": 0.00022129379337026262,
      "loss": 4.6581,
      "mean_token_accuracy": 0.26930470019578934,
      "num_tokens": 138649240.0,
      "step": 60455
    },
    {
      "entropy": 5.510700035095215,
      "epoch": 5.807877041306436,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022126050141237568,
      "loss": 4.603,
      "mean_token_accuracy": 0.26436216533184054,
      "num_tokens": 138661240.0,
      "step": 60460
    },
    {
      "entropy": 5.582611465454102,
      "epoch": 5.80835734870317,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002212272107021604,
      "loss": 4.7111,
      "mean_token_accuracy": 0.2605133682489395,
      "num_tokens": 138672595.0,
      "step": 60465
    },
    {
      "entropy": 5.609344339370727,
      "epoch": 5.808837656099904,
      "grad_norm": 1.71875,
      "learning_rate": 0.00022119392124038962,
      "loss": 4.716,
      "mean_token_accuracy": 0.2593295738101006,
      "num_tokens": 138682976.0,
      "step": 60470
    },
    {
      "entropy": 5.537750005722046,
      "epoch": 5.809317963496638,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022116063302783618,
      "loss": 4.6689,
      "mean_token_accuracy": 0.2648746266961098,
      "num_tokens": 138694311.0,
      "step": 60475
    },
    {
      "entropy": 5.535234260559082,
      "epoch": 5.8097982708933715,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022112734606527295,
      "loss": 4.6522,
      "mean_token_accuracy": 0.25745805352926254,
      "num_tokens": 138706675.0,
      "step": 60480
    },
    {
      "entropy": 5.52876877784729,
      "epoch": 5.810278578290106,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022109406035347273,
      "loss": 4.6189,
      "mean_token_accuracy": 0.2639111548662186,
      "num_tokens": 138717324.0,
      "step": 60485
    },
    {
      "entropy": 5.531757879257202,
      "epoch": 5.810758885686839,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022106077589320843,
      "loss": 4.6632,
      "mean_token_accuracy": 0.2597472950816154,
      "num_tokens": 138729402.0,
      "step": 60490
    },
    {
      "entropy": 5.5161780834198,
      "epoch": 5.811239193083574,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022102749268525275,
      "loss": 4.6607,
      "mean_token_accuracy": 0.26055106669664385,
      "num_tokens": 138740842.0,
      "step": 60495
    },
    {
      "entropy": 5.543690729141235,
      "epoch": 5.811719500480307,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022099421073037845,
      "loss": 4.6411,
      "mean_token_accuracy": 0.2626835510134697,
      "num_tokens": 138752842.0,
      "step": 60500
    },
    {
      "entropy": 5.613500261306763,
      "epoch": 5.812199807877041,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022096093002935818,
      "loss": 4.7555,
      "mean_token_accuracy": 0.2572041615843773,
      "num_tokens": 138764612.0,
      "step": 60505
    },
    {
      "entropy": 5.580539035797119,
      "epoch": 5.812680115273775,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022092765058296466,
      "loss": 4.7124,
      "mean_token_accuracy": 0.2585474520921707,
      "num_tokens": 138777425.0,
      "step": 60510
    },
    {
      "entropy": 5.629915523529053,
      "epoch": 5.813160422670509,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022089437239197047,
      "loss": 4.6985,
      "mean_token_accuracy": 0.25885362178087234,
      "num_tokens": 138788133.0,
      "step": 60515
    },
    {
      "entropy": 5.527203178405761,
      "epoch": 5.8136407300672435,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022086109545714833,
      "loss": 4.6445,
      "mean_token_accuracy": 0.2628727152943611,
      "num_tokens": 138800226.0,
      "step": 60520
    },
    {
      "entropy": 5.439180707931518,
      "epoch": 5.814121037463977,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002208278197792708,
      "loss": 4.5191,
      "mean_token_accuracy": 0.27090988755226136,
      "num_tokens": 138811718.0,
      "step": 60525
    },
    {
      "entropy": 5.514009046554565,
      "epoch": 5.814601344860711,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022079454535911048,
      "loss": 4.6358,
      "mean_token_accuracy": 0.26410684585571287,
      "num_tokens": 138823510.0,
      "step": 60530
    },
    {
      "entropy": 5.514532041549683,
      "epoch": 5.815081652257445,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002207612721974398,
      "loss": 4.5852,
      "mean_token_accuracy": 0.2663987085223198,
      "num_tokens": 138833696.0,
      "step": 60535
    },
    {
      "entropy": 5.516482400894165,
      "epoch": 5.815561959654179,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022072800029503137,
      "loss": 4.6019,
      "mean_token_accuracy": 0.2702731654047966,
      "num_tokens": 138843962.0,
      "step": 60540
    },
    {
      "entropy": 5.552116441726684,
      "epoch": 5.816042267050912,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002206947296526576,
      "loss": 4.6837,
      "mean_token_accuracy": 0.263187338411808,
      "num_tokens": 138854564.0,
      "step": 60545
    },
    {
      "entropy": 5.4913249015808105,
      "epoch": 5.816522574447647,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022066146027109091,
      "loss": 4.5667,
      "mean_token_accuracy": 0.27351194620132446,
      "num_tokens": 138865939.0,
      "step": 60550
    },
    {
      "entropy": 5.554841375350952,
      "epoch": 5.81700288184438,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022062819215110381,
      "loss": 4.6474,
      "mean_token_accuracy": 0.26225215196609497,
      "num_tokens": 138876766.0,
      "step": 60555
    },
    {
      "entropy": 5.411576604843139,
      "epoch": 5.8174831892411145,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022059492529346865,
      "loss": 4.495,
      "mean_token_accuracy": 0.28009536415338515,
      "num_tokens": 138886914.0,
      "step": 60560
    },
    {
      "entropy": 5.559987354278564,
      "epoch": 5.817963496637848,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022056165969895768,
      "loss": 4.7622,
      "mean_token_accuracy": 0.2544462189078331,
      "num_tokens": 138897832.0,
      "step": 60565
    },
    {
      "entropy": 5.576977729797363,
      "epoch": 5.818443804034582,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002205283953683434,
      "loss": 4.7245,
      "mean_token_accuracy": 0.2579704120755196,
      "num_tokens": 138910214.0,
      "step": 60570
    },
    {
      "entropy": 5.580975770950317,
      "epoch": 5.818924111431316,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022049513230239804,
      "loss": 4.6388,
      "mean_token_accuracy": 0.26988341361284257,
      "num_tokens": 138920059.0,
      "step": 60575
    },
    {
      "entropy": 5.601779651641846,
      "epoch": 5.81940441882805,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022046187050189388,
      "loss": 4.673,
      "mean_token_accuracy": 0.25855653584003446,
      "num_tokens": 138930900.0,
      "step": 60580
    },
    {
      "entropy": 5.5542220115661625,
      "epoch": 5.8198847262247835,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022042860996760306,
      "loss": 4.6835,
      "mean_token_accuracy": 0.2577186360955238,
      "num_tokens": 138942964.0,
      "step": 60585
    },
    {
      "entropy": 5.566652536392212,
      "epoch": 5.820365033621518,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022039535070029792,
      "loss": 4.765,
      "mean_token_accuracy": 0.2511377513408661,
      "num_tokens": 138955063.0,
      "step": 60590
    },
    {
      "entropy": 5.568023157119751,
      "epoch": 5.820845341018252,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022036209270075057,
      "loss": 4.6426,
      "mean_token_accuracy": 0.2654829978942871,
      "num_tokens": 138966020.0,
      "step": 60595
    },
    {
      "entropy": 5.622562456130981,
      "epoch": 5.821325648414986,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022032883596973324,
      "loss": 4.7485,
      "mean_token_accuracy": 0.2609182447195053,
      "num_tokens": 138977775.0,
      "step": 60600
    },
    {
      "entropy": 5.563694763183594,
      "epoch": 5.821805955811719,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022029558050801795,
      "loss": 4.7069,
      "mean_token_accuracy": 0.2493315279483795,
      "num_tokens": 138989810.0,
      "step": 60605
    },
    {
      "entropy": 5.54860897064209,
      "epoch": 5.822286263208453,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022026232631637682,
      "loss": 4.7314,
      "mean_token_accuracy": 0.256300713121891,
      "num_tokens": 139002258.0,
      "step": 60610
    },
    {
      "entropy": 5.495651483535767,
      "epoch": 5.822766570605188,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022022907339558202,
      "loss": 4.5615,
      "mean_token_accuracy": 0.26896418929100036,
      "num_tokens": 139013485.0,
      "step": 60615
    },
    {
      "entropy": 5.582965517044068,
      "epoch": 5.823246878001921,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002201958217464055,
      "loss": 4.7853,
      "mean_token_accuracy": 0.2501967951655388,
      "num_tokens": 139027103.0,
      "step": 60620
    },
    {
      "entropy": 5.507383012771607,
      "epoch": 5.823727185398655,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022016257136961924,
      "loss": 4.5965,
      "mean_token_accuracy": 0.2713489279150963,
      "num_tokens": 139037623.0,
      "step": 60625
    },
    {
      "entropy": 5.472547245025635,
      "epoch": 5.824207492795389,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002201293222659952,
      "loss": 4.607,
      "mean_token_accuracy": 0.2665613070130348,
      "num_tokens": 139048458.0,
      "step": 60630
    },
    {
      "entropy": 5.494236850738526,
      "epoch": 5.824687800192123,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022009607443630535,
      "loss": 4.6569,
      "mean_token_accuracy": 0.26549278348684313,
      "num_tokens": 139060904.0,
      "step": 60635
    },
    {
      "entropy": 5.513376379013062,
      "epoch": 5.825168107588857,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022006282788132164,
      "loss": 4.6882,
      "mean_token_accuracy": 0.25873049795627595,
      "num_tokens": 139071165.0,
      "step": 60640
    },
    {
      "entropy": 5.606014108657837,
      "epoch": 5.825648414985591,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022002958260181594,
      "loss": 4.699,
      "mean_token_accuracy": 0.26049362272024157,
      "num_tokens": 139082146.0,
      "step": 60645
    },
    {
      "entropy": 5.585106086730957,
      "epoch": 5.826128722382324,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002199963385985602,
      "loss": 4.7025,
      "mean_token_accuracy": 0.26243444830179213,
      "num_tokens": 139093931.0,
      "step": 60650
    },
    {
      "entropy": 5.537521409988403,
      "epoch": 5.826609029779059,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021996309587232605,
      "loss": 4.6474,
      "mean_token_accuracy": 0.25847636312246325,
      "num_tokens": 139104131.0,
      "step": 60655
    },
    {
      "entropy": 5.584187412261963,
      "epoch": 5.827089337175792,
      "grad_norm": 1.25,
      "learning_rate": 0.0002199298544238854,
      "loss": 4.6527,
      "mean_token_accuracy": 0.25331574976444243,
      "num_tokens": 139115088.0,
      "step": 60660
    },
    {
      "entropy": 5.4881195545196535,
      "epoch": 5.8275696445725265,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021989661425400998,
      "loss": 4.5576,
      "mean_token_accuracy": 0.26463344842195513,
      "num_tokens": 139125914.0,
      "step": 60665
    },
    {
      "entropy": 5.529412364959716,
      "epoch": 5.828049951969261,
      "grad_norm": 2.421875,
      "learning_rate": 0.00021986337536347155,
      "loss": 4.6833,
      "mean_token_accuracy": 0.26542852222919466,
      "num_tokens": 139136905.0,
      "step": 60670
    },
    {
      "entropy": 5.485814237594605,
      "epoch": 5.828530259365994,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002198301377530419,
      "loss": 4.6115,
      "mean_token_accuracy": 0.25980019122362136,
      "num_tokens": 139147952.0,
      "step": 60675
    },
    {
      "entropy": 5.5090522289276125,
      "epoch": 5.829010566762728,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021979690142349262,
      "loss": 4.6582,
      "mean_token_accuracy": 0.2565661832690239,
      "num_tokens": 139159090.0,
      "step": 60680
    },
    {
      "entropy": 5.543023872375488,
      "epoch": 5.829490874159462,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021976366637559532,
      "loss": 4.6525,
      "mean_token_accuracy": 0.2673694550991058,
      "num_tokens": 139169814.0,
      "step": 60685
    },
    {
      "entropy": 5.591088056564331,
      "epoch": 5.829971181556196,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002197304326101217,
      "loss": 4.7195,
      "mean_token_accuracy": 0.25263864547014236,
      "num_tokens": 139181854.0,
      "step": 60690
    },
    {
      "entropy": 5.478748893737793,
      "epoch": 5.83045148895293,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002196972001278433,
      "loss": 4.5923,
      "mean_token_accuracy": 0.27061418890953065,
      "num_tokens": 139192862.0,
      "step": 60695
    },
    {
      "entropy": 5.518581867218018,
      "epoch": 5.830931796349664,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021966396892953167,
      "loss": 4.68,
      "mean_token_accuracy": 0.2574854165315628,
      "num_tokens": 139203594.0,
      "step": 60700
    },
    {
      "entropy": 5.500273990631103,
      "epoch": 5.8314121037463975,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021963073901595838,
      "loss": 4.5855,
      "mean_token_accuracy": 0.2694326162338257,
      "num_tokens": 139216475.0,
      "step": 60705
    },
    {
      "entropy": 5.572375535964966,
      "epoch": 5.831892411143132,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021959751038789498,
      "loss": 4.7013,
      "mean_token_accuracy": 0.26050257831811907,
      "num_tokens": 139226626.0,
      "step": 60710
    },
    {
      "entropy": 5.545838928222656,
      "epoch": 5.832372718539865,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002195642830461128,
      "loss": 4.672,
      "mean_token_accuracy": 0.2601121425628662,
      "num_tokens": 139239713.0,
      "step": 60715
    },
    {
      "entropy": 5.5016350746154785,
      "epoch": 5.8328530259366,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021953105699138337,
      "loss": 4.6062,
      "mean_token_accuracy": 0.2655696883797646,
      "num_tokens": 139250948.0,
      "step": 60720
    },
    {
      "entropy": 5.521322965621948,
      "epoch": 5.833333333333333,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002194978322244781,
      "loss": 4.6648,
      "mean_token_accuracy": 0.2579079419374466,
      "num_tokens": 139263145.0,
      "step": 60725
    },
    {
      "entropy": 5.622298336029052,
      "epoch": 5.833813640730067,
      "grad_norm": 1.25,
      "learning_rate": 0.00021946460874616835,
      "loss": 4.8056,
      "mean_token_accuracy": 0.25031501054763794,
      "num_tokens": 139274960.0,
      "step": 60730
    },
    {
      "entropy": 5.624000453948975,
      "epoch": 5.834293948126801,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021943138655722549,
      "loss": 4.6604,
      "mean_token_accuracy": 0.2593493953347206,
      "num_tokens": 139285482.0,
      "step": 60735
    },
    {
      "entropy": 5.515123176574707,
      "epoch": 5.834774255523535,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021939816565842092,
      "loss": 4.6538,
      "mean_token_accuracy": 0.25910494178533555,
      "num_tokens": 139297092.0,
      "step": 60740
    },
    {
      "entropy": 5.4729938983917235,
      "epoch": 5.8352545629202694,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002193649460505257,
      "loss": 4.6477,
      "mean_token_accuracy": 0.2632166236639023,
      "num_tokens": 139308070.0,
      "step": 60745
    },
    {
      "entropy": 5.479848003387451,
      "epoch": 5.835734870317003,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002193317277343113,
      "loss": 4.5249,
      "mean_token_accuracy": 0.2762273192405701,
      "num_tokens": 139319514.0,
      "step": 60750
    },
    {
      "entropy": 5.500290775299073,
      "epoch": 5.836215177713736,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021929851071054884,
      "loss": 4.5131,
      "mean_token_accuracy": 0.2818739950656891,
      "num_tokens": 139329245.0,
      "step": 60755
    },
    {
      "entropy": 5.561311149597168,
      "epoch": 5.836695485110471,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021926529498000958,
      "loss": 4.7656,
      "mean_token_accuracy": 0.2506259813904762,
      "num_tokens": 139340801.0,
      "step": 60760
    },
    {
      "entropy": 5.5007397651672365,
      "epoch": 5.837175792507205,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021923208054346467,
      "loss": 4.5977,
      "mean_token_accuracy": 0.26631277799606323,
      "num_tokens": 139353023.0,
      "step": 60765
    },
    {
      "entropy": 5.540537548065186,
      "epoch": 5.837656099903938,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021919886740168532,
      "loss": 4.6756,
      "mean_token_accuracy": 0.25523336976766586,
      "num_tokens": 139366330.0,
      "step": 60770
    },
    {
      "entropy": 5.596071863174439,
      "epoch": 5.838136407300673,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021916565555544248,
      "loss": 4.6871,
      "mean_token_accuracy": 0.26073211133480073,
      "num_tokens": 139378032.0,
      "step": 60775
    },
    {
      "entropy": 5.5914207935333256,
      "epoch": 5.838616714697406,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021913244500550738,
      "loss": 4.6673,
      "mean_token_accuracy": 0.2553742125630379,
      "num_tokens": 139389386.0,
      "step": 60780
    },
    {
      "entropy": 5.5422128200531,
      "epoch": 5.8390970220941405,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021909923575265096,
      "loss": 4.6681,
      "mean_token_accuracy": 0.26709907203912736,
      "num_tokens": 139401508.0,
      "step": 60785
    },
    {
      "entropy": 5.566555547714233,
      "epoch": 5.839577329490874,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021906602779764433,
      "loss": 4.6286,
      "mean_token_accuracy": 0.2661320701241493,
      "num_tokens": 139412886.0,
      "step": 60790
    },
    {
      "entropy": 5.5056602478027346,
      "epoch": 5.840057636887608,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021903282114125845,
      "loss": 4.5637,
      "mean_token_accuracy": 0.27132882475852965,
      "num_tokens": 139423622.0,
      "step": 60795
    },
    {
      "entropy": 5.614078760147095,
      "epoch": 5.840537944284342,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021899961578426426,
      "loss": 4.7412,
      "mean_token_accuracy": 0.26227412223815916,
      "num_tokens": 139434597.0,
      "step": 60800
    },
    {
      "entropy": 5.545518112182617,
      "epoch": 5.841018251681076,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002189664117274327,
      "loss": 4.675,
      "mean_token_accuracy": 0.25801973044872284,
      "num_tokens": 139446373.0,
      "step": 60805
    },
    {
      "entropy": 5.484616041183472,
      "epoch": 5.8414985590778095,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021893320897153472,
      "loss": 4.6092,
      "mean_token_accuracy": 0.27126633673906325,
      "num_tokens": 139458427.0,
      "step": 60810
    },
    {
      "entropy": 5.540403127670288,
      "epoch": 5.841978866474544,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021890000751734106,
      "loss": 4.6483,
      "mean_token_accuracy": 0.2621729835867882,
      "num_tokens": 139470981.0,
      "step": 60815
    },
    {
      "entropy": 5.511498498916626,
      "epoch": 5.842459173871278,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021886680736562275,
      "loss": 4.5947,
      "mean_token_accuracy": 0.2632403433322906,
      "num_tokens": 139481703.0,
      "step": 60820
    },
    {
      "entropy": 5.637999200820923,
      "epoch": 5.8429394812680115,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002188336085171504,
      "loss": 4.6936,
      "mean_token_accuracy": 0.2547279864549637,
      "num_tokens": 139492411.0,
      "step": 60825
    },
    {
      "entropy": 5.562396287918091,
      "epoch": 5.843419788664745,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021880041097269493,
      "loss": 4.6494,
      "mean_token_accuracy": 0.26527799367904664,
      "num_tokens": 139503683.0,
      "step": 60830
    },
    {
      "entropy": 5.552141618728638,
      "epoch": 5.843900096061479,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021876721473302702,
      "loss": 4.6997,
      "mean_token_accuracy": 0.2553423196077347,
      "num_tokens": 139515439.0,
      "step": 60835
    },
    {
      "entropy": 5.460990762710571,
      "epoch": 5.844380403458214,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021873401979891738,
      "loss": 4.5928,
      "mean_token_accuracy": 0.26724348217248917,
      "num_tokens": 139527410.0,
      "step": 60840
    },
    {
      "entropy": 5.477114582061768,
      "epoch": 5.844860710854947,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021870082617113674,
      "loss": 4.5707,
      "mean_token_accuracy": 0.26776991188526156,
      "num_tokens": 139536824.0,
      "step": 60845
    },
    {
      "entropy": 5.512262296676636,
      "epoch": 5.845341018251681,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002186676338504558,
      "loss": 4.666,
      "mean_token_accuracy": 0.2602341756224632,
      "num_tokens": 139549811.0,
      "step": 60850
    },
    {
      "entropy": 5.600000333786011,
      "epoch": 5.845821325648415,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021863444283764505,
      "loss": 4.7495,
      "mean_token_accuracy": 0.2526373818516731,
      "num_tokens": 139561032.0,
      "step": 60855
    },
    {
      "entropy": 5.627175521850586,
      "epoch": 5.846301633045149,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021860125313347524,
      "loss": 4.7166,
      "mean_token_accuracy": 0.26188569515943527,
      "num_tokens": 139572590.0,
      "step": 60860
    },
    {
      "entropy": 5.569585943222046,
      "epoch": 5.846781940441883,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021856806473871681,
      "loss": 4.6608,
      "mean_token_accuracy": 0.26178358793258666,
      "num_tokens": 139585122.0,
      "step": 60865
    },
    {
      "entropy": 5.561471605300904,
      "epoch": 5.847262247838617,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021853487765414038,
      "loss": 4.7018,
      "mean_token_accuracy": 0.2598062291741371,
      "num_tokens": 139597337.0,
      "step": 60870
    },
    {
      "entropy": 5.533599138259888,
      "epoch": 5.84774255523535,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021850169188051638,
      "loss": 4.5656,
      "mean_token_accuracy": 0.26744211465120316,
      "num_tokens": 139608963.0,
      "step": 60875
    },
    {
      "entropy": 5.553767681121826,
      "epoch": 5.848222862632085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021846850741861535,
      "loss": 4.6605,
      "mean_token_accuracy": 0.263949953019619,
      "num_tokens": 139620950.0,
      "step": 60880
    },
    {
      "entropy": 5.5184471130371096,
      "epoch": 5.848703170028818,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002184353242692077,
      "loss": 4.6674,
      "mean_token_accuracy": 0.2603405073285103,
      "num_tokens": 139632066.0,
      "step": 60885
    },
    {
      "entropy": 5.583782768249511,
      "epoch": 5.8491834774255524,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021840214243306388,
      "loss": 4.6971,
      "mean_token_accuracy": 0.2592486947774887,
      "num_tokens": 139643046.0,
      "step": 60890
    },
    {
      "entropy": 5.5850786685943605,
      "epoch": 5.849663784822286,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002183689619109543,
      "loss": 4.6587,
      "mean_token_accuracy": 0.2584285452961922,
      "num_tokens": 139653546.0,
      "step": 60895
    },
    {
      "entropy": 5.554304313659668,
      "epoch": 5.85014409221902,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021833578270364918,
      "loss": 4.7566,
      "mean_token_accuracy": 0.2544987216591835,
      "num_tokens": 139665263.0,
      "step": 60900
    },
    {
      "entropy": 5.548329448699951,
      "epoch": 5.850624399615754,
      "grad_norm": 1.4375,
      "learning_rate": 0.000218302604811919,
      "loss": 4.6612,
      "mean_token_accuracy": 0.25820946544408796,
      "num_tokens": 139676469.0,
      "step": 60905
    },
    {
      "entropy": 5.589704799652099,
      "epoch": 5.851104707012488,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002182694282365339,
      "loss": 4.6703,
      "mean_token_accuracy": 0.26271192282438277,
      "num_tokens": 139687883.0,
      "step": 60910
    },
    {
      "entropy": 5.537579822540283,
      "epoch": 5.851585014409222,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021823625297826422,
      "loss": 4.6536,
      "mean_token_accuracy": 0.2589310839772224,
      "num_tokens": 139701234.0,
      "step": 60915
    },
    {
      "entropy": 5.656995582580566,
      "epoch": 5.852065321805956,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021820307903788023,
      "loss": 4.7735,
      "mean_token_accuracy": 0.24707613736391068,
      "num_tokens": 139713433.0,
      "step": 60920
    },
    {
      "entropy": 5.5966987133026125,
      "epoch": 5.85254562920269,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002181699064161522,
      "loss": 4.7158,
      "mean_token_accuracy": 0.25224104672670367,
      "num_tokens": 139724429.0,
      "step": 60925
    },
    {
      "entropy": 5.546548509597779,
      "epoch": 5.8530259365994235,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021813673511385001,
      "loss": 4.6505,
      "mean_token_accuracy": 0.2541789814829826,
      "num_tokens": 139735133.0,
      "step": 60930
    },
    {
      "entropy": 5.567511034011841,
      "epoch": 5.853506243996158,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021810356513174406,
      "loss": 4.6357,
      "mean_token_accuracy": 0.2725573152303696,
      "num_tokens": 139745139.0,
      "step": 60935
    },
    {
      "entropy": 5.578327417373657,
      "epoch": 5.853986551392891,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002180703964706044,
      "loss": 4.6974,
      "mean_token_accuracy": 0.25970028787851335,
      "num_tokens": 139755525.0,
      "step": 60940
    },
    {
      "entropy": 5.592256212234497,
      "epoch": 5.854466858789626,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021803722913120105,
      "loss": 4.7423,
      "mean_token_accuracy": 0.25607453733682634,
      "num_tokens": 139767961.0,
      "step": 60945
    },
    {
      "entropy": 5.489500284194946,
      "epoch": 5.854947166186359,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002180040631143041,
      "loss": 4.6163,
      "mean_token_accuracy": 0.26918024718761446,
      "num_tokens": 139778805.0,
      "step": 60950
    },
    {
      "entropy": 5.504235029220581,
      "epoch": 5.855427473583093,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021797089842068363,
      "loss": 4.621,
      "mean_token_accuracy": 0.26669666171073914,
      "num_tokens": 139790659.0,
      "step": 60955
    },
    {
      "entropy": 5.5341747283935545,
      "epoch": 5.855907780979827,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021793773505110952,
      "loss": 4.6393,
      "mean_token_accuracy": 0.2603244349360466,
      "num_tokens": 139801305.0,
      "step": 60960
    },
    {
      "entropy": 5.556881141662598,
      "epoch": 5.856388088376561,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002179045730063517,
      "loss": 4.6683,
      "mean_token_accuracy": 0.2609951809048653,
      "num_tokens": 139811503.0,
      "step": 60965
    },
    {
      "entropy": 5.496035575866699,
      "epoch": 5.8568683957732945,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021787141228718021,
      "loss": 4.5988,
      "mean_token_accuracy": 0.27208214700222016,
      "num_tokens": 139822010.0,
      "step": 60970
    },
    {
      "entropy": 5.527679014205932,
      "epoch": 5.857348703170029,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021783825289436483,
      "loss": 4.6581,
      "mean_token_accuracy": 0.25471733659505846,
      "num_tokens": 139834614.0,
      "step": 60975
    },
    {
      "entropy": 5.551214265823364,
      "epoch": 5.857829010566762,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021780509482867554,
      "loss": 4.6233,
      "mean_token_accuracy": 0.2630553930997849,
      "num_tokens": 139844514.0,
      "step": 60980
    },
    {
      "entropy": 5.617616796493531,
      "epoch": 5.858309317963497,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002177719380908822,
      "loss": 4.7195,
      "mean_token_accuracy": 0.25453522950410845,
      "num_tokens": 139855804.0,
      "step": 60985
    },
    {
      "entropy": 5.460003471374511,
      "epoch": 5.858789625360231,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002177387826817544,
      "loss": 4.5282,
      "mean_token_accuracy": 0.27622664272785186,
      "num_tokens": 139867963.0,
      "step": 60990
    },
    {
      "entropy": 5.6005267143249515,
      "epoch": 5.859269932756964,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021770562860206206,
      "loss": 4.7022,
      "mean_token_accuracy": 0.2608550161123276,
      "num_tokens": 139879010.0,
      "step": 60995
    },
    {
      "entropy": 5.434625959396362,
      "epoch": 5.859750240153699,
      "grad_norm": 1.984375,
      "learning_rate": 0.00021767247585257487,
      "loss": 4.5397,
      "mean_token_accuracy": 0.27088043093681335,
      "num_tokens": 139889546.0,
      "step": 61000
    },
    {
      "entropy": 5.52612886428833,
      "epoch": 5.860230547550432,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002176393244340626,
      "loss": 4.624,
      "mean_token_accuracy": 0.26402702182531357,
      "num_tokens": 139902366.0,
      "step": 61005
    },
    {
      "entropy": 5.571727275848389,
      "epoch": 5.8607108549471665,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021760617434729484,
      "loss": 4.6653,
      "mean_token_accuracy": 0.264787495136261,
      "num_tokens": 139913437.0,
      "step": 61010
    },
    {
      "entropy": 5.5795997142791744,
      "epoch": 5.8611911623439,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021757302559304142,
      "loss": 4.6669,
      "mean_token_accuracy": 0.2613757088780403,
      "num_tokens": 139925796.0,
      "step": 61015
    },
    {
      "entropy": 5.564776182174683,
      "epoch": 5.861671469740634,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002175398781720717,
      "loss": 4.6603,
      "mean_token_accuracy": 0.26164934039115906,
      "num_tokens": 139935929.0,
      "step": 61020
    },
    {
      "entropy": 5.571517467498779,
      "epoch": 5.862151777137368,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021750673208515542,
      "loss": 4.6755,
      "mean_token_accuracy": 0.25407036393880844,
      "num_tokens": 139946953.0,
      "step": 61025
    },
    {
      "entropy": 5.530568933486938,
      "epoch": 5.862632084534102,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021747358733306205,
      "loss": 4.6489,
      "mean_token_accuracy": 0.2681692034006119,
      "num_tokens": 139957445.0,
      "step": 61030
    },
    {
      "entropy": 5.5107102394104,
      "epoch": 5.863112391930835,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021744044391656126,
      "loss": 4.6818,
      "mean_token_accuracy": 0.2642258569598198,
      "num_tokens": 139968904.0,
      "step": 61035
    },
    {
      "entropy": 5.564255475997925,
      "epoch": 5.86359269932757,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021740730183642237,
      "loss": 4.7186,
      "mean_token_accuracy": 0.26182160079479216,
      "num_tokens": 139980406.0,
      "step": 61040
    },
    {
      "entropy": 5.5629143714904785,
      "epoch": 5.864073006724303,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021737416109341503,
      "loss": 4.6823,
      "mean_token_accuracy": 0.25659426748752595,
      "num_tokens": 139991071.0,
      "step": 61045
    },
    {
      "entropy": 5.624567365646362,
      "epoch": 5.8645533141210375,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021734102168830845,
      "loss": 4.7146,
      "mean_token_accuracy": 0.24901470094919204,
      "num_tokens": 140003080.0,
      "step": 61050
    },
    {
      "entropy": 5.549970102310181,
      "epoch": 5.865033621517771,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021730788362187215,
      "loss": 4.6318,
      "mean_token_accuracy": 0.27110588997602464,
      "num_tokens": 140014108.0,
      "step": 61055
    },
    {
      "entropy": 5.535363006591797,
      "epoch": 5.865513928914505,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021727474689487548,
      "loss": 4.6701,
      "mean_token_accuracy": 0.26337750554084777,
      "num_tokens": 140025863.0,
      "step": 61060
    },
    {
      "entropy": 5.592347002029419,
      "epoch": 5.86599423631124,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021724161150808775,
      "loss": 4.6623,
      "mean_token_accuracy": 0.2582484036684036,
      "num_tokens": 140035930.0,
      "step": 61065
    },
    {
      "entropy": 5.516900730133057,
      "epoch": 5.866474543707973,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021720847746227833,
      "loss": 4.6443,
      "mean_token_accuracy": 0.26386552304029465,
      "num_tokens": 140046985.0,
      "step": 61070
    },
    {
      "entropy": 5.542405223846435,
      "epoch": 5.866954851104707,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021717534475821648,
      "loss": 4.7033,
      "mean_token_accuracy": 0.2580243512988091,
      "num_tokens": 140059476.0,
      "step": 61075
    },
    {
      "entropy": 5.555456161499023,
      "epoch": 5.867435158501441,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021714221339667146,
      "loss": 4.684,
      "mean_token_accuracy": 0.2613572537899017,
      "num_tokens": 140069592.0,
      "step": 61080
    },
    {
      "entropy": 5.589885902404785,
      "epoch": 5.867915465898175,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021710908337841235,
      "loss": 4.7115,
      "mean_token_accuracy": 0.2558082267642021,
      "num_tokens": 140081386.0,
      "step": 61085
    },
    {
      "entropy": 5.462820911407471,
      "epoch": 5.868395773294909,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021707595470420843,
      "loss": 4.5468,
      "mean_token_accuracy": 0.2685136765241623,
      "num_tokens": 140092727.0,
      "step": 61090
    },
    {
      "entropy": 5.530139684677124,
      "epoch": 5.868876080691643,
      "grad_norm": 1.375,
      "learning_rate": 0.0002170428273748289,
      "loss": 4.6789,
      "mean_token_accuracy": 0.2583895355463028,
      "num_tokens": 140102813.0,
      "step": 61095
    },
    {
      "entropy": 5.506501913070679,
      "epoch": 5.869356388088376,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021700970139104276,
      "loss": 4.6644,
      "mean_token_accuracy": 0.2597545266151428,
      "num_tokens": 140114300.0,
      "step": 61100
    },
    {
      "entropy": 5.52150068283081,
      "epoch": 5.869836695485111,
      "grad_norm": 1.25,
      "learning_rate": 0.0002169765767536192,
      "loss": 4.5902,
      "mean_token_accuracy": 0.26237106919288633,
      "num_tokens": 140126817.0,
      "step": 61105
    },
    {
      "entropy": 5.561600303649902,
      "epoch": 5.870317002881844,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021694345346332723,
      "loss": 4.6932,
      "mean_token_accuracy": 0.2575021505355835,
      "num_tokens": 140139490.0,
      "step": 61110
    },
    {
      "entropy": 5.535038852691651,
      "epoch": 5.870797310278578,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021691033152093587,
      "loss": 4.6855,
      "mean_token_accuracy": 0.2632981166243553,
      "num_tokens": 140150333.0,
      "step": 61115
    },
    {
      "entropy": 5.6978870868682865,
      "epoch": 5.871277617675312,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021687721092721407,
      "loss": 4.7811,
      "mean_token_accuracy": 0.2528209686279297,
      "num_tokens": 140161070.0,
      "step": 61120
    },
    {
      "entropy": 5.5967213153839115,
      "epoch": 5.871757925072046,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021684409168293089,
      "loss": 4.7255,
      "mean_token_accuracy": 0.25565494000911715,
      "num_tokens": 140172513.0,
      "step": 61125
    },
    {
      "entropy": 5.50593409538269,
      "epoch": 5.87223823246878,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021681097378885522,
      "loss": 4.594,
      "mean_token_accuracy": 0.26504524499177934,
      "num_tokens": 140182585.0,
      "step": 61130
    },
    {
      "entropy": 5.517978382110596,
      "epoch": 5.872718539865514,
      "grad_norm": 1.375,
      "learning_rate": 0.000216777857245756,
      "loss": 4.7031,
      "mean_token_accuracy": 0.25998312681913377,
      "num_tokens": 140194624.0,
      "step": 61135
    },
    {
      "entropy": 5.571613788604736,
      "epoch": 5.873198847262248,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021674474205440197,
      "loss": 4.6951,
      "mean_token_accuracy": 0.2588188499212265,
      "num_tokens": 140205595.0,
      "step": 61140
    },
    {
      "entropy": 5.456858825683594,
      "epoch": 5.873679154658982,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021671162821556206,
      "loss": 4.5265,
      "mean_token_accuracy": 0.26940031796693803,
      "num_tokens": 140216474.0,
      "step": 61145
    },
    {
      "entropy": 5.659751129150391,
      "epoch": 5.874159462055716,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021667851573000508,
      "loss": 4.7932,
      "mean_token_accuracy": 0.2513751223683357,
      "num_tokens": 140229359.0,
      "step": 61150
    },
    {
      "entropy": 5.615337467193603,
      "epoch": 5.8746397694524495,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002166454045984998,
      "loss": 4.7827,
      "mean_token_accuracy": 0.251635754108429,
      "num_tokens": 140240228.0,
      "step": 61155
    },
    {
      "entropy": 5.612359952926636,
      "epoch": 5.875120076849184,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002166122948218149,
      "loss": 4.7559,
      "mean_token_accuracy": 0.2553471878170967,
      "num_tokens": 140250796.0,
      "step": 61160
    },
    {
      "entropy": 5.549535942077637,
      "epoch": 5.875600384245917,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021657918640071922,
      "loss": 4.6775,
      "mean_token_accuracy": 0.261450058221817,
      "num_tokens": 140262822.0,
      "step": 61165
    },
    {
      "entropy": 5.588044738769531,
      "epoch": 5.876080691642652,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002165460793359813,
      "loss": 4.7375,
      "mean_token_accuracy": 0.2590902239084244,
      "num_tokens": 140274846.0,
      "step": 61170
    },
    {
      "entropy": 5.503982305526733,
      "epoch": 5.876560999039385,
      "grad_norm": 1.375,
      "learning_rate": 0.0002165129736283698,
      "loss": 4.6043,
      "mean_token_accuracy": 0.2698937147855759,
      "num_tokens": 140286446.0,
      "step": 61175
    },
    {
      "entropy": 5.533860349655152,
      "epoch": 5.877041306436119,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021647986927865343,
      "loss": 4.6259,
      "mean_token_accuracy": 0.2638911843299866,
      "num_tokens": 140297315.0,
      "step": 61180
    },
    {
      "entropy": 5.458318614959717,
      "epoch": 5.877521613832853,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021644676628760075,
      "loss": 4.5768,
      "mean_token_accuracy": 0.27265052795410155,
      "num_tokens": 140308825.0,
      "step": 61185
    },
    {
      "entropy": 5.532498455047607,
      "epoch": 5.878001921229587,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021641366465598023,
      "loss": 4.6713,
      "mean_token_accuracy": 0.2587598755955696,
      "num_tokens": 140320189.0,
      "step": 61190
    },
    {
      "entropy": 5.639038562774658,
      "epoch": 5.8784822286263205,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002163805643845606,
      "loss": 4.7326,
      "mean_token_accuracy": 0.2601009994745255,
      "num_tokens": 140332426.0,
      "step": 61195
    },
    {
      "entropy": 5.554370403289795,
      "epoch": 5.878962536023055,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021634746547411006,
      "loss": 4.7056,
      "mean_token_accuracy": 0.2556776061654091,
      "num_tokens": 140345572.0,
      "step": 61200
    },
    {
      "entropy": 5.534821844100952,
      "epoch": 5.879442843419788,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002163143679253972,
      "loss": 4.6848,
      "mean_token_accuracy": 0.2598814249038696,
      "num_tokens": 140356323.0,
      "step": 61205
    },
    {
      "entropy": 5.532545280456543,
      "epoch": 5.879923150816523,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021628127173919056,
      "loss": 4.6251,
      "mean_token_accuracy": 0.26322783082723616,
      "num_tokens": 140367794.0,
      "step": 61210
    },
    {
      "entropy": 5.558782291412354,
      "epoch": 5.880403458213257,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002162481769162583,
      "loss": 4.6918,
      "mean_token_accuracy": 0.2569254443049431,
      "num_tokens": 140379884.0,
      "step": 61215
    },
    {
      "entropy": 5.534154939651489,
      "epoch": 5.88088376560999,
      "grad_norm": 1.375,
      "learning_rate": 0.00021621508345736896,
      "loss": 4.6364,
      "mean_token_accuracy": 0.2580986484885216,
      "num_tokens": 140390952.0,
      "step": 61220
    },
    {
      "entropy": 5.517537212371826,
      "epoch": 5.881364073006724,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000216181991363291,
      "loss": 4.6622,
      "mean_token_accuracy": 0.257511804997921,
      "num_tokens": 140401586.0,
      "step": 61225
    },
    {
      "entropy": 5.533567142486572,
      "epoch": 5.881844380403458,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021614890063479238,
      "loss": 4.6752,
      "mean_token_accuracy": 0.2567374959588051,
      "num_tokens": 140414961.0,
      "step": 61230
    },
    {
      "entropy": 5.500997400283813,
      "epoch": 5.8823246878001925,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021611581127264153,
      "loss": 4.6039,
      "mean_token_accuracy": 0.26062176674604415,
      "num_tokens": 140426463.0,
      "step": 61235
    },
    {
      "entropy": 5.566974878311157,
      "epoch": 5.882804995196926,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002160827232776067,
      "loss": 4.6421,
      "mean_token_accuracy": 0.2655740469694138,
      "num_tokens": 140438055.0,
      "step": 61240
    },
    {
      "entropy": 5.516263151168824,
      "epoch": 5.88328530259366,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021604963665045607,
      "loss": 4.5812,
      "mean_token_accuracy": 0.26309223622083666,
      "num_tokens": 140448772.0,
      "step": 61245
    },
    {
      "entropy": 5.5903074741363525,
      "epoch": 5.883765609990394,
      "grad_norm": 1.375,
      "learning_rate": 0.0002160165513919578,
      "loss": 4.7094,
      "mean_token_accuracy": 0.25638037025928495,
      "num_tokens": 140458919.0,
      "step": 61250
    },
    {
      "entropy": 5.5067192077636715,
      "epoch": 5.884245917387128,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021598346750288023,
      "loss": 4.6574,
      "mean_token_accuracy": 0.26212007105350493,
      "num_tokens": 140470238.0,
      "step": 61255
    },
    {
      "entropy": 5.520710611343384,
      "epoch": 5.884726224783861,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021595038498399116,
      "loss": 4.6024,
      "mean_token_accuracy": 0.2691893458366394,
      "num_tokens": 140481976.0,
      "step": 61260
    },
    {
      "entropy": 5.503508615493774,
      "epoch": 5.885206532180596,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021591730383605884,
      "loss": 4.6578,
      "mean_token_accuracy": 0.2648516118526459,
      "num_tokens": 140494058.0,
      "step": 61265
    },
    {
      "entropy": 5.55854754447937,
      "epoch": 5.885686839577329,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002158842240598513,
      "loss": 4.6845,
      "mean_token_accuracy": 0.26152886003255843,
      "num_tokens": 140504473.0,
      "step": 61270
    },
    {
      "entropy": 5.574283599853516,
      "epoch": 5.8861671469740635,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021585114565613652,
      "loss": 4.775,
      "mean_token_accuracy": 0.2516148641705513,
      "num_tokens": 140517104.0,
      "step": 61275
    },
    {
      "entropy": 5.56144347190857,
      "epoch": 5.886647454370797,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021581806862568247,
      "loss": 4.5633,
      "mean_token_accuracy": 0.2722728356719017,
      "num_tokens": 140528336.0,
      "step": 61280
    },
    {
      "entropy": 5.585961389541626,
      "epoch": 5.887127761767531,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002157849929692573,
      "loss": 4.68,
      "mean_token_accuracy": 0.2637165144085884,
      "num_tokens": 140539985.0,
      "step": 61285
    },
    {
      "entropy": 5.5200238704681395,
      "epoch": 5.887608069164266,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002157519186876286,
      "loss": 4.5984,
      "mean_token_accuracy": 0.27087369114160537,
      "num_tokens": 140551544.0,
      "step": 61290
    },
    {
      "entropy": 5.511615371704101,
      "epoch": 5.888088376560999,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002157188457815644,
      "loss": 4.6465,
      "mean_token_accuracy": 0.26375734657049177,
      "num_tokens": 140563219.0,
      "step": 61295
    },
    {
      "entropy": 5.557224130630493,
      "epoch": 5.8885686839577325,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021568577425183267,
      "loss": 4.5918,
      "mean_token_accuracy": 0.26815681904554367,
      "num_tokens": 140574357.0,
      "step": 61300
    },
    {
      "entropy": 5.541649055480957,
      "epoch": 5.889048991354467,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021565270409920108,
      "loss": 4.6568,
      "mean_token_accuracy": 0.2697086468338966,
      "num_tokens": 140586591.0,
      "step": 61305
    },
    {
      "entropy": 5.527759647369384,
      "epoch": 5.889529298751201,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021561963532443746,
      "loss": 4.6412,
      "mean_token_accuracy": 0.26136278063058854,
      "num_tokens": 140597177.0,
      "step": 61310
    },
    {
      "entropy": 5.529057836532592,
      "epoch": 5.890009606147935,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021558656792830967,
      "loss": 4.6529,
      "mean_token_accuracy": 0.2640748083591461,
      "num_tokens": 140608517.0,
      "step": 61315
    },
    {
      "entropy": 5.513595294952393,
      "epoch": 5.890489913544669,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002155535019115853,
      "loss": 4.6336,
      "mean_token_accuracy": 0.2621897265315056,
      "num_tokens": 140619386.0,
      "step": 61320
    },
    {
      "entropy": 5.468540954589844,
      "epoch": 5.890970220941402,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021552043727503207,
      "loss": 4.6105,
      "mean_token_accuracy": 0.2622104465961456,
      "num_tokens": 140632658.0,
      "step": 61325
    },
    {
      "entropy": 5.55156831741333,
      "epoch": 5.891450528338137,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021548737401941764,
      "loss": 4.6108,
      "mean_token_accuracy": 0.2669148415327072,
      "num_tokens": 140643393.0,
      "step": 61330
    },
    {
      "entropy": 5.562274026870727,
      "epoch": 5.89193083573487,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002154543121455097,
      "loss": 4.7401,
      "mean_token_accuracy": 0.25649765878915787,
      "num_tokens": 140655172.0,
      "step": 61335
    },
    {
      "entropy": 5.493505334854126,
      "epoch": 5.892411143131604,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021542125165407584,
      "loss": 4.5537,
      "mean_token_accuracy": 0.27107581347227094,
      "num_tokens": 140665252.0,
      "step": 61340
    },
    {
      "entropy": 5.592222213745117,
      "epoch": 5.892891450528338,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021538819254588365,
      "loss": 4.6849,
      "mean_token_accuracy": 0.26492991894483564,
      "num_tokens": 140675935.0,
      "step": 61345
    },
    {
      "entropy": 5.516985654830933,
      "epoch": 5.893371757925072,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021535513482170054,
      "loss": 4.6773,
      "mean_token_accuracy": 0.258355975151062,
      "num_tokens": 140687597.0,
      "step": 61350
    },
    {
      "entropy": 5.543391227722168,
      "epoch": 5.893852065321806,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021532207848229412,
      "loss": 4.6253,
      "mean_token_accuracy": 0.2692916452884674,
      "num_tokens": 140699524.0,
      "step": 61355
    },
    {
      "entropy": 5.58677306175232,
      "epoch": 5.89433237271854,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021528902352843183,
      "loss": 4.7623,
      "mean_token_accuracy": 0.24922287613153457,
      "num_tokens": 140711806.0,
      "step": 61360
    },
    {
      "entropy": 5.450173234939575,
      "epoch": 5.894812680115274,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021525596996088104,
      "loss": 4.5898,
      "mean_token_accuracy": 0.27095168828964233,
      "num_tokens": 140722619.0,
      "step": 61365
    },
    {
      "entropy": 5.520827388763427,
      "epoch": 5.895292987512008,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002152229177804093,
      "loss": 4.622,
      "mean_token_accuracy": 0.2545802041888237,
      "num_tokens": 140733858.0,
      "step": 61370
    },
    {
      "entropy": 5.524343109130859,
      "epoch": 5.895773294908741,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002151898669877839,
      "loss": 4.6326,
      "mean_token_accuracy": 0.263246887922287,
      "num_tokens": 140746433.0,
      "step": 61375
    },
    {
      "entropy": 5.613786888122559,
      "epoch": 5.8962536023054755,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021515681758377226,
      "loss": 4.7132,
      "mean_token_accuracy": 0.2592561960220337,
      "num_tokens": 140756154.0,
      "step": 61380
    },
    {
      "entropy": 5.579229116439819,
      "epoch": 5.89673390970221,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021512376956914155,
      "loss": 4.75,
      "mean_token_accuracy": 0.25170755088329316,
      "num_tokens": 140767331.0,
      "step": 61385
    },
    {
      "entropy": 5.5818115234375,
      "epoch": 5.897214217098943,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021509072294465903,
      "loss": 4.6275,
      "mean_token_accuracy": 0.26385619640350344,
      "num_tokens": 140777485.0,
      "step": 61390
    },
    {
      "entropy": 5.515258073806763,
      "epoch": 5.8976945244956775,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021505767771109213,
      "loss": 4.6385,
      "mean_token_accuracy": 0.2599785506725311,
      "num_tokens": 140788526.0,
      "step": 61395
    },
    {
      "entropy": 5.491332912445069,
      "epoch": 5.898174831892411,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021502463386920796,
      "loss": 4.6133,
      "mean_token_accuracy": 0.2692064180970192,
      "num_tokens": 140800028.0,
      "step": 61400
    },
    {
      "entropy": 5.523237991333008,
      "epoch": 5.898655139289145,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021499159141977366,
      "loss": 4.6227,
      "mean_token_accuracy": 0.2690308064222336,
      "num_tokens": 140810930.0,
      "step": 61405
    },
    {
      "entropy": 5.526254320144654,
      "epoch": 5.899135446685879,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021495855036355644,
      "loss": 4.6004,
      "mean_token_accuracy": 0.2611606359481812,
      "num_tokens": 140822785.0,
      "step": 61410
    },
    {
      "entropy": 5.505383777618408,
      "epoch": 5.899615754082613,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002149255107013234,
      "loss": 4.6757,
      "mean_token_accuracy": 0.26323762685060503,
      "num_tokens": 140835576.0,
      "step": 61415
    },
    {
      "entropy": 5.553416299819946,
      "epoch": 5.9000960614793465,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021489247243384164,
      "loss": 4.6469,
      "mean_token_accuracy": 0.2686583623290062,
      "num_tokens": 140846610.0,
      "step": 61420
    },
    {
      "entropy": 5.658548498153687,
      "epoch": 5.900576368876081,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002148594355618781,
      "loss": 4.7,
      "mean_token_accuracy": 0.25525894314050673,
      "num_tokens": 140857641.0,
      "step": 61425
    },
    {
      "entropy": 5.531138134002686,
      "epoch": 5.901056676272814,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021482640008619998,
      "loss": 4.5963,
      "mean_token_accuracy": 0.2640419155359268,
      "num_tokens": 140868752.0,
      "step": 61430
    },
    {
      "entropy": 5.489720439910888,
      "epoch": 5.901536983669549,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021479336600757414,
      "loss": 4.6304,
      "mean_token_accuracy": 0.26593522876501086,
      "num_tokens": 140879964.0,
      "step": 61435
    },
    {
      "entropy": 5.562710523605347,
      "epoch": 5.902017291066282,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021476033332676754,
      "loss": 4.7099,
      "mean_token_accuracy": 0.2577583357691765,
      "num_tokens": 140891372.0,
      "step": 61440
    },
    {
      "entropy": 5.576767539978027,
      "epoch": 5.902497598463016,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021472730204454715,
      "loss": 4.7286,
      "mean_token_accuracy": 0.25811557173728944,
      "num_tokens": 140903551.0,
      "step": 61445
    },
    {
      "entropy": 5.609074354171753,
      "epoch": 5.90297790585975,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002146942721616798,
      "loss": 4.7356,
      "mean_token_accuracy": 0.25829570144414904,
      "num_tokens": 140915688.0,
      "step": 61450
    },
    {
      "entropy": 5.5820472717285154,
      "epoch": 5.903458213256484,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021466124367893236,
      "loss": 4.721,
      "mean_token_accuracy": 0.2534581795334816,
      "num_tokens": 140929207.0,
      "step": 61455
    },
    {
      "entropy": 5.532747364044189,
      "epoch": 5.9039385206532184,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021462821659707176,
      "loss": 4.5524,
      "mean_token_accuracy": 0.2748356953263283,
      "num_tokens": 140940553.0,
      "step": 61460
    },
    {
      "entropy": 5.550593996047974,
      "epoch": 5.904418828049952,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002145951909168647,
      "loss": 4.6627,
      "mean_token_accuracy": 0.2559382975101471,
      "num_tokens": 140951826.0,
      "step": 61465
    },
    {
      "entropy": 5.6171228885650635,
      "epoch": 5.904899135446686,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002145621666390779,
      "loss": 4.6991,
      "mean_token_accuracy": 0.2523166611790657,
      "num_tokens": 140963441.0,
      "step": 61470
    },
    {
      "entropy": 5.6014691352844235,
      "epoch": 5.90537944284342,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021452914376447814,
      "loss": 4.704,
      "mean_token_accuracy": 0.2647988647222519,
      "num_tokens": 140974808.0,
      "step": 61475
    },
    {
      "entropy": 5.535645246505737,
      "epoch": 5.905859750240154,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021449612229383202,
      "loss": 4.6612,
      "mean_token_accuracy": 0.26510401666164396,
      "num_tokens": 140986910.0,
      "step": 61480
    },
    {
      "entropy": 5.593220567703247,
      "epoch": 5.906340057636887,
      "grad_norm": 1.375,
      "learning_rate": 0.0002144631022279064,
      "loss": 4.7483,
      "mean_token_accuracy": 0.25397634953260423,
      "num_tokens": 140998446.0,
      "step": 61485
    },
    {
      "entropy": 5.541077184677124,
      "epoch": 5.906820365033622,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002144300835674677,
      "loss": 4.6467,
      "mean_token_accuracy": 0.2626450791954994,
      "num_tokens": 141009198.0,
      "step": 61490
    },
    {
      "entropy": 5.540299892425537,
      "epoch": 5.907300672430355,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021439706631328276,
      "loss": 4.6196,
      "mean_token_accuracy": 0.26632470488548277,
      "num_tokens": 141020790.0,
      "step": 61495
    },
    {
      "entropy": 5.55344181060791,
      "epoch": 5.9077809798270895,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021436405046611779,
      "loss": 4.6567,
      "mean_token_accuracy": 0.25531993210315707,
      "num_tokens": 141031921.0,
      "step": 61500
    },
    {
      "entropy": 5.486753320693969,
      "epoch": 5.908261287223823,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002143310360267396,
      "loss": 4.5498,
      "mean_token_accuracy": 0.2681242272257805,
      "num_tokens": 141042999.0,
      "step": 61505
    },
    {
      "entropy": 5.489978361129761,
      "epoch": 5.908741594620557,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002142980229959146,
      "loss": 4.6549,
      "mean_token_accuracy": 0.2656893104314804,
      "num_tokens": 141055057.0,
      "step": 61510
    },
    {
      "entropy": 5.442947578430176,
      "epoch": 5.909221902017291,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021426501137440922,
      "loss": 4.524,
      "mean_token_accuracy": 0.27828930616378783,
      "num_tokens": 141066238.0,
      "step": 61515
    },
    {
      "entropy": 5.584312391281128,
      "epoch": 5.909702209414025,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021423200116298997,
      "loss": 4.7416,
      "mean_token_accuracy": 0.2594525009393692,
      "num_tokens": 141077426.0,
      "step": 61520
    },
    {
      "entropy": 5.5310546875,
      "epoch": 5.9101825168107585,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002141989923624233,
      "loss": 4.6329,
      "mean_token_accuracy": 0.2655400693416595,
      "num_tokens": 141087189.0,
      "step": 61525
    },
    {
      "entropy": 5.578344106674194,
      "epoch": 5.910662824207493,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002141659849734754,
      "loss": 4.6557,
      "mean_token_accuracy": 0.26811119318008425,
      "num_tokens": 141096715.0,
      "step": 61530
    },
    {
      "entropy": 5.4369957447052,
      "epoch": 5.911143131604227,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021413297899691263,
      "loss": 4.5471,
      "mean_token_accuracy": 0.2735160499811172,
      "num_tokens": 141107775.0,
      "step": 61535
    },
    {
      "entropy": 5.453733253479004,
      "epoch": 5.9116234390009605,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021409997443350143,
      "loss": 4.5403,
      "mean_token_accuracy": 0.2741753816604614,
      "num_tokens": 141118428.0,
      "step": 61540
    },
    {
      "entropy": 5.518336772918701,
      "epoch": 5.912103746397695,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021406697128400793,
      "loss": 4.6817,
      "mean_token_accuracy": 0.25365377515554427,
      "num_tokens": 141129270.0,
      "step": 61545
    },
    {
      "entropy": 5.505549144744873,
      "epoch": 5.912584053794428,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021403396954919842,
      "loss": 4.651,
      "mean_token_accuracy": 0.26058964133262635,
      "num_tokens": 141140898.0,
      "step": 61550
    },
    {
      "entropy": 5.548992443084717,
      "epoch": 5.913064361191163,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021400096922983918,
      "loss": 4.6425,
      "mean_token_accuracy": 0.26264303773641584,
      "num_tokens": 141153103.0,
      "step": 61555
    },
    {
      "entropy": 5.527091407775879,
      "epoch": 5.913544668587896,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002139679703266963,
      "loss": 4.6049,
      "mean_token_accuracy": 0.2713927462697029,
      "num_tokens": 141163876.0,
      "step": 61560
    },
    {
      "entropy": 5.6396894454956055,
      "epoch": 5.91402497598463,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021393497284053587,
      "loss": 4.8135,
      "mean_token_accuracy": 0.2524599328637123,
      "num_tokens": 141174145.0,
      "step": 61565
    },
    {
      "entropy": 5.529948568344116,
      "epoch": 5.914505283381364,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021390197677212402,
      "loss": 4.6003,
      "mean_token_accuracy": 0.268857279419899,
      "num_tokens": 141184572.0,
      "step": 61570
    },
    {
      "entropy": 5.634138727188111,
      "epoch": 5.914985590778098,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002138689821222269,
      "loss": 4.7652,
      "mean_token_accuracy": 0.24356772601604462,
      "num_tokens": 141197616.0,
      "step": 61575
    },
    {
      "entropy": 5.474820470809936,
      "epoch": 5.915465898174832,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021383598889161045,
      "loss": 4.5511,
      "mean_token_accuracy": 0.27278693914413454,
      "num_tokens": 141208264.0,
      "step": 61580
    },
    {
      "entropy": 5.484280586242676,
      "epoch": 5.915946205571566,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021380299708104078,
      "loss": 4.6245,
      "mean_token_accuracy": 0.2642259120941162,
      "num_tokens": 141219621.0,
      "step": 61585
    },
    {
      "entropy": 5.551556539535523,
      "epoch": 5.916426512968299,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021377000669128377,
      "loss": 4.7092,
      "mean_token_accuracy": 0.2661742433905602,
      "num_tokens": 141230924.0,
      "step": 61590
    },
    {
      "entropy": 5.518836736679077,
      "epoch": 5.916906820365034,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021373701772310538,
      "loss": 4.5721,
      "mean_token_accuracy": 0.27399367541074754,
      "num_tokens": 141241401.0,
      "step": 61595
    },
    {
      "entropy": 5.54680495262146,
      "epoch": 5.917387127761767,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021370403017727152,
      "loss": 4.7056,
      "mean_token_accuracy": 0.2565073177218437,
      "num_tokens": 141252030.0,
      "step": 61600
    },
    {
      "entropy": 5.569205331802368,
      "epoch": 5.917867435158501,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021367104405454806,
      "loss": 4.726,
      "mean_token_accuracy": 0.25163298547267915,
      "num_tokens": 141263678.0,
      "step": 61605
    },
    {
      "entropy": 5.558939743041992,
      "epoch": 5.918347742555236,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021363805935570085,
      "loss": 4.636,
      "mean_token_accuracy": 0.25931423902511597,
      "num_tokens": 141276207.0,
      "step": 61610
    },
    {
      "entropy": 5.549807834625244,
      "epoch": 5.918828049951969,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021360507608149577,
      "loss": 4.6191,
      "mean_token_accuracy": 0.27008501887321473,
      "num_tokens": 141288677.0,
      "step": 61615
    },
    {
      "entropy": 5.599237155914307,
      "epoch": 5.9193083573487035,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021357209423269847,
      "loss": 4.7423,
      "mean_token_accuracy": 0.2572954371571541,
      "num_tokens": 141299530.0,
      "step": 61620
    },
    {
      "entropy": 5.4919675350189205,
      "epoch": 5.919788664745437,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021353911381007473,
      "loss": 4.6797,
      "mean_token_accuracy": 0.2581424921751022,
      "num_tokens": 141311611.0,
      "step": 61625
    },
    {
      "entropy": 5.581992483139038,
      "epoch": 5.920268972142171,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021350613481439017,
      "loss": 4.6588,
      "mean_token_accuracy": 0.25978380143642427,
      "num_tokens": 141321446.0,
      "step": 61630
    },
    {
      "entropy": 5.546506929397583,
      "epoch": 5.920749279538905,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002134731572464107,
      "loss": 4.6237,
      "mean_token_accuracy": 0.26350701600313187,
      "num_tokens": 141333865.0,
      "step": 61635
    },
    {
      "entropy": 5.513963317871093,
      "epoch": 5.921229586935639,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002134401811069017,
      "loss": 4.5937,
      "mean_token_accuracy": 0.26694830507040024,
      "num_tokens": 141345654.0,
      "step": 61640
    },
    {
      "entropy": 5.5235827445983885,
      "epoch": 5.9217098943323725,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021340720639662902,
      "loss": 4.6252,
      "mean_token_accuracy": 0.2758466064929962,
      "num_tokens": 141356295.0,
      "step": 61645
    },
    {
      "entropy": 5.496722412109375,
      "epoch": 5.922190201729107,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021337423311635812,
      "loss": 4.6153,
      "mean_token_accuracy": 0.2695568189024925,
      "num_tokens": 141367743.0,
      "step": 61650
    },
    {
      "entropy": 5.548123550415039,
      "epoch": 5.92267050912584,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021334126126685448,
      "loss": 4.6219,
      "mean_token_accuracy": 0.2689269229769707,
      "num_tokens": 141378842.0,
      "step": 61655
    },
    {
      "entropy": 5.4516815662384035,
      "epoch": 5.923150816522575,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021330829084888366,
      "loss": 4.5644,
      "mean_token_accuracy": 0.26996535807847977,
      "num_tokens": 141389343.0,
      "step": 61660
    },
    {
      "entropy": 5.543186616897583,
      "epoch": 5.923631123919308,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021327532186321107,
      "loss": 4.6718,
      "mean_token_accuracy": 0.2654263094067574,
      "num_tokens": 141400255.0,
      "step": 61665
    },
    {
      "entropy": 5.515135669708252,
      "epoch": 5.924111431316042,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021324235431060236,
      "loss": 4.6279,
      "mean_token_accuracy": 0.2663297563791275,
      "num_tokens": 141412197.0,
      "step": 61670
    },
    {
      "entropy": 5.53442063331604,
      "epoch": 5.924591738712776,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021320938819182272,
      "loss": 4.6019,
      "mean_token_accuracy": 0.26700019687414167,
      "num_tokens": 141423486.0,
      "step": 61675
    },
    {
      "entropy": 5.576445388793945,
      "epoch": 5.92507204610951,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002131764235076376,
      "loss": 4.6842,
      "mean_token_accuracy": 0.2585097447037697,
      "num_tokens": 141434980.0,
      "step": 61680
    },
    {
      "entropy": 5.501513862609864,
      "epoch": 5.925552353506244,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021314346025881237,
      "loss": 4.6642,
      "mean_token_accuracy": 0.26320810317993165,
      "num_tokens": 141445491.0,
      "step": 61685
    },
    {
      "entropy": 5.551715850830078,
      "epoch": 5.926032660902978,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021311049844611234,
      "loss": 4.6338,
      "mean_token_accuracy": 0.26313233077526094,
      "num_tokens": 141456719.0,
      "step": 61690
    },
    {
      "entropy": 5.421948957443237,
      "epoch": 5.926512968299712,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021307753807030277,
      "loss": 4.5335,
      "mean_token_accuracy": 0.2814964443445206,
      "num_tokens": 141469176.0,
      "step": 61695
    },
    {
      "entropy": 5.580943298339844,
      "epoch": 5.926993275696446,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021304457913214885,
      "loss": 4.6833,
      "mean_token_accuracy": 0.2582657620310783,
      "num_tokens": 141479886.0,
      "step": 61700
    },
    {
      "entropy": 5.618884038925171,
      "epoch": 5.92747358309318,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021301162163241588,
      "loss": 4.7251,
      "mean_token_accuracy": 0.25936992317438123,
      "num_tokens": 141490416.0,
      "step": 61705
    },
    {
      "entropy": 5.528534269332885,
      "epoch": 5.927953890489913,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021297866557186895,
      "loss": 4.6326,
      "mean_token_accuracy": 0.2622279107570648,
      "num_tokens": 141501903.0,
      "step": 61710
    },
    {
      "entropy": 5.567389011383057,
      "epoch": 5.928434197886648,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002129457109512732,
      "loss": 4.7144,
      "mean_token_accuracy": 0.2579793319106102,
      "num_tokens": 141514151.0,
      "step": 61715
    },
    {
      "entropy": 5.58990707397461,
      "epoch": 5.928914505283381,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002129127577713938,
      "loss": 4.6582,
      "mean_token_accuracy": 0.25793192684650423,
      "num_tokens": 141524454.0,
      "step": 61720
    },
    {
      "entropy": 5.499668073654175,
      "epoch": 5.9293948126801155,
      "grad_norm": 1.375,
      "learning_rate": 0.00021287980603299585,
      "loss": 4.592,
      "mean_token_accuracy": 0.2687947615981102,
      "num_tokens": 141535778.0,
      "step": 61725
    },
    {
      "entropy": 5.577361011505127,
      "epoch": 5.929875120076849,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002128468557368443,
      "loss": 4.7255,
      "mean_token_accuracy": 0.25920871943235396,
      "num_tokens": 141547574.0,
      "step": 61730
    },
    {
      "entropy": 5.512040472030639,
      "epoch": 5.930355427473583,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021281390688370435,
      "loss": 4.6318,
      "mean_token_accuracy": 0.26464777886867524,
      "num_tokens": 141559590.0,
      "step": 61735
    },
    {
      "entropy": 5.601073694229126,
      "epoch": 5.930835734870317,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021278095947434067,
      "loss": 4.7037,
      "mean_token_accuracy": 0.26244949251413346,
      "num_tokens": 141571057.0,
      "step": 61740
    },
    {
      "entropy": 5.526744747161866,
      "epoch": 5.931316042267051,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002127480135095184,
      "loss": 4.6104,
      "mean_token_accuracy": 0.2703842014074326,
      "num_tokens": 141581160.0,
      "step": 61745
    },
    {
      "entropy": 5.597484731674195,
      "epoch": 5.931796349663784,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002127150689900023,
      "loss": 4.7458,
      "mean_token_accuracy": 0.25624427795410154,
      "num_tokens": 141592269.0,
      "step": 61750
    },
    {
      "entropy": 5.564985227584839,
      "epoch": 5.932276657060519,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021268212591655743,
      "loss": 4.6927,
      "mean_token_accuracy": 0.25433144718408585,
      "num_tokens": 141604071.0,
      "step": 61755
    },
    {
      "entropy": 5.621315908432007,
      "epoch": 5.932756964457253,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021264918428994857,
      "loss": 4.7542,
      "mean_token_accuracy": 0.25321154296398163,
      "num_tokens": 141614401.0,
      "step": 61760
    },
    {
      "entropy": 5.5518018245697025,
      "epoch": 5.9332372718539865,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002126162441109406,
      "loss": 4.6729,
      "mean_token_accuracy": 0.2636876553297043,
      "num_tokens": 141625901.0,
      "step": 61765
    },
    {
      "entropy": 5.529810190200806,
      "epoch": 5.93371757925072,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021258330538029808,
      "loss": 4.6141,
      "mean_token_accuracy": 0.26794005632400514,
      "num_tokens": 141636212.0,
      "step": 61770
    },
    {
      "entropy": 5.428330564498902,
      "epoch": 5.934197886647454,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021255036809878587,
      "loss": 4.5519,
      "mean_token_accuracy": 0.26569567918777465,
      "num_tokens": 141647457.0,
      "step": 61775
    },
    {
      "entropy": 5.4559863090515135,
      "epoch": 5.934678194044189,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021251743226716869,
      "loss": 4.5881,
      "mean_token_accuracy": 0.2617605417966843,
      "num_tokens": 141659132.0,
      "step": 61780
    },
    {
      "entropy": 5.502321100234985,
      "epoch": 5.935158501440922,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021248449788621122,
      "loss": 4.6671,
      "mean_token_accuracy": 0.25726415812969206,
      "num_tokens": 141671812.0,
      "step": 61785
    },
    {
      "entropy": 5.551694869995117,
      "epoch": 5.935638808837656,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021245156495667807,
      "loss": 4.5434,
      "mean_token_accuracy": 0.26720012426376344,
      "num_tokens": 141681844.0,
      "step": 61790
    },
    {
      "entropy": 5.56373701095581,
      "epoch": 5.93611911623439,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021241863347933387,
      "loss": 4.6988,
      "mean_token_accuracy": 0.2611274540424347,
      "num_tokens": 141693703.0,
      "step": 61795
    },
    {
      "entropy": 5.546082019805908,
      "epoch": 5.936599423631124,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021238570345494313,
      "loss": 4.672,
      "mean_token_accuracy": 0.2537045106291771,
      "num_tokens": 141705683.0,
      "step": 61800
    },
    {
      "entropy": 5.507821464538575,
      "epoch": 5.937079731027858,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021235277488427045,
      "loss": 4.6093,
      "mean_token_accuracy": 0.26528044492006303,
      "num_tokens": 141716759.0,
      "step": 61805
    },
    {
      "entropy": 5.553246736526489,
      "epoch": 5.937560038424592,
      "grad_norm": 1.375,
      "learning_rate": 0.00021231984776808028,
      "loss": 4.6036,
      "mean_token_accuracy": 0.2644908294081688,
      "num_tokens": 141727410.0,
      "step": 61810
    },
    {
      "entropy": 5.583773279190064,
      "epoch": 5.938040345821325,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021228692210713713,
      "loss": 4.7396,
      "mean_token_accuracy": 0.2617565467953682,
      "num_tokens": 141738503.0,
      "step": 61815
    },
    {
      "entropy": 5.545973539352417,
      "epoch": 5.93852065321806,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021225399790220545,
      "loss": 4.7335,
      "mean_token_accuracy": 0.2543173387646675,
      "num_tokens": 141749964.0,
      "step": 61820
    },
    {
      "entropy": 5.534628057479859,
      "epoch": 5.939000960614793,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002122210751540497,
      "loss": 4.6891,
      "mean_token_accuracy": 0.26030791997909547,
      "num_tokens": 141761164.0,
      "step": 61825
    },
    {
      "entropy": 5.5288163185119625,
      "epoch": 5.939481268011527,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002121881538634341,
      "loss": 4.6174,
      "mean_token_accuracy": 0.2645660534501076,
      "num_tokens": 141772545.0,
      "step": 61830
    },
    {
      "entropy": 5.599167251586914,
      "epoch": 5.939961575408262,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021215523403112302,
      "loss": 4.7087,
      "mean_token_accuracy": 0.26770423352718353,
      "num_tokens": 141783445.0,
      "step": 61835
    },
    {
      "entropy": 5.491690254211425,
      "epoch": 5.940441882804995,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021212231565788083,
      "loss": 4.6132,
      "mean_token_accuracy": 0.2675896853208542,
      "num_tokens": 141796084.0,
      "step": 61840
    },
    {
      "entropy": 5.502481985092163,
      "epoch": 5.940922190201729,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021208939874447176,
      "loss": 4.6532,
      "mean_token_accuracy": 0.26034881323575976,
      "num_tokens": 141807445.0,
      "step": 61845
    },
    {
      "entropy": 5.475606203079224,
      "epoch": 5.941402497598463,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021205648329166,
      "loss": 4.5889,
      "mean_token_accuracy": 0.26403638273477553,
      "num_tokens": 141819596.0,
      "step": 61850
    },
    {
      "entropy": 5.643578052520752,
      "epoch": 5.941882804995197,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021202356930020988,
      "loss": 4.7524,
      "mean_token_accuracy": 0.24811681509017944,
      "num_tokens": 141831750.0,
      "step": 61855
    },
    {
      "entropy": 5.564232349395752,
      "epoch": 5.942363112391931,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002119906567708854,
      "loss": 4.5842,
      "mean_token_accuracy": 0.2582828402519226,
      "num_tokens": 141842123.0,
      "step": 61860
    },
    {
      "entropy": 5.630542993545532,
      "epoch": 5.942843419788665,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002119577457044508,
      "loss": 4.6719,
      "mean_token_accuracy": 0.25991900861263273,
      "num_tokens": 141851854.0,
      "step": 61865
    },
    {
      "entropy": 5.530697917938232,
      "epoch": 5.9433237271853985,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021192483610167018,
      "loss": 4.6663,
      "mean_token_accuracy": 0.2582010507583618,
      "num_tokens": 141864508.0,
      "step": 61870
    },
    {
      "entropy": 5.442307233810425,
      "epoch": 5.943804034582133,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021189192796330754,
      "loss": 4.5729,
      "mean_token_accuracy": 0.2739676386117935,
      "num_tokens": 141876176.0,
      "step": 61875
    },
    {
      "entropy": 5.502853965759277,
      "epoch": 5.944284341978866,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021185902129012692,
      "loss": 4.6992,
      "mean_token_accuracy": 0.2628228336572647,
      "num_tokens": 141887954.0,
      "step": 61880
    },
    {
      "entropy": 5.547378635406494,
      "epoch": 5.944764649375601,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021182611608289243,
      "loss": 4.6164,
      "mean_token_accuracy": 0.2734766036272049,
      "num_tokens": 141899303.0,
      "step": 61885
    },
    {
      "entropy": 5.564608812332153,
      "epoch": 5.945244956772334,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002117932123423678,
      "loss": 4.661,
      "mean_token_accuracy": 0.2610130473971367,
      "num_tokens": 141911892.0,
      "step": 61890
    },
    {
      "entropy": 5.582547330856324,
      "epoch": 5.945725264169068,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021176031006931716,
      "loss": 4.7189,
      "mean_token_accuracy": 0.2554671436548233,
      "num_tokens": 141923650.0,
      "step": 61895
    },
    {
      "entropy": 5.568554067611695,
      "epoch": 5.946205571565802,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021172740926450435,
      "loss": 4.7557,
      "mean_token_accuracy": 0.2573845773935318,
      "num_tokens": 141936132.0,
      "step": 61900
    },
    {
      "entropy": 5.645052766799926,
      "epoch": 5.946685878962536,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021169450992869317,
      "loss": 4.7068,
      "mean_token_accuracy": 0.25573190599679946,
      "num_tokens": 141947044.0,
      "step": 61905
    },
    {
      "entropy": 5.488313436508179,
      "epoch": 5.94716618635927,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002116616120626475,
      "loss": 4.6189,
      "mean_token_accuracy": 0.2653516709804535,
      "num_tokens": 141958366.0,
      "step": 61910
    },
    {
      "entropy": 5.561206865310669,
      "epoch": 5.947646493756004,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021162871566713122,
      "loss": 4.7023,
      "mean_token_accuracy": 0.2565362885594368,
      "num_tokens": 141970219.0,
      "step": 61915
    },
    {
      "entropy": 5.632935380935669,
      "epoch": 5.948126801152737,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002115958207429079,
      "loss": 4.689,
      "mean_token_accuracy": 0.25579327493906023,
      "num_tokens": 141980680.0,
      "step": 61920
    },
    {
      "entropy": 5.599103164672852,
      "epoch": 5.948607108549472,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002115629272907414,
      "loss": 4.7182,
      "mean_token_accuracy": 0.2543082430958748,
      "num_tokens": 141991764.0,
      "step": 61925
    },
    {
      "entropy": 5.586736440658569,
      "epoch": 5.949087415946206,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002115300353113953,
      "loss": 4.7307,
      "mean_token_accuracy": 0.25855404436588286,
      "num_tokens": 142002628.0,
      "step": 61930
    },
    {
      "entropy": 5.5669654369354244,
      "epoch": 5.949567723342939,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002114971448056333,
      "loss": 4.7033,
      "mean_token_accuracy": 0.2564259722828865,
      "num_tokens": 142013868.0,
      "step": 61935
    },
    {
      "entropy": 5.531270456314087,
      "epoch": 5.950048030739674,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021146425577421907,
      "loss": 4.6555,
      "mean_token_accuracy": 0.2588251084089279,
      "num_tokens": 142025188.0,
      "step": 61940
    },
    {
      "entropy": 5.5605286121368405,
      "epoch": 5.950528338136407,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021143136821791613,
      "loss": 4.7335,
      "mean_token_accuracy": 0.260177481174469,
      "num_tokens": 142035355.0,
      "step": 61945
    },
    {
      "entropy": 5.594437551498413,
      "epoch": 5.9510086455331415,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021139848213748807,
      "loss": 4.7217,
      "mean_token_accuracy": 0.2555141791701317,
      "num_tokens": 142047930.0,
      "step": 61950
    },
    {
      "entropy": 5.61688494682312,
      "epoch": 5.951488952929875,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021136559753369844,
      "loss": 4.6894,
      "mean_token_accuracy": 0.26040235012769697,
      "num_tokens": 142058524.0,
      "step": 61955
    },
    {
      "entropy": 5.553113317489624,
      "epoch": 5.951969260326609,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021133271440731065,
      "loss": 4.6826,
      "mean_token_accuracy": 0.25835772752761843,
      "num_tokens": 142069687.0,
      "step": 61960
    },
    {
      "entropy": 5.630315732955933,
      "epoch": 5.952449567723343,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002112998327590882,
      "loss": 4.729,
      "mean_token_accuracy": 0.2540401041507721,
      "num_tokens": 142082534.0,
      "step": 61965
    },
    {
      "entropy": 5.514696931838989,
      "epoch": 5.952929875120077,
      "grad_norm": 1.5,
      "learning_rate": 0.00021126695258979445,
      "loss": 4.616,
      "mean_token_accuracy": 0.27182695269584656,
      "num_tokens": 142092985.0,
      "step": 61970
    },
    {
      "entropy": 5.5105774879455565,
      "epoch": 5.95341018251681,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021123407390019283,
      "loss": 4.6059,
      "mean_token_accuracy": 0.2624050796031952,
      "num_tokens": 142105022.0,
      "step": 61975
    },
    {
      "entropy": 5.584660243988037,
      "epoch": 5.953890489913545,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021120119669104664,
      "loss": 4.7237,
      "mean_token_accuracy": 0.2574622929096222,
      "num_tokens": 142116683.0,
      "step": 61980
    },
    {
      "entropy": 5.625707054138184,
      "epoch": 5.954370797310279,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021116832096311928,
      "loss": 4.7091,
      "mean_token_accuracy": 0.2618328809738159,
      "num_tokens": 142128083.0,
      "step": 61985
    },
    {
      "entropy": 5.566347646713257,
      "epoch": 5.9548511047070125,
      "grad_norm": 1.5,
      "learning_rate": 0.00021113544671717388,
      "loss": 4.6391,
      "mean_token_accuracy": 0.26483078896999357,
      "num_tokens": 142138807.0,
      "step": 61990
    },
    {
      "entropy": 5.526387929916382,
      "epoch": 5.955331412103746,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021110257395397385,
      "loss": 4.6393,
      "mean_token_accuracy": 0.2635545417666435,
      "num_tokens": 142150772.0,
      "step": 61995
    },
    {
      "entropy": 5.4835042476654055,
      "epoch": 5.95581171950048,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021106970267428227,
      "loss": 4.6437,
      "mean_token_accuracy": 0.2679118990898132,
      "num_tokens": 142161851.0,
      "step": 62000
    },
    {
      "entropy": 5.467178058624268,
      "epoch": 5.956292026897215,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021103683287886245,
      "loss": 4.6015,
      "mean_token_accuracy": 0.2692988678812981,
      "num_tokens": 142173720.0,
      "step": 62005
    },
    {
      "entropy": 5.557041072845459,
      "epoch": 5.956772334293948,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002110039645684773,
      "loss": 4.6767,
      "mean_token_accuracy": 0.26666124612092973,
      "num_tokens": 142184875.0,
      "step": 62010
    },
    {
      "entropy": 5.5295729637146,
      "epoch": 5.957252641690682,
      "grad_norm": 1.25,
      "learning_rate": 0.0002109710977438901,
      "loss": 4.6224,
      "mean_token_accuracy": 0.261667537689209,
      "num_tokens": 142196289.0,
      "step": 62015
    },
    {
      "entropy": 5.541493892669678,
      "epoch": 5.957732949087416,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002109382324058639,
      "loss": 4.6622,
      "mean_token_accuracy": 0.2624338924884796,
      "num_tokens": 142207711.0,
      "step": 62020
    },
    {
      "entropy": 5.5160280704498295,
      "epoch": 5.95821325648415,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021090536855516173,
      "loss": 4.5738,
      "mean_token_accuracy": 0.26832608729600904,
      "num_tokens": 142218753.0,
      "step": 62025
    },
    {
      "entropy": 5.54095573425293,
      "epoch": 5.958693563880884,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021087250619254662,
      "loss": 4.6068,
      "mean_token_accuracy": 0.26220899522304536,
      "num_tokens": 142229328.0,
      "step": 62030
    },
    {
      "entropy": 5.512529802322388,
      "epoch": 5.959173871277618,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021083964531878153,
      "loss": 4.5897,
      "mean_token_accuracy": 0.264933368563652,
      "num_tokens": 142240482.0,
      "step": 62035
    },
    {
      "entropy": 5.662607860565186,
      "epoch": 5.959654178674351,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002108067859346292,
      "loss": 4.7789,
      "mean_token_accuracy": 0.2458129793405533,
      "num_tokens": 142250428.0,
      "step": 62040
    },
    {
      "entropy": 5.614550638198852,
      "epoch": 5.960134486071086,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021077392804085278,
      "loss": 4.7162,
      "mean_token_accuracy": 0.2553910404443741,
      "num_tokens": 142262217.0,
      "step": 62045
    },
    {
      "entropy": 5.660085153579712,
      "epoch": 5.960614793467819,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000210741071638215,
      "loss": 4.7088,
      "mean_token_accuracy": 0.2578634679317474,
      "num_tokens": 142273250.0,
      "step": 62050
    },
    {
      "entropy": 5.5802733421325685,
      "epoch": 5.961095100864553,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021070821672747875,
      "loss": 4.6664,
      "mean_token_accuracy": 0.261997127532959,
      "num_tokens": 142284465.0,
      "step": 62055
    },
    {
      "entropy": 5.5009317874908445,
      "epoch": 5.961575408261287,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021067536330940685,
      "loss": 4.5748,
      "mean_token_accuracy": 0.26950313299894335,
      "num_tokens": 142295875.0,
      "step": 62060
    },
    {
      "entropy": 5.518031215667724,
      "epoch": 5.962055715658021,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021064251138476204,
      "loss": 4.6948,
      "mean_token_accuracy": 0.25427896678447726,
      "num_tokens": 142308000.0,
      "step": 62065
    },
    {
      "entropy": 5.606643724441528,
      "epoch": 5.962536023054755,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021060966095430689,
      "loss": 4.7064,
      "mean_token_accuracy": 0.2664404079318047,
      "num_tokens": 142319641.0,
      "step": 62070
    },
    {
      "entropy": 5.504209423065186,
      "epoch": 5.963016330451489,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021057681201880425,
      "loss": 4.5904,
      "mean_token_accuracy": 0.2626887708902359,
      "num_tokens": 142332909.0,
      "step": 62075
    },
    {
      "entropy": 5.52659387588501,
      "epoch": 5.963496637848223,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021054396457901675,
      "loss": 4.6358,
      "mean_token_accuracy": 0.26182585060596464,
      "num_tokens": 142344617.0,
      "step": 62080
    },
    {
      "entropy": 5.617606735229492,
      "epoch": 5.963976945244957,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021051111863570692,
      "loss": 4.6904,
      "mean_token_accuracy": 0.24939944744110107,
      "num_tokens": 142356024.0,
      "step": 62085
    },
    {
      "entropy": 5.571924495697021,
      "epoch": 5.964457252641691,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002104782741896375,
      "loss": 4.7037,
      "mean_token_accuracy": 0.26255389750003816,
      "num_tokens": 142367558.0,
      "step": 62090
    },
    {
      "entropy": 5.682452630996704,
      "epoch": 5.9649375600384245,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021044543124157107,
      "loss": 4.7745,
      "mean_token_accuracy": 0.2500956282019615,
      "num_tokens": 142379367.0,
      "step": 62095
    },
    {
      "entropy": 5.524271249771118,
      "epoch": 5.965417867435159,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002104125897922699,
      "loss": 4.6272,
      "mean_token_accuracy": 0.26705593019723894,
      "num_tokens": 142391449.0,
      "step": 62100
    },
    {
      "entropy": 5.543116664886474,
      "epoch": 5.965898174831892,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021037974984249665,
      "loss": 4.6564,
      "mean_token_accuracy": 0.2597732856869698,
      "num_tokens": 142403154.0,
      "step": 62105
    },
    {
      "entropy": 5.4900367736816404,
      "epoch": 5.9663784822286265,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021034691139301372,
      "loss": 4.6251,
      "mean_token_accuracy": 0.2624582201242447,
      "num_tokens": 142414762.0,
      "step": 62110
    },
    {
      "entropy": 5.602065753936768,
      "epoch": 5.96685878962536,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002103140744445835,
      "loss": 4.6923,
      "mean_token_accuracy": 0.25854717195034027,
      "num_tokens": 142425458.0,
      "step": 62115
    },
    {
      "entropy": 5.5642698287963865,
      "epoch": 5.967339097022094,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002102812389979684,
      "loss": 4.681,
      "mean_token_accuracy": 0.26355284452438354,
      "num_tokens": 142437565.0,
      "step": 62120
    },
    {
      "entropy": 5.487855291366577,
      "epoch": 5.967819404418828,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021024840505393084,
      "loss": 4.6134,
      "mean_token_accuracy": 0.26281909048557284,
      "num_tokens": 142449476.0,
      "step": 62125
    },
    {
      "entropy": 5.538434076309204,
      "epoch": 5.968299711815562,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021021557261323299,
      "loss": 4.6638,
      "mean_token_accuracy": 0.26210072338581086,
      "num_tokens": 142460231.0,
      "step": 62130
    },
    {
      "entropy": 5.572771120071411,
      "epoch": 5.9687800192122955,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002101827416766371,
      "loss": 4.6734,
      "mean_token_accuracy": 0.25860377252101896,
      "num_tokens": 142471358.0,
      "step": 62135
    },
    {
      "entropy": 5.491818428039551,
      "epoch": 5.96926032660903,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021014991224490553,
      "loss": 4.5728,
      "mean_token_accuracy": 0.26858940422534944,
      "num_tokens": 142481746.0,
      "step": 62140
    },
    {
      "entropy": 5.514382553100586,
      "epoch": 5.969740634005763,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021011708431880046,
      "loss": 4.6309,
      "mean_token_accuracy": 0.26370142996311186,
      "num_tokens": 142494031.0,
      "step": 62145
    },
    {
      "entropy": 5.560198354721069,
      "epoch": 5.970220941402498,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021008425789908398,
      "loss": 4.6587,
      "mean_token_accuracy": 0.2608669713139534,
      "num_tokens": 142506324.0,
      "step": 62150
    },
    {
      "entropy": 5.638709878921508,
      "epoch": 5.970701248799232,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021005143298651836,
      "loss": 4.705,
      "mean_token_accuracy": 0.26003327369689944,
      "num_tokens": 142517317.0,
      "step": 62155
    },
    {
      "entropy": 5.525492858886719,
      "epoch": 5.971181556195965,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021001860958186553,
      "loss": 4.5885,
      "mean_token_accuracy": 0.2614795133471489,
      "num_tokens": 142528171.0,
      "step": 62160
    },
    {
      "entropy": 5.56147952079773,
      "epoch": 5.9716618635927,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020998578768588767,
      "loss": 4.6618,
      "mean_token_accuracy": 0.2629563257098198,
      "num_tokens": 142539446.0,
      "step": 62165
    },
    {
      "entropy": 5.582225799560547,
      "epoch": 5.972142170989433,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002099529672993467,
      "loss": 4.6653,
      "mean_token_accuracy": 0.26836984753608706,
      "num_tokens": 142550100.0,
      "step": 62170
    },
    {
      "entropy": 5.626637697219849,
      "epoch": 5.972622478386167,
      "grad_norm": 1.5,
      "learning_rate": 0.00020992014842300479,
      "loss": 4.6601,
      "mean_token_accuracy": 0.2668990850448608,
      "num_tokens": 142560908.0,
      "step": 62175
    },
    {
      "entropy": 5.520214176177978,
      "epoch": 5.973102785782901,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020988733105762363,
      "loss": 4.6352,
      "mean_token_accuracy": 0.26479664593935015,
      "num_tokens": 142571971.0,
      "step": 62180
    },
    {
      "entropy": 5.489286613464356,
      "epoch": 5.973583093179635,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020985451520396552,
      "loss": 4.5972,
      "mean_token_accuracy": 0.26892402619123457,
      "num_tokens": 142582816.0,
      "step": 62185
    },
    {
      "entropy": 5.501842927932739,
      "epoch": 5.974063400576369,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020982170086279197,
      "loss": 4.6436,
      "mean_token_accuracy": 0.2616840168833733,
      "num_tokens": 142594376.0,
      "step": 62190
    },
    {
      "entropy": 5.6838172435760494,
      "epoch": 5.974543707973103,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020978888803486502,
      "loss": 4.8673,
      "mean_token_accuracy": 0.2519707947969437,
      "num_tokens": 142606379.0,
      "step": 62195
    },
    {
      "entropy": 5.550667572021484,
      "epoch": 5.975024015369836,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020975607672094642,
      "loss": 4.6287,
      "mean_token_accuracy": 0.2612584739923477,
      "num_tokens": 142616994.0,
      "step": 62200
    },
    {
      "entropy": 5.462632751464843,
      "epoch": 5.975504322766571,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000209723266921798,
      "loss": 4.6249,
      "mean_token_accuracy": 0.26632325202226637,
      "num_tokens": 142628043.0,
      "step": 62205
    },
    {
      "entropy": 5.467019939422608,
      "epoch": 5.975984630163304,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020969045863818148,
      "loss": 4.6367,
      "mean_token_accuracy": 0.26211577355861665,
      "num_tokens": 142638715.0,
      "step": 62210
    },
    {
      "entropy": 5.552647686004638,
      "epoch": 5.9764649375600385,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020965765187085862,
      "loss": 4.6876,
      "mean_token_accuracy": 0.25897659063339235,
      "num_tokens": 142649706.0,
      "step": 62215
    },
    {
      "entropy": 5.644637870788574,
      "epoch": 5.976945244956772,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020962484662059107,
      "loss": 4.8543,
      "mean_token_accuracy": 0.2544919416308403,
      "num_tokens": 142661896.0,
      "step": 62220
    },
    {
      "entropy": 5.495627546310425,
      "epoch": 5.977425552353506,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020959204288814042,
      "loss": 4.5555,
      "mean_token_accuracy": 0.2723546549677849,
      "num_tokens": 142674466.0,
      "step": 62225
    },
    {
      "entropy": 5.567780256271362,
      "epoch": 5.977905859750241,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020955924067426828,
      "loss": 4.6351,
      "mean_token_accuracy": 0.2609137862920761,
      "num_tokens": 142684936.0,
      "step": 62230
    },
    {
      "entropy": 5.581378316879272,
      "epoch": 5.978386167146974,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020952643997973632,
      "loss": 4.7173,
      "mean_token_accuracy": 0.24752013087272645,
      "num_tokens": 142696944.0,
      "step": 62235
    },
    {
      "entropy": 5.496189785003662,
      "epoch": 5.978866474543708,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020949364080530592,
      "loss": 4.633,
      "mean_token_accuracy": 0.2603339910507202,
      "num_tokens": 142708150.0,
      "step": 62240
    },
    {
      "entropy": 5.579437208175659,
      "epoch": 5.979346781940442,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002094608431517388,
      "loss": 4.6813,
      "mean_token_accuracy": 0.26317254453897476,
      "num_tokens": 142720084.0,
      "step": 62245
    },
    {
      "entropy": 5.561345338821411,
      "epoch": 5.979827089337176,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020942804701979618,
      "loss": 4.6502,
      "mean_token_accuracy": 0.26341464668512343,
      "num_tokens": 142731126.0,
      "step": 62250
    },
    {
      "entropy": 5.63978214263916,
      "epoch": 5.9803073967339095,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002093952524102396,
      "loss": 4.739,
      "mean_token_accuracy": 0.2563867121934891,
      "num_tokens": 142743412.0,
      "step": 62255
    },
    {
      "entropy": 5.508463716506958,
      "epoch": 5.980787704130644,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020936245932383054,
      "loss": 4.6003,
      "mean_token_accuracy": 0.2685361191630363,
      "num_tokens": 142755127.0,
      "step": 62260
    },
    {
      "entropy": 5.586277151107788,
      "epoch": 5.981268011527377,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020932966776133016,
      "loss": 4.7025,
      "mean_token_accuracy": 0.2556081533432007,
      "num_tokens": 142766315.0,
      "step": 62265
    },
    {
      "entropy": 5.470039653778076,
      "epoch": 5.981748318924112,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020929687772349998,
      "loss": 4.5368,
      "mean_token_accuracy": 0.2765215337276459,
      "num_tokens": 142776978.0,
      "step": 62270
    },
    {
      "entropy": 5.60791163444519,
      "epoch": 5.982228626320845,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020926408921110117,
      "loss": 4.7072,
      "mean_token_accuracy": 0.26145135015249255,
      "num_tokens": 142788825.0,
      "step": 62275
    },
    {
      "entropy": 5.474887037277222,
      "epoch": 5.982708933717579,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000209231302224895,
      "loss": 4.6093,
      "mean_token_accuracy": 0.2651356875896454,
      "num_tokens": 142800880.0,
      "step": 62280
    },
    {
      "entropy": 5.582542610168457,
      "epoch": 5.983189241114313,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002091985167656427,
      "loss": 4.7636,
      "mean_token_accuracy": 0.2526999399065971,
      "num_tokens": 142814652.0,
      "step": 62285
    },
    {
      "entropy": 5.566245985031128,
      "epoch": 5.983669548511047,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020916573283410545,
      "loss": 4.7026,
      "mean_token_accuracy": 0.25814050287008283,
      "num_tokens": 142827692.0,
      "step": 62290
    },
    {
      "entropy": 5.629412317276001,
      "epoch": 5.984149855907781,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020913295043104442,
      "loss": 4.7203,
      "mean_token_accuracy": 0.2573553681373596,
      "num_tokens": 142839603.0,
      "step": 62295
    },
    {
      "entropy": 5.598828125,
      "epoch": 5.984630163304515,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002091001695572207,
      "loss": 4.6781,
      "mean_token_accuracy": 0.2575645476579666,
      "num_tokens": 142851264.0,
      "step": 62300
    },
    {
      "entropy": 5.605097150802612,
      "epoch": 5.985110470701249,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020906739021339538,
      "loss": 4.7282,
      "mean_token_accuracy": 0.25425607413053514,
      "num_tokens": 142862900.0,
      "step": 62305
    },
    {
      "entropy": 5.557414484024048,
      "epoch": 5.985590778097983,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020903461240032945,
      "loss": 4.6283,
      "mean_token_accuracy": 0.26848878115415575,
      "num_tokens": 142873599.0,
      "step": 62310
    },
    {
      "entropy": 5.4906717300415036,
      "epoch": 5.986071085494717,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020900183611878394,
      "loss": 4.5606,
      "mean_token_accuracy": 0.26977211982011795,
      "num_tokens": 142884745.0,
      "step": 62315
    },
    {
      "entropy": 5.421318912506104,
      "epoch": 5.98655139289145,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020896906136951983,
      "loss": 4.5775,
      "mean_token_accuracy": 0.2785556077957153,
      "num_tokens": 142896909.0,
      "step": 62320
    },
    {
      "entropy": 5.517410469055176,
      "epoch": 5.987031700288185,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020893628815329802,
      "loss": 4.7161,
      "mean_token_accuracy": 0.26523578315973284,
      "num_tokens": 142908175.0,
      "step": 62325
    },
    {
      "entropy": 5.539705085754394,
      "epoch": 5.987512007684918,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002089035164708794,
      "loss": 4.6398,
      "mean_token_accuracy": 0.26091254949569703,
      "num_tokens": 142919616.0,
      "step": 62330
    },
    {
      "entropy": 5.543536710739136,
      "epoch": 5.9879923150816525,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000208870746323025,
      "loss": 4.6179,
      "mean_token_accuracy": 0.26398129016160965,
      "num_tokens": 142930515.0,
      "step": 62335
    },
    {
      "entropy": 5.612761306762695,
      "epoch": 5.988472622478386,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002088379777104954,
      "loss": 4.7556,
      "mean_token_accuracy": 0.2551178768277168,
      "num_tokens": 142941878.0,
      "step": 62340
    },
    {
      "entropy": 5.512419843673706,
      "epoch": 5.98895292987512,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002088052106340515,
      "loss": 4.6404,
      "mean_token_accuracy": 0.2652003765106201,
      "num_tokens": 142952305.0,
      "step": 62345
    },
    {
      "entropy": 5.506988191604615,
      "epoch": 5.989433237271854,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020877244509445402,
      "loss": 4.6297,
      "mean_token_accuracy": 0.26265636682510374,
      "num_tokens": 142965420.0,
      "step": 62350
    },
    {
      "entropy": 5.540214157104492,
      "epoch": 5.989913544668588,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020873968109246378,
      "loss": 4.6323,
      "mean_token_accuracy": 0.2643590927124023,
      "num_tokens": 142976622.0,
      "step": 62355
    },
    {
      "entropy": 5.5641203880310055,
      "epoch": 5.9903938520653215,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020870691862884128,
      "loss": 4.7083,
      "mean_token_accuracy": 0.26351575553417206,
      "num_tokens": 142987401.0,
      "step": 62360
    },
    {
      "entropy": 5.512170934677124,
      "epoch": 5.990874159462056,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002086741577043475,
      "loss": 4.547,
      "mean_token_accuracy": 0.28230844289064405,
      "num_tokens": 142998366.0,
      "step": 62365
    },
    {
      "entropy": 5.562621259689331,
      "epoch": 5.991354466858789,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020864139831974266,
      "loss": 4.7386,
      "mean_token_accuracy": 0.2554374158382416,
      "num_tokens": 143010283.0,
      "step": 62370
    },
    {
      "entropy": 5.600290298461914,
      "epoch": 5.991834774255524,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020860864047578752,
      "loss": 4.6769,
      "mean_token_accuracy": 0.2555079489946365,
      "num_tokens": 143021265.0,
      "step": 62375
    },
    {
      "entropy": 5.494988918304443,
      "epoch": 5.992315081652258,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020857588417324264,
      "loss": 4.6145,
      "mean_token_accuracy": 0.26646924912929537,
      "num_tokens": 143033004.0,
      "step": 62380
    },
    {
      "entropy": 5.51693205833435,
      "epoch": 5.992795389048991,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020854312941286841,
      "loss": 4.6144,
      "mean_token_accuracy": 0.2638750225305557,
      "num_tokens": 143044221.0,
      "step": 62385
    },
    {
      "entropy": 5.559898805618286,
      "epoch": 5.993275696445725,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020851037619542545,
      "loss": 4.7427,
      "mean_token_accuracy": 0.2594941794872284,
      "num_tokens": 143055220.0,
      "step": 62390
    },
    {
      "entropy": 5.5536095142364506,
      "epoch": 5.993756003842459,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002084776245216742,
      "loss": 4.6504,
      "mean_token_accuracy": 0.2595017790794373,
      "num_tokens": 143066327.0,
      "step": 62395
    },
    {
      "entropy": 5.581593418121338,
      "epoch": 5.994236311239193,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020844487439237487,
      "loss": 4.6489,
      "mean_token_accuracy": 0.2663262650370598,
      "num_tokens": 143076652.0,
      "step": 62400
    },
    {
      "entropy": 5.551470041275024,
      "epoch": 5.994716618635927,
      "grad_norm": 1.25,
      "learning_rate": 0.00020841212580828794,
      "loss": 4.7055,
      "mean_token_accuracy": 0.25657775402069094,
      "num_tokens": 143088665.0,
      "step": 62405
    },
    {
      "entropy": 5.562639379501343,
      "epoch": 5.995196926032661,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020837937877017377,
      "loss": 4.641,
      "mean_token_accuracy": 0.2647566869854927,
      "num_tokens": 143099215.0,
      "step": 62410
    },
    {
      "entropy": 5.548040914535522,
      "epoch": 5.995677233429395,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020834663327879248,
      "loss": 4.6724,
      "mean_token_accuracy": 0.2681772902607918,
      "num_tokens": 143110019.0,
      "step": 62415
    },
    {
      "entropy": 5.675546646118164,
      "epoch": 5.996157540826129,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020831388933490453,
      "loss": 4.8171,
      "mean_token_accuracy": 0.2519792065024376,
      "num_tokens": 143121397.0,
      "step": 62420
    },
    {
      "entropy": 5.523349809646606,
      "epoch": 5.996637848222862,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002082811469392702,
      "loss": 4.6074,
      "mean_token_accuracy": 0.2634592831134796,
      "num_tokens": 143132291.0,
      "step": 62425
    },
    {
      "entropy": 5.544467544555664,
      "epoch": 5.997118155619597,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020824840609264932,
      "loss": 4.67,
      "mean_token_accuracy": 0.2668112337589264,
      "num_tokens": 143143189.0,
      "step": 62430
    },
    {
      "entropy": 5.574091815948487,
      "epoch": 5.99759846301633,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002082156667958023,
      "loss": 4.7043,
      "mean_token_accuracy": 0.26179418861866,
      "num_tokens": 143154045.0,
      "step": 62435
    },
    {
      "entropy": 5.500385046005249,
      "epoch": 5.9980787704130645,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020818292904948915,
      "loss": 4.5997,
      "mean_token_accuracy": 0.26966788321733476,
      "num_tokens": 143165523.0,
      "step": 62440
    },
    {
      "entropy": 5.506670570373535,
      "epoch": 5.998559077809798,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020815019285447001,
      "loss": 4.6017,
      "mean_token_accuracy": 0.268097847700119,
      "num_tokens": 143177506.0,
      "step": 62445
    },
    {
      "entropy": 5.6350400924682615,
      "epoch": 5.999039385206532,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002081174582115049,
      "loss": 4.7385,
      "mean_token_accuracy": 0.25693571865558623,
      "num_tokens": 143188505.0,
      "step": 62450
    },
    {
      "entropy": 5.541305351257324,
      "epoch": 5.999519692603267,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020808472512135384,
      "loss": 4.6471,
      "mean_token_accuracy": 0.2636709213256836,
      "num_tokens": 143200836.0,
      "step": 62455
    },
    {
      "entropy": 5.548274421691895,
      "epoch": 6.0,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00020805199358477677,
      "loss": 4.6932,
      "mean_token_accuracy": 0.25246810615062715,
      "num_tokens": 143211216.0,
      "step": 62460
    },
    {
      "entropy": 5.584371328353882,
      "epoch": 6.000480307396734,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020801926360253353,
      "loss": 4.7042,
      "mean_token_accuracy": 0.2545191213488579,
      "num_tokens": 143223093.0,
      "step": 62465
    },
    {
      "entropy": 5.5378059387207035,
      "epoch": 6.000960614793468,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020798653517538414,
      "loss": 4.6498,
      "mean_token_accuracy": 0.27342658042907714,
      "num_tokens": 143234572.0,
      "step": 62470
    },
    {
      "entropy": 5.455716991424561,
      "epoch": 6.001440922190202,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020795380830408838,
      "loss": 4.5733,
      "mean_token_accuracy": 0.2700075447559357,
      "num_tokens": 143245423.0,
      "step": 62475
    },
    {
      "entropy": 5.525109004974365,
      "epoch": 6.0019212295869355,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020792108298940616,
      "loss": 4.6198,
      "mean_token_accuracy": 0.2690477713942528,
      "num_tokens": 143257822.0,
      "step": 62480
    },
    {
      "entropy": 5.547153425216675,
      "epoch": 6.00240153698367,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002078883592320972,
      "loss": 4.6695,
      "mean_token_accuracy": 0.25869294852018354,
      "num_tokens": 143269988.0,
      "step": 62485
    },
    {
      "entropy": 5.580815601348877,
      "epoch": 6.002881844380403,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020785563703292132,
      "loss": 4.6832,
      "mean_token_accuracy": 0.2603013142943382,
      "num_tokens": 143281585.0,
      "step": 62490
    },
    {
      "entropy": 5.636180830001831,
      "epoch": 6.003362151777138,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020782291639263807,
      "loss": 4.6755,
      "mean_token_accuracy": 0.26284333914518354,
      "num_tokens": 143294051.0,
      "step": 62495
    },
    {
      "entropy": 5.564570045471191,
      "epoch": 6.003842459173871,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020779019731200722,
      "loss": 4.6487,
      "mean_token_accuracy": 0.26291466057300567,
      "num_tokens": 143305314.0,
      "step": 62500
    },
    {
      "entropy": 5.462167358398437,
      "epoch": 6.004322766570605,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020775747979178845,
      "loss": 4.5842,
      "mean_token_accuracy": 0.2672311544418335,
      "num_tokens": 143316143.0,
      "step": 62505
    },
    {
      "entropy": 5.524545526504516,
      "epoch": 6.004803073967339,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002077247638327413,
      "loss": 4.6275,
      "mean_token_accuracy": 0.2642575800418854,
      "num_tokens": 143326430.0,
      "step": 62510
    },
    {
      "entropy": 5.5272197246551515,
      "epoch": 6.005283381364073,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020769204943562548,
      "loss": 4.6646,
      "mean_token_accuracy": 0.2564738854765892,
      "num_tokens": 143338729.0,
      "step": 62515
    },
    {
      "entropy": 5.483224153518677,
      "epoch": 6.005763688760807,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002076593366012003,
      "loss": 4.5826,
      "mean_token_accuracy": 0.27772766947746275,
      "num_tokens": 143350328.0,
      "step": 62520
    },
    {
      "entropy": 5.493165493011475,
      "epoch": 6.006243996157541,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020762662533022535,
      "loss": 4.5332,
      "mean_token_accuracy": 0.27892089933156966,
      "num_tokens": 143361295.0,
      "step": 62525
    },
    {
      "entropy": 5.500663995742798,
      "epoch": 6.006724303554274,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002075939156234602,
      "loss": 4.6043,
      "mean_token_accuracy": 0.266435506939888,
      "num_tokens": 143373569.0,
      "step": 62530
    },
    {
      "entropy": 5.570245742797852,
      "epoch": 6.007204610951009,
      "grad_norm": 1.375,
      "learning_rate": 0.00020756120748166413,
      "loss": 4.6085,
      "mean_token_accuracy": 0.2634695366024971,
      "num_tokens": 143383966.0,
      "step": 62535
    },
    {
      "entropy": 5.605322170257568,
      "epoch": 6.007684918347743,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020752850090559655,
      "loss": 4.6757,
      "mean_token_accuracy": 0.26474127173423767,
      "num_tokens": 143394870.0,
      "step": 62540
    },
    {
      "entropy": 5.518954753875732,
      "epoch": 6.008165225744476,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002074957958960168,
      "loss": 4.6015,
      "mean_token_accuracy": 0.27151014357805253,
      "num_tokens": 143407552.0,
      "step": 62545
    },
    {
      "entropy": 5.5254350185394285,
      "epoch": 6.008645533141211,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020746309245368435,
      "loss": 4.6111,
      "mean_token_accuracy": 0.27834982573986056,
      "num_tokens": 143419838.0,
      "step": 62550
    },
    {
      "entropy": 5.620533037185669,
      "epoch": 6.009125840537944,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002074303905793582,
      "loss": 4.7382,
      "mean_token_accuracy": 0.2556561976671219,
      "num_tokens": 143431612.0,
      "step": 62555
    },
    {
      "entropy": 5.541678953170776,
      "epoch": 6.0096061479346785,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002073976902737978,
      "loss": 4.6441,
      "mean_token_accuracy": 0.2606700763106346,
      "num_tokens": 143443783.0,
      "step": 62560
    },
    {
      "entropy": 5.56302433013916,
      "epoch": 6.010086455331412,
      "grad_norm": 1.375,
      "learning_rate": 0.00020736499153776227,
      "loss": 4.6797,
      "mean_token_accuracy": 0.2658719003200531,
      "num_tokens": 143456285.0,
      "step": 62565
    },
    {
      "entropy": 5.490819931030273,
      "epoch": 6.010566762728146,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020733229437201083,
      "loss": 4.6203,
      "mean_token_accuracy": 0.2578127279877663,
      "num_tokens": 143468338.0,
      "step": 62570
    },
    {
      "entropy": 5.558051252365113,
      "epoch": 6.01104707012488,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002072995987773026,
      "loss": 4.6169,
      "mean_token_accuracy": 0.2687546700239182,
      "num_tokens": 143479118.0,
      "step": 62575
    },
    {
      "entropy": 5.584839534759522,
      "epoch": 6.011527377521614,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020726690475439668,
      "loss": 4.6947,
      "mean_token_accuracy": 0.2635503485798836,
      "num_tokens": 143490653.0,
      "step": 62580
    },
    {
      "entropy": 5.440512752532959,
      "epoch": 6.0120076849183475,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020723421230405204,
      "loss": 4.5042,
      "mean_token_accuracy": 0.2826091766357422,
      "num_tokens": 143501685.0,
      "step": 62585
    },
    {
      "entropy": 5.495725440979004,
      "epoch": 6.012487992315082,
      "grad_norm": 1.375,
      "learning_rate": 0.00020720152142702781,
      "loss": 4.6258,
      "mean_token_accuracy": 0.25971042960882185,
      "num_tokens": 143512923.0,
      "step": 62590
    },
    {
      "entropy": 5.528156185150147,
      "epoch": 6.012968299711815,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002071688321240829,
      "loss": 4.5839,
      "mean_token_accuracy": 0.26804490238428114,
      "num_tokens": 143523333.0,
      "step": 62595
    },
    {
      "entropy": 5.522160625457763,
      "epoch": 6.01344860710855,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020713614439597629,
      "loss": 4.6235,
      "mean_token_accuracy": 0.26942972242832186,
      "num_tokens": 143534985.0,
      "step": 62600
    },
    {
      "entropy": 5.5745823860168455,
      "epoch": 6.013928914505283,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002071034582434669,
      "loss": 4.6463,
      "mean_token_accuracy": 0.2598835453391075,
      "num_tokens": 143544859.0,
      "step": 62605
    },
    {
      "entropy": 5.52847204208374,
      "epoch": 6.014409221902017,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020707077366731363,
      "loss": 4.6475,
      "mean_token_accuracy": 0.2639261156320572,
      "num_tokens": 143557586.0,
      "step": 62610
    },
    {
      "entropy": 5.487706899642944,
      "epoch": 6.014889529298751,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002070380906682753,
      "loss": 4.5642,
      "mean_token_accuracy": 0.2646970212459564,
      "num_tokens": 143568468.0,
      "step": 62615
    },
    {
      "entropy": 5.518429946899414,
      "epoch": 6.015369836695485,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020700540924711058,
      "loss": 4.6123,
      "mean_token_accuracy": 0.2619155153632164,
      "num_tokens": 143579849.0,
      "step": 62620
    },
    {
      "entropy": 5.547892713546753,
      "epoch": 6.015850144092219,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020697272940457844,
      "loss": 4.6058,
      "mean_token_accuracy": 0.26504750102758406,
      "num_tokens": 143591517.0,
      "step": 62625
    },
    {
      "entropy": 5.56696047782898,
      "epoch": 6.016330451488953,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002069400511414375,
      "loss": 4.7152,
      "mean_token_accuracy": 0.2625411957502365,
      "num_tokens": 143604492.0,
      "step": 62630
    },
    {
      "entropy": 5.582138776779175,
      "epoch": 6.016810758885687,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020690737445844655,
      "loss": 4.736,
      "mean_token_accuracy": 0.2650868520140648,
      "num_tokens": 143617663.0,
      "step": 62635
    },
    {
      "entropy": 5.5549623489379885,
      "epoch": 6.017291066282421,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020687469935636417,
      "loss": 4.6403,
      "mean_token_accuracy": 0.26967569887638093,
      "num_tokens": 143628728.0,
      "step": 62640
    },
    {
      "entropy": 5.445296049118042,
      "epoch": 6.017771373679155,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002068420258359489,
      "loss": 4.5335,
      "mean_token_accuracy": 0.27355162501335145,
      "num_tokens": 143639608.0,
      "step": 62645
    },
    {
      "entropy": 5.5562906742095945,
      "epoch": 6.018251681075888,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002068093538979594,
      "loss": 4.6773,
      "mean_token_accuracy": 0.2578597739338875,
      "num_tokens": 143652761.0,
      "step": 62650
    },
    {
      "entropy": 5.607173109054566,
      "epoch": 6.018731988472623,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00020677668354315428,
      "loss": 4.6272,
      "mean_token_accuracy": 0.26222542971372603,
      "num_tokens": 143663317.0,
      "step": 62655
    },
    {
      "entropy": 5.5209832191467285,
      "epoch": 6.019212295869356,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020674401477229192,
      "loss": 4.6664,
      "mean_token_accuracy": 0.2583839550614357,
      "num_tokens": 143674649.0,
      "step": 62660
    },
    {
      "entropy": 5.58603572845459,
      "epoch": 6.0196926032660905,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020671134758613085,
      "loss": 4.6464,
      "mean_token_accuracy": 0.261921526491642,
      "num_tokens": 143685521.0,
      "step": 62665
    },
    {
      "entropy": 5.542790222167969,
      "epoch": 6.020172910662824,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020667868198542965,
      "loss": 4.5591,
      "mean_token_accuracy": 0.2778633564710617,
      "num_tokens": 143695988.0,
      "step": 62670
    },
    {
      "entropy": 5.492027664184571,
      "epoch": 6.020653218059558,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002066460179709465,
      "loss": 4.543,
      "mean_token_accuracy": 0.2755208805203438,
      "num_tokens": 143708524.0,
      "step": 62675
    },
    {
      "entropy": 5.442368412017823,
      "epoch": 6.021133525456292,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020661335554343985,
      "loss": 4.5729,
      "mean_token_accuracy": 0.2780182808637619,
      "num_tokens": 143719449.0,
      "step": 62680
    },
    {
      "entropy": 5.56150598526001,
      "epoch": 6.021613832853026,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020658069470366798,
      "loss": 4.5998,
      "mean_token_accuracy": 0.2651869386434555,
      "num_tokens": 143729995.0,
      "step": 62685
    },
    {
      "entropy": 5.485789728164673,
      "epoch": 6.022094140249759,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002065480354523892,
      "loss": 4.5549,
      "mean_token_accuracy": 0.27181306183338166,
      "num_tokens": 143742271.0,
      "step": 62690
    },
    {
      "entropy": 5.552960348129273,
      "epoch": 6.022574447646494,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002065153777903618,
      "loss": 4.6654,
      "mean_token_accuracy": 0.2660981759428978,
      "num_tokens": 143753553.0,
      "step": 62695
    },
    {
      "entropy": 5.531838846206665,
      "epoch": 6.023054755043228,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002064827217183441,
      "loss": 4.6358,
      "mean_token_accuracy": 0.26575808227062225,
      "num_tokens": 143765076.0,
      "step": 62700
    },
    {
      "entropy": 5.585115098953247,
      "epoch": 6.0235350624399615,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020645006723709396,
      "loss": 4.6374,
      "mean_token_accuracy": 0.26883571594953537,
      "num_tokens": 143775838.0,
      "step": 62705
    },
    {
      "entropy": 5.599866724014282,
      "epoch": 6.024015369836696,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020641741434736978,
      "loss": 4.6597,
      "mean_token_accuracy": 0.2656903207302094,
      "num_tokens": 143787227.0,
      "step": 62710
    },
    {
      "entropy": 5.541287279129028,
      "epoch": 6.024495677233429,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020638476304992958,
      "loss": 4.6485,
      "mean_token_accuracy": 0.2589654356241226,
      "num_tokens": 143799071.0,
      "step": 62715
    },
    {
      "entropy": 5.460856819152832,
      "epoch": 6.024975984630164,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002063521133455314,
      "loss": 4.5189,
      "mean_token_accuracy": 0.2761397808790207,
      "num_tokens": 143809447.0,
      "step": 62720
    },
    {
      "entropy": 5.469598340988159,
      "epoch": 6.025456292026897,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020631946523493334,
      "loss": 4.5688,
      "mean_token_accuracy": 0.26668178886175153,
      "num_tokens": 143821262.0,
      "step": 62725
    },
    {
      "entropy": 5.548301124572754,
      "epoch": 6.025936599423631,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020628681871889343,
      "loss": 4.6824,
      "mean_token_accuracy": 0.2664347216486931,
      "num_tokens": 143834022.0,
      "step": 62730
    },
    {
      "entropy": 5.527257871627808,
      "epoch": 6.026416906820365,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020625417379816947,
      "loss": 4.5689,
      "mean_token_accuracy": 0.2708633095026016,
      "num_tokens": 143844658.0,
      "step": 62735
    },
    {
      "entropy": 5.527938175201416,
      "epoch": 6.026897214217099,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002062215304735195,
      "loss": 4.647,
      "mean_token_accuracy": 0.25669515430927276,
      "num_tokens": 143857130.0,
      "step": 62740
    },
    {
      "entropy": 5.602615118026733,
      "epoch": 6.027377521613833,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020618888874570135,
      "loss": 4.6428,
      "mean_token_accuracy": 0.26836990267038346,
      "num_tokens": 143867762.0,
      "step": 62745
    },
    {
      "entropy": 5.552262830734253,
      "epoch": 6.027857829010567,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020615624861547284,
      "loss": 4.6567,
      "mean_token_accuracy": 0.2630383729934692,
      "num_tokens": 143880382.0,
      "step": 62750
    },
    {
      "entropy": 5.549840688705444,
      "epoch": 6.0283381364073,
      "grad_norm": 1.375,
      "learning_rate": 0.0002061236100835918,
      "loss": 4.6808,
      "mean_token_accuracy": 0.2623315453529358,
      "num_tokens": 143891599.0,
      "step": 62755
    },
    {
      "entropy": 5.466345024108887,
      "epoch": 6.028818443804035,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020609097315081614,
      "loss": 4.5378,
      "mean_token_accuracy": 0.2700633451342583,
      "num_tokens": 143902805.0,
      "step": 62760
    },
    {
      "entropy": 5.602393722534179,
      "epoch": 6.029298751200768,
      "grad_norm": 1.375,
      "learning_rate": 0.0002060583378179034,
      "loss": 4.7056,
      "mean_token_accuracy": 0.25765957981348037,
      "num_tokens": 143913680.0,
      "step": 62765
    },
    {
      "entropy": 5.560571336746216,
      "epoch": 6.029779058597502,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020602570408561127,
      "loss": 4.6399,
      "mean_token_accuracy": 0.26703511625528337,
      "num_tokens": 143925095.0,
      "step": 62770
    },
    {
      "entropy": 5.560616397857666,
      "epoch": 6.030259365994237,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002059930719546975,
      "loss": 4.6375,
      "mean_token_accuracy": 0.2671928271651268,
      "num_tokens": 143935467.0,
      "step": 62775
    },
    {
      "entropy": 5.500977516174316,
      "epoch": 6.03073967339097,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002059604414259198,
      "loss": 4.5944,
      "mean_token_accuracy": 0.2707537367939949,
      "num_tokens": 143945188.0,
      "step": 62780
    },
    {
      "entropy": 5.5651082515716555,
      "epoch": 6.0312199807877045,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002059278125000355,
      "loss": 4.667,
      "mean_token_accuracy": 0.26197167336940763,
      "num_tokens": 143956997.0,
      "step": 62785
    },
    {
      "entropy": 5.603625059127808,
      "epoch": 6.031700288184438,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020589518517780242,
      "loss": 4.7257,
      "mean_token_accuracy": 0.254161661863327,
      "num_tokens": 143968691.0,
      "step": 62790
    },
    {
      "entropy": 5.6269432544708256,
      "epoch": 6.032180595581172,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002058625594599779,
      "loss": 4.6813,
      "mean_token_accuracy": 0.2574621796607971,
      "num_tokens": 143980128.0,
      "step": 62795
    },
    {
      "entropy": 5.471712923049926,
      "epoch": 6.032660902977906,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002058299353473194,
      "loss": 4.5255,
      "mean_token_accuracy": 0.27777641713619233,
      "num_tokens": 143991787.0,
      "step": 62800
    },
    {
      "entropy": 5.495617008209228,
      "epoch": 6.03314121037464,
      "grad_norm": 1.375,
      "learning_rate": 0.00020579731284058446,
      "loss": 4.6704,
      "mean_token_accuracy": 0.26168995201587675,
      "num_tokens": 144003753.0,
      "step": 62805
    },
    {
      "entropy": 5.460531949996948,
      "epoch": 6.0336215177713735,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002057646919405304,
      "loss": 4.5559,
      "mean_token_accuracy": 0.27234213799238205,
      "num_tokens": 144016092.0,
      "step": 62810
    },
    {
      "entropy": 5.5575837135314945,
      "epoch": 6.034101825168108,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002057320726479146,
      "loss": 4.6958,
      "mean_token_accuracy": 0.25228428691625593,
      "num_tokens": 144027677.0,
      "step": 62815
    },
    {
      "entropy": 5.633167839050293,
      "epoch": 6.034582132564841,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002056994549634944,
      "loss": 4.7094,
      "mean_token_accuracy": 0.2616530805826187,
      "num_tokens": 144038905.0,
      "step": 62820
    },
    {
      "entropy": 5.5749671459198,
      "epoch": 6.0350624399615755,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020566683888802707,
      "loss": 4.6724,
      "mean_token_accuracy": 0.26906911432743075,
      "num_tokens": 144050441.0,
      "step": 62825
    },
    {
      "entropy": 5.545311355590821,
      "epoch": 6.035542747358309,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020563422442226986,
      "loss": 4.6102,
      "mean_token_accuracy": 0.27386810481548307,
      "num_tokens": 144061448.0,
      "step": 62830
    },
    {
      "entropy": 5.507832431793213,
      "epoch": 6.036023054755043,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020560161156698005,
      "loss": 4.6475,
      "mean_token_accuracy": 0.2626535207033157,
      "num_tokens": 144071699.0,
      "step": 62835
    },
    {
      "entropy": 5.449597024917603,
      "epoch": 6.036503362151777,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002055690003229147,
      "loss": 4.5259,
      "mean_token_accuracy": 0.2750674068927765,
      "num_tokens": 144082793.0,
      "step": 62840
    },
    {
      "entropy": 5.567458057403565,
      "epoch": 6.036983669548511,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020553639069083096,
      "loss": 4.645,
      "mean_token_accuracy": 0.26055206805467607,
      "num_tokens": 144092990.0,
      "step": 62845
    },
    {
      "entropy": 5.556014680862427,
      "epoch": 6.037463976945245,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000205503782671486,
      "loss": 4.6449,
      "mean_token_accuracy": 0.2677556395530701,
      "num_tokens": 144104602.0,
      "step": 62850
    },
    {
      "entropy": 5.592458772659302,
      "epoch": 6.037944284341979,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002054711762656369,
      "loss": 4.6601,
      "mean_token_accuracy": 0.26355962455272675,
      "num_tokens": 144115155.0,
      "step": 62855
    },
    {
      "entropy": 5.4945018768310545,
      "epoch": 6.038424591738713,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020543857147404054,
      "loss": 4.5993,
      "mean_token_accuracy": 0.2683968558907509,
      "num_tokens": 144126720.0,
      "step": 62860
    },
    {
      "entropy": 5.561190557479859,
      "epoch": 6.038904899135447,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000205405968297454,
      "loss": 4.67,
      "mean_token_accuracy": 0.2670682221651077,
      "num_tokens": 144139217.0,
      "step": 62865
    },
    {
      "entropy": 5.547578573226929,
      "epoch": 6.039385206532181,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002053733667366342,
      "loss": 4.6203,
      "mean_token_accuracy": 0.27040044218301773,
      "num_tokens": 144150673.0,
      "step": 62870
    },
    {
      "entropy": 5.522335338592529,
      "epoch": 6.039865513928914,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020534076679233816,
      "loss": 4.6263,
      "mean_token_accuracy": 0.2709339052438736,
      "num_tokens": 144161557.0,
      "step": 62875
    },
    {
      "entropy": 5.610909175872803,
      "epoch": 6.040345821325649,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020530816846532269,
      "loss": 4.7372,
      "mean_token_accuracy": 0.25405035465955733,
      "num_tokens": 144173374.0,
      "step": 62880
    },
    {
      "entropy": 5.5849498271942135,
      "epoch": 6.040826128722382,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020527557175634454,
      "loss": 4.7083,
      "mean_token_accuracy": 0.2581766337156296,
      "num_tokens": 144186607.0,
      "step": 62885
    },
    {
      "entropy": 5.515845441818238,
      "epoch": 6.041306436119116,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020524297666616055,
      "loss": 4.6038,
      "mean_token_accuracy": 0.2671589121222496,
      "num_tokens": 144198762.0,
      "step": 62890
    },
    {
      "entropy": 5.495509099960327,
      "epoch": 6.04178674351585,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020521038319552755,
      "loss": 4.5628,
      "mean_token_accuracy": 0.2749801829457283,
      "num_tokens": 144209586.0,
      "step": 62895
    },
    {
      "entropy": 5.45864782333374,
      "epoch": 6.042267050912584,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002051777913452022,
      "loss": 4.564,
      "mean_token_accuracy": 0.27164370119571685,
      "num_tokens": 144221832.0,
      "step": 62900
    },
    {
      "entropy": 5.599259662628174,
      "epoch": 6.042747358309318,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002051452011159412,
      "loss": 4.7009,
      "mean_token_accuracy": 0.25232345014810564,
      "num_tokens": 144233756.0,
      "step": 62905
    },
    {
      "entropy": 5.497105884552002,
      "epoch": 6.043227665706052,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020511261250850133,
      "loss": 4.5609,
      "mean_token_accuracy": 0.26485236287117003,
      "num_tokens": 144245449.0,
      "step": 62910
    },
    {
      "entropy": 5.519883251190185,
      "epoch": 6.043707973102785,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002050800255236389,
      "loss": 4.6043,
      "mean_token_accuracy": 0.2639192879199982,
      "num_tokens": 144256298.0,
      "step": 62915
    },
    {
      "entropy": 5.585788631439209,
      "epoch": 6.04418828049952,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020504744016211076,
      "loss": 4.7041,
      "mean_token_accuracy": 0.2604124665260315,
      "num_tokens": 144267318.0,
      "step": 62920
    },
    {
      "entropy": 5.4740362644195555,
      "epoch": 6.044668587896253,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020501485642467326,
      "loss": 4.5268,
      "mean_token_accuracy": 0.27755056619644164,
      "num_tokens": 144278433.0,
      "step": 62925
    },
    {
      "entropy": 5.560106611251831,
      "epoch": 6.0451488952929875,
      "grad_norm": 1.34375,
      "learning_rate": 0.000204982274312083,
      "loss": 4.6465,
      "mean_token_accuracy": 0.26355959177017213,
      "num_tokens": 144289233.0,
      "step": 62930
    },
    {
      "entropy": 5.497397756576538,
      "epoch": 6.045629202689722,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020494969382509642,
      "loss": 4.5714,
      "mean_token_accuracy": 0.2694384127855301,
      "num_tokens": 144300717.0,
      "step": 62935
    },
    {
      "entropy": 5.591887617111206,
      "epoch": 6.046109510086455,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020491711496447006,
      "loss": 4.6535,
      "mean_token_accuracy": 0.2598041892051697,
      "num_tokens": 144312642.0,
      "step": 62940
    },
    {
      "entropy": 5.614377880096436,
      "epoch": 6.04658981748319,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020488453773096011,
      "loss": 4.7437,
      "mean_token_accuracy": 0.2612559363245964,
      "num_tokens": 144323170.0,
      "step": 62945
    },
    {
      "entropy": 5.650554513931274,
      "epoch": 6.047070124879923,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020485196212532291,
      "loss": 4.6898,
      "mean_token_accuracy": 0.26126588582992555,
      "num_tokens": 144334920.0,
      "step": 62950
    },
    {
      "entropy": 5.537757205963135,
      "epoch": 6.047550432276657,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020481938814831497,
      "loss": 4.6167,
      "mean_token_accuracy": 0.2657688707113266,
      "num_tokens": 144347341.0,
      "step": 62955
    },
    {
      "entropy": 5.578398132324219,
      "epoch": 6.048030739673391,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020478681580069237,
      "loss": 4.706,
      "mean_token_accuracy": 0.2549761801958084,
      "num_tokens": 144358821.0,
      "step": 62960
    },
    {
      "entropy": 5.583028888702392,
      "epoch": 6.048511047070125,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020475424508321145,
      "loss": 4.726,
      "mean_token_accuracy": 0.254068323969841,
      "num_tokens": 144370169.0,
      "step": 62965
    },
    {
      "entropy": 5.555161571502685,
      "epoch": 6.0489913544668585,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020472167599662843,
      "loss": 4.6315,
      "mean_token_accuracy": 0.2591047421097755,
      "num_tokens": 144380950.0,
      "step": 62970
    },
    {
      "entropy": 5.525082111358643,
      "epoch": 6.049471661863593,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020468910854169937,
      "loss": 4.6372,
      "mean_token_accuracy": 0.2653557166457176,
      "num_tokens": 144392170.0,
      "step": 62975
    },
    {
      "entropy": 5.530495023727417,
      "epoch": 6.049951969260326,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020465654271918033,
      "loss": 4.6392,
      "mean_token_accuracy": 0.25954277217388155,
      "num_tokens": 144403167.0,
      "step": 62980
    },
    {
      "entropy": 5.566528749465943,
      "epoch": 6.050432276657061,
      "grad_norm": 1.375,
      "learning_rate": 0.00020462397852982756,
      "loss": 4.6756,
      "mean_token_accuracy": 0.2597759798169136,
      "num_tokens": 144413791.0,
      "step": 62985
    },
    {
      "entropy": 5.548465347290039,
      "epoch": 6.050912584053794,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020459141597439708,
      "loss": 4.6186,
      "mean_token_accuracy": 0.2644553452730179,
      "num_tokens": 144424980.0,
      "step": 62990
    },
    {
      "entropy": 5.606929731369019,
      "epoch": 6.051392891450528,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020455885505364475,
      "loss": 4.7006,
      "mean_token_accuracy": 0.2610105127096176,
      "num_tokens": 144437003.0,
      "step": 62995
    },
    {
      "entropy": 5.581430959701538,
      "epoch": 6.051873198847262,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020452629576832675,
      "loss": 4.6987,
      "mean_token_accuracy": 0.25976104736328126,
      "num_tokens": 144449426.0,
      "step": 63000
    },
    {
      "epoch": 6.051873198847262,
      "eval_entropy": 5.264341967911443,
      "eval_loss": 4.758184909820557,
      "eval_mean_token_accuracy": 0.26538346073649705,
      "eval_num_tokens": 144449426.0,
      "eval_runtime": 20.649,
      "eval_samples_per_second": 1589.179,
      "eval_steps_per_second": 198.653,
      "step": 63000
    },
    {
      "entropy": 5.587174749374389,
      "epoch": 6.052353506243996,
      "grad_norm": 1.375,
      "learning_rate": 0.00020449373811919885,
      "loss": 4.6568,
      "mean_token_accuracy": 0.26138958930969236,
      "num_tokens": 144461211.0,
      "step": 63005
    },
    {
      "entropy": 5.591328191757202,
      "epoch": 6.0528338136407305,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020446118210701699,
      "loss": 4.7344,
      "mean_token_accuracy": 0.263718181848526,
      "num_tokens": 144473682.0,
      "step": 63010
    },
    {
      "entropy": 5.606502628326416,
      "epoch": 6.053314121037464,
      "grad_norm": 1.34375,
      "learning_rate": 0.000204428627732537,
      "loss": 4.6656,
      "mean_token_accuracy": 0.2599875718355179,
      "num_tokens": 144485028.0,
      "step": 63015
    },
    {
      "entropy": 5.5476758003234865,
      "epoch": 6.053794428434198,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020439607499651475,
      "loss": 4.6247,
      "mean_token_accuracy": 0.2619013711810112,
      "num_tokens": 144496955.0,
      "step": 63020
    },
    {
      "entropy": 5.513276147842407,
      "epoch": 6.054274735830932,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020436352389970597,
      "loss": 4.6476,
      "mean_token_accuracy": 0.2616622284054756,
      "num_tokens": 144508303.0,
      "step": 63025
    },
    {
      "entropy": 5.532259368896485,
      "epoch": 6.054755043227666,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020433097444286653,
      "loss": 4.653,
      "mean_token_accuracy": 0.26412136256694796,
      "num_tokens": 144519819.0,
      "step": 63030
    },
    {
      "entropy": 5.595178890228271,
      "epoch": 6.055235350624399,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002042984266267519,
      "loss": 4.6694,
      "mean_token_accuracy": 0.26194499582052233,
      "num_tokens": 144531018.0,
      "step": 63035
    },
    {
      "entropy": 5.566550588607788,
      "epoch": 6.055715658021134,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020426588045211787,
      "loss": 4.6477,
      "mean_token_accuracy": 0.2651625484228134,
      "num_tokens": 144542787.0,
      "step": 63040
    },
    {
      "entropy": 5.5998673915863035,
      "epoch": 6.056195965417867,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020423333591972005,
      "loss": 4.6345,
      "mean_token_accuracy": 0.26118944585323334,
      "num_tokens": 144553215.0,
      "step": 63045
    },
    {
      "entropy": 5.641121530532837,
      "epoch": 6.0566762728146015,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020420079303031403,
      "loss": 4.6744,
      "mean_token_accuracy": 0.2634767577052116,
      "num_tokens": 144563495.0,
      "step": 63050
    },
    {
      "entropy": 5.575661087036133,
      "epoch": 6.057156580211335,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020416825178465537,
      "loss": 4.6521,
      "mean_token_accuracy": 0.262600015103817,
      "num_tokens": 144574820.0,
      "step": 63055
    },
    {
      "entropy": 5.566269779205323,
      "epoch": 6.057636887608069,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020413571218349968,
      "loss": 4.6873,
      "mean_token_accuracy": 0.2569805756211281,
      "num_tokens": 144586138.0,
      "step": 63060
    },
    {
      "entropy": 5.643759679794312,
      "epoch": 6.058117195004803,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020410317422760223,
      "loss": 4.7319,
      "mean_token_accuracy": 0.2537429094314575,
      "num_tokens": 144598786.0,
      "step": 63065
    },
    {
      "entropy": 5.575222301483154,
      "epoch": 6.058597502401537,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020407063791771858,
      "loss": 4.6458,
      "mean_token_accuracy": 0.255424402654171,
      "num_tokens": 144608691.0,
      "step": 63070
    },
    {
      "entropy": 5.492836856842041,
      "epoch": 6.0590778097982705,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020403810325460409,
      "loss": 4.6226,
      "mean_token_accuracy": 0.26308201253414154,
      "num_tokens": 144619859.0,
      "step": 63075
    },
    {
      "entropy": 5.483706760406494,
      "epoch": 6.059558117195005,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020400557023901413,
      "loss": 4.5391,
      "mean_token_accuracy": 0.2791858375072479,
      "num_tokens": 144632972.0,
      "step": 63080
    },
    {
      "entropy": 5.6356299877166744,
      "epoch": 6.060038424591739,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020397303887170399,
      "loss": 4.7095,
      "mean_token_accuracy": 0.25515650659799577,
      "num_tokens": 144644890.0,
      "step": 63085
    },
    {
      "entropy": 5.537867498397827,
      "epoch": 6.060518731988473,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020394050915342906,
      "loss": 4.6489,
      "mean_token_accuracy": 0.26606784760951996,
      "num_tokens": 144657637.0,
      "step": 63090
    },
    {
      "entropy": 5.4720926761627195,
      "epoch": 6.060999039385207,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002039079810849444,
      "loss": 4.5785,
      "mean_token_accuracy": 0.2714178442955017,
      "num_tokens": 144669674.0,
      "step": 63095
    },
    {
      "entropy": 5.525311994552612,
      "epoch": 6.06147934678194,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020387545466700542,
      "loss": 4.6027,
      "mean_token_accuracy": 0.2737512469291687,
      "num_tokens": 144681356.0,
      "step": 63100
    },
    {
      "entropy": 5.58305344581604,
      "epoch": 6.061959654178675,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020384292990036713,
      "loss": 4.6546,
      "mean_token_accuracy": 0.26136543452739713,
      "num_tokens": 144692792.0,
      "step": 63105
    },
    {
      "entropy": 5.576237726211548,
      "epoch": 6.062439961575408,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002038104067857847,
      "loss": 4.6633,
      "mean_token_accuracy": 0.26030370891094207,
      "num_tokens": 144702901.0,
      "step": 63110
    },
    {
      "entropy": 5.482085084915161,
      "epoch": 6.062920268972142,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020377788532401325,
      "loss": 4.5353,
      "mean_token_accuracy": 0.2699751928448677,
      "num_tokens": 144714535.0,
      "step": 63115
    },
    {
      "entropy": 5.536666822433472,
      "epoch": 6.063400576368876,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020374536551580786,
      "loss": 4.5849,
      "mean_token_accuracy": 0.27424585968256,
      "num_tokens": 144726149.0,
      "step": 63120
    },
    {
      "entropy": 5.492267704010009,
      "epoch": 6.06388088376561,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020371284736192346,
      "loss": 4.5837,
      "mean_token_accuracy": 0.27072844803333285,
      "num_tokens": 144737500.0,
      "step": 63125
    },
    {
      "entropy": 5.6860222816467285,
      "epoch": 6.064361191162344,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020368033086311504,
      "loss": 4.7698,
      "mean_token_accuracy": 0.25303261578083036,
      "num_tokens": 144748975.0,
      "step": 63130
    },
    {
      "entropy": 5.568271017074585,
      "epoch": 6.064841498559078,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002036478160201376,
      "loss": 4.6411,
      "mean_token_accuracy": 0.2641108468174934,
      "num_tokens": 144760735.0,
      "step": 63135
    },
    {
      "entropy": 5.574754238128662,
      "epoch": 6.065321805955811,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020361530283374602,
      "loss": 4.7038,
      "mean_token_accuracy": 0.2591883957386017,
      "num_tokens": 144771951.0,
      "step": 63140
    },
    {
      "entropy": 5.547768449783325,
      "epoch": 6.065802113352546,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020358279130469509,
      "loss": 4.692,
      "mean_token_accuracy": 0.26449227780103685,
      "num_tokens": 144783863.0,
      "step": 63145
    },
    {
      "entropy": 5.582587862014771,
      "epoch": 6.066282420749279,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020355028143373972,
      "loss": 4.6588,
      "mean_token_accuracy": 0.2561593890190125,
      "num_tokens": 144795537.0,
      "step": 63150
    },
    {
      "entropy": 5.583914566040039,
      "epoch": 6.0667627281460135,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020351777322163463,
      "loss": 4.7242,
      "mean_token_accuracy": 0.2622089236974716,
      "num_tokens": 144806280.0,
      "step": 63155
    },
    {
      "entropy": 5.535638475418091,
      "epoch": 6.067243035542747,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020348526666913458,
      "loss": 4.6654,
      "mean_token_accuracy": 0.2633342817425728,
      "num_tokens": 144817394.0,
      "step": 63160
    },
    {
      "entropy": 5.4929876804351805,
      "epoch": 6.067723342939481,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020345276177699428,
      "loss": 4.6256,
      "mean_token_accuracy": 0.26706855446100236,
      "num_tokens": 144828701.0,
      "step": 63165
    },
    {
      "entropy": 5.588437843322754,
      "epoch": 6.068203650336216,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002034202585459684,
      "loss": 4.7216,
      "mean_token_accuracy": 0.2641924530267715,
      "num_tokens": 144840274.0,
      "step": 63170
    },
    {
      "entropy": 5.5708075046539305,
      "epoch": 6.068683957732949,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020338775697681155,
      "loss": 4.6401,
      "mean_token_accuracy": 0.2596262514591217,
      "num_tokens": 144852954.0,
      "step": 63175
    },
    {
      "entropy": 5.583873128890991,
      "epoch": 6.069164265129683,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020335525707027846,
      "loss": 4.6301,
      "mean_token_accuracy": 0.26790550351142883,
      "num_tokens": 144864618.0,
      "step": 63180
    },
    {
      "entropy": 5.611468648910522,
      "epoch": 6.069644572526417,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020332275882712348,
      "loss": 4.6891,
      "mean_token_accuracy": 0.26235680282115936,
      "num_tokens": 144877867.0,
      "step": 63185
    },
    {
      "entropy": 5.458426904678345,
      "epoch": 6.070124879923151,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002032902622481012,
      "loss": 4.5654,
      "mean_token_accuracy": 0.26795639246702196,
      "num_tokens": 144888753.0,
      "step": 63190
    },
    {
      "entropy": 5.500395202636719,
      "epoch": 6.0706051873198845,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020325776733396607,
      "loss": 4.6011,
      "mean_token_accuracy": 0.2667128682136536,
      "num_tokens": 144901483.0,
      "step": 63195
    },
    {
      "entropy": 5.505134868621826,
      "epoch": 6.071085494716619,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020322527408547255,
      "loss": 4.607,
      "mean_token_accuracy": 0.26688039153814314,
      "num_tokens": 144912065.0,
      "step": 63200
    },
    {
      "entropy": 5.546274137496948,
      "epoch": 6.071565802113352,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020319278250337502,
      "loss": 4.6493,
      "mean_token_accuracy": 0.26355180591344834,
      "num_tokens": 144923532.0,
      "step": 63205
    },
    {
      "entropy": 5.539047765731811,
      "epoch": 6.072046109510087,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020316029258842799,
      "loss": 4.6772,
      "mean_token_accuracy": 0.26280332505702975,
      "num_tokens": 144934553.0,
      "step": 63210
    },
    {
      "entropy": 5.481849527359008,
      "epoch": 6.07252641690682,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002031278043413855,
      "loss": 4.5818,
      "mean_token_accuracy": 0.27047395408153535,
      "num_tokens": 144945863.0,
      "step": 63215
    },
    {
      "entropy": 5.485468864440918,
      "epoch": 6.073006724303554,
      "grad_norm": 1.25,
      "learning_rate": 0.00020309531776300206,
      "loss": 4.5855,
      "mean_token_accuracy": 0.27976918816566465,
      "num_tokens": 144956622.0,
      "step": 63220
    },
    {
      "entropy": 5.478802824020386,
      "epoch": 6.073487031700288,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002030628328540318,
      "loss": 4.595,
      "mean_token_accuracy": 0.2674854457378387,
      "num_tokens": 144968653.0,
      "step": 63225
    },
    {
      "entropy": 5.604525804519653,
      "epoch": 6.073967339097022,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020303034961522892,
      "loss": 4.7411,
      "mean_token_accuracy": 0.2643617570400238,
      "num_tokens": 144981251.0,
      "step": 63230
    },
    {
      "entropy": 5.503230190277099,
      "epoch": 6.074447646493756,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020299786804734766,
      "loss": 4.5408,
      "mean_token_accuracy": 0.2753423810005188,
      "num_tokens": 144992396.0,
      "step": 63235
    },
    {
      "entropy": 5.584474563598633,
      "epoch": 6.07492795389049,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020296538815114213,
      "loss": 4.6627,
      "mean_token_accuracy": 0.2745090052485466,
      "num_tokens": 145003252.0,
      "step": 63240
    },
    {
      "entropy": 5.656764841079712,
      "epoch": 6.075408261287224,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020293290992736635,
      "loss": 4.7907,
      "mean_token_accuracy": 0.25589241534471513,
      "num_tokens": 145014762.0,
      "step": 63245
    },
    {
      "entropy": 5.504111242294312,
      "epoch": 6.075888568683958,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020290043337677438,
      "loss": 4.5854,
      "mean_token_accuracy": 0.2655637085437775,
      "num_tokens": 145026070.0,
      "step": 63250
    },
    {
      "entropy": 5.54669976234436,
      "epoch": 6.076368876080692,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020286795850012025,
      "loss": 4.6783,
      "mean_token_accuracy": 0.2642425671219826,
      "num_tokens": 145038477.0,
      "step": 63255
    },
    {
      "entropy": 5.521353816986084,
      "epoch": 6.076849183477425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020283548529815797,
      "loss": 4.5872,
      "mean_token_accuracy": 0.264971998333931,
      "num_tokens": 145050233.0,
      "step": 63260
    },
    {
      "entropy": 5.438601160049439,
      "epoch": 6.07732949087416,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002028030137716414,
      "loss": 4.4896,
      "mean_token_accuracy": 0.27903083264827727,
      "num_tokens": 145062099.0,
      "step": 63265
    },
    {
      "entropy": 5.471169948577881,
      "epoch": 6.077809798270893,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020277054392132455,
      "loss": 4.5709,
      "mean_token_accuracy": 0.2660563111305237,
      "num_tokens": 145074151.0,
      "step": 63270
    },
    {
      "entropy": 5.5453742980957035,
      "epoch": 6.0782901056676275,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020273807574796115,
      "loss": 4.6324,
      "mean_token_accuracy": 0.26387839764356613,
      "num_tokens": 145085523.0,
      "step": 63275
    },
    {
      "entropy": 5.613106298446655,
      "epoch": 6.078770413064361,
      "grad_norm": 1.5,
      "learning_rate": 0.000202705609252305,
      "loss": 4.6938,
      "mean_token_accuracy": 0.2579176604747772,
      "num_tokens": 145096456.0,
      "step": 63280
    },
    {
      "entropy": 5.594957494735718,
      "epoch": 6.079250720461095,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020267314443511003,
      "loss": 4.6935,
      "mean_token_accuracy": 0.25830467939376833,
      "num_tokens": 145106855.0,
      "step": 63285
    },
    {
      "entropy": 5.544490146636963,
      "epoch": 6.079731027857829,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020264068129712982,
      "loss": 4.6906,
      "mean_token_accuracy": 0.2554101675748825,
      "num_tokens": 145117442.0,
      "step": 63290
    },
    {
      "entropy": 5.448541450500488,
      "epoch": 6.080211335254563,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020260821983911808,
      "loss": 4.5227,
      "mean_token_accuracy": 0.26918595731258393,
      "num_tokens": 145128061.0,
      "step": 63295
    },
    {
      "entropy": 5.5055944442749025,
      "epoch": 6.0806916426512965,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020257576006182872,
      "loss": 4.6369,
      "mean_token_accuracy": 0.26376212686300277,
      "num_tokens": 145139486.0,
      "step": 63300
    },
    {
      "entropy": 5.571022272109985,
      "epoch": 6.081171950048031,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020254330196601495,
      "loss": 4.6798,
      "mean_token_accuracy": 0.26015686988830566,
      "num_tokens": 145150550.0,
      "step": 63305
    },
    {
      "entropy": 5.5445726871490475,
      "epoch": 6.081652257444764,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002025108455524306,
      "loss": 4.6569,
      "mean_token_accuracy": 0.26472638100385665,
      "num_tokens": 145162121.0,
      "step": 63310
    },
    {
      "entropy": 5.546026182174683,
      "epoch": 6.082132564841499,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020247839082182919,
      "loss": 4.6727,
      "mean_token_accuracy": 0.2672994375228882,
      "num_tokens": 145173516.0,
      "step": 63315
    },
    {
      "entropy": 5.584561681747436,
      "epoch": 6.082612872238233,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020244593777496417,
      "loss": 4.7174,
      "mean_token_accuracy": 0.2645305708050728,
      "num_tokens": 145185039.0,
      "step": 63320
    },
    {
      "entropy": 5.561371946334839,
      "epoch": 6.083093179634966,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020241348641258906,
      "loss": 4.6263,
      "mean_token_accuracy": 0.26684655994176865,
      "num_tokens": 145195844.0,
      "step": 63325
    },
    {
      "entropy": 5.508942747116089,
      "epoch": 6.083573487031701,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020238103673545733,
      "loss": 4.5842,
      "mean_token_accuracy": 0.26710212230682373,
      "num_tokens": 145207454.0,
      "step": 63330
    },
    {
      "entropy": 5.566770410537719,
      "epoch": 6.084053794428434,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002023485887443222,
      "loss": 4.6594,
      "mean_token_accuracy": 0.26579691618680956,
      "num_tokens": 145219568.0,
      "step": 63335
    },
    {
      "entropy": 5.499643850326538,
      "epoch": 6.084534101825168,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020231614243993712,
      "loss": 4.5934,
      "mean_token_accuracy": 0.26635933518409727,
      "num_tokens": 145230843.0,
      "step": 63340
    },
    {
      "entropy": 5.63680510520935,
      "epoch": 6.085014409221902,
      "grad_norm": 1.375,
      "learning_rate": 0.00020228369782305538,
      "loss": 4.7283,
      "mean_token_accuracy": 0.2530595734715462,
      "num_tokens": 145241759.0,
      "step": 63345
    },
    {
      "entropy": 5.498260545730591,
      "epoch": 6.085494716618636,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020225125489443024,
      "loss": 4.5604,
      "mean_token_accuracy": 0.26805648803710935,
      "num_tokens": 145253079.0,
      "step": 63350
    },
    {
      "entropy": 5.652970790863037,
      "epoch": 6.08597502401537,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020221881365481499,
      "loss": 4.7222,
      "mean_token_accuracy": 0.2495717376470566,
      "num_tokens": 145263733.0,
      "step": 63355
    },
    {
      "entropy": 5.589232683181763,
      "epoch": 6.086455331412104,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020218637410496278,
      "loss": 4.6805,
      "mean_token_accuracy": 0.264448806643486,
      "num_tokens": 145275754.0,
      "step": 63360
    },
    {
      "entropy": 5.514969205856323,
      "epoch": 6.086935638808837,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020215393624562668,
      "loss": 4.5444,
      "mean_token_accuracy": 0.2781086727976799,
      "num_tokens": 145288250.0,
      "step": 63365
    },
    {
      "entropy": 5.517730331420898,
      "epoch": 6.087415946205572,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020212150007755997,
      "loss": 4.6458,
      "mean_token_accuracy": 0.2587470695376396,
      "num_tokens": 145299069.0,
      "step": 63370
    },
    {
      "entropy": 5.619483375549317,
      "epoch": 6.087896253602305,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002020890656015155,
      "loss": 4.7009,
      "mean_token_accuracy": 0.26247870177030563,
      "num_tokens": 145311216.0,
      "step": 63375
    },
    {
      "entropy": 5.604793548583984,
      "epoch": 6.0883765609990395,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002020566328182465,
      "loss": 4.6738,
      "mean_token_accuracy": 0.2557791367173195,
      "num_tokens": 145322931.0,
      "step": 63380
    },
    {
      "entropy": 5.610116958618164,
      "epoch": 6.088856868395773,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002020242017285059,
      "loss": 4.7514,
      "mean_token_accuracy": 0.2523442402482033,
      "num_tokens": 145335358.0,
      "step": 63385
    },
    {
      "entropy": 5.605352592468262,
      "epoch": 6.089337175792507,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020199177233304666,
      "loss": 4.656,
      "mean_token_accuracy": 0.26102276891469955,
      "num_tokens": 145346392.0,
      "step": 63390
    },
    {
      "entropy": 5.581712865829468,
      "epoch": 6.0898174831892415,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002019593446326216,
      "loss": 4.6087,
      "mean_token_accuracy": 0.2680393993854523,
      "num_tokens": 145356711.0,
      "step": 63395
    },
    {
      "entropy": 5.5338774681091305,
      "epoch": 6.090297790585975,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002019269186279837,
      "loss": 4.7079,
      "mean_token_accuracy": 0.2572602316737175,
      "num_tokens": 145368279.0,
      "step": 63400
    },
    {
      "entropy": 5.411503267288208,
      "epoch": 6.090778097982709,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020189449431988577,
      "loss": 4.4543,
      "mean_token_accuracy": 0.28090626299381255,
      "num_tokens": 145379840.0,
      "step": 63405
    },
    {
      "entropy": 5.512851905822754,
      "epoch": 6.091258405379443,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020186207170908068,
      "loss": 4.5725,
      "mean_token_accuracy": 0.272024630010128,
      "num_tokens": 145392613.0,
      "step": 63410
    },
    {
      "entropy": 5.586895084381103,
      "epoch": 6.091738712776177,
      "grad_norm": 1.421875,
      "learning_rate": 0.000201829650796321,
      "loss": 4.6695,
      "mean_token_accuracy": 0.2605185627937317,
      "num_tokens": 145404369.0,
      "step": 63415
    },
    {
      "entropy": 5.481295585632324,
      "epoch": 6.0922190201729105,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020179723158235963,
      "loss": 4.5624,
      "mean_token_accuracy": 0.2696104273200035,
      "num_tokens": 145416214.0,
      "step": 63420
    },
    {
      "entropy": 5.542007732391357,
      "epoch": 6.092699327569645,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020176481406794917,
      "loss": 4.6181,
      "mean_token_accuracy": 0.26970224380493163,
      "num_tokens": 145427980.0,
      "step": 63425
    },
    {
      "entropy": 5.550861501693726,
      "epoch": 6.093179634966378,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020173239825384222,
      "loss": 4.6192,
      "mean_token_accuracy": 0.26732760965824126,
      "num_tokens": 145438608.0,
      "step": 63430
    },
    {
      "entropy": 5.58512601852417,
      "epoch": 6.093659942363113,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020169998414079138,
      "loss": 4.6818,
      "mean_token_accuracy": 0.26594310253858566,
      "num_tokens": 145450459.0,
      "step": 63435
    },
    {
      "entropy": 5.503444957733154,
      "epoch": 6.094140249759846,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002016675717295493,
      "loss": 4.6203,
      "mean_token_accuracy": 0.2686192154884338,
      "num_tokens": 145463539.0,
      "step": 63440
    },
    {
      "entropy": 5.558893442153931,
      "epoch": 6.09462055715658,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002016351610208685,
      "loss": 4.6529,
      "mean_token_accuracy": 0.26484771370887755,
      "num_tokens": 145474418.0,
      "step": 63445
    },
    {
      "entropy": 5.613425636291504,
      "epoch": 6.095100864553314,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020160275201550138,
      "loss": 4.7339,
      "mean_token_accuracy": 0.25541001707315447,
      "num_tokens": 145486532.0,
      "step": 63450
    },
    {
      "entropy": 5.55156569480896,
      "epoch": 6.095581171950048,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020157034471420044,
      "loss": 4.6499,
      "mean_token_accuracy": 0.26303938776254654,
      "num_tokens": 145498552.0,
      "step": 63455
    },
    {
      "entropy": 5.509260988235473,
      "epoch": 6.0960614793467816,
      "grad_norm": 1.390625,
      "learning_rate": 0.000201537939117718,
      "loss": 4.6406,
      "mean_token_accuracy": 0.2655620887875557,
      "num_tokens": 145509839.0,
      "step": 63460
    },
    {
      "entropy": 5.576595878601074,
      "epoch": 6.096541786743516,
      "grad_norm": 1.375,
      "learning_rate": 0.00020150553522680647,
      "loss": 4.7244,
      "mean_token_accuracy": 0.2568192005157471,
      "num_tokens": 145520895.0,
      "step": 63465
    },
    {
      "entropy": 5.54014892578125,
      "epoch": 6.09702209414025,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002014731330422182,
      "loss": 4.5742,
      "mean_token_accuracy": 0.2684568762779236,
      "num_tokens": 145531194.0,
      "step": 63470
    },
    {
      "entropy": 5.546180200576782,
      "epoch": 6.097502401536984,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020144073256470545,
      "loss": 4.7406,
      "mean_token_accuracy": 0.2576889216899872,
      "num_tokens": 145543086.0,
      "step": 63475
    },
    {
      "entropy": 5.559757328033447,
      "epoch": 6.097982708933718,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020140833379502055,
      "loss": 4.6715,
      "mean_token_accuracy": 0.25253279954195024,
      "num_tokens": 145554967.0,
      "step": 63480
    },
    {
      "entropy": 5.656436109542847,
      "epoch": 6.098463016330451,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002013759367339155,
      "loss": 4.7288,
      "mean_token_accuracy": 0.2601718157529831,
      "num_tokens": 145564892.0,
      "step": 63485
    },
    {
      "entropy": 5.61010708808899,
      "epoch": 6.098943323727186,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020134354138214266,
      "loss": 4.6614,
      "mean_token_accuracy": 0.26554731726646424,
      "num_tokens": 145576640.0,
      "step": 63490
    },
    {
      "entropy": 5.522754240036011,
      "epoch": 6.099423631123919,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000201311147740454,
      "loss": 4.6222,
      "mean_token_accuracy": 0.26703813672065735,
      "num_tokens": 145587910.0,
      "step": 63495
    },
    {
      "entropy": 5.505711269378662,
      "epoch": 6.0999039385206535,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002012787558096017,
      "loss": 4.6108,
      "mean_token_accuracy": 0.26398866176605223,
      "num_tokens": 145600360.0,
      "step": 63500
    },
    {
      "entropy": 5.545007944107056,
      "epoch": 6.100384245917387,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020124636559033782,
      "loss": 4.695,
      "mean_token_accuracy": 0.25825450867414473,
      "num_tokens": 145612319.0,
      "step": 63505
    },
    {
      "entropy": 5.512269687652588,
      "epoch": 6.100864553314121,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020121397708341437,
      "loss": 4.5823,
      "mean_token_accuracy": 0.2708490639925003,
      "num_tokens": 145623489.0,
      "step": 63510
    },
    {
      "entropy": 5.5148872375488285,
      "epoch": 6.101344860710855,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020118159028958315,
      "loss": 4.5657,
      "mean_token_accuracy": 0.2693127065896988,
      "num_tokens": 145634895.0,
      "step": 63515
    },
    {
      "entropy": 5.531505680084228,
      "epoch": 6.101825168107589,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002011492052095963,
      "loss": 4.6308,
      "mean_token_accuracy": 0.26685809940099714,
      "num_tokens": 145646458.0,
      "step": 63520
    },
    {
      "entropy": 5.5776121616363525,
      "epoch": 6.1023054755043225,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020111682184420555,
      "loss": 4.6152,
      "mean_token_accuracy": 0.26552532613277435,
      "num_tokens": 145656978.0,
      "step": 63525
    },
    {
      "entropy": 5.5253214836120605,
      "epoch": 6.102785782901057,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020108444019416283,
      "loss": 4.6442,
      "mean_token_accuracy": 0.25982756465673446,
      "num_tokens": 145668520.0,
      "step": 63530
    },
    {
      "entropy": 5.479861736297607,
      "epoch": 6.10326609029779,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002010520602602199,
      "loss": 4.5596,
      "mean_token_accuracy": 0.2679075330495834,
      "num_tokens": 145679555.0,
      "step": 63535
    },
    {
      "entropy": 5.492525243759156,
      "epoch": 6.1037463976945245,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020101968204312866,
      "loss": 4.577,
      "mean_token_accuracy": 0.2687950611114502,
      "num_tokens": 145689864.0,
      "step": 63540
    },
    {
      "entropy": 5.58464879989624,
      "epoch": 6.104226705091258,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020098730554364065,
      "loss": 4.6863,
      "mean_token_accuracy": 0.25965001434087753,
      "num_tokens": 145701236.0,
      "step": 63545
    },
    {
      "entropy": 5.483273029327393,
      "epoch": 6.104707012487992,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020095493076250758,
      "loss": 4.4911,
      "mean_token_accuracy": 0.2788840368390083,
      "num_tokens": 145712394.0,
      "step": 63550
    },
    {
      "entropy": 5.522745656967163,
      "epoch": 6.105187319884727,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020092255770048123,
      "loss": 4.6061,
      "mean_token_accuracy": 0.26490657180547716,
      "num_tokens": 145726023.0,
      "step": 63555
    },
    {
      "entropy": 5.557167100906372,
      "epoch": 6.10566762728146,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020089018635831301,
      "loss": 4.6567,
      "mean_token_accuracy": 0.2606297641992569,
      "num_tokens": 145737803.0,
      "step": 63560
    },
    {
      "entropy": 5.604971313476563,
      "epoch": 6.106147934678194,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020085781673675473,
      "loss": 4.6926,
      "mean_token_accuracy": 0.26156467348337176,
      "num_tokens": 145749630.0,
      "step": 63565
    },
    {
      "entropy": 5.593834638595581,
      "epoch": 6.106628242074928,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020082544883655778,
      "loss": 4.6682,
      "mean_token_accuracy": 0.2592995122075081,
      "num_tokens": 145761261.0,
      "step": 63570
    },
    {
      "entropy": 5.543236303329468,
      "epoch": 6.107108549471662,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020079308265847363,
      "loss": 4.5726,
      "mean_token_accuracy": 0.27413471043109894,
      "num_tokens": 145772620.0,
      "step": 63575
    },
    {
      "entropy": 5.483468818664551,
      "epoch": 6.107588856868396,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020076071820325375,
      "loss": 4.5705,
      "mean_token_accuracy": 0.273446099460125,
      "num_tokens": 145783938.0,
      "step": 63580
    },
    {
      "entropy": 5.5027546882629395,
      "epoch": 6.10806916426513,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020072835547164952,
      "loss": 4.6518,
      "mean_token_accuracy": 0.26717695891857146,
      "num_tokens": 145795601.0,
      "step": 63585
    },
    {
      "entropy": 5.582496213912964,
      "epoch": 6.108549471661863,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020069599446441232,
      "loss": 4.6741,
      "mean_token_accuracy": 0.2550822287797928,
      "num_tokens": 145807369.0,
      "step": 63590
    },
    {
      "entropy": 5.561588144302368,
      "epoch": 6.109029779058598,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020066363518229352,
      "loss": 4.646,
      "mean_token_accuracy": 0.2541254639625549,
      "num_tokens": 145818081.0,
      "step": 63595
    },
    {
      "entropy": 5.522036838531494,
      "epoch": 6.109510086455331,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002006312776260445,
      "loss": 4.6491,
      "mean_token_accuracy": 0.25812389999628066,
      "num_tokens": 145828712.0,
      "step": 63600
    },
    {
      "entropy": 5.585594320297242,
      "epoch": 6.109990393852065,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020059892179641626,
      "loss": 4.6481,
      "mean_token_accuracy": 0.26661555022001265,
      "num_tokens": 145840457.0,
      "step": 63605
    },
    {
      "entropy": 5.502903795242309,
      "epoch": 6.110470701248799,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0002005665676941601,
      "loss": 4.544,
      "mean_token_accuracy": 0.27286339104175567,
      "num_tokens": 145850092.0,
      "step": 63610
    },
    {
      "entropy": 5.493363952636718,
      "epoch": 6.110951008645533,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020053421532002726,
      "loss": 4.5723,
      "mean_token_accuracy": 0.2730685070157051,
      "num_tokens": 145861833.0,
      "step": 63615
    },
    {
      "entropy": 5.496992921829223,
      "epoch": 6.111431316042267,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020050186467476878,
      "loss": 4.5709,
      "mean_token_accuracy": 0.27343633621931074,
      "num_tokens": 145874316.0,
      "step": 63620
    },
    {
      "entropy": 5.604024124145508,
      "epoch": 6.111911623439001,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002004695157591358,
      "loss": 4.6892,
      "mean_token_accuracy": 0.25322683751583097,
      "num_tokens": 145885686.0,
      "step": 63625
    },
    {
      "entropy": 5.560802555084228,
      "epoch": 6.112391930835735,
      "grad_norm": 1.375,
      "learning_rate": 0.0002004371685738794,
      "loss": 4.5878,
      "mean_token_accuracy": 0.26590835899114607,
      "num_tokens": 145897451.0,
      "step": 63630
    },
    {
      "entropy": 5.534720802307129,
      "epoch": 6.112872238232469,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020040482311975056,
      "loss": 4.6392,
      "mean_token_accuracy": 0.26840159595012664,
      "num_tokens": 145909499.0,
      "step": 63635
    },
    {
      "entropy": 5.532949447631836,
      "epoch": 6.113352545629203,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020037247939750014,
      "loss": 4.6642,
      "mean_token_accuracy": 0.26453697979450225,
      "num_tokens": 145921025.0,
      "step": 63640
    },
    {
      "entropy": 5.521660995483399,
      "epoch": 6.1138328530259365,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002003401374078792,
      "loss": 4.6238,
      "mean_token_accuracy": 0.26747461706399916,
      "num_tokens": 145933094.0,
      "step": 63645
    },
    {
      "entropy": 5.457572412490845,
      "epoch": 6.114313160422671,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020030779715163855,
      "loss": 4.5703,
      "mean_token_accuracy": 0.27279768884181976,
      "num_tokens": 145944549.0,
      "step": 63650
    },
    {
      "entropy": 5.5490015029907225,
      "epoch": 6.114793467819404,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020027545862952905,
      "loss": 4.6443,
      "mean_token_accuracy": 0.2630264684557915,
      "num_tokens": 145955961.0,
      "step": 63655
    },
    {
      "entropy": 5.532362937927246,
      "epoch": 6.115273775216139,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020024312184230164,
      "loss": 4.6414,
      "mean_token_accuracy": 0.2620961904525757,
      "num_tokens": 145967789.0,
      "step": 63660
    },
    {
      "entropy": 5.580096626281739,
      "epoch": 6.115754082612872,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020021078679070681,
      "loss": 4.6623,
      "mean_token_accuracy": 0.26297716945409777,
      "num_tokens": 145978389.0,
      "step": 63665
    },
    {
      "entropy": 5.483082914352417,
      "epoch": 6.116234390009606,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020017845347549546,
      "loss": 4.5871,
      "mean_token_accuracy": 0.26701914370059965,
      "num_tokens": 145990545.0,
      "step": 63670
    },
    {
      "entropy": 5.490467739105225,
      "epoch": 6.11671469740634,
      "grad_norm": 1.25,
      "learning_rate": 0.00020014612189741832,
      "loss": 4.591,
      "mean_token_accuracy": 0.27358094453811643,
      "num_tokens": 146001234.0,
      "step": 63675
    },
    {
      "entropy": 5.496380090713501,
      "epoch": 6.117195004803074,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020011379205722585,
      "loss": 4.6147,
      "mean_token_accuracy": 0.2661309540271759,
      "num_tokens": 146013131.0,
      "step": 63680
    },
    {
      "entropy": 5.64533109664917,
      "epoch": 6.1176753121998075,
      "grad_norm": 1.265625,
      "learning_rate": 0.00020008146395566884,
      "loss": 4.7341,
      "mean_token_accuracy": 0.26095701158046725,
      "num_tokens": 146024774.0,
      "step": 63685
    },
    {
      "entropy": 5.67124080657959,
      "epoch": 6.118155619596542,
      "grad_norm": 1.625,
      "learning_rate": 0.0002000491375934978,
      "loss": 4.779,
      "mean_token_accuracy": 0.24983020275831222,
      "num_tokens": 146035059.0,
      "step": 63690
    },
    {
      "entropy": 5.632565593719482,
      "epoch": 6.118635926993275,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020001681297146317,
      "loss": 4.754,
      "mean_token_accuracy": 0.24682419896125793,
      "num_tokens": 146047057.0,
      "step": 63695
    },
    {
      "entropy": 5.578771924972534,
      "epoch": 6.11911623439001,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001999844900903155,
      "loss": 4.6565,
      "mean_token_accuracy": 0.260798779129982,
      "num_tokens": 146058269.0,
      "step": 63700
    },
    {
      "entropy": 5.5389237880706785,
      "epoch": 6.119596541786744,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001999521689508052,
      "loss": 4.594,
      "mean_token_accuracy": 0.261784528195858,
      "num_tokens": 146071121.0,
      "step": 63705
    },
    {
      "entropy": 5.533519983291626,
      "epoch": 6.120076849183477,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001999198495536827,
      "loss": 4.6192,
      "mean_token_accuracy": 0.2672332763671875,
      "num_tokens": 146082802.0,
      "step": 63710
    },
    {
      "entropy": 5.504478549957275,
      "epoch": 6.120557156580212,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019988753189969838,
      "loss": 4.597,
      "mean_token_accuracy": 0.26382258236408235,
      "num_tokens": 146094647.0,
      "step": 63715
    },
    {
      "entropy": 5.566830396652222,
      "epoch": 6.121037463976945,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019985521598960254,
      "loss": 4.6988,
      "mean_token_accuracy": 0.2679736778140068,
      "num_tokens": 146106631.0,
      "step": 63720
    },
    {
      "entropy": 5.603071832656861,
      "epoch": 6.1215177713736795,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019982290182414542,
      "loss": 4.712,
      "mean_token_accuracy": 0.2539738669991493,
      "num_tokens": 146118586.0,
      "step": 63725
    },
    {
      "entropy": 5.570912837982178,
      "epoch": 6.121998078770413,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019979058940407729,
      "loss": 4.5917,
      "mean_token_accuracy": 0.26435220837593076,
      "num_tokens": 146130500.0,
      "step": 63730
    },
    {
      "entropy": 5.422836780548096,
      "epoch": 6.122478386167147,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001997582787301483,
      "loss": 4.5232,
      "mean_token_accuracy": 0.2717283830046654,
      "num_tokens": 146141464.0,
      "step": 63735
    },
    {
      "entropy": 5.533276081085205,
      "epoch": 6.122958693563881,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019972596980310869,
      "loss": 4.6516,
      "mean_token_accuracy": 0.2622276872396469,
      "num_tokens": 146153975.0,
      "step": 63740
    },
    {
      "entropy": 5.477151393890381,
      "epoch": 6.123439000960615,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001996936626237086,
      "loss": 4.5708,
      "mean_token_accuracy": 0.274350406229496,
      "num_tokens": 146164840.0,
      "step": 63745
    },
    {
      "entropy": 5.479374551773072,
      "epoch": 6.123919308357348,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019966135719269807,
      "loss": 4.5457,
      "mean_token_accuracy": 0.26943138241767883,
      "num_tokens": 146176525.0,
      "step": 63750
    },
    {
      "entropy": 5.434027147293091,
      "epoch": 6.124399615754083,
      "grad_norm": 1.34375,
      "learning_rate": 0.000199629053510827,
      "loss": 4.5894,
      "mean_token_accuracy": 0.27221096158027647,
      "num_tokens": 146189050.0,
      "step": 63755
    },
    {
      "entropy": 5.600881290435791,
      "epoch": 6.124879923150816,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001995967515788456,
      "loss": 4.7014,
      "mean_token_accuracy": 0.2534781187772751,
      "num_tokens": 146198717.0,
      "step": 63760
    },
    {
      "entropy": 5.507308626174927,
      "epoch": 6.1253602305475505,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019956445139750363,
      "loss": 4.6023,
      "mean_token_accuracy": 0.27104726284742353,
      "num_tokens": 146209824.0,
      "step": 63765
    },
    {
      "entropy": 5.528431129455567,
      "epoch": 6.125840537944284,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019953215296755113,
      "loss": 4.6207,
      "mean_token_accuracy": 0.27025769650936127,
      "num_tokens": 146220528.0,
      "step": 63770
    },
    {
      "entropy": 5.46353235244751,
      "epoch": 6.126320845341018,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000199499856289738,
      "loss": 4.5518,
      "mean_token_accuracy": 0.26879429519176484,
      "num_tokens": 146232036.0,
      "step": 63775
    },
    {
      "entropy": 5.5950579166412355,
      "epoch": 6.126801152737752,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019946756136481408,
      "loss": 4.6635,
