{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6157813206585011,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.1113718893378972,
      "epoch": 0.0015394533016462528,
      "grad_norm": 23.25,
      "learning_rate": 2.666666666666667e-07,
      "loss": 1.4148,
      "mean_token_accuracy": 0.6499394655227662,
      "num_tokens": 379552.0,
      "step": 5
    },
    {
      "entropy": 1.1505246445536614,
      "epoch": 0.0030789066032925055,
      "grad_norm": 25.375,
      "learning_rate": 6.000000000000001e-07,
      "loss": 1.4742,
      "mean_token_accuracy": 0.6414433296769857,
      "num_tokens": 743248.0,
      "step": 10
    },
    {
      "entropy": 1.1535595946013928,
      "epoch": 0.004618359904938758,
      "grad_norm": 24.0,
      "learning_rate": 9.333333333333334e-07,
      "loss": 1.4559,
      "mean_token_accuracy": 0.6436704084277153,
      "num_tokens": 1124191.0,
      "step": 15
    },
    {
      "entropy": 1.1471380971372127,
      "epoch": 0.006157813206585011,
      "grad_norm": 21.5,
      "learning_rate": 1.2666666666666669e-06,
      "loss": 1.4609,
      "mean_token_accuracy": 0.642368745803833,
      "num_tokens": 1488145.0,
      "step": 20
    },
    {
      "entropy": 1.155402697250247,
      "epoch": 0.007697266508231265,
      "grad_norm": 18.25,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 1.4329,
      "mean_token_accuracy": 0.6484112951904535,
      "num_tokens": 1860647.0,
      "step": 25
    },
    {
      "entropy": 1.1452072311192751,
      "epoch": 0.009236719809877517,
      "grad_norm": 17.25,
      "learning_rate": 1.9333333333333336e-06,
      "loss": 1.3712,
      "mean_token_accuracy": 0.6529930120334029,
      "num_tokens": 2245973.0,
      "step": 30
    },
    {
      "entropy": 1.123310711979866,
      "epoch": 0.01077617311152377,
      "grad_norm": 13.8125,
      "learning_rate": 2.266666666666667e-06,
      "loss": 1.3082,
      "mean_token_accuracy": 0.6658382505178452,
      "num_tokens": 2631053.0,
      "step": 35
    },
    {
      "entropy": 1.1543800953775645,
      "epoch": 0.012315626413170022,
      "grad_norm": 8.75,
      "learning_rate": 2.6e-06,
      "loss": 1.2982,
      "mean_token_accuracy": 0.6618854686617851,
      "num_tokens": 2992829.0,
      "step": 40
    },
    {
      "entropy": 1.1680023059248925,
      "epoch": 0.013855079714816277,
      "grad_norm": 6.1875,
      "learning_rate": 2.9333333333333338e-06,
      "loss": 1.2948,
      "mean_token_accuracy": 0.6657806046307087,
      "num_tokens": 3382112.0,
      "step": 45
    },
    {
      "entropy": 1.147538623958826,
      "epoch": 0.01539453301646253,
      "grad_norm": 5.375,
      "learning_rate": 3.266666666666667e-06,
      "loss": 1.2509,
      "mean_token_accuracy": 0.675503570586443,
      "num_tokens": 3764785.0,
      "step": 50
    },
    {
      "entropy": 1.107958966679871,
      "epoch": 0.01693398631810878,
      "grad_norm": 4.90625,
      "learning_rate": 3.6000000000000003e-06,
      "loss": 1.1997,
      "mean_token_accuracy": 0.6811204344034195,
      "num_tokens": 4126394.0,
      "step": 55
    },
    {
      "entropy": 1.1157011151313783,
      "epoch": 0.018473439619755033,
      "grad_norm": 2.265625,
      "learning_rate": 3.9333333333333335e-06,
      "loss": 1.193,
      "mean_token_accuracy": 0.6808499224483967,
      "num_tokens": 4510149.0,
      "step": 60
    },
    {
      "entropy": 1.1392802488058806,
      "epoch": 0.020012892921401286,
      "grad_norm": 1.1953125,
      "learning_rate": 4.266666666666668e-06,
      "loss": 1.1903,
      "mean_token_accuracy": 0.6806150872260333,
      "num_tokens": 4875984.0,
      "step": 65
    },
    {
      "entropy": 1.1204315345734357,
      "epoch": 0.02155234622304754,
      "grad_norm": 1.0078125,
      "learning_rate": 4.600000000000001e-06,
      "loss": 1.1621,
      "mean_token_accuracy": 0.6908985383808612,
      "num_tokens": 5251973.0,
      "step": 70
    },
    {
      "entropy": 1.1343737579882145,
      "epoch": 0.02309179952469379,
      "grad_norm": 0.90234375,
      "learning_rate": 4.933333333333334e-06,
      "loss": 1.1367,
      "mean_token_accuracy": 0.6892836567014455,
      "num_tokens": 5622302.0,
      "step": 75
    },
    {
      "entropy": 1.138794292882085,
      "epoch": 0.024631252826340044,
      "grad_norm": 0.81640625,
      "learning_rate": 5.2666666666666665e-06,
      "loss": 1.1657,
      "mean_token_accuracy": 0.6916824176907539,
      "num_tokens": 5987467.0,
      "step": 80
    },
    {
      "entropy": 1.1140711281448603,
      "epoch": 0.0261707061279863,
      "grad_norm": 0.796875,
      "learning_rate": 5.600000000000001e-06,
      "loss": 1.1287,