      "mean_token_accuracy": 0.2646224156022072,
      "num_tokens": 146242853.0,
      "step": 63780
    },
    {
      "entropy": 5.596765422821045,
      "epoch": 6.127281460134486,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019943526819352897,
      "loss": 4.6855,
      "mean_token_accuracy": 0.2548292100429535,
      "num_tokens": 146254332.0,
      "step": 63785
    },
    {
      "entropy": 5.5556823253631595,
      "epoch": 6.12776176753122,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001994029767766326,
      "loss": 4.619,
      "mean_token_accuracy": 0.27016020119190215,
      "num_tokens": 146266226.0,
      "step": 63790
    },
    {
      "entropy": 5.511308336257935,
      "epoch": 6.128242074927954,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019937068711487456,
      "loss": 4.6735,
      "mean_token_accuracy": 0.2631280288100243,
      "num_tokens": 146279200.0,
      "step": 63795
    },
    {
      "entropy": 5.4985894680023195,
      "epoch": 6.128722382324688,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019933839920900466,
      "loss": 4.5886,
      "mean_token_accuracy": 0.26983089596033094,
      "num_tokens": 146290223.0,
      "step": 63800
    },
    {
      "entropy": 5.550282430648804,
      "epoch": 6.129202689721422,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019930611305977243,
      "loss": 4.5919,
      "mean_token_accuracy": 0.26524660736322403,
      "num_tokens": 146301284.0,
      "step": 63805
    },
    {
      "entropy": 5.546987819671631,
      "epoch": 6.129682997118156,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001992738286679276,
      "loss": 4.6653,
      "mean_token_accuracy": 0.256363245844841,
      "num_tokens": 146312899.0,
      "step": 63810
    },
    {
      "entropy": 5.5450633525848385,
      "epoch": 6.130163304514889,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001992415460342195,
      "loss": 4.6459,
      "mean_token_accuracy": 0.26749693006277087,
      "num_tokens": 146323985.0,
      "step": 63815
    },
    {
      "entropy": 5.566667461395264,
      "epoch": 6.130643611911624,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001992092651593978,
      "loss": 4.6838,
      "mean_token_accuracy": 0.26027183383703234,
      "num_tokens": 146335028.0,
      "step": 63820
    },
    {
      "entropy": 5.558127546310425,
      "epoch": 6.131123919308357,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001991769860442118,
      "loss": 4.6844,
      "mean_token_accuracy": 0.2660338431596756,
      "num_tokens": 146346729.0,
      "step": 63825
    },
    {
      "entropy": 5.557045269012451,
      "epoch": 6.131604226705091,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019914470868941114,
      "loss": 4.6996,
      "mean_token_accuracy": 0.2636779323220253,
      "num_tokens": 146359914.0,
      "step": 63830
    },
    {
      "entropy": 5.561152601242066,
      "epoch": 6.132084534101825,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019911243309574507,
      "loss": 4.6424,
      "mean_token_accuracy": 0.2672276392579079,
      "num_tokens": 146371956.0,
      "step": 63835
    },
    {
      "entropy": 5.543655157089233,
      "epoch": 6.132564841498559,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000199080159263963,
      "loss": 4.6174,
      "mean_token_accuracy": 0.2678207948803902,
      "num_tokens": 146382666.0,
      "step": 63840
    },
    {
      "entropy": 5.562052536010742,
      "epoch": 6.133045148895293,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019904788719481422,
      "loss": 4.6785,
      "mean_token_accuracy": 0.2610196828842163,
      "num_tokens": 146394752.0,
      "step": 63845
    },
    {
      "entropy": 5.538807392120361,
      "epoch": 6.133525456292027,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019901561688904785,
      "loss": 4.627,
      "mean_token_accuracy": 0.26350231617689135,
      "num_tokens": 146407043.0,
      "step": 63850
    },
    {
      "entropy": 5.469693613052368,
      "epoch": 6.13400576368876,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019898334834741327,
      "loss": 4.4824,
      "mean_token_accuracy": 0.27966841459274294,
      "num_tokens": 146418839.0,
      "step": 63855
    },
    {
      "entropy": 5.534380006790161,
      "epoch": 6.134486071085495,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001989510815706596,
      "loss": 4.6185,
      "mean_token_accuracy": 0.2665853381156921,
      "num_tokens": 146429812.0,
      "step": 63860
    },
    {
      "entropy": 5.498738145828247,
      "epoch": 6.134966378482229,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019891881655953607,
      "loss": 4.5836,
      "mean_token_accuracy": 0.27262035459280015,
      "num_tokens": 146440008.0,
      "step": 63865
    },
    {
      "entropy": 5.674598836898804,
      "epoch": 6.1354466858789625,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019888655331479178,
      "loss": 4.823,
      "mean_token_accuracy": 0.24996848553419113,
      "num_tokens": 146450679.0,
      "step": 63870
    },
    {
      "entropy": 5.6191541194915775,
      "epoch": 6.135926993275697,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019885429183717556,
      "loss": 4.7162,
      "mean_token_accuracy": 0.25421948730945587,
      "num_tokens": 146463022.0,
      "step": 63875
    },
    {
      "entropy": 5.503224420547485,
      "epoch": 6.13640730067243,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001988220321274366,
      "loss": 4.565,
      "mean_token_accuracy": 0.27044849544763566,
      "num_tokens": 146474234.0,
      "step": 63880
    },
    {
      "entropy": 5.506978940963745,
      "epoch": 6.136887608069165,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019878977418632388,
      "loss": 4.6292,
      "mean_token_accuracy": 0.2677585706114769,
      "num_tokens": 146485738.0,
      "step": 63885
    },
    {
      "entropy": 5.4935859680175785,
      "epoch": 6.137367915465898,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019875751801458626,
      "loss": 4.6821,
      "mean_token_accuracy": 0.2571256563067436,
      "num_tokens": 146496931.0,
      "step": 63890
    },
    {
      "entropy": 5.547782278060913,
      "epoch": 6.137848222862632,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019872526361297272,
      "loss": 4.6552,
      "mean_token_accuracy": 0.26887100785970686,
      "num_tokens": 146507731.0,
      "step": 63895
    },
    {
      "entropy": 5.598525476455689,
      "epoch": 6.138328530259366,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019869301098223214,
      "loss": 4.6864,
      "mean_token_accuracy": 0.2591820374131203,
      "num_tokens": 146519748.0,
      "step": 63900
    },
    {
      "entropy": 5.581119251251221,
      "epoch": 6.1388088376561,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001986607601231132,
      "loss": 4.579,
      "mean_token_accuracy": 0.26971314698457716,
      "num_tokens": 146529536.0,
      "step": 63905
    },
    {
      "entropy": 5.508720731735229,
      "epoch": 6.1392891450528335,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019862851103636466,
      "loss": 4.6104,
      "mean_token_accuracy": 0.2734218031167984,
      "num_tokens": 146541482.0,
      "step": 63910
    },
    {
      "entropy": 5.4992882251739506,
      "epoch": 6.139769452449568,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019859626372273538,
      "loss": 4.624,
      "mean_token_accuracy": 0.260304594039917,
      "num_tokens": 146552820.0,
      "step": 63915
    },
    {
      "entropy": 5.511483097076416,
      "epoch": 6.140249759846301,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019856401818297393,
      "loss": 4.5863,
      "mean_token_accuracy": 0.26006352454423903,
      "num_tokens": 146564247.0,
      "step": 63920
    },
    {
      "entropy": 5.605426359176636,
      "epoch": 6.140730067243036,
      "grad_norm": 1.375,
      "learning_rate": 0.00019853177441782904,
      "loss": 4.7293,
      "mean_token_accuracy": 0.26314439475536344,
      "num_tokens": 146575776.0,
      "step": 63925
    },
    {
      "entropy": 5.643743085861206,
      "epoch": 6.141210374639769,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019849953242804936,
      "loss": 4.642,
      "mean_token_accuracy": 0.26675753444433215,
      "num_tokens": 146586938.0,
      "step": 63930
    },
    {
      "entropy": 5.424413108825684,
      "epoch": 6.141690682036503,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001984672922143833,
      "loss": 4.5003,
      "mean_token_accuracy": 0.27608164995908735,
      "num_tokens": 146598579.0,
      "step": 63935
    },
    {
      "entropy": 5.536844158172608,
      "epoch": 6.142170989433238,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019843505377757954,
      "loss": 4.5746,
      "mean_token_accuracy": 0.27086873799562455,
      "num_tokens": 146610238.0,
      "step": 63940
    },
    {
      "entropy": 5.542253732681274,
      "epoch": 6.142651296829971,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019840281711838636,
      "loss": 4.6819,
      "mean_token_accuracy": 0.26121353805065156,
      "num_tokens": 146620581.0,
      "step": 63945
    },
    {
      "entropy": 5.5255396366119385,
      "epoch": 6.1431316042267055,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001983705822375524,
      "loss": 4.6218,
      "mean_token_accuracy": 0.2620323374867439,
      "num_tokens": 146631425.0,
      "step": 63950
    },
    {
      "entropy": 5.607959985733032,
      "epoch": 6.143611911623439,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019833834913582588,
      "loss": 4.7038,
      "mean_token_accuracy": 0.2548371359705925,
      "num_tokens": 146643867.0,
      "step": 63955
    },
    {
      "entropy": 5.557074451446534,
      "epoch": 6.144092219020173,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001983061178139554,
      "loss": 4.6023,
      "mean_token_accuracy": 0.2724242299795151,
      "num_tokens": 146655653.0,
      "step": 63960
    },
    {
      "entropy": 5.415797901153565,
      "epoch": 6.144572526416907,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019827388827268905,
      "loss": 4.562,
      "mean_token_accuracy": 0.273182275891304,
      "num_tokens": 146667830.0,
      "step": 63965
    },
    {
      "entropy": 5.532304859161377,
      "epoch": 6.145052833813641,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019824166051277516,
      "loss": 4.6262,
      "mean_token_accuracy": 0.26664263308048247,
      "num_tokens": 146679366.0,
      "step": 63970
    },
    {
      "entropy": 5.46289381980896,
      "epoch": 6.145533141210374,
      "grad_norm": 1.515625,
      "learning_rate": 0.000198209434534962,
      "loss": 4.621,
      "mean_token_accuracy": 0.2590521365404129,
      "num_tokens": 146689287.0,
      "step": 63975
    },
    {
      "entropy": 5.521263885498047,
      "epoch": 6.146013448607109,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019817721033999782,
      "loss": 4.6012,
      "mean_token_accuracy": 0.27647820562124253,
      "num_tokens": 146699845.0,
      "step": 63980
    },
    {
      "entropy": 5.639087677001953,
      "epoch": 6.146493756003842,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019814498792863062,
      "loss": 4.7643,
      "mean_token_accuracy": 0.2570952981710434,
      "num_tokens": 146710866.0,
      "step": 63985
    },
    {
      "entropy": 5.621717357635498,
      "epoch": 6.1469740634005765,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019811276730160866,
      "loss": 4.7174,
      "mean_token_accuracy": 0.26106147319078443,
      "num_tokens": 146723352.0,
      "step": 63990
    },
    {
      "entropy": 5.600099658966064,
      "epoch": 6.14745437079731,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019808054845967988,
      "loss": 4.658,
      "mean_token_accuracy": 0.263405080139637,
      "num_tokens": 146734979.0,
      "step": 63995
    },
    {
      "entropy": 5.547343349456787,
      "epoch": 6.147934678194044,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019804833140359235,
      "loss": 4.6538,
      "mean_token_accuracy": 0.27195014208555224,
      "num_tokens": 146747218.0,
      "step": 64000
    },
    {
      "entropy": 5.5178266048431395,
      "epoch": 6.148414985590778,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00019801611613409404,
      "loss": 4.6163,
      "mean_token_accuracy": 0.2632916897535324,
      "num_tokens": 146757178.0,
      "step": 64005
    },
    {
      "entropy": 5.475858211517334,
      "epoch": 6.148895292987512,
      "grad_norm": 1.265625,
      "learning_rate": 0.000197983902651933,
      "loss": 4.597,
      "mean_token_accuracy": 0.26462350636720655,
      "num_tokens": 146769441.0,
      "step": 64010
    },
    {
      "entropy": 5.59573712348938,
      "epoch": 6.149375600384246,
      "grad_norm": 1.453125,
      "learning_rate": 0.000197951690957857,
      "loss": 4.6848,
      "mean_token_accuracy": 0.26209335029125214,
      "num_tokens": 146781278.0,
      "step": 64015
    },
    {
      "entropy": 5.52979884147644,
      "epoch": 6.14985590778098,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019791948105261404,
      "loss": 4.6171,
      "mean_token_accuracy": 0.26507615447044375,
      "num_tokens": 146792580.0,
      "step": 64020
    },
    {
      "entropy": 5.430289602279663,
      "epoch": 6.150336215177714,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019788727293695174,
      "loss": 4.4906,
      "mean_token_accuracy": 0.2740024298429489,
      "num_tokens": 146804581.0,
      "step": 64025
    },
    {
      "entropy": 5.573318672180176,
      "epoch": 6.1508165225744476,
      "grad_norm": 1.359375,
      "learning_rate": 0.000197855066611618,
      "loss": 4.7238,
      "mean_token_accuracy": 0.26130145490169526,
      "num_tokens": 146816388.0,
      "step": 64030
    },
    {
      "entropy": 5.6373215675354,
      "epoch": 6.151296829971182,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019782286207736056,
      "loss": 4.7498,
      "mean_token_accuracy": 0.2565506145358086,
      "num_tokens": 146829326.0,
      "step": 64035
    },
    {
      "entropy": 5.412512540817261,
      "epoch": 6.151777137367915,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019779065933492706,
      "loss": 4.4972,
      "mean_token_accuracy": 0.2753343015909195,
      "num_tokens": 146841763.0,
      "step": 64040
    },
    {
      "entropy": 5.567430925369263,
      "epoch": 6.15225744476465,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019775845838506523,
      "loss": 4.6301,
      "mean_token_accuracy": 0.2579917937517166,
      "num_tokens": 146852598.0,
      "step": 64045
    },
    {
      "entropy": 5.634658002853394,
      "epoch": 6.152737752161383,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019772625922852266,
      "loss": 4.6787,
      "mean_token_accuracy": 0.2630211442708969,
      "num_tokens": 146865439.0,
      "step": 64050
    },
    {
      "entropy": 5.595598840713501,
      "epoch": 6.153218059558117,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001976940618660469,
      "loss": 4.6573,
      "mean_token_accuracy": 0.2622625008225441,
      "num_tokens": 146875824.0,
      "step": 64055
    },
    {
      "entropy": 5.595555353164673,
      "epoch": 6.153698366954851,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001976618662983854,
      "loss": 4.649,
      "mean_token_accuracy": 0.2605834499001503,
      "num_tokens": 146886584.0,
      "step": 64060
    },
    {
      "entropy": 5.527118301391601,
      "epoch": 6.154178674351585,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019762967252628572,
      "loss": 4.661,
      "mean_token_accuracy": 0.26257350742816926,
      "num_tokens": 146898441.0,
      "step": 64065
    },
    {
      "entropy": 5.4889506816864015,
      "epoch": 6.154658981748319,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019759748055049532,
      "loss": 4.6036,
      "mean_token_accuracy": 0.260887710750103,
      "num_tokens": 146910096.0,
      "step": 64070
    },
    {
      "entropy": 5.527692461013794,
      "epoch": 6.155139289145053,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019756529037176157,
      "loss": 4.5813,
      "mean_token_accuracy": 0.2681431233882904,
      "num_tokens": 146921280.0,
      "step": 64075
    },
    {
      "entropy": 5.596574497222901,
      "epoch": 6.155619596541786,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001975331019908319,
      "loss": 4.6919,
      "mean_token_accuracy": 0.25161612033843994,
      "num_tokens": 146932577.0,
      "step": 64080
    },
    {
      "entropy": 5.646720600128174,
      "epoch": 6.156099903938521,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019750091540845355,
      "loss": 4.7483,
      "mean_token_accuracy": 0.25563304871320724,
      "num_tokens": 146943868.0,
      "step": 64085
    },
    {
      "entropy": 5.515724849700928,
      "epoch": 6.156580211335255,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019746873062537385,
      "loss": 4.5842,
      "mean_token_accuracy": 0.27282573878765104,
      "num_tokens": 146955769.0,
      "step": 64090
    },
    {
      "entropy": 5.51161413192749,
      "epoch": 6.1570605187319885,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019743654764233987,
      "loss": 4.575,
      "mean_token_accuracy": 0.271124342083931,
      "num_tokens": 146966808.0,
      "step": 64095
    },
    {
      "entropy": 5.541619491577149,
      "epoch": 6.157540826128723,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019740436646009906,
      "loss": 4.6336,
      "mean_token_accuracy": 0.26306398808956144,
      "num_tokens": 146978478.0,
      "step": 64100
    },
    {
      "entropy": 5.583771848678589,
      "epoch": 6.158021133525456,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001973721870793984,
      "loss": 4.7195,
      "mean_token_accuracy": 0.2585771307349205,
      "num_tokens": 146991855.0,
      "step": 64105
    },
    {
      "entropy": 5.606616401672364,
      "epoch": 6.1585014409221905,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019734000950098513,
      "loss": 4.7026,
      "mean_token_accuracy": 0.2603223294019699,
      "num_tokens": 147002606.0,
      "step": 64110
    },
    {
      "entropy": 5.524806261062622,
      "epoch": 6.158981748318924,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019730783372560619,
      "loss": 4.6305,
      "mean_token_accuracy": 0.2648417964577675,
      "num_tokens": 147013323.0,
      "step": 64115
    },
    {
      "entropy": 5.477319240570068,
      "epoch": 6.159462055715658,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019727565975400857,
      "loss": 4.641,
      "mean_token_accuracy": 0.26025501638650894,
      "num_tokens": 147024128.0,
      "step": 64120
    },
    {
      "entropy": 5.497671270370484,
      "epoch": 6.159942363112392,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019724348758693943,
      "loss": 4.606,
      "mean_token_accuracy": 0.26109065413475036,
      "num_tokens": 147036747.0,
      "step": 64125
    },
    {
      "entropy": 5.497667455673218,
      "epoch": 6.160422670509126,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019721131722514557,
      "loss": 4.5492,
      "mean_token_accuracy": 0.2680392265319824,
      "num_tokens": 147046948.0,
      "step": 64130
    },
    {
      "entropy": 5.4867973804473875,
      "epoch": 6.1609029779058595,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019717914866937392,
      "loss": 4.6194,
      "mean_token_accuracy": 0.266625913977623,
      "num_tokens": 147058070.0,
      "step": 64135
    },
    {
      "entropy": 5.5030618667602536,
      "epoch": 6.161383285302594,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001971469819203715,
      "loss": 4.6397,
      "mean_token_accuracy": 0.259457066655159,
      "num_tokens": 147068177.0,
      "step": 64140
    },
    {
      "entropy": 5.501549482345581,
      "epoch": 6.161863592699327,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001971148169788849,
      "loss": 4.5723,
      "mean_token_accuracy": 0.26844760328531264,
      "num_tokens": 147078562.0,
      "step": 64145
    },
    {
      "entropy": 5.550351905822754,
      "epoch": 6.162343900096062,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019708265384566092,
      "loss": 4.6623,
      "mean_token_accuracy": 0.25681157857179643,
      "num_tokens": 147090543.0,
      "step": 64150
    },
    {
      "entropy": 5.581176996231079,
      "epoch": 6.162824207492795,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019705049252144636,
      "loss": 4.7192,
      "mean_token_accuracy": 0.2579334512352943,
      "num_tokens": 147102204.0,
      "step": 64155
    },
    {
      "entropy": 5.630045557022095,
      "epoch": 6.163304514889529,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019701833300698795,
      "loss": 4.7244,
      "mean_token_accuracy": 0.26078583151102064,
      "num_tokens": 147113587.0,
      "step": 64160
    },
    {
      "entropy": 5.664146089553833,
      "epoch": 6.163784822286263,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019698617530303225,
      "loss": 4.7392,
      "mean_token_accuracy": 0.25924297571182253,
      "num_tokens": 147124871.0,
      "step": 64165
    },
    {
      "entropy": 5.599732398986816,
      "epoch": 6.164265129682997,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000196954019410326,
      "loss": 4.6727,
      "mean_token_accuracy": 0.25438967943191526,
      "num_tokens": 147136184.0,
      "step": 64170
    },
    {
      "entropy": 5.476594018936157,
      "epoch": 6.164745437079731,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001969218653296156,
      "loss": 4.5846,
      "mean_token_accuracy": 0.27176597267389296,
      "num_tokens": 147148611.0,
      "step": 64175
    },
    {
      "entropy": 5.548592853546142,
      "epoch": 6.165225744476465,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001968897130616476,
      "loss": 4.6661,
      "mean_token_accuracy": 0.2648413419723511,
      "num_tokens": 147161490.0,
      "step": 64180
    },
    {
      "entropy": 5.421257591247558,
      "epoch": 6.165706051873199,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019685756260716858,
      "loss": 4.5196,
      "mean_token_accuracy": 0.27420028001070024,
      "num_tokens": 147173321.0,
      "step": 64185
    },
    {
      "entropy": 5.553285455703735,
      "epoch": 6.166186359269933,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001968254139669249,
      "loss": 4.6421,
      "mean_token_accuracy": 0.26262365728616716,
      "num_tokens": 147185100.0,
      "step": 64190
    },
    {
      "entropy": 5.522534990310669,
      "epoch": 6.166666666666667,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019679326714166297,
      "loss": 4.6072,
      "mean_token_accuracy": 0.26616947203874586,
      "num_tokens": 147196345.0,
      "step": 64195
    },
    {
      "entropy": 5.556355762481689,
      "epoch": 6.1671469740634,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001967611221321291,
      "loss": 4.6521,
      "mean_token_accuracy": 0.26411412805318835,
      "num_tokens": 147205868.0,
      "step": 64200
    },
    {
      "entropy": 5.560463905334473,
      "epoch": 6.167627281460135,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019672897893906977,
      "loss": 4.6519,
      "mean_token_accuracy": 0.2671646520495415,
      "num_tokens": 147217506.0,
      "step": 64205
    },
    {
      "entropy": 5.502682113647461,
      "epoch": 6.168107588856868,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001966968375632311,
      "loss": 4.5413,
      "mean_token_accuracy": 0.2757652685046196,
      "num_tokens": 147229393.0,
      "step": 64210
    },
    {
      "entropy": 5.53810224533081,
      "epoch": 6.1685878962536025,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001966646980053593,
      "loss": 4.6595,
      "mean_token_accuracy": 0.2640548601746559,
      "num_tokens": 147240504.0,
      "step": 64215
    },
    {
      "entropy": 5.487601470947266,
      "epoch": 6.169068203650336,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019663256026620064,
      "loss": 4.5802,
      "mean_token_accuracy": 0.27161485105752947,
      "num_tokens": 147252716.0,
      "step": 64220
    },
    {
      "entropy": 5.591001653671265,
      "epoch": 6.16954851104707,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019660042434650122,
      "loss": 4.7143,
      "mean_token_accuracy": 0.2569474056363106,
      "num_tokens": 147263402.0,
      "step": 64225
    },
    {
      "entropy": 5.538327312469482,
      "epoch": 6.170028818443804,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001965682902470071,
      "loss": 4.6056,
      "mean_token_accuracy": 0.2651172086596489,
      "num_tokens": 147273904.0,
      "step": 64230
    },
    {
      "entropy": 5.536508464813233,
      "epoch": 6.170509125840538,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019653615796846446,
      "loss": 4.58,
      "mean_token_accuracy": 0.2692959666252136,
      "num_tokens": 147285728.0,
      "step": 64235
    },
    {
      "entropy": 5.529067516326904,
      "epoch": 6.1709894332372714,
      "grad_norm": 1.375,
      "learning_rate": 0.00019650402751161916,
      "loss": 4.6047,
      "mean_token_accuracy": 0.2682346642017365,
      "num_tokens": 147297204.0,
      "step": 64240
    },
    {
      "entropy": 5.56314377784729,
      "epoch": 6.171469740634006,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019647189887721733,
      "loss": 4.7252,
      "mean_token_accuracy": 0.252009753882885,
      "num_tokens": 147307630.0,
      "step": 64245
    },
    {
      "entropy": 5.622680044174194,
      "epoch": 6.17195004803074,
      "grad_norm": 1.25,
      "learning_rate": 0.00019643977206600487,
      "loss": 4.6638,
      "mean_token_accuracy": 0.26404002457857134,
      "num_tokens": 147318576.0,
      "step": 64250
    },
    {
      "entropy": 5.500941181182862,
      "epoch": 6.1724303554274735,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019640764707872755,
      "loss": 4.5309,
      "mean_token_accuracy": 0.27788711041212083,
      "num_tokens": 147329281.0,
      "step": 64255
    },
    {
      "entropy": 5.423029136657715,
      "epoch": 6.172910662824208,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019637552391613128,
      "loss": 4.5687,
      "mean_token_accuracy": 0.27108335345983503,
      "num_tokens": 147340761.0,
      "step": 64260
    },
    {
      "entropy": 5.455651140213012,
      "epoch": 6.173390970220941,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019634340257896195,
      "loss": 4.6184,
      "mean_token_accuracy": 0.264813169836998,
      "num_tokens": 147351606.0,
      "step": 64265
    },
    {
      "entropy": 5.492382431030274,
      "epoch": 6.173871277617676,
      "grad_norm": 1.25,
      "learning_rate": 0.00019631128306796513,
      "loss": 4.6682,
      "mean_token_accuracy": 0.26036559641361234,
      "num_tokens": 147362975.0,
      "step": 64270
    },
    {
      "entropy": 5.5322349071502686,
      "epoch": 6.174351585014409,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019627916538388678,
      "loss": 4.6148,
      "mean_token_accuracy": 0.2631732702255249,
      "num_tokens": 147374009.0,
      "step": 64275
    },
    {
      "entropy": 5.600596141815186,
      "epoch": 6.174831892411143,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001962470495274724,
      "loss": 4.6522,
      "mean_token_accuracy": 0.2643610119819641,
      "num_tokens": 147385586.0,
      "step": 64280
    },
    {
      "entropy": 5.534956312179565,
      "epoch": 6.175312199807877,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019621493549946766,
      "loss": 4.5775,
      "mean_token_accuracy": 0.2653154477477074,
      "num_tokens": 147397569.0,
      "step": 64285
    },
    {
      "entropy": 5.659997272491455,
      "epoch": 6.175792507204611,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001961828233006182,
      "loss": 4.8107,
      "mean_token_accuracy": 0.2567427456378937,
      "num_tokens": 147410513.0,
      "step": 64290
    },
    {
      "entropy": 5.611063528060913,
      "epoch": 6.176272814601345,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019615071293166964,
      "loss": 4.6631,
      "mean_token_accuracy": 0.25903465449810026,
      "num_tokens": 147422044.0,
      "step": 64295
    },
    {
      "entropy": 5.561987018585205,
      "epoch": 6.176753121998079,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019611860439336726,
      "loss": 4.6649,
      "mean_token_accuracy": 0.2569535970687866,
      "num_tokens": 147434619.0,
      "step": 64300
    },
    {
      "entropy": 5.51857624053955,
      "epoch": 6.177233429394812,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001960864976864567,
      "loss": 4.5853,
      "mean_token_accuracy": 0.26457626074552537,
      "num_tokens": 147445957.0,
      "step": 64305
    },
    {
      "entropy": 5.572715330123901,
      "epoch": 6.177713736791547,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019605439281168336,
      "loss": 4.6882,
      "mean_token_accuracy": 0.2595442682504654,
      "num_tokens": 147457260.0,
      "step": 64310
    },
    {
      "entropy": 5.554123497009277,
      "epoch": 6.17819404418828,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019602228976979254,
      "loss": 4.6853,
      "mean_token_accuracy": 0.261368615925312,
      "num_tokens": 147468892.0,
      "step": 64315
    },
    {
      "entropy": 5.457979965209961,
      "epoch": 6.178674351585014,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019599018856152962,
      "loss": 4.631,
      "mean_token_accuracy": 0.2682849392294884,
      "num_tokens": 147480083.0,
      "step": 64320
    },
    {
      "entropy": 5.474446392059326,
      "epoch": 6.179154658981748,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019595808918764008,
      "loss": 4.5591,
      "mean_token_accuracy": 0.27226208448410033,
      "num_tokens": 147491214.0,
      "step": 64325
    },
    {
      "entropy": 5.553346872329712,
      "epoch": 6.179634966378482,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019592599164886887,
      "loss": 4.6352,
      "mean_token_accuracy": 0.26851012259721757,
      "num_tokens": 147502499.0,
      "step": 64330
    },
    {
      "entropy": 5.4913492679595945,
      "epoch": 6.1801152737752165,
      "grad_norm": 1.375,
      "learning_rate": 0.0001958938959459613,
      "loss": 4.6361,
      "mean_token_accuracy": 0.26961071491241456,
      "num_tokens": 147514338.0,
      "step": 64335
    },
    {
      "entropy": 5.4987804889678955,
      "epoch": 6.18059558117195,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019586180207966264,
      "loss": 4.5604,
      "mean_token_accuracy": 0.2692994147539139,
      "num_tokens": 147524749.0,
      "step": 64340
    },
    {
      "entropy": 5.566159439086914,
      "epoch": 6.181075888568684,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019582971005071791,
      "loss": 4.6639,
      "mean_token_accuracy": 0.2695245236158371,
      "num_tokens": 147536123.0,
      "step": 64345
    },
    {
      "entropy": 5.586336517333985,
      "epoch": 6.181556195965418,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019579761985987222,
      "loss": 4.7527,
      "mean_token_accuracy": 0.2628609657287598,
      "num_tokens": 147546324.0,
      "step": 64350
    },
    {
      "entropy": 5.554686117172241,
      "epoch": 6.182036503362152,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019576553150787068,
      "loss": 4.6121,
      "mean_token_accuracy": 0.2664113029837608,
      "num_tokens": 147559076.0,
      "step": 64355
    },
    {
      "entropy": 5.543859767913818,
      "epoch": 6.1825168107588855,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001957334449954582,
      "loss": 4.6044,
      "mean_token_accuracy": 0.27750795185565946,
      "num_tokens": 147570983.0,
      "step": 64360
    },
    {
      "entropy": 5.53080940246582,
      "epoch": 6.18299711815562,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019570136032337964,
      "loss": 4.6564,
      "mean_token_accuracy": 0.2610304936766624,
      "num_tokens": 147584264.0,
      "step": 64365
    },
    {
      "entropy": 5.520500135421753,
      "epoch": 6.183477425552353,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019566927749238016,
      "loss": 4.567,
      "mean_token_accuracy": 0.2760824665427208,
      "num_tokens": 147594780.0,
      "step": 64370
    },
    {
      "entropy": 5.4763427734375,
      "epoch": 6.183957732949088,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019563719650320438,
      "loss": 4.5553,
      "mean_token_accuracy": 0.27632002383470533,
      "num_tokens": 147607151.0,
      "step": 64375
    },
    {
      "entropy": 5.563276863098144,
      "epoch": 6.184438040345821,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001956051173565973,
      "loss": 4.6134,
      "mean_token_accuracy": 0.26283784210681915,
      "num_tokens": 147616982.0,
      "step": 64380
    },
    {
      "entropy": 5.471196413040161,
      "epoch": 6.184918347742555,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019557304005330374,
      "loss": 4.5279,
      "mean_token_accuracy": 0.2768677890300751,
      "num_tokens": 147627430.0,
      "step": 64385
    },
    {
      "entropy": 5.474202728271484,
      "epoch": 6.185398655139289,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019554096459406824,
      "loss": 4.574,
      "mean_token_accuracy": 0.2660449340939522,
      "num_tokens": 147639507.0,
      "step": 64390
    },
    {
      "entropy": 5.520092582702636,
      "epoch": 6.185878962536023,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019550889097963558,
      "loss": 4.6634,
      "mean_token_accuracy": 0.26385027915239334,
      "num_tokens": 147650439.0,
      "step": 64395
    },
    {
      "entropy": 5.580659484863281,
      "epoch": 6.1863592699327565,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019547681921075043,
      "loss": 4.6949,
      "mean_token_accuracy": 0.25517932921648023,
      "num_tokens": 147662884.0,
      "step": 64400
    },
    {
      "entropy": 5.560041093826294,
      "epoch": 6.186839577329491,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019544474928815742,
      "loss": 4.691,
      "mean_token_accuracy": 0.2709386318922043,
      "num_tokens": 147674434.0,
      "step": 64405
    },
    {
      "entropy": 5.585399198532104,
      "epoch": 6.187319884726225,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001954126812126011,
      "loss": 4.6738,
      "mean_token_accuracy": 0.2696186602115631,
      "num_tokens": 147686196.0,
      "step": 64410
    },
    {
      "entropy": 5.530363321304321,
      "epoch": 6.187800192122959,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019538061498482607,
      "loss": 4.68,
      "mean_token_accuracy": 0.2564491659402847,
      "num_tokens": 147698507.0,
      "step": 64415
    },
    {
      "entropy": 5.609235095977783,
      "epoch": 6.188280499519693,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019534855060557665,
      "loss": 4.6853,
      "mean_token_accuracy": 0.25701709389686583,
      "num_tokens": 147710387.0,
      "step": 64420
    },
    {
      "entropy": 5.542082405090332,
      "epoch": 6.188760806916426,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019531648807559737,
      "loss": 4.6121,
      "mean_token_accuracy": 0.2636721312999725,
      "num_tokens": 147722082.0,
      "step": 64425
    },
    {
      "entropy": 5.57137713432312,
      "epoch": 6.189241114313161,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019528442739563263,
      "loss": 4.6792,
      "mean_token_accuracy": 0.26152551025152204,
      "num_tokens": 147733836.0,
      "step": 64430
    },
    {
      "entropy": 5.531088733673096,
      "epoch": 6.189721421709894,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019525236856642676,
      "loss": 4.6193,
      "mean_token_accuracy": 0.2688797667622566,
      "num_tokens": 147745093.0,
      "step": 64435
    },
    {
      "entropy": 5.463660192489624,
      "epoch": 6.1902017291066285,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001952203115887241,
      "loss": 4.5797,
      "mean_token_accuracy": 0.262809219956398,
      "num_tokens": 147757427.0,
      "step": 64440
    },
    {
      "entropy": 5.572685956954956,
      "epoch": 6.190682036503362,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019518825646326905,
      "loss": 4.6645,
      "mean_token_accuracy": 0.26648886054754256,
      "num_tokens": 147768489.0,
      "step": 64445
    },
    {
      "entropy": 5.581135225296021,
      "epoch": 6.191162343900096,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019515620319080556,
      "loss": 4.6223,
      "mean_token_accuracy": 0.266724593937397,
      "num_tokens": 147779371.0,
      "step": 64450
    },
    {
      "entropy": 5.506928253173828,
      "epoch": 6.19164265129683,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019512415177207792,
      "loss": 4.5553,
      "mean_token_accuracy": 0.2729090854525566,
      "num_tokens": 147790826.0,
      "step": 64455
    },
    {
      "entropy": 5.492562580108642,
      "epoch": 6.192122958693564,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019509210220783032,
      "loss": 4.6572,
      "mean_token_accuracy": 0.26619541049003603,
      "num_tokens": 147801317.0,
      "step": 64460
    },
    {
      "entropy": 5.492790603637696,
      "epoch": 6.192603266090297,
      "grad_norm": 1.375,
      "learning_rate": 0.00019506005449880683,
      "loss": 4.5508,
      "mean_token_accuracy": 0.2729938730597496,
      "num_tokens": 147812045.0,
      "step": 64465
    },
    {
      "entropy": 5.634143304824829,
      "epoch": 6.193083573487032,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001950280086457515,
      "loss": 4.7519,
      "mean_token_accuracy": 0.253858382999897,
      "num_tokens": 147823377.0,
      "step": 64470
    },
    {
      "entropy": 5.519191694259644,
      "epoch": 6.193563880883765,
      "grad_norm": 1.375,
      "learning_rate": 0.00019499596464940845,
      "loss": 4.577,
      "mean_token_accuracy": 0.2658052623271942,
      "num_tokens": 147834937.0,
      "step": 64475
    },
    {
      "entropy": 5.403230381011963,
      "epoch": 6.1940441882804995,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001949639225105214,
      "loss": 4.4874,
      "mean_token_accuracy": 0.2776879325509071,
      "num_tokens": 147845485.0,
      "step": 64480
    },
    {
      "entropy": 5.494584751129151,
      "epoch": 6.194524495677234,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019493188222983443,
      "loss": 4.6714,
      "mean_token_accuracy": 0.2673429220914841,
      "num_tokens": 147856115.0,
      "step": 64485
    },
    {
      "entropy": 5.61151237487793,
      "epoch": 6.195004803073967,
      "grad_norm": 1.375,
      "learning_rate": 0.00019489984380809144,
      "loss": 4.6766,
      "mean_token_accuracy": 0.2606715723872185,
      "num_tokens": 147867209.0,
      "step": 64490
    },
    {
      "entropy": 5.449818134307861,
      "epoch": 6.195485110470702,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001948678072460362,
      "loss": 4.4921,
      "mean_token_accuracy": 0.2794219940900803,
      "num_tokens": 147878568.0,
      "step": 64495
    },
    {
      "entropy": 5.510498523712158,
      "epoch": 6.195965417867435,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019483577254441248,
      "loss": 4.5976,
      "mean_token_accuracy": 0.2674685835838318,
      "num_tokens": 147889709.0,
      "step": 64500
    },
    {
      "entropy": 5.566466236114502,
      "epoch": 6.196445725264169,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001948037397039642,
      "loss": 4.6876,
      "mean_token_accuracy": 0.2573281556367874,
      "num_tokens": 147901513.0,
      "step": 64505
    },
    {
      "entropy": 5.571881055831909,
      "epoch": 6.196926032660903,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019477170872543486,
      "loss": 4.6294,
      "mean_token_accuracy": 0.26393183022737504,
      "num_tokens": 147912796.0,
      "step": 64510
    },
    {
      "entropy": 5.593442058563232,
      "epoch": 6.197406340057637,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019473967960956828,
      "loss": 4.6778,
      "mean_token_accuracy": 0.26443364173173906,
      "num_tokens": 147924596.0,
      "step": 64515
    },
    {
      "entropy": 5.572287750244141,
      "epoch": 6.197886647454371,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019470765235710802,
      "loss": 4.6833,
      "mean_token_accuracy": 0.26601952612400054,
      "num_tokens": 147935430.0,
      "step": 64520
    },
    {
      "entropy": 5.57545313835144,
      "epoch": 6.198366954851105,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001946756269687976,
      "loss": 4.6572,
      "mean_token_accuracy": 0.26458934247493743,
      "num_tokens": 147946693.0,
      "step": 64525
    },
    {
      "entropy": 5.510348224639893,
      "epoch": 6.198847262247838,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019464360344538062,
      "loss": 4.6349,
      "mean_token_accuracy": 0.2632124796509743,
      "num_tokens": 147958155.0,
      "step": 64530
    },
    {
      "entropy": 5.524910974502563,
      "epoch": 6.199327569644573,
      "grad_norm": 1.375,
      "learning_rate": 0.00019461158178760057,
      "loss": 4.6544,
      "mean_token_accuracy": 0.25943855494260787,
      "num_tokens": 147969893.0,
      "step": 64535
    },
    {
      "entropy": 5.56368989944458,
      "epoch": 6.199807877041306,
      "grad_norm": 1.375,
      "learning_rate": 0.00019457956199620086,
      "loss": 4.6909,
      "mean_token_accuracy": 0.2574020490050316,
      "num_tokens": 147981361.0,
      "step": 64540
    },
    {
      "entropy": 5.566048860549927,
      "epoch": 6.20028818443804,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019454754407192493,
      "loss": 4.6365,
      "mean_token_accuracy": 0.26011005938053133,
      "num_tokens": 147992904.0,
      "step": 64545
    },
    {
      "entropy": 5.5893511295318605,
      "epoch": 6.200768491834774,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019451552801551613,
      "loss": 4.6624,
      "mean_token_accuracy": 0.2662329703569412,
      "num_tokens": 148003762.0,
      "step": 64550
    },
    {
      "entropy": 5.5438923835754395,
      "epoch": 6.201248799231508,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019448351382771783,
      "loss": 4.603,
      "mean_token_accuracy": 0.2679720759391785,
      "num_tokens": 148015129.0,
      "step": 64555
    },
    {
      "entropy": 5.5803831100463865,
      "epoch": 6.2017291066282425,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001944515015092732,
      "loss": 4.6901,
      "mean_token_accuracy": 0.2586836189031601,
      "num_tokens": 148026202.0,
      "step": 64560
    },
    {
      "entropy": 5.631416893005371,
      "epoch": 6.202209414024976,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001944194910609256,
      "loss": 4.6964,
      "mean_token_accuracy": 0.2622792825102806,
      "num_tokens": 148036599.0,
      "step": 64565
    },
    {
      "entropy": 5.528493595123291,
      "epoch": 6.20268972142171,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001943874824834181,
      "loss": 4.6033,
      "mean_token_accuracy": 0.2632400020956993,
      "num_tokens": 148049263.0,
      "step": 64570
    },
    {
      "entropy": 5.5119565486907955,
      "epoch": 6.203170028818444,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001943554757774938,
      "loss": 4.5663,
      "mean_token_accuracy": 0.26718822717666624,
      "num_tokens": 148059543.0,
      "step": 64575
    },
    {
      "entropy": 5.437027072906494,
      "epoch": 6.203650336215178,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019432347094389596,
      "loss": 4.5578,
      "mean_token_accuracy": 0.2744017854332924,
      "num_tokens": 148071479.0,
      "step": 64580
    },
    {
      "entropy": 5.544295358657837,
      "epoch": 6.2041306436119115,
      "grad_norm": 1.25,
      "learning_rate": 0.00019429146798336755,
      "loss": 4.6144,
      "mean_token_accuracy": 0.265205480158329,
      "num_tokens": 148083152.0,
      "step": 64585
    },
    {
      "entropy": 5.51594181060791,
      "epoch": 6.204610951008646,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019425946689665164,
      "loss": 4.596,
      "mean_token_accuracy": 0.26916255354881286,
      "num_tokens": 148095362.0,
      "step": 64590
    },
    {
      "entropy": 5.507626390457153,
      "epoch": 6.205091258405379,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019422746768449117,
      "loss": 4.5512,
      "mean_token_accuracy": 0.26901700645685195,
      "num_tokens": 148106188.0,
      "step": 64595
    },
    {
      "entropy": 5.5947418212890625,
      "epoch": 6.2055715658021136,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019419547034762902,
      "loss": 4.7041,
      "mean_token_accuracy": 0.2592119723558426,
      "num_tokens": 148117282.0,
      "step": 64600
    },
    {
      "entropy": 5.539407396316529,
      "epoch": 6.206051873198847,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001941634748868081,
      "loss": 4.6671,
      "mean_token_accuracy": 0.26534550487995145,
      "num_tokens": 148129400.0,
      "step": 64605
    },
    {
      "entropy": 5.578109407424927,
      "epoch": 6.206532180595581,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019413148130277124,
      "loss": 4.6668,
      "mean_token_accuracy": 0.26190226823091506,
      "num_tokens": 148141019.0,
      "step": 64610
    },
    {
      "entropy": 5.553600835800171,
      "epoch": 6.207012487992315,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019409948959626128,
      "loss": 4.6226,
      "mean_token_accuracy": 0.25839742422103884,
      "num_tokens": 148152297.0,
      "step": 64615
    },
    {
      "entropy": 5.572001600265503,
      "epoch": 6.207492795389049,
      "grad_norm": 1.375,
      "learning_rate": 0.0001940674997680209,
      "loss": 4.6025,
      "mean_token_accuracy": 0.26750746816396714,
      "num_tokens": 148163011.0,
      "step": 64620
    },
    {
      "entropy": 5.555718469619751,
      "epoch": 6.2079731027857825,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019403551181879298,
      "loss": 4.6322,
      "mean_token_accuracy": 0.2599629133939743,
      "num_tokens": 148173880.0,
      "step": 64625
    },
    {
      "entropy": 5.541150903701782,
      "epoch": 6.208453410182517,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019400352574932002,
      "loss": 4.7205,
      "mean_token_accuracy": 0.2516786873340607,
      "num_tokens": 148185838.0,
      "step": 64630
    },
    {
      "entropy": 5.571783018112183,
      "epoch": 6.208933717579251,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001939715415603446,
      "loss": 4.668,
      "mean_token_accuracy": 0.2638304069638252,
      "num_tokens": 148196585.0,
      "step": 64635
    },
    {
      "entropy": 5.530551385879517,
      "epoch": 6.209414024975985,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019393955925260938,
      "loss": 4.6259,
      "mean_token_accuracy": 0.2673601686954498,
      "num_tokens": 148208469.0,
      "step": 64640
    },
    {
      "entropy": 5.524638700485229,
      "epoch": 6.209894332372719,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019390757882685694,
      "loss": 4.6436,
      "mean_token_accuracy": 0.26256491243839264,
      "num_tokens": 148220341.0,
      "step": 64645
    },
    {
      "entropy": 5.57599573135376,
      "epoch": 6.210374639769452,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019387560028382968,
      "loss": 4.6047,
      "mean_token_accuracy": 0.2673928439617157,
      "num_tokens": 148230879.0,
      "step": 64650
    },
    {
      "entropy": 5.6028480052948,
      "epoch": 6.210854947166187,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019384362362427022,
      "loss": 4.6666,
      "mean_token_accuracy": 0.2620189696550369,
      "num_tokens": 148243363.0,
      "step": 64655
    },
    {
      "entropy": 5.58856348991394,
      "epoch": 6.21133525456292,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019381164884892072,
      "loss": 4.6948,
      "mean_token_accuracy": 0.2597568079829216,
      "num_tokens": 148254679.0,
      "step": 64660
    },
    {
      "entropy": 5.479299402236938,
      "epoch": 6.2118155619596545,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019377967595852368,
      "loss": 4.5889,
      "mean_token_accuracy": 0.2610502228140831,
      "num_tokens": 148268204.0,
      "step": 64665
    },
    {
      "entropy": 5.5319455623626705,
      "epoch": 6.212295869356388,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001937477049538214,
      "loss": 4.5938,
      "mean_token_accuracy": 0.2665530487895012,
      "num_tokens": 148278642.0,
      "step": 64670
    },
    {
      "entropy": 5.5256610870361325,
      "epoch": 6.212776176753122,
      "grad_norm": 1.5,
      "learning_rate": 0.00019371573583555613,
      "loss": 4.593,
      "mean_token_accuracy": 0.26914713382720945,
      "num_tokens": 148290555.0,
      "step": 64675
    },
    {
      "entropy": 5.546334218978882,
      "epoch": 6.213256484149856,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019368376860447008,
      "loss": 4.6713,
      "mean_token_accuracy": 0.26454060673713686,
      "num_tokens": 148300240.0,
      "step": 64680
    },
    {
      "entropy": 5.505762577056885,
      "epoch": 6.21373679154659,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019365180326130554,
      "loss": 4.5369,
      "mean_token_accuracy": 0.26926530003547666,
      "num_tokens": 148310725.0,
      "step": 64685
    },
    {
      "entropy": 5.5189642906188965,
      "epoch": 6.214217098943323,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019361983980680452,
      "loss": 4.5947,
      "mean_token_accuracy": 0.26882189214229585,
      "num_tokens": 148323230.0,
      "step": 64690
    },
    {
      "entropy": 5.475341796875,
      "epoch": 6.214697406340058,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019358787824170915,
      "loss": 4.6304,
      "mean_token_accuracy": 0.25892275869846343,
      "num_tokens": 148335456.0,
      "step": 64695
    },
    {
      "entropy": 5.500913381576538,
      "epoch": 6.215177713736791,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001935559185667615,
      "loss": 4.5998,
      "mean_token_accuracy": 0.2667180925607681,
      "num_tokens": 148347071.0,
      "step": 64700
    },
    {
      "entropy": 5.59953088760376,
      "epoch": 6.2156580211335255,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001935239607827036,
      "loss": 4.7032,
      "mean_token_accuracy": 0.2574209913611412,
      "num_tokens": 148358748.0,
      "step": 64705
    },
    {
      "entropy": 5.548409700393677,
      "epoch": 6.216138328530259,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019349200489027747,
      "loss": 4.6546,
      "mean_token_accuracy": 0.2682051122188568,
      "num_tokens": 148369271.0,
      "step": 64710
    },
    {
      "entropy": 5.6329419136047365,
      "epoch": 6.216618635926993,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019346005089022496,
      "loss": 4.7412,
      "mean_token_accuracy": 0.2582503318786621,
      "num_tokens": 148379575.0,
      "step": 64715
    },
    {
      "entropy": 5.470215368270874,
      "epoch": 6.217098943323728,
      "grad_norm": 1.375,
      "learning_rate": 0.00019342809878328783,
      "loss": 4.5588,
      "mean_token_accuracy": 0.2765799254179001,
      "num_tokens": 148391315.0,
      "step": 64720
    },
    {
      "entropy": 5.504700708389282,
      "epoch": 6.217579250720461,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019339614857020803,
      "loss": 4.6356,
      "mean_token_accuracy": 0.2653097063302994,
      "num_tokens": 148403137.0,
      "step": 64725
    },
    {
      "entropy": 5.545669984817505,
      "epoch": 6.218059558117195,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001933642002517274,
      "loss": 4.6791,
      "mean_token_accuracy": 0.2643677368760109,
      "num_tokens": 148414767.0,
      "step": 64730
    },
    {
      "entropy": 5.53811297416687,
      "epoch": 6.218539865513929,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019333225382858765,
      "loss": 4.6274,
      "mean_token_accuracy": 0.25902105271816256,
      "num_tokens": 148426474.0,
      "step": 64735
    },
    {
      "entropy": 5.576452207565308,
      "epoch": 6.219020172910663,
      "grad_norm": 1.5,
      "learning_rate": 0.00019330030930153036,
      "loss": 4.6279,
      "mean_token_accuracy": 0.26672418117523194,
      "num_tokens": 148437092.0,
      "step": 64740
    },
    {
      "entropy": 5.584133005142212,
      "epoch": 6.2195004803073966,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019326836667129748,
      "loss": 4.7143,
      "mean_token_accuracy": 0.2572834551334381,
      "num_tokens": 148448616.0,
      "step": 64745
    },
    {
      "entropy": 5.662796354293823,
      "epoch": 6.219980787704131,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019323642593863035,
      "loss": 4.7781,
      "mean_token_accuracy": 0.24737419486045836,
      "num_tokens": 148460153.0,
      "step": 64750
    },
    {
      "entropy": 5.605150604248047,
      "epoch": 6.220461095100864,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001932044871042705,
      "loss": 4.7099,
      "mean_token_accuracy": 0.2505455926060677,
      "num_tokens": 148470728.0,
      "step": 64755
    },
    {
      "entropy": 5.532744455337524,
      "epoch": 6.220941402497599,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019317255016895968,
      "loss": 4.5458,
      "mean_token_accuracy": 0.2720814123749733,
      "num_tokens": 148481312.0,
      "step": 64760
    },
    {
      "entropy": 5.49589958190918,
      "epoch": 6.221421709894332,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001931406151334393,
      "loss": 4.5606,
      "mean_token_accuracy": 0.26917386054992676,
      "num_tokens": 148491547.0,
      "step": 64765
    },
    {
      "entropy": 5.485755872726441,
      "epoch": 6.221902017291066,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019310868199845066,
      "loss": 4.5959,
      "mean_token_accuracy": 0.27283268570899966,
      "num_tokens": 148503380.0,
      "step": 64770
    },
    {
      "entropy": 5.426678657531738,
      "epoch": 6.2223823246878,
      "grad_norm": 1.375,
      "learning_rate": 0.00019307675076473535,
      "loss": 4.5465,
      "mean_token_accuracy": 0.27069038301706316,
      "num_tokens": 148514879.0,
      "step": 64775
    },
    {
      "entropy": 5.508114957809449,
      "epoch": 6.222862632084534,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019304482143303465,
      "loss": 4.5721,
      "mean_token_accuracy": 0.27379622459411623,
      "num_tokens": 148525390.0,
      "step": 64780
    },
    {
      "entropy": 5.618040132522583,
      "epoch": 6.223342939481268,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019301289400408978,
      "loss": 4.6196,
      "mean_token_accuracy": 0.2645293816924095,
      "num_tokens": 148536460.0,
      "step": 64785
    },
    {
      "entropy": 5.578921842575073,
      "epoch": 6.223823246878002,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019298096847864205,
      "loss": 4.6786,
      "mean_token_accuracy": 0.263154511153698,
      "num_tokens": 148546539.0,
      "step": 64790
    },
    {
      "entropy": 5.483044385910034,
      "epoch": 6.224303554274736,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019294904485743264,
      "loss": 4.5587,
      "mean_token_accuracy": 0.272451114654541,
      "num_tokens": 148556503.0,
      "step": 64795
    },
    {
      "entropy": 5.554987478256225,
      "epoch": 6.22478386167147,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001929171231412028,
      "loss": 4.6933,
      "mean_token_accuracy": 0.2576795548200607,
      "num_tokens": 148569459.0,
      "step": 64800
    },
    {
      "entropy": 5.58113579750061,
      "epoch": 6.225264169068204,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001928852033306937,
      "loss": 4.7047,
      "mean_token_accuracy": 0.2663125678896904,
      "num_tokens": 148580612.0,
      "step": 64805
    },
    {
      "entropy": 5.577726364135742,
      "epoch": 6.2257444764649374,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019285328542664626,
      "loss": 4.7297,
      "mean_token_accuracy": 0.25469988733530047,
      "num_tokens": 148592626.0,
      "step": 64810
    },
    {
      "entropy": 5.4933980941772464,
      "epoch": 6.226224783861672,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019282136942980167,
      "loss": 4.5373,
      "mean_token_accuracy": 0.2745485842227936,
      "num_tokens": 148603033.0,
      "step": 64815
    },
    {
      "entropy": 5.572376012802124,
      "epoch": 6.226705091258405,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019278945534090088,
      "loss": 4.6824,
      "mean_token_accuracy": 0.25322502553462983,
      "num_tokens": 148614781.0,
      "step": 64820
    },
    {
      "entropy": 5.517554140090942,
      "epoch": 6.2271853986551395,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019275754316068472,
      "loss": 4.6044,
      "mean_token_accuracy": 0.267966565489769,
      "num_tokens": 148626642.0,
      "step": 64825
    },
    {
      "entropy": 5.569476795196533,
      "epoch": 6.227665706051873,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019272563288989426,
      "loss": 4.6022,
      "mean_token_accuracy": 0.2720278725028038,
      "num_tokens": 148637509.0,
      "step": 64830
    },
    {
      "entropy": 5.478944492340088,
      "epoch": 6.228146013448607,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019269372452927037,
      "loss": 4.5772,
      "mean_token_accuracy": 0.2689176708459854,
      "num_tokens": 148649324.0,
      "step": 64835
    },
    {
      "entropy": 5.540562963485717,
      "epoch": 6.228626320845341,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019266181807955367,
      "loss": 4.6212,
      "mean_token_accuracy": 0.26403075754642485,
      "num_tokens": 148660639.0,
      "step": 64840
    },
    {
      "entropy": 5.579916620254517,
      "epoch": 6.229106628242075,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019262991354148512,
      "loss": 4.6623,
      "mean_token_accuracy": 0.26294576972723005,
      "num_tokens": 148671758.0,
      "step": 64845
    },
    {
      "entropy": 5.53311824798584,
      "epoch": 6.2295869356388085,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019259801091580536,
      "loss": 4.6618,
      "mean_token_accuracy": 0.25885886400938035,
      "num_tokens": 148683650.0,
      "step": 64850
    },
    {
      "entropy": 5.639308977127075,
      "epoch": 6.230067243035543,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001925661102032551,
      "loss": 4.6908,
      "mean_token_accuracy": 0.25314414203166963,
      "num_tokens": 148694678.0,
      "step": 64855
    },
    {
      "entropy": 5.5634300231933596,
      "epoch": 6.230547550432276,
      "grad_norm": 1.421875,
      "learning_rate": 0.000192534211404575,
      "loss": 4.6182,
      "mean_token_accuracy": 0.26662830710411073,
      "num_tokens": 148707034.0,
      "step": 64860
    },
    {
      "entropy": 5.617035007476806,
      "epoch": 6.231027857829011,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019250231452050572,
      "loss": 4.6526,
      "mean_token_accuracy": 0.2569588005542755,
      "num_tokens": 148717829.0,
      "step": 64865
    },
    {
      "entropy": 5.523245239257813,
      "epoch": 6.231508165225744,
      "grad_norm": 1.375,
      "learning_rate": 0.00019247041955178763,
      "loss": 4.6207,
      "mean_token_accuracy": 0.27424629032611847,
      "num_tokens": 148729241.0,
      "step": 64870
    },
    {
      "entropy": 5.503736734390259,
      "epoch": 6.231988472622478,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019243852649916137,
      "loss": 4.6087,
      "mean_token_accuracy": 0.27493046075105665,
      "num_tokens": 148740342.0,
      "step": 64875
    },
    {
      "entropy": 5.515902853012085,
      "epoch": 6.232468780019213,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019240663536336733,
      "loss": 4.6304,
      "mean_token_accuracy": 0.2647357195615768,
      "num_tokens": 148751411.0,
      "step": 64880
    },
    {
      "entropy": 5.531299781799317,
      "epoch": 6.232949087415946,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000192374746145146,
      "loss": 4.6014,
      "mean_token_accuracy": 0.2698360249400139,
      "num_tokens": 148763383.0,
      "step": 64885
    },
    {
      "entropy": 5.518218374252319,
      "epoch": 6.23342939481268,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019234285884523765,
      "loss": 4.6088,
      "mean_token_accuracy": 0.26510490775108336,
      "num_tokens": 148773900.0,
      "step": 64890
    },
    {
      "entropy": 5.569363212585449,
      "epoch": 6.233909702209414,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001923109734643828,
      "loss": 4.6679,
      "mean_token_accuracy": 0.2635533109307289,
      "num_tokens": 148784749.0,
      "step": 64895
    },
    {
      "entropy": 5.544369745254516,
      "epoch": 6.234390009606148,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019227909000332156,
      "loss": 4.7094,
      "mean_token_accuracy": 0.2607336759567261,
      "num_tokens": 148796273.0,
      "step": 64900
    },
    {
      "entropy": 5.507138872146607,
      "epoch": 6.234870317002882,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001922472084627942,
      "loss": 4.6175,
      "mean_token_accuracy": 0.2618619933724403,
      "num_tokens": 148807441.0,
      "step": 64905
    },
    {
      "entropy": 5.559746789932251,
      "epoch": 6.235350624399616,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019221532884354091,
      "loss": 4.6252,
      "mean_token_accuracy": 0.2673623874783516,
      "num_tokens": 148819622.0,
      "step": 64910
    },
    {
      "entropy": 5.5147926807403564,
      "epoch": 6.235830931796349,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019218345114630186,
      "loss": 4.6074,
      "mean_token_accuracy": 0.26034362614154816,
      "num_tokens": 148831368.0,
      "step": 64915
    },
    {
      "entropy": 5.56548171043396,
      "epoch": 6.236311239193084,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019215157537181722,
      "loss": 4.7164,
      "mean_token_accuracy": 0.2572261393070221,
      "num_tokens": 148843500.0,
      "step": 64920
    },
    {
      "entropy": 5.4976531028747555,
      "epoch": 6.236791546589817,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019211970152082703,
      "loss": 4.5788,
      "mean_token_accuracy": 0.27064095586538317,
      "num_tokens": 148855219.0,
      "step": 64925
    },
    {
      "entropy": 5.6055854797363285,
      "epoch": 6.2372718539865515,
      "grad_norm": 1.5,
      "learning_rate": 0.00019208782959407117,
      "loss": 4.7377,
      "mean_token_accuracy": 0.2629585936665535,
      "num_tokens": 148866021.0,
      "step": 64930
    },
    {
      "entropy": 5.530796194076538,
      "epoch": 6.237752161383285,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019205595959228972,
      "loss": 4.626,
      "mean_token_accuracy": 0.26474661082029344,
      "num_tokens": 148878283.0,
      "step": 64935
    },
    {
      "entropy": 5.4752116203308105,
      "epoch": 6.238232468780019,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019202409151622264,
      "loss": 4.5337,
      "mean_token_accuracy": 0.27224806100130083,
      "num_tokens": 148889443.0,
      "step": 64940
    },
    {
      "entropy": 5.546534633636474,
      "epoch": 6.238712776176753,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019199222536660972,
      "loss": 4.6812,
      "mean_token_accuracy": 0.2627295583486557,
      "num_tokens": 148901683.0,
      "step": 64945
    },
    {
      "entropy": 5.584082889556885,
      "epoch": 6.239193083573487,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019196036114419086,
      "loss": 4.6655,
      "mean_token_accuracy": 0.26436917632818224,
      "num_tokens": 148913558.0,
      "step": 64950
    },
    {
      "entropy": 5.531363010406494,
      "epoch": 6.239673390970221,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001919284988497059,
      "loss": 4.6273,
      "mean_token_accuracy": 0.2680807501077652,
      "num_tokens": 148925594.0,
      "step": 64955
    },
    {
      "entropy": 5.5920463562011715,
      "epoch": 6.240153698366955,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019189663848389444,
      "loss": 4.6845,
      "mean_token_accuracy": 0.26195879876613615,
      "num_tokens": 148936921.0,
      "step": 64960
    },
    {
      "entropy": 5.625844430923462,
      "epoch": 6.240634005763689,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001918647800474963,
      "loss": 4.6534,
      "mean_token_accuracy": 0.2648466780781746,
      "num_tokens": 148948415.0,
      "step": 64965
    },
    {
      "entropy": 5.5609596252441404,
      "epoch": 6.2411143131604225,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019183292354125109,
      "loss": 4.6978,
      "mean_token_accuracy": 0.25554660707712173,
      "num_tokens": 148960081.0,
      "step": 64970
    },
    {
      "entropy": 5.5505694389343265,
      "epoch": 6.241594620557157,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019180106896589844,
      "loss": 4.6558,
      "mean_token_accuracy": 0.26611038744449617,
      "num_tokens": 148972307.0,
      "step": 64975
    },
    {
      "entropy": 5.541330766677857,
      "epoch": 6.24207492795389,
      "grad_norm": 1.375,
      "learning_rate": 0.00019176921632217792,
      "loss": 4.6435,
      "mean_token_accuracy": 0.26451031267642977,
      "num_tokens": 148984479.0,
      "step": 64980
    },
    {
      "entropy": 5.5801324367523195,
      "epoch": 6.242555235350625,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019173736561082917,
      "loss": 4.6561,
      "mean_token_accuracy": 0.2597335621714592,
      "num_tokens": 148998111.0,
      "step": 64985
    },
    {
      "entropy": 5.54150619506836,
      "epoch": 6.243035542747358,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019170551683259142,
      "loss": 4.6609,
      "mean_token_accuracy": 0.2603402078151703,
      "num_tokens": 149009851.0,
      "step": 64990
    },
    {
      "entropy": 5.662200117111206,
      "epoch": 6.243515850144092,
      "grad_norm": 1.640625,
      "learning_rate": 0.00019167366998820425,
      "loss": 4.6624,
      "mean_token_accuracy": 0.26222196221351624,
      "num_tokens": 149020313.0,
      "step": 64995
    },
    {
      "entropy": 5.5059630393981935,
      "epoch": 6.243996157540826,
      "grad_norm": 1.34375,
      "learning_rate": 0.000191641825078407,
      "loss": 4.6028,
      "mean_token_accuracy": 0.27162969261407854,
      "num_tokens": 149032065.0,
      "step": 65000
    },
    {
      "entropy": 5.549830436706543,
      "epoch": 6.24447646493756,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001916099821039391,
      "loss": 4.6198,
      "mean_token_accuracy": 0.2709179311990738,
      "num_tokens": 149043952.0,
      "step": 65005
    },
    {
      "entropy": 5.545174980163575,
      "epoch": 6.244956772334294,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019157814106553978,
      "loss": 4.694,
      "mean_token_accuracy": 0.2632328301668167,
      "num_tokens": 149055939.0,
      "step": 65010
    },
    {
      "entropy": 5.511519479751587,
      "epoch": 6.245437079731028,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001915463019639484,
      "loss": 4.5874,
      "mean_token_accuracy": 0.26472805738449096,
      "num_tokens": 149067101.0,
      "step": 65015
    },
    {
      "entropy": 5.488394165039063,
      "epoch": 6.245917387127761,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019151446479990397,
      "loss": 4.5345,
      "mean_token_accuracy": 0.2770597696304321,
      "num_tokens": 149077688.0,
      "step": 65020
    },
    {
      "entropy": 5.448996639251709,
      "epoch": 6.246397694524496,
      "grad_norm": 1.375,
      "learning_rate": 0.0001914826295741458,
      "loss": 4.5615,
      "mean_token_accuracy": 0.2742885693907738,
      "num_tokens": 149089558.0,
      "step": 65025
    },
    {
      "entropy": 5.5017658233642575,
      "epoch": 6.24687800192123,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019145079628741295,
      "loss": 4.6498,
      "mean_token_accuracy": 0.2609818711876869,
      "num_tokens": 149100052.0,
      "step": 65030
    },
    {
      "entropy": 5.557335281372071,
      "epoch": 6.247358309317963,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019141896494044458,
      "loss": 4.6262,
      "mean_token_accuracy": 0.2642430305480957,
      "num_tokens": 149110998.0,
      "step": 65035
    },
    {
      "entropy": 5.584568643569947,
      "epoch": 6.247838616714698,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019138713553397958,
      "loss": 4.6976,
      "mean_token_accuracy": 0.2563799723982811,
      "num_tokens": 149122772.0,
      "step": 65040
    },
    {
      "entropy": 5.67080512046814,
      "epoch": 6.248318924111431,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019135530806875712,
      "loss": 4.7633,
      "mean_token_accuracy": 0.25562081336975095,
      "num_tokens": 149133916.0,
      "step": 65045
    },
    {
      "entropy": 5.473799562454223,
      "epoch": 6.2487992315081655,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019132348254551597,
      "loss": 4.5528,
      "mean_token_accuracy": 0.2619148790836334,
      "num_tokens": 149144909.0,
      "step": 65050
    },
    {
      "entropy": 5.469046306610108,
      "epoch": 6.249279538904899,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019129165896499513,
      "loss": 4.5319,
      "mean_token_accuracy": 0.27823989987373354,
      "num_tokens": 149156157.0,
      "step": 65055
    },
    {
      "entropy": 5.508300065994263,
      "epoch": 6.249759846301633,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019125983732793327,
      "loss": 4.6706,
      "mean_token_accuracy": 0.2614768370985985,
      "num_tokens": 149169897.0,
      "step": 65060
    },
    {
      "entropy": 5.498487186431885,
      "epoch": 6.250240153698367,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019122801763506948,
      "loss": 4.6283,
      "mean_token_accuracy": 0.263319531083107,
      "num_tokens": 149181458.0,
      "step": 65065
    },
    {
      "entropy": 5.495869398117065,
      "epoch": 6.250720461095101,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019119619988714232,
      "loss": 4.5682,
      "mean_token_accuracy": 0.280259470641613,
      "num_tokens": 149192590.0,
      "step": 65070
    },
    {
      "entropy": 5.519107151031494,
      "epoch": 6.2512007684918345,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019116438408489056,
      "loss": 4.5825,
      "mean_token_accuracy": 0.2604526177048683,
      "num_tokens": 149203785.0,
      "step": 65075
    },
    {
      "entropy": 5.551033210754395,
      "epoch": 6.251681075888569,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019113257022905283,
      "loss": 4.6745,
      "mean_token_accuracy": 0.26161772608757017,
      "num_tokens": 149215735.0,
      "step": 65080
    },
    {
      "entropy": 5.595971345901489,
      "epoch": 6.252161383285302,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019110075832036777,
      "loss": 4.6391,
      "mean_token_accuracy": 0.26292767375707626,
      "num_tokens": 149225944.0,
      "step": 65085
    },
    {
      "entropy": 5.540692663192749,
      "epoch": 6.252641690682037,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019106894835957405,
      "loss": 4.6121,
      "mean_token_accuracy": 0.2635891482234001,
      "num_tokens": 149236585.0,
      "step": 65090
    },
    {
      "entropy": 5.5710631847381595,
      "epoch": 6.25312199807877,
      "grad_norm": 1.375,
      "learning_rate": 0.00019103714034741004,
      "loss": 4.6768,
      "mean_token_accuracy": 0.26164509952068327,
      "num_tokens": 149248702.0,
      "step": 65095
    },
    {
      "entropy": 5.572843360900879,
      "epoch": 6.253602305475504,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019100533428461437,
      "loss": 4.6891,
      "mean_token_accuracy": 0.25634372383356097,
      "num_tokens": 149259871.0,
      "step": 65100
    },
    {
      "entropy": 5.498288679122925,
      "epoch": 6.254082612872239,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019097353017192542,
      "loss": 4.5889,
      "mean_token_accuracy": 0.2655647799372673,
      "num_tokens": 149272005.0,
      "step": 65105
    },
    {
      "entropy": 5.567382001876831,
      "epoch": 6.254562920268972,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019094172801008158,
      "loss": 4.6993,
      "mean_token_accuracy": 0.25863397866487503,
      "num_tokens": 149283648.0,
      "step": 65110
    },
    {
      "entropy": 5.560844945907593,
      "epoch": 6.255043227665706,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001909099277998212,
      "loss": 4.6258,
      "mean_token_accuracy": 0.26332270801067353,
      "num_tokens": 149294692.0,
      "step": 65115
    },
    {
      "entropy": 5.616438150405884,
      "epoch": 6.25552353506244,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001908781295418826,
      "loss": 4.7043,
      "mean_token_accuracy": 0.25944981873035433,
      "num_tokens": 149306719.0,
      "step": 65120
    },
    {
      "entropy": 5.505355358123779,
      "epoch": 6.256003842459174,
      "grad_norm": 1.375,
      "learning_rate": 0.00019084633323700407,
      "loss": 4.527,
      "mean_token_accuracy": 0.2742248222231865,
      "num_tokens": 149317859.0,
      "step": 65125
    },
    {
      "entropy": 5.529403495788574,
      "epoch": 6.256484149855908,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019081453888592377,
      "loss": 4.6535,
      "mean_token_accuracy": 0.25697475373744966,
      "num_tokens": 149329449.0,
      "step": 65130
    },
    {
      "entropy": 5.538867521286011,
      "epoch": 6.256964457252642,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019078274648937993,
      "loss": 4.6293,
      "mean_token_accuracy": 0.2605503275990486,
      "num_tokens": 149340307.0,
      "step": 65135
    },
    {
      "entropy": 5.469790601730347,
      "epoch": 6.257444764649375,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019075095604811066,
      "loss": 4.5601,
      "mean_token_accuracy": 0.27011622935533525,
      "num_tokens": 149351896.0,
      "step": 65140
    },
    {
      "entropy": 5.552295351028443,
      "epoch": 6.25792507204611,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019071916756285395,
      "loss": 4.5953,
      "mean_token_accuracy": 0.2678973823785782,
      "num_tokens": 149363002.0,
      "step": 65145
    },
    {
      "entropy": 5.496552324295044,
      "epoch": 6.258405379442843,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00019068738103434797,
      "loss": 4.5985,
      "mean_token_accuracy": 0.26755171120166776,
      "num_tokens": 149373498.0,
      "step": 65150
    },
    {
      "entropy": 5.587796878814697,
      "epoch": 6.2588856868395775,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019065559646333057,
      "loss": 4.6473,
      "mean_token_accuracy": 0.2644110471010208,
      "num_tokens": 149385342.0,
      "step": 65155
    },
    {
      "entropy": 5.546698141098022,
      "epoch": 6.259365994236311,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019062381385053978,
      "loss": 4.6454,
      "mean_token_accuracy": 0.2621238261461258,
      "num_tokens": 149396749.0,
      "step": 65160
    },
    {
      "entropy": 5.5421384334564205,
      "epoch": 6.259846301633045,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019059203319671364,
      "loss": 4.6637,
      "mean_token_accuracy": 0.2625812187790871,
      "num_tokens": 149408742.0,
      "step": 65165
    },
    {
      "entropy": 5.47341685295105,
      "epoch": 6.260326609029779,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019056025450258968,
      "loss": 4.5502,
      "mean_token_accuracy": 0.2761399209499359,
      "num_tokens": 149420274.0,
      "step": 65170
    },
    {
      "entropy": 5.591525888442993,
      "epoch": 6.260806916426513,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019052847776890593,
      "loss": 4.6581,
      "mean_token_accuracy": 0.25317687839269637,
      "num_tokens": 149431631.0,
      "step": 65175
    },
    {
      "entropy": 5.542814254760742,
      "epoch": 6.261287223823247,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001904967029964001,
      "loss": 4.6257,
      "mean_token_accuracy": 0.26749965250492097,
      "num_tokens": 149442419.0,
      "step": 65180
    },
    {
      "entropy": 5.505303382873535,
      "epoch": 6.261767531219981,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001904649301858099,
      "loss": 4.6205,
      "mean_token_accuracy": 0.2627886667847633,
      "num_tokens": 149454415.0,
      "step": 65185
    },
    {
      "entropy": 5.63346004486084,
      "epoch": 6.262247838616715,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019043315933787297,
      "loss": 4.7131,
      "mean_token_accuracy": 0.25782914012670516,
      "num_tokens": 149466282.0,
      "step": 65190
    },
    {
      "entropy": 5.527612209320068,
      "epoch": 6.2627281460134485,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019040139045332704,
      "loss": 4.6372,
      "mean_token_accuracy": 0.266141252219677,
      "num_tokens": 149478636.0,
      "step": 65195
    },
    {
      "entropy": 5.535066556930542,
      "epoch": 6.263208453410183,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001903696235329096,
      "loss": 4.6428,
      "mean_token_accuracy": 0.26456328779459,
      "num_tokens": 149489519.0,
      "step": 65200
    },
    {
      "entropy": 5.508727169036865,
      "epoch": 6.263688760806916,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001903378585773582,
      "loss": 4.6085,
      "mean_token_accuracy": 0.270836178958416,
      "num_tokens": 149499914.0,
      "step": 65205
    },
    {
      "entropy": 5.612763214111328,
      "epoch": 6.264169068203651,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019030609558741025,
      "loss": 4.7119,
      "mean_token_accuracy": 0.26138725578784944,
      "num_tokens": 149511225.0,
      "step": 65210
    },
    {
      "entropy": 5.541930055618286,
      "epoch": 6.264649375600384,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019027433456380333,
      "loss": 4.6353,
      "mean_token_accuracy": 0.2708174377679825,
      "num_tokens": 149522837.0,
      "step": 65215
    },
    {
      "entropy": 5.559955835342407,
      "epoch": 6.265129682997118,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019024257550727476,
      "loss": 4.6909,
      "mean_token_accuracy": 0.2600936084985733,
      "num_tokens": 149534178.0,
      "step": 65220
    },
    {
      "entropy": 5.533206653594971,
      "epoch": 6.265609990393852,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019021081841856202,
      "loss": 4.6107,
      "mean_token_accuracy": 0.2727965459227562,
      "num_tokens": 149544946.0,
      "step": 65225
    },
    {
      "entropy": 5.540114974975586,
      "epoch": 6.266090297790586,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019017906329840217,
      "loss": 4.6184,
      "mean_token_accuracy": 0.26456633061170576,
      "num_tokens": 149557968.0,
      "step": 65230
    },
    {
      "entropy": 5.4591978073120115,
      "epoch": 6.26657060518732,
      "grad_norm": 1.375,
      "learning_rate": 0.00019014731014753262,
      "loss": 4.5923,
      "mean_token_accuracy": 0.2693284884095192,
      "num_tokens": 149569991.0,
      "step": 65235
    },
    {
      "entropy": 5.519132947921753,
      "epoch": 6.267050912584054,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001901155589666906,
      "loss": 4.5598,
      "mean_token_accuracy": 0.2668400198221207,
      "num_tokens": 149581614.0,
      "step": 65240
    },
    {
      "entropy": 5.646364498138428,
      "epoch": 6.267531219980787,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001900838097566132,
      "loss": 4.7779,
      "mean_token_accuracy": 0.25494661182165146,
      "num_tokens": 149592887.0,
      "step": 65245
    },
    {
      "entropy": 5.607180070877075,
      "epoch": 6.268011527377522,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019005206251803764,
      "loss": 4.7386,
      "mean_token_accuracy": 0.2583717867732048,
      "num_tokens": 149604481.0,
      "step": 65250
    },
    {
      "entropy": 5.574615955352783,
      "epoch": 6.268491834774256,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019002031725170098,
      "loss": 4.6738,
      "mean_token_accuracy": 0.25581829994916916,
      "num_tokens": 149616187.0,
      "step": 65255
    },
    {
      "entropy": 5.404380893707275,
      "epoch": 6.268972142170989,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018998857395834018,
      "loss": 4.4848,
      "mean_token_accuracy": 0.278210012614727,
      "num_tokens": 149626663.0,
      "step": 65260
    },
    {
      "entropy": 5.539743423461914,
      "epoch": 6.269452449567724,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018995683263869223,
      "loss": 4.6277,
      "mean_token_accuracy": 0.26676924228668214,
      "num_tokens": 149638686.0,
      "step": 65265
    },
    {
      "entropy": 5.527234745025635,
      "epoch": 6.269932756964457,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001899250932934941,
      "loss": 4.6543,
      "mean_token_accuracy": 0.26237234473228455,
      "num_tokens": 149650941.0,
      "step": 65270
    },
    {
      "entropy": 5.5412774085998535,
      "epoch": 6.2704130643611915,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018989335592348272,
      "loss": 4.6271,
      "mean_token_accuracy": 0.271978123486042,
      "num_tokens": 149661673.0,
      "step": 65275
    },
    {
      "entropy": 5.558324384689331,
      "epoch": 6.270893371757925,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001898616205293948,
      "loss": 4.6621,
      "mean_token_accuracy": 0.2627242669463158,
      "num_tokens": 149673607.0,
      "step": 65280
    },
    {
      "entropy": 5.544007778167725,
      "epoch": 6.271373679154659,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001898298871119674,
      "loss": 4.6425,
      "mean_token_accuracy": 0.2652522727847099,
      "num_tokens": 149685213.0,
      "step": 65285
    },
    {
      "entropy": 5.515383052825928,
      "epoch": 6.271853986551393,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001897981556719371,
      "loss": 4.6283,
      "mean_token_accuracy": 0.26380297392606733,
      "num_tokens": 149696858.0,
      "step": 65290
    },
    {
      "entropy": 5.552501010894775,
      "epoch": 6.272334293948127,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018976642621004053,
      "loss": 4.6671,
      "mean_token_accuracy": 0.2594910427927971,
      "num_tokens": 149707573.0,
      "step": 65295
    },
    {
      "entropy": 5.518757963180542,
      "epoch": 6.2728146013448605,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018973469872701448,
      "loss": 4.5469,
      "mean_token_accuracy": 0.2770609676837921,
      "num_tokens": 149718696.0,
      "step": 65300
    },
    {
      "entropy": 5.460474061965942,
      "epoch": 6.273294908741595,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018970297322359552,
      "loss": 4.5806,
      "mean_token_accuracy": 0.2763232856988907,
      "num_tokens": 149731145.0,
      "step": 65305
    },
    {
      "entropy": 5.540302467346192,
      "epoch": 6.273775216138328,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018967124970052025,
      "loss": 4.6352,
      "mean_token_accuracy": 0.26192479133605956,
      "num_tokens": 149742266.0,
      "step": 65310
    },
    {
      "entropy": 5.498677587509155,
      "epoch": 6.2742555235350626,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018963952815852532,
      "loss": 4.5727,
      "mean_token_accuracy": 0.27353194206953046,
      "num_tokens": 149753418.0,
      "step": 65315
    },
    {
      "entropy": 5.591800308227539,
      "epoch": 6.274735830931796,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018960780859834692,
      "loss": 4.7178,
      "mean_token_accuracy": 0.26661486476659774,
      "num_tokens": 149766478.0,
      "step": 65320
    },
    {
      "entropy": 5.530780410766601,
      "epoch": 6.27521613832853,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001895760910207217,
      "loss": 4.6199,
      "mean_token_accuracy": 0.2683963432908058,
      "num_tokens": 149777945.0,
      "step": 65325
    },
    {
      "entropy": 5.577345180511474,
      "epoch": 6.275696445725265,
      "grad_norm": 1.375,
      "learning_rate": 0.00018954437542638598,
      "loss": 4.685,
      "mean_token_accuracy": 0.2609307661652565,
      "num_tokens": 149789547.0,
      "step": 65330
    },
    {
      "entropy": 5.529082298278809,
      "epoch": 6.276176753121998,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018951266181607606,
      "loss": 4.677,
      "mean_token_accuracy": 0.2608643889427185,
      "num_tokens": 149800813.0,
      "step": 65335
    },
    {
      "entropy": 5.55447940826416,
      "epoch": 6.276657060518732,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001894809501905283,
      "loss": 4.6628,
      "mean_token_accuracy": 0.2619768962264061,
      "num_tokens": 149810831.0,
      "step": 65340
    },
    {
      "entropy": 5.515576553344727,
      "epoch": 6.277137367915466,
      "grad_norm": 1.4375,
      "learning_rate": 0.000189449240550479,
      "loss": 4.6413,
      "mean_token_accuracy": 0.2625895872712135,
      "num_tokens": 149822487.0,
      "step": 65345
    },
    {
      "entropy": 5.551766061782837,
      "epoch": 6.2776176753122,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001894175328966643,
      "loss": 4.6746,
      "mean_token_accuracy": 0.2611315011978149,
      "num_tokens": 149834071.0,
      "step": 65350
    },
    {
      "entropy": 5.499827766418457,
      "epoch": 6.278097982708934,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001893858272298204,
      "loss": 4.6368,
      "mean_token_accuracy": 0.26575318425893785,
      "num_tokens": 149845953.0,
      "step": 65355
    },
    {
      "entropy": 5.707639169692993,
      "epoch": 6.278578290105668,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018935412355068328,
      "loss": 4.8102,
      "mean_token_accuracy": 0.25103742629289627,
      "num_tokens": 149858671.0,
      "step": 65360
    },
    {
      "entropy": 5.623979806900024,
      "epoch": 6.279058597502401,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001893224218599891,
      "loss": 4.7639,
      "mean_token_accuracy": 0.2550782486796379,
      "num_tokens": 149870358.0,
      "step": 65365
    },
    {
      "entropy": 5.571062421798706,
      "epoch": 6.279538904899136,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018929072215847386,
      "loss": 4.69,
      "mean_token_accuracy": 0.25946239829063417,
      "num_tokens": 149881131.0,
      "step": 65370
    },
    {
      "entropy": 5.586054849624634,
      "epoch": 6.280019212295869,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00018925902444687364,
      "loss": 4.6612,
      "mean_token_accuracy": 0.2658997043967247,
      "num_tokens": 149891538.0,
      "step": 65375
    },
    {
      "entropy": 5.5692277431488035,
      "epoch": 6.2804995196926034,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018922732872592418,
      "loss": 4.6913,
      "mean_token_accuracy": 0.26284450143575666,
      "num_tokens": 149903520.0,
      "step": 65380
    },
    {
      "entropy": 5.525498390197754,
      "epoch": 6.280979827089337,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001891956349963615,
      "loss": 4.645,
      "mean_token_accuracy": 0.26575535982847215,
      "num_tokens": 149914888.0,
      "step": 65385
    },
    {
      "entropy": 5.597358894348145,
      "epoch": 6.281460134486071,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018916394325892133,
      "loss": 4.7232,
      "mean_token_accuracy": 0.26385351866483686,
      "num_tokens": 149925574.0,
      "step": 65390
    },
    {
      "entropy": 5.489427757263184,
      "epoch": 6.281940441882805,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001891322535143396,
      "loss": 4.5294,
      "mean_token_accuracy": 0.2701012670993805,
      "num_tokens": 149938065.0,
      "step": 65395
    },
    {
      "entropy": 5.5481775283813475,
      "epoch": 6.282420749279539,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018910056576335193,
      "loss": 4.5976,
      "mean_token_accuracy": 0.26307033449411393,
      "num_tokens": 149949793.0,
      "step": 65400
    },
    {
      "entropy": 5.488281679153443,
      "epoch": 6.282901056676272,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001890688800066941,
      "loss": 4.5801,
      "mean_token_accuracy": 0.2740341067314148,
      "num_tokens": 149960963.0,
      "step": 65405
    },
    {
      "entropy": 5.6158324718475345,
      "epoch": 6.283381364073007,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018903719624510163,
      "loss": 4.7172,
      "mean_token_accuracy": 0.2629045382142067,
      "num_tokens": 149971298.0,
      "step": 65410
    },
    {
      "entropy": 5.55624418258667,
      "epoch": 6.28386167146974,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018900551447931024,
      "loss": 4.6444,
      "mean_token_accuracy": 0.2656639352440834,
      "num_tokens": 149981546.0,
      "step": 65415
    },
    {
      "entropy": 5.650677633285523,
      "epoch": 6.2843419788664745,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018897383471005546,
      "loss": 4.7516,
      "mean_token_accuracy": 0.2549508988857269,
      "num_tokens": 149992617.0,
      "step": 65420
    },
    {
      "entropy": 5.521064376831054,
      "epoch": 6.284822286263209,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018894215693807277,
      "loss": 4.6279,
      "mean_token_accuracy": 0.2635958522558212,
      "num_tokens": 150003548.0,
      "step": 65425
    },
    {
      "entropy": 5.572253942489624,
      "epoch": 6.285302593659942,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018891048116409764,
      "loss": 4.6865,
      "mean_token_accuracy": 0.26776101142168046,
      "num_tokens": 150014859.0,
      "step": 65430
    },
    {
      "entropy": 5.5831374168396,
      "epoch": 6.285782901056677,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018887880738886566,
      "loss": 4.6496,
      "mean_token_accuracy": 0.2601312413811684,
      "num_tokens": 150025783.0,
      "step": 65435
    },
    {
      "entropy": 5.556234216690063,
      "epoch": 6.28626320845341,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001888471356131119,
      "loss": 4.6451,
      "mean_token_accuracy": 0.2662109389901161,
      "num_tokens": 150038129.0,
      "step": 65440
    },
    {
      "entropy": 5.524358654022217,
      "epoch": 6.286743515850144,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001888154658375718,
      "loss": 4.5812,
      "mean_token_accuracy": 0.27163876593112946,
      "num_tokens": 150050668.0,
      "step": 65445
    },
    {
      "entropy": 5.489668083190918,
      "epoch": 6.287223823246878,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018878379806298073,
      "loss": 4.6538,
      "mean_token_accuracy": 0.2647062614560127,
      "num_tokens": 150062331.0,
      "step": 65450
    },
    {
      "entropy": 5.635817050933838,
      "epoch": 6.287704130643612,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018875213229007387,
      "loss": 4.7748,
      "mean_token_accuracy": 0.25181578695774076,
      "num_tokens": 150075071.0,
      "step": 65455
    },
    {
      "entropy": 5.533529663085938,
      "epoch": 6.2881844380403455,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018872046851958635,
      "loss": 4.629,
      "mean_token_accuracy": 0.2683826506137848,
      "num_tokens": 150087070.0,
      "step": 65460
    },
    {
      "entropy": 5.472302961349487,
      "epoch": 6.28866474543708,
      "grad_norm": 1.59375,
      "learning_rate": 0.00018868880675225347,
      "loss": 4.5381,
      "mean_token_accuracy": 0.27283705323934554,
      "num_tokens": 150096451.0,
      "step": 65465
    },
    {
      "entropy": 5.471967029571533,
      "epoch": 6.289145052833813,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001886571469888101,
      "loss": 4.5603,
      "mean_token_accuracy": 0.26622827649116515,
      "num_tokens": 150108570.0,
      "step": 65470
    },
    {
      "entropy": 5.514820194244384,
      "epoch": 6.289625360230548,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001886254892299914,
      "loss": 4.6387,
      "mean_token_accuracy": 0.26718979626893996,
      "num_tokens": 150119563.0,
      "step": 65475
    },
    {
      "entropy": 5.499587678909302,
      "epoch": 6.290105667627281,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018859383347653231,
      "loss": 4.6322,
      "mean_token_accuracy": 0.26815918684005735,
      "num_tokens": 150131218.0,
      "step": 65480
    },
    {
      "entropy": 5.552410984039307,
      "epoch": 6.290585975024015,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018856217972916794,
      "loss": 4.655,
      "mean_token_accuracy": 0.2638369306921959,
      "num_tokens": 150142251.0,
      "step": 65485
    },
    {
      "entropy": 5.625957775115967,
      "epoch": 6.291066282420749,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018853052798863302,
      "loss": 4.7026,
      "mean_token_accuracy": 0.25910849720239637,
      "num_tokens": 150153149.0,
      "step": 65490
    },
    {
      "entropy": 5.597996425628662,
      "epoch": 6.291546589817483,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018849887825566258,
      "loss": 4.6763,
      "mean_token_accuracy": 0.25831211507320406,
      "num_tokens": 150166264.0,
      "step": 65495
    },
    {
      "entropy": 5.681907510757446,
      "epoch": 6.2920268972142175,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018846723053099118,
      "loss": 4.7936,
      "mean_token_accuracy": 0.24753608405590058,
      "num_tokens": 150177810.0,
      "step": 65500
    },
    {
      "entropy": 5.58465723991394,
      "epoch": 6.292507204610951,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018843558481535384,
      "loss": 4.6372,
      "mean_token_accuracy": 0.2618625327944756,
      "num_tokens": 150189480.0,
      "step": 65505
    },
    {
      "entropy": 5.488057804107666,
      "epoch": 6.292987512007685,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018840394110948512,
      "loss": 4.6035,
      "mean_token_accuracy": 0.26839331835508345,
      "num_tokens": 150202315.0,
      "step": 65510
    },
    {
      "entropy": 5.60984959602356,
      "epoch": 6.293467819404419,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018837229941411976,
      "loss": 4.6674,
      "mean_token_accuracy": 0.26326774805784225,
      "num_tokens": 150213290.0,
      "step": 65515
    },
    {
      "entropy": 5.494002199172973,
      "epoch": 6.293948126801153,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018834065972999237,
      "loss": 4.5883,
      "mean_token_accuracy": 0.2680045276880264,
      "num_tokens": 150226194.0,
      "step": 65520
    },
    {
      "entropy": 5.530822467803955,
      "epoch": 6.2944284341978864,
      "grad_norm": 1.375,
      "learning_rate": 0.00018830902205783768,
      "loss": 4.5888,
      "mean_token_accuracy": 0.2679983705282211,
      "num_tokens": 150237874.0,
      "step": 65525
    },
    {
      "entropy": 5.565312004089355,
      "epoch": 6.294908741594621,
      "grad_norm": 1.5,
      "learning_rate": 0.00018827738639838991,
      "loss": 4.6237,
      "mean_token_accuracy": 0.26469955295324327,
      "num_tokens": 150248904.0,
      "step": 65530
    },
    {
      "entropy": 5.562710952758789,
      "epoch": 6.295389048991354,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018824575275238375,
      "loss": 4.6402,
      "mean_token_accuracy": 0.2649236500263214,
      "num_tokens": 150260840.0,
      "step": 65535
    },
    {
      "entropy": 5.544158506393432,
      "epoch": 6.2958693563880885,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018821412112055363,
      "loss": 4.6323,
      "mean_token_accuracy": 0.2651011824607849,
      "num_tokens": 150271959.0,
      "step": 65540
    },
    {
      "entropy": 5.535789203643799,
      "epoch": 6.296349663784822,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018818249150363386,
      "loss": 4.6845,
      "mean_token_accuracy": 0.2620638832449913,
      "num_tokens": 150283603.0,
      "step": 65545
    },
    {
      "entropy": 5.543155908584595,
      "epoch": 6.296829971181556,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018815086390235885,
      "loss": 4.6696,
      "mean_token_accuracy": 0.26510629057884216,
      "num_tokens": 150295072.0,
      "step": 65550
    },
    {
      "entropy": 5.55010404586792,
      "epoch": 6.29731027857829,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018811923831746303,
      "loss": 4.6851,
      "mean_token_accuracy": 0.256780332326889,
      "num_tokens": 150305808.0,
      "step": 65555
    },
    {
      "entropy": 5.581111335754395,
      "epoch": 6.297790585975024,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001880876147496804,
      "loss": 4.6647,
      "mean_token_accuracy": 0.2688105687499046,
      "num_tokens": 150317970.0,
      "step": 65560
    },
    {
      "entropy": 5.591871356964111,
      "epoch": 6.2982708933717575,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018805599319974525,
      "loss": 4.6634,
      "mean_token_accuracy": 0.2640883758664131,
      "num_tokens": 150329414.0,
      "step": 65565
    },
    {
      "entropy": 5.657427644729614,
      "epoch": 6.298751200768492,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018802437366839182,
      "loss": 4.7911,
      "mean_token_accuracy": 0.2512430340051651,
      "num_tokens": 150341666.0,
      "step": 65570
    },
    {
      "entropy": 5.5107934951782225,
      "epoch": 6.299231508165226,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018799275615635414,
      "loss": 4.6024,
      "mean_token_accuracy": 0.261115200817585,
      "num_tokens": 150353412.0,
      "step": 65575
    },
    {
      "entropy": 5.601131439208984,
      "epoch": 6.29971181556196,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001879611406643663,
      "loss": 4.7188,
      "mean_token_accuracy": 0.2548553556203842,
      "num_tokens": 150364486.0,
      "step": 65580
    },
    {
      "entropy": 5.53441309928894,
      "epoch": 6.300192122958694,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018792952719316243,
      "loss": 4.6086,
      "mean_token_accuracy": 0.26440462917089463,
      "num_tokens": 150375553.0,
      "step": 65585
    },
    {
      "entropy": 5.523419141769409,
      "epoch": 6.300672430355427,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018789791574347627,
      "loss": 4.6609,
      "mean_token_accuracy": 0.26276606172323225,
      "num_tokens": 150388349.0,
      "step": 65590
    },
    {
      "entropy": 5.537891387939453,
      "epoch": 6.301152737752162,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018786630631604191,
      "loss": 4.5966,
      "mean_token_accuracy": 0.269854237139225,
      "num_tokens": 150399533.0,
      "step": 65595
    },
    {
      "entropy": 5.463969373703003,
      "epoch": 6.301633045148895,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018783469891159316,
      "loss": 4.583,
      "mean_token_accuracy": 0.2688867062330246,
      "num_tokens": 150411526.0,
      "step": 65600
    },
    {
      "entropy": 5.430668592453003,
      "epoch": 6.302113352545629,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018780309353086388,
      "loss": 4.5802,
      "mean_token_accuracy": 0.2734509915113449,
      "num_tokens": 150423727.0,
      "step": 65605
    },
    {
      "entropy": 5.487943983078003,
      "epoch": 6.302593659942363,
      "grad_norm": 1.5,
      "learning_rate": 0.00018777149017458784,
      "loss": 4.583,
      "mean_token_accuracy": 0.2709418684244156,
      "num_tokens": 150434269.0,
      "step": 65610
    },
    {
      "entropy": 5.500328922271729,
      "epoch": 6.303073967339097,
      "grad_norm": 1.375,
      "learning_rate": 0.0001877398888434988,
      "loss": 4.616,
      "mean_token_accuracy": 0.26401911228895186,
      "num_tokens": 150446362.0,
      "step": 65615
    },
    {
      "entropy": 5.564226579666138,
      "epoch": 6.303554274735831,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018770828953833046,
      "loss": 4.6712,
      "mean_token_accuracy": 0.2592402949929237,
      "num_tokens": 150458865.0,
      "step": 65620
    },
    {
      "entropy": 5.530987024307251,
      "epoch": 6.304034582132565,
      "grad_norm": 1.5,
      "learning_rate": 0.00018767669225981638,
      "loss": 4.5892,
      "mean_token_accuracy": 0.2681091740727425,
      "num_tokens": 150470650.0,
      "step": 65625
    },
    {
      "entropy": 5.552479219436646,
      "epoch": 6.304514889529298,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001876450970086902,
      "loss": 4.6442,
      "mean_token_accuracy": 0.26872446238994596,
      "num_tokens": 150483745.0,
      "step": 65630
    },
    {
      "entropy": 5.496347427368164,
      "epoch": 6.304995196926033,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018761350378568547,
      "loss": 4.5773,
      "mean_token_accuracy": 0.2657984733581543,
      "num_tokens": 150494843.0,
      "step": 65635
    },
    {
      "entropy": 5.519354152679443,
      "epoch": 6.305475504322766,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018758191259153574,
      "loss": 4.6672,
      "mean_token_accuracy": 0.26067504584789275,
      "num_tokens": 150506576.0,
      "step": 65640
    },
    {
      "entropy": 5.447332334518433,
      "epoch": 6.3059558117195005,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018755032342697447,
      "loss": 4.5629,
      "mean_token_accuracy": 0.27248657792806624,
      "num_tokens": 150518045.0,
      "step": 65645
    },
    {
      "entropy": 5.477085781097412,
      "epoch": 6.306436119116235,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018751873629273497,
      "loss": 4.5275,
      "mean_token_accuracy": 0.27341141402721403,
      "num_tokens": 150528931.0,
      "step": 65650
    },
    {
      "entropy": 5.534451246261597,
      "epoch": 6.306916426512968,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018748715118955061,
      "loss": 4.6613,
      "mean_token_accuracy": 0.2661794498562813,
      "num_tokens": 150540487.0,
      "step": 65655
    },
    {
      "entropy": 5.585462379455566,
      "epoch": 6.307396733909703,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001874555681181548,
      "loss": 4.6671,
      "mean_token_accuracy": 0.2588772550225258,
      "num_tokens": 150551047.0,
      "step": 65660
    },
    {
      "entropy": 5.545774126052857,
      "epoch": 6.307877041306436,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018742398707928066,
      "loss": 4.6172,
      "mean_token_accuracy": 0.26971552073955535,
      "num_tokens": 150561743.0,
      "step": 65665
    },
    {
      "entropy": 5.570072317123413,
      "epoch": 6.30835734870317,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001873924080736616,
      "loss": 4.7116,
      "mean_token_accuracy": 0.2561634972691536,
      "num_tokens": 150574508.0,
      "step": 65670
    },
    {
      "entropy": 5.527748250961304,
      "epoch": 6.308837656099904,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018736083110203062,
      "loss": 4.6704,
      "mean_token_accuracy": 0.2617662400007248,
      "num_tokens": 150586417.0,
      "step": 65675
    },
    {
      "entropy": 5.598165607452392,
      "epoch": 6.309317963496638,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018732925616512093,
      "loss": 4.6294,
      "mean_token_accuracy": 0.2662899434566498,
      "num_tokens": 150598340.0,
      "step": 65680
    },
    {
      "entropy": 5.615917444229126,
      "epoch": 6.3097982708933715,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018729768326366554,
      "loss": 4.7402,
      "mean_token_accuracy": 0.24652252942323685,
      "num_tokens": 150610088.0,
      "step": 65685
    },
    {
      "entropy": 5.566122102737427,
      "epoch": 6.310278578290106,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018726611239839753,
      "loss": 4.6458,
      "mean_token_accuracy": 0.27017616778612136,
      "num_tokens": 150620727.0,
      "step": 65690
    },
    {
      "entropy": 5.479802656173706,
      "epoch": 6.310758885686839,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018723454357004988,
      "loss": 4.5538,
      "mean_token_accuracy": 0.2684061169624329,
      "num_tokens": 150632145.0,
      "step": 65695
    },
    {
      "entropy": 5.619988632202149,
      "epoch": 6.311239193083574,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018720297677935556,
      "loss": 4.7288,
      "mean_token_accuracy": 0.2573360338807106,
      "num_tokens": 150642625.0,
      "step": 65700
    },
    {
      "entropy": 5.569182395935059,
      "epoch": 6.311719500480307,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018717141202704746,
      "loss": 4.6595,
      "mean_token_accuracy": 0.2605747073888779,
      "num_tokens": 150654028.0,
      "step": 65705
    },
    {
      "entropy": 5.5222502708435055,
      "epoch": 6.312199807877041,
      "grad_norm": 1.5,
      "learning_rate": 0.00018713984931385824,
      "loss": 4.6555,
      "mean_token_accuracy": 0.2662147685885429,
      "num_tokens": 150665214.0,
      "step": 65710
    },
    {
      "entropy": 5.602643156051636,
      "epoch": 6.312680115273775,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018710828864052092,
      "loss": 4.6955,
      "mean_token_accuracy": 0.26406390219926834,
      "num_tokens": 150676512.0,
      "step": 65715
    },
    {
      "entropy": 5.5720282077789305,
      "epoch": 6.313160422670509,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001870767300077681,
      "loss": 4.6563,
      "mean_token_accuracy": 0.2602775260806084,
      "num_tokens": 150687410.0,
      "step": 65720
    },
    {
      "entropy": 5.582530784606933,
      "epoch": 6.3136407300672435,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018704517341633253,
      "loss": 4.6712,
      "mean_token_accuracy": 0.2545044243335724,
      "num_tokens": 150699743.0,
      "step": 65725
    },
    {
      "entropy": 5.474296283721924,
      "epoch": 6.314121037463977,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001870136188669469,
      "loss": 4.5858,
      "mean_token_accuracy": 0.2674019306898117,
      "num_tokens": 150711715.0,
      "step": 65730
    },
    {
      "entropy": 5.5945628643035885,
      "epoch": 6.314601344860711,
      "grad_norm": 1.375,
      "learning_rate": 0.00018698206636034388,
      "loss": 4.6445,
      "mean_token_accuracy": 0.26586409360170365,
      "num_tokens": 150722264.0,
      "step": 65735
    },
    {
      "entropy": 5.547099494934082,
      "epoch": 6.315081652257445,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018695051589725576,
      "loss": 4.6173,
      "mean_token_accuracy": 0.2706831723451614,
      "num_tokens": 150734123.0,
      "step": 65740
    },
    {
      "entropy": 5.569935274124146,
      "epoch": 6.315561959654179,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001869189674784152,
      "loss": 4.631,
      "mean_token_accuracy": 0.2656416058540344,
      "num_tokens": 150744137.0,
      "step": 65745
    },
    {
      "entropy": 5.596735620498658,
      "epoch": 6.316042267050912,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018688742110455477,
      "loss": 4.6836,
      "mean_token_accuracy": 0.26919974237680433,
      "num_tokens": 150756580.0,
      "step": 65750
    },
    {
      "entropy": 5.600152397155762,
      "epoch": 6.316522574447647,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001868558767764067,
      "loss": 4.6715,
      "mean_token_accuracy": 0.25748559832572937,
      "num_tokens": 150766650.0,
      "step": 65755
    },
    {
      "entropy": 5.598894929885864,
      "epoch": 6.31700288184438,
      "grad_norm": 1.5,
      "learning_rate": 0.00018682433449470343,
      "loss": 4.7469,
      "mean_token_accuracy": 0.25879001021385195,
      "num_tokens": 150777984.0,
      "step": 65760
    },
    {
      "entropy": 5.556213665008545,
      "epoch": 6.3174831892411145,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018679279426017737,
      "loss": 4.7015,
      "mean_token_accuracy": 0.2603309601545334,
      "num_tokens": 150790891.0,
      "step": 65765
    },
    {
      "entropy": 5.501380825042725,
      "epoch": 6.317963496637848,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018676125607356065,
      "loss": 4.5536,
      "mean_token_accuracy": 0.27430815249681473,
      "num_tokens": 150801473.0,
      "step": 65770
    },
    {
      "entropy": 5.510186624526978,
      "epoch": 6.318443804034582,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018672971993558557,
      "loss": 4.646,
      "mean_token_accuracy": 0.2606385603547096,
      "num_tokens": 150812224.0,
      "step": 65775
    },
    {
      "entropy": 5.479627323150635,
      "epoch": 6.318924111431316,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001866981858469842,
      "loss": 4.632,
      "mean_token_accuracy": 0.271913056075573,
      "num_tokens": 150823930.0,
      "step": 65780
    },
    {
      "entropy": 5.497575569152832,
      "epoch": 6.31940441882805,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018666665380848884,
      "loss": 4.5683,
      "mean_token_accuracy": 0.27538695335388186,
      "num_tokens": 150836000.0,
      "step": 65785
    },
    {
      "entropy": 5.5366991519927975,
      "epoch": 6.3198847262247835,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018663512382083148,
      "loss": 4.6103,
      "mean_token_accuracy": 0.2628632396459579,
      "num_tokens": 150847542.0,
      "step": 65790
    },
    {
      "entropy": 5.6604691505432125,
      "epoch": 6.320365033621518,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001866035958847442,
      "loss": 4.7848,
      "mean_token_accuracy": 0.25497095584869384,
      "num_tokens": 150860242.0,
      "step": 65795
    },
    {
      "entropy": 5.565968227386475,
      "epoch": 6.320845341018252,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001865720700009589,
      "loss": 4.6528,
      "mean_token_accuracy": 0.25904890298843386,
      "num_tokens": 150872399.0,
      "step": 65800
    },
    {
      "entropy": 5.52213454246521,
      "epoch": 6.321325648414986,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001865405461702075,
      "loss": 4.6424,
      "mean_token_accuracy": 0.26304813027381896,
      "num_tokens": 150884459.0,
      "step": 65805
    },
    {
      "entropy": 5.483789110183716,
      "epoch": 6.32180595581172,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000186509024393222,
      "loss": 4.5715,
      "mean_token_accuracy": 0.27618183344602587,
      "num_tokens": 150895861.0,
      "step": 65810
    },
    {
      "entropy": 5.547307443618775,
      "epoch": 6.322286263208453,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001864775046707341,
      "loss": 4.6312,
      "mean_token_accuracy": 0.2644404023885727,
      "num_tokens": 150907165.0,
      "step": 65815
    },
    {
      "entropy": 5.593939638137817,
      "epoch": 6.322766570605188,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018644598700347576,
      "loss": 4.6763,
      "mean_token_accuracy": 0.26487301141023634,
      "num_tokens": 150918774.0,
      "step": 65820
    },
    {
      "entropy": 5.5820701122283936,
      "epoch": 6.323246878001921,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018641447139217877,
      "loss": 4.6189,
      "mean_token_accuracy": 0.2669000059366226,
      "num_tokens": 150929695.0,
      "step": 65825
    },
    {
      "entropy": 5.4480788230896,
      "epoch": 6.323727185398655,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018638295783757452,
      "loss": 4.528,
      "mean_token_accuracy": 0.2809994757175446,
      "num_tokens": 150940830.0,
      "step": 65830
    },
    {
      "entropy": 5.465711832046509,
      "epoch": 6.324207492795389,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018635144634039492,
      "loss": 4.595,
      "mean_token_accuracy": 0.26116946190595625,
      "num_tokens": 150952097.0,
      "step": 65835
    },
    {
      "entropy": 5.638252878189087,
      "epoch": 6.324687800192123,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001863199369013715,
      "loss": 4.7529,
      "mean_token_accuracy": 0.2489137515425682,
      "num_tokens": 150963687.0,
      "step": 65840
    },
    {
      "entropy": 5.457390928268433,
      "epoch": 6.325168107588857,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001862884295212358,
      "loss": 4.5692,
      "mean_token_accuracy": 0.2644383326172829,
      "num_tokens": 150975433.0,
      "step": 65845
    },
    {
      "entropy": 5.479865598678589,
      "epoch": 6.325648414985591,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018625692420071939,
      "loss": 4.5713,
      "mean_token_accuracy": 0.2729009583592415,
      "num_tokens": 150986135.0,
      "step": 65850
    },
    {
      "entropy": 5.514183759689331,
      "epoch": 6.326128722382324,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001862254209405536,
      "loss": 4.6261,
      "mean_token_accuracy": 0.26098700016736986,
      "num_tokens": 150996696.0,
      "step": 65855
    },
    {
      "entropy": 5.5136878967285154,
      "epoch": 6.326609029779059,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001861939197414701,
      "loss": 4.5962,
      "mean_token_accuracy": 0.2692252740263939,
      "num_tokens": 151008001.0,
      "step": 65860
    },
    {
      "entropy": 5.531649875640869,
      "epoch": 6.327089337175792,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018616242060419996,
      "loss": 4.637,
      "mean_token_accuracy": 0.2588507801294327,
      "num_tokens": 151018474.0,
      "step": 65865
    },
    {
      "entropy": 5.498845767974854,
      "epoch": 6.3275696445725265,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018613092352947464,
      "loss": 4.5991,
      "mean_token_accuracy": 0.27568810880184175,
      "num_tokens": 151030588.0,
      "step": 65870
    },
    {
      "entropy": 5.5084991455078125,
      "epoch": 6.328049951969261,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018609942851802535,
      "loss": 4.592,
      "mean_token_accuracy": 0.273893740773201,
      "num_tokens": 151040929.0,
      "step": 65875
    },
    {
      "entropy": 5.504638385772705,
      "epoch": 6.328530259365994,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018606793557058337,
      "loss": 4.6262,
      "mean_token_accuracy": 0.26678169667720797,
      "num_tokens": 151052816.0,
      "step": 65880
    },
    {
      "entropy": 5.478301668167115,
      "epoch": 6.3290105667627286,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018603644468787988,
      "loss": 4.5463,
      "mean_token_accuracy": 0.2771549135446548,
      "num_tokens": 151063803.0,
      "step": 65885
    },
    {
      "entropy": 5.5136744499206545,
      "epoch": 6.329490874159462,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018600495587064604,
      "loss": 4.6274,
      "mean_token_accuracy": 0.26327747404575347,
      "num_tokens": 151075260.0,
      "step": 65890
    },
    {
      "entropy": 5.53675332069397,
      "epoch": 6.329971181556196,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018597346911961276,
      "loss": 4.6706,
      "mean_token_accuracy": 0.2611535206437111,
      "num_tokens": 151085350.0,
      "step": 65895
    },
    {
      "entropy": 5.553715944290161,
      "epoch": 6.33045148895293,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001859419844355112,
      "loss": 4.6943,
      "mean_token_accuracy": 0.2609265372157097,
      "num_tokens": 151097250.0,
      "step": 65900
    },
    {
      "entropy": 5.49323844909668,
      "epoch": 6.330931796349664,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001859105018190723,
      "loss": 4.6106,
      "mean_token_accuracy": 0.2666540414094925,
      "num_tokens": 151108558.0,
      "step": 65905
    },
    {
      "entropy": 5.486723709106445,
      "epoch": 6.3314121037463975,
      "grad_norm": 1.375,
      "learning_rate": 0.00018587902127102703,
      "loss": 4.5843,
      "mean_token_accuracy": 0.27299043983221055,
      "num_tokens": 151121027.0,
      "step": 65910
    },
    {
      "entropy": 5.501684761047363,
      "epoch": 6.331892411143132,
      "grad_norm": 1.25,
      "learning_rate": 0.00018584754279210618,
      "loss": 4.5457,
      "mean_token_accuracy": 0.2725046396255493,
      "num_tokens": 151132845.0,
      "step": 65915
    },
    {
      "entropy": 5.5947850227355955,
      "epoch": 6.332372718539865,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018581606638304077,
      "loss": 4.7298,
      "mean_token_accuracy": 0.24804289788007736,
      "num_tokens": 151144621.0,
      "step": 65920
    },
    {
      "entropy": 5.612072992324829,
      "epoch": 6.3328530259366,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018578459204456144,
      "loss": 4.7269,
      "mean_token_accuracy": 0.2531360313296318,
      "num_tokens": 151155496.0,
      "step": 65925
    },
    {
      "entropy": 5.587067270278931,
      "epoch": 6.333333333333333,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018575311977739895,
      "loss": 4.6331,
      "mean_token_accuracy": 0.2712651893496513,
      "num_tokens": 151166332.0,
      "step": 65930
    },
    {
      "entropy": 5.616136837005615,
      "epoch": 6.333813640730067,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018572164958228406,
      "loss": 4.7723,
      "mean_token_accuracy": 0.2582621186971664,
      "num_tokens": 151178929.0,
      "step": 65935
    },
    {
      "entropy": 5.564432573318482,
      "epoch": 6.334293948126801,
      "grad_norm": 1.375,
      "learning_rate": 0.00018569018145994727,
      "loss": 4.6478,
      "mean_token_accuracy": 0.26338449865579605,
      "num_tokens": 151190054.0,
      "step": 65940
    },
    {
      "entropy": 5.4544782638549805,
      "epoch": 6.334774255523535,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018565871541111935,
      "loss": 4.5557,
      "mean_token_accuracy": 0.2719812229275703,
      "num_tokens": 151200708.0,
      "step": 65945
    },
    {
      "entropy": 5.476980113983155,
      "epoch": 6.335254562920269,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018562725143653075,
      "loss": 4.535,
      "mean_token_accuracy": 0.27121034264564514,
      "num_tokens": 151211807.0,
      "step": 65950
    },
    {
      "entropy": 5.577582502365113,
      "epoch": 6.335734870317003,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018559578953691197,
      "loss": 4.6822,
      "mean_token_accuracy": 0.26415221095085145,
      "num_tokens": 151223870.0,
      "step": 65955
    },
    {
      "entropy": 5.540898847579956,
      "epoch": 6.336215177713736,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018556432971299347,
      "loss": 4.6451,
      "mean_token_accuracy": 0.26878360062837603,
      "num_tokens": 151235639.0,
      "step": 65960
    },
    {
      "entropy": 5.5096453666687015,
      "epoch": 6.336695485110471,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018553287196550566,
      "loss": 4.6417,
      "mean_token_accuracy": 0.27036363184452056,
      "num_tokens": 151247647.0,
      "step": 65965
    },
    {
      "entropy": 5.570425891876221,
      "epoch": 6.337175792507205,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018550141629517896,
      "loss": 4.6387,
      "mean_token_accuracy": 0.2575374871492386,
      "num_tokens": 151258539.0,
      "step": 65970
    },
    {
      "entropy": 5.566310167312622,
      "epoch": 6.337656099903938,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018546996270274357,
      "loss": 4.6648,
      "mean_token_accuracy": 0.26176778972148895,
      "num_tokens": 151270307.0,
      "step": 65975
    },
    {
      "entropy": 5.602149772644043,
      "epoch": 6.338136407300673,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018543851118892985,
      "loss": 4.687,
      "mean_token_accuracy": 0.254022678732872,
      "num_tokens": 151281015.0,
      "step": 65980
    },
    {
      "entropy": 5.541650009155274,
      "epoch": 6.338616714697406,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018540706175446787,
      "loss": 4.6507,
      "mean_token_accuracy": 0.2667512372136116,
      "num_tokens": 151293971.0,
      "step": 65985
    },
    {
      "entropy": 5.481200122833252,
      "epoch": 6.3390970220941405,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018537561440008793,
      "loss": 4.5584,
      "mean_token_accuracy": 0.273677396774292,
      "num_tokens": 151306328.0,
      "step": 65990
    },
    {
      "entropy": 5.606839895248413,
      "epoch": 6.339577329490874,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018534416912652007,
      "loss": 4.6986,
      "mean_token_accuracy": 0.2608008190989494,
      "num_tokens": 151317514.0,
      "step": 65995
    },
    {
      "entropy": 5.533851099014282,
      "epoch": 6.340057636887608,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018531272593449438,
      "loss": 4.6419,
      "mean_token_accuracy": 0.2663942530751228,
      "num_tokens": 151329121.0,
      "step": 66000
    },
    {
      "epoch": 6.340057636887608,
      "eval_entropy": 5.245151309720602,
      "eval_loss": 4.753321647644043,
      "eval_mean_token_accuracy": 0.2665031260149195,
      "eval_num_tokens": 151329121.0,
      "eval_runtime": 20.5006,
      "eval_samples_per_second": 1600.687,
      "eval_steps_per_second": 200.092,
      "step": 66000
    },
    {
      "entropy": 5.484074497222901,
      "epoch": 6.340537944284342,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018528128482474094,
      "loss": 4.5495,
      "mean_token_accuracy": 0.2705417349934578,
      "num_tokens": 151339918.0,
      "step": 66005
    },
    {
      "entropy": 5.511465120315552,
      "epoch": 6.341018251681076,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018524984579798966,
      "loss": 4.6079,
      "mean_token_accuracy": 0.2624133974313736,
      "num_tokens": 151351174.0,
      "step": 66010
    },
    {
      "entropy": 5.499097585678101,
      "epoch": 6.3414985590778095,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018521840885497037,
      "loss": 4.6117,
      "mean_token_accuracy": 0.26449282467365265,
      "num_tokens": 151361437.0,
      "step": 66015
    },
    {
      "entropy": 5.526341772079467,
      "epoch": 6.341978866474544,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001851869739964131,
      "loss": 4.6426,
      "mean_token_accuracy": 0.26139551401138306,
      "num_tokens": 151372658.0,
      "step": 66020
    },
    {
      "entropy": 5.525505065917969,
      "epoch": 6.342459173871277,
      "grad_norm": 1.375,
      "learning_rate": 0.0001851555412230475,
      "loss": 4.6683,
      "mean_token_accuracy": 0.2647404745221138,
      "num_tokens": 151383829.0,
      "step": 66025
    },
    {
      "entropy": 5.547568511962891,
      "epoch": 6.3429394812680115,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018512411053560356,
      "loss": 4.6032,
      "mean_token_accuracy": 0.2737890511751175,
      "num_tokens": 151394479.0,
      "step": 66030
    },
    {
      "entropy": 5.576145935058594,
      "epoch": 6.343419788664745,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018509268193481088,
      "loss": 4.7309,
      "mean_token_accuracy": 0.2527498185634613,
      "num_tokens": 151406536.0,
      "step": 66035
    },
    {
      "entropy": 5.578877925872803,
      "epoch": 6.343900096061479,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018506125542139918,
      "loss": 4.6575,
      "mean_token_accuracy": 0.2683799237012863,
      "num_tokens": 151418094.0,
      "step": 66040
    },
    {
      "entropy": 5.604692554473877,
      "epoch": 6.344380403458214,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000185029830996098,
      "loss": 4.7035,
      "mean_token_accuracy": 0.2528657361865044,
      "num_tokens": 151429906.0,
      "step": 66045
    },
    {
      "entropy": 5.532488203048706,
      "epoch": 6.344860710854947,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018499840865963702,
      "loss": 4.6996,
      "mean_token_accuracy": 0.2602324768900871,
      "num_tokens": 151442361.0,
      "step": 66050
    },
    {
      "entropy": 5.506442642211914,
      "epoch": 6.345341018251681,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018496698841274572,
      "loss": 4.639,
      "mean_token_accuracy": 0.26262266784906385,
      "num_tokens": 151453314.0,
      "step": 66055
    },
    {
      "entropy": 5.568798446655274,
      "epoch": 6.345821325648415,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001849355702561536,
      "loss": 4.6542,
      "mean_token_accuracy": 0.2657468467950821,
      "num_tokens": 151465181.0,
      "step": 66060
    },
    {
      "entropy": 5.563934803009033,
      "epoch": 6.346301633045149,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001849041541905901,
      "loss": 4.6555,
      "mean_token_accuracy": 0.2609781548380852,
      "num_tokens": 151477236.0,
      "step": 66065
    },
    {
      "entropy": 5.568157911300659,
      "epoch": 6.346781940441883,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001848727402167847,
      "loss": 4.6061,
      "mean_token_accuracy": 0.26552932560443876,
      "num_tokens": 151489961.0,
      "step": 66070
    },
    {
      "entropy": 5.580051565170288,
      "epoch": 6.347262247838617,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018484132833546658,
      "loss": 4.7037,
      "mean_token_accuracy": 0.2539836809039116,
      "num_tokens": 151501176.0,
      "step": 66075
    },
    {
      "entropy": 5.510925960540772,
      "epoch": 6.34774255523535,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018480991854736507,
      "loss": 4.57,
      "mean_token_accuracy": 0.2736889436841011,
      "num_tokens": 151511478.0,
      "step": 66080
    },
    {
      "entropy": 5.4911003589630125,
      "epoch": 6.348222862632085,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018477851085320945,
      "loss": 4.5503,
      "mean_token_accuracy": 0.277837285399437,
      "num_tokens": 151523123.0,
      "step": 66085
    },
    {
      "entropy": 5.466259527206421,
      "epoch": 6.348703170028818,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018474710525372896,
      "loss": 4.5865,
      "mean_token_accuracy": 0.26709639877080915,
      "num_tokens": 151534138.0,
      "step": 66090
    },
    {
      "entropy": 5.468119668960571,
      "epoch": 6.3491834774255524,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018471570174965256,
      "loss": 4.5378,
      "mean_token_accuracy": 0.27090739756822585,
      "num_tokens": 151547372.0,
      "step": 66095
    },
    {
      "entropy": 5.623589563369751,
      "epoch": 6.349663784822286,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018468430034170968,
      "loss": 4.7774,
      "mean_token_accuracy": 0.2531581461429596,
      "num_tokens": 151559930.0,
      "step": 66100
    },
    {
      "entropy": 5.49972767829895,
      "epoch": 6.35014409221902,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018465290103062903,
      "loss": 4.6094,
      "mean_token_accuracy": 0.27310357689857484,
      "num_tokens": 151571673.0,
      "step": 66105
    },
    {
      "entropy": 5.59212737083435,
      "epoch": 6.350624399615754,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018462150381713977,
      "loss": 4.7019,
      "mean_token_accuracy": 0.25724539160728455,
      "num_tokens": 151583395.0,
      "step": 66110
    },
    {
      "entropy": 5.474138402938843,
      "epoch": 6.351104707012488,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018459010870197078,
      "loss": 4.5799,
      "mean_token_accuracy": 0.26847514063119887,
      "num_tokens": 151594653.0,
      "step": 66115
    },
    {
      "entropy": 5.478215837478638,
      "epoch": 6.351585014409222,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018455871568585108,
      "loss": 4.5676,
      "mean_token_accuracy": 0.2716072559356689,
      "num_tokens": 151605870.0,
      "step": 66120
    },
    {
      "entropy": 5.442676305770874,
      "epoch": 6.352065321805956,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018452732476950938,
      "loss": 4.6108,
      "mean_token_accuracy": 0.2697299435734749,
      "num_tokens": 151619142.0,
      "step": 66125
    },
    {
      "entropy": 5.557912826538086,
      "epoch": 6.35254562920269,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018449593595367469,
      "loss": 4.7039,
      "mean_token_accuracy": 0.2616403788328171,
      "num_tokens": 151630281.0,
      "step": 66130
    },
    {
      "entropy": 5.512059879302979,
      "epoch": 6.3530259365994235,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018446454923907546,
      "loss": 4.6016,
      "mean_token_accuracy": 0.2666778266429901,
      "num_tokens": 151641476.0,
      "step": 66135
    },
    {
      "entropy": 5.635841703414917,
      "epoch": 6.353506243996158,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018443316462644067,
      "loss": 4.7905,
      "mean_token_accuracy": 0.2484616830945015,
      "num_tokens": 151652909.0,
      "step": 66140
    },
    {
      "entropy": 5.517461824417114,
      "epoch": 6.353986551392891,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018440178211649876,
      "loss": 4.5261,
      "mean_token_accuracy": 0.2876113712787628,
      "num_tokens": 151663643.0,
      "step": 66145
    },
    {
      "entropy": 5.5352071762084964,
      "epoch": 6.354466858789626,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018437040170997849,
      "loss": 4.6298,
      "mean_token_accuracy": 0.27398828864097596,
      "num_tokens": 151674463.0,
      "step": 66150
    },
    {
      "entropy": 5.534753036499024,
      "epoch": 6.354947166186359,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001843390234076084,
      "loss": 4.6244,
      "mean_token_accuracy": 0.2630961716175079,
      "num_tokens": 151685465.0,
      "step": 66155
    },
    {
      "entropy": 5.514005947113037,
      "epoch": 6.355427473583093,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018430764721011705,
      "loss": 4.6408,
      "mean_token_accuracy": 0.2606455937027931,
      "num_tokens": 151696886.0,
      "step": 66160
    },
    {
      "entropy": 5.563063859939575,
      "epoch": 6.355907780979827,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018427627311823268,
      "loss": 4.7039,
      "mean_token_accuracy": 0.2572352156043053,
      "num_tokens": 151707933.0,
      "step": 66165
    },
    {
      "entropy": 5.605059957504272,
      "epoch": 6.356388088376561,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018424490113268393,
      "loss": 4.6564,
      "mean_token_accuracy": 0.2630354166030884,
      "num_tokens": 151720143.0,
      "step": 66170
    },
    {
      "entropy": 5.568411922454834,
      "epoch": 6.3568683957732945,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001842135312541991,
      "loss": 4.6449,
      "mean_token_accuracy": 0.26695931553840635,
      "num_tokens": 151731105.0,
      "step": 66175
    },
    {
      "entropy": 5.4729503154754635,
      "epoch": 6.357348703170029,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018418216348350648,
      "loss": 4.5939,
      "mean_token_accuracy": 0.2659589186310768,
      "num_tokens": 151742961.0,
      "step": 66180
    },
    {
      "entropy": 5.554652166366577,
      "epoch": 6.357829010566762,
      "grad_norm": 1.375,
      "learning_rate": 0.0001841507978213343,
      "loss": 4.7061,
      "mean_token_accuracy": 0.25858000218868255,
      "num_tokens": 151755375.0,
      "step": 66185
    },
    {
      "entropy": 5.562044668197632,
      "epoch": 6.358309317963497,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018411943426841092,
      "loss": 4.6269,
      "mean_token_accuracy": 0.26370885223150253,
      "num_tokens": 151765854.0,
      "step": 66190
    },
    {
      "entropy": 5.5046485424041744,
      "epoch": 6.358789625360231,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001840880728254644,
      "loss": 4.6199,
      "mean_token_accuracy": 0.26902866959571836,
      "num_tokens": 151776984.0,
      "step": 66195
    },
    {
      "entropy": 5.464341545104981,
      "epoch": 6.359269932756964,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018405671349322285,
      "loss": 4.6053,
      "mean_token_accuracy": 0.2732784479856491,
      "num_tokens": 151788959.0,
      "step": 66200
    },
    {
      "entropy": 5.497906827926636,
      "epoch": 6.359750240153699,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018402535627241434,
      "loss": 4.5841,
      "mean_token_accuracy": 0.27182594537734983,
      "num_tokens": 151799466.0,
      "step": 66205
    },
    {
      "entropy": 5.505326461791992,
      "epoch": 6.360230547550432,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001839940011637669,
      "loss": 4.6009,
      "mean_token_accuracy": 0.27024245709180833,
      "num_tokens": 151811095.0,
      "step": 66210
    },
    {
      "entropy": 5.5133568286895756,
      "epoch": 6.3607108549471665,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018396264816800863,
      "loss": 4.5909,
      "mean_token_accuracy": 0.2722735613584518,
      "num_tokens": 151823358.0,
      "step": 66215
    },
    {
      "entropy": 5.515657186508179,
      "epoch": 6.3611911623439,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018393129728586728,
      "loss": 4.642,
      "mean_token_accuracy": 0.2603470325469971,
      "num_tokens": 151835926.0,
      "step": 66220
    },
    {
      "entropy": 5.464744329452515,
      "epoch": 6.361671469740634,
      "grad_norm": 1.25,
      "learning_rate": 0.00018389994851807074,
      "loss": 4.5561,
      "mean_token_accuracy": 0.2759879231452942,
      "num_tokens": 151847883.0,
      "step": 66225
    },
    {
      "entropy": 5.605698347091675,
      "epoch": 6.362151777137368,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018386860186534693,
      "loss": 4.7385,
      "mean_token_accuracy": 0.2543942466378212,
      "num_tokens": 151859215.0,
      "step": 66230
    },
    {
      "entropy": 5.498382806777954,
      "epoch": 6.362632084534102,
      "grad_norm": 1.5,
      "learning_rate": 0.0001838372573284236,
      "loss": 4.6363,
      "mean_token_accuracy": 0.26675044298171996,
      "num_tokens": 151871088.0,
      "step": 66235
    },
    {
      "entropy": 5.585180616378784,
      "epoch": 6.363112391930835,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018380591490802845,
      "loss": 4.7091,
      "mean_token_accuracy": 0.26224873811006544,
      "num_tokens": 151882568.0,
      "step": 66240
    },
    {
      "entropy": 5.49586820602417,
      "epoch": 6.36359269932757,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001837745746048891,
      "loss": 4.6034,
      "mean_token_accuracy": 0.2706134274601936,
      "num_tokens": 151893098.0,
      "step": 66245
    },
    {
      "entropy": 5.563578462600708,
      "epoch": 6.364073006724303,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001837432364197334,
      "loss": 4.6381,
      "mean_token_accuracy": 0.2633501023054123,
      "num_tokens": 151904766.0,
      "step": 66250
    },
    {
      "entropy": 5.499965572357178,
      "epoch": 6.3645533141210375,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018371190035328862,
      "loss": 4.6182,
      "mean_token_accuracy": 0.26913315802812576,
      "num_tokens": 151915534.0,
      "step": 66255
    },
    {
      "entropy": 5.575617408752441,
      "epoch": 6.365033621517771,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001836805664062825,
      "loss": 4.7162,
      "mean_token_accuracy": 0.2563780590891838,
      "num_tokens": 151927196.0,
      "step": 66260
    },
    {
      "entropy": 5.535867071151733,
      "epoch": 6.365513928914505,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018364923457944244,
      "loss": 4.6261,
      "mean_token_accuracy": 0.26041433960199356,
      "num_tokens": 151938509.0,
      "step": 66265
    },
    {
      "entropy": 5.574331903457642,
      "epoch": 6.36599423631124,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001836179048734959,
      "loss": 4.697,
      "mean_token_accuracy": 0.2621780648827553,
      "num_tokens": 151949592.0,
      "step": 66270
    },
    {
      "entropy": 5.605873870849609,
      "epoch": 6.366474543707973,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018358657728917028,
      "loss": 4.7472,
      "mean_token_accuracy": 0.2595200642943382,
      "num_tokens": 151961095.0,
      "step": 66275
    },
    {
      "entropy": 5.554869699478149,
      "epoch": 6.366954851104707,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018355525182719296,
      "loss": 4.6212,
      "mean_token_accuracy": 0.2637707948684692,
      "num_tokens": 151971799.0,
      "step": 66280
    },
    {
      "entropy": 5.443694639205932,
      "epoch": 6.367435158501441,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018352392848829102,
      "loss": 4.6101,
      "mean_token_accuracy": 0.2689604043960571,
      "num_tokens": 151982673.0,
      "step": 66285
    },
    {
      "entropy": 5.478324365615845,
      "epoch": 6.367915465898175,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018349260727319194,
      "loss": 4.6198,
      "mean_token_accuracy": 0.2725543022155762,
      "num_tokens": 151994771.0,
      "step": 66290
    },
    {
      "entropy": 5.5080084800720215,
      "epoch": 6.368395773294909,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018346128818262275,
      "loss": 4.6215,
      "mean_token_accuracy": 0.261769050359726,
      "num_tokens": 152006213.0,
      "step": 66295
    },
    {
      "entropy": 5.525676631927491,
      "epoch": 6.368876080691643,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018342997121731058,
      "loss": 4.6657,
      "mean_token_accuracy": 0.2662900134921074,
      "num_tokens": 152017559.0,
      "step": 66300
    },
    {
      "entropy": 5.538890838623047,
      "epoch": 6.369356388088376,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018339865637798265,
      "loss": 4.6443,
      "mean_token_accuracy": 0.2700124219059944,
      "num_tokens": 152029514.0,
      "step": 66305
    },
    {
      "entropy": 5.652804565429688,
      "epoch": 6.369836695485111,
      "grad_norm": 1.375,
      "learning_rate": 0.000183367343665366,
      "loss": 4.8014,
      "mean_token_accuracy": 0.2546264335513115,
      "num_tokens": 152040797.0,
      "step": 66310
    },
    {
      "entropy": 5.581482553482056,
      "epoch": 6.370317002881844,
      "grad_norm": 1.375,
      "learning_rate": 0.00018333603308018741,
      "loss": 4.6627,
      "mean_token_accuracy": 0.26333910375833514,
      "num_tokens": 152054275.0,
      "step": 66315
    },
    {
      "entropy": 5.528045654296875,
      "epoch": 6.370797310278578,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018330472462317393,
      "loss": 4.6268,
      "mean_token_accuracy": 0.26614891439676286,
      "num_tokens": 152066840.0,
      "step": 66320
    },
    {
      "entropy": 5.539571809768677,
      "epoch": 6.371277617675312,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001832734182950525,
      "loss": 4.629,
      "mean_token_accuracy": 0.2683915600180626,
      "num_tokens": 152078770.0,
      "step": 66325
    },
    {
      "entropy": 5.613544178009033,
      "epoch": 6.371757925072046,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018324211409654996,
      "loss": 4.7199,
      "mean_token_accuracy": 0.25299670100212096,
      "num_tokens": 152091144.0,
      "step": 66330
    },
    {
      "entropy": 5.524249267578125,
      "epoch": 6.37223823246878,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018321081202839302,
      "loss": 4.6069,
      "mean_token_accuracy": 0.2591278895735741,
      "num_tokens": 152101189.0,
      "step": 66335
    },
    {
      "entropy": 5.551014423370361,
      "epoch": 6.372718539865514,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018317951209130857,
      "loss": 4.6592,
      "mean_token_accuracy": 0.26777167022228243,
      "num_tokens": 152112429.0,
      "step": 66340
    },
    {
      "entropy": 5.5982459545135494,
      "epoch": 6.373198847262248,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018314821428602306,
      "loss": 4.7591,
      "mean_token_accuracy": 0.25635301470756533,
      "num_tokens": 152124369.0,
      "step": 66345
    },
    {
      "entropy": 5.443406677246093,
      "epoch": 6.373679154658982,
      "grad_norm": 1.59375,
      "learning_rate": 0.00018311691861326335,
      "loss": 4.5162,
      "mean_token_accuracy": 0.2734124198555946,
      "num_tokens": 152135073.0,
      "step": 66350
    },
    {
      "entropy": 5.470147705078125,
      "epoch": 6.374159462055716,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018308562507375593,
      "loss": 4.5465,
      "mean_token_accuracy": 0.2789193630218506,
      "num_tokens": 152147254.0,
      "step": 66355
    },
    {
      "entropy": 5.5426778316497805,
      "epoch": 6.3746397694524495,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001830543336682273,
      "loss": 4.6276,
      "mean_token_accuracy": 0.2657278269529343,
      "num_tokens": 152159905.0,
      "step": 66360
    },
    {
      "entropy": 5.533050298690796,
      "epoch": 6.375120076849184,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001830230443974041,
      "loss": 4.543,
      "mean_token_accuracy": 0.26845393925905225,
      "num_tokens": 152171188.0,
      "step": 66365
    },
    {
      "entropy": 5.573424196243286,
      "epoch": 6.375600384245917,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018299175726201278,
      "loss": 4.6427,
      "mean_token_accuracy": 0.2563684418797493,
      "num_tokens": 152182687.0,
      "step": 66370
    },
    {
      "entropy": 5.499977350234985,
      "epoch": 6.376080691642652,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001829604722627795,
      "loss": 4.6344,
      "mean_token_accuracy": 0.26477238088846206,
      "num_tokens": 152193546.0,
      "step": 66375
    },
    {
      "entropy": 5.494563245773316,
      "epoch": 6.376560999039385,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001829291894004308,
      "loss": 4.5642,
      "mean_token_accuracy": 0.26911176443099977,
      "num_tokens": 152203128.0,
      "step": 66380
    },
    {
      "entropy": 5.527571439743042,
      "epoch": 6.377041306436119,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018289790867569294,
      "loss": 4.7189,
      "mean_token_accuracy": 0.253514775633812,
      "num_tokens": 152214845.0,
      "step": 66385
    },
    {
      "entropy": 5.466054677963257,
      "epoch": 6.377521613832853,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001828666300892921,
      "loss": 4.4875,
      "mean_token_accuracy": 0.27615155577659606,
      "num_tokens": 152225621.0,
      "step": 66390
    },
    {
      "entropy": 5.54309720993042,
      "epoch": 6.378001921229587,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001828353536419546,
      "loss": 4.6238,
      "mean_token_accuracy": 0.26971338391304017,
      "num_tokens": 152237553.0,
      "step": 66395
    },
    {
      "entropy": 5.558413314819336,
      "epoch": 6.3784822286263205,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001828040793344065,
      "loss": 4.6346,
      "mean_token_accuracy": 0.27130394279956815,
      "num_tokens": 152248498.0,
      "step": 66400
    },
    {
      "entropy": 5.562586975097656,
      "epoch": 6.378962536023055,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018277280716737395,
      "loss": 4.5952,
      "mean_token_accuracy": 0.2713479146361351,
      "num_tokens": 152260109.0,
      "step": 66405
    },
    {
      "entropy": 5.47955904006958,
      "epoch": 6.379442843419788,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00018274153714158286,
      "loss": 4.5728,
      "mean_token_accuracy": 0.2775333166122437,
      "num_tokens": 152270890.0,
      "step": 66410
    },
    {
      "entropy": 5.503725051879883,
      "epoch": 6.379923150816523,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018271026925775936,
      "loss": 4.6335,
      "mean_token_accuracy": 0.2683076933026314,
      "num_tokens": 152282555.0,
      "step": 66415
    },
    {
      "entropy": 5.51844334602356,
      "epoch": 6.380403458213257,
      "grad_norm": 1.375,
      "learning_rate": 0.00018267900351662936,
      "loss": 4.6245,
      "mean_token_accuracy": 0.26379183679819107,
      "num_tokens": 152294805.0,
      "step": 66420
    },
    {
      "entropy": 5.579897546768189,
      "epoch": 6.38088376560999,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018264773991891875,
      "loss": 4.6481,
      "mean_token_accuracy": 0.26074049025774004,
      "num_tokens": 152306546.0,
      "step": 66425
    },
    {
      "entropy": 5.607895612716675,
      "epoch": 6.381364073006725,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018261647846535352,
      "loss": 4.7198,
      "mean_token_accuracy": 0.2513426810503006,
      "num_tokens": 152318032.0,
      "step": 66430
    },
    {
      "entropy": 5.492422342300415,
      "epoch": 6.381844380403458,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001825852191566592,
      "loss": 4.5571,
      "mean_token_accuracy": 0.2712961107492447,
      "num_tokens": 152329932.0,
      "step": 66435
    },
    {
      "entropy": 5.546331071853638,
      "epoch": 6.3823246878001925,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018255396199356173,
      "loss": 4.616,
      "mean_token_accuracy": 0.26321388483047486,
      "num_tokens": 152341871.0,
      "step": 66440
    },
    {
      "entropy": 5.495746612548828,
      "epoch": 6.382804995196926,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001825227069767867,
      "loss": 4.5773,
      "mean_token_accuracy": 0.27265861332416536,
      "num_tokens": 152352402.0,
      "step": 66445
    },
    {
      "entropy": 5.506417369842529,
      "epoch": 6.38328530259366,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018249145410705986,
      "loss": 4.5933,
      "mean_token_accuracy": 0.2727459132671356,
      "num_tokens": 152363803.0,
      "step": 66450
    },
    {
      "entropy": 5.599659109115601,
      "epoch": 6.383765609990394,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018246020338510673,
      "loss": 4.7018,
      "mean_token_accuracy": 0.2570579141378403,
      "num_tokens": 152373661.0,
      "step": 66455
    },
    {
      "entropy": 5.50138783454895,
      "epoch": 6.384245917387128,
      "grad_norm": 1.40625,
      "learning_rate": 0.000182428954811653,
      "loss": 4.6352,
      "mean_token_accuracy": 0.26748556792736056,
      "num_tokens": 152384435.0,
      "step": 66460
    },
    {
      "entropy": 5.508697986602783,
      "epoch": 6.384726224783861,
      "grad_norm": 1.375,
      "learning_rate": 0.0001823977083874239,
      "loss": 4.593,
      "mean_token_accuracy": 0.2618558719754219,
      "num_tokens": 152395452.0,
      "step": 66465
    },
    {
      "entropy": 5.473724508285523,
      "epoch": 6.385206532180596,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018236646411314513,
      "loss": 4.567,
      "mean_token_accuracy": 0.27453852593898775,
      "num_tokens": 152406999.0,
      "step": 66470
    },
    {
      "entropy": 5.58601746559143,
      "epoch": 6.385686839577329,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018233522198954195,
      "loss": 4.74,
      "mean_token_accuracy": 0.25709466338157655,
      "num_tokens": 152418990.0,
      "step": 66475
    },
    {
      "entropy": 5.491377592086792,
      "epoch": 6.3861671469740635,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018230398201733977,
      "loss": 4.5717,
      "mean_token_accuracy": 0.27123330235481263,
      "num_tokens": 152430218.0,
      "step": 66480
    },
    {
      "entropy": 5.552018880844116,
      "epoch": 6.386647454370797,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001822727441972639,
      "loss": 4.6518,
      "mean_token_accuracy": 0.26236843317747116,
      "num_tokens": 152442106.0,
      "step": 66485
    },
    {
      "entropy": 5.5536839962005615,
      "epoch": 6.387127761767531,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018224150853003956,
      "loss": 4.6457,
      "mean_token_accuracy": 0.2601133808493614,
      "num_tokens": 152452908.0,
      "step": 66490
    },
    {
      "entropy": 5.642125129699707,
      "epoch": 6.387608069164266,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000182210275016392,
      "loss": 4.6689,
      "mean_token_accuracy": 0.2574803277850151,
      "num_tokens": 152463253.0,
      "step": 66495
    },
    {
      "entropy": 5.568573951721191,
      "epoch": 6.388088376560999,
      "grad_norm": 1.5,
      "learning_rate": 0.0001821790436570463,
      "loss": 4.6823,
      "mean_token_accuracy": 0.26087375432252885,
      "num_tokens": 152473826.0,
      "step": 66500
    },
    {
      "entropy": 5.49469313621521,
      "epoch": 6.388568683957733,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018214781445272755,
      "loss": 4.5963,
      "mean_token_accuracy": 0.26892772912979124,
      "num_tokens": 152484366.0,
      "step": 66505
    },
    {
      "entropy": 5.525574827194214,
      "epoch": 6.389048991354467,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001821165874041608,
      "loss": 4.6268,
      "mean_token_accuracy": 0.26809539645910263,
      "num_tokens": 152495587.0,
      "step": 66510
    },
    {
      "entropy": 5.4901563167572025,
      "epoch": 6.389529298751201,
      "grad_norm": 1.453125,
      "learning_rate": 0.00018208536251207113,
      "loss": 4.4645,
      "mean_token_accuracy": 0.2827120065689087,
      "num_tokens": 152506587.0,
      "step": 66515
    },
    {
      "entropy": 5.485766220092773,
      "epoch": 6.390009606147935,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001820541397771835,
      "loss": 4.5284,
      "mean_token_accuracy": 0.2726543456315994,
      "num_tokens": 152517251.0,
      "step": 66520
    },
    {
      "entropy": 5.547025680541992,
      "epoch": 6.390489913544669,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001820229192002227,
      "loss": 4.6765,
      "mean_token_accuracy": 0.25498612225055695,
      "num_tokens": 152528650.0,
      "step": 66525
    },
    {
      "entropy": 5.537498140335083,
      "epoch": 6.390970220941402,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018199170078191362,
      "loss": 4.6353,
      "mean_token_accuracy": 0.2620469510555267,
      "num_tokens": 152539654.0,
      "step": 66530
    },
    {
      "entropy": 5.65944275856018,
      "epoch": 6.391450528338137,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018196048452298104,
      "loss": 4.7815,
      "mean_token_accuracy": 0.2591292053461075,
      "num_tokens": 152551056.0,
      "step": 66535
    },
    {
      "entropy": 5.530075645446777,
      "epoch": 6.39193083573487,
      "grad_norm": 1.375,
      "learning_rate": 0.00018192927042414981,
      "loss": 4.6543,
      "mean_token_accuracy": 0.2649324357509613,
      "num_tokens": 152563365.0,
      "step": 66540
    },
    {
      "entropy": 5.544871282577515,
      "epoch": 6.392411143131604,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018189805848614456,
      "loss": 4.6582,
      "mean_token_accuracy": 0.2597183272242546,
      "num_tokens": 152574568.0,
      "step": 66545
    },
    {
      "entropy": 5.61327018737793,
      "epoch": 6.392891450528338,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018186684870968983,
      "loss": 4.7588,
      "mean_token_accuracy": 0.2548930928111076,
      "num_tokens": 152586931.0,
      "step": 66550
    },
    {
      "entropy": 5.537481832504272,
      "epoch": 6.393371757925072,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018183564109551043,
      "loss": 4.6028,
      "mean_token_accuracy": 0.26699705570936205,
      "num_tokens": 152599489.0,
      "step": 66555
    },
    {
      "entropy": 5.565831661224365,
      "epoch": 6.393852065321806,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018180443564433073,
      "loss": 4.685,
      "mean_token_accuracy": 0.2630565628409386,
      "num_tokens": 152610121.0,
      "step": 66560
    },
    {
      "entropy": 5.560694456100464,
      "epoch": 6.39433237271854,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001817732323568753,
      "loss": 4.6263,
      "mean_token_accuracy": 0.27109200358390806,
      "num_tokens": 152622184.0,
      "step": 66565
    },
    {
      "entropy": 5.467597246170044,
      "epoch": 6.394812680115273,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018174203123386866,
      "loss": 4.5607,
      "mean_token_accuracy": 0.273808915913105,
      "num_tokens": 152634340.0,
      "step": 66570
    },
    {
      "entropy": 5.554424905776978,
      "epoch": 6.395292987512008,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018171083227603508,
      "loss": 4.6313,
      "mean_token_accuracy": 0.26514761745929716,
      "num_tokens": 152645697.0,
      "step": 66575
    },
    {
      "entropy": 5.562559700012207,
      "epoch": 6.395773294908741,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018167963548409904,
      "loss": 4.6452,
      "mean_token_accuracy": 0.2594499528408051,
      "num_tokens": 152657010.0,
      "step": 66580
    },
    {
      "entropy": 5.4993088722229,
      "epoch": 6.3962536023054755,
      "grad_norm": 1.5,
      "learning_rate": 0.00018164844085878475,
      "loss": 4.5122,
      "mean_token_accuracy": 0.2743947386741638,
      "num_tokens": 152668497.0,
      "step": 66585
    },
    {
      "entropy": 5.593475484848023,
      "epoch": 6.39673390970221,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001816172484008164,
      "loss": 4.7195,
      "mean_token_accuracy": 0.2551822602748871,
      "num_tokens": 152679795.0,
      "step": 66590
    },
    {
      "entropy": 5.541647291183471,
      "epoch": 6.397214217098943,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018158605811091822,
      "loss": 4.6388,
      "mean_token_accuracy": 0.26467067301273345,
      "num_tokens": 152692241.0,
      "step": 66595
    },
    {
      "entropy": 5.488809967041016,
      "epoch": 6.3976945244956775,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001815548699898144,
      "loss": 4.5567,
      "mean_token_accuracy": 0.27634555101394653,
      "num_tokens": 152704171.0,
      "step": 66600
    },
    {
      "entropy": 5.554880666732788,
      "epoch": 6.398174831892411,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001815236840382291,
      "loss": 4.6181,
      "mean_token_accuracy": 0.2718179315328598,
      "num_tokens": 152716179.0,
      "step": 66605
    },
    {
      "entropy": 5.491524696350098,
      "epoch": 6.398655139289145,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001814925002568863,
      "loss": 4.5815,
      "mean_token_accuracy": 0.2679037511348724,
      "num_tokens": 152727435.0,
      "step": 66610
    },
    {
      "entropy": 5.50878758430481,
      "epoch": 6.399135446685879,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018146131864650995,
      "loss": 4.5891,
      "mean_token_accuracy": 0.26826329082250594,
      "num_tokens": 152738854.0,
      "step": 66615
    },
    {
      "entropy": 5.522752571105957,
      "epoch": 6.399615754082613,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018143013920782398,
      "loss": 4.6087,
      "mean_token_accuracy": 0.2648699313402176,
      "num_tokens": 152750869.0,
      "step": 66620
    },
    {
      "entropy": 5.608761167526245,
      "epoch": 6.4000960614793465,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018139896194155236,
      "loss": 4.69,
      "mean_token_accuracy": 0.2604913994669914,
      "num_tokens": 152761610.0,
      "step": 66625
    },
    {
      "entropy": 5.537380123138428,
      "epoch": 6.400576368876081,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018136778684841893,
      "loss": 4.6193,
      "mean_token_accuracy": 0.2683445453643799,
      "num_tokens": 152772936.0,
      "step": 66630
    },
    {
      "entropy": 5.45739517211914,
      "epoch": 6.401056676272814,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018133661392914747,
      "loss": 4.5261,
      "mean_token_accuracy": 0.2786402225494385,
      "num_tokens": 152783998.0,
      "step": 66635
    },
    {
      "entropy": 5.600032997131348,
      "epoch": 6.401536983669549,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018130544318446164,
      "loss": 4.7212,
      "mean_token_accuracy": 0.25182133466005324,
      "num_tokens": 152797279.0,
      "step": 66640
    },
    {
      "entropy": 5.615682935714721,
      "epoch": 6.402017291066282,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018127427461508521,
      "loss": 4.7808,
      "mean_token_accuracy": 0.25094001591205595,
      "num_tokens": 152809532.0,
      "step": 66645
    },
    {
      "entropy": 5.595202970504761,
      "epoch": 6.402497598463016,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018124310822174188,
      "loss": 4.6667,
      "mean_token_accuracy": 0.2644464761018753,
      "num_tokens": 152820633.0,
      "step": 66650
    },
    {
      "entropy": 5.611115789413452,
      "epoch": 6.40297790585975,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018121194400515524,
      "loss": 4.7661,
      "mean_token_accuracy": 0.25431063771247864,
      "num_tokens": 152832417.0,
      "step": 66655
    },
    {
      "entropy": 5.591615533828735,
      "epoch": 6.403458213256484,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018118078196604866,
      "loss": 4.6514,
      "mean_token_accuracy": 0.2665770262479782,
      "num_tokens": 152843860.0,
      "step": 66660
    },
    {
      "entropy": 5.56947922706604,
      "epoch": 6.4039385206532184,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018114962210514574,
      "loss": 4.6852,
      "mean_token_accuracy": 0.2622928574681282,
      "num_tokens": 152855130.0,
      "step": 66665
    },
    {
      "entropy": 5.507465553283692,
      "epoch": 6.404418828049952,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018111846442316992,
      "loss": 4.6359,
      "mean_token_accuracy": 0.2623109221458435,
      "num_tokens": 152865734.0,
      "step": 66670
    },
    {
      "entropy": 5.502532911300659,
      "epoch": 6.404899135446686,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018108730892084458,
      "loss": 4.5707,
      "mean_token_accuracy": 0.2718650072813034,
      "num_tokens": 152876926.0,
      "step": 66675
    },
    {
      "entropy": 5.505537748336792,
      "epoch": 6.40537944284342,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018105615559889306,
      "loss": 4.5871,
      "mean_token_accuracy": 0.27527831196784974,
      "num_tokens": 152888201.0,
      "step": 66680
    },
    {
      "entropy": 5.522429037094116,
      "epoch": 6.405859750240154,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018102500445803864,
      "loss": 4.6355,
      "mean_token_accuracy": 0.2640791177749634,
      "num_tokens": 152900309.0,
      "step": 66685
    },
    {
      "entropy": 5.431570148468017,
      "epoch": 6.406340057636887,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018099385549900454,
      "loss": 4.553,
      "mean_token_accuracy": 0.2705784499645233,
      "num_tokens": 152912025.0,
      "step": 66690
    },
    {
      "entropy": 5.5775399684906,
      "epoch": 6.406820365033622,
      "grad_norm": 1.5,
      "learning_rate": 0.00018096270872251414,
      "loss": 4.7261,
      "mean_token_accuracy": 0.2604486346244812,
      "num_tokens": 152923734.0,
      "step": 66695
    },
    {
      "entropy": 5.538027334213257,
      "epoch": 6.407300672430355,
      "grad_norm": 1.375,
      "learning_rate": 0.0001809315641292903,
      "loss": 4.6266,
      "mean_token_accuracy": 0.26795669049024584,
      "num_tokens": 152935124.0,
      "step": 66700
    },
    {
      "entropy": 5.564821434020996,
      "epoch": 6.4077809798270895,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018090042172005615,
      "loss": 4.6333,
      "mean_token_accuracy": 0.2641993761062622,
      "num_tokens": 152947427.0,
      "step": 66705
    },
    {
      "entropy": 5.551565742492675,
      "epoch": 6.408261287223823,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001808692814955349,
      "loss": 4.648,
      "mean_token_accuracy": 0.26411387920379636,
      "num_tokens": 152959940.0,
      "step": 66710
    },
    {
      "entropy": 5.600613164901733,
      "epoch": 6.408741594620557,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001808381434564494,
      "loss": 4.7021,
      "mean_token_accuracy": 0.255855156481266,
      "num_tokens": 152972249.0,
      "step": 66715
    },
    {
      "entropy": 5.53734655380249,
      "epoch": 6.409221902017291,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018080700760352247,
      "loss": 4.6081,
      "mean_token_accuracy": 0.25826194882392883,
      "num_tokens": 152984555.0,
      "step": 66720
    },
    {
      "entropy": 5.496561050415039,
      "epoch": 6.409702209414025,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018077587393747725,
      "loss": 4.5423,
      "mean_token_accuracy": 0.27089803814888,
      "num_tokens": 152995063.0,
      "step": 66725
    },
    {
      "entropy": 5.5755743980407715,
      "epoch": 6.4101825168107585,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018074474245903654,
      "loss": 4.6901,
      "mean_token_accuracy": 0.2540290281176567,
      "num_tokens": 153006470.0,
      "step": 66730
    },
    {
      "entropy": 5.673631095886231,
      "epoch": 6.410662824207493,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018071361316892293,
      "loss": 4.7989,
      "mean_token_accuracy": 0.24847493022680284,
      "num_tokens": 153017072.0,
      "step": 66735
    },
    {
      "entropy": 5.600458097457886,
      "epoch": 6.411143131604227,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018068248606785926,
      "loss": 4.7185,
      "mean_token_accuracy": 0.2571298614144325,
      "num_tokens": 153028739.0,
      "step": 66740
    },
    {
      "entropy": 5.5033714294433596,
      "epoch": 6.4116234390009605,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018065136115656815,
      "loss": 4.6163,
      "mean_token_accuracy": 0.26296799927949904,
      "num_tokens": 153041659.0,
      "step": 66745
    },
    {
      "entropy": 5.531151151657104,
      "epoch": 6.412103746397695,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001806202384357723,
      "loss": 4.5914,
      "mean_token_accuracy": 0.2737041309475899,
      "num_tokens": 153052398.0,
      "step": 66750
    },
    {
      "entropy": 5.507239818572998,
      "epoch": 6.412584053794428,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018058911790619426,
      "loss": 4.6698,
      "mean_token_accuracy": 0.265424807369709,
      "num_tokens": 153064138.0,
      "step": 66755
    },
    {
      "entropy": 5.501428461074829,
      "epoch": 6.413064361191163,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018055799956855667,
      "loss": 4.5741,
      "mean_token_accuracy": 0.264825402200222,
      "num_tokens": 153075338.0,
      "step": 66760
    },
    {
      "entropy": 5.498278045654297,
      "epoch": 6.413544668587896,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018052688342358176,
      "loss": 4.5428,
      "mean_token_accuracy": 0.26874801963567735,
      "num_tokens": 153085279.0,
      "step": 66765
    },
    {
      "entropy": 5.522616243362426,
      "epoch": 6.41402497598463,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018049576947199214,
      "loss": 4.66,
      "mean_token_accuracy": 0.2689571186900139,
      "num_tokens": 153097998.0,
      "step": 66770
    },
    {
      "entropy": 5.498558712005615,
      "epoch": 6.414505283381364,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001804646577145101,
      "loss": 4.6282,
      "mean_token_accuracy": 0.26634082198143005,
      "num_tokens": 153108740.0,
      "step": 66775
    },
    {
      "entropy": 5.596980953216553,
      "epoch": 6.414985590778098,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018043354815185797,
      "loss": 4.6718,
      "mean_token_accuracy": 0.25633932650089264,
      "num_tokens": 153119970.0,
      "step": 66780
    },
    {
      "entropy": 5.54710431098938,
      "epoch": 6.415465898174832,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018040244078475814,
      "loss": 4.5665,
      "mean_token_accuracy": 0.2713458016514778,
      "num_tokens": 153132734.0,
      "step": 66785
    },
    {
      "entropy": 5.589692831039429,
      "epoch": 6.415946205571566,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018037133561393272,
      "loss": 4.6641,
      "mean_token_accuracy": 0.2657909125089645,
      "num_tokens": 153144510.0,
      "step": 66790
    },
    {
      "entropy": 5.583707571029663,
      "epoch": 6.416426512968299,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001803402326401039,
      "loss": 4.7202,
      "mean_token_accuracy": 0.2533646926283836,
      "num_tokens": 153155224.0,
      "step": 66795
    },
    {
      "entropy": 5.546238040924072,
      "epoch": 6.416906820365034,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018030913186399373,
      "loss": 4.6415,
      "mean_token_accuracy": 0.26897598803043365,
      "num_tokens": 153165512.0,
      "step": 66800
    },
    {
      "entropy": 5.545797634124756,
      "epoch": 6.417387127761767,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018027803328632435,
      "loss": 4.627,
      "mean_token_accuracy": 0.2639942213892937,
      "num_tokens": 153177349.0,
      "step": 66805
    },
    {
      "entropy": 5.540287685394287,
      "epoch": 6.417867435158501,
      "grad_norm": 1.375,
      "learning_rate": 0.0001802469369078178,
      "loss": 4.5893,
      "mean_token_accuracy": 0.2711742758750916,
      "num_tokens": 153188355.0,
      "step": 66810
    },
    {
      "entropy": 5.567296028137207,
      "epoch": 6.418347742555236,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018021584272919606,
      "loss": 4.6941,
      "mean_token_accuracy": 0.25180643051862717,
      "num_tokens": 153200939.0,
      "step": 66815
    },
    {
      "entropy": 5.632682180404663,
      "epoch": 6.418828049951969,
      "grad_norm": 1.328125,
      "learning_rate": 0.000180184750751181,
      "loss": 4.7501,
      "mean_token_accuracy": 0.2646998330950737,
      "num_tokens": 153211704.0,
      "step": 66820
    },
    {
      "entropy": 5.588235664367676,
      "epoch": 6.4193083573487035,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018015366097449444,
      "loss": 4.6069,
      "mean_token_accuracy": 0.2664744362235069,
      "num_tokens": 153222627.0,
      "step": 66825
    },
    {
      "entropy": 5.558996725082397,
      "epoch": 6.419788664745437,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001801225733998582,
      "loss": 4.6466,
      "mean_token_accuracy": 0.26756056398153305,
      "num_tokens": 153233598.0,
      "step": 66830
    },
    {
      "entropy": 5.54027419090271,
      "epoch": 6.420268972142171,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018009148802799414,
      "loss": 4.6833,
      "mean_token_accuracy": 0.2634829252958298,
      "num_tokens": 153244532.0,
      "step": 66835
    },
    {
      "entropy": 5.584281778335571,
      "epoch": 6.420749279538905,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001800604048596239,
      "loss": 4.6391,
      "mean_token_accuracy": 0.26731078028678895,
      "num_tokens": 153256989.0,
      "step": 66840
    },
    {
      "entropy": 5.5258323669433596,
      "epoch": 6.421229586935639,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001800293238954691,
      "loss": 4.5653,
      "mean_token_accuracy": 0.2735492378473282,
      "num_tokens": 153267734.0,
      "step": 66845
    },
    {
      "entropy": 5.587073707580567,
      "epoch": 6.4217098943323725,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017999824513625145,
      "loss": 4.6833,
      "mean_token_accuracy": 0.25675587058067323,
      "num_tokens": 153279298.0,
      "step": 66850
    },
    {
      "entropy": 5.523614358901978,
      "epoch": 6.422190201729107,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017996716858269246,
      "loss": 4.6298,
      "mean_token_accuracy": 0.2660213425755501,
      "num_tokens": 153290758.0,
      "step": 66855
    },
    {
      "entropy": 5.533997392654419,
      "epoch": 6.42267050912584,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017993609423551355,
      "loss": 4.6241,
      "mean_token_accuracy": 0.2656128257513046,
      "num_tokens": 153301924.0,
      "step": 66860
    },
    {
      "entropy": 5.559253311157226,
      "epoch": 6.423150816522575,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001799050220954363,
      "loss": 4.6054,
      "mean_token_accuracy": 0.2614621758460999,
      "num_tokens": 153312823.0,
      "step": 66865
    },
    {
      "entropy": 5.506985807418824,
      "epoch": 6.423631123919308,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017987395216318203,
      "loss": 4.6306,
      "mean_token_accuracy": 0.26247299164533616,
      "num_tokens": 153324992.0,
      "step": 66870
    },
    {
      "entropy": 5.544868612289429,
      "epoch": 6.424111431316042,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017984288443947212,
      "loss": 4.6379,
      "mean_token_accuracy": 0.2579761743545532,
      "num_tokens": 153335334.0,
      "step": 66875
    },
    {
      "entropy": 5.597603178024292,
      "epoch": 6.424591738712776,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000179811818925028,
      "loss": 4.6384,
      "mean_token_accuracy": 0.26547804325819013,
      "num_tokens": 153345889.0,
      "step": 66880
    },
    {
      "entropy": 5.565538930892944,
      "epoch": 6.42507204610951,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017978075562057066,
      "loss": 4.6494,
      "mean_token_accuracy": 0.26409745812416074,
      "num_tokens": 153356936.0,
      "step": 66885
    },
    {
      "entropy": 5.622570133209228,
      "epoch": 6.425552353506244,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001797496945268215,
      "loss": 4.7022,
      "mean_token_accuracy": 0.25950473845005034,
      "num_tokens": 153368260.0,
      "step": 66890
    },
    {
      "entropy": 5.556840515136718,
      "epoch": 6.426032660902978,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017971863564450151,
      "loss": 4.6483,
      "mean_token_accuracy": 0.2693421259522438,
      "num_tokens": 153380582.0,
      "step": 66895
    },
    {
      "entropy": 5.559394121170044,
      "epoch": 6.426512968299712,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017968757897433195,
      "loss": 4.6531,
      "mean_token_accuracy": 0.26566313952207565,
      "num_tokens": 153392075.0,
      "step": 66900
    },
    {
      "entropy": 5.539916849136352,
      "epoch": 6.426993275696446,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017965652451703374,
      "loss": 4.6205,
      "mean_token_accuracy": 0.2735665336251259,
      "num_tokens": 153404291.0,
      "step": 66905
    },
    {
      "entropy": 5.587897539138794,
      "epoch": 6.42747358309318,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017962547227332805,
      "loss": 4.6844,
      "mean_token_accuracy": 0.26467376947402954,
      "num_tokens": 153416296.0,
      "step": 66910
    },
    {
      "entropy": 5.626646900177002,
      "epoch": 6.427953890489913,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017959442224393558,
      "loss": 4.7164,
      "mean_token_accuracy": 0.2550771817564964,
      "num_tokens": 153428271.0,
      "step": 66915
    },
    {
      "entropy": 5.483522224426269,
      "epoch": 6.428434197886648,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001795633744295773,
      "loss": 4.5579,
      "mean_token_accuracy": 0.27532594949007033,
      "num_tokens": 153439674.0,
      "step": 66920
    },
    {
      "entropy": 5.5502386569976805,
      "epoch": 6.428914505283381,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001795323288309741,
      "loss": 4.6313,
      "mean_token_accuracy": 0.2645795553922653,
      "num_tokens": 153451185.0,
      "step": 66925
    },
    {
      "entropy": 5.483292102813721,
      "epoch": 6.4293948126801155,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017950128544884676,
      "loss": 4.567,
      "mean_token_accuracy": 0.275115829706192,
      "num_tokens": 153462156.0,
      "step": 66930
    },
    {
      "entropy": 5.54263825416565,
      "epoch": 6.429875120076849,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017947024428391595,
      "loss": 4.5923,
      "mean_token_accuracy": 0.2674952119588852,
      "num_tokens": 153473697.0,
      "step": 66935
    },
    {
      "entropy": 5.544269943237305,
      "epoch": 6.430355427473583,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017943920533690255,
      "loss": 4.6035,
      "mean_token_accuracy": 0.2656805217266083,
      "num_tokens": 153484806.0,
      "step": 66940
    },
    {
      "entropy": 5.532512426376343,
      "epoch": 6.430835734870317,
      "grad_norm": 1.375,
      "learning_rate": 0.00017940816860852692,
      "loss": 4.6005,
      "mean_token_accuracy": 0.2722944959998131,
      "num_tokens": 153495654.0,
      "step": 66945
    },
    {
      "entropy": 5.533222198486328,
      "epoch": 6.431316042267051,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00017937713409950978,
      "loss": 4.6112,
      "mean_token_accuracy": 0.2692937433719635,
      "num_tokens": 153505473.0,
      "step": 66950
    },
    {
      "entropy": 5.4476165771484375,
      "epoch": 6.431796349663784,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017934610181057165,
      "loss": 4.5646,
      "mean_token_accuracy": 0.2754641309380531,
      "num_tokens": 153516553.0,
      "step": 66955
    },
    {
      "entropy": 5.539608526229858,
      "epoch": 6.432276657060519,
      "grad_norm": 1.34375,
      "learning_rate": 0.000179315071742433,
      "loss": 4.6349,
      "mean_token_accuracy": 0.2666369676589966,
      "num_tokens": 153527523.0,
      "step": 66960
    },
    {
      "entropy": 5.582952404022217,
      "epoch": 6.432756964457253,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017928404389581427,
      "loss": 4.6577,
      "mean_token_accuracy": 0.26418003141880037,
      "num_tokens": 153539024.0,
      "step": 66965
    },
    {
      "entropy": 5.607680034637451,
      "epoch": 6.4332372718539865,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017925301827143586,
      "loss": 4.756,
      "mean_token_accuracy": 0.2546795606613159,
      "num_tokens": 153549596.0,
      "step": 66970
    },
    {
      "entropy": 5.609267616271973,
      "epoch": 6.433717579250721,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000179221994870018,
      "loss": 4.707,
      "mean_token_accuracy": 0.2550155147910118,
      "num_tokens": 153560274.0,
      "step": 66975
    },
    {
      "entropy": 5.5053504467010494,
      "epoch": 6.434197886647454,
      "grad_norm": 1.421875,
      "learning_rate": 0.000179190973692281,
      "loss": 4.6225,
      "mean_token_accuracy": 0.26355891227722167,
      "num_tokens": 153571561.0,
      "step": 66980
    },
    {
      "entropy": 5.52706618309021,
      "epoch": 6.434678194044189,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001791599547389451,
      "loss": 4.6165,
      "mean_token_accuracy": 0.27131480276584624,
      "num_tokens": 153582773.0,
      "step": 66985
    },
    {
      "entropy": 5.600967216491699,
      "epoch": 6.435158501440922,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001791289380107305,
      "loss": 4.7737,
      "mean_token_accuracy": 0.2591385766863823,
      "num_tokens": 153596016.0,
      "step": 66990
    },
    {
      "entropy": 5.562388324737549,
      "epoch": 6.435638808837656,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017909792350835724,
      "loss": 4.6434,
      "mean_token_accuracy": 0.26289332062006,
      "num_tokens": 153608146.0,
      "step": 66995
    },
    {
      "entropy": 5.614917039871216,
      "epoch": 6.43611911623439,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017906691123254552,
      "loss": 4.7481,
      "mean_token_accuracy": 0.2598439916968346,
      "num_tokens": 153621359.0,
      "step": 67000
    },
    {
      "entropy": 5.554977226257324,
      "epoch": 6.436599423631124,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017903590118401518,
      "loss": 4.6283,
      "mean_token_accuracy": 0.2711853995919228,
      "num_tokens": 153633823.0,
      "step": 67005
    },
    {
      "entropy": 5.5345580101013185,
      "epoch": 6.437079731027858,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001790048933634863,
      "loss": 4.6424,
      "mean_token_accuracy": 0.25879622548818587,
      "num_tokens": 153645062.0,
      "step": 67010
    },
    {
      "entropy": 5.5343475341796875,
      "epoch": 6.437560038424592,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017897388777167868,
      "loss": 4.5719,
      "mean_token_accuracy": 0.2680335834622383,
      "num_tokens": 153656177.0,
      "step": 67015
    },
    {
      "entropy": 5.54300708770752,
      "epoch": 6.438040345821325,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017894288440931227,
      "loss": 4.6364,
      "mean_token_accuracy": 0.27489208430051804,
      "num_tokens": 153666411.0,
      "step": 67020
    },
    {
      "entropy": 5.587017297744751,
      "epoch": 6.43852065321806,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001789118832771069,
      "loss": 4.6449,
      "mean_token_accuracy": 0.2613487899303436,
      "num_tokens": 153678373.0,
      "step": 67025
    },
    {
      "entropy": 5.53976149559021,
      "epoch": 6.439000960614793,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017888088437578232,
      "loss": 4.5794,
      "mean_token_accuracy": 0.26649772971868513,
      "num_tokens": 153689723.0,
      "step": 67030
    },
    {
      "entropy": 5.44994101524353,
      "epoch": 6.439481268011527,
      "grad_norm": 1.5,
      "learning_rate": 0.0001788498877060581,
      "loss": 4.5625,
      "mean_token_accuracy": 0.274663382768631,
      "num_tokens": 153699877.0,
      "step": 67035
    },
    {
      "entropy": 5.514984273910523,
      "epoch": 6.439961575408262,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017881889326865397,
      "loss": 4.6467,
      "mean_token_accuracy": 0.2621260270476341,
      "num_tokens": 153709940.0,
      "step": 67040
    },
    {
      "entropy": 5.560617351531983,
      "epoch": 6.440441882804995,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001787879010642896,
      "loss": 4.638,
      "mean_token_accuracy": 0.2637738913297653,
      "num_tokens": 153721043.0,
      "step": 67045
    },
    {
      "entropy": 5.558673477172851,
      "epoch": 6.4409221902017295,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001787569110936844,
      "loss": 4.6439,
      "mean_token_accuracy": 0.2683815285563469,
      "num_tokens": 153731611.0,
      "step": 67050
    },
    {
      "entropy": 5.5768145561218265,
      "epoch": 6.441402497598463,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000178725923357558,
      "loss": 4.6506,
      "mean_token_accuracy": 0.2674273759126663,
      "num_tokens": 153742015.0,
      "step": 67055
    },
    {
      "entropy": 5.483465337753296,
      "epoch": 6.441882804995197,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017869493785662983,
      "loss": 4.6179,
      "mean_token_accuracy": 0.26824810206890104,
      "num_tokens": 153752890.0,
      "step": 67060
    },
    {
      "entropy": 5.473289918899536,
      "epoch": 6.442363112391931,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017866395459161915,
      "loss": 4.537,
      "mean_token_accuracy": 0.2707737311720848,
      "num_tokens": 153764257.0,
      "step": 67065
    },
    {
      "entropy": 5.530167865753174,
      "epoch": 6.442843419788665,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017863297356324542,
      "loss": 4.6286,
      "mean_token_accuracy": 0.26680032908916473,
      "num_tokens": 153776295.0,
      "step": 67070
    },
    {
      "entropy": 5.495985603332519,
      "epoch": 6.4433237271853985,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017860199477222788,
      "loss": 4.6083,
      "mean_token_accuracy": 0.2685881108045578,
      "num_tokens": 153787185.0,
      "step": 67075
    },
    {
      "entropy": 5.5243308544158936,
      "epoch": 6.443804034582133,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001785710182192858,
      "loss": 4.64,
      "mean_token_accuracy": 0.26617230772972106,
      "num_tokens": 153799413.0,
      "step": 67080
    },
    {
      "entropy": 5.547128200531006,
      "epoch": 6.444284341978866,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017854004390513821,
      "loss": 4.5934,
      "mean_token_accuracy": 0.2706089556217194,
      "num_tokens": 153810036.0,
      "step": 67085
    },
    {
      "entropy": 5.5174360275268555,
      "epoch": 6.444764649375601,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017850907183050463,
      "loss": 4.5469,
      "mean_token_accuracy": 0.2628898903727531,
      "num_tokens": 153821098.0,
      "step": 67090
    },
    {
      "entropy": 5.569192361831665,
      "epoch": 6.445244956772334,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001784781019961037,
      "loss": 4.6235,
      "mean_token_accuracy": 0.266573403775692,
      "num_tokens": 153833121.0,
      "step": 67095
    },
    {
      "entropy": 5.546654796600341,
      "epoch": 6.445725264169068,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001784471344026547,
      "loss": 4.5852,
      "mean_token_accuracy": 0.2642074391245842,
      "num_tokens": 153845218.0,
      "step": 67100
    },
    {
      "entropy": 5.5585103034973145,
      "epoch": 6.446205571565802,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001784161690508765,
      "loss": 4.6424,
      "mean_token_accuracy": 0.2675533264875412,
      "num_tokens": 153856071.0,
      "step": 67105
    },
    {
      "entropy": 5.569208192825317,
      "epoch": 6.446685878962536,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017838520594148805,
      "loss": 4.7027,
      "mean_token_accuracy": 0.26038738489151003,
      "num_tokens": 153866721.0,
      "step": 67110
    },
    {
      "entropy": 5.553293609619141,
      "epoch": 6.4471661863592695,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001783542450752082,
      "loss": 4.6614,
      "mean_token_accuracy": 0.2596336901187897,
      "num_tokens": 153877017.0,
      "step": 67115
    },
    {
      "entropy": 5.479728412628174,
      "epoch": 6.447646493756004,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017832328645275592,
      "loss": 4.589,
      "mean_token_accuracy": 0.26347724795341493,
      "num_tokens": 153888115.0,
      "step": 67120
    },
    {
      "entropy": 5.581163120269776,
      "epoch": 6.448126801152737,
      "grad_norm": 1.5,
      "learning_rate": 0.00017829233007484978,
      "loss": 4.675,
      "mean_token_accuracy": 0.26370707899332047,
      "num_tokens": 153899015.0,
      "step": 67125
    },
    {
      "entropy": 5.534974765777588,
      "epoch": 6.448607108549472,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017826137594220855,
      "loss": 4.6443,
      "mean_token_accuracy": 0.26484186202287674,
      "num_tokens": 153911209.0,
      "step": 67130
    },
    {
      "entropy": 5.552356624603272,
      "epoch": 6.449087415946206,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017823042405555096,
      "loss": 4.5911,
      "mean_token_accuracy": 0.2690120816230774,
      "num_tokens": 153922271.0,
      "step": 67135
    },
    {
      "entropy": 5.524063301086426,
      "epoch": 6.449567723342939,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017819947441559552,
      "loss": 4.6042,
      "mean_token_accuracy": 0.26411831974983213,
      "num_tokens": 153935002.0,
      "step": 67140
    },
    {
      "entropy": 5.507423782348633,
      "epoch": 6.450048030739674,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017816852702306086,
      "loss": 4.6124,
      "mean_token_accuracy": 0.2642873257398605,
      "num_tokens": 153947163.0,
      "step": 67145
    },
    {
      "entropy": 5.576019763946533,
      "epoch": 6.450528338136407,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017813758187866564,
      "loss": 4.6326,
      "mean_token_accuracy": 0.26178126633167265,
      "num_tokens": 153957950.0,
      "step": 67150
    },
    {
      "entropy": 5.511912393569946,
      "epoch": 6.4510086455331415,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017810663898312794,
      "loss": 4.5785,
      "mean_token_accuracy": 0.2692445576190948,
      "num_tokens": 153969774.0,
      "step": 67155
    },
    {
      "entropy": 5.500415229797364,
      "epoch": 6.451488952929875,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017807569833716647,
      "loss": 4.5513,
      "mean_token_accuracy": 0.27271131575107577,
      "num_tokens": 153981052.0,
      "step": 67160
    },
    {
      "entropy": 5.555708408355713,
      "epoch": 6.451969260326609,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017804475994149938,
      "loss": 4.6235,
      "mean_token_accuracy": 0.26152700930833817,
      "num_tokens": 153991648.0,
      "step": 67165
    },
    {
      "entropy": 5.586214017868042,
      "epoch": 6.452449567723343,
      "grad_norm": 1.578125,
      "learning_rate": 0.00017801382379684517,
      "loss": 4.6445,
      "mean_token_accuracy": 0.25900267362594603,
      "num_tokens": 154001653.0,
      "step": 67170
    },
    {
      "entropy": 5.660612869262695,
      "epoch": 6.452929875120077,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001779828899039219,
      "loss": 4.7188,
      "mean_token_accuracy": 0.2581408262252808,
      "num_tokens": 154013581.0,
      "step": 67175
    },
    {
      "entropy": 5.568455171585083,
      "epoch": 6.45341018251681,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017795195826344784,
      "loss": 4.6779,
      "mean_token_accuracy": 0.26668906956911087,
      "num_tokens": 154025928.0,
      "step": 67180
    },
    {
      "entropy": 5.515613460540772,
      "epoch": 6.453890489913545,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001779210288761412,
      "loss": 4.626,
      "mean_token_accuracy": 0.2668168947100639,
      "num_tokens": 154037746.0,
      "step": 67185
    },
    {
      "entropy": 5.606559085845947,
      "epoch": 6.454370797310278,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001778901017427201,
      "loss": 4.6913,
      "mean_token_accuracy": 0.2648675739765167,
      "num_tokens": 154049353.0,
      "step": 67190
    },
    {
      "entropy": 5.623862934112549,
      "epoch": 6.4548511047070125,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017785917686390234,
      "loss": 4.714,
      "mean_token_accuracy": 0.25709168612957,
      "num_tokens": 154060233.0,
      "step": 67195
    },
    {
      "entropy": 5.548217010498047,
      "epoch": 6.455331412103746,
      "grad_norm": 1.5625,
      "learning_rate": 0.00017782825424040606,
      "loss": 4.6344,
      "mean_token_accuracy": 0.2689301520586014,
      "num_tokens": 154071273.0,
      "step": 67200
    },
    {
      "entropy": 5.526430082321167,
      "epoch": 6.45581171950048,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00017779733387294916,
      "loss": 4.5977,
      "mean_token_accuracy": 0.2728703022003174,
      "num_tokens": 154082433.0,
      "step": 67205
    },
    {
      "entropy": 5.5529622554779055,
      "epoch": 6.456292026897215,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017776641576224954,
      "loss": 4.7016,
      "mean_token_accuracy": 0.2571434110403061,
      "num_tokens": 154095060.0,
      "step": 67210
    },
    {
      "entropy": 5.548617792129517,
      "epoch": 6.456772334293948,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017773549990902506,
      "loss": 4.6328,
      "mean_token_accuracy": 0.25763437896966934,
      "num_tokens": 154108014.0,
      "step": 67215
    },
    {
      "entropy": 5.647882270812988,
      "epoch": 6.457252641690682,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001777045863139934,
      "loss": 4.6884,
      "mean_token_accuracy": 0.25868615210056306,
      "num_tokens": 154119459.0,
      "step": 67220
    },
    {
      "entropy": 5.558073568344116,
      "epoch": 6.457732949087416,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017767367497787233,
      "loss": 4.6848,
      "mean_token_accuracy": 0.25662347227334975,
      "num_tokens": 154132305.0,
      "step": 67225
    },
    {
      "entropy": 5.466446161270142,
      "epoch": 6.45821325648415,
      "grad_norm": 1.375,
      "learning_rate": 0.0001776427659013797,
      "loss": 4.5479,
      "mean_token_accuracy": 0.2738269537687302,
      "num_tokens": 154142873.0,
      "step": 67230
    },
    {
      "entropy": 5.5377521991729735,
      "epoch": 6.458693563880884,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017761185908523282,
      "loss": 4.6135,
      "mean_token_accuracy": 0.26233538538217543,
      "num_tokens": 154154664.0,
      "step": 67235
    },
    {
      "entropy": 5.582519960403443,
      "epoch": 6.459173871277618,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017758095453014933,
      "loss": 4.6702,
      "mean_token_accuracy": 0.2669148623943329,
      "num_tokens": 154166257.0,
      "step": 67240
    },
    {
      "entropy": 5.572408294677734,
      "epoch": 6.459654178674351,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017755005223684688,
      "loss": 4.6716,
      "mean_token_accuracy": 0.259099505841732,
      "num_tokens": 154177034.0,
      "step": 67245
    },
    {
      "entropy": 5.527988624572754,
      "epoch": 6.460134486071086,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001775191522060428,
      "loss": 4.6429,
      "mean_token_accuracy": 0.2615524545311928,
      "num_tokens": 154189087.0,
      "step": 67250
    },
    {
      "entropy": 5.542248964309692,
      "epoch": 6.460614793467819,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017748825443845458,
      "loss": 4.6781,
      "mean_token_accuracy": 0.26314410865306853,
      "num_tokens": 154200328.0,
      "step": 67255
    },
    {
      "entropy": 5.570888900756836,
      "epoch": 6.461095100864553,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017745735893479959,
      "loss": 4.6661,
      "mean_token_accuracy": 0.261383980512619,
      "num_tokens": 154211689.0,
      "step": 67260
    },
    {
      "entropy": 5.635349082946777,
      "epoch": 6.461575408261287,
      "grad_norm": 1.421875,
      "learning_rate": 0.000177426465695795,
      "loss": 4.6941,
      "mean_token_accuracy": 0.255916491150856,
      "num_tokens": 154222615.0,
      "step": 67265
    },
    {
      "entropy": 5.565058040618896,
      "epoch": 6.462055715658021,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017739557472215822,
      "loss": 4.6165,
      "mean_token_accuracy": 0.2670889303088188,
      "num_tokens": 154235111.0,
      "step": 67270
    },
    {
      "entropy": 5.545385074615479,
      "epoch": 6.462536023054755,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017736468601460632,
      "loss": 4.6154,
      "mean_token_accuracy": 0.2740654319524765,
      "num_tokens": 154245198.0,
      "step": 67275
    },
    {
      "entropy": 5.549708414077759,
      "epoch": 6.463016330451489,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017733379957385654,
      "loss": 4.611,
      "mean_token_accuracy": 0.2663852646946907,
      "num_tokens": 154256249.0,
      "step": 67280
    },
    {
      "entropy": 5.506419610977173,
      "epoch": 6.463496637848223,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017730291540062588,
      "loss": 4.6116,
      "mean_token_accuracy": 0.27277105748653413,
      "num_tokens": 154267115.0,
      "step": 67285
    },
    {
      "entropy": 5.498484802246094,
      "epoch": 6.463976945244957,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017727203349563136,
      "loss": 4.6191,
      "mean_token_accuracy": 0.2622630804777145,
      "num_tokens": 154278533.0,
      "step": 67290
    },
    {
      "entropy": 5.468664169311523,
      "epoch": 6.464457252641691,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001772411538595901,
      "loss": 4.6147,
      "mean_token_accuracy": 0.2669155403971672,
      "num_tokens": 154290575.0,
      "step": 67295
    },
    {
      "entropy": 5.57002067565918,
      "epoch": 6.4649375600384245,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017721027649321907,
      "loss": 4.6984,
      "mean_token_accuracy": 0.26266343593597413,
      "num_tokens": 154302859.0,
      "step": 67300
    },
    {
      "entropy": 5.576344633102417,
      "epoch": 6.465417867435159,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017717940139723494,
      "loss": 4.6778,
      "mean_token_accuracy": 0.26303054988384245,
      "num_tokens": 154314777.0,
      "step": 67305
    },
    {
      "entropy": 5.601679134368896,
      "epoch": 6.465898174831892,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017714852857235463,
      "loss": 4.6879,
      "mean_token_accuracy": 0.26471823304891584,
      "num_tokens": 154326203.0,
      "step": 67310
    },
    {
      "entropy": 5.638621854782104,
      "epoch": 6.4663784822286265,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017711765801929491,
      "loss": 4.6953,
      "mean_token_accuracy": 0.2529295653104782,
      "num_tokens": 154338767.0,
      "step": 67315
    },
    {
      "entropy": 5.579273986816406,
      "epoch": 6.46685878962536,
      "grad_norm": 1.375,
      "learning_rate": 0.00017708678973877258,
      "loss": 4.6481,
      "mean_token_accuracy": 0.2560842618346214,
      "num_tokens": 154349759.0,
      "step": 67320
    },
    {
      "entropy": 5.592537260055542,
      "epoch": 6.467339097022094,
      "grad_norm": 1.5,
      "learning_rate": 0.0001770559237315042,
      "loss": 4.7108,
      "mean_token_accuracy": 0.2590688481926918,
      "num_tokens": 154361302.0,
      "step": 67325
    },
    {
      "entropy": 5.546665859222412,
      "epoch": 6.467819404418828,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001770250599982065,
      "loss": 4.6307,
      "mean_token_accuracy": 0.2690189152956009,
      "num_tokens": 154372843.0,
      "step": 67330
    },
    {
      "entropy": 5.5253763675689695,
      "epoch": 6.468299711815562,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017699419853959593,
      "loss": 4.6164,
      "mean_token_accuracy": 0.27152280062437056,
      "num_tokens": 154383239.0,
      "step": 67335
    },
    {
      "entropy": 5.5195629596710205,
      "epoch": 6.4687800192122955,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017696333935638902,
      "loss": 4.6425,
      "mean_token_accuracy": 0.2665128901600838,
      "num_tokens": 154394708.0,
      "step": 67340
    },
    {
      "entropy": 5.458021259307861,
      "epoch": 6.46926032660903,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017693248244930233,
      "loss": 4.5799,
      "mean_token_accuracy": 0.2717313066124916,
      "num_tokens": 154406795.0,
      "step": 67345
    },
    {
      "entropy": 5.584364748001098,
      "epoch": 6.469740634005763,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017690162781905212,
      "loss": 4.6653,
      "mean_token_accuracy": 0.2646311327815056,
      "num_tokens": 154418929.0,
      "step": 67350
    },
    {
      "entropy": 5.606611251831055,
      "epoch": 6.470220941402498,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017687077546635487,
      "loss": 4.7365,
      "mean_token_accuracy": 0.2525714337825775,
      "num_tokens": 154430658.0,
      "step": 67355
    },
    {
      "entropy": 5.528782224655151,
      "epoch": 6.470701248799232,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017683992539192688,
      "loss": 4.6075,
      "mean_token_accuracy": 0.2672928750514984,
      "num_tokens": 154442095.0,
      "step": 67360
    },
    {
      "entropy": 5.547919511795044,
      "epoch": 6.471181556195965,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001768090775964843,
      "loss": 4.5793,
      "mean_token_accuracy": 0.2651969015598297,
      "num_tokens": 154453593.0,
      "step": 67365
    },
    {
      "entropy": 5.579581689834595,
      "epoch": 6.4716618635927,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001767782320807434,
      "loss": 4.6862,
      "mean_token_accuracy": 0.2618436127901077,
      "num_tokens": 154466123.0,
      "step": 67370
    },
    {
      "entropy": 5.530952501296997,
      "epoch": 6.472142170989433,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017674738884542018,
      "loss": 4.6203,
      "mean_token_accuracy": 0.2659789055585861,
      "num_tokens": 154477433.0,
      "step": 67375
    },
    {
      "entropy": 5.481872463226319,
      "epoch": 6.472622478386167,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017671654789123094,
      "loss": 4.5889,
      "mean_token_accuracy": 0.267284682393074,
      "num_tokens": 154487891.0,
      "step": 67380
    },
    {
      "entropy": 5.517545318603515,
      "epoch": 6.473102785782901,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001766857092188916,
      "loss": 4.5368,
      "mean_token_accuracy": 0.2755782574415207,
      "num_tokens": 154499994.0,
      "step": 67385
    },
    {
      "entropy": 5.542771673202514,
      "epoch": 6.473583093179635,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001766548728291182,
      "loss": 4.6787,
      "mean_token_accuracy": 0.267144538462162,
      "num_tokens": 154511535.0,
      "step": 67390
    },
    {
      "entropy": 5.607212162017822,
      "epoch": 6.474063400576369,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017662403872262657,
      "loss": 4.6934,
      "mean_token_accuracy": 0.268142381310463,
      "num_tokens": 154523471.0,
      "step": 67395
    },
    {
      "entropy": 5.507665395736694,
      "epoch": 6.474543707973103,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017659320690013264,
      "loss": 4.611,
      "mean_token_accuracy": 0.26430396884679797,
      "num_tokens": 154533937.0,
      "step": 67400
    },
    {
      "entropy": 5.579110097885132,
      "epoch": 6.475024015369836,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017656237736235226,
      "loss": 4.6973,
      "mean_token_accuracy": 0.2604848727583885,
      "num_tokens": 154545877.0,
      "step": 67405
    },
    {
      "entropy": 5.512388086318969,
      "epoch": 6.475504322766571,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017653155011000115,
      "loss": 4.5598,
      "mean_token_accuracy": 0.274099637567997,
      "num_tokens": 154557089.0,
      "step": 67410
    },
    {
      "entropy": 5.446384859085083,
      "epoch": 6.475984630163304,
      "grad_norm": 1.375,
      "learning_rate": 0.00017650072514379506,
      "loss": 4.506,
      "mean_token_accuracy": 0.27388349622488023,
      "num_tokens": 154569422.0,
      "step": 67415
    },
    {
      "entropy": 5.53115406036377,
      "epoch": 6.4764649375600385,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00017646990246444972,
      "loss": 4.6132,
      "mean_token_accuracy": 0.2653500363230705,
      "num_tokens": 154580720.0,
      "step": 67420
    },
    {
      "entropy": 5.559924077987671,
      "epoch": 6.476945244956772,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017643908207268062,
      "loss": 4.6423,
      "mean_token_accuracy": 0.2605688855051994,
      "num_tokens": 154592866.0,
      "step": 67425
    },
    {
      "entropy": 5.506445360183716,
      "epoch": 6.477425552353506,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017640826396920334,
      "loss": 4.5818,
      "mean_token_accuracy": 0.267400224506855,
      "num_tokens": 154604548.0,
      "step": 67430
    },
    {
      "entropy": 5.534429836273193,
      "epoch": 6.477905859750241,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017637744815473345,
      "loss": 4.6228,
      "mean_token_accuracy": 0.2684492200613022,
      "num_tokens": 154615609.0,
      "step": 67435
    },
    {
      "entropy": 5.559779930114746,
      "epoch": 6.478386167146974,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017634663462998633,
      "loss": 4.626,
      "mean_token_accuracy": 0.2695632457733154,
      "num_tokens": 154627868.0,
      "step": 67440
    },
    {
      "entropy": 5.5749608993530275,
      "epoch": 6.478866474543708,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017631582339567743,
      "loss": 4.6564,
      "mean_token_accuracy": 0.2608323931694031,
      "num_tokens": 154638522.0,
      "step": 67445
    },
    {
      "entropy": 5.596755313873291,
      "epoch": 6.479346781940442,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017628501445252214,
      "loss": 4.6887,
      "mean_token_accuracy": 0.25558072328567505,
      "num_tokens": 154650060.0,
      "step": 67450
    },
    {
      "entropy": 5.59959397315979,
      "epoch": 6.479827089337176,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017625420780123562,
      "loss": 4.6874,
      "mean_token_accuracy": 0.26012349873781204,
      "num_tokens": 154660773.0,
      "step": 67455
    },
    {
      "entropy": 5.559816980361939,
      "epoch": 6.4803073967339095,
      "grad_norm": 1.375,
      "learning_rate": 0.00017622340344253324,
      "loss": 4.7235,
      "mean_token_accuracy": 0.2582228481769562,
      "num_tokens": 154672306.0,
      "step": 67460
    },
    {
      "entropy": 5.472349691390991,
      "epoch": 6.480787704130644,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017619260137713,
      "loss": 4.6153,
      "mean_token_accuracy": 0.26457690745592116,
      "num_tokens": 154683930.0,
      "step": 67465
    },
    {
      "entropy": 5.585435009002685,
      "epoch": 6.481268011527377,
      "grad_norm": 1.578125,
      "learning_rate": 0.00017616180160574124,
      "loss": 4.6043,
      "mean_token_accuracy": 0.2635687798261642,
      "num_tokens": 154694678.0,
      "step": 67470
    },
    {
      "entropy": 5.6063148021698,
      "epoch": 6.481748318924112,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000176131004129082,
      "loss": 4.6009,
      "mean_token_accuracy": 0.27000546306371687,
      "num_tokens": 154704828.0,
      "step": 67475
    },
    {
      "entropy": 5.55430417060852,
      "epoch": 6.482228626320845,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017610020894786728,
      "loss": 4.6555,
      "mean_token_accuracy": 0.2661376729607582,
      "num_tokens": 154716462.0,
      "step": 67480
    },
    {
      "entropy": 5.498303985595703,
      "epoch": 6.482708933717579,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017606941606281197,
      "loss": 4.6538,
      "mean_token_accuracy": 0.26065070927143097,
      "num_tokens": 154728055.0,
      "step": 67485
    },
    {
      "entropy": 5.466910457611084,
      "epoch": 6.483189241114313,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017603862547463107,
      "loss": 4.5473,
      "mean_token_accuracy": 0.2678948312997818,
      "num_tokens": 154739584.0,
      "step": 67490
    },
    {
      "entropy": 5.542935419082641,
      "epoch": 6.483669548511047,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001760078371840394,
      "loss": 4.6658,
      "mean_token_accuracy": 0.2586966440081596,
      "num_tokens": 154750936.0,
      "step": 67495
    },
    {
      "entropy": 5.484054660797119,
      "epoch": 6.484149855907781,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017597705119175183,
      "loss": 4.5903,
      "mean_token_accuracy": 0.2726525843143463,
      "num_tokens": 154763293.0,
      "step": 67500
    },
    {
      "entropy": 5.5755438804626465,
      "epoch": 6.484630163304515,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001759462674984831,
      "loss": 4.6683,
      "mean_token_accuracy": 0.2586237072944641,
      "num_tokens": 154775143.0,
      "step": 67505
    },
    {
      "entropy": 5.568549489974975,
      "epoch": 6.485110470701249,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017591548610494795,
      "loss": 4.6659,
      "mean_token_accuracy": 0.2678973123431206,
      "num_tokens": 154785268.0,
      "step": 67510
    },
    {
      "entropy": 5.4849790096282955,
      "epoch": 6.485590778097983,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017588470701186093,
      "loss": 4.5385,
      "mean_token_accuracy": 0.2780755668878555,
      "num_tokens": 154798082.0,
      "step": 67515
    },
    {
      "entropy": 5.510526895523071,
      "epoch": 6.486071085494717,
      "grad_norm": 1.625,
      "learning_rate": 0.00017585393021993668,
      "loss": 4.6442,
      "mean_token_accuracy": 0.2662848591804504,
      "num_tokens": 154810171.0,
      "step": 67520
    },
    {
      "entropy": 5.594816780090332,
      "epoch": 6.48655139289145,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017582315572988976,
      "loss": 4.6607,
      "mean_token_accuracy": 0.26143784523010255,
      "num_tokens": 154821692.0,
      "step": 67525
    },
    {
      "entropy": 5.5294637203216555,
      "epoch": 6.487031700288185,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017579238354243476,
      "loss": 4.643,
      "mean_token_accuracy": 0.26770407259464263,
      "num_tokens": 154833267.0,
      "step": 67530
    },
    {
      "entropy": 5.560405111312866,
      "epoch": 6.487512007684918,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017576161365828597,
      "loss": 4.6973,
      "mean_token_accuracy": 0.2565897524356842,
      "num_tokens": 154844435.0,
      "step": 67535
    },
    {
      "entropy": 5.52112774848938,
      "epoch": 6.4879923150816525,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001757308460781578,
      "loss": 4.6486,
      "mean_token_accuracy": 0.2667852446436882,
      "num_tokens": 154855976.0,
      "step": 67540
    },
    {
      "entropy": 5.568934631347656,
      "epoch": 6.488472622478386,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017570008080276472,
      "loss": 4.6599,
      "mean_token_accuracy": 0.26136032193899156,
      "num_tokens": 154867053.0,
      "step": 67545
    },
    {
      "entropy": 5.558664703369141,
      "epoch": 6.48895292987512,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017566931783282087,
      "loss": 4.6163,
      "mean_token_accuracy": 0.26399491280317305,
      "num_tokens": 154877618.0,
      "step": 67550
    },
    {
      "entropy": 5.6001218318939205,
      "epoch": 6.489433237271854,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001756385571690405,
      "loss": 4.6984,
      "mean_token_accuracy": 0.25858682096004487,
      "num_tokens": 154887451.0,
      "step": 67555
    },
    {
      "entropy": 5.596087026596069,
      "epoch": 6.489913544668588,
      "grad_norm": 1.65625,
      "learning_rate": 0.00017560779881213773,
      "loss": 4.6867,
      "mean_token_accuracy": 0.26085562258958817,
      "num_tokens": 154898504.0,
      "step": 67560
    },
    {
      "entropy": 5.544309806823731,
      "epoch": 6.4903938520653215,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017557704276282676,
      "loss": 4.653,
      "mean_token_accuracy": 0.2590549647808075,
      "num_tokens": 154910064.0,
      "step": 67565
    },
    {
      "entropy": 5.514814996719361,
      "epoch": 6.490874159462056,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017554628902182162,
      "loss": 4.5794,
      "mean_token_accuracy": 0.2662051975727081,
      "num_tokens": 154921519.0,
      "step": 67570
    },
    {
      "entropy": 5.555480241775513,
      "epoch": 6.491354466858789,
      "grad_norm": 1.375,
      "learning_rate": 0.00017551553758983647,
      "loss": 4.6221,
      "mean_token_accuracy": 0.2659731149673462,
      "num_tokens": 154933065.0,
      "step": 67575
    },
    {
      "entropy": 5.5169471263885494,
      "epoch": 6.491834774255524,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017548478846758498,
      "loss": 4.5959,
      "mean_token_accuracy": 0.27349289655685427,
      "num_tokens": 154944793.0,
      "step": 67580
    },
    {
      "entropy": 5.560365152359009,
      "epoch": 6.492315081652258,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017545404165578122,
      "loss": 4.6638,
      "mean_token_accuracy": 0.2583503365516663,
      "num_tokens": 154955155.0,
      "step": 67585
    },
    {
      "entropy": 5.589911794662475,
      "epoch": 6.492795389048991,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017542329715513905,
      "loss": 4.6804,
      "mean_token_accuracy": 0.26396137326955793,
      "num_tokens": 154966542.0,
      "step": 67590
    },
    {
      "entropy": 5.4468201160430905,
      "epoch": 6.493275696445726,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017539255496637225,
      "loss": 4.4894,
      "mean_token_accuracy": 0.27338055819272994,
      "num_tokens": 154976390.0,
      "step": 67595
    },
    {
      "entropy": 5.469383430480957,
      "epoch": 6.493756003842459,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001753618150901945,
      "loss": 4.6069,
      "mean_token_accuracy": 0.2673968136310577,
      "num_tokens": 154987888.0,
      "step": 67600
    },
    {
      "entropy": 5.509808921813965,
      "epoch": 6.494236311239193,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017533107752731964,
      "loss": 4.6006,
      "mean_token_accuracy": 0.2696466699242592,
      "num_tokens": 154999602.0,
      "step": 67605
    },
    {
      "entropy": 5.5107005596160885,
      "epoch": 6.494716618635927,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017530034227846114,
      "loss": 4.6334,
      "mean_token_accuracy": 0.26352624893188475,
      "num_tokens": 155011172.0,
      "step": 67610
    },
    {
      "entropy": 5.569770336151123,
      "epoch": 6.495196926032661,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017526960934433262,
      "loss": 4.6962,
      "mean_token_accuracy": 0.2563869208097458,
      "num_tokens": 155023037.0,
      "step": 67615
    },
    {
      "entropy": 5.434990739822387,
      "epoch": 6.495677233429395,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017523887872564764,
      "loss": 4.528,
      "mean_token_accuracy": 0.27745865285396576,
      "num_tokens": 155035796.0,
      "step": 67620
    },
    {
      "entropy": 5.493896102905273,
      "epoch": 6.496157540826129,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017520815042311967,
      "loss": 4.59,
      "mean_token_accuracy": 0.27658569514751435,
      "num_tokens": 155047920.0,
      "step": 67625
    },
    {
      "entropy": 5.560424518585205,
      "epoch": 6.496637848222862,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017517742443746214,
      "loss": 4.6922,
      "mean_token_accuracy": 0.267881965637207,
      "num_tokens": 155058550.0,
      "step": 67630
    },
    {
      "entropy": 5.610449409484863,
      "epoch": 6.497118155619597,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017514670076938845,
      "loss": 4.6589,
      "mean_token_accuracy": 0.25769285708665846,
      "num_tokens": 155068471.0,
      "step": 67635
    },
    {
      "entropy": 5.511415624618531,
      "epoch": 6.49759846301633,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017511597941961184,
      "loss": 4.5917,
      "mean_token_accuracy": 0.270668987929821,
      "num_tokens": 155079822.0,
      "step": 67640
    },
    {
      "entropy": 5.529954147338867,
      "epoch": 6.4980787704130645,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017508526038884553,
      "loss": 4.6652,
      "mean_token_accuracy": 0.26270727068185806,
      "num_tokens": 155091763.0,
      "step": 67645
    },
    {
      "entropy": 5.510094404220581,
      "epoch": 6.498559077809798,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017505454367780282,
      "loss": 4.5541,
      "mean_token_accuracy": 0.269744499027729,
      "num_tokens": 155102192.0,
      "step": 67650
    },
    {
      "entropy": 5.594233226776123,
      "epoch": 6.499039385206532,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017502382928719678,
      "loss": 4.7706,
      "mean_token_accuracy": 0.26055672764778137,
      "num_tokens": 155114758.0,
      "step": 67655
    },
    {
      "entropy": 5.490715503692627,
      "epoch": 6.499519692603266,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017499311721774064,
      "loss": 4.5719,
      "mean_token_accuracy": 0.27186083048582077,
      "num_tokens": 155126645.0,
      "step": 67660
    },
    {
      "entropy": 5.46464319229126,
      "epoch": 6.5,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017496240747014738,
      "loss": 4.5103,
      "mean_token_accuracy": 0.2759902536869049,
      "num_tokens": 155137414.0,
      "step": 67665
    },
    {
      "entropy": 5.424281406402588,
      "epoch": 6.500480307396733,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017493170004512993,
      "loss": 4.4987,
      "mean_token_accuracy": 0.27706995606422424,
      "num_tokens": 155147876.0,
      "step": 67670
    },
    {
      "entropy": 5.532167625427246,
      "epoch": 6.500960614793468,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017490099494340124,
      "loss": 4.6607,
      "mean_token_accuracy": 0.2604469209909439,
      "num_tokens": 155161043.0,
      "step": 67675
    },
    {
      "entropy": 5.559666156768799,
      "epoch": 6.501440922190202,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017487029216567418,
      "loss": 4.6514,
      "mean_token_accuracy": 0.26328386217355726,
      "num_tokens": 155174018.0,
      "step": 67680
    },
    {
      "entropy": 5.518641328811645,
      "epoch": 6.5019212295869355,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017483959171266168,
      "loss": 4.5498,
      "mean_token_accuracy": 0.2698654443025589,
      "num_tokens": 155184850.0,
      "step": 67685
    },
    {
      "entropy": 5.5464221954345705,
      "epoch": 6.50240153698367,
      "grad_norm": 1.5,
      "learning_rate": 0.0001748088935850764,
      "loss": 4.6463,
      "mean_token_accuracy": 0.2724409341812134,
      "num_tokens": 155197035.0,
      "step": 67690
    },
    {
      "entropy": 5.554757738113404,
      "epoch": 6.502881844380403,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001747781977836312,
      "loss": 4.6617,
      "mean_token_accuracy": 0.26266122311353685,
      "num_tokens": 155208029.0,
      "step": 67695
    },
    {
      "entropy": 5.517664432525635,
      "epoch": 6.503362151777138,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017474750430903857,
      "loss": 4.6279,
      "mean_token_accuracy": 0.2671022072434425,
      "num_tokens": 155219277.0,
      "step": 67700
    },
    {
      "entropy": 5.585539197921753,
      "epoch": 6.503842459173871,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001747168131620112,
      "loss": 4.6885,
      "mean_token_accuracy": 0.25450124144554137,
      "num_tokens": 155230459.0,
      "step": 67705
    },
    {
      "entropy": 5.521709251403808,
      "epoch": 6.504322766570605,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001746861243432616,
      "loss": 4.6264,
      "mean_token_accuracy": 0.2724839895963669,
      "num_tokens": 155243163.0,
      "step": 67710
    },
    {
      "entropy": 5.504795551300049,
      "epoch": 6.504803073967339,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017465543785350238,
      "loss": 4.5689,
      "mean_token_accuracy": 0.2751519948244095,
      "num_tokens": 155254286.0,
      "step": 67715
    },
    {
      "entropy": 5.529293012619019,
      "epoch": 6.505283381364073,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017462475369344594,
      "loss": 4.6276,
      "mean_token_accuracy": 0.2684641435742378,
      "num_tokens": 155264892.0,
      "step": 67720
    },
    {
      "entropy": 5.517968654632568,
      "epoch": 6.505763688760807,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001745940718638047,
      "loss": 4.6195,
      "mean_token_accuracy": 0.26450823098421095,
      "num_tokens": 155276946.0,
      "step": 67725
    },
    {
      "entropy": 5.5482994556427006,
      "epoch": 6.506243996157541,
      "grad_norm": 1.375,
      "learning_rate": 0.0001745633923652909,
      "loss": 4.6856,
      "mean_token_accuracy": 0.2583813637495041,
      "num_tokens": 155290153.0,
      "step": 67730
    },
    {
      "entropy": 5.478854942321777,
      "epoch": 6.506724303554275,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001745327151986169,
      "loss": 4.5612,
      "mean_token_accuracy": 0.2717957064509392,
      "num_tokens": 155302113.0,
      "step": 67735
    },
    {
      "entropy": 5.527215480804443,
      "epoch": 6.507204610951009,
      "grad_norm": 1.515625,
      "learning_rate": 0.00017450204036449497,
      "loss": 4.6254,
      "mean_token_accuracy": 0.2672765046358109,
      "num_tokens": 155313238.0,
      "step": 67740
    },
    {
      "entropy": 5.617434787750244,
      "epoch": 6.507684918347742,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017447136786363722,
      "loss": 4.7059,
      "mean_token_accuracy": 0.25791052877902987,
      "num_tokens": 155324246.0,
      "step": 67745
    },
    {
      "entropy": 5.6016294956207275,
      "epoch": 6.508165225744476,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017444069769675575,
      "loss": 4.6801,
      "mean_token_accuracy": 0.26343217492103577,
      "num_tokens": 155336848.0,
      "step": 67750
    },
    {
      "entropy": 5.581860065460205,
      "epoch": 6.508645533141211,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017441002986456273,
      "loss": 4.6195,
      "mean_token_accuracy": 0.27189686000347135,
      "num_tokens": 155347932.0,
      "step": 67755
    },
    {
      "entropy": 5.511019039154053,
      "epoch": 6.509125840537944,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001743793643677701,
      "loss": 4.5842,
      "mean_token_accuracy": 0.26663699448108674,
      "num_tokens": 155359151.0,
      "step": 67760
    },
    {
      "entropy": 5.525539827346802,
      "epoch": 6.5096061479346785,
      "grad_norm": 1.375,
      "learning_rate": 0.00017434870120708993,
      "loss": 4.6098,
      "mean_token_accuracy": 0.26362535655498504,
      "num_tokens": 155371026.0,
      "step": 67765
    },
    {
      "entropy": 5.518289613723755,
      "epoch": 6.510086455331412,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017431804038323402,
      "loss": 4.6341,
      "mean_token_accuracy": 0.25807793736457824,
      "num_tokens": 155382080.0,
      "step": 67770
    },
    {
      "entropy": 5.490946626663208,
      "epoch": 6.510566762728146,
      "grad_norm": 1.453125,
      "learning_rate": 0.00017428738189691418,
      "loss": 4.5894,
      "mean_token_accuracy": 0.2682444155216217,
      "num_tokens": 155393479.0,
      "step": 67775
    },
    {
      "entropy": 5.529736661911011,
      "epoch": 6.51104707012488,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017425672574884228,
      "loss": 4.6425,
      "mean_token_accuracy": 0.2640245959162712,
      "num_tokens": 155405642.0,
      "step": 67780
    },
    {
      "entropy": 5.475625324249267,
      "epoch": 6.511527377521614,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00017422607193973006,
      "loss": 4.5172,
      "mean_token_accuracy": 0.2727954387664795,
      "num_tokens": 155416589.0,
      "step": 67785
    },
    {
      "entropy": 5.567519903182983,
      "epoch": 6.5120076849183475,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017419542047028924,
      "loss": 4.6029,
      "mean_token_accuracy": 0.264498707652092,
      "num_tokens": 155427482.0,
      "step": 67790
    },
    {
      "entropy": 5.565842199325561,
      "epoch": 6.512487992315082,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017416477134123143,
      "loss": 4.5906,
      "mean_token_accuracy": 0.2698490276932716,
      "num_tokens": 155439822.0,
      "step": 67795
    },
    {
      "entropy": 5.488489818572998,
      "epoch": 6.512968299711815,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001741341245532682,
      "loss": 4.5389,
      "mean_token_accuracy": 0.27276863157749176,
      "num_tokens": 155451084.0,
      "step": 67800
    },
    {
      "entropy": 5.497422409057617,
      "epoch": 6.51344860710855,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017410348010711118,
      "loss": 4.6008,
      "mean_token_accuracy": 0.2610479682683945,
      "num_tokens": 155462169.0,
      "step": 67805
    },
    {
      "entropy": 5.550080728530884,
      "epoch": 6.513928914505283,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017407283800347165,
      "loss": 4.6201,
      "mean_token_accuracy": 0.2746462419629097,
      "num_tokens": 155475123.0,
      "step": 67810
    },
    {
      "entropy": 5.498548412322998,
      "epoch": 6.514409221902017,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017404219824306117,
      "loss": 4.5814,
      "mean_token_accuracy": 0.26959064453840254,
      "num_tokens": 155486192.0,
      "step": 67815
    },
    {
      "entropy": 5.57283124923706,
      "epoch": 6.514889529298751,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017401156082659102,
      "loss": 4.6169,
      "mean_token_accuracy": 0.27020419389009476,
      "num_tokens": 155496491.0,
      "step": 67820
    },
    {
      "entropy": 5.59962420463562,
      "epoch": 6.515369836695485,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017398092575477258,
      "loss": 4.6943,
      "mean_token_accuracy": 0.2591943323612213,
      "num_tokens": 155508447.0,
      "step": 67825
    },
    {
      "entropy": 5.509496259689331,
      "epoch": 6.515850144092219,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017395029302831705,
      "loss": 4.6549,
      "mean_token_accuracy": 0.2627624779939651,
      "num_tokens": 155518850.0,
      "step": 67830
    },
    {
      "entropy": 5.506197261810303,
      "epoch": 6.516330451488953,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017391966264793568,
      "loss": 4.6054,
      "mean_token_accuracy": 0.27365767359733584,
      "num_tokens": 155530891.0,
      "step": 67835
    },
    {
      "entropy": 5.508092164993286,
      "epoch": 6.516810758885687,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017388903461433957,
      "loss": 4.6609,
      "mean_token_accuracy": 0.2623383328318596,
      "num_tokens": 155542599.0,
      "step": 67840
    },
    {
      "entropy": 5.549112606048584,
      "epoch": 6.517291066282421,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017385840892823994,
      "loss": 4.6561,
      "mean_token_accuracy": 0.27138647735118865,
      "num_tokens": 155554791.0,
      "step": 67845
    },
    {
      "entropy": 5.614932489395142,
      "epoch": 6.517771373679155,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017382778559034767,
      "loss": 4.6834,
      "mean_token_accuracy": 0.25991224199533464,
      "num_tokens": 155566241.0,
      "step": 67850
    },
    {
      "entropy": 5.464278793334961,
      "epoch": 6.518251681075888,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017379716460137377,
      "loss": 4.5655,
      "mean_token_accuracy": 0.27078562080860136,
      "num_tokens": 155577201.0,
      "step": 67855
    },
    {
      "entropy": 5.590586471557617,
      "epoch": 6.518731988472623,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001737665459620292,
      "loss": 4.6966,
      "mean_token_accuracy": 0.25843224972486495,
      "num_tokens": 155589170.0,
      "step": 67860
    },
    {
      "entropy": 5.566259384155273,
      "epoch": 6.519212295869356,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017373592967302482,
      "loss": 4.7052,
      "mean_token_accuracy": 0.2549753084778786,
      "num_tokens": 155601136.0,
      "step": 67865
    },
    {
      "entropy": 5.61032395362854,
      "epoch": 6.5196926032660905,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001737053157350715,
      "loss": 4.7299,
      "mean_token_accuracy": 0.2559367373585701,
      "num_tokens": 155612646.0,
      "step": 67870
    },
    {
      "entropy": 5.536223649978638,
      "epoch": 6.520172910662824,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017367470414888004,
      "loss": 4.612,
      "mean_token_accuracy": 0.26385392993688583,
      "num_tokens": 155624403.0,
      "step": 67875
    },
    {
      "entropy": 5.573596382141114,
      "epoch": 6.520653218059558,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017364409491516094,
      "loss": 4.6547,
      "mean_token_accuracy": 0.26333710700273516,
      "num_tokens": 155635704.0,
      "step": 67880
    },
    {
      "entropy": 5.512998771667481,
      "epoch": 6.521133525456292,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017361348803462503,
      "loss": 4.6383,
      "mean_token_accuracy": 0.2646451696753502,
      "num_tokens": 155648081.0,
      "step": 67885
    },
    {
      "entropy": 5.477452516555786,
      "epoch": 6.521613832853026,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001735828835079829,
      "loss": 4.5314,
      "mean_token_accuracy": 0.27508158087730405,
      "num_tokens": 155658777.0,
      "step": 67890
    },
    {
      "entropy": 5.550367975234986,
      "epoch": 6.522094140249759,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017355228133594504,
      "loss": 4.6314,
      "mean_token_accuracy": 0.26448087990283964,
      "num_tokens": 155670262.0,
      "step": 67895
    },
    {
      "entropy": 5.533347415924072,
      "epoch": 6.522574447646494,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00017352168151922194,
      "loss": 4.6265,
      "mean_token_accuracy": 0.2615464672446251,
      "num_tokens": 155681813.0,
      "step": 67900
    },
    {
      "entropy": 5.4850745677948,
      "epoch": 6.523054755043228,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017349108405852415,
      "loss": 4.5837,
      "mean_token_accuracy": 0.2659697115421295,
      "num_tokens": 155694122.0,
      "step": 67905
    },
    {
      "entropy": 5.6408754825592045,
      "epoch": 6.5235350624399615,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017346048895456193,
      "loss": 4.7237,
      "mean_token_accuracy": 0.2589033991098404,
      "num_tokens": 155705141.0,
      "step": 67910
    },
    {
      "entropy": 5.524169540405273,
      "epoch": 6.524015369836696,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017342989620804564,
      "loss": 4.5418,
      "mean_token_accuracy": 0.2716947212815285,
      "num_tokens": 155716095.0,
      "step": 67915
    },
    {
      "entropy": 5.509609174728394,
      "epoch": 6.524495677233429,
      "grad_norm": 1.375,
      "learning_rate": 0.00017339930581968555,
      "loss": 4.6454,
      "mean_token_accuracy": 0.2589000672101974,
      "num_tokens": 155729118.0,
      "step": 67920
    },
    {
      "entropy": 5.495949363708496,
      "epoch": 6.524975984630164,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017336871779019187,
      "loss": 4.6057,
      "mean_token_accuracy": 0.26864875108003616,
      "num_tokens": 155739564.0,
      "step": 67925
    },
    {
      "entropy": 5.479231309890747,
      "epoch": 6.525456292026897,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017333813212027478,
      "loss": 4.6257,
      "mean_token_accuracy": 0.2654461249709129,
      "num_tokens": 155752225.0,
      "step": 67930
    },
    {
      "entropy": 5.552003049850464,
      "epoch": 6.525936599423631,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017330754881064448,
      "loss": 4.6834,
      "mean_token_accuracy": 0.2582033649086952,
      "num_tokens": 155764641.0,
      "step": 67935
    },
    {
      "entropy": 5.516009998321533,
      "epoch": 6.526416906820365,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017327696786201087,
      "loss": 4.5987,
      "mean_token_accuracy": 0.27070930004119875,
      "num_tokens": 155775182.0,
      "step": 67940
    },
    {
      "entropy": 5.5388706684112545,
      "epoch": 6.526897214217099,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017324638927508398,
      "loss": 4.6387,
      "mean_token_accuracy": 0.2595921203494072,
      "num_tokens": 155786829.0,
      "step": 67945
    },
    {
      "entropy": 5.563060092926025,
      "epoch": 6.527377521613833,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017321581305057383,
      "loss": 4.6489,
      "mean_token_accuracy": 0.26519507318735125,
      "num_tokens": 155798116.0,
      "step": 67950
    },
    {
      "entropy": 5.618709135055542,
      "epoch": 6.527857829010567,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017318523918919023,
      "loss": 4.7911,
      "mean_token_accuracy": 0.25685528814792635,
      "num_tokens": 155809701.0,
      "step": 67955
    },
    {
      "entropy": 5.601193046569824,
      "epoch": 6.5283381364073,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017315466769164304,
      "loss": 4.7063,
      "mean_token_accuracy": 0.2567026600241661,
      "num_tokens": 155821203.0,
      "step": 67960
    },
    {
      "entropy": 5.537944698333741,
      "epoch": 6.528818443804035,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00017312409855864213,
      "loss": 4.5707,
      "mean_token_accuracy": 0.26579972058534623,
      "num_tokens": 155833524.0,
      "step": 67965
    },
    {
      "entropy": 5.550785779953003,
      "epoch": 6.529298751200768,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001730935317908971,
      "loss": 4.711,
      "mean_token_accuracy": 0.2596357211470604,
      "num_tokens": 155845465.0,
      "step": 67970
    },
    {
      "entropy": 5.486488485336304,
      "epoch": 6.529779058597502,
      "grad_norm": 1.53125,
      "learning_rate": 0.00017306296738911763,
      "loss": 4.5903,
      "mean_token_accuracy": 0.2678314059972763,
      "num_tokens": 155856108.0,
      "step": 67975
    },
    {
      "entropy": 5.574110651016236,
      "epoch": 6.530259365994237,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017303240535401337,
      "loss": 4.7074,
      "mean_token_accuracy": 0.2612038403749466,
      "num_tokens": 155867587.0,
      "step": 67980
    },
    {
      "entropy": 5.506482315063477,
      "epoch": 6.53073967339097,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017300184568629385,
      "loss": 4.5634,
      "mean_token_accuracy": 0.27728813886642456,
      "num_tokens": 155878037.0,
      "step": 67985
    },
    {
      "entropy": 5.511858034133911,
      "epoch": 6.5312199807877045,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017297128838666864,
      "loss": 4.6134,
      "mean_token_accuracy": 0.2710674285888672,
      "num_tokens": 155889431.0,
      "step": 67990
    },
    {
      "entropy": 5.5135256290435795,
      "epoch": 6.531700288184438,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001729407334558472,
      "loss": 4.6005,
      "mean_token_accuracy": 0.2745009809732437,
      "num_tokens": 155900658.0,
      "step": 67995
    },
    {
      "entropy": 5.459759283065796,
      "epoch": 6.532180595581172,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001729101808945388,
      "loss": 4.5678,
      "mean_token_accuracy": 0.2733603149652481,
      "num_tokens": 155911657.0,
      "step": 68000
    },
    {
      "entropy": 5.544164562225342,
      "epoch": 6.532660902977906,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017287963070345292,
      "loss": 4.6634,
      "mean_token_accuracy": 0.25784466564655306,
      "num_tokens": 155923376.0,
      "step": 68005
    },
    {
      "entropy": 5.524686288833618,
      "epoch": 6.53314121037464,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001728490828832987,
      "loss": 4.6312,
      "mean_token_accuracy": 0.2661459296941757,
      "num_tokens": 155934781.0,
      "step": 68010
    },
    {
      "entropy": 5.58263931274414,
      "epoch": 6.5336215177713735,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001728185374347855,
      "loss": 4.6611,
      "mean_token_accuracy": 0.2629502534866333,
      "num_tokens": 155945310.0,
      "step": 68015
    },
    {
      "entropy": 5.588219833374024,
      "epoch": 6.534101825168108,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017278799435862242,
      "loss": 4.7075,
      "mean_token_accuracy": 0.26060733050107954,
      "num_tokens": 155956743.0,
      "step": 68020
    },
    {
      "entropy": 5.524760389328003,
      "epoch": 6.534582132564841,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017275745365551864,
      "loss": 4.5715,
      "mean_token_accuracy": 0.27187318354845047,
      "num_tokens": 155967806.0,
      "step": 68025
    },
    {
      "entropy": 5.535934972763061,
      "epoch": 6.5350624399615755,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017272691532618317,
      "loss": 4.5868,
      "mean_token_accuracy": 0.2693667247891426,
      "num_tokens": 155980575.0,
      "step": 68030
    },
    {
      "entropy": 5.563226842880249,
      "epoch": 6.535542747358309,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017269637937132503,
      "loss": 4.6441,
      "mean_token_accuracy": 0.2643414169549942,
      "num_tokens": 155991275.0,
      "step": 68035
    },
    {
      "entropy": 5.64766902923584,
      "epoch": 6.536023054755043,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017266584579165317,
      "loss": 4.6946,
      "mean_token_accuracy": 0.255583593249321,
      "num_tokens": 156002448.0,
      "step": 68040
    },
    {
      "entropy": 5.495423030853272,
      "epoch": 6.536503362151777,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001726353145878765,
      "loss": 4.6392,
      "mean_token_accuracy": 0.27112081199884414,
      "num_tokens": 156013777.0,
      "step": 68045
    },
    {
      "entropy": 5.547961711883545,
      "epoch": 6.536983669548511,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017260478576070387,
      "loss": 4.6173,
      "mean_token_accuracy": 0.26885559856891633,
      "num_tokens": 156023682.0,
      "step": 68050
    },
    {
      "entropy": 5.5495484352111815,
      "epoch": 6.537463976945245,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017257425931084414,
      "loss": 4.5834,
      "mean_token_accuracy": 0.27707321345806124,
      "num_tokens": 156035355.0,
      "step": 68055
    },
    {
      "entropy": 5.561171436309815,
      "epoch": 6.537944284341979,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017254373523900587,
      "loss": 4.6139,
      "mean_token_accuracy": 0.26930879056453705,
      "num_tokens": 156046486.0,
      "step": 68060
    },
    {
      "entropy": 5.475201797485352,
      "epoch": 6.538424591738713,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017251321354589788,
      "loss": 4.601,
      "mean_token_accuracy": 0.2692804381251335,
      "num_tokens": 156057394.0,
      "step": 68065
    },
    {
      "entropy": 5.529809236526489,
      "epoch": 6.538904899135447,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001724826942322287,
      "loss": 4.6554,
      "mean_token_accuracy": 0.2588812351226807,
      "num_tokens": 156069661.0,
      "step": 68070
    },
    {
      "entropy": 5.5212544918060305,
      "epoch": 6.539385206532181,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000172452177298707,
      "loss": 4.612,
      "mean_token_accuracy": 0.2636133745312691,
      "num_tokens": 156080825.0,
      "step": 68075
    },
    {
      "entropy": 5.560997581481933,
      "epoch": 6.539865513928914,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001724216627460412,
      "loss": 4.597,
      "mean_token_accuracy": 0.27012866288423537,
      "num_tokens": 156091982.0,
      "step": 68080
    },
    {
      "entropy": 5.574079084396362,
      "epoch": 6.540345821325649,
      "grad_norm": 1.5,
      "learning_rate": 0.00017239115057493987,
      "loss": 4.6826,
      "mean_token_accuracy": 0.25823543071746824,
      "num_tokens": 156102664.0,
      "step": 68085
    },
    {
      "entropy": 5.58871603012085,
      "epoch": 6.540826128722382,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017236064078611128,
      "loss": 4.7394,
      "mean_token_accuracy": 0.2617736905813217,
      "num_tokens": 156113998.0,
      "step": 68090
    },
    {
      "entropy": 5.554805088043213,
      "epoch": 6.541306436119116,
      "grad_norm": 1.375,
      "learning_rate": 0.00017233013338026384,
      "loss": 4.6242,
      "mean_token_accuracy": 0.2671095699071884,
      "num_tokens": 156125525.0,
      "step": 68095
    },
    {
      "entropy": 5.534127807617187,
      "epoch": 6.54178674351585,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001722996283581058,
      "loss": 4.6151,
      "mean_token_accuracy": 0.27006075978279115,
      "num_tokens": 156135652.0,
      "step": 68100
    },
    {
      "entropy": 5.516559839248657,
      "epoch": 6.542267050912584,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017226912572034544,
      "loss": 4.6226,
      "mean_token_accuracy": 0.2694422841072083,
      "num_tokens": 156147204.0,
      "step": 68105
    },
    {
      "entropy": 5.479529666900635,
      "epoch": 6.542747358309318,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017223862546769095,
      "loss": 4.5983,
      "mean_token_accuracy": 0.26859467327594755,
      "num_tokens": 156158624.0,
      "step": 68110
    },
    {
      "entropy": 5.54798903465271,
      "epoch": 6.543227665706052,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017220812760085051,
      "loss": 4.6788,
      "mean_token_accuracy": 0.267075414955616,
      "num_tokens": 156170275.0,
      "step": 68115
    },
    {
      "entropy": 5.520459938049316,
      "epoch": 6.543707973102785,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017217763212053205,
      "loss": 4.6488,
      "mean_token_accuracy": 0.25624485313892365,
      "num_tokens": 156181746.0,
      "step": 68120
    },
    {
      "entropy": 5.489982414245605,
      "epoch": 6.54418828049952,
      "grad_norm": 1.5,
      "learning_rate": 0.00017214713902744365,
      "loss": 4.5576,
      "mean_token_accuracy": 0.2708659589290619,
      "num_tokens": 156191950.0,
      "step": 68125
    },
    {
      "entropy": 5.574555778503418,
      "epoch": 6.544668587896254,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00017211664832229327,
      "loss": 4.6649,
      "mean_token_accuracy": 0.25619874596595765,
      "num_tokens": 156202338.0,
      "step": 68130
    },
    {
      "entropy": 5.5769411563873295,
      "epoch": 6.5451488952929875,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00017208616000578884,
      "loss": 4.6488,
      "mean_token_accuracy": 0.26635689437389376,
      "num_tokens": 156212613.0,
      "step": 68135
    },
    {
      "entropy": 5.548558902740479,
      "epoch": 6.545629202689721,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017205567407863813,
      "loss": 4.6714,
      "mean_token_accuracy": 0.25950454771518705,
      "num_tokens": 156224773.0,
      "step": 68140
    },
    {
      "entropy": 5.478812742233276,
      "epoch": 6.546109510086455,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017202519054154909,
      "loss": 4.5793,
      "mean_token_accuracy": 0.2769743949174881,
      "num_tokens": 156235820.0,
      "step": 68145
    },
    {
      "entropy": 5.491660118103027,
      "epoch": 6.54658981748319,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017199470939522927,
      "loss": 4.5559,
      "mean_token_accuracy": 0.27477896213531494,
      "num_tokens": 156247273.0,
      "step": 68150
    },
    {
      "entropy": 5.515896129608154,
      "epoch": 6.547070124879923,
      "grad_norm": 1.375,
      "learning_rate": 0.00017196423064038646,
      "loss": 4.6084,
      "mean_token_accuracy": 0.2623714402318001,
      "num_tokens": 156259295.0,
      "step": 68155
    },
    {
      "entropy": 5.54987268447876,
      "epoch": 6.547550432276657,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017193375427772823,
      "loss": 4.6407,
      "mean_token_accuracy": 0.26543592363595964,
      "num_tokens": 156270128.0,
      "step": 68160
    },
    {
      "entropy": 5.4455554485321045,
      "epoch": 6.548030739673391,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001719032803079622,
      "loss": 4.519,
      "mean_token_accuracy": 0.2786490634083748,
      "num_tokens": 156282788.0,
      "step": 68165
    },
    {
      "entropy": 5.5463909149169925,
      "epoch": 6.548511047070125,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017187280873179583,
      "loss": 4.6392,
      "mean_token_accuracy": 0.2652574822306633,
      "num_tokens": 156293643.0,
      "step": 68170
    },
    {
      "entropy": 5.501885318756104,
      "epoch": 6.5489913544668585,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001718423395499367,
      "loss": 4.6099,
      "mean_token_accuracy": 0.2643371492624283,
      "num_tokens": 156305071.0,
      "step": 68175
    },
    {
      "entropy": 5.552946329116821,
      "epoch": 6.549471661863593,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001718118727630921,
      "loss": 4.5958,
      "mean_token_accuracy": 0.2629432827234268,
      "num_tokens": 156315395.0,
      "step": 68180
    },
    {
      "entropy": 5.602946376800537,
      "epoch": 6.549951969260326,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001717814083719694,
      "loss": 4.719,
      "mean_token_accuracy": 0.2586641117930412,
      "num_tokens": 156326360.0,
      "step": 68185
    },
    {
      "entropy": 5.55166220664978,
      "epoch": 6.550432276657061,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001717509463772759,
      "loss": 4.6748,
      "mean_token_accuracy": 0.2637017861008644,
      "num_tokens": 156337958.0,
      "step": 68190
    },
    {
      "entropy": 5.564894580841065,
      "epoch": 6.550912584053794,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001717204867797188,
      "loss": 4.6361,
      "mean_token_accuracy": 0.2693284720182419,
      "num_tokens": 156349024.0,
      "step": 68195
    },
    {
      "entropy": 5.575805044174194,
      "epoch": 6.551392891450528,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017169002958000533,
      "loss": 4.6849,
      "mean_token_accuracy": 0.2639420241117477,
      "num_tokens": 156360155.0,
      "step": 68200
    },
    {
      "entropy": 5.473696041107178,
      "epoch": 6.551873198847263,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017165957477884272,
      "loss": 4.6086,
      "mean_token_accuracy": 0.27037582397460935,
      "num_tokens": 156371638.0,
      "step": 68205
    },
    {
      "entropy": 5.526395082473755,
      "epoch": 6.552353506243996,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017162912237693784,
      "loss": 4.6412,
      "mean_token_accuracy": 0.25935757011175153,
      "num_tokens": 156383895.0,
      "step": 68210
    },
    {
      "entropy": 5.534279155731201,
      "epoch": 6.55283381364073,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017159867237499777,
      "loss": 4.6687,
      "mean_token_accuracy": 0.26778786480426786,
      "num_tokens": 156397540.0,
      "step": 68215
    },
    {
      "entropy": 5.3677609920501705,
      "epoch": 6.553314121037464,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001715682247737295,
      "loss": 4.4772,
      "mean_token_accuracy": 0.2744288980960846,
      "num_tokens": 156407745.0,
      "step": 68220
    },
    {
      "entropy": 5.499505138397216,
      "epoch": 6.553794428434198,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001715377795738399,
      "loss": 4.6539,
      "mean_token_accuracy": 0.2694837749004364,
      "num_tokens": 156418842.0,
      "step": 68225
    },
    {
      "entropy": 5.499263095855713,
      "epoch": 6.554274735830932,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017150733677603585,
      "loss": 4.5741,
      "mean_token_accuracy": 0.27816465944051744,
      "num_tokens": 156431065.0,
      "step": 68230
    },
    {
      "entropy": 5.5568907260894775,
      "epoch": 6.554755043227666,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001714768963810242,
      "loss": 4.6477,
      "mean_token_accuracy": 0.259365639090538,
      "num_tokens": 156444205.0,
      "step": 68235
    },
    {
      "entropy": 5.67081241607666,
      "epoch": 6.555235350624399,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017144645838951155,
      "loss": 4.732,
      "mean_token_accuracy": 0.2581228852272034,
      "num_tokens": 156455377.0,
      "step": 68240
    },
    {
      "entropy": 5.608661556243897,
      "epoch": 6.555715658021134,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017141602280220468,
      "loss": 4.6607,
      "mean_token_accuracy": 0.2578150376677513,
      "num_tokens": 156466050.0,
      "step": 68245
    },
    {
      "entropy": 5.482802391052246,
      "epoch": 6.556195965417867,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017138558961981014,
      "loss": 4.5215,
      "mean_token_accuracy": 0.2706895172595978,
      "num_tokens": 156477597.0,
      "step": 68250
    },
    {
      "entropy": 5.497862863540649,
      "epoch": 6.5566762728146015,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017135515884303455,
      "loss": 4.6199,
      "mean_token_accuracy": 0.2633408397436142,
      "num_tokens": 156489505.0,
      "step": 68255
    },
    {
      "entropy": 5.485463237762451,
      "epoch": 6.557156580211335,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001713247304725844,
      "loss": 4.6013,
      "mean_token_accuracy": 0.2693441569805145,
      "num_tokens": 156499740.0,
      "step": 68260
    },
    {
      "entropy": 5.508782052993775,
      "epoch": 6.557636887608069,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00017129430450916623,
      "loss": 4.63,
      "mean_token_accuracy": 0.2650704011321068,
      "num_tokens": 156510439.0,
      "step": 68265
    },
    {
      "entropy": 5.594641256332397,
      "epoch": 6.558117195004803,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001712638809534863,
      "loss": 4.6833,
      "mean_token_accuracy": 0.2582543924450874,
      "num_tokens": 156521084.0,
      "step": 68270
    },
    {
      "entropy": 5.569587659835816,
      "epoch": 6.558597502401537,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017123345980625098,
      "loss": 4.6819,
      "mean_token_accuracy": 0.2660505801439285,
      "num_tokens": 156533216.0,
      "step": 68275
    },
    {
      "entropy": 5.577699613571167,
      "epoch": 6.559077809798271,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017120304106816664,
      "loss": 4.7301,
      "mean_token_accuracy": 0.25016123056411743,
      "num_tokens": 156543041.0,
      "step": 68280
    },
    {
      "entropy": 5.543671560287476,
      "epoch": 6.559558117195005,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001711726247399395,
      "loss": 4.6205,
      "mean_token_accuracy": 0.262258443236351,
      "num_tokens": 156554421.0,
      "step": 68285
    },
    {
      "entropy": 5.558346176147461,
      "epoch": 6.560038424591738,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017114221082227565,
      "loss": 4.6607,
      "mean_token_accuracy": 0.2676048129796982,
      "num_tokens": 156565900.0,
      "step": 68290
    },
    {
      "entropy": 5.485808324813843,
      "epoch": 6.560518731988473,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017111179931588138,
      "loss": 4.605,
      "mean_token_accuracy": 0.26922188699245453,
      "num_tokens": 156576984.0,
      "step": 68295
    },
    {
      "entropy": 5.574205017089843,
      "epoch": 6.560999039385207,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017108139022146258,
      "loss": 4.6871,
      "mean_token_accuracy": 0.2651195332407951,
      "num_tokens": 156588151.0,
      "step": 68300
    },
    {
      "entropy": 5.654614734649658,
      "epoch": 6.56147934678194,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00017105098353972531,
      "loss": 4.7495,
      "mean_token_accuracy": 0.25672087967395785,
      "num_tokens": 156600136.0,
      "step": 68305
    },
    {
      "entropy": 5.560108757019043,
      "epoch": 6.561959654178675,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017102057927137555,
      "loss": 4.662,
      "mean_token_accuracy": 0.25940441340208054,
      "num_tokens": 156612159.0,
      "step": 68310
    },
    {
      "entropy": 5.563811635971069,
      "epoch": 6.562439961575408,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017099017741711918,
      "loss": 4.5961,
      "mean_token_accuracy": 0.27109573781490326,
      "num_tokens": 156623861.0,
      "step": 68315
    },
    {
      "entropy": 5.56148886680603,
      "epoch": 6.562920268972142,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017095977797766202,
      "loss": 4.6185,
      "mean_token_accuracy": 0.2646134078502655,
      "num_tokens": 156634257.0,
      "step": 68320
    },
    {
      "entropy": 5.614301872253418,
      "epoch": 6.563400576368876,
      "grad_norm": 1.59375,
      "learning_rate": 0.00017092938095370985,
      "loss": 4.672,
      "mean_token_accuracy": 0.2730659648776054,
      "num_tokens": 156644921.0,
      "step": 68325
    },
    {
      "entropy": 5.619295740127564,
      "epoch": 6.56388088376561,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001708989863459685,
      "loss": 4.752,
      "mean_token_accuracy": 0.25914126038551333,
      "num_tokens": 156656671.0,
      "step": 68330
    },
    {
      "entropy": 5.594583129882812,
      "epoch": 6.564361191162344,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017086859415514361,
      "loss": 4.6335,
      "mean_token_accuracy": 0.2682907611131668,
      "num_tokens": 156667689.0,
      "step": 68335
    },
    {
      "entropy": 5.577763891220092,
      "epoch": 6.564841498559078,
      "grad_norm": 1.375,
      "learning_rate": 0.00017083820438194069,
      "loss": 4.6333,
      "mean_token_accuracy": 0.2709270179271698,
      "num_tokens": 156678693.0,
      "step": 68340
    },
    {
      "entropy": 5.4603087425231935,
      "epoch": 6.565321805955811,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001708078170270654,
      "loss": 4.5874,
      "mean_token_accuracy": 0.2630434975028038,
      "num_tokens": 156689991.0,
      "step": 68345
    },
    {
      "entropy": 5.515113782882691,
      "epoch": 6.565802113352546,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00017077743209122315,
      "loss": 4.7108,
      "mean_token_accuracy": 0.26553526520729065,
      "num_tokens": 156700332.0,
      "step": 68350
    },
    {
      "entropy": 5.5202234268188475,
      "epoch": 6.56628242074928,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001707470495751195,
      "loss": 4.6519,
      "mean_token_accuracy": 0.2607721403241158,
      "num_tokens": 156712284.0,
      "step": 68355
    },
    {
      "entropy": 5.574809455871582,
      "epoch": 6.5667627281460135,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017071666947945975,
      "loss": 4.6403,
      "mean_token_accuracy": 0.2620344489812851,
      "num_tokens": 156724203.0,
      "step": 68360
    },
    {
      "entropy": 5.521551704406738,
      "epoch": 6.567243035542747,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017068629180494926,
      "loss": 4.6383,
      "mean_token_accuracy": 0.26503159701824186,
      "num_tokens": 156737328.0,
      "step": 68365
    },
    {
      "entropy": 5.592178678512573,
      "epoch": 6.567723342939481,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001706559165522934,
      "loss": 4.7078,
      "mean_token_accuracy": 0.2548080861568451,
      "num_tokens": 156749260.0,
      "step": 68370
    },
    {
      "entropy": 5.437304878234864,
      "epoch": 6.568203650336216,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017062554372219735,
      "loss": 4.4621,
      "mean_token_accuracy": 0.2793129473924637,
      "num_tokens": 156760952.0,
      "step": 68375
    },
    {
      "entropy": 5.526464509963989,
      "epoch": 6.568683957732949,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017059517331536618,
      "loss": 4.5942,
      "mean_token_accuracy": 0.26514860838651655,
      "num_tokens": 156771391.0,
      "step": 68380
    },
    {
      "entropy": 5.499074983596802,
      "epoch": 6.569164265129683,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017056480533250506,
      "loss": 4.6444,
      "mean_token_accuracy": 0.27338754683732985,
      "num_tokens": 156782807.0,
      "step": 68385
    },
    {
      "entropy": 5.568312740325927,
      "epoch": 6.569644572526417,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017053443977431903,
      "loss": 4.6724,
      "mean_token_accuracy": 0.26504970490932467,
      "num_tokens": 156795428.0,
      "step": 68390
    },
    {
      "entropy": 5.561947298049927,
      "epoch": 6.570124879923151,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017050407664151318,
      "loss": 4.6012,
      "mean_token_accuracy": 0.26929829716682435,
      "num_tokens": 156806531.0,
      "step": 68395
    },
    {
      "entropy": 5.543382167816162,
      "epoch": 6.5706051873198845,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017047371593479233,
      "loss": 4.6386,
      "mean_token_accuracy": 0.2709738492965698,
      "num_tokens": 156819182.0,
      "step": 68400
    },
    {
      "entropy": 5.572841501235962,
      "epoch": 6.571085494716619,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001704433576548614,
      "loss": 4.715,
      "mean_token_accuracy": 0.2642086401581764,
      "num_tokens": 156830847.0,
      "step": 68405
    },
    {
      "entropy": 5.4857385635375975,
      "epoch": 6.571565802113352,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017041300180242524,
      "loss": 4.5949,
      "mean_token_accuracy": 0.2646335244178772,
      "num_tokens": 156842579.0,
      "step": 68410
    },
    {
      "entropy": 5.559309339523315,
      "epoch": 6.572046109510087,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017038264837818874,
      "loss": 4.6994,
      "mean_token_accuracy": 0.2589599430561066,
      "num_tokens": 156855827.0,
      "step": 68415
    },
    {
      "entropy": 5.570414447784424,
      "epoch": 6.57252641690682,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017035229738285638,
      "loss": 4.6338,
      "mean_token_accuracy": 0.2646701708436012,
      "num_tokens": 156867335.0,
      "step": 68420
    },
    {
      "entropy": 5.591010189056396,
      "epoch": 6.573006724303554,
      "grad_norm": 1.375,
      "learning_rate": 0.00017032194881713296,
      "loss": 4.647,
      "mean_token_accuracy": 0.2552787512540817,
      "num_tokens": 156879434.0,
      "step": 68425
    },
    {
      "entropy": 5.490574502944947,
      "epoch": 6.573487031700288,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017029160268172306,
      "loss": 4.5751,
      "mean_token_accuracy": 0.2720642194151878,
      "num_tokens": 156890797.0,
      "step": 68430
    },
    {
      "entropy": 5.5580878257751465,
      "epoch": 6.573967339097022,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001702612589773312,
      "loss": 4.6978,
      "mean_token_accuracy": 0.26283103972673416,
      "num_tokens": 156904045.0,
      "step": 68435
    },
    {
      "entropy": 5.617723751068115,
      "epoch": 6.574447646493756,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017023091770466188,
      "loss": 4.6873,
      "mean_token_accuracy": 0.261137031018734,
      "num_tokens": 156917165.0,
      "step": 68440
    },
    {
      "entropy": 5.609929466247559,
      "epoch": 6.57492795389049,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017020057886441969,
      "loss": 4.6058,
      "mean_token_accuracy": 0.2689911276102066,
      "num_tokens": 156927746.0,
      "step": 68445
    },
    {
      "entropy": 5.5428690910339355,
      "epoch": 6.575408261287224,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00017017024245730874,
      "loss": 4.5672,
      "mean_token_accuracy": 0.2726146072149277,
      "num_tokens": 156938525.0,
      "step": 68450
    },
    {
      "entropy": 5.506717872619629,
      "epoch": 6.575888568683958,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017013990848403348,
      "loss": 4.6536,
      "mean_token_accuracy": 0.2607381701469421,
      "num_tokens": 156950758.0,
      "step": 68455
    },
    {
      "entropy": 5.566954278945923,
      "epoch": 6.576368876080692,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00017010957694529822,
      "loss": 4.7087,
      "mean_token_accuracy": 0.2594147175550461,
      "num_tokens": 156961440.0,
      "step": 68460
    },
    {
      "entropy": 5.519272041320801,
      "epoch": 6.576849183477425,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017007924784180706,
      "loss": 4.5883,
      "mean_token_accuracy": 0.2722401976585388,
      "num_tokens": 156972957.0,
      "step": 68465
    },
    {
      "entropy": 5.616933536529541,
      "epoch": 6.57732949087416,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017004892117426426,
      "loss": 4.7674,
      "mean_token_accuracy": 0.25949208736419677,
      "num_tokens": 156983754.0,
      "step": 68470
    },
    {
      "entropy": 5.532073259353638,
      "epoch": 6.577809798270893,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000170018596943374,
      "loss": 4.6494,
      "mean_token_accuracy": 0.262882699072361,
      "num_tokens": 156994761.0,
      "step": 68475
    },
    {
      "entropy": 5.544251918792725,
      "epoch": 6.5782901056676275,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016998827514984,
      "loss": 4.6314,
      "mean_token_accuracy": 0.26319183558225634,
      "num_tokens": 157004905.0,
      "step": 68480
    },
    {
      "entropy": 5.543581581115722,
      "epoch": 6.578770413064361,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016995795579436652,
      "loss": 4.6533,
      "mean_token_accuracy": 0.2659570321440697,
      "num_tokens": 157017400.0,
      "step": 68485
    },
    {
      "entropy": 5.5797182559967045,
      "epoch": 6.579250720461095,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016992763887765732,
      "loss": 4.6281,
      "mean_token_accuracy": 0.2584346175193787,
      "num_tokens": 157028344.0,
      "step": 68490
    },
    {
      "entropy": 5.534253263473511,
      "epoch": 6.579731027857829,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016989732440041638,
      "loss": 4.5851,
      "mean_token_accuracy": 0.2679125517606735,
      "num_tokens": 157038871.0,
      "step": 68495
    },
    {
      "entropy": 5.465820741653443,
      "epoch": 6.580211335254563,
      "grad_norm": 1.59375,
      "learning_rate": 0.00016986701236334744,
      "loss": 4.5357,
      "mean_token_accuracy": 0.26951477527618406,
      "num_tokens": 157050188.0,
      "step": 68500
    },
    {
      "entropy": 5.539918565750122,
      "epoch": 6.5806916426512965,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016983670276715443,
      "loss": 4.6453,
      "mean_token_accuracy": 0.265122526884079,
      "num_tokens": 157061097.0,
      "step": 68505
    },
    {
      "entropy": 5.538680601119995,
      "epoch": 6.581171950048031,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001698063956125408,
      "loss": 4.5787,
      "mean_token_accuracy": 0.269234000146389,
      "num_tokens": 157072047.0,
      "step": 68510
    },
    {
      "entropy": 5.651409435272217,
      "epoch": 6.581652257444764,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016977609090021033,
      "loss": 4.7661,
      "mean_token_accuracy": 0.2545360416173935,
      "num_tokens": 157084235.0,
      "step": 68515
    },
    {
      "entropy": 5.587276935577393,
      "epoch": 6.582132564841499,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016974578863086654,
      "loss": 4.6923,
      "mean_token_accuracy": 0.26208489686250686,
      "num_tokens": 157095833.0,
      "step": 68520
    },
    {
      "entropy": 5.582407093048095,
      "epoch": 6.582612872238233,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016971548880521305,
      "loss": 4.7182,
      "mean_token_accuracy": 0.2603034436702728,
      "num_tokens": 157107672.0,
      "step": 68525
    },
    {
      "entropy": 5.519202470779419,
      "epoch": 6.583093179634966,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016968519142395323,
      "loss": 4.5423,
      "mean_token_accuracy": 0.2686887621879578,
      "num_tokens": 157119080.0,
      "step": 68530
    },
    {
      "entropy": 5.551149511337281,
      "epoch": 6.583573487031701,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016965489648779062,
      "loss": 4.6657,
      "mean_token_accuracy": 0.26225791722536085,
      "num_tokens": 157130670.0,
      "step": 68535
    },
    {
      "entropy": 5.516753101348877,
      "epoch": 6.584053794428434,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016962460399742843,
      "loss": 4.5681,
      "mean_token_accuracy": 0.27543850243091583,
      "num_tokens": 157142391.0,
      "step": 68540
    },
    {
      "entropy": 5.502665853500366,
      "epoch": 6.584534101825168,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016959431395357005,
      "loss": 4.5939,
      "mean_token_accuracy": 0.2690931499004364,
      "num_tokens": 157153269.0,
      "step": 68545
    },
    {
      "entropy": 5.444040727615357,
      "epoch": 6.585014409221902,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016956402635691871,
      "loss": 4.5856,
      "mean_token_accuracy": 0.2716400384902954,
      "num_tokens": 157165102.0,
      "step": 68550
    },
    {
      "entropy": 5.52481427192688,
      "epoch": 6.585494716618636,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001695337412081776,
      "loss": 4.6211,
      "mean_token_accuracy": 0.2636388808488846,
      "num_tokens": 157175770.0,
      "step": 68555
    },
    {
      "entropy": 5.547509574890137,
      "epoch": 6.58597502401537,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016950345850804985,
      "loss": 4.6485,
      "mean_token_accuracy": 0.26665669232606887,
      "num_tokens": 157188315.0,
      "step": 68560
    },
    {
      "entropy": 5.627558326721191,
      "epoch": 6.586455331412104,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001694731782572386,
      "loss": 4.7175,
      "mean_token_accuracy": 0.261460679769516,
      "num_tokens": 157198429.0,
      "step": 68565
    },
    {
      "entropy": 5.605136966705322,
      "epoch": 6.586935638808837,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016944290045644672,
      "loss": 4.7352,
      "mean_token_accuracy": 0.25964120775461197,
      "num_tokens": 157210055.0,
      "step": 68570
    },
    {
      "entropy": 5.478120374679565,
      "epoch": 6.587415946205572,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016941262510637728,
      "loss": 4.5388,
      "mean_token_accuracy": 0.26858623921871183,
      "num_tokens": 157222449.0,
      "step": 68575
    },
    {
      "entropy": 5.501484775543213,
      "epoch": 6.587896253602305,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016938235220773312,
      "loss": 4.6093,
      "mean_token_accuracy": 0.26549035757780076,
      "num_tokens": 157234648.0,
      "step": 68580
    },
    {
      "entropy": 5.5074159622192385,
      "epoch": 6.5883765609990395,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016935208176121713,
      "loss": 4.5565,
      "mean_token_accuracy": 0.27181870192289354,
      "num_tokens": 157245233.0,
      "step": 68585
    },
    {
      "entropy": 5.471854162216187,
      "epoch": 6.588856868395773,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001693218137675321,
      "loss": 4.575,
      "mean_token_accuracy": 0.27485713064670564,
      "num_tokens": 157257618.0,
      "step": 68590
    },
    {
      "entropy": 5.468523693084717,
      "epoch": 6.589337175792507,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016929154822738083,
      "loss": 4.5986,
      "mean_token_accuracy": 0.26784230321645736,
      "num_tokens": 157268857.0,
      "step": 68595
    },
    {
      "entropy": 5.507498264312744,
      "epoch": 6.5898174831892415,
      "grad_norm": 1.5,
      "learning_rate": 0.00016926128514146584,
      "loss": 4.5915,
      "mean_token_accuracy": 0.26851126700639727,
      "num_tokens": 157278979.0,
      "step": 68600
    },
    {
      "entropy": 5.5895459175109865,
      "epoch": 6.590297790585975,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016923102451048982,
      "loss": 4.6837,
      "mean_token_accuracy": 0.25658933222293856,
      "num_tokens": 157293226.0,
      "step": 68605
    },
    {
      "entropy": 5.624797105789185,
      "epoch": 6.590778097982709,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016920076633515535,
      "loss": 4.7639,
      "mean_token_accuracy": 0.24794536232948303,
      "num_tokens": 157304615.0,
      "step": 68610
    },
    {
      "entropy": 5.573349237442017,
      "epoch": 6.591258405379443,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016917051061616495,
      "loss": 4.6935,
      "mean_token_accuracy": 0.2612724632024765,
      "num_tokens": 157316887.0,
      "step": 68615
    },
    {
      "entropy": 5.590266609191895,
      "epoch": 6.591738712776177,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000169140257354221,
      "loss": 4.7096,
      "mean_token_accuracy": 0.2545085445046425,
      "num_tokens": 157328043.0,
      "step": 68620
    },
    {
      "entropy": 5.5498854637146,
      "epoch": 6.5922190201729105,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016911000655002605,
      "loss": 4.6419,
      "mean_token_accuracy": 0.2649378404021263,
      "num_tokens": 157339951.0,
      "step": 68625
    },
    {
      "entropy": 5.541872119903564,
      "epoch": 6.592699327569645,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016907975820428222,
      "loss": 4.6015,
      "mean_token_accuracy": 0.2710425227880478,
      "num_tokens": 157350208.0,
      "step": 68630
    },
    {
      "entropy": 5.568590927124023,
      "epoch": 6.593179634966378,
      "grad_norm": 1.375,
      "learning_rate": 0.00016904951231769188,
      "loss": 4.7168,
      "mean_token_accuracy": 0.2607915535569191,
      "num_tokens": 157363503.0,
      "step": 68635
    },
    {
      "entropy": 5.580610656738282,
      "epoch": 6.593659942363113,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001690192688909573,
      "loss": 4.6843,
      "mean_token_accuracy": 0.26762567907571794,
      "num_tokens": 157375612.0,
      "step": 68640
    },
    {
      "entropy": 5.565935707092285,
      "epoch": 6.594140249759846,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016898902792478055,
      "loss": 4.597,
      "mean_token_accuracy": 0.2690336674451828,
      "num_tokens": 157385984.0,
      "step": 68645
    },
    {
      "entropy": 5.5762214183807375,
      "epoch": 6.59462055715658,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00016895878941986382,
      "loss": 4.6717,
      "mean_token_accuracy": 0.2637089803814888,
      "num_tokens": 157396737.0,
      "step": 68650
    },
    {
      "entropy": 5.465801668167114,
      "epoch": 6.595100864553314,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016892855337690915,
      "loss": 4.5203,
      "mean_token_accuracy": 0.28358458280563353,
      "num_tokens": 157408571.0,
      "step": 68655
    },
    {
      "entropy": 5.464648628234864,
      "epoch": 6.595581171950048,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016889831979661845,
      "loss": 4.6034,
      "mean_token_accuracy": 0.26626466661691667,
      "num_tokens": 157420170.0,
      "step": 68660
    },
    {
      "entropy": 5.585205364227295,
      "epoch": 6.5960614793467816,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016886808867969374,
      "loss": 4.7047,
      "mean_token_accuracy": 0.26057256013154984,
      "num_tokens": 157430899.0,
      "step": 68665
    },
    {
      "entropy": 5.481502676010132,
      "epoch": 6.596541786743516,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016883786002683683,
      "loss": 4.5681,
      "mean_token_accuracy": 0.2746858596801758,
      "num_tokens": 157442613.0,
      "step": 68670
    },
    {
      "entropy": 5.514785099029541,
      "epoch": 6.59702209414025,
      "grad_norm": 1.5,
      "learning_rate": 0.0001688076338387496,
      "loss": 4.5734,
      "mean_token_accuracy": 0.2695698067545891,
      "num_tokens": 157454533.0,
      "step": 68675
    },
    {
      "entropy": 5.5041056156158445,
      "epoch": 6.597502401536984,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016877741011613373,
      "loss": 4.666,
      "mean_token_accuracy": 0.2549891427159309,
      "num_tokens": 157467058.0,
      "step": 68680
    },
    {
      "entropy": 5.5036208152771,
      "epoch": 6.597982708933718,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016874718885969109,
      "loss": 4.5783,
      "mean_token_accuracy": 0.2721652537584305,
      "num_tokens": 157478634.0,
      "step": 68685
    },
    {
      "entropy": 5.578709983825684,
      "epoch": 6.598463016330451,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016871697007012315,
      "loss": 4.6034,
      "mean_token_accuracy": 0.2659753516316414,
      "num_tokens": 157490917.0,
      "step": 68690
    },
    {
      "entropy": 5.536720037460327,
      "epoch": 6.598943323727186,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001686867537481316,
      "loss": 4.5476,
      "mean_token_accuracy": 0.2723558947443962,
      "num_tokens": 157502261.0,
      "step": 68695
    },
    {
      "entropy": 5.561313676834106,
      "epoch": 6.599423631123919,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001686565398944179,
      "loss": 4.683,
      "mean_token_accuracy": 0.26089982837438586,
      "num_tokens": 157513395.0,
      "step": 68700
    },
    {
      "entropy": 5.543047714233398,
      "epoch": 6.5999039385206535,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016862632850968358,
      "loss": 4.6908,
      "mean_token_accuracy": 0.25925855338573456,
      "num_tokens": 157526137.0,
      "step": 68705
    },
    {
      "entropy": 5.510346746444702,
      "epoch": 6.600384245917387,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016859611959463004,
      "loss": 4.5845,
      "mean_token_accuracy": 0.2726954653859138,
      "num_tokens": 157537117.0,
      "step": 68710
    },
    {
      "entropy": 5.48246021270752,
      "epoch": 6.600864553314121,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016856591314995872,
      "loss": 4.5797,
      "mean_token_accuracy": 0.2690275937318802,
      "num_tokens": 157548852.0,
      "step": 68715
    },
    {
      "entropy": 5.679568195343018,
      "epoch": 6.601344860710855,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001685357091763708,
      "loss": 4.8355,
      "mean_token_accuracy": 0.2602518931031227,
      "num_tokens": 157560914.0,
      "step": 68720
    },
    {
      "entropy": 5.557421588897705,
      "epoch": 6.601825168107589,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016850550767456756,
      "loss": 4.6929,
      "mean_token_accuracy": 0.2573664754629135,
      "num_tokens": 157572598.0,
      "step": 68725
    },
    {
      "entropy": 5.563958930969238,
      "epoch": 6.6023054755043225,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001684753086452502,
      "loss": 4.5971,
      "mean_token_accuracy": 0.27323103696107864,
      "num_tokens": 157585049.0,
      "step": 68730
    },
    {
      "entropy": 5.585254907608032,
      "epoch": 6.602785782901057,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001684451120891199,
      "loss": 4.6533,
      "mean_token_accuracy": 0.25984794199466704,
      "num_tokens": 157596817.0,
      "step": 68735
    },
    {
      "entropy": 5.523770952224732,
      "epoch": 6.60326609029779,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016841491800687763,
      "loss": 4.6285,
      "mean_token_accuracy": 0.2627594545483589,
      "num_tokens": 157608800.0,
      "step": 68740
    },
    {
      "entropy": 5.5956056118011475,
      "epoch": 6.6037463976945245,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016838472639922462,
      "loss": 4.6658,
      "mean_token_accuracy": 0.2647575870156288,
      "num_tokens": 157619618.0,
      "step": 68745
    },
    {
      "entropy": 5.495684289932251,
      "epoch": 6.604226705091259,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016835453726686154,
      "loss": 4.6295,
      "mean_token_accuracy": 0.2640072599053383,
      "num_tokens": 157630874.0,
      "step": 68750
    },
    {
      "entropy": 5.536849880218506,
      "epoch": 6.604707012487992,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016832435061048945,
      "loss": 4.6692,
      "mean_token_accuracy": 0.2620534017682076,
      "num_tokens": 157644629.0,
      "step": 68755
    },
    {
      "entropy": 5.605214929580688,
      "epoch": 6.605187319884726,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001682941664308092,
      "loss": 4.6894,
      "mean_token_accuracy": 0.2583914250135422,
      "num_tokens": 157654894.0,
      "step": 68760
    },
    {
      "entropy": 5.606161403656006,
      "epoch": 6.60566762728146,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001682639847285215,
      "loss": 4.6825,
      "mean_token_accuracy": 0.26342997401952745,
      "num_tokens": 157666570.0,
      "step": 68765
    },
    {
      "entropy": 5.551444053649902,
      "epoch": 6.606147934678194,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016823380550432722,
      "loss": 4.6426,
      "mean_token_accuracy": 0.2655872732400894,
      "num_tokens": 157678069.0,
      "step": 68770
    },
    {
      "entropy": 5.554088306427002,
      "epoch": 6.606628242074928,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016820362875892697,
      "loss": 4.6598,
      "mean_token_accuracy": 0.26177156865596773,
      "num_tokens": 157689785.0,
      "step": 68775
    },
    {
      "entropy": 5.486487674713135,
      "epoch": 6.607108549471662,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016817345449302127,
      "loss": 4.5197,
      "mean_token_accuracy": 0.2811383455991745,
      "num_tokens": 157701157.0,
      "step": 68780
    },
    {
      "entropy": 5.575623416900635,
      "epoch": 6.607588856868396,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001681432827073108,
      "loss": 4.6632,
      "mean_token_accuracy": 0.2718476474285126,
      "num_tokens": 157713614.0,
      "step": 68785
    },
    {
      "entropy": 5.617337274551391,
      "epoch": 6.60806916426513,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001681131134024959,
      "loss": 4.7056,
      "mean_token_accuracy": 0.2595022290945053,
      "num_tokens": 157725737.0,
      "step": 68790
    },
    {
      "entropy": 5.564959335327148,
      "epoch": 6.608549471661863,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016808294657927718,
      "loss": 4.6868,
      "mean_token_accuracy": 0.26383148580789567,
      "num_tokens": 157736899.0,
      "step": 68795
    },
    {
      "entropy": 5.618646907806396,
      "epoch": 6.609029779058598,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016805278223835495,
      "loss": 4.6621,
      "mean_token_accuracy": 0.26557164788246157,
      "num_tokens": 157748157.0,
      "step": 68800
    },
    {
      "entropy": 5.550740337371826,
      "epoch": 6.609510086455331,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016802262038042966,
      "loss": 4.6204,
      "mean_token_accuracy": 0.2682378217577934,
      "num_tokens": 157759256.0,
      "step": 68805
    },
    {
      "entropy": 5.7249438762664795,
      "epoch": 6.609990393852065,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016799246100620134,
      "loss": 4.8851,
      "mean_token_accuracy": 0.2458641842007637,
      "num_tokens": 157771221.0,
      "step": 68810
    },
    {
      "entropy": 5.524044227600098,
      "epoch": 6.610470701248799,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016796230411637037,
      "loss": 4.71,
      "mean_token_accuracy": 0.2613599717617035,
      "num_tokens": 157782234.0,
      "step": 68815
    },
    {
      "entropy": 5.555283737182617,
      "epoch": 6.610951008645533,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001679321497116368,
      "loss": 4.6485,
      "mean_token_accuracy": 0.2670139387249947,
      "num_tokens": 157792168.0,
      "step": 68820
    },
    {
      "entropy": 5.567356395721435,
      "epoch": 6.6114313160422675,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016790199779270082,
      "loss": 4.644,
      "mean_token_accuracy": 0.25619198083877565,
      "num_tokens": 157803339.0,
      "step": 68825
    },
    {
      "entropy": 5.494337558746338,
      "epoch": 6.611911623439001,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016787184836026248,
      "loss": 4.6154,
      "mean_token_accuracy": 0.2635000482201576,
      "num_tokens": 157815305.0,
      "step": 68830
    },
    {
      "entropy": 5.5127174854278564,
      "epoch": 6.612391930835734,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016784170141502175,
      "loss": 4.646,
      "mean_token_accuracy": 0.2640627145767212,
      "num_tokens": 157827307.0,
      "step": 68835
    },
    {
      "entropy": 5.501062345504761,
      "epoch": 6.612872238232469,
      "grad_norm": 1.25,
      "learning_rate": 0.00016781155695767846,
      "loss": 4.6209,
      "mean_token_accuracy": 0.26528209894895555,
      "num_tokens": 157840334.0,
      "step": 68840
    },
    {
      "entropy": 5.553457546234131,
      "epoch": 6.613352545629203,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016778141498893253,
      "loss": 4.6043,
      "mean_token_accuracy": 0.26752398908138275,
      "num_tokens": 157851463.0,
      "step": 68845
    },
    {
      "entropy": 5.5091872215271,
      "epoch": 6.6138328530259365,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016775127550948372,
      "loss": 4.5771,
      "mean_token_accuracy": 0.27032897919416427,
      "num_tokens": 157862803.0,
      "step": 68850
    },
    {
      "entropy": 5.5417074203491214,
      "epoch": 6.614313160422671,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016772113852003191,
      "loss": 4.6095,
      "mean_token_accuracy": 0.26788846403360367,
      "num_tokens": 157874160.0,
      "step": 68855
    },
    {
      "entropy": 5.546793508529663,
      "epoch": 6.614793467819404,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016769100402127664,
      "loss": 4.627,
      "mean_token_accuracy": 0.2591719627380371,
      "num_tokens": 157884786.0,
      "step": 68860
    },
    {
      "entropy": 5.600137090682983,
      "epoch": 6.615273775216139,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016766087201391765,
      "loss": 4.693,
      "mean_token_accuracy": 0.253128345310688,
      "num_tokens": 157897371.0,
      "step": 68865
    },
    {
      "entropy": 5.614685153961181,
      "epoch": 6.615754082612872,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016763074249865457,
      "loss": 4.7061,
      "mean_token_accuracy": 0.26003539115190505,
      "num_tokens": 157908997.0,
      "step": 68870
    },
    {
      "entropy": 5.633772087097168,
      "epoch": 6.616234390009606,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001676006154761867,
      "loss": 4.7141,
      "mean_token_accuracy": 0.25981463491916656,
      "num_tokens": 157919151.0,
      "step": 68875
    },
    {
      "entropy": 5.577950382232666,
      "epoch": 6.61671469740634,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016757049094721368,
      "loss": 4.7068,
      "mean_token_accuracy": 0.2583244428038597,
      "num_tokens": 157931753.0,
      "step": 68880
    },
    {
      "entropy": 5.478216934204101,
      "epoch": 6.617195004803074,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016754036891243485,
      "loss": 4.5129,
      "mean_token_accuracy": 0.2731946125626564,
      "num_tokens": 157943571.0,
      "step": 68885
    },
    {
      "entropy": 5.582665967941284,
      "epoch": 6.6176753121998075,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016751024937254955,
      "loss": 4.6061,
      "mean_token_accuracy": 0.2592634841799736,
      "num_tokens": 157955333.0,
      "step": 68890
    },
    {
      "entropy": 5.531170082092285,
      "epoch": 6.618155619596542,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016748013232825706,
      "loss": 4.66,
      "mean_token_accuracy": 0.2611585304141045,
      "num_tokens": 157967991.0,
      "step": 68895
    },
    {
      "entropy": 5.5500304222106935,
      "epoch": 6.618635926993276,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016745001778025668,
      "loss": 4.6591,
      "mean_token_accuracy": 0.27185830622911455,
      "num_tokens": 157979375.0,
      "step": 68900
    },
    {
      "entropy": 5.659171485900879,
      "epoch": 6.61911623439001,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016741990572924752,
      "loss": 4.7897,
      "mean_token_accuracy": 0.25271878242492674,
      "num_tokens": 157991530.0,
      "step": 68905
    },
    {
      "entropy": 5.522030067443848,
      "epoch": 6.619596541786743,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016738979617592875,
      "loss": 4.5974,
      "mean_token_accuracy": 0.2696236059069633,
      "num_tokens": 158002043.0,
      "step": 68910
    },
    {
      "entropy": 5.592045593261719,
      "epoch": 6.620076849183477,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016735968912099934,
      "loss": 4.6369,
      "mean_token_accuracy": 0.26886607706546783,
      "num_tokens": 158013605.0,
      "step": 68915
    },
    {
      "entropy": 5.518778276443482,
      "epoch": 6.620557156580212,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016732958456515824,
      "loss": 4.6157,
      "mean_token_accuracy": 0.26574262529611586,
      "num_tokens": 158024273.0,
      "step": 68920
    },
    {
      "entropy": 5.495540618896484,
      "epoch": 6.621037463976945,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016729948250910454,
      "loss": 4.6095,
      "mean_token_accuracy": 0.2680290281772614,
      "num_tokens": 158035435.0,
      "step": 68925
    },
    {
      "entropy": 5.496201705932617,
      "epoch": 6.6215177713736795,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016726938295353705,
      "loss": 4.6017,
      "mean_token_accuracy": 0.2698610797524452,
      "num_tokens": 158048800.0,
      "step": 68930
    },
    {
      "entropy": 5.584016370773315,
      "epoch": 6.621998078770413,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016723928589915457,
      "loss": 4.6916,
      "mean_token_accuracy": 0.26025123745203016,
      "num_tokens": 158060735.0,
      "step": 68935
    },
    {
      "entropy": 5.563517665863037,
      "epoch": 6.622478386167147,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016720919134665592,
      "loss": 4.6185,
      "mean_token_accuracy": 0.2679632991552353,
      "num_tokens": 158072017.0,
      "step": 68940
    },
    {
      "entropy": 5.511303806304932,
      "epoch": 6.622958693563881,
      "grad_norm": 1.25,
      "learning_rate": 0.00016717909929673974,
      "loss": 4.5461,
      "mean_token_accuracy": 0.2696761041879654,
      "num_tokens": 158084040.0,
      "step": 68945
    },
    {
      "entropy": 5.5654560089111325,
      "epoch": 6.623439000960615,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016714900975010484,
      "loss": 4.6367,
      "mean_token_accuracy": 0.26401654332876207,
      "num_tokens": 158095349.0,
      "step": 68950
    },
    {
      "entropy": 5.480224323272705,
      "epoch": 6.623919308357348,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016711892270744954,
      "loss": 4.5419,
      "mean_token_accuracy": 0.27280848622322085,
      "num_tokens": 158106912.0,
      "step": 68955
    },
    {
      "entropy": 5.575611114501953,
      "epoch": 6.624399615754083,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016708883816947252,
      "loss": 4.6649,
      "mean_token_accuracy": 0.26454448103904726,
      "num_tokens": 158116751.0,
      "step": 68960
    },
    {
      "entropy": 5.520468235015869,
      "epoch": 6.624879923150816,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016705875613687226,
      "loss": 4.6277,
      "mean_token_accuracy": 0.2653045117855072,
      "num_tokens": 158128192.0,
      "step": 68965
    },
    {
      "entropy": 5.493208932876587,
      "epoch": 6.6253602305475505,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001670286766103472,
      "loss": 4.5761,
      "mean_token_accuracy": 0.27646357715129855,
      "num_tokens": 158140952.0,
      "step": 68970
    },
    {
      "entropy": 5.547638988494873,
      "epoch": 6.625840537944284,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016699859959059564,
      "loss": 4.6655,
      "mean_token_accuracy": 0.2597607418894768,
      "num_tokens": 158153732.0,
      "step": 68975
    },
    {
      "entropy": 5.578320932388306,
      "epoch": 6.626320845341018,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001669685250783159,
      "loss": 4.6991,
      "mean_token_accuracy": 0.2635035887360573,
      "num_tokens": 158164947.0,
      "step": 68980
    },
    {
      "entropy": 5.606412601470947,
      "epoch": 6.626801152737752,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016693845307420624,
      "loss": 4.6412,
      "mean_token_accuracy": 0.2601527884602547,
      "num_tokens": 158176259.0,
      "step": 68985
    },
    {
      "entropy": 5.59907169342041,
      "epoch": 6.627281460134486,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016690838357896482,
      "loss": 4.6901,
      "mean_token_accuracy": 0.2537083089351654,
      "num_tokens": 158188261.0,
      "step": 68990
    },
    {
      "entropy": 5.540513277053833,
      "epoch": 6.62776176753122,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016687831659328975,
      "loss": 4.6029,
      "mean_token_accuracy": 0.2703482389450073,
      "num_tokens": 158200018.0,
      "step": 68995
    },
    {
      "entropy": 5.561776351928711,
      "epoch": 6.628242074927954,
      "grad_norm": 1.53125,
      "learning_rate": 0.00016684825211787908,
      "loss": 4.7006,
      "mean_token_accuracy": 0.25872285962104796,
      "num_tokens": 158210227.0,
      "step": 69000
    },
    {
      "epoch": 6.628242074927954,
      "eval_entropy": 5.235162813799489,
      "eval_loss": 4.7528767585754395,
      "eval_mean_token_accuracy": 0.26660762477152117,
      "eval_num_tokens": 158210227.0,
      "eval_runtime": 20.632,
      "eval_samples_per_second": 1590.491,
      "eval_steps_per_second": 198.817,
      "step": 69000
    },
    {
      "entropy": 5.4788895606994625,
      "epoch": 6.628722382324688,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016681819015343087,
      "loss": 4.6116,
      "mean_token_accuracy": 0.26331334859132766,
      "num_tokens": 158222893.0,
      "step": 69005
    },
    {
      "entropy": 5.491120672225952,
      "epoch": 6.629202689721422,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001667881307006431,
      "loss": 4.6102,
      "mean_token_accuracy": 0.26209473311901094,
      "num_tokens": 158234918.0,
      "step": 69010
    },
    {
      "entropy": 5.535275316238403,
      "epoch": 6.629682997118156,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016675807376021363,
      "loss": 4.6465,
      "mean_token_accuracy": 0.2670057132840157,
      "num_tokens": 158246307.0,
      "step": 69015
    },
    {
      "entropy": 5.552132892608642,
      "epoch": 6.630163304514889,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001667280193328402,
      "loss": 4.6499,
      "mean_token_accuracy": 0.26648586988449097,
      "num_tokens": 158257323.0,
      "step": 69020
    },
    {
      "entropy": 5.548651933670044,
      "epoch": 6.630643611911624,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016669796741922065,
      "loss": 4.6582,
      "mean_token_accuracy": 0.26180976927280425,
      "num_tokens": 158268205.0,
      "step": 69025
    },
    {
      "entropy": 5.557918357849121,
      "epoch": 6.631123919308357,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016666791802005278,
      "loss": 4.6389,
      "mean_token_accuracy": 0.26368548423051835,
      "num_tokens": 158279558.0,
      "step": 69030
    },
    {
      "entropy": 5.510062408447266,
      "epoch": 6.631604226705091,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001666378711360341,
      "loss": 4.606,
      "mean_token_accuracy": 0.263804192841053,
      "num_tokens": 158291803.0,
      "step": 69035
    },
    {
      "entropy": 5.6035017490386965,
      "epoch": 6.632084534101825,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001666078267678623,
      "loss": 4.7108,
      "mean_token_accuracy": 0.26416987031698225,
      "num_tokens": 158302814.0,
      "step": 69040
    },
    {
      "entropy": 5.627990341186523,
      "epoch": 6.632564841498559,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016657778491623498,
      "loss": 4.6883,
      "mean_token_accuracy": 0.2618901371955872,
      "num_tokens": 158314378.0,
      "step": 69045
    },
    {
      "entropy": 5.5670746803283695,
      "epoch": 6.633045148895293,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016654774558184948,
      "loss": 4.6859,
      "mean_token_accuracy": 0.2571896806359291,
      "num_tokens": 158326246.0,
      "step": 69050
    },
    {
      "entropy": 5.538031530380249,
      "epoch": 6.633525456292027,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016651770876540328,
      "loss": 4.5911,
      "mean_token_accuracy": 0.2770868629217148,
      "num_tokens": 158336476.0,
      "step": 69055
    },
    {
      "entropy": 5.56445689201355,
      "epoch": 6.63400576368876,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016648767446759378,
      "loss": 4.6788,
      "mean_token_accuracy": 0.2565046831965446,
      "num_tokens": 158348919.0,
      "step": 69060
    },
    {
      "entropy": 5.49154257774353,
      "epoch": 6.634486071085495,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00016645764268911821,
      "loss": 4.4656,
      "mean_token_accuracy": 0.2787931650876999,
      "num_tokens": 158362100.0,
      "step": 69065
    },
    {
      "entropy": 5.640294790267944,
      "epoch": 6.634966378482229,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016642761343067393,
      "loss": 4.7367,
      "mean_token_accuracy": 0.25696817338466643,
      "num_tokens": 158373086.0,
      "step": 69070
    },
    {
      "entropy": 5.571170282363892,
      "epoch": 6.6354466858789625,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016639758669295812,
      "loss": 4.6775,
      "mean_token_accuracy": 0.2648494005203247,
      "num_tokens": 158386466.0,
      "step": 69075
    },
    {
      "entropy": 5.49235577583313,
      "epoch": 6.635926993275697,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00016636756247666778,
      "loss": 4.5903,
      "mean_token_accuracy": 0.27155870497226714,
      "num_tokens": 158397089.0,
      "step": 69080
    },
    {
      "entropy": 5.5856414318084715,
      "epoch": 6.63640730067243,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001663375407825001,
      "loss": 4.6848,
      "mean_token_accuracy": 0.26021881997585294,
      "num_tokens": 158408770.0,
      "step": 69085
    },
    {
      "entropy": 5.522451162338257,
      "epoch": 6.636887608069165,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016630752161115204,
      "loss": 4.5655,
      "mean_token_accuracy": 0.26893245428800583,
      "num_tokens": 158420249.0,
      "step": 69090
    },
    {
      "entropy": 5.560948514938355,
      "epoch": 6.637367915465898,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016627750496332058,
      "loss": 4.637,
      "mean_token_accuracy": 0.2620751246809959,
      "num_tokens": 158431227.0,
      "step": 69095
    },
    {
      "entropy": 5.528344345092774,
      "epoch": 6.637848222862632,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001662474908397026,
      "loss": 4.633,
      "mean_token_accuracy": 0.2644712910056114,
      "num_tokens": 158442977.0,
      "step": 69100
    },
    {
      "entropy": 5.531970548629761,
      "epoch": 6.638328530259366,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016621747924099506,
      "loss": 4.6227,
      "mean_token_accuracy": 0.26129990071058273,
      "num_tokens": 158453345.0,
      "step": 69105
    },
    {
      "entropy": 5.604838514328003,
      "epoch": 6.6388088376561,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016618747016789454,
      "loss": 4.6874,
      "mean_token_accuracy": 0.26080563217401503,
      "num_tokens": 158463438.0,
      "step": 69110
    },
    {
      "entropy": 5.550107860565186,
      "epoch": 6.6392891450528335,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016615746362109787,
      "loss": 4.6833,
      "mean_token_accuracy": 0.26230600774288176,
      "num_tokens": 158475542.0,
      "step": 69115
    },
    {
      "entropy": 5.626029634475708,
      "epoch": 6.639769452449568,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016612745960130172,
      "loss": 4.7355,
      "mean_token_accuracy": 0.251746267080307,
      "num_tokens": 158487126.0,
      "step": 69120
    },
    {
      "entropy": 5.4977175235748295,
      "epoch": 6.640249759846301,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016609745810920262,
      "loss": 4.5622,
      "mean_token_accuracy": 0.2742354333400726,
      "num_tokens": 158497786.0,
      "step": 69125
    },
    {
      "entropy": 5.5650022506713865,
      "epoch": 6.640730067243036,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016606745914549717,
      "loss": 4.6993,
      "mean_token_accuracy": 0.25754719972610474,
      "num_tokens": 158508541.0,
      "step": 69130
    },
    {
      "entropy": 5.574152326583862,
      "epoch": 6.641210374639769,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016603746271088196,
      "loss": 4.703,
      "mean_token_accuracy": 0.2614338159561157,
      "num_tokens": 158520570.0,
      "step": 69135
    },
    {
      "entropy": 5.56692214012146,
      "epoch": 6.641690682036503,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016600746880605322,
      "loss": 4.68,
      "mean_token_accuracy": 0.26433215290308,
      "num_tokens": 158531216.0,
      "step": 69140
    },
    {
      "entropy": 5.597336721420288,
      "epoch": 6.642170989433238,
      "grad_norm": 1.609375,
      "learning_rate": 0.00016597747743170744,
      "loss": 4.6888,
      "mean_token_accuracy": 0.26491312831640246,
      "num_tokens": 158542549.0,
      "step": 69145
    },
    {
      "entropy": 5.515254354476928,
      "epoch": 6.642651296829971,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016594748858854088,
      "loss": 4.6381,
      "mean_token_accuracy": 0.2745270699262619,
      "num_tokens": 158555302.0,
      "step": 69150
    },
    {
      "entropy": 5.520387887954712,
      "epoch": 6.6431316042267055,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016591750227724984,
      "loss": 4.574,
      "mean_token_accuracy": 0.27196422815322874,
      "num_tokens": 158566597.0,
      "step": 69155
    },
    {
      "entropy": 5.563843822479248,
      "epoch": 6.643611911623439,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016588751849853047,
      "loss": 4.674,
      "mean_token_accuracy": 0.2600968614220619,
      "num_tokens": 158578699.0,
      "step": 69160
    },
    {
      "entropy": 5.547764539718628,
      "epoch": 6.644092219020173,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016585753725307901,
      "loss": 4.5935,
      "mean_token_accuracy": 0.2725073412060738,
      "num_tokens": 158590093.0,
      "step": 69165
    },
    {
      "entropy": 5.492608594894409,
      "epoch": 6.644572526416907,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016582755854159135,
      "loss": 4.603,
      "mean_token_accuracy": 0.2703173279762268,
      "num_tokens": 158601212.0,
      "step": 69170
    },
    {
      "entropy": 5.556228351593018,
      "epoch": 6.645052833813641,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016579758236476362,
      "loss": 4.6302,
      "mean_token_accuracy": 0.25980129837989807,
      "num_tokens": 158612401.0,
      "step": 69175
    },
    {
      "entropy": 5.572423505783081,
      "epoch": 6.645533141210374,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016576760872329176,
      "loss": 4.6682,
      "mean_token_accuracy": 0.26408752202987673,
      "num_tokens": 158624078.0,
      "step": 69180
    },
    {
      "entropy": 5.490637683868409,
      "epoch": 6.646013448607109,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00016573763761787164,
      "loss": 4.5374,
      "mean_token_accuracy": 0.27585008293390273,
      "num_tokens": 158635100.0,
      "step": 69185
    },
    {
      "entropy": 5.518730401992798,
      "epoch": 6.646493756003842,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00016570766904919917,
      "loss": 4.6169,
      "mean_token_accuracy": 0.2727417707443237,
      "num_tokens": 158646710.0,
      "step": 69190
    },
    {
      "entropy": 5.52503924369812,
      "epoch": 6.6469740634005765,
      "grad_norm": 1.578125,
      "learning_rate": 0.00016567770301797007,
      "loss": 4.6009,
      "mean_token_accuracy": 0.276397830247879,
      "num_tokens": 158658321.0,
      "step": 69195
    },
    {
      "entropy": 5.505974340438843,
      "epoch": 6.64745437079731,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001656477395248802,
      "loss": 4.6567,
      "mean_token_accuracy": 0.26057289242744447,
      "num_tokens": 158669130.0,
      "step": 69200
    },
    {
      "entropy": 5.560281276702881,
      "epoch": 6.647934678194044,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016561777857062503,
      "loss": 4.6522,
      "mean_token_accuracy": 0.26753502935171125,
      "num_tokens": 158680762.0,
      "step": 69205
    },
    {
      "entropy": 5.512847423553467,
      "epoch": 6.648414985590778,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016558782015590022,
      "loss": 4.5777,
      "mean_token_accuracy": 0.27549574077129363,
      "num_tokens": 158691580.0,
      "step": 69210
    },
    {
      "entropy": 5.475575733184814,
      "epoch": 6.648895292987512,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016555786428140134,
      "loss": 4.6216,
      "mean_token_accuracy": 0.2699723556637764,
      "num_tokens": 158703716.0,
      "step": 69215
    },
    {
      "entropy": 5.528531455993653,
      "epoch": 6.649375600384246,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016552791094782388,
      "loss": 4.552,
      "mean_token_accuracy": 0.27198525220155717,
      "num_tokens": 158715862.0,
      "step": 69220
    },
    {
      "entropy": 5.577464580535889,
      "epoch": 6.64985590778098,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016549796015586328,
      "loss": 4.6455,
      "mean_token_accuracy": 0.26247874796390536,
      "num_tokens": 158726552.0,
      "step": 69225
    },
    {
      "entropy": 5.4790459156036375,
      "epoch": 6.650336215177714,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016546801190621494,
      "loss": 4.5738,
      "mean_token_accuracy": 0.27266696095466614,
      "num_tokens": 158737726.0,
      "step": 69230
    },
    {
      "entropy": 5.550645637512207,
      "epoch": 6.6508165225744476,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016543806619957405,
      "loss": 4.6658,
      "mean_token_accuracy": 0.2613871231675148,
      "num_tokens": 158749994.0,
      "step": 69235
    },
    {
      "entropy": 5.464734220504761,
      "epoch": 6.651296829971182,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016540812303663595,
      "loss": 4.553,
      "mean_token_accuracy": 0.2779473543167114,
      "num_tokens": 158761077.0,
      "step": 69240
    },
    {
      "entropy": 5.530900859832764,
      "epoch": 6.651777137367915,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016537818241809578,
      "loss": 4.6203,
      "mean_token_accuracy": 0.2626966252923012,
      "num_tokens": 158773146.0,
      "step": 69245
    },
    {
      "entropy": 5.580292797088623,
      "epoch": 6.65225744476465,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016534824434464868,
      "loss": 4.7414,
      "mean_token_accuracy": 0.2593535676598549,
      "num_tokens": 158785534.0,
      "step": 69250
    },
    {
      "entropy": 5.613668203353882,
      "epoch": 6.652737752161383,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016531830881698982,
      "loss": 4.6923,
      "mean_token_accuracy": 0.25375126898288725,
      "num_tokens": 158796868.0,
      "step": 69255
    },
    {
      "entropy": 5.506757354736328,
      "epoch": 6.653218059558117,
      "grad_norm": 1.25,
      "learning_rate": 0.0001652883758358142,
      "loss": 4.6029,
      "mean_token_accuracy": 0.26638130098581314,
      "num_tokens": 158808006.0,
      "step": 69260
    },
    {
      "entropy": 5.5276776313781735,
      "epoch": 6.653698366954851,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001652584454018166,
      "loss": 4.6345,
      "mean_token_accuracy": 0.26993103325366974,
      "num_tokens": 158819588.0,
      "step": 69265
    },
    {
      "entropy": 5.50198335647583,
      "epoch": 6.654178674351585,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016522851751569204,
      "loss": 4.5971,
      "mean_token_accuracy": 0.2696480557322502,
      "num_tokens": 158831714.0,
      "step": 69270
    },
    {
      "entropy": 5.600333309173584,
      "epoch": 6.654658981748319,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016519859217813538,
      "loss": 4.6527,
      "mean_token_accuracy": 0.26464369595050813,
      "num_tokens": 158844072.0,
      "step": 69275
    },
    {
      "entropy": 5.572503471374512,
      "epoch": 6.655139289145053,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016516866938984137,
      "loss": 4.6849,
      "mean_token_accuracy": 0.2590324103832245,
      "num_tokens": 158855347.0,
      "step": 69280
    },
    {
      "entropy": 5.538154077529907,
      "epoch": 6.655619596541786,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016513874915150467,
      "loss": 4.6471,
      "mean_token_accuracy": 0.2618099495768547,
      "num_tokens": 158867945.0,
      "step": 69285
    },
    {
      "entropy": 5.553250741958618,
      "epoch": 6.656099903938521,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016510883146382018,
      "loss": 4.6269,
      "mean_token_accuracy": 0.26979687809944153,
      "num_tokens": 158880885.0,
      "step": 69290
    },
    {
      "entropy": 5.480870151519776,
      "epoch": 6.656580211335255,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001650789163274821,
      "loss": 4.5912,
      "mean_token_accuracy": 0.2658148244023323,
      "num_tokens": 158892414.0,
      "step": 69295
    },
    {
      "entropy": 5.573120403289795,
      "epoch": 6.6570605187319885,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016504900374318527,
      "loss": 4.6504,
      "mean_token_accuracy": 0.261319537460804,
      "num_tokens": 158903703.0,
      "step": 69300
    },
    {
      "entropy": 5.59841103553772,
      "epoch": 6.657540826128722,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016501909371162405,
      "loss": 4.6879,
      "mean_token_accuracy": 0.2616792231798172,
      "num_tokens": 158915471.0,
      "step": 69305
    },
    {
      "entropy": 5.529654169082642,
      "epoch": 6.658021133525456,
      "grad_norm": 1.375,
      "learning_rate": 0.00016498918623349294,
      "loss": 4.6298,
      "mean_token_accuracy": 0.26430384516716005,
      "num_tokens": 158926454.0,
      "step": 69310
    },
    {
      "entropy": 5.5508544921875,
      "epoch": 6.6585014409221905,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016495928130948623,
      "loss": 4.6611,
      "mean_token_accuracy": 0.2608819857239723,
      "num_tokens": 158937634.0,
      "step": 69315
    },
    {
      "entropy": 5.533236694335938,
      "epoch": 6.658981748318924,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016492937894029832,
      "loss": 4.6855,
      "mean_token_accuracy": 0.26425969153642653,
      "num_tokens": 158949599.0,
      "step": 69320
    },
    {
      "entropy": 5.4887782573699955,
      "epoch": 6.659462055715658,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016489947912662334,
      "loss": 4.6052,
      "mean_token_accuracy": 0.2676428332924843,
      "num_tokens": 158960723.0,
      "step": 69325
    },
    {
      "entropy": 5.489761829376221,
      "epoch": 6.659942363112392,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016486958186915553,
      "loss": 4.5828,
      "mean_token_accuracy": 0.275131431221962,
      "num_tokens": 158971466.0,
      "step": 69330
    },
    {
      "entropy": 5.422856712341309,
      "epoch": 6.660422670509126,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000164839687168589,
      "loss": 4.5205,
      "mean_token_accuracy": 0.27859459817409515,
      "num_tokens": 158983127.0,
      "step": 69335
    },
    {
      "entropy": 5.527157258987427,
      "epoch": 6.6609029779058595,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016480979502561777,
      "loss": 4.6312,
      "mean_token_accuracy": 0.2639795258641243,
      "num_tokens": 158993937.0,
      "step": 69340
    },
    {
      "entropy": 5.49750657081604,
      "epoch": 6.661383285302594,
      "grad_norm": 1.375,
      "learning_rate": 0.00016477990544093592,
      "loss": 4.6259,
      "mean_token_accuracy": 0.26884745955467226,
      "num_tokens": 159004257.0,
      "step": 69345
    },
    {
      "entropy": 5.4938520908355715,
      "epoch": 6.661863592699327,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001647500184152375,
      "loss": 4.5902,
      "mean_token_accuracy": 0.26149034649133684,
      "num_tokens": 159015330.0,
      "step": 69350
    },
    {
      "entropy": 5.578207206726074,
      "epoch": 6.662343900096062,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001647201339492162,
      "loss": 4.7182,
      "mean_token_accuracy": 0.25932855159044266,
      "num_tokens": 159027526.0,
      "step": 69355
    },
    {
      "entropy": 5.53261890411377,
      "epoch": 6.662824207492795,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001646902520435659,
      "loss": 4.6562,
      "mean_token_accuracy": 0.26918700337409973,
      "num_tokens": 159040000.0,
      "step": 69360
    },
    {
      "entropy": 5.590804672241211,
      "epoch": 6.663304514889529,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016466037269898041,
      "loss": 4.7142,
      "mean_token_accuracy": 0.25914849191904066,
      "num_tokens": 159053418.0,
      "step": 69365
    },
    {
      "entropy": 5.467402410507202,
      "epoch": 6.663784822286264,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001646304959161534,
      "loss": 4.5805,
      "mean_token_accuracy": 0.26592351496219635,
      "num_tokens": 159065672.0,
      "step": 69370
    },
    {
      "entropy": 5.521137380599976,
      "epoch": 6.664265129682997,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001646006216957786,
      "loss": 4.6581,
      "mean_token_accuracy": 0.2644515708088875,
      "num_tokens": 159078180.0,
      "step": 69375
    },
    {
      "entropy": 5.508445930480957,
      "epoch": 6.6647454370797306,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016457075003854956,
      "loss": 4.6098,
      "mean_token_accuracy": 0.2626370668411255,
      "num_tokens": 159088272.0,
      "step": 69380
    },
    {
      "entropy": 5.568877553939819,
      "epoch": 6.665225744476465,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016454088094515975,
      "loss": 4.6406,
      "mean_token_accuracy": 0.26532770693302155,
      "num_tokens": 159099345.0,
      "step": 69385
    },
    {
      "entropy": 5.5520501136779785,
      "epoch": 6.665706051873199,
      "grad_norm": 1.375,
      "learning_rate": 0.00016451101441630268,
      "loss": 4.6916,
      "mean_token_accuracy": 0.2571020066738129,
      "num_tokens": 159111221.0,
      "step": 69390
    },
    {
      "entropy": 5.557260751724243,
      "epoch": 6.666186359269933,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016448115045267176,
      "loss": 4.7441,
      "mean_token_accuracy": 0.25756377130746844,
      "num_tokens": 159122367.0,
      "step": 69395
    },
    {
      "entropy": 5.4820794582366945,
      "epoch": 6.666666666666667,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016445128905496036,
      "loss": 4.5585,
      "mean_token_accuracy": 0.2717552661895752,
      "num_tokens": 159133821.0,
      "step": 69400
    },
    {
      "entropy": 5.573512315750122,
      "epoch": 6.6671469740634,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001644214302238618,
      "loss": 4.6808,
      "mean_token_accuracy": 0.2622342362999916,
      "num_tokens": 159145415.0,
      "step": 69405
    },
    {
      "entropy": 5.570688676834107,
      "epoch": 6.667627281460135,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001643915739600693,
      "loss": 4.6019,
      "mean_token_accuracy": 0.26787666231393814,
      "num_tokens": 159156696.0,
      "step": 69410
    },
    {
      "entropy": 5.579158353805542,
      "epoch": 6.668107588856868,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016436172026427598,
      "loss": 4.6807,
      "mean_token_accuracy": 0.2650223597884178,
      "num_tokens": 159168103.0,
      "step": 69415
    },
    {
      "entropy": 5.518838262557983,
      "epoch": 6.6685878962536025,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016433186913717497,
      "loss": 4.6194,
      "mean_token_accuracy": 0.26914599984884263,
      "num_tokens": 159179529.0,
      "step": 69420
    },
    {
      "entropy": 5.545850133895874,
      "epoch": 6.669068203650336,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016430202057945935,
      "loss": 4.6225,
      "mean_token_accuracy": 0.27218843400478365,
      "num_tokens": 159191159.0,
      "step": 69425
    },
    {
      "entropy": 5.508128595352173,
      "epoch": 6.66954851104707,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001642721745918221,
      "loss": 4.576,
      "mean_token_accuracy": 0.27193067967891693,
      "num_tokens": 159201753.0,
      "step": 69430
    },
    {
      "entropy": 5.5493066787719725,
      "epoch": 6.670028818443804,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016424233117495613,
      "loss": 4.6745,
      "mean_token_accuracy": 0.26470799595117567,
      "num_tokens": 159212731.0,
      "step": 69435
    },
    {
      "entropy": 5.5571982860565186,
      "epoch": 6.670509125840538,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016421249032955436,
      "loss": 4.6444,
      "mean_token_accuracy": 0.26920155584812167,
      "num_tokens": 159224432.0,
      "step": 69440
    },
    {
      "entropy": 5.502919101715088,
      "epoch": 6.670989433237272,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001641826520563097,
      "loss": 4.5692,
      "mean_token_accuracy": 0.26778155863285064,
      "num_tokens": 159237608.0,
      "step": 69445
    },
    {
      "entropy": 5.51903281211853,
      "epoch": 6.671469740634006,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016415281635591476,
      "loss": 4.6362,
      "mean_token_accuracy": 0.2666430905461311,
      "num_tokens": 159249943.0,
      "step": 69450
    },
    {
      "entropy": 5.5417852878570555,
      "epoch": 6.671950048030739,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001641229832290622,
      "loss": 4.6593,
      "mean_token_accuracy": 0.2596232458949089,
      "num_tokens": 159261737.0,
      "step": 69455
    },
    {
      "entropy": 5.560056734085083,
      "epoch": 6.6724303554274735,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016409315267644476,
      "loss": 4.6302,
      "mean_token_accuracy": 0.26490108370780946,
      "num_tokens": 159273882.0,
      "step": 69460
    },
    {
      "entropy": 5.541607999801636,
      "epoch": 6.672910662824208,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000164063324698755,
      "loss": 4.5894,
      "mean_token_accuracy": 0.27407838851213456,
      "num_tokens": 159285199.0,
      "step": 69465
    },
    {
      "entropy": 5.53671407699585,
      "epoch": 6.673390970220941,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001640334992966854,
      "loss": 4.6404,
      "mean_token_accuracy": 0.27656998485326767,
      "num_tokens": 159296054.0,
      "step": 69470
    },
    {
      "entropy": 5.53811411857605,
      "epoch": 6.673871277617676,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016400367647092845,
      "loss": 4.6293,
      "mean_token_accuracy": 0.26527937799692153,
      "num_tokens": 159308423.0,
      "step": 69475
    },
    {
      "entropy": 5.525744915008545,
      "epoch": 6.674351585014409,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016397385622217658,
      "loss": 4.6473,
      "mean_token_accuracy": 0.2641667813062668,
      "num_tokens": 159318717.0,
      "step": 69480
    },
    {
      "entropy": 5.602201271057129,
      "epoch": 6.674831892411143,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001639440385511221,
      "loss": 4.695,
      "mean_token_accuracy": 0.2590313509106636,
      "num_tokens": 159330343.0,
      "step": 69485
    },
    {
      "entropy": 5.462579441070557,
      "epoch": 6.675312199807877,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016391422345845728,
      "loss": 4.5717,
      "mean_token_accuracy": 0.2715607389807701,
      "num_tokens": 159341437.0,
      "step": 69490
    },
    {
      "entropy": 5.561572408676147,
      "epoch": 6.675792507204611,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016388441094487429,
      "loss": 4.6854,
      "mean_token_accuracy": 0.2644413039088249,
      "num_tokens": 159352670.0,
      "step": 69495
    },
    {
      "entropy": 5.620968532562256,
      "epoch": 6.676272814601345,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001638546010110653,
      "loss": 4.7318,
      "mean_token_accuracy": 0.2542466074228287,
      "num_tokens": 159366028.0,
      "step": 69500
    },
    {
      "entropy": 5.570926141738892,
      "epoch": 6.676753121998079,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001638247936577225,
      "loss": 4.7018,
      "mean_token_accuracy": 0.26006115823984144,
      "num_tokens": 159377505.0,
      "step": 69505
    },
    {
      "entropy": 5.566264533996582,
      "epoch": 6.677233429394812,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00016379498888553778,
      "loss": 4.6547,
      "mean_token_accuracy": 0.26025648415088654,
      "num_tokens": 159389171.0,
      "step": 69510
    },
    {
      "entropy": 5.556874227523804,
      "epoch": 6.677713736791547,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016376518669520328,
      "loss": 4.6447,
      "mean_token_accuracy": 0.25868666023015974,
      "num_tokens": 159400807.0,
      "step": 69515
    },
    {
      "entropy": 5.53340516090393,
      "epoch": 6.678194044188281,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016373538708741088,
      "loss": 4.6561,
      "mean_token_accuracy": 0.2614533990621567,
      "num_tokens": 159412492.0,
      "step": 69520
    },
    {
      "entropy": 5.542261266708374,
      "epoch": 6.678674351585014,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001637055900628524,
      "loss": 4.6355,
      "mean_token_accuracy": 0.2642805963754654,
      "num_tokens": 159424530.0,
      "step": 69525
    },
    {
      "entropy": 5.509677791595459,
      "epoch": 6.679154658981748,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016367579562221962,
      "loss": 4.5782,
      "mean_token_accuracy": 0.26814003437757494,
      "num_tokens": 159435709.0,
      "step": 69530
    },
    {
      "entropy": 5.55436110496521,
      "epoch": 6.679634966378482,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016364600376620427,
      "loss": 4.6952,
      "mean_token_accuracy": 0.25845115035772326,
      "num_tokens": 159448958.0,
      "step": 69535
    },
    {
      "entropy": 5.5978538513183596,
      "epoch": 6.6801152737752165,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016361621449549808,
      "loss": 4.6518,
      "mean_token_accuracy": 0.2662371024489403,
      "num_tokens": 159459943.0,
      "step": 69540
    },
    {
      "entropy": 5.458105659484863,
      "epoch": 6.68059558117195,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016358642781079264,
      "loss": 4.577,
      "mean_token_accuracy": 0.27173047214746476,
      "num_tokens": 159471441.0,
      "step": 69545
    },
    {
      "entropy": 5.526128911972046,
      "epoch": 6.681075888568684,
      "grad_norm": 1.25,
      "learning_rate": 0.0001635566437127795,
      "loss": 4.5674,
      "mean_token_accuracy": 0.27252580374479296,
      "num_tokens": 159482540.0,
      "step": 69550
    },
    {
      "entropy": 5.467961931228638,
      "epoch": 6.681556195965418,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016352686220215016,
      "loss": 4.4777,
      "mean_token_accuracy": 0.2787435933947563,
      "num_tokens": 159493792.0,
      "step": 69555
    },
    {
      "entropy": 5.482243061065674,
      "epoch": 6.682036503362152,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016349708327959621,
      "loss": 4.6049,
      "mean_token_accuracy": 0.26702774614095687,
      "num_tokens": 159505449.0,
      "step": 69560
    },
    {
      "entropy": 5.525392198562622,
      "epoch": 6.6825168107588855,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016346730694580876,
      "loss": 4.6474,
      "mean_token_accuracy": 0.26573692113161085,
      "num_tokens": 159516301.0,
      "step": 69565
    },
    {
      "entropy": 5.616410636901856,
      "epoch": 6.68299711815562,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016343753320147928,
      "loss": 4.7238,
      "mean_token_accuracy": 0.2524976685643196,
      "num_tokens": 159527459.0,
      "step": 69570
    },
    {
      "entropy": 5.6090038299560545,
      "epoch": 6.683477425552353,
      "grad_norm": 1.671875,
      "learning_rate": 0.000163407762047299,
      "loss": 4.7069,
      "mean_token_accuracy": 0.2547245115041733,
      "num_tokens": 159539711.0,
      "step": 69575
    },
    {
      "entropy": 5.578363466262817,
      "epoch": 6.683957732949088,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016337799348395914,
      "loss": 4.7175,
      "mean_token_accuracy": 0.2590574875473976,
      "num_tokens": 159550524.0,
      "step": 69580
    },
    {
      "entropy": 5.565977716445923,
      "epoch": 6.684438040345821,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001633482275121508,
      "loss": 4.6776,
      "mean_token_accuracy": 0.2626384750008583,
      "num_tokens": 159562274.0,
      "step": 69585
    },
    {
      "entropy": 5.5502301216125485,
      "epoch": 6.684918347742555,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001633184641325652,
      "loss": 4.6857,
      "mean_token_accuracy": 0.2624508529901505,
      "num_tokens": 159572531.0,
      "step": 69590
    },
    {
      "entropy": 5.5378824234008786,
      "epoch": 6.685398655139289,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001632887033458931,
      "loss": 4.661,
      "mean_token_accuracy": 0.2597489610314369,
      "num_tokens": 159583923.0,
      "step": 69595
    },
    {
      "entropy": 5.489324045181275,
      "epoch": 6.685878962536023,
      "grad_norm": 1.375,
      "learning_rate": 0.00016325894515282563,
      "loss": 4.6007,
      "mean_token_accuracy": 0.2704952895641327,
      "num_tokens": 159594811.0,
      "step": 69600
    },
    {
      "entropy": 5.566879415512085,
      "epoch": 6.6863592699327565,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016322918955405364,
      "loss": 4.6215,
      "mean_token_accuracy": 0.2702565982937813,
      "num_tokens": 159606688.0,
      "step": 69605
    },
    {
      "entropy": 5.518069553375244,
      "epoch": 6.686839577329491,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000163199436550268,
      "loss": 4.618,
      "mean_token_accuracy": 0.2615752935409546,
      "num_tokens": 159618911.0,
      "step": 69610
    },
    {
      "entropy": 5.532121706008911,
      "epoch": 6.687319884726225,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016316968614215943,
      "loss": 4.5954,
      "mean_token_accuracy": 0.265307354927063,
      "num_tokens": 159631027.0,
      "step": 69615
    },
    {
      "entropy": 5.48723030090332,
      "epoch": 6.687800192122959,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016313993833041876,
      "loss": 4.5267,
      "mean_token_accuracy": 0.27789601534605024,
      "num_tokens": 159643046.0,
      "step": 69620
    },
    {
      "entropy": 5.512260675430298,
      "epoch": 6.688280499519693,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016311019311573648,
      "loss": 4.5922,
      "mean_token_accuracy": 0.2753074377775192,
      "num_tokens": 159653561.0,
      "step": 69625
    },
    {
      "entropy": 5.530674648284912,
      "epoch": 6.688760806916426,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016308045049880332,
      "loss": 4.6249,
      "mean_token_accuracy": 0.2691976338624954,
      "num_tokens": 159664390.0,
      "step": 69630
    },
    {
      "entropy": 5.59000940322876,
      "epoch": 6.689241114313161,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016305071048030976,
      "loss": 4.6923,
      "mean_token_accuracy": 0.2601704865694046,
      "num_tokens": 159676028.0,
      "step": 69635
    },
    {
      "entropy": 5.521459436416626,
      "epoch": 6.689721421709894,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016302097306094624,
      "loss": 4.5895,
      "mean_token_accuracy": 0.26556291729211806,
      "num_tokens": 159687497.0,
      "step": 69640
    },
    {
      "entropy": 5.606122207641602,
      "epoch": 6.6902017291066285,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016299123824140322,
      "loss": 4.7686,
      "mean_token_accuracy": 0.2543942630290985,
      "num_tokens": 159698799.0,
      "step": 69645
    },
    {
      "entropy": 5.5237291812896725,
      "epoch": 6.690682036503362,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016296150602237115,
      "loss": 4.6129,
      "mean_token_accuracy": 0.2624008685350418,
      "num_tokens": 159709436.0,
      "step": 69650
    },
    {
      "entropy": 5.55457444190979,
      "epoch": 6.691162343900096,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016293177640454016,
      "loss": 4.6948,
      "mean_token_accuracy": 0.2606357306241989,
      "num_tokens": 159721214.0,
      "step": 69655
    },
    {
      "entropy": 5.565516519546509,
      "epoch": 6.69164265129683,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001629020493886005,
      "loss": 4.7293,
      "mean_token_accuracy": 0.25696091502904894,
      "num_tokens": 159733811.0,
      "step": 69660
    },
    {
      "entropy": 5.54701189994812,
      "epoch": 6.692122958693564,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016287232497524246,
      "loss": 4.6195,
      "mean_token_accuracy": 0.2623715981841087,
      "num_tokens": 159745699.0,
      "step": 69665
    },
    {
      "entropy": 5.547164630889893,
      "epoch": 6.692603266090297,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016284260316515604,
      "loss": 4.5949,
      "mean_token_accuracy": 0.27277383804321287,
      "num_tokens": 159757288.0,
      "step": 69670
    },
    {
      "entropy": 5.559226942062378,
      "epoch": 6.693083573487032,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016281288395903132,
      "loss": 4.6129,
      "mean_token_accuracy": 0.26862446367740633,
      "num_tokens": 159768388.0,
      "step": 69675
    },
    {
      "entropy": 5.4591141700744625,
      "epoch": 6.693563880883765,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016278316735755837,
      "loss": 4.57,
      "mean_token_accuracy": 0.2679256424307823,
      "num_tokens": 159779867.0,
      "step": 69680
    },
    {
      "entropy": 5.51080322265625,
      "epoch": 6.6940441882804995,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016275345336142706,
      "loss": 4.591,
      "mean_token_accuracy": 0.2711101248860359,
      "num_tokens": 159790890.0,
      "step": 69685
    },
    {
      "entropy": 5.529190921783448,
      "epoch": 6.694524495677234,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016272374197132719,
      "loss": 4.6741,
      "mean_token_accuracy": 0.2638984277844429,
      "num_tokens": 159802450.0,
      "step": 69690
    },
    {
      "entropy": 5.48934588432312,
      "epoch": 6.695004803073967,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016269403318794868,
      "loss": 4.5419,
      "mean_token_accuracy": 0.2748701602220535,
      "num_tokens": 159812703.0,
      "step": 69695
    },
    {
      "entropy": 5.578380155563354,
      "epoch": 6.695485110470702,
      "grad_norm": 1.375,
      "learning_rate": 0.0001626643270119812,
      "loss": 4.7079,
      "mean_token_accuracy": 0.26039294004440305,
      "num_tokens": 159824050.0,
      "step": 69700
    },
    {
      "entropy": 5.471708154678344,
      "epoch": 6.695965417867435,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001626346234441145,
      "loss": 4.5548,
      "mean_token_accuracy": 0.27597388029098513,
      "num_tokens": 159835362.0,
      "step": 69705
    },
    {
      "entropy": 5.5629185199737545,
      "epoch": 6.696445725264169,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016260492248503828,
      "loss": 4.6463,
      "mean_token_accuracy": 0.26588855087757113,
      "num_tokens": 159845827.0,
      "step": 69710
    },
    {
      "entropy": 5.490452718734741,
      "epoch": 6.696926032660903,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001625752241354419,
      "loss": 4.5986,
      "mean_token_accuracy": 0.2655805706977844,
      "num_tokens": 159857819.0,
      "step": 69715
    },
    {
      "entropy": 5.485313034057617,
      "epoch": 6.697406340057637,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000162545528396015,
      "loss": 4.5527,
      "mean_token_accuracy": 0.2731858715415001,
      "num_tokens": 159869104.0,
      "step": 69720
    },
    {
      "entropy": 5.489582633972168,
      "epoch": 6.697886647454371,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016251583526744702,
      "loss": 4.541,
      "mean_token_accuracy": 0.2730959251523018,
      "num_tokens": 159880980.0,
      "step": 69725
    },
    {
      "entropy": 5.548802375793457,
      "epoch": 6.698366954851105,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016248614475042735,
      "loss": 4.6301,
      "mean_token_accuracy": 0.2622397497296333,
      "num_tokens": 159893179.0,
      "step": 69730
    },
    {
      "entropy": 5.503396034240723,
      "epoch": 6.698847262247838,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016245645684564525,
      "loss": 4.5494,
      "mean_token_accuracy": 0.26811895370483396,
      "num_tokens": 159904842.0,
      "step": 69735
    },
    {
      "entropy": 5.558327770233154,
      "epoch": 6.699327569644573,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016242677155379014,
      "loss": 4.6858,
      "mean_token_accuracy": 0.25566666573286057,
      "num_tokens": 159917414.0,
      "step": 69740
    },
    {
      "entropy": 5.54866361618042,
      "epoch": 6.699807877041306,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016239708887555107,
      "loss": 4.7167,
      "mean_token_accuracy": 0.2697520643472672,
      "num_tokens": 159929098.0,
      "step": 69745
    },
    {
      "entropy": 5.591428470611572,
      "epoch": 6.70028818443804,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016236740881161722,
      "loss": 4.7297,
      "mean_token_accuracy": 0.25210813730955123,
      "num_tokens": 159942292.0,
      "step": 69750
    },
    {
      "entropy": 5.527241945266724,
      "epoch": 6.700768491834774,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016233773136267764,
      "loss": 4.6206,
      "mean_token_accuracy": 0.2618432879447937,
      "num_tokens": 159954326.0,
      "step": 69755
    },
    {
      "entropy": 5.4935492992401125,
      "epoch": 6.701248799231508,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016230805652942148,
      "loss": 4.5917,
      "mean_token_accuracy": 0.26698943227529526,
      "num_tokens": 159964742.0,
      "step": 69760
    },
    {
      "entropy": 5.6195250988006595,
      "epoch": 6.7017291066282425,
      "grad_norm": 1.5625,
      "learning_rate": 0.00016227838431253754,
      "loss": 4.6951,
      "mean_token_accuracy": 0.2655677482485771,
      "num_tokens": 159975396.0,
      "step": 69765
    },
    {
      "entropy": 5.540630245208741,
      "epoch": 6.702209414024976,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016224871471271496,
      "loss": 4.6621,
      "mean_token_accuracy": 0.2677322790026665,
      "num_tokens": 159987069.0,
      "step": 69770
    },
    {
      "entropy": 5.642499399185181,
      "epoch": 6.70268972142171,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016221904773064231,
      "loss": 4.7161,
      "mean_token_accuracy": 0.2624735027551651,
      "num_tokens": 159999015.0,
      "step": 69775
    },
    {
      "entropy": 5.607342195510864,
      "epoch": 6.703170028818444,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016218938336700844,
      "loss": 4.6438,
      "mean_token_accuracy": 0.2676179140806198,
      "num_tokens": 160011850.0,
      "step": 69780
    },
    {
      "entropy": 5.5529947757720945,
      "epoch": 6.703650336215178,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016215972162250217,
      "loss": 4.574,
      "mean_token_accuracy": 0.27579513043165205,
      "num_tokens": 160022652.0,
      "step": 69785
    },
    {
      "entropy": 5.516988134384155,
      "epoch": 6.7041306436119115,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016213006249781207,
      "loss": 4.6324,
      "mean_token_accuracy": 0.2688401088118553,
      "num_tokens": 160033256.0,
      "step": 69790
    },
    {
      "entropy": 5.4728460788726805,
      "epoch": 6.704610951008646,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016210040599362675,
      "loss": 4.6104,
      "mean_token_accuracy": 0.2686884254217148,
      "num_tokens": 160044094.0,
      "step": 69795
    },
    {
      "entropy": 5.535929918289185,
      "epoch": 6.705091258405379,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016207075211063486,
      "loss": 4.6082,
      "mean_token_accuracy": 0.2660929083824158,
      "num_tokens": 160056140.0,
      "step": 69800
    },
    {
      "entropy": 5.581789350509643,
      "epoch": 6.7055715658021136,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016204110084952472,
      "loss": 4.7144,
      "mean_token_accuracy": 0.26076831966638564,
      "num_tokens": 160066728.0,
      "step": 69805
    },
    {
      "entropy": 5.564633989334107,
      "epoch": 6.706051873198847,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016201145221098474,
      "loss": 4.6194,
      "mean_token_accuracy": 0.26765458583831786,
      "num_tokens": 160079542.0,
      "step": 69810
    },
    {
      "entropy": 5.471935415267945,
      "epoch": 6.706532180595581,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00016198180619570338,
      "loss": 4.5178,
      "mean_token_accuracy": 0.27033158391714096,
      "num_tokens": 160090367.0,
      "step": 69815
    },
    {
      "entropy": 5.478032159805298,
      "epoch": 6.707012487992315,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016195216280436887,
      "loss": 4.5303,
      "mean_token_accuracy": 0.2719599217176437,
      "num_tokens": 160102233.0,
      "step": 69820
    },
    {
      "entropy": 5.5496419906616214,
      "epoch": 6.707492795389049,
      "grad_norm": 1.5625,
      "learning_rate": 0.00016192252203766944,
      "loss": 4.6916,
      "mean_token_accuracy": 0.26203186959028246,
      "num_tokens": 160114241.0,
      "step": 69825
    },
    {
      "entropy": 5.545086097717285,
      "epoch": 6.7079731027857825,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016189288389629336,
      "loss": 4.6305,
      "mean_token_accuracy": 0.26625710874795916,
      "num_tokens": 160126235.0,
      "step": 69830
    },
    {
      "entropy": 5.576013898849487,
      "epoch": 6.708453410182517,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001618632483809286,
      "loss": 4.6876,
      "mean_token_accuracy": 0.26431805193424224,
      "num_tokens": 160137588.0,
      "step": 69835
    },
    {
      "entropy": 5.549488067626953,
      "epoch": 6.708933717579251,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016183361549226326,
      "loss": 4.6494,
      "mean_token_accuracy": 0.2661689713597298,
      "num_tokens": 160149149.0,
      "step": 69840
    },
    {
      "entropy": 5.520613098144532,
      "epoch": 6.709414024975985,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016180398523098533,
      "loss": 4.6443,
      "mean_token_accuracy": 0.2640922427177429,
      "num_tokens": 160160790.0,
      "step": 69845
    },
    {
      "entropy": 5.521372985839844,
      "epoch": 6.709894332372719,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016177435759778274,
      "loss": 4.6354,
      "mean_token_accuracy": 0.2582969218492508,
      "num_tokens": 160171904.0,
      "step": 69850
    },
    {
      "entropy": 5.543326997756958,
      "epoch": 6.710374639769452,
      "grad_norm": 1.375,
      "learning_rate": 0.00016174473259334332,
      "loss": 4.5883,
      "mean_token_accuracy": 0.27139000594615936,
      "num_tokens": 160182686.0,
      "step": 69855
    },
    {
      "entropy": 5.533744239807129,
      "epoch": 6.710854947166187,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00016171511021835502,
      "loss": 4.6185,
      "mean_token_accuracy": 0.2716316804289818,
      "num_tokens": 160195654.0,
      "step": 69860
    },
    {
      "entropy": 5.557735586166382,
      "epoch": 6.71133525456292,
      "grad_norm": 1.5,
      "learning_rate": 0.00016168549047350543,
      "loss": 4.6236,
      "mean_token_accuracy": 0.2745110601186752,
      "num_tokens": 160207312.0,
      "step": 69865
    },
    {
      "entropy": 5.50056939125061,
      "epoch": 6.7118155619596545,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016165587335948222,
      "loss": 4.6695,
      "mean_token_accuracy": 0.2599162369966507,
      "num_tokens": 160219944.0,
      "step": 69870
    },
    {
      "entropy": 5.528270959854126,
      "epoch": 6.712295869356388,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016162625887697308,
      "loss": 4.6445,
      "mean_token_accuracy": 0.26428467631340025,
      "num_tokens": 160232508.0,
      "step": 69875
    },
    {
      "entropy": 5.52138843536377,
      "epoch": 6.712776176753122,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016159664702666558,
      "loss": 4.5469,
      "mean_token_accuracy": 0.2692390888929367,
      "num_tokens": 160244481.0,
      "step": 69880
    },
    {
      "entropy": 5.6046936988830565,
      "epoch": 6.713256484149856,
      "grad_norm": 1.65625,
      "learning_rate": 0.00016156703780924718,
      "loss": 4.7122,
      "mean_token_accuracy": 0.25663304179906843,
      "num_tokens": 160255885.0,
      "step": 69885
    },
    {
      "entropy": 5.574739551544189,
      "epoch": 6.71373679154659,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016153743122540544,
      "loss": 4.6807,
      "mean_token_accuracy": 0.2506943494081497,
      "num_tokens": 160267100.0,
      "step": 69890
    },
    {
      "entropy": 5.539762115478515,
      "epoch": 6.714217098943323,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001615078272758275,
      "loss": 4.6171,
      "mean_token_accuracy": 0.26958947628736496,
      "num_tokens": 160278178.0,
      "step": 69895
    },
    {
      "entropy": 5.606568336486816,
      "epoch": 6.714697406340058,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016147822596120087,
      "loss": 4.7154,
      "mean_token_accuracy": 0.2609461277723312,
      "num_tokens": 160290637.0,
      "step": 69900
    },
    {
      "entropy": 5.62084813117981,
      "epoch": 6.715177713736791,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001614486272822127,
      "loss": 4.7186,
      "mean_token_accuracy": 0.25512965172529223,
      "num_tokens": 160302954.0,
      "step": 69905
    },
    {
      "entropy": 5.539444303512573,
      "epoch": 6.7156580211335255,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016141903123955027,
      "loss": 4.6638,
      "mean_token_accuracy": 0.264400714635849,
      "num_tokens": 160314177.0,
      "step": 69910
    },
    {
      "entropy": 5.527236080169677,
      "epoch": 6.71613832853026,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016138943783390063,
      "loss": 4.6431,
      "mean_token_accuracy": 0.26934377402067183,
      "num_tokens": 160324318.0,
      "step": 69915
    },
    {
      "entropy": 5.473287963867188,
      "epoch": 6.716618635926993,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016135984706595096,
      "loss": 4.5582,
      "mean_token_accuracy": 0.26802490204572677,
      "num_tokens": 160335991.0,
      "step": 69920
    },
    {
      "entropy": 5.523803281784057,
      "epoch": 6.717098943323727,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016133025893638815,
      "loss": 4.6271,
      "mean_token_accuracy": 0.2620951935648918,
      "num_tokens": 160347612.0,
      "step": 69925
    },
    {
      "entropy": 5.508708429336548,
      "epoch": 6.717579250720461,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001613006734458992,
      "loss": 4.613,
      "mean_token_accuracy": 0.26811044961214064,
      "num_tokens": 160358816.0,
      "step": 69930
    },
    {
      "entropy": 5.530815315246582,
      "epoch": 6.718059558117195,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016127109059517098,
      "loss": 4.6309,
      "mean_token_accuracy": 0.26918611377477647,
      "num_tokens": 160370982.0,
      "step": 69935
    },
    {
      "entropy": 5.554703950881958,
      "epoch": 6.718539865513929,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001612415103848903,
      "loss": 4.6827,
      "mean_token_accuracy": 0.2624793082475662,
      "num_tokens": 160382192.0,
      "step": 69940
    },
    {
      "entropy": 5.570268344879151,
      "epoch": 6.719020172910663,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016121193281574397,
      "loss": 4.6501,
      "mean_token_accuracy": 0.2578760594129562,
      "num_tokens": 160394864.0,
      "step": 69945
    },
    {
      "entropy": 5.587197875976562,
      "epoch": 6.7195004803073966,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016118235788841877,
      "loss": 4.6668,
      "mean_token_accuracy": 0.26059727668762206,
      "num_tokens": 160406928.0,
      "step": 69950
    },
    {
      "entropy": 5.476836633682251,
      "epoch": 6.719980787704131,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016115278560360123,
      "loss": 4.5167,
      "mean_token_accuracy": 0.2726746752858162,
      "num_tokens": 160417171.0,
      "step": 69955
    },
    {
      "entropy": 5.474330425262451,
      "epoch": 6.720461095100864,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016112321596197787,
      "loss": 4.5746,
      "mean_token_accuracy": 0.2720606952905655,
      "num_tokens": 160429982.0,
      "step": 69960
    },
    {
      "entropy": 5.590037107467651,
      "epoch": 6.720941402497599,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001610936489642353,
      "loss": 4.6888,
      "mean_token_accuracy": 0.2568307802081108,
      "num_tokens": 160441247.0,
      "step": 69965
    },
    {
      "entropy": 5.473558044433593,
      "epoch": 6.721421709894332,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016106408461106,
      "loss": 4.5646,
      "mean_token_accuracy": 0.2681679934263229,
      "num_tokens": 160452426.0,
      "step": 69970
    },
    {
      "entropy": 5.521333837509156,
      "epoch": 6.721902017291066,
      "grad_norm": 1.59375,
      "learning_rate": 0.00016103452290313833,
      "loss": 4.5773,
      "mean_token_accuracy": 0.2723096191883087,
      "num_tokens": 160463104.0,
      "step": 69975
    },
    {
      "entropy": 5.557624578475952,
      "epoch": 6.7223823246878,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001610049638411567,
      "loss": 4.678,
      "mean_token_accuracy": 0.2622938483953476,
      "num_tokens": 160475020.0,
      "step": 69980
    },
    {
      "entropy": 5.4915379047393795,
      "epoch": 6.722862632084534,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016097540742580124,
      "loss": 4.5929,
      "mean_token_accuracy": 0.2681494548916817,
      "num_tokens": 160486421.0,
      "step": 69985
    },
    {
      "entropy": 5.620865726470948,
      "epoch": 6.7233429394812685,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016094585365775824,
      "loss": 4.7153,
      "mean_token_accuracy": 0.26495301723480225,
      "num_tokens": 160496935.0,
      "step": 69990
    },
    {
      "entropy": 5.542581796646118,
      "epoch": 6.723823246878002,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016091630253771384,
      "loss": 4.6453,
      "mean_token_accuracy": 0.26505800038576127,
      "num_tokens": 160507452.0,
      "step": 69995
    },
    {
      "entropy": 5.5537951469421385,
      "epoch": 6.724303554274735,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016088675406635415,
      "loss": 4.6364,
      "mean_token_accuracy": 0.2666326895356178,
      "num_tokens": 160518579.0,
      "step": 70000
    },
    {
      "entropy": 5.624405479431152,
      "epoch": 6.72478386167147,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016085720824436515,
      "loss": 4.6745,
      "mean_token_accuracy": 0.2623208165168762,
      "num_tokens": 160530510.0,
      "step": 70005
    },
    {
      "entropy": 5.546865463256836,
      "epoch": 6.725264169068204,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016082766507243286,
      "loss": 4.6928,
      "mean_token_accuracy": 0.26404906511306764,
      "num_tokens": 160542530.0,
      "step": 70010
    },
    {
      "entropy": 5.635730123519897,
      "epoch": 6.7257444764649374,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016079812455124326,
      "loss": 4.7382,
      "mean_token_accuracy": 0.25354182422161103,
      "num_tokens": 160554267.0,
      "step": 70015
    },
    {
      "entropy": 5.534042882919311,
      "epoch": 6.726224783861672,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016076858668148198,
      "loss": 4.6222,
      "mean_token_accuracy": 0.2696492627263069,
      "num_tokens": 160564913.0,
      "step": 70020
    },
    {
      "entropy": 5.491554832458496,
      "epoch": 6.726705091258405,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016073905146383495,
      "loss": 4.5829,
      "mean_token_accuracy": 0.2612353265285492,
      "num_tokens": 160575925.0,
      "step": 70025
    },
    {
      "entropy": 5.547077608108521,
      "epoch": 6.7271853986551395,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016070951889898783,
      "loss": 4.6219,
      "mean_token_accuracy": 0.25884530544281004,
      "num_tokens": 160586684.0,
      "step": 70030
    },
    {
      "entropy": 5.489803981781006,
      "epoch": 6.727665706051873,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016067998898762633,
      "loss": 4.5812,
      "mean_token_accuracy": 0.2680170089006424,
      "num_tokens": 160598064.0,
      "step": 70035
    },
    {
      "entropy": 5.479488277435303,
      "epoch": 6.728146013448607,
      "grad_norm": 1.375,
      "learning_rate": 0.00016065046173043603,
      "loss": 4.5871,
      "mean_token_accuracy": 0.2683908209204674,
      "num_tokens": 160609720.0,
      "step": 70040
    },
    {
      "entropy": 5.440539026260376,
      "epoch": 6.728626320845341,
      "grad_norm": 1.375,
      "learning_rate": 0.00016062093712810248,
      "loss": 4.5156,
      "mean_token_accuracy": 0.28347412794828414,
      "num_tokens": 160621279.0,
      "step": 70045
    },
    {
      "entropy": 5.58558874130249,
      "epoch": 6.729106628242075,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001605914151813111,
      "loss": 4.7138,
      "mean_token_accuracy": 0.26053631603717803,
      "num_tokens": 160633749.0,
      "step": 70050
    },
    {
      "entropy": 5.551487779617309,
      "epoch": 6.7295869356388085,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016056189589074742,
      "loss": 4.6885,
      "mean_token_accuracy": 0.26368962079286573,
      "num_tokens": 160645849.0,
      "step": 70055
    },
    {
      "entropy": 5.5562738418579105,
      "epoch": 6.730067243035543,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00016053237925709663,
      "loss": 4.6066,
      "mean_token_accuracy": 0.26781429499387743,
      "num_tokens": 160657160.0,
      "step": 70060
    },
    {
      "entropy": 5.480229616165161,
      "epoch": 6.730547550432277,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001605028652810441,
      "loss": 4.5735,
      "mean_token_accuracy": 0.2737630158662796,
      "num_tokens": 160668002.0,
      "step": 70065
    },
    {
      "entropy": 5.610695552825928,
      "epoch": 6.731027857829011,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016047335396327507,
      "loss": 4.6693,
      "mean_token_accuracy": 0.2624616399407387,
      "num_tokens": 160679985.0,
      "step": 70070
    },
    {
      "entropy": 5.500142049789429,
      "epoch": 6.731508165225744,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016044384530447463,
      "loss": 4.6077,
      "mean_token_accuracy": 0.2697420731186867,
      "num_tokens": 160692230.0,
      "step": 70075
    },
    {
      "entropy": 5.562235498428345,
      "epoch": 6.731988472622478,
      "grad_norm": 1.34375,
      "learning_rate": 0.000160414339305328,
      "loss": 4.7178,
      "mean_token_accuracy": 0.2502016946673393,
      "num_tokens": 160705348.0,
      "step": 70080
    },
    {
      "entropy": 5.533374071121216,
      "epoch": 6.732468780019213,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016038483596652012,
      "loss": 4.6358,
      "mean_token_accuracy": 0.26839960366487503,
      "num_tokens": 160717357.0,
      "step": 70085
    },
    {
      "entropy": 5.548200941085815,
      "epoch": 6.732949087415946,
      "grad_norm": 1.4375,
      "learning_rate": 0.00016035533528873603,
      "loss": 4.6053,
      "mean_token_accuracy": 0.26104090064764024,
      "num_tokens": 160727881.0,
      "step": 70090
    },
    {
      "entropy": 5.580974483489991,
      "epoch": 6.73342939481268,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001603258372726607,
      "loss": 4.6903,
      "mean_token_accuracy": 0.2606777399778366,
      "num_tokens": 160740733.0,
      "step": 70095
    },
    {
      "entropy": 5.559756708145142,
      "epoch": 6.733909702209414,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016029634191897885,
      "loss": 4.5777,
      "mean_token_accuracy": 0.27104204297065737,
      "num_tokens": 160751040.0,
      "step": 70100
    },
    {
      "entropy": 5.4779744148254395,
      "epoch": 6.734390009606148,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016026684922837534,
      "loss": 4.5427,
      "mean_token_accuracy": 0.27318121045827864,
      "num_tokens": 160761174.0,
      "step": 70105
    },
    {
      "entropy": 5.49974479675293,
      "epoch": 6.734870317002882,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016023735920153487,
      "loss": 4.6137,
      "mean_token_accuracy": 0.267936135828495,
      "num_tokens": 160773413.0,
      "step": 70110
    },
    {
      "entropy": 5.554651069641113,
      "epoch": 6.735350624399616,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016020787183914221,
      "loss": 4.7273,
      "mean_token_accuracy": 0.25503929853439333,
      "num_tokens": 160784771.0,
      "step": 70115
    },
    {
      "entropy": 5.479927015304566,
      "epoch": 6.735830931796349,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016017838714188186,
      "loss": 4.5573,
      "mean_token_accuracy": 0.2704597055912018,
      "num_tokens": 160795632.0,
      "step": 70120
    },
    {
      "entropy": 5.549260425567627,
      "epoch": 6.736311239193084,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016014890511043838,
      "loss": 4.6198,
      "mean_token_accuracy": 0.26760813891887664,
      "num_tokens": 160806438.0,
      "step": 70125
    },
    {
      "entropy": 5.501552581787109,
      "epoch": 6.736791546589817,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016011942574549644,
      "loss": 4.5542,
      "mean_token_accuracy": 0.26764976531267165,
      "num_tokens": 160817290.0,
      "step": 70130
    },
    {
      "entropy": 5.450233030319214,
      "epoch": 6.7372718539865515,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016008994904774017,
      "loss": 4.5745,
      "mean_token_accuracy": 0.26356320083141327,
      "num_tokens": 160829634.0,
      "step": 70135
    },
    {
      "entropy": 5.402971982955933,
      "epoch": 6.737752161383285,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001600604750178541,
      "loss": 4.492,
      "mean_token_accuracy": 0.27559252381324767,
      "num_tokens": 160839867.0,
      "step": 70140
    },
    {
      "entropy": 5.557228660583496,
      "epoch": 6.738232468780019,
      "grad_norm": 1.375,
      "learning_rate": 0.0001600310036565225,
      "loss": 4.6729,
      "mean_token_accuracy": 0.2606890946626663,
      "num_tokens": 160851827.0,
      "step": 70145
    },
    {
      "entropy": 5.5512512683868405,
      "epoch": 6.738712776176753,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001600015349644296,
      "loss": 4.6213,
      "mean_token_accuracy": 0.26708784103393557,
      "num_tokens": 160864206.0,
      "step": 70150
    },
    {
      "entropy": 5.464783954620361,
      "epoch": 6.739193083573487,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015997206894225957,
      "loss": 4.6206,
      "mean_token_accuracy": 0.2684222474694252,
      "num_tokens": 160876350.0,
      "step": 70155
    },
    {
      "entropy": 5.570570421218872,
      "epoch": 6.739673390970221,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015994260559069662,
      "loss": 4.6506,
      "mean_token_accuracy": 0.265093594789505,
      "num_tokens": 160887617.0,
      "step": 70160
    },
    {
      "entropy": 5.556715393066407,
      "epoch": 6.740153698366955,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00015991314491042467,
      "loss": 4.7023,
      "mean_token_accuracy": 0.2650743737816811,
      "num_tokens": 160900250.0,
      "step": 70165
    },
    {
      "entropy": 5.500756359100341,
      "epoch": 6.740634005763689,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015988368690212772,
      "loss": 4.5726,
      "mean_token_accuracy": 0.27094386518001556,
      "num_tokens": 160912080.0,
      "step": 70170
    },
    {
      "entropy": 5.516406488418579,
      "epoch": 6.7411143131604225,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015985423156648977,
      "loss": 4.6194,
      "mean_token_accuracy": 0.26524966210126877,
      "num_tokens": 160923557.0,
      "step": 70175
    },
    {
      "entropy": 5.518215131759644,
      "epoch": 6.741594620557157,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015982477890419462,
      "loss": 4.6178,
      "mean_token_accuracy": 0.26896006911993026,
      "num_tokens": 160935197.0,
      "step": 70180
    },
    {
      "entropy": 5.599180603027344,
      "epoch": 6.74207492795389,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015979532891592608,
      "loss": 4.7462,
      "mean_token_accuracy": 0.25890687555074693,
      "num_tokens": 160946328.0,
      "step": 70185
    },
    {
      "entropy": 5.540931606292725,
      "epoch": 6.742555235350625,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015976588160236804,
      "loss": 4.5826,
      "mean_token_accuracy": 0.267069673538208,
      "num_tokens": 160957868.0,
      "step": 70190
    },
    {
      "entropy": 5.540746355056763,
      "epoch": 6.743035542747358,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015973643696420397,
      "loss": 4.6064,
      "mean_token_accuracy": 0.26535639017820356,
      "num_tokens": 160969798.0,
      "step": 70195
    },
    {
      "entropy": 5.555325078964233,
      "epoch": 6.743515850144092,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015970699500211756,
      "loss": 4.658,
      "mean_token_accuracy": 0.26851325631141665,
      "num_tokens": 160980873.0,
      "step": 70200
    },
    {
      "entropy": 5.524410724639893,
      "epoch": 6.743996157540826,
      "grad_norm": 1.5,
      "learning_rate": 0.00015967755571679238,
      "loss": 4.6496,
      "mean_token_accuracy": 0.2650640994310379,
      "num_tokens": 160991327.0,
      "step": 70205
    },
    {
      "entropy": 5.493148326873779,
      "epoch": 6.74447646493756,
      "grad_norm": 1.375,
      "learning_rate": 0.00015964811910891192,
      "loss": 4.6029,
      "mean_token_accuracy": 0.2680926710367203,
      "num_tokens": 161002798.0,
      "step": 70210
    },
    {
      "entropy": 5.6310944080352785,
      "epoch": 6.744956772334294,
      "grad_norm": 1.5,
      "learning_rate": 0.00015961868517915963,
      "loss": 4.7095,
      "mean_token_accuracy": 0.25797266513109207,
      "num_tokens": 161013712.0,
      "step": 70215
    },
    {
      "entropy": 5.54282054901123,
      "epoch": 6.745437079731028,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015958925392821889,
      "loss": 4.6629,
      "mean_token_accuracy": 0.26592428982257843,
      "num_tokens": 161025128.0,
      "step": 70220
    },
    {
      "entropy": 5.472967624664307,
      "epoch": 6.745917387127761,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015955982535677293,
      "loss": 4.558,
      "mean_token_accuracy": 0.2693632811307907,
      "num_tokens": 161036703.0,
      "step": 70225
    },
    {
      "entropy": 5.485302925109863,
      "epoch": 6.746397694524496,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000159530399465505,
      "loss": 4.5399,
      "mean_token_accuracy": 0.27433956116437913,
      "num_tokens": 161049063.0,
      "step": 70230
    },
    {
      "entropy": 5.5795002460479735,
      "epoch": 6.74687800192123,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015950097625509838,
      "loss": 4.7234,
      "mean_token_accuracy": 0.25710154324769974,
      "num_tokens": 161060479.0,
      "step": 70235
    },
    {
      "entropy": 5.576026153564453,
      "epoch": 6.747358309317963,
      "grad_norm": 1.5,
      "learning_rate": 0.00015947155572623606,
      "loss": 4.6725,
      "mean_token_accuracy": 0.26233800053596495,
      "num_tokens": 161071617.0,
      "step": 70240
    },
    {
      "entropy": 5.582014036178589,
      "epoch": 6.747838616714698,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015944213787960125,
      "loss": 4.7105,
      "mean_token_accuracy": 0.2603239968419075,
      "num_tokens": 161083675.0,
      "step": 70245
    },
    {
      "entropy": 5.533619451522827,
      "epoch": 6.748318924111431,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015941272271587686,
      "loss": 4.6345,
      "mean_token_accuracy": 0.26431943476200104,
      "num_tokens": 161096003.0,
      "step": 70250
    },
    {
      "entropy": 5.62551736831665,
      "epoch": 6.7487992315081655,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001593833102357458,
      "loss": 4.7175,
      "mean_token_accuracy": 0.2577503383159637,
      "num_tokens": 161106883.0,
      "step": 70255
    },
    {
      "entropy": 5.537777996063232,
      "epoch": 6.749279538904899,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015935390043989096,
      "loss": 4.5975,
      "mean_token_accuracy": 0.267477510869503,
      "num_tokens": 161117938.0,
      "step": 70260
    },
    {
      "entropy": 5.500053930282593,
      "epoch": 6.749759846301633,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015932449332899514,
      "loss": 4.5926,
      "mean_token_accuracy": 0.25968945026397705,
      "num_tokens": 161128526.0,
      "step": 70265
    },
    {
      "entropy": 5.575167512893676,
      "epoch": 6.750240153698367,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015929508890374108,
      "loss": 4.7182,
      "mean_token_accuracy": 0.25488104224205016,
      "num_tokens": 161140889.0,
      "step": 70270
    },
    {
      "entropy": 5.5627013683319095,
      "epoch": 6.750720461095101,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015926568716481156,
      "loss": 4.6645,
      "mean_token_accuracy": 0.25887651592493055,
      "num_tokens": 161152070.0,
      "step": 70275
    },
    {
      "entropy": 5.50691442489624,
      "epoch": 6.7512007684918345,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001592362881128891,
      "loss": 4.5944,
      "mean_token_accuracy": 0.26469837725162504,
      "num_tokens": 161162655.0,
      "step": 70280
    },
    {
      "entropy": 5.510850429534912,
      "epoch": 6.751681075888569,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015920689174865626,
      "loss": 4.5967,
      "mean_token_accuracy": 0.26763464510440826,
      "num_tokens": 161173844.0,
      "step": 70285
    },
    {
      "entropy": 5.5253581523895265,
      "epoch": 6.752161383285302,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015917749807279556,
      "loss": 4.6514,
      "mean_token_accuracy": 0.263621611893177,
      "num_tokens": 161184254.0,
      "step": 70290
    },
    {
      "entropy": 5.613457584381104,
      "epoch": 6.752641690682037,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00015914810708598942,
      "loss": 4.7243,
      "mean_token_accuracy": 0.26233811378479005,
      "num_tokens": 161194847.0,
      "step": 70295
    },
    {
      "entropy": 5.521867799758911,
      "epoch": 6.75312199807877,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001591187187889202,
      "loss": 4.6215,
      "mean_token_accuracy": 0.2665195196866989,
      "num_tokens": 161205691.0,
      "step": 70300
    },
    {
      "entropy": 5.602354860305786,
      "epoch": 6.753602305475504,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015908933318227025,
      "loss": 4.6949,
      "mean_token_accuracy": 0.26219747215509415,
      "num_tokens": 161216444.0,
      "step": 70305
    },
    {
      "entropy": 5.621389770507813,
      "epoch": 6.754082612872239,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015905995026672192,
      "loss": 4.6908,
      "mean_token_accuracy": 0.2638693392276764,
      "num_tokens": 161228554.0,
      "step": 70310
    },
    {
      "entropy": 5.553325176239014,
      "epoch": 6.754562920268972,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015903057004295715,
      "loss": 4.6513,
      "mean_token_accuracy": 0.2677394852042198,
      "num_tokens": 161239690.0,
      "step": 70315
    },
    {
      "entropy": 5.447436141967773,
      "epoch": 6.755043227665706,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015900119251165818,
      "loss": 4.5473,
      "mean_token_accuracy": 0.28023826479911806,
      "num_tokens": 161251774.0,
      "step": 70320
    },
    {
      "entropy": 5.516789817810059,
      "epoch": 6.75552353506244,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001589718176735071,
      "loss": 4.6077,
      "mean_token_accuracy": 0.2711781248450279,
      "num_tokens": 161263059.0,
      "step": 70325
    },
    {
      "entropy": 5.485160255432129,
      "epoch": 6.756003842459174,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001589424455291859,
      "loss": 4.5856,
      "mean_token_accuracy": 0.2723499462008476,
      "num_tokens": 161273544.0,
      "step": 70330
    },
    {
      "entropy": 5.469732332229614,
      "epoch": 6.756484149855908,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015891307607937644,
      "loss": 4.5792,
      "mean_token_accuracy": 0.2721803277730942,
      "num_tokens": 161284936.0,
      "step": 70335
    },
    {
      "entropy": 5.460144138336181,
      "epoch": 6.756964457252642,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015888370932476076,
      "loss": 4.5493,
      "mean_token_accuracy": 0.26851639747619627,
      "num_tokens": 161296299.0,
      "step": 70340
    },
    {
      "entropy": 5.51510910987854,
      "epoch": 6.757444764649375,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015885434526602044,
      "loss": 4.5772,
      "mean_token_accuracy": 0.2706740409135818,
      "num_tokens": 161307646.0,
      "step": 70345
    },
    {
      "entropy": 5.474975490570069,
      "epoch": 6.75792507204611,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015882498390383737,
      "loss": 4.5752,
      "mean_token_accuracy": 0.2675823047757149,
      "num_tokens": 161319497.0,
      "step": 70350
    },
    {
      "entropy": 5.526910209655762,
      "epoch": 6.758405379442843,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015879562523889317,
      "loss": 4.6394,
      "mean_token_accuracy": 0.2599660515785217,
      "num_tokens": 161330439.0,
      "step": 70355
    },
    {
      "entropy": 5.467951059341431,
      "epoch": 6.7588856868395775,
      "grad_norm": 1.640625,
      "learning_rate": 0.00015876626927186958,
      "loss": 4.6482,
      "mean_token_accuracy": 0.2635582447052002,
      "num_tokens": 161341145.0,
      "step": 70360
    },
    {
      "entropy": 5.54323902130127,
      "epoch": 6.759365994236311,
      "grad_norm": 1.375,
      "learning_rate": 0.00015873691600344804,
      "loss": 4.6833,
      "mean_token_accuracy": 0.2630552724003792,
      "num_tokens": 161352491.0,
      "step": 70365
    },
    {
      "entropy": 5.50614743232727,
      "epoch": 6.759846301633045,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015870756543431012,
      "loss": 4.6005,
      "mean_token_accuracy": 0.2622902363538742,
      "num_tokens": 161364963.0,
      "step": 70370
    },
    {
      "entropy": 5.559150838851929,
      "epoch": 6.760326609029779,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015867821756513717,
      "loss": 4.6741,
      "mean_token_accuracy": 0.2665116682648659,
      "num_tokens": 161376109.0,
      "step": 70375
    },
    {
      "entropy": 5.50013427734375,
      "epoch": 6.760806916426513,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015864887239661058,
      "loss": 4.571,
      "mean_token_accuracy": 0.2700946509838104,
      "num_tokens": 161386601.0,
      "step": 70380
    },
    {
      "entropy": 5.5820556640625,
      "epoch": 6.761287223823247,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001586195299294117,
      "loss": 4.7515,
      "mean_token_accuracy": 0.25831920504570005,
      "num_tokens": 161398437.0,
      "step": 70385
    },
    {
      "entropy": 5.57558536529541,
      "epoch": 6.761767531219981,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001585901901642217,
      "loss": 4.6306,
      "mean_token_accuracy": 0.26254093945026397,
      "num_tokens": 161408470.0,
      "step": 70390
    },
    {
      "entropy": 5.565252065658569,
      "epoch": 6.762247838616715,
      "grad_norm": 1.375,
      "learning_rate": 0.0001585608531017219,
      "loss": 4.6597,
      "mean_token_accuracy": 0.26127553880214693,
      "num_tokens": 161420137.0,
      "step": 70395
    },
    {
      "entropy": 5.567739868164063,
      "epoch": 6.7627281460134485,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001585315187425934,
      "loss": 4.6483,
      "mean_token_accuracy": 0.26051474958658216,
      "num_tokens": 161430939.0,
      "step": 70400
    },
    {
      "entropy": 5.499185419082641,
      "epoch": 6.763208453410183,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015850218708751706,
      "loss": 4.6475,
      "mean_token_accuracy": 0.25701753944158556,
      "num_tokens": 161442102.0,
      "step": 70405
    },
    {
      "entropy": 5.643463611602783,
      "epoch": 6.763688760806916,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015847285813717408,
      "loss": 4.7418,
      "mean_token_accuracy": 0.2564899563789368,
      "num_tokens": 161452887.0,
      "step": 70410
    },
    {
      "entropy": 5.5194016456604,
      "epoch": 6.764169068203651,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015844353189224526,
      "loss": 4.6237,
      "mean_token_accuracy": 0.2742712140083313,
      "num_tokens": 161463182.0,
      "step": 70415
    },
    {
      "entropy": 5.510123348236084,
      "epoch": 6.764649375600384,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015841420835341157,
      "loss": 4.5888,
      "mean_token_accuracy": 0.2758236795663834,
      "num_tokens": 161473824.0,
      "step": 70420
    },
    {
      "entropy": 5.4475609302520756,
      "epoch": 6.765129682997118,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001583848875213537,
      "loss": 4.5484,
      "mean_token_accuracy": 0.27116160094738007,
      "num_tokens": 161484932.0,
      "step": 70425
    },
    {
      "entropy": 5.4652632713317875,
      "epoch": 6.765609990393852,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015835556939675267,
      "loss": 4.587,
      "mean_token_accuracy": 0.2679688796401024,
      "num_tokens": 161495812.0,
      "step": 70430
    },
    {
      "entropy": 5.548689651489258,
      "epoch": 6.766090297790586,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001583262539802888,
      "loss": 4.6868,
      "mean_token_accuracy": 0.2595934748649597,
      "num_tokens": 161507201.0,
      "step": 70435
    },
    {
      "entropy": 5.578254127502442,
      "epoch": 6.76657060518732,
      "grad_norm": 1.5,
      "learning_rate": 0.00015829694127264285,
      "loss": 4.6506,
      "mean_token_accuracy": 0.2578978046774864,
      "num_tokens": 161517548.0,
      "step": 70440
    },
    {
      "entropy": 5.549007749557495,
      "epoch": 6.767050912584054,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001582676312744954,
      "loss": 4.6686,
      "mean_token_accuracy": 0.26550923883914945,
      "num_tokens": 161529417.0,
      "step": 70445
    },
    {
      "entropy": 5.511301851272583,
      "epoch": 6.767531219980787,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015823832398652692,
      "loss": 4.6009,
      "mean_token_accuracy": 0.26615088284015653,
      "num_tokens": 161542007.0,
      "step": 70450
    },
    {
      "entropy": 5.5259183883667,
      "epoch": 6.768011527377522,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001582090194094179,
      "loss": 4.6048,
      "mean_token_accuracy": 0.25821964591741564,
      "num_tokens": 161553545.0,
      "step": 70455
    },
    {
      "entropy": 5.496645879745484,
      "epoch": 6.768491834774256,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015817971754384868,
      "loss": 4.5407,
      "mean_token_accuracy": 0.270071367919445,
      "num_tokens": 161566207.0,
      "step": 70460
    },
    {
      "entropy": 5.506757020950317,
      "epoch": 6.768972142170989,
      "grad_norm": 1.375,
      "learning_rate": 0.0001581504183904995,
      "loss": 4.5905,
      "mean_token_accuracy": 0.2642368242144585,
      "num_tokens": 161576915.0,
      "step": 70465
    },
    {
      "entropy": 5.532187843322754,
      "epoch": 6.769452449567723,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015812112195005056,
      "loss": 4.6245,
      "mean_token_accuracy": 0.2623932495713234,
      "num_tokens": 161588447.0,
      "step": 70470
    },
    {
      "entropy": 5.5675647258758545,
      "epoch": 6.769932756964457,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015809182822318216,
      "loss": 4.6695,
      "mean_token_accuracy": 0.26118344217538836,
      "num_tokens": 161599548.0,
      "step": 70475
    },
    {
      "entropy": 5.559930086135864,
      "epoch": 6.7704130643611915,
      "grad_norm": 1.375,
      "learning_rate": 0.00015806253721057433,
      "loss": 4.6127,
      "mean_token_accuracy": 0.2686581969261169,
      "num_tokens": 161611099.0,
      "step": 70480
    },
    {
      "entropy": 5.506870222091675,
      "epoch": 6.770893371757925,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001580332489129072,
      "loss": 4.6304,
      "mean_token_accuracy": 0.2643526911735535,
      "num_tokens": 161623802.0,
      "step": 70485
    },
    {
      "entropy": 5.559726905822754,
      "epoch": 6.771373679154659,
      "grad_norm": 1.578125,
      "learning_rate": 0.00015800396333086076,
      "loss": 4.6532,
      "mean_token_accuracy": 0.26550513356924055,
      "num_tokens": 161633993.0,
      "step": 70490
    },
    {
      "entropy": 5.470870590209961,
      "epoch": 6.771853986551393,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015797468046511483,
      "loss": 4.6215,
      "mean_token_accuracy": 0.2703682407736778,
      "num_tokens": 161646564.0,
      "step": 70495
    },
    {
      "entropy": 5.6195151805877686,
      "epoch": 6.772334293948127,
      "grad_norm": 1.375,
      "learning_rate": 0.0001579454003163493,
      "loss": 4.7604,
      "mean_token_accuracy": 0.25463211685419085,
      "num_tokens": 161659080.0,
      "step": 70500
    },
    {
      "entropy": 5.554838514328003,
      "epoch": 6.7728146013448605,
      "grad_norm": 1.28125,
      "learning_rate": 0.000157916122885244,
      "loss": 4.6631,
      "mean_token_accuracy": 0.2613470897078514,
      "num_tokens": 161670927.0,
      "step": 70505
    },
    {
      "entropy": 5.585502290725708,
      "epoch": 6.773294908741595,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001578868481724786,
      "loss": 4.6971,
      "mean_token_accuracy": 0.2613661989569664,
      "num_tokens": 161682035.0,
      "step": 70510
    },
    {
      "entropy": 5.574312591552735,
      "epoch": 6.773775216138328,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015785757617873294,
      "loss": 4.6836,
      "mean_token_accuracy": 0.2660604193806648,
      "num_tokens": 161692812.0,
      "step": 70515
    },
    {
      "entropy": 5.5219450950622555,
      "epoch": 6.7742555235350626,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015782830690468653,
      "loss": 4.6623,
      "mean_token_accuracy": 0.25923632830381393,
      "num_tokens": 161703774.0,
      "step": 70520
    },
    {
      "entropy": 5.507394552230835,
      "epoch": 6.774735830931796,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015779904035101886,
      "loss": 4.5836,
      "mean_token_accuracy": 0.2733612865209579,
      "num_tokens": 161713707.0,
      "step": 70525
    },
    {
      "entropy": 5.498645401000976,
      "epoch": 6.77521613832853,
      "grad_norm": 1.5,
      "learning_rate": 0.00015776977651840946,
      "loss": 4.5927,
      "mean_token_accuracy": 0.27054265737533567,
      "num_tokens": 161723603.0,
      "step": 70530
    },
    {
      "entropy": 5.577858543395996,
      "epoch": 6.775696445725265,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001577405154075377,
      "loss": 4.6917,
      "mean_token_accuracy": 0.259021857380867,
      "num_tokens": 161735340.0,
      "step": 70535
    },
    {
      "entropy": 5.590332937240601,
      "epoch": 6.776176753121998,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000157711257019083,
      "loss": 4.7563,
      "mean_token_accuracy": 0.25835304856300356,
      "num_tokens": 161746959.0,
      "step": 70540
    },
    {
      "entropy": 5.543303489685059,
      "epoch": 6.7766570605187315,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015768200135372467,
      "loss": 4.7182,
      "mean_token_accuracy": 0.2597611710429192,
      "num_tokens": 161759349.0,
      "step": 70545
    },
    {
      "entropy": 5.606799173355102,
      "epoch": 6.777137367915466,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015765274841214195,
      "loss": 4.6472,
      "mean_token_accuracy": 0.270685613155365,
      "num_tokens": 161771094.0,
      "step": 70550
    },
    {
      "entropy": 5.538992643356323,
      "epoch": 6.7776176753122,
      "grad_norm": 1.375,
      "learning_rate": 0.00015762349819501393,
      "loss": 4.6287,
      "mean_token_accuracy": 0.2574970290064812,
      "num_tokens": 161781494.0,
      "step": 70555
    },
    {
      "entropy": 5.564244127273559,
      "epoch": 6.778097982708934,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001575942507030197,
      "loss": 4.7316,
      "mean_token_accuracy": 0.24866251796483993,
      "num_tokens": 161793062.0,
      "step": 70560
    },
    {
      "entropy": 5.624715042114258,
      "epoch": 6.778578290105668,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015756500593683837,
      "loss": 4.6906,
      "mean_token_accuracy": 0.26519800573587415,
      "num_tokens": 161803566.0,
      "step": 70565
    },
    {
      "entropy": 5.428567600250244,
      "epoch": 6.779058597502401,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015753576389714887,
      "loss": 4.4685,
      "mean_token_accuracy": 0.2795730084180832,
      "num_tokens": 161814958.0,
      "step": 70570
    },
    {
      "entropy": 5.516906976699829,
      "epoch": 6.779538904899136,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001575065245846301,
      "loss": 4.64,
      "mean_token_accuracy": 0.26126119643449786,
      "num_tokens": 161826562.0,
      "step": 70575
    },
    {
      "entropy": 5.473510789871216,
      "epoch": 6.780019212295869,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015747728799996103,
      "loss": 4.5648,
      "mean_token_accuracy": 0.27369713187217715,
      "num_tokens": 161837361.0,
      "step": 70580
    },
    {
      "entropy": 5.503241920471192,
      "epoch": 6.7804995196926034,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001574480541438203,
      "loss": 4.6584,
      "mean_token_accuracy": 0.26199175864458085,
      "num_tokens": 161849831.0,
      "step": 70585
    },
    {
      "entropy": 5.537655067443848,
      "epoch": 6.780979827089337,
      "grad_norm": 1.375,
      "learning_rate": 0.0001574188230168868,
      "loss": 4.6979,
      "mean_token_accuracy": 0.2626100018620491,
      "num_tokens": 161862715.0,
      "step": 70590
    },
    {
      "entropy": 5.511038160324096,
      "epoch": 6.781460134486071,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000157389594619839,
      "loss": 4.5303,
      "mean_token_accuracy": 0.2722605764865875,
      "num_tokens": 161873756.0,
      "step": 70595
    },
    {
      "entropy": 5.510903072357178,
      "epoch": 6.781940441882805,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001573603689533556,
      "loss": 4.5612,
      "mean_token_accuracy": 0.2670850858092308,
      "num_tokens": 161886041.0,
      "step": 70600
    },
    {
      "entropy": 5.474885892868042,
      "epoch": 6.782420749279539,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001573311460181151,
      "loss": 4.5492,
      "mean_token_accuracy": 0.27098310440778733,
      "num_tokens": 161897638.0,
      "step": 70605
    },
    {
      "entropy": 5.504368305206299,
      "epoch": 6.782901056676273,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015730192581479597,
      "loss": 4.6182,
      "mean_token_accuracy": 0.26793684512376786,
      "num_tokens": 161908320.0,
      "step": 70610
    },
    {
      "entropy": 5.5045562267303465,
      "epoch": 6.783381364073007,
      "grad_norm": 1.5,
      "learning_rate": 0.00015727270834407664,
      "loss": 4.6021,
      "mean_token_accuracy": 0.26707864105701445,
      "num_tokens": 161919099.0,
      "step": 70615
    },
    {
      "entropy": 5.492359066009522,
      "epoch": 6.78386167146974,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00015724349360663546,
      "loss": 4.6224,
      "mean_token_accuracy": 0.2651988387107849,
      "num_tokens": 161928972.0,
      "step": 70620
    },
    {
      "entropy": 5.5626473903656,
      "epoch": 6.7843419788664745,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015721428160315066,
      "loss": 4.6861,
      "mean_token_accuracy": 0.26644949465990064,
      "num_tokens": 161940894.0,
      "step": 70625
    },
    {
      "entropy": 5.51218729019165,
      "epoch": 6.784822286263209,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015718507233430064,
      "loss": 4.6691,
      "mean_token_accuracy": 0.2648956671357155,
      "num_tokens": 161952955.0,
      "step": 70630
    },
    {
      "entropy": 5.578524923324585,
      "epoch": 6.785302593659942,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015715586580076325,
      "loss": 4.6486,
      "mean_token_accuracy": 0.2575831800699234,
      "num_tokens": 161964367.0,
      "step": 70635
    },
    {
      "entropy": 5.467731142044068,
      "epoch": 6.785782901056677,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015712666200321673,
      "loss": 4.5194,
      "mean_token_accuracy": 0.27754788994789126,
      "num_tokens": 161975396.0,
      "step": 70640
    },
    {
      "entropy": 5.5638659477233885,
      "epoch": 6.78626320845341,
      "grad_norm": 1.578125,
      "learning_rate": 0.00015709746094233918,
      "loss": 4.7121,
      "mean_token_accuracy": 0.25808198004961014,
      "num_tokens": 161986261.0,
      "step": 70645
    },
    {
      "entropy": 5.523604822158814,
      "epoch": 6.786743515850144,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001570682626188084,
      "loss": 4.626,
      "mean_token_accuracy": 0.26568560898303983,
      "num_tokens": 161997329.0,
      "step": 70650
    },
    {
      "entropy": 5.5602698802948,
      "epoch": 6.787223823246878,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001570390670333024,
      "loss": 4.658,
      "mean_token_accuracy": 0.2570517435669899,
      "num_tokens": 162009356.0,
      "step": 70655
    },
    {
      "entropy": 5.494174337387085,
      "epoch": 6.787704130643612,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015700987418649904,
      "loss": 4.6084,
      "mean_token_accuracy": 0.26591404974460603,
      "num_tokens": 162020065.0,
      "step": 70660
    },
    {
      "entropy": 5.5619104385375975,
      "epoch": 6.7881844380403455,
      "grad_norm": 1.375,
      "learning_rate": 0.000156980684079076,
      "loss": 4.6497,
      "mean_token_accuracy": 0.2594912499189377,
      "num_tokens": 162031191.0,
      "step": 70665
    },
    {
      "entropy": 5.525353860855103,
      "epoch": 6.78866474543708,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001569514967117111,
      "loss": 4.6281,
      "mean_token_accuracy": 0.26787602007389066,
      "num_tokens": 162042576.0,
      "step": 70670
    },
    {
      "entropy": 5.6493896484375,
      "epoch": 6.789145052833813,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015692231208508186,
      "loss": 4.7132,
      "mean_token_accuracy": 0.2608063742518425,
      "num_tokens": 162054530.0,
      "step": 70675
    },
    {
      "entropy": 5.60210952758789,
      "epoch": 6.789625360230548,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001568931301998659,
      "loss": 4.7605,
      "mean_token_accuracy": 0.255485363304615,
      "num_tokens": 162065568.0,
      "step": 70680
    },
    {
      "entropy": 5.503751039505005,
      "epoch": 6.790105667627281,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015686395105674077,
      "loss": 4.5825,
      "mean_token_accuracy": 0.2700812056660652,
      "num_tokens": 162078333.0,
      "step": 70685
    },
    {
      "entropy": 5.513127756118775,
      "epoch": 6.790585975024015,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015683477465638392,
      "loss": 4.6315,
      "mean_token_accuracy": 0.2739291563630104,
      "num_tokens": 162089494.0,
      "step": 70690
    },
    {
      "entropy": 5.53506326675415,
      "epoch": 6.791066282420749,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015680560099947267,
      "loss": 4.673,
      "mean_token_accuracy": 0.26538159549236295,
      "num_tokens": 162101600.0,
      "step": 70695
    },
    {
      "entropy": 5.527905035018921,
      "epoch": 6.791546589817483,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015677643008668452,
      "loss": 4.5536,
      "mean_token_accuracy": 0.2721947118639946,
      "num_tokens": 162112647.0,
      "step": 70700
    },
    {
      "entropy": 5.622681522369385,
      "epoch": 6.7920268972142175,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015674726191869652,
      "loss": 4.7349,
      "mean_token_accuracy": 0.2576187789440155,
      "num_tokens": 162124819.0,
      "step": 70705
    },
    {
      "entropy": 5.51520562171936,
      "epoch": 6.792507204610951,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015671809649618596,
      "loss": 4.6008,
      "mean_token_accuracy": 0.26592584103345873,
      "num_tokens": 162135177.0,
      "step": 70710
    },
    {
      "entropy": 5.479788684844971,
      "epoch": 6.792987512007685,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015668893381982996,
      "loss": 4.6208,
      "mean_token_accuracy": 0.2639024555683136,
      "num_tokens": 162145464.0,
      "step": 70715
    },
    {
      "entropy": 5.657188415527344,
      "epoch": 6.793467819404419,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015665977389030562,
      "loss": 4.7964,
      "mean_token_accuracy": 0.2558334141969681,
      "num_tokens": 162156657.0,
      "step": 70720
    },
    {
      "entropy": 5.5266945362091064,
      "epoch": 6.793948126801153,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015663061670828993,
      "loss": 4.5712,
      "mean_token_accuracy": 0.2713169664144516,
      "num_tokens": 162168153.0,
      "step": 70725
    },
    {
      "entropy": 5.553949785232544,
      "epoch": 6.7944284341978864,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015660146227445992,
      "loss": 4.6975,
      "mean_token_accuracy": 0.26330935806035993,
      "num_tokens": 162179888.0,
      "step": 70730
    },
    {
      "entropy": 5.5655231952667235,
      "epoch": 6.794908741594621,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001565723105894923,
      "loss": 4.6435,
      "mean_token_accuracy": 0.26947763562202454,
      "num_tokens": 162190380.0,
      "step": 70735
    },
    {
      "entropy": 5.588271427154541,
      "epoch": 6.795389048991354,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015654316165406398,
      "loss": 4.7102,
      "mean_token_accuracy": 0.2613353535532951,
      "num_tokens": 162201929.0,
      "step": 70740
    },
    {
      "entropy": 5.575709056854248,
      "epoch": 6.7958693563880885,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001565140154688517,
      "loss": 4.6612,
      "mean_token_accuracy": 0.2630887687206268,
      "num_tokens": 162213394.0,
      "step": 70745
    },
    {
      "entropy": 5.586643266677856,
      "epoch": 6.796349663784822,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015648487203453215,
      "loss": 4.6766,
      "mean_token_accuracy": 0.26306650042533875,
      "num_tokens": 162223666.0,
      "step": 70750
    },
    {
      "entropy": 5.514015245437622,
      "epoch": 6.796829971181556,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00015645573135178193,
      "loss": 4.619,
      "mean_token_accuracy": 0.26690651923418046,
      "num_tokens": 162233277.0,
      "step": 70755
    },
    {
      "entropy": 5.500656461715698,
      "epoch": 6.79731027857829,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015642659342127773,
      "loss": 4.6112,
      "mean_token_accuracy": 0.26393335610628127,
      "num_tokens": 162244433.0,
      "step": 70760
    },
    {
      "entropy": 5.534213829040527,
      "epoch": 6.797790585975024,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015639745824369585,
      "loss": 4.6345,
      "mean_token_accuracy": 0.26086017191410066,
      "num_tokens": 162255550.0,
      "step": 70765
    },
    {
      "entropy": 5.531907367706299,
      "epoch": 6.7982708933717575,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015636832581971283,
      "loss": 4.6199,
      "mean_token_accuracy": 0.2653211370110512,
      "num_tokens": 162266406.0,
      "step": 70770
    },
    {
      "entropy": 5.517665529251099,
      "epoch": 6.798751200768492,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015633919615000498,
      "loss": 4.5667,
      "mean_token_accuracy": 0.2730701595544815,
      "num_tokens": 162277623.0,
      "step": 70775
    },
    {
      "entropy": 5.545091581344605,
      "epoch": 6.799231508165226,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015631006923524866,
      "loss": 4.6392,
      "mean_token_accuracy": 0.26715577095746995,
      "num_tokens": 162289816.0,
      "step": 70780
    },
    {
      "entropy": 5.527380657196045,
      "epoch": 6.79971181556196,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001562809450761201,
      "loss": 4.6204,
      "mean_token_accuracy": 0.26718587428331375,
      "num_tokens": 162301162.0,
      "step": 70785
    },
    {
      "entropy": 5.532938241958618,
      "epoch": 6.800192122958694,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001562518236732956,
      "loss": 4.6082,
      "mean_token_accuracy": 0.2709589093923569,
      "num_tokens": 162312619.0,
      "step": 70790
    },
    {
      "entropy": 5.467314434051514,
      "epoch": 6.800672430355427,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000156222705027451,
      "loss": 4.6147,
      "mean_token_accuracy": 0.2757339358329773,
      "num_tokens": 162323821.0,
      "step": 70795
    },
    {
      "entropy": 5.472148656845093,
      "epoch": 6.801152737752162,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015619358913926252,
      "loss": 4.4873,
      "mean_token_accuracy": 0.27320210337638856,
      "num_tokens": 162334322.0,
      "step": 70800
    },
    {
      "entropy": 5.574123907089233,
      "epoch": 6.801633045148895,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001561644760094061,
      "loss": 4.7009,
      "mean_token_accuracy": 0.25515635013580323,
      "num_tokens": 162345353.0,
      "step": 70805
    },
    {
      "entropy": 5.525879383087158,
      "epoch": 6.802113352545629,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001561353656385577,
      "loss": 4.6074,
      "mean_token_accuracy": 0.2754709303379059,
      "num_tokens": 162357188.0,
      "step": 70810
    },
    {
      "entropy": 5.567653942108154,
      "epoch": 6.802593659942363,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015610625802739314,
      "loss": 4.6642,
      "mean_token_accuracy": 0.26208713948726653,
      "num_tokens": 162368348.0,
      "step": 70815
    },
    {
      "entropy": 5.407168960571289,
      "epoch": 6.803073967339097,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015607715317658832,
      "loss": 4.5678,
      "mean_token_accuracy": 0.2694502905011177,
      "num_tokens": 162381205.0,
      "step": 70820
    },
    {
      "entropy": 5.4694184303283695,
      "epoch": 6.803554274735831,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001560480510868188,
      "loss": 4.59,
      "mean_token_accuracy": 0.2703677713871002,
      "num_tokens": 162393863.0,
      "step": 70825
    },
    {
      "entropy": 5.578182649612427,
      "epoch": 6.804034582132565,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001560189517587603,
      "loss": 4.7104,
      "mean_token_accuracy": 0.254748971760273,
      "num_tokens": 162406917.0,
      "step": 70830
    },
    {
      "entropy": 5.475598764419556,
      "epoch": 6.804514889529298,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015598985519308842,
      "loss": 4.5721,
      "mean_token_accuracy": 0.2722257345914841,
      "num_tokens": 162417745.0,
      "step": 70835
    },
    {
      "entropy": 5.551700782775879,
      "epoch": 6.804995196926033,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015596076139047873,
      "loss": 4.6534,
      "mean_token_accuracy": 0.2662879467010498,
      "num_tokens": 162429868.0,
      "step": 70840
    },
    {
      "entropy": 5.472982168197632,
      "epoch": 6.805475504322766,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001559316703516067,
      "loss": 4.5831,
      "mean_token_accuracy": 0.2705939456820488,
      "num_tokens": 162441838.0,
      "step": 70845
    },
    {
      "entropy": 5.439602470397949,
      "epoch": 6.8059558117195005,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015590258207714774,
      "loss": 4.5797,
      "mean_token_accuracy": 0.27426624596118926,
      "num_tokens": 162452448.0,
      "step": 70850
    },
    {
      "entropy": 5.570477533340454,
      "epoch": 6.806436119116235,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015587349656777722,
      "loss": 4.6137,
      "mean_token_accuracy": 0.26300616562366486,
      "num_tokens": 162463593.0,
      "step": 70855
    },
    {
      "entropy": 5.5209818363189695,
      "epoch": 6.806916426512968,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001558444138241703,
      "loss": 4.6722,
      "mean_token_accuracy": 0.271072256565094,
      "num_tokens": 162473604.0,
      "step": 70860
    },
    {
      "entropy": 5.554390287399292,
      "epoch": 6.807396733909703,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001558153338470023,
      "loss": 4.6972,
      "mean_token_accuracy": 0.26038604229688644,
      "num_tokens": 162485330.0,
      "step": 70865
    },
    {
      "entropy": 5.560159301757812,
      "epoch": 6.807877041306436,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015578625663694833,
      "loss": 4.7361,
      "mean_token_accuracy": 0.2533180519938469,
      "num_tokens": 162497280.0,
      "step": 70870
    },
    {
      "entropy": 5.539905977249146,
      "epoch": 6.80835734870317,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001557571821946835,
      "loss": 4.6563,
      "mean_token_accuracy": 0.262447264790535,
      "num_tokens": 162510423.0,
      "step": 70875
    },
    {
      "entropy": 5.540040779113769,
      "epoch": 6.808837656099904,
      "grad_norm": 1.375,
      "learning_rate": 0.0001557281105208828,
      "loss": 4.578,
      "mean_token_accuracy": 0.2709879770874977,
      "num_tokens": 162521912.0,
      "step": 70880
    },
    {
      "entropy": 5.521214437484741,
      "epoch": 6.809317963496638,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015569904161622132,
      "loss": 4.627,
      "mean_token_accuracy": 0.269944529235363,
      "num_tokens": 162533534.0,
      "step": 70885
    },
    {
      "entropy": 5.5063011169433596,
      "epoch": 6.8097982708933715,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015566997548137373,
      "loss": 4.5992,
      "mean_token_accuracy": 0.27002048790454863,
      "num_tokens": 162545529.0,
      "step": 70890
    },
    {
      "entropy": 5.595167398452759,
      "epoch": 6.810278578290106,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015564091211701497,
      "loss": 4.7251,
      "mean_token_accuracy": 0.25459723770618437,
      "num_tokens": 162557679.0,
      "step": 70895
    },
    {
      "entropy": 5.607602500915528,
      "epoch": 6.810758885686839,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015561185152381982,
      "loss": 4.6616,
      "mean_token_accuracy": 0.2551655188202858,
      "num_tokens": 162568602.0,
      "step": 70900
    },
    {
      "entropy": 5.528606319427491,
      "epoch": 6.811239193083574,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015558279370246294,
      "loss": 4.6275,
      "mean_token_accuracy": 0.2676772341132164,
      "num_tokens": 162580757.0,
      "step": 70905
    },
    {
      "entropy": 5.497388172149658,
      "epoch": 6.811719500480307,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015555373865361904,
      "loss": 4.6134,
      "mean_token_accuracy": 0.2699264228343964,
      "num_tokens": 162591749.0,
      "step": 70910
    },
    {
      "entropy": 5.535013246536255,
      "epoch": 6.812199807877041,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00015552468637796272,
      "loss": 4.6382,
      "mean_token_accuracy": 0.26528599113225937,
      "num_tokens": 162601385.0,
      "step": 70915
    },
    {
      "entropy": 5.579871320724488,
      "epoch": 6.812680115273775,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001554956368761683,
      "loss": 4.7122,
      "mean_token_accuracy": 0.25872643291950226,
      "num_tokens": 162613041.0,
      "step": 70920
    },
    {
      "entropy": 5.546609497070312,
      "epoch": 6.813160422670509,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015546659014891036,
      "loss": 4.6513,
      "mean_token_accuracy": 0.2631898432970047,
      "num_tokens": 162625700.0,
      "step": 70925
    },
    {
      "entropy": 5.512362909317017,
      "epoch": 6.8136407300672435,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015543754619686326,
      "loss": 4.6038,
      "mean_token_accuracy": 0.27362017035484315,
      "num_tokens": 162637097.0,
      "step": 70930
    },
    {
      "entropy": 5.688719701766968,
      "epoch": 6.814121037463977,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001554085050207013,
      "loss": 4.7929,
      "mean_token_accuracy": 0.2558764427900314,
      "num_tokens": 162648852.0,
      "step": 70935
    },
    {
      "entropy": 5.50422887802124,
      "epoch": 6.814601344860711,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015537946662109875,
      "loss": 4.5262,
      "mean_token_accuracy": 0.27610218822956084,
      "num_tokens": 162660059.0,
      "step": 70940
    },
    {
      "entropy": 5.490576219558716,
      "epoch": 6.815081652257445,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015535043099872985,
      "loss": 4.5905,
      "mean_token_accuracy": 0.2782669022679329,
      "num_tokens": 162672244.0,
      "step": 70945
    },
    {
      "entropy": 5.587637901306152,
      "epoch": 6.815561959654179,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001553213981542686,
      "loss": 4.6486,
      "mean_token_accuracy": 0.26178584098815916,
      "num_tokens": 162683294.0,
      "step": 70950
    },
    {
      "entropy": 5.519561862945556,
      "epoch": 6.816042267050912,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001552923680883891,
      "loss": 4.6604,
      "mean_token_accuracy": 0.26212034225463865,
      "num_tokens": 162694690.0,
      "step": 70955
    },
    {
      "entropy": 5.49985523223877,
      "epoch": 6.816522574447647,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00015526334080176536,
      "loss": 4.6151,
      "mean_token_accuracy": 0.2602711319923401,
      "num_tokens": 162705877.0,
      "step": 70960
    },
    {
      "entropy": 5.542924785614014,
      "epoch": 6.81700288184438,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015523431629507127,
      "loss": 4.6199,
      "mean_token_accuracy": 0.2600203216075897,
      "num_tokens": 162717575.0,
      "step": 70965
    },
    {
      "entropy": 5.547339677810669,
      "epoch": 6.8174831892411145,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015520529456898076,
      "loss": 4.5993,
      "mean_token_accuracy": 0.2690749645233154,
      "num_tokens": 162728665.0,
      "step": 70970
    },
    {
      "entropy": 5.5663981437683105,
      "epoch": 6.817963496637848,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015517627562416765,
      "loss": 4.6488,
      "mean_token_accuracy": 0.2690432772040367,
      "num_tokens": 162740656.0,
      "step": 70975
    },
    {
      "entropy": 5.541835832595825,
      "epoch": 6.818443804034582,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015514725946130553,
      "loss": 4.6337,
      "mean_token_accuracy": 0.2602178260684013,
      "num_tokens": 162752889.0,
      "step": 70980
    },
    {
      "entropy": 5.453948068618774,
      "epoch": 6.818924111431316,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015511824608106816,
      "loss": 4.4976,
      "mean_token_accuracy": 0.2751708969473839,
      "num_tokens": 162764392.0,
      "step": 70985
    },
    {
      "entropy": 5.522738790512085,
      "epoch": 6.81940441882805,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015508923548412918,
      "loss": 4.6259,
      "mean_token_accuracy": 0.2642743542790413,
      "num_tokens": 162775769.0,
      "step": 70990
    },
    {
      "entropy": 5.497921419143677,
      "epoch": 6.8198847262247835,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015506022767116206,
      "loss": 4.5537,
      "mean_token_accuracy": 0.2656716242432594,
      "num_tokens": 162786486.0,
      "step": 70995
    },
    {
      "entropy": 5.539846038818359,
      "epoch": 6.820365033621518,
      "grad_norm": 1.609375,
      "learning_rate": 0.0001550312226428403,
      "loss": 4.6508,
      "mean_token_accuracy": 0.2701194956898689,
      "num_tokens": 162797032.0,
      "step": 71000
    },
    {
      "entropy": 5.615135335922242,
      "epoch": 6.820845341018252,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001550022203998374,
      "loss": 4.7734,
      "mean_token_accuracy": 0.2548702031373978,
      "num_tokens": 162809436.0,
      "step": 71005
    },
    {
      "entropy": 5.60467562675476,
      "epoch": 6.821325648414986,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015497322094282654,
      "loss": 4.698,
      "mean_token_accuracy": 0.2626178488135338,
      "num_tokens": 162819935.0,
      "step": 71010
    },
    {
      "entropy": 5.5642753601074215,
      "epoch": 6.821805955811719,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001549442242724811,
      "loss": 4.6146,
      "mean_token_accuracy": 0.2646833062171936,
      "num_tokens": 162832156.0,
      "step": 71015
    },
    {
      "entropy": 5.531971979141235,
      "epoch": 6.822286263208453,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015491523038947428,
      "loss": 4.5753,
      "mean_token_accuracy": 0.27104842364788057,
      "num_tokens": 162844405.0,
      "step": 71020
    },
    {
      "entropy": 5.585798120498657,
      "epoch": 6.822766570605188,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001548862392944792,
      "loss": 4.729,
      "mean_token_accuracy": 0.2550159052014351,
      "num_tokens": 162856073.0,
      "step": 71025
    },
    {
      "entropy": 5.513704538345337,
      "epoch": 6.823246878001921,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000154857250988169,
      "loss": 4.5945,
      "mean_token_accuracy": 0.2693341881036758,
      "num_tokens": 162867343.0,
      "step": 71030
    },
    {
      "entropy": 5.548765373229981,
      "epoch": 6.823727185398655,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015482826547121676,
      "loss": 4.6455,
      "mean_token_accuracy": 0.2540882736444473,
      "num_tokens": 162879201.0,
      "step": 71035
    },
    {
      "entropy": 5.50544753074646,
      "epoch": 6.824207492795389,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015479928274429527,
      "loss": 4.6136,
      "mean_token_accuracy": 0.26557886749506,
      "num_tokens": 162890020.0,
      "step": 71040
    },
    {
      "entropy": 5.495974206924439,
      "epoch": 6.824687800192123,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015477030280807753,
      "loss": 4.6278,
      "mean_token_accuracy": 0.2692234873771667,
      "num_tokens": 162902022.0,
      "step": 71045
    },
    {
      "entropy": 5.558561897277832,
      "epoch": 6.825168107588857,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015474132566323638,
      "loss": 4.6629,
      "mean_token_accuracy": 0.2593816116452217,
      "num_tokens": 162913997.0,
      "step": 71050
    },
    {
      "entropy": 5.531975746154785,
      "epoch": 6.825648414985591,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015471235131044449,
      "loss": 4.5976,
      "mean_token_accuracy": 0.26879052072763443,
      "num_tokens": 162925359.0,
      "step": 71055
    },
    {
      "entropy": 5.566509485244751,
      "epoch": 6.826128722382324,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015468337975037467,
      "loss": 4.6663,
      "mean_token_accuracy": 0.26379244923591616,
      "num_tokens": 162937694.0,
      "step": 71060
    },
    {
      "entropy": 5.479130744934082,
      "epoch": 6.826609029779059,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00015465441098369955,
      "loss": 4.456,
      "mean_token_accuracy": 0.27028398364782336,
      "num_tokens": 162947435.0,
      "step": 71065
    },
    {
      "entropy": 5.512561416625976,
      "epoch": 6.827089337175792,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015462544501109157,
      "loss": 4.6862,
      "mean_token_accuracy": 0.2656083583831787,
      "num_tokens": 162959128.0,
      "step": 71070
    },
    {
      "entropy": 5.525627279281617,
      "epoch": 6.8275696445725265,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015459648183322334,
      "loss": 4.6191,
      "mean_token_accuracy": 0.26364307552576066,
      "num_tokens": 162969832.0,
      "step": 71075
    },
    {
      "entropy": 5.478419303894043,
      "epoch": 6.828049951969261,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015456752145076724,
      "loss": 4.5267,
      "mean_token_accuracy": 0.28260705024003985,
      "num_tokens": 162981490.0,
      "step": 71080
    },
    {
      "entropy": 5.599829626083374,
      "epoch": 6.828530259365994,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015453856386439573,
      "loss": 4.692,
      "mean_token_accuracy": 0.2589452683925629,
      "num_tokens": 162993349.0,
      "step": 71085
    },
    {
      "entropy": 5.634712839126587,
      "epoch": 6.829010566762728,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015450960907478106,
      "loss": 4.7206,
      "mean_token_accuracy": 0.2604272037744522,
      "num_tokens": 163004599.0,
      "step": 71090
    },
    {
      "entropy": 5.529178428649902,
      "epoch": 6.829490874159462,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015448065708259555,
      "loss": 4.6516,
      "mean_token_accuracy": 0.26834541410207746,
      "num_tokens": 163016490.0,
      "step": 71095
    },
    {
      "entropy": 5.615412092208862,
      "epoch": 6.829971181556196,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015445170788851127,
      "loss": 4.6773,
      "mean_token_accuracy": 0.2596674695611,
      "num_tokens": 163027198.0,
      "step": 71100
    },
    {
      "entropy": 5.478446912765503,
      "epoch": 6.83045148895293,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015442276149320034,
      "loss": 4.5891,
      "mean_token_accuracy": 0.27741125226020813,
      "num_tokens": 163036940.0,
      "step": 71105
    },
    {
      "entropy": 5.5993406772613525,
      "epoch": 6.830931796349664,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015439381789733483,
      "loss": 4.7288,
      "mean_token_accuracy": 0.2595232889056206,
      "num_tokens": 163049145.0,
      "step": 71110
    },
    {
      "entropy": 5.501508140563965,
      "epoch": 6.8314121037463975,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015436487710158677,
      "loss": 4.6004,
      "mean_token_accuracy": 0.272073370218277,
      "num_tokens": 163059579.0,
      "step": 71115
    },
    {
      "entropy": 5.50522747039795,
      "epoch": 6.831892411143132,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015433593910662807,
      "loss": 4.659,
      "mean_token_accuracy": 0.26195213496685027,
      "num_tokens": 163071453.0,
      "step": 71120
    },
    {
      "entropy": 5.557209014892578,
      "epoch": 6.832372718539865,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015430700391313063,
      "loss": 4.6493,
      "mean_token_accuracy": 0.26861563622951506,
      "num_tokens": 163084068.0,
      "step": 71125
    },
    {
      "entropy": 5.489830207824707,
      "epoch": 6.8328530259366,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001542780715217661,
      "loss": 4.5204,
      "mean_token_accuracy": 0.2797777622938156,
      "num_tokens": 163095294.0,
      "step": 71130
    },
    {
      "entropy": 5.573320627212524,
      "epoch": 6.833333333333333,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015424914193320626,
      "loss": 4.6309,
      "mean_token_accuracy": 0.25866119414567945,
      "num_tokens": 163106805.0,
      "step": 71135
    },
    {
      "entropy": 5.580056238174438,
      "epoch": 6.833813640730067,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015422021514812278,
      "loss": 4.638,
      "mean_token_accuracy": 0.26784050315618513,
      "num_tokens": 163117812.0,
      "step": 71140
    },
    {
      "entropy": 5.519908666610718,
      "epoch": 6.834293948126801,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015419129116718729,
      "loss": 4.6638,
      "mean_token_accuracy": 0.2633627071976662,
      "num_tokens": 163128406.0,
      "step": 71145
    },
    {
      "entropy": 5.5200052738189695,
      "epoch": 6.834774255523535,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001541623699910713,
      "loss": 4.5944,
      "mean_token_accuracy": 0.2655971273779869,
      "num_tokens": 163140660.0,
      "step": 71150
    },
    {
      "entropy": 5.485721635818481,
      "epoch": 6.8352545629202694,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015413345162044624,
      "loss": 4.6424,
      "mean_token_accuracy": 0.2678700968623161,
      "num_tokens": 163153293.0,
      "step": 71155
    },
    {
      "entropy": 5.516875982284546,
      "epoch": 6.835734870317003,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015410453605598352,
      "loss": 4.5903,
      "mean_token_accuracy": 0.2678742751479149,
      "num_tokens": 163163956.0,
      "step": 71160
    },
    {
      "entropy": 5.487786483764649,
      "epoch": 6.836215177713736,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015407562329835457,
      "loss": 4.5668,
      "mean_token_accuracy": 0.26745830923318864,
      "num_tokens": 163175182.0,
      "step": 71165
    },
    {
      "entropy": 5.49892144203186,
      "epoch": 6.836695485110471,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001540467133482305,
      "loss": 4.5781,
      "mean_token_accuracy": 0.2616309329867363,
      "num_tokens": 163186914.0,
      "step": 71170
    },
    {
      "entropy": 5.523708438873291,
      "epoch": 6.837175792507205,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015401780620628259,
      "loss": 4.7016,
      "mean_token_accuracy": 0.2548889622092247,
      "num_tokens": 163197070.0,
      "step": 71175
    },
    {
      "entropy": 5.561715126037598,
      "epoch": 6.837656099903938,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000153988901873182,
      "loss": 4.6993,
      "mean_token_accuracy": 0.254059998691082,
      "num_tokens": 163210223.0,
      "step": 71180
    },
    {
      "entropy": 5.472588777542114,
      "epoch": 6.838136407300673,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001539600003495998,
      "loss": 4.5752,
      "mean_token_accuracy": 0.2680208086967468,
      "num_tokens": 163221534.0,
      "step": 71185
    },
    {
      "entropy": 5.536988735198975,
      "epoch": 6.838616714697406,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015393110163620688,
      "loss": 4.6504,
      "mean_token_accuracy": 0.2624038204550743,
      "num_tokens": 163231995.0,
      "step": 71190
    },
    {
      "entropy": 5.476819181442261,
      "epoch": 6.8390970220941405,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015390220573367435,
      "loss": 4.5941,
      "mean_token_accuracy": 0.26917762160301206,
      "num_tokens": 163243796.0,
      "step": 71195
    },
    {
      "entropy": 5.574246501922607,
      "epoch": 6.839577329490874,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015387331264267302,
      "loss": 4.6939,
      "mean_token_accuracy": 0.2616629794239998,
      "num_tokens": 163254647.0,
      "step": 71200
    },
    {
      "entropy": 5.495556926727295,
      "epoch": 6.840057636887608,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015384442236387376,
      "loss": 4.6337,
      "mean_token_accuracy": 0.2656491339206696,
      "num_tokens": 163266347.0,
      "step": 71205
    },
    {
      "entropy": 5.479841947555542,
      "epoch": 6.840537944284342,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001538155348979472,
      "loss": 4.6311,
      "mean_token_accuracy": 0.2728609174489975,
      "num_tokens": 163278719.0,
      "step": 71210
    },
    {
      "entropy": 5.566596412658692,
      "epoch": 6.841018251681076,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015378665024556404,
      "loss": 4.6684,
      "mean_token_accuracy": 0.2587817445397377,
      "num_tokens": 163290601.0,
      "step": 71215
    },
    {
      "entropy": 5.563589239120484,
      "epoch": 6.8414985590778095,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015375776840739495,
      "loss": 4.6491,
      "mean_token_accuracy": 0.26532773226499556,
      "num_tokens": 163300901.0,
      "step": 71220
    },
    {
      "entropy": 5.487699890136719,
      "epoch": 6.841978866474544,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015372888938411044,
      "loss": 4.5332,
      "mean_token_accuracy": 0.2706977307796478,
      "num_tokens": 163312019.0,
      "step": 71225
    },
    {
      "entropy": 5.490850496292114,
      "epoch": 6.842459173871278,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015370001317638104,
      "loss": 4.666,
      "mean_token_accuracy": 0.25996638685464857,
      "num_tokens": 163324074.0,
      "step": 71230
    },
    {
      "entropy": 5.540451574325561,
      "epoch": 6.8429394812680115,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015367113978487714,
      "loss": 4.6991,
      "mean_token_accuracy": 0.26191223710775374,
      "num_tokens": 163335732.0,
      "step": 71235
    },
    {
      "entropy": 5.5357866287231445,
      "epoch": 6.843419788664745,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001536422692102691,
      "loss": 4.6169,
      "mean_token_accuracy": 0.2689633473753929,
      "num_tokens": 163346705.0,
      "step": 71240
    },
    {
      "entropy": 5.526152610778809,
      "epoch": 6.843900096061479,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001536134014532273,
      "loss": 4.5793,
      "mean_token_accuracy": 0.27567996084690094,
      "num_tokens": 163358072.0,
      "step": 71245
    },
    {
      "entropy": 5.513291931152343,
      "epoch": 6.844380403458214,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015358453651442177,
      "loss": 4.6885,
      "mean_token_accuracy": 0.25209795236587523,
      "num_tokens": 163370984.0,
      "step": 71250
    },
    {
      "entropy": 5.515302991867065,
      "epoch": 6.844860710854947,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001535556743945228,
      "loss": 4.6102,
      "mean_token_accuracy": 0.2678855374455452,
      "num_tokens": 163381764.0,
      "step": 71255
    },
    {
      "entropy": 5.508322286605835,
      "epoch": 6.845341018251681,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015352681509420046,
      "loss": 4.6442,
      "mean_token_accuracy": 0.2631602302193642,
      "num_tokens": 163392339.0,
      "step": 71260
    },
    {
      "entropy": 5.463661861419678,
      "epoch": 6.845821325648415,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015349795861412474,
      "loss": 4.5614,
      "mean_token_accuracy": 0.266728387773037,
      "num_tokens": 163403104.0,
      "step": 71265
    },
    {
      "entropy": 5.6079333305358885,
      "epoch": 6.846301633045149,
      "grad_norm": 1.46875,
      "learning_rate": 0.00015346910495496566,
      "loss": 4.7055,
      "mean_token_accuracy": 0.2575897365808487,
      "num_tokens": 163413837.0,
      "step": 71270
    },
    {
      "entropy": 5.539429950714111,
      "epoch": 6.846781940441883,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001534402541173932,
      "loss": 4.6207,
      "mean_token_accuracy": 0.2623226761817932,
      "num_tokens": 163425023.0,
      "step": 71275
    },
    {
      "entropy": 5.539066457748413,
      "epoch": 6.847262247838617,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015341140610207693,
      "loss": 4.6426,
      "mean_token_accuracy": 0.26558271795511246,
      "num_tokens": 163436700.0,
      "step": 71280
    },
    {
      "entropy": 5.5546674728393555,
      "epoch": 6.84774255523535,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001533825609096868,
      "loss": 4.6284,
      "mean_token_accuracy": 0.27023616284132,
      "num_tokens": 163447646.0,
      "step": 71285
    },
    {
      "entropy": 5.567237377166748,
      "epoch": 6.848222862632085,
      "grad_norm": 1.375,
      "learning_rate": 0.00015335371854089248,
      "loss": 4.6703,
      "mean_token_accuracy": 0.2685726389288902,
      "num_tokens": 163458370.0,
      "step": 71290
    },
    {
      "entropy": 5.529361629486084,
      "epoch": 6.848703170028818,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001533248789963636,
      "loss": 4.6845,
      "mean_token_accuracy": 0.2639315590262413,
      "num_tokens": 163470959.0,
      "step": 71295
    },
    {
      "entropy": 5.521088790893555,
      "epoch": 6.8491834774255524,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015329604227676974,
      "loss": 4.6096,
      "mean_token_accuracy": 0.2682741716504097,
      "num_tokens": 163482385.0,
      "step": 71300
    },
    {
      "entropy": 5.515385389328003,
      "epoch": 6.849663784822286,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015326720838278047,
      "loss": 4.5313,
      "mean_token_accuracy": 0.26876397281885145,
      "num_tokens": 163492621.0,
      "step": 71305
    },
    {
      "entropy": 5.506097650527954,
      "epoch": 6.85014409221902,
      "grad_norm": 1.359375,
      "learning_rate": 0.000153238377315065,
      "loss": 4.5854,
      "mean_token_accuracy": 0.2704681113362312,
      "num_tokens": 163503912.0,
      "step": 71310
    },
    {
      "entropy": 5.562208271026611,
      "epoch": 6.850624399615754,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015320954907429297,
      "loss": 4.6616,
      "mean_token_accuracy": 0.2599916309118271,
      "num_tokens": 163515594.0,
      "step": 71315
    },
    {
      "entropy": 5.53976583480835,
      "epoch": 6.851104707012488,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015318072366113348,
      "loss": 4.5804,
      "mean_token_accuracy": 0.266039215028286,
      "num_tokens": 163527298.0,
      "step": 71320
    },
    {
      "entropy": 5.5286932468414305,
      "epoch": 6.851585014409222,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015315190107625587,
      "loss": 4.562,
      "mean_token_accuracy": 0.26707056909799576,
      "num_tokens": 163539283.0,
      "step": 71325
    },
    {
      "entropy": 5.577171087265015,
      "epoch": 6.852065321805956,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001531230813203293,
      "loss": 4.7116,
      "mean_token_accuracy": 0.2581640616059303,
      "num_tokens": 163550720.0,
      "step": 71330
    },
    {
      "entropy": 5.579672765731812,
      "epoch": 6.85254562920269,
      "grad_norm": 1.34375,
      "learning_rate": 0.000153094264394023,
      "loss": 4.7266,
      "mean_token_accuracy": 0.2521899625658989,
      "num_tokens": 163562858.0,
      "step": 71335
    },
    {
      "entropy": 5.572727012634277,
      "epoch": 6.8530259365994235,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015306545029800578,
      "loss": 4.7288,
      "mean_token_accuracy": 0.2650096550583839,
      "num_tokens": 163574702.0,
      "step": 71340
    },
    {
      "entropy": 5.555712699890137,
      "epoch": 6.853506243996158,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015303663903294676,
      "loss": 4.6548,
      "mean_token_accuracy": 0.2641847595572472,
      "num_tokens": 163586075.0,
      "step": 71345
    },
    {
      "entropy": 5.524639701843261,
      "epoch": 6.853986551392891,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015300783059951482,
      "loss": 4.6438,
      "mean_token_accuracy": 0.2667123407125473,
      "num_tokens": 163598151.0,
      "step": 71350
    },
    {
      "entropy": 5.509214639663696,
      "epoch": 6.854466858789626,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001529790249983788,
      "loss": 4.629,
      "mean_token_accuracy": 0.2668333649635315,
      "num_tokens": 163609778.0,
      "step": 71355
    },
    {
      "entropy": 5.488419532775879,
      "epoch": 6.854947166186359,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001529502222302075,
      "loss": 4.5439,
      "mean_token_accuracy": 0.2701927497982979,
      "num_tokens": 163621799.0,
      "step": 71360
    },
    {
      "entropy": 5.554777145385742,
      "epoch": 6.855427473583093,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001529214222956697,
      "loss": 4.6384,
      "mean_token_accuracy": 0.2664796680212021,
      "num_tokens": 163632701.0,
      "step": 71365
    },
    {
      "entropy": 5.557388973236084,
      "epoch": 6.855907780979827,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015289262519543385,
      "loss": 4.6259,
      "mean_token_accuracy": 0.2595252335071564,
      "num_tokens": 163644606.0,
      "step": 71370
    },
    {
      "entropy": 5.539832592010498,
      "epoch": 6.856388088376561,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001528638309301687,
      "loss": 4.5983,
      "mean_token_accuracy": 0.2710599109530449,
      "num_tokens": 163656462.0,
      "step": 71375
    },
    {
      "entropy": 5.4440265655517575,
      "epoch": 6.8568683957732945,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015283503950054277,
      "loss": 4.511,
      "mean_token_accuracy": 0.27912358939647675,
      "num_tokens": 163667547.0,
      "step": 71380
    },
    {
      "entropy": 5.556212615966797,
      "epoch": 6.857348703170029,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001528062509072244,
      "loss": 4.6626,
      "mean_token_accuracy": 0.2666967123746872,
      "num_tokens": 163678876.0,
      "step": 71385
    },
    {
      "entropy": 5.510605955123902,
      "epoch": 6.857829010566762,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015277746515088208,
      "loss": 4.5992,
      "mean_token_accuracy": 0.26916526407003405,
      "num_tokens": 163689466.0,
      "step": 71390
    },
    {
      "entropy": 5.629361915588379,
      "epoch": 6.858309317963497,
      "grad_norm": 1.375,
      "learning_rate": 0.00015274868223218415,
      "loss": 4.7515,
      "mean_token_accuracy": 0.2554721966385841,
      "num_tokens": 163700964.0,
      "step": 71395
    },
    {
      "entropy": 5.526620864868164,
      "epoch": 6.858789625360231,
      "grad_norm": 1.515625,
      "learning_rate": 0.00015271990215179877,
      "loss": 4.5787,
      "mean_token_accuracy": 0.2684818640351295,
      "num_tokens": 163712268.0,
      "step": 71400
    },
    {
      "entropy": 5.538753175735474,
      "epoch": 6.859269932756964,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001526911249103941,
      "loss": 4.6179,
      "mean_token_accuracy": 0.2648764356970787,
      "num_tokens": 163722910.0,
      "step": 71405
    },
    {
      "entropy": 5.560775184631348,
      "epoch": 6.859750240153699,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015266235050863836,
      "loss": 4.7205,
      "mean_token_accuracy": 0.2550479382276535,
      "num_tokens": 163734912.0,
      "step": 71410
    },
    {
      "entropy": 5.512862300872802,
      "epoch": 6.860230547550432,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001526335789471996,
      "loss": 4.5874,
      "mean_token_accuracy": 0.2705429121851921,
      "num_tokens": 163745733.0,
      "step": 71415
    },
    {
      "entropy": 5.564819478988648,
      "epoch": 6.8607108549471665,
      "grad_norm": 1.625,
      "learning_rate": 0.00015260481022674575,
      "loss": 4.6837,
      "mean_token_accuracy": 0.2604906365275383,
      "num_tokens": 163757347.0,
      "step": 71420
    },
    {
      "entropy": 5.538030672073364,
      "epoch": 6.8611911623439,
      "grad_norm": 1.5,
      "learning_rate": 0.00015257604434794486,
      "loss": 4.6196,
      "mean_token_accuracy": 0.27229451537132265,
      "num_tokens": 163769131.0,
      "step": 71425
    },
    {
      "entropy": 5.556623697280884,
      "epoch": 6.861671469740634,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001525472813114646,
      "loss": 4.6718,
      "mean_token_accuracy": 0.259023879468441,
      "num_tokens": 163780852.0,
      "step": 71430
    },
    {
      "entropy": 5.528094148635864,
      "epoch": 6.862151777137368,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00015251852111797286,
      "loss": 4.5577,
      "mean_token_accuracy": 0.27253199815750123,
      "num_tokens": 163791106.0,
      "step": 71435
    },
    {
      "entropy": 5.558023929595947,
      "epoch": 6.862632084534102,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001524897637681374,
      "loss": 4.6772,
      "mean_token_accuracy": 0.2554916664958,
      "num_tokens": 163801572.0,
      "step": 71440
    },
    {
      "entropy": 5.577737045288086,
      "epoch": 6.863112391930835,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001524610092626258,
      "loss": 4.6898,
      "mean_token_accuracy": 0.2576914831995964,
      "num_tokens": 163812315.0,
      "step": 71445
    },
    {
      "entropy": 5.555098819732666,
      "epoch": 6.86359269932757,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015243225760210572,
      "loss": 4.7448,
      "mean_token_accuracy": 0.25522268265485765,
      "num_tokens": 163823238.0,
      "step": 71450
    },
    {
      "entropy": 5.658917093276978,
      "epoch": 6.864073006724303,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015240350878724474,
      "loss": 4.753,
      "mean_token_accuracy": 0.26265209913253784,
      "num_tokens": 163835404.0,
      "step": 71455
    },
    {
      "entropy": 5.6439464569091795,
      "epoch": 6.8645533141210375,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015237476281871015,
      "loss": 4.6686,
      "mean_token_accuracy": 0.26048228442668914,
      "num_tokens": 163846097.0,
      "step": 71460
    },
    {
      "entropy": 5.572615385055542,
      "epoch": 6.865033621517771,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001523460196971695,
      "loss": 4.6018,
      "mean_token_accuracy": 0.27562940418720244,
      "num_tokens": 163857011.0,
      "step": 71465
    },
    {
      "entropy": 5.5505836486816404,
      "epoch": 6.865513928914505,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015231727942329005,
      "loss": 4.6436,
      "mean_token_accuracy": 0.26932083517313005,
      "num_tokens": 163869132.0,
      "step": 71470
    },
    {
      "entropy": 5.574595022201538,
      "epoch": 6.86599423631124,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015228854199773905,
      "loss": 4.6767,
      "mean_token_accuracy": 0.26482724994421003,
      "num_tokens": 163880532.0,
      "step": 71475
    },
    {
      "entropy": 5.53836612701416,
      "epoch": 6.866474543707973,
      "grad_norm": 1.4375,
      "learning_rate": 0.00015225980742118373,
      "loss": 4.6739,
      "mean_token_accuracy": 0.2575585827231407,
      "num_tokens": 163892319.0,
      "step": 71480
    },
    {
      "entropy": 5.555846166610718,
      "epoch": 6.866954851104707,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015223107569429132,
      "loss": 4.5907,
      "mean_token_accuracy": 0.2690947115421295,
      "num_tokens": 163902158.0,
      "step": 71485
    },
    {
      "entropy": 5.496968603134155,
      "epoch": 6.867435158501441,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001522023468177287,
      "loss": 4.5707,
      "mean_token_accuracy": 0.2740286231040955,
      "num_tokens": 163913852.0,
      "step": 71490
    },
    {
      "entropy": 5.594629716873169,
      "epoch": 6.867915465898175,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015217362079216292,
      "loss": 4.7652,
      "mean_token_accuracy": 0.25237085670232773,
      "num_tokens": 163925431.0,
      "step": 71495
    },
    {
      "entropy": 5.575824165344239,
      "epoch": 6.868395773294909,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015214489761826094,
      "loss": 4.6084,
      "mean_token_accuracy": 0.2723877876996994,
      "num_tokens": 163935600.0,
      "step": 71500
    },
    {
      "entropy": 5.5121523380279545,
      "epoch": 6.868876080691643,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001521161772966897,
      "loss": 4.5967,
      "mean_token_accuracy": 0.2647134825587273,
      "num_tokens": 163946701.0,
      "step": 71505
    },
    {
      "entropy": 5.546230030059815,
      "epoch": 6.869356388088376,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015208745982811587,
      "loss": 4.6364,
      "mean_token_accuracy": 0.2634963482618332,
      "num_tokens": 163957933.0,
      "step": 71510
    },
    {
      "entropy": 5.511762619018555,
      "epoch": 6.869836695485111,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015205874521320636,
      "loss": 4.717,
      "mean_token_accuracy": 0.25930941551923753,
      "num_tokens": 163970933.0,
      "step": 71515
    },
    {
      "entropy": 5.601847314834595,
      "epoch": 6.870317002881844,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015203003345262762,
      "loss": 4.7349,
      "mean_token_accuracy": 0.2559596046805382,
      "num_tokens": 163981073.0,
      "step": 71520
    },
    {
      "entropy": 5.654232692718506,
      "epoch": 6.870797310278578,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001520013245470464,
      "loss": 4.7857,
      "mean_token_accuracy": 0.25272048115730283,
      "num_tokens": 163992940.0,
      "step": 71525
    },
    {
      "entropy": 5.55561580657959,
      "epoch": 6.871277617675312,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001519726184971292,
      "loss": 4.6088,
      "mean_token_accuracy": 0.2657206758856773,
      "num_tokens": 164005956.0,
      "step": 71530
    },
    {
      "entropy": 5.474674367904663,
      "epoch": 6.871757925072046,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015194391530354244,
      "loss": 4.5092,
      "mean_token_accuracy": 0.27607413828372956,
      "num_tokens": 164017232.0,
      "step": 71535
    },
    {
      "entropy": 5.497896337509156,
      "epoch": 6.87223823246878,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015191521496695255,
      "loss": 4.6305,
      "mean_token_accuracy": 0.2649719312787056,
      "num_tokens": 164027341.0,
      "step": 71540
    },
    {
      "entropy": 5.550157070159912,
      "epoch": 6.872718539865514,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000151886517488026,
      "loss": 4.6412,
      "mean_token_accuracy": 0.2654786095023155,
      "num_tokens": 164040071.0,
      "step": 71545
    },
    {
      "entropy": 5.60293231010437,
      "epoch": 6.873198847262248,
      "grad_norm": 1.5,
      "learning_rate": 0.00015185782286742882,
      "loss": 4.7002,
      "mean_token_accuracy": 0.26192061603069305,
      "num_tokens": 164051096.0,
      "step": 71550
    },
    {
      "entropy": 5.478853607177735,
      "epoch": 6.873679154658982,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001518291311058274,
      "loss": 4.5661,
      "mean_token_accuracy": 0.2748462572693825,
      "num_tokens": 164062349.0,
      "step": 71555
    },
    {
      "entropy": 5.605139207839966,
      "epoch": 6.874159462055716,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015180044220388777,
      "loss": 4.7717,
      "mean_token_accuracy": 0.25434947162866595,
      "num_tokens": 164074023.0,
      "step": 71560
    },
    {
      "entropy": 5.5782488822937015,
      "epoch": 6.8746397694524495,
      "grad_norm": 1.375,
      "learning_rate": 0.00015177175616227606,
      "loss": 4.6926,
      "mean_token_accuracy": 0.26363767236471175,
      "num_tokens": 164086792.0,
      "step": 71565
    },
    {
      "entropy": 5.517696523666382,
      "epoch": 6.875120076849184,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00015174307298165823,
      "loss": 4.5888,
      "mean_token_accuracy": 0.2658486500382423,
      "num_tokens": 164097149.0,
      "step": 71570
    },
    {
      "entropy": 5.4697425365448,
      "epoch": 6.875600384245917,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015171439266270037,
      "loss": 4.5328,
      "mean_token_accuracy": 0.2749337494373322,
      "num_tokens": 164109254.0,
      "step": 71575
    },
    {
      "entropy": 5.540575218200684,
      "epoch": 6.876080691642652,
      "grad_norm": 1.375,
      "learning_rate": 0.0001516857152060681,
      "loss": 4.6797,
      "mean_token_accuracy": 0.26386154592037203,
      "num_tokens": 164121648.0,
      "step": 71580
    },
    {
      "entropy": 5.551023483276367,
      "epoch": 6.876560999039385,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001516570406124274,
      "loss": 4.6616,
      "mean_token_accuracy": 0.26303297728300096,
      "num_tokens": 164132659.0,
      "step": 71585
    },
    {
      "entropy": 5.5352977275848385,
      "epoch": 6.877041306436119,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015162836888244397,
      "loss": 4.5926,
      "mean_token_accuracy": 0.2723348468542099,
      "num_tokens": 164144194.0,
      "step": 71590
    },
    {
      "entropy": 5.587139034271241,
      "epoch": 6.877521613832853,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015159970001678348,
      "loss": 4.7248,
      "mean_token_accuracy": 0.2537796348333359,
      "num_tokens": 164156820.0,
      "step": 71595
    },
    {
      "entropy": 5.5662816047668455,
      "epoch": 6.878001921229587,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015157103401611144,
      "loss": 4.656,
      "mean_token_accuracy": 0.267615994811058,
      "num_tokens": 164168358.0,
      "step": 71600
    },
    {
      "entropy": 5.600099229812622,
      "epoch": 6.8784822286263205,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015154237088109367,
      "loss": 4.6926,
      "mean_token_accuracy": 0.25435169488191606,
      "num_tokens": 164180133.0,
      "step": 71605
    },
    {
      "entropy": 5.64367413520813,
      "epoch": 6.878962536023055,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015151371061239533,
      "loss": 4.7767,
      "mean_token_accuracy": 0.25557137131690977,
      "num_tokens": 164191338.0,
      "step": 71610
    },
    {
      "entropy": 5.492652034759521,
      "epoch": 6.879442843419788,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015148505321068196,
      "loss": 4.5677,
      "mean_token_accuracy": 0.28088370263576506,
      "num_tokens": 164202964.0,
      "step": 71615
    },
    {
      "entropy": 5.533126544952393,
      "epoch": 6.879923150816523,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015145639867661888,
      "loss": 4.6053,
      "mean_token_accuracy": 0.26508253514766694,
      "num_tokens": 164214365.0,
      "step": 71620
    },
    {
      "entropy": 5.556800270080567,
      "epoch": 6.880403458213257,
      "grad_norm": 1.53125,
      "learning_rate": 0.00015142774701087136,
      "loss": 4.6114,
      "mean_token_accuracy": 0.26794544905424117,
      "num_tokens": 164225891.0,
      "step": 71625
    },
    {
      "entropy": 5.577627420425415,
      "epoch": 6.88088376560999,
      "grad_norm": 1.5,
      "learning_rate": 0.00015139909821410463,
      "loss": 4.6531,
      "mean_token_accuracy": 0.26422095894813535,
      "num_tokens": 164237065.0,
      "step": 71630
    },
    {
      "entropy": 5.567802095413208,
      "epoch": 6.881364073006724,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015137045228698392,
      "loss": 4.6463,
      "mean_token_accuracy": 0.2667135924100876,
      "num_tokens": 164248820.0,
      "step": 71635
    },
    {
      "entropy": 5.490375709533692,
      "epoch": 6.881844380403458,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001513418092301741,
      "loss": 4.5819,
      "mean_token_accuracy": 0.26922163665294646,
      "num_tokens": 164260935.0,
      "step": 71640
    },
    {
      "entropy": 5.48764214515686,
      "epoch": 6.8823246878001925,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015131316904434027,
      "loss": 4.6242,
      "mean_token_accuracy": 0.2709803760051727,
      "num_tokens": 164273257.0,
      "step": 71645
    },
    {
      "entropy": 5.519076538085938,
      "epoch": 6.882804995196926,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015128453173014744,
      "loss": 4.5995,
      "mean_token_accuracy": 0.2644487202167511,
      "num_tokens": 164285185.0,
      "step": 71650
    },
    {
      "entropy": 5.490129613876343,
      "epoch": 6.88328530259366,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001512558972882604,
      "loss": 4.5508,
      "mean_token_accuracy": 0.2752844899892807,
      "num_tokens": 164297149.0,
      "step": 71655
    },
    {
      "entropy": 5.508622646331787,
      "epoch": 6.883765609990394,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015122726571934392,
      "loss": 4.5862,
      "mean_token_accuracy": 0.2742466926574707,
      "num_tokens": 164309490.0,
      "step": 71660
    },
    {
      "entropy": 5.515450096130371,
      "epoch": 6.884245917387128,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015119863702406293,
      "loss": 4.6165,
      "mean_token_accuracy": 0.2668618530035019,
      "num_tokens": 164320547.0,
      "step": 71665
    },
    {
      "entropy": 5.430152511596679,
      "epoch": 6.884726224783861,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001511700112030819,
      "loss": 4.5327,
      "mean_token_accuracy": 0.2755524605512619,
      "num_tokens": 164333783.0,
      "step": 71670
    },
    {
      "entropy": 5.492672777175903,
      "epoch": 6.885206532180596,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015114138825706553,
      "loss": 4.5917,
      "mean_token_accuracy": 0.2699321836233139,
      "num_tokens": 164344512.0,
      "step": 71675
    },
    {
      "entropy": 5.601951217651367,
      "epoch": 6.885686839577329,
      "grad_norm": 1.375,
      "learning_rate": 0.0001511127681866783,
      "loss": 4.7096,
      "mean_token_accuracy": 0.26038440465927126,
      "num_tokens": 164356543.0,
      "step": 71680
    },
    {
      "entropy": 5.561308670043945,
      "epoch": 6.8861671469740635,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001510841509925847,
      "loss": 4.6424,
      "mean_token_accuracy": 0.26927307695150376,
      "num_tokens": 164368137.0,
      "step": 71685
    },
    {
      "entropy": 5.533357000350952,
      "epoch": 6.886647454370797,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015105553667544918,
      "loss": 4.6587,
      "mean_token_accuracy": 0.25679012984037397,
      "num_tokens": 164380168.0,
      "step": 71690
    },
    {
      "entropy": 5.57384843826294,
      "epoch": 6.887127761767531,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001510269252359362,
      "loss": 4.7201,
      "mean_token_accuracy": 0.2624179035425186,
      "num_tokens": 164392342.0,
      "step": 71695
    },
    {
      "entropy": 5.569413948059082,
      "epoch": 6.887608069164266,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015099831667470976,
      "loss": 4.6653,
      "mean_token_accuracy": 0.261773394048214,
      "num_tokens": 164404403.0,
      "step": 71700
    },
    {
      "entropy": 5.501697444915772,
      "epoch": 6.888088376560999,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015096971099243418,
      "loss": 4.5631,
      "mean_token_accuracy": 0.270651513338089,
      "num_tokens": 164417351.0,
      "step": 71705
    },
    {
      "entropy": 5.500424480438232,
      "epoch": 6.8885686839577325,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015094110818977366,
      "loss": 4.6027,
      "mean_token_accuracy": 0.264645017683506,
      "num_tokens": 164428930.0,
      "step": 71710
    },
    {
      "entropy": 5.624823808670044,
      "epoch": 6.889048991354467,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001509125082673922,
      "loss": 4.7292,
      "mean_token_accuracy": 0.2551347494125366,
      "num_tokens": 164440387.0,
      "step": 71715
    },
    {
      "entropy": 5.620701932907105,
      "epoch": 6.889529298751201,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015088391122595386,
      "loss": 4.7097,
      "mean_token_accuracy": 0.2624679446220398,
      "num_tokens": 164450763.0,
      "step": 71720
    },
    {
      "entropy": 5.533255863189697,
      "epoch": 6.890009606147935,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015085531706612252,
      "loss": 4.6321,
      "mean_token_accuracy": 0.26415856182575226,
      "num_tokens": 164462365.0,
      "step": 71725
    },
    {
      "entropy": 5.50114254951477,
      "epoch": 6.890489913544669,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001508267257885621,
      "loss": 4.5714,
      "mean_token_accuracy": 0.26668711453676225,
      "num_tokens": 164473914.0,
      "step": 71730
    },
    {
      "entropy": 5.510266208648682,
      "epoch": 6.890970220941402,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015079813739393648,
      "loss": 4.6298,
      "mean_token_accuracy": 0.26527395397424697,
      "num_tokens": 164485372.0,
      "step": 71735
    },
    {
      "entropy": 5.490761613845825,
      "epoch": 6.891450528338137,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001507695518829092,
      "loss": 4.5984,
      "mean_token_accuracy": 0.27484927475452425,
      "num_tokens": 164496472.0,
      "step": 71740
    },
    {
      "entropy": 5.635164499282837,
      "epoch": 6.89193083573487,
      "grad_norm": 1.5,
      "learning_rate": 0.00015074096925614402,
      "loss": 4.734,
      "mean_token_accuracy": 0.2560160025954247,
      "num_tokens": 164507442.0,
      "step": 71745
    },
    {
      "entropy": 5.511887550354004,
      "epoch": 6.892411143131604,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001507123895143046,
      "loss": 4.6006,
      "mean_token_accuracy": 0.270758281648159,
      "num_tokens": 164518881.0,
      "step": 71750
    },
    {
      "entropy": 5.465102672576904,
      "epoch": 6.892891450528338,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001506838126580544,
      "loss": 4.4954,
      "mean_token_accuracy": 0.2776475787162781,
      "num_tokens": 164530357.0,
      "step": 71755
    },
    {
      "entropy": 5.569883394241333,
      "epoch": 6.893371757925072,
      "grad_norm": 1.375,
      "learning_rate": 0.00015065523868805698,
      "loss": 4.6545,
      "mean_token_accuracy": 0.25957771986722944,
      "num_tokens": 164542136.0,
      "step": 71760
    },
    {
      "entropy": 5.540154886245728,
      "epoch": 6.893852065321806,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015062666760497567,
      "loss": 4.6313,
      "mean_token_accuracy": 0.26425803154706956,
      "num_tokens": 164554794.0,
      "step": 71765
    },
    {
      "entropy": 5.542927360534668,
      "epoch": 6.89433237271854,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015059809940947381,
      "loss": 4.6575,
      "mean_token_accuracy": 0.26615930944681165,
      "num_tokens": 164565631.0,
      "step": 71770
    },
    {
      "entropy": 5.552306747436523,
      "epoch": 6.894812680115274,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015056953410221474,
      "loss": 4.6276,
      "mean_token_accuracy": 0.26319530159235,
      "num_tokens": 164576506.0,
      "step": 71775
    },
    {
      "entropy": 5.628633069992065,
      "epoch": 6.895292987512008,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001505409716838616,
      "loss": 4.7247,
      "mean_token_accuracy": 0.2542562961578369,
      "num_tokens": 164589132.0,
      "step": 71780
    },
    {
      "entropy": 5.504726123809815,
      "epoch": 6.895773294908741,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015051241215507748,
      "loss": 4.6399,
      "mean_token_accuracy": 0.2628180250525475,
      "num_tokens": 164600708.0,
      "step": 71785
    },
    {
      "entropy": 5.50949912071228,
      "epoch": 6.8962536023054755,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001504838555165255,
      "loss": 4.6123,
      "mean_token_accuracy": 0.26586395800113677,
      "num_tokens": 164611523.0,
      "step": 71790
    },
    {
      "entropy": 5.56024022102356,
      "epoch": 6.89673390970221,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015045530176886865,
      "loss": 4.6433,
      "mean_token_accuracy": 0.26386098712682726,
      "num_tokens": 164623533.0,
      "step": 71795
    },
    {
      "entropy": 5.564570522308349,
      "epoch": 6.897214217098943,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001504267509127699,
      "loss": 4.6498,
      "mean_token_accuracy": 0.26911988854408264,
      "num_tokens": 164634779.0,
      "step": 71800
    },
    {
      "entropy": 5.534595775604248,
      "epoch": 6.8976945244956775,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015039820294889206,
      "loss": 4.6836,
      "mean_token_accuracy": 0.25239784121513364,
      "num_tokens": 164646355.0,
      "step": 71805
    },
    {
      "entropy": 5.5411779403686525,
      "epoch": 6.898174831892411,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015036965787789796,
      "loss": 4.6533,
      "mean_token_accuracy": 0.2645678550004959,
      "num_tokens": 164657449.0,
      "step": 71810
    },
    {
      "entropy": 5.571731472015381,
      "epoch": 6.898655139289145,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001503411157004504,
      "loss": 4.6734,
      "mean_token_accuracy": 0.26227420270442964,
      "num_tokens": 164669896.0,
      "step": 71815
    },
    {
      "entropy": 5.538686800003052,
      "epoch": 6.899135446685879,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015031257641721194,
      "loss": 4.6651,
      "mean_token_accuracy": 0.2626635268330574,
      "num_tokens": 164682133.0,
      "step": 71820
    },
    {
      "entropy": 5.519080305099488,
      "epoch": 6.899615754082613,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015028404002884516,
      "loss": 4.6569,
      "mean_token_accuracy": 0.2630732163786888,
      "num_tokens": 164692774.0,
      "step": 71825
    },
    {
      "entropy": 5.500117635726928,
      "epoch": 6.9000960614793465,
      "grad_norm": 1.375,
      "learning_rate": 0.00015025550653601267,
      "loss": 4.6442,
      "mean_token_accuracy": 0.2598843932151794,
      "num_tokens": 164704044.0,
      "step": 71830
    },
    {
      "entropy": 5.527959108352661,
      "epoch": 6.900576368876081,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015022697593937686,
      "loss": 4.6331,
      "mean_token_accuracy": 0.2692868307232857,
      "num_tokens": 164715437.0,
      "step": 71835
    },
    {
      "entropy": 5.540354251861572,
      "epoch": 6.901056676272814,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001501984482396002,
      "loss": 4.6333,
      "mean_token_accuracy": 0.264498645067215,
      "num_tokens": 164726429.0,
      "step": 71840
    },
    {
      "entropy": 5.519151401519776,
      "epoch": 6.901536983669549,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015016992343734508,
      "loss": 4.6296,
      "mean_token_accuracy": 0.26632282733917234,
      "num_tokens": 164738002.0,
      "step": 71845
    },
    {
      "entropy": 5.469437122344971,
      "epoch": 6.902017291066282,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001501414015332736,
      "loss": 4.5576,
      "mean_token_accuracy": 0.272123184800148,
      "num_tokens": 164750307.0,
      "step": 71850
    },
    {
      "entropy": 5.551602077484131,
      "epoch": 6.902497598463016,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015011288252804797,
      "loss": 4.5663,
      "mean_token_accuracy": 0.2703551083803177,
      "num_tokens": 164761502.0,
      "step": 71855
    },
    {
      "entropy": 5.606405925750733,
      "epoch": 6.90297790585975,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015008436642233042,
      "loss": 4.713,
      "mean_token_accuracy": 0.2549747437238693,
      "num_tokens": 164773829.0,
      "step": 71860
    },
    {
      "entropy": 5.517340135574341,
      "epoch": 6.903458213256484,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015005585321678294,
      "loss": 4.5806,
      "mean_token_accuracy": 0.2731436133384705,
      "num_tokens": 164785511.0,
      "step": 71865
    },
    {
      "entropy": 5.57279634475708,
      "epoch": 6.9039385206532184,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001500273429120675,
      "loss": 4.656,
      "mean_token_accuracy": 0.2685400515794754,
      "num_tokens": 164796214.0,
      "step": 71870
    },
    {
      "entropy": 5.445075845718383,
      "epoch": 6.904418828049952,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001499988355088462,
      "loss": 4.5532,
      "mean_token_accuracy": 0.27598741054534914,
      "num_tokens": 164808338.0,
      "step": 71875
    },
    {
      "entropy": 5.52422866821289,
      "epoch": 6.904899135446686,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014997033100778068,
      "loss": 4.6747,
      "mean_token_accuracy": 0.25979725569486617,
      "num_tokens": 164818813.0,
      "step": 71880
    },
    {
      "entropy": 5.466674375534057,
      "epoch": 6.90537944284342,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001499418294095328,
      "loss": 4.55,
      "mean_token_accuracy": 0.2724357098340988,
      "num_tokens": 164831204.0,
      "step": 71885
    },
    {
      "entropy": 5.563847398757934,
      "epoch": 6.905859750240154,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014991333071476428,
      "loss": 4.6428,
      "mean_token_accuracy": 0.2648319140076637,
      "num_tokens": 164841799.0,
      "step": 71890
    },
    {
      "entropy": 5.503853034973145,
      "epoch": 6.906340057636887,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001498848349241368,
      "loss": 4.5908,
      "mean_token_accuracy": 0.27405530512332915,
      "num_tokens": 164852885.0,
      "step": 71895
    },
    {
      "entropy": 5.573763847351074,
      "epoch": 6.906820365033622,
      "grad_norm": 1.375,
      "learning_rate": 0.00014985634203831195,
      "loss": 4.6466,
      "mean_token_accuracy": 0.2657577246427536,
      "num_tokens": 164865153.0,
      "step": 71900
    },
    {
      "entropy": 5.486956691741943,
      "epoch": 6.907300672430355,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001498278520579513,
      "loss": 4.5531,
      "mean_token_accuracy": 0.2714957594871521,
      "num_tokens": 164875573.0,
      "step": 71905
    },
    {
      "entropy": 5.587463331222534,
      "epoch": 6.9077809798270895,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014979936498371615,
      "loss": 4.6868,
      "mean_token_accuracy": 0.259995673596859,
      "num_tokens": 164886771.0,
      "step": 71910
    },
    {
      "entropy": 5.505510711669922,
      "epoch": 6.908261287223823,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014977088081626798,
      "loss": 4.5923,
      "mean_token_accuracy": 0.27054040879011154,
      "num_tokens": 164898358.0,
      "step": 71915
    },
    {
      "entropy": 5.564299535751343,
      "epoch": 6.908741594620557,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001497423995562681,
      "loss": 4.602,
      "mean_token_accuracy": 0.2666640028357506,
      "num_tokens": 164910077.0,
      "step": 71920
    },
    {
      "entropy": 5.611558485031128,
      "epoch": 6.909221902017291,
      "grad_norm": 1.375,
      "learning_rate": 0.00014971392120437783,
      "loss": 4.6193,
      "mean_token_accuracy": 0.2655856966972351,
      "num_tokens": 164920040.0,
      "step": 71925
    },
    {
      "entropy": 5.556548929214477,
      "epoch": 6.909702209414025,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001496854457612582,
      "loss": 4.672,
      "mean_token_accuracy": 0.26497085988521574,
      "num_tokens": 164931766.0,
      "step": 71930
    },
    {
      "entropy": 5.493348455429077,
      "epoch": 6.9101825168107585,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014965697322757055,
      "loss": 4.547,
      "mean_token_accuracy": 0.2736771315336227,
      "num_tokens": 164943793.0,
      "step": 71935
    },
    {
      "entropy": 5.54929895401001,
      "epoch": 6.910662824207493,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014962850360397567,
      "loss": 4.6956,
      "mean_token_accuracy": 0.25666994601488113,
      "num_tokens": 164955767.0,
      "step": 71940
    },
    {
      "entropy": 5.581461191177368,
      "epoch": 6.911143131604227,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014960003689113466,
      "loss": 4.6466,
      "mean_token_accuracy": 0.2573492392897606,
      "num_tokens": 164966381.0,
      "step": 71945
    },
    {
      "entropy": 5.529603147506714,
      "epoch": 6.9116234390009605,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014957157308970848,
      "loss": 4.5674,
      "mean_token_accuracy": 0.2716151922941208,
      "num_tokens": 164977556.0,
      "step": 71950
    },
    {
      "entropy": 5.5133997917175295,
      "epoch": 6.912103746397695,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014954311220035787,
      "loss": 4.596,
      "mean_token_accuracy": 0.26704874634742737,
      "num_tokens": 164989369.0,
      "step": 71955
    },
    {
      "entropy": 5.615029764175415,
      "epoch": 6.912584053794428,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001495146542237437,
      "loss": 4.7509,
      "mean_token_accuracy": 0.2528818964958191,
      "num_tokens": 165000904.0,
      "step": 71960
    },
    {
      "entropy": 5.613379669189453,
      "epoch": 6.913064361191163,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014948619916052675,
      "loss": 4.7172,
      "mean_token_accuracy": 0.255932354927063,
      "num_tokens": 165013892.0,
      "step": 71965
    },
    {
      "entropy": 5.623828935623169,
      "epoch": 6.913544668587896,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014945774701136743,
      "loss": 4.6898,
      "mean_token_accuracy": 0.265678745508194,
      "num_tokens": 165023884.0,
      "step": 71970
    },
    {
      "entropy": 5.55617504119873,
      "epoch": 6.91402497598463,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014942929777692643,
      "loss": 4.6723,
      "mean_token_accuracy": 0.26582648754119875,
      "num_tokens": 165035572.0,
      "step": 71975
    },
    {
      "entropy": 5.533938884735107,
      "epoch": 6.914505283381364,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014940085145786428,
      "loss": 4.6631,
      "mean_token_accuracy": 0.2566011905670166,
      "num_tokens": 165047882.0,
      "step": 71980
    },
    {
      "entropy": 5.600078010559082,
      "epoch": 6.914985590778098,
      "grad_norm": 1.375,
      "learning_rate": 0.00014937240805484143,
      "loss": 4.6615,
      "mean_token_accuracy": 0.2627962276339531,
      "num_tokens": 165059323.0,
      "step": 71985
    },
    {
      "entropy": 5.503631830215454,
      "epoch": 6.915465898174832,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001493439675685182,
      "loss": 4.5838,
      "mean_token_accuracy": 0.2743931129574776,
      "num_tokens": 165070851.0,
      "step": 71990
    },
    {
      "entropy": 5.552213287353515,
      "epoch": 6.915946205571566,
      "grad_norm": 1.546875,
      "learning_rate": 0.000149315529999555,
      "loss": 4.6068,
      "mean_token_accuracy": 0.2655432179570198,
      "num_tokens": 165083924.0,
      "step": 71995
    },
    {
      "entropy": 5.588787508010864,
      "epoch": 6.916426512968299,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014928709534861194,
      "loss": 4.7063,
      "mean_token_accuracy": 0.26138064116239546,
      "num_tokens": 165094945.0,
      "step": 72000
    },
    {
      "epoch": 6.916426512968299,
      "eval_entropy": 5.256048509052822,
      "eval_loss": 4.746522903442383,
      "eval_mean_token_accuracy": 0.2673139179371541,
      "eval_num_tokens": 165094945.0,
      "eval_runtime": 21.134,
      "eval_samples_per_second": 1552.714,
      "eval_steps_per_second": 194.095,
      "step": 72000
    },
    {
      "entropy": 5.512285614013672,
      "epoch": 6.916906820365034,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014925866361634922,
      "loss": 4.6052,
      "mean_token_accuracy": 0.2625088572502136,
      "num_tokens": 165105361.0,
      "step": 72005
    },
    {
      "entropy": 5.570732736587525,
      "epoch": 6.917387127761767,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00014923023480342692,
      "loss": 4.5963,
      "mean_token_accuracy": 0.2673541694879532,
      "num_tokens": 165115137.0,
      "step": 72010
    },
    {
      "entropy": 5.518689966201782,
      "epoch": 6.917867435158501,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014920180891050516,
      "loss": 4.5813,
      "mean_token_accuracy": 0.2724625036120415,
      "num_tokens": 165125944.0,
      "step": 72015
    },
    {
      "entropy": 5.524832820892334,
      "epoch": 6.918347742555236,
      "grad_norm": 1.375,
      "learning_rate": 0.00014917338593824382,
      "loss": 4.5766,
      "mean_token_accuracy": 0.26612224280834196,
      "num_tokens": 165136688.0,
      "step": 72020
    },
    {
      "entropy": 5.566816329956055,
      "epoch": 6.918828049951969,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014914496588730293,
      "loss": 4.6493,
      "mean_token_accuracy": 0.2608286142349243,
      "num_tokens": 165148224.0,
      "step": 72025
    },
    {
      "entropy": 5.5781067371368405,
      "epoch": 6.9193083573487035,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014911654875834213,
      "loss": 4.66,
      "mean_token_accuracy": 0.26714586466550827,
      "num_tokens": 165160131.0,
      "step": 72030
    },
    {
      "entropy": 5.533166694641113,
      "epoch": 6.919788664745437,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014908813455202126,
      "loss": 4.6587,
      "mean_token_accuracy": 0.2690616726875305,
      "num_tokens": 165172086.0,
      "step": 72035
    },
    {
      "entropy": 5.496481323242188,
      "epoch": 6.920268972142171,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014905972326900005,
      "loss": 4.6289,
      "mean_token_accuracy": 0.2738476559519768,
      "num_tokens": 165183302.0,
      "step": 72040
    },
    {
      "entropy": 5.49518461227417,
      "epoch": 6.920749279538905,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001490313149099381,
      "loss": 4.5169,
      "mean_token_accuracy": 0.2691544950008392,
      "num_tokens": 165194955.0,
      "step": 72045
    },
    {
      "entropy": 5.548827314376831,
      "epoch": 6.921229586935639,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014900290947549494,
      "loss": 4.6784,
      "mean_token_accuracy": 0.25969909876585007,
      "num_tokens": 165206275.0,
      "step": 72050
    },
    {
      "entropy": 5.544446182250977,
      "epoch": 6.9217098943323725,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001489745069663302,
      "loss": 4.6085,
      "mean_token_accuracy": 0.2638499796390533,
      "num_tokens": 165217337.0,
      "step": 72055
    },
    {
      "entropy": 5.578777551651001,
      "epoch": 6.922190201729107,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014894610738310307,
      "loss": 4.6635,
      "mean_token_accuracy": 0.26592067182064055,
      "num_tokens": 165228300.0,
      "step": 72060
    },
    {
      "entropy": 5.61888861656189,
      "epoch": 6.92267050912584,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014891771072647302,
      "loss": 4.6943,
      "mean_token_accuracy": 0.26236984729766843,
      "num_tokens": 165239157.0,
      "step": 72065
    },
    {
      "entropy": 5.546820163726807,
      "epoch": 6.923150816522575,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001488893169970994,
      "loss": 4.6075,
      "mean_token_accuracy": 0.2636877715587616,
      "num_tokens": 165249937.0,
      "step": 72070
    },
    {
      "entropy": 5.585377168655396,
      "epoch": 6.923631123919308,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014886092619564135,
      "loss": 4.664,
      "mean_token_accuracy": 0.26111972630023955,
      "num_tokens": 165261979.0,
      "step": 72075
    },
    {
      "entropy": 5.5659050941467285,
      "epoch": 6.924111431316042,
      "grad_norm": 1.46875,
      "learning_rate": 0.000148832538322758,
      "loss": 4.6437,
      "mean_token_accuracy": 0.2684280812740326,
      "num_tokens": 165273028.0,
      "step": 72080
    },
    {
      "entropy": 5.517032957077026,
      "epoch": 6.924591738712776,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001488041533791086,
      "loss": 4.6378,
      "mean_token_accuracy": 0.26573132872581484,
      "num_tokens": 165285297.0,
      "step": 72085
    },
    {
      "entropy": 5.487856483459472,
      "epoch": 6.92507204610951,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00014877577136535188,
      "loss": 4.5748,
      "mean_token_accuracy": 0.27714437395334246,
      "num_tokens": 165296672.0,
      "step": 72090
    },
    {
      "entropy": 5.556535959243774,
      "epoch": 6.925552353506244,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000148747392282147,
      "loss": 4.6055,
      "mean_token_accuracy": 0.2651883035898209,
      "num_tokens": 165306914.0,
      "step": 72095
    },
    {
      "entropy": 5.586099243164062,
      "epoch": 6.926032660902978,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014871901613015275,
      "loss": 4.7409,
      "mean_token_accuracy": 0.2636206239461899,
      "num_tokens": 165319467.0,
      "step": 72100
    },
    {
      "entropy": 5.447433614730835,
      "epoch": 6.926512968299712,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000148690642910028,
      "loss": 4.5258,
      "mean_token_accuracy": 0.2783461809158325,
      "num_tokens": 165330703.0,
      "step": 72105
    },
    {
      "entropy": 5.543246936798096,
      "epoch": 6.926993275696446,
      "grad_norm": 1.375,
      "learning_rate": 0.00014866227262243142,
      "loss": 4.6453,
      "mean_token_accuracy": 0.2658283054828644,
      "num_tokens": 165342926.0,
      "step": 72110
    },
    {
      "entropy": 5.6239522933959964,
      "epoch": 6.92747358309318,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014863390526802185,
      "loss": 4.7065,
      "mean_token_accuracy": 0.2552871122956276,
      "num_tokens": 165354725.0,
      "step": 72115
    },
    {
      "entropy": 5.556834173202515,
      "epoch": 6.927953890489913,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014860554084745764,
      "loss": 4.5918,
      "mean_token_accuracy": 0.2679504707455635,
      "num_tokens": 165366310.0,
      "step": 72120
    },
    {
      "entropy": 5.460710048675537,
      "epoch": 6.928434197886648,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014857717936139748,
      "loss": 4.5758,
      "mean_token_accuracy": 0.2715784177184105,
      "num_tokens": 165377920.0,
      "step": 72125
    },
    {
      "entropy": 5.501307773590088,
      "epoch": 6.928914505283381,
      "grad_norm": 1.375,
      "learning_rate": 0.0001485488208104998,
      "loss": 4.6363,
      "mean_token_accuracy": 0.26183520555496215,
      "num_tokens": 165390020.0,
      "step": 72130
    },
    {
      "entropy": 5.556384086608887,
      "epoch": 6.9293948126801155,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014852046519542306,
      "loss": 4.6738,
      "mean_token_accuracy": 0.26468349397182467,
      "num_tokens": 165400814.0,
      "step": 72135
    },
    {
      "entropy": 5.55340895652771,
      "epoch": 6.929875120076849,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014849211251682554,
      "loss": 4.6768,
      "mean_token_accuracy": 0.25793414413928983,
      "num_tokens": 165414116.0,
      "step": 72140
    },
    {
      "entropy": 5.506587171554566,
      "epoch": 6.930355427473583,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014846376277536554,
      "loss": 4.5748,
      "mean_token_accuracy": 0.26601044088602066,
      "num_tokens": 165426937.0,
      "step": 72145
    },
    {
      "entropy": 5.6198780059814455,
      "epoch": 6.930835734870317,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014843541597170124,
      "loss": 4.7126,
      "mean_token_accuracy": 0.26260355710983274,
      "num_tokens": 165437829.0,
      "step": 72150
    },
    {
      "entropy": 5.615439462661743,
      "epoch": 6.931316042267051,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014840707210649068,
      "loss": 4.6136,
      "mean_token_accuracy": 0.26484693586826324,
      "num_tokens": 165448309.0,
      "step": 72155
    },
    {
      "entropy": 5.631065511703492,
      "epoch": 6.931796349663784,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014837873118039204,
      "loss": 4.747,
      "mean_token_accuracy": 0.25771512240171435,
      "num_tokens": 165460041.0,
      "step": 72160
    },
    {
      "entropy": 5.498062562942505,
      "epoch": 6.932276657060519,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014835039319406327,
      "loss": 4.6076,
      "mean_token_accuracy": 0.2682263195514679,
      "num_tokens": 165471421.0,
      "step": 72165
    },
    {
      "entropy": 5.562574863433838,
      "epoch": 6.932756964457253,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001483220581481623,
      "loss": 4.7049,
      "mean_token_accuracy": 0.2553224593400955,
      "num_tokens": 165482679.0,
      "step": 72170
    },
    {
      "entropy": 5.5856232166290285,
      "epoch": 6.9332372718539865,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014829372604334708,
      "loss": 4.6206,
      "mean_token_accuracy": 0.2616842433810234,
      "num_tokens": 165493612.0,
      "step": 72175
    },
    {
      "entropy": 5.576513624191284,
      "epoch": 6.93371757925072,
      "grad_norm": 1.375,
      "learning_rate": 0.0001482653968802752,
      "loss": 4.6512,
      "mean_token_accuracy": 0.2670958340167999,
      "num_tokens": 165504938.0,
      "step": 72180
    },
    {
      "entropy": 5.583228588104248,
      "epoch": 6.934197886647454,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014823707065960448,
      "loss": 4.6543,
      "mean_token_accuracy": 0.26530994921922685,
      "num_tokens": 165515968.0,
      "step": 72185
    },
    {
      "entropy": 5.498240184783936,
      "epoch": 6.934678194044189,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014820874738199257,
      "loss": 4.5822,
      "mean_token_accuracy": 0.2699716240167618,
      "num_tokens": 165525903.0,
      "step": 72190
    },
    {
      "entropy": 5.557437562942505,
      "epoch": 6.935158501440922,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014818042704809706,
      "loss": 4.6828,
      "mean_token_accuracy": 0.26119688600301744,
      "num_tokens": 165537863.0,
      "step": 72195
    },
    {
      "entropy": 5.5262336254119875,
      "epoch": 6.935638808837656,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014815210965857546,
      "loss": 4.627,
      "mean_token_accuracy": 0.2654326051473618,
      "num_tokens": 165547461.0,
      "step": 72200
    },
    {
      "entropy": 5.485876989364624,
      "epoch": 6.93611911623439,
      "grad_norm": 1.375,
      "learning_rate": 0.0001481237952140853,
      "loss": 4.5956,
      "mean_token_accuracy": 0.2645938813686371,
      "num_tokens": 165558986.0,
      "step": 72205
    },
    {
      "entropy": 5.610422039031983,
      "epoch": 6.936599423631124,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014809548371528375,
      "loss": 4.6842,
      "mean_token_accuracy": 0.26061290204524995,
      "num_tokens": 165570617.0,
      "step": 72210
    },
    {
      "entropy": 5.587245082855224,
      "epoch": 6.937079731027858,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001480671751628283,
      "loss": 4.6646,
      "mean_token_accuracy": 0.2739460229873657,
      "num_tokens": 165582418.0,
      "step": 72215
    },
    {
      "entropy": 5.450488471984864,
      "epoch": 6.937560038424592,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001480388695573761,
      "loss": 4.4968,
      "mean_token_accuracy": 0.28126625120639803,
      "num_tokens": 165593741.0,
      "step": 72220
    },
    {
      "entropy": 5.558525419235229,
      "epoch": 6.938040345821325,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014801056689958434,
      "loss": 4.6775,
      "mean_token_accuracy": 0.26696076691150666,
      "num_tokens": 165604755.0,
      "step": 72225
    },
    {
      "entropy": 5.542083215713501,
      "epoch": 6.93852065321806,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014798226719011016,
      "loss": 4.6589,
      "mean_token_accuracy": 0.2598339393734932,
      "num_tokens": 165615242.0,
      "step": 72230
    },
    {
      "entropy": 5.553343009948731,
      "epoch": 6.939000960614793,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014795397042961062,
      "loss": 4.6725,
      "mean_token_accuracy": 0.26044662594795226,
      "num_tokens": 165626056.0,
      "step": 72235
    },
    {
      "entropy": 5.524865818023682,
      "epoch": 6.939481268011527,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014792567661874257,
      "loss": 4.614,
      "mean_token_accuracy": 0.26873716711997986,
      "num_tokens": 165638136.0,
      "step": 72240
    },
    {
      "entropy": 5.538531875610351,
      "epoch": 6.939961575408262,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014789738575816296,
      "loss": 4.6319,
      "mean_token_accuracy": 0.27155986428260803,
      "num_tokens": 165650876.0,
      "step": 72245
    },
    {
      "entropy": 5.559043216705322,
      "epoch": 6.940441882804995,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014786909784852864,
      "loss": 4.6141,
      "mean_token_accuracy": 0.26539619714021684,
      "num_tokens": 165662150.0,
      "step": 72250
    },
    {
      "entropy": 5.507083606719971,
      "epoch": 6.940922190201729,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001478408128904964,
      "loss": 4.5547,
      "mean_token_accuracy": 0.2732487842440605,
      "num_tokens": 165673164.0,
      "step": 72255
    },
    {
      "entropy": 5.457926893234253,
      "epoch": 6.941402497598463,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014781253088472292,
      "loss": 4.5703,
      "mean_token_accuracy": 0.2665087267756462,
      "num_tokens": 165684132.0,
      "step": 72260
    },
    {
      "entropy": 5.519528532028199,
      "epoch": 6.941882804995197,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00014778425183186472,
      "loss": 4.6037,
      "mean_token_accuracy": 0.2694368839263916,
      "num_tokens": 165694611.0,
      "step": 72265
    },
    {
      "entropy": 5.517513704299927,
      "epoch": 6.942363112391931,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001477559757325786,
      "loss": 4.6248,
      "mean_token_accuracy": 0.25750691890716554,
      "num_tokens": 165707097.0,
      "step": 72270
    },
    {
      "entropy": 5.4626318454742435,
      "epoch": 6.942843419788665,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014772770258752077,
      "loss": 4.5129,
      "mean_token_accuracy": 0.273998886346817,
      "num_tokens": 165716852.0,
      "step": 72275
    },
    {
      "entropy": 5.56989016532898,
      "epoch": 6.9433237271853985,
      "grad_norm": 1.5,
      "learning_rate": 0.00014769943239734774,
      "loss": 4.6019,
      "mean_token_accuracy": 0.26739634573459625,
      "num_tokens": 165727479.0,
      "step": 72280
    },
    {
      "entropy": 5.573316287994385,
      "epoch": 6.943804034582133,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001476711651627159,
      "loss": 4.6751,
      "mean_token_accuracy": 0.25946711003780365,
      "num_tokens": 165738448.0,
      "step": 72285
    },
    {
      "entropy": 5.5505589008331295,
      "epoch": 6.944284341978866,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014764290088428156,
      "loss": 4.6305,
      "mean_token_accuracy": 0.26818657964468,
      "num_tokens": 165748253.0,
      "step": 72290
    },
    {
      "entropy": 5.49556336402893,
      "epoch": 6.944764649375601,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014761463956270083,
      "loss": 4.5228,
      "mean_token_accuracy": 0.2761464357376099,
      "num_tokens": 165759777.0,
      "step": 72295
    },
    {
      "entropy": 5.461954259872437,
      "epoch": 6.945244956772334,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014758638119862993,
      "loss": 4.5606,
      "mean_token_accuracy": 0.26996910125017165,
      "num_tokens": 165770969.0,
      "step": 72300
    },
    {
      "entropy": 5.45676703453064,
      "epoch": 6.945725264169068,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014755812579272492,
      "loss": 4.5569,
      "mean_token_accuracy": 0.2725673630833626,
      "num_tokens": 165781939.0,
      "step": 72305
    },
    {
      "entropy": 5.47939395904541,
      "epoch": 6.946205571565802,
      "grad_norm": 1.375,
      "learning_rate": 0.0001475298733456419,
      "loss": 4.5667,
      "mean_token_accuracy": 0.2710769295692444,
      "num_tokens": 165791992.0,
      "step": 72310
    },
    {
      "entropy": 5.526808452606201,
      "epoch": 6.946685878962536,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014750162385803662,
      "loss": 4.6516,
      "mean_token_accuracy": 0.2747504755854607,
      "num_tokens": 165804212.0,
      "step": 72315
    },
    {
      "entropy": 5.464246320724487,
      "epoch": 6.94716618635927,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014747337733056503,
      "loss": 4.5225,
      "mean_token_accuracy": 0.27546036392450335,
      "num_tokens": 165814858.0,
      "step": 72320
    },
    {
      "entropy": 5.4850257396697994,
      "epoch": 6.947646493756004,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014744513376388295,
      "loss": 4.6442,
      "mean_token_accuracy": 0.2667358383536339,
      "num_tokens": 165827616.0,
      "step": 72325
    },
    {
      "entropy": 5.513969755172729,
      "epoch": 6.948126801152737,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014741689315864615,
      "loss": 4.6205,
      "mean_token_accuracy": 0.2639049932360649,
      "num_tokens": 165839736.0,
      "step": 72330
    },
    {
      "entropy": 5.527528381347656,
      "epoch": 6.948607108549472,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001473886555155102,
      "loss": 4.5768,
      "mean_token_accuracy": 0.26266513019800186,
      "num_tokens": 165850161.0,
      "step": 72335
    },
    {
      "entropy": 5.515808820724487,
      "epoch": 6.949087415946206,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014736042083513074,
      "loss": 4.5745,
      "mean_token_accuracy": 0.27035142183303834,
      "num_tokens": 165861929.0,
      "step": 72340
    },
    {
      "entropy": 5.59441909790039,
      "epoch": 6.949567723342939,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001473321891181633,
      "loss": 4.7317,
      "mean_token_accuracy": 0.24917576014995574,
      "num_tokens": 165872790.0,
      "step": 72345
    },
    {
      "entropy": 5.512671756744385,
      "epoch": 6.950048030739674,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014730396036526338,
      "loss": 4.6016,
      "mean_token_accuracy": 0.2657285988330841,
      "num_tokens": 165884398.0,
      "step": 72350
    },
    {
      "entropy": 5.553928375244141,
      "epoch": 6.950528338136407,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014727573457708632,
      "loss": 4.6888,
      "mean_token_accuracy": 0.26045253723859785,
      "num_tokens": 165895495.0,
      "step": 72355
    },
    {
      "entropy": 5.560635375976562,
      "epoch": 6.9510086455331415,
      "grad_norm": 1.59375,
      "learning_rate": 0.00014724751175428742,
      "loss": 4.65,
      "mean_token_accuracy": 0.26884703040122987,
      "num_tokens": 165906677.0,
      "step": 72360
    },
    {
      "entropy": 5.570590209960938,
      "epoch": 6.951488952929875,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001472192918975219,
      "loss": 4.6442,
      "mean_token_accuracy": 0.26642917543649675,
      "num_tokens": 165917192.0,
      "step": 72365
    },
    {
      "entropy": 5.53935489654541,
      "epoch": 6.951969260326609,
      "grad_norm": 1.421875,
      "learning_rate": 0.000147191075007445,
      "loss": 4.6427,
      "mean_token_accuracy": 0.2686604708433151,
      "num_tokens": 165928954.0,
      "step": 72370
    },
    {
      "entropy": 5.557026147842407,
      "epoch": 6.952449567723343,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014716286108471186,
      "loss": 4.7251,
      "mean_token_accuracy": 0.2562288448214531,
      "num_tokens": 165940784.0,
      "step": 72375
    },
    {
      "entropy": 5.525177145004273,
      "epoch": 6.952929875120077,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014713465012997746,
      "loss": 4.6404,
      "mean_token_accuracy": 0.2732794627547264,
      "num_tokens": 165952020.0,
      "step": 72380
    },
    {
      "entropy": 5.544010305404663,
      "epoch": 6.95341018251681,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014710644214389692,
      "loss": 4.6056,
      "mean_token_accuracy": 0.2685826003551483,
      "num_tokens": 165962701.0,
      "step": 72385
    },
    {
      "entropy": 5.5034651279449465,
      "epoch": 6.953890489913545,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014707823712712493,
      "loss": 4.6271,
      "mean_token_accuracy": 0.27074563801288604,
      "num_tokens": 165976339.0,
      "step": 72390
    },
    {
      "entropy": 5.541717147827148,
      "epoch": 6.954370797310279,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001470500350803164,
      "loss": 4.6436,
      "mean_token_accuracy": 0.2651332333683968,
      "num_tokens": 165988314.0,
      "step": 72395
    },
    {
      "entropy": 5.5830991744995115,
      "epoch": 6.9548511047070125,
      "grad_norm": 1.59375,
      "learning_rate": 0.00014702183600412612,
      "loss": 4.6625,
      "mean_token_accuracy": 0.26587338745594025,
      "num_tokens": 165999046.0,
      "step": 72400
    },
    {
      "entropy": 5.561027336120605,
      "epoch": 6.955331412103746,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001469936398992088,
      "loss": 4.6325,
      "mean_token_accuracy": 0.2696981757879257,
      "num_tokens": 166010391.0,
      "step": 72405
    },
    {
      "entropy": 5.54158763885498,
      "epoch": 6.95581171950048,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014696544676621909,
      "loss": 4.6216,
      "mean_token_accuracy": 0.26432740688323975,
      "num_tokens": 166020954.0,
      "step": 72410
    },
    {
      "entropy": 5.59707555770874,
      "epoch": 6.956292026897215,
      "grad_norm": 1.5,
      "learning_rate": 0.00014693725660581158,
      "loss": 4.7113,
      "mean_token_accuracy": 0.2545066997408867,
      "num_tokens": 166031991.0,
      "step": 72415
    },
    {
      "entropy": 5.530090618133545,
      "epoch": 6.956772334293948,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014690906941864062,
      "loss": 4.605,
      "mean_token_accuracy": 0.2735243320465088,
      "num_tokens": 166043495.0,
      "step": 72420
    },
    {
      "entropy": 5.509286260604858,
      "epoch": 6.957252641690682,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014688088520536076,
      "loss": 4.617,
      "mean_token_accuracy": 0.2671402096748352,
      "num_tokens": 166054599.0,
      "step": 72425
    },
    {
      "entropy": 5.502990579605102,
      "epoch": 6.957732949087416,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014685270396662623,
      "loss": 4.5671,
      "mean_token_accuracy": 0.27646092772483827,
      "num_tokens": 166066444.0,
      "step": 72430
    },
    {
      "entropy": 5.541307592391968,
      "epoch": 6.95821325648415,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014682452570309143,
      "loss": 4.6837,
      "mean_token_accuracy": 0.26322786659002306,
      "num_tokens": 166077275.0,
      "step": 72435
    },
    {
      "entropy": 5.508466386795044,
      "epoch": 6.958693563880884,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014679635041541056,
      "loss": 4.5665,
      "mean_token_accuracy": 0.27330227196216583,
      "num_tokens": 166088836.0,
      "step": 72440
    },
    {
      "entropy": 5.51628155708313,
      "epoch": 6.959173871277618,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014676817810423782,
      "loss": 4.6238,
      "mean_token_accuracy": 0.26704501509666445,
      "num_tokens": 166101205.0,
      "step": 72445
    },
    {
      "entropy": 5.487334966659546,
      "epoch": 6.959654178674351,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014674000877022708,
      "loss": 4.5441,
      "mean_token_accuracy": 0.2755450367927551,
      "num_tokens": 166112256.0,
      "step": 72450
    },
    {
      "entropy": 5.464765739440918,
      "epoch": 6.960134486071086,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014671184241403255,
      "loss": 4.6274,
      "mean_token_accuracy": 0.26531669944524766,
      "num_tokens": 166123448.0,
      "step": 72455
    },
    {
      "entropy": 5.4950034618377686,
      "epoch": 6.960614793467819,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014668367903630808,
      "loss": 4.6436,
      "mean_token_accuracy": 0.25978897213935853,
      "num_tokens": 166135588.0,
      "step": 72460
    },
    {
      "entropy": 5.603880882263184,
      "epoch": 6.961095100864553,
      "grad_norm": 1.375,
      "learning_rate": 0.0001466555186377076,
      "loss": 4.7281,
      "mean_token_accuracy": 0.2571778029203415,
      "num_tokens": 166147161.0,
      "step": 72465
    },
    {
      "entropy": 5.56515326499939,
      "epoch": 6.961575408261287,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014662736121888478,
      "loss": 4.5707,
      "mean_token_accuracy": 0.2700156822800636,
      "num_tokens": 166159668.0,
      "step": 72470
    },
    {
      "entropy": 5.6346738815307615,
      "epoch": 6.962055715658021,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001465992067804936,
      "loss": 4.69,
      "mean_token_accuracy": 0.2588014855980873,
      "num_tokens": 166171483.0,
      "step": 72475
    },
    {
      "entropy": 5.523130750656128,
      "epoch": 6.962536023054755,
      "grad_norm": 1.5,
      "learning_rate": 0.00014657105532318745,
      "loss": 4.531,
      "mean_token_accuracy": 0.2717684954404831,
      "num_tokens": 166182753.0,
      "step": 72480
    },
    {
      "entropy": 5.5556563377380375,
      "epoch": 6.963016330451489,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014654290684762005,
      "loss": 4.6291,
      "mean_token_accuracy": 0.2734114572405815,
      "num_tokens": 166195152.0,
      "step": 72485
    },
    {
      "entropy": 5.480248308181762,
      "epoch": 6.963496637848223,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014651476135444494,
      "loss": 4.6033,
      "mean_token_accuracy": 0.2698461726307869,
      "num_tokens": 166207375.0,
      "step": 72490
    },
    {
      "entropy": 5.439777946472168,
      "epoch": 6.963976945244957,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014648661884431553,
      "loss": 4.5566,
      "mean_token_accuracy": 0.27437774538993837,
      "num_tokens": 166219565.0,
      "step": 72495
    },
    {
      "entropy": 5.511311054229736,
      "epoch": 6.964457252641691,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001464584793178852,
      "loss": 4.5935,
      "mean_token_accuracy": 0.26969464868307114,
      "num_tokens": 166231364.0,
      "step": 72500
    },
    {
      "entropy": 5.545021009445191,
      "epoch": 6.9649375600384245,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001464303427758073,
      "loss": 4.6084,
      "mean_token_accuracy": 0.2641223564743996,
      "num_tokens": 166242875.0,
      "step": 72505
    },
    {
      "entropy": 5.6492509841918945,
      "epoch": 6.965417867435159,
      "grad_norm": 1.375,
      "learning_rate": 0.00014640220921873518,
      "loss": 4.7563,
      "mean_token_accuracy": 0.2527687311172485,
      "num_tokens": 166255439.0,
      "step": 72510
    },
    {
      "entropy": 5.53351058959961,
      "epoch": 6.965898174831892,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014637407864732184,
      "loss": 4.6118,
      "mean_token_accuracy": 0.26168515235185624,
      "num_tokens": 166267035.0,
      "step": 72515
    },
    {
      "entropy": 5.6185239315032955,
      "epoch": 6.9663784822286265,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014634595106222036,
      "loss": 4.7654,
      "mean_token_accuracy": 0.25928658843040464,
      "num_tokens": 166278367.0,
      "step": 72520
    },
    {
      "entropy": 5.510808038711548,
      "epoch": 6.96685878962536,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014631782646408397,
      "loss": 4.5247,
      "mean_token_accuracy": 0.27554147094488146,
      "num_tokens": 166290255.0,
      "step": 72525
    },
    {
      "entropy": 5.572258615493775,
      "epoch": 6.967339097022094,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014628970485356548,
      "loss": 4.6725,
      "mean_token_accuracy": 0.2628546193242073,
      "num_tokens": 166301266.0,
      "step": 72530
    },
    {
      "entropy": 5.531632709503174,
      "epoch": 6.967819404418828,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014626158623131787,
      "loss": 4.5595,
      "mean_token_accuracy": 0.2702999204397202,
      "num_tokens": 166312069.0,
      "step": 72535
    },
    {
      "entropy": 5.493241453170777,
      "epoch": 6.968299711815562,
      "grad_norm": 1.34375,
      "learning_rate": 0.000146233470597994,
      "loss": 4.5313,
      "mean_token_accuracy": 0.26833837032318114,
      "num_tokens": 166323986.0,
      "step": 72540
    },
    {
      "entropy": 5.451448965072632,
      "epoch": 6.9687800192122955,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014620535795424662,
      "loss": 4.6012,
      "mean_token_accuracy": 0.2695758804678917,
      "num_tokens": 166335998.0,
      "step": 72545
    },
    {
      "entropy": 5.469877910614014,
      "epoch": 6.96926032660903,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014617724830072825,
      "loss": 4.5475,
      "mean_token_accuracy": 0.2689938828349113,
      "num_tokens": 166347406.0,
      "step": 72550
    },
    {
      "entropy": 5.624242067337036,
      "epoch": 6.969740634005763,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014614914163809173,
      "loss": 4.7257,
      "mean_token_accuracy": 0.261394040286541,
      "num_tokens": 166357586.0,
      "step": 72555
    },
    {
      "entropy": 5.520079803466797,
      "epoch": 6.970220941402498,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001461210379669895,
      "loss": 4.5572,
      "mean_token_accuracy": 0.27478277683258057,
      "num_tokens": 166368419.0,
      "step": 72560
    },
    {
      "entropy": 5.596768474578857,
      "epoch": 6.970701248799232,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001460929372880741,
      "loss": 4.727,
      "mean_token_accuracy": 0.25372845232486724,
      "num_tokens": 166379251.0,
      "step": 72565
    },
    {
      "entropy": 5.587683629989624,
      "epoch": 6.971181556195965,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014606483960199796,
      "loss": 4.6892,
      "mean_token_accuracy": 0.26015848517417905,
      "num_tokens": 166389228.0,
      "step": 72570
    },
    {
      "entropy": 5.535424709320068,
      "epoch": 6.9716618635927,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001460367449094133,
      "loss": 4.6288,
      "mean_token_accuracy": 0.26650130450725557,
      "num_tokens": 166402096.0,
      "step": 72575
    },
    {
      "entropy": 5.533633422851563,
      "epoch": 6.972142170989433,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014600865321097255,
      "loss": 4.6112,
      "mean_token_accuracy": 0.2706857085227966,
      "num_tokens": 166414199.0,
      "step": 72580
    },
    {
      "entropy": 5.504445362091064,
      "epoch": 6.972622478386167,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014598056450732783,
      "loss": 4.5618,
      "mean_token_accuracy": 0.27108547687530515,
      "num_tokens": 166425260.0,
      "step": 72585
    },
    {
      "entropy": 5.5466114521026615,
      "epoch": 6.973102785782901,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00014595247879913126,
      "loss": 4.6097,
      "mean_token_accuracy": 0.27186382561922073,
      "num_tokens": 166436274.0,
      "step": 72590
    },
    {
      "entropy": 5.554885196685791,
      "epoch": 6.973583093179635,
      "grad_norm": 1.453125,
      "learning_rate": 0.000145924396087035,
      "loss": 4.658,
      "mean_token_accuracy": 0.2578905388712883,
      "num_tokens": 166448276.0,
      "step": 72595
    },
    {
      "entropy": 5.501657247543335,
      "epoch": 6.974063400576369,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014589631637169098,
      "loss": 4.634,
      "mean_token_accuracy": 0.26489107459783556,
      "num_tokens": 166460343.0,
      "step": 72600
    },
    {
      "entropy": 5.536774349212647,
      "epoch": 6.974543707973103,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014586823965375112,
      "loss": 4.5638,
      "mean_token_accuracy": 0.26989350765943526,
      "num_tokens": 166471904.0,
      "step": 72605
    },
    {
      "entropy": 5.538299608230591,
      "epoch": 6.975024015369836,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014584016593386723,
      "loss": 4.6224,
      "mean_token_accuracy": 0.2674254313111305,
      "num_tokens": 166483639.0,
      "step": 72610
    },
    {
      "entropy": 5.522238922119141,
      "epoch": 6.975504322766571,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014581209521269124,
      "loss": 4.616,
      "mean_token_accuracy": 0.26493047177791595,
      "num_tokens": 166493527.0,
      "step": 72615
    },
    {
      "entropy": 5.501427173614502,
      "epoch": 6.975984630163304,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014578402749087474,
      "loss": 4.6359,
      "mean_token_accuracy": 0.26740531623363495,
      "num_tokens": 166504604.0,
      "step": 72620
    },
    {
      "entropy": 5.5110921382904055,
      "epoch": 6.9764649375600385,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014575596276906943,
      "loss": 4.6412,
      "mean_token_accuracy": 0.2623103767633438,
      "num_tokens": 166515654.0,
      "step": 72625
    },
    {
      "entropy": 5.532003498077392,
      "epoch": 6.976945244956772,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014572790104792698,
      "loss": 4.6833,
      "mean_token_accuracy": 0.2628752112388611,
      "num_tokens": 166527507.0,
      "step": 72630
    },
    {
      "entropy": 5.637643194198608,
      "epoch": 6.977425552353506,
      "grad_norm": 1.5,
      "learning_rate": 0.00014569984232809874,
      "loss": 4.7799,
      "mean_token_accuracy": 0.2509211480617523,
      "num_tokens": 166539207.0,
      "step": 72635
    },
    {
      "entropy": 5.519262838363647,
      "epoch": 6.977905859750241,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001456717866102362,
      "loss": 4.6493,
      "mean_token_accuracy": 0.26506989598274233,
      "num_tokens": 166550580.0,
      "step": 72640
    },
    {
      "entropy": 5.539969539642334,
      "epoch": 6.978386167146974,
      "grad_norm": 1.625,
      "learning_rate": 0.00014564373389499075,
      "loss": 4.6027,
      "mean_token_accuracy": 0.2692824497818947,
      "num_tokens": 166562670.0,
      "step": 72645
    },
    {
      "entropy": 5.584159851074219,
      "epoch": 6.978866474543708,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001456156841830137,
      "loss": 4.6903,
      "mean_token_accuracy": 0.26000415831804274,
      "num_tokens": 166573597.0,
      "step": 72650
    },
    {
      "entropy": 5.549161911010742,
      "epoch": 6.979346781940442,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014558763747495623,
      "loss": 4.5909,
      "mean_token_accuracy": 0.2681578129529953,
      "num_tokens": 166584673.0,
      "step": 72655
    },
    {
      "entropy": 5.587115621566772,
      "epoch": 6.979827089337176,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014555959377146968,
      "loss": 4.6465,
      "mean_token_accuracy": 0.2634041115641594,
      "num_tokens": 166595107.0,
      "step": 72660
    },
    {
      "entropy": 5.503247928619385,
      "epoch": 6.9803073967339095,
      "grad_norm": 1.25,
      "learning_rate": 0.00014553155307320487,
      "loss": 4.5684,
      "mean_token_accuracy": 0.27326408922672274,
      "num_tokens": 166607004.0,
      "step": 72665
    },
    {
      "entropy": 5.535565662384033,
      "epoch": 6.980787704130644,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000145503515380813,
      "loss": 4.6651,
      "mean_token_accuracy": 0.25569523125886917,
      "num_tokens": 166618906.0,
      "step": 72670
    },
    {
      "entropy": 5.524456214904785,
      "epoch": 6.981268011527377,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014547548069494495,
      "loss": 4.6527,
      "mean_token_accuracy": 0.26641190350055693,
      "num_tokens": 166631633.0,
      "step": 72675
    },
    {
      "entropy": 5.514929580688476,
      "epoch": 6.981748318924112,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001454474490162517,
      "loss": 4.5905,
      "mean_token_accuracy": 0.27155856490135194,
      "num_tokens": 166643247.0,
      "step": 72680
    },
    {
      "entropy": 5.603456211090088,
      "epoch": 6.982228626320845,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014541942034538392,
      "loss": 4.7187,
      "mean_token_accuracy": 0.25327134132385254,
      "num_tokens": 166654148.0,
      "step": 72685
    },
    {
      "entropy": 5.61089859008789,
      "epoch": 6.982708933717579,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001453913946829925,
      "loss": 4.7626,
      "mean_token_accuracy": 0.2587238773703575,
      "num_tokens": 166665351.0,
      "step": 72690
    },
    {
      "entropy": 5.528999328613281,
      "epoch": 6.983189241114313,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014536337202972798,
      "loss": 4.6883,
      "mean_token_accuracy": 0.26789138466119766,
      "num_tokens": 166676562.0,
      "step": 72695
    },
    {
      "entropy": 5.554622459411621,
      "epoch": 6.983669548511047,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014533535238624103,
      "loss": 4.6238,
      "mean_token_accuracy": 0.2718269035220146,
      "num_tokens": 166687308.0,
      "step": 72700
    },
    {
      "entropy": 5.510891580581665,
      "epoch": 6.984149855907781,
      "grad_norm": 1.25,
      "learning_rate": 0.00014530733575318213,
      "loss": 4.613,
      "mean_token_accuracy": 0.2665402889251709,
      "num_tokens": 166699725.0,
      "step": 72705
    },
    {
      "entropy": 5.535503673553467,
      "epoch": 6.984630163304515,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014527932213120182,
      "loss": 4.6451,
      "mean_token_accuracy": 0.2605778530240059,
      "num_tokens": 166711850.0,
      "step": 72710
    },
    {
      "entropy": 5.469921731948853,
      "epoch": 6.985110470701249,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014525131152095044,
      "loss": 4.59,
      "mean_token_accuracy": 0.2707383245229721,
      "num_tokens": 166724118.0,
      "step": 72715
    },
    {
      "entropy": 5.547807836532593,
      "epoch": 6.985590778097983,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014522330392307842,
      "loss": 4.6343,
      "mean_token_accuracy": 0.2642398938536644,
      "num_tokens": 166735242.0,
      "step": 72720
    },
    {
      "entropy": 5.521790075302124,
      "epoch": 6.986071085494717,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014519529933823587,
      "loss": 4.5771,
      "mean_token_accuracy": 0.27585970163345336,
      "num_tokens": 166747372.0,
      "step": 72725
    },
    {
      "entropy": 5.514885187149048,
      "epoch": 6.98655139289145,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000145167297767073,
      "loss": 4.5601,
      "mean_token_accuracy": 0.2676199391484261,
      "num_tokens": 166758815.0,
      "step": 72730
    },
    {
      "entropy": 5.57382493019104,
      "epoch": 6.987031700288185,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014513929921023997,
      "loss": 4.6465,
      "mean_token_accuracy": 0.2667263001203537,
      "num_tokens": 166769052.0,
      "step": 72735
    },
    {
      "entropy": 5.558717727661133,
      "epoch": 6.987512007684918,
      "grad_norm": 1.375,
      "learning_rate": 0.00014511130366838678,
      "loss": 4.6544,
      "mean_token_accuracy": 0.2622929200530052,
      "num_tokens": 166779686.0,
      "step": 72740
    },
    {
      "entropy": 5.564115905761719,
      "epoch": 6.9879923150816525,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014508331114216345,
      "loss": 4.7309,
      "mean_token_accuracy": 0.2614737808704376,
      "num_tokens": 166791139.0,
      "step": 72745
    },
    {
      "entropy": 5.59360899925232,
      "epoch": 6.988472622478386,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014505532163221993,
      "loss": 4.6935,
      "mean_token_accuracy": 0.2573151275515556,
      "num_tokens": 166802897.0,
      "step": 72750
    },
    {
      "entropy": 5.559649801254272,
      "epoch": 6.98895292987512,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001450273351392059,
      "loss": 4.6706,
      "mean_token_accuracy": 0.2613823816180229,
      "num_tokens": 166813759.0,
      "step": 72755
    },
    {
      "entropy": 5.514453268051147,
      "epoch": 6.989433237271854,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014499935166377125,
      "loss": 4.5798,
      "mean_token_accuracy": 0.28276345431804656,
      "num_tokens": 166825635.0,
      "step": 72760
    },
    {
      "entropy": 5.561342525482178,
      "epoch": 6.989913544668588,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001449713712065656,
      "loss": 4.6749,
      "mean_token_accuracy": 0.2620834246277809,
      "num_tokens": 166837004.0,
      "step": 72765
    },
    {
      "entropy": 5.524118804931641,
      "epoch": 6.9903938520653215,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014494339376823858,
      "loss": 4.6024,
      "mean_token_accuracy": 0.2667835682630539,
      "num_tokens": 166848219.0,
      "step": 72770
    },
    {
      "entropy": 5.5420115947723385,
      "epoch": 6.990874159462056,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001449154193494398,
      "loss": 4.6232,
      "mean_token_accuracy": 0.2697190895676613,
      "num_tokens": 166859477.0,
      "step": 72775
    },
    {
      "entropy": 5.576909065246582,
      "epoch": 6.991354466858789,
      "grad_norm": 1.375,
      "learning_rate": 0.00014488744795081878,
      "loss": 4.6635,
      "mean_token_accuracy": 0.26409322768449783,
      "num_tokens": 166870810.0,
      "step": 72780
    },
    {
      "entropy": 5.445016670227051,
      "epoch": 6.991834774255524,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014485947957302482,
      "loss": 4.5672,
      "mean_token_accuracy": 0.2703013464808464,
      "num_tokens": 166883335.0,
      "step": 72785
    },
    {
      "entropy": 5.563136768341065,
      "epoch": 6.992315081652258,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014483151421670727,
      "loss": 4.703,
      "mean_token_accuracy": 0.26656845062971113,
      "num_tokens": 166894409.0,
      "step": 72790
    },
    {
      "entropy": 5.528215503692627,
      "epoch": 6.992795389048991,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014480355188251542,
      "loss": 4.637,
      "mean_token_accuracy": 0.2629809632897377,
      "num_tokens": 166906932.0,
      "step": 72795
    },
    {
      "entropy": 5.506413507461548,
      "epoch": 6.993275696445725,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014477559257109854,
      "loss": 4.5743,
      "mean_token_accuracy": 0.271475587785244,
      "num_tokens": 166918651.0,
      "step": 72800
    },
    {
      "entropy": 5.535728693008423,
      "epoch": 6.993756003842459,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001447476362831057,
      "loss": 4.6615,
      "mean_token_accuracy": 0.2645168364048004,
      "num_tokens": 166930141.0,
      "step": 72805
    },
    {
      "entropy": 5.5657671928405765,
      "epoch": 6.994236311239193,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014471968301918604,
      "loss": 4.6681,
      "mean_token_accuracy": 0.2695573538541794,
      "num_tokens": 166940800.0,
      "step": 72810
    },
    {
      "entropy": 5.607204914093018,
      "epoch": 6.994716618635927,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014469173277998842,
      "loss": 4.6813,
      "mean_token_accuracy": 0.259944187104702,
      "num_tokens": 166952588.0,
      "step": 72815
    },
    {
      "entropy": 5.519180202484131,
      "epoch": 6.995196926032661,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001446637855661618,
      "loss": 4.5869,
      "mean_token_accuracy": 0.2717531830072403,
      "num_tokens": 166964173.0,
      "step": 72820
    },
    {
      "entropy": 5.502686929702759,
      "epoch": 6.995677233429395,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001446358413783551,
      "loss": 4.6194,
      "mean_token_accuracy": 0.25809974074363706,
      "num_tokens": 166976044.0,
      "step": 72825
    },
    {
      "entropy": 5.527952861785889,
      "epoch": 6.996157540826129,
      "grad_norm": 1.359375,
      "learning_rate": 0.000144607900217217,
      "loss": 4.6324,
      "mean_token_accuracy": 0.27078024446964266,
      "num_tokens": 166988678.0,
      "step": 72830
    },
    {
      "entropy": 5.558516931533814,
      "epoch": 6.996637848222862,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014457996208339636,
      "loss": 4.6081,
      "mean_token_accuracy": 0.266258043050766,
      "num_tokens": 166999004.0,
      "step": 72835
    },
    {
      "entropy": 5.509972476959229,
      "epoch": 6.997118155619597,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014455202697754167,
      "loss": 4.672,
      "mean_token_accuracy": 0.25983576476573944,
      "num_tokens": 167010519.0,
      "step": 72840
    },
    {
      "entropy": 5.590714740753174,
      "epoch": 6.99759846301633,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00014452409490030166,
      "loss": 4.652,
      "mean_token_accuracy": 0.26935725212097167,
      "num_tokens": 167021921.0,
      "step": 72845
    },
    {
      "entropy": 5.4609545230865475,
      "epoch": 6.9980787704130645,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014449616585232464,
      "loss": 4.5281,
      "mean_token_accuracy": 0.2697518154978752,
      "num_tokens": 167033042.0,
      "step": 72850
    },
    {
      "entropy": 5.539589691162109,
      "epoch": 6.998559077809798,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014446823983425914,
      "loss": 4.6504,
      "mean_token_accuracy": 0.2613439902663231,
      "num_tokens": 167045126.0,
      "step": 72855
    },
    {
      "entropy": 5.633802700042724,
      "epoch": 6.999039385206532,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014444031684675353,
      "loss": 4.7334,
      "mean_token_accuracy": 0.25616529732942583,
      "num_tokens": 167056885.0,
      "step": 72860
    },
    {
      "entropy": 5.494436311721802,
      "epoch": 6.999519692603267,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014441239689045606,
      "loss": 4.5775,
      "mean_token_accuracy": 0.2672291651368141,
      "num_tokens": 167069409.0,
      "step": 72865
    },
    {
      "entropy": 5.506060647964477,
      "epoch": 7.0,
      "grad_norm": 1.65625,
      "learning_rate": 0.00014438447996601494,
      "loss": 4.5952,
      "mean_token_accuracy": 0.2802064403891563,
      "num_tokens": 167079752.0,
      "step": 72870
    },
    {
      "entropy": 5.594519996643067,
      "epoch": 7.000480307396734,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001443565660740784,
      "loss": 4.6513,
      "mean_token_accuracy": 0.26384918093681337,
      "num_tokens": 167089861.0,
      "step": 72875
    },
    {
      "entropy": 5.599524021148682,
      "epoch": 7.000960614793468,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00014432865521529447,
      "loss": 4.6991,
      "mean_token_accuracy": 0.259360970556736,
      "num_tokens": 167101604.0,
      "step": 72880
    },
    {
      "entropy": 5.588574647903442,
      "epoch": 7.001440922190202,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001443007473903112,
      "loss": 4.6754,
      "mean_token_accuracy": 0.26012282967567446,
      "num_tokens": 167113195.0,
      "step": 72885
    },
    {
      "entropy": 5.577700662612915,
      "epoch": 7.0019212295869355,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014427284259977641,
      "loss": 4.5989,
      "mean_token_accuracy": 0.2686824068427086,
      "num_tokens": 167124692.0,
      "step": 72890
    },
    {
      "entropy": 5.507524538040161,
      "epoch": 7.00240153698367,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014424494084433804,
      "loss": 4.5409,
      "mean_token_accuracy": 0.2723156288266182,
      "num_tokens": 167137458.0,
      "step": 72895
    },
    {
      "entropy": 5.548329401016235,
      "epoch": 7.002881844380403,
      "grad_norm": 1.375,
      "learning_rate": 0.00014421704212464387,
      "loss": 4.646,
      "mean_token_accuracy": 0.26528634428977965,
      "num_tokens": 167149949.0,
      "step": 72900
    },
    {
      "entropy": 5.5807051181793215,
      "epoch": 7.003362151777138,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014418914644134164,
      "loss": 4.6726,
      "mean_token_accuracy": 0.2605944812297821,
      "num_tokens": 167162024.0,
      "step": 72905
    },
    {
      "entropy": 5.5194212913513185,
      "epoch": 7.003842459173871,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00014416125379507896,
      "loss": 4.5796,
      "mean_token_accuracy": 0.27106691598892213,
      "num_tokens": 167173101.0,
      "step": 72910
    },
    {
      "entropy": 5.520035171508789,
      "epoch": 7.004322766570605,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014413336418650353,
      "loss": 4.6137,
      "mean_token_accuracy": 0.269783191382885,
      "num_tokens": 167185717.0,
      "step": 72915
    },
    {
      "entropy": 5.55257215499878,
      "epoch": 7.004803073967339,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014410547761626276,
      "loss": 4.5932,
      "mean_token_accuracy": 0.26727106124162675,
      "num_tokens": 167197163.0,
      "step": 72920
    },
    {
      "entropy": 5.526814031600952,
      "epoch": 7.005283381364073,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001440775940850042,
      "loss": 4.6392,
      "mean_token_accuracy": 0.26461361944675443,
      "num_tokens": 167208577.0,
      "step": 72925
    },
    {
      "entropy": 5.539435720443725,
      "epoch": 7.005763688760807,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014404971359337505,
      "loss": 4.6263,
      "mean_token_accuracy": 0.26847977191209793,
      "num_tokens": 167219731.0,
      "step": 72930
    },
    {
      "entropy": 5.525163984298706,
      "epoch": 7.006243996157541,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001440218361420227,
      "loss": 4.6478,
      "mean_token_accuracy": 0.2603356152772903,
      "num_tokens": 167230759.0,
      "step": 72935
    },
    {
      "entropy": 5.557791233062744,
      "epoch": 7.006724303554274,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014399396173159443,
      "loss": 4.6115,
      "mean_token_accuracy": 0.26873240172863005,
      "num_tokens": 167241576.0,
      "step": 72940
    },
    {
      "entropy": 5.599612760543823,
      "epoch": 7.007204610951009,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014396609036273732,
      "loss": 4.6976,
      "mean_token_accuracy": 0.2546194076538086,
      "num_tokens": 167252717.0,
      "step": 72945
    },
    {
      "entropy": 5.6085875034332275,
      "epoch": 7.007684918347743,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014393822203609852,
      "loss": 4.6882,
      "mean_token_accuracy": 0.2582130029797554,
      "num_tokens": 167263207.0,
      "step": 72950
    },
    {
      "entropy": 5.561701440811158,
      "epoch": 7.008165225744476,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014391035675232505,
      "loss": 4.6281,
      "mean_token_accuracy": 0.26827426105737684,
      "num_tokens": 167273943.0,
      "step": 72955
    },
    {
      "entropy": 5.616114473342895,
      "epoch": 7.008645533141211,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001438824945120639,
      "loss": 4.7637,
      "mean_token_accuracy": 0.25394180566072466,
      "num_tokens": 167286218.0,
      "step": 72960
    },
    {
      "entropy": 5.553945398330688,
      "epoch": 7.009125840537944,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014385463531596182,
      "loss": 4.6126,
      "mean_token_accuracy": 0.26867216229438784,
      "num_tokens": 167298323.0,
      "step": 72965
    },
    {
      "entropy": 5.570003366470337,
      "epoch": 7.0096061479346785,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014382677916466568,
      "loss": 4.6204,
      "mean_token_accuracy": 0.26778624653816224,
      "num_tokens": 167310273.0,
      "step": 72970
    },
    {
      "entropy": 5.540704250335693,
      "epoch": 7.010086455331412,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014379892605882223,
      "loss": 4.6077,
      "mean_token_accuracy": 0.2714583158493042,
      "num_tokens": 167321940.0,
      "step": 72975
    },
    {
      "entropy": 5.558940792083741,
      "epoch": 7.010566762728146,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014377107599907813,
      "loss": 4.6794,
      "mean_token_accuracy": 0.26106309741735456,
      "num_tokens": 167333120.0,
      "step": 72980
    },
    {
      "entropy": 5.550432920455933,
      "epoch": 7.01104707012488,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014374322898607993,
      "loss": 4.6515,
      "mean_token_accuracy": 0.2730477601289749,
      "num_tokens": 167344600.0,
      "step": 72985
    },
    {
      "entropy": 5.554981708526611,
      "epoch": 7.011527377521614,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001437153850204743,
      "loss": 4.6068,
      "mean_token_accuracy": 0.26940464079380033,
      "num_tokens": 167355963.0,
      "step": 72990
    },
    {
      "entropy": 5.5383766174316404,
      "epoch": 7.0120076849183475,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014368754410290757,
      "loss": 4.6792,
      "mean_token_accuracy": 0.2608115583658218,
      "num_tokens": 167367926.0,
      "step": 72995
    },
    {
      "entropy": 5.46279616355896,
      "epoch": 7.012487992315082,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014365970623402612,
      "loss": 4.5059,
      "mean_token_accuracy": 0.27242964804172515,
      "num_tokens": 167379966.0,
      "step": 73000
    },
    {
      "entropy": 5.50424165725708,
      "epoch": 7.012968299711815,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014363187141447624,
      "loss": 4.5445,
      "mean_token_accuracy": 0.27934519946575165,
      "num_tokens": 167391685.0,
      "step": 73005
    },
    {
      "entropy": 5.518012523651123,
      "epoch": 7.01344860710855,
      "grad_norm": 1.515625,
      "learning_rate": 0.00014360403964490425,
      "loss": 4.5749,
      "mean_token_accuracy": 0.27312192916870115,
      "num_tokens": 167403033.0,
      "step": 73010
    },
    {
      "entropy": 5.558325338363647,
      "epoch": 7.013928914505283,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014357621092595636,
      "loss": 4.6581,
      "mean_token_accuracy": 0.2644228249788284,
      "num_tokens": 167413614.0,
      "step": 73015
    },
    {
      "entropy": 5.6167443752288815,
      "epoch": 7.014409221902017,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00014354838525827864,
      "loss": 4.6773,
      "mean_token_accuracy": 0.2578620210289955,
      "num_tokens": 167423670.0,
      "step": 73020
    },
    {
      "entropy": 5.505486917495728,
      "epoch": 7.014889529298751,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014352056264251695,
      "loss": 4.5921,
      "mean_token_accuracy": 0.26510191410779954,
      "num_tokens": 167434855.0,
      "step": 73025
    },
    {
      "entropy": 5.550540781021118,
      "epoch": 7.015369836695485,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014349274307931743,
      "loss": 4.6787,
      "mean_token_accuracy": 0.2588001132011414,
      "num_tokens": 167446360.0,
      "step": 73030
    },
    {
      "entropy": 5.535346937179566,
      "epoch": 7.015850144092219,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00014346492656932595,
      "loss": 4.63,
      "mean_token_accuracy": 0.2736825183033943,
      "num_tokens": 167458295.0,
      "step": 73035
    },
    {
      "entropy": 5.597954368591308,
      "epoch": 7.016330451488953,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014343711311318825,
      "loss": 4.6366,
      "mean_token_accuracy": 0.2665732875466347,
      "num_tokens": 167469057.0,
      "step": 73040
    },
    {
      "entropy": 5.526576662063599,
      "epoch": 7.016810758885687,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001434093027115501,
      "loss": 4.6572,
      "mean_token_accuracy": 0.260408490896225,
      "num_tokens": 167481175.0,
      "step": 73045
    },
    {
      "entropy": 5.4994055271148685,
      "epoch": 7.017291066282421,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014338149536505734,
      "loss": 4.5726,
      "mean_token_accuracy": 0.2670094192028046,
      "num_tokens": 167492557.0,
      "step": 73050
    },
    {
      "entropy": 5.488433361053467,
      "epoch": 7.017771373679155,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001433536910743553,
      "loss": 4.5714,
      "mean_token_accuracy": 0.27065039426088333,
      "num_tokens": 167503479.0,
      "step": 73055
    },
    {
      "entropy": 5.622025203704834,
      "epoch": 7.018251681075888,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001433258898400897,
      "loss": 4.703,
      "mean_token_accuracy": 0.26386208534240724,
      "num_tokens": 167514579.0,
      "step": 73060
    },
    {
      "entropy": 5.6094073295593265,
      "epoch": 7.018731988472623,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014329809166290593,
      "loss": 4.6803,
      "mean_token_accuracy": 0.2559779182076454,
      "num_tokens": 167526759.0,
      "step": 73065
    },
    {
      "entropy": 5.539759778976441,
      "epoch": 7.019212295869356,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014327029654344936,
      "loss": 4.6349,
      "mean_token_accuracy": 0.2663307026028633,
      "num_tokens": 167537739.0,
      "step": 73070
    },
    {
      "entropy": 5.563729667663575,
      "epoch": 7.0196926032660905,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014324250448236536,
      "loss": 4.5909,
      "mean_token_accuracy": 0.2718434542417526,
      "num_tokens": 167549293.0,
      "step": 73075
    },
    {
      "entropy": 5.589991712570191,
      "epoch": 7.020172910662824,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014321471548029924,
      "loss": 4.6897,
      "mean_token_accuracy": 0.25906629413366317,
      "num_tokens": 167559903.0,
      "step": 73080
    },
    {
      "entropy": 5.507767105102539,
      "epoch": 7.020653218059558,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00014318692953789602,
      "loss": 4.5867,
      "mean_token_accuracy": 0.2706139847636223,
      "num_tokens": 167570375.0,
      "step": 73085
    },
    {
      "entropy": 5.495127868652344,
      "epoch": 7.021133525456292,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014315914665580094,
      "loss": 4.5896,
      "mean_token_accuracy": 0.27185508608818054,
      "num_tokens": 167581735.0,
      "step": 73090
    },
    {
      "entropy": 5.6090600967407225,
      "epoch": 7.021613832853026,
      "grad_norm": 1.375,
      "learning_rate": 0.00014313136683465893,
      "loss": 4.6948,
      "mean_token_accuracy": 0.26735878437757493,
      "num_tokens": 167592973.0,
      "step": 73095
    },
    {
      "entropy": 5.458469438552856,
      "epoch": 7.022094140249759,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014310359007511505,
      "loss": 4.5258,
      "mean_token_accuracy": 0.27307885736227033,
      "num_tokens": 167603891.0,
      "step": 73100
    },
    {
      "entropy": 5.502515459060669,
      "epoch": 7.022574447646494,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014307581637781414,
      "loss": 4.5644,
      "mean_token_accuracy": 0.27432534098625183,
      "num_tokens": 167614793.0,
      "step": 73105
    },
    {
      "entropy": 5.4523763179779055,
      "epoch": 7.023054755043228,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00014304804574340112,
      "loss": 4.5805,
      "mean_token_accuracy": 0.2746123865246773,
      "num_tokens": 167625851.0,
      "step": 73110
    },
    {
      "entropy": 5.552269506454468,
      "epoch": 7.0235350624399615,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001430202781725206,
      "loss": 4.6235,
      "mean_token_accuracy": 0.2696494162082672,
      "num_tokens": 167636930.0,
      "step": 73115
    },
    {
      "entropy": 5.584697294235229,
      "epoch": 7.024015369836696,
      "grad_norm": 1.5,
      "learning_rate": 0.0001429925136658173,
      "loss": 4.6815,
      "mean_token_accuracy": 0.26596923917531967,
      "num_tokens": 167650298.0,
      "step": 73120
    },
    {
      "entropy": 5.630866432189942,
      "epoch": 7.024495677233429,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014296475222393585,
      "loss": 4.7966,
      "mean_token_accuracy": 0.2573527008295059,
      "num_tokens": 167662556.0,
      "step": 73125
    },
    {
      "entropy": 5.5302989959716795,
      "epoch": 7.024975984630164,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014293699384752083,
      "loss": 4.6021,
      "mean_token_accuracy": 0.26952509880065917,
      "num_tokens": 167674004.0,
      "step": 73130
    },
    {
      "entropy": 5.574158573150635,
      "epoch": 7.025456292026897,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001429092385372166,
      "loss": 4.6493,
      "mean_token_accuracy": 0.2624647498130798,
      "num_tokens": 167684818.0,
      "step": 73135
    },
    {
      "entropy": 5.5917047500610355,
      "epoch": 7.025936599423631,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014288148629366777,
      "loss": 4.6885,
      "mean_token_accuracy": 0.261457534134388,
      "num_tokens": 167695983.0,
      "step": 73140
    },
    {
      "entropy": 5.585774707794189,
      "epoch": 7.026416906820365,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001428537371175184,
      "loss": 4.7263,
      "mean_token_accuracy": 0.2613465413451195,
      "num_tokens": 167708403.0,
      "step": 73145
    },
    {
      "entropy": 5.533954095840454,
      "epoch": 7.026897214217099,
      "grad_norm": 1.625,
      "learning_rate": 0.00014282599100941284,
      "loss": 4.6313,
      "mean_token_accuracy": 0.2705230042338371,
      "num_tokens": 167719794.0,
      "step": 73150
    },
    {
      "entropy": 5.620970916748047,
      "epoch": 7.027377521613833,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014279824796999534,
      "loss": 4.7322,
      "mean_token_accuracy": 0.2593259274959564,
      "num_tokens": 167732061.0,
      "step": 73155
    },
    {
      "entropy": 5.530067348480225,
      "epoch": 7.027857829010567,
      "grad_norm": 1.828125,
      "learning_rate": 0.00014277050799990992,
      "loss": 4.6242,
      "mean_token_accuracy": 0.27415349930524824,
      "num_tokens": 167743681.0,
      "step": 73160
    },
    {
      "entropy": 5.555681562423706,
      "epoch": 7.0283381364073,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00014274277109980072,
      "loss": 4.6173,
      "mean_token_accuracy": 0.271116641163826,
      "num_tokens": 167753869.0,
      "step": 73165
    },
    {
      "entropy": 5.592002868652344,
      "epoch": 7.028818443804035,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001427150372703117,
      "loss": 4.6515,
      "mean_token_accuracy": 0.2594101130962372,
      "num_tokens": 167764857.0,
      "step": 73170
    },
    {
      "entropy": 5.484890079498291,
      "epoch": 7.029298751200768,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001426873065120866,
      "loss": 4.5458,
      "mean_token_accuracy": 0.27586956322193146,
      "num_tokens": 167774297.0,
      "step": 73175
    },
    {
      "entropy": 5.5607483863830565,
      "epoch": 7.029779058597502,
      "grad_norm": 1.625,
      "learning_rate": 0.00014265957882576938,
      "loss": 4.583,
      "mean_token_accuracy": 0.2751629829406738,
      "num_tokens": 167785688.0,
      "step": 73180
    },
    {
      "entropy": 5.488081312179565,
      "epoch": 7.030259365994237,
      "grad_norm": 1.59375,
      "learning_rate": 0.00014263185421200375,
      "loss": 4.6337,
      "mean_token_accuracy": 0.2708441033959389,
      "num_tokens": 167796625.0,
      "step": 73185
    },
    {
      "entropy": 5.565041208267212,
      "epoch": 7.03073967339097,
      "grad_norm": 1.90625,
      "learning_rate": 0.00014260413267143347,
      "loss": 4.6517,
      "mean_token_accuracy": 0.2691453292965889,
      "num_tokens": 167806835.0,
      "step": 73190
    },
    {
      "entropy": 5.48363766670227,
      "epoch": 7.0312199807877045,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014257641420470203,
      "loss": 4.5676,
      "mean_token_accuracy": 0.2728249326348305,
      "num_tokens": 167818231.0,
      "step": 73195
    },
    {
      "entropy": 5.632121896743774,
      "epoch": 7.031700288184438,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014254869881245313,
      "loss": 4.725,
      "mean_token_accuracy": 0.2609554469585419,
      "num_tokens": 167830211.0,
      "step": 73200
    },
    {
      "entropy": 5.560349416732788,
      "epoch": 7.032180595581172,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014252098649533007,
      "loss": 4.718,
      "mean_token_accuracy": 0.2554594561457634,
      "num_tokens": 167840637.0,
      "step": 73205
    },
    {
      "entropy": 5.605376529693603,
      "epoch": 7.032660902977906,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001424932772539763,
      "loss": 4.6637,
      "mean_token_accuracy": 0.26641237288713454,
      "num_tokens": 167851854.0,
      "step": 73210
    },
    {
      "entropy": 5.551298475265503,
      "epoch": 7.03314121037464,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014246557108903523,
      "loss": 4.6184,
      "mean_token_accuracy": 0.2646568238735199,
      "num_tokens": 167862646.0,
      "step": 73215
    },
    {
      "entropy": 5.5420907497406,
      "epoch": 7.0336215177713735,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014243786800115002,
      "loss": 4.6439,
      "mean_token_accuracy": 0.26504864990711213,
      "num_tokens": 167874752.0,
      "step": 73220
    },
    {
      "entropy": 5.558241844177246,
      "epoch": 7.034101825168108,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014241016799096387,
      "loss": 4.6469,
      "mean_token_accuracy": 0.26574368476867677,
      "num_tokens": 167887001.0,
      "step": 73225
    },
    {
      "entropy": 5.5988945960998535,
      "epoch": 7.034582132564841,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014238247105912,
      "loss": 4.6671,
      "mean_token_accuracy": 0.2601494133472443,
      "num_tokens": 167898497.0,
      "step": 73230
    },
    {
      "entropy": 5.530678749084473,
      "epoch": 7.0350624399615755,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001423547772062613,
      "loss": 4.6379,
      "mean_token_accuracy": 0.26875818371772764,
      "num_tokens": 167910488.0,
      "step": 73235
    },
    {
      "entropy": 5.633678817749024,
      "epoch": 7.035542747358309,
      "grad_norm": 1.59375,
      "learning_rate": 0.00014232708643303077,
      "loss": 4.7451,
      "mean_token_accuracy": 0.2563159614801407,
      "num_tokens": 167920750.0,
      "step": 73240
    },
    {
      "entropy": 5.605683898925781,
      "epoch": 7.036023054755043,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014229939874007137,
      "loss": 4.6805,
      "mean_token_accuracy": 0.2589926227927208,
      "num_tokens": 167932637.0,
      "step": 73245
    },
    {
      "entropy": 5.477867841720581,
      "epoch": 7.036503362151777,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014227171412802583,
      "loss": 4.5785,
      "mean_token_accuracy": 0.269228395819664,
      "num_tokens": 167944295.0,
      "step": 73250
    },
    {
      "entropy": 5.551488924026489,
      "epoch": 7.036983669548511,
      "grad_norm": 1.359375,
      "learning_rate": 0.000142244032597537,
      "loss": 4.6337,
      "mean_token_accuracy": 0.2649577558040619,
      "num_tokens": 167955326.0,
      "step": 73255
    },
    {
      "entropy": 5.568483114242554,
      "epoch": 7.037463976945245,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014221635414924766,
      "loss": 4.7091,
      "mean_token_accuracy": 0.26163764894008634,
      "num_tokens": 167966466.0,
      "step": 73260
    },
    {
      "entropy": 5.552259397506714,
      "epoch": 7.037944284341979,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014218867878380012,
      "loss": 4.6298,
      "mean_token_accuracy": 0.2670425623655319,
      "num_tokens": 167978927.0,
      "step": 73265
    },
    {
      "entropy": 5.542652130126953,
      "epoch": 7.038424591738713,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014216100650183718,
      "loss": 4.6134,
      "mean_token_accuracy": 0.2697275340557098,
      "num_tokens": 167990773.0,
      "step": 73270
    },
    {
      "entropy": 5.6348950386047365,
      "epoch": 7.038904899135447,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014213333730400113,
      "loss": 4.7625,
      "mean_token_accuracy": 0.2569560155272484,
      "num_tokens": 168002710.0,
      "step": 73275
    },
    {
      "entropy": 5.4441005229949955,
      "epoch": 7.039385206532181,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001421056711909345,
      "loss": 4.5248,
      "mean_token_accuracy": 0.275781637430191,
      "num_tokens": 168014615.0,
      "step": 73280
    },
    {
      "entropy": 5.524303245544433,
      "epoch": 7.039865513928914,
      "grad_norm": 1.5,
      "learning_rate": 0.00014207800816327956,
      "loss": 4.5762,
      "mean_token_accuracy": 0.2698636084794998,
      "num_tokens": 168025309.0,
      "step": 73285
    },
    {
      "entropy": 5.544626474380493,
      "epoch": 7.040345821325649,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014205034822167866,
      "loss": 4.6028,
      "mean_token_accuracy": 0.2698934614658356,
      "num_tokens": 168037254.0,
      "step": 73290
    },
    {
      "entropy": 5.547358512878418,
      "epoch": 7.040826128722382,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001420226913667738,
      "loss": 4.6782,
      "mean_token_accuracy": 0.2674862533807755,
      "num_tokens": 168050584.0,
      "step": 73295
    },
    {
      "entropy": 5.641821241378784,
      "epoch": 7.041306436119116,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00014199503759920723,
      "loss": 4.7304,
      "mean_token_accuracy": 0.25513156652450564,
      "num_tokens": 168060600.0,
      "step": 73300
    },
    {
      "entropy": 5.55968017578125,
      "epoch": 7.04178674351585,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001419673869196209,
      "loss": 4.6193,
      "mean_token_accuracy": 0.2677605807781219,
      "num_tokens": 168071831.0,
      "step": 73305
    },
    {
      "entropy": 5.5727921485900875,
      "epoch": 7.042267050912584,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014193973932865683,
      "loss": 4.6184,
      "mean_token_accuracy": 0.2711756587028503,
      "num_tokens": 168083010.0,
      "step": 73310
    },
    {
      "entropy": 5.483346891403198,
      "epoch": 7.042747358309318,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014191209482695692,
      "loss": 4.5569,
      "mean_token_accuracy": 0.2732117295265198,
      "num_tokens": 168095848.0,
      "step": 73315
    },
    {
      "entropy": 5.508819103240967,
      "epoch": 7.043227665706052,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014188445341516305,
      "loss": 4.6526,
      "mean_token_accuracy": 0.2681246057152748,
      "num_tokens": 168107752.0,
      "step": 73320
    },
    {
      "entropy": 5.604783010482788,
      "epoch": 7.043707973102785,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014185681509391678,
      "loss": 4.7486,
      "mean_token_accuracy": 0.260158883035183,
      "num_tokens": 168119109.0,
      "step": 73325
    },
    {
      "entropy": 5.55135760307312,
      "epoch": 7.04418828049952,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014182917986385997,
      "loss": 4.5869,
      "mean_token_accuracy": 0.2708472490310669,
      "num_tokens": 168130690.0,
      "step": 73330
    },
    {
      "entropy": 5.611652183532715,
      "epoch": 7.044668587896253,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00014180154772563413,
      "loss": 4.731,
      "mean_token_accuracy": 0.25464636236429217,
      "num_tokens": 168141970.0,
      "step": 73335
    },
    {
      "entropy": 5.546359586715698,
      "epoch": 7.0451488952929875,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014177391867988082,
      "loss": 4.6121,
      "mean_token_accuracy": 0.27019740641117096,
      "num_tokens": 168153424.0,
      "step": 73340
    },
    {
      "entropy": 5.487613201141357,
      "epoch": 7.045629202689722,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014174629272724156,
      "loss": 4.5642,
      "mean_token_accuracy": 0.27474803030490874,
      "num_tokens": 168164124.0,
      "step": 73345
    },
    {
      "entropy": 5.57571930885315,
      "epoch": 7.046109510086455,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014171866986835774,
      "loss": 4.633,
      "mean_token_accuracy": 0.26379009187221525,
      "num_tokens": 168174528.0,
      "step": 73350
    },
    {
      "entropy": 5.582575798034668,
      "epoch": 7.04658981748319,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014169105010387062,
      "loss": 4.7019,
      "mean_token_accuracy": 0.2594164803624153,
      "num_tokens": 168184907.0,
      "step": 73355
    },
    {
      "entropy": 5.502667999267578,
      "epoch": 7.047070124879923,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014166343343442143,
      "loss": 4.5894,
      "mean_token_accuracy": 0.2636879995465279,
      "num_tokens": 168196030.0,
      "step": 73360
    },
    {
      "entropy": 5.506070327758789,
      "epoch": 7.047550432276657,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014163581986065138,
      "loss": 4.5799,
      "mean_token_accuracy": 0.2744167685508728,
      "num_tokens": 168207721.0,
      "step": 73365
    },
    {
      "entropy": 5.503181076049804,
      "epoch": 7.048030739673391,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014160820938320163,
      "loss": 4.5646,
      "mean_token_accuracy": 0.2638531893491745,
      "num_tokens": 168219287.0,
      "step": 73370
    },
    {
      "entropy": 5.479534578323364,
      "epoch": 7.048511047070125,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001415806020027131,
      "loss": 4.5961,
      "mean_token_accuracy": 0.2703237995505333,
      "num_tokens": 168230232.0,
      "step": 73375
    },
    {
      "entropy": 5.520877027511597,
      "epoch": 7.0489913544668585,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014155299771982692,
      "loss": 4.5994,
      "mean_token_accuracy": 0.2691867783665657,
      "num_tokens": 168241467.0,
      "step": 73380
    },
    {
      "entropy": 5.495223665237427,
      "epoch": 7.049471661863593,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014152539653518382,
      "loss": 4.6223,
      "mean_token_accuracy": 0.27071294784545896,
      "num_tokens": 168254693.0,
      "step": 73385
    },
    {
      "entropy": 5.538338041305542,
      "epoch": 7.049951969260326,
      "grad_norm": 1.375,
      "learning_rate": 0.00014149779844942466,
      "loss": 4.6732,
      "mean_token_accuracy": 0.2604970186948776,
      "num_tokens": 168266985.0,
      "step": 73390
    },
    {
      "entropy": 5.519969320297241,
      "epoch": 7.050432276657061,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001414702034631902,
      "loss": 4.5729,
      "mean_token_accuracy": 0.26989112347364425,
      "num_tokens": 168279419.0,
      "step": 73395
    },
    {
      "entropy": 5.505685329437256,
      "epoch": 7.050912584053794,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014144261157712112,
      "loss": 4.5681,
      "mean_token_accuracy": 0.27045102417469025,
      "num_tokens": 168290204.0,
      "step": 73400
    },
    {
      "entropy": 5.519434213638306,
      "epoch": 7.051392891450528,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014141502279185796,
      "loss": 4.629,
      "mean_token_accuracy": 0.2628548592329025,
      "num_tokens": 168301727.0,
      "step": 73405
    },
    {
      "entropy": 5.528857088088989,
      "epoch": 7.051873198847262,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001413874371080414,
      "loss": 4.6666,
      "mean_token_accuracy": 0.2604815915226936,
      "num_tokens": 168313352.0,
      "step": 73410
    },
    {
      "entropy": 5.463309478759766,
      "epoch": 7.052353506243996,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014135985452631183,
      "loss": 4.5511,
      "mean_token_accuracy": 0.2772036507725716,
      "num_tokens": 168324928.0,
      "step": 73415
    },
    {
      "entropy": 5.535771369934082,
      "epoch": 7.0528338136407305,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014133227504730953,
      "loss": 4.5908,
      "mean_token_accuracy": 0.2687584772706032,
      "num_tokens": 168335539.0,
      "step": 73420
    },
    {
      "entropy": 5.558785486221313,
      "epoch": 7.053314121037464,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001413046986716749,
      "loss": 4.6492,
      "mean_token_accuracy": 0.2653864949941635,
      "num_tokens": 168346628.0,
      "step": 73425
    },
    {
      "entropy": 5.487405109405517,
      "epoch": 7.053794428434198,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014127712540004816,
      "loss": 4.5992,
      "mean_token_accuracy": 0.27292275726795195,
      "num_tokens": 168357977.0,
      "step": 73430
    },
    {
      "entropy": 5.604959869384766,
      "epoch": 7.054274735830932,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014124955523306948,
      "loss": 4.7364,
      "mean_token_accuracy": 0.2574183687567711,
      "num_tokens": 168370263.0,
      "step": 73435
    },
    {
      "entropy": 5.6021278381347654,
      "epoch": 7.054755043227666,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014122198817137903,
      "loss": 4.6844,
      "mean_token_accuracy": 0.2601161807775497,
      "num_tokens": 168382330.0,
      "step": 73440
    },
    {
      "entropy": 5.505945539474487,
      "epoch": 7.055235350624399,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014119442421561674,
      "loss": 4.5867,
      "mean_token_accuracy": 0.27358112037181853,
      "num_tokens": 168393591.0,
      "step": 73445
    },
    {
      "entropy": 5.468271017074585,
      "epoch": 7.055715658021134,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001411668633664226,
      "loss": 4.5395,
      "mean_token_accuracy": 0.27080490440130234,
      "num_tokens": 168403798.0,
      "step": 73450
    },
    {
      "entropy": 5.5076923847198485,
      "epoch": 7.056195965417867,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014113930562443654,
      "loss": 4.6135,
      "mean_token_accuracy": 0.2633629307150841,
      "num_tokens": 168416233.0,
      "step": 73455
    },
    {
      "entropy": 5.515041399002075,
      "epoch": 7.0566762728146015,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014111175099029827,
      "loss": 4.6164,
      "mean_token_accuracy": 0.27362459897994995,
      "num_tokens": 168427544.0,
      "step": 73460
    },
    {
      "entropy": 5.590503168106079,
      "epoch": 7.057156580211335,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014108419946464758,
      "loss": 4.7101,
      "mean_token_accuracy": 0.2615038052201271,
      "num_tokens": 168438487.0,
      "step": 73465
    },
    {
      "entropy": 5.466888666152954,
      "epoch": 7.057636887608069,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014105665104812413,
      "loss": 4.5784,
      "mean_token_accuracy": 0.2703955233097076,
      "num_tokens": 168450475.0,
      "step": 73470
    },
    {
      "entropy": 5.546050691604615,
      "epoch": 7.058117195004803,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00014102910574136752,
      "loss": 4.6376,
      "mean_token_accuracy": 0.26974252313375474,
      "num_tokens": 168460675.0,
      "step": 73475
    },
    {
      "entropy": 5.472871351242065,
      "epoch": 7.058597502401537,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014100156354501724,
      "loss": 4.5435,
      "mean_token_accuracy": 0.27615095525979994,
      "num_tokens": 168473205.0,
      "step": 73480
    },
    {
      "entropy": 5.491340780258179,
      "epoch": 7.0590778097982705,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014097402445971275,
      "loss": 4.6007,
      "mean_token_accuracy": 0.27039945870637894,
      "num_tokens": 168484015.0,
      "step": 73485
    },
    {
      "entropy": 5.49952712059021,
      "epoch": 7.059558117195005,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014094648848609348,
      "loss": 4.5871,
      "mean_token_accuracy": 0.27358085960149764,
      "num_tokens": 168495412.0,
      "step": 73490
    },
    {
      "entropy": 5.537569665908814,
      "epoch": 7.060038424591739,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014091895562479872,
      "loss": 4.6831,
      "mean_token_accuracy": 0.26004534512758254,
      "num_tokens": 168507780.0,
      "step": 73495
    },
    {
      "entropy": 5.527877473831177,
      "epoch": 7.060518731988473,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014089142587646764,
      "loss": 4.6256,
      "mean_token_accuracy": 0.26804621070623397,
      "num_tokens": 168519184.0,
      "step": 73500
    },
    {
      "entropy": 5.553924560546875,
      "epoch": 7.060999039385207,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001408638992417394,
      "loss": 4.6493,
      "mean_token_accuracy": 0.26484169363975524,
      "num_tokens": 168531945.0,
      "step": 73505
    },
    {
      "entropy": 5.4220795154571535,
      "epoch": 7.06147934678194,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001408363757212531,
      "loss": 4.5553,
      "mean_token_accuracy": 0.2761289432644844,
      "num_tokens": 168542760.0,
      "step": 73510
    },
    {
      "entropy": 5.425263452529907,
      "epoch": 7.061959654178675,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001408088553156478,
      "loss": 4.5241,
      "mean_token_accuracy": 0.27093071192502977,
      "num_tokens": 168552860.0,
      "step": 73515
    },
    {
      "entropy": 5.477583885192871,
      "epoch": 7.062439961575408,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014078133802556238,
      "loss": 4.5562,
      "mean_token_accuracy": 0.27833130210638046,
      "num_tokens": 168563368.0,
      "step": 73520
    },
    {
      "entropy": 5.513938522338867,
      "epoch": 7.062920268972142,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001407538238516358,
      "loss": 4.6421,
      "mean_token_accuracy": 0.26652141362428666,
      "num_tokens": 168574393.0,
      "step": 73525
    },
    {
      "entropy": 5.6041553020477295,
      "epoch": 7.063400576368876,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014072631279450688,
      "loss": 4.652,
      "mean_token_accuracy": 0.2670684114098549,
      "num_tokens": 168586479.0,
      "step": 73530
    },
    {
      "entropy": 5.521082258224487,
      "epoch": 7.06388088376561,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014069880485481412,
      "loss": 4.6375,
      "mean_token_accuracy": 0.2696401163935661,
      "num_tokens": 168597488.0,
      "step": 73535
    },
    {
      "entropy": 5.547782707214355,
      "epoch": 7.064361191162344,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001406713000331964,
      "loss": 4.6375,
      "mean_token_accuracy": 0.2717360079288483,
      "num_tokens": 168608392.0,
      "step": 73540
    },
    {
      "entropy": 5.530754232406617,
      "epoch": 7.064841498559078,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014064379833029217,
      "loss": 4.6853,
      "mean_token_accuracy": 0.2546174317598343,
      "num_tokens": 168620995.0,
      "step": 73545
    },
    {
      "entropy": 5.53896484375,
      "epoch": 7.065321805955811,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014061629974673997,
      "loss": 4.546,
      "mean_token_accuracy": 0.2732239216566086,
      "num_tokens": 168631239.0,
      "step": 73550
    },
    {
      "entropy": 5.475737047195435,
      "epoch": 7.065802113352546,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014058880428317828,
      "loss": 4.5533,
      "mean_token_accuracy": 0.2728662997484207,
      "num_tokens": 168643411.0,
      "step": 73555
    },
    {
      "entropy": 5.585183238983154,
      "epoch": 7.066282420749279,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001405613119402455,
      "loss": 4.7308,
      "mean_token_accuracy": 0.2595270201563835,
      "num_tokens": 168654629.0,
      "step": 73560
    },
    {
      "entropy": 5.522166728973389,
      "epoch": 7.0667627281460135,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001405338227185798,
      "loss": 4.608,
      "mean_token_accuracy": 0.2662870541214943,
      "num_tokens": 168666496.0,
      "step": 73565
    },
    {
      "entropy": 5.581201028823853,
      "epoch": 7.067243035542747,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014050633661881945,
      "loss": 4.6443,
      "mean_token_accuracy": 0.2574384406208992,
      "num_tokens": 168678564.0,
      "step": 73570
    },
    {
      "entropy": 5.485375547409058,
      "epoch": 7.067723342939481,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014047885364160256,
      "loss": 4.4812,
      "mean_token_accuracy": 0.2731946587562561,
      "num_tokens": 168689450.0,
      "step": 73575
    },
    {
      "entropy": 5.5436110496521,
      "epoch": 7.068203650336216,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001404513737875673,
      "loss": 4.5982,
      "mean_token_accuracy": 0.2606070801615715,
      "num_tokens": 168701611.0,
      "step": 73580
    },
    {
      "entropy": 5.626401376724243,
      "epoch": 7.068683957732949,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014042389705735159,
      "loss": 4.6362,
      "mean_token_accuracy": 0.2587163746356964,
      "num_tokens": 168712626.0,
      "step": 73585
    },
    {
      "entropy": 5.453693866729736,
      "epoch": 7.069164265129683,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014039642345159344,
      "loss": 4.5391,
      "mean_token_accuracy": 0.2741744458675385,
      "num_tokens": 168724352.0,
      "step": 73590
    },
    {
      "entropy": 5.424973583221435,
      "epoch": 7.069644572526417,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001403689529709306,
      "loss": 4.5496,
      "mean_token_accuracy": 0.27321096062660216,
      "num_tokens": 168736301.0,
      "step": 73595
    },
    {
      "entropy": 5.59266505241394,
      "epoch": 7.070124879923151,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014034148561600088,
      "loss": 4.7593,
      "mean_token_accuracy": 0.25465050637722014,
      "num_tokens": 168747782.0,
      "step": 73600
    },
    {
      "entropy": 5.5472015857696535,
      "epoch": 7.0706051873198845,
      "grad_norm": 1.546875,
      "learning_rate": 0.00014031402138744203,
      "loss": 4.6259,
      "mean_token_accuracy": 0.27129580825567245,
      "num_tokens": 168758785.0,
      "step": 73605
    },
    {
      "entropy": 5.541590213775635,
      "epoch": 7.071085494716619,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014028656028589165,
      "loss": 4.5861,
      "mean_token_accuracy": 0.2671747475862503,
      "num_tokens": 168769454.0,
      "step": 73610
    },
    {
      "entropy": 5.537990140914917,
      "epoch": 7.071565802113352,
      "grad_norm": 1.625,
      "learning_rate": 0.00014025910231198736,
      "loss": 4.5967,
      "mean_token_accuracy": 0.26318683475255966,
      "num_tokens": 168779938.0,
      "step": 73615
    },
    {
      "entropy": 5.590822362899781,
      "epoch": 7.072046109510087,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00014023164746636662,
      "loss": 4.7339,
      "mean_token_accuracy": 0.2588916555047035,
      "num_tokens": 168791743.0,
      "step": 73620
    },
    {
      "entropy": 5.581443214416504,
      "epoch": 7.07252641690682,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001402041957496668,
      "loss": 4.692,
      "mean_token_accuracy": 0.26017486602067946,
      "num_tokens": 168802293.0,
      "step": 73625
    },
    {
      "entropy": 5.556732511520385,
      "epoch": 7.073006724303554,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00014017674716252534,
      "loss": 4.6221,
      "mean_token_accuracy": 0.26546196490526197,
      "num_tokens": 168816717.0,
      "step": 73630
    },
    {
      "entropy": 5.480586910247803,
      "epoch": 7.073487031700288,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001401493017055794,
      "loss": 4.583,
      "mean_token_accuracy": 0.268784898519516,
      "num_tokens": 168830108.0,
      "step": 73635
    },
    {
      "entropy": 5.616439437866211,
      "epoch": 7.073967339097022,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014012185937946632,
      "loss": 4.664,
      "mean_token_accuracy": 0.2601202577352524,
      "num_tokens": 168842550.0,
      "step": 73640
    },
    {
      "entropy": 5.533539533615112,
      "epoch": 7.074447646493756,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001400944201848231,
      "loss": 4.6297,
      "mean_token_accuracy": 0.2587776154279709,
      "num_tokens": 168855282.0,
      "step": 73645
    },
    {
      "entropy": 5.504299974441528,
      "epoch": 7.07492795389049,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00014006698412228697,
      "loss": 4.6011,
      "mean_token_accuracy": 0.26788805723190307,
      "num_tokens": 168867121.0,
      "step": 73650
    },
    {
      "entropy": 5.533366298675537,
      "epoch": 7.075408261287224,
      "grad_norm": 1.484375,
      "learning_rate": 0.00014003955119249467,
      "loss": 4.5454,
      "mean_token_accuracy": 0.275897254049778,
      "num_tokens": 168877986.0,
      "step": 73655
    },
    {
      "entropy": 5.542065715789795,
      "epoch": 7.075888568683958,
      "grad_norm": 1.5,
      "learning_rate": 0.0001400121213960833,
      "loss": 4.6136,
      "mean_token_accuracy": 0.26321711838245393,
      "num_tokens": 168889252.0,
      "step": 73660
    },
    {
      "entropy": 5.530004978179932,
      "epoch": 7.076368876080692,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013998469473368964,
      "loss": 4.6195,
      "mean_token_accuracy": 0.2692072555422783,
      "num_tokens": 168900509.0,
      "step": 73665
    },
    {
      "entropy": 5.467665767669677,
      "epoch": 7.076849183477425,
      "grad_norm": 1.234375,
      "learning_rate": 0.00013995727120595043,
      "loss": 4.5919,
      "mean_token_accuracy": 0.2662634700536728,
      "num_tokens": 168914136.0,
      "step": 73670
    },
    {
      "entropy": 5.445248079299927,
      "epoch": 7.07732949087416,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00013992985081350237,
      "loss": 4.5096,
      "mean_token_accuracy": 0.2842234879732132,
      "num_tokens": 168925730.0,
      "step": 73675
    },
    {
      "entropy": 5.5143688201904295,
      "epoch": 7.077809798270893,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013990243355698222,
      "loss": 4.6105,
      "mean_token_accuracy": 0.2616979330778122,
      "num_tokens": 168936663.0,
      "step": 73680
    },
    {
      "entropy": 5.515757894515991,
      "epoch": 7.0782901056676275,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001398750194370263,
      "loss": 4.6182,
      "mean_token_accuracy": 0.2668797269463539,
      "num_tokens": 168948894.0,
      "step": 73685
    },
    {
      "entropy": 5.5379753589630125,
      "epoch": 7.078770413064361,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001398476084542712,
      "loss": 4.6589,
      "mean_token_accuracy": 0.269814071059227,
      "num_tokens": 168961213.0,
      "step": 73690
    },
    {
      "entropy": 5.5441468238830565,
      "epoch": 7.079250720461095,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001398202006093533,
      "loss": 4.6023,
      "mean_token_accuracy": 0.27515115439891813,
      "num_tokens": 168972609.0,
      "step": 73695
    },
    {
      "entropy": 5.547084760665894,
      "epoch": 7.079731027857829,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013979279590290895,
      "loss": 4.6318,
      "mean_token_accuracy": 0.26259041130542754,
      "num_tokens": 168983898.0,
      "step": 73700
    },
    {
      "entropy": 5.504147863388061,
      "epoch": 7.080211335254563,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001397653943355744,
      "loss": 4.6202,
      "mean_token_accuracy": 0.2632964804768562,
      "num_tokens": 168995606.0,
      "step": 73705
    },
    {
      "entropy": 5.5504734992980955,
      "epoch": 7.0806916426512965,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013973799590798587,
      "loss": 4.6814,
      "mean_token_accuracy": 0.26052730679512026,
      "num_tokens": 169007533.0,
      "step": 73710
    },
    {
      "entropy": 5.456449556350708,
      "epoch": 7.081171950048031,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001397106006207794,
      "loss": 4.5371,
      "mean_token_accuracy": 0.2750008672475815,
      "num_tokens": 169017137.0,
      "step": 73715
    },
    {
      "entropy": 5.519888973236084,
      "epoch": 7.081652257444764,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013968320847459104,
      "loss": 4.594,
      "mean_token_accuracy": 0.2694978594779968,
      "num_tokens": 169029128.0,
      "step": 73720
    },
    {
      "entropy": 5.523782396316529,
      "epoch": 7.082132564841499,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013965581947005676,
      "loss": 4.625,
      "mean_token_accuracy": 0.2678512543439865,
      "num_tokens": 169041361.0,
      "step": 73725
    },
    {
      "entropy": 5.569071245193482,
      "epoch": 7.082612872238233,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013962843360781246,
      "loss": 4.6339,
      "mean_token_accuracy": 0.2636341616511345,
      "num_tokens": 169052641.0,
      "step": 73730
    },
    {
      "entropy": 5.52393798828125,
      "epoch": 7.083093179634966,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013960105088849401,
      "loss": 4.5739,
      "mean_token_accuracy": 0.2673016518354416,
      "num_tokens": 169065407.0,
      "step": 73735
    },
    {
      "entropy": 5.499409532546997,
      "epoch": 7.083573487031701,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013957367131273713,
      "loss": 4.5544,
      "mean_token_accuracy": 0.27444659024477003,
      "num_tokens": 169076231.0,
      "step": 73740
    },
    {
      "entropy": 5.53809757232666,
      "epoch": 7.084053794428434,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013954629488117741,
      "loss": 4.5642,
      "mean_token_accuracy": 0.2738802582025528,
      "num_tokens": 169087175.0,
      "step": 73745
    },
    {
      "entropy": 5.5036386013031,
      "epoch": 7.084534101825168,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001395189215944505,
      "loss": 4.6627,
      "mean_token_accuracy": 0.2698191523551941,
      "num_tokens": 169099064.0,
      "step": 73750
    },
    {
      "entropy": 5.509459638595581,
      "epoch": 7.085014409221902,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013949155145319192,
      "loss": 4.5688,
      "mean_token_accuracy": 0.27328169643878936,
      "num_tokens": 169110316.0,
      "step": 73755
    },
    {
      "entropy": 5.520721006393432,
      "epoch": 7.085494716618636,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001394641844580372,
      "loss": 4.621,
      "mean_token_accuracy": 0.2676038101315498,
      "num_tokens": 169121712.0,
      "step": 73760
    },
    {
      "entropy": 5.640804481506348,
      "epoch": 7.08597502401537,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001394368206096216,
      "loss": 4.7506,
      "mean_token_accuracy": 0.25757564753293993,
      "num_tokens": 169136103.0,
      "step": 73765
    },
    {
      "entropy": 5.544728326797485,
      "epoch": 7.086455331412104,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001394094599085806,
      "loss": 4.6484,
      "mean_token_accuracy": 0.2661354333162308,
      "num_tokens": 169148976.0,
      "step": 73770
    },
    {
      "entropy": 5.54124813079834,
      "epoch": 7.086935638808837,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001393821023555492,
      "loss": 4.6765,
      "mean_token_accuracy": 0.2637198343873024,
      "num_tokens": 169159807.0,
      "step": 73775
    },
    {
      "entropy": 5.55276050567627,
      "epoch": 7.087415946205572,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001393547479511627,
      "loss": 4.6556,
      "mean_token_accuracy": 0.25674339532852175,
      "num_tokens": 169171625.0,
      "step": 73780
    },
    {
      "entropy": 5.555880689620972,
      "epoch": 7.087896253602305,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013932739669605617,
      "loss": 4.6164,
      "mean_token_accuracy": 0.2692781284451485,
      "num_tokens": 169183673.0,
      "step": 73785
    },
    {
      "entropy": 5.617931270599366,
      "epoch": 7.0883765609990395,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013930004859086465,
      "loss": 4.6951,
      "mean_token_accuracy": 0.25675651878118516,
      "num_tokens": 169194753.0,
      "step": 73790
    },
    {
      "entropy": 5.491146278381348,
      "epoch": 7.088856868395773,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000139272703636223,
      "loss": 4.5504,
      "mean_token_accuracy": 0.2686678469181061,
      "num_tokens": 169206683.0,
      "step": 73795
    },
    {
      "entropy": 5.546147727966309,
      "epoch": 7.089337175792507,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013924536183276625,
      "loss": 4.6454,
      "mean_token_accuracy": 0.2642544522881508,
      "num_tokens": 169217241.0,
      "step": 73800
    },
    {
      "entropy": 5.510651063919068,
      "epoch": 7.0898174831892415,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013921802318112903,
      "loss": 4.6408,
      "mean_token_accuracy": 0.26478707492351533,
      "num_tokens": 169228553.0,
      "step": 73805
    },
    {
      "entropy": 5.4496846199035645,
      "epoch": 7.090297790585975,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001391906876819461,
      "loss": 4.6299,
      "mean_token_accuracy": 0.2669177696108818,
      "num_tokens": 169238865.0,
      "step": 73810
    },
    {
      "entropy": 5.512781095504761,
      "epoch": 7.090778097982709,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013916335533585214,
      "loss": 4.5829,
      "mean_token_accuracy": 0.26202044636011124,
      "num_tokens": 169249856.0,
      "step": 73815
    },
    {
      "entropy": 5.493684387207031,
      "epoch": 7.091258405379443,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013913602614348174,
      "loss": 4.5913,
      "mean_token_accuracy": 0.2708647966384888,
      "num_tokens": 169261682.0,
      "step": 73820
    },
    {
      "entropy": 5.489960098266602,
      "epoch": 7.091738712776177,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013910870010546932,
      "loss": 4.5838,
      "mean_token_accuracy": 0.2652413100004196,
      "num_tokens": 169273581.0,
      "step": 73825
    },
    {
      "entropy": 5.569044923782348,
      "epoch": 7.0922190201729105,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013908137722244948,
      "loss": 4.6684,
      "mean_token_accuracy": 0.26213457733392714,
      "num_tokens": 169284257.0,
      "step": 73830
    },
    {
      "entropy": 5.520832109451294,
      "epoch": 7.092699327569645,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001390540574950564,
      "loss": 4.6088,
      "mean_token_accuracy": 0.26127859950065613,
      "num_tokens": 169296125.0,
      "step": 73835
    },
    {
      "entropy": 5.51398491859436,
      "epoch": 7.093179634966378,
      "grad_norm": 1.375,
      "learning_rate": 0.00013902674092392442,
      "loss": 4.6325,
      "mean_token_accuracy": 0.2664514183998108,
      "num_tokens": 169308463.0,
      "step": 73840
    },
    {
      "entropy": 5.458111047744751,
      "epoch": 7.093659942363113,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013899942750968775,
      "loss": 4.5889,
      "mean_token_accuracy": 0.27047374546527864,
      "num_tokens": 169319432.0,
      "step": 73845
    },
    {
      "entropy": 5.481448698043823,
      "epoch": 7.094140249759846,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013897211725298053,
      "loss": 4.5626,
      "mean_token_accuracy": 0.2692851394414902,
      "num_tokens": 169329399.0,
      "step": 73850
    },
    {
      "entropy": 5.493642711639405,
      "epoch": 7.09462055715658,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013894481015443687,
      "loss": 4.5767,
      "mean_token_accuracy": 0.27574796974658966,
      "num_tokens": 169340540.0,
      "step": 73855
    },
    {
      "entropy": 5.5379030227661135,
      "epoch": 7.095100864553314,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013891750621469073,
      "loss": 4.6083,
      "mean_token_accuracy": 0.2687259778380394,
      "num_tokens": 169351044.0,
      "step": 73860
    },
    {
      "entropy": 5.655591058731079,
      "epoch": 7.095581171950048,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000138890205434376,
      "loss": 4.8086,
      "mean_token_accuracy": 0.25294742435216905,
      "num_tokens": 169363570.0,
      "step": 73865
    },
    {
      "entropy": 5.516164064407349,
      "epoch": 7.0960614793467816,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001388629078141265,
      "loss": 4.5887,
      "mean_token_accuracy": 0.2741409823298454,
      "num_tokens": 169375175.0,
      "step": 73870
    },
    {
      "entropy": 5.572156190872192,
      "epoch": 7.096541786743516,
      "grad_norm": 1.5,
      "learning_rate": 0.00013883561335457605,
      "loss": 4.701,
      "mean_token_accuracy": 0.2517304092645645,
      "num_tokens": 169387065.0,
      "step": 73875
    },
    {
      "entropy": 5.526774311065674,
      "epoch": 7.09702209414025,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013880832205635834,
      "loss": 4.6383,
      "mean_token_accuracy": 0.25664026737213136,
      "num_tokens": 169397770.0,
      "step": 73880
    },
    {
      "entropy": 5.551918649673462,
      "epoch": 7.097502401536984,
      "grad_norm": 1.578125,
      "learning_rate": 0.00013878103392010694,
      "loss": 4.6072,
      "mean_token_accuracy": 0.2715077489614487,
      "num_tokens": 169409872.0,
      "step": 73885
    },
    {
      "entropy": 5.562070083618164,
      "epoch": 7.097982708933718,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013875374894645556,
      "loss": 4.661,
      "mean_token_accuracy": 0.26186527907848356,
      "num_tokens": 169420227.0,
      "step": 73890
    },
    {
      "entropy": 5.464760208129883,
      "epoch": 7.098463016330451,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001387264671360375,
      "loss": 4.5284,
      "mean_token_accuracy": 0.2718514442443848,
      "num_tokens": 169431166.0,
      "step": 73895
    },
    {
      "entropy": 5.519573450088501,
      "epoch": 7.098943323727186,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013869918848948617,
      "loss": 4.5808,
      "mean_token_accuracy": 0.2710700213909149,
      "num_tokens": 169442460.0,
      "step": 73900
    },
    {
      "entropy": 5.515498971939087,
      "epoch": 7.099423631123919,
      "grad_norm": 1.46875,
      "learning_rate": 0.000138671913007435,
      "loss": 4.6492,
      "mean_token_accuracy": 0.2647509858012199,
      "num_tokens": 169454472.0,
      "step": 73905
    },
    {
      "entropy": 5.5409743309021,
      "epoch": 7.0999039385206535,
      "grad_norm": 1.375,
      "learning_rate": 0.00013864464069051718,
      "loss": 4.6033,
      "mean_token_accuracy": 0.26840486377477646,
      "num_tokens": 169466784.0,
      "step": 73910
    },
    {
      "entropy": 5.572575521469116,
      "epoch": 7.100384245917387,
      "grad_norm": 1.5,
      "learning_rate": 0.00013861737153936593,
      "loss": 4.6595,
      "mean_token_accuracy": 0.26398933082818987,
      "num_tokens": 169479661.0,
      "step": 73915
    },
    {
      "entropy": 5.6307929992675785,
      "epoch": 7.100864553314121,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013859010555461438,
      "loss": 4.6895,
      "mean_token_accuracy": 0.2562789648771286,
      "num_tokens": 169490873.0,
      "step": 73920
    },
    {
      "entropy": 5.490722274780273,
      "epoch": 7.101344860710855,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001385628427368955,
      "loss": 4.5847,
      "mean_token_accuracy": 0.27238104939460756,
      "num_tokens": 169502867.0,
      "step": 73925
    },
    {
      "entropy": 5.510533571243286,
      "epoch": 7.101825168107589,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013853558308684222,
      "loss": 4.5716,
      "mean_token_accuracy": 0.278498911857605,
      "num_tokens": 169515205.0,
      "step": 73930
    },
    {
      "entropy": 5.5160901069641115,
      "epoch": 7.1023054755043225,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013850832660508753,
      "loss": 4.6096,
      "mean_token_accuracy": 0.26527927815914154,
      "num_tokens": 169525641.0,
      "step": 73935
    },
    {
      "entropy": 5.489575815200806,
      "epoch": 7.102785782901057,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013848107329226416,
      "loss": 4.5696,
      "mean_token_accuracy": 0.2699206382036209,
      "num_tokens": 169538343.0,
      "step": 73940
    },
    {
      "entropy": 5.533459663391113,
      "epoch": 7.10326609029779,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013845382314900493,
      "loss": 4.6421,
      "mean_token_accuracy": 0.26352368146181104,
      "num_tokens": 169549615.0,
      "step": 73945
    },
    {
      "entropy": 5.574195861816406,
      "epoch": 7.1037463976945245,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013842657617594252,
      "loss": 4.6683,
      "mean_token_accuracy": 0.26424478739500046,
      "num_tokens": 169561395.0,
      "step": 73950
    },
    {
      "entropy": 5.478045511245727,
      "epoch": 7.104226705091258,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013839933237370938,
      "loss": 4.5595,
      "mean_token_accuracy": 0.2755463719367981,
      "num_tokens": 169572648.0,
      "step": 73955
    },
    {
      "entropy": 5.46611795425415,
      "epoch": 7.104707012487992,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013837209174293817,
      "loss": 4.5599,
      "mean_token_accuracy": 0.26834983229637144,
      "num_tokens": 169583025.0,
      "step": 73960
    },
    {
      "entropy": 5.537231349945069,
      "epoch": 7.105187319884727,
      "grad_norm": 1.578125,
      "learning_rate": 0.00013834485428426124,
      "loss": 4.644,
      "mean_token_accuracy": 0.27239468693733215,
      "num_tokens": 169593488.0,
      "step": 73965
    },
    {
      "entropy": 5.484343194961548,
      "epoch": 7.10566762728146,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013831761999831104,
      "loss": 4.6355,
      "mean_token_accuracy": 0.26537704169750215,
      "num_tokens": 169604489.0,
      "step": 73970
    },
    {
      "entropy": 5.491749334335327,
      "epoch": 7.106147934678194,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013829038888571983,
      "loss": 4.705,
      "mean_token_accuracy": 0.26181718707084656,
      "num_tokens": 169615833.0,
      "step": 73975
    },
    {
      "entropy": 5.522173357009888,
      "epoch": 7.106628242074928,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013826316094711982,
      "loss": 4.6464,
      "mean_token_accuracy": 0.26303552091121674,
      "num_tokens": 169627891.0,
      "step": 73980
    },
    {
      "entropy": 5.538308811187744,
      "epoch": 7.107108549471662,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013823593618314322,
      "loss": 4.5975,
      "mean_token_accuracy": 0.2669689267873764,
      "num_tokens": 169638327.0,
      "step": 73985
    },
    {
      "entropy": 5.474299955368042,
      "epoch": 7.107588856868396,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001382087145944222,
      "loss": 4.5676,
      "mean_token_accuracy": 0.26598956286907194,
      "num_tokens": 169651064.0,
      "step": 73990
    },
    {
      "entropy": 5.6002637386322025,
      "epoch": 7.10806916426513,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001381814961815885,
      "loss": 4.6581,
      "mean_token_accuracy": 0.2620506823062897,
      "num_tokens": 169661183.0,
      "step": 73995
    },
    {
      "entropy": 5.498781633377075,
      "epoch": 7.108549471661863,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013815428094527418,
      "loss": 4.5189,
      "mean_token_accuracy": 0.2782899707555771,
      "num_tokens": 169672711.0,
      "step": 74000
    },
    {
      "entropy": 5.445368051528931,
      "epoch": 7.109029779058598,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013812706888611113,
      "loss": 4.5502,
      "mean_token_accuracy": 0.273337060213089,
      "num_tokens": 169684155.0,
      "step": 74005
    },
    {
      "entropy": 5.453190469741822,
      "epoch": 7.109510086455331,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001380998600047311,
      "loss": 4.5752,
      "mean_token_accuracy": 0.27055991590023043,
      "num_tokens": 169695981.0,
      "step": 74010
    },
    {
      "entropy": 5.509819269180298,
      "epoch": 7.109990393852065,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013807265430176586,
      "loss": 4.5986,
      "mean_token_accuracy": 0.263154199719429,
      "num_tokens": 169707485.0,
      "step": 74015
    },
    {
      "entropy": 5.541237258911133,
      "epoch": 7.110470701248799,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013804545177784695,
      "loss": 4.5879,
      "mean_token_accuracy": 0.27379643023014066,
      "num_tokens": 169719008.0,
      "step": 74020
    },
    {
      "entropy": 5.547253036499024,
      "epoch": 7.110951008645533,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013801825243360598,
      "loss": 4.6344,
      "mean_token_accuracy": 0.2657721772789955,
      "num_tokens": 169731916.0,
      "step": 74025
    },
    {
      "entropy": 5.549623775482178,
      "epoch": 7.111431316042267,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001379910562696745,
      "loss": 4.6074,
      "mean_token_accuracy": 0.2679978430271149,
      "num_tokens": 169743614.0,
      "step": 74030
    },
    {
      "entropy": 5.527208757400513,
      "epoch": 7.111911623439001,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00013796386328668382,
      "loss": 4.6413,
      "mean_token_accuracy": 0.2705214127898216,
      "num_tokens": 169756510.0,
      "step": 74035
    },
    {
      "entropy": 5.527632856369019,
      "epoch": 7.112391930835735,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001379366734852653,
      "loss": 4.5489,
      "mean_token_accuracy": 0.28398258090019224,
      "num_tokens": 169766459.0,
      "step": 74040
    },
    {
      "entropy": 5.538580799102784,
      "epoch": 7.112872238232469,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013790948686605024,
      "loss": 4.6141,
      "mean_token_accuracy": 0.26709558814764023,
      "num_tokens": 169778176.0,
      "step": 74045
    },
    {
      "entropy": 5.495402002334595,
      "epoch": 7.113352545629203,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001378823034296698,
      "loss": 4.5741,
      "mean_token_accuracy": 0.2724573716521263,
      "num_tokens": 169790020.0,
      "step": 74050
    },
    {
      "entropy": 5.529724979400635,
      "epoch": 7.1138328530259365,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013785512317675513,
      "loss": 4.5429,
      "mean_token_accuracy": 0.270048488676548,
      "num_tokens": 169801260.0,
      "step": 74055
    },
    {
      "entropy": 5.47285943031311,
      "epoch": 7.114313160422671,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013782794610793725,
      "loss": 4.5511,
      "mean_token_accuracy": 0.273667311668396,
      "num_tokens": 169812994.0,
      "step": 74060
    },
    {
      "entropy": 5.505134201049804,
      "epoch": 7.114793467819404,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001378007722238472,
      "loss": 4.6178,
      "mean_token_accuracy": 0.27267240881919863,
      "num_tokens": 169823722.0,
      "step": 74065
    },
    {
      "entropy": 5.590714597702027,
      "epoch": 7.115273775216139,
      "grad_norm": 1.375,
      "learning_rate": 0.00013777360152511582,
      "loss": 4.7129,
      "mean_token_accuracy": 0.253943695127964,
      "num_tokens": 169835480.0,
      "step": 74070
    },
    {
      "entropy": 5.60672435760498,
      "epoch": 7.115754082612872,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013774643401237392,
      "loss": 4.6504,
      "mean_token_accuracy": 0.27036159932613374,
      "num_tokens": 169845958.0,
      "step": 74075
    },
    {
      "entropy": 5.572702360153198,
      "epoch": 7.116234390009606,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001377192696862522,
      "loss": 4.6699,
      "mean_token_accuracy": 0.26414471566677095,
      "num_tokens": 169858825.0,
      "step": 74080
    },
    {
      "entropy": 5.459206867218017,
      "epoch": 7.11671469740634,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013769210854738145,
      "loss": 4.5429,
      "mean_token_accuracy": 0.27571929842233656,
      "num_tokens": 169870213.0,
      "step": 74085
    },
    {
      "entropy": 5.523894882202148,
      "epoch": 7.117195004803074,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001376649505963922,
      "loss": 4.5428,
      "mean_token_accuracy": 0.27995885610580445,
      "num_tokens": 169881429.0,
      "step": 74090
    },
    {
      "entropy": 5.449127674102783,
      "epoch": 7.1176753121998075,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013763779583391501,
      "loss": 4.5736,
      "mean_token_accuracy": 0.2663209721446037,
      "num_tokens": 169893847.0,
      "step": 74095
    },
    {
      "entropy": 5.501451539993286,
      "epoch": 7.118155619596542,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001376106442605804,
      "loss": 4.572,
      "mean_token_accuracy": 0.27160927653312683,
      "num_tokens": 169904653.0,
      "step": 74100
    },
    {
      "entropy": 5.572364377975464,
      "epoch": 7.118635926993275,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00013758349587701858,
      "loss": 4.6419,
      "mean_token_accuracy": 0.2723774343729019,
      "num_tokens": 169914761.0,
      "step": 74105
    },
    {
      "entropy": 5.569084024429321,
      "epoch": 7.11911623439001,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001375563506838599,
      "loss": 4.6927,
      "mean_token_accuracy": 0.2616042003035545,
      "num_tokens": 169926407.0,
      "step": 74110
    },
    {
      "entropy": 5.532780265808105,
      "epoch": 7.119596541786744,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013752920868173468,
      "loss": 4.6149,
      "mean_token_accuracy": 0.2728362172842026,
      "num_tokens": 169937831.0,
      "step": 74115
    },
    {
      "entropy": 5.489981937408447,
      "epoch": 7.120076849183477,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013750206987127303,
      "loss": 4.5465,
      "mean_token_accuracy": 0.27461793571710585,
      "num_tokens": 169948429.0,
      "step": 74120
    },
    {
      "entropy": 5.5687071800231935,
      "epoch": 7.120557156580212,
      "grad_norm": 1.328125,
      "learning_rate": 0.000137474934253105,
      "loss": 4.6535,
      "mean_token_accuracy": 0.2613865822553635,
      "num_tokens": 169959934.0,
      "step": 74125
    },
    {
      "entropy": 5.584298706054687,
      "epoch": 7.121037463976945,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013744780182786077,
      "loss": 4.6932,
      "mean_token_accuracy": 0.2588234320282936,
      "num_tokens": 169970736.0,
      "step": 74130
    },
    {
      "entropy": 5.531216859817505,
      "epoch": 7.1215177713736795,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013742067259617003,
      "loss": 4.6117,
      "mean_token_accuracy": 0.26302527636289597,
      "num_tokens": 169981993.0,
      "step": 74135
    },
    {
      "entropy": 5.4761241436004635,
      "epoch": 7.121998078770413,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001373935465586627,
      "loss": 4.5412,
      "mean_token_accuracy": 0.2642862468957901,
      "num_tokens": 169993183.0,
      "step": 74140
    },
    {
      "entropy": 5.543015766143799,
      "epoch": 7.122478386167147,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013736642371596863,
      "loss": 4.6116,
      "mean_token_accuracy": 0.2699196681380272,
      "num_tokens": 170004846.0,
      "step": 74145
    },
    {
      "entropy": 5.556795120239258,
      "epoch": 7.122958693563881,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013733930406871752,
      "loss": 4.6578,
      "mean_token_accuracy": 0.26815953850746155,
      "num_tokens": 170015532.0,
      "step": 74150
    },
    {
      "entropy": 5.596641635894775,
      "epoch": 7.123439000960615,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013731218761753905,
      "loss": 4.7571,
      "mean_token_accuracy": 0.2580200955271721,
      "num_tokens": 170026366.0,
      "step": 74155
    },
    {
      "entropy": 5.533947563171386,
      "epoch": 7.123919308357348,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013728507436306276,
      "loss": 4.5841,
      "mean_token_accuracy": 0.262365385890007,
      "num_tokens": 170036429.0,
      "step": 74160
    },
    {
      "entropy": 5.606025648117066,
      "epoch": 7.124399615754083,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013725796430591803,
      "loss": 4.6965,
      "mean_token_accuracy": 0.2600237011909485,
      "num_tokens": 170048518.0,
      "step": 74165
    },
    {
      "entropy": 5.48764853477478,
      "epoch": 7.124879923150816,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00013723085744673442,
      "loss": 4.5649,
      "mean_token_accuracy": 0.2698725700378418,
      "num_tokens": 170058596.0,
      "step": 74170
    },
    {
      "entropy": 5.584135913848877,
      "epoch": 7.1253602305475505,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013720375378614114,
      "loss": 4.7142,
      "mean_token_accuracy": 0.2516046553850174,
      "num_tokens": 170070216.0,
      "step": 74175
    },
    {
      "entropy": 5.622352266311646,
      "epoch": 7.125840537944284,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013717665332476758,
      "loss": 4.7666,
      "mean_token_accuracy": 0.25390070378780366,
      "num_tokens": 170082329.0,
      "step": 74180
    },
    {
      "entropy": 5.5245575428009035,
      "epoch": 7.126320845341018,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013714955606324285,
      "loss": 4.5714,
      "mean_token_accuracy": 0.26387827545404435,
      "num_tokens": 170093732.0,
      "step": 74185
    },
    {
      "entropy": 5.515815591812133,
      "epoch": 7.126801152737752,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013712246200219607,
      "loss": 4.5687,
      "mean_token_accuracy": 0.27151616513729093,
      "num_tokens": 170106444.0,
      "step": 74190
    },
    {
      "entropy": 5.511297273635864,
      "epoch": 7.127281460134486,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013709537114225644,
      "loss": 4.5789,
      "mean_token_accuracy": 0.26977607905864714,
      "num_tokens": 170118300.0,
      "step": 74195
    },
    {
      "entropy": 5.5406805038452145,
      "epoch": 7.12776176753122,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001370682834840527,
      "loss": 4.5744,
      "mean_token_accuracy": 0.28055537343025205,
      "num_tokens": 170129582.0,
      "step": 74200
    },
    {
      "entropy": 5.5932111740112305,
      "epoch": 7.128242074927954,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013704119902821385,
      "loss": 4.6611,
      "mean_token_accuracy": 0.2598853513598442,
      "num_tokens": 170140498.0,
      "step": 74205
    },
    {
      "entropy": 5.589874029159546,
      "epoch": 7.128722382324688,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001370141177753687,
      "loss": 4.7249,
      "mean_token_accuracy": 0.25749265402555466,
      "num_tokens": 170152210.0,
      "step": 74210
    },
    {
      "entropy": 5.592243099212647,
      "epoch": 7.129202689721422,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013698703972614602,
      "loss": 4.6341,
      "mean_token_accuracy": 0.2661929249763489,
      "num_tokens": 170161882.0,
      "step": 74215
    },
    {
      "entropy": 5.51217360496521,
      "epoch": 7.129682997118156,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013695996488117444,
      "loss": 4.5255,
      "mean_token_accuracy": 0.2788235113024712,
      "num_tokens": 170173871.0,
      "step": 74220
    },
    {
      "entropy": 5.591884422302246,
      "epoch": 7.130163304514889,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013693289324108265,
      "loss": 4.7006,
      "mean_token_accuracy": 0.26235922873020173,
      "num_tokens": 170185195.0,
      "step": 74225
    },
    {
      "entropy": 5.523777770996094,
      "epoch": 7.130643611911624,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013690582480649905,
      "loss": 4.6122,
      "mean_token_accuracy": 0.27221751064062116,
      "num_tokens": 170198295.0,
      "step": 74230
    },
    {
      "entropy": 5.535372877120972,
      "epoch": 7.131123919308357,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013687875957805212,
      "loss": 4.632,
      "mean_token_accuracy": 0.2674989715218544,
      "num_tokens": 170210635.0,
      "step": 74235
    },
    {
      "entropy": 5.509183740615844,
      "epoch": 7.131604226705091,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013685169755637029,
      "loss": 4.5873,
      "mean_token_accuracy": 0.2708041310310364,
      "num_tokens": 170221740.0,
      "step": 74240
    },
    {
      "entropy": 5.489975881576538,
      "epoch": 7.132084534101825,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013682463874208179,
      "loss": 4.5381,
      "mean_token_accuracy": 0.2700631842017174,
      "num_tokens": 170233482.0,
      "step": 74245
    },
    {
      "entropy": 5.681890964508057,
      "epoch": 7.132564841498559,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013679758313581492,
      "loss": 4.7184,
      "mean_token_accuracy": 0.2589510381221771,
      "num_tokens": 170245227.0,
      "step": 74250
    },
    {
      "entropy": 5.524737453460693,
      "epoch": 7.133045148895293,
      "grad_norm": 1.375,
      "learning_rate": 0.00013677053073819784,
      "loss": 4.6666,
      "mean_token_accuracy": 0.26582775861024854,
      "num_tokens": 170257774.0,
      "step": 74255
    },
    {
      "entropy": 5.502507305145263,
      "epoch": 7.133525456292027,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001367434815498585,
      "loss": 4.624,
      "mean_token_accuracy": 0.2683500826358795,
      "num_tokens": 170268903.0,
      "step": 74260
    },
    {
      "entropy": 5.544528007507324,
      "epoch": 7.13400576368876,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013671643557142498,
      "loss": 4.6855,
      "mean_token_accuracy": 0.2655554339289665,
      "num_tokens": 170281726.0,
      "step": 74265
    },
    {
      "entropy": 5.556994199752808,
      "epoch": 7.134486071085495,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013668939280352525,
      "loss": 4.6499,
      "mean_token_accuracy": 0.2618187054991722,
      "num_tokens": 170292719.0,
      "step": 74270
    },
    {
      "entropy": 5.487546396255493,
      "epoch": 7.134966378482229,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001366623532467871,
      "loss": 4.5333,
      "mean_token_accuracy": 0.27400256395339967,
      "num_tokens": 170303223.0,
      "step": 74275
    },
    {
      "entropy": 5.527951860427857,
      "epoch": 7.1354466858789625,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001366353169018383,
      "loss": 4.6687,
      "mean_token_accuracy": 0.26300536245107653,
      "num_tokens": 170314532.0,
      "step": 74280
    },
    {
      "entropy": 5.5400574684143065,
      "epoch": 7.135926993275697,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013660828376930667,
      "loss": 4.6392,
      "mean_token_accuracy": 0.2712987497448921,
      "num_tokens": 170326386.0,
      "step": 74285
    },
    {
      "entropy": 5.461184692382813,
      "epoch": 7.13640730067243,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013658125384981968,
      "loss": 4.5381,
      "mean_token_accuracy": 0.27994937598705294,
      "num_tokens": 170337718.0,
      "step": 74290
    },
    {
      "entropy": 5.557984972000122,
      "epoch": 7.136887608069165,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013655422714400496,
      "loss": 4.6115,
      "mean_token_accuracy": 0.26847357004880906,
      "num_tokens": 170349215.0,
      "step": 74295
    },
    {
      "entropy": 5.559256744384766,
      "epoch": 7.137367915465898,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013652720365249,
      "loss": 4.6161,
      "mean_token_accuracy": 0.26352272033691404,
      "num_tokens": 170359777.0,
      "step": 74300
    },
    {
      "entropy": 5.541814184188842,
      "epoch": 7.137848222862632,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013650018337590215,
      "loss": 4.6369,
      "mean_token_accuracy": 0.26828493624925615,
      "num_tokens": 170371848.0,
      "step": 74305
    },
    {
      "entropy": 5.597704029083252,
      "epoch": 7.138328530259366,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013647316631486882,
      "loss": 4.6632,
      "mean_token_accuracy": 0.26907008737325666,
      "num_tokens": 170382926.0,
      "step": 74310
    },
    {
      "entropy": 5.56742262840271,
      "epoch": 7.1388088376561,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001364461524700173,
      "loss": 4.706,
      "mean_token_accuracy": 0.25503335148096085,
      "num_tokens": 170394439.0,
      "step": 74315
    },
    {
      "entropy": 5.427092170715332,
      "epoch": 7.1392891450528335,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013641914184197462,
      "loss": 4.4833,
      "mean_token_accuracy": 0.2782762497663498,
      "num_tokens": 170406005.0,
      "step": 74320
    },
    {
      "entropy": 5.530260992050171,
      "epoch": 7.139769452449568,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013639213443136797,
      "loss": 4.6163,
      "mean_token_accuracy": 0.2630910336971283,
      "num_tokens": 170417523.0,
      "step": 74325
    },
    {
      "entropy": 5.608069133758545,
      "epoch": 7.140249759846301,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013636513023882436,
      "loss": 4.6707,
      "mean_token_accuracy": 0.2617097944021225,
      "num_tokens": 170430304.0,
      "step": 74330
    },
    {
      "entropy": 5.6168921947479244,
      "epoch": 7.140730067243036,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001363381292649708,
      "loss": 4.7249,
      "mean_token_accuracy": 0.25587560385465624,
      "num_tokens": 170441779.0,
      "step": 74335
    },
    {
      "entropy": 5.512508249282837,
      "epoch": 7.141210374639769,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001363111315104341,
      "loss": 4.5828,
      "mean_token_accuracy": 0.2720150247216225,
      "num_tokens": 170453569.0,
      "step": 74340
    },
    {
      "entropy": 5.468924522399902,
      "epoch": 7.141690682036503,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013628413697584118,
      "loss": 4.5436,
      "mean_token_accuracy": 0.2722680658102036,
      "num_tokens": 170466433.0,
      "step": 74345
    },
    {
      "entropy": 5.529798412322998,
      "epoch": 7.142170989433238,
      "grad_norm": 1.375,
      "learning_rate": 0.00013625714566181863,
      "loss": 4.645,
      "mean_token_accuracy": 0.2740336075425148,
      "num_tokens": 170478235.0,
      "step": 74350
    },
    {
      "entropy": 5.569037818908692,
      "epoch": 7.142651296829971,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013623015756899321,
      "loss": 4.6963,
      "mean_token_accuracy": 0.2629708513617516,
      "num_tokens": 170490056.0,
      "step": 74355
    },
    {
      "entropy": 5.474824094772339,
      "epoch": 7.1431316042267055,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013620317269799143,
      "loss": 4.5393,
      "mean_token_accuracy": 0.27319301664829254,
      "num_tokens": 170501275.0,
      "step": 74360
    },
    {
      "entropy": 5.514108419418335,
      "epoch": 7.143611911623439,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013617619104943984,
      "loss": 4.6157,
      "mean_token_accuracy": 0.2683121681213379,
      "num_tokens": 170512129.0,
      "step": 74365
    },
    {
      "entropy": 5.52210545539856,
      "epoch": 7.144092219020173,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013614921262396486,
      "loss": 4.6001,
      "mean_token_accuracy": 0.2683114856481552,
      "num_tokens": 170524445.0,
      "step": 74370
    },
    {
      "entropy": 5.483611822128296,
      "epoch": 7.144572526416907,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013612223742219299,
      "loss": 4.5781,
      "mean_token_accuracy": 0.27091933935880663,
      "num_tokens": 170536290.0,
      "step": 74375
    },
    {
      "entropy": 5.488154077529908,
      "epoch": 7.145052833813641,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013609526544475025,
      "loss": 4.5738,
      "mean_token_accuracy": 0.2691407844424248,
      "num_tokens": 170547628.0,
      "step": 74380
    },
    {
      "entropy": 5.54821081161499,
      "epoch": 7.145533141210374,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013606829669226306,
      "loss": 4.6138,
      "mean_token_accuracy": 0.2725048840045929,
      "num_tokens": 170559758.0,
      "step": 74385
    },
    {
      "entropy": 5.490788555145263,
      "epoch": 7.146013448607109,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001360413311653574,
      "loss": 4.5419,
      "mean_token_accuracy": 0.2718944475054741,
      "num_tokens": 170572165.0,
      "step": 74390
    },
    {
      "entropy": 5.579222822189331,
      "epoch": 7.146493756003842,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013601436886465944,
      "loss": 4.6611,
      "mean_token_accuracy": 0.2632534667849541,
      "num_tokens": 170582998.0,
      "step": 74395
    },
    {
      "entropy": 5.657668161392212,
      "epoch": 7.1469740634005765,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013598740979079514,
      "loss": 4.7534,
      "mean_token_accuracy": 0.26007615923881533,
      "num_tokens": 170594147.0,
      "step": 74400
    },
    {
      "entropy": 5.580942583084107,
      "epoch": 7.14745437079731,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013596045394439045,
      "loss": 4.6379,
      "mean_token_accuracy": 0.2716700851917267,
      "num_tokens": 170606033.0,
      "step": 74405
    },
    {
      "entropy": 5.493118476867676,
      "epoch": 7.147934678194044,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001359335013260711,
      "loss": 4.5311,
      "mean_token_accuracy": 0.2806247740983963,
      "num_tokens": 170617997.0,
      "step": 74410
    },
    {
      "entropy": 5.567634773254395,
      "epoch": 7.148414985590778,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001359065519364629,
      "loss": 4.6883,
      "mean_token_accuracy": 0.2593674331903458,
      "num_tokens": 170629767.0,
      "step": 74415
    },
    {
      "entropy": 5.509501791000366,
      "epoch": 7.148895292987512,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013587960577619157,
      "loss": 4.633,
      "mean_token_accuracy": 0.27021795213222505,
      "num_tokens": 170641245.0,
      "step": 74420
    },
    {
      "entropy": 5.564255237579346,
      "epoch": 7.149375600384246,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013585266284588265,
      "loss": 4.6085,
      "mean_token_accuracy": 0.2696210861206055,
      "num_tokens": 170652373.0,
      "step": 74425
    },
    {
      "entropy": 5.5224165439605715,
      "epoch": 7.14985590778098,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013582572314616175,
      "loss": 4.6087,
      "mean_token_accuracy": 0.26784548759460447,
      "num_tokens": 170663437.0,
      "step": 74430
    },
    {
      "entropy": 5.54425048828125,
      "epoch": 7.150336215177714,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013579878667765435,
      "loss": 4.6637,
      "mean_token_accuracy": 0.2641939461231232,
      "num_tokens": 170674422.0,
      "step": 74435
    },
    {
      "entropy": 5.57803783416748,
      "epoch": 7.1508165225744476,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001357718534409857,
      "loss": 4.6482,
      "mean_token_accuracy": 0.2648308351635933,
      "num_tokens": 170687182.0,
      "step": 74440
    },
    {
      "entropy": 5.60358510017395,
      "epoch": 7.151296829971182,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013574492343678123,
      "loss": 4.6362,
      "mean_token_accuracy": 0.2712384179234505,
      "num_tokens": 170697057.0,
      "step": 74445
    },
    {
      "entropy": 5.592615222930908,
      "epoch": 7.151777137367915,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013571799666566608,
      "loss": 4.6684,
      "mean_token_accuracy": 0.26609219461679456,
      "num_tokens": 170707676.0,
      "step": 74450
    },
    {
      "entropy": 5.521679401397705,
      "epoch": 7.15225744476465,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001356910731282655,
      "loss": 4.6146,
      "mean_token_accuracy": 0.26552299708127974,
      "num_tokens": 170718703.0,
      "step": 74455
    },
    {
      "entropy": 5.419411325454712,
      "epoch": 7.152737752161383,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013566415282520453,
      "loss": 4.4681,
      "mean_token_accuracy": 0.2805450588464737,
      "num_tokens": 170730771.0,
      "step": 74460
    },
    {
      "entropy": 5.5669941902160645,
      "epoch": 7.153218059558117,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013563723575710826,
      "loss": 4.6794,
      "mean_token_accuracy": 0.26424430310726166,
      "num_tokens": 170742282.0,
      "step": 74465
    },
    {
      "entropy": 5.464341402053833,
      "epoch": 7.153698366954851,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001356103219246015,
      "loss": 4.5414,
      "mean_token_accuracy": 0.27317859828472135,
      "num_tokens": 170752952.0,
      "step": 74470
    },
    {
      "entropy": 5.589394569396973,
      "epoch": 7.154178674351585,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013558341132830916,
      "loss": 4.6574,
      "mean_token_accuracy": 0.27024552822113035,
      "num_tokens": 170764454.0,
      "step": 74475
    },
    {
      "entropy": 5.6093803405761715,
      "epoch": 7.154658981748319,
      "grad_norm": 1.34375,
      "learning_rate": 0.000135556503968856,
      "loss": 4.7304,
      "mean_token_accuracy": 0.25641033947467806,
      "num_tokens": 170776264.0,
      "step": 74480
    },
    {
      "entropy": 5.578007936477661,
      "epoch": 7.155139289145053,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013552959984686674,
      "loss": 4.6314,
      "mean_token_accuracy": 0.26731411516666415,
      "num_tokens": 170789114.0,
      "step": 74485
    },
    {
      "entropy": 5.462963581085205,
      "epoch": 7.155619596541786,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013550269896296608,
      "loss": 4.5428,
      "mean_token_accuracy": 0.2749914273619652,
      "num_tokens": 170801742.0,
      "step": 74490
    },
    {
      "entropy": 5.529713249206543,
      "epoch": 7.156099903938521,
      "grad_norm": 1.5,
      "learning_rate": 0.0001354758013177786,
      "loss": 4.5796,
      "mean_token_accuracy": 0.26923408955335615,
      "num_tokens": 170813284.0,
      "step": 74495
    },
    {
      "entropy": 5.582540035247803,
      "epoch": 7.156580211335255,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001354489069119286,
      "loss": 4.6666,
      "mean_token_accuracy": 0.2633258804678917,
      "num_tokens": 170823627.0,
      "step": 74500
    },
    {
      "entropy": 5.553247404098511,
      "epoch": 7.1570605187319885,
      "grad_norm": 1.5,
      "learning_rate": 0.00013542201574604063,
      "loss": 4.6698,
      "mean_token_accuracy": 0.2587997689843178,
      "num_tokens": 170834841.0,
      "step": 74505
    },
    {
      "entropy": 5.484581851959229,
      "epoch": 7.157540826128723,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013539512782073898,
      "loss": 4.5298,
      "mean_token_accuracy": 0.2740281894803047,
      "num_tokens": 170847303.0,
      "step": 74510
    },
    {
      "entropy": 5.574166917800904,
      "epoch": 7.158021133525456,
      "grad_norm": 1.5,
      "learning_rate": 0.0001353682431366479,
      "loss": 4.7213,
      "mean_token_accuracy": 0.259699909389019,
      "num_tokens": 170858760.0,
      "step": 74515
    },
    {
      "entropy": 5.551106119155884,
      "epoch": 7.1585014409221905,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013534136169439162,
      "loss": 4.6504,
      "mean_token_accuracy": 0.2586995601654053,
      "num_tokens": 170870304.0,
      "step": 74520
    },
    {
      "entropy": 5.522233343124389,
      "epoch": 7.158981748318924,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001353144834945943,
      "loss": 4.6025,
      "mean_token_accuracy": 0.2674132466316223,
      "num_tokens": 170882958.0,
      "step": 74525
    },
    {
      "entropy": 5.547688102722168,
      "epoch": 7.159462055715658,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001352876085378798,
      "loss": 4.6037,
      "mean_token_accuracy": 0.26616885513067245,
      "num_tokens": 170893730.0,
      "step": 74530
    },
    {
      "entropy": 5.509203195571899,
      "epoch": 7.159942363112392,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013526073682487217,
      "loss": 4.5833,
      "mean_token_accuracy": 0.2707981109619141,
      "num_tokens": 170906121.0,
      "step": 74535
    },
    {
      "entropy": 5.516079425811768,
      "epoch": 7.160422670509126,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001352338683561953,
      "loss": 4.5632,
      "mean_token_accuracy": 0.26775495111942293,
      "num_tokens": 170917113.0,
      "step": 74540
    },
    {
      "entropy": 5.506854581832886,
      "epoch": 7.1609029779058595,
      "grad_norm": 1.453125,
      "learning_rate": 0.000135207003132473,
      "loss": 4.6564,
      "mean_token_accuracy": 0.2663440376520157,
      "num_tokens": 170928820.0,
      "step": 74545
    },
    {
      "entropy": 5.514790296554565,
      "epoch": 7.161383285302594,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013518014115432896,
      "loss": 4.5792,
      "mean_token_accuracy": 0.26761104315519335,
      "num_tokens": 170939182.0,
      "step": 74550
    },
    {
      "entropy": 5.560434627532959,
      "epoch": 7.161863592699327,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013515328242238688,
      "loss": 4.6546,
      "mean_token_accuracy": 0.26518302261829374,
      "num_tokens": 170950373.0,
      "step": 74555
    },
    {
      "entropy": 5.507522392272949,
      "epoch": 7.162343900096062,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013512642693727032,
      "loss": 4.6153,
      "mean_token_accuracy": 0.26548485904932023,
      "num_tokens": 170962626.0,
      "step": 74560
    },
    {
      "entropy": 5.590258264541626,
      "epoch": 7.162824207492795,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001350995746996029,
      "loss": 4.6056,
      "mean_token_accuracy": 0.2765309542417526,
      "num_tokens": 170974580.0,
      "step": 74565
    },
    {
      "entropy": 5.661219310760498,
      "epoch": 7.163304514889529,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001350727257100078,
      "loss": 4.69,
      "mean_token_accuracy": 0.25618603974580767,
      "num_tokens": 170985081.0,
      "step": 74570
    },
    {
      "entropy": 5.552891540527344,
      "epoch": 7.163784822286263,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001350458799691085,
      "loss": 4.6333,
      "mean_token_accuracy": 0.2701585844159126,
      "num_tokens": 170996297.0,
      "step": 74575
    },
    {
      "entropy": 5.52807765007019,
      "epoch": 7.164265129682997,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013501903747752833,
      "loss": 4.5815,
      "mean_token_accuracy": 0.2657694682478905,
      "num_tokens": 171006788.0,
      "step": 74580
    },
    {
      "entropy": 5.51499924659729,
      "epoch": 7.164745437079731,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013499219823589044,
      "loss": 4.5782,
      "mean_token_accuracy": 0.2712701290845871,
      "num_tokens": 171018459.0,
      "step": 74585
    },
    {
      "entropy": 5.4656617641448975,
      "epoch": 7.165225744476465,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013496536224481798,
      "loss": 4.5869,
      "mean_token_accuracy": 0.26871854811906815,
      "num_tokens": 171030854.0,
      "step": 74590
    },
    {
      "entropy": 5.536802482604981,
      "epoch": 7.165706051873199,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013493852950493396,
      "loss": 4.6401,
      "mean_token_accuracy": 0.26658297926187513,
      "num_tokens": 171042925.0,
      "step": 74595
    },
    {
      "entropy": 5.527561092376709,
      "epoch": 7.166186359269933,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013491170001686144,
      "loss": 4.5399,
      "mean_token_accuracy": 0.27644524574279783,
      "num_tokens": 171055248.0,
      "step": 74600
    },
    {
      "entropy": 5.561631679534912,
      "epoch": 7.166666666666667,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001348848737812233,
      "loss": 4.6062,
      "mean_token_accuracy": 0.2693852916359901,
      "num_tokens": 171067196.0,
      "step": 74605
    },
    {
      "entropy": 5.574583435058594,
      "epoch": 7.1671469740634,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013485805079864227,
      "loss": 4.6774,
      "mean_token_accuracy": 0.2606257349252701,
      "num_tokens": 171078227.0,
      "step": 74610
    },
    {
      "entropy": 5.534097623825073,
      "epoch": 7.167627281460135,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013483123106974116,
      "loss": 4.6366,
      "mean_token_accuracy": 0.2650590971112251,
      "num_tokens": 171089916.0,
      "step": 74615
    },
    {
      "entropy": 5.542708158493042,
      "epoch": 7.168107588856868,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013480441459514265,
      "loss": 4.5928,
      "mean_token_accuracy": 0.26632120013237,
      "num_tokens": 171103215.0,
      "step": 74620
    },
    {
      "entropy": 5.5545440196990965,
      "epoch": 7.1685878962536025,
      "grad_norm": 1.5,
      "learning_rate": 0.00013477760137546935,
      "loss": 4.5995,
      "mean_token_accuracy": 0.26841898411512377,
      "num_tokens": 171113408.0,
      "step": 74625
    },
    {
      "entropy": 5.558678531646729,
      "epoch": 7.169068203650336,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013475079141134376,
      "loss": 4.6938,
      "mean_token_accuracy": 0.25889148861169814,
      "num_tokens": 171125584.0,
      "step": 74630
    },
    {
      "entropy": 5.54181318283081,
      "epoch": 7.16954851104707,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013472398470338836,
      "loss": 4.6115,
      "mean_token_accuracy": 0.2648227483034134,
      "num_tokens": 171136574.0,
      "step": 74635
    },
    {
      "entropy": 5.527110576629639,
      "epoch": 7.170028818443804,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001346971812522255,
      "loss": 4.6052,
      "mean_token_accuracy": 0.2771770805120468,
      "num_tokens": 171148136.0,
      "step": 74640
    },
    {
      "entropy": 5.536635398864746,
      "epoch": 7.170509125840538,
      "grad_norm": 1.375,
      "learning_rate": 0.00013467038105847753,
      "loss": 4.6214,
      "mean_token_accuracy": 0.2667711481451988,
      "num_tokens": 171160741.0,
      "step": 74645
    },
    {
      "entropy": 5.519650602340699,
      "epoch": 7.1709894332372714,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013464358412276656,
      "loss": 4.6172,
      "mean_token_accuracy": 0.27034674435853956,
      "num_tokens": 171172117.0,
      "step": 74650
    },
    {
      "entropy": 5.517412853240967,
      "epoch": 7.171469740634006,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001346167904457148,
      "loss": 4.6348,
      "mean_token_accuracy": 0.2675958961248398,
      "num_tokens": 171183235.0,
      "step": 74655
    },
    {
      "entropy": 5.528929710388184,
      "epoch": 7.17195004803074,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001345900000279443,
      "loss": 4.6098,
      "mean_token_accuracy": 0.27092467695474626,
      "num_tokens": 171195742.0,
      "step": 74660
    },
    {
      "entropy": 5.438409757614136,
      "epoch": 7.1724303554274735,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013456321287007707,
      "loss": 4.5176,
      "mean_token_accuracy": 0.2692127704620361,
      "num_tokens": 171207652.0,
      "step": 74665
    },
    {
      "entropy": 5.429952096939087,
      "epoch": 7.172910662824208,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013453642897273498,
      "loss": 4.4726,
      "mean_token_accuracy": 0.2815813809633255,
      "num_tokens": 171218699.0,
      "step": 74670
    },
    {
      "entropy": 5.587338781356811,
      "epoch": 7.173390970220941,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013450964833654005,
      "loss": 4.6876,
      "mean_token_accuracy": 0.2664379641413689,
      "num_tokens": 171230835.0,
      "step": 74675
    },
    {
      "entropy": 5.4450843811035154,
      "epoch": 7.173871277617676,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013448287096211385,
      "loss": 4.528,
      "mean_token_accuracy": 0.2768238440155983,
      "num_tokens": 171242905.0,
      "step": 74680
    },
    {
      "entropy": 5.522226715087891,
      "epoch": 7.174351585014409,
      "grad_norm": 1.375,
      "learning_rate": 0.0001344560968500781,
      "loss": 4.6447,
      "mean_token_accuracy": 0.26650245040655135,
      "num_tokens": 171254480.0,
      "step": 74685
    },
    {
      "entropy": 5.564349603652954,
      "epoch": 7.174831892411143,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013442932600105445,
      "loss": 4.5991,
      "mean_token_accuracy": 0.26587320417165755,
      "num_tokens": 171264685.0,
      "step": 74690
    },
    {
      "entropy": 5.489644765853882,
      "epoch": 7.175312199807877,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013440255841566445,
      "loss": 4.5599,
      "mean_token_accuracy": 0.2704439043998718,
      "num_tokens": 171276837.0,
      "step": 74695
    },
    {
      "entropy": 5.52646918296814,
      "epoch": 7.175792507204611,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013437579409452955,
      "loss": 4.5926,
      "mean_token_accuracy": 0.2650066137313843,
      "num_tokens": 171287698.0,
      "step": 74700
    },
    {
      "entropy": 5.476342058181762,
      "epoch": 7.176272814601345,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013434903303827123,
      "loss": 4.615,
      "mean_token_accuracy": 0.27129998952150347,
      "num_tokens": 171298898.0,
      "step": 74705
    },
    {
      "entropy": 5.498172855377197,
      "epoch": 7.176753121998079,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013432227524751063,
      "loss": 4.5547,
      "mean_token_accuracy": 0.2790578961372375,
      "num_tokens": 171311720.0,
      "step": 74710
    },
    {
      "entropy": 5.4818892002105715,
      "epoch": 7.177233429394812,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001342955207228691,
      "loss": 4.5414,
      "mean_token_accuracy": 0.27432461827993393,
      "num_tokens": 171323689.0,
      "step": 74715
    },
    {
      "entropy": 5.651065731048584,
      "epoch": 7.177713736791547,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013426876946496773,
      "loss": 4.7522,
      "mean_token_accuracy": 0.2584566161036491,
      "num_tokens": 171333698.0,
      "step": 74720
    },
    {
      "entropy": 5.537358379364013,
      "epoch": 7.17819404418828,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001342420214744277,
      "loss": 4.615,
      "mean_token_accuracy": 0.2680540308356285,
      "num_tokens": 171345634.0,
      "step": 74725
    },
    {
      "entropy": 5.501626062393188,
      "epoch": 7.178674351585014,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013421527675186993,
      "loss": 4.6154,
      "mean_token_accuracy": 0.26579149067401886,
      "num_tokens": 171356485.0,
      "step": 74730
    },
    {
      "entropy": 5.560934829711914,
      "epoch": 7.179154658981748,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013418853529791546,
      "loss": 4.6706,
      "mean_token_accuracy": 0.26969567984342574,
      "num_tokens": 171368712.0,
      "step": 74735
    },
    {
      "entropy": 5.4977916240692135,
      "epoch": 7.179634966378482,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013416179711318502,
      "loss": 4.6604,
      "mean_token_accuracy": 0.26647282391786575,
      "num_tokens": 171381466.0,
      "step": 74740
    },
    {
      "entropy": 5.557861995697022,
      "epoch": 7.1801152737752165,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001341350621982994,
      "loss": 4.6081,
      "mean_token_accuracy": 0.26936824023723605,
      "num_tokens": 171393788.0,
      "step": 74745
    },
    {
      "entropy": 5.5362207889556885,
      "epoch": 7.18059558117195,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013410833055387943,
      "loss": 4.623,
      "mean_token_accuracy": 0.2726891696453094,
      "num_tokens": 171405197.0,
      "step": 74750
    },
    {
      "entropy": 5.463277816772461,
      "epoch": 7.181075888568684,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013408160218054561,
      "loss": 4.5349,
      "mean_token_accuracy": 0.27682808339595794,
      "num_tokens": 171415863.0,
      "step": 74755
    },
    {
      "entropy": 5.54933009147644,
      "epoch": 7.181556195965418,
      "grad_norm": 1.578125,
      "learning_rate": 0.00013405487707891857,
      "loss": 4.6103,
      "mean_token_accuracy": 0.2694155603647232,
      "num_tokens": 171427437.0,
      "step": 74760
    },
    {
      "entropy": 5.588200473785401,
      "epoch": 7.182036503362152,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001340281552496188,
      "loss": 4.6887,
      "mean_token_accuracy": 0.26552030593156817,
      "num_tokens": 171439121.0,
      "step": 74765
    },
    {
      "entropy": 5.559072637557984,
      "epoch": 7.1825168107588855,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013400143669326664,
      "loss": 4.6709,
      "mean_token_accuracy": 0.2676408067345619,
      "num_tokens": 171451442.0,
      "step": 74770
    },
    {
      "entropy": 5.570770978927612,
      "epoch": 7.18299711815562,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013397472141048238,
      "loss": 4.6119,
      "mean_token_accuracy": 0.27182871401309966,
      "num_tokens": 171463481.0,
      "step": 74775
    },
    {
      "entropy": 5.522018909454346,
      "epoch": 7.183477425552353,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013394800940188639,
      "loss": 4.6039,
      "mean_token_accuracy": 0.2664687216281891,
      "num_tokens": 171474036.0,
      "step": 74780
    },
    {
      "entropy": 5.446891832351684,
      "epoch": 7.183957732949088,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013392130066809877,
      "loss": 4.5564,
      "mean_token_accuracy": 0.274871788918972,
      "num_tokens": 171485000.0,
      "step": 74785
    },
    {
      "entropy": 5.507531929016113,
      "epoch": 7.184438040345821,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001338945952097396,
      "loss": 4.5475,
      "mean_token_accuracy": 0.2863434225320816,
      "num_tokens": 171496082.0,
      "step": 74790
    },
    {
      "entropy": 5.576309680938721,
      "epoch": 7.184918347742555,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013386789302742908,
      "loss": 4.67,
      "mean_token_accuracy": 0.2690550908446312,
      "num_tokens": 171507668.0,
      "step": 74795
    },
    {
      "entropy": 5.5322428226470945,
      "epoch": 7.185398655139289,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013384119412178686,
      "loss": 4.5815,
      "mean_token_accuracy": 0.2715823531150818,
      "num_tokens": 171519223.0,
      "step": 74800
    },
    {
      "entropy": 5.579528570175171,
      "epoch": 7.185878962536023,
      "grad_norm": 1.375,
      "learning_rate": 0.00013381449849343298,
      "loss": 4.7027,
      "mean_token_accuracy": 0.2627569153904915,
      "num_tokens": 171530701.0,
      "step": 74805
    },
    {
      "entropy": 5.464916944503784,
      "epoch": 7.1863592699327565,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001337878061429872,
      "loss": 4.5637,
      "mean_token_accuracy": 0.2684487044811249,
      "num_tokens": 171541726.0,
      "step": 74810
    },
    {
      "entropy": 5.441355180740357,
      "epoch": 7.186839577329491,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013376111707106924,
      "loss": 4.5784,
      "mean_token_accuracy": 0.27118022441864015,
      "num_tokens": 171552537.0,
      "step": 74815
    },
    {
      "entropy": 5.505514192581177,
      "epoch": 7.187319884726225,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013373443127829876,
      "loss": 4.6435,
      "mean_token_accuracy": 0.26512376219034195,
      "num_tokens": 171564955.0,
      "step": 74820
    },
    {
      "entropy": 5.534683513641357,
      "epoch": 7.187800192122959,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001337077487652954,
      "loss": 4.571,
      "mean_token_accuracy": 0.2784298241138458,
      "num_tokens": 171576122.0,
      "step": 74825
    },
    {
      "entropy": 5.540063762664795,
      "epoch": 7.188280499519693,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013368106953267847,
      "loss": 4.5887,
      "mean_token_accuracy": 0.266245137155056,
      "num_tokens": 171587179.0,
      "step": 74830
    },
    {
      "entropy": 5.58066873550415,
      "epoch": 7.188760806916426,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001336543935810675,
      "loss": 4.6565,
      "mean_token_accuracy": 0.26207460910081865,
      "num_tokens": 171598983.0,
      "step": 74835
    },
    {
      "entropy": 5.612930059432983,
      "epoch": 7.189241114313161,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013362772091108177,
      "loss": 4.7167,
      "mean_token_accuracy": 0.26247469484806063,
      "num_tokens": 171609899.0,
      "step": 74840
    },
    {
      "entropy": 5.48357195854187,
      "epoch": 7.189721421709894,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013360105152334062,
      "loss": 4.5523,
      "mean_token_accuracy": 0.2775997221469879,
      "num_tokens": 171620226.0,
      "step": 74845
    },
    {
      "entropy": 5.559530210494995,
      "epoch": 7.1902017291066285,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013357438541846311,
      "loss": 4.6676,
      "mean_token_accuracy": 0.25395167618989944,
      "num_tokens": 171631877.0,
      "step": 74850
    },
    {
      "entropy": 5.502202558517456,
      "epoch": 7.190682036503362,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013354772259706855,
      "loss": 4.575,
      "mean_token_accuracy": 0.2694039911031723,
      "num_tokens": 171644773.0,
      "step": 74855
    },
    {
      "entropy": 5.528346347808838,
      "epoch": 7.191162343900096,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013352106305977577,
      "loss": 4.6493,
      "mean_token_accuracy": 0.26712394654750826,
      "num_tokens": 171657184.0,
      "step": 74860
    },
    {
      "entropy": 5.481803226470947,
      "epoch": 7.19164265129683,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013349440680720376,
      "loss": 4.5462,
      "mean_token_accuracy": 0.2763221740722656,
      "num_tokens": 171667597.0,
      "step": 74865
    },
    {
      "entropy": 5.578953838348388,
      "epoch": 7.192122958693564,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013346775383997147,
      "loss": 4.6484,
      "mean_token_accuracy": 0.2676400065422058,
      "num_tokens": 171679097.0,
      "step": 74870
    },
    {
      "entropy": 5.575450038909912,
      "epoch": 7.192603266090297,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013344110415869768,
      "loss": 4.6056,
      "mean_token_accuracy": 0.26396534144878386,
      "num_tokens": 171690475.0,
      "step": 74875
    },
    {
      "entropy": 5.601915073394776,
      "epoch": 7.193083573487032,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013341445776400107,
      "loss": 4.6975,
      "mean_token_accuracy": 0.26467924267053605,
      "num_tokens": 171702286.0,
      "step": 74880
    },
    {
      "entropy": 5.4039161682128904,
      "epoch": 7.193563880883765,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013338781465650046,
      "loss": 4.5136,
      "mean_token_accuracy": 0.2812206134200096,
      "num_tokens": 171713957.0,
      "step": 74885
    },
    {
      "entropy": 5.5554533958435055,
      "epoch": 7.1940441882804995,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013336117483681422,
      "loss": 4.6689,
      "mean_token_accuracy": 0.26201873272657394,
      "num_tokens": 171725449.0,
      "step": 74890
    },
    {
      "entropy": 5.5731007099151615,
      "epoch": 7.194524495677234,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013333453830556083,
      "loss": 4.643,
      "mean_token_accuracy": 0.2636016845703125,
      "num_tokens": 171737056.0,
      "step": 74895
    },
    {
      "entropy": 5.549427270889282,
      "epoch": 7.195004803073967,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013330790506335888,
      "loss": 4.6373,
      "mean_token_accuracy": 0.2677264854311943,
      "num_tokens": 171747808.0,
      "step": 74900
    },
    {
      "entropy": 5.507622909545899,
      "epoch": 7.195485110470702,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001332812751108266,
      "loss": 4.5833,
      "mean_token_accuracy": 0.2696107983589172,
      "num_tokens": 171759086.0,
      "step": 74905
    },
    {
      "entropy": 5.488352203369141,
      "epoch": 7.195965417867435,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013325464844858231,
      "loss": 4.6206,
      "mean_token_accuracy": 0.26660939007997514,
      "num_tokens": 171770039.0,
      "step": 74910
    },
    {
      "entropy": 5.54852442741394,
      "epoch": 7.196445725264169,
      "grad_norm": 1.375,
      "learning_rate": 0.00013322802507724423,
      "loss": 4.6235,
      "mean_token_accuracy": 0.2690539970993996,
      "num_tokens": 171780337.0,
      "step": 74915
    },
    {
      "entropy": 5.51436858177185,
      "epoch": 7.196926032660903,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001332014049974304,
      "loss": 4.587,
      "mean_token_accuracy": 0.26598624140024185,
      "num_tokens": 171792180.0,
      "step": 74920
    },
    {
      "entropy": 5.571781778335572,
      "epoch": 7.197406340057637,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013317478820975886,
      "loss": 4.644,
      "mean_token_accuracy": 0.26445821225643157,
      "num_tokens": 171803707.0,
      "step": 74925
    },
    {
      "entropy": 5.483949661254883,
      "epoch": 7.197886647454371,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013314817471484763,
      "loss": 4.5435,
      "mean_token_accuracy": 0.2776844710111618,
      "num_tokens": 171815074.0,
      "step": 74930
    },
    {
      "entropy": 5.476941108703613,
      "epoch": 7.198366954851105,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013312156451331455,
      "loss": 4.5377,
      "mean_token_accuracy": 0.27081696391105653,
      "num_tokens": 171825982.0,
      "step": 74935
    },
    {
      "entropy": 5.562270450592041,
      "epoch": 7.198847262247838,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013309495760577744,
      "loss": 4.6948,
      "mean_token_accuracy": 0.2604784697294235,
      "num_tokens": 171837109.0,
      "step": 74940
    },
    {
      "entropy": 5.613301181793213,
      "epoch": 7.199327569644573,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013306835399285417,
      "loss": 4.6832,
      "mean_token_accuracy": 0.2612863078713417,
      "num_tokens": 171848284.0,
      "step": 74945
    },
    {
      "entropy": 5.49329686164856,
      "epoch": 7.199807877041306,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013304175367516213,
      "loss": 4.5591,
      "mean_token_accuracy": 0.27083341032266617,
      "num_tokens": 171859096.0,
      "step": 74950
    },
    {
      "entropy": 5.497457361221313,
      "epoch": 7.20028818443804,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001330151566533191,
      "loss": 4.5972,
      "mean_token_accuracy": 0.26862486600875857,
      "num_tokens": 171870843.0,
      "step": 74955
    },
    {
      "entropy": 5.563615846633911,
      "epoch": 7.200768491834774,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013298856292794247,
      "loss": 4.6599,
      "mean_token_accuracy": 0.2620293781161308,
      "num_tokens": 171881399.0,
      "step": 74960
    },
    {
      "entropy": 5.51402587890625,
      "epoch": 7.201248799231508,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013296197249964976,
      "loss": 4.5947,
      "mean_token_accuracy": 0.2674530118703842,
      "num_tokens": 171892711.0,
      "step": 74965
    },
    {
      "entropy": 5.532382011413574,
      "epoch": 7.2017291066282425,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013293538536905824,
      "loss": 4.6315,
      "mean_token_accuracy": 0.2590166822075844,
      "num_tokens": 171902850.0,
      "step": 74970
    },
    {
      "entropy": 5.573034381866455,
      "epoch": 7.202209414024976,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013290880153678534,
      "loss": 4.7659,
      "mean_token_accuracy": 0.25908599197864535,
      "num_tokens": 171914989.0,
      "step": 74975
    },
    {
      "entropy": 5.533785009384156,
      "epoch": 7.20268972142171,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013288222100344804,
      "loss": 4.6088,
      "mean_token_accuracy": 0.26473528891801834,
      "num_tokens": 171926660.0,
      "step": 74980
    },
    {
      "entropy": 5.495807456970215,
      "epoch": 7.203170028818444,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013285564376966358,
      "loss": 4.5356,
      "mean_token_accuracy": 0.270518983900547,
      "num_tokens": 171938310.0,
      "step": 74985
    },
    {
      "entropy": 5.644392013549805,
      "epoch": 7.203650336215178,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013282906983604902,
      "loss": 4.7321,
      "mean_token_accuracy": 0.2524718463420868,
      "num_tokens": 171948973.0,
      "step": 74990
    },
    {
      "entropy": 5.559797906875611,
      "epoch": 7.2041306436119115,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013280249920322125,
      "loss": 4.6414,
      "mean_token_accuracy": 0.26115371435880663,
      "num_tokens": 171959976.0,
      "step": 74995
    },
    {
      "entropy": 5.5061239242553714,
      "epoch": 7.204610951008646,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013277593187179727,
      "loss": 4.5741,
      "mean_token_accuracy": 0.27525960505008695,
      "num_tokens": 171972006.0,
      "step": 75000
    },
    {
      "epoch": 7.204610951008646,
      "eval_entropy": 5.254498226360133,
      "eval_loss": 4.744600296020508,
      "eval_mean_token_accuracy": 0.2676159508175774,
      "eval_num_tokens": 171972006.0,
      "eval_runtime": 20.6786,
      "eval_samples_per_second": 1586.909,
      "eval_steps_per_second": 198.37,
      "step": 75000
    },
    {
      "entropy": 5.554872465133667,
      "epoch": 7.205091258405379,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013274936784239382,
      "loss": 4.5958,
      "mean_token_accuracy": 0.2680622383952141,
      "num_tokens": 171984014.0,
      "step": 75005
    },
    {
      "entropy": 5.5901023864746096,
      "epoch": 7.2055715658021136,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001327228071156276,
      "loss": 4.6544,
      "mean_token_accuracy": 0.26694094240665434,
      "num_tokens": 171995384.0,
      "step": 75010
    },
    {
      "entropy": 5.575534343719482,
      "epoch": 7.206051873198847,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001326962496921154,
      "loss": 4.7045,
      "mean_token_accuracy": 0.26499656587839127,
      "num_tokens": 172006387.0,
      "step": 75015
    },
    {
      "entropy": 5.5078558921813965,
      "epoch": 7.206532180595581,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013266969557247367,
      "loss": 4.5764,
      "mean_token_accuracy": 0.2688931480050087,
      "num_tokens": 172018276.0,
      "step": 75020
    },
    {
      "entropy": 5.586534452438355,
      "epoch": 7.207012487992315,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013264314475731898,
      "loss": 4.6076,
      "mean_token_accuracy": 0.27070665657520293,
      "num_tokens": 172028236.0,
      "step": 75025
    },
    {
      "entropy": 5.492084598541259,
      "epoch": 7.207492795389049,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001326165972472677,
      "loss": 4.5247,
      "mean_token_accuracy": 0.2745358407497406,
      "num_tokens": 172038588.0,
      "step": 75030
    },
    {
      "entropy": 5.495838451385498,
      "epoch": 7.2079731027857825,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001325900530429364,
      "loss": 4.6001,
      "mean_token_accuracy": 0.2682051554322243,
      "num_tokens": 172050414.0,
      "step": 75035
    },
    {
      "entropy": 5.569799852371216,
      "epoch": 7.208453410182517,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013256351214494108,
      "loss": 4.6757,
      "mean_token_accuracy": 0.26761086732149125,
      "num_tokens": 172060694.0,
      "step": 75040
    },
    {
      "entropy": 5.518804121017456,
      "epoch": 7.208933717579251,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013253697455389808,
      "loss": 4.5572,
      "mean_token_accuracy": 0.2741913482546806,
      "num_tokens": 172071197.0,
      "step": 75045
    },
    {
      "entropy": 5.602867889404297,
      "epoch": 7.209414024975985,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013251044027042346,
      "loss": 4.6893,
      "mean_token_accuracy": 0.25269905626773836,
      "num_tokens": 172081199.0,
      "step": 75050
    },
    {
      "entropy": 5.470126390457153,
      "epoch": 7.209894332372719,
      "grad_norm": 1.375,
      "learning_rate": 0.00013248390929513338,
      "loss": 4.5371,
      "mean_token_accuracy": 0.27429878413677217,
      "num_tokens": 172092218.0,
      "step": 75055
    },
    {
      "entropy": 5.435133457183838,
      "epoch": 7.210374639769452,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001324573816286437,
      "loss": 4.4955,
      "mean_token_accuracy": 0.2761994287371635,
      "num_tokens": 172102778.0,
      "step": 75060
    },
    {
      "entropy": 5.491444110870361,
      "epoch": 7.210854947166187,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013243085727157039,
      "loss": 4.6164,
      "mean_token_accuracy": 0.26755997389554975,
      "num_tokens": 172114963.0,
      "step": 75065
    },
    {
      "entropy": 5.526160860061646,
      "epoch": 7.21133525456292,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013240433622452923,
      "loss": 4.5662,
      "mean_token_accuracy": 0.2732578873634338,
      "num_tokens": 172126280.0,
      "step": 75070
    },
    {
      "entropy": 5.496904420852661,
      "epoch": 7.2118155619596545,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013237781848813592,
      "loss": 4.5391,
      "mean_token_accuracy": 0.27160650193691255,
      "num_tokens": 172137675.0,
      "step": 75075
    },
    {
      "entropy": 5.6643119812011715,
      "epoch": 7.212295869356388,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013235130406300617,
      "loss": 4.7386,
      "mean_token_accuracy": 0.2619451001286507,
      "num_tokens": 172148799.0,
      "step": 75080
    },
    {
      "entropy": 5.557811689376831,
      "epoch": 7.212776176753122,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013232479294975557,
      "loss": 4.665,
      "mean_token_accuracy": 0.26478424072265627,
      "num_tokens": 172159646.0,
      "step": 75085
    },
    {
      "entropy": 5.5155665397644045,
      "epoch": 7.213256484149856,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001322982851489996,
      "loss": 4.6215,
      "mean_token_accuracy": 0.2675774320960045,
      "num_tokens": 172172005.0,
      "step": 75090
    },
    {
      "entropy": 5.537699127197266,
      "epoch": 7.21373679154659,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013227178066135372,
      "loss": 4.6231,
      "mean_token_accuracy": 0.2719566896557808,
      "num_tokens": 172183022.0,
      "step": 75095
    },
    {
      "entropy": 5.576747226715088,
      "epoch": 7.214217098943323,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013224527948743338,
      "loss": 4.6365,
      "mean_token_accuracy": 0.27134649008512496,
      "num_tokens": 172195235.0,
      "step": 75100
    },
    {
      "entropy": 5.5555259704589846,
      "epoch": 7.214697406340058,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013221878162785361,
      "loss": 4.5798,
      "mean_token_accuracy": 0.2636964201927185,
      "num_tokens": 172206431.0,
      "step": 75105
    },
    {
      "entropy": 5.578658819198608,
      "epoch": 7.215177713736791,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013219228708322983,
      "loss": 4.6727,
      "mean_token_accuracy": 0.26558141112327577,
      "num_tokens": 172217140.0,
      "step": 75110
    },
    {
      "entropy": 5.598310327529907,
      "epoch": 7.2156580211335255,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013216579585417703,
      "loss": 4.6817,
      "mean_token_accuracy": 0.2618088498711586,
      "num_tokens": 172229891.0,
      "step": 75115
    },
    {
      "entropy": 5.523516368865967,
      "epoch": 7.216138328530259,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013213930794131037,
      "loss": 4.6379,
      "mean_token_accuracy": 0.26957767754793166,
      "num_tokens": 172241768.0,
      "step": 75120
    },
    {
      "entropy": 5.548958730697632,
      "epoch": 7.216618635926993,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013211282334524478,
      "loss": 4.6266,
      "mean_token_accuracy": 0.2675631567835808,
      "num_tokens": 172253262.0,
      "step": 75125
    },
    {
      "entropy": 5.662068176269531,
      "epoch": 7.217098943323728,
      "grad_norm": 1.5,
      "learning_rate": 0.00013208634206659508,
      "loss": 4.711,
      "mean_token_accuracy": 0.2566552832722664,
      "num_tokens": 172264870.0,
      "step": 75130
    },
    {
      "entropy": 5.617045783996582,
      "epoch": 7.217579250720461,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001320598641059762,
      "loss": 4.7024,
      "mean_token_accuracy": 0.25807405412197115,
      "num_tokens": 172276117.0,
      "step": 75135
    },
    {
      "entropy": 5.578080940246582,
      "epoch": 7.218059558117195,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013203338946400292,
      "loss": 4.6105,
      "mean_token_accuracy": 0.277401465177536,
      "num_tokens": 172287182.0,
      "step": 75140
    },
    {
      "entropy": 5.521456718444824,
      "epoch": 7.218539865513929,
      "grad_norm": 1.609375,
      "learning_rate": 0.00013200691814128974,
      "loss": 4.5508,
      "mean_token_accuracy": 0.2722635269165039,
      "num_tokens": 172298449.0,
      "step": 75145
    },
    {
      "entropy": 5.54841194152832,
      "epoch": 7.219020172910663,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013198045013845127,
      "loss": 4.661,
      "mean_token_accuracy": 0.2687112808227539,
      "num_tokens": 172308228.0,
      "step": 75150
    },
    {
      "entropy": 5.512120246887207,
      "epoch": 7.2195004803073966,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013195398545610208,
      "loss": 4.5843,
      "mean_token_accuracy": 0.2686687082052231,
      "num_tokens": 172318289.0,
      "step": 75155
    },
    {
      "entropy": 5.494325590133667,
      "epoch": 7.219980787704131,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001319275240948566,
      "loss": 4.5612,
      "mean_token_accuracy": 0.26934049725532533,
      "num_tokens": 172329932.0,
      "step": 75160
    },
    {
      "entropy": 5.526544523239136,
      "epoch": 7.220461095100864,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013190106605532917,
      "loss": 4.5571,
      "mean_token_accuracy": 0.2614936947822571,
      "num_tokens": 172340590.0,
      "step": 75165
    },
    {
      "entropy": 5.512213039398193,
      "epoch": 7.220941402497599,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001318746113381341,
      "loss": 4.572,
      "mean_token_accuracy": 0.26444190442562104,
      "num_tokens": 172351672.0,
      "step": 75170
    },
    {
      "entropy": 5.5444958209991455,
      "epoch": 7.221421709894332,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001318481599438855,
      "loss": 4.644,
      "mean_token_accuracy": 0.26777190864086153,
      "num_tokens": 172363285.0,
      "step": 75175
    },
    {
      "entropy": 5.50796070098877,
      "epoch": 7.221902017291066,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013182171187319763,
      "loss": 4.6138,
      "mean_token_accuracy": 0.26576379388570787,
      "num_tokens": 172375316.0,
      "step": 75180
    },
    {
      "entropy": 5.491513299942016,
      "epoch": 7.2223823246878,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013179526712668444,
      "loss": 4.5746,
      "mean_token_accuracy": 0.2763786897063255,
      "num_tokens": 172385890.0,
      "step": 75185
    },
    {
      "entropy": 5.525764036178589,
      "epoch": 7.222862632084534,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00013176882570495988,
      "loss": 4.5561,
      "mean_token_accuracy": 0.27779840379953386,
      "num_tokens": 172397422.0,
      "step": 75190
    },
    {
      "entropy": 5.558621215820312,
      "epoch": 7.223342939481268,
      "grad_norm": 1.5,
      "learning_rate": 0.00013174238760863788,
      "loss": 4.6313,
      "mean_token_accuracy": 0.2619284108281136,
      "num_tokens": 172408709.0,
      "step": 75195
    },
    {
      "entropy": 5.4823816299438475,
      "epoch": 7.223823246878002,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013171595283833226,
      "loss": 4.5845,
      "mean_token_accuracy": 0.27841341197490693,
      "num_tokens": 172421050.0,
      "step": 75200
    },
    {
      "entropy": 5.502104043960571,
      "epoch": 7.224303554274736,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00013168952139465674,
      "loss": 4.575,
      "mean_token_accuracy": 0.2706716895103455,
      "num_tokens": 172431142.0,
      "step": 75205
    },
    {
      "entropy": 5.527717351913452,
      "epoch": 7.22478386167147,
      "grad_norm": 1.59375,
      "learning_rate": 0.000131663093278225,
      "loss": 4.592,
      "mean_token_accuracy": 0.27420038878917696,
      "num_tokens": 172442211.0,
      "step": 75210
    },
    {
      "entropy": 5.593891429901123,
      "epoch": 7.225264169068204,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013163666848965068,
      "loss": 4.6597,
      "mean_token_accuracy": 0.25676277428865435,
      "num_tokens": 172453452.0,
      "step": 75215
    },
    {
      "entropy": 5.530690336227417,
      "epoch": 7.2257444764649374,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013161024702954715,
      "loss": 4.5752,
      "mean_token_accuracy": 0.27332723736763,
      "num_tokens": 172464557.0,
      "step": 75220
    },
    {
      "entropy": 5.550651502609253,
      "epoch": 7.226224783861672,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001315838288985279,
      "loss": 4.6303,
      "mean_token_accuracy": 0.2676075041294098,
      "num_tokens": 172475881.0,
      "step": 75225
    },
    {
      "entropy": 5.616596984863281,
      "epoch": 7.226705091258405,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001315574140972063,
      "loss": 4.8039,
      "mean_token_accuracy": 0.24781890362501144,
      "num_tokens": 172488623.0,
      "step": 75230
    },
    {
      "entropy": 5.528106212615967,
      "epoch": 7.2271853986551395,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001315310026261956,
      "loss": 4.6341,
      "mean_token_accuracy": 0.26772569864988327,
      "num_tokens": 172500452.0,
      "step": 75235
    },
    {
      "entropy": 5.530389595031738,
      "epoch": 7.227665706051873,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013150459448610897,
      "loss": 4.5338,
      "mean_token_accuracy": 0.27407066226005555,
      "num_tokens": 172512045.0,
      "step": 75240
    },
    {
      "entropy": 5.540942239761352,
      "epoch": 7.228146013448607,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013147818967755963,
      "loss": 4.6119,
      "mean_token_accuracy": 0.2686929330229759,
      "num_tokens": 172523477.0,
      "step": 75245
    },
    {
      "entropy": 5.52734146118164,
      "epoch": 7.228626320845341,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013145178820116052,
      "loss": 4.6602,
      "mean_token_accuracy": 0.2600407376885414,
      "num_tokens": 172534961.0,
      "step": 75250
    },
    {
      "entropy": 5.591213703155518,
      "epoch": 7.229106628242075,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013142539005752462,
      "loss": 4.7197,
      "mean_token_accuracy": 0.2526647537946701,
      "num_tokens": 172546160.0,
      "step": 75255
    },
    {
      "entropy": 5.494704341888427,
      "epoch": 7.2295869356388085,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013139899524726483,
      "loss": 4.5866,
      "mean_token_accuracy": 0.26657048910856246,
      "num_tokens": 172557243.0,
      "step": 75260
    },
    {
      "entropy": 5.507984256744384,
      "epoch": 7.230067243035543,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013137260377099396,
      "loss": 4.5981,
      "mean_token_accuracy": 0.26718044430017474,
      "num_tokens": 172568843.0,
      "step": 75265
    },
    {
      "entropy": 5.596411085128784,
      "epoch": 7.230547550432276,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013134621562932474,
      "loss": 4.7148,
      "mean_token_accuracy": 0.2546523600816727,
      "num_tokens": 172580860.0,
      "step": 75270
    },
    {
      "entropy": 5.588078451156616,
      "epoch": 7.231027857829011,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013131983082286997,
      "loss": 4.7015,
      "mean_token_accuracy": 0.2684033587574959,
      "num_tokens": 172591368.0,
      "step": 75275
    },
    {
      "entropy": 5.531862449645996,
      "epoch": 7.231508165225744,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00013129344935224193,
      "loss": 4.674,
      "mean_token_accuracy": 0.26577567756175996,
      "num_tokens": 172602145.0,
      "step": 75280
    },
    {
      "entropy": 5.561032438278199,
      "epoch": 7.231988472622478,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013126707121805333,
      "loss": 4.6789,
      "mean_token_accuracy": 0.2650584027171135,
      "num_tokens": 172612387.0,
      "step": 75285
    },
    {
      "entropy": 5.606153059005737,
      "epoch": 7.232468780019213,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001312406964209165,
      "loss": 4.598,
      "mean_token_accuracy": 0.2716933205723763,
      "num_tokens": 172624153.0,
      "step": 75290
    },
    {
      "entropy": 5.488755178451538,
      "epoch": 7.232949087415946,
      "grad_norm": 1.375,
      "learning_rate": 0.00013121432496144384,
      "loss": 4.5603,
      "mean_token_accuracy": 0.26896225214004515,
      "num_tokens": 172636116.0,
      "step": 75295
    },
    {
      "entropy": 5.533460807800293,
      "epoch": 7.23342939481268,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013118795684024758,
      "loss": 4.5467,
      "mean_token_accuracy": 0.26511377692222593,
      "num_tokens": 172645892.0,
      "step": 75300
    },
    {
      "entropy": 5.53065710067749,
      "epoch": 7.233909702209414,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013116159205794003,
      "loss": 4.5727,
      "mean_token_accuracy": 0.2726557567715645,
      "num_tokens": 172657487.0,
      "step": 75305
    },
    {
      "entropy": 5.549449110031128,
      "epoch": 7.234390009606148,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013113523061513313,
      "loss": 4.6409,
      "mean_token_accuracy": 0.2673132672905922,
      "num_tokens": 172668188.0,
      "step": 75310
    },
    {
      "entropy": 5.571653604507446,
      "epoch": 7.234870317002882,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013110887251243896,
      "loss": 4.6649,
      "mean_token_accuracy": 0.26520247757434845,
      "num_tokens": 172680113.0,
      "step": 75315
    },
    {
      "entropy": 5.4771185398101805,
      "epoch": 7.235350624399616,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013108251775046948,
      "loss": 4.5414,
      "mean_token_accuracy": 0.26766578406095504,
      "num_tokens": 172692315.0,
      "step": 75320
    },
    {
      "entropy": 5.562510442733765,
      "epoch": 7.235830931796349,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013105616632983662,
      "loss": 4.7232,
      "mean_token_accuracy": 0.25568685233592986,
      "num_tokens": 172705144.0,
      "step": 75325
    },
    {
      "entropy": 5.46970157623291,
      "epoch": 7.236311239193084,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013102981825115217,
      "loss": 4.5954,
      "mean_token_accuracy": 0.27392629235982896,
      "num_tokens": 172716793.0,
      "step": 75330
    },
    {
      "entropy": 5.453766584396362,
      "epoch": 7.236791546589817,
      "grad_norm": 1.375,
      "learning_rate": 0.0001310034735150279,
      "loss": 4.5445,
      "mean_token_accuracy": 0.27038819789886476,
      "num_tokens": 172727542.0,
      "step": 75335
    },
    {
      "entropy": 5.60413556098938,
      "epoch": 7.2372718539865515,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001309771321220753,
      "loss": 4.677,
      "mean_token_accuracy": 0.26169459372758863,
      "num_tokens": 172738053.0,
      "step": 75340
    },
    {
      "entropy": 5.597728443145752,
      "epoch": 7.237752161383285,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013095079407290605,
      "loss": 4.6396,
      "mean_token_accuracy": 0.2609253376722336,
      "num_tokens": 172750022.0,
      "step": 75345
    },
    {
      "entropy": 5.5973752498626705,
      "epoch": 7.238232468780019,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013092445936813162,
      "loss": 4.7009,
      "mean_token_accuracy": 0.262450809776783,
      "num_tokens": 172762433.0,
      "step": 75350
    },
    {
      "entropy": 5.5482653141021725,
      "epoch": 7.238712776176753,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013089812800836344,
      "loss": 4.6375,
      "mean_token_accuracy": 0.2648943826556206,
      "num_tokens": 172774491.0,
      "step": 75355
    },
    {
      "entropy": 5.523399496078492,
      "epoch": 7.239193083573487,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001308717999942128,
      "loss": 4.5739,
      "mean_token_accuracy": 0.2720367297530174,
      "num_tokens": 172784909.0,
      "step": 75360
    },
    {
      "entropy": 5.492518568038941,
      "epoch": 7.239673390970221,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001308454753262911,
      "loss": 4.5539,
      "mean_token_accuracy": 0.28252457082271576,
      "num_tokens": 172796086.0,
      "step": 75365
    },
    {
      "entropy": 5.4044318199157715,
      "epoch": 7.240153698366955,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001308191540052093,
      "loss": 4.5112,
      "mean_token_accuracy": 0.27953178286552427,
      "num_tokens": 172807212.0,
      "step": 75370
    },
    {
      "entropy": 5.519930028915406,
      "epoch": 7.240634005763689,
      "grad_norm": 1.609375,
      "learning_rate": 0.00013079283603157862,
      "loss": 4.6175,
      "mean_token_accuracy": 0.26841750741004944,
      "num_tokens": 172817453.0,
      "step": 75375
    },
    {
      "entropy": 5.512042427062989,
      "epoch": 7.2411143131604225,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013076652140601013,
      "loss": 4.5803,
      "mean_token_accuracy": 0.27147836834192274,
      "num_tokens": 172830156.0,
      "step": 75380
    },
    {
      "entropy": 5.6391044616699215,
      "epoch": 7.241594620557157,
      "grad_norm": 1.5,
      "learning_rate": 0.0001307402101291147,
      "loss": 4.6771,
      "mean_token_accuracy": 0.26187364012002945,
      "num_tokens": 172842185.0,
      "step": 75385
    },
    {
      "entropy": 5.5418437957763675,
      "epoch": 7.24207492795389,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001307139022015032,
      "loss": 4.5752,
      "mean_token_accuracy": 0.26660145074129105,
      "num_tokens": 172854386.0,
      "step": 75390
    },
    {
      "entropy": 5.519415283203125,
      "epoch": 7.242555235350625,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013068759762378655,
      "loss": 4.5717,
      "mean_token_accuracy": 0.27773753106594085,
      "num_tokens": 172865373.0,
      "step": 75395
    },
    {
      "entropy": 5.548176527023315,
      "epoch": 7.243035542747358,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00013066129639657527,
      "loss": 4.6501,
      "mean_token_accuracy": 0.27757813930511477,
      "num_tokens": 172876439.0,
      "step": 75400
    },
    {
      "entropy": 5.532443952560425,
      "epoch": 7.243515850144092,
      "grad_norm": 1.5,
      "learning_rate": 0.0001306349985204801,
      "loss": 4.6101,
      "mean_token_accuracy": 0.26810785084962846,
      "num_tokens": 172887332.0,
      "step": 75405
    },
    {
      "entropy": 5.455933046340943,
      "epoch": 7.243996157540826,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001306087039961116,
      "loss": 4.554,
      "mean_token_accuracy": 0.27001273185014724,
      "num_tokens": 172900667.0,
      "step": 75410
    },
    {
      "entropy": 5.543501234054565,
      "epoch": 7.24447646493756,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001305824128240802,
      "loss": 4.6122,
      "mean_token_accuracy": 0.2669500455260277,
      "num_tokens": 172912614.0,
      "step": 75415
    },
    {
      "entropy": 5.570844888687134,
      "epoch": 7.244956772334294,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013055612500499641,
      "loss": 4.6534,
      "mean_token_accuracy": 0.26714906096458435,
      "num_tokens": 172925888.0,
      "step": 75420
    },
    {
      "entropy": 5.54092435836792,
      "epoch": 7.245437079731028,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013052984053947054,
      "loss": 4.6289,
      "mean_token_accuracy": 0.25912155359983446,
      "num_tokens": 172938273.0,
      "step": 75425
    },
    {
      "entropy": 5.48435492515564,
      "epoch": 7.245917387127761,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013050355942811266,
      "loss": 4.5849,
      "mean_token_accuracy": 0.26523740142583846,
      "num_tokens": 172949965.0,
      "step": 75430
    },
    {
      "entropy": 5.471442222595215,
      "epoch": 7.246397694524496,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013047728167153312,
      "loss": 4.5842,
      "mean_token_accuracy": 0.27298593074083327,
      "num_tokens": 172961568.0,
      "step": 75435
    },
    {
      "entropy": 5.5375062942504885,
      "epoch": 7.24687800192123,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013045100727034192,
      "loss": 4.6322,
      "mean_token_accuracy": 0.2667647212743759,
      "num_tokens": 172974295.0,
      "step": 75440
    },
    {
      "entropy": 5.574265480041504,
      "epoch": 7.247358309317963,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013042473622514912,
      "loss": 4.6919,
      "mean_token_accuracy": 0.2622604086995125,
      "num_tokens": 172986307.0,
      "step": 75445
    },
    {
      "entropy": 5.6019978523254395,
      "epoch": 7.247838616714698,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00013039846853656463,
      "loss": 4.6901,
      "mean_token_accuracy": 0.26535348892211913,
      "num_tokens": 172997268.0,
      "step": 75450
    },
    {
      "entropy": 5.567543411254883,
      "epoch": 7.248318924111431,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001303722042051984,
      "loss": 4.6068,
      "mean_token_accuracy": 0.2758058547973633,
      "num_tokens": 173008951.0,
      "step": 75455
    },
    {
      "entropy": 5.606078386306763,
      "epoch": 7.2487992315081655,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001303459432316601,
      "loss": 4.6398,
      "mean_token_accuracy": 0.2695565477013588,
      "num_tokens": 173020961.0,
      "step": 75460
    },
    {
      "entropy": 5.525909566879273,
      "epoch": 7.249279538904899,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013031968561655943,
      "loss": 4.6004,
      "mean_token_accuracy": 0.27055488228797914,
      "num_tokens": 173032399.0,
      "step": 75465
    },
    {
      "entropy": 5.53682279586792,
      "epoch": 7.249759846301633,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013029343136050607,
      "loss": 4.5691,
      "mean_token_accuracy": 0.2710370302200317,
      "num_tokens": 173043008.0,
      "step": 75470
    },
    {
      "entropy": 5.573597860336304,
      "epoch": 7.250240153698367,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013026718046410952,
      "loss": 4.6865,
      "mean_token_accuracy": 0.26597209721803666,
      "num_tokens": 173053911.0,
      "step": 75475
    },
    {
      "entropy": 5.534584093093872,
      "epoch": 7.250720461095101,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013024093292797928,
      "loss": 4.6981,
      "mean_token_accuracy": 0.2543087124824524,
      "num_tokens": 173065485.0,
      "step": 75480
    },
    {
      "entropy": 5.602439308166504,
      "epoch": 7.2512007684918345,
      "grad_norm": 1.5,
      "learning_rate": 0.0001302146887527248,
      "loss": 4.5999,
      "mean_token_accuracy": 0.26761448234319685,
      "num_tokens": 173076434.0,
      "step": 75485
    },
    {
      "entropy": 5.569420576095581,
      "epoch": 7.251681075888569,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013018844793895524,
      "loss": 4.6301,
      "mean_token_accuracy": 0.26489530354738233,
      "num_tokens": 173087869.0,
      "step": 75490
    },
    {
      "entropy": 5.5836187362670895,
      "epoch": 7.252161383285302,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013016221048727991,
      "loss": 4.6205,
      "mean_token_accuracy": 0.2726381465792656,
      "num_tokens": 173098756.0,
      "step": 75495
    },
    {
      "entropy": 5.612384223937989,
      "epoch": 7.252641690682037,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00013013597639830802,
      "loss": 4.7074,
      "mean_token_accuracy": 0.25992610305547714,
      "num_tokens": 173111168.0,
      "step": 75500
    },
    {
      "entropy": 5.482774162292481,
      "epoch": 7.25312199807877,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013010974567264854,
      "loss": 4.5688,
      "mean_token_accuracy": 0.27434810549020766,
      "num_tokens": 173121982.0,
      "step": 75505
    },
    {
      "entropy": 5.563203907012939,
      "epoch": 7.253602305475504,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001300835183109106,
      "loss": 4.6613,
      "mean_token_accuracy": 0.2660913586616516,
      "num_tokens": 173133810.0,
      "step": 75510
    },
    {
      "entropy": 5.508498430252075,
      "epoch": 7.254082612872239,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001300572943137031,
      "loss": 4.5625,
      "mean_token_accuracy": 0.27608813941478727,
      "num_tokens": 173145441.0,
      "step": 75515
    },
    {
      "entropy": 5.503763341903687,
      "epoch": 7.254562920268972,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013003107368163473,
      "loss": 4.6181,
      "mean_token_accuracy": 0.2711683377623558,
      "num_tokens": 173155914.0,
      "step": 75520
    },
    {
      "entropy": 5.542408037185669,
      "epoch": 7.255043227665706,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013000485641531437,
      "loss": 4.6025,
      "mean_token_accuracy": 0.2631997838616371,
      "num_tokens": 173168415.0,
      "step": 75525
    },
    {
      "entropy": 5.576706266403198,
      "epoch": 7.25552353506244,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012997864251535068,
      "loss": 4.6369,
      "mean_token_accuracy": 0.27162805795669553,
      "num_tokens": 173179554.0,
      "step": 75530
    },
    {
      "entropy": 5.554383325576782,
      "epoch": 7.256003842459174,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012995243198235233,
      "loss": 4.6316,
      "mean_token_accuracy": 0.2707413375377655,
      "num_tokens": 173191922.0,
      "step": 75535
    },
    {
      "entropy": 5.538602304458618,
      "epoch": 7.256484149855908,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012992622481692779,
      "loss": 4.6383,
      "mean_token_accuracy": 0.2624629959464073,
      "num_tokens": 173204358.0,
      "step": 75540
    },
    {
      "entropy": 5.591030120849609,
      "epoch": 7.256964457252642,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012990002101968557,
      "loss": 4.7112,
      "mean_token_accuracy": 0.25487740635871886,
      "num_tokens": 173216703.0,
      "step": 75545
    },
    {
      "entropy": 5.48977427482605,
      "epoch": 7.257444764649375,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012987382059123398,
      "loss": 4.5474,
      "mean_token_accuracy": 0.2807105451822281,
      "num_tokens": 173226704.0,
      "step": 75550
    },
    {
      "entropy": 5.59058313369751,
      "epoch": 7.25792507204611,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012984762353218134,
      "loss": 4.6249,
      "mean_token_accuracy": 0.2711916044354439,
      "num_tokens": 173238048.0,
      "step": 75555
    },
    {
      "entropy": 5.553730773925781,
      "epoch": 7.258405379442843,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012982142984313582,
      "loss": 4.5922,
      "mean_token_accuracy": 0.27057134062051774,
      "num_tokens": 173248504.0,
      "step": 75560
    },
    {
      "entropy": 5.506544685363769,
      "epoch": 7.2588856868395775,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012979523952470564,
      "loss": 4.5868,
      "mean_token_accuracy": 0.2716060936450958,
      "num_tokens": 173259078.0,
      "step": 75565
    },
    {
      "entropy": 5.55144624710083,
      "epoch": 7.259365994236311,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012976905257749882,
      "loss": 4.6676,
      "mean_token_accuracy": 0.27062235921621325,
      "num_tokens": 173270907.0,
      "step": 75570
    },
    {
      "entropy": 5.5246483325958256,
      "epoch": 7.259846301633045,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012974286900212345,
      "loss": 4.6355,
      "mean_token_accuracy": 0.2633200168609619,
      "num_tokens": 173283313.0,
      "step": 75575
    },
    {
      "entropy": 5.520747661590576,
      "epoch": 7.260326609029779,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00012971668879918725,
      "loss": 4.5853,
      "mean_token_accuracy": 0.2717532441020012,
      "num_tokens": 173295182.0,
      "step": 75580
    },
    {
      "entropy": 5.541589546203613,
      "epoch": 7.260806916426513,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001296905119692981,
      "loss": 4.6097,
      "mean_token_accuracy": 0.2644726946949959,
      "num_tokens": 173308392.0,
      "step": 75585
    },
    {
      "entropy": 5.437935543060303,
      "epoch": 7.261287223823247,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001296643385130638,
      "loss": 4.5516,
      "mean_token_accuracy": 0.2703853636980057,
      "num_tokens": 173318448.0,
      "step": 75590
    },
    {
      "entropy": 5.572103214263916,
      "epoch": 7.261767531219981,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012963816843109203,
      "loss": 4.714,
      "mean_token_accuracy": 0.26305299401283266,
      "num_tokens": 173329443.0,
      "step": 75595
    },
    {
      "entropy": 5.503599739074707,
      "epoch": 7.262247838616715,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012961200172399029,
      "loss": 4.5814,
      "mean_token_accuracy": 0.2730345636606216,
      "num_tokens": 173339951.0,
      "step": 75600
    },
    {
      "entropy": 5.482914924621582,
      "epoch": 7.2627281460134485,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012958583839236631,
      "loss": 4.6017,
      "mean_token_accuracy": 0.26814324855804444,
      "num_tokens": 173352679.0,
      "step": 75605
    },
    {
      "entropy": 5.558354473114013,
      "epoch": 7.263208453410183,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00012955967843682726,
      "loss": 4.6285,
      "mean_token_accuracy": 0.262769515812397,
      "num_tokens": 173362786.0,
      "step": 75610
    },
    {
      "entropy": 5.610446357727051,
      "epoch": 7.263688760806916,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012953352185798058,
      "loss": 4.7016,
      "mean_token_accuracy": 0.25493704080581664,
      "num_tokens": 173374982.0,
      "step": 75615
    },
    {
      "entropy": 5.5948902606964115,
      "epoch": 7.264169068203651,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012950736865643362,
      "loss": 4.7041,
      "mean_token_accuracy": 0.2621935620903969,
      "num_tokens": 173387108.0,
      "step": 75620
    },
    {
      "entropy": 5.536550998687744,
      "epoch": 7.264649375600384,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012948121883279352,
      "loss": 4.5869,
      "mean_token_accuracy": 0.2620421588420868,
      "num_tokens": 173399472.0,
      "step": 75625
    },
    {
      "entropy": 5.551709461212158,
      "epoch": 7.265129682997118,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001294550723876674,
      "loss": 4.5925,
      "mean_token_accuracy": 0.2743805006146431,
      "num_tokens": 173411609.0,
      "step": 75630
    },
    {
      "entropy": 5.509026288986206,
      "epoch": 7.265609990393852,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001294289293216624,
      "loss": 4.6008,
      "mean_token_accuracy": 0.2628676533699036,
      "num_tokens": 173424336.0,
      "step": 75635
    },
    {
      "entropy": 5.65768666267395,
      "epoch": 7.266090297790586,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012940278963538538,
      "loss": 4.7085,
      "mean_token_accuracy": 0.2633839905261993,
      "num_tokens": 173435598.0,
      "step": 75640
    },
    {
      "entropy": 5.533663177490235,
      "epoch": 7.26657060518732,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012937665332944317,
      "loss": 4.63,
      "mean_token_accuracy": 0.2625950247049332,
      "num_tokens": 173447416.0,
      "step": 75645
    },
    {
      "entropy": 5.537890958786011,
      "epoch": 7.267050912584054,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001293505204044427,
      "loss": 4.609,
      "mean_token_accuracy": 0.2740715086460114,
      "num_tokens": 173458883.0,
      "step": 75650
    },
    {
      "entropy": 5.546895980834961,
      "epoch": 7.267531219980787,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012932439086099068,
      "loss": 4.6633,
      "mean_token_accuracy": 0.26343909800052645,
      "num_tokens": 173469970.0,
      "step": 75655
    },
    {
      "entropy": 5.507574558258057,
      "epoch": 7.268011527377522,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001292982646996937,
      "loss": 4.5868,
      "mean_token_accuracy": 0.27122215628623964,
      "num_tokens": 173480779.0,
      "step": 75660
    },
    {
      "entropy": 5.51743745803833,
      "epoch": 7.268491834774256,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012927214192115836,
      "loss": 4.591,
      "mean_token_accuracy": 0.27440559715032575,
      "num_tokens": 173492738.0,
      "step": 75665
    },
    {
      "entropy": 5.5324197769165036,
      "epoch": 7.268972142170989,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012924602252599126,
      "loss": 4.5946,
      "mean_token_accuracy": 0.2685330376029015,
      "num_tokens": 173503578.0,
      "step": 75670
    },
    {
      "entropy": 5.508429479598999,
      "epoch": 7.269452449567724,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012921990651479863,
      "loss": 4.5889,
      "mean_token_accuracy": 0.26628764271736144,
      "num_tokens": 173514009.0,
      "step": 75675
    },
    {
      "entropy": 5.494171667098999,
      "epoch": 7.269932756964457,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012919379388818693,
      "loss": 4.5408,
      "mean_token_accuracy": 0.2783857673406601,
      "num_tokens": 173526402.0,
      "step": 75680
    },
    {
      "entropy": 5.59805588722229,
      "epoch": 7.2704130643611915,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012916768464676232,
      "loss": 4.6824,
      "mean_token_accuracy": 0.2608129680156708,
      "num_tokens": 173538235.0,
      "step": 75685
    },
    {
      "entropy": 5.503923940658569,
      "epoch": 7.270893371757925,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012914157879113106,
      "loss": 4.5974,
      "mean_token_accuracy": 0.269058033823967,
      "num_tokens": 173550503.0,
      "step": 75690
    },
    {
      "entropy": 5.533040523529053,
      "epoch": 7.271373679154659,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012911547632189923,
      "loss": 4.5436,
      "mean_token_accuracy": 0.2732329085469246,
      "num_tokens": 173561567.0,
      "step": 75695
    },
    {
      "entropy": 5.554868936538696,
      "epoch": 7.271853986551393,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012908937723967283,
      "loss": 4.6542,
      "mean_token_accuracy": 0.2657729983329773,
      "num_tokens": 173572807.0,
      "step": 75700
    },
    {
      "entropy": 5.48948302268982,
      "epoch": 7.272334293948127,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012906328154505786,
      "loss": 4.6069,
      "mean_token_accuracy": 0.2711788237094879,
      "num_tokens": 173583567.0,
      "step": 75705
    },
    {
      "entropy": 5.4453150749206545,
      "epoch": 7.2728146013448605,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012903718923866026,
      "loss": 4.4895,
      "mean_token_accuracy": 0.27398366928100587,
      "num_tokens": 173594219.0,
      "step": 75710
    },
    {
      "entropy": 5.44126443862915,
      "epoch": 7.273294908741595,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001290111003210856,
      "loss": 4.5348,
      "mean_token_accuracy": 0.27644055485725405,
      "num_tokens": 173604984.0,
      "step": 75715
    },
    {
      "entropy": 5.474061059951782,
      "epoch": 7.273775216138328,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012898501479293963,
      "loss": 4.6516,
      "mean_token_accuracy": 0.26822247356176376,
      "num_tokens": 173616368.0,
      "step": 75720
    },
    {
      "entropy": 5.546813201904297,
      "epoch": 7.2742555235350626,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001289589326548281,
      "loss": 4.6491,
      "mean_token_accuracy": 0.26778415143489837,
      "num_tokens": 173628477.0,
      "step": 75725
    },
    {
      "entropy": 5.532682943344116,
      "epoch": 7.274735830931796,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012893285390735648,
      "loss": 4.6181,
      "mean_token_accuracy": 0.267703877389431,
      "num_tokens": 173640991.0,
      "step": 75730
    },
    {
      "entropy": 5.593256902694702,
      "epoch": 7.27521613832853,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001289067785511303,
      "loss": 4.7212,
      "mean_token_accuracy": 0.26089287996292115,
      "num_tokens": 173653933.0,
      "step": 75735
    },
    {
      "entropy": 5.548925828933716,
      "epoch": 7.275696445725265,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012888070658675487,
      "loss": 4.603,
      "mean_token_accuracy": 0.2634889826178551,
      "num_tokens": 173665561.0,
      "step": 75740
    },
    {
      "entropy": 5.525576543807984,
      "epoch": 7.276176753121998,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001288546380148356,
      "loss": 4.606,
      "mean_token_accuracy": 0.2730468824505806,
      "num_tokens": 173677682.0,
      "step": 75745
    },
    {
      "entropy": 5.5475489616394045,
      "epoch": 7.276657060518732,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001288285728359777,
      "loss": 4.6694,
      "mean_token_accuracy": 0.2729084402322769,
      "num_tokens": 173689214.0,
      "step": 75750
    },
    {
      "entropy": 5.583007335662842,
      "epoch": 7.277137367915466,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001288025110507862,
      "loss": 4.5945,
      "mean_token_accuracy": 0.26669144034385683,
      "num_tokens": 173699277.0,
      "step": 75755
    },
    {
      "entropy": 5.5155596256256105,
      "epoch": 7.2776176753122,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012877645265986637,
      "loss": 4.5849,
      "mean_token_accuracy": 0.2690996274352074,
      "num_tokens": 173712244.0,
      "step": 75760
    },
    {
      "entropy": 5.493745470046997,
      "epoch": 7.278097982708934,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012875039766382305,
      "loss": 4.5927,
      "mean_token_accuracy": 0.27161201685667036,
      "num_tokens": 173723164.0,
      "step": 75765
    },
    {
      "entropy": 5.581380558013916,
      "epoch": 7.278578290105668,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012872434606326123,
      "loss": 4.6731,
      "mean_token_accuracy": 0.26534601449966433,
      "num_tokens": 173734609.0,
      "step": 75770
    },
    {
      "entropy": 5.5804771900177,
      "epoch": 7.279058597502401,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012869829785878574,
      "loss": 4.6479,
      "mean_token_accuracy": 0.26002236604690554,
      "num_tokens": 173746546.0,
      "step": 75775
    },
    {
      "entropy": 5.4679632663726805,
      "epoch": 7.279538904899136,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012867225305100138,
      "loss": 4.5579,
      "mean_token_accuracy": 0.27203588634729386,
      "num_tokens": 173757173.0,
      "step": 75780
    },
    {
      "entropy": 5.550504684448242,
      "epoch": 7.280019212295869,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012864621164051281,
      "loss": 4.668,
      "mean_token_accuracy": 0.25922356694936755,
      "num_tokens": 173767836.0,
      "step": 75785
    },
    {
      "entropy": 5.491434526443482,
      "epoch": 7.2804995196926034,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012862017362792462,
      "loss": 4.612,
      "mean_token_accuracy": 0.2688512414693832,
      "num_tokens": 173779220.0,
      "step": 75790
    },
    {
      "entropy": 5.549924516677857,
      "epoch": 7.280979827089337,
      "grad_norm": 1.375,
      "learning_rate": 0.00012859413901384131,
      "loss": 4.636,
      "mean_token_accuracy": 0.2602373957633972,
      "num_tokens": 173790388.0,
      "step": 75795
    },
    {
      "entropy": 5.476679801940918,
      "epoch": 7.281460134486071,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012856810779886735,
      "loss": 4.5578,
      "mean_token_accuracy": 0.2740861654281616,
      "num_tokens": 173802383.0,
      "step": 75800
    },
    {
      "entropy": 5.6027368068695065,
      "epoch": 7.281940441882805,
      "grad_norm": 1.5,
      "learning_rate": 0.00012854207998360713,
      "loss": 4.5911,
      "mean_token_accuracy": 0.25904066413640975,
      "num_tokens": 173813181.0,
      "step": 75805
    },
    {
      "entropy": 5.551294660568237,
      "epoch": 7.282420749279539,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012851605556866492,
      "loss": 4.5892,
      "mean_token_accuracy": 0.2641820162534714,
      "num_tokens": 173825461.0,
      "step": 75810
    },
    {
      "entropy": 5.433237981796265,
      "epoch": 7.282901056676272,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012849003455464496,
      "loss": 4.5554,
      "mean_token_accuracy": 0.27260631918907163,
      "num_tokens": 173835745.0,
      "step": 75815
    },
    {
      "entropy": 5.543279027938842,
      "epoch": 7.283381364073007,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012846401694215133,
      "loss": 4.6694,
      "mean_token_accuracy": 0.26516512781381607,
      "num_tokens": 173847336.0,
      "step": 75820
    },
    {
      "entropy": 5.543687629699707,
      "epoch": 7.28386167146974,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001284380027317881,
      "loss": 4.6453,
      "mean_token_accuracy": 0.2626883268356323,
      "num_tokens": 173858864.0,
      "step": 75825
    },
    {
      "entropy": 5.4567896842956545,
      "epoch": 7.2843419788664745,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012841199192415925,
      "loss": 4.6,
      "mean_token_accuracy": 0.27246933430433273,
      "num_tokens": 173870867.0,
      "step": 75830
    },
    {
      "entropy": 5.462248754501343,
      "epoch": 7.284822286263209,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012838598451986866,
      "loss": 4.5617,
      "mean_token_accuracy": 0.2689273700118065,
      "num_tokens": 173883537.0,
      "step": 75835
    },
    {
      "entropy": 5.45311541557312,
      "epoch": 7.285302593659942,
      "grad_norm": 1.5,
      "learning_rate": 0.00012835998051952013,
      "loss": 4.5584,
      "mean_token_accuracy": 0.26836226880550385,
      "num_tokens": 173894317.0,
      "step": 75840
    },
    {
      "entropy": 5.4945765972137455,
      "epoch": 7.285782901056677,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00012833397992371743,
      "loss": 4.5534,
      "mean_token_accuracy": 0.27545480728149413,
      "num_tokens": 173906218.0,
      "step": 75845
    },
    {
      "entropy": 5.575456953048706,
      "epoch": 7.28626320845341,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001283079827330643,
      "loss": 4.5934,
      "mean_token_accuracy": 0.26876224726438525,
      "num_tokens": 173917180.0,
      "step": 75850
    },
    {
      "entropy": 5.49236159324646,
      "epoch": 7.286743515850144,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012828198894816416,
      "loss": 4.5945,
      "mean_token_accuracy": 0.2713487669825554,
      "num_tokens": 173930034.0,
      "step": 75855
    },
    {
      "entropy": 5.4764796733856205,
      "epoch": 7.287223823246878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012825599856962057,
      "loss": 4.5613,
      "mean_token_accuracy": 0.26823406368494035,
      "num_tokens": 173940938.0,
      "step": 75860
    },
    {
      "entropy": 5.498720264434814,
      "epoch": 7.287704130643612,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012823001159803694,
      "loss": 4.554,
      "mean_token_accuracy": 0.27039234191179273,
      "num_tokens": 173951099.0,
      "step": 75865
    },
    {
      "entropy": 5.499125719070435,
      "epoch": 7.2881844380403455,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001282040280340166,
      "loss": 4.5855,
      "mean_token_accuracy": 0.2577754735946655,
      "num_tokens": 173962708.0,
      "step": 75870
    },
    {
      "entropy": 5.455015039443969,
      "epoch": 7.28866474543708,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001281780478781629,
      "loss": 4.5785,
      "mean_token_accuracy": 0.27166987806558607,
      "num_tokens": 173975936.0,
      "step": 75875
    },
    {
      "entropy": 5.476438570022583,
      "epoch": 7.289145052833813,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000128152071131079,
      "loss": 4.5843,
      "mean_token_accuracy": 0.2681459248065948,
      "num_tokens": 173987129.0,
      "step": 75880
    },
    {
      "entropy": 5.467551898956299,
      "epoch": 7.289625360230548,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012812609779336783,
      "loss": 4.5571,
      "mean_token_accuracy": 0.2667397201061249,
      "num_tokens": 173998827.0,
      "step": 75885
    },
    {
      "entropy": 5.564424133300781,
      "epoch": 7.290105667627281,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001281001278656326,
      "loss": 4.667,
      "mean_token_accuracy": 0.2668531075119972,
      "num_tokens": 174010488.0,
      "step": 75890
    },
    {
      "entropy": 5.578599691390991,
      "epoch": 7.290585975024015,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001280741613484762,
      "loss": 4.6044,
      "mean_token_accuracy": 0.2693801075220108,
      "num_tokens": 174021702.0,
      "step": 75895
    },
    {
      "entropy": 5.536367416381836,
      "epoch": 7.291066282420749,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001280481982425015,
      "loss": 4.6348,
      "mean_token_accuracy": 0.26225627958774567,
      "num_tokens": 174034190.0,
      "step": 75900
    },
    {
      "entropy": 5.480269098281861,
      "epoch": 7.291546589817483,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012802223854831125,
      "loss": 4.5999,
      "mean_token_accuracy": 0.2670634135603905,
      "num_tokens": 174046916.0,
      "step": 75905
    },
    {
      "entropy": 5.545184755325318,
      "epoch": 7.2920268972142175,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012799628226650828,
      "loss": 4.72,
      "mean_token_accuracy": 0.25999900698661804,
      "num_tokens": 174058236.0,
      "step": 75910
    },
    {
      "entropy": 5.517314672470093,
      "epoch": 7.292507204610951,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012797032939769503,
      "loss": 4.5936,
      "mean_token_accuracy": 0.26832080632448196,
      "num_tokens": 174070101.0,
      "step": 75915
    },
    {
      "entropy": 5.426311254501343,
      "epoch": 7.292987512007685,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012794437994247416,
      "loss": 4.5723,
      "mean_token_accuracy": 0.2713765114545822,
      "num_tokens": 174081880.0,
      "step": 75920
    },
    {
      "entropy": 5.580863666534424,
      "epoch": 7.293467819404419,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012791843390144814,
      "loss": 4.6636,
      "mean_token_accuracy": 0.26901928186416624,
      "num_tokens": 174092610.0,
      "step": 75925
    },
    {
      "entropy": 5.5621541976928714,
      "epoch": 7.293948126801153,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001278924912752193,
      "loss": 4.6312,
      "mean_token_accuracy": 0.27244449555873873,
      "num_tokens": 174103206.0,
      "step": 75930
    },
    {
      "entropy": 5.609616708755493,
      "epoch": 7.2944284341978864,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012786655206439002,
      "loss": 4.6695,
      "mean_token_accuracy": 0.26560672372579575,
      "num_tokens": 174113667.0,
      "step": 75935
    },
    {
      "entropy": 5.534660148620605,
      "epoch": 7.294908741594621,
      "grad_norm": 1.375,
      "learning_rate": 0.00012784061626956256,
      "loss": 4.6423,
      "mean_token_accuracy": 0.2677114546298981,
      "num_tokens": 174127038.0,
      "step": 75940
    },
    {
      "entropy": 5.551913404464722,
      "epoch": 7.295389048991354,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00012781468389133892,
      "loss": 4.6037,
      "mean_token_accuracy": 0.2699416846036911,
      "num_tokens": 174138109.0,
      "step": 75945
    },
    {
      "entropy": 5.598496818542481,
      "epoch": 7.2958693563880885,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012778875493032132,
      "loss": 4.6544,
      "mean_token_accuracy": 0.2601450592279434,
      "num_tokens": 174149935.0,
      "step": 75950
    },
    {
      "entropy": 5.551746368408203,
      "epoch": 7.296349663784822,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001277628293871117,
      "loss": 4.675,
      "mean_token_accuracy": 0.2611996173858643,
      "num_tokens": 174162957.0,
      "step": 75955
    },
    {
      "entropy": 5.619849300384521,
      "epoch": 7.296829971181556,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012773690726231197,
      "loss": 4.6939,
      "mean_token_accuracy": 0.25752335637807844,
      "num_tokens": 174174284.0,
      "step": 75960
    },
    {
      "entropy": 5.490240812301636,
      "epoch": 7.29731027857829,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012771098855652398,
      "loss": 4.5947,
      "mean_token_accuracy": 0.27059673368930814,
      "num_tokens": 174187230.0,
      "step": 75965
    },
    {
      "entropy": 5.532300043106079,
      "epoch": 7.297790585975024,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012768507327034953,
      "loss": 4.567,
      "mean_token_accuracy": 0.2762021511793137,
      "num_tokens": 174197731.0,
      "step": 75970
    },
    {
      "entropy": 5.515860557556152,
      "epoch": 7.2982708933717575,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012765916140439017,
      "loss": 4.6058,
      "mean_token_accuracy": 0.26673107743263247,
      "num_tokens": 174209847.0,
      "step": 75975
    },
    {
      "entropy": 5.583535432815552,
      "epoch": 7.298751200768492,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012763325295924759,
      "loss": 4.66,
      "mean_token_accuracy": 0.2624252140522003,
      "num_tokens": 174221133.0,
      "step": 75980
    },
    {
      "entropy": 5.586118078231811,
      "epoch": 7.299231508165226,
      "grad_norm": 1.703125,
      "learning_rate": 0.00012760734793552326,
      "loss": 4.6826,
      "mean_token_accuracy": 0.25650039315223694,
      "num_tokens": 174232125.0,
      "step": 75985
    },
    {
      "entropy": 5.50619478225708,
      "epoch": 7.29971181556196,
      "grad_norm": 1.578125,
      "learning_rate": 0.00012758144633381868,
      "loss": 4.5789,
      "mean_token_accuracy": 0.26775696128606796,
      "num_tokens": 174242721.0,
      "step": 75990
    },
    {
      "entropy": 5.583796644210816,
      "epoch": 7.300192122958694,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012755554815473513,
      "loss": 4.6448,
      "mean_token_accuracy": 0.2641008958220482,
      "num_tokens": 174254266.0,
      "step": 75995
    },
    {
      "entropy": 5.640284299850464,
      "epoch": 7.300672430355427,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012752965339887402,
      "loss": 4.7298,
      "mean_token_accuracy": 0.2636964187026024,
      "num_tokens": 174265420.0,
      "step": 76000
    },
    {
      "entropy": 5.629212999343872,
      "epoch": 7.301152737752162,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001275037620668364,
      "loss": 4.6979,
      "mean_token_accuracy": 0.266973452270031,
      "num_tokens": 174275118.0,
      "step": 76005
    },
    {
      "entropy": 5.5767569065094,
      "epoch": 7.301633045148895,
      "grad_norm": 1.375,
      "learning_rate": 0.00012747787415922344,
      "loss": 4.6581,
      "mean_token_accuracy": 0.27159487307071684,
      "num_tokens": 174286341.0,
      "step": 76010
    },
    {
      "entropy": 5.543702793121338,
      "epoch": 7.302113352545629,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001274519896766362,
      "loss": 4.6181,
      "mean_token_accuracy": 0.2566261038184166,
      "num_tokens": 174297761.0,
      "step": 76015
    },
    {
      "entropy": 5.508248996734619,
      "epoch": 7.302593659942363,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012742610861967563,
      "loss": 4.6224,
      "mean_token_accuracy": 0.2646869316697121,
      "num_tokens": 174309967.0,
      "step": 76020
    },
    {
      "entropy": 5.486123085021973,
      "epoch": 7.303073967339097,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001274002309889426,
      "loss": 4.6025,
      "mean_token_accuracy": 0.2731409728527069,
      "num_tokens": 174320991.0,
      "step": 76025
    },
    {
      "entropy": 5.54752287864685,
      "epoch": 7.303554274735831,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012737435678503801,
      "loss": 4.6148,
      "mean_token_accuracy": 0.26751789152622224,
      "num_tokens": 174331675.0,
      "step": 76030
    },
    {
      "entropy": 5.495284366607666,
      "epoch": 7.304034582132565,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012734848600856242,
      "loss": 4.588,
      "mean_token_accuracy": 0.26724617183208466,
      "num_tokens": 174343290.0,
      "step": 76035
    },
    {
      "entropy": 5.494834232330322,
      "epoch": 7.304514889529298,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012732261866011657,
      "loss": 4.5882,
      "mean_token_accuracy": 0.27252885401248933,
      "num_tokens": 174354645.0,
      "step": 76040
    },
    {
      "entropy": 5.49804196357727,
      "epoch": 7.304995196926033,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012729675474030101,
      "loss": 4.6031,
      "mean_token_accuracy": 0.27021496891975405,
      "num_tokens": 174366567.0,
      "step": 76045
    },
    {
      "entropy": 5.566899108886719,
      "epoch": 7.305475504322766,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012727089424971615,
      "loss": 4.6681,
      "mean_token_accuracy": 0.263342010974884,
      "num_tokens": 174378526.0,
      "step": 76050
    },
    {
      "entropy": 5.550774478912354,
      "epoch": 7.3059558117195005,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012724503718896253,
      "loss": 4.6845,
      "mean_token_accuracy": 0.26012512296438217,
      "num_tokens": 174389904.0,
      "step": 76055
    },
    {
      "entropy": 5.6376721382141115,
      "epoch": 7.306436119116235,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012721918355864046,
      "loss": 4.6909,
      "mean_token_accuracy": 0.2623948141932487,
      "num_tokens": 174400225.0,
      "step": 76060
    },
    {
      "entropy": 5.692605257034302,
      "epoch": 7.306916426512968,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012719333335935005,
      "loss": 4.7635,
      "mean_token_accuracy": 0.25877188742160795,
      "num_tokens": 174411788.0,
      "step": 76065
    },
    {
      "entropy": 5.46589207649231,
      "epoch": 7.307396733909703,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012716748659169152,
      "loss": 4.5692,
      "mean_token_accuracy": 0.2733179584145546,
      "num_tokens": 174423861.0,
      "step": 76070
    },
    {
      "entropy": 5.489527797698974,
      "epoch": 7.307877041306436,
      "grad_norm": 1.390625,
      "learning_rate": 0.000127141643256265,
      "loss": 4.5573,
      "mean_token_accuracy": 0.2794171556830406,
      "num_tokens": 174434771.0,
      "step": 76075
    },
    {
      "entropy": 5.549934196472168,
      "epoch": 7.30835734870317,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012711580335367052,
      "loss": 4.6024,
      "mean_token_accuracy": 0.25927418321371076,
      "num_tokens": 174445654.0,
      "step": 76080
    },
    {
      "entropy": 5.475153112411499,
      "epoch": 7.308837656099904,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012708996688450786,
      "loss": 4.5569,
      "mean_token_accuracy": 0.2718913435935974,
      "num_tokens": 174456483.0,
      "step": 76085
    },
    {
      "entropy": 5.4969494342803955,
      "epoch": 7.309317963496638,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012706413384937709,
      "loss": 4.5776,
      "mean_token_accuracy": 0.2705193907022476,
      "num_tokens": 174467617.0,
      "step": 76090
    },
    {
      "entropy": 5.486099481582642,
      "epoch": 7.3097982708933715,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001270383042488778,
      "loss": 4.6006,
      "mean_token_accuracy": 0.26551152765750885,
      "num_tokens": 174477777.0,
      "step": 76095
    },
    {
      "entropy": 5.500007677078247,
      "epoch": 7.310278578290106,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012701247808360972,
      "loss": 4.614,
      "mean_token_accuracy": 0.2645556956529617,
      "num_tokens": 174488090.0,
      "step": 76100
    },
    {
      "entropy": 5.5642565250396725,
      "epoch": 7.310758885686839,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012698665535417248,
      "loss": 4.6543,
      "mean_token_accuracy": 0.2671560123562813,
      "num_tokens": 174499968.0,
      "step": 76105
    },
    {
      "entropy": 5.4434422016143795,
      "epoch": 7.311239193083574,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012696083606116557,
      "loss": 4.5582,
      "mean_token_accuracy": 0.27657311260700224,
      "num_tokens": 174511068.0,
      "step": 76110
    },
    {
      "entropy": 5.511576175689697,
      "epoch": 7.311719500480307,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012693502020518844,
      "loss": 4.5747,
      "mean_token_accuracy": 0.2654511511325836,
      "num_tokens": 174522103.0,
      "step": 76115
    },
    {
      "entropy": 5.565602827072143,
      "epoch": 7.312199807877041,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001269092077868406,
      "loss": 4.6012,
      "mean_token_accuracy": 0.2663873568177223,
      "num_tokens": 174534246.0,
      "step": 76120
    },
    {
      "entropy": 5.4818767547607425,
      "epoch": 7.312680115273775,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012688339880672112,
      "loss": 4.5893,
      "mean_token_accuracy": 0.2770807534456253,
      "num_tokens": 174544713.0,
      "step": 76125
    },
    {
      "entropy": 5.619908142089844,
      "epoch": 7.313160422670509,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001268575932654293,
      "loss": 4.7301,
      "mean_token_accuracy": 0.2583503484725952,
      "num_tokens": 174556942.0,
      "step": 76130
    },
    {
      "entropy": 5.533699321746826,
      "epoch": 7.3136407300672435,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012683179116356426,
      "loss": 4.5981,
      "mean_token_accuracy": 0.27166115045547484,
      "num_tokens": 174567882.0,
      "step": 76135
    },
    {
      "entropy": 5.432317733764648,
      "epoch": 7.314121037463977,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012680599250172508,
      "loss": 4.5297,
      "mean_token_accuracy": 0.27492102831602094,
      "num_tokens": 174578410.0,
      "step": 76140
    },
    {
      "entropy": 5.5120687007904055,
      "epoch": 7.314601344860711,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001267801972805107,
      "loss": 4.5826,
      "mean_token_accuracy": 0.2737463817000389,
      "num_tokens": 174588676.0,
      "step": 76145
    },
    {
      "entropy": 5.53970308303833,
      "epoch": 7.315081652257445,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001267544055005201,
      "loss": 4.6136,
      "mean_token_accuracy": 0.2677567511796951,
      "num_tokens": 174599929.0,
      "step": 76150
    },
    {
      "entropy": 5.500543737411499,
      "epoch": 7.315561959654179,
      "grad_norm": 1.5,
      "learning_rate": 0.0001267286171623519,
      "loss": 4.6129,
      "mean_token_accuracy": 0.26760795563459394,
      "num_tokens": 174610109.0,
      "step": 76155
    },
    {
      "entropy": 5.5081346988677975,
      "epoch": 7.316042267050912,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012670283226660497,
      "loss": 4.6044,
      "mean_token_accuracy": 0.26762391030788424,
      "num_tokens": 174620502.0,
      "step": 76160
    },
    {
      "entropy": 5.506991958618164,
      "epoch": 7.316522574447647,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012667705081387793,
      "loss": 4.6088,
      "mean_token_accuracy": 0.263354717195034,
      "num_tokens": 174632122.0,
      "step": 76165
    },
    {
      "entropy": 5.524147939682007,
      "epoch": 7.31700288184438,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001266512728047693,
      "loss": 4.5606,
      "mean_token_accuracy": 0.26757015883922575,
      "num_tokens": 174643067.0,
      "step": 76170
    },
    {
      "entropy": 5.543981695175171,
      "epoch": 7.3174831892411145,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012662549823987766,
      "loss": 4.7077,
      "mean_token_accuracy": 0.256437686085701,
      "num_tokens": 174654978.0,
      "step": 76175
    },
    {
      "entropy": 5.5310407161712645,
      "epoch": 7.317963496637848,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012659972711980138,
      "loss": 4.6705,
      "mean_token_accuracy": 0.2643521085381508,
      "num_tokens": 174666477.0,
      "step": 76180
    },
    {
      "entropy": 5.589647102355957,
      "epoch": 7.318443804034582,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012657395944513878,
      "loss": 4.727,
      "mean_token_accuracy": 0.2596920907497406,
      "num_tokens": 174679524.0,
      "step": 76185
    },
    {
      "entropy": 5.576356983184814,
      "epoch": 7.318924111431316,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012654819521648803,
      "loss": 4.6326,
      "mean_token_accuracy": 0.26473285257816315,
      "num_tokens": 174690583.0,
      "step": 76190
    },
    {
      "entropy": 5.618145418167114,
      "epoch": 7.31940441882805,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012652243443444743,
      "loss": 4.7103,
      "mean_token_accuracy": 0.25924838036298753,
      "num_tokens": 174700971.0,
      "step": 76195
    },
    {
      "entropy": 5.430175399780273,
      "epoch": 7.3198847262247835,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012649667709961499,
      "loss": 4.4722,
      "mean_token_accuracy": 0.2795081675052643,
      "num_tokens": 174711226.0,
      "step": 76200
    },
    {
      "entropy": 5.438708019256592,
      "epoch": 7.320365033621518,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00012647092321258873,
      "loss": 4.5315,
      "mean_token_accuracy": 0.27170724868774415,
      "num_tokens": 174721744.0,
      "step": 76205
    },
    {
      "entropy": 5.5835339546203615,
      "epoch": 7.320845341018252,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012644517277396672,
      "loss": 4.6768,
      "mean_token_accuracy": 0.26396760195493696,
      "num_tokens": 174735181.0,
      "step": 76210
    },
    {
      "entropy": 5.624518632888794,
      "epoch": 7.321325648414986,
      "grad_norm": 1.375,
      "learning_rate": 0.00012641942578434657,
      "loss": 4.7152,
      "mean_token_accuracy": 0.2538110092282295,
      "num_tokens": 174747831.0,
      "step": 76215
    },
    {
      "entropy": 5.584916496276856,
      "epoch": 7.32180595581172,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012639368224432615,
      "loss": 4.6267,
      "mean_token_accuracy": 0.2669665053486824,
      "num_tokens": 174758858.0,
      "step": 76220
    },
    {
      "entropy": 5.511748361587524,
      "epoch": 7.322286263208453,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001263679421545032,
      "loss": 4.5496,
      "mean_token_accuracy": 0.2739315778017044,
      "num_tokens": 174770196.0,
      "step": 76225
    },
    {
      "entropy": 5.487383604049683,
      "epoch": 7.322766570605188,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012634220551547526,
      "loss": 4.6141,
      "mean_token_accuracy": 0.26895503252744674,
      "num_tokens": 174782795.0,
      "step": 76230
    },
    {
      "entropy": 5.5344886302948,
      "epoch": 7.323246878001921,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012631647232783986,
      "loss": 4.6362,
      "mean_token_accuracy": 0.2670832112431526,
      "num_tokens": 174794613.0,
      "step": 76235
    },
    {
      "entropy": 5.5680365562438965,
      "epoch": 7.323727185398655,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012629074259219454,
      "loss": 4.6566,
      "mean_token_accuracy": 0.26350370049476624,
      "num_tokens": 174806380.0,
      "step": 76240
    },
    {
      "entropy": 5.519626522064209,
      "epoch": 7.324207492795389,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001262650163091366,
      "loss": 4.5797,
      "mean_token_accuracy": 0.26953449547290803,
      "num_tokens": 174817130.0,
      "step": 76245
    },
    {
      "entropy": 5.542593717575073,
      "epoch": 7.324687800192123,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012623929347926327,
      "loss": 4.6165,
      "mean_token_accuracy": 0.2714542135596275,
      "num_tokens": 174829147.0,
      "step": 76250
    },
    {
      "entropy": 5.497519159317017,
      "epoch": 7.325168107588857,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012621357410317187,
      "loss": 4.608,
      "mean_token_accuracy": 0.26790174245834353,
      "num_tokens": 174841984.0,
      "step": 76255
    },
    {
      "entropy": 5.489471673965454,
      "epoch": 7.325648414985591,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012618785818145942,
      "loss": 4.6096,
      "mean_token_accuracy": 0.26836974173784256,
      "num_tokens": 174853843.0,
      "step": 76260
    },
    {
      "entropy": 5.598964023590088,
      "epoch": 7.326128722382324,
      "grad_norm": 1.609375,
      "learning_rate": 0.000126162145714723,
      "loss": 4.656,
      "mean_token_accuracy": 0.2695327877998352,
      "num_tokens": 174863442.0,
      "step": 76265
    },
    {
      "entropy": 5.538924789428711,
      "epoch": 7.326609029779059,
      "grad_norm": 1.578125,
      "learning_rate": 0.00012613643670355968,
      "loss": 4.6924,
      "mean_token_accuracy": 0.26177300959825517,
      "num_tokens": 174874393.0,
      "step": 76270
    },
    {
      "entropy": 5.564909601211548,
      "epoch": 7.327089337175792,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001261107311485662,
      "loss": 4.6615,
      "mean_token_accuracy": 0.26993915140628816,
      "num_tokens": 174885431.0,
      "step": 76275
    },
    {
      "entropy": 5.4533977031707765,
      "epoch": 7.3275696445725265,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012608502905033948,
      "loss": 4.5275,
      "mean_token_accuracy": 0.2748842969536781,
      "num_tokens": 174897194.0,
      "step": 76280
    },
    {
      "entropy": 5.538880491256714,
      "epoch": 7.328049951969261,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012605933040947626,
      "loss": 4.5793,
      "mean_token_accuracy": 0.2670562297105789,
      "num_tokens": 174909347.0,
      "step": 76285
    },
    {
      "entropy": 5.506159019470215,
      "epoch": 7.328530259365994,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000126033635226573,
      "loss": 4.5696,
      "mean_token_accuracy": 0.27459904849529265,
      "num_tokens": 174919562.0,
      "step": 76290
    },
    {
      "entropy": 5.478295087814331,
      "epoch": 7.3290105667627286,
      "grad_norm": 1.375,
      "learning_rate": 0.00012600794350222646,
      "loss": 4.6113,
      "mean_token_accuracy": 0.2639650821685791,
      "num_tokens": 174931720.0,
      "step": 76295
    },
    {
      "entropy": 5.45707688331604,
      "epoch": 7.329490874159462,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012598225523703305,
      "loss": 4.5306,
      "mean_token_accuracy": 0.2719028443098068,
      "num_tokens": 174942163.0,
      "step": 76300
    },
    {
      "entropy": 5.537959861755371,
      "epoch": 7.329971181556196,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012595657043158916,
      "loss": 4.5895,
      "mean_token_accuracy": 0.27315675020217894,
      "num_tokens": 174952352.0,
      "step": 76305
    },
    {
      "entropy": 5.605618906021118,
      "epoch": 7.33045148895293,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012593088908649112,
      "loss": 4.7542,
      "mean_token_accuracy": 0.2612104579806328,
      "num_tokens": 174964427.0,
      "step": 76310
    },
    {
      "entropy": 5.5084480285644535,
      "epoch": 7.330931796349664,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012590521120233518,
      "loss": 4.627,
      "mean_token_accuracy": 0.2691408395767212,
      "num_tokens": 174975726.0,
      "step": 76315
    },
    {
      "entropy": 5.5648294448852536,
      "epoch": 7.3314121037463975,
      "grad_norm": 1.703125,
      "learning_rate": 0.0001258795367797176,
      "loss": 4.6556,
      "mean_token_accuracy": 0.2636221915483475,
      "num_tokens": 174987003.0,
      "step": 76320
    },
    {
      "entropy": 5.558211088180542,
      "epoch": 7.331892411143132,
      "grad_norm": 1.625,
      "learning_rate": 0.00012585386581923436,
      "loss": 4.6262,
      "mean_token_accuracy": 0.2647143378853798,
      "num_tokens": 174998333.0,
      "step": 76325
    },
    {
      "entropy": 5.52296781539917,
      "epoch": 7.332372718539865,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012582819832148145,
      "loss": 4.606,
      "mean_token_accuracy": 0.26604788154363634,
      "num_tokens": 175009911.0,
      "step": 76330
    },
    {
      "entropy": 5.512455940246582,
      "epoch": 7.3328530259366,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001258025342870548,
      "loss": 4.6135,
      "mean_token_accuracy": 0.268030546605587,
      "num_tokens": 175021435.0,
      "step": 76335
    },
    {
      "entropy": 5.576320648193359,
      "epoch": 7.333333333333333,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012577687371655033,
      "loss": 4.6054,
      "mean_token_accuracy": 0.2686627462506294,
      "num_tokens": 175031996.0,
      "step": 76340
    },
    {
      "entropy": 5.57458086013794,
      "epoch": 7.333813640730067,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012575121661056372,
      "loss": 4.6816,
      "mean_token_accuracy": 0.26157149076461794,
      "num_tokens": 175042945.0,
      "step": 76345
    },
    {
      "entropy": 5.542572641372681,
      "epoch": 7.334293948126801,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012572556296969067,
      "loss": 4.6998,
      "mean_token_accuracy": 0.2608522802591324,
      "num_tokens": 175055443.0,
      "step": 76350
    },
    {
      "entropy": 5.513788414001465,
      "epoch": 7.334774255523535,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012569991279452682,
      "loss": 4.5815,
      "mean_token_accuracy": 0.2658440977334976,
      "num_tokens": 175066863.0,
      "step": 76355
    },
    {
      "entropy": 5.52627649307251,
      "epoch": 7.335254562920269,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001256742660856677,
      "loss": 4.6243,
      "mean_token_accuracy": 0.26865973621606826,
      "num_tokens": 175077027.0,
      "step": 76360
    },
    {
      "entropy": 5.509033107757569,
      "epoch": 7.335734870317003,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012564862284370867,
      "loss": 4.662,
      "mean_token_accuracy": 0.2584828197956085,
      "num_tokens": 175087725.0,
      "step": 76365
    },
    {
      "entropy": 5.527083253860473,
      "epoch": 7.336215177713736,
      "grad_norm": 1.5,
      "learning_rate": 0.00012562298306924513,
      "loss": 4.6209,
      "mean_token_accuracy": 0.26583455204963685,
      "num_tokens": 175098896.0,
      "step": 76370
    },
    {
      "entropy": 5.496304893493653,
      "epoch": 7.336695485110471,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012559734676287237,
      "loss": 4.6005,
      "mean_token_accuracy": 0.2805350810289383,
      "num_tokens": 175108491.0,
      "step": 76375
    },
    {
      "entropy": 5.549818420410157,
      "epoch": 7.337175792507205,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012557171392518557,
      "loss": 4.6233,
      "mean_token_accuracy": 0.26666012555360796,
      "num_tokens": 175118378.0,
      "step": 76380
    },
    {
      "entropy": 5.577137565612793,
      "epoch": 7.337656099903938,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001255460845567799,
      "loss": 4.7038,
      "mean_token_accuracy": 0.26072280257940295,
      "num_tokens": 175130282.0,
      "step": 76385
    },
    {
      "entropy": 5.524727821350098,
      "epoch": 7.338136407300673,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001255204586582504,
      "loss": 4.5855,
      "mean_token_accuracy": 0.26699379831552505,
      "num_tokens": 175141330.0,
      "step": 76390
    },
    {
      "entropy": 5.471242952346802,
      "epoch": 7.338616714697406,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012549483623019192,
      "loss": 4.5048,
      "mean_token_accuracy": 0.2764063239097595,
      "num_tokens": 175153037.0,
      "step": 76395
    },
    {
      "entropy": 5.461909341812134,
      "epoch": 7.3390970220941405,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001254692172731994,
      "loss": 4.5257,
      "mean_token_accuracy": 0.27616270929574965,
      "num_tokens": 175164534.0,
      "step": 76400
    },
    {
      "entropy": 5.525546646118164,
      "epoch": 7.339577329490874,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012544360178786762,
      "loss": 4.6288,
      "mean_token_accuracy": 0.26860991716384885,
      "num_tokens": 175176543.0,
      "step": 76405
    },
    {
      "entropy": 5.457051086425781,
      "epoch": 7.340057636887608,
      "grad_norm": 1.5,
      "learning_rate": 0.00012541798977479135,
      "loss": 4.581,
      "mean_token_accuracy": 0.27220394313335416,
      "num_tokens": 175188369.0,
      "step": 76410
    },
    {
      "entropy": 5.480927848815918,
      "epoch": 7.340537944284342,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00012539238123456518,
      "loss": 4.5695,
      "mean_token_accuracy": 0.2655743211507797,
      "num_tokens": 175199981.0,
      "step": 76415
    },
    {
      "entropy": 5.517491102218628,
      "epoch": 7.341018251681076,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012536677616778374,
      "loss": 4.5841,
      "mean_token_accuracy": 0.27788545191287994,
      "num_tokens": 175211850.0,
      "step": 76420
    },
    {
      "entropy": 5.5024083137512205,
      "epoch": 7.3414985590778095,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012534117457504139,
      "loss": 4.6201,
      "mean_token_accuracy": 0.2627014249563217,
      "num_tokens": 175222771.0,
      "step": 76425
    },
    {
      "entropy": 5.55044641494751,
      "epoch": 7.341978866474544,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012531557645693254,
      "loss": 4.6317,
      "mean_token_accuracy": 0.2673605367541313,
      "num_tokens": 175234942.0,
      "step": 76430
    },
    {
      "entropy": 5.4817639827728275,
      "epoch": 7.342459173871277,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012528998181405155,
      "loss": 4.5752,
      "mean_token_accuracy": 0.26991914957761765,
      "num_tokens": 175245392.0,
      "step": 76435
    },
    {
      "entropy": 5.529428148269654,
      "epoch": 7.3429394812680115,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00012526439064699263,
      "loss": 4.5939,
      "mean_token_accuracy": 0.2712163105607033,
      "num_tokens": 175255124.0,
      "step": 76440
    },
    {
      "entropy": 5.517915344238281,
      "epoch": 7.343419788664745,
      "grad_norm": 1.234375,
      "learning_rate": 0.00012523880295634994,
      "loss": 4.667,
      "mean_token_accuracy": 0.2670596525073051,
      "num_tokens": 175268533.0,
      "step": 76445
    },
    {
      "entropy": 5.527147674560547,
      "epoch": 7.343900096061479,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001252132187427176,
      "loss": 4.5438,
      "mean_token_accuracy": 0.2759313404560089,
      "num_tokens": 175279316.0,
      "step": 76450
    },
    {
      "entropy": 5.505368566513061,
      "epoch": 7.344380403458214,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012518763800668946,
      "loss": 4.5778,
      "mean_token_accuracy": 0.2742560550570488,
      "num_tokens": 175290869.0,
      "step": 76455
    },
    {
      "entropy": 5.5425464630126955,
      "epoch": 7.344860710854947,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012516206074885956,
      "loss": 4.6352,
      "mean_token_accuracy": 0.2580261558294296,
      "num_tokens": 175302057.0,
      "step": 76460
    },
    {
      "entropy": 5.478969860076904,
      "epoch": 7.345341018251681,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012513648696982167,
      "loss": 4.5659,
      "mean_token_accuracy": 0.2628646045923233,
      "num_tokens": 175313564.0,
      "step": 76465
    },
    {
      "entropy": 5.527493238449097,
      "epoch": 7.345821325648415,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012511091667016954,
      "loss": 4.5538,
      "mean_token_accuracy": 0.27924413084983823,
      "num_tokens": 175324729.0,
      "step": 76470
    },
    {
      "entropy": 5.503084373474121,
      "epoch": 7.346301633045149,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012508534985049687,
      "loss": 4.5875,
      "mean_token_accuracy": 0.2753132045269012,
      "num_tokens": 175335634.0,
      "step": 76475
    },
    {
      "entropy": 5.527726745605468,
      "epoch": 7.346781940441883,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012505978651139723,
      "loss": 4.6173,
      "mean_token_accuracy": 0.26498199105262754,
      "num_tokens": 175347234.0,
      "step": 76480
    },
    {
      "entropy": 5.497065830230713,
      "epoch": 7.347262247838617,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012503422665346411,
      "loss": 4.597,
      "mean_token_accuracy": 0.2679297402501106,
      "num_tokens": 175358498.0,
      "step": 76485
    },
    {
      "entropy": 5.571463584899902,
      "epoch": 7.34774255523535,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001250086702772909,
      "loss": 4.6153,
      "mean_token_accuracy": 0.26716437190771103,
      "num_tokens": 175370123.0,
      "step": 76490
    },
    {
      "entropy": 5.505539894104004,
      "epoch": 7.348222862632085,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000124983117383471,
      "loss": 4.5987,
      "mean_token_accuracy": 0.27322116047143935,
      "num_tokens": 175381050.0,
      "step": 76495
    },
    {
      "entropy": 5.5299829006195065,
      "epoch": 7.348703170028818,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012495756797259767,
      "loss": 4.5744,
      "mean_token_accuracy": 0.26623977422714235,
      "num_tokens": 175391476.0,
      "step": 76500
    },
    {
      "entropy": 5.499986457824707,
      "epoch": 7.3491834774255524,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012493202204526407,
      "loss": 4.5813,
      "mean_token_accuracy": 0.2684733852744102,
      "num_tokens": 175402831.0,
      "step": 76505
    },
    {
      "entropy": 5.529048490524292,
      "epoch": 7.349663784822286,
      "grad_norm": 1.375,
      "learning_rate": 0.00012490647960206337,
      "loss": 4.5934,
      "mean_token_accuracy": 0.2671955287456512,
      "num_tokens": 175414259.0,
      "step": 76510
    },
    {
      "entropy": 5.484806203842163,
      "epoch": 7.35014409221902,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012488094064358845,
      "loss": 4.5937,
      "mean_token_accuracy": 0.27418558597564696,
      "num_tokens": 175424551.0,
      "step": 76515
    },
    {
      "entropy": 5.521051597595215,
      "epoch": 7.350624399615754,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012485540517043233,
      "loss": 4.611,
      "mean_token_accuracy": 0.2701309844851494,
      "num_tokens": 175435862.0,
      "step": 76520
    },
    {
      "entropy": 5.459927272796631,
      "epoch": 7.351104707012488,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012482987318318789,
      "loss": 4.5605,
      "mean_token_accuracy": 0.27337819784879686,
      "num_tokens": 175447086.0,
      "step": 76525
    },
    {
      "entropy": 5.535019588470459,
      "epoch": 7.351585014409222,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012480434468244785,
      "loss": 4.6501,
      "mean_token_accuracy": 0.2714104026556015,
      "num_tokens": 175458260.0,
      "step": 76530
    },
    {
      "entropy": 5.509312725067138,
      "epoch": 7.352065321805956,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012477881966880492,
      "loss": 4.6102,
      "mean_token_accuracy": 0.26891896277666094,
      "num_tokens": 175468781.0,
      "step": 76535
    },
    {
      "entropy": 5.620385694503784,
      "epoch": 7.35254562920269,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012475329814285184,
      "loss": 4.7269,
      "mean_token_accuracy": 0.2599880203604698,
      "num_tokens": 175480238.0,
      "step": 76540
    },
    {
      "entropy": 5.488522052764893,
      "epoch": 7.3530259365994235,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012472778010518096,
      "loss": 4.5388,
      "mean_token_accuracy": 0.2752400875091553,
      "num_tokens": 175490867.0,
      "step": 76545
    },
    {
      "entropy": 5.508225965499878,
      "epoch": 7.353506243996158,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001247022655563848,
      "loss": 4.56,
      "mean_token_accuracy": 0.277467343211174,
      "num_tokens": 175503027.0,
      "step": 76550
    },
    {
      "entropy": 5.546572208404541,
      "epoch": 7.353986551392891,
      "grad_norm": 1.609375,
      "learning_rate": 0.00012467675449705576,
      "loss": 4.633,
      "mean_token_accuracy": 0.2716399982571602,
      "num_tokens": 175514012.0,
      "step": 76555
    },
    {
      "entropy": 5.580609083175659,
      "epoch": 7.354466858789626,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012465124692778606,
      "loss": 4.7059,
      "mean_token_accuracy": 0.26144794225692747,
      "num_tokens": 175524620.0,
      "step": 76560
    },
    {
      "entropy": 5.520032262802124,
      "epoch": 7.354947166186359,
      "grad_norm": 1.34375,
      "learning_rate": 0.000124625742849168,
      "loss": 4.6339,
      "mean_token_accuracy": 0.268134506046772,
      "num_tokens": 175535622.0,
      "step": 76565
    },
    {
      "entropy": 5.576312685012818,
      "epoch": 7.355427473583093,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012460024226179376,
      "loss": 4.6581,
      "mean_token_accuracy": 0.2598222404718399,
      "num_tokens": 175547549.0,
      "step": 76570
    },
    {
      "entropy": 5.5648435115814205,
      "epoch": 7.355907780979827,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001245747451662552,
      "loss": 4.6416,
      "mean_token_accuracy": 0.26476603746414185,
      "num_tokens": 175559429.0,
      "step": 76575
    },
    {
      "entropy": 5.514224481582642,
      "epoch": 7.356388088376561,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001245492515631444,
      "loss": 4.6424,
      "mean_token_accuracy": 0.2640830770134926,
      "num_tokens": 175570873.0,
      "step": 76580
    },
    {
      "entropy": 5.559652423858642,
      "epoch": 7.3568683957732945,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012452376145305323,
      "loss": 4.6216,
      "mean_token_accuracy": 0.267521496117115,
      "num_tokens": 175583203.0,
      "step": 76585
    },
    {
      "entropy": 5.553902626037598,
      "epoch": 7.357348703170029,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001244982748365735,
      "loss": 4.6043,
      "mean_token_accuracy": 0.2626766338944435,
      "num_tokens": 175593645.0,
      "step": 76590
    },
    {
      "entropy": 5.515901374816894,
      "epoch": 7.357829010566762,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012447279171429694,
      "loss": 4.6146,
      "mean_token_accuracy": 0.2656849518418312,
      "num_tokens": 175605503.0,
      "step": 76595
    },
    {
      "entropy": 5.548572254180908,
      "epoch": 7.358309317963497,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001244473120868152,
      "loss": 4.6893,
      "mean_token_accuracy": 0.25923922210931777,
      "num_tokens": 175616809.0,
      "step": 76600
    },
    {
      "entropy": 5.607307958602905,
      "epoch": 7.358789625360231,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001244218359547198,
      "loss": 4.6855,
      "mean_token_accuracy": 0.2573160409927368,
      "num_tokens": 175627459.0,
      "step": 76605
    },
    {
      "entropy": 5.5426350116729735,
      "epoch": 7.359269932756964,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012439636331860223,
      "loss": 4.5936,
      "mean_token_accuracy": 0.2712543398141861,
      "num_tokens": 175639596.0,
      "step": 76610
    },
    {
      "entropy": 5.503322172164917,
      "epoch": 7.359750240153699,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001243708941790539,
      "loss": 4.6024,
      "mean_token_accuracy": 0.2670115426182747,
      "num_tokens": 175650540.0,
      "step": 76615
    },
    {
      "entropy": 5.482146549224853,
      "epoch": 7.360230547550432,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012434542853666616,
      "loss": 4.5495,
      "mean_token_accuracy": 0.27810574769973756,
      "num_tokens": 175663292.0,
      "step": 76620
    },
    {
      "entropy": 5.505754375457764,
      "epoch": 7.3607108549471665,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001243199663920302,
      "loss": 4.6451,
      "mean_token_accuracy": 0.26571243554353713,
      "num_tokens": 175675082.0,
      "step": 76625
    },
    {
      "entropy": 5.576863622665405,
      "epoch": 7.3611911623439,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00012429450774573725,
      "loss": 4.6624,
      "mean_token_accuracy": 0.26439346075057985,
      "num_tokens": 175685898.0,
      "step": 76630
    },
    {
      "entropy": 5.538249397277832,
      "epoch": 7.361671469740634,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012426905259837825,
      "loss": 4.5834,
      "mean_token_accuracy": 0.2718898385763168,
      "num_tokens": 175697040.0,
      "step": 76635
    },
    {
      "entropy": 5.567992258071899,
      "epoch": 7.362151777137368,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001242436009505443,
      "loss": 4.6619,
      "mean_token_accuracy": 0.2669712319970131,
      "num_tokens": 175707276.0,
      "step": 76640
    },
    {
      "entropy": 5.479863405227661,
      "epoch": 7.362632084534102,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012421815280282624,
      "loss": 4.5291,
      "mean_token_accuracy": 0.27903186827898024,
      "num_tokens": 175719142.0,
      "step": 76645
    },
    {
      "entropy": 5.560622692108154,
      "epoch": 7.363112391930835,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000124192708155815,
      "loss": 4.6791,
      "mean_token_accuracy": 0.25841859430074693,
      "num_tokens": 175730759.0,
      "step": 76650
    },
    {
      "entropy": 5.4718200206756595,
      "epoch": 7.36359269932757,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012416726701010124,
      "loss": 4.5921,
      "mean_token_accuracy": 0.2675833225250244,
      "num_tokens": 175743653.0,
      "step": 76655
    },
    {
      "entropy": 5.522088146209716,
      "epoch": 7.364073006724303,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012414182936627572,
      "loss": 4.6771,
      "mean_token_accuracy": 0.26768717616796495,
      "num_tokens": 175755886.0,
      "step": 76660
    },
    {
      "entropy": 5.53211236000061,
      "epoch": 7.3645533141210375,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001241163952249289,
      "loss": 4.6176,
      "mean_token_accuracy": 0.26733954101800916,
      "num_tokens": 175767834.0,
      "step": 76665
    },
    {
      "entropy": 5.546552991867065,
      "epoch": 7.365033621517771,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012409096458665134,
      "loss": 4.5956,
      "mean_token_accuracy": 0.27680066227912903,
      "num_tokens": 175778215.0,
      "step": 76670
    },
    {
      "entropy": 5.491183853149414,
      "epoch": 7.365513928914505,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012406553745203348,
      "loss": 4.5963,
      "mean_token_accuracy": 0.26682561337947847,
      "num_tokens": 175790937.0,
      "step": 76675
    },
    {
      "entropy": 5.54368257522583,
      "epoch": 7.36599423631124,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001240401138216657,
      "loss": 4.7045,
      "mean_token_accuracy": 0.25423123240470885,
      "num_tokens": 175802725.0,
      "step": 76680
    },
    {
      "entropy": 5.58168215751648,
      "epoch": 7.366474543707973,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012401469369613817,
      "loss": 4.7103,
      "mean_token_accuracy": 0.24968115985393524,
      "num_tokens": 175814468.0,
      "step": 76685
    },
    {
      "entropy": 5.5542370796203615,
      "epoch": 7.366954851104707,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001239892770760412,
      "loss": 4.6742,
      "mean_token_accuracy": 0.26065069288015363,
      "num_tokens": 175825206.0,
      "step": 76690
    },
    {
      "entropy": 5.541395473480224,
      "epoch": 7.367435158501441,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012396386396196473,
      "loss": 4.6475,
      "mean_token_accuracy": 0.2684959515929222,
      "num_tokens": 175835915.0,
      "step": 76695
    },
    {
      "entropy": 5.54014105796814,
      "epoch": 7.367915465898175,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012393845435449888,
      "loss": 4.6306,
      "mean_token_accuracy": 0.2691194176673889,
      "num_tokens": 175846610.0,
      "step": 76700
    },
    {
      "entropy": 5.507621431350708,
      "epoch": 7.368395773294909,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012391304825423353,
      "loss": 4.6149,
      "mean_token_accuracy": 0.2701083317399025,
      "num_tokens": 175858590.0,
      "step": 76705
    },
    {
      "entropy": 5.467441129684448,
      "epoch": 7.368876080691643,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012388764566175858,
      "loss": 4.5559,
      "mean_token_accuracy": 0.2747277170419693,
      "num_tokens": 175870091.0,
      "step": 76710
    },
    {
      "entropy": 5.4686699390411375,
      "epoch": 7.369356388088376,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012386224657766375,
      "loss": 4.5331,
      "mean_token_accuracy": 0.2854066133499146,
      "num_tokens": 175882796.0,
      "step": 76715
    },
    {
      "entropy": 5.478076601028443,
      "epoch": 7.369836695485111,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001238368510025389,
      "loss": 4.6297,
      "mean_token_accuracy": 0.2637812927365303,
      "num_tokens": 175894810.0,
      "step": 76720
    },
    {
      "entropy": 5.597141408920288,
      "epoch": 7.370317002881844,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001238114589369734,
      "loss": 4.702,
      "mean_token_accuracy": 0.25980985164642334,
      "num_tokens": 175906159.0,
      "step": 76725
    },
    {
      "entropy": 5.543131494522095,
      "epoch": 7.370797310278578,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001237860703815569,
      "loss": 4.613,
      "mean_token_accuracy": 0.2746706515550613,
      "num_tokens": 175916818.0,
      "step": 76730
    },
    {
      "entropy": 5.511009931564331,
      "epoch": 7.371277617675312,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012376068533687882,
      "loss": 4.5841,
      "mean_token_accuracy": 0.27221019715070727,
      "num_tokens": 175928120.0,
      "step": 76735
    },
    {
      "entropy": 5.512326669692993,
      "epoch": 7.371757925072046,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012373530380352854,
      "loss": 4.6105,
      "mean_token_accuracy": 0.27045400738716124,
      "num_tokens": 175938961.0,
      "step": 76740
    },
    {
      "entropy": 5.6711701393127445,
      "epoch": 7.37223823246878,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012370992578209533,
      "loss": 4.8073,
      "mean_token_accuracy": 0.2514166757464409,
      "num_tokens": 175950430.0,
      "step": 76745
    },
    {
      "entropy": 5.59734582901001,
      "epoch": 7.372718539865514,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012368455127316852,
      "loss": 4.6347,
      "mean_token_accuracy": 0.2636130556464195,
      "num_tokens": 175963360.0,
      "step": 76750
    },
    {
      "entropy": 5.546596097946167,
      "epoch": 7.373198847262248,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012365918027733698,
      "loss": 4.6614,
      "mean_token_accuracy": 0.2572838142514229,
      "num_tokens": 175976987.0,
      "step": 76755
    },
    {
      "entropy": 5.521928310394287,
      "epoch": 7.373679154658982,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012363381279518993,
      "loss": 4.6273,
      "mean_token_accuracy": 0.2708470895886421,
      "num_tokens": 175988524.0,
      "step": 76760
    },
    {
      "entropy": 5.453183937072754,
      "epoch": 7.374159462055716,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012360844882731628,
      "loss": 4.5644,
      "mean_token_accuracy": 0.27290751338005065,
      "num_tokens": 176000163.0,
      "step": 76765
    },
    {
      "entropy": 5.6040033340454105,
      "epoch": 7.3746397694524495,
      "grad_norm": 1.375,
      "learning_rate": 0.00012358308837430486,
      "loss": 4.7208,
      "mean_token_accuracy": 0.25532193332910535,
      "num_tokens": 176012920.0,
      "step": 76770
    },
    {
      "entropy": 5.523635578155518,
      "epoch": 7.375120076849184,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012355773143674458,
      "loss": 4.6153,
      "mean_token_accuracy": 0.26974946111440656,
      "num_tokens": 176023737.0,
      "step": 76775
    },
    {
      "entropy": 5.495352268218994,
      "epoch": 7.375600384245917,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001235323780152241,
      "loss": 4.5621,
      "mean_token_accuracy": 0.2716190040111542,
      "num_tokens": 176035416.0,
      "step": 76780
    },
    {
      "entropy": 5.514281606674194,
      "epoch": 7.376080691642652,
      "grad_norm": 1.5,
      "learning_rate": 0.000123507028110332,
      "loss": 4.5674,
      "mean_token_accuracy": 0.27438582926988603,
      "num_tokens": 176046296.0,
      "step": 76785
    },
    {
      "entropy": 5.612879848480224,
      "epoch": 7.376560999039385,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012348168172265686,
      "loss": 4.6805,
      "mean_token_accuracy": 0.2613286763429642,
      "num_tokens": 176056956.0,
      "step": 76790
    },
    {
      "entropy": 5.594086217880249,
      "epoch": 7.377041306436119,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012345633885278714,
      "loss": 4.6598,
      "mean_token_accuracy": 0.25599246472120285,
      "num_tokens": 176069394.0,
      "step": 76795
    },
    {
      "entropy": 5.495570325851441,
      "epoch": 7.377521613832853,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00012343099950131125,
      "loss": 4.5356,
      "mean_token_accuracy": 0.269488599896431,
      "num_tokens": 176079634.0,
      "step": 76800
    },
    {
      "entropy": 5.524545955657959,
      "epoch": 7.378001921229587,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012340566366881748,
      "loss": 4.6717,
      "mean_token_accuracy": 0.25496692061424253,
      "num_tokens": 176093381.0,
      "step": 76805
    },
    {
      "entropy": 5.498708581924438,
      "epoch": 7.3784822286263205,
      "grad_norm": 1.375,
      "learning_rate": 0.00012338033135589406,
      "loss": 4.5315,
      "mean_token_accuracy": 0.27588191628456116,
      "num_tokens": 176103757.0,
      "step": 76810
    },
    {
      "entropy": 5.4963523864746096,
      "epoch": 7.378962536023055,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012335500256312914,
      "loss": 4.6126,
      "mean_token_accuracy": 0.26735698580741885,
      "num_tokens": 176115394.0,
      "step": 76815
    },
    {
      "entropy": 5.541108655929565,
      "epoch": 7.379442843419788,
      "grad_norm": 1.5,
      "learning_rate": 0.00012332967729111082,
      "loss": 4.6876,
      "mean_token_accuracy": 0.2618061825633049,
      "num_tokens": 176126586.0,
      "step": 76820
    },
    {
      "entropy": 5.506294536590576,
      "epoch": 7.379923150816523,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012330435554042698,
      "loss": 4.6128,
      "mean_token_accuracy": 0.2702059030532837,
      "num_tokens": 176139011.0,
      "step": 76825
    },
    {
      "entropy": 5.568774747848511,
      "epoch": 7.380403458213257,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001232790373116655,
      "loss": 4.6358,
      "mean_token_accuracy": 0.2631305634975433,
      "num_tokens": 176149159.0,
      "step": 76830
    },
    {
      "entropy": 5.517100715637207,
      "epoch": 7.38088376560999,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012325372260541431,
      "loss": 4.5919,
      "mean_token_accuracy": 0.2694935411214828,
      "num_tokens": 176161537.0,
      "step": 76835
    },
    {
      "entropy": 5.613601446151733,
      "epoch": 7.381364073006725,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012322841142226108,
      "loss": 4.682,
      "mean_token_accuracy": 0.26398875266313554,
      "num_tokens": 176172611.0,
      "step": 76840
    },
    {
      "entropy": 5.561581802368164,
      "epoch": 7.381844380403458,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012320310376279348,
      "loss": 4.6615,
      "mean_token_accuracy": 0.26330033540725706,
      "num_tokens": 176184689.0,
      "step": 76845
    },
    {
      "entropy": 5.497640752792359,
      "epoch": 7.3823246878001925,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012317779962759903,
      "loss": 4.5789,
      "mean_token_accuracy": 0.2694458901882172,
      "num_tokens": 176194797.0,
      "step": 76850
    },
    {
      "entropy": 5.625890064239502,
      "epoch": 7.382804995196926,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012315249901726527,
      "loss": 4.7597,
      "mean_token_accuracy": 0.2587196320295334,
      "num_tokens": 176206375.0,
      "step": 76855
    },
    {
      "entropy": 5.499942588806152,
      "epoch": 7.38328530259366,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012312720193237965,
      "loss": 4.5597,
      "mean_token_accuracy": 0.2698747530579567,
      "num_tokens": 176218487.0,
      "step": 76860
    },
    {
      "entropy": 5.602021789550781,
      "epoch": 7.383765609990394,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001231019083735294,
      "loss": 4.6808,
      "mean_token_accuracy": 0.26031937897205354,
      "num_tokens": 176229309.0,
      "step": 76865
    },
    {
      "entropy": 5.483977317810059,
      "epoch": 7.384245917387128,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012307661834130173,
      "loss": 4.6109,
      "mean_token_accuracy": 0.2622386574745178,
      "num_tokens": 176242038.0,
      "step": 76870
    },
    {
      "entropy": 5.509524154663086,
      "epoch": 7.384726224783861,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012305133183628393,
      "loss": 4.6106,
      "mean_token_accuracy": 0.2652278959751129,
      "num_tokens": 176251976.0,
      "step": 76875
    },
    {
      "entropy": 5.512401485443116,
      "epoch": 7.385206532180596,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012302604885906294,
      "loss": 4.5724,
      "mean_token_accuracy": 0.27043892443180084,
      "num_tokens": 176263294.0,
      "step": 76880
    },
    {
      "entropy": 5.53942379951477,
      "epoch": 7.385686839577329,
      "grad_norm": 1.5,
      "learning_rate": 0.00012300076941022586,
      "loss": 4.6252,
      "mean_token_accuracy": 0.26833302676677706,
      "num_tokens": 176274355.0,
      "step": 76885
    },
    {
      "entropy": 5.549146223068237,
      "epoch": 7.3861671469740635,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001229754934903596,
      "loss": 4.6504,
      "mean_token_accuracy": 0.26339775919914243,
      "num_tokens": 176286986.0,
      "step": 76890
    },
    {
      "entropy": 5.463040828704834,
      "epoch": 7.386647454370797,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001229502211000509,
      "loss": 4.5894,
      "mean_token_accuracy": 0.2706568315625191,
      "num_tokens": 176297811.0,
      "step": 76895
    },
    {
      "entropy": 5.533137035369873,
      "epoch": 7.387127761767531,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001229249522398867,
      "loss": 4.6577,
      "mean_token_accuracy": 0.2629494398832321,
      "num_tokens": 176309640.0,
      "step": 76900
    },
    {
      "entropy": 5.530860328674317,
      "epoch": 7.387608069164266,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012289968691045342,
      "loss": 4.6168,
      "mean_token_accuracy": 0.26072089076042176,
      "num_tokens": 176320236.0,
      "step": 76905
    },
    {
      "entropy": 5.496640682220459,
      "epoch": 7.388088376560999,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012287442511233778,
      "loss": 4.5377,
      "mean_token_accuracy": 0.27968843579292296,
      "num_tokens": 176332134.0,
      "step": 76910
    },
    {
      "entropy": 5.5059370517730715,
      "epoch": 7.388568683957733,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001228491668461263,
      "loss": 4.5713,
      "mean_token_accuracy": 0.2718709230422974,
      "num_tokens": 176343149.0,
      "step": 76915
    },
    {
      "entropy": 5.563431119918823,
      "epoch": 7.389048991354467,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012282391211240534,
      "loss": 4.6491,
      "mean_token_accuracy": 0.2675041064620018,
      "num_tokens": 176354438.0,
      "step": 76920
    },
    {
      "entropy": 5.56791615486145,
      "epoch": 7.389529298751201,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012279866091176127,
      "loss": 4.6407,
      "mean_token_accuracy": 0.2700399771332741,
      "num_tokens": 176365260.0,
      "step": 76925
    },
    {
      "entropy": 5.519742059707641,
      "epoch": 7.390009606147935,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012277341324478042,
      "loss": 4.5973,
      "mean_token_accuracy": 0.2717962354421616,
      "num_tokens": 176376215.0,
      "step": 76930
    },
    {
      "entropy": 5.514758729934693,
      "epoch": 7.390489913544669,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012274816911204883,
      "loss": 4.6315,
      "mean_token_accuracy": 0.2634970277547836,
      "num_tokens": 176387722.0,
      "step": 76935
    },
    {
      "entropy": 5.608729267120362,
      "epoch": 7.390970220941402,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012272292851415269,
      "loss": 4.6646,
      "mean_token_accuracy": 0.2669728472828865,
      "num_tokens": 176398508.0,
      "step": 76940
    },
    {
      "entropy": 5.607793998718262,
      "epoch": 7.391450528338137,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012269769145167793,
      "loss": 4.6734,
      "mean_token_accuracy": 0.2613354727625847,
      "num_tokens": 176409648.0,
      "step": 76945
    },
    {
      "entropy": 5.6218396663665775,
      "epoch": 7.39193083573487,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00012267245792521055,
      "loss": 4.6976,
      "mean_token_accuracy": 0.2539172276854515,
      "num_tokens": 176419978.0,
      "step": 76950
    },
    {
      "entropy": 5.590926885604858,
      "epoch": 7.392411143131604,
      "grad_norm": 1.375,
      "learning_rate": 0.00012264722793533636,
      "loss": 4.7226,
      "mean_token_accuracy": 0.2621404483914375,
      "num_tokens": 176431708.0,
      "step": 76955
    },
    {
      "entropy": 5.587695980072022,
      "epoch": 7.392891450528338,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012262200148264123,
      "loss": 4.6196,
      "mean_token_accuracy": 0.25945882350206373,
      "num_tokens": 176442356.0,
      "step": 76960
    },
    {
      "entropy": 5.578138780593872,
      "epoch": 7.393371757925072,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012259677856771067,
      "loss": 4.6393,
      "mean_token_accuracy": 0.2706207290291786,
      "num_tokens": 176453818.0,
      "step": 76965
    },
    {
      "entropy": 5.500357294082642,
      "epoch": 7.393852065321806,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001225715591911304,
      "loss": 4.51,
      "mean_token_accuracy": 0.2801367461681366,
      "num_tokens": 176464374.0,
      "step": 76970
    },
    {
      "entropy": 5.574135446548462,
      "epoch": 7.39433237271854,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001225463433534859,
      "loss": 4.7313,
      "mean_token_accuracy": 0.26149821281433105,
      "num_tokens": 176476098.0,
      "step": 76975
    },
    {
      "entropy": 5.558745098114014,
      "epoch": 7.394812680115273,
      "grad_norm": 1.375,
      "learning_rate": 0.00012252113105536258,
      "loss": 4.6206,
      "mean_token_accuracy": 0.268041780591011,
      "num_tokens": 176487751.0,
      "step": 76980
    },
    {
      "entropy": 5.554134941101074,
      "epoch": 7.395292987512008,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012249592229734583,
      "loss": 4.6481,
      "mean_token_accuracy": 0.2593346104025841,
      "num_tokens": 176500049.0,
      "step": 76985
    },
    {
      "entropy": 5.421144151687622,
      "epoch": 7.395773294908741,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012247071708002105,
      "loss": 4.4931,
      "mean_token_accuracy": 0.28634233474731446,
      "num_tokens": 176511665.0,
      "step": 76990
    },
    {
      "entropy": 5.532313108444214,
      "epoch": 7.3962536023054755,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012244551540397316,
      "loss": 4.5728,
      "mean_token_accuracy": 0.2643143504858017,
      "num_tokens": 176522146.0,
      "step": 76995
    },
    {
      "entropy": 5.506574869155884,
      "epoch": 7.39673390970221,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012242031726978744,
      "loss": 4.5761,
      "mean_token_accuracy": 0.27240302562713625,
      "num_tokens": 176533458.0,
      "step": 77000
    },
    {
      "entropy": 5.436460781097412,
      "epoch": 7.397214217098943,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00012239512267804887,
      "loss": 4.4631,
      "mean_token_accuracy": 0.2797147870063782,
      "num_tokens": 176544098.0,
      "step": 77005
    },
    {
      "entropy": 5.501794290542603,
      "epoch": 7.3976945244956775,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012236993162934243,
      "loss": 4.5497,
      "mean_token_accuracy": 0.2758283495903015,
      "num_tokens": 176556169.0,
      "step": 77010
    },
    {
      "entropy": 5.480987977981568,
      "epoch": 7.398174831892411,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012234474412425295,
      "loss": 4.6153,
      "mean_token_accuracy": 0.2636081874370575,
      "num_tokens": 176567312.0,
      "step": 77015
    },
    {
      "entropy": 5.506535291671753,
      "epoch": 7.398655139289145,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012231956016336528,
      "loss": 4.5882,
      "mean_token_accuracy": 0.2632965981960297,
      "num_tokens": 176576853.0,
      "step": 77020
    },
    {
      "entropy": 5.491471195220948,
      "epoch": 7.399135446685879,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012229437974726402,
      "loss": 4.5597,
      "mean_token_accuracy": 0.27073664516210555,
      "num_tokens": 176587195.0,
      "step": 77025
    },
    {
      "entropy": 5.502240610122681,
      "epoch": 7.399615754082613,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001222692028765338,
      "loss": 4.6229,
      "mean_token_accuracy": 0.2638338327407837,
      "num_tokens": 176597535.0,
      "step": 77030
    },
    {
      "entropy": 5.483839273452759,
      "epoch": 7.4000960614793465,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012224402955175913,
      "loss": 4.6197,
      "mean_token_accuracy": 0.2660699784755707,
      "num_tokens": 176608916.0,
      "step": 77035
    },
    {
      "entropy": 5.551679277420044,
      "epoch": 7.400576368876081,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012221885977352454,
      "loss": 4.6448,
      "mean_token_accuracy": 0.2640024036169052,
      "num_tokens": 176620925.0,
      "step": 77040
    },
    {
      "entropy": 5.500045251846314,
      "epoch": 7.401056676272814,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012219369354241435,
      "loss": 4.612,
      "mean_token_accuracy": 0.26503877639770507,
      "num_tokens": 176632853.0,
      "step": 77045
    },
    {
      "entropy": 5.469888544082641,
      "epoch": 7.401536983669549,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012216853085901294,
      "loss": 4.4832,
      "mean_token_accuracy": 0.2744519144296646,
      "num_tokens": 176644793.0,
      "step": 77050
    },
    {
      "entropy": 5.565685510635376,
      "epoch": 7.402017291066282,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012214337172390433,
      "loss": 4.631,
      "mean_token_accuracy": 0.2632502019405365,
      "num_tokens": 176656198.0,
      "step": 77055
    },
    {
      "entropy": 5.575326013565063,
      "epoch": 7.402497598463016,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012211821613767275,
      "loss": 4.642,
      "mean_token_accuracy": 0.2694235533475876,
      "num_tokens": 176667670.0,
      "step": 77060
    },
    {
      "entropy": 5.507529449462891,
      "epoch": 7.40297790585975,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012209306410090225,
      "loss": 4.6627,
      "mean_token_accuracy": 0.2547886401414871,
      "num_tokens": 176678462.0,
      "step": 77065
    },
    {
      "entropy": 5.488834571838379,
      "epoch": 7.403458213256484,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012206791561417674,
      "loss": 4.5225,
      "mean_token_accuracy": 0.27354924380779266,
      "num_tokens": 176689537.0,
      "step": 77070
    },
    {
      "entropy": 5.511465215682984,
      "epoch": 7.4039385206532184,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001220427706780801,
      "loss": 4.6683,
      "mean_token_accuracy": 0.2616133227944374,
      "num_tokens": 176701258.0,
      "step": 77075
    },
    {
      "entropy": 5.513262557983398,
      "epoch": 7.404418828049952,
      "grad_norm": 1.5,
      "learning_rate": 0.00012201762929319624,
      "loss": 4.5364,
      "mean_token_accuracy": 0.2735149681568146,
      "num_tokens": 176711406.0,
      "step": 77080
    },
    {
      "entropy": 5.527090978622437,
      "epoch": 7.404899135446686,
      "grad_norm": 1.375,
      "learning_rate": 0.00012199249146010869,
      "loss": 4.6172,
      "mean_token_accuracy": 0.26767676025629045,
      "num_tokens": 176723533.0,
      "step": 77085
    },
    {
      "entropy": 5.502191209793091,
      "epoch": 7.40537944284342,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012196735717940111,
      "loss": 4.6095,
      "mean_token_accuracy": 0.2670235961675644,
      "num_tokens": 176734725.0,
      "step": 77090
    },
    {
      "entropy": 5.536881637573242,
      "epoch": 7.405859750240154,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012194222645165711,
      "loss": 4.6429,
      "mean_token_accuracy": 0.2635518506169319,
      "num_tokens": 176745819.0,
      "step": 77095
    },
    {
      "entropy": 5.521086359024048,
      "epoch": 7.406340057636887,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012191709927746013,
      "loss": 4.6488,
      "mean_token_accuracy": 0.2695117354393005,
      "num_tokens": 176756875.0,
      "step": 77100
    },
    {
      "entropy": 5.5817193508148195,
      "epoch": 7.406820365033622,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001218919756573935,
      "loss": 4.6623,
      "mean_token_accuracy": 0.261855748295784,
      "num_tokens": 176768158.0,
      "step": 77105
    },
    {
      "entropy": 5.571627044677735,
      "epoch": 7.407300672430355,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012186685559204069,
      "loss": 4.6691,
      "mean_token_accuracy": 0.2573924005031586,
      "num_tokens": 176779424.0,
      "step": 77110
    },
    {
      "entropy": 5.4797553539276125,
      "epoch": 7.4077809798270895,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012184173908198466,
      "loss": 4.5442,
      "mean_token_accuracy": 0.2716261327266693,
      "num_tokens": 176790176.0,
      "step": 77115
    },
    {
      "entropy": 5.535060930252075,
      "epoch": 7.408261287223823,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001218166261278087,
      "loss": 4.6452,
      "mean_token_accuracy": 0.26522432267665863,
      "num_tokens": 176801528.0,
      "step": 77120
    },
    {
      "entropy": 5.568867111206055,
      "epoch": 7.408741594620557,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012179151673009579,
      "loss": 4.6565,
      "mean_token_accuracy": 0.25872943699359896,
      "num_tokens": 176813671.0,
      "step": 77125
    },
    {
      "entropy": 5.557069587707519,
      "epoch": 7.409221902017291,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012176641088942894,
      "loss": 4.6043,
      "mean_token_accuracy": 0.2667640134692192,
      "num_tokens": 176826024.0,
      "step": 77130
    },
    {
      "entropy": 5.514448976516723,
      "epoch": 7.409702209414025,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012174130860639101,
      "loss": 4.5356,
      "mean_token_accuracy": 0.2766598552465439,
      "num_tokens": 176837481.0,
      "step": 77135
    },
    {
      "entropy": 5.511081886291504,
      "epoch": 7.4101825168107585,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012171620988156491,
      "loss": 4.6267,
      "mean_token_accuracy": 0.2620175272226334,
      "num_tokens": 176849695.0,
      "step": 77140
    },
    {
      "entropy": 5.585691308975219,
      "epoch": 7.410662824207493,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012169111471553317,
      "loss": 4.6674,
      "mean_token_accuracy": 0.26587062776088716,
      "num_tokens": 176860509.0,
      "step": 77145
    },
    {
      "entropy": 5.488864517211914,
      "epoch": 7.411143131604227,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001216660231088785,
      "loss": 4.6146,
      "mean_token_accuracy": 0.2656100928783417,
      "num_tokens": 176872392.0,
      "step": 77150
    },
    {
      "entropy": 5.486457347869873,
      "epoch": 7.4116234390009605,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012164093506218348,
      "loss": 4.5026,
      "mean_token_accuracy": 0.2798673868179321,
      "num_tokens": 176882955.0,
      "step": 77155
    },
    {
      "entropy": 5.589533710479737,
      "epoch": 7.412103746397695,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00012161585057603056,
      "loss": 4.7454,
      "mean_token_accuracy": 0.25903378129005433,
      "num_tokens": 176893927.0,
      "step": 77160
    },
    {
      "entropy": 5.608587121963501,
      "epoch": 7.412584053794428,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00012159076965100212,
      "loss": 4.7238,
      "mean_token_accuracy": 0.2632043719291687,
      "num_tokens": 176904774.0,
      "step": 77165
    },
    {
      "entropy": 5.514708995819092,
      "epoch": 7.413064361191163,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012156569228768056,
      "loss": 4.6251,
      "mean_token_accuracy": 0.27094183266162875,
      "num_tokens": 176916329.0,
      "step": 77170
    },
    {
      "entropy": 5.55232720375061,
      "epoch": 7.413544668587896,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012154061848664793,
      "loss": 4.5628,
      "mean_token_accuracy": 0.2715796664357185,
      "num_tokens": 176927584.0,
      "step": 77175
    },
    {
      "entropy": 5.507074022293091,
      "epoch": 7.41402497598463,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012151554824848648,
      "loss": 4.5754,
      "mean_token_accuracy": 0.2705894947052002,
      "num_tokens": 176939693.0,
      "step": 77180
    },
    {
      "entropy": 5.55054726600647,
      "epoch": 7.414505283381364,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012149048157377823,
      "loss": 4.6186,
      "mean_token_accuracy": 0.2711751669645309,
      "num_tokens": 176950075.0,
      "step": 77185
    },
    {
      "entropy": 5.535124158859253,
      "epoch": 7.414985590778098,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012146541846310518,
      "loss": 4.6742,
      "mean_token_accuracy": 0.26722975075244904,
      "num_tokens": 176961761.0,
      "step": 77190
    },
    {
      "entropy": 5.568575239181518,
      "epoch": 7.415465898174832,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001214403589170492,
      "loss": 4.7086,
      "mean_token_accuracy": 0.2595836207270622,
      "num_tokens": 176973149.0,
      "step": 77195
    },
    {
      "entropy": 5.419738388061523,
      "epoch": 7.415946205571566,
      "grad_norm": 1.375,
      "learning_rate": 0.00012141530293619219,
      "loss": 4.553,
      "mean_token_accuracy": 0.27431823313236237,
      "num_tokens": 176984878.0,
      "step": 77200
    },
    {
      "entropy": 5.511426639556885,
      "epoch": 7.416426512968299,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012139025052111572,
      "loss": 4.631,
      "mean_token_accuracy": 0.27228339314460753,
      "num_tokens": 176996544.0,
      "step": 77205
    },
    {
      "entropy": 5.488936901092529,
      "epoch": 7.416906820365034,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012136520167240152,
      "loss": 4.525,
      "mean_token_accuracy": 0.278300404548645,
      "num_tokens": 177007617.0,
      "step": 77210
    },
    {
      "entropy": 5.546442174911499,
      "epoch": 7.417387127761767,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012134015639063109,
      "loss": 4.6006,
      "mean_token_accuracy": 0.2701179921627045,
      "num_tokens": 177018774.0,
      "step": 77215
    },
    {
      "entropy": 5.618250131607056,
      "epoch": 7.417867435158501,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000121315114676386,
      "loss": 4.7507,
      "mean_token_accuracy": 0.2560040220618248,
      "num_tokens": 177031718.0,
      "step": 77220
    },
    {
      "entropy": 5.5301597118377686,
      "epoch": 7.418347742555236,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012129007653024755,
      "loss": 4.5994,
      "mean_token_accuracy": 0.2647125095129013,
      "num_tokens": 177042544.0,
      "step": 77225
    },
    {
      "entropy": 5.599887895584106,
      "epoch": 7.418828049951969,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001212650419527972,
      "loss": 4.7379,
      "mean_token_accuracy": 0.26383222043514254,
      "num_tokens": 177053470.0,
      "step": 77230
    },
    {
      "entropy": 5.560779237747193,
      "epoch": 7.4193083573487035,
      "grad_norm": 1.5,
      "learning_rate": 0.00012124001094461601,
      "loss": 4.617,
      "mean_token_accuracy": 0.26518380641937256,
      "num_tokens": 177065324.0,
      "step": 77235
    },
    {
      "entropy": 5.624463415145874,
      "epoch": 7.419788664745437,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012121498350628518,
      "loss": 4.6768,
      "mean_token_accuracy": 0.2654169112443924,
      "num_tokens": 177076763.0,
      "step": 77240
    },
    {
      "entropy": 5.551996564865112,
      "epoch": 7.420268972142171,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012118995963838579,
      "loss": 4.6692,
      "mean_token_accuracy": 0.2634863629937172,
      "num_tokens": 177088543.0,
      "step": 77245
    },
    {
      "entropy": 5.498275279998779,
      "epoch": 7.420749279538905,
      "grad_norm": 1.609375,
      "learning_rate": 0.00012116493934149882,
      "loss": 4.5958,
      "mean_token_accuracy": 0.2696526288986206,
      "num_tokens": 177100207.0,
      "step": 77250
    },
    {
      "entropy": 5.549862670898437,
      "epoch": 7.421229586935639,
      "grad_norm": 1.625,
      "learning_rate": 0.00012113992261620516,
      "loss": 4.5923,
      "mean_token_accuracy": 0.26751860678195954,
      "num_tokens": 177110728.0,
      "step": 77255
    },
    {
      "entropy": 5.531507539749145,
      "epoch": 7.4217098943323725,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001211149094630857,
      "loss": 4.5624,
      "mean_token_accuracy": 0.2662960574030876,
      "num_tokens": 177122428.0,
      "step": 77260
    },
    {
      "entropy": 5.536641311645508,
      "epoch": 7.422190201729107,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012108989988272101,
      "loss": 4.59,
      "mean_token_accuracy": 0.27491766661405564,
      "num_tokens": 177133925.0,
      "step": 77265
    },
    {
      "entropy": 5.516088247299194,
      "epoch": 7.42267050912584,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012106489387569185,
      "loss": 4.5883,
      "mean_token_accuracy": 0.2755108490586281,
      "num_tokens": 177144521.0,
      "step": 77270
    },
    {
      "entropy": 5.494770860671997,
      "epoch": 7.423150816522575,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00012103989144257872,
      "loss": 4.5586,
      "mean_token_accuracy": 0.27359454333782196,
      "num_tokens": 177154946.0,
      "step": 77275
    },
    {
      "entropy": 5.5479827404022215,
      "epoch": 7.423631123919308,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012101489258396218,
      "loss": 4.591,
      "mean_token_accuracy": 0.27043323665857316,
      "num_tokens": 177166340.0,
      "step": 77280
    },
    {
      "entropy": 5.5210230350494385,
      "epoch": 7.424111431316042,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012098989730042258,
      "loss": 4.6025,
      "mean_token_accuracy": 0.264816877245903,
      "num_tokens": 177177707.0,
      "step": 77285
    },
    {
      "entropy": 5.510782146453858,
      "epoch": 7.424591738712776,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012096490559254033,
      "loss": 4.5611,
      "mean_token_accuracy": 0.2748164474964142,
      "num_tokens": 177188406.0,
      "step": 77290
    },
    {
      "entropy": 5.5187335968017575,
      "epoch": 7.42507204610951,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012093991746089549,
      "loss": 4.5788,
      "mean_token_accuracy": 0.2686475902795792,
      "num_tokens": 177199797.0,
      "step": 77295
    },
    {
      "entropy": 5.507018375396728,
      "epoch": 7.425552353506244,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012091493290606833,
      "loss": 4.5492,
      "mean_token_accuracy": 0.27539946883916855,
      "num_tokens": 177211116.0,
      "step": 77300
    },
    {
      "entropy": 5.511601448059082,
      "epoch": 7.426032660902978,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00012088995192863888,
      "loss": 4.6033,
      "mean_token_accuracy": 0.2745028868317604,
      "num_tokens": 177221567.0,
      "step": 77305
    },
    {
      "entropy": 5.498178243637085,
      "epoch": 7.426512968299712,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012086497452918712,
      "loss": 4.6787,
      "mean_token_accuracy": 0.2643149822950363,
      "num_tokens": 177233770.0,
      "step": 77310
    },
    {
      "entropy": 5.528987169265747,
      "epoch": 7.426993275696446,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012084000070829298,
      "loss": 4.596,
      "mean_token_accuracy": 0.2720886215567589,
      "num_tokens": 177245540.0,
      "step": 77315
    },
    {
      "entropy": 5.551324319839478,
      "epoch": 7.42747358309318,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012081503046653632,
      "loss": 4.6464,
      "mean_token_accuracy": 0.2640131011605263,
      "num_tokens": 177256021.0,
      "step": 77320
    },
    {
      "entropy": 5.5395862579345705,
      "epoch": 7.427953890489913,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012079006380449675,
      "loss": 4.5855,
      "mean_token_accuracy": 0.26617407351732253,
      "num_tokens": 177266432.0,
      "step": 77325
    },
    {
      "entropy": 5.527892303466797,
      "epoch": 7.428434197886648,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012076510072275399,
      "loss": 4.5763,
      "mean_token_accuracy": 0.27198814749717715,
      "num_tokens": 177278225.0,
      "step": 77330
    },
    {
      "entropy": 5.595567321777343,
      "epoch": 7.428914505283381,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012074014122188762,
      "loss": 4.6417,
      "mean_token_accuracy": 0.2653343126177788,
      "num_tokens": 177288398.0,
      "step": 77335
    },
    {
      "entropy": 5.48531928062439,
      "epoch": 7.4293948126801155,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012071518530247711,
      "loss": 4.5566,
      "mean_token_accuracy": 0.273794549703598,
      "num_tokens": 177299341.0,
      "step": 77340
    },
    {
      "entropy": 5.520271444320679,
      "epoch": 7.429875120076849,
      "grad_norm": 1.671875,
      "learning_rate": 0.00012069023296510184,
      "loss": 4.5486,
      "mean_token_accuracy": 0.2783812299370766,
      "num_tokens": 177309287.0,
      "step": 77345
    },
    {
      "entropy": 5.519012451171875,
      "epoch": 7.430355427473583,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012066528421034125,
      "loss": 4.6374,
      "mean_token_accuracy": 0.26113116294145583,
      "num_tokens": 177321335.0,
      "step": 77350
    },
    {
      "entropy": 5.469300556182861,
      "epoch": 7.430835734870317,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00012064033903877441,
      "loss": 4.5436,
      "mean_token_accuracy": 0.275494784116745,
      "num_tokens": 177332161.0,
      "step": 77355
    },
    {
      "entropy": 5.573672914505005,
      "epoch": 7.431316042267051,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00012061539745098054,
      "loss": 4.6413,
      "mean_token_accuracy": 0.2688645914196968,
      "num_tokens": 177343343.0,
      "step": 77360
    },
    {
      "entropy": 5.556908226013183,
      "epoch": 7.431796349663784,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012059045944753869,
      "loss": 4.6138,
      "mean_token_accuracy": 0.2687139168381691,
      "num_tokens": 177354829.0,
      "step": 77365
    },
    {
      "entropy": 5.521419525146484,
      "epoch": 7.432276657060519,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012056552502902788,
      "loss": 4.5909,
      "mean_token_accuracy": 0.2687852710485458,
      "num_tokens": 177366274.0,
      "step": 77370
    },
    {
      "entropy": 5.529905462265015,
      "epoch": 7.432756964457253,
      "grad_norm": 1.609375,
      "learning_rate": 0.00012054059419602703,
      "loss": 4.666,
      "mean_token_accuracy": 0.25976079851388933,
      "num_tokens": 177377201.0,
      "step": 77375
    },
    {
      "entropy": 5.526751375198364,
      "epoch": 7.4332372718539865,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012051566694911492,
      "loss": 4.5869,
      "mean_token_accuracy": 0.2677151933312416,
      "num_tokens": 177388583.0,
      "step": 77380
    },
    {
      "entropy": 5.528580141067505,
      "epoch": 7.433717579250721,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001204907432888703,
      "loss": 4.6397,
      "mean_token_accuracy": 0.2694658741354942,
      "num_tokens": 177399057.0,
      "step": 77385
    },
    {
      "entropy": 5.598992872238159,
      "epoch": 7.434197886647454,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012046582321587187,
      "loss": 4.5993,
      "mean_token_accuracy": 0.266717204451561,
      "num_tokens": 177409179.0,
      "step": 77390
    },
    {
      "entropy": 5.539132881164551,
      "epoch": 7.434678194044189,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012044090673069811,
      "loss": 4.5983,
      "mean_token_accuracy": 0.26941029280424117,
      "num_tokens": 177420017.0,
      "step": 77395
    },
    {
      "entropy": 5.512806844711304,
      "epoch": 7.435158501440922,
      "grad_norm": 1.375,
      "learning_rate": 0.00012041599383392754,
      "loss": 4.6298,
      "mean_token_accuracy": 0.25367934852838514,
      "num_tokens": 177432957.0,
      "step": 77400
    },
    {
      "entropy": 5.443612194061279,
      "epoch": 7.435638808837656,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012039108452613856,
      "loss": 4.4997,
      "mean_token_accuracy": 0.28100008964538575,
      "num_tokens": 177443859.0,
      "step": 77405
    },
    {
      "entropy": 5.563088607788086,
      "epoch": 7.43611911623439,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012036617880790954,
      "loss": 4.6947,
      "mean_token_accuracy": 0.2631051644682884,
      "num_tokens": 177455873.0,
      "step": 77410
    },
    {
      "entropy": 5.486434745788574,
      "epoch": 7.436599423631124,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012034127667981864,
      "loss": 4.5502,
      "mean_token_accuracy": 0.27232208997011187,
      "num_tokens": 177467151.0,
      "step": 77415
    },
    {
      "entropy": 5.529820680618286,
      "epoch": 7.437079731027858,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012031637814244407,
      "loss": 4.5387,
      "mean_token_accuracy": 0.27699326872825625,
      "num_tokens": 177478291.0,
      "step": 77420
    },
    {
      "entropy": 5.532017040252685,
      "epoch": 7.437560038424592,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012029148319636389,
      "loss": 4.6265,
      "mean_token_accuracy": 0.26795136630535127,
      "num_tokens": 177491215.0,
      "step": 77425
    },
    {
      "entropy": 5.509214973449707,
      "epoch": 7.438040345821325,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012026659184215612,
      "loss": 4.6208,
      "mean_token_accuracy": 0.26550738960504533,
      "num_tokens": 177502075.0,
      "step": 77430
    },
    {
      "entropy": 5.540668201446533,
      "epoch": 7.43852065321806,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001202417040803986,
      "loss": 4.6218,
      "mean_token_accuracy": 0.26791299879550934,
      "num_tokens": 177513854.0,
      "step": 77435
    },
    {
      "entropy": 5.4925124168396,
      "epoch": 7.439000960614793,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012021681991166913,
      "loss": 4.5534,
      "mean_token_accuracy": 0.27197564840316774,
      "num_tokens": 177525135.0,
      "step": 77440
    },
    {
      "entropy": 5.539172601699829,
      "epoch": 7.439481268011527,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012019193933654553,
      "loss": 4.6057,
      "mean_token_accuracy": 0.26550271511077883,
      "num_tokens": 177536775.0,
      "step": 77445
    },
    {
      "entropy": 5.509156036376953,
      "epoch": 7.439961575408262,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012016706235560538,
      "loss": 4.5837,
      "mean_token_accuracy": 0.26708705723285675,
      "num_tokens": 177548595.0,
      "step": 77450
    },
    {
      "entropy": 5.527478504180908,
      "epoch": 7.440441882804995,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001201421889694263,
      "loss": 4.658,
      "mean_token_accuracy": 0.26180369555950167,
      "num_tokens": 177560271.0,
      "step": 77455
    },
    {
      "entropy": 5.464867067337036,
      "epoch": 7.4409221902017295,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012011731917858576,
      "loss": 4.558,
      "mean_token_accuracy": 0.26764669865369795,
      "num_tokens": 177571130.0,
      "step": 77460
    },
    {
      "entropy": 5.4945290088653564,
      "epoch": 7.441402497598463,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012009245298366115,
      "loss": 4.5282,
      "mean_token_accuracy": 0.27979364395141604,
      "num_tokens": 177582114.0,
      "step": 77465
    },
    {
      "entropy": 5.445371007919311,
      "epoch": 7.441882804995197,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012006759038522988,
      "loss": 4.5219,
      "mean_token_accuracy": 0.27540864050388336,
      "num_tokens": 177593752.0,
      "step": 77470
    },
    {
      "entropy": 5.5430199146270756,
      "epoch": 7.442363112391931,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012004273138386902,
      "loss": 4.6182,
      "mean_token_accuracy": 0.26390170603990554,
      "num_tokens": 177606582.0,
      "step": 77475
    },
    {
      "entropy": 5.523727512359619,
      "epoch": 7.442843419788665,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012001787598015583,
      "loss": 4.5908,
      "mean_token_accuracy": 0.27050958275794984,
      "num_tokens": 177619019.0,
      "step": 77480
    },
    {
      "entropy": 5.533151578903198,
      "epoch": 7.4433237271853985,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011999302417466736,
      "loss": 4.583,
      "mean_token_accuracy": 0.2660277232527733,
      "num_tokens": 177630495.0,
      "step": 77485
    },
    {
      "entropy": 5.593477201461792,
      "epoch": 7.443804034582133,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011996817596798058,
      "loss": 4.6368,
      "mean_token_accuracy": 0.2632936179637909,
      "num_tokens": 177643006.0,
      "step": 77490
    },
    {
      "entropy": 5.508943653106689,
      "epoch": 7.444284341978866,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011994333136067241,
      "loss": 4.5644,
      "mean_token_accuracy": 0.2683599203824997,
      "num_tokens": 177653615.0,
      "step": 77495
    },
    {
      "entropy": 5.612348461151123,
      "epoch": 7.444764649375601,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011991849035331965,
      "loss": 4.7193,
      "mean_token_accuracy": 0.25620150864124297,
      "num_tokens": 177666852.0,
      "step": 77500
    },
    {
      "entropy": 5.482387828826904,
      "epoch": 7.445244956772334,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011989365294649913,
      "loss": 4.5859,
      "mean_token_accuracy": 0.27114867269992826,
      "num_tokens": 177678296.0,
      "step": 77505
    },
    {
      "entropy": 5.527215433120728,
      "epoch": 7.445725264169068,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011986881914078734,
      "loss": 4.6649,
      "mean_token_accuracy": 0.2629913672804832,
      "num_tokens": 177690503.0,
      "step": 77510
    },
    {
      "entropy": 5.6098527908325195,
      "epoch": 7.446205571565802,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011984398893676093,
      "loss": 4.7151,
      "mean_token_accuracy": 0.25697618871927264,
      "num_tokens": 177701150.0,
      "step": 77515
    },
    {
      "entropy": 5.550714111328125,
      "epoch": 7.446685878962536,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011981916233499637,
      "loss": 4.6759,
      "mean_token_accuracy": 0.2649750754237175,
      "num_tokens": 177712680.0,
      "step": 77520
    },
    {
      "entropy": 5.484586000442505,
      "epoch": 7.4471661863592695,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011979433933607008,
      "loss": 4.5507,
      "mean_token_accuracy": 0.2751176953315735,
      "num_tokens": 177724087.0,
      "step": 77525
    },
    {
      "entropy": 5.501762247085571,
      "epoch": 7.447646493756004,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011976951994055835,
      "loss": 4.544,
      "mean_token_accuracy": 0.27030420303344727,
      "num_tokens": 177736103.0,
      "step": 77530
    },
    {
      "entropy": 5.567268896102905,
      "epoch": 7.448126801152737,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011974470414903751,
      "loss": 4.7436,
      "mean_token_accuracy": 0.26152022182941437,
      "num_tokens": 177749764.0,
      "step": 77535
    },
    {
      "entropy": 5.587038516998291,
      "epoch": 7.448607108549472,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011971989196208356,
      "loss": 4.7252,
      "mean_token_accuracy": 0.262559674680233,
      "num_tokens": 177760880.0,
      "step": 77540
    },
    {
      "entropy": 5.56146650314331,
      "epoch": 7.449087415946206,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011969508338027261,
      "loss": 4.6355,
      "mean_token_accuracy": 0.27005145847797396,
      "num_tokens": 177772643.0,
      "step": 77545
    },
    {
      "entropy": 5.537431383132935,
      "epoch": 7.449567723342939,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011967027840418069,
      "loss": 4.5748,
      "mean_token_accuracy": 0.2647475928068161,
      "num_tokens": 177783951.0,
      "step": 77550
    },
    {
      "entropy": 5.472904920578003,
      "epoch": 7.450048030739674,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011964547703438366,
      "loss": 4.5718,
      "mean_token_accuracy": 0.2641502723097801,
      "num_tokens": 177796755.0,
      "step": 77555
    },
    {
      "entropy": 5.571767377853393,
      "epoch": 7.450528338136407,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011962067927145733,
      "loss": 4.6524,
      "mean_token_accuracy": 0.2637211471796036,
      "num_tokens": 177809147.0,
      "step": 77560
    },
    {
      "entropy": 5.456592130661011,
      "epoch": 7.4510086455331415,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011959588511597754,
      "loss": 4.58,
      "mean_token_accuracy": 0.2732953026890755,
      "num_tokens": 177821163.0,
      "step": 77565
    },
    {
      "entropy": 5.498871088027954,
      "epoch": 7.451488952929875,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011957109456851976,
      "loss": 4.5534,
      "mean_token_accuracy": 0.27050175964832307,
      "num_tokens": 177831953.0,
      "step": 77570
    },
    {
      "entropy": 5.562982702255249,
      "epoch": 7.451969260326609,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011954630762965962,
      "loss": 4.6385,
      "mean_token_accuracy": 0.26943222135305406,
      "num_tokens": 177843316.0,
      "step": 77575
    },
    {
      "entropy": 5.530693244934082,
      "epoch": 7.452449567723343,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011952152429997264,
      "loss": 4.5797,
      "mean_token_accuracy": 0.27178299278020857,
      "num_tokens": 177854908.0,
      "step": 77580
    },
    {
      "entropy": 5.529840040206909,
      "epoch": 7.452929875120077,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011949674458003418,
      "loss": 4.5339,
      "mean_token_accuracy": 0.279093936085701,
      "num_tokens": 177865255.0,
      "step": 77585
    },
    {
      "entropy": 5.497775268554688,
      "epoch": 7.45341018251681,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011947196847041955,
      "loss": 4.5518,
      "mean_token_accuracy": 0.2705265864729881,
      "num_tokens": 177877431.0,
      "step": 77590
    },
    {
      "entropy": 5.563421821594238,
      "epoch": 7.453890489913545,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011944719597170407,
      "loss": 4.6792,
      "mean_token_accuracy": 0.2606868579983711,
      "num_tokens": 177887704.0,
      "step": 77595
    },
    {
      "entropy": 5.463846206665039,
      "epoch": 7.454370797310278,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011942242708446275,
      "loss": 4.5306,
      "mean_token_accuracy": 0.27424083948135375,
      "num_tokens": 177898192.0,
      "step": 77600
    },
    {
      "entropy": 5.500065660476684,
      "epoch": 7.4548511047070125,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011939766180927067,
      "loss": 4.6569,
      "mean_token_accuracy": 0.2604189604520798,
      "num_tokens": 177910656.0,
      "step": 77605
    },
    {
      "entropy": 5.492850589752197,
      "epoch": 7.455331412103746,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011937290014670287,
      "loss": 4.6221,
      "mean_token_accuracy": 0.2673456221818924,
      "num_tokens": 177923091.0,
      "step": 77610
    },
    {
      "entropy": 5.549402189254761,
      "epoch": 7.45581171950048,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001193481420973342,
      "loss": 4.5748,
      "mean_token_accuracy": 0.2700383022427559,
      "num_tokens": 177934544.0,
      "step": 77615
    },
    {
      "entropy": 5.5832602977752686,
      "epoch": 7.456292026897215,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011932338766173946,
      "loss": 4.6717,
      "mean_token_accuracy": 0.26185364425182345,
      "num_tokens": 177945355.0,
      "step": 77620
    },
    {
      "entropy": 5.522974157333374,
      "epoch": 7.456772334293948,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011929863684049348,
      "loss": 4.5662,
      "mean_token_accuracy": 0.2783267229795456,
      "num_tokens": 177957200.0,
      "step": 77625
    },
    {
      "entropy": 5.529144811630249,
      "epoch": 7.457252641690682,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011927388963417073,
      "loss": 4.598,
      "mean_token_accuracy": 0.2741657182574272,
      "num_tokens": 177967941.0,
      "step": 77630
    },
    {
      "entropy": 5.558833122253418,
      "epoch": 7.457732949087416,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011924914604334587,
      "loss": 4.642,
      "mean_token_accuracy": 0.2652043759822845,
      "num_tokens": 177979173.0,
      "step": 77635
    },
    {
      "entropy": 5.547058629989624,
      "epoch": 7.45821325648415,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011922440606859335,
      "loss": 4.6339,
      "mean_token_accuracy": 0.26333134472370145,
      "num_tokens": 177991342.0,
      "step": 77640
    },
    {
      "entropy": 5.474408960342407,
      "epoch": 7.458693563880884,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011919966971048756,
      "loss": 4.6064,
      "mean_token_accuracy": 0.2723473638296127,
      "num_tokens": 178002836.0,
      "step": 77645
    },
    {
      "entropy": 5.575658798217773,
      "epoch": 7.459173871277618,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001191749369696028,
      "loss": 4.6602,
      "mean_token_accuracy": 0.26545230597257613,
      "num_tokens": 178015174.0,
      "step": 77650
    },
    {
      "entropy": 5.48419303894043,
      "epoch": 7.459654178674351,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011915020784651338,
      "loss": 4.6076,
      "mean_token_accuracy": 0.2682896092534065,
      "num_tokens": 178026604.0,
      "step": 77655
    },
    {
      "entropy": 5.613191986083985,
      "epoch": 7.460134486071086,
      "grad_norm": 1.375,
      "learning_rate": 0.00011912548234179328,
      "loss": 4.7523,
      "mean_token_accuracy": 0.2650311917066574,
      "num_tokens": 178038936.0,
      "step": 77660
    },
    {
      "entropy": 5.569421195983887,
      "epoch": 7.460614793467819,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011910076045601662,
      "loss": 4.608,
      "mean_token_accuracy": 0.2718140795826912,
      "num_tokens": 178050484.0,
      "step": 77665
    },
    {
      "entropy": 5.546824645996094,
      "epoch": 7.461095100864553,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011907604218975738,
      "loss": 4.5357,
      "mean_token_accuracy": 0.2741492211818695,
      "num_tokens": 178061181.0,
      "step": 77670
    },
    {
      "entropy": 5.499111127853394,
      "epoch": 7.461575408261287,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011905132754358947,
      "loss": 4.6136,
      "mean_token_accuracy": 0.2653109699487686,
      "num_tokens": 178072602.0,
      "step": 77675
    },
    {
      "entropy": 5.558745288848877,
      "epoch": 7.462055715658021,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011902661651808663,
      "loss": 4.6782,
      "mean_token_accuracy": 0.2618281155824661,
      "num_tokens": 178083958.0,
      "step": 77680
    },
    {
      "entropy": 5.52507848739624,
      "epoch": 7.462536023054755,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011900190911382268,
      "loss": 4.608,
      "mean_token_accuracy": 0.2632731005549431,
      "num_tokens": 178095221.0,
      "step": 77685
    },
    {
      "entropy": 5.594831752777099,
      "epoch": 7.463016330451489,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011897720533137112,
      "loss": 4.71,
      "mean_token_accuracy": 0.25866654962301255,
      "num_tokens": 178106837.0,
      "step": 77690
    },
    {
      "entropy": 5.511279058456421,
      "epoch": 7.463496637848223,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011895250517130555,
      "loss": 4.553,
      "mean_token_accuracy": 0.2707278400659561,
      "num_tokens": 178118308.0,
      "step": 77695
    },
    {
      "entropy": 5.562321996688842,
      "epoch": 7.463976945244957,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011892780863419946,
      "loss": 4.625,
      "mean_token_accuracy": 0.2693986713886261,
      "num_tokens": 178127838.0,
      "step": 77700
    },
    {
      "entropy": 5.453148651123047,
      "epoch": 7.464457252641691,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001189031157206262,
      "loss": 4.5188,
      "mean_token_accuracy": 0.27196047455072403,
      "num_tokens": 178139646.0,
      "step": 77705
    },
    {
      "entropy": 5.522477006912231,
      "epoch": 7.4649375600384245,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001188784264311591,
      "loss": 4.6387,
      "mean_token_accuracy": 0.268601693212986,
      "num_tokens": 178150517.0,
      "step": 77710
    },
    {
      "entropy": 5.632756519317627,
      "epoch": 7.465417867435159,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011885374076637142,
      "loss": 4.789,
      "mean_token_accuracy": 0.2559943661093712,
      "num_tokens": 178162149.0,
      "step": 77715
    },
    {
      "entropy": 5.504691314697266,
      "epoch": 7.465898174831892,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011882905872683612,
      "loss": 4.6027,
      "mean_token_accuracy": 0.2723626971244812,
      "num_tokens": 178173693.0,
      "step": 77720
    },
    {
      "entropy": 5.465426588058472,
      "epoch": 7.4663784822286265,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011880438031312637,
      "loss": 4.5513,
      "mean_token_accuracy": 0.262085422873497,
      "num_tokens": 178184572.0,
      "step": 77725
    },
    {
      "entropy": 5.572068786621093,
      "epoch": 7.46685878962536,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001187797055258151,
      "loss": 4.6415,
      "mean_token_accuracy": 0.268338517844677,
      "num_tokens": 178196722.0,
      "step": 77730
    },
    {
      "entropy": 5.588969659805298,
      "epoch": 7.467339097022094,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011875503436547518,
      "loss": 4.656,
      "mean_token_accuracy": 0.2640321508049965,
      "num_tokens": 178207590.0,
      "step": 77735
    },
    {
      "entropy": 5.55291223526001,
      "epoch": 7.467819404418828,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011873036683267942,
      "loss": 4.6705,
      "mean_token_accuracy": 0.26754265278577805,
      "num_tokens": 178219250.0,
      "step": 77740
    },
    {
      "entropy": 5.502286195755005,
      "epoch": 7.468299711815562,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011870570292800058,
      "loss": 4.5805,
      "mean_token_accuracy": 0.2736409530043602,
      "num_tokens": 178230619.0,
      "step": 77745
    },
    {
      "entropy": 5.491658163070679,
      "epoch": 7.4687800192122955,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011868104265201115,
      "loss": 4.6122,
      "mean_token_accuracy": 0.26816707402467727,
      "num_tokens": 178241063.0,
      "step": 77750
    },
    {
      "entropy": 5.486335706710816,
      "epoch": 7.46926032660903,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011865638600528374,
      "loss": 4.5829,
      "mean_token_accuracy": 0.2680300146341324,
      "num_tokens": 178251200.0,
      "step": 77755
    },
    {
      "entropy": 5.584149169921875,
      "epoch": 7.469740634005763,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011863173298839079,
      "loss": 4.7072,
      "mean_token_accuracy": 0.2655773341655731,
      "num_tokens": 178264471.0,
      "step": 77760
    },
    {
      "entropy": 5.569104862213135,
      "epoch": 7.470220941402498,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001186070836019047,
      "loss": 4.5846,
      "mean_token_accuracy": 0.27371534109115603,
      "num_tokens": 178275379.0,
      "step": 77765
    },
    {
      "entropy": 5.5494927883148195,
      "epoch": 7.470701248799232,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011858243784639773,
      "loss": 4.5611,
      "mean_token_accuracy": 0.26673501282930373,
      "num_tokens": 178285568.0,
      "step": 77770
    },
    {
      "entropy": 5.589699411392212,
      "epoch": 7.471181556195965,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011855779572244216,
      "loss": 4.7255,
      "mean_token_accuracy": 0.25976099371910094,
      "num_tokens": 178297482.0,
      "step": 77775
    },
    {
      "entropy": 5.6191659450531,
      "epoch": 7.4716618635927,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011853315723060997,
      "loss": 4.7326,
      "mean_token_accuracy": 0.2551561281085014,
      "num_tokens": 178308685.0,
      "step": 77780
    },
    {
      "entropy": 5.522055673599243,
      "epoch": 7.472142170989433,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011850852237147325,
      "loss": 4.5856,
      "mean_token_accuracy": 0.2735472500324249,
      "num_tokens": 178320562.0,
      "step": 77785
    },
    {
      "entropy": 5.605477619171142,
      "epoch": 7.472622478386167,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011848389114560395,
      "loss": 4.7163,
      "mean_token_accuracy": 0.2577843427658081,
      "num_tokens": 178331194.0,
      "step": 77790
    },
    {
      "entropy": 5.558308839797974,
      "epoch": 7.473102785782901,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011845926355357394,
      "loss": 4.6097,
      "mean_token_accuracy": 0.26993207782506945,
      "num_tokens": 178343585.0,
      "step": 77795
    },
    {
      "entropy": 5.586913347244263,
      "epoch": 7.473583093179635,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011843463959595501,
      "loss": 4.6639,
      "mean_token_accuracy": 0.2676133468747139,
      "num_tokens": 178354514.0,
      "step": 77800
    },
    {
      "entropy": 5.591069221496582,
      "epoch": 7.474063400576369,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011841001927331893,
      "loss": 4.7286,
      "mean_token_accuracy": 0.2544096663594246,
      "num_tokens": 178365821.0,
      "step": 77805
    },
    {
      "entropy": 5.478343725204468,
      "epoch": 7.474543707973103,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011838540258623715,
      "loss": 4.5448,
      "mean_token_accuracy": 0.27529742419719694,
      "num_tokens": 178375800.0,
      "step": 77810
    },
    {
      "entropy": 5.579397773742675,
      "epoch": 7.475024015369836,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011836078953528125,
      "loss": 4.699,
      "mean_token_accuracy": 0.26001348346471786,
      "num_tokens": 178387482.0,
      "step": 77815
    },
    {
      "entropy": 5.490932893753052,
      "epoch": 7.475504322766571,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001183361801210227,
      "loss": 4.5735,
      "mean_token_accuracy": 0.2653514638543129,
      "num_tokens": 178399492.0,
      "step": 77820
    },
    {
      "entropy": 5.511504697799682,
      "epoch": 7.475984630163304,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00011831157434403286,
      "loss": 4.6087,
      "mean_token_accuracy": 0.2710350900888443,
      "num_tokens": 178411754.0,
      "step": 77825
    },
    {
      "entropy": 5.552990913391113,
      "epoch": 7.4764649375600385,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011828697220488299,
      "loss": 4.6537,
      "mean_token_accuracy": 0.26650360226631165,
      "num_tokens": 178423019.0,
      "step": 77830
    },
    {
      "entropy": 5.532704925537109,
      "epoch": 7.476945244956772,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011826237370414437,
      "loss": 4.6215,
      "mean_token_accuracy": 0.2577653005719185,
      "num_tokens": 178434170.0,
      "step": 77835
    },
    {
      "entropy": 5.590952730178833,
      "epoch": 7.477425552353506,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011823777884238792,
      "loss": 4.7075,
      "mean_token_accuracy": 0.26273461878299714,
      "num_tokens": 178444985.0,
      "step": 77840
    },
    {
      "entropy": 5.564501810073852,
      "epoch": 7.477905859750241,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011821318762018478,
      "loss": 4.5738,
      "mean_token_accuracy": 0.27104205042123797,
      "num_tokens": 178455702.0,
      "step": 77845
    },
    {
      "entropy": 5.5842643737792965,
      "epoch": 7.478386167146974,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011818860003810586,
      "loss": 4.6675,
      "mean_token_accuracy": 0.2656619057059288,
      "num_tokens": 178468122.0,
      "step": 77850
    },
    {
      "entropy": 5.481122303009033,
      "epoch": 7.478866474543708,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011816401609672201,
      "loss": 4.5421,
      "mean_token_accuracy": 0.2697379946708679,
      "num_tokens": 178479732.0,
      "step": 77855
    },
    {
      "entropy": 5.532065439224243,
      "epoch": 7.479346781940442,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011813943579660398,
      "loss": 4.6652,
      "mean_token_accuracy": 0.2618196696043015,
      "num_tokens": 178492125.0,
      "step": 77860
    },
    {
      "entropy": 5.532980918884277,
      "epoch": 7.479827089337176,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011811485913832255,
      "loss": 4.6166,
      "mean_token_accuracy": 0.2701090916991234,
      "num_tokens": 178503174.0,
      "step": 77865
    },
    {
      "entropy": 5.677554273605347,
      "epoch": 7.4803073967339095,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00011809028612244817,
      "loss": 4.7633,
      "mean_token_accuracy": 0.2525856629014015,
      "num_tokens": 178514803.0,
      "step": 77870
    },
    {
      "entropy": 5.546356868743897,
      "epoch": 7.480787704130644,
      "grad_norm": 1.59375,
      "learning_rate": 0.00011806571674955144,
      "loss": 4.6104,
      "mean_token_accuracy": 0.27196831405162813,
      "num_tokens": 178525985.0,
      "step": 77875
    },
    {
      "entropy": 5.533011817932129,
      "epoch": 7.481268011527377,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011804115102020272,
      "loss": 4.5562,
      "mean_token_accuracy": 0.26558915227651597,
      "num_tokens": 178537695.0,
      "step": 77880
    },
    {
      "entropy": 5.469656133651734,
      "epoch": 7.481748318924112,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00011801658893497243,
      "loss": 4.5454,
      "mean_token_accuracy": 0.27546420097351076,
      "num_tokens": 178549318.0,
      "step": 77885
    },
    {
      "entropy": 5.562287282943726,
      "epoch": 7.482228626320845,
      "grad_norm": 1.375,
      "learning_rate": 0.00011799203049443082,
      "loss": 4.6965,
      "mean_token_accuracy": 0.26135674566030503,
      "num_tokens": 178560471.0,
      "step": 77890
    },
    {
      "entropy": 5.63423867225647,
      "epoch": 7.482708933717579,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001179674756991481,
      "loss": 4.7693,
      "mean_token_accuracy": 0.2597285628318787,
      "num_tokens": 178574250.0,
      "step": 77895
    },
    {
      "entropy": 5.55570273399353,
      "epoch": 7.483189241114313,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011794292454969424,
      "loss": 4.5947,
      "mean_token_accuracy": 0.2743424952030182,
      "num_tokens": 178585216.0,
      "step": 77900
    },
    {
      "entropy": 5.596237659454346,
      "epoch": 7.483669548511047,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011791837704663929,
      "loss": 4.6353,
      "mean_token_accuracy": 0.2698796570301056,
      "num_tokens": 178595991.0,
      "step": 77905
    },
    {
      "entropy": 5.540353727340698,
      "epoch": 7.484149855907781,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011789383319055319,
      "loss": 4.6013,
      "mean_token_accuracy": 0.2647435873746872,
      "num_tokens": 178607771.0,
      "step": 77910
    },
    {
      "entropy": 5.512701463699341,
      "epoch": 7.484630163304515,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001178692929820058,
      "loss": 4.5552,
      "mean_token_accuracy": 0.27415467947721484,
      "num_tokens": 178618541.0,
      "step": 77915
    },
    {
      "entropy": 5.506408214569092,
      "epoch": 7.485110470701249,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001178447564215668,
      "loss": 4.6958,
      "mean_token_accuracy": 0.26217477172613146,
      "num_tokens": 178629914.0,
      "step": 77920
    },
    {
      "entropy": 5.481174039840698,
      "epoch": 7.485590778097983,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011782022350980601,
      "loss": 4.5852,
      "mean_token_accuracy": 0.2804880067706108,
      "num_tokens": 178641945.0,
      "step": 77925
    },
    {
      "entropy": 5.492533493041992,
      "epoch": 7.486071085494717,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011779569424729281,
      "loss": 4.5673,
      "mean_token_accuracy": 0.27453933358192445,
      "num_tokens": 178653007.0,
      "step": 77930
    },
    {
      "entropy": 5.547990322113037,
      "epoch": 7.48655139289145,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001177711686345968,
      "loss": 4.6443,
      "mean_token_accuracy": 0.2687486261129379,
      "num_tokens": 178664355.0,
      "step": 77935
    },
    {
      "entropy": 5.570827150344849,
      "epoch": 7.487031700288185,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011774664667228738,
      "loss": 4.7017,
      "mean_token_accuracy": 0.26032121330499647,
      "num_tokens": 178676543.0,
      "step": 77940
    },
    {
      "entropy": 5.49367618560791,
      "epoch": 7.487512007684918,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011772212836093388,
      "loss": 4.5467,
      "mean_token_accuracy": 0.2701458051800728,
      "num_tokens": 178688823.0,
      "step": 77945
    },
    {
      "entropy": 5.58208441734314,
      "epoch": 7.4879923150816525,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011769761370110551,
      "loss": 4.6852,
      "mean_token_accuracy": 0.2643438562750816,
      "num_tokens": 178700406.0,
      "step": 77950
    },
    {
      "entropy": 5.485629844665527,
      "epoch": 7.488472622478386,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001176731026933715,
      "loss": 4.5667,
      "mean_token_accuracy": 0.26643392592668536,
      "num_tokens": 178710799.0,
      "step": 77955
    },
    {
      "entropy": 5.56054916381836,
      "epoch": 7.48895292987512,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011764859533830087,
      "loss": 4.6402,
      "mean_token_accuracy": 0.26958245784044266,
      "num_tokens": 178722788.0,
      "step": 77960
    },
    {
      "entropy": 5.553530025482178,
      "epoch": 7.489433237271854,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011762409163646267,
      "loss": 4.62,
      "mean_token_accuracy": 0.2672366738319397,
      "num_tokens": 178734629.0,
      "step": 77965
    },
    {
      "entropy": 5.545094156265259,
      "epoch": 7.489913544668588,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011759959158842572,
      "loss": 4.6628,
      "mean_token_accuracy": 0.256755430996418,
      "num_tokens": 178746267.0,
      "step": 77970
    },
    {
      "entropy": 5.550877618789673,
      "epoch": 7.4903938520653215,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011757509519475884,
      "loss": 4.6167,
      "mean_token_accuracy": 0.2699581295251846,
      "num_tokens": 178756931.0,
      "step": 77975
    },
    {
      "entropy": 5.564462947845459,
      "epoch": 7.490874159462056,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001175506024560308,
      "loss": 4.6621,
      "mean_token_accuracy": 0.2598510071635246,
      "num_tokens": 178769786.0,
      "step": 77980
    },
    {
      "entropy": 5.491206645965576,
      "epoch": 7.491354466858789,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011752611337281023,
      "loss": 4.5588,
      "mean_token_accuracy": 0.2755330890417099,
      "num_tokens": 178780742.0,
      "step": 77985
    },
    {
      "entropy": 5.558912658691407,
      "epoch": 7.491834774255524,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001175016279456657,
      "loss": 4.7011,
      "mean_token_accuracy": 0.25940436869859695,
      "num_tokens": 178792674.0,
      "step": 77990
    },
    {
      "entropy": 5.517195081710815,
      "epoch": 7.492315081652258,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011747714617516571,
      "loss": 4.5914,
      "mean_token_accuracy": 0.26696107983589173,
      "num_tokens": 178804889.0,
      "step": 77995
    },
    {
      "entropy": 5.501047897338867,
      "epoch": 7.492795389048991,
      "grad_norm": 1.53125,
      "learning_rate": 0.00011745266806187864,
      "loss": 4.5355,
      "mean_token_accuracy": 0.27514599412679674,
      "num_tokens": 178816084.0,
      "step": 78000
    },
    {
      "epoch": 7.492795389048991,
      "eval_entropy": 5.2420047689565274,
      "eval_loss": 4.742284297943115,
      "eval_mean_token_accuracy": 0.26790906561821015,
      "eval_num_tokens": 178816084.0,
      "eval_runtime": 20.6222,
      "eval_samples_per_second": 1591.248,
      "eval_steps_per_second": 198.912,
      "step": 78000
    },
    {
      "entropy": 5.487389135360718,
      "epoch": 7.493275696445726,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011742819360637289,
      "loss": 4.6214,
      "mean_token_accuracy": 0.26713291108608245,
      "num_tokens": 178828124.0,
      "step": 78005
    },
    {
      "entropy": 5.483064031600952,
      "epoch": 7.493756003842459,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011740372280921649,
      "loss": 4.5559,
      "mean_token_accuracy": 0.2734342575073242,
      "num_tokens": 178839335.0,
      "step": 78010
    },
    {
      "entropy": 5.60186996459961,
      "epoch": 7.494236311239193,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011737925567097769,
      "loss": 4.7019,
      "mean_token_accuracy": 0.2609479263424873,
      "num_tokens": 178850494.0,
      "step": 78015
    },
    {
      "entropy": 5.543515872955322,
      "epoch": 7.494716618635927,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011735479219222455,
      "loss": 4.6153,
      "mean_token_accuracy": 0.2674731403589249,
      "num_tokens": 178861171.0,
      "step": 78020
    },
    {
      "entropy": 5.557148838043213,
      "epoch": 7.495196926032661,
      "grad_norm": 1.671875,
      "learning_rate": 0.000117330332373525,
      "loss": 4.5812,
      "mean_token_accuracy": 0.267902472615242,
      "num_tokens": 178871480.0,
      "step": 78025
    },
    {
      "entropy": 5.556967687606812,
      "epoch": 7.495677233429395,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011730587621544696,
      "loss": 4.619,
      "mean_token_accuracy": 0.2707656189799309,
      "num_tokens": 178883187.0,
      "step": 78030
    },
    {
      "entropy": 5.52000207901001,
      "epoch": 7.496157540826129,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011728142371855822,
      "loss": 4.6186,
      "mean_token_accuracy": 0.2714358612895012,
      "num_tokens": 178894847.0,
      "step": 78035
    },
    {
      "entropy": 5.543948602676392,
      "epoch": 7.496637848222862,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001172569748834265,
      "loss": 4.6231,
      "mean_token_accuracy": 0.26732450872659685,
      "num_tokens": 178906826.0,
      "step": 78040
    },
    {
      "entropy": 5.5186604976654055,
      "epoch": 7.497118155619597,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001172325297106195,
      "loss": 4.5897,
      "mean_token_accuracy": 0.26146008670330045,
      "num_tokens": 178918210.0,
      "step": 78045
    },
    {
      "entropy": 5.489210367202759,
      "epoch": 7.49759846301633,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001172080882007046,
      "loss": 4.5498,
      "mean_token_accuracy": 0.27503417432308197,
      "num_tokens": 178929476.0,
      "step": 78050
    },
    {
      "entropy": 5.52728304862976,
      "epoch": 7.4980787704130645,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011718365035424934,
      "loss": 4.6075,
      "mean_token_accuracy": 0.2648928686976433,
      "num_tokens": 178942033.0,
      "step": 78055
    },
    {
      "entropy": 5.512071657180786,
      "epoch": 7.498559077809798,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011715921617182112,
      "loss": 4.5828,
      "mean_token_accuracy": 0.2669952854514122,
      "num_tokens": 178954285.0,
      "step": 78060
    },
    {
      "entropy": 5.536767482757568,
      "epoch": 7.499039385206532,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001171347856539872,
      "loss": 4.6307,
      "mean_token_accuracy": 0.26664677560329436,
      "num_tokens": 178965825.0,
      "step": 78065
    },
    {
      "entropy": 5.506334257125855,
      "epoch": 7.499519692603266,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001171103588013148,
      "loss": 4.5845,
      "mean_token_accuracy": 0.2701448142528534,
      "num_tokens": 178977112.0,
      "step": 78070
    },
    {
      "entropy": 5.569579315185547,
      "epoch": 7.5,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001170859356143711,
      "loss": 4.6829,
      "mean_token_accuracy": 0.2609786674380302,
      "num_tokens": 178988870.0,
      "step": 78075
    },
    {
      "entropy": 5.510657501220703,
      "epoch": 7.500480307396733,
      "grad_norm": 1.359375,
      "learning_rate": 0.000117061516093723,
      "loss": 4.5736,
      "mean_token_accuracy": 0.2698140576481819,
      "num_tokens": 178999908.0,
      "step": 78080
    },
    {
      "entropy": 5.548113393783569,
      "epoch": 7.500960614793468,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011703710023993751,
      "loss": 4.6402,
      "mean_token_accuracy": 0.26933884620666504,
      "num_tokens": 179011572.0,
      "step": 78085
    },
    {
      "entropy": 5.52212815284729,
      "epoch": 7.501440922190202,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001170126880535815,
      "loss": 4.5692,
      "mean_token_accuracy": 0.26407052874565123,
      "num_tokens": 179021802.0,
      "step": 78090
    },
    {
      "entropy": 5.557010746002197,
      "epoch": 7.5019212295869355,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011698827953522176,
      "loss": 4.66,
      "mean_token_accuracy": 0.26290650814771654,
      "num_tokens": 179032885.0,
      "step": 78095
    },
    {
      "entropy": 5.474430513381958,
      "epoch": 7.50240153698367,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011696387468542493,
      "loss": 4.5736,
      "mean_token_accuracy": 0.27011816650629045,
      "num_tokens": 179045077.0,
      "step": 78100
    },
    {
      "entropy": 5.485526037216187,
      "epoch": 7.502881844380403,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011693947350475777,
      "loss": 4.5588,
      "mean_token_accuracy": 0.26669475585222246,
      "num_tokens": 179055868.0,
      "step": 78105
    },
    {
      "entropy": 5.567382860183716,
      "epoch": 7.503362151777138,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011691507599378662,
      "loss": 4.6821,
      "mean_token_accuracy": 0.26398793011903765,
      "num_tokens": 179067213.0,
      "step": 78110
    },
    {
      "entropy": 5.571899461746216,
      "epoch": 7.503842459173871,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011689068215307796,
      "loss": 4.6261,
      "mean_token_accuracy": 0.25749070942401886,
      "num_tokens": 179078248.0,
      "step": 78115
    },
    {
      "entropy": 5.576285886764526,
      "epoch": 7.504322766570605,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001168662919831982,
      "loss": 4.6423,
      "mean_token_accuracy": 0.2630542665719986,
      "num_tokens": 179089642.0,
      "step": 78120
    },
    {
      "entropy": 5.488516092300415,
      "epoch": 7.504803073967339,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011684190548471356,
      "loss": 4.552,
      "mean_token_accuracy": 0.2784141585230827,
      "num_tokens": 179100224.0,
      "step": 78125
    },
    {
      "entropy": 5.508522653579712,
      "epoch": 7.505283381364073,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011681752265819026,
      "loss": 4.5642,
      "mean_token_accuracy": 0.2730681523680687,
      "num_tokens": 179111855.0,
      "step": 78130
    },
    {
      "entropy": 5.623974370956421,
      "epoch": 7.505763688760807,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011679314350419444,
      "loss": 4.757,
      "mean_token_accuracy": 0.2604340597987175,
      "num_tokens": 179123803.0,
      "step": 78135
    },
    {
      "entropy": 5.501619625091553,
      "epoch": 7.506243996157541,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011676876802329198,
      "loss": 4.5736,
      "mean_token_accuracy": 0.2704528018832207,
      "num_tokens": 179136101.0,
      "step": 78140
    },
    {
      "entropy": 5.501300191879272,
      "epoch": 7.506724303554275,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011674439621604888,
      "loss": 4.5674,
      "mean_token_accuracy": 0.2726980924606323,
      "num_tokens": 179146576.0,
      "step": 78145
    },
    {
      "entropy": 5.491797828674317,
      "epoch": 7.507204610951009,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011672002808303097,
      "loss": 4.5724,
      "mean_token_accuracy": 0.2725539982318878,
      "num_tokens": 179157846.0,
      "step": 78150
    },
    {
      "entropy": 5.57291579246521,
      "epoch": 7.507684918347742,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011669566362480403,
      "loss": 4.6765,
      "mean_token_accuracy": 0.267888106405735,
      "num_tokens": 179169822.0,
      "step": 78155
    },
    {
      "entropy": 5.5361837387084964,
      "epoch": 7.508165225744476,
      "grad_norm": 1.28125,
      "learning_rate": 0.00011667130284193368,
      "loss": 4.64,
      "mean_token_accuracy": 0.2589723750948906,
      "num_tokens": 179180188.0,
      "step": 78160
    },
    {
      "entropy": 5.552768278121948,
      "epoch": 7.508645533141211,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011664694573498564,
      "loss": 4.6742,
      "mean_token_accuracy": 0.25949864983558657,
      "num_tokens": 179191190.0,
      "step": 78165
    },
    {
      "entropy": 5.602953910827637,
      "epoch": 7.509125840537944,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001166225923045252,
      "loss": 4.746,
      "mean_token_accuracy": 0.25696408897638323,
      "num_tokens": 179202944.0,
      "step": 78170
    },
    {
      "entropy": 5.570067453384399,
      "epoch": 7.5096061479346785,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011659824255111793,
      "loss": 4.5746,
      "mean_token_accuracy": 0.2618630021810532,
      "num_tokens": 179214434.0,
      "step": 78175
    },
    {
      "entropy": 5.6143640041351315,
      "epoch": 7.510086455331412,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011657389647532908,
      "loss": 4.7075,
      "mean_token_accuracy": 0.2604984328150749,
      "num_tokens": 179226255.0,
      "step": 78180
    },
    {
      "entropy": 5.64035849571228,
      "epoch": 7.510566762728146,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011654955407772395,
      "loss": 4.7795,
      "mean_token_accuracy": 0.2581579312682152,
      "num_tokens": 179238789.0,
      "step": 78185
    },
    {
      "entropy": 5.495444202423096,
      "epoch": 7.51104707012488,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011652521535886766,
      "loss": 4.5157,
      "mean_token_accuracy": 0.27664482444524763,
      "num_tokens": 179249523.0,
      "step": 78190
    },
    {
      "entropy": 5.501522254943848,
      "epoch": 7.511527377521614,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011650088031932537,
      "loss": 4.5455,
      "mean_token_accuracy": 0.26980644166469575,
      "num_tokens": 179260177.0,
      "step": 78195
    },
    {
      "entropy": 5.585860681533814,
      "epoch": 7.5120076849183475,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011647654895966191,
      "loss": 4.6829,
      "mean_token_accuracy": 0.26773969233036043,
      "num_tokens": 179270967.0,
      "step": 78200
    },
    {
      "entropy": 5.494057703018188,
      "epoch": 7.512487992315082,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011645222128044227,
      "loss": 4.5916,
      "mean_token_accuracy": 0.2651565715670586,
      "num_tokens": 179282206.0,
      "step": 78205
    },
    {
      "entropy": 5.501889038085937,
      "epoch": 7.512968299711815,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011642789728223127,
      "loss": 4.6044,
      "mean_token_accuracy": 0.2643053710460663,
      "num_tokens": 179293893.0,
      "step": 78210
    },
    {
      "entropy": 5.602098560333252,
      "epoch": 7.51344860710855,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011640357696559363,
      "loss": 4.6321,
      "mean_token_accuracy": 0.2661884129047394,
      "num_tokens": 179304828.0,
      "step": 78215
    },
    {
      "entropy": 5.532146692276001,
      "epoch": 7.513928914505283,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000116379260331094,
      "loss": 4.5794,
      "mean_token_accuracy": 0.2730631649494171,
      "num_tokens": 179315821.0,
      "step": 78220
    },
    {
      "entropy": 5.643733501434326,
      "epoch": 7.514409221902017,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011635494737929699,
      "loss": 4.6913,
      "mean_token_accuracy": 0.26450350880622864,
      "num_tokens": 179327592.0,
      "step": 78225
    },
    {
      "entropy": 5.5523582935333256,
      "epoch": 7.514889529298751,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011633063811076696,
      "loss": 4.6066,
      "mean_token_accuracy": 0.2644233569502831,
      "num_tokens": 179339181.0,
      "step": 78230
    },
    {
      "entropy": 5.477328729629517,
      "epoch": 7.515369836695485,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011630633252606837,
      "loss": 4.5711,
      "mean_token_accuracy": 0.269455087184906,
      "num_tokens": 179351186.0,
      "step": 78235
    },
    {
      "entropy": 5.501113700866699,
      "epoch": 7.515850144092219,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00011628203062576554,
      "loss": 4.5257,
      "mean_token_accuracy": 0.2794559746980667,
      "num_tokens": 179362501.0,
      "step": 78240
    },
    {
      "entropy": 5.598853302001953,
      "epoch": 7.516330451488953,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011625773241042263,
      "loss": 4.7122,
      "mean_token_accuracy": 0.25746507197618484,
      "num_tokens": 179374634.0,
      "step": 78245
    },
    {
      "entropy": 5.542302370071411,
      "epoch": 7.516810758885687,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011623343788060382,
      "loss": 4.5685,
      "mean_token_accuracy": 0.2743497177958488,
      "num_tokens": 179386350.0,
      "step": 78250
    },
    {
      "entropy": 5.551734924316406,
      "epoch": 7.517291066282421,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001162091470368732,
      "loss": 4.643,
      "mean_token_accuracy": 0.2656000554561615,
      "num_tokens": 179397584.0,
      "step": 78255
    },
    {
      "entropy": 5.553012657165527,
      "epoch": 7.517771373679155,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001161848598797946,
      "loss": 4.6174,
      "mean_token_accuracy": 0.2721098601818085,
      "num_tokens": 179408599.0,
      "step": 78260
    },
    {
      "entropy": 5.493256330490112,
      "epoch": 7.518251681075888,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011616057640993199,
      "loss": 4.6045,
      "mean_token_accuracy": 0.26889567375183104,
      "num_tokens": 179420791.0,
      "step": 78265
    },
    {
      "entropy": 5.598184061050415,
      "epoch": 7.518731988472623,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011613629662784914,
      "loss": 4.6547,
      "mean_token_accuracy": 0.26367377042770385,
      "num_tokens": 179431636.0,
      "step": 78270
    },
    {
      "entropy": 5.556986951828003,
      "epoch": 7.519212295869356,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011611202053410973,
      "loss": 4.605,
      "mean_token_accuracy": 0.2709451600909233,
      "num_tokens": 179443851.0,
      "step": 78275
    },
    {
      "entropy": 5.476762962341309,
      "epoch": 7.5196926032660905,
      "grad_norm": 1.375,
      "learning_rate": 0.0001160877481292774,
      "loss": 4.5887,
      "mean_token_accuracy": 0.2706350237131119,
      "num_tokens": 179455372.0,
      "step": 78280
    },
    {
      "entropy": 5.511824035644532,
      "epoch": 7.520172910662824,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001160634794139158,
      "loss": 4.6063,
      "mean_token_accuracy": 0.27220906019210817,
      "num_tokens": 179466292.0,
      "step": 78285
    },
    {
      "entropy": 5.515906810760498,
      "epoch": 7.520653218059558,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011603921438858815,
      "loss": 4.6207,
      "mean_token_accuracy": 0.2656466782093048,
      "num_tokens": 179476997.0,
      "step": 78290
    },
    {
      "entropy": 5.494460773468018,
      "epoch": 7.521133525456292,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011601495305385794,
      "loss": 4.5316,
      "mean_token_accuracy": 0.27734531462192535,
      "num_tokens": 179487372.0,
      "step": 78295
    },
    {
      "entropy": 5.436023759841919,
      "epoch": 7.521613832853026,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011599069541028843,
      "loss": 4.4856,
      "mean_token_accuracy": 0.27919953167438505,
      "num_tokens": 179497722.0,
      "step": 78300
    },
    {
      "entropy": 5.507812070846557,
      "epoch": 7.522094140249759,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011596644145844282,
      "loss": 4.6126,
      "mean_token_accuracy": 0.26965736597776413,
      "num_tokens": 179510151.0,
      "step": 78305
    },
    {
      "entropy": 5.474165534973144,
      "epoch": 7.522574447646494,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011594219119888421,
      "loss": 4.5667,
      "mean_token_accuracy": 0.27516586184501646,
      "num_tokens": 179521073.0,
      "step": 78310
    },
    {
      "entropy": 5.58180422782898,
      "epoch": 7.523054755043228,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011591794463217567,
      "loss": 4.63,
      "mean_token_accuracy": 0.2687695547938347,
      "num_tokens": 179532400.0,
      "step": 78315
    },
    {
      "entropy": 5.590230941772461,
      "epoch": 7.5235350624399615,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011589370175888002,
      "loss": 4.662,
      "mean_token_accuracy": 0.26544856280088425,
      "num_tokens": 179543759.0,
      "step": 78320
    },
    {
      "entropy": 5.521355867385864,
      "epoch": 7.524015369836696,
      "grad_norm": 1.375,
      "learning_rate": 0.00011586946257956016,
      "loss": 4.5426,
      "mean_token_accuracy": 0.27661575973033903,
      "num_tokens": 179554962.0,
      "step": 78325
    },
    {
      "entropy": 5.576153182983399,
      "epoch": 7.524495677233429,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011584522709477885,
      "loss": 4.7112,
      "mean_token_accuracy": 0.25604735910892484,
      "num_tokens": 179566796.0,
      "step": 78330
    },
    {
      "entropy": 5.576178407669067,
      "epoch": 7.524975984630164,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001158209953050988,
      "loss": 4.6677,
      "mean_token_accuracy": 0.26331147998571397,
      "num_tokens": 179578361.0,
      "step": 78335
    },
    {
      "entropy": 5.556678724288941,
      "epoch": 7.525456292026897,
      "grad_norm": 1.375,
      "learning_rate": 0.00011579676721108256,
      "loss": 4.6449,
      "mean_token_accuracy": 0.27282922714948654,
      "num_tokens": 179589690.0,
      "step": 78340
    },
    {
      "entropy": 5.504484367370606,
      "epoch": 7.525936599423631,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001157725428132927,
      "loss": 4.6519,
      "mean_token_accuracy": 0.2592116311192513,
      "num_tokens": 179601930.0,
      "step": 78345
    },
    {
      "entropy": 5.596656560897827,
      "epoch": 7.526416906820365,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011574832211229154,
      "loss": 4.7351,
      "mean_token_accuracy": 0.26063770800828934,
      "num_tokens": 179613770.0,
      "step": 78350
    },
    {
      "entropy": 5.4916832447052,
      "epoch": 7.526897214217099,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011572410510864147,
      "loss": 4.5753,
      "mean_token_accuracy": 0.26149909049272535,
      "num_tokens": 179624952.0,
      "step": 78355
    },
    {
      "entropy": 5.557597970962524,
      "epoch": 7.527377521613833,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011569989180290471,
      "loss": 4.6416,
      "mean_token_accuracy": 0.2641674026846886,
      "num_tokens": 179636855.0,
      "step": 78360
    },
    {
      "entropy": 5.53564863204956,
      "epoch": 7.527857829010567,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011567568219564342,
      "loss": 4.5806,
      "mean_token_accuracy": 0.26840720772743226,
      "num_tokens": 179648156.0,
      "step": 78365
    },
    {
      "entropy": 5.583444833755493,
      "epoch": 7.5283381364073,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011565147628741972,
      "loss": 4.6871,
      "mean_token_accuracy": 0.2662746459245682,
      "num_tokens": 179660342.0,
      "step": 78370
    },
    {
      "entropy": 5.657597160339355,
      "epoch": 7.528818443804035,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011562727407879564,
      "loss": 4.7651,
      "mean_token_accuracy": 0.24994631260633468,
      "num_tokens": 179670701.0,
      "step": 78375
    },
    {
      "entropy": 5.6036299705505375,
      "epoch": 7.529298751200768,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011560307557033295,
      "loss": 4.7704,
      "mean_token_accuracy": 0.2596877411007881,
      "num_tokens": 179682766.0,
      "step": 78380
    },
    {
      "entropy": 5.58765664100647,
      "epoch": 7.529779058597502,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011557888076259352,
      "loss": 4.6569,
      "mean_token_accuracy": 0.265694397687912,
      "num_tokens": 179695661.0,
      "step": 78385
    },
    {
      "entropy": 5.543644380569458,
      "epoch": 7.530259365994237,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011555468965613911,
      "loss": 4.5556,
      "mean_token_accuracy": 0.273456284403801,
      "num_tokens": 179708018.0,
      "step": 78390
    },
    {
      "entropy": 5.533721303939819,
      "epoch": 7.53073967339097,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011553050225153135,
      "loss": 4.6283,
      "mean_token_accuracy": 0.2631498083472252,
      "num_tokens": 179720662.0,
      "step": 78395
    },
    {
      "entropy": 5.5130220413208,
      "epoch": 7.5312199807877045,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011550631854933178,
      "loss": 4.5863,
      "mean_token_accuracy": 0.2695863083004951,
      "num_tokens": 179732736.0,
      "step": 78400
    },
    {
      "entropy": 5.489157342910767,
      "epoch": 7.531700288184438,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000115482138550102,
      "loss": 4.5847,
      "mean_token_accuracy": 0.26562894880771637,
      "num_tokens": 179745913.0,
      "step": 78405
    },
    {
      "entropy": 5.475177049636841,
      "epoch": 7.532180595581172,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011545796225440321,
      "loss": 4.5477,
      "mean_token_accuracy": 0.2734752371907234,
      "num_tokens": 179757255.0,
      "step": 78410
    },
    {
      "entropy": 5.533751535415649,
      "epoch": 7.532660902977906,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011543378966279677,
      "loss": 4.624,
      "mean_token_accuracy": 0.2579867348074913,
      "num_tokens": 179769076.0,
      "step": 78415
    },
    {
      "entropy": 5.546472787857056,
      "epoch": 7.53314121037464,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011540962077584394,
      "loss": 4.6798,
      "mean_token_accuracy": 0.2634598881006241,
      "num_tokens": 179780506.0,
      "step": 78420
    },
    {
      "entropy": 5.599309778213501,
      "epoch": 7.5336215177713735,
      "grad_norm": 1.75,
      "learning_rate": 0.00011538545559410582,
      "loss": 4.6634,
      "mean_token_accuracy": 0.26325574666261675,
      "num_tokens": 179790898.0,
      "step": 78425
    },
    {
      "entropy": 5.529374647140503,
      "epoch": 7.534101825168108,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011536129411814346,
      "loss": 4.6294,
      "mean_token_accuracy": 0.26721516251564026,
      "num_tokens": 179802165.0,
      "step": 78430
    },
    {
      "entropy": 5.611709022521973,
      "epoch": 7.534582132564841,
      "grad_norm": 1.5,
      "learning_rate": 0.00011533713634851792,
      "loss": 4.7666,
      "mean_token_accuracy": 0.2513264328241348,
      "num_tokens": 179812684.0,
      "step": 78435
    },
    {
      "entropy": 5.596878290176392,
      "epoch": 7.5350624399615755,
      "grad_norm": 1.5,
      "learning_rate": 0.00011531298228578985,
      "loss": 4.6982,
      "mean_token_accuracy": 0.26163673102855683,
      "num_tokens": 179825365.0,
      "step": 78440
    },
    {
      "entropy": 5.539650917053223,
      "epoch": 7.535542747358309,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011528883193052021,
      "loss": 4.5908,
      "mean_token_accuracy": 0.2708847403526306,
      "num_tokens": 179837416.0,
      "step": 78445
    },
    {
      "entropy": 5.534697484970093,
      "epoch": 7.536023054755043,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011526468528326961,
      "loss": 4.6962,
      "mean_token_accuracy": 0.2598336473107338,
      "num_tokens": 179848942.0,
      "step": 78450
    },
    {
      "entropy": 5.524177312850952,
      "epoch": 7.536503362151777,
      "grad_norm": 1.28125,
      "learning_rate": 0.00011524054234459873,
      "loss": 4.6255,
      "mean_token_accuracy": 0.268820983171463,
      "num_tokens": 179860405.0,
      "step": 78455
    },
    {
      "entropy": 5.5633703708648685,
      "epoch": 7.536983669548511,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011521640311506806,
      "loss": 4.6484,
      "mean_token_accuracy": 0.2662363171577454,
      "num_tokens": 179872980.0,
      "step": 78460
    },
    {
      "entropy": 5.5104810237884525,
      "epoch": 7.537463976945245,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011519226759523813,
      "loss": 4.6154,
      "mean_token_accuracy": 0.2703523188829422,
      "num_tokens": 179885476.0,
      "step": 78465
    },
    {
      "entropy": 5.482981729507446,
      "epoch": 7.537944284341979,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001151681357856691,
      "loss": 4.5185,
      "mean_token_accuracy": 0.27732882350683213,
      "num_tokens": 179896763.0,
      "step": 78470
    },
    {
      "entropy": 5.409803342819214,
      "epoch": 7.538424591738713,
      "grad_norm": 1.5,
      "learning_rate": 0.00011514400768692142,
      "loss": 4.5157,
      "mean_token_accuracy": 0.2807216390967369,
      "num_tokens": 179908453.0,
      "step": 78475
    },
    {
      "entropy": 5.6198090553283695,
      "epoch": 7.538904899135447,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011511988329955516,
      "loss": 4.686,
      "mean_token_accuracy": 0.2631077066063881,
      "num_tokens": 179918702.0,
      "step": 78480
    },
    {
      "entropy": 5.46167402267456,
      "epoch": 7.539385206532181,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011509576262413048,
      "loss": 4.5477,
      "mean_token_accuracy": 0.2809975355863571,
      "num_tokens": 179929247.0,
      "step": 78485
    },
    {
      "entropy": 5.45731520652771,
      "epoch": 7.539865513928914,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011507164566120737,
      "loss": 4.5587,
      "mean_token_accuracy": 0.27122727781534195,
      "num_tokens": 179940563.0,
      "step": 78490
    },
    {
      "entropy": 5.535559368133545,
      "epoch": 7.540345821325649,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011504753241134575,
      "loss": 4.6782,
      "mean_token_accuracy": 0.2564093768596649,
      "num_tokens": 179951561.0,
      "step": 78495
    },
    {
      "entropy": 5.430089235305786,
      "epoch": 7.540826128722382,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011502342287510556,
      "loss": 4.5161,
      "mean_token_accuracy": 0.2762096017599106,
      "num_tokens": 179964487.0,
      "step": 78500
    },
    {
      "entropy": 5.460230159759521,
      "epoch": 7.541306436119116,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011499931705304636,
      "loss": 4.522,
      "mean_token_accuracy": 0.27808583676815035,
      "num_tokens": 179975982.0,
      "step": 78505
    },
    {
      "entropy": 5.570805549621582,
      "epoch": 7.54178674351585,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011497521494572794,
      "loss": 4.5956,
      "mean_token_accuracy": 0.26613897979259493,
      "num_tokens": 179986521.0,
      "step": 78510
    },
    {
      "entropy": 5.546423721313476,
      "epoch": 7.542267050912584,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011495111655370983,
      "loss": 4.6449,
      "mean_token_accuracy": 0.26181812435388563,
      "num_tokens": 179998557.0,
      "step": 78515
    },
    {
      "entropy": 5.510932397842407,
      "epoch": 7.542747358309318,
      "grad_norm": 1.375,
      "learning_rate": 0.00011492702187755156,
      "loss": 4.6296,
      "mean_token_accuracy": 0.2660122960805893,
      "num_tokens": 180009752.0,
      "step": 78520
    },
    {
      "entropy": 5.532959604263306,
      "epoch": 7.543227665706052,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011490293091781253,
      "loss": 4.6308,
      "mean_token_accuracy": 0.2673857122659683,
      "num_tokens": 180022074.0,
      "step": 78525
    },
    {
      "entropy": 5.6329957962036135,
      "epoch": 7.543707973102785,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011487884367505203,
      "loss": 4.6696,
      "mean_token_accuracy": 0.2607701286673546,
      "num_tokens": 180033546.0,
      "step": 78530
    },
    {
      "entropy": 5.564144134521484,
      "epoch": 7.54418828049952,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011485476014982933,
      "loss": 4.6548,
      "mean_token_accuracy": 0.26404875218868257,
      "num_tokens": 180044618.0,
      "step": 78535
    },
    {
      "entropy": 5.5360606670379635,
      "epoch": 7.544668587896254,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011483068034270364,
      "loss": 4.6426,
      "mean_token_accuracy": 0.2708636909723282,
      "num_tokens": 180056907.0,
      "step": 78540
    },
    {
      "entropy": 5.471135997772217,
      "epoch": 7.5451488952929875,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011480660425423387,
      "loss": 4.5901,
      "mean_token_accuracy": 0.2678396224975586,
      "num_tokens": 180068713.0,
      "step": 78545
    },
    {
      "entropy": 5.554118061065674,
      "epoch": 7.545629202689721,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011478253188497908,
      "loss": 4.6208,
      "mean_token_accuracy": 0.2674970611929893,
      "num_tokens": 180079244.0,
      "step": 78550
    },
    {
      "entropy": 5.51852240562439,
      "epoch": 7.546109510086455,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011475846323549815,
      "loss": 4.6897,
      "mean_token_accuracy": 0.26703097820281985,
      "num_tokens": 180090026.0,
      "step": 78555
    },
    {
      "entropy": 5.499582052230835,
      "epoch": 7.54658981748319,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011473439830634986,
      "loss": 4.6189,
      "mean_token_accuracy": 0.2654464140534401,
      "num_tokens": 180101199.0,
      "step": 78560
    },
    {
      "entropy": 5.497898244857788,
      "epoch": 7.547070124879923,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011471033709809299,
      "loss": 4.5923,
      "mean_token_accuracy": 0.2724624007940292,
      "num_tokens": 180113443.0,
      "step": 78565
    },
    {
      "entropy": 5.481863880157471,
      "epoch": 7.547550432276657,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011468627961128609,
      "loss": 4.5277,
      "mean_token_accuracy": 0.2822971135377884,
      "num_tokens": 180123135.0,
      "step": 78570
    },
    {
      "entropy": 5.481062269210815,
      "epoch": 7.548030739673391,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011466222584648777,
      "loss": 4.5822,
      "mean_token_accuracy": 0.272249761223793,
      "num_tokens": 180134130.0,
      "step": 78575
    },
    {
      "entropy": 5.472975349426269,
      "epoch": 7.548511047070125,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011463817580425654,
      "loss": 4.5368,
      "mean_token_accuracy": 0.2714802920818329,
      "num_tokens": 180144579.0,
      "step": 78580
    },
    {
      "entropy": 5.466867065429687,
      "epoch": 7.5489913544668585,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001146141294851506,
      "loss": 4.4908,
      "mean_token_accuracy": 0.27810807526111603,
      "num_tokens": 180154572.0,
      "step": 78585
    },
    {
      "entropy": 5.527522087097168,
      "epoch": 7.549471661863593,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011459008688972834,
      "loss": 4.6467,
      "mean_token_accuracy": 0.26686985343694686,
      "num_tokens": 180166031.0,
      "step": 78590
    },
    {
      "entropy": 5.506479406356812,
      "epoch": 7.549951969260326,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011456604801854792,
      "loss": 4.6859,
      "mean_token_accuracy": 0.25742626637220384,
      "num_tokens": 180179837.0,
      "step": 78595
    },
    {
      "entropy": 5.605707025527954,
      "epoch": 7.550432276657061,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001145420128721675,
      "loss": 4.6633,
      "mean_token_accuracy": 0.2665267542004585,
      "num_tokens": 180191692.0,
      "step": 78600
    },
    {
      "entropy": 5.5595602035522464,
      "epoch": 7.550912584053794,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011451798145114506,
      "loss": 4.6498,
      "mean_token_accuracy": 0.2658149257302284,
      "num_tokens": 180202948.0,
      "step": 78605
    },
    {
      "entropy": 5.601134443283081,
      "epoch": 7.551392891450528,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011449395375603857,
      "loss": 4.6643,
      "mean_token_accuracy": 0.2644840970635414,
      "num_tokens": 180214008.0,
      "step": 78610
    },
    {
      "entropy": 5.539640188217163,
      "epoch": 7.551873198847263,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011446992978740593,
      "loss": 4.5768,
      "mean_token_accuracy": 0.267328105866909,
      "num_tokens": 180224489.0,
      "step": 78615
    },
    {
      "entropy": 5.5259459018707275,
      "epoch": 7.552353506243996,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001144459095458048,
      "loss": 4.6681,
      "mean_token_accuracy": 0.26761066913604736,
      "num_tokens": 180236891.0,
      "step": 78620
    },
    {
      "entropy": 5.4935366153717045,
      "epoch": 7.55283381364073,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011442189303179288,
      "loss": 4.5767,
      "mean_token_accuracy": 0.27153509855270386,
      "num_tokens": 180248569.0,
      "step": 78625
    },
    {
      "entropy": 5.522446966171264,
      "epoch": 7.553314121037464,
      "grad_norm": 1.53125,
      "learning_rate": 0.00011439788024592778,
      "loss": 4.5821,
      "mean_token_accuracy": 0.2655811980366707,
      "num_tokens": 180259429.0,
      "step": 78630
    },
    {
      "entropy": 5.562025451660157,
      "epoch": 7.553794428434198,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000114373871188767,
      "loss": 4.6942,
      "mean_token_accuracy": 0.2639037773013115,
      "num_tokens": 180270100.0,
      "step": 78635
    },
    {
      "entropy": 5.518457460403442,
      "epoch": 7.554274735830932,
      "grad_norm": 1.375,
      "learning_rate": 0.00011434986586086799,
      "loss": 4.6344,
      "mean_token_accuracy": 0.26670292019844055,
      "num_tokens": 180281122.0,
      "step": 78640
    },
    {
      "entropy": 5.643044090270996,
      "epoch": 7.554755043227666,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011432586426278811,
      "loss": 4.7932,
      "mean_token_accuracy": 0.24303440600633622,
      "num_tokens": 180294246.0,
      "step": 78645
    },
    {
      "entropy": 5.485005331039429,
      "epoch": 7.555235350624399,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011430186639508453,
      "loss": 4.5197,
      "mean_token_accuracy": 0.27622997015714645,
      "num_tokens": 180305928.0,
      "step": 78650
    },
    {
      "entropy": 5.585308694839478,
      "epoch": 7.555715658021134,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011427787225831438,
      "loss": 4.6941,
      "mean_token_accuracy": 0.26714626997709273,
      "num_tokens": 180316772.0,
      "step": 78655
    },
    {
      "entropy": 5.567402744293213,
      "epoch": 7.556195965417867,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011425388185303486,
      "loss": 4.6989,
      "mean_token_accuracy": 0.27109378576278687,
      "num_tokens": 180329202.0,
      "step": 78660
    },
    {
      "entropy": 5.448240327835083,
      "epoch": 7.5566762728146015,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011422989517980282,
      "loss": 4.5701,
      "mean_token_accuracy": 0.2658624887466431,
      "num_tokens": 180341585.0,
      "step": 78665
    },
    {
      "entropy": 5.506641149520874,
      "epoch": 7.557156580211335,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011420591223917525,
      "loss": 4.5865,
      "mean_token_accuracy": 0.27179997265338895,
      "num_tokens": 180352956.0,
      "step": 78670
    },
    {
      "entropy": 5.54478874206543,
      "epoch": 7.557636887608069,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011418193303170899,
      "loss": 4.6188,
      "mean_token_accuracy": 0.2656882658600807,
      "num_tokens": 180364891.0,
      "step": 78675
    },
    {
      "entropy": 5.5469283103942875,
      "epoch": 7.558117195004803,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011415795755796067,
      "loss": 4.707,
      "mean_token_accuracy": 0.26389378160238264,
      "num_tokens": 180376176.0,
      "step": 78680
    },
    {
      "entropy": 5.595888710021972,
      "epoch": 7.558597502401537,
      "grad_norm": 1.59375,
      "learning_rate": 0.00011413398581848698,
      "loss": 4.6691,
      "mean_token_accuracy": 0.2703205272555351,
      "num_tokens": 180387870.0,
      "step": 78685
    },
    {
      "entropy": 5.57194242477417,
      "epoch": 7.559077809798271,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011411001781384445,
      "loss": 4.6567,
      "mean_token_accuracy": 0.2729285567998886,
      "num_tokens": 180399368.0,
      "step": 78690
    },
    {
      "entropy": 5.58526759147644,
      "epoch": 7.559558117195005,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011408605354458957,
      "loss": 4.7115,
      "mean_token_accuracy": 0.26143939048051834,
      "num_tokens": 180411209.0,
      "step": 78695
    },
    {
      "entropy": 5.543239879608154,
      "epoch": 7.560038424591738,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011406209301127871,
      "loss": 4.5099,
      "mean_token_accuracy": 0.27172366082668303,
      "num_tokens": 180422487.0,
      "step": 78700
    },
    {
      "entropy": 5.656872940063477,
      "epoch": 7.560518731988473,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001140381362144682,
      "loss": 4.7638,
      "mean_token_accuracy": 0.2542490303516388,
      "num_tokens": 180434460.0,
      "step": 78705
    },
    {
      "entropy": 5.531317090988159,
      "epoch": 7.560999039385207,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011401418315471418,
      "loss": 4.616,
      "mean_token_accuracy": 0.2602966070175171,
      "num_tokens": 180445597.0,
      "step": 78710
    },
    {
      "entropy": 5.486549520492554,
      "epoch": 7.56147934678194,
      "grad_norm": 1.28125,
      "learning_rate": 0.00011399023383257278,
      "loss": 4.5779,
      "mean_token_accuracy": 0.2699623018503189,
      "num_tokens": 180458341.0,
      "step": 78715
    },
    {
      "entropy": 5.5033485889434814,
      "epoch": 7.561959654178675,
      "grad_norm": 1.5,
      "learning_rate": 0.00011396628824860004,
      "loss": 4.5434,
      "mean_token_accuracy": 0.2716266468167305,
      "num_tokens": 180469750.0,
      "step": 78720
    },
    {
      "entropy": 5.45830340385437,
      "epoch": 7.562439961575408,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011394234640335193,
      "loss": 4.5168,
      "mean_token_accuracy": 0.2869134575128555,
      "num_tokens": 180480518.0,
      "step": 78725
    },
    {
      "entropy": 5.469973945617676,
      "epoch": 7.562920268972142,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011391840829738428,
      "loss": 4.4973,
      "mean_token_accuracy": 0.2767973214387894,
      "num_tokens": 180491893.0,
      "step": 78730
    },
    {
      "entropy": 5.471096038818359,
      "epoch": 7.563400576368876,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011389447393125296,
      "loss": 4.4764,
      "mean_token_accuracy": 0.2798026591539383,
      "num_tokens": 180501467.0,
      "step": 78735
    },
    {
      "entropy": 5.518263149261474,
      "epoch": 7.56388088376561,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001138705433055135,
      "loss": 4.6897,
      "mean_token_accuracy": 0.26593857556581496,
      "num_tokens": 180513134.0,
      "step": 78740
    },
    {
      "entropy": 5.553814935684204,
      "epoch": 7.564361191162344,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011384661642072155,
      "loss": 4.6805,
      "mean_token_accuracy": 0.26069949716329577,
      "num_tokens": 180525265.0,
      "step": 78745
    },
    {
      "entropy": 5.518943548202515,
      "epoch": 7.564841498559078,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011382269327743266,
      "loss": 4.6103,
      "mean_token_accuracy": 0.2642407059669495,
      "num_tokens": 180537045.0,
      "step": 78750
    },
    {
      "entropy": 5.628894424438476,
      "epoch": 7.565321805955811,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011379877387620224,
      "loss": 4.7216,
      "mean_token_accuracy": 0.25589887648820875,
      "num_tokens": 180548453.0,
      "step": 78755
    },
    {
      "entropy": 5.532553434371948,
      "epoch": 7.565802113352546,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001137748582175856,
      "loss": 4.636,
      "mean_token_accuracy": 0.2681162774562836,
      "num_tokens": 180560158.0,
      "step": 78760
    },
    {
      "entropy": 5.579592180252075,
      "epoch": 7.56628242074928,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00011375094630213809,
      "loss": 4.6699,
      "mean_token_accuracy": 0.26997365802526474,
      "num_tokens": 180571468.0,
      "step": 78765
    },
    {
      "entropy": 5.51166181564331,
      "epoch": 7.5667627281460135,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011372703813041472,
      "loss": 4.5815,
      "mean_token_accuracy": 0.27647086679935456,
      "num_tokens": 180583326.0,
      "step": 78770
    },
    {
      "entropy": 5.569271802902222,
      "epoch": 7.567243035542747,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011370313370297064,
      "loss": 4.639,
      "mean_token_accuracy": 0.26942090392112733,
      "num_tokens": 180595177.0,
      "step": 78775
    },
    {
      "entropy": 5.545388841629029,
      "epoch": 7.567723342939481,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011367923302036087,
      "loss": 4.6082,
      "mean_token_accuracy": 0.2673124775290489,
      "num_tokens": 180606502.0,
      "step": 78780
    },
    {
      "entropy": 5.594648838043213,
      "epoch": 7.568203650336216,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011365533608314026,
      "loss": 4.6945,
      "mean_token_accuracy": 0.25762745141983034,
      "num_tokens": 180618747.0,
      "step": 78785
    },
    {
      "entropy": 5.577661561965942,
      "epoch": 7.568683957732949,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011363144289186366,
      "loss": 4.6821,
      "mean_token_accuracy": 0.26528672873973846,
      "num_tokens": 180629786.0,
      "step": 78790
    },
    {
      "entropy": 5.545656824111939,
      "epoch": 7.569164265129683,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011360755344708586,
      "loss": 4.6401,
      "mean_token_accuracy": 0.2659398764371872,
      "num_tokens": 180642330.0,
      "step": 78795
    },
    {
      "entropy": 5.479640007019043,
      "epoch": 7.569644572526417,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011358366774936134,
      "loss": 4.5466,
      "mean_token_accuracy": 0.27461263686418536,
      "num_tokens": 180652332.0,
      "step": 78800
    },
    {
      "entropy": 5.520191335678101,
      "epoch": 7.570124879923151,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011355978579924478,
      "loss": 4.6146,
      "mean_token_accuracy": 0.26372024416923523,
      "num_tokens": 180663317.0,
      "step": 78805
    },
    {
      "entropy": 5.495601797103882,
      "epoch": 7.5706051873198845,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001135359075972906,
      "loss": 4.593,
      "mean_token_accuracy": 0.26564489454030993,
      "num_tokens": 180674521.0,
      "step": 78810
    },
    {
      "entropy": 5.430765581130982,
      "epoch": 7.571085494716619,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011351203314405317,
      "loss": 4.557,
      "mean_token_accuracy": 0.2742757350206375,
      "num_tokens": 180686758.0,
      "step": 78815
    },
    {
      "entropy": 5.561042308807373,
      "epoch": 7.571565802113352,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011348816244008685,
      "loss": 4.666,
      "mean_token_accuracy": 0.27156869471073153,
      "num_tokens": 180698604.0,
      "step": 78820
    },
    {
      "entropy": 5.6015723705291744,
      "epoch": 7.572046109510087,
      "grad_norm": 1.375,
      "learning_rate": 0.00011346429548594586,
      "loss": 4.7256,
      "mean_token_accuracy": 0.25947294384241104,
      "num_tokens": 180710273.0,
      "step": 78825
    },
    {
      "entropy": 5.599677991867066,
      "epoch": 7.57252641690682,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011344043228218421,
      "loss": 4.6704,
      "mean_token_accuracy": 0.26378281116485597,
      "num_tokens": 180721913.0,
      "step": 78830
    },
    {
      "entropy": 5.520884084701538,
      "epoch": 7.573006724303554,
      "grad_norm": 1.234375,
      "learning_rate": 0.00011341657282935597,
      "loss": 4.5989,
      "mean_token_accuracy": 0.2726454198360443,
      "num_tokens": 180732771.0,
      "step": 78835
    },
    {
      "entropy": 5.498121213912964,
      "epoch": 7.573487031700288,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011339271712801511,
      "loss": 4.5437,
      "mean_token_accuracy": 0.2758928120136261,
      "num_tokens": 180745009.0,
      "step": 78840
    },
    {
      "entropy": 5.5386505126953125,
      "epoch": 7.573967339097022,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00011336886517871547,
      "loss": 4.6257,
      "mean_token_accuracy": 0.2690087929368019,
      "num_tokens": 180756123.0,
      "step": 78845
    },
    {
      "entropy": 5.492664194107055,
      "epoch": 7.574447646493756,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011334501698201087,
      "loss": 4.5995,
      "mean_token_accuracy": 0.27437758892774583,
      "num_tokens": 180768014.0,
      "step": 78850
    },
    {
      "entropy": 5.514504098892212,
      "epoch": 7.57492795389049,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000113321172538455,
      "loss": 4.63,
      "mean_token_accuracy": 0.2652719423174858,
      "num_tokens": 180779325.0,
      "step": 78855
    },
    {
      "entropy": 5.473705196380616,
      "epoch": 7.575408261287224,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011329733184860136,
      "loss": 4.5645,
      "mean_token_accuracy": 0.2716944247484207,
      "num_tokens": 180791954.0,
      "step": 78860
    },
    {
      "entropy": 5.585518169403076,
      "epoch": 7.575888568683958,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011327349491300351,
      "loss": 4.655,
      "mean_token_accuracy": 0.2606149882078171,
      "num_tokens": 180804145.0,
      "step": 78865
    },
    {
      "entropy": 5.452359771728515,
      "epoch": 7.576368876080692,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011324966173221488,
      "loss": 4.4393,
      "mean_token_accuracy": 0.2892660856246948,
      "num_tokens": 180814982.0,
      "step": 78870
    },
    {
      "entropy": 5.537131500244141,
      "epoch": 7.576849183477425,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001132258323067888,
      "loss": 4.5921,
      "mean_token_accuracy": 0.2691638648509979,
      "num_tokens": 180827019.0,
      "step": 78875
    },
    {
      "entropy": 5.574140214920044,
      "epoch": 7.57732949087416,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011320200663727854,
      "loss": 4.7253,
      "mean_token_accuracy": 0.2553164929151535,
      "num_tokens": 180840006.0,
      "step": 78880
    },
    {
      "entropy": 5.514279365539551,
      "epoch": 7.577809798270893,
      "grad_norm": 1.375,
      "learning_rate": 0.00011317818472423728,
      "loss": 4.6085,
      "mean_token_accuracy": 0.2625275686383247,
      "num_tokens": 180853224.0,
      "step": 78885
    },
    {
      "entropy": 5.612389755249024,
      "epoch": 7.5782901056676275,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00011315436656821802,
      "loss": 4.7336,
      "mean_token_accuracy": 0.26030156314373015,
      "num_tokens": 180863520.0,
      "step": 78890
    },
    {
      "entropy": 5.551396131515503,
      "epoch": 7.578770413064361,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011313055216977378,
      "loss": 4.6115,
      "mean_token_accuracy": 0.2699619323015213,
      "num_tokens": 180874460.0,
      "step": 78895
    },
    {
      "entropy": 5.533211994171142,
      "epoch": 7.579250720461095,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011310674152945748,
      "loss": 4.6558,
      "mean_token_accuracy": 0.2680956915020943,
      "num_tokens": 180886001.0,
      "step": 78900
    },
    {
      "entropy": 5.536896991729736,
      "epoch": 7.579731027857829,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011308293464782189,
      "loss": 4.5898,
      "mean_token_accuracy": 0.27096216976642606,
      "num_tokens": 180897374.0,
      "step": 78905
    },
    {
      "entropy": 5.472412490844727,
      "epoch": 7.580211335254563,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011305913152541976,
      "loss": 4.588,
      "mean_token_accuracy": 0.2693932384252548,
      "num_tokens": 180908739.0,
      "step": 78910
    },
    {
      "entropy": 5.593320083618164,
      "epoch": 7.5806916426512965,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001130353321628038,
      "loss": 4.7062,
      "mean_token_accuracy": 0.25618912279605865,
      "num_tokens": 180920272.0,
      "step": 78915
    },
    {
      "entropy": 5.48904275894165,
      "epoch": 7.581171950048031,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011301153656052644,
      "loss": 4.5527,
      "mean_token_accuracy": 0.2669686362147331,
      "num_tokens": 180931879.0,
      "step": 78920
    },
    {
      "entropy": 5.559494495391846,
      "epoch": 7.581652257444764,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001129877447191402,
      "loss": 4.6986,
      "mean_token_accuracy": 0.268300487101078,
      "num_tokens": 180944290.0,
      "step": 78925
    },
    {
      "entropy": 5.571070289611816,
      "epoch": 7.582132564841499,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011296395663919742,
      "loss": 4.6091,
      "mean_token_accuracy": 0.2647917717695236,
      "num_tokens": 180955202.0,
      "step": 78930
    },
    {
      "entropy": 5.454857254028321,
      "epoch": 7.582612872238233,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011294017232125044,
      "loss": 4.5502,
      "mean_token_accuracy": 0.2711860164999962,
      "num_tokens": 180966858.0,
      "step": 78935
    },
    {
      "entropy": 5.475846862792968,
      "epoch": 7.583093179634966,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011291639176585145,
      "loss": 4.6273,
      "mean_token_accuracy": 0.26812585443258286,
      "num_tokens": 180978101.0,
      "step": 78940
    },
    {
      "entropy": 5.55639271736145,
      "epoch": 7.583573487031701,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011289261497355262,
      "loss": 4.6556,
      "mean_token_accuracy": 0.2621735230088234,
      "num_tokens": 180988840.0,
      "step": 78945
    },
    {
      "entropy": 5.558872222900391,
      "epoch": 7.584053794428434,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011286884194490582,
      "loss": 4.665,
      "mean_token_accuracy": 0.258784282207489,
      "num_tokens": 181000220.0,
      "step": 78950
    },
    {
      "entropy": 5.651305770874023,
      "epoch": 7.584534101825168,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001128450726804631,
      "loss": 4.7513,
      "mean_token_accuracy": 0.25709015876054764,
      "num_tokens": 181011100.0,
      "step": 78955
    },
    {
      "entropy": 5.559948301315307,
      "epoch": 7.585014409221902,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011282130718077628,
      "loss": 4.6327,
      "mean_token_accuracy": 0.2649137988686562,
      "num_tokens": 181022498.0,
      "step": 78960
    },
    {
      "entropy": 5.5019816875457765,
      "epoch": 7.585494716618636,
      "grad_norm": 1.375,
      "learning_rate": 0.00011279754544639713,
      "loss": 4.5717,
      "mean_token_accuracy": 0.27566893249750135,
      "num_tokens": 181034095.0,
      "step": 78965
    },
    {
      "entropy": 5.554961919784546,
      "epoch": 7.58597502401537,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011277378747787734,
      "loss": 4.6696,
      "mean_token_accuracy": 0.2656422331929207,
      "num_tokens": 181045348.0,
      "step": 78970
    },
    {
      "entropy": 5.486689901351928,
      "epoch": 7.586455331412104,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011275003327576856,
      "loss": 4.6129,
      "mean_token_accuracy": 0.26662461906671525,
      "num_tokens": 181057916.0,
      "step": 78975
    },
    {
      "entropy": 5.561379146575928,
      "epoch": 7.586935638808837,
      "grad_norm": 1.375,
      "learning_rate": 0.00011272628284062214,
      "loss": 4.6835,
      "mean_token_accuracy": 0.2641612946987152,
      "num_tokens": 181068342.0,
      "step": 78980
    },
    {
      "entropy": 5.555734252929687,
      "epoch": 7.587415946205572,
      "grad_norm": 1.53125,
      "learning_rate": 0.00011270253617298961,
      "loss": 4.6309,
      "mean_token_accuracy": 0.26462866067886354,
      "num_tokens": 181079084.0,
      "step": 78985
    },
    {
      "entropy": 5.568646097183228,
      "epoch": 7.587896253602305,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011267879327342226,
      "loss": 4.6737,
      "mean_token_accuracy": 0.25706829130649567,
      "num_tokens": 181091505.0,
      "step": 78990
    },
    {
      "entropy": 5.491316509246826,
      "epoch": 7.5883765609990395,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001126550541424713,
      "loss": 4.6342,
      "mean_token_accuracy": 0.2646765798330307,
      "num_tokens": 181102529.0,
      "step": 78995
    },
    {
      "entropy": 5.526732778549194,
      "epoch": 7.588856868395773,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00011263131878068794,
      "loss": 4.6168,
      "mean_token_accuracy": 0.2741875186562538,
      "num_tokens": 181115908.0,
      "step": 79000
    },
    {
      "entropy": 5.67940239906311,
      "epoch": 7.589337175792507,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011260758718862331,
      "loss": 4.7676,
      "mean_token_accuracy": 0.25300118029117585,
      "num_tokens": 181127871.0,
      "step": 79005
    },
    {
      "entropy": 5.513166046142578,
      "epoch": 7.5898174831892415,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001125838593668282,
      "loss": 4.5891,
      "mean_token_accuracy": 0.26585167795419695,
      "num_tokens": 181139475.0,
      "step": 79010
    },
    {
      "entropy": 5.563820314407349,
      "epoch": 7.590297790585975,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011256013531585363,
      "loss": 4.6356,
      "mean_token_accuracy": 0.2667722165584564,
      "num_tokens": 181151225.0,
      "step": 79015
    },
    {
      "entropy": 5.582314109802246,
      "epoch": 7.590778097982709,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011253641503625034,
      "loss": 4.6065,
      "mean_token_accuracy": 0.26734889298677444,
      "num_tokens": 181162515.0,
      "step": 79020
    },
    {
      "entropy": 5.450829982757568,
      "epoch": 7.591258405379443,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011251269852856908,
      "loss": 4.5037,
      "mean_token_accuracy": 0.28022018671035764,
      "num_tokens": 181173333.0,
      "step": 79025
    },
    {
      "entropy": 5.558090353012085,
      "epoch": 7.591738712776177,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001124889857933605,
      "loss": 4.6624,
      "mean_token_accuracy": 0.2696232169866562,
      "num_tokens": 181185700.0,
      "step": 79030
    },
    {
      "entropy": 5.588767719268799,
      "epoch": 7.5922190201729105,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011246527683117513,
      "loss": 4.6398,
      "mean_token_accuracy": 0.26968623548746107,
      "num_tokens": 181196330.0,
      "step": 79035
    },
    {
      "entropy": 5.609730911254883,
      "epoch": 7.592699327569645,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011244157164256338,
      "loss": 4.6828,
      "mean_token_accuracy": 0.2671284019947052,
      "num_tokens": 181208729.0,
      "step": 79040
    },
    {
      "entropy": 5.502044343948365,
      "epoch": 7.593179634966378,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011241787022807561,
      "loss": 4.5629,
      "mean_token_accuracy": 0.2710377246141434,
      "num_tokens": 181220480.0,
      "step": 79045
    },
    {
      "entropy": 5.537319898605347,
      "epoch": 7.593659942363113,
      "grad_norm": 1.59375,
      "learning_rate": 0.00011239417258826214,
      "loss": 4.5911,
      "mean_token_accuracy": 0.2709709480404854,
      "num_tokens": 181231306.0,
      "step": 79050
    },
    {
      "entropy": 5.489433526992798,
      "epoch": 7.594140249759846,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011237047872367314,
      "loss": 4.5673,
      "mean_token_accuracy": 0.27464977353811265,
      "num_tokens": 181242151.0,
      "step": 79055
    },
    {
      "entropy": 5.466402769088745,
      "epoch": 7.59462055715658,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011234678863485872,
      "loss": 4.5476,
      "mean_token_accuracy": 0.2732745498418808,
      "num_tokens": 181253933.0,
      "step": 79060
    },
    {
      "entropy": 5.4816650390625,
      "epoch": 7.595100864553314,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011232310232236888,
      "loss": 4.6458,
      "mean_token_accuracy": 0.2674422889947891,
      "num_tokens": 181265879.0,
      "step": 79065
    },
    {
      "entropy": 5.5034324645996096,
      "epoch": 7.595581171950048,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011229941978675362,
      "loss": 4.5602,
      "mean_token_accuracy": 0.27902557253837584,
      "num_tokens": 181278574.0,
      "step": 79070
    },
    {
      "entropy": 5.543645572662354,
      "epoch": 7.5960614793467816,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011227574102856264,
      "loss": 4.6151,
      "mean_token_accuracy": 0.2620904788374901,
      "num_tokens": 181290257.0,
      "step": 79075
    },
    {
      "entropy": 5.605256748199463,
      "epoch": 7.596541786743516,
      "grad_norm": 1.375,
      "learning_rate": 0.00011225206604834576,
      "loss": 4.7182,
      "mean_token_accuracy": 0.2558312088251114,
      "num_tokens": 181302300.0,
      "step": 79080
    },
    {
      "entropy": 5.51648530960083,
      "epoch": 7.59702209414025,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011222839484665264,
      "loss": 4.6351,
      "mean_token_accuracy": 0.26482999622821807,
      "num_tokens": 181315087.0,
      "step": 79085
    },
    {
      "entropy": 5.607822132110596,
      "epoch": 7.597502401536984,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011220472742403287,
      "loss": 4.7208,
      "mean_token_accuracy": 0.2567916542291641,
      "num_tokens": 181326947.0,
      "step": 79090
    },
    {
      "entropy": 5.594553756713867,
      "epoch": 7.597982708933718,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011218106378103593,
      "loss": 4.6907,
      "mean_token_accuracy": 0.2675647482275963,
      "num_tokens": 181337274.0,
      "step": 79095
    },
    {
      "entropy": 5.53327169418335,
      "epoch": 7.598463016330451,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011215740391821122,
      "loss": 4.6075,
      "mean_token_accuracy": 0.2677896678447723,
      "num_tokens": 181348290.0,
      "step": 79100
    },
    {
      "entropy": 5.5607531547546385,
      "epoch": 7.598943323727186,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011213374783610803,
      "loss": 4.6099,
      "mean_token_accuracy": 0.26370156854391097,
      "num_tokens": 181359232.0,
      "step": 79105
    },
    {
      "entropy": 5.506102848052978,
      "epoch": 7.599423631123919,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011211009553527566,
      "loss": 4.6045,
      "mean_token_accuracy": 0.26962760984897616,
      "num_tokens": 181370671.0,
      "step": 79110
    },
    {
      "entropy": 5.536318445205689,
      "epoch": 7.5999039385206535,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011208644701626315,
      "loss": 4.6381,
      "mean_token_accuracy": 0.2643285498023033,
      "num_tokens": 181382597.0,
      "step": 79115
    },
    {
      "entropy": 5.564944934844971,
      "epoch": 7.600384245917387,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011206280227961957,
      "loss": 4.7215,
      "mean_token_accuracy": 0.2645706072449684,
      "num_tokens": 181393090.0,
      "step": 79120
    },
    {
      "entropy": 5.618915462493897,
      "epoch": 7.600864553314121,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011203916132589389,
      "loss": 4.7095,
      "mean_token_accuracy": 0.26307264864444735,
      "num_tokens": 181404402.0,
      "step": 79125
    },
    {
      "entropy": 5.618213033676147,
      "epoch": 7.601344860710855,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011201552415563502,
      "loss": 4.6607,
      "mean_token_accuracy": 0.2643963351845741,
      "num_tokens": 181414789.0,
      "step": 79130
    },
    {
      "entropy": 5.604283571243286,
      "epoch": 7.601825168107589,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001119918907693917,
      "loss": 4.6495,
      "mean_token_accuracy": 0.26744032055139544,
      "num_tokens": 181426467.0,
      "step": 79135
    },
    {
      "entropy": 5.540360021591186,
      "epoch": 7.6023054755043225,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011196826116771264,
      "loss": 4.6475,
      "mean_token_accuracy": 0.26527752727270126,
      "num_tokens": 181437696.0,
      "step": 79140
    },
    {
      "entropy": 5.563962316513061,
      "epoch": 7.602785782901057,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011194463535114643,
      "loss": 4.6254,
      "mean_token_accuracy": 0.2705245837569237,
      "num_tokens": 181449866.0,
      "step": 79145
    },
    {
      "entropy": 5.459997749328613,
      "epoch": 7.60326609029779,
      "grad_norm": 1.5,
      "learning_rate": 0.0001119210133202417,
      "loss": 4.5188,
      "mean_token_accuracy": 0.2774031534790993,
      "num_tokens": 181462080.0,
      "step": 79150
    },
    {
      "entropy": 5.49266791343689,
      "epoch": 7.6037463976945245,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011189739507554673,
      "loss": 4.6058,
      "mean_token_accuracy": 0.26847572177648543,
      "num_tokens": 181473976.0,
      "step": 79155
    },
    {
      "entropy": 5.528505659103393,
      "epoch": 7.604226705091259,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011187378061760992,
      "loss": 4.5938,
      "mean_token_accuracy": 0.2715308889746666,
      "num_tokens": 181486122.0,
      "step": 79160
    },
    {
      "entropy": 5.528113794326782,
      "epoch": 7.604707012487992,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011185016994697955,
      "loss": 4.5288,
      "mean_token_accuracy": 0.2837370127439499,
      "num_tokens": 181497712.0,
      "step": 79165
    },
    {
      "entropy": 5.571224546432495,
      "epoch": 7.605187319884726,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011182656306420378,
      "loss": 4.7315,
      "mean_token_accuracy": 0.2607993006706238,
      "num_tokens": 181509219.0,
      "step": 79170
    },
    {
      "entropy": 5.526147127151489,
      "epoch": 7.60566762728146,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011180295996983069,
      "loss": 4.6348,
      "mean_token_accuracy": 0.26687469482421877,
      "num_tokens": 181520627.0,
      "step": 79175
    },
    {
      "entropy": 5.528483200073242,
      "epoch": 7.606147934678194,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011177936066440828,
      "loss": 4.6022,
      "mean_token_accuracy": 0.27345981299877165,
      "num_tokens": 181532814.0,
      "step": 79180
    },
    {
      "entropy": 5.562067890167237,
      "epoch": 7.606628242074928,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011175576514848442,
      "loss": 4.6503,
      "mean_token_accuracy": 0.2648743614554405,
      "num_tokens": 181544554.0,
      "step": 79185
    },
    {
      "entropy": 5.650294399261474,
      "epoch": 7.607108549471662,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011173217342260705,
      "loss": 4.7272,
      "mean_token_accuracy": 0.25276884287595747,
      "num_tokens": 181556420.0,
      "step": 79190
    },
    {
      "entropy": 5.5991363525390625,
      "epoch": 7.607588856868396,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011170858548732374,
      "loss": 4.7179,
      "mean_token_accuracy": 0.2615563109517097,
      "num_tokens": 181568742.0,
      "step": 79195
    },
    {
      "entropy": 5.494099569320679,
      "epoch": 7.60806916426513,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011168500134318218,
      "loss": 4.6163,
      "mean_token_accuracy": 0.26639108210802076,
      "num_tokens": 181581235.0,
      "step": 79200
    },
    {
      "entropy": 5.5809845447540285,
      "epoch": 7.608549471661863,
      "grad_norm": 1.5,
      "learning_rate": 0.00011166142099072996,
      "loss": 4.6501,
      "mean_token_accuracy": 0.26067582666873934,
      "num_tokens": 181592034.0,
      "step": 79205
    },
    {
      "entropy": 5.514943075180054,
      "epoch": 7.609029779058598,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011163784443051451,
      "loss": 4.5599,
      "mean_token_accuracy": 0.27231901586055757,
      "num_tokens": 181602770.0,
      "step": 79210
    },
    {
      "entropy": 5.499369764328003,
      "epoch": 7.609510086455331,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011161427166308324,
      "loss": 4.5366,
      "mean_token_accuracy": 0.2706191033124924,
      "num_tokens": 181613885.0,
      "step": 79215
    },
    {
      "entropy": 5.522282218933105,
      "epoch": 7.609990393852065,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011159070268898352,
      "loss": 4.595,
      "mean_token_accuracy": 0.2690237507224083,
      "num_tokens": 181625389.0,
      "step": 79220
    },
    {
      "entropy": 5.548044681549072,
      "epoch": 7.610470701248799,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011156713750876235,
      "loss": 4.6479,
      "mean_token_accuracy": 0.26715758442878723,
      "num_tokens": 181637300.0,
      "step": 79225
    },
    {
      "entropy": 5.502521896362305,
      "epoch": 7.610951008645533,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011154357612296697,
      "loss": 4.6006,
      "mean_token_accuracy": 0.2651588603854179,
      "num_tokens": 181649427.0,
      "step": 79230
    },
    {
      "entropy": 5.499505043029785,
      "epoch": 7.6114313160422675,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011152001853214438,
      "loss": 4.6228,
      "mean_token_accuracy": 0.26507880389690397,
      "num_tokens": 181661584.0,
      "step": 79235
    },
    {
      "entropy": 5.493945646286011,
      "epoch": 7.611911623439001,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011149646473684154,
      "loss": 4.5971,
      "mean_token_accuracy": 0.27264865189790727,
      "num_tokens": 181672295.0,
      "step": 79240
    },
    {
      "entropy": 5.502749967575073,
      "epoch": 7.612391930835734,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00011147291473760526,
      "loss": 4.5988,
      "mean_token_accuracy": 0.27108662128448485,
      "num_tokens": 181684295.0,
      "step": 79245
    },
    {
      "entropy": 5.574969053268433,
      "epoch": 7.612872238232469,
      "grad_norm": 1.375,
      "learning_rate": 0.00011144936853498241,
      "loss": 4.6417,
      "mean_token_accuracy": 0.2724490791559219,
      "num_tokens": 181695908.0,
      "step": 79250
    },
    {
      "entropy": 5.602262687683106,
      "epoch": 7.613352545629203,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011142582612951949,
      "loss": 4.7108,
      "mean_token_accuracy": 0.26210874021053315,
      "num_tokens": 181707669.0,
      "step": 79255
    },
    {
      "entropy": 5.493008804321289,
      "epoch": 7.6138328530259365,
      "grad_norm": 1.28125,
      "learning_rate": 0.00011140228752176318,
      "loss": 4.5892,
      "mean_token_accuracy": 0.2664280816912651,
      "num_tokens": 181719323.0,
      "step": 79260
    },
    {
      "entropy": 5.5362121105194095,
      "epoch": 7.614313160422671,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011137875271225997,
      "loss": 4.6273,
      "mean_token_accuracy": 0.2673512473702431,
      "num_tokens": 181730496.0,
      "step": 79265
    },
    {
      "entropy": 5.51537504196167,
      "epoch": 7.614793467819404,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011135522170155626,
      "loss": 4.5038,
      "mean_token_accuracy": 0.28066670298576357,
      "num_tokens": 181741271.0,
      "step": 79270
    },
    {
      "entropy": 5.567674779891968,
      "epoch": 7.615273775216139,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011133169449019839,
      "loss": 4.6389,
      "mean_token_accuracy": 0.26174142211675644,
      "num_tokens": 181753122.0,
      "step": 79275
    },
    {
      "entropy": 5.50764684677124,
      "epoch": 7.615754082612872,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011130817107873264,
      "loss": 4.6251,
      "mean_token_accuracy": 0.2713292151689529,
      "num_tokens": 181765246.0,
      "step": 79280
    },
    {
      "entropy": 5.4809528350830075,
      "epoch": 7.616234390009606,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011128465146770503,
      "loss": 4.6102,
      "mean_token_accuracy": 0.27251405119895933,
      "num_tokens": 181776889.0,
      "step": 79285
    },
    {
      "entropy": 5.527331352233887,
      "epoch": 7.61671469740634,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011126113565766166,
      "loss": 4.5607,
      "mean_token_accuracy": 0.2747654691338539,
      "num_tokens": 181788923.0,
      "step": 79290
    },
    {
      "entropy": 5.468953227996826,
      "epoch": 7.617195004803074,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011123762364914855,
      "loss": 4.5175,
      "mean_token_accuracy": 0.272426900267601,
      "num_tokens": 181801489.0,
      "step": 79295
    },
    {
      "entropy": 5.55304970741272,
      "epoch": 7.6176753121998075,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011121411544271151,
      "loss": 4.6236,
      "mean_token_accuracy": 0.2702735781669617,
      "num_tokens": 181813023.0,
      "step": 79300
    },
    {
      "entropy": 5.547876787185669,
      "epoch": 7.618155619596542,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00011119061103889639,
      "loss": 4.6501,
      "mean_token_accuracy": 0.2665713921189308,
      "num_tokens": 181824837.0,
      "step": 79305
    },
    {
      "entropy": 5.5039286613464355,
      "epoch": 7.618635926993276,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011116711043824892,
      "loss": 4.6003,
      "mean_token_accuracy": 0.2650326192378998,
      "num_tokens": 181836451.0,
      "step": 79310
    },
    {
      "entropy": 5.510843181610108,
      "epoch": 7.61911623439001,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011114361364131459,
      "loss": 4.6002,
      "mean_token_accuracy": 0.2673339247703552,
      "num_tokens": 181848309.0,
      "step": 79315
    },
    {
      "entropy": 5.538144111633301,
      "epoch": 7.619596541786743,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011112012064863902,
      "loss": 4.6528,
      "mean_token_accuracy": 0.25650544464588165,
      "num_tokens": 181859230.0,
      "step": 79320
    },
    {
      "entropy": 5.583040523529053,
      "epoch": 7.620076849183477,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001110966314607676,
      "loss": 4.6549,
      "mean_token_accuracy": 0.26418543308973313,
      "num_tokens": 181871260.0,
      "step": 79325
    },
    {
      "entropy": 5.518413305282593,
      "epoch": 7.620557156580212,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011107314607824573,
      "loss": 4.5702,
      "mean_token_accuracy": 0.2678869754076004,
      "num_tokens": 181883604.0,
      "step": 79330
    },
    {
      "entropy": 5.49209771156311,
      "epoch": 7.621037463976945,
      "grad_norm": 1.5625,
      "learning_rate": 0.00011104966450161863,
      "loss": 4.5634,
      "mean_token_accuracy": 0.2731957048177719,
      "num_tokens": 181893328.0,
      "step": 79335
    },
    {
      "entropy": 5.541071367263794,
      "epoch": 7.6215177713736795,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011102618673143155,
      "loss": 4.6256,
      "mean_token_accuracy": 0.2607275888323784,
      "num_tokens": 181905078.0,
      "step": 79340
    },
    {
      "entropy": 5.517868232727051,
      "epoch": 7.621998078770413,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011100271276822945,
      "loss": 4.5751,
      "mean_token_accuracy": 0.28056624829769133,
      "num_tokens": 181916331.0,
      "step": 79345
    },
    {
      "entropy": 5.460095167160034,
      "epoch": 7.622478386167147,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001109792426125574,
      "loss": 4.5268,
      "mean_token_accuracy": 0.2725417479872704,
      "num_tokens": 181928178.0,
      "step": 79350
    },
    {
      "entropy": 5.466282176971435,
      "epoch": 7.622958693563881,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011095577626496028,
      "loss": 4.5807,
      "mean_token_accuracy": 0.2677400216460228,
      "num_tokens": 181939756.0,
      "step": 79355
    },
    {
      "entropy": 5.543509435653687,
      "epoch": 7.623439000960615,
      "grad_norm": 1.265625,
      "learning_rate": 0.00011093231372598292,
      "loss": 4.6639,
      "mean_token_accuracy": 0.2620297268033028,
      "num_tokens": 181952329.0,
      "step": 79360
    },
    {
      "entropy": 5.514752054214478,
      "epoch": 7.623919308357348,
      "grad_norm": 1.5,
      "learning_rate": 0.00011090885499617007,
      "loss": 4.574,
      "mean_token_accuracy": 0.2697790160775185,
      "num_tokens": 181963438.0,
      "step": 79365
    },
    {
      "entropy": 5.564290523529053,
      "epoch": 7.624399615754083,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011088540007606643,
      "loss": 4.6603,
      "mean_token_accuracy": 0.2689312011003494,
      "num_tokens": 181975525.0,
      "step": 79370
    },
    {
      "entropy": 5.593834495544433,
      "epoch": 7.624879923150816,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001108619489662164,
      "loss": 4.6728,
      "mean_token_accuracy": 0.2679836377501488,
      "num_tokens": 181986478.0,
      "step": 79375
    },
    {
      "entropy": 5.561084318161011,
      "epoch": 7.6253602305475505,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00011083850166716453,
      "loss": 4.6645,
      "mean_token_accuracy": 0.2640442118048668,
      "num_tokens": 181997992.0,
      "step": 79380
    },
    {
      "entropy": 5.481447982788086,
      "epoch": 7.625840537944284,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011081505817945517,
      "loss": 4.5358,
      "mean_token_accuracy": 0.27190086543560027,
      "num_tokens": 182007742.0,
      "step": 79385
    },
    {
      "entropy": 5.549991273880005,
      "epoch": 7.626320845341018,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011079161850363265,
      "loss": 4.6831,
      "mean_token_accuracy": 0.2645321309566498,
      "num_tokens": 182020671.0,
      "step": 79390
    },
    {
      "entropy": 5.488533067703247,
      "epoch": 7.626801152737752,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011076818264024113,
      "loss": 4.5676,
      "mean_token_accuracy": 0.26709062308073045,
      "num_tokens": 182031924.0,
      "step": 79395
    },
    {
      "entropy": 5.467839050292969,
      "epoch": 7.627281460134486,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011074475058982483,
      "loss": 4.5467,
      "mean_token_accuracy": 0.27037870436906813,
      "num_tokens": 182044617.0,
      "step": 79400
    },
    {
      "entropy": 5.539542293548584,
      "epoch": 7.62776176753122,
      "grad_norm": 1.53125,
      "learning_rate": 0.00011072132235292762,
      "loss": 4.588,
      "mean_token_accuracy": 0.2680670455098152,
      "num_tokens": 182055204.0,
      "step": 79405
    },
    {
      "entropy": 5.564290189743042,
      "epoch": 7.628242074927954,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001106978979300935,
      "loss": 4.6813,
      "mean_token_accuracy": 0.26098931282758714,
      "num_tokens": 182066982.0,
      "step": 79410
    },
    {
      "entropy": 5.530998086929321,
      "epoch": 7.628722382324688,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011067447732186628,
      "loss": 4.703,
      "mean_token_accuracy": 0.26555139422416685,
      "num_tokens": 182078454.0,
      "step": 79415
    },
    {
      "entropy": 5.430376386642456,
      "epoch": 7.629202689721422,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011065106052878977,
      "loss": 4.5372,
      "mean_token_accuracy": 0.2741936415433884,
      "num_tokens": 182090520.0,
      "step": 79420
    },
    {
      "entropy": 5.464127826690674,
      "epoch": 7.629682997118156,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011062764755140762,
      "loss": 4.5694,
      "mean_token_accuracy": 0.26618890315294264,
      "num_tokens": 182101931.0,
      "step": 79425
    },
    {
      "entropy": 5.5239519596099855,
      "epoch": 7.630163304514889,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00011060423839026349,
      "loss": 4.6443,
      "mean_token_accuracy": 0.26916290670633314,
      "num_tokens": 182114700.0,
      "step": 79430
    },
    {
      "entropy": 5.490394067764282,
      "epoch": 7.630643611911624,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001105808330459007,
      "loss": 4.5698,
      "mean_token_accuracy": 0.2690826267004013,
      "num_tokens": 182126343.0,
      "step": 79435
    },
    {
      "entropy": 5.4919572353363035,
      "epoch": 7.631123919308357,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011055743151886275,
      "loss": 4.5883,
      "mean_token_accuracy": 0.2637495070695877,
      "num_tokens": 182136960.0,
      "step": 79440
    },
    {
      "entropy": 5.581543779373169,
      "epoch": 7.631604226705091,
      "grad_norm": 1.5,
      "learning_rate": 0.00011053403380969293,
      "loss": 4.7309,
      "mean_token_accuracy": 0.2570267006754875,
      "num_tokens": 182148439.0,
      "step": 79445
    },
    {
      "entropy": 5.533446884155273,
      "epoch": 7.632084534101825,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001105106399189345,
      "loss": 4.5939,
      "mean_token_accuracy": 0.2718494847416878,
      "num_tokens": 182159610.0,
      "step": 79450
    },
    {
      "entropy": 5.4906041622161865,
      "epoch": 7.632564841498559,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011048724984713058,
      "loss": 4.5993,
      "mean_token_accuracy": 0.26665627360343935,
      "num_tokens": 182170098.0,
      "step": 79455
    },
    {
      "entropy": 5.550370836257935,
      "epoch": 7.633045148895293,
      "grad_norm": 1.65625,
      "learning_rate": 0.00011046386359482428,
      "loss": 4.6338,
      "mean_token_accuracy": 0.27306551337242124,
      "num_tokens": 182181644.0,
      "step": 79460
    },
    {
      "entropy": 5.507027292251587,
      "epoch": 7.633525456292027,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011044048116255843,
      "loss": 4.5368,
      "mean_token_accuracy": 0.2727375775575638,
      "num_tokens": 182192623.0,
      "step": 79465
    },
    {
      "entropy": 5.526354551315308,
      "epoch": 7.63400576368876,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011041710255087595,
      "loss": 4.6173,
      "mean_token_accuracy": 0.26819154918193816,
      "num_tokens": 182203692.0,
      "step": 79470
    },
    {
      "entropy": 5.462149906158447,
      "epoch": 7.634486071085495,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011039372776031966,
      "loss": 4.5255,
      "mean_token_accuracy": 0.27693947553634646,
      "num_tokens": 182215556.0,
      "step": 79475
    },
    {
      "entropy": 5.488791465759277,
      "epoch": 7.634966378482229,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011037035679143222,
      "loss": 4.612,
      "mean_token_accuracy": 0.2626036435365677,
      "num_tokens": 182226870.0,
      "step": 79480
    },
    {
      "entropy": 5.577980709075928,
      "epoch": 7.6354466858789625,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011034698964475626,
      "loss": 4.6831,
      "mean_token_accuracy": 0.26507865488529203,
      "num_tokens": 182238342.0,
      "step": 79485
    },
    {
      "entropy": 5.543226861953736,
      "epoch": 7.635926993275697,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011032362632083434,
      "loss": 4.6506,
      "mean_token_accuracy": 0.26691529601812364,
      "num_tokens": 182251281.0,
      "step": 79490
    },
    {
      "entropy": 5.514912557601929,
      "epoch": 7.63640730067243,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011030026682020875,
      "loss": 4.6392,
      "mean_token_accuracy": 0.26522359251976013,
      "num_tokens": 182262173.0,
      "step": 79495
    },
    {
      "entropy": 5.4536988735198975,
      "epoch": 7.636887608069165,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011027691114342193,
      "loss": 4.5811,
      "mean_token_accuracy": 0.268124932050705,
      "num_tokens": 182273207.0,
      "step": 79500
    },
    {
      "entropy": 5.477387475967407,
      "epoch": 7.637367915465898,
      "grad_norm": 1.375,
      "learning_rate": 0.00011025355929101605,
      "loss": 4.5971,
      "mean_token_accuracy": 0.264689627289772,
      "num_tokens": 182283804.0,
      "step": 79505
    },
    {
      "entropy": 5.541228199005127,
      "epoch": 7.637848222862632,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011023021126353335,
      "loss": 4.5956,
      "mean_token_accuracy": 0.26289119720458987,
      "num_tokens": 182294981.0,
      "step": 79510
    },
    {
      "entropy": 5.5105854034423825,
      "epoch": 7.638328530259366,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00011020686706151592,
      "loss": 4.5835,
      "mean_token_accuracy": 0.2663647323846817,
      "num_tokens": 182305541.0,
      "step": 79515
    },
    {
      "entropy": 5.476885366439819,
      "epoch": 7.6388088376561,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011018352668550574,
      "loss": 4.5643,
      "mean_token_accuracy": 0.2630235493183136,
      "num_tokens": 182317277.0,
      "step": 79520
    },
    {
      "entropy": 5.519849443435669,
      "epoch": 7.6392891450528335,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011016019013604459,
      "loss": 4.6434,
      "mean_token_accuracy": 0.2673695132136345,
      "num_tokens": 182328781.0,
      "step": 79525
    },
    {
      "entropy": 5.435085201263428,
      "epoch": 7.639769452449568,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011013685741367439,
      "loss": 4.538,
      "mean_token_accuracy": 0.2778972566127777,
      "num_tokens": 182340116.0,
      "step": 79530
    },
    {
      "entropy": 5.545442199707031,
      "epoch": 7.640249759846301,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011011352851893681,
      "loss": 4.7125,
      "mean_token_accuracy": 0.2611470803618431,
      "num_tokens": 182353216.0,
      "step": 79535
    },
    {
      "entropy": 5.520909261703491,
      "epoch": 7.640730067243036,
      "grad_norm": 1.578125,
      "learning_rate": 0.00011009020345237346,
      "loss": 4.6196,
      "mean_token_accuracy": 0.2661642625927925,
      "num_tokens": 182365148.0,
      "step": 79540
    },
    {
      "entropy": 5.5264991283416744,
      "epoch": 7.641210374639769,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001100668822145259,
      "loss": 4.6045,
      "mean_token_accuracy": 0.26124423146247866,
      "num_tokens": 182377576.0,
      "step": 79545
    },
    {
      "entropy": 5.522027826309204,
      "epoch": 7.641690682036503,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011004356480593567,
      "loss": 4.593,
      "mean_token_accuracy": 0.264550644159317,
      "num_tokens": 182390277.0,
      "step": 79550
    },
    {
      "entropy": 5.5052502155303955,
      "epoch": 7.642170989433238,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011002025122714395,
      "loss": 4.5897,
      "mean_token_accuracy": 0.26901935040950775,
      "num_tokens": 182401762.0,
      "step": 79555
    },
    {
      "entropy": 5.528386068344116,
      "epoch": 7.642651296829971,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010999694147869216,
      "loss": 4.6345,
      "mean_token_accuracy": 0.259711979329586,
      "num_tokens": 182414445.0,
      "step": 79560
    },
    {
      "entropy": 5.5469427585601805,
      "epoch": 7.6431316042267055,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010997363556112139,
      "loss": 4.7101,
      "mean_token_accuracy": 0.2616821125149727,
      "num_tokens": 182426064.0,
      "step": 79565
    },
    {
      "entropy": 5.556279277801513,
      "epoch": 7.643611911623439,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010995033347497276,
      "loss": 4.6565,
      "mean_token_accuracy": 0.2610418230295181,
      "num_tokens": 182437370.0,
      "step": 79570
    },
    {
      "entropy": 5.517169761657715,
      "epoch": 7.644092219020173,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010992703522078733,
      "loss": 4.6456,
      "mean_token_accuracy": 0.267591167986393,
      "num_tokens": 182448011.0,
      "step": 79575
    },
    {
      "entropy": 5.531075096130371,
      "epoch": 7.644572526416907,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010990374079910604,
      "loss": 4.6527,
      "mean_token_accuracy": 0.2640190124511719,
      "num_tokens": 182459388.0,
      "step": 79580
    },
    {
      "entropy": 5.483650827407837,
      "epoch": 7.645052833813641,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010988045021046956,
      "loss": 4.5522,
      "mean_token_accuracy": 0.2695238173007965,
      "num_tokens": 182470132.0,
      "step": 79585
    },
    {
      "entropy": 5.545689678192138,
      "epoch": 7.645533141210374,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010985716345541876,
      "loss": 4.6587,
      "mean_token_accuracy": 0.2636184200644493,
      "num_tokens": 182481350.0,
      "step": 79590
    },
    {
      "entropy": 5.622100400924682,
      "epoch": 7.646013448607109,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010983388053449423,
      "loss": 4.7376,
      "mean_token_accuracy": 0.2594169393181801,
      "num_tokens": 182493422.0,
      "step": 79595
    },
    {
      "entropy": 5.4762297630310055,
      "epoch": 7.646493756003842,
      "grad_norm": 1.375,
      "learning_rate": 0.00010981060144823655,
      "loss": 4.5445,
      "mean_token_accuracy": 0.2695756644010544,
      "num_tokens": 182505468.0,
      "step": 79600
    },
    {
      "entropy": 5.504778003692627,
      "epoch": 7.6469740634005765,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010978732619718618,
      "loss": 4.6013,
      "mean_token_accuracy": 0.2734478384256363,
      "num_tokens": 182516604.0,
      "step": 79605
    },
    {
      "entropy": 5.486139345169067,
      "epoch": 7.64745437079731,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010976405478188361,
      "loss": 4.5542,
      "mean_token_accuracy": 0.2747410774230957,
      "num_tokens": 182526996.0,
      "step": 79610
    },
    {
      "entropy": 5.542107725143433,
      "epoch": 7.647934678194044,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010974078720286892,
      "loss": 4.6521,
      "mean_token_accuracy": 0.2650338351726532,
      "num_tokens": 182539020.0,
      "step": 79615
    },
    {
      "entropy": 5.531624174118042,
      "epoch": 7.648414985590778,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010971752346068248,
      "loss": 4.6593,
      "mean_token_accuracy": 0.2621752917766571,
      "num_tokens": 182551083.0,
      "step": 79620
    },
    {
      "entropy": 5.540549087524414,
      "epoch": 7.648895292987512,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010969426355586434,
      "loss": 4.6102,
      "mean_token_accuracy": 0.26033976674079895,
      "num_tokens": 182561521.0,
      "step": 79625
    },
    {
      "entropy": 5.621037006378174,
      "epoch": 7.649375600384246,
      "grad_norm": 1.5,
      "learning_rate": 0.00010967100748895454,
      "loss": 4.7576,
      "mean_token_accuracy": 0.2535313665866852,
      "num_tokens": 182574180.0,
      "step": 79630
    },
    {
      "entropy": 5.493536567687988,
      "epoch": 7.64985590778098,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010964775526049306,
      "loss": 4.5648,
      "mean_token_accuracy": 0.27116797119379044,
      "num_tokens": 182586055.0,
      "step": 79635
    },
    {
      "entropy": 5.542221164703369,
      "epoch": 7.650336215177714,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010962450687101966,
      "loss": 4.638,
      "mean_token_accuracy": 0.2684953808784485,
      "num_tokens": 182596457.0,
      "step": 79640
    },
    {
      "entropy": 5.5171253204345705,
      "epoch": 7.6508165225744476,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010960126232107421,
      "loss": 4.595,
      "mean_token_accuracy": 0.271727454662323,
      "num_tokens": 182608469.0,
      "step": 79645
    },
    {
      "entropy": 5.600767660140991,
      "epoch": 7.651296829971182,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010957802161119629,
      "loss": 4.7268,
      "mean_token_accuracy": 0.2548773795366287,
      "num_tokens": 182619810.0,
      "step": 79650
    },
    {
      "entropy": 5.521220922470093,
      "epoch": 7.651777137367915,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00010955478474192546,
      "loss": 4.6334,
      "mean_token_accuracy": 0.27076219767332077,
      "num_tokens": 182629666.0,
      "step": 79655
    },
    {
      "entropy": 5.524881362915039,
      "epoch": 7.65225744476465,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010953155171380129,
      "loss": 4.6208,
      "mean_token_accuracy": 0.269991272687912,
      "num_tokens": 182640835.0,
      "step": 79660
    },
    {
      "entropy": 5.620256662368774,
      "epoch": 7.652737752161383,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010950832252736313,
      "loss": 4.7689,
      "mean_token_accuracy": 0.2549885779619217,
      "num_tokens": 182653290.0,
      "step": 79665
    },
    {
      "entropy": 5.503326368331909,
      "epoch": 7.653218059558117,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010948509718315032,
      "loss": 4.5699,
      "mean_token_accuracy": 0.2694094032049179,
      "num_tokens": 182665372.0,
      "step": 79670
    },
    {
      "entropy": 5.466679286956787,
      "epoch": 7.653698366954851,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00010946187568170209,
      "loss": 4.5221,
      "mean_token_accuracy": 0.27254725694656373,
      "num_tokens": 182675972.0,
      "step": 79675
    },
    {
      "entropy": 5.530858945846558,
      "epoch": 7.654178674351585,
      "grad_norm": 1.375,
      "learning_rate": 0.00010943865802355752,
      "loss": 4.69,
      "mean_token_accuracy": 0.2632018491625786,
      "num_tokens": 182687476.0,
      "step": 79680
    },
    {
      "entropy": 5.5410158157348635,
      "epoch": 7.654658981748319,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010941544420925581,
      "loss": 4.6516,
      "mean_token_accuracy": 0.26698445826768874,
      "num_tokens": 182698319.0,
      "step": 79685
    },
    {
      "entropy": 5.493991088867188,
      "epoch": 7.655139289145053,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010939223423933572,
      "loss": 4.5206,
      "mean_token_accuracy": 0.27713309079408643,
      "num_tokens": 182709386.0,
      "step": 79690
    },
    {
      "entropy": 5.527962350845337,
      "epoch": 7.655619596541786,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010936902811433617,
      "loss": 4.6427,
      "mean_token_accuracy": 0.2667836010456085,
      "num_tokens": 182720992.0,
      "step": 79695
    },
    {
      "entropy": 5.450780010223388,
      "epoch": 7.656099903938521,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010934582583479596,
      "loss": 4.5403,
      "mean_token_accuracy": 0.2724055558443069,
      "num_tokens": 182732694.0,
      "step": 79700
    },
    {
      "entropy": 5.528377199172974,
      "epoch": 7.656580211335255,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010932262740125381,
      "loss": 4.637,
      "mean_token_accuracy": 0.26905506253242495,
      "num_tokens": 182744991.0,
      "step": 79705
    },
    {
      "entropy": 5.523064804077149,
      "epoch": 7.6570605187319885,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010929943281424828,
      "loss": 4.6484,
      "mean_token_accuracy": 0.2721130445599556,
      "num_tokens": 182758033.0,
      "step": 79710
    },
    {
      "entropy": 5.565268421173096,
      "epoch": 7.657540826128722,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010927624207431791,
      "loss": 4.6763,
      "mean_token_accuracy": 0.2658912941813469,
      "num_tokens": 182769959.0,
      "step": 79715
    },
    {
      "entropy": 5.558621454238891,
      "epoch": 7.658021133525456,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001092530551820011,
      "loss": 4.6214,
      "mean_token_accuracy": 0.2654190957546234,
      "num_tokens": 182781362.0,
      "step": 79720
    },
    {
      "entropy": 5.483369398117065,
      "epoch": 7.6585014409221905,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010922987213783623,
      "loss": 4.525,
      "mean_token_accuracy": 0.2793955713510513,
      "num_tokens": 182793164.0,
      "step": 79725
    },
    {
      "entropy": 5.48122091293335,
      "epoch": 7.658981748318924,
      "grad_norm": 1.5,
      "learning_rate": 0.00010920669294236141,
      "loss": 4.5796,
      "mean_token_accuracy": 0.27688014805316924,
      "num_tokens": 182804445.0,
      "step": 79730
    },
    {
      "entropy": 5.558041334152222,
      "epoch": 7.659462055715658,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010918351759611491,
      "loss": 4.679,
      "mean_token_accuracy": 0.2598605901002884,
      "num_tokens": 182816446.0,
      "step": 79735
    },
    {
      "entropy": 5.646425437927246,
      "epoch": 7.659942363112392,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010916034609963474,
      "loss": 4.7693,
      "mean_token_accuracy": 0.24996655136346818,
      "num_tokens": 182829088.0,
      "step": 79740
    },
    {
      "entropy": 5.490480184555054,
      "epoch": 7.660422670509126,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010913717845345887,
      "loss": 4.5493,
      "mean_token_accuracy": 0.27062030136585236,
      "num_tokens": 182841073.0,
      "step": 79745
    },
    {
      "entropy": 5.550104284286499,
      "epoch": 7.6609029779058595,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010911401465812524,
      "loss": 4.613,
      "mean_token_accuracy": 0.26721980720758437,
      "num_tokens": 182852829.0,
      "step": 79750
    },
    {
      "entropy": 5.551048755645752,
      "epoch": 7.661383285302594,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001090908547141716,
      "loss": 4.6455,
      "mean_token_accuracy": 0.26894629150629046,
      "num_tokens": 182865203.0,
      "step": 79755
    },
    {
      "entropy": 5.585287809371948,
      "epoch": 7.661863592699327,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010906769862213572,
      "loss": 4.7092,
      "mean_token_accuracy": 0.2606659397482872,
      "num_tokens": 182877544.0,
      "step": 79760
    },
    {
      "entropy": 5.4869513511657715,
      "epoch": 7.662343900096062,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001090445463825551,
      "loss": 4.5319,
      "mean_token_accuracy": 0.2754721909761429,
      "num_tokens": 182888750.0,
      "step": 79765
    },
    {
      "entropy": 5.569169902801514,
      "epoch": 7.662824207492795,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010902139799596733,
      "loss": 4.633,
      "mean_token_accuracy": 0.26750215142965317,
      "num_tokens": 182899581.0,
      "step": 79770
    },
    {
      "entropy": 5.6038830280303955,
      "epoch": 7.663304514889529,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010899825346290984,
      "loss": 4.6758,
      "mean_token_accuracy": 0.2622648343443871,
      "num_tokens": 182910528.0,
      "step": 79775
    },
    {
      "entropy": 5.420995759963989,
      "epoch": 7.663784822286264,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010897511278391997,
      "loss": 4.4865,
      "mean_token_accuracy": 0.27729442417621614,
      "num_tokens": 182921557.0,
      "step": 79780
    },
    {
      "entropy": 5.523232269287109,
      "epoch": 7.664265129682997,
      "grad_norm": 1.28125,
      "learning_rate": 0.00010895197595953502,
      "loss": 4.6257,
      "mean_token_accuracy": 0.2705099806189537,
      "num_tokens": 182933263.0,
      "step": 79785
    },
    {
      "entropy": 5.463152599334717,
      "epoch": 7.6647454370797306,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010892884299029218,
      "loss": 4.5497,
      "mean_token_accuracy": 0.27306444346904757,
      "num_tokens": 182943711.0,
      "step": 79790
    },
    {
      "entropy": 5.463147830963135,
      "epoch": 7.665225744476465,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001089057138767284,
      "loss": 4.5708,
      "mean_token_accuracy": 0.26998473703861237,
      "num_tokens": 182954749.0,
      "step": 79795
    },
    {
      "entropy": 5.48923659324646,
      "epoch": 7.665706051873199,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010888258861938074,
      "loss": 4.5952,
      "mean_token_accuracy": 0.26638136357069014,
      "num_tokens": 182966106.0,
      "step": 79800
    },
    {
      "entropy": 5.51689076423645,
      "epoch": 7.666186359269933,
      "grad_norm": 1.59375,
      "learning_rate": 0.00010885946721878612,
      "loss": 4.6255,
      "mean_token_accuracy": 0.26809030026197433,
      "num_tokens": 182976351.0,
      "step": 79805
    },
    {
      "entropy": 5.525360012054444,
      "epoch": 7.666666666666667,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010883634967548133,
      "loss": 4.6446,
      "mean_token_accuracy": 0.2655484572052956,
      "num_tokens": 182988093.0,
      "step": 79810
    },
    {
      "entropy": 5.495399236679077,
      "epoch": 7.6671469740634,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001088132359900031,
      "loss": 4.6069,
      "mean_token_accuracy": 0.27502452582120895,
      "num_tokens": 182999347.0,
      "step": 79815
    },
    {
      "entropy": 5.502859687805175,
      "epoch": 7.667627281460135,
      "grad_norm": 1.375,
      "learning_rate": 0.00010879012616288813,
      "loss": 4.5648,
      "mean_token_accuracy": 0.27038607001304626,
      "num_tokens": 183010047.0,
      "step": 79820
    },
    {
      "entropy": 5.54498143196106,
      "epoch": 7.668107588856868,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001087670201946728,
      "loss": 4.656,
      "mean_token_accuracy": 0.25907588452100755,
      "num_tokens": 183023637.0,
      "step": 79825
    },
    {
      "entropy": 5.534092092514038,
      "epoch": 7.6685878962536025,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010874391808589366,
      "loss": 4.6317,
      "mean_token_accuracy": 0.2688953012228012,
      "num_tokens": 183034661.0,
      "step": 79830
    },
    {
      "entropy": 5.546588516235351,
      "epoch": 7.669068203650336,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010872081983708707,
      "loss": 4.566,
      "mean_token_accuracy": 0.2680308982729912,
      "num_tokens": 183046399.0,
      "step": 79835
    },
    {
      "entropy": 5.508970594406128,
      "epoch": 7.66954851104707,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010869772544878928,
      "loss": 4.5878,
      "mean_token_accuracy": 0.2769902914762497,
      "num_tokens": 183058408.0,
      "step": 79840
    },
    {
      "entropy": 5.5533819675445555,
      "epoch": 7.670028818443804,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001086746349215365,
      "loss": 4.6518,
      "mean_token_accuracy": 0.2664509117603302,
      "num_tokens": 183070403.0,
      "step": 79845
    },
    {
      "entropy": 5.451980257034302,
      "epoch": 7.670509125840538,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010865154825586487,
      "loss": 4.5363,
      "mean_token_accuracy": 0.2781476631760597,
      "num_tokens": 183082552.0,
      "step": 79850
    },
    {
      "entropy": 5.457070016860962,
      "epoch": 7.670989433237272,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010862846545231028,
      "loss": 4.5407,
      "mean_token_accuracy": 0.27713934779167176,
      "num_tokens": 183093971.0,
      "step": 79855
    },
    {
      "entropy": 5.572148323059082,
      "epoch": 7.671469740634006,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010860538651140869,
      "loss": 4.7365,
      "mean_token_accuracy": 0.25477424412965777,
      "num_tokens": 183106108.0,
      "step": 79860
    },
    {
      "entropy": 5.524488496780395,
      "epoch": 7.671950048030739,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010858231143369596,
      "loss": 4.6048,
      "mean_token_accuracy": 0.26960730701684954,
      "num_tokens": 183117102.0,
      "step": 79865
    },
    {
      "entropy": 5.542973804473877,
      "epoch": 7.6724303554274735,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010855924021970777,
      "loss": 4.6314,
      "mean_token_accuracy": 0.2652950346469879,
      "num_tokens": 183129311.0,
      "step": 79870
    },
    {
      "entropy": 5.508196115493774,
      "epoch": 7.672910662824208,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010853617286997982,
      "loss": 4.5854,
      "mean_token_accuracy": 0.27099101394414904,
      "num_tokens": 183140040.0,
      "step": 79875
    },
    {
      "entropy": 5.481262683868408,
      "epoch": 7.673390970220941,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010851310938504769,
      "loss": 4.604,
      "mean_token_accuracy": 0.26569440960884094,
      "num_tokens": 183151718.0,
      "step": 79880
    },
    {
      "entropy": 5.519116306304932,
      "epoch": 7.673871277617676,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010849004976544674,
      "loss": 4.6609,
      "mean_token_accuracy": 0.26225848197937013,
      "num_tokens": 183164324.0,
      "step": 79885
    },
    {
      "entropy": 5.489718818664551,
      "epoch": 7.674351585014409,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010846699401171237,
      "loss": 4.5566,
      "mean_token_accuracy": 0.2689977496862411,
      "num_tokens": 183176127.0,
      "step": 79890
    },
    {
      "entropy": 5.516720819473266,
      "epoch": 7.674831892411143,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010844394212437994,
      "loss": 4.5798,
      "mean_token_accuracy": 0.26813667714595796,
      "num_tokens": 183186851.0,
      "step": 79895
    },
    {
      "entropy": 5.569597339630127,
      "epoch": 7.675312199807877,
      "grad_norm": 1.375,
      "learning_rate": 0.00010842089410398458,
      "loss": 4.6386,
      "mean_token_accuracy": 0.26972910314798354,
      "num_tokens": 183198572.0,
      "step": 79900
    },
    {
      "entropy": 5.524479007720947,
      "epoch": 7.675792507204611,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010839784995106142,
      "loss": 4.5762,
      "mean_token_accuracy": 0.2741483971476555,
      "num_tokens": 183210366.0,
      "step": 79905
    },
    {
      "entropy": 5.591636371612549,
      "epoch": 7.676272814601345,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010837480966614556,
      "loss": 4.6853,
      "mean_token_accuracy": 0.2645750492811203,
      "num_tokens": 183221469.0,
      "step": 79910
    },
    {
      "entropy": 5.449346542358398,
      "epoch": 7.676753121998079,
      "grad_norm": 1.5,
      "learning_rate": 0.00010835177324977177,
      "loss": 4.5498,
      "mean_token_accuracy": 0.2733680486679077,
      "num_tokens": 183232712.0,
      "step": 79915
    },
    {
      "entropy": 5.5105737209320065,
      "epoch": 7.677233429394812,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010832874070247496,
      "loss": 4.5997,
      "mean_token_accuracy": 0.2697826877236366,
      "num_tokens": 183244463.0,
      "step": 79920
    },
    {
      "entropy": 5.590111970901489,
      "epoch": 7.677713736791547,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010830571202478986,
      "loss": 4.6835,
      "mean_token_accuracy": 0.26750879287719725,
      "num_tokens": 183256263.0,
      "step": 79925
    },
    {
      "entropy": 5.484096813201904,
      "epoch": 7.678194044188281,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010828268721725118,
      "loss": 4.6296,
      "mean_token_accuracy": 0.26846968531608584,
      "num_tokens": 183268514.0,
      "step": 79930
    },
    {
      "entropy": 5.486729526519776,
      "epoch": 7.678674351585014,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010825966628039344,
      "loss": 4.5464,
      "mean_token_accuracy": 0.2809972375631332,
      "num_tokens": 183280502.0,
      "step": 79935
    },
    {
      "entropy": 5.572740125656128,
      "epoch": 7.679154658981748,
      "grad_norm": 1.375,
      "learning_rate": 0.00010823664921475118,
      "loss": 4.6743,
      "mean_token_accuracy": 0.2598456904292107,
      "num_tokens": 183291906.0,
      "step": 79940
    },
    {
      "entropy": 5.5778160572052,
      "epoch": 7.679634966378482,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001082136360208587,
      "loss": 4.6844,
      "mean_token_accuracy": 0.26473683416843413,
      "num_tokens": 183303231.0,
      "step": 79945
    },
    {
      "entropy": 5.46283745765686,
      "epoch": 7.6801152737752165,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010819062669925031,
      "loss": 4.5387,
      "mean_token_accuracy": 0.2777345061302185,
      "num_tokens": 183314198.0,
      "step": 79950
    },
    {
      "entropy": 5.527227592468262,
      "epoch": 7.68059558117195,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010816762125046029,
      "loss": 4.6827,
      "mean_token_accuracy": 0.262746599316597,
      "num_tokens": 183325041.0,
      "step": 79955
    },
    {
      "entropy": 5.491460084915161,
      "epoch": 7.681075888568684,
      "grad_norm": 1.5,
      "learning_rate": 0.00010814461967502271,
      "loss": 4.5843,
      "mean_token_accuracy": 0.2725193440914154,
      "num_tokens": 183336862.0,
      "step": 79960
    },
    {
      "entropy": 5.563893890380859,
      "epoch": 7.681556195965418,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010812162197347155,
      "loss": 4.6788,
      "mean_token_accuracy": 0.2622549936175346,
      "num_tokens": 183347490.0,
      "step": 79965
    },
    {
      "entropy": 5.558601713180542,
      "epoch": 7.682036503362152,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010809862814634091,
      "loss": 4.5927,
      "mean_token_accuracy": 0.2695405945181847,
      "num_tokens": 183358725.0,
      "step": 79970
    },
    {
      "entropy": 5.563027811050415,
      "epoch": 7.6825168107588855,
      "grad_norm": 1.5,
      "learning_rate": 0.00010807563819416444,
      "loss": 4.6017,
      "mean_token_accuracy": 0.27427230179309847,
      "num_tokens": 183369904.0,
      "step": 79975
    },
    {
      "entropy": 5.538242340087891,
      "epoch": 7.68299711815562,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010805265211747599,
      "loss": 4.5915,
      "mean_token_accuracy": 0.26932506412267687,
      "num_tokens": 183381201.0,
      "step": 79980
    },
    {
      "entropy": 5.541587066650391,
      "epoch": 7.683477425552353,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010802966991680923,
      "loss": 4.6458,
      "mean_token_accuracy": 0.2628519520163536,
      "num_tokens": 183392741.0,
      "step": 79985
    },
    {
      "entropy": 5.5122308254241945,
      "epoch": 7.683957732949088,
      "grad_norm": 1.375,
      "learning_rate": 0.00010800669159269775,
      "loss": 4.6052,
      "mean_token_accuracy": 0.2729429006576538,
      "num_tokens": 183404666.0,
      "step": 79990
    },
    {
      "entropy": 5.499029064178467,
      "epoch": 7.684438040345821,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010798371714567501,
      "loss": 4.6192,
      "mean_token_accuracy": 0.2722224473953247,
      "num_tokens": 183415253.0,
      "step": 79995
    },
    {
      "entropy": 5.466496229171753,
      "epoch": 7.684918347742555,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001079607465762745,
      "loss": 4.624,
      "mean_token_accuracy": 0.2696531772613525,
      "num_tokens": 183426252.0,
      "step": 80000
    },
    {
      "entropy": 5.5243080139160154,
      "epoch": 7.685398655139289,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010793777988502937,
      "loss": 4.6029,
      "mean_token_accuracy": 0.2684288263320923,
      "num_tokens": 183438724.0,
      "step": 80005
    },
    {
      "entropy": 5.562285375595093,
      "epoch": 7.685878962536023,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010791481707247291,
      "loss": 4.6546,
      "mean_token_accuracy": 0.2625546395778656,
      "num_tokens": 183451528.0,
      "step": 80010
    },
    {
      "entropy": 5.505018520355224,
      "epoch": 7.6863592699327565,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001078918581391383,
      "loss": 4.5714,
      "mean_token_accuracy": 0.2742940798401833,
      "num_tokens": 183463258.0,
      "step": 80015
    },
    {
      "entropy": 5.472805833816528,
      "epoch": 7.686839577329491,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001078689030855585,
      "loss": 4.5453,
      "mean_token_accuracy": 0.276374576985836,
      "num_tokens": 183474755.0,
      "step": 80020
    },
    {
      "entropy": 5.523767137527466,
      "epoch": 7.687319884726225,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010784595191226648,
      "loss": 4.6158,
      "mean_token_accuracy": 0.2662804827094078,
      "num_tokens": 183485391.0,
      "step": 80025
    },
    {
      "entropy": 5.479464960098267,
      "epoch": 7.687800192122959,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010782300461979519,
      "loss": 4.5606,
      "mean_token_accuracy": 0.27160156816244124,
      "num_tokens": 183495960.0,
      "step": 80030
    },
    {
      "entropy": 5.51237530708313,
      "epoch": 7.688280499519693,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010780006120867728,
      "loss": 4.6121,
      "mean_token_accuracy": 0.26929259300231934,
      "num_tokens": 183506908.0,
      "step": 80035
    },
    {
      "entropy": 5.591991567611695,
      "epoch": 7.688760806916426,
      "grad_norm": 1.609375,
      "learning_rate": 0.00010777712167944543,
      "loss": 4.716,
      "mean_token_accuracy": 0.2608791202306747,
      "num_tokens": 183517832.0,
      "step": 80040
    },
    {
      "entropy": 5.593932962417602,
      "epoch": 7.689241114313161,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010775418603263226,
      "loss": 4.6703,
      "mean_token_accuracy": 0.26597338914871216,
      "num_tokens": 183529480.0,
      "step": 80045
    },
    {
      "entropy": 5.484368991851807,
      "epoch": 7.689721421709894,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010773125426877031,
      "loss": 4.538,
      "mean_token_accuracy": 0.26972628831863404,
      "num_tokens": 183540898.0,
      "step": 80050
    },
    {
      "entropy": 5.51813850402832,
      "epoch": 7.6902017291066285,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010770832638839192,
      "loss": 4.6645,
      "mean_token_accuracy": 0.26355465352535246,
      "num_tokens": 183552057.0,
      "step": 80055
    },
    {
      "entropy": 5.504279947280883,
      "epoch": 7.690682036503362,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00010768540239202951,
      "loss": 4.614,
      "mean_token_accuracy": 0.272217221558094,
      "num_tokens": 183563588.0,
      "step": 80060
    },
    {
      "entropy": 5.5196850299835205,
      "epoch": 7.691162343900096,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010766248228021518,
      "loss": 4.594,
      "mean_token_accuracy": 0.27020968347787855,
      "num_tokens": 183575703.0,
      "step": 80065
    },
    {
      "entropy": 5.539829301834106,
      "epoch": 7.69164265129683,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010763956605348112,
      "loss": 4.6348,
      "mean_token_accuracy": 0.26112128496170045,
      "num_tokens": 183587539.0,
      "step": 80070
    },
    {
      "entropy": 5.483985376358032,
      "epoch": 7.692122958693564,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010761665371235934,
      "loss": 4.583,
      "mean_token_accuracy": 0.2659306928515434,
      "num_tokens": 183599365.0,
      "step": 80075
    },
    {
      "entropy": 5.451831293106079,
      "epoch": 7.692603266090297,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010759374525738189,
      "loss": 4.5322,
      "mean_token_accuracy": 0.274740269780159,
      "num_tokens": 183610551.0,
      "step": 80080
    },
    {
      "entropy": 5.5853844165802,
      "epoch": 7.693083573487032,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010757084068908055,
      "loss": 4.6531,
      "mean_token_accuracy": 0.2637693271040916,
      "num_tokens": 183621380.0,
      "step": 80085
    },
    {
      "entropy": 5.5063214778900145,
      "epoch": 7.693563880883765,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001075479400079872,
      "loss": 4.5998,
      "mean_token_accuracy": 0.2640496611595154,
      "num_tokens": 183632685.0,
      "step": 80090
    },
    {
      "entropy": 5.4841163635253904,
      "epoch": 7.6940441882804995,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010752504321463338,
      "loss": 4.566,
      "mean_token_accuracy": 0.2663381636142731,
      "num_tokens": 183642641.0,
      "step": 80095
    },
    {
      "entropy": 5.495046091079712,
      "epoch": 7.694524495677234,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010750215030955076,
      "loss": 4.553,
      "mean_token_accuracy": 0.2744944840669632,
      "num_tokens": 183653437.0,
      "step": 80100
    },
    {
      "entropy": 5.486144495010376,
      "epoch": 7.695004803073967,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010747926129327087,
      "loss": 4.5654,
      "mean_token_accuracy": 0.2689700245857239,
      "num_tokens": 183665650.0,
      "step": 80105
    },
    {
      "entropy": 5.549837875366211,
      "epoch": 7.695485110470702,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010745637616632508,
      "loss": 4.657,
      "mean_token_accuracy": 0.27224338203668597,
      "num_tokens": 183676565.0,
      "step": 80110
    },
    {
      "entropy": 5.529689788818359,
      "epoch": 7.695965417867435,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010743349492924474,
      "loss": 4.6021,
      "mean_token_accuracy": 0.26322357505559923,
      "num_tokens": 183688153.0,
      "step": 80115
    },
    {
      "entropy": 5.624380445480346,
      "epoch": 7.696445725264169,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001074106175825611,
      "loss": 4.7276,
      "mean_token_accuracy": 0.2623990163207054,
      "num_tokens": 183700808.0,
      "step": 80120
    },
    {
      "entropy": 5.517118453979492,
      "epoch": 7.696926032660903,
      "grad_norm": 1.5,
      "learning_rate": 0.00010738774412680527,
      "loss": 4.604,
      "mean_token_accuracy": 0.26920960396528243,
      "num_tokens": 183712943.0,
      "step": 80125
    },
    {
      "entropy": 5.525986528396606,
      "epoch": 7.697406340057637,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010736487456250827,
      "loss": 4.6276,
      "mean_token_accuracy": 0.26540369391441343,
      "num_tokens": 183725379.0,
      "step": 80130
    },
    {
      "entropy": 5.509902477264404,
      "epoch": 7.697886647454371,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010734200889020115,
      "loss": 4.6155,
      "mean_token_accuracy": 0.2590750530362129,
      "num_tokens": 183736503.0,
      "step": 80135
    },
    {
      "entropy": 5.573500061035157,
      "epoch": 7.698366954851105,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001073191471104147,
      "loss": 4.6317,
      "mean_token_accuracy": 0.26774545311927794,
      "num_tokens": 183747073.0,
      "step": 80140
    },
    {
      "entropy": 5.450863075256348,
      "epoch": 7.698847262247838,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010729628922367977,
      "loss": 4.541,
      "mean_token_accuracy": 0.2734984219074249,
      "num_tokens": 183758288.0,
      "step": 80145
    },
    {
      "entropy": 5.517054605484009,
      "epoch": 7.699327569644573,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001072734352305271,
      "loss": 4.6436,
      "mean_token_accuracy": 0.26284263730049134,
      "num_tokens": 183770289.0,
      "step": 80150
    },
    {
      "entropy": 5.55720362663269,
      "epoch": 7.699807877041306,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010725058513148716,
      "loss": 4.6324,
      "mean_token_accuracy": 0.26971179246902466,
      "num_tokens": 183782023.0,
      "step": 80155
    },
    {
      "entropy": 5.543245124816894,
      "epoch": 7.70028818443804,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010722773892709048,
      "loss": 4.6297,
      "mean_token_accuracy": 0.26057397574186325,
      "num_tokens": 183792796.0,
      "step": 80160
    },
    {
      "entropy": 5.521967029571533,
      "epoch": 7.700768491834774,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010720489661786753,
      "loss": 4.6074,
      "mean_token_accuracy": 0.2649917036294937,
      "num_tokens": 183804801.0,
      "step": 80165
    },
    {
      "entropy": 5.578129816055298,
      "epoch": 7.701248799231508,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010718205820434865,
      "loss": 4.6722,
      "mean_token_accuracy": 0.26343432664871214,
      "num_tokens": 183816610.0,
      "step": 80170
    },
    {
      "entropy": 5.53313102722168,
      "epoch": 7.7017291066282425,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010715922368706408,
      "loss": 4.6258,
      "mean_token_accuracy": 0.26390956789255143,
      "num_tokens": 183827602.0,
      "step": 80175
    },
    {
      "entropy": 5.558081483840942,
      "epoch": 7.702209414024976,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010713639306654398,
      "loss": 4.6222,
      "mean_token_accuracy": 0.26815055012702943,
      "num_tokens": 183839442.0,
      "step": 80180
    },
    {
      "entropy": 5.523237514495849,
      "epoch": 7.70268972142171,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00010711356634331834,
      "loss": 4.6321,
      "mean_token_accuracy": 0.26831458508968353,
      "num_tokens": 183851207.0,
      "step": 80185
    },
    {
      "entropy": 5.535186719894409,
      "epoch": 7.703170028818444,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010709074351791715,
      "loss": 4.6488,
      "mean_token_accuracy": 0.26843591332435607,
      "num_tokens": 183863101.0,
      "step": 80190
    },
    {
      "entropy": 5.491516637802124,
      "epoch": 7.703650336215178,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001070679245908703,
      "loss": 4.5862,
      "mean_token_accuracy": 0.2695355385541916,
      "num_tokens": 183874262.0,
      "step": 80195
    },
    {
      "entropy": 5.496145248413086,
      "epoch": 7.7041306436119115,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010704510956270761,
      "loss": 4.5514,
      "mean_token_accuracy": 0.2748636528849602,
      "num_tokens": 183886305.0,
      "step": 80200
    },
    {
      "entropy": 5.49645209312439,
      "epoch": 7.704610951008646,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010702229843395874,
      "loss": 4.6273,
      "mean_token_accuracy": 0.2687380760908127,
      "num_tokens": 183897220.0,
      "step": 80205
    },
    {
      "entropy": 5.48027195930481,
      "epoch": 7.705091258405379,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010699949120515332,
      "loss": 4.5181,
      "mean_token_accuracy": 0.2687672898173332,
      "num_tokens": 183907061.0,
      "step": 80210
    },
    {
      "entropy": 5.4945385456085205,
      "epoch": 7.7055715658021136,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010697668787682085,
      "loss": 4.5635,
      "mean_token_accuracy": 0.2679160624742508,
      "num_tokens": 183918533.0,
      "step": 80215
    },
    {
      "entropy": 5.578485202789307,
      "epoch": 7.706051873198847,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010695388844949082,
      "loss": 4.6771,
      "mean_token_accuracy": 0.26214939504861834,
      "num_tokens": 183929887.0,
      "step": 80220
    },
    {
      "entropy": 5.556217908859253,
      "epoch": 7.706532180595581,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010693109292369243,
      "loss": 4.6927,
      "mean_token_accuracy": 0.26680581718683244,
      "num_tokens": 183941309.0,
      "step": 80225
    },
    {
      "entropy": 5.531804084777832,
      "epoch": 7.707012487992315,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010690830129995504,
      "loss": 4.5947,
      "mean_token_accuracy": 0.2656926244497299,
      "num_tokens": 183954128.0,
      "step": 80230
    },
    {
      "entropy": 5.549935054779053,
      "epoch": 7.707492795389049,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010688551357880773,
      "loss": 4.6036,
      "mean_token_accuracy": 0.26593125611543655,
      "num_tokens": 183965678.0,
      "step": 80235
    },
    {
      "entropy": 5.488738489151001,
      "epoch": 7.7079731027857825,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001068627297607796,
      "loss": 4.56,
      "mean_token_accuracy": 0.2708927094936371,
      "num_tokens": 183977031.0,
      "step": 80240
    },
    {
      "entropy": 5.482057332992554,
      "epoch": 7.708453410182517,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010683994984639964,
      "loss": 4.6,
      "mean_token_accuracy": 0.2657077983021736,
      "num_tokens": 183988108.0,
      "step": 80245
    },
    {
      "entropy": 5.501138639450073,
      "epoch": 7.708933717579251,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010681717383619666,
      "loss": 4.5699,
      "mean_token_accuracy": 0.27348951399326327,
      "num_tokens": 183999710.0,
      "step": 80250
    },
    {
      "entropy": 5.52293815612793,
      "epoch": 7.709414024975985,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010679440173069955,
      "loss": 4.5735,
      "mean_token_accuracy": 0.2693258851766586,
      "num_tokens": 184011398.0,
      "step": 80255
    },
    {
      "entropy": 5.509024953842163,
      "epoch": 7.709894332372719,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010677163353043704,
      "loss": 4.5936,
      "mean_token_accuracy": 0.27044808864593506,
      "num_tokens": 184022667.0,
      "step": 80260
    },
    {
      "entropy": 5.586342859268188,
      "epoch": 7.710374639769452,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010674886923593754,
      "loss": 4.6758,
      "mean_token_accuracy": 0.26859264969825747,
      "num_tokens": 184034243.0,
      "step": 80265
    },
    {
      "entropy": 5.482898712158203,
      "epoch": 7.710854947166187,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010672610884772976,
      "loss": 4.6012,
      "mean_token_accuracy": 0.2698423594236374,
      "num_tokens": 184045537.0,
      "step": 80270
    },
    {
      "entropy": 5.5304077625274655,
      "epoch": 7.71133525456292,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010670335236634204,
      "loss": 4.5913,
      "mean_token_accuracy": 0.26959712952375414,
      "num_tokens": 184056748.0,
      "step": 80275
    },
    {
      "entropy": 5.506354093551636,
      "epoch": 7.7118155619596545,
      "grad_norm": 1.5,
      "learning_rate": 0.00010668059979230273,
      "loss": 4.6213,
      "mean_token_accuracy": 0.26520259976387023,
      "num_tokens": 184068603.0,
      "step": 80280
    },
    {
      "entropy": 5.427571821212768,
      "epoch": 7.712295869356388,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010665785112614007,
      "loss": 4.5059,
      "mean_token_accuracy": 0.2803068280220032,
      "num_tokens": 184080672.0,
      "step": 80285
    },
    {
      "entropy": 5.5354907512664795,
      "epoch": 7.712776176753122,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010663510636838225,
      "loss": 4.6028,
      "mean_token_accuracy": 0.27306451052427294,
      "num_tokens": 184092524.0,
      "step": 80290
    },
    {
      "entropy": 5.576852464675904,
      "epoch": 7.713256484149856,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010661236551955733,
      "loss": 4.6459,
      "mean_token_accuracy": 0.26584055870771406,
      "num_tokens": 184104085.0,
      "step": 80295
    },
    {
      "entropy": 5.52792501449585,
      "epoch": 7.71373679154659,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010658962858019333,
      "loss": 4.5855,
      "mean_token_accuracy": 0.2655741572380066,
      "num_tokens": 184115379.0,
      "step": 80300
    },
    {
      "entropy": 5.538973236083985,
      "epoch": 7.714217098943323,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000106566895550818,
      "loss": 4.6184,
      "mean_token_accuracy": 0.2594107449054718,
      "num_tokens": 184128553.0,
      "step": 80305
    },
    {
      "entropy": 5.521293973922729,
      "epoch": 7.714697406340058,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010654416643195924,
      "loss": 4.6096,
      "mean_token_accuracy": 0.2684806615114212,
      "num_tokens": 184139557.0,
      "step": 80310
    },
    {
      "entropy": 5.498114776611328,
      "epoch": 7.715177713736791,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001065214412241447,
      "loss": 4.5846,
      "mean_token_accuracy": 0.2749483048915863,
      "num_tokens": 184149921.0,
      "step": 80315
    },
    {
      "entropy": 5.45651068687439,
      "epoch": 7.7156580211335255,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010649871992790202,
      "loss": 4.5604,
      "mean_token_accuracy": 0.2730944961309433,
      "num_tokens": 184161313.0,
      "step": 80320
    },
    {
      "entropy": 5.511893081665039,
      "epoch": 7.71613832853026,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010647600254375874,
      "loss": 4.6223,
      "mean_token_accuracy": 0.266419318318367,
      "num_tokens": 184173137.0,
      "step": 80325
    },
    {
      "entropy": 5.5532513618469235,
      "epoch": 7.716618635926993,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001064532890722423,
      "loss": 4.6726,
      "mean_token_accuracy": 0.2608289524912834,
      "num_tokens": 184185054.0,
      "step": 80330
    },
    {
      "entropy": 5.5093855381011965,
      "epoch": 7.717098943323727,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010643057951387997,
      "loss": 4.5865,
      "mean_token_accuracy": 0.2761713147163391,
      "num_tokens": 184195103.0,
      "step": 80335
    },
    {
      "entropy": 5.512271308898926,
      "epoch": 7.717579250720461,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010640787386919903,
      "loss": 4.6047,
      "mean_token_accuracy": 0.2722137585282326,
      "num_tokens": 184208061.0,
      "step": 80340
    },
    {
      "entropy": 5.515231418609619,
      "epoch": 7.718059558117195,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010638517213872665,
      "loss": 4.6386,
      "mean_token_accuracy": 0.26824546307325364,
      "num_tokens": 184219778.0,
      "step": 80345
    },
    {
      "entropy": 5.558420038223266,
      "epoch": 7.718539865513929,
      "grad_norm": 1.71875,
      "learning_rate": 0.00010636247432298984,
      "loss": 4.6041,
      "mean_token_accuracy": 0.26580062359571455,
      "num_tokens": 184230641.0,
      "step": 80350
    },
    {
      "entropy": 5.578493547439575,
      "epoch": 7.719020172910663,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010633978042251569,
      "loss": 4.7052,
      "mean_token_accuracy": 0.25697180777788164,
      "num_tokens": 184242693.0,
      "step": 80355
    },
    {
      "entropy": 5.5736712455749515,
      "epoch": 7.7195004803073966,
      "grad_norm": 1.625,
      "learning_rate": 0.00010631709043783106,
      "loss": 4.691,
      "mean_token_accuracy": 0.2618876755237579,
      "num_tokens": 184254922.0,
      "step": 80360
    },
    {
      "entropy": 5.493878412246704,
      "epoch": 7.719980787704131,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010629440436946263,
      "loss": 4.5689,
      "mean_token_accuracy": 0.2751865446567535,
      "num_tokens": 184266877.0,
      "step": 80365
    },
    {
      "entropy": 5.539554595947266,
      "epoch": 7.720461095100864,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010627172221793719,
      "loss": 4.6023,
      "mean_token_accuracy": 0.2732731521129608,
      "num_tokens": 184279681.0,
      "step": 80370
    },
    {
      "entropy": 5.54700608253479,
      "epoch": 7.720941402497599,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010624904398378132,
      "loss": 4.5994,
      "mean_token_accuracy": 0.26395758986473083,
      "num_tokens": 184290715.0,
      "step": 80375
    },
    {
      "entropy": 5.525628232955933,
      "epoch": 7.721421709894332,
      "grad_norm": 1.59375,
      "learning_rate": 0.00010622636966752158,
      "loss": 4.6326,
      "mean_token_accuracy": 0.2706016004085541,
      "num_tokens": 184301591.0,
      "step": 80380
    },
    {
      "entropy": 5.476874876022339,
      "epoch": 7.721902017291066,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010620369926968438,
      "loss": 4.6321,
      "mean_token_accuracy": 0.26852713376283643,
      "num_tokens": 184312408.0,
      "step": 80385
    },
    {
      "entropy": 5.532886552810669,
      "epoch": 7.7223823246878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010618103279079614,
      "loss": 4.6373,
      "mean_token_accuracy": 0.26345517337322233,
      "num_tokens": 184325202.0,
      "step": 80390
    },
    {
      "entropy": 5.582065391540527,
      "epoch": 7.722862632084534,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010615837023138293,
      "loss": 4.6292,
      "mean_token_accuracy": 0.26456527709960936,
      "num_tokens": 184337396.0,
      "step": 80395
    },
    {
      "entropy": 5.51645040512085,
      "epoch": 7.7233429394812685,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010613571159197103,
      "loss": 4.6036,
      "mean_token_accuracy": 0.2656456485390663,
      "num_tokens": 184348731.0,
      "step": 80400
    },
    {
      "entropy": 5.550070810317993,
      "epoch": 7.723823246878002,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010611305687308647,
      "loss": 4.6417,
      "mean_token_accuracy": 0.2658614695072174,
      "num_tokens": 184359016.0,
      "step": 80405
    },
    {
      "entropy": 5.534515285491944,
      "epoch": 7.724303554274735,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010609040607525524,
      "loss": 4.5901,
      "mean_token_accuracy": 0.2743246525526047,
      "num_tokens": 184370350.0,
      "step": 80410
    },
    {
      "entropy": 5.472051954269409,
      "epoch": 7.72478386167147,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010606775919900324,
      "loss": 4.5483,
      "mean_token_accuracy": 0.2701514706015587,
      "num_tokens": 184381077.0,
      "step": 80415
    },
    {
      "entropy": 5.571776342391968,
      "epoch": 7.725264169068204,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001060451162448563,
      "loss": 4.6665,
      "mean_token_accuracy": 0.26086620390415194,
      "num_tokens": 184392409.0,
      "step": 80420
    },
    {
      "entropy": 5.515854263305664,
      "epoch": 7.7257444764649374,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010602247721334001,
      "loss": 4.6269,
      "mean_token_accuracy": 0.2706715792417526,
      "num_tokens": 184404854.0,
      "step": 80425
    },
    {
      "entropy": 5.541769695281983,
      "epoch": 7.726224783861672,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010599984210498004,
      "loss": 4.6704,
      "mean_token_accuracy": 0.2636167660355568,
      "num_tokens": 184415704.0,
      "step": 80430
    },
    {
      "entropy": 5.4533318996429445,
      "epoch": 7.726705091258405,
      "grad_norm": 1.375,
      "learning_rate": 0.00010597721092030189,
      "loss": 4.5067,
      "mean_token_accuracy": 0.28096548169851304,
      "num_tokens": 184428065.0,
      "step": 80435
    },
    {
      "entropy": 5.474973440170288,
      "epoch": 7.7271853986551395,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010595458365983102,
      "loss": 4.5676,
      "mean_token_accuracy": 0.27155022919178007,
      "num_tokens": 184440173.0,
      "step": 80440
    },
    {
      "entropy": 5.59303765296936,
      "epoch": 7.727665706051873,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010593196032409276,
      "loss": 4.6454,
      "mean_token_accuracy": 0.25988302677869796,
      "num_tokens": 184450128.0,
      "step": 80445
    },
    {
      "entropy": 5.577547550201416,
      "epoch": 7.728146013448607,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001059093409136124,
      "loss": 4.7175,
      "mean_token_accuracy": 0.25962127596139906,
      "num_tokens": 184460753.0,
      "step": 80450
    },
    {
      "entropy": 5.599573945999145,
      "epoch": 7.728626320845341,
      "grad_norm": 1.359375,
      "learning_rate": 0.000105886725428915,
      "loss": 4.7026,
      "mean_token_accuracy": 0.2573524698615074,
      "num_tokens": 184473651.0,
      "step": 80455
    },
    {
      "entropy": 5.478924942016602,
      "epoch": 7.729106628242075,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010586411387052567,
      "loss": 4.5411,
      "mean_token_accuracy": 0.2789858877658844,
      "num_tokens": 184484716.0,
      "step": 80460
    },
    {
      "entropy": 5.488550806045533,
      "epoch": 7.7295869356388085,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010584150623896937,
      "loss": 4.5095,
      "mean_token_accuracy": 0.27229908406734465,
      "num_tokens": 184495902.0,
      "step": 80465
    },
    {
      "entropy": 5.52181715965271,
      "epoch": 7.730067243035543,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010581890253477102,
      "loss": 4.5947,
      "mean_token_accuracy": 0.26797316670417787,
      "num_tokens": 184507946.0,
      "step": 80470
    },
    {
      "entropy": 5.543420267105103,
      "epoch": 7.730547550432277,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010579630275845539,
      "loss": 4.7428,
      "mean_token_accuracy": 0.2554979547858238,
      "num_tokens": 184520471.0,
      "step": 80475
    },
    {
      "entropy": 5.53973069190979,
      "epoch": 7.731027857829011,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010577370691054722,
      "loss": 4.648,
      "mean_token_accuracy": 0.2607017010450363,
      "num_tokens": 184531648.0,
      "step": 80480
    },
    {
      "entropy": 5.526231002807617,
      "epoch": 7.731508165225744,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010575111499157103,
      "loss": 4.6105,
      "mean_token_accuracy": 0.26959048211574554,
      "num_tokens": 184542667.0,
      "step": 80485
    },
    {
      "entropy": 5.5065545558929445,
      "epoch": 7.731988472622478,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010572852700205137,
      "loss": 4.5496,
      "mean_token_accuracy": 0.2701434686779976,
      "num_tokens": 184553741.0,
      "step": 80490
    },
    {
      "entropy": 5.580609321594238,
      "epoch": 7.732468780019213,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010570594294251268,
      "loss": 4.6499,
      "mean_token_accuracy": 0.26161194294691087,
      "num_tokens": 184565531.0,
      "step": 80495
    },
    {
      "entropy": 5.5903538227081295,
      "epoch": 7.732949087415946,
      "grad_norm": 1.625,
      "learning_rate": 0.00010568336281347929,
      "loss": 4.7161,
      "mean_token_accuracy": 0.2561507061123848,
      "num_tokens": 184576544.0,
      "step": 80500
    },
    {
      "entropy": 5.517499399185181,
      "epoch": 7.73342939481268,
      "grad_norm": 1.375,
      "learning_rate": 0.00010566078661547548,
      "loss": 4.576,
      "mean_token_accuracy": 0.27692378908395765,
      "num_tokens": 184588419.0,
      "step": 80505
    },
    {
      "entropy": 5.559843635559082,
      "epoch": 7.733909702209414,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010563821434902543,
      "loss": 4.6987,
      "mean_token_accuracy": 0.260859040915966,
      "num_tokens": 184600403.0,
      "step": 80510
    },
    {
      "entropy": 5.550307941436768,
      "epoch": 7.734390009606148,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010561564601465304,
      "loss": 4.6056,
      "mean_token_accuracy": 0.27367421984672546,
      "num_tokens": 184612002.0,
      "step": 80515
    },
    {
      "entropy": 5.571071815490723,
      "epoch": 7.734870317002882,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010559308161288242,
      "loss": 4.6561,
      "mean_token_accuracy": 0.2714370369911194,
      "num_tokens": 184622881.0,
      "step": 80520
    },
    {
      "entropy": 5.563169193267822,
      "epoch": 7.735350624399616,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010557052114423739,
      "loss": 4.6598,
      "mean_token_accuracy": 0.2665164351463318,
      "num_tokens": 184632939.0,
      "step": 80525
    },
    {
      "entropy": 5.448678874969483,
      "epoch": 7.735830931796349,
      "grad_norm": 1.375,
      "learning_rate": 0.00010554796460924176,
      "loss": 4.5577,
      "mean_token_accuracy": 0.27256465554237364,
      "num_tokens": 184643798.0,
      "step": 80530
    },
    {
      "entropy": 5.550955963134766,
      "epoch": 7.736311239193084,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010552541200841922,
      "loss": 4.6654,
      "mean_token_accuracy": 0.2652120068669319,
      "num_tokens": 184656770.0,
      "step": 80535
    },
    {
      "entropy": 5.43323130607605,
      "epoch": 7.736791546589817,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010550286334229344,
      "loss": 4.5099,
      "mean_token_accuracy": 0.28017172068357465,
      "num_tokens": 184669487.0,
      "step": 80540
    },
    {
      "entropy": 5.517778778076172,
      "epoch": 7.7372718539865515,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010548031861138782,
      "loss": 4.5903,
      "mean_token_accuracy": 0.26812210977077483,
      "num_tokens": 184681764.0,
      "step": 80545
    },
    {
      "entropy": 5.511574554443359,
      "epoch": 7.737752161383285,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010545777781622585,
      "loss": 4.6229,
      "mean_token_accuracy": 0.26027249693870547,
      "num_tokens": 184694355.0,
      "step": 80550
    },
    {
      "entropy": 5.507906723022461,
      "epoch": 7.738232468780019,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001054352409573308,
      "loss": 4.5582,
      "mean_token_accuracy": 0.27022181153297425,
      "num_tokens": 184705484.0,
      "step": 80555
    },
    {
      "entropy": 5.49801344871521,
      "epoch": 7.738712776176753,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010541270803522596,
      "loss": 4.5381,
      "mean_token_accuracy": 0.2743335887789726,
      "num_tokens": 184715962.0,
      "step": 80560
    },
    {
      "entropy": 5.491722106933594,
      "epoch": 7.739193083573487,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010539017905043446,
      "loss": 4.6143,
      "mean_token_accuracy": 0.27256215512752535,
      "num_tokens": 184727593.0,
      "step": 80565
    },
    {
      "entropy": 5.486928033828735,
      "epoch": 7.739673390970221,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010536765400347947,
      "loss": 4.5628,
      "mean_token_accuracy": 0.2702593594789505,
      "num_tokens": 184739572.0,
      "step": 80570
    },
    {
      "entropy": 5.551372957229614,
      "epoch": 7.740153698366955,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010534513289488375,
      "loss": 4.6573,
      "mean_token_accuracy": 0.2619842976331711,
      "num_tokens": 184752581.0,
      "step": 80575
    },
    {
      "entropy": 5.429301500320435,
      "epoch": 7.740634005763689,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010532261572517026,
      "loss": 4.4623,
      "mean_token_accuracy": 0.28289371728897095,
      "num_tokens": 184763652.0,
      "step": 80580
    },
    {
      "entropy": 5.5709593296051025,
      "epoch": 7.7411143131604225,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010530010249486183,
      "loss": 4.6495,
      "mean_token_accuracy": 0.2630657210946083,
      "num_tokens": 184773574.0,
      "step": 80585
    },
    {
      "entropy": 5.550573873519897,
      "epoch": 7.741594620557157,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001052775932044811,
      "loss": 4.6375,
      "mean_token_accuracy": 0.26653235256671903,
      "num_tokens": 184784436.0,
      "step": 80590
    },
    {
      "entropy": 5.529950475692749,
      "epoch": 7.74207492795389,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001052550878545507,
      "loss": 4.6122,
      "mean_token_accuracy": 0.2655377432703972,
      "num_tokens": 184795893.0,
      "step": 80595
    },
    {
      "entropy": 5.559251403808593,
      "epoch": 7.742555235350625,
      "grad_norm": 1.375,
      "learning_rate": 0.00010523258644559316,
      "loss": 4.6135,
      "mean_token_accuracy": 0.2650794520974159,
      "num_tokens": 184807045.0,
      "step": 80600
    },
    {
      "entropy": 5.509832286834717,
      "epoch": 7.743035542747358,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001052100889781308,
      "loss": 4.6374,
      "mean_token_accuracy": 0.2763575464487076,
      "num_tokens": 184819192.0,
      "step": 80605
    },
    {
      "entropy": 5.5600221157073975,
      "epoch": 7.743515850144092,
      "grad_norm": 1.640625,
      "learning_rate": 0.00010518759545268603,
      "loss": 4.6063,
      "mean_token_accuracy": 0.2731913775205612,
      "num_tokens": 184830268.0,
      "step": 80610
    },
    {
      "entropy": 5.534974575042725,
      "epoch": 7.743996157540826,
      "grad_norm": 1.65625,
      "learning_rate": 0.00010516510586978101,
      "loss": 4.5772,
      "mean_token_accuracy": 0.2730641454458237,
      "num_tokens": 184840489.0,
      "step": 80615
    },
    {
      "entropy": 5.53709864616394,
      "epoch": 7.74447646493756,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010514262022993796,
      "loss": 4.5895,
      "mean_token_accuracy": 0.262706932425499,
      "num_tokens": 184852255.0,
      "step": 80620
    },
    {
      "entropy": 5.538714122772217,
      "epoch": 7.744956772334294,
      "grad_norm": 1.5,
      "learning_rate": 0.00010512013853367888,
      "loss": 4.6985,
      "mean_token_accuracy": 0.2580198675394058,
      "num_tokens": 184863752.0,
      "step": 80625
    },
    {
      "entropy": 5.511666584014892,
      "epoch": 7.745437079731028,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001050976607815258,
      "loss": 4.6039,
      "mean_token_accuracy": 0.271416375041008,
      "num_tokens": 184874669.0,
      "step": 80630
    },
    {
      "entropy": 5.614974451065064,
      "epoch": 7.745917387127761,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010507518697400046,
      "loss": 4.714,
      "mean_token_accuracy": 0.2577840566635132,
      "num_tokens": 184886664.0,
      "step": 80635
    },
    {
      "entropy": 5.608202219009399,
      "epoch": 7.746397694524496,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010505271711162473,
      "loss": 4.6974,
      "mean_token_accuracy": 0.2620038568973541,
      "num_tokens": 184897699.0,
      "step": 80640
    },
    {
      "entropy": 5.500785875320434,
      "epoch": 7.74687800192123,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010503025119492024,
      "loss": 4.5066,
      "mean_token_accuracy": 0.27897932678461074,
      "num_tokens": 184909274.0,
      "step": 80645
    },
    {
      "entropy": 5.544729900360108,
      "epoch": 7.747358309317963,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010500778922440865,
      "loss": 4.6623,
      "mean_token_accuracy": 0.2640905737876892,
      "num_tokens": 184920310.0,
      "step": 80650
    },
    {
      "entropy": 5.570060062408447,
      "epoch": 7.747838616714698,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010498533120061141,
      "loss": 4.693,
      "mean_token_accuracy": 0.2635408863425255,
      "num_tokens": 184931846.0,
      "step": 80655
    },
    {
      "entropy": 5.5683441162109375,
      "epoch": 7.748318924111431,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010496287712404998,
      "loss": 4.6775,
      "mean_token_accuracy": 0.26399578899145126,
      "num_tokens": 184943393.0,
      "step": 80660
    },
    {
      "entropy": 5.491556167602539,
      "epoch": 7.7487992315081655,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010494042699524561,
      "loss": 4.5426,
      "mean_token_accuracy": 0.2762404039502144,
      "num_tokens": 184953632.0,
      "step": 80665
    },
    {
      "entropy": 5.549554300308228,
      "epoch": 7.749279538904899,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010491798081471951,
      "loss": 4.595,
      "mean_token_accuracy": 0.2656602054834366,
      "num_tokens": 184965051.0,
      "step": 80670
    },
    {
      "entropy": 5.529583263397217,
      "epoch": 7.749759846301633,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010489553858299288,
      "loss": 4.6227,
      "mean_token_accuracy": 0.2708225652575493,
      "num_tokens": 184977238.0,
      "step": 80675
    },
    {
      "entropy": 5.511842441558838,
      "epoch": 7.750240153698367,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010487310030058675,
      "loss": 4.5745,
      "mean_token_accuracy": 0.26942268162965777,
      "num_tokens": 184989256.0,
      "step": 80680
    },
    {
      "entropy": 5.600909948348999,
      "epoch": 7.750720461095101,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010485066596802204,
      "loss": 4.7611,
      "mean_token_accuracy": 0.2496345743536949,
      "num_tokens": 185001259.0,
      "step": 80685
    },
    {
      "entropy": 5.5536994457244875,
      "epoch": 7.7512007684918345,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001048282355858197,
      "loss": 4.6387,
      "mean_token_accuracy": 0.2684241980314255,
      "num_tokens": 185012126.0,
      "step": 80690
    },
    {
      "entropy": 5.529705667495728,
      "epoch": 7.751681075888569,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010480580915450033,
      "loss": 4.6174,
      "mean_token_accuracy": 0.26362231373786926,
      "num_tokens": 185023949.0,
      "step": 80695
    },
    {
      "entropy": 5.5437846183776855,
      "epoch": 7.752161383285302,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010478338667458472,
      "loss": 4.6173,
      "mean_token_accuracy": 0.26634140461683276,
      "num_tokens": 185036766.0,
      "step": 80700
    },
    {
      "entropy": 5.566598224639892,
      "epoch": 7.752641690682037,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010476096814659343,
      "loss": 4.6401,
      "mean_token_accuracy": 0.26023897975683213,
      "num_tokens": 185050328.0,
      "step": 80705
    },
    {
      "entropy": 5.575076675415039,
      "epoch": 7.75312199807877,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010473855357104695,
      "loss": 4.6804,
      "mean_token_accuracy": 0.26134220510721207,
      "num_tokens": 185060837.0,
      "step": 80710
    },
    {
      "entropy": 5.4633313655853275,
      "epoch": 7.753602305475504,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010471614294846568,
      "loss": 4.5665,
      "mean_token_accuracy": 0.2725845232605934,
      "num_tokens": 185073086.0,
      "step": 80715
    },
    {
      "entropy": 5.5495884895324705,
      "epoch": 7.754082612872239,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010469373627937001,
      "loss": 4.6234,
      "mean_token_accuracy": 0.27514071464538575,
      "num_tokens": 185084478.0,
      "step": 80720
    },
    {
      "entropy": 5.537157392501831,
      "epoch": 7.754562920268972,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010467133356427999,
      "loss": 4.6436,
      "mean_token_accuracy": 0.26512407064437865,
      "num_tokens": 185095561.0,
      "step": 80725
    },
    {
      "entropy": 5.5898542404174805,
      "epoch": 7.755043227665706,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010464893480371587,
      "loss": 4.6608,
      "mean_token_accuracy": 0.26887340694665907,
      "num_tokens": 185106442.0,
      "step": 80730
    },
    {
      "entropy": 5.557669067382813,
      "epoch": 7.75552353506244,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010462653999819762,
      "loss": 4.6783,
      "mean_token_accuracy": 0.2548289015889168,
      "num_tokens": 185118356.0,
      "step": 80735
    },
    {
      "entropy": 5.53000807762146,
      "epoch": 7.756003842459174,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00010460414914824522,
      "loss": 4.5927,
      "mean_token_accuracy": 0.2698141872882843,
      "num_tokens": 185130606.0,
      "step": 80740
    },
    {
      "entropy": 5.481638383865357,
      "epoch": 7.756484149855908,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010458176225437852,
      "loss": 4.5415,
      "mean_token_accuracy": 0.283090078830719,
      "num_tokens": 185142000.0,
      "step": 80745
    },
    {
      "entropy": 5.537296104431152,
      "epoch": 7.756964457252642,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001045593793171173,
      "loss": 4.6332,
      "mean_token_accuracy": 0.26221868991851804,
      "num_tokens": 185152951.0,
      "step": 80750
    },
    {
      "entropy": 5.578528261184692,
      "epoch": 7.757444764649375,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00010453700033698111,
      "loss": 4.6638,
      "mean_token_accuracy": 0.2603528931736946,
      "num_tokens": 185163193.0,
      "step": 80755
    },
    {
      "entropy": 5.571477746963501,
      "epoch": 7.75792507204611,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001045146253144896,
      "loss": 4.6648,
      "mean_token_accuracy": 0.26049634963274004,
      "num_tokens": 185174029.0,
      "step": 80760
    },
    {
      "entropy": 5.545762062072754,
      "epoch": 7.758405379442843,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001044922542501623,
      "loss": 4.6624,
      "mean_token_accuracy": 0.2696756273508072,
      "num_tokens": 185184486.0,
      "step": 80765
    },
    {
      "entropy": 5.5004462718963625,
      "epoch": 7.7588856868395775,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001044698871445185,
      "loss": 4.5294,
      "mean_token_accuracy": 0.27353364378213885,
      "num_tokens": 185196131.0,
      "step": 80770
    },
    {
      "entropy": 5.537823915481567,
      "epoch": 7.759365994236311,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010444752399807758,
      "loss": 4.6565,
      "mean_token_accuracy": 0.264360174536705,
      "num_tokens": 185207452.0,
      "step": 80775
    },
    {
      "entropy": 5.534722137451172,
      "epoch": 7.759846301633045,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001044251648113587,
      "loss": 4.6219,
      "mean_token_accuracy": 0.2630938977003098,
      "num_tokens": 185219865.0,
      "step": 80780
    },
    {
      "entropy": 5.5177911758422855,
      "epoch": 7.760326609029779,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000104402809584881,
      "loss": 4.5682,
      "mean_token_accuracy": 0.2715673357248306,
      "num_tokens": 185231206.0,
      "step": 80785
    },
    {
      "entropy": 5.506876039505005,
      "epoch": 7.760806916426513,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010438045831916357,
      "loss": 4.5767,
      "mean_token_accuracy": 0.27153135985136034,
      "num_tokens": 185242623.0,
      "step": 80790
    },
    {
      "entropy": 5.5665611743927,
      "epoch": 7.761287223823247,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010435811101472517,
      "loss": 4.6483,
      "mean_token_accuracy": 0.2587078377604485,
      "num_tokens": 185253439.0,
      "step": 80795
    },
    {
      "entropy": 5.461405277252197,
      "epoch": 7.761767531219981,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010433576767208475,
      "loss": 4.555,
      "mean_token_accuracy": 0.27362772822380066,
      "num_tokens": 185264714.0,
      "step": 80800
    },
    {
      "entropy": 5.486602163314819,
      "epoch": 7.762247838616715,
      "grad_norm": 1.5,
      "learning_rate": 0.00010431342829176105,
      "loss": 4.5275,
      "mean_token_accuracy": 0.27018555998802185,
      "num_tokens": 185276813.0,
      "step": 80805
    },
    {
      "entropy": 5.477140474319458,
      "epoch": 7.7627281460134485,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010429109287427265,
      "loss": 4.555,
      "mean_token_accuracy": 0.276094526052475,
      "num_tokens": 185288525.0,
      "step": 80810
    },
    {
      "entropy": 5.455356311798096,
      "epoch": 7.763208453410183,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010426876142013822,
      "loss": 4.5345,
      "mean_token_accuracy": 0.2711879849433899,
      "num_tokens": 185299370.0,
      "step": 80815
    },
    {
      "entropy": 5.45035605430603,
      "epoch": 7.763688760806916,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010424643392987618,
      "loss": 4.5579,
      "mean_token_accuracy": 0.27361485809087754,
      "num_tokens": 185310603.0,
      "step": 80820
    },
    {
      "entropy": 5.553579092025757,
      "epoch": 7.764169068203651,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001042241104040049,
      "loss": 4.6191,
      "mean_token_accuracy": 0.26722281575202944,
      "num_tokens": 185321144.0,
      "step": 80825
    },
    {
      "entropy": 5.521544981002807,
      "epoch": 7.764649375600384,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010420179084304275,
      "loss": 4.5482,
      "mean_token_accuracy": 0.2770368427038193,
      "num_tokens": 185332912.0,
      "step": 80830
    },
    {
      "entropy": 5.587537431716919,
      "epoch": 7.765129682997118,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010417947524750778,
      "loss": 4.636,
      "mean_token_accuracy": 0.2654685601592064,
      "num_tokens": 185344614.0,
      "step": 80835
    },
    {
      "entropy": 5.573302841186523,
      "epoch": 7.765609990393852,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010415716361791815,
      "loss": 4.6684,
      "mean_token_accuracy": 0.26292489618062975,
      "num_tokens": 185355566.0,
      "step": 80840
    },
    {
      "entropy": 5.540123510360718,
      "epoch": 7.766090297790586,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010413485595479191,
      "loss": 4.6031,
      "mean_token_accuracy": 0.27245512455701826,
      "num_tokens": 185367595.0,
      "step": 80845
    },
    {
      "entropy": 5.594246530532837,
      "epoch": 7.76657060518732,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010411255225864693,
      "loss": 4.6848,
      "mean_token_accuracy": 0.26474253833293915,
      "num_tokens": 185378978.0,
      "step": 80850
    },
    {
      "entropy": 5.578521299362182,
      "epoch": 7.767050912584054,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00010409025253000105,
      "loss": 4.6533,
      "mean_token_accuracy": 0.2668703243136406,
      "num_tokens": 185390359.0,
      "step": 80855
    },
    {
      "entropy": 5.6552716255187985,
      "epoch": 7.767531219980787,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010406795676937203,
      "loss": 4.7384,
      "mean_token_accuracy": 0.2515922635793686,
      "num_tokens": 185401814.0,
      "step": 80860
    },
    {
      "entropy": 5.571403837203979,
      "epoch": 7.768011527377522,
      "grad_norm": 1.5,
      "learning_rate": 0.00010404566497727746,
      "loss": 4.6699,
      "mean_token_accuracy": 0.2603390470147133,
      "num_tokens": 185414425.0,
      "step": 80865
    },
    {
      "entropy": 5.588761711120606,
      "epoch": 7.768491834774256,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010402337715423498,
      "loss": 4.6761,
      "mean_token_accuracy": 0.2594642907381058,
      "num_tokens": 185425365.0,
      "step": 80870
    },
    {
      "entropy": 5.5115619659423825,
      "epoch": 7.768972142170989,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010400109330076198,
      "loss": 4.6252,
      "mean_token_accuracy": 0.2620170548558235,
      "num_tokens": 185436687.0,
      "step": 80875
    },
    {
      "entropy": 5.493004703521729,
      "epoch": 7.769452449567723,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010397881341737579,
      "loss": 4.5514,
      "mean_token_accuracy": 0.2683560878038406,
      "num_tokens": 185448036.0,
      "step": 80880
    },
    {
      "entropy": 5.591364955902099,
      "epoch": 7.769932756964457,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00010395653750459372,
      "loss": 4.6573,
      "mean_token_accuracy": 0.2702783107757568,
      "num_tokens": 185459068.0,
      "step": 80885
    },
    {
      "entropy": 5.447332429885864,
      "epoch": 7.7704130643611915,
      "grad_norm": 1.5,
      "learning_rate": 0.00010393426556293297,
      "loss": 4.5152,
      "mean_token_accuracy": 0.27314233034849167,
      "num_tokens": 185470578.0,
      "step": 80890
    },
    {
      "entropy": 5.533141469955444,
      "epoch": 7.770893371757925,
      "grad_norm": 1.375,
      "learning_rate": 0.00010391199759291058,
      "loss": 4.655,
      "mean_token_accuracy": 0.26605349630117414,
      "num_tokens": 185482490.0,
      "step": 80895
    },
    {
      "entropy": 5.465753126144409,
      "epoch": 7.771373679154659,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001038897335950436,
      "loss": 4.5284,
      "mean_token_accuracy": 0.27766057252883913,
      "num_tokens": 185494199.0,
      "step": 80900
    },
    {
      "entropy": 5.648331499099731,
      "epoch": 7.771853986551393,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010386747356984899,
      "loss": 4.7224,
      "mean_token_accuracy": 0.2596989780664444,
      "num_tokens": 185506585.0,
      "step": 80905
    },
    {
      "entropy": 5.491866874694824,
      "epoch": 7.772334293948127,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010384521751784342,
      "loss": 4.5521,
      "mean_token_accuracy": 0.2688756436109543,
      "num_tokens": 185518112.0,
      "step": 80910
    },
    {
      "entropy": 5.4871033191680905,
      "epoch": 7.7728146013448605,
      "grad_norm": 1.25,
      "learning_rate": 0.00010382296543954367,
      "loss": 4.5863,
      "mean_token_accuracy": 0.2712739497423172,
      "num_tokens": 185530679.0,
      "step": 80915
    },
    {
      "entropy": 5.454814720153808,
      "epoch": 7.773294908741595,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010380071733546637,
      "loss": 4.5299,
      "mean_token_accuracy": 0.27564767152071,
      "num_tokens": 185541870.0,
      "step": 80920
    },
    {
      "entropy": 5.496362781524658,
      "epoch": 7.773775216138328,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010377847320612806,
      "loss": 4.5248,
      "mean_token_accuracy": 0.27468681037425996,
      "num_tokens": 185552788.0,
      "step": 80925
    },
    {
      "entropy": 5.4036823272705075,
      "epoch": 7.7742555235350626,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010375623305204516,
      "loss": 4.4674,
      "mean_token_accuracy": 0.27897894084453584,
      "num_tokens": 185564382.0,
      "step": 80930
    },
    {
      "entropy": 5.515308570861817,
      "epoch": 7.774735830931796,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001037339968737341,
      "loss": 4.5089,
      "mean_token_accuracy": 0.280517989397049,
      "num_tokens": 185574770.0,
      "step": 80935
    },
    {
      "entropy": 5.568727111816406,
      "epoch": 7.77521613832853,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010371176467171104,
      "loss": 4.6538,
      "mean_token_accuracy": 0.2635163366794586,
      "num_tokens": 185586307.0,
      "step": 80940
    },
    {
      "entropy": 5.548581838607788,
      "epoch": 7.775696445725265,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010368953644649218,
      "loss": 4.625,
      "mean_token_accuracy": 0.26866733133792875,
      "num_tokens": 185597065.0,
      "step": 80945
    },
    {
      "entropy": 5.5203429698944095,
      "epoch": 7.776176753121998,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010366731219859358,
      "loss": 4.583,
      "mean_token_accuracy": 0.2686708584427834,
      "num_tokens": 185607169.0,
      "step": 80950
    },
    {
      "entropy": 5.438231897354126,
      "epoch": 7.7766570605187315,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010364509192853125,
      "loss": 4.5428,
      "mean_token_accuracy": 0.27546869218349457,
      "num_tokens": 185618278.0,
      "step": 80955
    },
    {
      "entropy": 5.493487215042114,
      "epoch": 7.777137367915466,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010362287563682105,
      "loss": 4.6291,
      "mean_token_accuracy": 0.2675952836871147,
      "num_tokens": 185629906.0,
      "step": 80960
    },
    {
      "entropy": 5.529599380493164,
      "epoch": 7.7776176753122,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00010360066332397889,
      "loss": 4.6697,
      "mean_token_accuracy": 0.2671845808625221,
      "num_tokens": 185642120.0,
      "step": 80965
    },
    {
      "entropy": 5.538376998901367,
      "epoch": 7.778097982708934,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010357845499052027,
      "loss": 4.692,
      "mean_token_accuracy": 0.26176566183567046,
      "num_tokens": 185654682.0,
      "step": 80970
    },
    {
      "entropy": 5.466158246994018,
      "epoch": 7.778578290105668,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010355625063696092,
      "loss": 4.5545,
      "mean_token_accuracy": 0.26876587718725203,
      "num_tokens": 185666500.0,
      "step": 80975
    },
    {
      "entropy": 5.533692312240601,
      "epoch": 7.779058597502401,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010353405026381634,
      "loss": 4.5758,
      "mean_token_accuracy": 0.2724940672516823,
      "num_tokens": 185677519.0,
      "step": 80980
    },
    {
      "entropy": 5.547539281845093,
      "epoch": 7.779538904899136,
      "grad_norm": 1.296875,
      "learning_rate": 0.000103511853871602,
      "loss": 4.6507,
      "mean_token_accuracy": 0.265374581515789,
      "num_tokens": 185689441.0,
      "step": 80985
    },
    {
      "entropy": 5.531675624847412,
      "epoch": 7.780019212295869,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00010348966146083317,
      "loss": 4.6024,
      "mean_token_accuracy": 0.2733334258198738,
      "num_tokens": 185701212.0,
      "step": 80990
    },
    {
      "entropy": 5.521816873550415,
      "epoch": 7.7804995196926034,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010346747303202519,
      "loss": 4.5484,
      "mean_token_accuracy": 0.2766282185912132,
      "num_tokens": 185711566.0,
      "step": 80995
    },
    {
      "entropy": 5.517161703109741,
      "epoch": 7.780979827089337,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010344528858569304,
      "loss": 4.5858,
      "mean_token_accuracy": 0.275600066781044,
      "num_tokens": 185722582.0,
      "step": 81000
    },
    {
      "epoch": 7.780979827089337,
      "eval_entropy": 5.219889696605725,
      "eval_loss": 4.738593101501465,
      "eval_mean_token_accuracy": 0.2686447554709765,
      "eval_num_tokens": 185722582.0,
      "eval_runtime": 20.7133,
      "eval_samples_per_second": 1584.245,
      "eval_steps_per_second": 198.037,
      "step": 81000
    },
    {
      "entropy": 5.484673976898193,
      "epoch": 7.781460134486071,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010342310812235195,
      "loss": 4.6306,
      "mean_token_accuracy": 0.2723671913146973,
      "num_tokens": 185735254.0,
      "step": 81005
    },
    {
      "entropy": 5.51634635925293,
      "epoch": 7.781940441882805,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010340093164251675,
      "loss": 4.6221,
      "mean_token_accuracy": 0.2629276692867279,
      "num_tokens": 185746380.0,
      "step": 81010
    },
    {
      "entropy": 5.5471619129180905,
      "epoch": 7.782420749279539,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010337875914670241,
      "loss": 4.5964,
      "mean_token_accuracy": 0.27270431220531466,
      "num_tokens": 185757274.0,
      "step": 81015
    },
    {
      "entropy": 5.53946590423584,
      "epoch": 7.782901056676273,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010335659063542363,
      "loss": 4.6574,
      "mean_token_accuracy": 0.26073697209358215,
      "num_tokens": 185768837.0,
      "step": 81020
    },
    {
      "entropy": 5.5838963985443115,
      "epoch": 7.783381364073007,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010333442610919525,
      "loss": 4.6785,
      "mean_token_accuracy": 0.2695322334766388,
      "num_tokens": 185778424.0,
      "step": 81025
    },
    {
      "entropy": 5.598317384719849,
      "epoch": 7.78386167146974,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010331226556853166,
      "loss": 4.7267,
      "mean_token_accuracy": 0.2583281949162483,
      "num_tokens": 185788790.0,
      "step": 81030
    },
    {
      "entropy": 5.506079578399659,
      "epoch": 7.7843419788664745,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010329010901394748,
      "loss": 4.6008,
      "mean_token_accuracy": 0.26304649859666823,
      "num_tokens": 185800389.0,
      "step": 81035
    },
    {
      "entropy": 5.554199504852295,
      "epoch": 7.784822286263209,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010326795644595706,
      "loss": 4.6219,
      "mean_token_accuracy": 0.266652999818325,
      "num_tokens": 185811501.0,
      "step": 81040
    },
    {
      "entropy": 5.599745655059815,
      "epoch": 7.785302593659942,
      "grad_norm": 1.609375,
      "learning_rate": 0.00010324580786507475,
      "loss": 4.6587,
      "mean_token_accuracy": 0.2643289566040039,
      "num_tokens": 185822275.0,
      "step": 81045
    },
    {
      "entropy": 5.5374932289123535,
      "epoch": 7.785782901056677,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010322366327181482,
      "loss": 4.6094,
      "mean_token_accuracy": 0.26496992260217667,
      "num_tokens": 185832621.0,
      "step": 81050
    },
    {
      "entropy": 5.505383443832398,
      "epoch": 7.78626320845341,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00010320152266669138,
      "loss": 4.5635,
      "mean_token_accuracy": 0.27356603890657427,
      "num_tokens": 185842912.0,
      "step": 81055
    },
    {
      "entropy": 5.486770105361939,
      "epoch": 7.786743515850144,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010317938605021841,
      "loss": 4.5694,
      "mean_token_accuracy": 0.2733764052391052,
      "num_tokens": 185854155.0,
      "step": 81060
    },
    {
      "entropy": 5.528371095657349,
      "epoch": 7.787223823246878,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010315725342290988,
      "loss": 4.5801,
      "mean_token_accuracy": 0.2658785551786423,
      "num_tokens": 185866178.0,
      "step": 81065
    },
    {
      "entropy": 5.538595962524414,
      "epoch": 7.787704130643612,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010313512478527966,
      "loss": 4.63,
      "mean_token_accuracy": 0.2675804287195206,
      "num_tokens": 185878449.0,
      "step": 81070
    },
    {
      "entropy": 5.522545289993286,
      "epoch": 7.7881844380403455,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010311300013784153,
      "loss": 4.6867,
      "mean_token_accuracy": 0.25986241549253464,
      "num_tokens": 185891502.0,
      "step": 81075
    },
    {
      "entropy": 5.492435121536255,
      "epoch": 7.78866474543708,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010309087948110912,
      "loss": 4.5491,
      "mean_token_accuracy": 0.272907130420208,
      "num_tokens": 185903861.0,
      "step": 81080
    },
    {
      "entropy": 5.56805419921875,
      "epoch": 7.789145052833813,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010306876281559607,
      "loss": 4.6408,
      "mean_token_accuracy": 0.2651448264718056,
      "num_tokens": 185914976.0,
      "step": 81085
    },
    {
      "entropy": 5.478685474395752,
      "epoch": 7.789625360230548,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010304665014181578,
      "loss": 4.5202,
      "mean_token_accuracy": 0.27726341784000397,
      "num_tokens": 185926142.0,
      "step": 81090
    },
    {
      "entropy": 5.488747406005859,
      "epoch": 7.790105667627281,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010302454146028168,
      "loss": 4.5588,
      "mean_token_accuracy": 0.27819547057151794,
      "num_tokens": 185937122.0,
      "step": 81095
    },
    {
      "entropy": 5.563781595230102,
      "epoch": 7.790585975024015,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010300243677150705,
      "loss": 4.6486,
      "mean_token_accuracy": 0.26939554810523986,
      "num_tokens": 185948199.0,
      "step": 81100
    },
    {
      "entropy": 5.575008964538574,
      "epoch": 7.791066282420749,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001029803360760051,
      "loss": 4.6145,
      "mean_token_accuracy": 0.27139536291360855,
      "num_tokens": 185958441.0,
      "step": 81105
    },
    {
      "entropy": 5.538266038894653,
      "epoch": 7.791546589817483,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010295823937428897,
      "loss": 4.6311,
      "mean_token_accuracy": 0.2632451668381691,
      "num_tokens": 185969839.0,
      "step": 81110
    },
    {
      "entropy": 5.4598658084869385,
      "epoch": 7.7920268972142175,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010293614666687173,
      "loss": 4.5712,
      "mean_token_accuracy": 0.2721046954393387,
      "num_tokens": 185981504.0,
      "step": 81115
    },
    {
      "entropy": 5.581509160995483,
      "epoch": 7.792507204610951,
      "grad_norm": 1.375,
      "learning_rate": 0.00010291405795426614,
      "loss": 4.7298,
      "mean_token_accuracy": 0.2543839141726494,
      "num_tokens": 185993694.0,
      "step": 81120
    },
    {
      "entropy": 5.536433267593384,
      "epoch": 7.792987512007685,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010289197323698518,
      "loss": 4.6349,
      "mean_token_accuracy": 0.26849460452795026,
      "num_tokens": 186005695.0,
      "step": 81125
    },
    {
      "entropy": 5.581405544281006,
      "epoch": 7.793467819404419,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010286989251554151,
      "loss": 4.7043,
      "mean_token_accuracy": 0.2614233687520027,
      "num_tokens": 186016614.0,
      "step": 81130
    },
    {
      "entropy": 5.4356893539428714,
      "epoch": 7.793948126801153,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010284781579044784,
      "loss": 4.5325,
      "mean_token_accuracy": 0.2781180769205093,
      "num_tokens": 186029282.0,
      "step": 81135
    },
    {
      "entropy": 5.556849479675293,
      "epoch": 7.7944284341978864,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010282574306221668,
      "loss": 4.639,
      "mean_token_accuracy": 0.2679125964641571,
      "num_tokens": 186040438.0,
      "step": 81140
    },
    {
      "entropy": 5.589427185058594,
      "epoch": 7.794908741594621,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001028036743313606,
      "loss": 4.6294,
      "mean_token_accuracy": 0.2695938363671303,
      "num_tokens": 186050220.0,
      "step": 81145
    },
    {
      "entropy": 5.479658031463623,
      "epoch": 7.795389048991354,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001027816095983918,
      "loss": 4.5493,
      "mean_token_accuracy": 0.2697453945875168,
      "num_tokens": 186061477.0,
      "step": 81150
    },
    {
      "entropy": 5.508383464813233,
      "epoch": 7.7958693563880885,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010275954886382266,
      "loss": 4.5681,
      "mean_token_accuracy": 0.2722103863954544,
      "num_tokens": 186073478.0,
      "step": 81155
    },
    {
      "entropy": 5.530997276306152,
      "epoch": 7.796349663784822,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010273749212816534,
      "loss": 4.5937,
      "mean_token_accuracy": 0.26746189296245576,
      "num_tokens": 186083968.0,
      "step": 81160
    },
    {
      "entropy": 5.48038067817688,
      "epoch": 7.796829971181556,
      "grad_norm": 1.375,
      "learning_rate": 0.00010271543939193196,
      "loss": 4.6173,
      "mean_token_accuracy": 0.26774311512708665,
      "num_tokens": 186095679.0,
      "step": 81165
    },
    {
      "entropy": 5.484781217575073,
      "epoch": 7.79731027857829,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010269339065563446,
      "loss": 4.5588,
      "mean_token_accuracy": 0.2733698859810829,
      "num_tokens": 186106913.0,
      "step": 81170
    },
    {
      "entropy": 5.51414532661438,
      "epoch": 7.797790585975024,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010267134591978486,
      "loss": 4.6408,
      "mean_token_accuracy": 0.26384390741586683,
      "num_tokens": 186118254.0,
      "step": 81175
    },
    {
      "entropy": 5.465154409408569,
      "epoch": 7.7982708933717575,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010264930518489482,
      "loss": 4.5972,
      "mean_token_accuracy": 0.26874847412109376,
      "num_tokens": 186130860.0,
      "step": 81180
    },
    {
      "entropy": 5.587178802490234,
      "epoch": 7.798751200768492,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010262726845147616,
      "loss": 4.6344,
      "mean_token_accuracy": 0.26054950803518295,
      "num_tokens": 186144911.0,
      "step": 81185
    },
    {
      "entropy": 5.602948379516602,
      "epoch": 7.799231508165226,
      "grad_norm": 1.375,
      "learning_rate": 0.00010260523572004046,
      "loss": 4.7023,
      "mean_token_accuracy": 0.25409339368343353,
      "num_tokens": 186156872.0,
      "step": 81190
    },
    {
      "entropy": 5.469465065002441,
      "epoch": 7.79971181556196,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001025832069910993,
      "loss": 4.5405,
      "mean_token_accuracy": 0.2784587174654007,
      "num_tokens": 186169706.0,
      "step": 81195
    },
    {
      "entropy": 5.552135276794433,
      "epoch": 7.800192122958694,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010256118226516411,
      "loss": 4.6258,
      "mean_token_accuracy": 0.26973360776901245,
      "num_tokens": 186181793.0,
      "step": 81200
    },
    {
      "entropy": 5.47746787071228,
      "epoch": 7.800672430355427,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00010253916154274625,
      "loss": 4.5959,
      "mean_token_accuracy": 0.2761701807379723,
      "num_tokens": 186193286.0,
      "step": 81205
    },
    {
      "entropy": 5.534044790267944,
      "epoch": 7.801152737752162,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010251714482435694,
      "loss": 4.6522,
      "mean_token_accuracy": 0.268588000535965,
      "num_tokens": 186203642.0,
      "step": 81210
    },
    {
      "entropy": 5.513017272949218,
      "epoch": 7.801633045148895,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010249513211050735,
      "loss": 4.6453,
      "mean_token_accuracy": 0.26555069386959074,
      "num_tokens": 186214994.0,
      "step": 81215
    },
    {
      "entropy": 5.676836824417114,
      "epoch": 7.802113352545629,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00010247312340170854,
      "loss": 4.7893,
      "mean_token_accuracy": 0.24671194702386856,
      "num_tokens": 186225688.0,
      "step": 81220
    },
    {
      "entropy": 5.551518249511719,
      "epoch": 7.802593659942363,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001024511186984715,
      "loss": 4.6055,
      "mean_token_accuracy": 0.2718664065003395,
      "num_tokens": 186236658.0,
      "step": 81225
    },
    {
      "entropy": 5.570572185516357,
      "epoch": 7.803073967339097,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010242911800130714,
      "loss": 4.6254,
      "mean_token_accuracy": 0.2629427805542946,
      "num_tokens": 186248358.0,
      "step": 81230
    },
    {
      "entropy": 5.564500999450684,
      "epoch": 7.803554274735831,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010240712131072627,
      "loss": 4.662,
      "mean_token_accuracy": 0.26779225617647173,
      "num_tokens": 186259315.0,
      "step": 81235
    },
    {
      "entropy": 5.504116249084473,
      "epoch": 7.804034582132565,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001023851286272395,
      "loss": 4.6058,
      "mean_token_accuracy": 0.26957480013370516,
      "num_tokens": 186270635.0,
      "step": 81240
    },
    {
      "entropy": 5.5660723686218265,
      "epoch": 7.804514889529298,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001023631399513575,
      "loss": 4.6203,
      "mean_token_accuracy": 0.26967343538999555,
      "num_tokens": 186282092.0,
      "step": 81245
    },
    {
      "entropy": 5.493328094482422,
      "epoch": 7.804995196926033,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010234115528359074,
      "loss": 4.5753,
      "mean_token_accuracy": 0.2779138535261154,
      "num_tokens": 186295627.0,
      "step": 81250
    },
    {
      "entropy": 5.632744073867798,
      "epoch": 7.805475504322766,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010231917462444966,
      "loss": 4.7369,
      "mean_token_accuracy": 0.26139990240335464,
      "num_tokens": 186307044.0,
      "step": 81255
    },
    {
      "entropy": 5.566123199462891,
      "epoch": 7.8059558117195005,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001022971979744446,
      "loss": 4.5809,
      "mean_token_accuracy": 0.2760099425911903,
      "num_tokens": 186317672.0,
      "step": 81260
    },
    {
      "entropy": 5.525916719436646,
      "epoch": 7.806436119116235,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010227522533408584,
      "loss": 4.5886,
      "mean_token_accuracy": 0.2670632556080818,
      "num_tokens": 186328797.0,
      "step": 81265
    },
    {
      "entropy": 5.502487564086914,
      "epoch": 7.806916426512968,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001022532567038834,
      "loss": 4.5831,
      "mean_token_accuracy": 0.26968168616294863,
      "num_tokens": 186341139.0,
      "step": 81270
    },
    {
      "entropy": 5.521794128417969,
      "epoch": 7.807396733909703,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001022312920843474,
      "loss": 4.633,
      "mean_token_accuracy": 0.2671863794326782,
      "num_tokens": 186352793.0,
      "step": 81275
    },
    {
      "entropy": 5.561620855331421,
      "epoch": 7.807877041306436,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010220933147598774,
      "loss": 4.6508,
      "mean_token_accuracy": 0.25862538516521455,
      "num_tokens": 186363839.0,
      "step": 81280
    },
    {
      "entropy": 5.550662088394165,
      "epoch": 7.80835734870317,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010218737487931436,
      "loss": 4.689,
      "mean_token_accuracy": 0.25709679871797564,
      "num_tokens": 186376489.0,
      "step": 81285
    },
    {
      "entropy": 5.560999202728271,
      "epoch": 7.808837656099904,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010216542229483697,
      "loss": 4.6174,
      "mean_token_accuracy": 0.2618746355175972,
      "num_tokens": 186387828.0,
      "step": 81290
    },
    {
      "entropy": 5.540353202819825,
      "epoch": 7.809317963496638,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010214347372306533,
      "loss": 4.6296,
      "mean_token_accuracy": 0.2625975161790848,
      "num_tokens": 186399010.0,
      "step": 81295
    },
    {
      "entropy": 5.524852848052978,
      "epoch": 7.8097982708933715,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001021215291645089,
      "loss": 4.6556,
      "mean_token_accuracy": 0.260277071595192,
      "num_tokens": 186410573.0,
      "step": 81300
    },
    {
      "entropy": 5.5375816822052,
      "epoch": 7.810278578290106,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001020995886196772,
      "loss": 4.5581,
      "mean_token_accuracy": 0.2781038388609886,
      "num_tokens": 186421806.0,
      "step": 81305
    },
    {
      "entropy": 5.496550798416138,
      "epoch": 7.810758885686839,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010207765208907962,
      "loss": 4.5626,
      "mean_token_accuracy": 0.2762739956378937,
      "num_tokens": 186433812.0,
      "step": 81310
    },
    {
      "entropy": 5.546178150177002,
      "epoch": 7.811239193083574,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010205571957322555,
      "loss": 4.6254,
      "mean_token_accuracy": 0.2633053302764893,
      "num_tokens": 186444299.0,
      "step": 81315
    },
    {
      "entropy": 5.425344467163086,
      "epoch": 7.811719500480307,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010203379107262409,
      "loss": 4.4982,
      "mean_token_accuracy": 0.27718197703361513,
      "num_tokens": 186455770.0,
      "step": 81320
    },
    {
      "entropy": 5.499709558486939,
      "epoch": 7.812199807877041,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00010201186658778444,
      "loss": 4.6048,
      "mean_token_accuracy": 0.2727509021759033,
      "num_tokens": 186466966.0,
      "step": 81325
    },
    {
      "entropy": 5.508368444442749,
      "epoch": 7.812680115273775,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010198994611921557,
      "loss": 4.5953,
      "mean_token_accuracy": 0.2705291107296944,
      "num_tokens": 186477913.0,
      "step": 81330
    },
    {
      "entropy": 5.499016332626343,
      "epoch": 7.813160422670509,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010196802966742638,
      "loss": 4.6201,
      "mean_token_accuracy": 0.26861328482627866,
      "num_tokens": 186488692.0,
      "step": 81335
    },
    {
      "entropy": 5.473350381851196,
      "epoch": 7.8136407300672435,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010194611723292575,
      "loss": 4.5658,
      "mean_token_accuracy": 0.2707598924636841,
      "num_tokens": 186500137.0,
      "step": 81340
    },
    {
      "entropy": 5.420165252685547,
      "epoch": 7.814121037463977,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010192420881622242,
      "loss": 4.5071,
      "mean_token_accuracy": 0.27336566150188446,
      "num_tokens": 186511082.0,
      "step": 81345
    },
    {
      "entropy": 5.5265909194946286,
      "epoch": 7.814601344860711,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010190230441782503,
      "loss": 4.6322,
      "mean_token_accuracy": 0.267674033343792,
      "num_tokens": 186523712.0,
      "step": 81350
    },
    {
      "entropy": 5.531818914413452,
      "epoch": 7.815081652257445,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010188040403824212,
      "loss": 4.6568,
      "mean_token_accuracy": 0.2650557115674019,
      "num_tokens": 186535565.0,
      "step": 81355
    },
    {
      "entropy": 5.560288190841675,
      "epoch": 7.815561959654179,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010185850767798219,
      "loss": 4.6294,
      "mean_token_accuracy": 0.2618128642439842,
      "num_tokens": 186547276.0,
      "step": 81360
    },
    {
      "entropy": 5.556744194030761,
      "epoch": 7.816042267050912,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010183661533755362,
      "loss": 4.6176,
      "mean_token_accuracy": 0.2674076735973358,
      "num_tokens": 186559665.0,
      "step": 81365
    },
    {
      "entropy": 5.545628023147583,
      "epoch": 7.816522574447647,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010181472701746463,
      "loss": 4.6203,
      "mean_token_accuracy": 0.2679457351565361,
      "num_tokens": 186570471.0,
      "step": 81370
    },
    {
      "entropy": 5.47947211265564,
      "epoch": 7.81700288184438,
      "grad_norm": 1.375,
      "learning_rate": 0.00010179284271822341,
      "loss": 4.5713,
      "mean_token_accuracy": 0.26761995404958727,
      "num_tokens": 186580960.0,
      "step": 81375
    },
    {
      "entropy": 5.526927709579468,
      "epoch": 7.8174831892411145,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010177096244033803,
      "loss": 4.6056,
      "mean_token_accuracy": 0.270926396548748,
      "num_tokens": 186594114.0,
      "step": 81380
    },
    {
      "entropy": 5.485074853897094,
      "epoch": 7.817963496637848,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010174908618431655,
      "loss": 4.5465,
      "mean_token_accuracy": 0.272643418610096,
      "num_tokens": 186606061.0,
      "step": 81385
    },
    {
      "entropy": 5.52259669303894,
      "epoch": 7.818443804034582,
      "grad_norm": 1.25,
      "learning_rate": 0.00010172721395066683,
      "loss": 4.6137,
      "mean_token_accuracy": 0.2700554996728897,
      "num_tokens": 186619176.0,
      "step": 81390
    },
    {
      "entropy": 5.45236349105835,
      "epoch": 7.818924111431316,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001017053457398967,
      "loss": 4.579,
      "mean_token_accuracy": 0.2778779983520508,
      "num_tokens": 186631004.0,
      "step": 81395
    },
    {
      "entropy": 5.500771713256836,
      "epoch": 7.81940441882805,
      "grad_norm": 1.5,
      "learning_rate": 0.00010168348155251386,
      "loss": 4.5946,
      "mean_token_accuracy": 0.2644282251596451,
      "num_tokens": 186642741.0,
      "step": 81400
    },
    {
      "entropy": 5.589228868484497,
      "epoch": 7.8198847262247835,
      "grad_norm": 1.296875,
      "learning_rate": 0.000101661621389026,
      "loss": 4.7095,
      "mean_token_accuracy": 0.2544361039996147,
      "num_tokens": 186656471.0,
      "step": 81405
    },
    {
      "entropy": 5.555806684494018,
      "epoch": 7.820365033621518,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010163976524994048,
      "loss": 4.6151,
      "mean_token_accuracy": 0.268628303706646,
      "num_tokens": 186667685.0,
      "step": 81410
    },
    {
      "entropy": 5.5910333633422855,
      "epoch": 7.820845341018252,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010161791313576488,
      "loss": 4.6377,
      "mean_token_accuracy": 0.2661697521805763,
      "num_tokens": 186677799.0,
      "step": 81415
    },
    {
      "entropy": 5.515089511871338,
      "epoch": 7.821325648414986,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001015960650470065,
      "loss": 4.5845,
      "mean_token_accuracy": 0.27366636842489245,
      "num_tokens": 186688221.0,
      "step": 81420
    },
    {
      "entropy": 5.517623710632324,
      "epoch": 7.821805955811719,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010157422098417256,
      "loss": 4.6698,
      "mean_token_accuracy": 0.2645576775074005,
      "num_tokens": 186698458.0,
      "step": 81425
    },
    {
      "entropy": 5.526386022567749,
      "epoch": 7.822286263208453,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010155238094777026,
      "loss": 4.6355,
      "mean_token_accuracy": 0.2622980892658234,
      "num_tokens": 186710280.0,
      "step": 81430
    },
    {
      "entropy": 5.546472883224487,
      "epoch": 7.822766570605188,
      "grad_norm": 1.203125,
      "learning_rate": 0.00010153054493830664,
      "loss": 4.6603,
      "mean_token_accuracy": 0.2643910527229309,
      "num_tokens": 186722624.0,
      "step": 81435
    },
    {
      "entropy": 5.510320663452148,
      "epoch": 7.823246878001921,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010150871295628864,
      "loss": 4.6129,
      "mean_token_accuracy": 0.26578067541122435,
      "num_tokens": 186733741.0,
      "step": 81440
    },
    {
      "entropy": 5.480048513412475,
      "epoch": 7.823727185398655,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010148688500222325,
      "loss": 4.5714,
      "mean_token_accuracy": 0.277038699388504,
      "num_tokens": 186745149.0,
      "step": 81445
    },
    {
      "entropy": 5.564575147628784,
      "epoch": 7.824207492795389,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001014650610766171,
      "loss": 4.6696,
      "mean_token_accuracy": 0.26185106337070463,
      "num_tokens": 186757309.0,
      "step": 81450
    },
    {
      "entropy": 5.530782651901245,
      "epoch": 7.824687800192123,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010144324117997693,
      "loss": 4.6308,
      "mean_token_accuracy": 0.27013797760009767,
      "num_tokens": 186768717.0,
      "step": 81455
    },
    {
      "entropy": 5.515321588516235,
      "epoch": 7.825168107588857,
      "grad_norm": 1.640625,
      "learning_rate": 0.00010142142531280935,
      "loss": 4.5889,
      "mean_token_accuracy": 0.26981385201215746,
      "num_tokens": 186779947.0,
      "step": 81460
    },
    {
      "entropy": 5.540974140167236,
      "epoch": 7.825648414985591,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010139961347562084,
      "loss": 4.6951,
      "mean_token_accuracy": 0.2624167248606682,
      "num_tokens": 186791097.0,
      "step": 81465
    },
    {
      "entropy": 5.504434394836426,
      "epoch": 7.826128722382324,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010137780566891783,
      "loss": 4.616,
      "mean_token_accuracy": 0.2641109198331833,
      "num_tokens": 186804132.0,
      "step": 81470
    },
    {
      "entropy": 5.520612144470215,
      "epoch": 7.826609029779059,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010135600189320668,
      "loss": 4.6616,
      "mean_token_accuracy": 0.26366226375102997,
      "num_tokens": 186816680.0,
      "step": 81475
    },
    {
      "entropy": 5.514740610122681,
      "epoch": 7.827089337175792,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001013342021489935,
      "loss": 4.5919,
      "mean_token_accuracy": 0.2679034858942032,
      "num_tokens": 186827659.0,
      "step": 81480
    },
    {
      "entropy": 5.538385534286499,
      "epoch": 7.8275696445725265,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010131240643678444,
      "loss": 4.6452,
      "mean_token_accuracy": 0.27414070069789886,
      "num_tokens": 186840305.0,
      "step": 81485
    },
    {
      "entropy": 5.608678102493286,
      "epoch": 7.828049951969261,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010129061475708556,
      "loss": 4.6719,
      "mean_token_accuracy": 0.26147150099277494,
      "num_tokens": 186852550.0,
      "step": 81490
    },
    {
      "entropy": 5.476624298095703,
      "epoch": 7.828530259365994,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010126882711040281,
      "loss": 4.5607,
      "mean_token_accuracy": 0.27020406872034075,
      "num_tokens": 186864371.0,
      "step": 81495
    },
    {
      "entropy": 5.551012897491455,
      "epoch": 7.829010566762728,
      "grad_norm": 1.375,
      "learning_rate": 0.00010124704349724198,
      "loss": 4.5841,
      "mean_token_accuracy": 0.271035248041153,
      "num_tokens": 186876043.0,
      "step": 81500
    },
    {
      "entropy": 5.541033220291138,
      "epoch": 7.829490874159462,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010122526391810897,
      "loss": 4.6211,
      "mean_token_accuracy": 0.2669347584247589,
      "num_tokens": 186887814.0,
      "step": 81505
    },
    {
      "entropy": 5.5386542797088625,
      "epoch": 7.829971181556196,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010120348837350925,
      "loss": 4.6292,
      "mean_token_accuracy": 0.2675372093915939,
      "num_tokens": 186899008.0,
      "step": 81510
    },
    {
      "entropy": 5.512109422683716,
      "epoch": 7.83045148895293,
      "grad_norm": 1.375,
      "learning_rate": 0.00010118171686394844,
      "loss": 4.6199,
      "mean_token_accuracy": 0.2692327469587326,
      "num_tokens": 186910457.0,
      "step": 81515
    },
    {
      "entropy": 5.547292852401734,
      "epoch": 7.830931796349664,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010115994938993205,
      "loss": 4.6472,
      "mean_token_accuracy": 0.27106847763061526,
      "num_tokens": 186922419.0,
      "step": 81520
    },
    {
      "entropy": 5.512323093414307,
      "epoch": 7.8314121037463975,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00010113818595196545,
      "loss": 4.602,
      "mean_token_accuracy": 0.2713282942771912,
      "num_tokens": 186934068.0,
      "step": 81525
    },
    {
      "entropy": 5.508616399765015,
      "epoch": 7.831892411143132,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010111642655055387,
      "loss": 4.5884,
      "mean_token_accuracy": 0.2773835748434067,
      "num_tokens": 186944638.0,
      "step": 81530
    },
    {
      "entropy": 5.510360860824585,
      "epoch": 7.832372718539865,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010109467118620264,
      "loss": 4.559,
      "mean_token_accuracy": 0.27549957633018496,
      "num_tokens": 186955896.0,
      "step": 81535
    },
    {
      "entropy": 5.503798580169677,
      "epoch": 7.8328530259366,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010107291985941666,
      "loss": 4.611,
      "mean_token_accuracy": 0.26273554265499116,
      "num_tokens": 186967094.0,
      "step": 81540
    },
    {
      "entropy": 5.544091796875,
      "epoch": 7.833333333333333,
      "grad_norm": 1.375,
      "learning_rate": 0.00010105117257070103,
      "loss": 4.6544,
      "mean_token_accuracy": 0.26429521590471267,
      "num_tokens": 186978067.0,
      "step": 81545
    },
    {
      "entropy": 5.549781084060669,
      "epoch": 7.833813640730067,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010102942932056067,
      "loss": 4.6555,
      "mean_token_accuracy": 0.2674077108502388,
      "num_tokens": 186989602.0,
      "step": 81550
    },
    {
      "entropy": 5.515760326385498,
      "epoch": 7.834293948126801,
      "grad_norm": 1.375,
      "learning_rate": 0.00010100769010950035,
      "loss": 4.5655,
      "mean_token_accuracy": 0.27681067287921907,
      "num_tokens": 187000785.0,
      "step": 81555
    },
    {
      "entropy": 5.623289918899536,
      "epoch": 7.834774255523535,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010098595493802484,
      "loss": 4.74,
      "mean_token_accuracy": 0.25913273841142653,
      "num_tokens": 187012474.0,
      "step": 81560
    },
    {
      "entropy": 5.557417726516723,
      "epoch": 7.8352545629202694,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010096422380663881,
      "loss": 4.5856,
      "mean_token_accuracy": 0.26865483075380325,
      "num_tokens": 187023351.0,
      "step": 81565
    },
    {
      "entropy": 5.420019483566284,
      "epoch": 7.835734870317003,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001009424967158466,
      "loss": 4.53,
      "mean_token_accuracy": 0.276714226603508,
      "num_tokens": 187034716.0,
      "step": 81570
    },
    {
      "entropy": 5.513358640670776,
      "epoch": 7.836215177713736,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010092077366615282,
      "loss": 4.5997,
      "mean_token_accuracy": 0.2659653007984161,
      "num_tokens": 187046833.0,
      "step": 81575
    },
    {
      "entropy": 5.5458083152771,
      "epoch": 7.836695485110471,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010089905465806175,
      "loss": 4.6875,
      "mean_token_accuracy": 0.26037041246891024,
      "num_tokens": 187058677.0,
      "step": 81580
    },
    {
      "entropy": 5.494491863250732,
      "epoch": 7.837175792507205,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010087733969207767,
      "loss": 4.5471,
      "mean_token_accuracy": 0.271082903444767,
      "num_tokens": 187070647.0,
      "step": 81585
    },
    {
      "entropy": 5.542911052703857,
      "epoch": 7.837656099903938,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010085562876870474,
      "loss": 4.6531,
      "mean_token_accuracy": 0.2729394882917404,
      "num_tokens": 187083664.0,
      "step": 81590
    },
    {
      "entropy": 5.541182184219361,
      "epoch": 7.838136407300673,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010083392188844705,
      "loss": 4.6055,
      "mean_token_accuracy": 0.26721378415822983,
      "num_tokens": 187094016.0,
      "step": 81595
    },
    {
      "entropy": 5.5049666404724125,
      "epoch": 7.838616714697406,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010081221905180844,
      "loss": 4.5703,
      "mean_token_accuracy": 0.26870280504226685,
      "num_tokens": 187105189.0,
      "step": 81600
    },
    {
      "entropy": 5.501814365386963,
      "epoch": 7.8390970220941405,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0001007905202592929,
      "loss": 4.5853,
      "mean_token_accuracy": 0.2681757315993309,
      "num_tokens": 187115706.0,
      "step": 81605
    },
    {
      "entropy": 5.576114892959595,
      "epoch": 7.839577329490874,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010076882551140418,
      "loss": 4.6708,
      "mean_token_accuracy": 0.2576023146510124,
      "num_tokens": 187126095.0,
      "step": 81610
    },
    {
      "entropy": 5.570911073684693,
      "epoch": 7.840057636887608,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010074713480864598,
      "loss": 4.6843,
      "mean_token_accuracy": 0.2644941106438637,
      "num_tokens": 187137735.0,
      "step": 81615
    },
    {
      "entropy": 5.451881742477417,
      "epoch": 7.840537944284342,
      "grad_norm": 1.5,
      "learning_rate": 0.00010072544815152187,
      "loss": 4.5365,
      "mean_token_accuracy": 0.27223353683948515,
      "num_tokens": 187149299.0,
      "step": 81620
    },
    {
      "entropy": 5.511653137207031,
      "epoch": 7.841018251681076,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010070376554053542,
      "loss": 4.6112,
      "mean_token_accuracy": 0.2649811640381813,
      "num_tokens": 187161374.0,
      "step": 81625
    },
    {
      "entropy": 5.528215122222901,
      "epoch": 7.8414985590778095,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010068208697618995,
      "loss": 4.6272,
      "mean_token_accuracy": 0.2601693466305733,
      "num_tokens": 187172221.0,
      "step": 81630
    },
    {
      "entropy": 5.570185899734497,
      "epoch": 7.841978866474544,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010066041245898877,
      "loss": 4.7079,
      "mean_token_accuracy": 0.2624524042010307,
      "num_tokens": 187183815.0,
      "step": 81635
    },
    {
      "entropy": 5.453647327423096,
      "epoch": 7.842459173871278,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010063874198943515,
      "loss": 4.5751,
      "mean_token_accuracy": 0.2802478536963463,
      "num_tokens": 187195634.0,
      "step": 81640
    },
    {
      "entropy": 5.534964084625244,
      "epoch": 7.8429394812680115,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010061707556803218,
      "loss": 4.5625,
      "mean_token_accuracy": 0.2740906193852425,
      "num_tokens": 187206641.0,
      "step": 81645
    },
    {
      "entropy": 5.476779174804688,
      "epoch": 7.843419788664745,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010059541319528293,
      "loss": 4.598,
      "mean_token_accuracy": 0.2666224330663681,
      "num_tokens": 187216788.0,
      "step": 81650
    },
    {
      "entropy": 5.52952094078064,
      "epoch": 7.843900096061479,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010057375487169034,
      "loss": 4.6017,
      "mean_token_accuracy": 0.270405738055706,
      "num_tokens": 187228489.0,
      "step": 81655
    },
    {
      "entropy": 5.54585485458374,
      "epoch": 7.844380403458214,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001005521005977572,
      "loss": 4.6848,
      "mean_token_accuracy": 0.263845157623291,
      "num_tokens": 187239862.0,
      "step": 81660
    },
    {
      "entropy": 5.503619718551636,
      "epoch": 7.844860710854947,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010053045037398622,
      "loss": 4.5932,
      "mean_token_accuracy": 0.27269532084465026,
      "num_tokens": 187251016.0,
      "step": 81665
    },
    {
      "entropy": 5.489659357070923,
      "epoch": 7.845341018251681,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010050880420088019,
      "loss": 4.5908,
      "mean_token_accuracy": 0.26539310067892075,
      "num_tokens": 187260988.0,
      "step": 81670
    },
    {
      "entropy": 5.631889867782593,
      "epoch": 7.845821325648415,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010048716207894154,
      "loss": 4.7089,
      "mean_token_accuracy": 0.26226541101932527,
      "num_tokens": 187272096.0,
      "step": 81675
    },
    {
      "entropy": 5.564923286437988,
      "epoch": 7.846301633045149,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010046552400867282,
      "loss": 4.6941,
      "mean_token_accuracy": 0.26799991726875305,
      "num_tokens": 187283873.0,
      "step": 81680
    },
    {
      "entropy": 5.5167001247406,
      "epoch": 7.846781940441883,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001004438899905764,
      "loss": 4.6188,
      "mean_token_accuracy": 0.27329549491405486,
      "num_tokens": 187295639.0,
      "step": 81685
    },
    {
      "entropy": 5.475968551635742,
      "epoch": 7.847262247838617,
      "grad_norm": 1.5,
      "learning_rate": 0.00010042226002515448,
      "loss": 4.5284,
      "mean_token_accuracy": 0.2766260474920273,
      "num_tokens": 187307115.0,
      "step": 81690
    },
    {
      "entropy": 5.482641410827637,
      "epoch": 7.84774255523535,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010040063411290932,
      "loss": 4.6079,
      "mean_token_accuracy": 0.26986519545316695,
      "num_tokens": 187318174.0,
      "step": 81695
    },
    {
      "entropy": 5.552752447128296,
      "epoch": 7.848222862632085,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010037901225434295,
      "loss": 4.6547,
      "mean_token_accuracy": 0.2638910174369812,
      "num_tokens": 187329955.0,
      "step": 81700
    },
    {
      "entropy": 5.502349042892456,
      "epoch": 7.848703170028818,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010035739444995743,
      "loss": 4.5739,
      "mean_token_accuracy": 0.26680516749620437,
      "num_tokens": 187341629.0,
      "step": 81705
    },
    {
      "entropy": 5.6611754417419435,
      "epoch": 7.8491834774255524,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010033578070025459,
      "loss": 4.6919,
      "mean_token_accuracy": 0.26641506105661394,
      "num_tokens": 187352773.0,
      "step": 81710
    },
    {
      "entropy": 5.537397003173828,
      "epoch": 7.849663784822286,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010031417100573636,
      "loss": 4.6115,
      "mean_token_accuracy": 0.26589563935995103,
      "num_tokens": 187365053.0,
      "step": 81715
    },
    {
      "entropy": 5.508825492858887,
      "epoch": 7.85014409221902,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010029256536690429,
      "loss": 4.592,
      "mean_token_accuracy": 0.26527789086103437,
      "num_tokens": 187377098.0,
      "step": 81720
    },
    {
      "entropy": 5.57076735496521,
      "epoch": 7.850624399615754,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001002709637842601,
      "loss": 4.6634,
      "mean_token_accuracy": 0.2671664372086525,
      "num_tokens": 187388686.0,
      "step": 81725
    },
    {
      "entropy": 5.444347286224366,
      "epoch": 7.851104707012488,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010024936625830527,
      "loss": 4.5375,
      "mean_token_accuracy": 0.273052316904068,
      "num_tokens": 187399983.0,
      "step": 81730
    },
    {
      "entropy": 5.5393256664276125,
      "epoch": 7.851585014409222,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010022777278954128,
      "loss": 4.6715,
      "mean_token_accuracy": 0.2593299448490143,
      "num_tokens": 187412268.0,
      "step": 81735
    },
    {
      "entropy": 5.553453397750855,
      "epoch": 7.852065321805956,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010020618337846941,
      "loss": 4.6412,
      "mean_token_accuracy": 0.26723902374506,
      "num_tokens": 187423230.0,
      "step": 81740
    },
    {
      "entropy": 5.542643308639526,
      "epoch": 7.85254562920269,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010018459802559102,
      "loss": 4.6242,
      "mean_token_accuracy": 0.2646590769290924,
      "num_tokens": 187434633.0,
      "step": 81745
    },
    {
      "entropy": 5.6407325744628904,
      "epoch": 7.8530259365994235,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010016301673140708,
      "loss": 4.6864,
      "mean_token_accuracy": 0.2622524335980415,
      "num_tokens": 187445661.0,
      "step": 81750
    },
    {
      "entropy": 5.591559934616089,
      "epoch": 7.853506243996158,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010014143949641877,
      "loss": 4.6898,
      "mean_token_accuracy": 0.2575438529253006,
      "num_tokens": 187456218.0,
      "step": 81755
    },
    {
      "entropy": 5.609073781967163,
      "epoch": 7.853986551392891,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010011986632112696,
      "loss": 4.7032,
      "mean_token_accuracy": 0.260616147518158,
      "num_tokens": 187468331.0,
      "step": 81760
    },
    {
      "entropy": 5.532809782028198,
      "epoch": 7.854466858789626,
      "grad_norm": 1.640625,
      "learning_rate": 0.00010009829720603261,
      "loss": 4.6138,
      "mean_token_accuracy": 0.2697667464613914,
      "num_tokens": 187478673.0,
      "step": 81765
    },
    {
      "entropy": 5.527470541000366,
      "epoch": 7.854947166186359,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00010007673215163643,
      "loss": 4.5557,
      "mean_token_accuracy": 0.27455989122390745,
      "num_tokens": 187490465.0,
      "step": 81770
    },
    {
      "entropy": 5.62494125366211,
      "epoch": 7.855427473583093,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010005517115843918,
      "loss": 4.7168,
      "mean_token_accuracy": 0.2582068920135498,
      "num_tokens": 187501725.0,
      "step": 81775
    },
    {
      "entropy": 5.580956077575683,
      "epoch": 7.855907780979827,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010003361422694129,
      "loss": 4.6589,
      "mean_token_accuracy": 0.2663057342171669,
      "num_tokens": 187512545.0,
      "step": 81780
    },
    {
      "entropy": 5.535506916046143,
      "epoch": 7.856388088376561,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010001206135764338,
      "loss": 4.612,
      "mean_token_accuracy": 0.2655406355857849,
      "num_tokens": 187524457.0,
      "step": 81785
    },
    {
      "entropy": 5.556659603118897,
      "epoch": 7.8568683957732945,
      "grad_norm": 1.3984375,
      "learning_rate": 9.999051255104576e-05,
      "loss": 4.7448,
      "mean_token_accuracy": 0.26290324330329895,
      "num_tokens": 187536376.0,
      "step": 81790
    },
    {
      "entropy": 5.521308135986328,
      "epoch": 7.857348703170029,
      "grad_norm": 1.4765625,
      "learning_rate": 9.996896780764879e-05,
      "loss": 4.5898,
      "mean_token_accuracy": 0.27306167781352997,
      "num_tokens": 187547016.0,
      "step": 81795
    },
    {
      "entropy": 5.553483343124389,
      "epoch": 7.857829010566762,
      "grad_norm": 1.40625,
      "learning_rate": 9.994742712795263e-05,
      "loss": 4.6363,
      "mean_token_accuracy": 0.2631507232785225,
      "num_tokens": 187557704.0,
      "step": 81800
    },
    {
      "entropy": 5.53595495223999,
      "epoch": 7.858309317963497,
      "grad_norm": 1.3125,
      "learning_rate": 9.992589051245752e-05,
      "loss": 4.6376,
      "mean_token_accuracy": 0.2606348246335983,
      "num_tokens": 187570471.0,
      "step": 81805
    },
    {
      "entropy": 5.626683235168457,
      "epoch": 7.858789625360231,
      "grad_norm": 1.4765625,
      "learning_rate": 9.990435796166325e-05,
      "loss": 4.7197,
      "mean_token_accuracy": 0.25690711140632627,
      "num_tokens": 187581347.0,
      "step": 81810
    },
    {
      "entropy": 5.474719524383545,
      "epoch": 7.859269932756964,
      "grad_norm": 1.375,
      "learning_rate": 9.988282947606992e-05,
      "loss": 4.5876,
      "mean_token_accuracy": 0.2733455091714859,
      "num_tokens": 187592609.0,
      "step": 81815
    },
    {
      "entropy": 5.589661407470703,
      "epoch": 7.859750240153699,
      "grad_norm": 1.421875,
      "learning_rate": 9.986130505617724e-05,
      "loss": 4.6945,
      "mean_token_accuracy": 0.2619194030761719,
      "num_tokens": 187603220.0,
      "step": 81820
    },
    {
      "entropy": 5.528939580917358,
      "epoch": 7.860230547550432,
      "grad_norm": 1.3671875,
      "learning_rate": 9.983978470248503e-05,
      "loss": 4.65,
      "mean_token_accuracy": 0.26152504831552503,
      "num_tokens": 187614564.0,
      "step": 81825
    },
    {
      "entropy": 5.5159735679626465,
      "epoch": 7.8607108549471665,
      "grad_norm": 1.546875,
      "learning_rate": 9.981826841549289e-05,
      "loss": 4.6206,
      "mean_token_accuracy": 0.26489160656929017,
      "num_tokens": 187626415.0,
      "step": 81830
    },
    {
      "entropy": 5.4986976146697994,
      "epoch": 7.8611911623439,
      "grad_norm": 1.4140625,
      "learning_rate": 9.979675619570045e-05,
      "loss": 4.5378,
      "mean_token_accuracy": 0.2781826853752136,
      "num_tokens": 187637480.0,
      "step": 81835
    },
    {
      "entropy": 5.530438470840454,
      "epoch": 7.861671469740634,
      "grad_norm": 1.421875,
      "learning_rate": 9.977524804360703e-05,
      "loss": 4.6009,
      "mean_token_accuracy": 0.2696570619940758,
      "num_tokens": 187648939.0,
      "step": 81840
    },
    {
      "entropy": 5.558337879180908,
      "epoch": 7.862151777137368,
      "grad_norm": 1.40625,
      "learning_rate": 9.975374395971201e-05,
      "loss": 4.6615,
      "mean_token_accuracy": 0.2620045647025108,
      "num_tokens": 187662030.0,
      "step": 81845
    },
    {
      "entropy": 5.4065734386444095,
      "epoch": 7.862632084534102,
      "grad_norm": 1.40625,
      "learning_rate": 9.973224394451472e-05,
      "loss": 4.4903,
      "mean_token_accuracy": 0.27434021681547166,
      "num_tokens": 187674082.0,
      "step": 81850
    },
    {
      "entropy": 5.53447995185852,
      "epoch": 7.863112391930835,
      "grad_norm": 1.4921875,
      "learning_rate": 9.971074799851426e-05,
      "loss": 4.6242,
      "mean_token_accuracy": 0.2720614656805992,
      "num_tokens": 187684364.0,
      "step": 81855
    },
    {
      "entropy": 5.49135274887085,
      "epoch": 7.86359269932757,
      "grad_norm": 1.515625,
      "learning_rate": 9.968925612220972e-05,
      "loss": 4.6283,
      "mean_token_accuracy": 0.2611041933298111,
      "num_tokens": 187696426.0,
      "step": 81860
    },
    {
      "entropy": 5.447522687911987,
      "epoch": 7.864073006724303,
      "grad_norm": 1.328125,
      "learning_rate": 9.966776831610016e-05,
      "loss": 4.4704,
      "mean_token_accuracy": 0.28636264204978945,
      "num_tokens": 187707985.0,
      "step": 81865
    },
    {
      "entropy": 5.559007072448731,
      "epoch": 7.8645533141210375,
      "grad_norm": 1.4609375,
      "learning_rate": 9.964628458068432e-05,
      "loss": 4.6472,
      "mean_token_accuracy": 0.2624396219849586,
      "num_tokens": 187719564.0,
      "step": 81870
    },
    {
      "entropy": 5.500686311721802,
      "epoch": 7.865033621517771,
      "grad_norm": 1.3828125,
      "learning_rate": 9.962480491646107e-05,
      "loss": 4.5608,
      "mean_token_accuracy": 0.27547193616628646,
      "num_tokens": 187731343.0,
      "step": 81875
    },
    {
      "entropy": 5.555851459503174,
      "epoch": 7.865513928914505,
      "grad_norm": 1.390625,
      "learning_rate": 9.96033293239291e-05,
      "loss": 4.6788,
      "mean_token_accuracy": 0.26088224053382875,
      "num_tokens": 187742722.0,
      "step": 81880
    },
    {
      "entropy": 5.55864691734314,
      "epoch": 7.86599423631124,
      "grad_norm": 1.3046875,
      "learning_rate": 9.9581857803587e-05,
      "loss": 4.6675,
      "mean_token_accuracy": 0.2614060193300247,
      "num_tokens": 187754789.0,
      "step": 81885
    },
    {
      "entropy": 5.570469999313355,
      "epoch": 7.866474543707973,
      "grad_norm": 1.421875,
      "learning_rate": 9.956039035593328e-05,
      "loss": 4.6503,
      "mean_token_accuracy": 0.2650793120265007,
      "num_tokens": 187765905.0,
      "step": 81890
    },
    {
      "entropy": 5.489630889892578,
      "epoch": 7.866954851104707,
      "grad_norm": 1.4375,
      "learning_rate": 9.953892698146633e-05,
      "loss": 4.5654,
      "mean_token_accuracy": 0.2716288536787033,
      "num_tokens": 187777295.0,
      "step": 81895
    },
    {
      "entropy": 5.586165046691894,
      "epoch": 7.867435158501441,
      "grad_norm": 1.640625,
      "learning_rate": 9.951746768068453e-05,
      "loss": 4.6368,
      "mean_token_accuracy": 0.26976011991500853,
      "num_tokens": 187788069.0,
      "step": 81900
    },
    {
      "entropy": 5.513994598388672,
      "epoch": 7.867915465898175,
      "grad_norm": 1.4140625,
      "learning_rate": 9.949601245408605e-05,
      "loss": 4.5598,
      "mean_token_accuracy": 0.27390662133693694,
      "num_tokens": 187798411.0,
      "step": 81905
    },
    {
      "entropy": 5.556591272354126,
      "epoch": 7.868395773294909,
      "grad_norm": 1.3515625,
      "learning_rate": 9.947456130216897e-05,
      "loss": 4.6642,
      "mean_token_accuracy": 0.26414810717105863,
      "num_tokens": 187809045.0,
      "step": 81910
    },
    {
      "entropy": 5.502201890945434,
      "epoch": 7.868876080691643,
      "grad_norm": 1.3203125,
      "learning_rate": 9.945311422543138e-05,
      "loss": 4.5627,
      "mean_token_accuracy": 0.2720388486981392,
      "num_tokens": 187821977.0,
      "step": 81915
    },
    {
      "entropy": 5.530204486846924,
      "epoch": 7.869356388088376,
      "grad_norm": 1.484375,
      "learning_rate": 9.943167122437123e-05,
      "loss": 4.621,
      "mean_token_accuracy": 0.2617656782269478,
      "num_tokens": 187833867.0,
      "step": 81920
    },
    {
      "entropy": 5.508363914489746,
      "epoch": 7.869836695485111,
      "grad_norm": 1.5390625,
      "learning_rate": 9.941023229948634e-05,
      "loss": 4.6011,
      "mean_token_accuracy": 0.2675512135028839,
      "num_tokens": 187844403.0,
      "step": 81925
    },
    {
      "entropy": 5.546095323562622,
      "epoch": 7.870317002881844,
      "grad_norm": 1.453125,
      "learning_rate": 9.938879745127447e-05,
      "loss": 4.6767,
      "mean_token_accuracy": 0.26488994657993314,
      "num_tokens": 187856524.0,
      "step": 81930
    },
    {
      "entropy": 5.487834739685058,
      "epoch": 7.870797310278578,
      "grad_norm": 1.546875,
      "learning_rate": 9.936736668023323e-05,
      "loss": 4.5654,
      "mean_token_accuracy": 0.27406144589185716,
      "num_tokens": 187866354.0,
      "step": 81935
    },
    {
      "entropy": 5.451440334320068,
      "epoch": 7.871277617675312,
      "grad_norm": 1.3671875,
      "learning_rate": 9.934593998686031e-05,
      "loss": 4.5641,
      "mean_token_accuracy": 0.26955062001943586,
      "num_tokens": 187877759.0,
      "step": 81940
    },
    {
      "entropy": 5.531260442733765,
      "epoch": 7.871757925072046,
      "grad_norm": 1.3828125,
      "learning_rate": 9.932451737165299e-05,
      "loss": 4.6156,
      "mean_token_accuracy": 0.2682308405637741,
      "num_tokens": 187890126.0,
      "step": 81945
    },
    {
      "entropy": 5.564579916000366,
      "epoch": 7.87223823246878,
      "grad_norm": 1.4609375,
      "learning_rate": 9.930309883510874e-05,
      "loss": 4.619,
      "mean_token_accuracy": 0.2634873777627945,
      "num_tokens": 187901551.0,
      "step": 81950
    },
    {
      "entropy": 5.476612901687622,
      "epoch": 7.872718539865514,
      "grad_norm": 1.3984375,
      "learning_rate": 9.928168437772483e-05,
      "loss": 4.5654,
      "mean_token_accuracy": 0.2721452504396439,
      "num_tokens": 187913367.0,
      "step": 81955
    },
    {
      "entropy": 5.5210946559906,
      "epoch": 7.873198847262248,
      "grad_norm": 1.5546875,
      "learning_rate": 9.926027399999841e-05,
      "loss": 4.5765,
      "mean_token_accuracy": 0.2728807896375656,
      "num_tokens": 187925043.0,
      "step": 81960
    },
    {
      "entropy": 5.572312068939209,
      "epoch": 7.873679154658982,
      "grad_norm": 1.390625,
      "learning_rate": 9.923886770242662e-05,
      "loss": 4.6848,
      "mean_token_accuracy": 0.25906564593315123,
      "num_tokens": 187937140.0,
      "step": 81965
    },
    {
      "entropy": 5.556784057617188,
      "epoch": 7.874159462055716,
      "grad_norm": 1.65625,
      "learning_rate": 9.921746548550639e-05,
      "loss": 4.6596,
      "mean_token_accuracy": 0.2640139475464821,
      "num_tokens": 187947356.0,
      "step": 81970
    },
    {
      "entropy": 5.654638719558716,
      "epoch": 7.8746397694524495,
      "grad_norm": 1.4765625,
      "learning_rate": 9.919606734973467e-05,
      "loss": 4.758,
      "mean_token_accuracy": 0.2638107791543007,
      "num_tokens": 187958293.0,
      "step": 81975
    },
    {
      "entropy": 5.473989391326905,
      "epoch": 7.875120076849184,
      "grad_norm": 1.53125,
      "learning_rate": 9.917467329560828e-05,
      "loss": 4.5305,
      "mean_token_accuracy": 0.27912798821926116,
      "num_tokens": 187969816.0,
      "step": 81980
    },
    {
      "entropy": 5.560119199752807,
      "epoch": 7.875600384245917,
      "grad_norm": 1.46875,
      "learning_rate": 9.91532833236238e-05,
      "loss": 4.6466,
      "mean_token_accuracy": 0.2696546643972397,
      "num_tokens": 187981455.0,
      "step": 81985
    },
    {
      "entropy": 5.4988502025604244,
      "epoch": 7.876080691642652,
      "grad_norm": 1.4765625,
      "learning_rate": 9.913189743427797e-05,
      "loss": 4.5896,
      "mean_token_accuracy": 0.26672624349594115,
      "num_tokens": 187993550.0,
      "step": 81990
    },
    {
      "entropy": 5.603810977935791,
      "epoch": 7.876560999039385,
      "grad_norm": 1.4609375,
      "learning_rate": 9.911051562806724e-05,
      "loss": 4.6352,
      "mean_token_accuracy": 0.2661183923482895,
      "num_tokens": 188004860.0,
      "step": 81995
    },
    {
      "entropy": 5.582941007614136,
      "epoch": 7.877041306436119,
      "grad_norm": 1.421875,
      "learning_rate": 9.908913790548804e-05,
      "loss": 4.6767,
      "mean_token_accuracy": 0.2690474569797516,
      "num_tokens": 188016995.0,
      "step": 82000
    },
    {
      "entropy": 5.566486263275147,
      "epoch": 7.877521613832853,
      "grad_norm": 1.4375,
      "learning_rate": 9.906776426703672e-05,
      "loss": 4.6139,
      "mean_token_accuracy": 0.2663489133119583,
      "num_tokens": 188029147.0,
      "step": 82005
    },
    {
      "entropy": 5.493472957611084,
      "epoch": 7.878001921229587,
      "grad_norm": 1.484375,
      "learning_rate": 9.904639471320952e-05,
      "loss": 4.5381,
      "mean_token_accuracy": 0.2748530119657516,
      "num_tokens": 188039906.0,
      "step": 82010
    },
    {
      "entropy": 5.555211305618286,
      "epoch": 7.8784822286263205,
      "grad_norm": 1.6015625,
      "learning_rate": 9.902502924450261e-05,
      "loss": 4.634,
      "mean_token_accuracy": 0.26456251591444013,
      "num_tokens": 188052992.0,
      "step": 82015
    },
    {
      "entropy": 5.53145842552185,
      "epoch": 7.878962536023055,
      "grad_norm": 1.4609375,
      "learning_rate": 9.900366786141189e-05,
      "loss": 4.6149,
      "mean_token_accuracy": 0.26200028508901596,
      "num_tokens": 188063981.0,
      "step": 82020
    },
    {
      "entropy": 5.500498294830322,
      "epoch": 7.879442843419788,
      "grad_norm": 1.5078125,
      "learning_rate": 9.898231056443345e-05,
      "loss": 4.6132,
      "mean_token_accuracy": 0.2707889899611473,
      "num_tokens": 188075519.0,
      "step": 82025
    },
    {
      "entropy": 5.535169219970703,
      "epoch": 7.879923150816523,
      "grad_norm": 1.4453125,
      "learning_rate": 9.896095735406306e-05,
      "loss": 4.6173,
      "mean_token_accuracy": 0.2630413442850113,
      "num_tokens": 188087292.0,
      "step": 82030
    },
    {
      "entropy": 5.508405113220215,
      "epoch": 7.880403458213257,
      "grad_norm": 1.3203125,
      "learning_rate": 9.893960823079653e-05,
      "loss": 4.5746,
      "mean_token_accuracy": 0.2766424804925919,
      "num_tokens": 188098549.0,
      "step": 82035
    },
    {
      "entropy": 5.56712794303894,
      "epoch": 7.88088376560999,
      "grad_norm": 1.3828125,
      "learning_rate": 9.89182631951295e-05,
      "loss": 4.6845,
      "mean_token_accuracy": 0.26396460831165314,
      "num_tokens": 188110760.0,
      "step": 82040
    },
    {
      "entropy": 5.45865159034729,
      "epoch": 7.881364073006724,
      "grad_norm": 1.296875,
      "learning_rate": 9.889692224755759e-05,
      "loss": 4.5761,
      "mean_token_accuracy": 0.27634447067976,
      "num_tokens": 188122876.0,
      "step": 82045
    },
    {
      "entropy": 5.457508230209351,
      "epoch": 7.881844380403458,
      "grad_norm": 1.3828125,
      "learning_rate": 9.887558538857619e-05,
      "loss": 4.5061,
      "mean_token_accuracy": 0.2780029386281967,
      "num_tokens": 188133625.0,
      "step": 82050
    },
    {
      "entropy": 5.502433586120605,
      "epoch": 7.8823246878001925,
      "grad_norm": 1.375,
      "learning_rate": 9.885425261868069e-05,
      "loss": 4.5433,
      "mean_token_accuracy": 0.27320015877485276,
      "num_tokens": 188145109.0,
      "step": 82055
    },
    {
      "entropy": 5.492188310623169,
      "epoch": 7.882804995196926,
      "grad_norm": 1.359375,
      "learning_rate": 9.88329239383664e-05,
      "loss": 4.591,
      "mean_token_accuracy": 0.2690177232027054,
      "num_tokens": 188155982.0,
      "step": 82060
    },
    {
      "entropy": 5.518042659759521,
      "epoch": 7.88328530259366,
      "grad_norm": 1.359375,
      "learning_rate": 9.881159934812853e-05,
      "loss": 4.5779,
      "mean_token_accuracy": 0.2683809459209442,
      "num_tokens": 188169028.0,
      "step": 82065
    },
    {
      "entropy": 5.580500364303589,
      "epoch": 7.883765609990394,
      "grad_norm": 1.40625,
      "learning_rate": 9.879027884846215e-05,
      "loss": 4.6724,
      "mean_token_accuracy": 0.2603022038936615,
      "num_tokens": 188180488.0,
      "step": 82070
    },
    {
      "entropy": 5.604614114761352,
      "epoch": 7.884245917387128,
      "grad_norm": 1.5234375,
      "learning_rate": 9.87689624398623e-05,
      "loss": 4.7064,
      "mean_token_accuracy": 0.2624202027916908,
      "num_tokens": 188192070.0,
      "step": 82075
    },
    {
      "entropy": 5.6111695766448975,
      "epoch": 7.884726224783861,
      "grad_norm": 1.421875,
      "learning_rate": 9.874765012282379e-05,
      "loss": 4.7005,
      "mean_token_accuracy": 0.25740617364645,
      "num_tokens": 188204468.0,
      "step": 82080
    },
    {
      "entropy": 5.4755748271942135,
      "epoch": 7.885206532180596,
      "grad_norm": 1.390625,
      "learning_rate": 9.872634189784149e-05,
      "loss": 4.5553,
      "mean_token_accuracy": 0.2757469147443771,
      "num_tokens": 188215821.0,
      "step": 82085
    },
    {
      "entropy": 5.61968994140625,
      "epoch": 7.885686839577329,
      "grad_norm": 1.3671875,
      "learning_rate": 9.870503776541013e-05,
      "loss": 4.6982,
      "mean_token_accuracy": 0.26315333545207975,
      "num_tokens": 188227030.0,
      "step": 82090
    },
    {
      "entropy": 5.542966747283936,
      "epoch": 7.8861671469740635,
      "grad_norm": 1.3515625,
      "learning_rate": 9.868373772602423e-05,
      "loss": 4.6977,
      "mean_token_accuracy": 0.2699266329407692,
      "num_tokens": 188237922.0,
      "step": 82095
    },
    {
      "entropy": 5.631543111801148,
      "epoch": 7.886647454370797,
      "grad_norm": 1.375,
      "learning_rate": 9.866244178017844e-05,
      "loss": 4.6797,
      "mean_token_accuracy": 0.26743755787611007,
      "num_tokens": 188248914.0,
      "step": 82100
    },
    {
      "entropy": 5.539115858078003,
      "epoch": 7.887127761767531,
      "grad_norm": 1.4921875,
      "learning_rate": 9.864114992836721e-05,
      "loss": 4.6186,
      "mean_token_accuracy": 0.26735239624977114,
      "num_tokens": 188260143.0,
      "step": 82105
    },
    {
      "entropy": 5.524415779113769,
      "epoch": 7.887608069164266,
      "grad_norm": 1.5859375,
      "learning_rate": 9.861986217108473e-05,
      "loss": 4.6281,
      "mean_token_accuracy": 0.25850734710693357,
      "num_tokens": 188271409.0,
      "step": 82110
    },
    {
      "entropy": 5.519557094573974,
      "epoch": 7.888088376560999,
      "grad_norm": 1.34375,
      "learning_rate": 9.859857850882528e-05,
      "loss": 4.5733,
      "mean_token_accuracy": 0.2707141265273094,
      "num_tokens": 188282042.0,
      "step": 82115
    },
    {
      "entropy": 5.48105902671814,
      "epoch": 7.8885686839577325,
      "grad_norm": 1.484375,
      "learning_rate": 9.857729894208306e-05,
      "loss": 4.5025,
      "mean_token_accuracy": 0.28198714554309845,
      "num_tokens": 188293482.0,
      "step": 82120
    },
    {
      "entropy": 5.502406740188599,
      "epoch": 7.889048991354467,
      "grad_norm": 1.671875,
      "learning_rate": 9.855602347135207e-05,
      "loss": 4.5815,
      "mean_token_accuracy": 0.26589069068431853,
      "num_tokens": 188305475.0,
      "step": 82125
    },
    {
      "entropy": 5.520367527008057,
      "epoch": 7.889529298751201,
      "grad_norm": 1.3671875,
      "learning_rate": 9.85347520971263e-05,
      "loss": 4.6538,
      "mean_token_accuracy": 0.26935974359512327,
      "num_tokens": 188318593.0,
      "step": 82130
    },
    {
      "entropy": 5.5437132835388185,
      "epoch": 7.890009606147935,
      "grad_norm": 1.5078125,
      "learning_rate": 9.851348481989966e-05,
      "loss": 4.6243,
      "mean_token_accuracy": 0.2653393417596817,
      "num_tokens": 188329678.0,
      "step": 82135
    },
    {
      "entropy": 5.6046826362609865,
      "epoch": 7.890489913544669,
      "grad_norm": 1.2890625,
      "learning_rate": 9.84922216401658e-05,
      "loss": 4.6922,
      "mean_token_accuracy": 0.25991230905056,
      "num_tokens": 188341798.0,
      "step": 82140
    },
    {
      "entropy": 5.460928869247437,
      "epoch": 7.890970220941402,
      "grad_norm": 1.4296875,
      "learning_rate": 9.847096255841839e-05,
      "loss": 4.5351,
      "mean_token_accuracy": 0.2738616779446602,
      "num_tokens": 188352375.0,
      "step": 82145
    },
    {
      "entropy": 5.4663208484649655,
      "epoch": 7.891450528338137,
      "grad_norm": 1.46875,
      "learning_rate": 9.844970757515106e-05,
      "loss": 4.5545,
      "mean_token_accuracy": 0.2694524645805359,
      "num_tokens": 188362762.0,
      "step": 82150
    },
    {
      "entropy": 5.464228391647339,
      "epoch": 7.89193083573487,
      "grad_norm": 1.5859375,
      "learning_rate": 9.842845669085729e-05,
      "loss": 4.6198,
      "mean_token_accuracy": 0.2656606063246727,
      "num_tokens": 188374392.0,
      "step": 82155
    },
    {
      "entropy": 5.452756023406982,
      "epoch": 7.892411143131604,
      "grad_norm": 1.3984375,
      "learning_rate": 9.840720990603042e-05,
      "loss": 4.5834,
      "mean_token_accuracy": 0.2718399852514267,
      "num_tokens": 188385460.0,
      "step": 82160
    },
    {
      "entropy": 5.550899982452393,
      "epoch": 7.892891450528338,
      "grad_norm": 1.3203125,
      "learning_rate": 9.838596722116385e-05,
      "loss": 4.6831,
      "mean_token_accuracy": 0.25840199142694475,
      "num_tokens": 188398162.0,
      "step": 82165
    },
    {
      "entropy": 5.522959661483765,
      "epoch": 7.893371757925072,
      "grad_norm": 1.3984375,
      "learning_rate": 9.836472863675061e-05,
      "loss": 4.659,
      "mean_token_accuracy": 0.27383606135845184,
      "num_tokens": 188409380.0,
      "step": 82170
    },
    {
      "entropy": 5.5218061923980715,
      "epoch": 7.893852065321806,
      "grad_norm": 1.46875,
      "learning_rate": 9.834349415328389e-05,
      "loss": 4.6368,
      "mean_token_accuracy": 0.26561654508113863,
      "num_tokens": 188420237.0,
      "step": 82175
    },
    {
      "entropy": 5.535035085678101,
      "epoch": 7.89433237271854,
      "grad_norm": 1.484375,
      "learning_rate": 9.832226377125664e-05,
      "loss": 4.6243,
      "mean_token_accuracy": 0.2667803630232811,
      "num_tokens": 188431189.0,
      "step": 82180
    },
    {
      "entropy": 5.547817993164062,
      "epoch": 7.894812680115274,
      "grad_norm": 1.3828125,
      "learning_rate": 9.830103749116182e-05,
      "loss": 4.7029,
      "mean_token_accuracy": 0.2637927979230881,
      "num_tokens": 188443425.0,
      "step": 82185
    },
    {
      "entropy": 5.500384664535522,
      "epoch": 7.895292987512008,
      "grad_norm": 1.75,
      "learning_rate": 9.827981531349221e-05,
      "loss": 4.5295,
      "mean_token_accuracy": 0.271441662311554,
      "num_tokens": 188454216.0,
      "step": 82190
    },
    {
      "entropy": 5.533672094345093,
      "epoch": 7.895773294908741,
      "grad_norm": 1.4375,
      "learning_rate": 9.825859723874062e-05,
      "loss": 4.6671,
      "mean_token_accuracy": 0.26463685035705564,
      "num_tokens": 188466270.0,
      "step": 82195
    },
    {
      "entropy": 5.576497650146484,
      "epoch": 7.8962536023054755,
      "grad_norm": 1.4375,
      "learning_rate": 9.823738326739951e-05,
      "loss": 4.6958,
      "mean_token_accuracy": 0.26382278054952624,
      "num_tokens": 188477467.0,
      "step": 82200
    },
    {
      "entropy": 5.508753871917724,
      "epoch": 7.89673390970221,
      "grad_norm": 1.5234375,
      "learning_rate": 9.82161733999615e-05,
      "loss": 4.5765,
      "mean_token_accuracy": 0.2740107148885727,
      "num_tokens": 188489337.0,
      "step": 82205
    },
    {
      "entropy": 5.479061603546143,
      "epoch": 7.897214217098943,
      "grad_norm": 1.2734375,
      "learning_rate": 9.819496763691898e-05,
      "loss": 4.5835,
      "mean_token_accuracy": 0.2688998028635979,
      "num_tokens": 188501371.0,
      "step": 82210
    },
    {
      "entropy": 5.495891427993774,
      "epoch": 7.8976945244956775,
      "grad_norm": 1.46875,
      "learning_rate": 9.817376597876431e-05,
      "loss": 4.5785,
      "mean_token_accuracy": 0.2696996062994003,
      "num_tokens": 188512819.0,
      "step": 82215
    },
    {
      "entropy": 5.565117359161377,
      "epoch": 7.898174831892411,
      "grad_norm": 1.4296875,
      "learning_rate": 9.815256842598974e-05,
      "loss": 4.6663,
      "mean_token_accuracy": 0.2641526535153389,
      "num_tokens": 188523859.0,
      "step": 82220
    },
    {
      "entropy": 5.509538412094116,
      "epoch": 7.898655139289145,
      "grad_norm": 1.3828125,
      "learning_rate": 9.813137497908747e-05,
      "loss": 4.6355,
      "mean_token_accuracy": 0.26656286269426344,
      "num_tokens": 188535385.0,
      "step": 82225
    },
    {
      "entropy": 5.498469066619873,
      "epoch": 7.899135446685879,
      "grad_norm": 1.3359375,
      "learning_rate": 9.811018563854942e-05,
      "loss": 4.598,
      "mean_token_accuracy": 0.26038877964019774,
      "num_tokens": 188545841.0,
      "step": 82230
    },
    {
      "entropy": 5.560430574417114,
      "epoch": 7.899615754082613,
      "grad_norm": 1.4609375,
      "learning_rate": 9.80890004048676e-05,
      "loss": 4.6151,
      "mean_token_accuracy": 0.2748557925224304,
      "num_tokens": 188556909.0,
      "step": 82235
    },
    {
      "entropy": 5.5223135471344,
      "epoch": 7.9000960614793465,
      "grad_norm": 1.2578125,
      "learning_rate": 9.806781927853387e-05,
      "loss": 4.6724,
      "mean_token_accuracy": 0.26772354543209076,
      "num_tokens": 188569941.0,
      "step": 82240
    },
    {
      "entropy": 5.540192461013794,
      "epoch": 7.900576368876081,
      "grad_norm": 1.328125,
      "learning_rate": 9.804664226004e-05,
      "loss": 4.6267,
      "mean_token_accuracy": 0.27109228521585466,
      "num_tokens": 188582784.0,
      "step": 82245
    },
    {
      "entropy": 5.454261827468872,
      "epoch": 7.901056676272814,
      "grad_norm": 1.3671875,
      "learning_rate": 9.802546934987763e-05,
      "loss": 4.5442,
      "mean_token_accuracy": 0.2782483994960785,
      "num_tokens": 188595085.0,
      "step": 82250
    },
    {
      "entropy": 5.52954888343811,
      "epoch": 7.901536983669549,
      "grad_norm": 1.46875,
      "learning_rate": 9.800430054853845e-05,
      "loss": 4.6082,
      "mean_token_accuracy": 0.27117965221405027,
      "num_tokens": 188605593.0,
      "step": 82255
    },
    {
      "entropy": 5.532173204421997,
      "epoch": 7.902017291066282,
      "grad_norm": 1.375,
      "learning_rate": 9.798313585651376e-05,
      "loss": 4.6178,
      "mean_token_accuracy": 0.2704765051603317,
      "num_tokens": 188617666.0,
      "step": 82260
    },
    {
      "entropy": 5.502093744277954,
      "epoch": 7.902497598463016,
      "grad_norm": 1.34375,
      "learning_rate": 9.796197527429505e-05,
      "loss": 4.6293,
      "mean_token_accuracy": 0.26438758224248887,
      "num_tokens": 188630268.0,
      "step": 82265
    },
    {
      "entropy": 5.516003179550171,
      "epoch": 7.90297790585975,
      "grad_norm": 1.34375,
      "learning_rate": 9.794081880237352e-05,
      "loss": 4.6049,
      "mean_token_accuracy": 0.2671751037240028,
      "num_tokens": 188642381.0,
      "step": 82270
    },
    {
      "entropy": 5.554571962356567,
      "epoch": 7.903458213256484,
      "grad_norm": 1.34375,
      "learning_rate": 9.791966644124047e-05,
      "loss": 4.6631,
      "mean_token_accuracy": 0.2644675984978676,
      "num_tokens": 188654914.0,
      "step": 82275
    },
    {
      "entropy": 5.536477184295654,
      "epoch": 7.9039385206532184,
      "grad_norm": 1.515625,
      "learning_rate": 9.789851819138693e-05,
      "loss": 4.6184,
      "mean_token_accuracy": 0.27140242606401443,
      "num_tokens": 188667298.0,
      "step": 82280
    },
    {
      "entropy": 5.484553384780884,
      "epoch": 7.904418828049952,
      "grad_norm": 1.3671875,
      "learning_rate": 9.7877374053304e-05,
      "loss": 4.5576,
      "mean_token_accuracy": 0.2716121211647987,
      "num_tokens": 188678507.0,
      "step": 82285
    },
    {
      "entropy": 5.506836557388306,
      "epoch": 7.904899135446686,
      "grad_norm": 1.375,
      "learning_rate": 9.785623402748239e-05,
      "loss": 4.6441,
      "mean_token_accuracy": 0.26863896399736403,
      "num_tokens": 188690371.0,
      "step": 82290
    },
    {
      "entropy": 5.466916608810425,
      "epoch": 7.90537944284342,
      "grad_norm": 1.484375,
      "learning_rate": 9.783509811441305e-05,
      "loss": 4.5738,
      "mean_token_accuracy": 0.27602519989013674,
      "num_tokens": 188702687.0,
      "step": 82295
    },
    {
      "entropy": 5.562116384506226,
      "epoch": 7.905859750240154,
      "grad_norm": 1.3359375,
      "learning_rate": 9.781396631458664e-05,
      "loss": 4.6282,
      "mean_token_accuracy": 0.2639226451516151,
      "num_tokens": 188714191.0,
      "step": 82300
    },
    {
      "entropy": 5.480126190185547,
      "epoch": 7.906340057636887,
      "grad_norm": 1.3515625,
      "learning_rate": 9.779283862849384e-05,
      "loss": 4.5378,
      "mean_token_accuracy": 0.2742892488837242,
      "num_tokens": 188725422.0,
      "step": 82305
    },
    {
      "entropy": 5.506340360641479,
      "epoch": 7.906820365033622,
      "grad_norm": 1.5625,
      "learning_rate": 9.777171505662507e-05,
      "loss": 4.5994,
      "mean_token_accuracy": 0.274086719751358,
      "num_tokens": 188737005.0,
      "step": 82310
    },
    {
      "entropy": 5.427354383468628,
      "epoch": 7.907300672430355,
      "grad_norm": 1.296875,
      "learning_rate": 9.775059559947091e-05,
      "loss": 4.4656,
      "mean_token_accuracy": 0.27739237546920775,
      "num_tokens": 188748598.0,
      "step": 82315
    },
    {
      "entropy": 5.36921534538269,
      "epoch": 7.9077809798270895,
      "grad_norm": 1.421875,
      "learning_rate": 9.772948025752153e-05,
      "loss": 4.4483,
      "mean_token_accuracy": 0.2858505010604858,
      "num_tokens": 188760160.0,
      "step": 82320
    },
    {
      "entropy": 5.432531881332397,
      "epoch": 7.908261287223823,
      "grad_norm": 1.5078125,
      "learning_rate": 9.770836903126725e-05,
      "loss": 4.5372,
      "mean_token_accuracy": 0.2791636049747467,
      "num_tokens": 188771157.0,
      "step": 82325
    },
    {
      "entropy": 5.529798793792724,
      "epoch": 7.908741594620557,
      "grad_norm": 1.484375,
      "learning_rate": 9.768726192119818e-05,
      "loss": 4.6137,
      "mean_token_accuracy": 0.2654179722070694,
      "num_tokens": 188782370.0,
      "step": 82330
    },
    {
      "entropy": 5.525809478759766,
      "epoch": 7.909221902017291,
      "grad_norm": 1.34375,
      "learning_rate": 9.766615892780437e-05,
      "loss": 4.629,
      "mean_token_accuracy": 0.26418224424123765,
      "num_tokens": 188793197.0,
      "step": 82335
    },
    {
      "entropy": 5.540893125534057,
      "epoch": 7.909702209414025,
      "grad_norm": 1.3515625,
      "learning_rate": 9.76450600515758e-05,
      "loss": 4.6166,
      "mean_token_accuracy": 0.2672824814915657,
      "num_tokens": 188804967.0,
      "step": 82340
    },
    {
      "entropy": 5.453608274459839,
      "epoch": 7.9101825168107585,
      "grad_norm": 1.3984375,
      "learning_rate": 9.762396529300235e-05,
      "loss": 4.5095,
      "mean_token_accuracy": 0.2777609437704086,
      "num_tokens": 188816666.0,
      "step": 82345
    },
    {
      "entropy": 5.482584285736084,
      "epoch": 7.910662824207493,
      "grad_norm": 1.6015625,
      "learning_rate": 9.760287465257364e-05,
      "loss": 4.5432,
      "mean_token_accuracy": 0.2708784356713295,
      "num_tokens": 188828264.0,
      "step": 82350
    },
    {
      "entropy": 5.537404870986938,
      "epoch": 7.911143131604227,
      "grad_norm": 1.484375,
      "learning_rate": 9.758178813077944e-05,
      "loss": 4.6183,
      "mean_token_accuracy": 0.2629910409450531,
      "num_tokens": 188839982.0,
      "step": 82355
    },
    {
      "entropy": 5.510191011428833,
      "epoch": 7.9116234390009605,
      "grad_norm": 1.421875,
      "learning_rate": 9.756070572810931e-05,
      "loss": 4.5577,
      "mean_token_accuracy": 0.2831081062555313,
      "num_tokens": 188851575.0,
      "step": 82360
    },
    {
      "entropy": 5.58344841003418,
      "epoch": 7.912103746397695,
      "grad_norm": 1.609375,
      "learning_rate": 9.753962744505268e-05,
      "loss": 4.6707,
      "mean_token_accuracy": 0.2640236601233482,
      "num_tokens": 188861768.0,
      "step": 82365
    },
    {
      "entropy": 5.481209087371826,
      "epoch": 7.912584053794428,
      "grad_norm": 1.6328125,
      "learning_rate": 9.751855328209897e-05,
      "loss": 4.5269,
      "mean_token_accuracy": 0.27811134308576585,
      "num_tokens": 188871656.0,
      "step": 82370
    },
    {
      "entropy": 5.576238059997559,
      "epoch": 7.913064361191163,
      "grad_norm": 1.34375,
      "learning_rate": 9.74974832397375e-05,
      "loss": 4.6366,
      "mean_token_accuracy": 0.27431033700704577,
      "num_tokens": 188883133.0,
      "step": 82375
    },
    {
      "entropy": 5.493227195739746,
      "epoch": 7.913544668587896,
      "grad_norm": 1.484375,
      "learning_rate": 9.747641731845735e-05,
      "loss": 4.5549,
      "mean_token_accuracy": 0.277620741724968,
      "num_tokens": 188894126.0,
      "step": 82380
    },
    {
      "entropy": 5.560912656784057,
      "epoch": 7.91402497598463,
      "grad_norm": 1.375,
      "learning_rate": 9.745535551874761e-05,
      "loss": 4.6487,
      "mean_token_accuracy": 0.26553214341402054,
      "num_tokens": 188905860.0,
      "step": 82385
    },
    {
      "entropy": 5.4603574752807615,
      "epoch": 7.914505283381364,
      "grad_norm": 1.2890625,
      "learning_rate": 9.743429784109734e-05,
      "loss": 4.544,
      "mean_token_accuracy": 0.2714642360806465,
      "num_tokens": 188917396.0,
      "step": 82390
    },
    {
      "entropy": 5.508357000350952,
      "epoch": 7.914985590778098,
      "grad_norm": 1.3203125,
      "learning_rate": 9.741324428599543e-05,
      "loss": 4.6228,
      "mean_token_accuracy": 0.26522613912820814,
      "num_tokens": 188928921.0,
      "step": 82395
    },
    {
      "entropy": 5.571076774597168,
      "epoch": 7.915465898174832,
      "grad_norm": 1.5703125,
      "learning_rate": 9.739219485393064e-05,
      "loss": 4.643,
      "mean_token_accuracy": 0.26466925591230395,
      "num_tokens": 188939711.0,
      "step": 82400
    },
    {
      "entropy": 5.487125730514526,
      "epoch": 7.915946205571566,
      "grad_norm": 1.4453125,
      "learning_rate": 9.737114954539175e-05,
      "loss": 4.5596,
      "mean_token_accuracy": 0.27450092434883117,
      "num_tokens": 188950316.0,
      "step": 82405
    },
    {
      "entropy": 5.47510256767273,
      "epoch": 7.916426512968299,
      "grad_norm": 1.3828125,
      "learning_rate": 9.735010836086727e-05,
      "loss": 4.5274,
      "mean_token_accuracy": 0.27872185558080675,
      "num_tokens": 188960889.0,
      "step": 82410
    },
    {
      "entropy": 5.656755495071411,
      "epoch": 7.916906820365034,
      "grad_norm": 1.359375,
      "learning_rate": 9.732907130084578e-05,
      "loss": 4.7204,
      "mean_token_accuracy": 0.2587967276573181,
      "num_tokens": 188972187.0,
      "step": 82415
    },
    {
      "entropy": 5.502543306350708,
      "epoch": 7.917387127761767,
      "grad_norm": 1.3515625,
      "learning_rate": 9.730803836581568e-05,
      "loss": 4.6092,
      "mean_token_accuracy": 0.27531781047582626,
      "num_tokens": 188983836.0,
      "step": 82420
    },
    {
      "entropy": 5.53965654373169,
      "epoch": 7.917867435158501,
      "grad_norm": 1.328125,
      "learning_rate": 9.728700955626526e-05,
      "loss": 4.6171,
      "mean_token_accuracy": 0.2649003744125366,
      "num_tokens": 188995715.0,
      "step": 82425
    },
    {
      "entropy": 5.545845794677734,
      "epoch": 7.918347742555236,
      "grad_norm": 1.4765625,
      "learning_rate": 9.726598487268281e-05,
      "loss": 4.6492,
      "mean_token_accuracy": 0.2710152745246887,
      "num_tokens": 189007633.0,
      "step": 82430
    },
    {
      "entropy": 5.50166883468628,
      "epoch": 7.918828049951969,
      "grad_norm": 1.5859375,
      "learning_rate": 9.724496431555646e-05,
      "loss": 4.6389,
      "mean_token_accuracy": 0.264043727517128,
      "num_tokens": 189017898.0,
      "step": 82435
    },
    {
      "entropy": 5.509866189956665,
      "epoch": 7.9193083573487035,
      "grad_norm": 1.5390625,
      "learning_rate": 9.72239478853742e-05,
      "loss": 4.6083,
      "mean_token_accuracy": 0.27035783380270006,
      "num_tokens": 189029137.0,
      "step": 82440
    },
    {
      "entropy": 5.551314640045166,
      "epoch": 7.919788664745437,
      "grad_norm": 1.390625,
      "learning_rate": 9.720293558262394e-05,
      "loss": 4.6561,
      "mean_token_accuracy": 0.2702859342098236,
      "num_tokens": 189040915.0,
      "step": 82445
    },
    {
      "entropy": 5.526090526580811,
      "epoch": 7.920268972142171,
      "grad_norm": 1.40625,
      "learning_rate": 9.718192740779358e-05,
      "loss": 4.5989,
      "mean_token_accuracy": 0.26510756462812424,
      "num_tokens": 189051303.0,
      "step": 82450
    },
    {
      "entropy": 5.573249387741089,
      "epoch": 7.920749279538905,
      "grad_norm": 1.3671875,
      "learning_rate": 9.716092336137084e-05,
      "loss": 4.7257,
      "mean_token_accuracy": 0.25815337598323823,
      "num_tokens": 189062545.0,
      "step": 82455
    },
    {
      "entropy": 5.542607831954956,
      "epoch": 7.921229586935639,
      "grad_norm": 1.2421875,
      "learning_rate": 9.713992344384341e-05,
      "loss": 4.5825,
      "mean_token_accuracy": 0.2701923385262489,
      "num_tokens": 189074640.0,
      "step": 82460
    },
    {
      "entropy": 5.53741865158081,
      "epoch": 7.9217098943323725,
      "grad_norm": 1.4296875,
      "learning_rate": 9.711892765569883e-05,
      "loss": 4.6191,
      "mean_token_accuracy": 0.2703919306397438,
      "num_tokens": 189086016.0,
      "step": 82465
    },
    {
      "entropy": 5.452317476272583,
      "epoch": 7.922190201729107,
      "grad_norm": 1.46875,
      "learning_rate": 9.70979359974245e-05,
      "loss": 4.5402,
      "mean_token_accuracy": 0.28064998984336853,
      "num_tokens": 189095596.0,
      "step": 82470
    },
    {
      "entropy": 5.626569128036499,
      "epoch": 7.92267050912584,
      "grad_norm": 1.5,
      "learning_rate": 9.707694846950796e-05,
      "loss": 4.6428,
      "mean_token_accuracy": 0.26030689626932146,
      "num_tokens": 189106839.0,
      "step": 82475
    },
    {
      "entropy": 5.580554008483887,
      "epoch": 7.923150816522575,
      "grad_norm": 1.484375,
      "learning_rate": 9.705596507243626e-05,
      "loss": 4.6447,
      "mean_token_accuracy": 0.26172816902399065,
      "num_tokens": 189117813.0,
      "step": 82480
    },
    {
      "entropy": 5.5096324443817135,
      "epoch": 7.923631123919308,
      "grad_norm": 1.421875,
      "learning_rate": 9.703498580669667e-05,
      "loss": 4.5602,
      "mean_token_accuracy": 0.27255226075649264,
      "num_tokens": 189128912.0,
      "step": 82485
    },
    {
      "entropy": 5.5646219730377195,
      "epoch": 7.924111431316042,
      "grad_norm": 1.4609375,
      "learning_rate": 9.701401067277626e-05,
      "loss": 4.6794,
      "mean_token_accuracy": 0.26438482403755187,
      "num_tokens": 189141067.0,
      "step": 82490
    },
    {
      "entropy": 5.47529559135437,
      "epoch": 7.924591738712776,
      "grad_norm": 1.4453125,
      "learning_rate": 9.699303967116202e-05,
      "loss": 4.5948,
      "mean_token_accuracy": 0.27307233661413194,
      "num_tokens": 189151901.0,
      "step": 82495
    },
    {
      "entropy": 5.53689169883728,
      "epoch": 7.92507204610951,
      "grad_norm": 1.421875,
      "learning_rate": 9.697207280234082e-05,
      "loss": 4.593,
      "mean_token_accuracy": 0.26870758533477784,
      "num_tokens": 189162509.0,
      "step": 82500
    },
    {
      "entropy": 5.606018924713135,
      "epoch": 7.925552353506244,
      "grad_norm": 1.4609375,
      "learning_rate": 9.695111006679947e-05,
      "loss": 4.7144,
      "mean_token_accuracy": 0.2590272814035416,
      "num_tokens": 189174216.0,
      "step": 82505
    },
    {
      "entropy": 5.56426739692688,
      "epoch": 7.926032660902978,
      "grad_norm": 1.5703125,
      "learning_rate": 9.693015146502463e-05,
      "loss": 4.6346,
      "mean_token_accuracy": 0.26080443412065507,
      "num_tokens": 189184461.0,
      "step": 82510
    },
    {
      "entropy": 5.551883172988892,
      "epoch": 7.926512968299712,
      "grad_norm": 1.484375,
      "learning_rate": 9.690919699750298e-05,
      "loss": 4.6314,
      "mean_token_accuracy": 0.25887503772974013,
      "num_tokens": 189195434.0,
      "step": 82515
    },
    {
      "entropy": 5.554142808914184,
      "epoch": 7.926993275696446,
      "grad_norm": 1.375,
      "learning_rate": 9.688824666472087e-05,
      "loss": 4.575,
      "mean_token_accuracy": 0.272611166536808,
      "num_tokens": 189206366.0,
      "step": 82520
    },
    {
      "entropy": 5.539286518096924,
      "epoch": 7.92747358309318,
      "grad_norm": 1.4140625,
      "learning_rate": 9.686730046716483e-05,
      "loss": 4.5964,
      "mean_token_accuracy": 0.25992528945207594,
      "num_tokens": 189216953.0,
      "step": 82525
    },
    {
      "entropy": 5.494785928726197,
      "epoch": 7.927953890489913,
      "grad_norm": 1.5625,
      "learning_rate": 9.68463584053211e-05,
      "loss": 4.5971,
      "mean_token_accuracy": 0.27319387048482896,
      "num_tokens": 189229673.0,
      "step": 82530
    },
    {
      "entropy": 5.435425853729248,
      "epoch": 7.928434197886648,
      "grad_norm": 1.421875,
      "learning_rate": 9.682542047967593e-05,
      "loss": 4.5534,
      "mean_token_accuracy": 0.27289972007274627,
      "num_tokens": 189240650.0,
      "step": 82535
    },
    {
      "entropy": 5.560023498535156,
      "epoch": 7.928914505283381,
      "grad_norm": 1.4375,
      "learning_rate": 9.680448669071543e-05,
      "loss": 4.6641,
      "mean_token_accuracy": 0.26655333787202834,
      "num_tokens": 189251058.0,
      "step": 82540
    },
    {
      "entropy": 5.564132833480835,
      "epoch": 7.9293948126801155,
      "grad_norm": 1.3671875,
      "learning_rate": 9.678355703892562e-05,
      "loss": 4.6494,
      "mean_token_accuracy": 0.2624301970005035,
      "num_tokens": 189262824.0,
      "step": 82545
    },
    {
      "entropy": 5.50698561668396,
      "epoch": 7.929875120076849,
      "grad_norm": 1.4375,
      "learning_rate": 9.676263152479241e-05,
      "loss": 4.588,
      "mean_token_accuracy": 0.2702495023608208,
      "num_tokens": 189274556.0,
      "step": 82550
    },
    {
      "entropy": 5.50175929069519,
      "epoch": 7.930355427473583,
      "grad_norm": 1.40625,
      "learning_rate": 9.674171014880173e-05,
      "loss": 4.5975,
      "mean_token_accuracy": 0.270205020904541,
      "num_tokens": 189286623.0,
      "step": 82555
    },
    {
      "entropy": 5.5276096820831295,
      "epoch": 7.930835734870317,
      "grad_norm": 1.5859375,
      "learning_rate": 9.672079291143913e-05,
      "loss": 4.6379,
      "mean_token_accuracy": 0.2711223572492599,
      "num_tokens": 189297861.0,
      "step": 82560
    },
    {
      "entropy": 5.488161182403564,
      "epoch": 7.931316042267051,
      "grad_norm": 1.4453125,
      "learning_rate": 9.669987981319034e-05,
      "loss": 4.5769,
      "mean_token_accuracy": 0.2717899054288864,
      "num_tokens": 189309151.0,
      "step": 82565
    },
    {
      "entropy": 5.418610525131226,
      "epoch": 7.931796349663784,
      "grad_norm": 1.46875,
      "learning_rate": 9.667897085454091e-05,
      "loss": 4.4561,
      "mean_token_accuracy": 0.27894211411476133,
      "num_tokens": 189320814.0,
      "step": 82570
    },
    {
      "entropy": 5.614761590957642,
      "epoch": 7.932276657060519,
      "grad_norm": 1.4296875,
      "learning_rate": 9.665806603597626e-05,
      "loss": 4.705,
      "mean_token_accuracy": 0.26120457947254183,
      "num_tokens": 189331508.0,
      "step": 82575
    },
    {
      "entropy": 5.568064594268799,
      "epoch": 7.932756964457253,
      "grad_norm": 1.4375,
      "learning_rate": 9.663716535798176e-05,
      "loss": 4.6487,
      "mean_token_accuracy": 0.26938346773386,
      "num_tokens": 189342623.0,
      "step": 82580
    },
    {
      "entropy": 5.468441915512085,
      "epoch": 7.9332372718539865,
      "grad_norm": 1.3046875,
      "learning_rate": 9.661626882104264e-05,
      "loss": 4.4676,
      "mean_token_accuracy": 0.2805987775325775,
      "num_tokens": 189353746.0,
      "step": 82585
    },
    {
      "entropy": 5.557177925109864,
      "epoch": 7.93371757925072,
      "grad_norm": 1.359375,
      "learning_rate": 9.659537642564415e-05,
      "loss": 4.6155,
      "mean_token_accuracy": 0.2690621197223663,
      "num_tokens": 189365614.0,
      "step": 82590
    },
    {
      "entropy": 5.499996042251587,
      "epoch": 7.934197886647454,
      "grad_norm": 1.3515625,
      "learning_rate": 9.657448817227118e-05,
      "loss": 4.5736,
      "mean_token_accuracy": 0.280072346329689,
      "num_tokens": 189377880.0,
      "step": 82595
    },
    {
      "entropy": 5.4474499225616455,
      "epoch": 7.934678194044189,
      "grad_norm": 1.4375,
      "learning_rate": 9.65536040614088e-05,
      "loss": 4.5174,
      "mean_token_accuracy": 0.271109813451767,
      "num_tokens": 189390165.0,
      "step": 82600
    },
    {
      "entropy": 5.477090978622437,
      "epoch": 7.935158501440922,
      "grad_norm": 1.515625,
      "learning_rate": 9.653272409354184e-05,
      "loss": 4.5721,
      "mean_token_accuracy": 0.2745537102222443,
      "num_tokens": 189401470.0,
      "step": 82605
    },
    {
      "entropy": 5.542542839050293,
      "epoch": 7.935638808837656,
      "grad_norm": 1.515625,
      "learning_rate": 9.65118482691551e-05,
      "loss": 4.6047,
      "mean_token_accuracy": 0.26744936406612396,
      "num_tokens": 189413206.0,
      "step": 82610
    },
    {
      "entropy": 5.548848295211792,
      "epoch": 7.93611911623439,
      "grad_norm": 1.53125,
      "learning_rate": 9.649097658873324e-05,
      "loss": 4.6273,
      "mean_token_accuracy": 0.2672612100839615,
      "num_tokens": 189423735.0,
      "step": 82615
    },
    {
      "entropy": 5.54577317237854,
      "epoch": 7.936599423631124,
      "grad_norm": 1.359375,
      "learning_rate": 9.647010905276091e-05,
      "loss": 4.5912,
      "mean_token_accuracy": 0.2752802580595016,
      "num_tokens": 189434363.0,
      "step": 82620
    },
    {
      "entropy": 5.475636005401611,
      "epoch": 7.937079731027858,
      "grad_norm": 1.515625,
      "learning_rate": 9.644924566172245e-05,
      "loss": 4.4999,
      "mean_token_accuracy": 0.2767212837934494,
      "num_tokens": 189445950.0,
      "step": 82625
    },
    {
      "entropy": 5.433478593826294,
      "epoch": 7.937560038424592,
      "grad_norm": 1.28125,
      "learning_rate": 9.64283864161023e-05,
      "loss": 4.5231,
      "mean_token_accuracy": 0.27704363167285917,
      "num_tokens": 189457775.0,
      "step": 82630
    },
    {
      "entropy": 5.5523059368133545,
      "epoch": 7.938040345821325,
      "grad_norm": 1.4140625,
      "learning_rate": 9.640753131638478e-05,
      "loss": 4.64,
      "mean_token_accuracy": 0.26734236180782317,
      "num_tokens": 189469930.0,
      "step": 82635
    },
    {
      "entropy": 5.5112096786499025,
      "epoch": 7.93852065321806,
      "grad_norm": 1.5390625,
      "learning_rate": 9.638668036305411e-05,
      "loss": 4.5847,
      "mean_token_accuracy": 0.2705824002623558,
      "num_tokens": 189481230.0,
      "step": 82640
    },
    {
      "entropy": 5.509421491622925,
      "epoch": 7.939000960614793,
      "grad_norm": 1.3046875,
      "learning_rate": 9.636583355659432e-05,
      "loss": 4.6124,
      "mean_token_accuracy": 0.27239805459976196,
      "num_tokens": 189494201.0,
      "step": 82645
    },
    {
      "entropy": 5.5009886741638185,
      "epoch": 7.939481268011527,
      "grad_norm": 1.28125,
      "learning_rate": 9.634499089748949e-05,
      "loss": 4.5746,
      "mean_token_accuracy": 0.27127144634723666,
      "num_tokens": 189505980.0,
      "step": 82650
    },
    {
      "entropy": 5.496173906326294,
      "epoch": 7.939961575408262,
      "grad_norm": 1.4765625,
      "learning_rate": 9.632415238622341e-05,
      "loss": 4.6774,
      "mean_token_accuracy": 0.2623604714870453,
      "num_tokens": 189517894.0,
      "step": 82655
    },
    {
      "entropy": 5.502114105224609,
      "epoch": 7.940441882804995,
      "grad_norm": 1.46875,
      "learning_rate": 9.630331802328e-05,
      "loss": 4.5068,
      "mean_token_accuracy": 0.2700021341443062,
      "num_tokens": 189529961.0,
      "step": 82660
    },
    {
      "entropy": 5.495048761367798,
      "epoch": 7.940922190201729,
      "grad_norm": 1.4765625,
      "learning_rate": 9.628248780914288e-05,
      "loss": 4.5704,
      "mean_token_accuracy": 0.27620292007923125,
      "num_tokens": 189540425.0,
      "step": 82665
    },
    {
      "entropy": 5.579979658126831,
      "epoch": 7.941402497598463,
      "grad_norm": 1.5390625,
      "learning_rate": 9.626166174429572e-05,
      "loss": 4.6293,
      "mean_token_accuracy": 0.26477812230587006,
      "num_tokens": 189552769.0,
      "step": 82670
    },
    {
      "entropy": 5.4966285705566404,
      "epoch": 7.941882804995197,
      "grad_norm": 1.4140625,
      "learning_rate": 9.624083982922205e-05,
      "loss": 4.5714,
      "mean_token_accuracy": 0.27030584365129473,
      "num_tokens": 189564359.0,
      "step": 82675
    },
    {
      "entropy": 5.517149639129639,
      "epoch": 7.942363112391931,
      "grad_norm": 1.4140625,
      "learning_rate": 9.622002206440533e-05,
      "loss": 4.593,
      "mean_token_accuracy": 0.2751706540584564,
      "num_tokens": 189576335.0,
      "step": 82680
    },
    {
      "entropy": 5.531369590759278,
      "epoch": 7.942843419788665,
      "grad_norm": 1.5703125,
      "learning_rate": 9.619920845032875e-05,
      "loss": 4.6096,
      "mean_token_accuracy": 0.27060900926589965,
      "num_tokens": 189587467.0,
      "step": 82685
    },
    {
      "entropy": 5.483508586883545,
      "epoch": 7.9433237271853985,
      "grad_norm": 1.46875,
      "learning_rate": 9.617839898747563e-05,
      "loss": 4.5022,
      "mean_token_accuracy": 0.2815698355436325,
      "num_tokens": 189597866.0,
      "step": 82690
    },
    {
      "entropy": 5.514473533630371,
      "epoch": 7.943804034582133,
      "grad_norm": 1.4296875,
      "learning_rate": 9.61575936763291e-05,
      "loss": 4.6418,
      "mean_token_accuracy": 0.2714236840605736,
      "num_tokens": 189610312.0,
      "step": 82695
    },
    {
      "entropy": 5.534240293502807,
      "epoch": 7.944284341978866,
      "grad_norm": 1.5,
      "learning_rate": 9.613679251737218e-05,
      "loss": 4.5893,
      "mean_token_accuracy": 0.2688780978322029,
      "num_tokens": 189619777.0,
      "step": 82700
    },
    {
      "entropy": 5.659024858474732,
      "epoch": 7.944764649375601,
      "grad_norm": 1.5,
      "learning_rate": 9.611599551108785e-05,
      "loss": 4.7748,
      "mean_token_accuracy": 0.26070494651794435,
      "num_tokens": 189631054.0,
      "step": 82705
    },
    {
      "entropy": 5.435055732727051,
      "epoch": 7.945244956772334,
      "grad_norm": 1.4765625,
      "learning_rate": 9.609520265795896e-05,
      "loss": 4.5152,
      "mean_token_accuracy": 0.2752004384994507,
      "num_tokens": 189641984.0,
      "step": 82710
    },
    {
      "entropy": 5.516033458709717,
      "epoch": 7.945725264169068,
      "grad_norm": 1.3828125,
      "learning_rate": 9.607441395846816e-05,
      "loss": 4.5768,
      "mean_token_accuracy": 0.2730414107441902,
      "num_tokens": 189653572.0,
      "step": 82715
    },
    {
      "entropy": 5.51661057472229,
      "epoch": 7.946205571565802,
      "grad_norm": 1.4140625,
      "learning_rate": 9.60536294130982e-05,
      "loss": 4.6213,
      "mean_token_accuracy": 0.2608450651168823,
      "num_tokens": 189665268.0,
      "step": 82720
    },
    {
      "entropy": 5.421278047561645,
      "epoch": 7.946685878962536,
      "grad_norm": 1.5546875,
      "learning_rate": 9.603284902233158e-05,
      "loss": 4.4592,
      "mean_token_accuracy": 0.2789406135678291,
      "num_tokens": 189676441.0,
      "step": 82725
    },
    {
      "entropy": 5.536655569076538,
      "epoch": 7.94716618635927,
      "grad_norm": 1.515625,
      "learning_rate": 9.601207278665077e-05,
      "loss": 4.6115,
      "mean_token_accuracy": 0.2665073424577713,
      "num_tokens": 189687365.0,
      "step": 82730
    },
    {
      "entropy": 5.427224922180176,
      "epoch": 7.947646493756004,
      "grad_norm": 1.4453125,
      "learning_rate": 9.59913007065382e-05,
      "loss": 4.5158,
      "mean_token_accuracy": 0.2718413844704628,
      "num_tokens": 189698517.0,
      "step": 82735
    },
    {
      "entropy": 5.550660228729248,
      "epoch": 7.948126801152737,
      "grad_norm": 1.703125,
      "learning_rate": 9.597053278247609e-05,
      "loss": 4.73,
      "mean_token_accuracy": 0.25924246907234194,
      "num_tokens": 189711863.0,
      "step": 82740
    },
    {
      "entropy": 5.478043413162231,
      "epoch": 7.948607108549472,
      "grad_norm": 1.34375,
      "learning_rate": 9.594976901494657e-05,
      "loss": 4.5749,
      "mean_token_accuracy": 0.2754921823740005,
      "num_tokens": 189723780.0,
      "step": 82745
    },
    {
      "entropy": 5.533266162872314,
      "epoch": 7.949087415946206,
      "grad_norm": 1.359375,
      "learning_rate": 9.592900940443172e-05,
      "loss": 4.6053,
      "mean_token_accuracy": 0.2638883888721466,
      "num_tokens": 189735309.0,
      "step": 82750
    },
    {
      "entropy": 5.566657161712646,
      "epoch": 7.949567723342939,
      "grad_norm": 1.4140625,
      "learning_rate": 9.590825395141355e-05,
      "loss": 4.6982,
      "mean_token_accuracy": 0.266585636138916,
      "num_tokens": 189747433.0,
      "step": 82755
    },
    {
      "entropy": 5.501752853393555,
      "epoch": 7.950048030739674,
      "grad_norm": 1.4296875,
      "learning_rate": 9.588750265637392e-05,
      "loss": 4.5586,
      "mean_token_accuracy": 0.26964863687753676,
      "num_tokens": 189758433.0,
      "step": 82760
    },
    {
      "entropy": 5.537489938735962,
      "epoch": 7.950528338136407,
      "grad_norm": 1.4921875,
      "learning_rate": 9.586675551979463e-05,
      "loss": 4.5976,
      "mean_token_accuracy": 0.2656642809510231,
      "num_tokens": 189768969.0,
      "step": 82765
    },
    {
      "entropy": 5.48874077796936,
      "epoch": 7.9510086455331415,
      "grad_norm": 1.4296875,
      "learning_rate": 9.584601254215743e-05,
      "loss": 4.533,
      "mean_token_accuracy": 0.27273771315813067,
      "num_tokens": 189779356.0,
      "step": 82770
    },
    {
      "entropy": 5.510634088516236,
      "epoch": 7.951488952929875,
      "grad_norm": 1.4296875,
      "learning_rate": 9.582527372394375e-05,
      "loss": 4.6392,
      "mean_token_accuracy": 0.2653057977557182,
      "num_tokens": 189790362.0,
      "step": 82775
    },
    {
      "entropy": 5.503455114364624,
      "epoch": 7.951969260326609,
      "grad_norm": 1.46875,
      "learning_rate": 9.580453906563522e-05,
      "loss": 4.5916,
      "mean_token_accuracy": 0.2750804886221886,
      "num_tokens": 189800781.0,
      "step": 82780
    },
    {
      "entropy": 5.461524772644043,
      "epoch": 7.952449567723343,
      "grad_norm": 1.34375,
      "learning_rate": 9.578380856771315e-05,
      "loss": 4.5803,
      "mean_token_accuracy": 0.26621070355176923,
      "num_tokens": 189812771.0,
      "step": 82785
    },
    {
      "entropy": 5.500088500976562,
      "epoch": 7.952929875120077,
      "grad_norm": 1.3984375,
      "learning_rate": 9.576308223065894e-05,
      "loss": 4.5123,
      "mean_token_accuracy": 0.278148877620697,
      "num_tokens": 189823855.0,
      "step": 82790
    },
    {
      "entropy": 5.5228550910949705,
      "epoch": 7.95341018251681,
      "grad_norm": 1.5078125,
      "learning_rate": 9.574236005495369e-05,
      "loss": 4.6305,
      "mean_token_accuracy": 0.2622277557849884,
      "num_tokens": 189835051.0,
      "step": 82795
    },
    {
      "entropy": 5.4817736625671385,
      "epoch": 7.953890489913545,
      "grad_norm": 1.578125,
      "learning_rate": 9.572164204107863e-05,
      "loss": 4.5873,
      "mean_token_accuracy": 0.28055463433265687,
      "num_tokens": 189846063.0,
      "step": 82800
    },
    {
      "entropy": 5.533479690551758,
      "epoch": 7.954370797310279,
      "grad_norm": 1.546875,
      "learning_rate": 9.570092818951462e-05,
      "loss": 4.6698,
      "mean_token_accuracy": 0.26386995017528536,
      "num_tokens": 189857440.0,
      "step": 82805
    },
    {
      "entropy": 5.507966184616089,
      "epoch": 7.9548511047070125,
      "grad_norm": 1.3125,
      "learning_rate": 9.568021850074265e-05,
      "loss": 4.6405,
      "mean_token_accuracy": 0.2701869323849678,
      "num_tokens": 189870660.0,
      "step": 82810
    },
    {
      "entropy": 5.504028940200806,
      "epoch": 7.955331412103746,
      "grad_norm": 1.4375,
      "learning_rate": 9.565951297524355e-05,
      "loss": 4.5884,
      "mean_token_accuracy": 0.2768410384654999,
      "num_tokens": 189882430.0,
      "step": 82815
    },
    {
      "entropy": 5.547930955886841,
      "epoch": 7.95581171950048,
      "grad_norm": 1.4296875,
      "learning_rate": 9.563881161349803e-05,
      "loss": 4.5881,
      "mean_token_accuracy": 0.2675489917397499,
      "num_tokens": 189893222.0,
      "step": 82820
    },
    {
      "entropy": 5.559457683563233,
      "epoch": 7.956292026897215,
      "grad_norm": 1.4453125,
      "learning_rate": 9.561811441598669e-05,
      "loss": 4.659,
      "mean_token_accuracy": 0.2630183443427086,
      "num_tokens": 189904257.0,
      "step": 82825
    },
    {
      "entropy": 5.563534164428711,
      "epoch": 7.956772334293948,
      "grad_norm": 1.4375,
      "learning_rate": 9.559742138319015e-05,
      "loss": 4.6278,
      "mean_token_accuracy": 0.268134143948555,
      "num_tokens": 189916054.0,
      "step": 82830
    },
    {
      "entropy": 5.469750881195068,
      "epoch": 7.957252641690682,
      "grad_norm": 1.34375,
      "learning_rate": 9.557673251558871e-05,
      "loss": 4.5239,
      "mean_token_accuracy": 0.274444916844368,
      "num_tokens": 189926946.0,
      "step": 82835
    },
    {
      "entropy": 5.55845365524292,
      "epoch": 7.957732949087416,
      "grad_norm": 1.390625,
      "learning_rate": 9.555604781366276e-05,
      "loss": 4.6832,
      "mean_token_accuracy": 0.26358333975076675,
      "num_tokens": 189938389.0,
      "step": 82840
    },
    {
      "entropy": 5.576261377334594,
      "epoch": 7.95821325648415,
      "grad_norm": 1.4765625,
      "learning_rate": 9.553536727789253e-05,
      "loss": 4.6768,
      "mean_token_accuracy": 0.25794200748205187,
      "num_tokens": 189950532.0,
      "step": 82845
    },
    {
      "entropy": 5.588772535324097,
      "epoch": 7.958693563880884,
      "grad_norm": 1.4921875,
      "learning_rate": 9.551469090875819e-05,
      "loss": 4.6673,
      "mean_token_accuracy": 0.25881763845682143,
      "num_tokens": 189961437.0,
      "step": 82850
    },
    {
      "entropy": 5.520328140258789,
      "epoch": 7.959173871277618,
      "grad_norm": 1.4765625,
      "learning_rate": 9.549401870673973e-05,
      "loss": 4.6135,
      "mean_token_accuracy": 0.26779684275388715,
      "num_tokens": 189973766.0,
      "step": 82855
    },
    {
      "entropy": 5.436557722091675,
      "epoch": 7.959654178674351,
      "grad_norm": 1.46875,
      "learning_rate": 9.547335067231726e-05,
      "loss": 4.5273,
      "mean_token_accuracy": 0.27896171510219575,
      "num_tokens": 189984707.0,
      "step": 82860
    },
    {
      "entropy": 5.596513986587524,
      "epoch": 7.960134486071086,
      "grad_norm": 1.578125,
      "learning_rate": 9.545268680597038e-05,
      "loss": 4.6427,
      "mean_token_accuracy": 0.2654342234134674,
      "num_tokens": 189995968.0,
      "step": 82865
    },
    {
      "entropy": 5.557560873031616,
      "epoch": 7.960614793467819,
      "grad_norm": 1.46875,
      "learning_rate": 9.5432027108179e-05,
      "loss": 4.6524,
      "mean_token_accuracy": 0.26262136548757553,
      "num_tokens": 190007097.0,
      "step": 82870
    },
    {
      "entropy": 5.499539995193482,
      "epoch": 7.961095100864553,
      "grad_norm": 1.4375,
      "learning_rate": 9.541137157942273e-05,
      "loss": 4.5459,
      "mean_token_accuracy": 0.273965460062027,
      "num_tokens": 190017717.0,
      "step": 82875
    },
    {
      "entropy": 5.554472827911377,
      "epoch": 7.961575408261287,
      "grad_norm": 1.453125,
      "learning_rate": 9.539072022018114e-05,
      "loss": 4.6156,
      "mean_token_accuracy": 0.26872392296791076,
      "num_tokens": 190028706.0,
      "step": 82880
    },
    {
      "entropy": 5.506856203079224,
      "epoch": 7.962055715658021,
      "grad_norm": 1.4765625,
      "learning_rate": 9.537007303093373e-05,
      "loss": 4.6339,
      "mean_token_accuracy": 0.2642479196190834,
      "num_tokens": 190039781.0,
      "step": 82885
    },
    {
      "entropy": 5.565738487243652,
      "epoch": 7.962536023054755,
      "grad_norm": 1.28125,
      "learning_rate": 9.534943001215984e-05,
      "loss": 4.6811,
      "mean_token_accuracy": 0.26412405967712405,
      "num_tokens": 190050500.0,
      "step": 82890
    },
    {
      "entropy": 5.502006769180298,
      "epoch": 7.963016330451489,
      "grad_norm": 1.484375,
      "learning_rate": 9.53287911643387e-05,
      "loss": 4.5999,
      "mean_token_accuracy": 0.2698483467102051,
      "num_tokens": 190062107.0,
      "step": 82895
    },
    {
      "entropy": 5.464374589920044,
      "epoch": 7.963496637848223,
      "grad_norm": 1.5078125,
      "learning_rate": 9.530815648794953e-05,
      "loss": 4.5402,
      "mean_token_accuracy": 0.2680538445711136,
      "num_tokens": 190073453.0,
      "step": 82900
    },
    {
      "entropy": 5.542019271850586,
      "epoch": 7.963976945244957,
      "grad_norm": 1.5625,
      "learning_rate": 9.528752598347135e-05,
      "loss": 4.6689,
      "mean_token_accuracy": 0.2568853199481964,
      "num_tokens": 190083452.0,
      "step": 82905
    },
    {
      "entropy": 5.5204390525817875,
      "epoch": 7.964457252641691,
      "grad_norm": 1.5,
      "learning_rate": 9.526689965138324e-05,
      "loss": 4.5547,
      "mean_token_accuracy": 0.270270599424839,
      "num_tokens": 190095733.0,
      "step": 82910
    },
    {
      "entropy": 5.514967060089111,
      "epoch": 7.9649375600384245,
      "grad_norm": 1.40625,
      "learning_rate": 9.524627749216399e-05,
      "loss": 4.6428,
      "mean_token_accuracy": 0.25951665043830874,
      "num_tokens": 190106784.0,
      "step": 82915
    },
    {
      "entropy": 5.49053258895874,
      "epoch": 7.965417867435159,
      "grad_norm": 1.390625,
      "learning_rate": 9.522565950629244e-05,
      "loss": 4.5262,
      "mean_token_accuracy": 0.2754605144262314,
      "num_tokens": 190118552.0,
      "step": 82920
    },
    {
      "entropy": 5.555972957611084,
      "epoch": 7.965898174831892,
      "grad_norm": 1.421875,
      "learning_rate": 9.520504569424726e-05,
      "loss": 4.6567,
      "mean_token_accuracy": 0.26207291930913923,
      "num_tokens": 190129465.0,
      "step": 82925
    },
    {
      "entropy": 5.526697969436645,
      "epoch": 7.9663784822286265,
      "grad_norm": 1.609375,
      "learning_rate": 9.518443605650701e-05,
      "loss": 4.603,
      "mean_token_accuracy": 0.2681398972868919,
      "num_tokens": 190141543.0,
      "step": 82930
    },
    {
      "entropy": 5.492100620269776,
      "epoch": 7.96685878962536,
      "grad_norm": 1.453125,
      "learning_rate": 9.516383059355021e-05,
      "loss": 4.5973,
      "mean_token_accuracy": 0.270967124402523,
      "num_tokens": 190152658.0,
      "step": 82935
    },
    {
      "entropy": 5.562976598739624,
      "epoch": 7.967339097022094,
      "grad_norm": 1.4140625,
      "learning_rate": 9.514322930585525e-05,
      "loss": 4.6382,
      "mean_token_accuracy": 0.262666417658329,
      "num_tokens": 190164213.0,
      "step": 82940
    },
    {
      "entropy": 5.451498317718506,
      "epoch": 7.967819404418828,
      "grad_norm": 1.453125,
      "learning_rate": 9.512263219390045e-05,
      "loss": 4.4822,
      "mean_token_accuracy": 0.27627420127391816,
      "num_tokens": 190175667.0,
      "step": 82945
    },
    {
      "entropy": 5.532528686523437,
      "epoch": 7.968299711815562,
      "grad_norm": 1.4453125,
      "learning_rate": 9.510203925816407e-05,
      "loss": 4.5984,
      "mean_token_accuracy": 0.2702605277299881,
      "num_tokens": 190187298.0,
      "step": 82950
    },
    {
      "entropy": 5.465533351898193,
      "epoch": 7.9687800192122955,
      "grad_norm": 1.3828125,
      "learning_rate": 9.50814504991241e-05,
      "loss": 4.5776,
      "mean_token_accuracy": 0.2729266732931137,
      "num_tokens": 190198759.0,
      "step": 82955
    },
    {
      "entropy": 5.521111345291137,
      "epoch": 7.96926032660903,
      "grad_norm": 1.578125,
      "learning_rate": 9.506086591725858e-05,
      "loss": 4.5941,
      "mean_token_accuracy": 0.2719318151473999,
      "num_tokens": 190210054.0,
      "step": 82960
    },
    {
      "entropy": 5.502058506011963,
      "epoch": 7.969740634005763,
      "grad_norm": 1.3125,
      "learning_rate": 9.504028551304547e-05,
      "loss": 4.5352,
      "mean_token_accuracy": 0.27590700685977937,
      "num_tokens": 190221835.0,
      "step": 82965
    },
    {
      "entropy": 5.529926061630249,
      "epoch": 7.970220941402498,
      "grad_norm": 1.296875,
      "learning_rate": 9.501970928696253e-05,
      "loss": 4.6138,
      "mean_token_accuracy": 0.27235486954450605,
      "num_tokens": 190233332.0,
      "step": 82970
    },
    {
      "entropy": 5.513177156448364,
      "epoch": 7.970701248799232,
      "grad_norm": 1.4453125,
      "learning_rate": 9.499913723948752e-05,
      "loss": 4.5528,
      "mean_token_accuracy": 0.2705823078751564,
      "num_tokens": 190244644.0,
      "step": 82975
    },
    {
      "entropy": 5.528625726699829,
      "epoch": 7.971181556195965,
      "grad_norm": 1.3828125,
      "learning_rate": 9.49785693710981e-05,
      "loss": 4.6099,
      "mean_token_accuracy": 0.261967396736145,
      "num_tokens": 190256964.0,
      "step": 82980
    },
    {
      "entropy": 5.533294057846069,
      "epoch": 7.9716618635927,
      "grad_norm": 1.3828125,
      "learning_rate": 9.495800568227173e-05,
      "loss": 4.5938,
      "mean_token_accuracy": 0.27245792001485825,
      "num_tokens": 190268963.0,
      "step": 82985
    },
    {
      "entropy": 5.5312987804412845,
      "epoch": 7.972142170989433,
      "grad_norm": 1.5,
      "learning_rate": 9.493744617348581e-05,
      "loss": 4.6297,
      "mean_token_accuracy": 0.26118639707565305,
      "num_tokens": 190280571.0,
      "step": 82990
    },
    {
      "entropy": 5.571421766281128,
      "epoch": 7.972622478386167,
      "grad_norm": 1.546875,
      "learning_rate": 9.491689084521771e-05,
      "loss": 4.6577,
      "mean_token_accuracy": 0.2627340734004974,
      "num_tokens": 190291255.0,
      "step": 82995
    },
    {
      "entropy": 5.4750901222229,
      "epoch": 7.973102785782901,
      "grad_norm": 1.5,
      "learning_rate": 9.489633969794468e-05,
      "loss": 4.5468,
      "mean_token_accuracy": 0.2644790053367615,
      "num_tokens": 190303156.0,
      "step": 83000
    },
    {
      "entropy": 5.4984945297241214,
      "epoch": 7.973583093179635,
      "grad_norm": 1.453125,
      "learning_rate": 9.487579273214386e-05,
      "loss": 4.6215,
      "mean_token_accuracy": 0.265743562579155,
      "num_tokens": 190315513.0,
      "step": 83005
    },
    {
      "entropy": 5.410772514343262,
      "epoch": 7.974063400576369,
      "grad_norm": 1.421875,
      "learning_rate": 9.485524994829231e-05,
      "loss": 4.4662,
      "mean_token_accuracy": 0.27758583426475525,
      "num_tokens": 190326966.0,
      "step": 83010
    },
    {
      "entropy": 5.483120918273926,
      "epoch": 7.974543707973103,
      "grad_norm": 1.5078125,
      "learning_rate": 9.483471134686689e-05,
      "loss": 4.5196,
      "mean_token_accuracy": 0.28193030655384066,
      "num_tokens": 190337681.0,
      "step": 83015
    },
    {
      "entropy": 5.523654890060425,
      "epoch": 7.975024015369836,
      "grad_norm": 1.4765625,
      "learning_rate": 9.48141769283445e-05,
      "loss": 4.5431,
      "mean_token_accuracy": 0.2774844288825989,
      "num_tokens": 190348247.0,
      "step": 83020
    },
    {
      "entropy": 5.527198219299317,
      "epoch": 7.975504322766571,
      "grad_norm": 1.40625,
      "learning_rate": 9.479364669320185e-05,
      "loss": 4.5793,
      "mean_token_accuracy": 0.26933019757270815,
      "num_tokens": 190359914.0,
      "step": 83025
    },
    {
      "entropy": 5.511746597290039,
      "epoch": 7.975984630163304,
      "grad_norm": 1.546875,
      "learning_rate": 9.477312064191567e-05,
      "loss": 4.5804,
      "mean_token_accuracy": 0.2684378162026405,
      "num_tokens": 190371263.0,
      "step": 83030
    },
    {
      "entropy": 5.45330662727356,
      "epoch": 7.9764649375600385,
      "grad_norm": 1.390625,
      "learning_rate": 9.475259877496241e-05,
      "loss": 4.5465,
      "mean_token_accuracy": 0.26895183473825457,
      "num_tokens": 190382917.0,
      "step": 83035
    },
    {
      "entropy": 5.504774570465088,
      "epoch": 7.976945244956772,
      "grad_norm": 1.46875,
      "learning_rate": 9.473208109281863e-05,
      "loss": 4.663,
      "mean_token_accuracy": 0.26441835016012194,
      "num_tokens": 190395240.0,
      "step": 83040
    },
    {
      "entropy": 5.530741739273071,
      "epoch": 7.977425552353506,
      "grad_norm": 1.3984375,
      "learning_rate": 9.471156759596066e-05,
      "loss": 4.6115,
      "mean_token_accuracy": 0.2611451968550682,
      "num_tokens": 190407271.0,
      "step": 83045
    },
    {
      "entropy": 5.562998628616333,
      "epoch": 7.977905859750241,
      "grad_norm": 1.5234375,
      "learning_rate": 9.46910582848647e-05,
      "loss": 4.6823,
      "mean_token_accuracy": 0.2619055315852165,
      "num_tokens": 190419809.0,
      "step": 83050
    },
    {
      "entropy": 5.4766936779022215,
      "epoch": 7.978386167146974,
      "grad_norm": 1.5,
      "learning_rate": 9.467055316000696e-05,
      "loss": 4.5486,
      "mean_token_accuracy": 0.27973096668720243,
      "num_tokens": 190431554.0,
      "step": 83055
    },
    {
      "entropy": 5.536663770675659,
      "epoch": 7.978866474543708,
      "grad_norm": 1.4765625,
      "learning_rate": 9.465005222186351e-05,
      "loss": 4.6027,
      "mean_token_accuracy": 0.26812389492988586,
      "num_tokens": 190442800.0,
      "step": 83060
    },
    {
      "entropy": 5.482669639587402,
      "epoch": 7.979346781940442,
      "grad_norm": 1.4140625,
      "learning_rate": 9.462955547091032e-05,
      "loss": 4.5677,
      "mean_token_accuracy": 0.26543882936239244,
      "num_tokens": 190453613.0,
      "step": 83065
    },
    {
      "entropy": 5.478753280639649,
      "epoch": 7.979827089337176,
      "grad_norm": 1.359375,
      "learning_rate": 9.460906290762324e-05,
      "loss": 4.6351,
      "mean_token_accuracy": 0.2749989092350006,
      "num_tokens": 190466746.0,
      "step": 83070
    },
    {
      "entropy": 5.618063592910767,
      "epoch": 7.9803073967339095,
      "grad_norm": 1.421875,
      "learning_rate": 9.458857453247811e-05,
      "loss": 4.7305,
      "mean_token_accuracy": 0.26357523500919344,
      "num_tokens": 190478018.0,
      "step": 83075
    },
    {
      "entropy": 5.497332286834717,
      "epoch": 7.980787704130644,
      "grad_norm": 1.59375,
      "learning_rate": 9.456809034595054e-05,
      "loss": 4.5572,
      "mean_token_accuracy": 0.269611956179142,
      "num_tokens": 190488667.0,
      "step": 83080
    },
    {
      "entropy": 5.552968311309814,
      "epoch": 7.981268011527377,
      "grad_norm": 1.3828125,
      "learning_rate": 9.45476103485162e-05,
      "loss": 4.7006,
      "mean_token_accuracy": 0.2642415001988411,
      "num_tokens": 190500085.0,
      "step": 83085
    },
    {
      "entropy": 5.5616419315338135,
      "epoch": 7.981748318924112,
      "grad_norm": 1.5,
      "learning_rate": 9.452713454065044e-05,
      "loss": 4.5843,
      "mean_token_accuracy": 0.27289851009845734,
      "num_tokens": 190510412.0,
      "step": 83090
    },
    {
      "entropy": 5.476388311386108,
      "epoch": 7.982228626320845,
      "grad_norm": 1.3125,
      "learning_rate": 9.450666292282877e-05,
      "loss": 4.5613,
      "mean_token_accuracy": 0.27463105171918867,
      "num_tokens": 190522052.0,
      "step": 83095
    },
    {
      "entropy": 5.5218620777130125,
      "epoch": 7.982708933717579,
      "grad_norm": 1.5234375,
      "learning_rate": 9.44861954955264e-05,
      "loss": 4.6965,
      "mean_token_accuracy": 0.2650661677122116,
      "num_tokens": 190533064.0,
      "step": 83100
    },
    {
      "entropy": 5.492741441726684,
      "epoch": 7.983189241114313,
      "grad_norm": 1.2890625,
      "learning_rate": 9.446573225921855e-05,
      "loss": 4.534,
      "mean_token_accuracy": 0.2699793800711632,
      "num_tokens": 190545147.0,
      "step": 83105
    },
    {
      "entropy": 5.530161571502686,
      "epoch": 7.983669548511047,
      "grad_norm": 1.3359375,
      "learning_rate": 9.44452732143803e-05,
      "loss": 4.6512,
      "mean_token_accuracy": 0.27307891845703125,
      "num_tokens": 190556975.0,
      "step": 83110
    },
    {
      "entropy": 5.536562204360962,
      "epoch": 7.984149855907781,
      "grad_norm": 1.5078125,
      "learning_rate": 9.442481836148669e-05,
      "loss": 4.6545,
      "mean_token_accuracy": 0.2650346249341965,
      "num_tokens": 190568530.0,
      "step": 83115
    },
    {
      "entropy": 5.530839920043945,
      "epoch": 7.984630163304515,
      "grad_norm": 1.5703125,
      "learning_rate": 9.440436770101262e-05,
      "loss": 4.581,
      "mean_token_accuracy": 0.277764168381691,
      "num_tokens": 190580500.0,
      "step": 83120
    },
    {
      "entropy": 5.58803071975708,
      "epoch": 7.985110470701249,
      "grad_norm": 1.4140625,
      "learning_rate": 9.438392123343293e-05,
      "loss": 4.6817,
      "mean_token_accuracy": 0.262839612364769,
      "num_tokens": 190592865.0,
      "step": 83125
    },
    {
      "entropy": 5.542111778259278,
      "epoch": 7.985590778097983,
      "grad_norm": 1.4609375,
      "learning_rate": 9.436347895922218e-05,
      "loss": 4.5961,
      "mean_token_accuracy": 0.26570947617292406,
      "num_tokens": 190603775.0,
      "step": 83130
    },
    {
      "entropy": 5.556442546844482,
      "epoch": 7.986071085494717,
      "grad_norm": 1.4765625,
      "learning_rate": 9.434304087885508e-05,
      "loss": 4.6089,
      "mean_token_accuracy": 0.2680871531367302,
      "num_tokens": 190614872.0,
      "step": 83135
    },
    {
      "entropy": 5.426616191864014,
      "epoch": 7.98655139289145,
      "grad_norm": 1.3515625,
      "learning_rate": 9.432260699280615e-05,
      "loss": 4.4821,
      "mean_token_accuracy": 0.28118243515491487,
      "num_tokens": 190625105.0,
      "step": 83140
    },
    {
      "entropy": 5.504283332824707,
      "epoch": 7.987031700288185,
      "grad_norm": 1.2890625,
      "learning_rate": 9.430217730154976e-05,
      "loss": 4.6405,
      "mean_token_accuracy": 0.26277495324611666,
      "num_tokens": 190636736.0,
      "step": 83145
    },
    {
      "entropy": 5.564868927001953,
      "epoch": 7.987512007684918,
      "grad_norm": 1.5546875,
      "learning_rate": 9.428175180556026e-05,
      "loss": 4.6672,
      "mean_token_accuracy": 0.26982906758785247,
      "num_tokens": 190649098.0,
      "step": 83150
    },
    {
      "entropy": 5.515558242797852,
      "epoch": 7.9879923150816525,
      "grad_norm": 1.4140625,
      "learning_rate": 9.426133050531186e-05,
      "loss": 4.6025,
      "mean_token_accuracy": 0.26842846274375914,
      "num_tokens": 190661236.0,
      "step": 83155
    },
    {
      "entropy": 5.480433320999145,
      "epoch": 7.988472622478386,
      "grad_norm": 1.3359375,
      "learning_rate": 9.424091340127874e-05,
      "loss": 4.563,
      "mean_token_accuracy": 0.27343574315309527,
      "num_tokens": 190673645.0,
      "step": 83160
    },
    {
      "entropy": 5.57380781173706,
      "epoch": 7.98895292987512,
      "grad_norm": 1.453125,
      "learning_rate": 9.422050049393479e-05,
      "loss": 4.7022,
      "mean_token_accuracy": 0.2593066915869713,
      "num_tokens": 190685417.0,
      "step": 83165
    },
    {
      "entropy": 5.518992948532104,
      "epoch": 7.989433237271854,
      "grad_norm": 1.4921875,
      "learning_rate": 9.420009178375404e-05,
      "loss": 4.6219,
      "mean_token_accuracy": 0.26897754669189455,
      "num_tokens": 190696507.0,
      "step": 83170
    },
    {
      "entropy": 5.5139271259307865,
      "epoch": 7.989913544668588,
      "grad_norm": 1.34375,
      "learning_rate": 9.41796872712103e-05,
      "loss": 4.5603,
      "mean_token_accuracy": 0.27265792340040207,
      "num_tokens": 190708502.0,
      "step": 83175
    },
    {
      "entropy": 5.444479131698609,
      "epoch": 7.9903938520653215,
      "grad_norm": 1.3984375,
      "learning_rate": 9.415928695677727e-05,
      "loss": 4.5427,
      "mean_token_accuracy": 0.27407185435295106,
      "num_tokens": 190719761.0,
      "step": 83180
    },
    {
      "entropy": 5.44104232788086,
      "epoch": 7.990874159462056,
      "grad_norm": 1.3828125,
      "learning_rate": 9.413889084092864e-05,
      "loss": 4.551,
      "mean_token_accuracy": 0.2691401407122612,
      "num_tokens": 190731876.0,
      "step": 83185
    },
    {
      "entropy": 5.566063356399536,
      "epoch": 7.991354466858789,
      "grad_norm": 1.5,
      "learning_rate": 9.411849892413795e-05,
      "loss": 4.7085,
      "mean_token_accuracy": 0.26090342849493026,
      "num_tokens": 190744118.0,
      "step": 83190
    },
    {
      "entropy": 5.557207870483398,
      "epoch": 7.991834774255524,
      "grad_norm": 1.5,
      "learning_rate": 9.409811120687855e-05,
      "loss": 4.6171,
      "mean_token_accuracy": 0.26490843296051025,
      "num_tokens": 190754493.0,
      "step": 83195
    },
    {
      "entropy": 5.55987958908081,
      "epoch": 7.992315081652258,
      "grad_norm": 1.5078125,
      "learning_rate": 9.407772768962386e-05,
      "loss": 4.598,
      "mean_token_accuracy": 0.2657634988427162,
      "num_tokens": 190765625.0,
      "step": 83200
    },
    {
      "entropy": 5.553909969329834,
      "epoch": 7.992795389048991,
      "grad_norm": 1.3671875,
      "learning_rate": 9.405734837284707e-05,
      "loss": 4.6043,
      "mean_token_accuracy": 0.27041284143924715,
      "num_tokens": 190776523.0,
      "step": 83205
    },
    {
      "entropy": 5.515041923522949,
      "epoch": 7.993275696445725,
      "grad_norm": 1.3203125,
      "learning_rate": 9.403697325702141e-05,
      "loss": 4.6112,
      "mean_token_accuracy": 0.2682774528861046,
      "num_tokens": 190788400.0,
      "step": 83210
    },
    {
      "entropy": 5.50836672782898,
      "epoch": 7.993756003842459,
      "grad_norm": 1.3671875,
      "learning_rate": 9.401660234261982e-05,
      "loss": 4.6311,
      "mean_token_accuracy": 0.26410272121429446,
      "num_tokens": 190800234.0,
      "step": 83215
    },
    {
      "entropy": 5.523421335220337,
      "epoch": 7.994236311239193,
      "grad_norm": 1.3515625,
      "learning_rate": 9.399623563011543e-05,
      "loss": 4.5971,
      "mean_token_accuracy": 0.26629688292741777,
      "num_tokens": 190812217.0,
      "step": 83220
    },
    {
      "entropy": 5.531105327606201,
      "epoch": 7.994716618635927,
      "grad_norm": 1.5078125,
      "learning_rate": 9.397587311998089e-05,
      "loss": 4.5674,
      "mean_token_accuracy": 0.275209042429924,
      "num_tokens": 190822942.0,
      "step": 83225
    },
    {
      "entropy": 5.607560825347901,
      "epoch": 7.995196926032661,
      "grad_norm": 1.359375,
      "learning_rate": 9.395551481268905e-05,
      "loss": 4.7341,
      "mean_token_accuracy": 0.2534781858325005,
      "num_tokens": 190834987.0,
      "step": 83230
    },
    {
      "entropy": 5.458277750015259,
      "epoch": 7.995677233429395,
      "grad_norm": 1.453125,
      "learning_rate": 9.393516070871255e-05,
      "loss": 4.6032,
      "mean_token_accuracy": 0.26847670823335645,
      "num_tokens": 190846160.0,
      "step": 83235
    },
    {
      "entropy": 5.553926801681518,
      "epoch": 7.996157540826129,
      "grad_norm": 1.7109375,
      "learning_rate": 9.391481080852398e-05,
      "loss": 4.6627,
      "mean_token_accuracy": 0.26688518524169924,
      "num_tokens": 190856868.0,
      "step": 83240
    },
    {
      "entropy": 5.531262350082398,
      "epoch": 7.996637848222862,
      "grad_norm": 1.3203125,
      "learning_rate": 9.389446511259581e-05,
      "loss": 4.6141,
      "mean_token_accuracy": 0.26704005151987076,
      "num_tokens": 190869451.0,
      "step": 83245
    },
    {
      "entropy": 5.49871187210083,
      "epoch": 7.997118155619597,
      "grad_norm": 1.3984375,
      "learning_rate": 9.387412362140043e-05,
      "loss": 4.5839,
      "mean_token_accuracy": 0.2639370128512383,
      "num_tokens": 190881093.0,
      "step": 83250
    },
    {
      "entropy": 5.600408267974854,
      "epoch": 7.99759846301633,
      "grad_norm": 1.5390625,
      "learning_rate": 9.385378633540999e-05,
      "loss": 4.7123,
      "mean_token_accuracy": 0.256695057451725,
      "num_tokens": 190892071.0,
      "step": 83255
    },
    {
      "entropy": 5.521434783935547,
      "epoch": 7.9980787704130645,
      "grad_norm": 1.453125,
      "learning_rate": 9.383345325509675e-05,
      "loss": 4.6054,
      "mean_token_accuracy": 0.2719884395599365,
      "num_tokens": 190902927.0,
      "step": 83260
    },
    {
      "entropy": 5.540793991088867,
      "epoch": 7.998559077809798,
      "grad_norm": 1.3046875,
      "learning_rate": 9.381312438093276e-05,
      "loss": 4.644,
      "mean_token_accuracy": 0.2634677693247795,
      "num_tokens": 190915161.0,
      "step": 83265
    },
    {
      "entropy": 5.5211896896362305,
      "epoch": 7.999039385206532,
      "grad_norm": 1.4453125,
      "learning_rate": 9.379279971339003e-05,
      "loss": 4.5942,
      "mean_token_accuracy": 0.2711396008729935,
      "num_tokens": 190926181.0,
      "step": 83270
    },
    {
      "entropy": 5.534595203399658,
      "epoch": 7.999519692603267,
      "grad_norm": 1.515625,
      "learning_rate": 9.377247925294042e-05,
      "loss": 4.5588,
      "mean_token_accuracy": 0.2753533661365509,
      "num_tokens": 190937332.0,
      "step": 83275
    },
    {
      "entropy": 5.595667171478271,
      "epoch": 8.0,
      "grad_norm": 1.796875,
      "learning_rate": 9.375216300005573e-05,
      "loss": 4.6322,
      "mean_token_accuracy": 0.273957234621048,
      "num_tokens": 190948288.0,
      "step": 83280
    },
    {
      "entropy": 5.561625909805298,
      "epoch": 8.000480307396733,
      "grad_norm": 1.546875,
      "learning_rate": 9.37318509552076e-05,
      "loss": 4.5821,
      "mean_token_accuracy": 0.2747158944606781,
      "num_tokens": 190959470.0,
      "step": 83285
    },
    {
      "entropy": 5.573621416091919,
      "epoch": 8.000960614793469,
      "grad_norm": 1.546875,
      "learning_rate": 9.371154311886763e-05,
      "loss": 4.723,
      "mean_token_accuracy": 0.260632848739624,
      "num_tokens": 190971225.0,
      "step": 83290
    },
    {
      "entropy": 5.510426187515259,
      "epoch": 8.001440922190202,
      "grad_norm": 1.5625,
      "learning_rate": 9.36912394915073e-05,
      "loss": 4.584,
      "mean_token_accuracy": 0.27008612006902694,
      "num_tokens": 190983696.0,
      "step": 83295
    },
    {
      "entropy": 5.486597681045533,
      "epoch": 8.001921229586936,
      "grad_norm": 1.453125,
      "learning_rate": 9.3670940073598e-05,
      "loss": 4.5486,
      "mean_token_accuracy": 0.27375943660736085,
      "num_tokens": 190995192.0,
      "step": 83300
    },
    {
      "entropy": 5.511933422088623,
      "epoch": 8.002401536983669,
      "grad_norm": 1.578125,
      "learning_rate": 9.365064486561106e-05,
      "loss": 4.5889,
      "mean_token_accuracy": 0.2710007160902023,
      "num_tokens": 191005957.0,
      "step": 83305
    },
    {
      "entropy": 5.51794924736023,
      "epoch": 8.002881844380404,
      "grad_norm": 1.40625,
      "learning_rate": 9.363035386801769e-05,
      "loss": 4.6118,
      "mean_token_accuracy": 0.2667169079184532,
      "num_tokens": 191018143.0,
      "step": 83310
    },
    {
      "entropy": 5.548024702072143,
      "epoch": 8.003362151777138,
      "grad_norm": 1.3515625,
      "learning_rate": 9.361006708128888e-05,
      "loss": 4.6255,
      "mean_token_accuracy": 0.2715615779161453,
      "num_tokens": 191029496.0,
      "step": 83315
    },
    {
      "entropy": 5.545863103866577,
      "epoch": 8.003842459173871,
      "grad_norm": 1.5703125,
      "learning_rate": 9.358978450589573e-05,
      "loss": 4.6386,
      "mean_token_accuracy": 0.26261777281761167,
      "num_tokens": 191040705.0,
      "step": 83320
    },
    {
      "entropy": 5.50573353767395,
      "epoch": 8.004322766570604,
      "grad_norm": 1.421875,
      "learning_rate": 9.356950614230905e-05,
      "loss": 4.5419,
      "mean_token_accuracy": 0.2753614544868469,
      "num_tokens": 191052506.0,
      "step": 83325
    },
    {
      "entropy": 5.572109889984131,
      "epoch": 8.00480307396734,
      "grad_norm": 1.6640625,
      "learning_rate": 9.354923199099974e-05,
      "loss": 4.6634,
      "mean_token_accuracy": 0.26638660579919815,
      "num_tokens": 191063786.0,
      "step": 83330
    },
    {
      "entropy": 5.535009288787842,
      "epoch": 8.005283381364073,
      "grad_norm": 1.46875,
      "learning_rate": 9.352896205243846e-05,
      "loss": 4.5849,
      "mean_token_accuracy": 0.26508568674325944,
      "num_tokens": 191074971.0,
      "step": 83335
    },
    {
      "entropy": 5.534331035614014,
      "epoch": 8.005763688760807,
      "grad_norm": 1.4609375,
      "learning_rate": 9.350869632709589e-05,
      "loss": 4.6317,
      "mean_token_accuracy": 0.26410990208387375,
      "num_tokens": 191086745.0,
      "step": 83340
    },
    {
      "entropy": 5.584292268753051,
      "epoch": 8.00624399615754,
      "grad_norm": 1.375,
      "learning_rate": 9.348843481544242e-05,
      "loss": 4.6821,
      "mean_token_accuracy": 0.263800311088562,
      "num_tokens": 191099038.0,
      "step": 83345
    },
    {
      "entropy": 5.45754976272583,
      "epoch": 8.006724303554275,
      "grad_norm": 1.5,
      "learning_rate": 9.34681775179485e-05,
      "loss": 4.5323,
      "mean_token_accuracy": 0.27254307866096494,
      "num_tokens": 191109885.0,
      "step": 83350
    },
    {
      "entropy": 5.580998659133911,
      "epoch": 8.007204610951009,
      "grad_norm": 1.5390625,
      "learning_rate": 9.344792443508445e-05,
      "loss": 4.6623,
      "mean_token_accuracy": 0.2673787221312523,
      "num_tokens": 191120028.0,
      "step": 83355
    },
    {
      "entropy": 5.458289337158203,
      "epoch": 8.007684918347742,
      "grad_norm": 1.59375,
      "learning_rate": 9.342767556732048e-05,
      "loss": 4.5338,
      "mean_token_accuracy": 0.2769989281892776,
      "num_tokens": 191132079.0,
      "step": 83360
    },
    {
      "entropy": 5.535205602645874,
      "epoch": 8.008165225744477,
      "grad_norm": 1.484375,
      "learning_rate": 9.340743091512676e-05,
      "loss": 4.6241,
      "mean_token_accuracy": 0.27353774607181547,
      "num_tokens": 191142675.0,
      "step": 83365
    },
    {
      "entropy": 5.614702749252319,
      "epoch": 8.00864553314121,
      "grad_norm": 1.4765625,
      "learning_rate": 9.338719047897334e-05,
      "loss": 4.6569,
      "mean_token_accuracy": 0.26774856001138686,
      "num_tokens": 191153257.0,
      "step": 83370
    },
    {
      "entropy": 5.500810003280639,
      "epoch": 8.009125840537944,
      "grad_norm": 1.6875,
      "learning_rate": 9.336695425932998e-05,
      "loss": 4.6378,
      "mean_token_accuracy": 0.2682988852262497,
      "num_tokens": 191165098.0,
      "step": 83375
    },
    {
      "entropy": 5.509009981155396,
      "epoch": 8.009606147934678,
      "grad_norm": 1.4609375,
      "learning_rate": 9.334672225666662e-05,
      "loss": 4.5943,
      "mean_token_accuracy": 0.271239173412323,
      "num_tokens": 191175855.0,
      "step": 83380
    },
    {
      "entropy": 5.534763145446777,
      "epoch": 8.010086455331413,
      "grad_norm": 1.46875,
      "learning_rate": 9.332649447145297e-05,
      "loss": 4.636,
      "mean_token_accuracy": 0.2694947823882103,
      "num_tokens": 191187028.0,
      "step": 83385
    },
    {
      "entropy": 5.581253576278686,
      "epoch": 8.010566762728146,
      "grad_norm": 1.515625,
      "learning_rate": 9.330627090415866e-05,
      "loss": 4.6286,
      "mean_token_accuracy": 0.2662922218441963,
      "num_tokens": 191197192.0,
      "step": 83390
    },
    {
      "entropy": 5.521291875839234,
      "epoch": 8.01104707012488,
      "grad_norm": 1.4453125,
      "learning_rate": 9.328605155525325e-05,
      "loss": 4.5936,
      "mean_token_accuracy": 0.2690448954701424,
      "num_tokens": 191208317.0,
      "step": 83395
    },
    {
      "entropy": 5.614150905609131,
      "epoch": 8.011527377521613,
      "grad_norm": 1.4609375,
      "learning_rate": 9.326583642520616e-05,
      "loss": 4.7029,
      "mean_token_accuracy": 0.26103067547082903,
      "num_tokens": 191220588.0,
      "step": 83400
    },
    {
      "entropy": 5.475686073303223,
      "epoch": 8.012007684918348,
      "grad_norm": 1.421875,
      "learning_rate": 9.324562551448668e-05,
      "loss": 4.526,
      "mean_token_accuracy": 0.2774422526359558,
      "num_tokens": 191231469.0,
      "step": 83405
    },
    {
      "entropy": 5.613299036026001,
      "epoch": 8.012487992315082,
      "grad_norm": 1.4921875,
      "learning_rate": 9.322541882356406e-05,
      "loss": 4.6434,
      "mean_token_accuracy": 0.2639468804001808,
      "num_tokens": 191241596.0,
      "step": 83410
    },
    {
      "entropy": 5.567828130722046,
      "epoch": 8.012968299711815,
      "grad_norm": 1.5078125,
      "learning_rate": 9.320521635290746e-05,
      "loss": 4.5762,
      "mean_token_accuracy": 0.2729702591896057,
      "num_tokens": 191251706.0,
      "step": 83415
    },
    {
      "entropy": 5.484910106658935,
      "epoch": 8.013448607108549,
      "grad_norm": 1.4765625,
      "learning_rate": 9.31850181029859e-05,
      "loss": 4.5168,
      "mean_token_accuracy": 0.2768603712320328,
      "num_tokens": 191262050.0,
      "step": 83420
    },
    {
      "entropy": 5.496414279937744,
      "epoch": 8.013928914505284,
      "grad_norm": 1.4453125,
      "learning_rate": 9.316482407426836e-05,
      "loss": 4.5275,
      "mean_token_accuracy": 0.2797435998916626,
      "num_tokens": 191274560.0,
      "step": 83425
    },
    {
      "entropy": 5.456106853485108,
      "epoch": 8.014409221902017,
      "grad_norm": 1.2890625,
      "learning_rate": 9.314463426722372e-05,
      "loss": 4.5254,
      "mean_token_accuracy": 0.2806327879428864,
      "num_tokens": 191285657.0,
      "step": 83430
    },
    {
      "entropy": 5.565062618255615,
      "epoch": 8.01488952929875,
      "grad_norm": 1.546875,
      "learning_rate": 9.312444868232061e-05,
      "loss": 4.6493,
      "mean_token_accuracy": 0.26855417191982267,
      "num_tokens": 191297558.0,
      "step": 83435
    },
    {
      "entropy": 5.452277183532715,
      "epoch": 8.015369836695486,
      "grad_norm": 1.6484375,
      "learning_rate": 9.310426732002775e-05,
      "loss": 4.5428,
      "mean_token_accuracy": 0.27033893167972567,
      "num_tokens": 191308448.0,
      "step": 83440
    },
    {
      "entropy": 5.533351707458496,
      "epoch": 8.01585014409222,
      "grad_norm": 1.46875,
      "learning_rate": 9.308409018081368e-05,
      "loss": 4.5833,
      "mean_token_accuracy": 0.26946651488542556,
      "num_tokens": 191320155.0,
      "step": 83445
    },
    {
      "entropy": 5.384952211380005,
      "epoch": 8.016330451488953,
      "grad_norm": 1.40625,
      "learning_rate": 9.306391726514687e-05,
      "loss": 4.4563,
      "mean_token_accuracy": 0.2849612295627594,
      "num_tokens": 191332337.0,
      "step": 83450
    },
    {
      "entropy": 5.563327407836914,
      "epoch": 8.016810758885686,
      "grad_norm": 1.4453125,
      "learning_rate": 9.304374857349563e-05,
      "loss": 4.666,
      "mean_token_accuracy": 0.25912812501192095,
      "num_tokens": 191343417.0,
      "step": 83455
    },
    {
      "entropy": 5.490362977981567,
      "epoch": 8.017291066282421,
      "grad_norm": 1.4609375,
      "learning_rate": 9.302358410632832e-05,
      "loss": 4.5447,
      "mean_token_accuracy": 0.27466493993997576,
      "num_tokens": 191354644.0,
      "step": 83460
    },
    {
      "entropy": 5.454006338119507,
      "epoch": 8.017771373679155,
      "grad_norm": 1.4609375,
      "learning_rate": 9.300342386411295e-05,
      "loss": 4.5445,
      "mean_token_accuracy": 0.2683520823717117,
      "num_tokens": 191366627.0,
      "step": 83465
    },
    {
      "entropy": 5.548182201385498,
      "epoch": 8.018251681075888,
      "grad_norm": 1.375,
      "learning_rate": 9.29832678473177e-05,
      "loss": 4.6164,
      "mean_token_accuracy": 0.2668395295739174,
      "num_tokens": 191377682.0,
      "step": 83470
    },
    {
      "entropy": 5.46372799873352,
      "epoch": 8.018731988472622,
      "grad_norm": 1.3359375,
      "learning_rate": 9.296311605641048e-05,
      "loss": 4.5568,
      "mean_token_accuracy": 0.28185021579265596,
      "num_tokens": 191388303.0,
      "step": 83475
    },
    {
      "entropy": 5.525523710250854,
      "epoch": 8.019212295869357,
      "grad_norm": 1.5,
      "learning_rate": 9.294296849185916e-05,
      "loss": 4.6094,
      "mean_token_accuracy": 0.2716655284166336,
      "num_tokens": 191399739.0,
      "step": 83480
    },
    {
      "entropy": 5.555505895614624,
      "epoch": 8.01969260326609,
      "grad_norm": 1.4765625,
      "learning_rate": 9.292282515413154e-05,
      "loss": 4.6467,
      "mean_token_accuracy": 0.26769723147153857,
      "num_tokens": 191412274.0,
      "step": 83485
    },
    {
      "entropy": 5.451489496231079,
      "epoch": 8.020172910662824,
      "grad_norm": 1.5078125,
      "learning_rate": 9.290268604369527e-05,
      "loss": 4.5461,
      "mean_token_accuracy": 0.2674117311835289,
      "num_tokens": 191423735.0,
      "step": 83490
    },
    {
      "entropy": 5.463139200210572,
      "epoch": 8.020653218059557,
      "grad_norm": 1.390625,
      "learning_rate": 9.288255116101788e-05,
      "loss": 4.5241,
      "mean_token_accuracy": 0.2773057922720909,
      "num_tokens": 191434772.0,
      "step": 83495
    },
    {
      "entropy": 5.4869019985198975,
      "epoch": 8.021133525456293,
      "grad_norm": 1.5625,
      "learning_rate": 9.286242050656687e-05,
      "loss": 4.5764,
      "mean_token_accuracy": 0.27790224403142927,
      "num_tokens": 191445286.0,
      "step": 83500
    },
    {
      "entropy": 5.463088274002075,
      "epoch": 8.021613832853026,
      "grad_norm": 1.4765625,
      "learning_rate": 9.284229408080963e-05,
      "loss": 4.5105,
      "mean_token_accuracy": 0.2777282059192657,
      "num_tokens": 191455764.0,
      "step": 83505
    },
    {
      "entropy": 5.464240837097168,
      "epoch": 8.02209414024976,
      "grad_norm": 1.40625,
      "learning_rate": 9.28221718842134e-05,
      "loss": 4.6143,
      "mean_token_accuracy": 0.27069924771785736,
      "num_tokens": 191467563.0,
      "step": 83510
    },
    {
      "entropy": 5.606807661056519,
      "epoch": 8.022574447646495,
      "grad_norm": 1.4609375,
      "learning_rate": 9.280205391724541e-05,
      "loss": 4.7089,
      "mean_token_accuracy": 0.25537123829126357,
      "num_tokens": 191479505.0,
      "step": 83515
    },
    {
      "entropy": 5.546534872055053,
      "epoch": 8.023054755043228,
      "grad_norm": 1.34375,
      "learning_rate": 9.278194018037276e-05,
      "loss": 4.5922,
      "mean_token_accuracy": 0.2769012004137039,
      "num_tokens": 191490401.0,
      "step": 83520
    },
    {
      "entropy": 5.542569828033447,
      "epoch": 8.023535062439962,
      "grad_norm": 1.4453125,
      "learning_rate": 9.276183067406232e-05,
      "loss": 4.6067,
      "mean_token_accuracy": 0.2677088648080826,
      "num_tokens": 191501970.0,
      "step": 83525
    },
    {
      "entropy": 5.569790887832641,
      "epoch": 8.024015369836695,
      "grad_norm": 1.3203125,
      "learning_rate": 9.274172539878104e-05,
      "loss": 4.6098,
      "mean_token_accuracy": 0.266112194955349,
      "num_tokens": 191514079.0,
      "step": 83530
    },
    {
      "entropy": 5.521232032775879,
      "epoch": 8.02449567723343,
      "grad_norm": 1.390625,
      "learning_rate": 9.272162435499568e-05,
      "loss": 4.5808,
      "mean_token_accuracy": 0.26774997264146805,
      "num_tokens": 191526043.0,
      "step": 83535
    },
    {
      "entropy": 5.528161811828613,
      "epoch": 8.024975984630164,
      "grad_norm": 1.3984375,
      "learning_rate": 9.270152754317296e-05,
      "loss": 4.6816,
      "mean_token_accuracy": 0.26074465811252595,
      "num_tokens": 191538847.0,
      "step": 83540
    },
    {
      "entropy": 5.486400175094604,
      "epoch": 8.025456292026897,
      "grad_norm": 1.5078125,
      "learning_rate": 9.268143496377949e-05,
      "loss": 4.576,
      "mean_token_accuracy": 0.27428008913993834,
      "num_tokens": 191549813.0,
      "step": 83545
    },
    {
      "entropy": 5.478278112411499,
      "epoch": 8.02593659942363,
      "grad_norm": 1.515625,
      "learning_rate": 9.266134661728174e-05,
      "loss": 4.5356,
      "mean_token_accuracy": 0.27632421553134917,
      "num_tokens": 191560614.0,
      "step": 83550
    },
    {
      "entropy": 5.577210235595703,
      "epoch": 8.026416906820366,
      "grad_norm": 1.5078125,
      "learning_rate": 9.264126250414605e-05,
      "loss": 4.6705,
      "mean_token_accuracy": 0.26248865872621535,
      "num_tokens": 191572388.0,
      "step": 83555
    },
    {
      "entropy": 5.548244571685791,
      "epoch": 8.0268972142171,
      "grad_norm": 1.421875,
      "learning_rate": 9.262118262483873e-05,
      "loss": 4.6971,
      "mean_token_accuracy": 0.265985780954361,
      "num_tokens": 191584140.0,
      "step": 83560
    },
    {
      "entropy": 5.48846025466919,
      "epoch": 8.027377521613833,
      "grad_norm": 1.53125,
      "learning_rate": 9.260110697982602e-05,
      "loss": 4.5219,
      "mean_token_accuracy": 0.2742289945483208,
      "num_tokens": 191594128.0,
      "step": 83565
    },
    {
      "entropy": 5.510998916625977,
      "epoch": 8.027857829010566,
      "grad_norm": 1.6328125,
      "learning_rate": 9.258103556957398e-05,
      "loss": 4.6069,
      "mean_token_accuracy": 0.26846869587898253,
      "num_tokens": 191605008.0,
      "step": 83570
    },
    {
      "entropy": 5.4380974769592285,
      "epoch": 8.028338136407301,
      "grad_norm": 1.2734375,
      "learning_rate": 9.256096839454863e-05,
      "loss": 4.514,
      "mean_token_accuracy": 0.2833043336868286,
      "num_tokens": 191617047.0,
      "step": 83575
    },
    {
      "entropy": 5.592174673080445,
      "epoch": 8.028818443804035,
      "grad_norm": 1.46875,
      "learning_rate": 9.25409054552159e-05,
      "loss": 4.7023,
      "mean_token_accuracy": 0.26038536727428435,
      "num_tokens": 191627316.0,
      "step": 83580
    },
    {
      "entropy": 5.514005136489868,
      "epoch": 8.029298751200768,
      "grad_norm": 1.796875,
      "learning_rate": 9.25208467520415e-05,
      "loss": 4.5977,
      "mean_token_accuracy": 0.26238508969545365,
      "num_tokens": 191639184.0,
      "step": 83585
    },
    {
      "entropy": 5.535818290710449,
      "epoch": 8.029779058597502,
      "grad_norm": 1.4921875,
      "learning_rate": 9.250079228549119e-05,
      "loss": 4.6381,
      "mean_token_accuracy": 0.26986327171325686,
      "num_tokens": 191650977.0,
      "step": 83590
    },
    {
      "entropy": 5.5799384117126465,
      "epoch": 8.030259365994237,
      "grad_norm": 1.484375,
      "learning_rate": 9.248074205603058e-05,
      "loss": 4.6357,
      "mean_token_accuracy": 0.2646437853574753,
      "num_tokens": 191663245.0,
      "step": 83595
    },
    {
      "entropy": 5.5916095733642575,
      "epoch": 8.03073967339097,
      "grad_norm": 1.5625,
      "learning_rate": 9.246069606412515e-05,
      "loss": 4.6454,
      "mean_token_accuracy": 0.26232237219810484,
      "num_tokens": 191673768.0,
      "step": 83600
    },
    {
      "entropy": 5.597424554824829,
      "epoch": 8.031219980787704,
      "grad_norm": 1.375,
      "learning_rate": 9.244065431024035e-05,
      "loss": 4.6704,
      "mean_token_accuracy": 0.2653399884700775,
      "num_tokens": 191684655.0,
      "step": 83605
    },
    {
      "entropy": 5.433764839172364,
      "epoch": 8.031700288184439,
      "grad_norm": 1.3828125,
      "learning_rate": 9.242061679484145e-05,
      "loss": 4.4876,
      "mean_token_accuracy": 0.279572120308876,
      "num_tokens": 191697155.0,
      "step": 83610
    },
    {
      "entropy": 5.459165239334107,
      "epoch": 8.032180595581172,
      "grad_norm": 1.3515625,
      "learning_rate": 9.24005835183937e-05,
      "loss": 4.5384,
      "mean_token_accuracy": 0.27114823311567304,
      "num_tokens": 191708469.0,
      "step": 83615
    },
    {
      "entropy": 5.480085277557373,
      "epoch": 8.032660902977906,
      "grad_norm": 1.515625,
      "learning_rate": 9.238055448136224e-05,
      "loss": 4.5568,
      "mean_token_accuracy": 0.27226781100034714,
      "num_tokens": 191719092.0,
      "step": 83620
    },
    {
      "entropy": 5.552248477935791,
      "epoch": 8.03314121037464,
      "grad_norm": 1.4140625,
      "learning_rate": 9.236052968421198e-05,
      "loss": 4.6995,
      "mean_token_accuracy": 0.261690653860569,
      "num_tokens": 191731625.0,
      "step": 83625
    },
    {
      "entropy": 5.498490905761718,
      "epoch": 8.033621517771374,
      "grad_norm": 1.359375,
      "learning_rate": 9.234050912740788e-05,
      "loss": 4.6238,
      "mean_token_accuracy": 0.27061064839363097,
      "num_tokens": 191742839.0,
      "step": 83630
    },
    {
      "entropy": 5.577027177810669,
      "epoch": 8.034101825168108,
      "grad_norm": 1.3984375,
      "learning_rate": 9.232049281141477e-05,
      "loss": 4.684,
      "mean_token_accuracy": 0.2664466366171837,
      "num_tokens": 191753630.0,
      "step": 83635
    },
    {
      "entropy": 5.5026181697845455,
      "epoch": 8.034582132564841,
      "grad_norm": 1.7421875,
      "learning_rate": 9.230048073669739e-05,
      "loss": 4.6559,
      "mean_token_accuracy": 0.2676150768995285,
      "num_tokens": 191765548.0,
      "step": 83640
    },
    {
      "entropy": 5.58566460609436,
      "epoch": 8.035062439961575,
      "grad_norm": 1.546875,
      "learning_rate": 9.228047290372032e-05,
      "loss": 4.694,
      "mean_token_accuracy": 0.2643905937671661,
      "num_tokens": 191776911.0,
      "step": 83645
    },
    {
      "entropy": 5.520675420761108,
      "epoch": 8.03554274735831,
      "grad_norm": 1.546875,
      "learning_rate": 9.226046931294811e-05,
      "loss": 4.5973,
      "mean_token_accuracy": 0.2657265141606331,
      "num_tokens": 191788336.0,
      "step": 83650
    },
    {
      "entropy": 5.457088136672974,
      "epoch": 8.036023054755043,
      "grad_norm": 1.40625,
      "learning_rate": 9.22404699648452e-05,
      "loss": 4.548,
      "mean_token_accuracy": 0.27996402978897095,
      "num_tokens": 191801017.0,
      "step": 83655
    },
    {
      "entropy": 5.521833086013794,
      "epoch": 8.036503362151777,
      "grad_norm": 1.390625,
      "learning_rate": 9.222047485987592e-05,
      "loss": 4.6293,
      "mean_token_accuracy": 0.26716116070747375,
      "num_tokens": 191813287.0,
      "step": 83660
    },
    {
      "entropy": 5.488739442825318,
      "epoch": 8.03698366954851,
      "grad_norm": 1.5390625,
      "learning_rate": 9.220048399850445e-05,
      "loss": 4.602,
      "mean_token_accuracy": 0.2653929486870766,
      "num_tokens": 191823664.0,
      "step": 83665
    },
    {
      "entropy": 5.431840276718139,
      "epoch": 8.037463976945245,
      "grad_norm": 1.515625,
      "learning_rate": 9.218049738119488e-05,
      "loss": 4.4922,
      "mean_token_accuracy": 0.28220740854740145,
      "num_tokens": 191835929.0,
      "step": 83670
    },
    {
      "entropy": 5.516669130325317,
      "epoch": 8.037944284341979,
      "grad_norm": 1.375,
      "learning_rate": 9.216051500841134e-05,
      "loss": 4.6553,
      "mean_token_accuracy": 0.26764140874147413,
      "num_tokens": 191849344.0,
      "step": 83675
    },
    {
      "entropy": 5.600926208496094,
      "epoch": 8.038424591738712,
      "grad_norm": 1.4453125,
      "learning_rate": 9.214053688061772e-05,
      "loss": 4.6724,
      "mean_token_accuracy": 0.255774661898613,
      "num_tokens": 191861985.0,
      "step": 83680
    },
    {
      "entropy": 5.58397479057312,
      "epoch": 8.038904899135447,
      "grad_norm": 1.5078125,
      "learning_rate": 9.212056299827785e-05,
      "loss": 4.6615,
      "mean_token_accuracy": 0.2689576029777527,
      "num_tokens": 191873205.0,
      "step": 83685
    },
    {
      "entropy": 5.505596494674682,
      "epoch": 8.039385206532181,
      "grad_norm": 1.421875,
      "learning_rate": 9.210059336185546e-05,
      "loss": 4.5778,
      "mean_token_accuracy": 0.27430294305086134,
      "num_tokens": 191884881.0,
      "step": 83690
    },
    {
      "entropy": 5.574382257461548,
      "epoch": 8.039865513928914,
      "grad_norm": 1.53125,
      "learning_rate": 9.208062797181418e-05,
      "loss": 4.6492,
      "mean_token_accuracy": 0.26252026557922364,
      "num_tokens": 191896015.0,
      "step": 83695
    },
    {
      "entropy": 5.4954609870910645,
      "epoch": 8.040345821325648,
      "grad_norm": 1.3671875,
      "learning_rate": 9.206066682861763e-05,
      "loss": 4.5357,
      "mean_token_accuracy": 0.27482594549655914,
      "num_tokens": 191907288.0,
      "step": 83700
    },
    {
      "entropy": 5.541153383255005,
      "epoch": 8.040826128722383,
      "grad_norm": 1.3984375,
      "learning_rate": 9.204070993272912e-05,
      "loss": 4.6444,
      "mean_token_accuracy": 0.2672965243458748,
      "num_tokens": 191918412.0,
      "step": 83705
    },
    {
      "entropy": 5.564008569717407,
      "epoch": 8.041306436119116,
      "grad_norm": 1.4453125,
      "learning_rate": 9.202075728461206e-05,
      "loss": 4.6454,
      "mean_token_accuracy": 0.25485320538282397,
      "num_tokens": 191929117.0,
      "step": 83710
    },
    {
      "entropy": 5.4585041999816895,
      "epoch": 8.04178674351585,
      "grad_norm": 1.2890625,
      "learning_rate": 9.200080888472966e-05,
      "loss": 4.5772,
      "mean_token_accuracy": 0.27602037489414216,
      "num_tokens": 191941551.0,
      "step": 83715
    },
    {
      "entropy": 5.6442896842956545,
      "epoch": 8.042267050912583,
      "grad_norm": 1.3671875,
      "learning_rate": 9.198086473354509e-05,
      "loss": 4.7726,
      "mean_token_accuracy": 0.25297749489545823,
      "num_tokens": 191954078.0,
      "step": 83720
    },
    {
      "entropy": 5.494951820373535,
      "epoch": 8.042747358309319,
      "grad_norm": 1.4609375,
      "learning_rate": 9.19609248315214e-05,
      "loss": 4.5234,
      "mean_token_accuracy": 0.2731934353709221,
      "num_tokens": 191965664.0,
      "step": 83725
    },
    {
      "entropy": 5.495788717269898,
      "epoch": 8.043227665706052,
      "grad_norm": 1.5078125,
      "learning_rate": 9.194098917912155e-05,
      "loss": 4.5842,
      "mean_token_accuracy": 0.2754505008459091,
      "num_tokens": 191977234.0,
      "step": 83730
    },
    {
      "entropy": 5.560746145248413,
      "epoch": 8.043707973102785,
      "grad_norm": 1.703125,
      "learning_rate": 9.192105777680831e-05,
      "loss": 4.661,
      "mean_token_accuracy": 0.2672446250915527,
      "num_tokens": 191988164.0,
      "step": 83735
    },
    {
      "entropy": 5.5448277473449705,
      "epoch": 8.044188280499519,
      "grad_norm": 1.390625,
      "learning_rate": 9.190113062504449e-05,
      "loss": 4.5809,
      "mean_token_accuracy": 0.2683079794049263,
      "num_tokens": 191999399.0,
      "step": 83740
    },
    {
      "entropy": 5.5717205047607425,
      "epoch": 8.044668587896254,
      "grad_norm": 1.3984375,
      "learning_rate": 9.18812077242927e-05,
      "loss": 4.6941,
      "mean_token_accuracy": 0.26651568412780763,
      "num_tokens": 192010381.0,
      "step": 83745
    },
    {
      "entropy": 5.5926069736480715,
      "epoch": 8.045148895292987,
      "grad_norm": 1.4765625,
      "learning_rate": 9.186128907501553e-05,
      "loss": 4.6612,
      "mean_token_accuracy": 0.2627770572900772,
      "num_tokens": 192021128.0,
      "step": 83750
    },
    {
      "entropy": 5.516658592224121,
      "epoch": 8.045629202689721,
      "grad_norm": 1.4375,
      "learning_rate": 9.184137467767545e-05,
      "loss": 4.5746,
      "mean_token_accuracy": 0.2720016002655029,
      "num_tokens": 192032968.0,
      "step": 83755
    },
    {
      "entropy": 5.53009524345398,
      "epoch": 8.046109510086456,
      "grad_norm": 1.4453125,
      "learning_rate": 9.18214645327348e-05,
      "loss": 4.6302,
      "mean_token_accuracy": 0.2652864843606949,
      "num_tokens": 192045012.0,
      "step": 83760
    },
    {
      "entropy": 5.52898473739624,
      "epoch": 8.04658981748319,
      "grad_norm": 1.3359375,
      "learning_rate": 9.180155864065576e-05,
      "loss": 4.5954,
      "mean_token_accuracy": 0.2630010813474655,
      "num_tokens": 192055902.0,
      "step": 83765
    },
    {
      "entropy": 5.493992471694947,
      "epoch": 8.047070124879923,
      "grad_norm": 1.75,
      "learning_rate": 9.178165700190055e-05,
      "loss": 4.5588,
      "mean_token_accuracy": 0.27289028465747833,
      "num_tokens": 192066707.0,
      "step": 83770
    },
    {
      "entropy": 5.488883066177368,
      "epoch": 8.047550432276656,
      "grad_norm": 1.3203125,
      "learning_rate": 9.176175961693121e-05,
      "loss": 4.573,
      "mean_token_accuracy": 0.27554250359535215,
      "num_tokens": 192078130.0,
      "step": 83775
    },
    {
      "entropy": 5.495991182327271,
      "epoch": 8.048030739673392,
      "grad_norm": 1.4140625,
      "learning_rate": 9.174186648620972e-05,
      "loss": 4.561,
      "mean_token_accuracy": 0.2767424643039703,
      "num_tokens": 192088491.0,
      "step": 83780
    },
    {
      "entropy": 5.520651340484619,
      "epoch": 8.048511047070125,
      "grad_norm": 1.4765625,
      "learning_rate": 9.172197761019788e-05,
      "loss": 4.5669,
      "mean_token_accuracy": 0.2597303420305252,
      "num_tokens": 192100674.0,
      "step": 83785
    },
    {
      "entropy": 5.550838041305542,
      "epoch": 8.048991354466859,
      "grad_norm": 1.515625,
      "learning_rate": 9.170209298935761e-05,
      "loss": 4.5728,
      "mean_token_accuracy": 0.26819113790988924,
      "num_tokens": 192112268.0,
      "step": 83790
    },
    {
      "entropy": 5.4798847198486325,
      "epoch": 8.049471661863592,
      "grad_norm": 1.4140625,
      "learning_rate": 9.168221262415039e-05,
      "loss": 4.6075,
      "mean_token_accuracy": 0.26470488011837007,
      "num_tokens": 192124892.0,
      "step": 83795
    },
    {
      "entropy": 5.557526063919068,
      "epoch": 8.049951969260327,
      "grad_norm": 1.4140625,
      "learning_rate": 9.166233651503783e-05,
      "loss": 4.6921,
      "mean_token_accuracy": 0.26536825448274615,
      "num_tokens": 192136665.0,
      "step": 83800
    },
    {
      "entropy": 5.523958587646485,
      "epoch": 8.05043227665706,
      "grad_norm": 1.4765625,
      "learning_rate": 9.164246466248141e-05,
      "loss": 4.6601,
      "mean_token_accuracy": 0.269507098197937,
      "num_tokens": 192147580.0,
      "step": 83805
    },
    {
      "entropy": 5.505958938598633,
      "epoch": 8.050912584053794,
      "grad_norm": 1.5390625,
      "learning_rate": 9.162259706694254e-05,
      "loss": 4.5333,
      "mean_token_accuracy": 0.2770512893795967,
      "num_tokens": 192158120.0,
      "step": 83810
    },
    {
      "entropy": 5.591593503952026,
      "epoch": 8.051392891450528,
      "grad_norm": 1.484375,
      "learning_rate": 9.160273372888239e-05,
      "loss": 4.7166,
      "mean_token_accuracy": 0.2635667070746422,
      "num_tokens": 192170144.0,
      "step": 83815
    },
    {
      "entropy": 5.52413763999939,
      "epoch": 8.051873198847263,
      "grad_norm": 1.328125,
      "learning_rate": 9.158287464876228e-05,
      "loss": 4.5769,
      "mean_token_accuracy": 0.27073743641376496,
      "num_tokens": 192183028.0,
      "step": 83820
    },
    {
      "entropy": 5.490774822235108,
      "epoch": 8.052353506243996,
      "grad_norm": 1.5859375,
      "learning_rate": 9.156301982704313e-05,
      "loss": 4.5023,
      "mean_token_accuracy": 0.2728767141699791,
      "num_tokens": 192193609.0,
      "step": 83825
    },
    {
      "entropy": 5.535236072540283,
      "epoch": 8.05283381364073,
      "grad_norm": 1.3359375,
      "learning_rate": 9.154316926418595e-05,
      "loss": 4.5757,
      "mean_token_accuracy": 0.2785253465175629,
      "num_tokens": 192204182.0,
      "step": 83830
    },
    {
      "entropy": 5.525649976730347,
      "epoch": 8.053314121037465,
      "grad_norm": 1.5546875,
      "learning_rate": 9.152332296065161e-05,
      "loss": 4.5865,
      "mean_token_accuracy": 0.275822751224041,
      "num_tokens": 192217014.0,
      "step": 83835
    },
    {
      "entropy": 5.557937479019165,
      "epoch": 8.053794428434198,
      "grad_norm": 1.46875,
      "learning_rate": 9.150348091690094e-05,
      "loss": 4.612,
      "mean_token_accuracy": 0.26473443061113355,
      "num_tokens": 192228713.0,
      "step": 83840
    },
    {
      "entropy": 5.467742013931274,
      "epoch": 8.054274735830932,
      "grad_norm": 1.3515625,
      "learning_rate": 9.148364313339455e-05,
      "loss": 4.6219,
      "mean_token_accuracy": 0.2724530100822449,
      "num_tokens": 192240195.0,
      "step": 83845
    },
    {
      "entropy": 5.586493015289307,
      "epoch": 8.054755043227665,
      "grad_norm": 1.5,
      "learning_rate": 9.146380961059307e-05,
      "loss": 4.6529,
      "mean_token_accuracy": 0.2604678139090538,
      "num_tokens": 192251895.0,
      "step": 83850
    },
    {
      "entropy": 5.459809970855713,
      "epoch": 8.0552353506244,
      "grad_norm": 1.3515625,
      "learning_rate": 9.144398034895692e-05,
      "loss": 4.5896,
      "mean_token_accuracy": 0.2647983506321907,
      "num_tokens": 192262875.0,
      "step": 83855
    },
    {
      "entropy": 5.4499430656433105,
      "epoch": 8.055715658021134,
      "grad_norm": 1.3984375,
      "learning_rate": 9.142415534894649e-05,
      "loss": 4.527,
      "mean_token_accuracy": 0.26967945545911787,
      "num_tokens": 192273701.0,
      "step": 83860
    },
    {
      "entropy": 5.475140762329102,
      "epoch": 8.056195965417867,
      "grad_norm": 1.421875,
      "learning_rate": 9.140433461102204e-05,
      "loss": 4.5728,
      "mean_token_accuracy": 0.27397834658622744,
      "num_tokens": 192285309.0,
      "step": 83865
    },
    {
      "entropy": 5.5261674404144285,
      "epoch": 8.0566762728146,
      "grad_norm": 1.46875,
      "learning_rate": 9.138451813564377e-05,
      "loss": 4.6109,
      "mean_token_accuracy": 0.2623490199446678,
      "num_tokens": 192297695.0,
      "step": 83870
    },
    {
      "entropy": 5.509649038314819,
      "epoch": 8.057156580211336,
      "grad_norm": 1.4921875,
      "learning_rate": 9.136470592327178e-05,
      "loss": 4.606,
      "mean_token_accuracy": 0.2693545460700989,
      "num_tokens": 192308898.0,
      "step": 83875
    },
    {
      "entropy": 5.620950222015381,
      "epoch": 8.05763688760807,
      "grad_norm": 1.4609375,
      "learning_rate": 9.134489797436613e-05,
      "loss": 4.7538,
      "mean_token_accuracy": 0.25674074739217756,
      "num_tokens": 192321312.0,
      "step": 83880
    },
    {
      "entropy": 5.686242341995239,
      "epoch": 8.058117195004803,
      "grad_norm": 1.6484375,
      "learning_rate": 9.132509428938649e-05,
      "loss": 4.7723,
      "mean_token_accuracy": 0.25576903522014616,
      "num_tokens": 192334417.0,
      "step": 83885
    },
    {
      "entropy": 5.602407789230346,
      "epoch": 8.058597502401536,
      "grad_norm": 1.4609375,
      "learning_rate": 9.130529486879277e-05,
      "loss": 4.6799,
      "mean_token_accuracy": 0.2658845528960228,
      "num_tokens": 192345612.0,
      "step": 83890
    },
    {
      "entropy": 5.587365531921387,
      "epoch": 8.059077809798271,
      "grad_norm": 1.375,
      "learning_rate": 9.128549971304466e-05,
      "loss": 4.6836,
      "mean_token_accuracy": 0.2593871891498566,
      "num_tokens": 192359461.0,
      "step": 83895
    },
    {
      "entropy": 5.562079811096192,
      "epoch": 8.059558117195005,
      "grad_norm": 1.4375,
      "learning_rate": 9.12657088226017e-05,
      "loss": 4.6438,
      "mean_token_accuracy": 0.2713285058736801,
      "num_tokens": 192371615.0,
      "step": 83900
    },
    {
      "entropy": 5.57713131904602,
      "epoch": 8.060038424591738,
      "grad_norm": 1.5390625,
      "learning_rate": 9.124592219792344e-05,
      "loss": 4.659,
      "mean_token_accuracy": 0.26214643716812136,
      "num_tokens": 192382682.0,
      "step": 83905
    },
    {
      "entropy": 5.571987295150757,
      "epoch": 8.060518731988473,
      "grad_norm": 1.5,
      "learning_rate": 9.122613983946925e-05,
      "loss": 4.6973,
      "mean_token_accuracy": 0.2553922191262245,
      "num_tokens": 192395055.0,
      "step": 83910
    },
    {
      "entropy": 5.579578828811646,
      "epoch": 8.060999039385207,
      "grad_norm": 1.5078125,
      "learning_rate": 9.120636174769833e-05,
      "loss": 4.6643,
      "mean_token_accuracy": 0.2710287973284721,
      "num_tokens": 192405673.0,
      "step": 83915
    },
    {
      "entropy": 5.462378644943238,
      "epoch": 8.06147934678194,
      "grad_norm": 1.46875,
      "learning_rate": 9.118658792306997e-05,
      "loss": 4.4829,
      "mean_token_accuracy": 0.2781299203634262,
      "num_tokens": 192416028.0,
      "step": 83920
    },
    {
      "entropy": 5.495112752914428,
      "epoch": 8.061959654178674,
      "grad_norm": 1.671875,
      "learning_rate": 9.116681836604321e-05,
      "loss": 4.5387,
      "mean_token_accuracy": 0.2706594794988632,
      "num_tokens": 192426841.0,
      "step": 83925
    },
    {
      "entropy": 5.498195219039917,
      "epoch": 8.062439961575409,
      "grad_norm": 1.5078125,
      "learning_rate": 9.114705307707708e-05,
      "loss": 4.5558,
      "mean_token_accuracy": 0.26958163529634477,
      "num_tokens": 192437547.0,
      "step": 83930
    },
    {
      "entropy": 5.50390396118164,
      "epoch": 8.062920268972142,
      "grad_norm": 1.40625,
      "learning_rate": 9.112729205663041e-05,
      "loss": 4.5842,
      "mean_token_accuracy": 0.2685898199677467,
      "num_tokens": 192448508.0,
      "step": 83935
    },
    {
      "entropy": 5.430561923980713,
      "epoch": 8.063400576368876,
      "grad_norm": 1.5078125,
      "learning_rate": 9.110753530516212e-05,
      "loss": 4.5392,
      "mean_token_accuracy": 0.2674103543162346,
      "num_tokens": 192459924.0,
      "step": 83940
    },
    {
      "entropy": 5.4282910346984865,
      "epoch": 8.06388088376561,
      "grad_norm": 1.5,
      "learning_rate": 9.108778282313074e-05,
      "loss": 4.4918,
      "mean_token_accuracy": 0.28163838386535645,
      "num_tokens": 192471934.0,
      "step": 83945
    },
    {
      "entropy": 5.567105102539062,
      "epoch": 8.064361191162345,
      "grad_norm": 1.4921875,
      "learning_rate": 9.106803461099491e-05,
      "loss": 4.6578,
      "mean_token_accuracy": 0.2695566192269325,
      "num_tokens": 192482765.0,
      "step": 83950
    },
    {
      "entropy": 5.541204881668091,
      "epoch": 8.064841498559078,
      "grad_norm": 1.40625,
      "learning_rate": 9.104829066921319e-05,
      "loss": 4.6115,
      "mean_token_accuracy": 0.2791746884584427,
      "num_tokens": 192494519.0,
      "step": 83955
    },
    {
      "entropy": 5.563522100448608,
      "epoch": 8.065321805955811,
      "grad_norm": 1.578125,
      "learning_rate": 9.102855099824392e-05,
      "loss": 4.5866,
      "mean_token_accuracy": 0.2732430651783943,
      "num_tokens": 192506483.0,
      "step": 83960
    },
    {
      "entropy": 5.480102300643921,
      "epoch": 8.065802113352545,
      "grad_norm": 1.390625,
      "learning_rate": 9.10088155985454e-05,
      "loss": 4.5426,
      "mean_token_accuracy": 0.27000013291835784,
      "num_tokens": 192517925.0,
      "step": 83965
    },
    {
      "entropy": 5.506524658203125,
      "epoch": 8.06628242074928,
      "grad_norm": 1.453125,
      "learning_rate": 9.098908447057591e-05,
      "loss": 4.5573,
      "mean_token_accuracy": 0.2714375302195549,
      "num_tokens": 192528793.0,
      "step": 83970
    },
    {
      "entropy": 5.540863466262818,
      "epoch": 8.066762728146013,
      "grad_norm": 1.703125,
      "learning_rate": 9.096935761479341e-05,
      "loss": 4.6037,
      "mean_token_accuracy": 0.2608863443136215,
      "num_tokens": 192539328.0,
      "step": 83975
    },
    {
      "entropy": 5.502819585800171,
      "epoch": 8.067243035542747,
      "grad_norm": 1.4765625,
      "learning_rate": 9.094963503165599e-05,
      "loss": 4.5932,
      "mean_token_accuracy": 0.2751583755016327,
      "num_tokens": 192550508.0,
      "step": 83980
    },
    {
      "entropy": 5.508764600753784,
      "epoch": 8.067723342939482,
      "grad_norm": 1.546875,
      "learning_rate": 9.092991672162152e-05,
      "loss": 4.5848,
      "mean_token_accuracy": 0.2701616123318672,
      "num_tokens": 192562209.0,
      "step": 83985
    },
    {
      "entropy": 5.55151629447937,
      "epoch": 8.068203650336216,
      "grad_norm": 1.4140625,
      "learning_rate": 9.09102026851478e-05,
      "loss": 4.6552,
      "mean_token_accuracy": 0.27089678347110746,
      "num_tokens": 192574630.0,
      "step": 83990
    },
    {
      "entropy": 5.549516582489014,
      "epoch": 8.068683957732949,
      "grad_norm": 1.34375,
      "learning_rate": 9.089049292269254e-05,
      "loss": 4.6331,
      "mean_token_accuracy": 0.2680306762456894,
      "num_tokens": 192588918.0,
      "step": 83995
    },
    {
      "entropy": 5.53296127319336,
      "epoch": 8.069164265129682,
      "grad_norm": 1.515625,
      "learning_rate": 9.08707874347134e-05,
      "loss": 4.5984,
      "mean_token_accuracy": 0.2677631974220276,
      "num_tokens": 192599661.0,
      "step": 84000
    },
    {
      "epoch": 8.069164265129682,
      "eval_entropy": 5.236136082844988,
      "eval_loss": 4.736738204956055,
      "eval_mean_token_accuracy": 0.2687668114862446,
      "eval_num_tokens": 192599661.0,
      "eval_runtime": 20.6384,
      "eval_samples_per_second": 1589.998,
      "eval_steps_per_second": 198.756,
      "step": 84000
    },
    {
      "entropy": 5.540999889373779,
      "epoch": 8.069644572526418,
      "grad_norm": 1.4375,
      "learning_rate": 9.085108622166777e-05,
      "loss": 4.5921,
      "mean_token_accuracy": 0.2757189214229584,
      "num_tokens": 192610780.0,
      "step": 84005
    },
    {
      "entropy": 5.477624177932739,
      "epoch": 8.070124879923151,
      "grad_norm": 1.3671875,
      "learning_rate": 9.08313892840131e-05,
      "loss": 4.5175,
      "mean_token_accuracy": 0.2815058767795563,
      "num_tokens": 192622306.0,
      "step": 84010
    },
    {
      "entropy": 5.492802858352661,
      "epoch": 8.070605187319885,
      "grad_norm": 1.3984375,
      "learning_rate": 9.08116966222067e-05,
      "loss": 4.6009,
      "mean_token_accuracy": 0.27064897418022155,
      "num_tokens": 192633908.0,
      "step": 84015
    },
    {
      "entropy": 5.5614441394805905,
      "epoch": 8.071085494716618,
      "grad_norm": 1.3984375,
      "learning_rate": 9.07920082367058e-05,
      "loss": 4.6274,
      "mean_token_accuracy": 0.2733959123492241,
      "num_tokens": 192645879.0,
      "step": 84020
    },
    {
      "entropy": 5.517910766601562,
      "epoch": 8.071565802113353,
      "grad_norm": 1.40625,
      "learning_rate": 9.077232412796746e-05,
      "loss": 4.6547,
      "mean_token_accuracy": 0.2690082788467407,
      "num_tokens": 192657714.0,
      "step": 84025
    },
    {
      "entropy": 5.556331920623779,
      "epoch": 8.072046109510087,
      "grad_norm": 1.546875,
      "learning_rate": 9.075264429644878e-05,
      "loss": 4.6674,
      "mean_token_accuracy": 0.2651951476931572,
      "num_tokens": 192669847.0,
      "step": 84030
    },
    {
      "entropy": 5.498369836807251,
      "epoch": 8.07252641690682,
      "grad_norm": 1.515625,
      "learning_rate": 9.073296874260654e-05,
      "loss": 4.6368,
      "mean_token_accuracy": 0.2643096104264259,
      "num_tokens": 192680940.0,
      "step": 84035
    },
    {
      "entropy": 5.534420442581177,
      "epoch": 8.073006724303553,
      "grad_norm": 1.484375,
      "learning_rate": 9.07132974668976e-05,
      "loss": 4.552,
      "mean_token_accuracy": 0.2724806934595108,
      "num_tokens": 192692015.0,
      "step": 84040
    },
    {
      "entropy": 5.571306228637695,
      "epoch": 8.073487031700289,
      "grad_norm": 1.515625,
      "learning_rate": 9.069363046977866e-05,
      "loss": 4.6628,
      "mean_token_accuracy": 0.2616443380713463,
      "num_tokens": 192704419.0,
      "step": 84045
    },
    {
      "entropy": 5.513591384887695,
      "epoch": 8.073967339097022,
      "grad_norm": 1.3828125,
      "learning_rate": 9.067396775170637e-05,
      "loss": 4.6286,
      "mean_token_accuracy": 0.26284928172826766,
      "num_tokens": 192716742.0,
      "step": 84050
    },
    {
      "entropy": 5.460582733154297,
      "epoch": 8.074447646493756,
      "grad_norm": 1.53125,
      "learning_rate": 9.06543093131372e-05,
      "loss": 4.5629,
      "mean_token_accuracy": 0.27455626130104066,
      "num_tokens": 192728240.0,
      "step": 84055
    },
    {
      "entropy": 5.433889532089234,
      "epoch": 8.07492795389049,
      "grad_norm": 1.28125,
      "learning_rate": 9.063465515452765e-05,
      "loss": 4.5163,
      "mean_token_accuracy": 0.2763267457485199,
      "num_tokens": 192739309.0,
      "step": 84060
    },
    {
      "entropy": 5.538053417205811,
      "epoch": 8.075408261287224,
      "grad_norm": 1.4140625,
      "learning_rate": 9.061500527633387e-05,
      "loss": 4.615,
      "mean_token_accuracy": 0.268973246216774,
      "num_tokens": 192751655.0,
      "step": 84065
    },
    {
      "entropy": 5.523343420028686,
      "epoch": 8.075888568683958,
      "grad_norm": 1.3828125,
      "learning_rate": 9.059535967901217e-05,
      "loss": 4.6393,
      "mean_token_accuracy": 0.26814412176609037,
      "num_tokens": 192763614.0,
      "step": 84070
    },
    {
      "entropy": 5.656684923171997,
      "epoch": 8.076368876080691,
      "grad_norm": 1.5390625,
      "learning_rate": 9.057571836301864e-05,
      "loss": 4.7783,
      "mean_token_accuracy": 0.2538502424955368,
      "num_tokens": 192775342.0,
      "step": 84075
    },
    {
      "entropy": 5.49628005027771,
      "epoch": 8.076849183477426,
      "grad_norm": 1.5,
      "learning_rate": 9.055608132880932e-05,
      "loss": 4.5489,
      "mean_token_accuracy": 0.2770990788936615,
      "num_tokens": 192787561.0,
      "step": 84080
    },
    {
      "entropy": 5.465510988235474,
      "epoch": 8.07732949087416,
      "grad_norm": 1.46875,
      "learning_rate": 9.053644857684006e-05,
      "loss": 4.5055,
      "mean_token_accuracy": 0.2712444573640823,
      "num_tokens": 192798953.0,
      "step": 84085
    },
    {
      "entropy": 5.4854480743408205,
      "epoch": 8.077809798270893,
      "grad_norm": 1.484375,
      "learning_rate": 9.051682010756683e-05,
      "loss": 4.5473,
      "mean_token_accuracy": 0.27487661838531496,
      "num_tokens": 192810741.0,
      "step": 84090
    },
    {
      "entropy": 5.487144184112549,
      "epoch": 8.078290105667627,
      "grad_norm": 1.40625,
      "learning_rate": 9.049719592144518e-05,
      "loss": 4.5598,
      "mean_token_accuracy": 0.26723127216100695,
      "num_tokens": 192822740.0,
      "step": 84095
    },
    {
      "entropy": 5.540175056457519,
      "epoch": 8.078770413064362,
      "grad_norm": 1.34375,
      "learning_rate": 9.047757601893075e-05,
      "loss": 4.5936,
      "mean_token_accuracy": 0.2669388234615326,
      "num_tokens": 192835793.0,
      "step": 84100
    },
    {
      "entropy": 5.5683225154876705,
      "epoch": 8.079250720461095,
      "grad_norm": 1.7109375,
      "learning_rate": 9.045796040047908e-05,
      "loss": 4.6731,
      "mean_token_accuracy": 0.2648232221603394,
      "num_tokens": 192846029.0,
      "step": 84105
    },
    {
      "entropy": 5.5620804786682125,
      "epoch": 8.079731027857829,
      "grad_norm": 1.40625,
      "learning_rate": 9.043834906654561e-05,
      "loss": 4.6069,
      "mean_token_accuracy": 0.2745808094739914,
      "num_tokens": 192857259.0,
      "step": 84110
    },
    {
      "entropy": 5.536179828643799,
      "epoch": 8.080211335254562,
      "grad_norm": 1.5078125,
      "learning_rate": 9.041874201758563e-05,
      "loss": 4.6017,
      "mean_token_accuracy": 0.27113409489393236,
      "num_tokens": 192867712.0,
      "step": 84115
    },
    {
      "entropy": 5.5150481224060055,
      "epoch": 8.080691642651297,
      "grad_norm": 1.4765625,
      "learning_rate": 9.039913925405444e-05,
      "loss": 4.4909,
      "mean_token_accuracy": 0.27521292567253114,
      "num_tokens": 192879080.0,
      "step": 84120
    },
    {
      "entropy": 5.525774621963501,
      "epoch": 8.08117195004803,
      "grad_norm": 1.359375,
      "learning_rate": 9.037954077640701e-05,
      "loss": 4.6384,
      "mean_token_accuracy": 0.27213838547468183,
      "num_tokens": 192891202.0,
      "step": 84125
    },
    {
      "entropy": 5.587141466140747,
      "epoch": 8.081652257444764,
      "grad_norm": 1.375,
      "learning_rate": 9.035994658509844e-05,
      "loss": 4.6908,
      "mean_token_accuracy": 0.25712289065122607,
      "num_tokens": 192902601.0,
      "step": 84130
    },
    {
      "entropy": 5.482535934448242,
      "epoch": 8.082132564841498,
      "grad_norm": 1.484375,
      "learning_rate": 9.034035668058363e-05,
      "loss": 4.5177,
      "mean_token_accuracy": 0.28406750559806826,
      "num_tokens": 192913556.0,
      "step": 84135
    },
    {
      "entropy": 5.567949724197388,
      "epoch": 8.082612872238233,
      "grad_norm": 1.4453125,
      "learning_rate": 9.032077106331743e-05,
      "loss": 4.6601,
      "mean_token_accuracy": 0.25962253510951994,
      "num_tokens": 192925086.0,
      "step": 84140
    },
    {
      "entropy": 5.436152791976928,
      "epoch": 8.083093179634966,
      "grad_norm": 1.3359375,
      "learning_rate": 9.030118973375453e-05,
      "loss": 4.5096,
      "mean_token_accuracy": 0.2747693434357643,
      "num_tokens": 192937112.0,
      "step": 84145
    },
    {
      "entropy": 5.57336916923523,
      "epoch": 8.0835734870317,
      "grad_norm": 1.359375,
      "learning_rate": 9.028161269234959e-05,
      "loss": 4.742,
      "mean_token_accuracy": 0.25753289759159087,
      "num_tokens": 192948584.0,
      "step": 84150
    },
    {
      "entropy": 5.571690940856934,
      "epoch": 8.084053794428435,
      "grad_norm": 1.421875,
      "learning_rate": 9.026203993955711e-05,
      "loss": 4.6241,
      "mean_token_accuracy": 0.2707853585481644,
      "num_tokens": 192959047.0,
      "step": 84155
    },
    {
      "entropy": 5.578487586975098,
      "epoch": 8.084534101825168,
      "grad_norm": 1.5,
      "learning_rate": 9.024247147583147e-05,
      "loss": 4.6106,
      "mean_token_accuracy": 0.2710350975394249,
      "num_tokens": 192969403.0,
      "step": 84160
    },
    {
      "entropy": 5.546055650711059,
      "epoch": 8.085014409221902,
      "grad_norm": 1.34375,
      "learning_rate": 9.022290730162705e-05,
      "loss": 4.6283,
      "mean_token_accuracy": 0.26974950432777406,
      "num_tokens": 192980742.0,
      "step": 84165
    },
    {
      "entropy": 5.498474264144898,
      "epoch": 8.085494716618635,
      "grad_norm": 1.4921875,
      "learning_rate": 9.020334741739802e-05,
      "loss": 4.567,
      "mean_token_accuracy": 0.27267482727766035,
      "num_tokens": 192991920.0,
      "step": 84170
    },
    {
      "entropy": 5.4579833984375,
      "epoch": 8.08597502401537,
      "grad_norm": 1.6015625,
      "learning_rate": 9.018379182359854e-05,
      "loss": 4.4504,
      "mean_token_accuracy": 0.2864324599504471,
      "num_tokens": 193003018.0,
      "step": 84175
    },
    {
      "entropy": 5.508915567398072,
      "epoch": 8.086455331412104,
      "grad_norm": 1.2265625,
      "learning_rate": 9.016424052068262e-05,
      "loss": 4.5725,
      "mean_token_accuracy": 0.27337910830974577,
      "num_tokens": 193016790.0,
      "step": 84180
    },
    {
      "entropy": 5.514945411682129,
      "epoch": 8.086935638808837,
      "grad_norm": 1.421875,
      "learning_rate": 9.014469350910417e-05,
      "loss": 4.5731,
      "mean_token_accuracy": 0.2754438415169716,
      "num_tokens": 193027326.0,
      "step": 84185
    },
    {
      "entropy": 5.545679521560669,
      "epoch": 8.08741594620557,
      "grad_norm": 1.453125,
      "learning_rate": 9.012515078931705e-05,
      "loss": 4.6428,
      "mean_token_accuracy": 0.26476057022809985,
      "num_tokens": 193038741.0,
      "step": 84190
    },
    {
      "entropy": 5.496673250198365,
      "epoch": 8.087896253602306,
      "grad_norm": 1.4140625,
      "learning_rate": 9.010561236177502e-05,
      "loss": 4.5573,
      "mean_token_accuracy": 0.266383658349514,
      "num_tokens": 193049662.0,
      "step": 84195
    },
    {
      "entropy": 5.465846824645996,
      "epoch": 8.08837656099904,
      "grad_norm": 1.3671875,
      "learning_rate": 9.008607822693156e-05,
      "loss": 4.5507,
      "mean_token_accuracy": 0.27517669051885607,
      "num_tokens": 193061625.0,
      "step": 84200
    },
    {
      "entropy": 5.539892482757568,
      "epoch": 8.088856868395773,
      "grad_norm": 1.4140625,
      "learning_rate": 9.006654838524031e-05,
      "loss": 4.635,
      "mean_token_accuracy": 0.26284138262271883,
      "num_tokens": 193071753.0,
      "step": 84205
    },
    {
      "entropy": 5.414966154098511,
      "epoch": 8.089337175792506,
      "grad_norm": 1.484375,
      "learning_rate": 9.004702283715465e-05,
      "loss": 4.4715,
      "mean_token_accuracy": 0.27728925347328187,
      "num_tokens": 193082480.0,
      "step": 84210
    },
    {
      "entropy": 5.517971515655518,
      "epoch": 8.089817483189242,
      "grad_norm": 1.4765625,
      "learning_rate": 9.002750158312795e-05,
      "loss": 4.6558,
      "mean_token_accuracy": 0.2647035837173462,
      "num_tokens": 193093715.0,
      "step": 84215
    },
    {
      "entropy": 5.468937063217163,
      "epoch": 8.090297790585975,
      "grad_norm": 1.40625,
      "learning_rate": 9.000798462361338e-05,
      "loss": 4.5353,
      "mean_token_accuracy": 0.27408905029296876,
      "num_tokens": 193104750.0,
      "step": 84220
    },
    {
      "entropy": 5.536163759231568,
      "epoch": 8.090778097982708,
      "grad_norm": 1.5859375,
      "learning_rate": 8.998847195906412e-05,
      "loss": 4.6207,
      "mean_token_accuracy": 0.26355637460947035,
      "num_tokens": 193116344.0,
      "step": 84225
    },
    {
      "entropy": 5.550725984573364,
      "epoch": 8.091258405379444,
      "grad_norm": 1.5390625,
      "learning_rate": 8.996896358993315e-05,
      "loss": 4.6837,
      "mean_token_accuracy": 0.260216149687767,
      "num_tokens": 193127888.0,
      "step": 84230
    },
    {
      "entropy": 5.451743412017822,
      "epoch": 8.091738712776177,
      "grad_norm": 1.4140625,
      "learning_rate": 8.994945951667347e-05,
      "loss": 4.5123,
      "mean_token_accuracy": 0.27323407828807833,
      "num_tokens": 193139365.0,
      "step": 84235
    },
    {
      "entropy": 5.509535789489746,
      "epoch": 8.09221902017291,
      "grad_norm": 1.4921875,
      "learning_rate": 8.992995973973781e-05,
      "loss": 4.6417,
      "mean_token_accuracy": 0.2638474851846695,
      "num_tokens": 193151465.0,
      "step": 84240
    },
    {
      "entropy": 5.60322060585022,
      "epoch": 8.092699327569644,
      "grad_norm": 1.4453125,
      "learning_rate": 8.991046425957894e-05,
      "loss": 4.622,
      "mean_token_accuracy": 0.2708221435546875,
      "num_tokens": 193161748.0,
      "step": 84245
    },
    {
      "entropy": 5.531693601608277,
      "epoch": 8.09317963496638,
      "grad_norm": 1.4609375,
      "learning_rate": 8.989097307664949e-05,
      "loss": 4.632,
      "mean_token_accuracy": 0.26538914293050764,
      "num_tokens": 193173196.0,
      "step": 84250
    },
    {
      "entropy": 5.502986669540405,
      "epoch": 8.093659942363113,
      "grad_norm": 1.4296875,
      "learning_rate": 8.987148619140195e-05,
      "loss": 4.5696,
      "mean_token_accuracy": 0.2782996416091919,
      "num_tokens": 193184946.0,
      "step": 84255
    },
    {
      "entropy": 5.51466999053955,
      "epoch": 8.094140249759846,
      "grad_norm": 1.3203125,
      "learning_rate": 8.98520036042888e-05,
      "loss": 4.6162,
      "mean_token_accuracy": 0.27002316415309907,
      "num_tokens": 193196691.0,
      "step": 84260
    },
    {
      "entropy": 5.546251440048218,
      "epoch": 8.09462055715658,
      "grad_norm": 1.5,
      "learning_rate": 8.983252531576238e-05,
      "loss": 4.5924,
      "mean_token_accuracy": 0.26741712987422944,
      "num_tokens": 193207517.0,
      "step": 84265
    },
    {
      "entropy": 5.481799840927124,
      "epoch": 8.095100864553315,
      "grad_norm": 1.4765625,
      "learning_rate": 8.981305132627487e-05,
      "loss": 4.5065,
      "mean_token_accuracy": 0.27568162977695465,
      "num_tokens": 193219148.0,
      "step": 84270
    },
    {
      "entropy": 5.508148241043091,
      "epoch": 8.095581171950048,
      "grad_norm": 1.484375,
      "learning_rate": 8.979358163627845e-05,
      "loss": 4.625,
      "mean_token_accuracy": 0.2673797994852066,
      "num_tokens": 193230929.0,
      "step": 84275
    },
    {
      "entropy": 5.501679992675781,
      "epoch": 8.096061479346782,
      "grad_norm": 1.484375,
      "learning_rate": 8.977411624622509e-05,
      "loss": 4.5839,
      "mean_token_accuracy": 0.2736998379230499,
      "num_tokens": 193242521.0,
      "step": 84280
    },
    {
      "entropy": 5.5530540466308596,
      "epoch": 8.096541786743515,
      "grad_norm": 1.421875,
      "learning_rate": 8.975465515656673e-05,
      "loss": 4.6259,
      "mean_token_accuracy": 0.2639204040169716,
      "num_tokens": 193253958.0,
      "step": 84285
    },
    {
      "entropy": 5.520121002197266,
      "epoch": 8.09702209414025,
      "grad_norm": 1.359375,
      "learning_rate": 8.97351983677552e-05,
      "loss": 4.6186,
      "mean_token_accuracy": 0.266881899535656,
      "num_tokens": 193266279.0,
      "step": 84290
    },
    {
      "entropy": 5.554859399795532,
      "epoch": 8.097502401536984,
      "grad_norm": 1.5,
      "learning_rate": 8.971574588024224e-05,
      "loss": 4.6068,
      "mean_token_accuracy": 0.2674616023898125,
      "num_tokens": 193278281.0,
      "step": 84295
    },
    {
      "entropy": 5.566569375991821,
      "epoch": 8.097982708933717,
      "grad_norm": 1.3828125,
      "learning_rate": 8.969629769447949e-05,
      "loss": 4.6362,
      "mean_token_accuracy": 0.26432667672634125,
      "num_tokens": 193290366.0,
      "step": 84300
    },
    {
      "entropy": 5.541655158996582,
      "epoch": 8.098463016330452,
      "grad_norm": 1.4140625,
      "learning_rate": 8.967685381091853e-05,
      "loss": 4.5568,
      "mean_token_accuracy": 0.2743175119161606,
      "num_tokens": 193300828.0,
      "step": 84305
    },
    {
      "entropy": 5.468429851531982,
      "epoch": 8.098943323727186,
      "grad_norm": 1.4453125,
      "learning_rate": 8.965741423001066e-05,
      "loss": 4.5438,
      "mean_token_accuracy": 0.27497767508029936,
      "num_tokens": 193311996.0,
      "step": 84310
    },
    {
      "entropy": 5.494264364242554,
      "epoch": 8.09942363112392,
      "grad_norm": 1.3828125,
      "learning_rate": 8.963797895220728e-05,
      "loss": 4.5547,
      "mean_token_accuracy": 0.2678613856434822,
      "num_tokens": 193324286.0,
      "step": 84315
    },
    {
      "entropy": 5.471817064285278,
      "epoch": 8.099903938520653,
      "grad_norm": 1.4296875,
      "learning_rate": 8.961854797795965e-05,
      "loss": 4.5483,
      "mean_token_accuracy": 0.2782577395439148,
      "num_tokens": 193335462.0,
      "step": 84320
    },
    {
      "entropy": 5.537335920333862,
      "epoch": 8.100384245917388,
      "grad_norm": 1.4609375,
      "learning_rate": 8.95991213077188e-05,
      "loss": 4.5887,
      "mean_token_accuracy": 0.27581228464841845,
      "num_tokens": 193346387.0,
      "step": 84325
    },
    {
      "entropy": 5.521111726760864,
      "epoch": 8.100864553314121,
      "grad_norm": 1.296875,
      "learning_rate": 8.957969894193588e-05,
      "loss": 4.6397,
      "mean_token_accuracy": 0.271431103348732,
      "num_tokens": 193358458.0,
      "step": 84330
    },
    {
      "entropy": 5.604614734649658,
      "epoch": 8.101344860710855,
      "grad_norm": 1.5234375,
      "learning_rate": 8.956028088106182e-05,
      "loss": 4.6779,
      "mean_token_accuracy": 0.2620885208249092,
      "num_tokens": 193369998.0,
      "step": 84335
    },
    {
      "entropy": 5.552932834625244,
      "epoch": 8.101825168107588,
      "grad_norm": 1.4140625,
      "learning_rate": 8.954086712554732e-05,
      "loss": 4.5687,
      "mean_token_accuracy": 0.27361889034509657,
      "num_tokens": 193382232.0,
      "step": 84340
    },
    {
      "entropy": 5.503530263900757,
      "epoch": 8.102305475504323,
      "grad_norm": 1.5703125,
      "learning_rate": 8.952145767584317e-05,
      "loss": 4.5927,
      "mean_token_accuracy": 0.2734806627035141,
      "num_tokens": 193393367.0,
      "step": 84345
    },
    {
      "entropy": 5.487341451644897,
      "epoch": 8.102785782901057,
      "grad_norm": 1.5390625,
      "learning_rate": 8.950205253240003e-05,
      "loss": 4.5946,
      "mean_token_accuracy": 0.2681694135069847,
      "num_tokens": 193404972.0,
      "step": 84350
    },
    {
      "entropy": 5.45516185760498,
      "epoch": 8.10326609029779,
      "grad_norm": 1.4375,
      "learning_rate": 8.948265169566843e-05,
      "loss": 4.5322,
      "mean_token_accuracy": 0.27751549780368806,
      "num_tokens": 193416519.0,
      "step": 84355
    },
    {
      "entropy": 5.5809601783752445,
      "epoch": 8.103746397694524,
      "grad_norm": 1.5703125,
      "learning_rate": 8.946325516609876e-05,
      "loss": 4.6236,
      "mean_token_accuracy": 0.27070059776306155,
      "num_tokens": 193426787.0,
      "step": 84360
    },
    {
      "entropy": 5.531977748870849,
      "epoch": 8.104226705091259,
      "grad_norm": 1.4296875,
      "learning_rate": 8.944386294414143e-05,
      "loss": 4.6138,
      "mean_token_accuracy": 0.27566270530223846,
      "num_tokens": 193439171.0,
      "step": 84365
    },
    {
      "entropy": 5.50751748085022,
      "epoch": 8.104707012487992,
      "grad_norm": 1.3515625,
      "learning_rate": 8.942447503024657e-05,
      "loss": 4.6134,
      "mean_token_accuracy": 0.27057713121175764,
      "num_tokens": 193450347.0,
      "step": 84370
    },
    {
      "entropy": 5.540218830108643,
      "epoch": 8.105187319884726,
      "grad_norm": 1.2890625,
      "learning_rate": 8.940509142486436e-05,
      "loss": 4.6443,
      "mean_token_accuracy": 0.27916384041309356,
      "num_tokens": 193462717.0,
      "step": 84375
    },
    {
      "entropy": 5.512127208709717,
      "epoch": 8.105667627281461,
      "grad_norm": 1.578125,
      "learning_rate": 8.938571212844482e-05,
      "loss": 4.5867,
      "mean_token_accuracy": 0.26418326497077943,
      "num_tokens": 193474087.0,
      "step": 84380
    },
    {
      "entropy": 5.490594434738159,
      "epoch": 8.106147934678194,
      "grad_norm": 1.5625,
      "learning_rate": 8.936633714143788e-05,
      "loss": 4.5871,
      "mean_token_accuracy": 0.26602543890476227,
      "num_tokens": 193485463.0,
      "step": 84385
    },
    {
      "entropy": 5.388792514801025,
      "epoch": 8.106628242074928,
      "grad_norm": 1.5546875,
      "learning_rate": 8.934696646429337e-05,
      "loss": 4.4715,
      "mean_token_accuracy": 0.28514143228530886,
      "num_tokens": 193497019.0,
      "step": 84390
    },
    {
      "entropy": 5.526931858062744,
      "epoch": 8.107108549471661,
      "grad_norm": 1.53125,
      "learning_rate": 8.932760009746107e-05,
      "loss": 4.5849,
      "mean_token_accuracy": 0.2769397243857384,
      "num_tokens": 193507166.0,
      "step": 84395
    },
    {
      "entropy": 5.517662906646729,
      "epoch": 8.107588856868396,
      "grad_norm": 1.359375,
      "learning_rate": 8.930823804139052e-05,
      "loss": 4.5963,
      "mean_token_accuracy": 0.2709119960665703,
      "num_tokens": 193518812.0,
      "step": 84400
    },
    {
      "entropy": 5.504832458496094,
      "epoch": 8.10806916426513,
      "grad_norm": 1.4296875,
      "learning_rate": 8.928888029653128e-05,
      "loss": 4.5961,
      "mean_token_accuracy": 0.2709937259554863,
      "num_tokens": 193530327.0,
      "step": 84405
    },
    {
      "entropy": 5.491122245788574,
      "epoch": 8.108549471661863,
      "grad_norm": 1.4921875,
      "learning_rate": 8.926952686333282e-05,
      "loss": 4.5161,
      "mean_token_accuracy": 0.2784626170992851,
      "num_tokens": 193540727.0,
      "step": 84410
    },
    {
      "entropy": 5.584831857681275,
      "epoch": 8.109029779058597,
      "grad_norm": 1.375,
      "learning_rate": 8.925017774224443e-05,
      "loss": 4.6368,
      "mean_token_accuracy": 0.2661813646554947,
      "num_tokens": 193552441.0,
      "step": 84415
    },
    {
      "entropy": 5.483498620986938,
      "epoch": 8.109510086455332,
      "grad_norm": 1.4921875,
      "learning_rate": 8.923083293371531e-05,
      "loss": 4.5287,
      "mean_token_accuracy": 0.2738815113902092,
      "num_tokens": 193564548.0,
      "step": 84420
    },
    {
      "entropy": 5.531073093414307,
      "epoch": 8.109990393852065,
      "grad_norm": 1.5234375,
      "learning_rate": 8.921149243819474e-05,
      "loss": 4.6089,
      "mean_token_accuracy": 0.27456180602312086,
      "num_tokens": 193576470.0,
      "step": 84425
    },
    {
      "entropy": 5.510556602478028,
      "epoch": 8.110470701248799,
      "grad_norm": 1.46875,
      "learning_rate": 8.919215625613155e-05,
      "loss": 4.626,
      "mean_token_accuracy": 0.26976933628320693,
      "num_tokens": 193587584.0,
      "step": 84430
    },
    {
      "entropy": 5.526956462860108,
      "epoch": 8.110951008645532,
      "grad_norm": 1.3984375,
      "learning_rate": 8.917282438797475e-05,
      "loss": 4.6129,
      "mean_token_accuracy": 0.2671120584011078,
      "num_tokens": 193600166.0,
      "step": 84435
    },
    {
      "entropy": 5.452800607681274,
      "epoch": 8.111431316042268,
      "grad_norm": 1.4453125,
      "learning_rate": 8.915349683417319e-05,
      "loss": 4.5277,
      "mean_token_accuracy": 0.2787966191768646,
      "num_tokens": 193611330.0,
      "step": 84440
    },
    {
      "entropy": 5.561955976486206,
      "epoch": 8.111911623439001,
      "grad_norm": 1.4453125,
      "learning_rate": 8.913417359517556e-05,
      "loss": 4.6106,
      "mean_token_accuracy": 0.27128954529762267,
      "num_tokens": 193623362.0,
      "step": 84445
    },
    {
      "entropy": 5.578621673583984,
      "epoch": 8.112391930835734,
      "grad_norm": 1.53125,
      "learning_rate": 8.911485467143053e-05,
      "loss": 4.6147,
      "mean_token_accuracy": 0.2645523682236671,
      "num_tokens": 193634477.0,
      "step": 84450
    },
    {
      "entropy": 5.475225639343262,
      "epoch": 8.11287223823247,
      "grad_norm": 1.4609375,
      "learning_rate": 8.909554006338666e-05,
      "loss": 4.5431,
      "mean_token_accuracy": 0.27232865691185,
      "num_tokens": 193644336.0,
      "step": 84455
    },
    {
      "entropy": 5.684483337402344,
      "epoch": 8.113352545629203,
      "grad_norm": 1.7265625,
      "learning_rate": 8.907622977149229e-05,
      "loss": 4.8445,
      "mean_token_accuracy": 0.25057041496038435,
      "num_tokens": 193656935.0,
      "step": 84460
    },
    {
      "entropy": 5.566836404800415,
      "epoch": 8.113832853025936,
      "grad_norm": 1.4609375,
      "learning_rate": 8.905692379619576e-05,
      "loss": 4.6064,
      "mean_token_accuracy": 0.26014402508735657,
      "num_tokens": 193669016.0,
      "step": 84465
    },
    {
      "entropy": 5.615290546417237,
      "epoch": 8.11431316042267,
      "grad_norm": 1.625,
      "learning_rate": 8.903762213794534e-05,
      "loss": 4.6818,
      "mean_token_accuracy": 0.2634595394134521,
      "num_tokens": 193680445.0,
      "step": 84470
    },
    {
      "entropy": 5.453089094161987,
      "epoch": 8.114793467819405,
      "grad_norm": 1.421875,
      "learning_rate": 8.901832479718912e-05,
      "loss": 4.4943,
      "mean_token_accuracy": 0.27799856662750244,
      "num_tokens": 193691235.0,
      "step": 84475
    },
    {
      "entropy": 5.464501571655274,
      "epoch": 8.115273775216139,
      "grad_norm": 1.3984375,
      "learning_rate": 8.899903177437516e-05,
      "loss": 4.5588,
      "mean_token_accuracy": 0.27652690559625626,
      "num_tokens": 193703697.0,
      "step": 84480
    },
    {
      "entropy": 5.566814947128296,
      "epoch": 8.115754082612872,
      "grad_norm": 1.5078125,
      "learning_rate": 8.897974306995141e-05,
      "loss": 4.6952,
      "mean_token_accuracy": 0.26632480770349504,
      "num_tokens": 193715038.0,
      "step": 84485
    },
    {
      "entropy": 5.50474705696106,
      "epoch": 8.116234390009605,
      "grad_norm": 1.4453125,
      "learning_rate": 8.896045868436564e-05,
      "loss": 4.5599,
      "mean_token_accuracy": 0.2723591536283493,
      "num_tokens": 193727775.0,
      "step": 84490
    },
    {
      "entropy": 5.533016729354858,
      "epoch": 8.11671469740634,
      "grad_norm": 1.3828125,
      "learning_rate": 8.894117861806556e-05,
      "loss": 4.6222,
      "mean_token_accuracy": 0.26754132062196734,
      "num_tokens": 193739101.0,
      "step": 84495
    },
    {
      "entropy": 5.565924119949341,
      "epoch": 8.117195004803074,
      "grad_norm": 1.5390625,
      "learning_rate": 8.892190287149885e-05,
      "loss": 4.6801,
      "mean_token_accuracy": 0.25491299778223037,
      "num_tokens": 193750898.0,
      "step": 84500
    },
    {
      "entropy": 5.511015367507935,
      "epoch": 8.117675312199808,
      "grad_norm": 1.4375,
      "learning_rate": 8.890263144511302e-05,
      "loss": 4.5467,
      "mean_token_accuracy": 0.2701493248343468,
      "num_tokens": 193762339.0,
      "step": 84505
    },
    {
      "entropy": 5.528918790817261,
      "epoch": 8.118155619596541,
      "grad_norm": 1.4140625,
      "learning_rate": 8.888336433935549e-05,
      "loss": 4.6497,
      "mean_token_accuracy": 0.2641819566488266,
      "num_tokens": 193773526.0,
      "step": 84510
    },
    {
      "entropy": 5.459507846832276,
      "epoch": 8.118635926993276,
      "grad_norm": 1.390625,
      "learning_rate": 8.886410155467361e-05,
      "loss": 4.4821,
      "mean_token_accuracy": 0.27481296062469485,
      "num_tokens": 193784945.0,
      "step": 84515
    },
    {
      "entropy": 5.434965515136719,
      "epoch": 8.11911623439001,
      "grad_norm": 1.40625,
      "learning_rate": 8.884484309151455e-05,
      "loss": 4.4889,
      "mean_token_accuracy": 0.28637961447238924,
      "num_tokens": 193796791.0,
      "step": 84520
    },
    {
      "entropy": 5.5307371616363525,
      "epoch": 8.119596541786743,
      "grad_norm": 1.4375,
      "learning_rate": 8.882558895032547e-05,
      "loss": 4.6216,
      "mean_token_accuracy": 0.26171208173036575,
      "num_tokens": 193808604.0,
      "step": 84525
    },
    {
      "entropy": 5.598495149612427,
      "epoch": 8.120076849183478,
      "grad_norm": 1.453125,
      "learning_rate": 8.88063391315534e-05,
      "loss": 4.6505,
      "mean_token_accuracy": 0.2584605172276497,
      "num_tokens": 193819099.0,
      "step": 84530
    },
    {
      "entropy": 5.508911561965943,
      "epoch": 8.120557156580212,
      "grad_norm": 1.40625,
      "learning_rate": 8.878709363564525e-05,
      "loss": 4.5715,
      "mean_token_accuracy": 0.27684385925531385,
      "num_tokens": 193830045.0,
      "step": 84535
    },
    {
      "entropy": 5.584780168533325,
      "epoch": 8.121037463976945,
      "grad_norm": 1.5390625,
      "learning_rate": 8.876785246304784e-05,
      "loss": 4.5876,
      "mean_token_accuracy": 0.2678858324885368,
      "num_tokens": 193840612.0,
      "step": 84540
    },
    {
      "entropy": 5.542997217178344,
      "epoch": 8.121517771373679,
      "grad_norm": 1.3984375,
      "learning_rate": 8.874861561420794e-05,
      "loss": 4.6159,
      "mean_token_accuracy": 0.2720154970884323,
      "num_tokens": 193853057.0,
      "step": 84545
    },
    {
      "entropy": 5.506246280670166,
      "epoch": 8.121998078770414,
      "grad_norm": 1.453125,
      "learning_rate": 8.872938308957209e-05,
      "loss": 4.626,
      "mean_token_accuracy": 0.2698985695838928,
      "num_tokens": 193865126.0,
      "step": 84550
    },
    {
      "entropy": 5.561378145217896,
      "epoch": 8.122478386167147,
      "grad_norm": 1.34375,
      "learning_rate": 8.871015488958687e-05,
      "loss": 4.7132,
      "mean_token_accuracy": 0.26120131015777587,
      "num_tokens": 193877395.0,
      "step": 84555
    },
    {
      "entropy": 5.53751392364502,
      "epoch": 8.12295869356388,
      "grad_norm": 1.4375,
      "learning_rate": 8.869093101469866e-05,
      "loss": 4.5852,
      "mean_token_accuracy": 0.27573115527629855,
      "num_tokens": 193888517.0,
      "step": 84560
    },
    {
      "entropy": 5.5753734588623045,
      "epoch": 8.123439000960614,
      "grad_norm": 1.4765625,
      "learning_rate": 8.867171146535383e-05,
      "loss": 4.6419,
      "mean_token_accuracy": 0.271712501347065,
      "num_tokens": 193899827.0,
      "step": 84565
    },
    {
      "entropy": 5.509001350402832,
      "epoch": 8.12391930835735,
      "grad_norm": 1.390625,
      "learning_rate": 8.865249624199856e-05,
      "loss": 4.5951,
      "mean_token_accuracy": 0.2710789993405342,
      "num_tokens": 193910844.0,
      "step": 84570
    },
    {
      "entropy": 5.44781084060669,
      "epoch": 8.124399615754083,
      "grad_norm": 1.328125,
      "learning_rate": 8.863328534507906e-05,
      "loss": 4.4601,
      "mean_token_accuracy": 0.2793524980545044,
      "num_tokens": 193922395.0,
      "step": 84575
    },
    {
      "entropy": 5.440123605728149,
      "epoch": 8.124879923150816,
      "grad_norm": 1.3984375,
      "learning_rate": 8.861407877504121e-05,
      "loss": 4.5061,
      "mean_token_accuracy": 0.277667498588562,
      "num_tokens": 193933793.0,
      "step": 84580
    },
    {
      "entropy": 5.46043586730957,
      "epoch": 8.12536023054755,
      "grad_norm": 1.46875,
      "learning_rate": 8.8594876532331e-05,
      "loss": 4.5316,
      "mean_token_accuracy": 0.2785508632659912,
      "num_tokens": 193944682.0,
      "step": 84585
    },
    {
      "entropy": 5.462030553817749,
      "epoch": 8.125840537944285,
      "grad_norm": 1.3203125,
      "learning_rate": 8.857567861739426e-05,
      "loss": 4.5813,
      "mean_token_accuracy": 0.27228794544935225,
      "num_tokens": 193956974.0,
      "step": 84590
    },
    {
      "entropy": 5.541478252410888,
      "epoch": 8.126320845341018,
      "grad_norm": 1.5625,
      "learning_rate": 8.85564850306767e-05,
      "loss": 4.6404,
      "mean_token_accuracy": 0.2689012184739113,
      "num_tokens": 193968401.0,
      "step": 84595
    },
    {
      "entropy": 5.5472417831420895,
      "epoch": 8.126801152737752,
      "grad_norm": 1.484375,
      "learning_rate": 8.853729577262392e-05,
      "loss": 4.6108,
      "mean_token_accuracy": 0.2764489710330963,
      "num_tokens": 193978431.0,
      "step": 84600
    },
    {
      "entropy": 5.513146829605103,
      "epoch": 8.127281460134487,
      "grad_norm": 1.40625,
      "learning_rate": 8.85181108436815e-05,
      "loss": 4.6468,
      "mean_token_accuracy": 0.2611275494098663,
      "num_tokens": 193989336.0,
      "step": 84605
    },
    {
      "entropy": 5.521677780151367,
      "epoch": 8.12776176753122,
      "grad_norm": 1.40625,
      "learning_rate": 8.849893024429478e-05,
      "loss": 4.6321,
      "mean_token_accuracy": 0.268670080602169,
      "num_tokens": 193999914.0,
      "step": 84610
    },
    {
      "entropy": 5.549580097198486,
      "epoch": 8.128242074927954,
      "grad_norm": 1.375,
      "learning_rate": 8.84797539749091e-05,
      "loss": 4.5971,
      "mean_token_accuracy": 0.28102810084819796,
      "num_tokens": 194011432.0,
      "step": 84615
    },
    {
      "entropy": 5.5600488662719725,
      "epoch": 8.128722382324687,
      "grad_norm": 1.3203125,
      "learning_rate": 8.846058203596965e-05,
      "loss": 4.6254,
      "mean_token_accuracy": 0.26639800816774367,
      "num_tokens": 194023843.0,
      "step": 84620
    },
    {
      "entropy": 5.486083316802978,
      "epoch": 8.129202689721422,
      "grad_norm": 1.4296875,
      "learning_rate": 8.84414144279216e-05,
      "loss": 4.5457,
      "mean_token_accuracy": 0.275239597260952,
      "num_tokens": 194034634.0,
      "step": 84625
    },
    {
      "entropy": 5.506525039672852,
      "epoch": 8.129682997118156,
      "grad_norm": 1.46875,
      "learning_rate": 8.842225115120994e-05,
      "loss": 4.5964,
      "mean_token_accuracy": 0.26784533858299253,
      "num_tokens": 194045735.0,
      "step": 84630
    },
    {
      "entropy": 5.515843725204467,
      "epoch": 8.13016330451489,
      "grad_norm": 1.5,
      "learning_rate": 8.840309220627965e-05,
      "loss": 4.5636,
      "mean_token_accuracy": 0.26696925461292265,
      "num_tokens": 194057124.0,
      "step": 84635
    },
    {
      "entropy": 5.493158674240112,
      "epoch": 8.130643611911623,
      "grad_norm": 1.4140625,
      "learning_rate": 8.838393759357541e-05,
      "loss": 4.6051,
      "mean_token_accuracy": 0.2684337332844734,
      "num_tokens": 194067752.0,
      "step": 84640
    },
    {
      "entropy": 5.609033203125,
      "epoch": 8.131123919308358,
      "grad_norm": 1.5546875,
      "learning_rate": 8.836478731354202e-05,
      "loss": 4.7366,
      "mean_token_accuracy": 0.25775035172700883,
      "num_tokens": 194077685.0,
      "step": 84645
    },
    {
      "entropy": 5.493514442443848,
      "epoch": 8.131604226705091,
      "grad_norm": 1.453125,
      "learning_rate": 8.834564136662404e-05,
      "loss": 4.5521,
      "mean_token_accuracy": 0.27914465963840485,
      "num_tokens": 194088881.0,
      "step": 84650
    },
    {
      "entropy": 5.482553911209107,
      "epoch": 8.132084534101825,
      "grad_norm": 1.71875,
      "learning_rate": 8.832649975326607e-05,
      "loss": 4.5585,
      "mean_token_accuracy": 0.2714832276105881,
      "num_tokens": 194099588.0,
      "step": 84655
    },
    {
      "entropy": 5.657704830169678,
      "epoch": 8.132564841498558,
      "grad_norm": 1.4765625,
      "learning_rate": 8.830736247391242e-05,
      "loss": 4.7206,
      "mean_token_accuracy": 0.25674007683992384,
      "num_tokens": 194109889.0,
      "step": 84660
    },
    {
      "entropy": 5.458453035354614,
      "epoch": 8.133045148895294,
      "grad_norm": 1.3125,
      "learning_rate": 8.828822952900753e-05,
      "loss": 4.5003,
      "mean_token_accuracy": 0.2786297306418419,
      "num_tokens": 194120851.0,
      "step": 84665
    },
    {
      "entropy": 5.510049343109131,
      "epoch": 8.133525456292027,
      "grad_norm": 1.421875,
      "learning_rate": 8.826910091899548e-05,
      "loss": 4.6317,
      "mean_token_accuracy": 0.2686959981918335,
      "num_tokens": 194133973.0,
      "step": 84670
    },
    {
      "entropy": 5.533246660232544,
      "epoch": 8.13400576368876,
      "grad_norm": 1.4140625,
      "learning_rate": 8.824997664432043e-05,
      "loss": 4.6399,
      "mean_token_accuracy": 0.25704778730869293,
      "num_tokens": 194145745.0,
      "step": 84675
    },
    {
      "entropy": 5.414880466461182,
      "epoch": 8.134486071085494,
      "grad_norm": 1.4453125,
      "learning_rate": 8.823085670542638e-05,
      "loss": 4.4919,
      "mean_token_accuracy": 0.2798508942127228,
      "num_tokens": 194156983.0,
      "step": 84680
    },
    {
      "entropy": 5.504521274566651,
      "epoch": 8.134966378482229,
      "grad_norm": 1.25,
      "learning_rate": 8.821174110275725e-05,
      "loss": 4.5775,
      "mean_token_accuracy": 0.26906220316886903,
      "num_tokens": 194169281.0,
      "step": 84685
    },
    {
      "entropy": 5.464554786682129,
      "epoch": 8.135446685878962,
      "grad_norm": 1.296875,
      "learning_rate": 8.819262983675683e-05,
      "loss": 4.5319,
      "mean_token_accuracy": 0.27654402405023576,
      "num_tokens": 194181479.0,
      "step": 84690
    },
    {
      "entropy": 5.505063438415528,
      "epoch": 8.135926993275696,
      "grad_norm": 1.5234375,
      "learning_rate": 8.817352290786893e-05,
      "loss": 4.5431,
      "mean_token_accuracy": 0.2735218957066536,
      "num_tokens": 194191353.0,
      "step": 84695
    },
    {
      "entropy": 5.522047472000122,
      "epoch": 8.136407300672431,
      "grad_norm": 1.4375,
      "learning_rate": 8.815442031653697e-05,
      "loss": 4.6446,
      "mean_token_accuracy": 0.26683999598026276,
      "num_tokens": 194202742.0,
      "step": 84700
    },
    {
      "entropy": 5.548709106445313,
      "epoch": 8.136887608069165,
      "grad_norm": 1.375,
      "learning_rate": 8.813532206320459e-05,
      "loss": 4.7081,
      "mean_token_accuracy": 0.2588270530104637,
      "num_tokens": 194215009.0,
      "step": 84705
    },
    {
      "entropy": 5.5011790752410885,
      "epoch": 8.137367915465898,
      "grad_norm": 1.4609375,
      "learning_rate": 8.811622814831514e-05,
      "loss": 4.5696,
      "mean_token_accuracy": 0.27124602198600767,
      "num_tokens": 194226047.0,
      "step": 84710
    },
    {
      "entropy": 5.5603758811950685,
      "epoch": 8.137848222862631,
      "grad_norm": 1.5390625,
      "learning_rate": 8.809713857231194e-05,
      "loss": 4.6115,
      "mean_token_accuracy": 0.2663195624947548,
      "num_tokens": 194237281.0,
      "step": 84715
    },
    {
      "entropy": 5.549579334259033,
      "epoch": 8.138328530259367,
      "grad_norm": 1.3359375,
      "learning_rate": 8.80780533356382e-05,
      "loss": 4.6088,
      "mean_token_accuracy": 0.2703708842396736,
      "num_tokens": 194250018.0,
      "step": 84720
    },
    {
      "entropy": 5.585869646072387,
      "epoch": 8.1388088376561,
      "grad_norm": 1.46875,
      "learning_rate": 8.805897243873707e-05,
      "loss": 4.6659,
      "mean_token_accuracy": 0.26454524844884875,
      "num_tokens": 194260930.0,
      "step": 84725
    },
    {
      "entropy": 5.544190359115601,
      "epoch": 8.139289145052834,
      "grad_norm": 1.703125,
      "learning_rate": 8.803989588205143e-05,
      "loss": 4.5396,
      "mean_token_accuracy": 0.2747069910168648,
      "num_tokens": 194272205.0,
      "step": 84730
    },
    {
      "entropy": 5.467393445968628,
      "epoch": 8.139769452449567,
      "grad_norm": 1.46875,
      "learning_rate": 8.802082366602426e-05,
      "loss": 4.4886,
      "mean_token_accuracy": 0.27550441920757296,
      "num_tokens": 194282710.0,
      "step": 84735
    },
    {
      "entropy": 5.5663103580474855,
      "epoch": 8.140249759846302,
      "grad_norm": 1.421875,
      "learning_rate": 8.800175579109837e-05,
      "loss": 4.5936,
      "mean_token_accuracy": 0.27070670276880265,
      "num_tokens": 194294113.0,
      "step": 84740
    },
    {
      "entropy": 5.573969173431396,
      "epoch": 8.140730067243036,
      "grad_norm": 1.453125,
      "learning_rate": 8.798269225771644e-05,
      "loss": 4.6545,
      "mean_token_accuracy": 0.26112647354602814,
      "num_tokens": 194304971.0,
      "step": 84745
    },
    {
      "entropy": 5.541416263580322,
      "epoch": 8.141210374639769,
      "grad_norm": 1.2578125,
      "learning_rate": 8.796363306632104e-05,
      "loss": 4.6497,
      "mean_token_accuracy": 0.25848820358514785,
      "num_tokens": 194316894.0,
      "step": 84750
    },
    {
      "entropy": 5.519479894638062,
      "epoch": 8.141690682036504,
      "grad_norm": 1.3359375,
      "learning_rate": 8.794457821735472e-05,
      "loss": 4.5883,
      "mean_token_accuracy": 0.2704417660832405,
      "num_tokens": 194328878.0,
      "step": 84755
    },
    {
      "entropy": 5.557454776763916,
      "epoch": 8.142170989433238,
      "grad_norm": 1.6484375,
      "learning_rate": 8.792552771125984e-05,
      "loss": 4.6341,
      "mean_token_accuracy": 0.2654304146766663,
      "num_tokens": 194339392.0,
      "step": 84760
    },
    {
      "entropy": 5.485590934753418,
      "epoch": 8.142651296829971,
      "grad_norm": 1.421875,
      "learning_rate": 8.790648154847877e-05,
      "loss": 4.5358,
      "mean_token_accuracy": 0.26853986382484435,
      "num_tokens": 194350321.0,
      "step": 84765
    },
    {
      "entropy": 5.533183240890503,
      "epoch": 8.143131604226705,
      "grad_norm": 1.3984375,
      "learning_rate": 8.78874397294536e-05,
      "loss": 4.6024,
      "mean_token_accuracy": 0.2702406167984009,
      "num_tokens": 194362889.0,
      "step": 84770
    },
    {
      "entropy": 5.529180669784546,
      "epoch": 8.14361191162344,
      "grad_norm": 1.4921875,
      "learning_rate": 8.786840225462645e-05,
      "loss": 4.6084,
      "mean_token_accuracy": 0.27095661759376527,
      "num_tokens": 194374291.0,
      "step": 84775
    },
    {
      "entropy": 5.601191854476928,
      "epoch": 8.144092219020173,
      "grad_norm": 1.546875,
      "learning_rate": 8.784936912443933e-05,
      "loss": 4.6384,
      "mean_token_accuracy": 0.2679922431707382,
      "num_tokens": 194385659.0,
      "step": 84780
    },
    {
      "entropy": 5.547211742401123,
      "epoch": 8.144572526416907,
      "grad_norm": 1.453125,
      "learning_rate": 8.783034033933415e-05,
      "loss": 4.6153,
      "mean_token_accuracy": 0.2636114567518234,
      "num_tokens": 194397924.0,
      "step": 84785
    },
    {
      "entropy": 5.614024353027344,
      "epoch": 8.14505283381364,
      "grad_norm": 1.5625,
      "learning_rate": 8.781131589975269e-05,
      "loss": 4.6144,
      "mean_token_accuracy": 0.27645795345306395,
      "num_tokens": 194408060.0,
      "step": 84790
    },
    {
      "entropy": 5.53236985206604,
      "epoch": 8.145533141210375,
      "grad_norm": 1.421875,
      "learning_rate": 8.779229580613663e-05,
      "loss": 4.6058,
      "mean_token_accuracy": 0.26981639862060547,
      "num_tokens": 194419806.0,
      "step": 84795
    },
    {
      "entropy": 5.589656066894531,
      "epoch": 8.146013448607109,
      "grad_norm": 1.328125,
      "learning_rate": 8.777328005892757e-05,
      "loss": 4.6607,
      "mean_token_accuracy": 0.268681101500988,
      "num_tokens": 194431031.0,
      "step": 84800
    },
    {
      "entropy": 5.510431289672852,
      "epoch": 8.146493756003842,
      "grad_norm": 1.53125,
      "learning_rate": 8.775426865856706e-05,
      "loss": 4.6274,
      "mean_token_accuracy": 0.26522684991359713,
      "num_tokens": 194442547.0,
      "step": 84805
    },
    {
      "entropy": 5.55216794013977,
      "epoch": 8.146974063400576,
      "grad_norm": 1.5,
      "learning_rate": 8.773526160549635e-05,
      "loss": 4.6055,
      "mean_token_accuracy": 0.26862964034080505,
      "num_tokens": 194454272.0,
      "step": 84810
    },
    {
      "entropy": 5.502371883392334,
      "epoch": 8.14745437079731,
      "grad_norm": 1.4140625,
      "learning_rate": 8.771625890015683e-05,
      "loss": 4.6217,
      "mean_token_accuracy": 0.2728405833244324,
      "num_tokens": 194465541.0,
      "step": 84815
    },
    {
      "entropy": 5.469659233093262,
      "epoch": 8.147934678194044,
      "grad_norm": 1.421875,
      "learning_rate": 8.769726054298966e-05,
      "loss": 4.5525,
      "mean_token_accuracy": 0.26771233081817625,
      "num_tokens": 194477051.0,
      "step": 84820
    },
    {
      "entropy": 5.527687406539917,
      "epoch": 8.148414985590778,
      "grad_norm": 1.5546875,
      "learning_rate": 8.767826653443592e-05,
      "loss": 4.6368,
      "mean_token_accuracy": 0.2680936112999916,
      "num_tokens": 194488845.0,
      "step": 84825
    },
    {
      "entropy": 5.515675020217896,
      "epoch": 8.148895292987511,
      "grad_norm": 1.5078125,
      "learning_rate": 8.765927687493657e-05,
      "loss": 4.6012,
      "mean_token_accuracy": 0.2711345285177231,
      "num_tokens": 194499838.0,
      "step": 84830
    },
    {
      "entropy": 5.538311243057251,
      "epoch": 8.149375600384246,
      "grad_norm": 1.3828125,
      "learning_rate": 8.764029156493255e-05,
      "loss": 4.6412,
      "mean_token_accuracy": 0.2705144464969635,
      "num_tokens": 194512261.0,
      "step": 84835
    },
    {
      "entropy": 5.4443755626678465,
      "epoch": 8.14985590778098,
      "grad_norm": 1.34375,
      "learning_rate": 8.762131060486462e-05,
      "loss": 4.535,
      "mean_token_accuracy": 0.27681293338537216,
      "num_tokens": 194523942.0,
      "step": 84840
    },
    {
      "entropy": 5.50292935371399,
      "epoch": 8.150336215177713,
      "grad_norm": 1.4375,
      "learning_rate": 8.760233399517351e-05,
      "loss": 4.6147,
      "mean_token_accuracy": 0.26877670735120773,
      "num_tokens": 194535920.0,
      "step": 84845
    },
    {
      "entropy": 5.506275463104248,
      "epoch": 8.150816522574448,
      "grad_norm": 1.5234375,
      "learning_rate": 8.758336173629969e-05,
      "loss": 4.5636,
      "mean_token_accuracy": 0.28061968088150024,
      "num_tokens": 194547107.0,
      "step": 84850
    },
    {
      "entropy": 5.465729141235352,
      "epoch": 8.151296829971182,
      "grad_norm": 1.4609375,
      "learning_rate": 8.756439382868368e-05,
      "loss": 4.5692,
      "mean_token_accuracy": 0.269480761885643,
      "num_tokens": 194558877.0,
      "step": 84855
    },
    {
      "entropy": 5.523971509933472,
      "epoch": 8.151777137367915,
      "grad_norm": 1.453125,
      "learning_rate": 8.75454302727659e-05,
      "loss": 4.6037,
      "mean_token_accuracy": 0.26971833109855653,
      "num_tokens": 194571256.0,
      "step": 84860
    },
    {
      "entropy": 5.455829668045044,
      "epoch": 8.152257444764649,
      "grad_norm": 1.265625,
      "learning_rate": 8.75264710689866e-05,
      "loss": 4.5581,
      "mean_token_accuracy": 0.27263342440128324,
      "num_tokens": 194583317.0,
      "step": 84865
    },
    {
      "entropy": 5.547043514251709,
      "epoch": 8.152737752161384,
      "grad_norm": 1.421875,
      "learning_rate": 8.750751621778597e-05,
      "loss": 4.6209,
      "mean_token_accuracy": 0.26672241240739825,
      "num_tokens": 194596165.0,
      "step": 84870
    },
    {
      "entropy": 5.55217547416687,
      "epoch": 8.153218059558117,
      "grad_norm": 1.5078125,
      "learning_rate": 8.748856571960411e-05,
      "loss": 4.6574,
      "mean_token_accuracy": 0.2641418620944023,
      "num_tokens": 194606317.0,
      "step": 84875
    },
    {
      "entropy": 5.59804892539978,
      "epoch": 8.15369836695485,
      "grad_norm": 1.3984375,
      "learning_rate": 8.746961957488089e-05,
      "loss": 4.7036,
      "mean_token_accuracy": 0.26668681502342223,
      "num_tokens": 194617891.0,
      "step": 84880
    },
    {
      "entropy": 5.5702770233154295,
      "epoch": 8.154178674351584,
      "grad_norm": 1.5390625,
      "learning_rate": 8.745067778405627e-05,
      "loss": 4.6773,
      "mean_token_accuracy": 0.2604890808463097,
      "num_tokens": 194630109.0,
      "step": 84885
    },
    {
      "entropy": 5.551334095001221,
      "epoch": 8.15465898174832,
      "grad_norm": 1.421875,
      "learning_rate": 8.743174034757002e-05,
      "loss": 4.6358,
      "mean_token_accuracy": 0.27586915642023085,
      "num_tokens": 194641595.0,
      "step": 84890
    },
    {
      "entropy": 5.51468858718872,
      "epoch": 8.155139289145053,
      "grad_norm": 1.40625,
      "learning_rate": 8.741280726586179e-05,
      "loss": 4.5646,
      "mean_token_accuracy": 0.2723284810781479,
      "num_tokens": 194652622.0,
      "step": 84895
    },
    {
      "entropy": 5.493188953399658,
      "epoch": 8.155619596541786,
      "grad_norm": 1.5,
      "learning_rate": 8.739387853937114e-05,
      "loss": 4.551,
      "mean_token_accuracy": 0.2743737667798996,
      "num_tokens": 194664434.0,
      "step": 84900
    },
    {
      "entropy": 5.452864742279052,
      "epoch": 8.15609990393852,
      "grad_norm": 1.5703125,
      "learning_rate": 8.737495416853763e-05,
      "loss": 4.589,
      "mean_token_accuracy": 0.2687187120318413,
      "num_tokens": 194675524.0,
      "step": 84905
    },
    {
      "entropy": 5.509461355209351,
      "epoch": 8.156580211335255,
      "grad_norm": 1.5390625,
      "learning_rate": 8.735603415380047e-05,
      "loss": 4.6004,
      "mean_token_accuracy": 0.26612572073936464,
      "num_tokens": 194687399.0,
      "step": 84910
    },
    {
      "entropy": 5.461695766448974,
      "epoch": 8.157060518731988,
      "grad_norm": 1.640625,
      "learning_rate": 8.733711849559903e-05,
      "loss": 4.5184,
      "mean_token_accuracy": 0.2785542458295822,
      "num_tokens": 194698259.0,
      "step": 84915
    },
    {
      "entropy": 5.499809265136719,
      "epoch": 8.157540826128722,
      "grad_norm": 1.359375,
      "learning_rate": 8.731820719437245e-05,
      "loss": 4.5165,
      "mean_token_accuracy": 0.2815883070230484,
      "num_tokens": 194708517.0,
      "step": 84920
    },
    {
      "entropy": 5.499628686904908,
      "epoch": 8.158021133525457,
      "grad_norm": 1.515625,
      "learning_rate": 8.72993002505598e-05,
      "loss": 4.5991,
      "mean_token_accuracy": 0.27661667913198473,
      "num_tokens": 194718962.0,
      "step": 84925
    },
    {
      "entropy": 5.573189401626587,
      "epoch": 8.15850144092219,
      "grad_norm": 1.2734375,
      "learning_rate": 8.728039766460002e-05,
      "loss": 4.7076,
      "mean_token_accuracy": 0.26198755353689196,
      "num_tokens": 194731164.0,
      "step": 84930
    },
    {
      "entropy": 5.545356559753418,
      "epoch": 8.158981748318924,
      "grad_norm": 1.3046875,
      "learning_rate": 8.72614994369321e-05,
      "loss": 4.618,
      "mean_token_accuracy": 0.2644848421216011,
      "num_tokens": 194743142.0,
      "step": 84935
    },
    {
      "entropy": 5.538501834869384,
      "epoch": 8.159462055715657,
      "grad_norm": 1.765625,
      "learning_rate": 8.724260556799457e-05,
      "loss": 4.6217,
      "mean_token_accuracy": 0.26590457558631897,
      "num_tokens": 194756427.0,
      "step": 84940
    },
    {
      "entropy": 5.581996870040894,
      "epoch": 8.159942363112393,
      "grad_norm": 1.703125,
      "learning_rate": 8.722371605822626e-05,
      "loss": 4.6703,
      "mean_token_accuracy": 0.27027757167816163,
      "num_tokens": 194767026.0,
      "step": 84945
    },
    {
      "entropy": 5.5847251415252686,
      "epoch": 8.160422670509126,
      "grad_norm": 1.5546875,
      "learning_rate": 8.720483090806569e-05,
      "loss": 4.6745,
      "mean_token_accuracy": 0.2557385668158531,
      "num_tokens": 194777211.0,
      "step": 84950
    },
    {
      "entropy": 5.53679084777832,
      "epoch": 8.16090297790586,
      "grad_norm": 1.4921875,
      "learning_rate": 8.71859501179513e-05,
      "loss": 4.6013,
      "mean_token_accuracy": 0.2720074325799942,
      "num_tokens": 194788697.0,
      "step": 84955
    },
    {
      "entropy": 5.5820728778839115,
      "epoch": 8.161383285302593,
      "grad_norm": 1.5078125,
      "learning_rate": 8.716707368832144e-05,
      "loss": 4.6477,
      "mean_token_accuracy": 0.2600206583738327,
      "num_tokens": 194799211.0,
      "step": 84960
    },
    {
      "entropy": 5.500585889816284,
      "epoch": 8.161863592699328,
      "grad_norm": 1.4296875,
      "learning_rate": 8.714820161961445e-05,
      "loss": 4.5596,
      "mean_token_accuracy": 0.2688427597284317,
      "num_tokens": 194810677.0,
      "step": 84965
    },
    {
      "entropy": 5.519262075424194,
      "epoch": 8.162343900096062,
      "grad_norm": 1.546875,
      "learning_rate": 8.712933391226832e-05,
      "loss": 4.5984,
      "mean_token_accuracy": 0.26830458343029023,
      "num_tokens": 194821639.0,
      "step": 84970
    },
    {
      "entropy": 5.480032348632813,
      "epoch": 8.162824207492795,
      "grad_norm": 1.34375,
      "learning_rate": 8.711047056672125e-05,
      "loss": 4.5336,
      "mean_token_accuracy": 0.2733038291335106,
      "num_tokens": 194832830.0,
      "step": 84975
    },
    {
      "entropy": 5.490379762649536,
      "epoch": 8.163304514889528,
      "grad_norm": 1.3671875,
      "learning_rate": 8.709161158341111e-05,
      "loss": 4.6041,
      "mean_token_accuracy": 0.26522603183984755,
      "num_tokens": 194845103.0,
      "step": 84980
    },
    {
      "entropy": 5.600762557983399,
      "epoch": 8.163784822286264,
      "grad_norm": 1.53125,
      "learning_rate": 8.707275696277578e-05,
      "loss": 4.6736,
      "mean_token_accuracy": 0.25726398825645447,
      "num_tokens": 194856345.0,
      "step": 84985
    },
    {
      "entropy": 5.497563123703003,
      "epoch": 8.164265129682997,
      "grad_norm": 1.59375,
      "learning_rate": 8.7053906705253e-05,
      "loss": 4.5645,
      "mean_token_accuracy": 0.27313906997442244,
      "num_tokens": 194865864.0,
      "step": 84990
    },
    {
      "entropy": 5.557281160354615,
      "epoch": 8.16474543707973,
      "grad_norm": 1.4609375,
      "learning_rate": 8.703506081128049e-05,
      "loss": 4.6375,
      "mean_token_accuracy": 0.2659267395734787,
      "num_tokens": 194877536.0,
      "step": 84995
    },
    {
      "entropy": 5.495945739746094,
      "epoch": 8.165225744476466,
      "grad_norm": 1.3984375,
      "learning_rate": 8.701621928129564e-05,
      "loss": 4.5802,
      "mean_token_accuracy": 0.2697813600301743,
      "num_tokens": 194888325.0,
      "step": 85000
    },
    {
      "entropy": 5.5123584270477295,
      "epoch": 8.1657060518732,
      "grad_norm": 1.46875,
      "learning_rate": 8.6997382115736e-05,
      "loss": 4.5811,
      "mean_token_accuracy": 0.2699752405285835,
      "num_tokens": 194899532.0,
      "step": 85005
    },
    {
      "entropy": 5.527110195159912,
      "epoch": 8.166186359269933,
      "grad_norm": 1.390625,
      "learning_rate": 8.69785493150389e-05,
      "loss": 4.6273,
      "mean_token_accuracy": 0.27291772663593294,
      "num_tokens": 194912376.0,
      "step": 85010
    },
    {
      "entropy": 5.496196985244751,
      "epoch": 8.166666666666666,
      "grad_norm": 1.4453125,
      "learning_rate": 8.695972087964158e-05,
      "loss": 4.5687,
      "mean_token_accuracy": 0.27840819507837294,
      "num_tokens": 194923971.0,
      "step": 85015
    },
    {
      "entropy": 5.6017763137817385,
      "epoch": 8.167146974063401,
      "grad_norm": 1.4140625,
      "learning_rate": 8.694089680998115e-05,
      "loss": 4.6625,
      "mean_token_accuracy": 0.26449627578258517,
      "num_tokens": 194934836.0,
      "step": 85020
    },
    {
      "entropy": 5.475258874893188,
      "epoch": 8.167627281460135,
      "grad_norm": 1.3046875,
      "learning_rate": 8.692207710649475e-05,
      "loss": 4.5071,
      "mean_token_accuracy": 0.28572235107421873,
      "num_tokens": 194946824.0,
      "step": 85025
    },
    {
      "entropy": 5.554562282562256,
      "epoch": 8.168107588856868,
      "grad_norm": 1.6171875,
      "learning_rate": 8.690326176961918e-05,
      "loss": 4.6436,
      "mean_token_accuracy": 0.267314700782299,
      "num_tokens": 194959448.0,
      "step": 85030
    },
    {
      "entropy": 5.504151344299316,
      "epoch": 8.168587896253602,
      "grad_norm": 1.390625,
      "learning_rate": 8.688445079979135e-05,
      "loss": 4.5605,
      "mean_token_accuracy": 0.274833419919014,
      "num_tokens": 194971163.0,
      "step": 85035
    },
    {
      "entropy": 5.544068717956543,
      "epoch": 8.169068203650337,
      "grad_norm": 1.484375,
      "learning_rate": 8.686564419744799e-05,
      "loss": 4.526,
      "mean_token_accuracy": 0.2741539731621742,
      "num_tokens": 194982277.0,
      "step": 85040
    },
    {
      "entropy": 5.552682685852051,
      "epoch": 8.16954851104707,
      "grad_norm": 1.34375,
      "learning_rate": 8.684684196302571e-05,
      "loss": 4.7131,
      "mean_token_accuracy": 0.25926531106233597,
      "num_tokens": 194994367.0,
      "step": 85045
    },
    {
      "entropy": 5.5129656314849855,
      "epoch": 8.170028818443804,
      "grad_norm": 1.546875,
      "learning_rate": 8.682804409696107e-05,
      "loss": 4.612,
      "mean_token_accuracy": 0.2671024277806282,
      "num_tokens": 195006530.0,
      "step": 85050
    },
    {
      "entropy": 5.477945470809937,
      "epoch": 8.170509125840537,
      "grad_norm": 1.375,
      "learning_rate": 8.680925059969054e-05,
      "loss": 4.5392,
      "mean_token_accuracy": 0.27882391810417173,
      "num_tokens": 195017911.0,
      "step": 85055
    },
    {
      "entropy": 5.452488851547241,
      "epoch": 8.170989433237272,
      "grad_norm": 1.375,
      "learning_rate": 8.679046147165036e-05,
      "loss": 4.5299,
      "mean_token_accuracy": 0.2806150019168854,
      "num_tokens": 195028875.0,
      "step": 85060
    },
    {
      "entropy": 5.537070178985596,
      "epoch": 8.171469740634006,
      "grad_norm": 1.53125,
      "learning_rate": 8.677167671327679e-05,
      "loss": 4.6406,
      "mean_token_accuracy": 0.2644072949886322,
      "num_tokens": 195040027.0,
      "step": 85065
    },
    {
      "entropy": 5.4889555931091305,
      "epoch": 8.17195004803074,
      "grad_norm": 1.3828125,
      "learning_rate": 8.675289632500594e-05,
      "loss": 4.5675,
      "mean_token_accuracy": 0.2726415678858757,
      "num_tokens": 195051569.0,
      "step": 85070
    },
    {
      "entropy": 5.5096845626831055,
      "epoch": 8.172430355427474,
      "grad_norm": 1.5859375,
      "learning_rate": 8.673412030727387e-05,
      "loss": 4.6265,
      "mean_token_accuracy": 0.2609466016292572,
      "num_tokens": 195062518.0,
      "step": 85075
    },
    {
      "entropy": 5.553704881668091,
      "epoch": 8.172910662824208,
      "grad_norm": 1.34375,
      "learning_rate": 8.67153486605165e-05,
      "loss": 4.6899,
      "mean_token_accuracy": 0.2678368330001831,
      "num_tokens": 195075924.0,
      "step": 85080
    },
    {
      "entropy": 5.479230737686157,
      "epoch": 8.173390970220941,
      "grad_norm": 1.4296875,
      "learning_rate": 8.669658138516969e-05,
      "loss": 4.5887,
      "mean_token_accuracy": 0.27414117455482484,
      "num_tokens": 195087467.0,
      "step": 85085
    },
    {
      "entropy": 5.484036588668824,
      "epoch": 8.173871277617675,
      "grad_norm": 1.3203125,
      "learning_rate": 8.667781848166906e-05,
      "loss": 4.5605,
      "mean_token_accuracy": 0.27066163867712023,
      "num_tokens": 195099105.0,
      "step": 85090
    },
    {
      "entropy": 5.465292310714721,
      "epoch": 8.17435158501441,
      "grad_norm": 1.5234375,
      "learning_rate": 8.665905995045028e-05,
      "loss": 4.5177,
      "mean_token_accuracy": 0.27580786049365996,
      "num_tokens": 195110737.0,
      "step": 85095
    },
    {
      "entropy": 5.530630159378052,
      "epoch": 8.174831892411143,
      "grad_norm": 1.390625,
      "learning_rate": 8.66403057919489e-05,
      "loss": 4.6029,
      "mean_token_accuracy": 0.2685830742120743,
      "num_tokens": 195121607.0,
      "step": 85100
    },
    {
      "entropy": 5.444610548019409,
      "epoch": 8.175312199807877,
      "grad_norm": 1.5078125,
      "learning_rate": 8.662155600660027e-05,
      "loss": 4.4599,
      "mean_token_accuracy": 0.2790436878800392,
      "num_tokens": 195132875.0,
      "step": 85105
    },
    {
      "entropy": 5.49889988899231,
      "epoch": 8.17579250720461,
      "grad_norm": 1.5234375,
      "learning_rate": 8.660281059483977e-05,
      "loss": 4.5767,
      "mean_token_accuracy": 0.2737630918622017,
      "num_tokens": 195145231.0,
      "step": 85110
    },
    {
      "entropy": 5.441727924346924,
      "epoch": 8.176272814601345,
      "grad_norm": 1.609375,
      "learning_rate": 8.658406955710263e-05,
      "loss": 4.4932,
      "mean_token_accuracy": 0.27992257475852966,
      "num_tokens": 195156025.0,
      "step": 85115
    },
    {
      "entropy": 5.536169004440308,
      "epoch": 8.176753121998079,
      "grad_norm": 1.4609375,
      "learning_rate": 8.656533289382382e-05,
      "loss": 4.6074,
      "mean_token_accuracy": 0.2681393727660179,
      "num_tokens": 195166951.0,
      "step": 85120
    },
    {
      "entropy": 5.554251670837402,
      "epoch": 8.177233429394812,
      "grad_norm": 1.484375,
      "learning_rate": 8.65466006054385e-05,
      "loss": 4.6085,
      "mean_token_accuracy": 0.2729603976011276,
      "num_tokens": 195177399.0,
      "step": 85125
    },
    {
      "entropy": 5.479535245895386,
      "epoch": 8.177713736791546,
      "grad_norm": 1.4765625,
      "learning_rate": 8.652787269238147e-05,
      "loss": 4.6136,
      "mean_token_accuracy": 0.2556384772062302,
      "num_tokens": 195189805.0,
      "step": 85130
    },
    {
      "entropy": 5.473015546798706,
      "epoch": 8.178194044188281,
      "grad_norm": 1.3125,
      "learning_rate": 8.650914915508764e-05,
      "loss": 4.5405,
      "mean_token_accuracy": 0.27033665776252747,
      "num_tokens": 195201887.0,
      "step": 85135
    },
    {
      "entropy": 5.505325078964233,
      "epoch": 8.178674351585014,
      "grad_norm": 1.4140625,
      "learning_rate": 8.649042999399165e-05,
      "loss": 4.543,
      "mean_token_accuracy": 0.2724686235189438,
      "num_tokens": 195213022.0,
      "step": 85140
    },
    {
      "entropy": 5.575730466842652,
      "epoch": 8.179154658981748,
      "grad_norm": 1.6171875,
      "learning_rate": 8.647171520952815e-05,
      "loss": 4.6559,
      "mean_token_accuracy": 0.2689074769616127,
      "num_tokens": 195223911.0,
      "step": 85145
    },
    {
      "entropy": 5.584693861007691,
      "epoch": 8.179634966378483,
      "grad_norm": 1.5,
      "learning_rate": 8.645300480213159e-05,
      "loss": 4.6241,
      "mean_token_accuracy": 0.27581063359975816,
      "num_tokens": 195235121.0,
      "step": 85150
    },
    {
      "entropy": 5.537451553344726,
      "epoch": 8.180115273775217,
      "grad_norm": 1.34375,
      "learning_rate": 8.643429877223633e-05,
      "loss": 4.6454,
      "mean_token_accuracy": 0.26732234209775924,
      "num_tokens": 195247382.0,
      "step": 85155
    },
    {
      "entropy": 5.523480606079102,
      "epoch": 8.18059558117195,
      "grad_norm": 1.4609375,
      "learning_rate": 8.641559712027677e-05,
      "loss": 4.6551,
      "mean_token_accuracy": 0.26447148621082306,
      "num_tokens": 195259867.0,
      "step": 85160
    },
    {
      "entropy": 5.45618691444397,
      "epoch": 8.181075888568683,
      "grad_norm": 1.46875,
      "learning_rate": 8.639689984668708e-05,
      "loss": 4.5463,
      "mean_token_accuracy": 0.2728238463401794,
      "num_tokens": 195273189.0,
      "step": 85165
    },
    {
      "entropy": 5.483478307723999,
      "epoch": 8.181556195965419,
      "grad_norm": 1.390625,
      "learning_rate": 8.637820695190131e-05,
      "loss": 4.6242,
      "mean_token_accuracy": 0.26362263709306716,
      "num_tokens": 195286487.0,
      "step": 85170
    },
    {
      "entropy": 5.446978759765625,
      "epoch": 8.182036503362152,
      "grad_norm": 1.421875,
      "learning_rate": 8.635951843635353e-05,
      "loss": 4.52,
      "mean_token_accuracy": 0.2765354007482529,
      "num_tokens": 195297106.0,
      "step": 85175
    },
    {
      "entropy": 5.519970798492432,
      "epoch": 8.182516810758885,
      "grad_norm": 1.4296875,
      "learning_rate": 8.634083430047754e-05,
      "loss": 4.658,
      "mean_token_accuracy": 0.26908432841300967,
      "num_tokens": 195307988.0,
      "step": 85180
    },
    {
      "entropy": 5.455465078353882,
      "epoch": 8.182997118155619,
      "grad_norm": 1.46875,
      "learning_rate": 8.632215454470721e-05,
      "loss": 4.4788,
      "mean_token_accuracy": 0.27906905114650726,
      "num_tokens": 195319832.0,
      "step": 85185
    },
    {
      "entropy": 5.507658290863037,
      "epoch": 8.183477425552354,
      "grad_norm": 1.3828125,
      "learning_rate": 8.630347916947614e-05,
      "loss": 4.5878,
      "mean_token_accuracy": 0.2677398532629013,
      "num_tokens": 195332333.0,
      "step": 85190
    },
    {
      "entropy": 5.532572078704834,
      "epoch": 8.183957732949088,
      "grad_norm": 1.515625,
      "learning_rate": 8.6284808175218e-05,
      "loss": 4.5867,
      "mean_token_accuracy": 0.26816471815109255,
      "num_tokens": 195343678.0,
      "step": 85195
    },
    {
      "entropy": 5.540688419342041,
      "epoch": 8.184438040345821,
      "grad_norm": 1.421875,
      "learning_rate": 8.626614156236623e-05,
      "loss": 4.6547,
      "mean_token_accuracy": 0.2626257658004761,
      "num_tokens": 195355282.0,
      "step": 85200
    },
    {
      "entropy": 5.50931396484375,
      "epoch": 8.184918347742554,
      "grad_norm": 1.453125,
      "learning_rate": 8.62474793313543e-05,
      "loss": 4.5616,
      "mean_token_accuracy": 0.27280468344688413,
      "num_tokens": 195367436.0,
      "step": 85205
    },
    {
      "entropy": 5.558471298217773,
      "epoch": 8.18539865513929,
      "grad_norm": 1.5,
      "learning_rate": 8.622882148261536e-05,
      "loss": 4.6505,
      "mean_token_accuracy": 0.2688245579600334,
      "num_tokens": 195378432.0,
      "step": 85210
    },
    {
      "entropy": 5.527683973312378,
      "epoch": 8.185878962536023,
      "grad_norm": 1.421875,
      "learning_rate": 8.621016801658262e-05,
      "loss": 4.5762,
      "mean_token_accuracy": 0.2624863818287849,
      "num_tokens": 195390084.0,
      "step": 85215
    },
    {
      "entropy": 5.641809558868408,
      "epoch": 8.186359269932757,
      "grad_norm": 1.3984375,
      "learning_rate": 8.619151893368923e-05,
      "loss": 4.7112,
      "mean_token_accuracy": 0.259801359474659,
      "num_tokens": 195402059.0,
      "step": 85220
    },
    {
      "entropy": 5.4896481990814205,
      "epoch": 8.186839577329492,
      "grad_norm": 1.3984375,
      "learning_rate": 8.617287423436813e-05,
      "loss": 4.5094,
      "mean_token_accuracy": 0.28316432535648345,
      "num_tokens": 195413145.0,
      "step": 85225
    },
    {
      "entropy": 5.514341592788696,
      "epoch": 8.187319884726225,
      "grad_norm": 1.359375,
      "learning_rate": 8.615423391905215e-05,
      "loss": 4.5167,
      "mean_token_accuracy": 0.2684875190258026,
      "num_tokens": 195424773.0,
      "step": 85230
    },
    {
      "entropy": 5.501185417175293,
      "epoch": 8.187800192122959,
      "grad_norm": 1.4140625,
      "learning_rate": 8.613559798817418e-05,
      "loss": 4.5769,
      "mean_token_accuracy": 0.2719056487083435,
      "num_tokens": 195436983.0,
      "step": 85235
    },
    {
      "entropy": 5.45141077041626,
      "epoch": 8.188280499519692,
      "grad_norm": 1.5546875,
      "learning_rate": 8.611696644216677e-05,
      "loss": 4.4962,
      "mean_token_accuracy": 0.27893100380897523,
      "num_tokens": 195446857.0,
      "step": 85240
    },
    {
      "entropy": 5.514728021621704,
      "epoch": 8.188760806916427,
      "grad_norm": 1.390625,
      "learning_rate": 8.60983392814625e-05,
      "loss": 4.5578,
      "mean_token_accuracy": 0.2754864037036896,
      "num_tokens": 195458818.0,
      "step": 85245
    },
    {
      "entropy": 5.4769810199737545,
      "epoch": 8.18924111431316,
      "grad_norm": 1.375,
      "learning_rate": 8.60797165064939e-05,
      "loss": 4.6321,
      "mean_token_accuracy": 0.265154567360878,
      "num_tokens": 195471339.0,
      "step": 85250
    },
    {
      "entropy": 5.442590570449829,
      "epoch": 8.189721421709894,
      "grad_norm": 1.4375,
      "learning_rate": 8.606109811769328e-05,
      "loss": 4.5523,
      "mean_token_accuracy": 0.2753768861293793,
      "num_tokens": 195482619.0,
      "step": 85255
    },
    {
      "entropy": 5.518635606765747,
      "epoch": 8.190201729106628,
      "grad_norm": 1.4765625,
      "learning_rate": 8.604248411549292e-05,
      "loss": 4.6403,
      "mean_token_accuracy": 0.26494899243116377,
      "num_tokens": 195493816.0,
      "step": 85260
    },
    {
      "entropy": 5.541833734512329,
      "epoch": 8.190682036503363,
      "grad_norm": 1.5859375,
      "learning_rate": 8.602387450032509e-05,
      "loss": 4.6085,
      "mean_token_accuracy": 0.26910405308008195,
      "num_tokens": 195505375.0,
      "step": 85265
    },
    {
      "entropy": 5.503495407104492,
      "epoch": 8.191162343900096,
      "grad_norm": 1.40625,
      "learning_rate": 8.600526927262167e-05,
      "loss": 4.5387,
      "mean_token_accuracy": 0.2795481562614441,
      "num_tokens": 195515952.0,
      "step": 85270
    },
    {
      "entropy": 5.518688440322876,
      "epoch": 8.19164265129683,
      "grad_norm": 1.5625,
      "learning_rate": 8.598666843281469e-05,
      "loss": 4.6101,
      "mean_token_accuracy": 0.26779562681913377,
      "num_tokens": 195526561.0,
      "step": 85275
    },
    {
      "entropy": 5.514497423171997,
      "epoch": 8.192122958693563,
      "grad_norm": 1.640625,
      "learning_rate": 8.596807198133603e-05,
      "loss": 4.611,
      "mean_token_accuracy": 0.27123548835515976,
      "num_tokens": 195538461.0,
      "step": 85280
    },
    {
      "entropy": 5.579106283187866,
      "epoch": 8.192603266090298,
      "grad_norm": 1.4453125,
      "learning_rate": 8.59494799186174e-05,
      "loss": 4.7309,
      "mean_token_accuracy": 0.2576924368739128,
      "num_tokens": 195550338.0,
      "step": 85285
    },
    {
      "entropy": 5.584515190124511,
      "epoch": 8.193083573487032,
      "grad_norm": 1.3515625,
      "learning_rate": 8.593089224509051e-05,
      "loss": 4.6672,
      "mean_token_accuracy": 0.26746219098567964,
      "num_tokens": 195561635.0,
      "step": 85290
    },
    {
      "entropy": 5.492901849746704,
      "epoch": 8.193563880883765,
      "grad_norm": 1.4765625,
      "learning_rate": 8.591230896118688e-05,
      "loss": 4.5782,
      "mean_token_accuracy": 0.2663190051913261,
      "num_tokens": 195572962.0,
      "step": 85295
    },
    {
      "entropy": 5.494809007644653,
      "epoch": 8.1940441882805,
      "grad_norm": 1.375,
      "learning_rate": 8.5893730067338e-05,
      "loss": 4.6153,
      "mean_token_accuracy": 0.26519097983837125,
      "num_tokens": 195584770.0,
      "step": 85300
    },
    {
      "entropy": 5.582520484924316,
      "epoch": 8.194524495677234,
      "grad_norm": 1.4609375,
      "learning_rate": 8.587515556397515e-05,
      "loss": 4.7184,
      "mean_token_accuracy": 0.25750131756067274,
      "num_tokens": 195597706.0,
      "step": 85305
    },
    {
      "entropy": 5.561565208435058,
      "epoch": 8.195004803073967,
      "grad_norm": 1.5390625,
      "learning_rate": 8.585658545152958e-05,
      "loss": 4.6132,
      "mean_token_accuracy": 0.2682109624147415,
      "num_tokens": 195608496.0,
      "step": 85310
    },
    {
      "entropy": 5.55720329284668,
      "epoch": 8.1954851104707,
      "grad_norm": 1.484375,
      "learning_rate": 8.583801973043246e-05,
      "loss": 4.5999,
      "mean_token_accuracy": 0.2664457097649574,
      "num_tokens": 195619130.0,
      "step": 85315
    },
    {
      "entropy": 5.499793577194214,
      "epoch": 8.195965417867436,
      "grad_norm": 1.390625,
      "learning_rate": 8.581945840111482e-05,
      "loss": 4.5322,
      "mean_token_accuracy": 0.2679421856999397,
      "num_tokens": 195630352.0,
      "step": 85320
    },
    {
      "entropy": 5.556524705886841,
      "epoch": 8.19644572526417,
      "grad_norm": 1.328125,
      "learning_rate": 8.580090146400762e-05,
      "loss": 4.6746,
      "mean_token_accuracy": 0.2593113720417023,
      "num_tokens": 195641687.0,
      "step": 85325
    },
    {
      "entropy": 5.451600170135498,
      "epoch": 8.196926032660903,
      "grad_norm": 1.4296875,
      "learning_rate": 8.578234891954167e-05,
      "loss": 4.5409,
      "mean_token_accuracy": 0.2831770867109299,
      "num_tokens": 195653000.0,
      "step": 85330
    },
    {
      "entropy": 5.564843416213989,
      "epoch": 8.197406340057636,
      "grad_norm": 1.53125,
      "learning_rate": 8.576380076814771e-05,
      "loss": 4.6956,
      "mean_token_accuracy": 0.2641251251101494,
      "num_tokens": 195664302.0,
      "step": 85335
    },
    {
      "entropy": 5.522328042984009,
      "epoch": 8.197886647454371,
      "grad_norm": 1.4296875,
      "learning_rate": 8.574525701025647e-05,
      "loss": 4.6142,
      "mean_token_accuracy": 0.26725687086582184,
      "num_tokens": 195675674.0,
      "step": 85340
    },
    {
      "entropy": 5.460185194015503,
      "epoch": 8.198366954851105,
      "grad_norm": 1.2734375,
      "learning_rate": 8.57267176462983e-05,
      "loss": 4.5793,
      "mean_token_accuracy": 0.2737333938479424,
      "num_tokens": 195688148.0,
      "step": 85345
    },
    {
      "entropy": 5.500152730941773,
      "epoch": 8.198847262247838,
      "grad_norm": 1.3515625,
      "learning_rate": 8.570818267670375e-05,
      "loss": 4.5527,
      "mean_token_accuracy": 0.2728447660803795,
      "num_tokens": 195699406.0,
      "step": 85350
    },
    {
      "entropy": 5.583339881896973,
      "epoch": 8.199327569644572,
      "grad_norm": 1.5234375,
      "learning_rate": 8.568965210190307e-05,
      "loss": 4.6232,
      "mean_token_accuracy": 0.2682013168931007,
      "num_tokens": 195711314.0,
      "step": 85355
    },
    {
      "entropy": 5.551823997497559,
      "epoch": 8.199807877041307,
      "grad_norm": 1.3984375,
      "learning_rate": 8.567112592232656e-05,
      "loss": 4.6375,
      "mean_token_accuracy": 0.2687870800495148,
      "num_tokens": 195723240.0,
      "step": 85360
    },
    {
      "entropy": 5.497257280349731,
      "epoch": 8.20028818443804,
      "grad_norm": 1.3125,
      "learning_rate": 8.565260413840434e-05,
      "loss": 4.6018,
      "mean_token_accuracy": 0.26818618029356,
      "num_tokens": 195736036.0,
      "step": 85365
    },
    {
      "entropy": 5.529100036621093,
      "epoch": 8.200768491834774,
      "grad_norm": 1.4296875,
      "learning_rate": 8.563408675056635e-05,
      "loss": 4.6635,
      "mean_token_accuracy": 0.2706352904438972,
      "num_tokens": 195747276.0,
      "step": 85370
    },
    {
      "entropy": 5.472634267807007,
      "epoch": 8.201248799231507,
      "grad_norm": 1.6328125,
      "learning_rate": 8.561557375924263e-05,
      "loss": 4.5228,
      "mean_token_accuracy": 0.27713131010532377,
      "num_tokens": 195758818.0,
      "step": 85375
    },
    {
      "entropy": 5.488768768310547,
      "epoch": 8.201729106628243,
      "grad_norm": 1.4765625,
      "learning_rate": 8.559706516486295e-05,
      "loss": 4.5286,
      "mean_token_accuracy": 0.2790542259812355,
      "num_tokens": 195769801.0,
      "step": 85380
    },
    {
      "entropy": 5.569213724136352,
      "epoch": 8.202209414024976,
      "grad_norm": 1.4609375,
      "learning_rate": 8.557856096785694e-05,
      "loss": 4.7185,
      "mean_token_accuracy": 0.2638195276260376,
      "num_tokens": 195781247.0,
      "step": 85385
    },
    {
      "entropy": 5.458194351196289,
      "epoch": 8.20268972142171,
      "grad_norm": 1.484375,
      "learning_rate": 8.556006116865429e-05,
      "loss": 4.5358,
      "mean_token_accuracy": 0.2755283534526825,
      "num_tokens": 195792660.0,
      "step": 85390
    },
    {
      "entropy": 5.533109140396118,
      "epoch": 8.203170028818445,
      "grad_norm": 1.4609375,
      "learning_rate": 8.554156576768452e-05,
      "loss": 4.6175,
      "mean_token_accuracy": 0.26507746428251266,
      "num_tokens": 195803326.0,
      "step": 85395
    },
    {
      "entropy": 5.538921976089478,
      "epoch": 8.203650336215178,
      "grad_norm": 1.3359375,
      "learning_rate": 8.5523074765377e-05,
      "loss": 4.6245,
      "mean_token_accuracy": 0.27063003480434417,
      "num_tokens": 195814906.0,
      "step": 85400
    },
    {
      "entropy": 5.421991157531738,
      "epoch": 8.204130643611911,
      "grad_norm": 1.375,
      "learning_rate": 8.550458816216102e-05,
      "loss": 4.4963,
      "mean_token_accuracy": 0.2809551268815994,
      "num_tokens": 195826482.0,
      "step": 85405
    },
    {
      "entropy": 5.519819593429565,
      "epoch": 8.204610951008645,
      "grad_norm": 1.3125,
      "learning_rate": 8.548610595846587e-05,
      "loss": 4.6206,
      "mean_token_accuracy": 0.2705475017428398,
      "num_tokens": 195839242.0,
      "step": 85410
    },
    {
      "entropy": 5.4730541706085205,
      "epoch": 8.20509125840538,
      "grad_norm": 1.390625,
      "learning_rate": 8.546762815472064e-05,
      "loss": 4.5289,
      "mean_token_accuracy": 0.27728709429502485,
      "num_tokens": 195851688.0,
      "step": 85415
    },
    {
      "entropy": 5.5418877601623535,
      "epoch": 8.205571565802114,
      "grad_norm": 1.5,
      "learning_rate": 8.54491547513542e-05,
      "loss": 4.5726,
      "mean_token_accuracy": 0.2682240352034569,
      "num_tokens": 195862134.0,
      "step": 85420
    },
    {
      "entropy": 5.567940950393677,
      "epoch": 8.206051873198847,
      "grad_norm": 1.4453125,
      "learning_rate": 8.543068574879557e-05,
      "loss": 4.6555,
      "mean_token_accuracy": 0.26902425438165667,
      "num_tokens": 195873619.0,
      "step": 85425
    },
    {
      "entropy": 5.473570919036865,
      "epoch": 8.20653218059558,
      "grad_norm": 1.4453125,
      "learning_rate": 8.541222114747352e-05,
      "loss": 4.5433,
      "mean_token_accuracy": 0.2750301033258438,
      "num_tokens": 195886345.0,
      "step": 85430
    },
    {
      "entropy": 5.543347120285034,
      "epoch": 8.207012487992316,
      "grad_norm": 1.25,
      "learning_rate": 8.539376094781672e-05,
      "loss": 4.6676,
      "mean_token_accuracy": 0.2594754949212074,
      "num_tokens": 195899045.0,
      "step": 85435
    },
    {
      "entropy": 5.509253931045532,
      "epoch": 8.207492795389049,
      "grad_norm": 1.453125,
      "learning_rate": 8.537530515025375e-05,
      "loss": 4.6088,
      "mean_token_accuracy": 0.26413109749555586,
      "num_tokens": 195910582.0,
      "step": 85440
    },
    {
      "entropy": 5.588964462280273,
      "epoch": 8.207973102785783,
      "grad_norm": 1.3203125,
      "learning_rate": 8.535685375521321e-05,
      "loss": 4.6729,
      "mean_token_accuracy": 0.2603873461484909,
      "num_tokens": 195923512.0,
      "step": 85445
    },
    {
      "entropy": 5.586812210083008,
      "epoch": 8.208453410182516,
      "grad_norm": 1.484375,
      "learning_rate": 8.533840676312335e-05,
      "loss": 4.6442,
      "mean_token_accuracy": 0.2677170723676682,
      "num_tokens": 195934728.0,
      "step": 85450
    },
    {
      "entropy": 5.556794691085815,
      "epoch": 8.208933717579251,
      "grad_norm": 1.4375,
      "learning_rate": 8.531996417441251e-05,
      "loss": 4.6558,
      "mean_token_accuracy": 0.2602480351924896,
      "num_tokens": 195946832.0,
      "step": 85455
    },
    {
      "entropy": 5.522459936141968,
      "epoch": 8.209414024975985,
      "grad_norm": 1.40625,
      "learning_rate": 8.530152598950885e-05,
      "loss": 4.598,
      "mean_token_accuracy": 0.2681323617696762,
      "num_tokens": 195959756.0,
      "step": 85460
    },
    {
      "entropy": 5.606183242797852,
      "epoch": 8.209894332372718,
      "grad_norm": 1.6328125,
      "learning_rate": 8.528309220884047e-05,
      "loss": 4.7263,
      "mean_token_accuracy": 0.2557064682245255,
      "num_tokens": 195971377.0,
      "step": 85465
    },
    {
      "entropy": 5.522199440002441,
      "epoch": 8.210374639769453,
      "grad_norm": 1.6171875,
      "learning_rate": 8.526466283283535e-05,
      "loss": 4.5856,
      "mean_token_accuracy": 0.2691436395049095,
      "num_tokens": 195981887.0,
      "step": 85470
    },
    {
      "entropy": 5.527181673049927,
      "epoch": 8.210854947166187,
      "grad_norm": 1.4375,
      "learning_rate": 8.52462378619214e-05,
      "loss": 4.6745,
      "mean_token_accuracy": 0.26229420155286787,
      "num_tokens": 195993293.0,
      "step": 85475
    },
    {
      "entropy": 5.495506143569946,
      "epoch": 8.21133525456292,
      "grad_norm": 1.4765625,
      "learning_rate": 8.522781729652632e-05,
      "loss": 4.5307,
      "mean_token_accuracy": 0.27637368738651275,
      "num_tokens": 196004487.0,
      "step": 85480
    },
    {
      "entropy": 5.518042755126953,
      "epoch": 8.211815561959654,
      "grad_norm": 1.4609375,
      "learning_rate": 8.520940113707781e-05,
      "loss": 4.5884,
      "mean_token_accuracy": 0.2751075968146324,
      "num_tokens": 196015224.0,
      "step": 85485
    },
    {
      "entropy": 5.629325294494629,
      "epoch": 8.212295869356389,
      "grad_norm": 1.3515625,
      "learning_rate": 8.519098938400344e-05,
      "loss": 4.6991,
      "mean_token_accuracy": 0.2599652484059334,
      "num_tokens": 196027187.0,
      "step": 85490
    },
    {
      "entropy": 5.396455574035644,
      "epoch": 8.212776176753122,
      "grad_norm": 1.4375,
      "learning_rate": 8.517258203773068e-05,
      "loss": 4.5114,
      "mean_token_accuracy": 0.2732197538018227,
      "num_tokens": 196038524.0,
      "step": 85495
    },
    {
      "entropy": 5.479953145980835,
      "epoch": 8.213256484149856,
      "grad_norm": 1.3671875,
      "learning_rate": 8.51541790986869e-05,
      "loss": 4.5976,
      "mean_token_accuracy": 0.275064954161644,
      "num_tokens": 196052163.0,
      "step": 85500
    },
    {
      "entropy": 5.571578454971314,
      "epoch": 8.213736791546589,
      "grad_norm": 1.5859375,
      "learning_rate": 8.51357805672994e-05,
      "loss": 4.6453,
      "mean_token_accuracy": 0.2684711545705795,
      "num_tokens": 196062970.0,
      "step": 85505
    },
    {
      "entropy": 5.521791410446167,
      "epoch": 8.214217098943324,
      "grad_norm": 1.296875,
      "learning_rate": 8.511738644399524e-05,
      "loss": 4.6688,
      "mean_token_accuracy": 0.26512532234191893,
      "num_tokens": 196075181.0,
      "step": 85510
    },
    {
      "entropy": 5.510292816162109,
      "epoch": 8.214697406340058,
      "grad_norm": 1.28125,
      "learning_rate": 8.509899672920151e-05,
      "loss": 4.6065,
      "mean_token_accuracy": 0.2689089924097061,
      "num_tokens": 196087031.0,
      "step": 85515
    },
    {
      "entropy": 5.569386768341064,
      "epoch": 8.215177713736791,
      "grad_norm": 1.5546875,
      "learning_rate": 8.508061142334524e-05,
      "loss": 4.6728,
      "mean_token_accuracy": 0.262909422814846,
      "num_tokens": 196097443.0,
      "step": 85520
    },
    {
      "entropy": 5.54618091583252,
      "epoch": 8.215658021133525,
      "grad_norm": 1.453125,
      "learning_rate": 8.506223052685319e-05,
      "loss": 4.6006,
      "mean_token_accuracy": 0.26392257362604143,
      "num_tokens": 196108667.0,
      "step": 85525
    },
    {
      "entropy": 5.495671606063842,
      "epoch": 8.21613832853026,
      "grad_norm": 1.4609375,
      "learning_rate": 8.504385404015216e-05,
      "loss": 4.566,
      "mean_token_accuracy": 0.27133436053991317,
      "num_tokens": 196120274.0,
      "step": 85530
    },
    {
      "entropy": 5.6047704219818115,
      "epoch": 8.216618635926993,
      "grad_norm": 1.6953125,
      "learning_rate": 8.502548196366885e-05,
      "loss": 4.6639,
      "mean_token_accuracy": 0.2697494313120842,
      "num_tokens": 196130986.0,
      "step": 85535
    },
    {
      "entropy": 5.407101440429687,
      "epoch": 8.217098943323727,
      "grad_norm": 1.484375,
      "learning_rate": 8.500711429782965e-05,
      "loss": 4.4424,
      "mean_token_accuracy": 0.289959704875946,
      "num_tokens": 196141010.0,
      "step": 85540
    },
    {
      "entropy": 5.4980250835418705,
      "epoch": 8.217579250720462,
      "grad_norm": 1.3984375,
      "learning_rate": 8.498875104306111e-05,
      "loss": 4.5697,
      "mean_token_accuracy": 0.2730893805623055,
      "num_tokens": 196152410.0,
      "step": 85545
    },
    {
      "entropy": 5.500788831710816,
      "epoch": 8.218059558117195,
      "grad_norm": 1.40625,
      "learning_rate": 8.497039219978958e-05,
      "loss": 4.5826,
      "mean_token_accuracy": 0.2703178718686104,
      "num_tokens": 196164728.0,
      "step": 85550
    },
    {
      "entropy": 5.456562232971192,
      "epoch": 8.218539865513929,
      "grad_norm": 1.3515625,
      "learning_rate": 8.495203776844124e-05,
      "loss": 4.5477,
      "mean_token_accuracy": 0.27683512568473817,
      "num_tokens": 196177235.0,
      "step": 85555
    },
    {
      "entropy": 5.504849004745483,
      "epoch": 8.219020172910662,
      "grad_norm": 1.3828125,
      "learning_rate": 8.493368774944227e-05,
      "loss": 4.6309,
      "mean_token_accuracy": 0.2638359621167183,
      "num_tokens": 196188680.0,
      "step": 85560
    },
    {
      "entropy": 5.53066611289978,
      "epoch": 8.219500480307397,
      "grad_norm": 1.2578125,
      "learning_rate": 8.491534214321873e-05,
      "loss": 4.6939,
      "mean_token_accuracy": 0.2670872151851654,
      "num_tokens": 196201605.0,
      "step": 85565
    },
    {
      "entropy": 5.514100885391235,
      "epoch": 8.21998078770413,
      "grad_norm": 1.4140625,
      "learning_rate": 8.489700095019648e-05,
      "loss": 4.586,
      "mean_token_accuracy": 0.2670822635293007,
      "num_tokens": 196212697.0,
      "step": 85570
    },
    {
      "entropy": 5.579127502441406,
      "epoch": 8.220461095100864,
      "grad_norm": 1.4453125,
      "learning_rate": 8.487866417080137e-05,
      "loss": 4.6505,
      "mean_token_accuracy": 0.2660963535308838,
      "num_tokens": 196224094.0,
      "step": 85575
    },
    {
      "entropy": 5.539628648757935,
      "epoch": 8.220941402497598,
      "grad_norm": 1.4296875,
      "learning_rate": 8.486033180545912e-05,
      "loss": 4.6121,
      "mean_token_accuracy": 0.26525899320840834,
      "num_tokens": 196236090.0,
      "step": 85580
    },
    {
      "entropy": 5.569957256317139,
      "epoch": 8.221421709894333,
      "grad_norm": 1.46875,
      "learning_rate": 8.484200385459537e-05,
      "loss": 4.5842,
      "mean_token_accuracy": 0.2710429459810257,
      "num_tokens": 196247330.0,
      "step": 85585
    },
    {
      "entropy": 5.593755483627319,
      "epoch": 8.221902017291066,
      "grad_norm": 1.5,
      "learning_rate": 8.482368031863565e-05,
      "loss": 4.67,
      "mean_token_accuracy": 0.26900199353694915,
      "num_tokens": 196259109.0,
      "step": 85590
    },
    {
      "entropy": 5.49825086593628,
      "epoch": 8.2223823246878,
      "grad_norm": 1.359375,
      "learning_rate": 8.480536119800541e-05,
      "loss": 4.5586,
      "mean_token_accuracy": 0.27471124529838564,
      "num_tokens": 196271087.0,
      "step": 85595
    },
    {
      "entropy": 5.435248994827271,
      "epoch": 8.222862632084533,
      "grad_norm": 1.3046875,
      "learning_rate": 8.478704649312985e-05,
      "loss": 4.44,
      "mean_token_accuracy": 0.2811268031597137,
      "num_tokens": 196282530.0,
      "step": 85600
    },
    {
      "entropy": 5.528384828567505,
      "epoch": 8.223342939481268,
      "grad_norm": 1.4609375,
      "learning_rate": 8.476873620443427e-05,
      "loss": 4.5842,
      "mean_token_accuracy": 0.2740258857607841,
      "num_tokens": 196293951.0,
      "step": 85605
    },
    {
      "entropy": 5.567494869232178,
      "epoch": 8.223823246878002,
      "grad_norm": 1.484375,
      "learning_rate": 8.475043033234377e-05,
      "loss": 4.6498,
      "mean_token_accuracy": 0.2622440531849861,
      "num_tokens": 196304435.0,
      "step": 85610
    },
    {
      "entropy": 5.6432051181793215,
      "epoch": 8.224303554274735,
      "grad_norm": 1.5,
      "learning_rate": 8.473212887728337e-05,
      "loss": 4.6684,
      "mean_token_accuracy": 0.26796513199806216,
      "num_tokens": 196316533.0,
      "step": 85615
    },
    {
      "entropy": 5.562815189361572,
      "epoch": 8.22478386167147,
      "grad_norm": 1.3828125,
      "learning_rate": 8.471383183967792e-05,
      "loss": 4.6814,
      "mean_token_accuracy": 0.2634744241833687,
      "num_tokens": 196327952.0,
      "step": 85620
    },
    {
      "entropy": 5.523191833496094,
      "epoch": 8.225264169068204,
      "grad_norm": 1.3203125,
      "learning_rate": 8.469553921995235e-05,
      "loss": 4.56,
      "mean_token_accuracy": 0.2692148879170418,
      "num_tokens": 196340497.0,
      "step": 85625
    },
    {
      "entropy": 5.562609052658081,
      "epoch": 8.225744476464937,
      "grad_norm": 1.4921875,
      "learning_rate": 8.46772510185312e-05,
      "loss": 4.6173,
      "mean_token_accuracy": 0.27127507627010344,
      "num_tokens": 196351333.0,
      "step": 85630
    },
    {
      "entropy": 5.530902147293091,
      "epoch": 8.22622478386167,
      "grad_norm": 1.421875,
      "learning_rate": 8.465896723583913e-05,
      "loss": 4.5544,
      "mean_token_accuracy": 0.26947706043720243,
      "num_tokens": 196362832.0,
      "step": 85635
    },
    {
      "entropy": 5.511087369918823,
      "epoch": 8.226705091258406,
      "grad_norm": 1.3046875,
      "learning_rate": 8.464068787230064e-05,
      "loss": 4.6216,
      "mean_token_accuracy": 0.27340648472309115,
      "num_tokens": 196375560.0,
      "step": 85640
    },
    {
      "entropy": 5.493729496002198,
      "epoch": 8.22718539865514,
      "grad_norm": 1.390625,
      "learning_rate": 8.462241292834015e-05,
      "loss": 4.5997,
      "mean_token_accuracy": 0.26752253621816635,
      "num_tokens": 196387957.0,
      "step": 85645
    },
    {
      "entropy": 5.459388303756714,
      "epoch": 8.227665706051873,
      "grad_norm": 1.5,
      "learning_rate": 8.460414240438191e-05,
      "loss": 4.5279,
      "mean_token_accuracy": 0.27323202788829803,
      "num_tokens": 196399582.0,
      "step": 85650
    },
    {
      "entropy": 5.565158748626709,
      "epoch": 8.228146013448606,
      "grad_norm": 1.453125,
      "learning_rate": 8.458587630085018e-05,
      "loss": 4.6332,
      "mean_token_accuracy": 0.26512746810913085,
      "num_tokens": 196412027.0,
      "step": 85655
    },
    {
      "entropy": 5.47750563621521,
      "epoch": 8.228626320845342,
      "grad_norm": 1.4375,
      "learning_rate": 8.456761461816895e-05,
      "loss": 4.5807,
      "mean_token_accuracy": 0.2747137576341629,
      "num_tokens": 196423243.0,
      "step": 85660
    },
    {
      "entropy": 5.504264163970947,
      "epoch": 8.229106628242075,
      "grad_norm": 1.5859375,
      "learning_rate": 8.454935735676223e-05,
      "loss": 4.5407,
      "mean_token_accuracy": 0.27886347472667694,
      "num_tokens": 196434255.0,
      "step": 85665
    },
    {
      "entropy": 5.544213342666626,
      "epoch": 8.229586935638808,
      "grad_norm": 1.4375,
      "learning_rate": 8.453110451705393e-05,
      "loss": 4.6149,
      "mean_token_accuracy": 0.2716964021325111,
      "num_tokens": 196446110.0,
      "step": 85670
    },
    {
      "entropy": 5.538180303573609,
      "epoch": 8.230067243035542,
      "grad_norm": 1.546875,
      "learning_rate": 8.451285609946777e-05,
      "loss": 4.5962,
      "mean_token_accuracy": 0.26592586636543275,
      "num_tokens": 196456662.0,
      "step": 85675
    },
    {
      "entropy": 5.489198827743531,
      "epoch": 8.230547550432277,
      "grad_norm": 1.421875,
      "learning_rate": 8.449461210442747e-05,
      "loss": 4.6089,
      "mean_token_accuracy": 0.2703761547803879,
      "num_tokens": 196468788.0,
      "step": 85680
    },
    {
      "entropy": 5.555047035217285,
      "epoch": 8.23102785782901,
      "grad_norm": 1.3515625,
      "learning_rate": 8.447637253235666e-05,
      "loss": 4.6652,
      "mean_token_accuracy": 0.2666927918791771,
      "num_tokens": 196479961.0,
      "step": 85685
    },
    {
      "entropy": 5.479959630966187,
      "epoch": 8.231508165225744,
      "grad_norm": 1.328125,
      "learning_rate": 8.445813738367866e-05,
      "loss": 4.5417,
      "mean_token_accuracy": 0.2715353459119797,
      "num_tokens": 196493297.0,
      "step": 85690
    },
    {
      "entropy": 5.485637998580932,
      "epoch": 8.23198847262248,
      "grad_norm": 1.4765625,
      "learning_rate": 8.443990665881696e-05,
      "loss": 4.5612,
      "mean_token_accuracy": 0.27121142446994784,
      "num_tokens": 196504834.0,
      "step": 85695
    },
    {
      "entropy": 5.481137371063232,
      "epoch": 8.232468780019213,
      "grad_norm": 1.421875,
      "learning_rate": 8.442168035819475e-05,
      "loss": 4.5374,
      "mean_token_accuracy": 0.2755940333008766,
      "num_tokens": 196516322.0,
      "step": 85700
    },
    {
      "entropy": 5.511570787429809,
      "epoch": 8.232949087415946,
      "grad_norm": 1.453125,
      "learning_rate": 8.440345848223521e-05,
      "loss": 4.5939,
      "mean_token_accuracy": 0.26423742175102233,
      "num_tokens": 196528638.0,
      "step": 85705
    },
    {
      "entropy": 5.591562271118164,
      "epoch": 8.23342939481268,
      "grad_norm": 1.515625,
      "learning_rate": 8.438524103136144e-05,
      "loss": 4.6596,
      "mean_token_accuracy": 0.2641114994883537,
      "num_tokens": 196540451.0,
      "step": 85710
    },
    {
      "entropy": 5.522779178619385,
      "epoch": 8.233909702209415,
      "grad_norm": 1.4609375,
      "learning_rate": 8.436702800599641e-05,
      "loss": 4.5961,
      "mean_token_accuracy": 0.2707219928503036,
      "num_tokens": 196551093.0,
      "step": 85715
    },
    {
      "entropy": 5.48381781578064,
      "epoch": 8.234390009606148,
      "grad_norm": 1.5546875,
      "learning_rate": 8.434881940656287e-05,
      "loss": 4.575,
      "mean_token_accuracy": 0.26681980341672895,
      "num_tokens": 196561634.0,
      "step": 85720
    },
    {
      "entropy": 5.533366537094116,
      "epoch": 8.234870317002882,
      "grad_norm": 1.5078125,
      "learning_rate": 8.43306152334836e-05,
      "loss": 4.6565,
      "mean_token_accuracy": 0.2591936722397804,
      "num_tokens": 196573587.0,
      "step": 85725
    },
    {
      "entropy": 5.5501933097839355,
      "epoch": 8.235350624399615,
      "grad_norm": 1.3203125,
      "learning_rate": 8.43124154871813e-05,
      "loss": 4.6464,
      "mean_token_accuracy": 0.2636237651109695,
      "num_tokens": 196585364.0,
      "step": 85730
    },
    {
      "entropy": 5.444081830978393,
      "epoch": 8.23583093179635,
      "grad_norm": 1.53125,
      "learning_rate": 8.429422016807847e-05,
      "loss": 4.4686,
      "mean_token_accuracy": 0.2753048211336136,
      "num_tokens": 196597023.0,
      "step": 85735
    },
    {
      "entropy": 5.530288171768189,
      "epoch": 8.236311239193084,
      "grad_norm": 1.5859375,
      "learning_rate": 8.427602927659758e-05,
      "loss": 4.6574,
      "mean_token_accuracy": 0.26403434574604034,
      "num_tokens": 196608846.0,
      "step": 85740
    },
    {
      "entropy": 5.429499816894531,
      "epoch": 8.236791546589817,
      "grad_norm": 1.3984375,
      "learning_rate": 8.425784281316101e-05,
      "loss": 4.4791,
      "mean_token_accuracy": 0.27679352164268495,
      "num_tokens": 196620664.0,
      "step": 85745
    },
    {
      "entropy": 5.53154993057251,
      "epoch": 8.23727185398655,
      "grad_norm": 1.46875,
      "learning_rate": 8.423966077819089e-05,
      "loss": 4.6374,
      "mean_token_accuracy": 0.27305111289024353,
      "num_tokens": 196631324.0,
      "step": 85750
    },
    {
      "entropy": 5.590594720840454,
      "epoch": 8.237752161383286,
      "grad_norm": 1.46875,
      "learning_rate": 8.42214831721094e-05,
      "loss": 4.6705,
      "mean_token_accuracy": 0.26772174537181853,
      "num_tokens": 196641623.0,
      "step": 85755
    },
    {
      "entropy": 5.505659580230713,
      "epoch": 8.23823246878002,
      "grad_norm": 1.390625,
      "learning_rate": 8.420330999533858e-05,
      "loss": 4.6695,
      "mean_token_accuracy": 0.2695920869708061,
      "num_tokens": 196652880.0,
      "step": 85760
    },
    {
      "entropy": 5.54866590499878,
      "epoch": 8.238712776176753,
      "grad_norm": 1.328125,
      "learning_rate": 8.418514124830035e-05,
      "loss": 4.6197,
      "mean_token_accuracy": 0.27078564912080766,
      "num_tokens": 196664998.0,
      "step": 85765
    },
    {
      "entropy": 5.521077394485474,
      "epoch": 8.239193083573488,
      "grad_norm": 1.6484375,
      "learning_rate": 8.416697693141653e-05,
      "loss": 4.6147,
      "mean_token_accuracy": 0.26206685304641725,
      "num_tokens": 196675301.0,
      "step": 85770
    },
    {
      "entropy": 5.548341608047485,
      "epoch": 8.239673390970221,
      "grad_norm": 1.4375,
      "learning_rate": 8.41488170451089e-05,
      "loss": 4.6301,
      "mean_token_accuracy": 0.2659137979149818,
      "num_tokens": 196687888.0,
      "step": 85775
    },
    {
      "entropy": 5.516104745864868,
      "epoch": 8.240153698366955,
      "grad_norm": 1.546875,
      "learning_rate": 8.4130661589799e-05,
      "loss": 4.5343,
      "mean_token_accuracy": 0.27259727716445925,
      "num_tokens": 196698979.0,
      "step": 85780
    },
    {
      "entropy": 5.584949350357055,
      "epoch": 8.240634005763688,
      "grad_norm": 1.2890625,
      "learning_rate": 8.411251056590834e-05,
      "loss": 4.6584,
      "mean_token_accuracy": 0.26327411979436877,
      "num_tokens": 196712887.0,
      "step": 85785
    },
    {
      "entropy": 5.47816686630249,
      "epoch": 8.241114313160423,
      "grad_norm": 1.5703125,
      "learning_rate": 8.409436397385839e-05,
      "loss": 4.5622,
      "mean_token_accuracy": 0.27606991678476334,
      "num_tokens": 196724053.0,
      "step": 85790
    },
    {
      "entropy": 5.503707838058472,
      "epoch": 8.241594620557157,
      "grad_norm": 1.4921875,
      "learning_rate": 8.40762218140704e-05,
      "loss": 4.5714,
      "mean_token_accuracy": 0.272924667596817,
      "num_tokens": 196733717.0,
      "step": 85795
    },
    {
      "entropy": 5.528609991073608,
      "epoch": 8.24207492795389,
      "grad_norm": 1.5,
      "learning_rate": 8.405808408696564e-05,
      "loss": 4.6342,
      "mean_token_accuracy": 0.26617713570594786,
      "num_tokens": 196744565.0,
      "step": 85800
    },
    {
      "entropy": 5.5256446361541744,
      "epoch": 8.242555235350624,
      "grad_norm": 1.5,
      "learning_rate": 8.403995079296519e-05,
      "loss": 4.6434,
      "mean_token_accuracy": 0.2625319823622704,
      "num_tokens": 196755921.0,
      "step": 85805
    },
    {
      "entropy": 5.52718677520752,
      "epoch": 8.243035542747359,
      "grad_norm": 1.53125,
      "learning_rate": 8.402182193249011e-05,
      "loss": 4.5993,
      "mean_token_accuracy": 0.27530261129140854,
      "num_tokens": 196766439.0,
      "step": 85810
    },
    {
      "entropy": 5.482557535171509,
      "epoch": 8.243515850144092,
      "grad_norm": 1.4609375,
      "learning_rate": 8.400369750596115e-05,
      "loss": 4.5718,
      "mean_token_accuracy": 0.27478078603744505,
      "num_tokens": 196778670.0,
      "step": 85815
    },
    {
      "entropy": 5.448480463027954,
      "epoch": 8.243996157540826,
      "grad_norm": 1.4140625,
      "learning_rate": 8.398557751379921e-05,
      "loss": 4.5195,
      "mean_token_accuracy": 0.2761835753917694,
      "num_tokens": 196789868.0,
      "step": 85820
    },
    {
      "entropy": 5.5292142868042,
      "epoch": 8.24447646493756,
      "grad_norm": 1.3515625,
      "learning_rate": 8.396746195642497e-05,
      "loss": 4.637,
      "mean_token_accuracy": 0.26928853541612624,
      "num_tokens": 196802165.0,
      "step": 85825
    },
    {
      "entropy": 5.521185398101807,
      "epoch": 8.244956772334294,
      "grad_norm": 1.3984375,
      "learning_rate": 8.3949350834259e-05,
      "loss": 4.6007,
      "mean_token_accuracy": 0.2628257155418396,
      "num_tokens": 196814702.0,
      "step": 85830
    },
    {
      "entropy": 5.47588210105896,
      "epoch": 8.245437079731028,
      "grad_norm": 1.3515625,
      "learning_rate": 8.393124414772181e-05,
      "loss": 4.5377,
      "mean_token_accuracy": 0.27057969123125075,
      "num_tokens": 196826652.0,
      "step": 85835
    },
    {
      "entropy": 5.553370189666748,
      "epoch": 8.245917387127761,
      "grad_norm": 1.3828125,
      "learning_rate": 8.391314189723381e-05,
      "loss": 4.6451,
      "mean_token_accuracy": 0.26176590770483016,
      "num_tokens": 196837612.0,
      "step": 85840
    },
    {
      "entropy": 5.504872512817383,
      "epoch": 8.246397694524497,
      "grad_norm": 1.5,
      "learning_rate": 8.389504408321521e-05,
      "loss": 4.6188,
      "mean_token_accuracy": 0.26320619136095047,
      "num_tokens": 196848335.0,
      "step": 85845
    },
    {
      "entropy": 5.602827167510986,
      "epoch": 8.24687800192123,
      "grad_norm": 1.46875,
      "learning_rate": 8.387695070608622e-05,
      "loss": 4.713,
      "mean_token_accuracy": 0.2631349891424179,
      "num_tokens": 196859592.0,
      "step": 85850
    },
    {
      "entropy": 5.518063116073608,
      "epoch": 8.247358309317963,
      "grad_norm": 1.453125,
      "learning_rate": 8.385886176626689e-05,
      "loss": 4.5536,
      "mean_token_accuracy": 0.27235018610954287,
      "num_tokens": 196870861.0,
      "step": 85855
    },
    {
      "entropy": 5.507609939575195,
      "epoch": 8.247838616714697,
      "grad_norm": 1.4296875,
      "learning_rate": 8.384077726417724e-05,
      "loss": 4.5333,
      "mean_token_accuracy": 0.27401872873306277,
      "num_tokens": 196882796.0,
      "step": 85860
    },
    {
      "entropy": 5.506363201141357,
      "epoch": 8.248318924111432,
      "grad_norm": 1.6015625,
      "learning_rate": 8.382269720023713e-05,
      "loss": 4.5675,
      "mean_token_accuracy": 0.2724496006965637,
      "num_tokens": 196893691.0,
      "step": 85865
    },
    {
      "entropy": 5.580466604232788,
      "epoch": 8.248799231508166,
      "grad_norm": 1.515625,
      "learning_rate": 8.38046215748663e-05,
      "loss": 4.6603,
      "mean_token_accuracy": 0.270893520116806,
      "num_tokens": 196904519.0,
      "step": 85870
    },
    {
      "entropy": 5.539703130722046,
      "epoch": 8.249279538904899,
      "grad_norm": 1.515625,
      "learning_rate": 8.378655038848445e-05,
      "loss": 4.6004,
      "mean_token_accuracy": 0.2642316594719887,
      "num_tokens": 196916120.0,
      "step": 85875
    },
    {
      "entropy": 5.490382099151612,
      "epoch": 8.249759846301632,
      "grad_norm": 1.4375,
      "learning_rate": 8.376848364151108e-05,
      "loss": 4.5416,
      "mean_token_accuracy": 0.2755592823028564,
      "num_tokens": 196927093.0,
      "step": 85880
    },
    {
      "entropy": 5.602187204360962,
      "epoch": 8.250240153698368,
      "grad_norm": 1.5859375,
      "learning_rate": 8.375042133436569e-05,
      "loss": 4.6392,
      "mean_token_accuracy": 0.26574267596006396,
      "num_tokens": 196939182.0,
      "step": 85885
    },
    {
      "entropy": 5.470438241958618,
      "epoch": 8.250720461095101,
      "grad_norm": 1.4921875,
      "learning_rate": 8.37323634674676e-05,
      "loss": 4.632,
      "mean_token_accuracy": 0.2679363012313843,
      "num_tokens": 196950199.0,
      "step": 85890
    },
    {
      "entropy": 5.463978481292725,
      "epoch": 8.251200768491834,
      "grad_norm": 1.5,
      "learning_rate": 8.371431004123608e-05,
      "loss": 4.5917,
      "mean_token_accuracy": 0.26789843291044235,
      "num_tokens": 196960966.0,
      "step": 85895
    },
    {
      "entropy": 5.5048198223114015,
      "epoch": 8.251681075888568,
      "grad_norm": 1.4296875,
      "learning_rate": 8.36962610560903e-05,
      "loss": 4.6177,
      "mean_token_accuracy": 0.26281833946704863,
      "num_tokens": 196973468.0,
      "step": 85900
    },
    {
      "entropy": 5.488169097900391,
      "epoch": 8.252161383285303,
      "grad_norm": 1.4609375,
      "learning_rate": 8.367821651244928e-05,
      "loss": 4.5758,
      "mean_token_accuracy": 0.27307667434215543,
      "num_tokens": 196985926.0,
      "step": 85905
    },
    {
      "entropy": 5.52816367149353,
      "epoch": 8.252641690682037,
      "grad_norm": 1.4453125,
      "learning_rate": 8.366017641073195e-05,
      "loss": 4.6346,
      "mean_token_accuracy": 0.25944941192865373,
      "num_tokens": 196997446.0,
      "step": 85910
    },
    {
      "entropy": 5.424450063705445,
      "epoch": 8.25312199807877,
      "grad_norm": 1.4140625,
      "learning_rate": 8.364214075135723e-05,
      "loss": 4.4787,
      "mean_token_accuracy": 0.2759236693382263,
      "num_tokens": 197008683.0,
      "step": 85915
    },
    {
      "entropy": 5.4929718494415285,
      "epoch": 8.253602305475503,
      "grad_norm": 1.484375,
      "learning_rate": 8.362410953474371e-05,
      "loss": 4.5422,
      "mean_token_accuracy": 0.2778706654906273,
      "num_tokens": 197020134.0,
      "step": 85920
    },
    {
      "entropy": 5.49965181350708,
      "epoch": 8.254082612872239,
      "grad_norm": 1.5859375,
      "learning_rate": 8.360608276131012e-05,
      "loss": 4.5486,
      "mean_token_accuracy": 0.2792858242988586,
      "num_tokens": 197031095.0,
      "step": 85925
    },
    {
      "entropy": 5.5307739734649655,
      "epoch": 8.254562920268972,
      "grad_norm": 1.5625,
      "learning_rate": 8.358806043147494e-05,
      "loss": 4.6668,
      "mean_token_accuracy": 0.26692967861890793,
      "num_tokens": 197043568.0,
      "step": 85930
    },
    {
      "entropy": 5.49798321723938,
      "epoch": 8.255043227665706,
      "grad_norm": 1.5078125,
      "learning_rate": 8.357004254565664e-05,
      "loss": 4.568,
      "mean_token_accuracy": 0.2717917010188103,
      "num_tokens": 197055157.0,
      "step": 85935
    },
    {
      "entropy": 5.502383232116699,
      "epoch": 8.25552353506244,
      "grad_norm": 1.7109375,
      "learning_rate": 8.355202910427353e-05,
      "loss": 4.5686,
      "mean_token_accuracy": 0.2706145241856575,
      "num_tokens": 197065050.0,
      "step": 85940
    },
    {
      "entropy": 5.495261907577515,
      "epoch": 8.256003842459174,
      "grad_norm": 1.4453125,
      "learning_rate": 8.353402010774382e-05,
      "loss": 4.5657,
      "mean_token_accuracy": 0.269625386595726,
      "num_tokens": 197075757.0,
      "step": 85945
    },
    {
      "entropy": 5.554506254196167,
      "epoch": 8.256484149855908,
      "grad_norm": 1.3828125,
      "learning_rate": 8.35160155564856e-05,
      "loss": 4.5873,
      "mean_token_accuracy": 0.2786561965942383,
      "num_tokens": 197085525.0,
      "step": 85950
    },
    {
      "entropy": 5.5787629127502445,
      "epoch": 8.256964457252641,
      "grad_norm": 1.46875,
      "learning_rate": 8.349801545091699e-05,
      "loss": 4.6814,
      "mean_token_accuracy": 0.25666002929210663,
      "num_tokens": 197097083.0,
      "step": 85955
    },
    {
      "entropy": 5.530234050750733,
      "epoch": 8.257444764649376,
      "grad_norm": 1.515625,
      "learning_rate": 8.348001979145574e-05,
      "loss": 4.617,
      "mean_token_accuracy": 0.26384112387895586,
      "num_tokens": 197108327.0,
      "step": 85960
    },
    {
      "entropy": 5.592846965789795,
      "epoch": 8.25792507204611,
      "grad_norm": 1.515625,
      "learning_rate": 8.346202857851975e-05,
      "loss": 4.735,
      "mean_token_accuracy": 0.2651759535074234,
      "num_tokens": 197118648.0,
      "step": 85965
    },
    {
      "entropy": 5.552652835845947,
      "epoch": 8.258405379442843,
      "grad_norm": 1.46875,
      "learning_rate": 8.34440418125267e-05,
      "loss": 4.6569,
      "mean_token_accuracy": 0.2660650357604027,
      "num_tokens": 197129192.0,
      "step": 85970
    },
    {
      "entropy": 5.59385461807251,
      "epoch": 8.258885686839577,
      "grad_norm": 1.5,
      "learning_rate": 8.34260594938942e-05,
      "loss": 4.6506,
      "mean_token_accuracy": 0.26392724066972734,
      "num_tokens": 197141305.0,
      "step": 85975
    },
    {
      "entropy": 5.510876560211182,
      "epoch": 8.259365994236312,
      "grad_norm": 1.4140625,
      "learning_rate": 8.340808162303973e-05,
      "loss": 4.6389,
      "mean_token_accuracy": 0.2682137072086334,
      "num_tokens": 197153551.0,
      "step": 85980
    },
    {
      "entropy": 5.5680928230285645,
      "epoch": 8.259846301633045,
      "grad_norm": 1.46875,
      "learning_rate": 8.33901082003807e-05,
      "loss": 4.6416,
      "mean_token_accuracy": 0.26553458273410796,
      "num_tokens": 197165348.0,
      "step": 85985
    },
    {
      "entropy": 5.5171043395996096,
      "epoch": 8.260326609029779,
      "grad_norm": 1.3671875,
      "learning_rate": 8.337213922633444e-05,
      "loss": 4.6251,
      "mean_token_accuracy": 0.2667085438966751,
      "num_tokens": 197176275.0,
      "step": 85990
    },
    {
      "entropy": 5.481961441040039,
      "epoch": 8.260806916426514,
      "grad_norm": 1.3203125,
      "learning_rate": 8.335417470131806e-05,
      "loss": 4.527,
      "mean_token_accuracy": 0.2761286646127701,
      "num_tokens": 197189320.0,
      "step": 85995
    },
    {
      "entropy": 5.508462381362915,
      "epoch": 8.261287223823247,
      "grad_norm": 1.4609375,
      "learning_rate": 8.333621462574864e-05,
      "loss": 4.6587,
      "mean_token_accuracy": 0.2666309744119644,
      "num_tokens": 197202038.0,
      "step": 86000
    },
    {
      "entropy": 5.535906171798706,
      "epoch": 8.26176753121998,
      "grad_norm": 1.4921875,
      "learning_rate": 8.331825900004322e-05,
      "loss": 4.6683,
      "mean_token_accuracy": 0.2632962137460709,
      "num_tokens": 197214587.0,
      "step": 86005
    },
    {
      "entropy": 5.5327057361602785,
      "epoch": 8.262247838616714,
      "grad_norm": 1.421875,
      "learning_rate": 8.330030782461861e-05,
      "loss": 4.5931,
      "mean_token_accuracy": 0.26351550668478013,
      "num_tokens": 197225798.0,
      "step": 86010
    },
    {
      "entropy": 5.502474689483643,
      "epoch": 8.26272814601345,
      "grad_norm": 1.40625,
      "learning_rate": 8.328236109989165e-05,
      "loss": 4.6116,
      "mean_token_accuracy": 0.26646713465452193,
      "num_tokens": 197237887.0,
      "step": 86015
    },
    {
      "entropy": 5.607561302185059,
      "epoch": 8.263208453410183,
      "grad_norm": 1.359375,
      "learning_rate": 8.326441882627903e-05,
      "loss": 4.6946,
      "mean_token_accuracy": 0.26528625637292863,
      "num_tokens": 197250220.0,
      "step": 86020
    },
    {
      "entropy": 5.5285499572753904,
      "epoch": 8.263688760806916,
      "grad_norm": 1.5234375,
      "learning_rate": 8.324648100419722e-05,
      "loss": 4.6093,
      "mean_token_accuracy": 0.2709269553422928,
      "num_tokens": 197261120.0,
      "step": 86025
    },
    {
      "entropy": 5.5828502655029295,
      "epoch": 8.26416906820365,
      "grad_norm": 1.5,
      "learning_rate": 8.32285476340627e-05,
      "loss": 4.6763,
      "mean_token_accuracy": 0.26540988087654116,
      "num_tokens": 197270933.0,
      "step": 86030
    },
    {
      "entropy": 5.518478441238403,
      "epoch": 8.264649375600385,
      "grad_norm": 1.59375,
      "learning_rate": 8.32106187162919e-05,
      "loss": 4.5623,
      "mean_token_accuracy": 0.2728344798088074,
      "num_tokens": 197280790.0,
      "step": 86035
    },
    {
      "entropy": 5.489998054504395,
      "epoch": 8.265129682997118,
      "grad_norm": 1.4375,
      "learning_rate": 8.319269425130102e-05,
      "loss": 4.51,
      "mean_token_accuracy": 0.2784385621547699,
      "num_tokens": 197291741.0,
      "step": 86040
    },
    {
      "entropy": 5.516088056564331,
      "epoch": 8.265609990393852,
      "grad_norm": 1.421875,
      "learning_rate": 8.317477423950622e-05,
      "loss": 4.6357,
      "mean_token_accuracy": 0.2693154662847519,
      "num_tokens": 197302495.0,
      "step": 86045
    },
    {
      "entropy": 5.483275938034057,
      "epoch": 8.266090297790585,
      "grad_norm": 1.3671875,
      "learning_rate": 8.315685868132364e-05,
      "loss": 4.5082,
      "mean_token_accuracy": 0.2684525534510612,
      "num_tokens": 197313662.0,
      "step": 86050
    },
    {
      "entropy": 5.497531080245972,
      "epoch": 8.26657060518732,
      "grad_norm": 1.375,
      "learning_rate": 8.313894757716905e-05,
      "loss": 4.5505,
      "mean_token_accuracy": 0.2687826856970787,
      "num_tokens": 197325586.0,
      "step": 86055
    },
    {
      "entropy": 5.521406936645508,
      "epoch": 8.267050912584054,
      "grad_norm": 1.3671875,
      "learning_rate": 8.312104092745841e-05,
      "loss": 4.6191,
      "mean_token_accuracy": 0.26763536632061,
      "num_tokens": 197337647.0,
      "step": 86060
    },
    {
      "entropy": 5.500399160385132,
      "epoch": 8.267531219980787,
      "grad_norm": 1.515625,
      "learning_rate": 8.310313873260744e-05,
      "loss": 4.6295,
      "mean_token_accuracy": 0.26726947724819183,
      "num_tokens": 197348774.0,
      "step": 86065
    },
    {
      "entropy": 5.47406415939331,
      "epoch": 8.26801152737752,
      "grad_norm": 1.34375,
      "learning_rate": 8.308524099303175e-05,
      "loss": 4.5342,
      "mean_token_accuracy": 0.2786698192358017,
      "num_tokens": 197361709.0,
      "step": 86070
    },
    {
      "entropy": 5.481706428527832,
      "epoch": 8.268491834774256,
      "grad_norm": 1.3515625,
      "learning_rate": 8.306734770914693e-05,
      "loss": 4.6463,
      "mean_token_accuracy": 0.26779884546995164,
      "num_tokens": 197374693.0,
      "step": 86075
    },
    {
      "entropy": 5.435806751251221,
      "epoch": 8.26897214217099,
      "grad_norm": 1.421875,
      "learning_rate": 8.30494588813684e-05,
      "loss": 4.4989,
      "mean_token_accuracy": 0.28325988054275514,
      "num_tokens": 197388198.0,
      "step": 86080
    },
    {
      "entropy": 5.500381708145142,
      "epoch": 8.269452449567723,
      "grad_norm": 1.4140625,
      "learning_rate": 8.30315745101114e-05,
      "loss": 4.5617,
      "mean_token_accuracy": 0.27212030738592147,
      "num_tokens": 197400728.0,
      "step": 86085
    },
    {
      "entropy": 5.533185386657715,
      "epoch": 8.269932756964458,
      "grad_norm": 1.5078125,
      "learning_rate": 8.301369459579123e-05,
      "loss": 4.5371,
      "mean_token_accuracy": 0.27300033569335935,
      "num_tokens": 197411317.0,
      "step": 86090
    },
    {
      "entropy": 5.577410125732422,
      "epoch": 8.270413064361192,
      "grad_norm": 1.53125,
      "learning_rate": 8.299581913882298e-05,
      "loss": 4.6912,
      "mean_token_accuracy": 0.2560177847743034,
      "num_tokens": 197421249.0,
      "step": 86095
    },
    {
      "entropy": 5.5440507411956785,
      "epoch": 8.270893371757925,
      "grad_norm": 1.5,
      "learning_rate": 8.297794813962167e-05,
      "loss": 4.6398,
      "mean_token_accuracy": 0.2650471180677414,
      "num_tokens": 197432364.0,
      "step": 86100
    },
    {
      "entropy": 5.4581938743591305,
      "epoch": 8.271373679154658,
      "grad_norm": 1.375,
      "learning_rate": 8.296008159860223e-05,
      "loss": 4.4905,
      "mean_token_accuracy": 0.2827514708042145,
      "num_tokens": 197443734.0,
      "step": 86105
    },
    {
      "entropy": 5.516400766372681,
      "epoch": 8.271853986551394,
      "grad_norm": 1.609375,
      "learning_rate": 8.29422195161795e-05,
      "loss": 4.5607,
      "mean_token_accuracy": 0.2815706104040146,
      "num_tokens": 197454467.0,
      "step": 86110
    },
    {
      "entropy": 5.576061534881592,
      "epoch": 8.272334293948127,
      "grad_norm": 1.5078125,
      "learning_rate": 8.292436189276809e-05,
      "loss": 4.643,
      "mean_token_accuracy": 0.2585069119930267,
      "num_tokens": 197466364.0,
      "step": 86115
    },
    {
      "entropy": 5.534551382064819,
      "epoch": 8.27281460134486,
      "grad_norm": 1.46875,
      "learning_rate": 8.290650872878265e-05,
      "loss": 4.5946,
      "mean_token_accuracy": 0.2703008934855461,
      "num_tokens": 197477263.0,
      "step": 86120
    },
    {
      "entropy": 5.512544727325439,
      "epoch": 8.273294908741594,
      "grad_norm": 1.546875,
      "learning_rate": 8.288866002463768e-05,
      "loss": 4.6038,
      "mean_token_accuracy": 0.2599830895662308,
      "num_tokens": 197487867.0,
      "step": 86125
    },
    {
      "entropy": 5.505478811264038,
      "epoch": 8.27377521613833,
      "grad_norm": 1.5078125,
      "learning_rate": 8.287081578074754e-05,
      "loss": 4.658,
      "mean_token_accuracy": 0.2662856444716454,
      "num_tokens": 197501305.0,
      "step": 86130
    },
    {
      "entropy": 5.530484247207641,
      "epoch": 8.274255523535063,
      "grad_norm": 1.5703125,
      "learning_rate": 8.285297599752657e-05,
      "loss": 4.6145,
      "mean_token_accuracy": 0.2728727325797081,
      "num_tokens": 197512466.0,
      "step": 86135
    },
    {
      "entropy": 5.458147954940796,
      "epoch": 8.274735830931796,
      "grad_norm": 1.6015625,
      "learning_rate": 8.2835140675389e-05,
      "loss": 4.5082,
      "mean_token_accuracy": 0.2773930788040161,
      "num_tokens": 197523719.0,
      "step": 86140
    },
    {
      "entropy": 5.597205400466919,
      "epoch": 8.27521613832853,
      "grad_norm": 1.4609375,
      "learning_rate": 8.281730981474881e-05,
      "loss": 4.678,
      "mean_token_accuracy": 0.2704206556081772,
      "num_tokens": 197536917.0,
      "step": 86145
    },
    {
      "entropy": 5.487534141540527,
      "epoch": 8.275696445725265,
      "grad_norm": 1.484375,
      "learning_rate": 8.279948341601999e-05,
      "loss": 4.5606,
      "mean_token_accuracy": 0.26891830414533613,
      "num_tokens": 197547468.0,
      "step": 86150
    },
    {
      "entropy": 5.5682785987854,
      "epoch": 8.276176753121998,
      "grad_norm": 1.4765625,
      "learning_rate": 8.278166147961646e-05,
      "loss": 4.6512,
      "mean_token_accuracy": 0.26642140448093415,
      "num_tokens": 197558711.0,
      "step": 86155
    },
    {
      "entropy": 5.38364086151123,
      "epoch": 8.276657060518732,
      "grad_norm": 1.4375,
      "learning_rate": 8.2763844005952e-05,
      "loss": 4.4318,
      "mean_token_accuracy": 0.29148960411548613,
      "num_tokens": 197569895.0,
      "step": 86160
    },
    {
      "entropy": 5.523050117492676,
      "epoch": 8.277137367915467,
      "grad_norm": 1.59375,
      "learning_rate": 8.274603099544022e-05,
      "loss": 4.5953,
      "mean_token_accuracy": 0.2813923329114914,
      "num_tokens": 197581297.0,
      "step": 86165
    },
    {
      "entropy": 5.4845664501190186,
      "epoch": 8.2776176753122,
      "grad_norm": 1.328125,
      "learning_rate": 8.272822244849479e-05,
      "loss": 4.546,
      "mean_token_accuracy": 0.2742637485265732,
      "num_tokens": 197592907.0,
      "step": 86170
    },
    {
      "entropy": 5.539804935455322,
      "epoch": 8.278097982708934,
      "grad_norm": 1.5390625,
      "learning_rate": 8.271041836552905e-05,
      "loss": 4.6096,
      "mean_token_accuracy": 0.273016083240509,
      "num_tokens": 197604206.0,
      "step": 86175
    },
    {
      "entropy": 5.528600358963013,
      "epoch": 8.278578290105667,
      "grad_norm": 1.5390625,
      "learning_rate": 8.269261874695641e-05,
      "loss": 4.6203,
      "mean_token_accuracy": 0.2732224762439728,
      "num_tokens": 197614687.0,
      "step": 86180
    },
    {
      "entropy": 5.539444065093994,
      "epoch": 8.279058597502402,
      "grad_norm": 1.4296875,
      "learning_rate": 8.267482359319015e-05,
      "loss": 4.6024,
      "mean_token_accuracy": 0.2683785483241081,
      "num_tokens": 197626924.0,
      "step": 86185
    },
    {
      "entropy": 5.496171665191651,
      "epoch": 8.279538904899136,
      "grad_norm": 1.6484375,
      "learning_rate": 8.265703290464338e-05,
      "loss": 4.52,
      "mean_token_accuracy": 0.27212291657924653,
      "num_tokens": 197638180.0,
      "step": 86190
    },
    {
      "entropy": 5.5898613929748535,
      "epoch": 8.28001921229587,
      "grad_norm": 1.5703125,
      "learning_rate": 8.263924668172918e-05,
      "loss": 4.6743,
      "mean_token_accuracy": 0.2588492944836617,
      "num_tokens": 197650025.0,
      "step": 86195
    },
    {
      "entropy": 5.560570192337036,
      "epoch": 8.280499519692603,
      "grad_norm": 1.4140625,
      "learning_rate": 8.262146492486053e-05,
      "loss": 4.6431,
      "mean_token_accuracy": 0.26008853763341905,
      "num_tokens": 197661224.0,
      "step": 86200
    },
    {
      "entropy": 5.456741762161255,
      "epoch": 8.280979827089338,
      "grad_norm": 1.328125,
      "learning_rate": 8.260368763445015e-05,
      "loss": 4.5797,
      "mean_token_accuracy": 0.2717523813247681,
      "num_tokens": 197673485.0,
      "step": 86205
    },
    {
      "entropy": 5.547867488861084,
      "epoch": 8.281460134486071,
      "grad_norm": 1.5234375,
      "learning_rate": 8.258591481091084e-05,
      "loss": 4.6261,
      "mean_token_accuracy": 0.2678986996412277,
      "num_tokens": 197685138.0,
      "step": 86210
    },
    {
      "entropy": 5.479272317886353,
      "epoch": 8.281940441882805,
      "grad_norm": 1.46875,
      "learning_rate": 8.256814645465525e-05,
      "loss": 4.5355,
      "mean_token_accuracy": 0.2786008194088936,
      "num_tokens": 197696289.0,
      "step": 86215
    },
    {
      "entropy": 5.537321710586548,
      "epoch": 8.282420749279538,
      "grad_norm": 1.4375,
      "learning_rate": 8.255038256609587e-05,
      "loss": 4.5893,
      "mean_token_accuracy": 0.2719172537326813,
      "num_tokens": 197706978.0,
      "step": 86220
    },
    {
      "entropy": 5.53784966468811,
      "epoch": 8.282901056676273,
      "grad_norm": 1.3359375,
      "learning_rate": 8.253262314564517e-05,
      "loss": 4.6023,
      "mean_token_accuracy": 0.2693585500121117,
      "num_tokens": 197719232.0,
      "step": 86225
    },
    {
      "entropy": 5.474792718887329,
      "epoch": 8.283381364073007,
      "grad_norm": 1.484375,
      "learning_rate": 8.25148681937155e-05,
      "loss": 4.577,
      "mean_token_accuracy": 0.27760600447654726,
      "num_tokens": 197731704.0,
      "step": 86230
    },
    {
      "entropy": 5.4823119163513185,
      "epoch": 8.28386167146974,
      "grad_norm": 1.5078125,
      "learning_rate": 8.249711771071897e-05,
      "loss": 4.539,
      "mean_token_accuracy": 0.26592080295085907,
      "num_tokens": 197742574.0,
      "step": 86235
    },
    {
      "entropy": 5.488392925262451,
      "epoch": 8.284341978866475,
      "grad_norm": 1.34375,
      "learning_rate": 8.247937169706776e-05,
      "loss": 4.6057,
      "mean_token_accuracy": 0.2640624389052391,
      "num_tokens": 197754029.0,
      "step": 86240
    },
    {
      "entropy": 5.564331674575806,
      "epoch": 8.284822286263209,
      "grad_norm": 1.4296875,
      "learning_rate": 8.246163015317385e-05,
      "loss": 4.6173,
      "mean_token_accuracy": 0.2708691135048866,
      "num_tokens": 197766218.0,
      "step": 86245
    },
    {
      "entropy": 5.4983251094818115,
      "epoch": 8.285302593659942,
      "grad_norm": 1.765625,
      "learning_rate": 8.244389307944919e-05,
      "loss": 4.5253,
      "mean_token_accuracy": 0.275181283056736,
      "num_tokens": 197776666.0,
      "step": 86250
    },
    {
      "entropy": 5.570951080322265,
      "epoch": 8.285782901056676,
      "grad_norm": 1.6171875,
      "learning_rate": 8.242616047630556e-05,
      "loss": 4.6609,
      "mean_token_accuracy": 0.25942868292331694,
      "num_tokens": 197788229.0,
      "step": 86255
    },
    {
      "entropy": 5.4811101913452145,
      "epoch": 8.286263208453411,
      "grad_norm": 1.46875,
      "learning_rate": 8.240843234415469e-05,
      "loss": 4.5566,
      "mean_token_accuracy": 0.27719674110412595,
      "num_tokens": 197798980.0,
      "step": 86260
    },
    {
      "entropy": 5.485958671569824,
      "epoch": 8.286743515850144,
      "grad_norm": 1.4765625,
      "learning_rate": 8.23907086834081e-05,
      "loss": 4.5819,
      "mean_token_accuracy": 0.2675266370177269,
      "num_tokens": 197810028.0,
      "step": 86265
    },
    {
      "entropy": 5.505485677719117,
      "epoch": 8.287223823246878,
      "grad_norm": 1.4921875,
      "learning_rate": 8.23729894944773e-05,
      "loss": 4.6056,
      "mean_token_accuracy": 0.2712463140487671,
      "num_tokens": 197821129.0,
      "step": 86270
    },
    {
      "entropy": 5.5436948299407955,
      "epoch": 8.287704130643611,
      "grad_norm": 1.328125,
      "learning_rate": 8.235527477777372e-05,
      "loss": 4.6272,
      "mean_token_accuracy": 0.26475114077329637,
      "num_tokens": 197833534.0,
      "step": 86275
    },
    {
      "entropy": 5.506768083572387,
      "epoch": 8.288184438040346,
      "grad_norm": 1.453125,
      "learning_rate": 8.233756453370863e-05,
      "loss": 4.5364,
      "mean_token_accuracy": 0.2739600300788879,
      "num_tokens": 197844857.0,
      "step": 86280
    },
    {
      "entropy": 5.523018789291382,
      "epoch": 8.28866474543708,
      "grad_norm": 1.359375,
      "learning_rate": 8.231985876269321e-05,
      "loss": 4.647,
      "mean_token_accuracy": 0.2699079617857933,
      "num_tokens": 197857937.0,
      "step": 86285
    },
    {
      "entropy": 5.561858463287353,
      "epoch": 8.289145052833813,
      "grad_norm": 1.328125,
      "learning_rate": 8.230215746513856e-05,
      "loss": 4.6162,
      "mean_token_accuracy": 0.2612989991903305,
      "num_tokens": 197869019.0,
      "step": 86290
    },
    {
      "entropy": 5.491109943389892,
      "epoch": 8.289625360230547,
      "grad_norm": 1.265625,
      "learning_rate": 8.228446064145556e-05,
      "loss": 4.5657,
      "mean_token_accuracy": 0.27366385906934737,
      "num_tokens": 197881974.0,
      "step": 86295
    },
    {
      "entropy": 5.491173267364502,
      "epoch": 8.290105667627282,
      "grad_norm": 1.40625,
      "learning_rate": 8.226676829205515e-05,
      "loss": 4.5654,
      "mean_token_accuracy": 0.2747225254774094,
      "num_tokens": 197894519.0,
      "step": 86300
    },
    {
      "entropy": 5.604162693023682,
      "epoch": 8.290585975024015,
      "grad_norm": 1.453125,
      "learning_rate": 8.224908041734808e-05,
      "loss": 4.7451,
      "mean_token_accuracy": 0.2545284956693649,
      "num_tokens": 197907002.0,
      "step": 86305
    },
    {
      "entropy": 5.49401888847351,
      "epoch": 8.291066282420749,
      "grad_norm": 1.4765625,
      "learning_rate": 8.2231397017745e-05,
      "loss": 4.6423,
      "mean_token_accuracy": 0.26320343315601347,
      "num_tokens": 197919566.0,
      "step": 86310
    },
    {
      "entropy": 5.552671146392822,
      "epoch": 8.291546589817484,
      "grad_norm": 1.4609375,
      "learning_rate": 8.221371809365646e-05,
      "loss": 4.6132,
      "mean_token_accuracy": 0.2632928043603897,
      "num_tokens": 197931664.0,
      "step": 86315
    },
    {
      "entropy": 5.492520332336426,
      "epoch": 8.292026897214217,
      "grad_norm": 1.25,
      "learning_rate": 8.219604364549302e-05,
      "loss": 4.5501,
      "mean_token_accuracy": 0.2750731334090233,
      "num_tokens": 197944197.0,
      "step": 86320
    },
    {
      "entropy": 5.554562044143677,
      "epoch": 8.292507204610951,
      "grad_norm": 1.3984375,
      "learning_rate": 8.217837367366487e-05,
      "loss": 4.6111,
      "mean_token_accuracy": 0.2687013730406761,
      "num_tokens": 197955350.0,
      "step": 86325
    },
    {
      "entropy": 5.537165832519531,
      "epoch": 8.292987512007684,
      "grad_norm": 1.390625,
      "learning_rate": 8.216070817858232e-05,
      "loss": 4.602,
      "mean_token_accuracy": 0.267085987329483,
      "num_tokens": 197967251.0,
      "step": 86330
    },
    {
      "entropy": 5.573562335968018,
      "epoch": 8.29346781940442,
      "grad_norm": 1.4921875,
      "learning_rate": 8.214304716065551e-05,
      "loss": 4.5935,
      "mean_token_accuracy": 0.2759979546070099,
      "num_tokens": 197977243.0,
      "step": 86335
    },
    {
      "entropy": 5.521477508544922,
      "epoch": 8.293948126801153,
      "grad_norm": 1.453125,
      "learning_rate": 8.212539062029448e-05,
      "loss": 4.5836,
      "mean_token_accuracy": 0.2738640680909157,
      "num_tokens": 197987883.0,
      "step": 86340
    },
    {
      "entropy": 5.574412870407104,
      "epoch": 8.294428434197886,
      "grad_norm": 1.390625,
      "learning_rate": 8.210773855790914e-05,
      "loss": 4.6932,
      "mean_token_accuracy": 0.2569715052843094,
      "num_tokens": 197999679.0,
      "step": 86345
    },
    {
      "entropy": 5.546269369125366,
      "epoch": 8.29490874159462,
      "grad_norm": 1.4375,
      "learning_rate": 8.20900909739094e-05,
      "loss": 4.638,
      "mean_token_accuracy": 0.2622259438037872,
      "num_tokens": 198010255.0,
      "step": 86350
    },
    {
      "entropy": 5.546977996826172,
      "epoch": 8.295389048991355,
      "grad_norm": 1.5234375,
      "learning_rate": 8.207244786870487e-05,
      "loss": 4.6496,
      "mean_token_accuracy": 0.26474207937717437,
      "num_tokens": 198021359.0,
      "step": 86355
    },
    {
      "entropy": 5.53995418548584,
      "epoch": 8.295869356388089,
      "grad_norm": 1.59375,
      "learning_rate": 8.205480924270525e-05,
      "loss": 4.6545,
      "mean_token_accuracy": 0.26944682002067566,
      "num_tokens": 198031982.0,
      "step": 86360
    },
    {
      "entropy": 5.484172344207764,
      "epoch": 8.296349663784822,
      "grad_norm": 1.453125,
      "learning_rate": 8.203717509632003e-05,
      "loss": 4.4968,
      "mean_token_accuracy": 0.2760023221373558,
      "num_tokens": 198042887.0,
      "step": 86365
    },
    {
      "entropy": 5.563906621932984,
      "epoch": 8.296829971181555,
      "grad_norm": 1.3515625,
      "learning_rate": 8.201954542995858e-05,
      "loss": 4.6699,
      "mean_token_accuracy": 0.26527138203382494,
      "num_tokens": 198054351.0,
      "step": 86370
    },
    {
      "entropy": 5.663556289672852,
      "epoch": 8.29731027857829,
      "grad_norm": 1.5859375,
      "learning_rate": 8.200192024403027e-05,
      "loss": 4.7249,
      "mean_token_accuracy": 0.25500150173902514,
      "num_tokens": 198066254.0,
      "step": 86375
    },
    {
      "entropy": 5.5532910346984865,
      "epoch": 8.297790585975024,
      "grad_norm": 1.5703125,
      "learning_rate": 8.198429953894434e-05,
      "loss": 4.6628,
      "mean_token_accuracy": 0.2640195444226265,
      "num_tokens": 198078883.0,
      "step": 86380
    },
    {
      "entropy": 5.630030012130737,
      "epoch": 8.298270893371757,
      "grad_norm": 1.3359375,
      "learning_rate": 8.196668331510978e-05,
      "loss": 4.7597,
      "mean_token_accuracy": 0.2583616837859154,
      "num_tokens": 198090955.0,
      "step": 86385
    },
    {
      "entropy": 5.530924558639526,
      "epoch": 8.298751200768493,
      "grad_norm": 1.4296875,
      "learning_rate": 8.194907157293566e-05,
      "loss": 4.5932,
      "mean_token_accuracy": 0.27244241535663605,
      "num_tokens": 198102229.0,
      "step": 86390
    },
    {
      "entropy": 5.540205192565918,
      "epoch": 8.299231508165226,
      "grad_norm": 1.5,
      "learning_rate": 8.193146431283084e-05,
      "loss": 4.5819,
      "mean_token_accuracy": 0.273526468873024,
      "num_tokens": 198113435.0,
      "step": 86395
    },
    {
      "entropy": 5.485333728790283,
      "epoch": 8.29971181556196,
      "grad_norm": 1.375,
      "learning_rate": 8.191386153520411e-05,
      "loss": 4.5687,
      "mean_token_accuracy": 0.272857078909874,
      "num_tokens": 198125493.0,
      "step": 86400
    },
    {
      "entropy": 5.591176128387451,
      "epoch": 8.300192122958693,
      "grad_norm": 1.5078125,
      "learning_rate": 8.18962632404642e-05,
      "loss": 4.6833,
      "mean_token_accuracy": 0.26869805753231046,
      "num_tokens": 198136046.0,
      "step": 86405
    },
    {
      "entropy": 5.566222620010376,
      "epoch": 8.300672430355428,
      "grad_norm": 1.4140625,
      "learning_rate": 8.187866942901967e-05,
      "loss": 4.6215,
      "mean_token_accuracy": 0.2651155084371567,
      "num_tokens": 198149708.0,
      "step": 86410
    },
    {
      "entropy": 5.6028800964355465,
      "epoch": 8.301152737752162,
      "grad_norm": 1.328125,
      "learning_rate": 8.186108010127898e-05,
      "loss": 4.7049,
      "mean_token_accuracy": 0.26075070053339006,
      "num_tokens": 198161052.0,
      "step": 86415
    },
    {
      "entropy": 5.4970540523529055,
      "epoch": 8.301633045148895,
      "grad_norm": 1.5078125,
      "learning_rate": 8.184349525765048e-05,
      "loss": 4.6146,
      "mean_token_accuracy": 0.27225276827812195,
      "num_tokens": 198172491.0,
      "step": 86420
    },
    {
      "entropy": 5.540571832656861,
      "epoch": 8.302113352545629,
      "grad_norm": 1.3515625,
      "learning_rate": 8.182591489854246e-05,
      "loss": 4.6194,
      "mean_token_accuracy": 0.2687088370323181,
      "num_tokens": 198184257.0,
      "step": 86425
    },
    {
      "entropy": 5.529975557327271,
      "epoch": 8.302593659942364,
      "grad_norm": 1.53125,
      "learning_rate": 8.180833902436308e-05,
      "loss": 4.6118,
      "mean_token_accuracy": 0.2675145998597145,
      "num_tokens": 198194974.0,
      "step": 86430
    },
    {
      "entropy": 5.6226318836212155,
      "epoch": 8.303073967339097,
      "grad_norm": 1.5,
      "learning_rate": 8.17907676355204e-05,
      "loss": 4.7066,
      "mean_token_accuracy": 0.2606735467910767,
      "num_tokens": 198207100.0,
      "step": 86435
    },
    {
      "entropy": 5.484433126449585,
      "epoch": 8.30355427473583,
      "grad_norm": 1.390625,
      "learning_rate": 8.17732007324224e-05,
      "loss": 4.5676,
      "mean_token_accuracy": 0.27008219957351687,
      "num_tokens": 198219963.0,
      "step": 86440
    },
    {
      "entropy": 5.503993511199951,
      "epoch": 8.304034582132564,
      "grad_norm": 1.484375,
      "learning_rate": 8.175563831547696e-05,
      "loss": 4.5964,
      "mean_token_accuracy": 0.27056487202644347,
      "num_tokens": 198231972.0,
      "step": 86445
    },
    {
      "entropy": 5.495936584472656,
      "epoch": 8.3045148895293,
      "grad_norm": 1.34375,
      "learning_rate": 8.173808038509175e-05,
      "loss": 4.5955,
      "mean_token_accuracy": 0.27717320472002027,
      "num_tokens": 198245031.0,
      "step": 86450
    },
    {
      "entropy": 5.552648448944092,
      "epoch": 8.304995196926033,
      "grad_norm": 1.40625,
      "learning_rate": 8.172052694167443e-05,
      "loss": 4.5908,
      "mean_token_accuracy": 0.2653530344367027,
      "num_tokens": 198256648.0,
      "step": 86455
    },
    {
      "entropy": 5.59535961151123,
      "epoch": 8.305475504322766,
      "grad_norm": 1.453125,
      "learning_rate": 8.170297798563253e-05,
      "loss": 4.7427,
      "mean_token_accuracy": 0.2591294005513191,
      "num_tokens": 198268160.0,
      "step": 86460
    },
    {
      "entropy": 5.508961486816406,
      "epoch": 8.3059558117195,
      "grad_norm": 1.4921875,
      "learning_rate": 8.168543351737352e-05,
      "loss": 4.6254,
      "mean_token_accuracy": 0.26478690207004546,
      "num_tokens": 198279314.0,
      "step": 86465
    },
    {
      "entropy": 5.505003309249878,
      "epoch": 8.306436119116235,
      "grad_norm": 1.46875,
      "learning_rate": 8.166789353730472e-05,
      "loss": 4.5991,
      "mean_token_accuracy": 0.2770008623600006,
      "num_tokens": 198290250.0,
      "step": 86470
    },
    {
      "entropy": 5.54933762550354,
      "epoch": 8.306916426512968,
      "grad_norm": 1.453125,
      "learning_rate": 8.165035804583337e-05,
      "loss": 4.6395,
      "mean_token_accuracy": 0.26613605618476865,
      "num_tokens": 198303251.0,
      "step": 86475
    },
    {
      "entropy": 5.557490825653076,
      "epoch": 8.307396733909702,
      "grad_norm": 1.703125,
      "learning_rate": 8.163282704336656e-05,
      "loss": 4.6621,
      "mean_token_accuracy": 0.2637137666344643,
      "num_tokens": 198312757.0,
      "step": 86480
    },
    {
      "entropy": 5.541711759567261,
      "epoch": 8.307877041306437,
      "grad_norm": 1.453125,
      "learning_rate": 8.161530053031138e-05,
      "loss": 4.5686,
      "mean_token_accuracy": 0.28114054501056673,
      "num_tokens": 198323944.0,
      "step": 86485
    },
    {
      "entropy": 5.531555652618408,
      "epoch": 8.30835734870317,
      "grad_norm": 1.4765625,
      "learning_rate": 8.159777850707461e-05,
      "loss": 4.6315,
      "mean_token_accuracy": 0.27046829760074614,
      "num_tokens": 198334739.0,
      "step": 86490
    },
    {
      "entropy": 5.5270942687988285,
      "epoch": 8.308837656099904,
      "grad_norm": 1.4140625,
      "learning_rate": 8.158026097406316e-05,
      "loss": 4.5955,
      "mean_token_accuracy": 0.2666316732764244,
      "num_tokens": 198346044.0,
      "step": 86495
    },
    {
      "entropy": 5.467145299911499,
      "epoch": 8.309317963496637,
      "grad_norm": 1.3828125,
      "learning_rate": 8.156274793168373e-05,
      "loss": 4.5469,
      "mean_token_accuracy": 0.2725500136613846,
      "num_tokens": 198357513.0,
      "step": 86500
    },
    {
      "entropy": 5.585055541992188,
      "epoch": 8.309798270893372,
      "grad_norm": 1.4140625,
      "learning_rate": 8.154523938034288e-05,
      "loss": 4.6486,
      "mean_token_accuracy": 0.2668084755539894,
      "num_tokens": 198368385.0,
      "step": 86505
    },
    {
      "entropy": 5.477368307113648,
      "epoch": 8.310278578290106,
      "grad_norm": 1.484375,
      "learning_rate": 8.152773532044712e-05,
      "loss": 4.4741,
      "mean_token_accuracy": 0.2789320319890976,
      "num_tokens": 198380126.0,
      "step": 86510
    },
    {
      "entropy": 5.522501993179321,
      "epoch": 8.31075888568684,
      "grad_norm": 1.375,
      "learning_rate": 8.151023575240288e-05,
      "loss": 4.6149,
      "mean_token_accuracy": 0.26724991351366045,
      "num_tokens": 198391837.0,
      "step": 86515
    },
    {
      "entropy": 5.540769004821778,
      "epoch": 8.311239193083573,
      "grad_norm": 1.484375,
      "learning_rate": 8.149274067661637e-05,
      "loss": 4.6394,
      "mean_token_accuracy": 0.2686233133077621,
      "num_tokens": 198404057.0,
      "step": 86520
    },
    {
      "entropy": 5.566540336608886,
      "epoch": 8.311719500480308,
      "grad_norm": 1.578125,
      "learning_rate": 8.14752500934939e-05,
      "loss": 4.6195,
      "mean_token_accuracy": 0.2648524448275566,
      "num_tokens": 198415465.0,
      "step": 86525
    },
    {
      "entropy": 5.464203023910523,
      "epoch": 8.312199807877041,
      "grad_norm": 1.4296875,
      "learning_rate": 8.145776400344143e-05,
      "loss": 4.5308,
      "mean_token_accuracy": 0.27256359457969664,
      "num_tokens": 198426950.0,
      "step": 86530
    },
    {
      "entropy": 5.550205993652344,
      "epoch": 8.312680115273775,
      "grad_norm": 1.3828125,
      "learning_rate": 8.144028240686499e-05,
      "loss": 4.6398,
      "mean_token_accuracy": 0.26990248411893847,
      "num_tokens": 198438330.0,
      "step": 86535
    },
    {
      "entropy": 5.522909355163574,
      "epoch": 8.31316042267051,
      "grad_norm": 1.53125,
      "learning_rate": 8.142280530417043e-05,
      "loss": 4.6133,
      "mean_token_accuracy": 0.2706077963113785,
      "num_tokens": 198449498.0,
      "step": 86540
    },
    {
      "entropy": 5.511595296859741,
      "epoch": 8.313640730067243,
      "grad_norm": 1.4296875,
      "learning_rate": 8.14053326957635e-05,
      "loss": 4.5423,
      "mean_token_accuracy": 0.2695436954498291,
      "num_tokens": 198459499.0,
      "step": 86545
    },
    {
      "entropy": 5.463170528411865,
      "epoch": 8.314121037463977,
      "grad_norm": 1.4296875,
      "learning_rate": 8.138786458204991e-05,
      "loss": 4.5778,
      "mean_token_accuracy": 0.2682070031762123,
      "num_tokens": 198471192.0,
      "step": 86550
    },
    {
      "entropy": 5.493385744094849,
      "epoch": 8.31460134486071,
      "grad_norm": 1.4609375,
      "learning_rate": 8.137040096343517e-05,
      "loss": 4.516,
      "mean_token_accuracy": 0.2779823035001755,
      "num_tokens": 198482511.0,
      "step": 86555
    },
    {
      "entropy": 5.523624038696289,
      "epoch": 8.315081652257446,
      "grad_norm": 1.59375,
      "learning_rate": 8.135294184032485e-05,
      "loss": 4.5799,
      "mean_token_accuracy": 0.2685663789510727,
      "num_tokens": 198493938.0,
      "step": 86560
    },
    {
      "entropy": 5.499951791763306,
      "epoch": 8.315561959654179,
      "grad_norm": 1.609375,
      "learning_rate": 8.133548721312411e-05,
      "loss": 4.5974,
      "mean_token_accuracy": 0.27000306695699694,
      "num_tokens": 198504314.0,
      "step": 86565
    },
    {
      "entropy": 5.506947755813599,
      "epoch": 8.316042267050912,
      "grad_norm": 1.4140625,
      "learning_rate": 8.131803708223835e-05,
      "loss": 4.5719,
      "mean_token_accuracy": 0.2720347836613655,
      "num_tokens": 198515373.0,
      "step": 86570
    },
    {
      "entropy": 5.486040210723877,
      "epoch": 8.316522574447646,
      "grad_norm": 1.5390625,
      "learning_rate": 8.130059144807262e-05,
      "loss": 4.564,
      "mean_token_accuracy": 0.2730781689286232,
      "num_tokens": 198525910.0,
      "step": 86575
    },
    {
      "entropy": 5.5887542247772215,
      "epoch": 8.317002881844381,
      "grad_norm": 1.3984375,
      "learning_rate": 8.128315031103195e-05,
      "loss": 4.7064,
      "mean_token_accuracy": 0.2558890491724014,
      "num_tokens": 198536866.0,
      "step": 86580
    },
    {
      "entropy": 5.530715465545654,
      "epoch": 8.317483189241115,
      "grad_norm": 1.4765625,
      "learning_rate": 8.126571367152136e-05,
      "loss": 4.5572,
      "mean_token_accuracy": 0.2722329646348953,
      "num_tokens": 198546862.0,
      "step": 86585
    },
    {
      "entropy": 5.563357400894165,
      "epoch": 8.317963496637848,
      "grad_norm": 1.5078125,
      "learning_rate": 8.124828152994567e-05,
      "loss": 4.6275,
      "mean_token_accuracy": 0.26492873579263687,
      "num_tokens": 198557826.0,
      "step": 86590
    },
    {
      "entropy": 5.608719396591186,
      "epoch": 8.318443804034581,
      "grad_norm": 1.578125,
      "learning_rate": 8.123085388670947e-05,
      "loss": 4.7562,
      "mean_token_accuracy": 0.26285273432731626,
      "num_tokens": 198569857.0,
      "step": 86595
    },
    {
      "entropy": 5.48262882232666,
      "epoch": 8.318924111431317,
      "grad_norm": 1.6328125,
      "learning_rate": 8.121343074221751e-05,
      "loss": 4.5695,
      "mean_token_accuracy": 0.27624782621860505,
      "num_tokens": 198581440.0,
      "step": 86600
    },
    {
      "entropy": 5.557262754440307,
      "epoch": 8.31940441882805,
      "grad_norm": 1.3515625,
      "learning_rate": 8.119601209687424e-05,
      "loss": 4.5988,
      "mean_token_accuracy": 0.2791834995150566,
      "num_tokens": 198593305.0,
      "step": 86605
    },
    {
      "entropy": 5.520156478881836,
      "epoch": 8.319884726224783,
      "grad_norm": 1.4453125,
      "learning_rate": 8.117859795108409e-05,
      "loss": 4.5645,
      "mean_token_accuracy": 0.2699425593018532,
      "num_tokens": 198603553.0,
      "step": 86610
    },
    {
      "entropy": 5.491294622421265,
      "epoch": 8.320365033621517,
      "grad_norm": 1.40625,
      "learning_rate": 8.116118830525137e-05,
      "loss": 4.6157,
      "mean_token_accuracy": 0.269508421421051,
      "num_tokens": 198616007.0,
      "step": 86615
    },
    {
      "entropy": 5.51040506362915,
      "epoch": 8.320845341018252,
      "grad_norm": 1.4140625,
      "learning_rate": 8.114378315978033e-05,
      "loss": 4.5773,
      "mean_token_accuracy": 0.27373181730508805,
      "num_tokens": 198627423.0,
      "step": 86620
    },
    {
      "entropy": 5.5901689529418945,
      "epoch": 8.321325648414986,
      "grad_norm": 1.4453125,
      "learning_rate": 8.112638251507495e-05,
      "loss": 4.6792,
      "mean_token_accuracy": 0.261505500972271,
      "num_tokens": 198639543.0,
      "step": 86625
    },
    {
      "entropy": 5.477462387084961,
      "epoch": 8.321805955811719,
      "grad_norm": 1.5390625,
      "learning_rate": 8.110898637153928e-05,
      "loss": 4.5476,
      "mean_token_accuracy": 0.26921986043453217,
      "num_tokens": 198651841.0,
      "step": 86630
    },
    {
      "entropy": 5.511936235427856,
      "epoch": 8.322286263208454,
      "grad_norm": 1.5078125,
      "learning_rate": 8.10915947295772e-05,
      "loss": 4.6368,
      "mean_token_accuracy": 0.2668431490659714,
      "num_tokens": 198662386.0,
      "step": 86635
    },
    {
      "entropy": 5.589300489425659,
      "epoch": 8.322766570605188,
      "grad_norm": 1.421875,
      "learning_rate": 8.107420758959252e-05,
      "loss": 4.6522,
      "mean_token_accuracy": 0.26769642233848573,
      "num_tokens": 198673213.0,
      "step": 86640
    },
    {
      "entropy": 5.562517881393433,
      "epoch": 8.323246878001921,
      "grad_norm": 1.484375,
      "learning_rate": 8.105682495198891e-05,
      "loss": 4.6206,
      "mean_token_accuracy": 0.27113906294107437,
      "num_tokens": 198683689.0,
      "step": 86645
    },
    {
      "entropy": 5.472719717025757,
      "epoch": 8.323727185398655,
      "grad_norm": 1.3359375,
      "learning_rate": 8.103944681716997e-05,
      "loss": 4.5861,
      "mean_token_accuracy": 0.26682529151439666,
      "num_tokens": 198695763.0,
      "step": 86650
    },
    {
      "entropy": 5.517943572998047,
      "epoch": 8.32420749279539,
      "grad_norm": 1.421875,
      "learning_rate": 8.102207318553909e-05,
      "loss": 4.6308,
      "mean_token_accuracy": 0.25850946009159087,
      "num_tokens": 198706836.0,
      "step": 86655
    },
    {
      "entropy": 5.509043216705322,
      "epoch": 8.324687800192123,
      "grad_norm": 1.4609375,
      "learning_rate": 8.100470405749962e-05,
      "loss": 4.5779,
      "mean_token_accuracy": 0.2728607714176178,
      "num_tokens": 198717892.0,
      "step": 86660
    },
    {
      "entropy": 5.5687315464019775,
      "epoch": 8.325168107588857,
      "grad_norm": 1.3671875,
      "learning_rate": 8.098733943345494e-05,
      "loss": 4.6305,
      "mean_token_accuracy": 0.26487350910902024,
      "num_tokens": 198728845.0,
      "step": 86665
    },
    {
      "entropy": 5.526576948165894,
      "epoch": 8.32564841498559,
      "grad_norm": 1.5703125,
      "learning_rate": 8.09699793138081e-05,
      "loss": 4.5715,
      "mean_token_accuracy": 0.2715014308691025,
      "num_tokens": 198739586.0,
      "step": 86670
    },
    {
      "entropy": 5.551091814041138,
      "epoch": 8.326128722382325,
      "grad_norm": 1.4296875,
      "learning_rate": 8.095262369896222e-05,
      "loss": 4.6046,
      "mean_token_accuracy": 0.2699943408370018,
      "num_tokens": 198751047.0,
      "step": 86675
    },
    {
      "entropy": 5.5748913288116455,
      "epoch": 8.326609029779059,
      "grad_norm": 1.4375,
      "learning_rate": 8.093527258932025e-05,
      "loss": 4.7105,
      "mean_token_accuracy": 0.25950256586074827,
      "num_tokens": 198762394.0,
      "step": 86680
    },
    {
      "entropy": 5.5002760887146,
      "epoch": 8.327089337175792,
      "grad_norm": 1.40625,
      "learning_rate": 8.091792598528493e-05,
      "loss": 4.5637,
      "mean_token_accuracy": 0.2661926791071892,
      "num_tokens": 198772736.0,
      "step": 86685
    },
    {
      "entropy": 5.522284507751465,
      "epoch": 8.327569644572526,
      "grad_norm": 1.6328125,
      "learning_rate": 8.090058388725907e-05,
      "loss": 4.5749,
      "mean_token_accuracy": 0.27096221446990965,
      "num_tokens": 198782936.0,
      "step": 86690
    },
    {
      "entropy": 5.458114814758301,
      "epoch": 8.32804995196926,
      "grad_norm": 1.6015625,
      "learning_rate": 8.088324629564533e-05,
      "loss": 4.5158,
      "mean_token_accuracy": 0.2764135718345642,
      "num_tokens": 198794101.0,
      "step": 86695
    },
    {
      "entropy": 5.5281587600708,
      "epoch": 8.328530259365994,
      "grad_norm": 1.59375,
      "learning_rate": 8.086591321084616e-05,
      "loss": 4.597,
      "mean_token_accuracy": 0.27104474008083346,
      "num_tokens": 198804167.0,
      "step": 86700
    },
    {
      "entropy": 5.662689971923828,
      "epoch": 8.329010566762728,
      "grad_norm": 1.4140625,
      "learning_rate": 8.084858463326402e-05,
      "loss": 4.7749,
      "mean_token_accuracy": 0.25403836369514465,
      "num_tokens": 198817317.0,
      "step": 86705
    },
    {
      "entropy": 5.491024971008301,
      "epoch": 8.329490874159463,
      "grad_norm": 1.53125,
      "learning_rate": 8.083126056330134e-05,
      "loss": 4.5701,
      "mean_token_accuracy": 0.27154699712991714,
      "num_tokens": 198828296.0,
      "step": 86710
    },
    {
      "entropy": 5.483888816833496,
      "epoch": 8.329971181556196,
      "grad_norm": 1.3671875,
      "learning_rate": 8.081394100136011e-05,
      "loss": 4.5304,
      "mean_token_accuracy": 0.2714651256799698,
      "num_tokens": 198840540.0,
      "step": 86715
    },
    {
      "entropy": 5.566613674163818,
      "epoch": 8.33045148895293,
      "grad_norm": 1.4453125,
      "learning_rate": 8.07966259478426e-05,
      "loss": 4.6536,
      "mean_token_accuracy": 0.2646555408835411,
      "num_tokens": 198851397.0,
      "step": 86720
    },
    {
      "entropy": 5.499383735656738,
      "epoch": 8.330931796349663,
      "grad_norm": 1.4453125,
      "learning_rate": 8.077931540315076e-05,
      "loss": 4.5519,
      "mean_token_accuracy": 0.2762279242277145,
      "num_tokens": 198863365.0,
      "step": 86725
    },
    {
      "entropy": 5.488224363327026,
      "epoch": 8.331412103746398,
      "grad_norm": 1.5390625,
      "learning_rate": 8.076200936768649e-05,
      "loss": 4.5093,
      "mean_token_accuracy": 0.2761889636516571,
      "num_tokens": 198873895.0,
      "step": 86730
    },
    {
      "entropy": 5.517120504379273,
      "epoch": 8.331892411143132,
      "grad_norm": 1.59375,
      "learning_rate": 8.074470784185162e-05,
      "loss": 4.5812,
      "mean_token_accuracy": 0.27255482226610184,
      "num_tokens": 198884181.0,
      "step": 86735
    },
    {
      "entropy": 5.484014415740967,
      "epoch": 8.332372718539865,
      "grad_norm": 1.46875,
      "learning_rate": 8.072741082604781e-05,
      "loss": 4.5602,
      "mean_token_accuracy": 0.27150759994983675,
      "num_tokens": 198895401.0,
      "step": 86740
    },
    {
      "entropy": 5.501725101470948,
      "epoch": 8.332853025936599,
      "grad_norm": 1.3671875,
      "learning_rate": 8.071011832067667e-05,
      "loss": 4.5829,
      "mean_token_accuracy": 0.27218695133924486,
      "num_tokens": 198906555.0,
      "step": 86745
    },
    {
      "entropy": 5.530845403671265,
      "epoch": 8.333333333333334,
      "grad_norm": 1.546875,
      "learning_rate": 8.069283032613961e-05,
      "loss": 4.5506,
      "mean_token_accuracy": 0.28231600522994993,
      "num_tokens": 198917203.0,
      "step": 86750
    },
    {
      "entropy": 5.505600023269653,
      "epoch": 8.333813640730067,
      "grad_norm": 1.5625,
      "learning_rate": 8.06755468428381e-05,
      "loss": 4.6367,
      "mean_token_accuracy": 0.2648101821541786,
      "num_tokens": 198928624.0,
      "step": 86755
    },
    {
      "entropy": 5.4878502368927,
      "epoch": 8.3342939481268,
      "grad_norm": 1.4375,
      "learning_rate": 8.065826787117331e-05,
      "loss": 4.5404,
      "mean_token_accuracy": 0.2760020405054092,
      "num_tokens": 198940576.0,
      "step": 86760
    },
    {
      "entropy": 5.512754011154175,
      "epoch": 8.334774255523534,
      "grad_norm": 1.5,
      "learning_rate": 8.064099341154649e-05,
      "loss": 4.5761,
      "mean_token_accuracy": 0.28018692433834075,
      "num_tokens": 198950573.0,
      "step": 86765
    },
    {
      "entropy": 5.4940135955810545,
      "epoch": 8.33525456292027,
      "grad_norm": 1.4453125,
      "learning_rate": 8.062372346435872e-05,
      "loss": 4.5953,
      "mean_token_accuracy": 0.26694034934043886,
      "num_tokens": 198962788.0,
      "step": 86770
    },
    {
      "entropy": 5.496091604232788,
      "epoch": 8.335734870317003,
      "grad_norm": 1.578125,
      "learning_rate": 8.060645803001089e-05,
      "loss": 4.5873,
      "mean_token_accuracy": 0.2710358053445816,
      "num_tokens": 198972775.0,
      "step": 86775
    },
    {
      "entropy": 5.449498033523559,
      "epoch": 8.336215177713736,
      "grad_norm": 1.359375,
      "learning_rate": 8.058919710890386e-05,
      "loss": 4.5526,
      "mean_token_accuracy": 0.28136534094810484,
      "num_tokens": 198984251.0,
      "step": 86780
    },
    {
      "entropy": 5.513584423065185,
      "epoch": 8.336695485110472,
      "grad_norm": 1.3359375,
      "learning_rate": 8.057194070143836e-05,
      "loss": 4.6428,
      "mean_token_accuracy": 0.26789259165525436,
      "num_tokens": 198996908.0,
      "step": 86785
    },
    {
      "entropy": 5.557885551452637,
      "epoch": 8.337175792507205,
      "grad_norm": 1.3515625,
      "learning_rate": 8.055468880801512e-05,
      "loss": 4.6259,
      "mean_token_accuracy": 0.2607306256890297,
      "num_tokens": 199008792.0,
      "step": 86790
    },
    {
      "entropy": 5.527619743347168,
      "epoch": 8.337656099903938,
      "grad_norm": 1.453125,
      "learning_rate": 8.053744142903458e-05,
      "loss": 4.6254,
      "mean_token_accuracy": 0.2640097364783287,
      "num_tokens": 199021519.0,
      "step": 86795
    },
    {
      "entropy": 5.6338285446167,
      "epoch": 8.338136407300672,
      "grad_norm": 1.3984375,
      "learning_rate": 8.052019856489728e-05,
      "loss": 4.7505,
      "mean_token_accuracy": 0.2574210435152054,
      "num_tokens": 199032432.0,
      "step": 86800
    },
    {
      "entropy": 5.5988517761230465,
      "epoch": 8.338616714697407,
      "grad_norm": 1.359375,
      "learning_rate": 8.050296021600343e-05,
      "loss": 4.6637,
      "mean_token_accuracy": 0.25714494436979296,
      "num_tokens": 199044163.0,
      "step": 86805
    },
    {
      "entropy": 5.5357767105102536,
      "epoch": 8.33909702209414,
      "grad_norm": 1.53125,
      "learning_rate": 8.04857263827533e-05,
      "loss": 4.6064,
      "mean_token_accuracy": 0.26032442301511766,
      "num_tokens": 199055697.0,
      "step": 86810
    },
    {
      "entropy": 5.5966509819030765,
      "epoch": 8.339577329490874,
      "grad_norm": 1.53125,
      "learning_rate": 8.046849706554701e-05,
      "loss": 4.6472,
      "mean_token_accuracy": 0.26166423559188845,
      "num_tokens": 199066236.0,
      "step": 86815
    },
    {
      "entropy": 5.4392914295196535,
      "epoch": 8.340057636887607,
      "grad_norm": 1.40625,
      "learning_rate": 8.04512722647846e-05,
      "loss": 4.4772,
      "mean_token_accuracy": 0.2792302221059799,
      "num_tokens": 199077764.0,
      "step": 86820
    },
    {
      "entropy": 5.532966661453247,
      "epoch": 8.340537944284343,
      "grad_norm": 1.421875,
      "learning_rate": 8.043405198086595e-05,
      "loss": 4.5886,
      "mean_token_accuracy": 0.27160166800022123,
      "num_tokens": 199088089.0,
      "step": 86825
    },
    {
      "entropy": 5.4402184009552,
      "epoch": 8.341018251681076,
      "grad_norm": 1.46875,
      "learning_rate": 8.041683621419092e-05,
      "loss": 4.4844,
      "mean_token_accuracy": 0.28213316202163696,
      "num_tokens": 199099141.0,
      "step": 86830
    },
    {
      "entropy": 5.534687900543213,
      "epoch": 8.34149855907781,
      "grad_norm": 1.5234375,
      "learning_rate": 8.03996249651591e-05,
      "loss": 4.6521,
      "mean_token_accuracy": 0.2708974152803421,
      "num_tokens": 199110403.0,
      "step": 86835
    },
    {
      "entropy": 5.453223323822021,
      "epoch": 8.341978866474543,
      "grad_norm": 1.5,
      "learning_rate": 8.038241823417016e-05,
      "loss": 4.5677,
      "mean_token_accuracy": 0.2693876788020134,
      "num_tokens": 199121016.0,
      "step": 86840
    },
    {
      "entropy": 5.4611187934875485,
      "epoch": 8.342459173871278,
      "grad_norm": 1.359375,
      "learning_rate": 8.036521602162356e-05,
      "loss": 4.565,
      "mean_token_accuracy": 0.2703811749815941,
      "num_tokens": 199132098.0,
      "step": 86845
    },
    {
      "entropy": 5.466056776046753,
      "epoch": 8.342939481268012,
      "grad_norm": 1.4296875,
      "learning_rate": 8.034801832791869e-05,
      "loss": 4.5582,
      "mean_token_accuracy": 0.2728357926011086,
      "num_tokens": 199142895.0,
      "step": 86850
    },
    {
      "entropy": 5.549064064025879,
      "epoch": 8.343419788664745,
      "grad_norm": 1.453125,
      "learning_rate": 8.033082515345485e-05,
      "loss": 4.5831,
      "mean_token_accuracy": 0.27492394745349885,
      "num_tokens": 199153174.0,
      "step": 86855
    },
    {
      "entropy": 5.521081018447876,
      "epoch": 8.34390009606148,
      "grad_norm": 1.4296875,
      "learning_rate": 8.031363649863126e-05,
      "loss": 4.5595,
      "mean_token_accuracy": 0.2722556248307228,
      "num_tokens": 199165914.0,
      "step": 86860
    },
    {
      "entropy": 5.534788751602173,
      "epoch": 8.344380403458214,
      "grad_norm": 1.546875,
      "learning_rate": 8.029645236384688e-05,
      "loss": 4.6309,
      "mean_token_accuracy": 0.2680793583393097,
      "num_tokens": 199177586.0,
      "step": 86865
    },
    {
      "entropy": 5.540863275527954,
      "epoch": 8.344860710854947,
      "grad_norm": 1.5625,
      "learning_rate": 8.027927274950071e-05,
      "loss": 4.6193,
      "mean_token_accuracy": 0.2698363170027733,
      "num_tokens": 199187723.0,
      "step": 86870
    },
    {
      "entropy": 5.617053365707397,
      "epoch": 8.34534101825168,
      "grad_norm": 1.375,
      "learning_rate": 8.026209765599162e-05,
      "loss": 4.6881,
      "mean_token_accuracy": 0.2610252648591995,
      "num_tokens": 199197930.0,
      "step": 86875
    },
    {
      "entropy": 5.616630697250367,
      "epoch": 8.345821325648416,
      "grad_norm": 1.4921875,
      "learning_rate": 8.024492708371839e-05,
      "loss": 4.6564,
      "mean_token_accuracy": 0.26117105036973953,
      "num_tokens": 199209650.0,
      "step": 86880
    },
    {
      "entropy": 5.470194911956787,
      "epoch": 8.34630163304515,
      "grad_norm": 1.3046875,
      "learning_rate": 8.022776103307964e-05,
      "loss": 4.5837,
      "mean_token_accuracy": 0.28026278913021085,
      "num_tokens": 199222167.0,
      "step": 86885
    },
    {
      "entropy": 5.538981580734253,
      "epoch": 8.346781940441883,
      "grad_norm": 1.5546875,
      "learning_rate": 8.021059950447396e-05,
      "loss": 4.6236,
      "mean_token_accuracy": 0.27591414749622345,
      "num_tokens": 199233828.0,
      "step": 86890
    },
    {
      "entropy": 5.5419848442077635,
      "epoch": 8.347262247838616,
      "grad_norm": 1.7109375,
      "learning_rate": 8.019344249829974e-05,
      "loss": 4.6363,
      "mean_token_accuracy": 0.2641681969165802,
      "num_tokens": 199246536.0,
      "step": 86895
    },
    {
      "entropy": 5.4708030223846436,
      "epoch": 8.347742555235351,
      "grad_norm": 1.53125,
      "learning_rate": 8.01762900149553e-05,
      "loss": 4.5459,
      "mean_token_accuracy": 0.27293231189250944,
      "num_tokens": 199257253.0,
      "step": 86900
    },
    {
      "entropy": 5.51427812576294,
      "epoch": 8.348222862632085,
      "grad_norm": 1.4921875,
      "learning_rate": 8.01591420548389e-05,
      "loss": 4.611,
      "mean_token_accuracy": 0.2657969698309898,
      "num_tokens": 199268828.0,
      "step": 86905
    },
    {
      "entropy": 5.514757299423218,
      "epoch": 8.348703170028818,
      "grad_norm": 1.46875,
      "learning_rate": 8.014199861834866e-05,
      "loss": 4.5738,
      "mean_token_accuracy": 0.26848467588424685,
      "num_tokens": 199280297.0,
      "step": 86910
    },
    {
      "entropy": 5.566936445236206,
      "epoch": 8.349183477425552,
      "grad_norm": 1.4453125,
      "learning_rate": 8.012485970588262e-05,
      "loss": 4.6776,
      "mean_token_accuracy": 0.2605183333158493,
      "num_tokens": 199292755.0,
      "step": 86915
    },
    {
      "entropy": 5.599282693862915,
      "epoch": 8.349663784822287,
      "grad_norm": 1.3671875,
      "learning_rate": 8.010772531783872e-05,
      "loss": 4.721,
      "mean_token_accuracy": 0.2575355753302574,
      "num_tokens": 199305589.0,
      "step": 86920
    },
    {
      "entropy": 5.562983798980713,
      "epoch": 8.35014409221902,
      "grad_norm": 1.453125,
      "learning_rate": 8.009059545461468e-05,
      "loss": 4.6354,
      "mean_token_accuracy": 0.27398273050785066,
      "num_tokens": 199316761.0,
      "step": 86925
    },
    {
      "entropy": 5.53808274269104,
      "epoch": 8.350624399615754,
      "grad_norm": 1.453125,
      "learning_rate": 8.007347011660824e-05,
      "loss": 4.5964,
      "mean_token_accuracy": 0.2745092213153839,
      "num_tokens": 199327430.0,
      "step": 86930
    },
    {
      "entropy": 5.510611963272095,
      "epoch": 8.351104707012489,
      "grad_norm": 1.4296875,
      "learning_rate": 8.005634930421701e-05,
      "loss": 4.6269,
      "mean_token_accuracy": 0.2725817859172821,
      "num_tokens": 199339066.0,
      "step": 86935
    },
    {
      "entropy": 5.443560361862183,
      "epoch": 8.351585014409222,
      "grad_norm": 1.3828125,
      "learning_rate": 8.00392330178385e-05,
      "loss": 4.5561,
      "mean_token_accuracy": 0.2741565048694611,
      "num_tokens": 199351848.0,
      "step": 86940
    },
    {
      "entropy": 5.5261533737182615,
      "epoch": 8.352065321805956,
      "grad_norm": 1.484375,
      "learning_rate": 8.002212125787009e-05,
      "loss": 4.6226,
      "mean_token_accuracy": 0.2699927747249603,
      "num_tokens": 199363596.0,
      "step": 86945
    },
    {
      "entropy": 5.491322422027588,
      "epoch": 8.35254562920269,
      "grad_norm": 1.453125,
      "learning_rate": 8.000501402470909e-05,
      "loss": 4.5957,
      "mean_token_accuracy": 0.2690961122512817,
      "num_tokens": 199374800.0,
      "step": 86950
    },
    {
      "entropy": 5.552658271789551,
      "epoch": 8.353025936599424,
      "grad_norm": 1.5703125,
      "learning_rate": 7.99879113187526e-05,
      "loss": 4.6978,
      "mean_token_accuracy": 0.26489573121070864,
      "num_tokens": 199386266.0,
      "step": 86955
    },
    {
      "entropy": 5.53824028968811,
      "epoch": 8.353506243996158,
      "grad_norm": 1.40625,
      "learning_rate": 7.997081314039774e-05,
      "loss": 4.5969,
      "mean_token_accuracy": 0.27585513740777967,
      "num_tokens": 199397730.0,
      "step": 86960
    },
    {
      "entropy": 5.5063879013061525,
      "epoch": 8.353986551392891,
      "grad_norm": 1.375,
      "learning_rate": 7.995371949004152e-05,
      "loss": 4.5634,
      "mean_token_accuracy": 0.27085714042186737,
      "num_tokens": 199409182.0,
      "step": 86965
    },
    {
      "entropy": 5.575491523742675,
      "epoch": 8.354466858789625,
      "grad_norm": 1.453125,
      "learning_rate": 7.993663036808072e-05,
      "loss": 4.7098,
      "mean_token_accuracy": 0.26921204179525376,
      "num_tokens": 199419794.0,
      "step": 86970
    },
    {
      "entropy": 5.586758089065552,
      "epoch": 8.35494716618636,
      "grad_norm": 1.359375,
      "learning_rate": 7.991954577491215e-05,
      "loss": 4.6443,
      "mean_token_accuracy": 0.26834153681993483,
      "num_tokens": 199430470.0,
      "step": 86975
    },
    {
      "entropy": 5.572679281234741,
      "epoch": 8.355427473583093,
      "grad_norm": 1.4609375,
      "learning_rate": 7.990246571093253e-05,
      "loss": 4.7039,
      "mean_token_accuracy": 0.2647045090794563,
      "num_tokens": 199441047.0,
      "step": 86980
    },
    {
      "entropy": 5.507415819168091,
      "epoch": 8.355907780979827,
      "grad_norm": 1.5546875,
      "learning_rate": 7.988539017653824e-05,
      "loss": 4.5435,
      "mean_token_accuracy": 0.273683999478817,
      "num_tokens": 199452095.0,
      "step": 86985
    },
    {
      "entropy": 5.538294553756714,
      "epoch": 8.35638808837656,
      "grad_norm": 1.46875,
      "learning_rate": 7.986831917212586e-05,
      "loss": 4.5448,
      "mean_token_accuracy": 0.26979468315839766,
      "num_tokens": 199463059.0,
      "step": 86990
    },
    {
      "entropy": 5.521485805511475,
      "epoch": 8.356868395773295,
      "grad_norm": 1.3828125,
      "learning_rate": 7.985125269809168e-05,
      "loss": 4.6181,
      "mean_token_accuracy": 0.26788748502731324,
      "num_tokens": 199474669.0,
      "step": 86995
    },
    {
      "entropy": 5.524040365219117,
      "epoch": 8.357348703170029,
      "grad_norm": 1.375,
      "learning_rate": 7.983419075483194e-05,
      "loss": 4.6249,
      "mean_token_accuracy": 0.26238711923360825,
      "num_tokens": 199485892.0,
      "step": 87000
    },
    {
      "epoch": 8.357348703170029,
      "eval_entropy": 5.219507513948326,
      "eval_loss": 4.736200332641602,
      "eval_mean_token_accuracy": 0.2688292877976558,
      "eval_num_tokens": 199485892.0,
      "eval_runtime": 20.6685,
      "eval_samples_per_second": 1587.685,
      "eval_steps_per_second": 198.467,
      "step": 87000
    }
  ],
  "logging_steps": 5,
  "max_steps": 104090,
  "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": 4.519087786708992e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}