      "mean_token_accuracy": 0.6946257088333369,
      "num_tokens": 6369083.0,
      "step": 85
    },
    {
      "entropy": 1.1451053522527217,
      "epoch": 0.027710159429632553,
      "grad_norm": 0.78515625,
      "learning_rate": 5.933333333333335e-06,
      "loss": 1.1546,
      "mean_token_accuracy": 0.6903918959200382,
      "num_tokens": 6749350.0,
      "step": 90
    },
    {
      "entropy": 1.1201870743185283,
      "epoch": 0.029249612731278806,
      "grad_norm": 0.7890625,
      "learning_rate": 6.266666666666668e-06,
      "loss": 1.1166,
      "mean_token_accuracy": 0.6962686065584421,
      "num_tokens": 7118012.0,
      "step": 95
    },
    {
      "entropy": 1.149271610751748,
      "epoch": 0.03078906603292506,
      "grad_norm": 0.796875,
      "learning_rate": 6.600000000000001e-06,
      "loss": 1.1534,
      "mean_token_accuracy": 0.6903159897774458,
      "num_tokens": 7486407.0,
      "step": 100
    },
    {
      "entropy": 1.1452004179358481,
      "epoch": 0.03232851933457131,
      "grad_norm": 0.74609375,
      "learning_rate": 6.9333333333333344e-06,
      "loss": 1.1658,
      "mean_token_accuracy": 0.6880020551383496,
      "num_tokens": 7853929.0,
      "step": 105
    },
    {
      "entropy": 1.133518392033875,
      "epoch": 0.03386797263621756,
      "grad_norm": 0.71484375,
      "learning_rate": 7.266666666666668e-06,
      "loss": 1.1477,
      "mean_token_accuracy": 0.691348098218441,
      "num_tokens": 8224258.0,
      "step": 110
    },
    {
      "entropy": 1.1419388929381966,
      "epoch": 0.035407425937863814,
      "grad_norm": 0.7734375,
      "learning_rate": 7.600000000000001e-06,
      "loss": 1.1499,
      "mean_token_accuracy": 0.6935681894421577,
      "num_tokens": 8601374.0,
      "step": 115
    },
    {
      "entropy": 1.1601579703390599,
      "epoch": 0.036946879239510066,
      "grad_norm": 0.7265625,
      "learning_rate": 7.933333333333334e-06,
      "loss": 1.1652,
      "mean_token_accuracy": 0.6893212374299764,
      "num_tokens": 8976860.0,
      "step": 120
    },
    {
      "entropy": 1.1541021421551705,
      "epoch": 0.03848633254115632,
      "grad_norm": 0.75,
      "learning_rate": 8.266666666666667e-06,
      "loss": 1.1553,
      "mean_token_accuracy": 0.6880058489739895,
      "num_tokens": 9342172.0,
      "step": 125
    },
    {
      "entropy": 1.1215011529624461,
      "epoch": 0.04002578584280257,
      "grad_norm": 0.73828125,
      "learning_rate": 8.6e-06,
      "loss": 1.11,
      "mean_token_accuracy": 0.6971393920481205,
      "num_tokens": 9711234.0,
      "step": 130
    },
    {
      "entropy": 1.1288955546915531,
      "epoch": 0.041565239144448825,
      "grad_norm": 0.69921875,
      "learning_rate": 8.933333333333333e-06,
      "loss": 1.1331,
      "mean_token_accuracy": 0.69485286436975,
      "num_tokens": 10092041.0,
      "step": 135
    },
    {
      "entropy": 1.1383998703211546,
      "epoch": 0.04310469244609508,
      "grad_norm": 0.71484375,
      "learning_rate": 9.266666666666667e-06,
      "loss": 1.1608,
      "mean_token_accuracy": 0.6917846284806728,
      "num_tokens": 10470367.0,
      "step": 140
    },
    {
      "entropy": 1.1720476493239402,
      "epoch": 0.04464414574774133,
      "grad_norm": 0.7890625,
      "learning_rate": 9.600000000000001e-06,
      "loss": 1.1643,
      "mean_token_accuracy": 0.6900354858487845,
      "num_tokens": 10840227.0,
      "step": 145
    },
    {
      "entropy": 1.1642087884247303,
      "epoch": 0.04618359904938758,
      "grad_norm": 0.765625,
      "learning_rate": 9.933333333333334e-06,
      "loss": 1.1662,
      "mean_token_accuracy": 0.6894872546195984,
      "num_tokens": 11212704.0,
      "step": 150
    },
    {
      "entropy": 1.1504485111683607,
      "epoch": 0.047723052351033836,
      "grad_norm": 0.734375,
      "learning_rate": 1.0266666666666668e-05,
      "loss": 1.1618,
      "mean_token_accuracy": 0.6902470916509629,
      "num_tokens": 11585050.0,
      "step": 155
    },
    {
      "entropy": 1.139546208642423,
      "epoch": 0.04926250565268009,
      "grad_norm": 0.71484375,
      "learning_rate": 1.0600000000000002e-05,
      "loss": 1.1371,
      "mean_token_accuracy": 0.6940638959407807,
      "num_tokens": 11955470.0,
      "step": 160
    },
    {
      "entropy": 1.1310970352962613,
      "epoch": 0.05080195895432634,
      "grad_norm": 0.73828125,
      "learning_rate": 1.0933333333333334e-05,
      "loss": 1.1217,
      "mean_token_accuracy": 0.697192519530654,
      "num_tokens": 12326414.0,
      "step": 165
    },
    {
      "entropy": 1.1263095822185278,
      "epoch": 0.0523414122559726,
      "grad_norm": 0.7109375,
      "learning_rate": 1.1266666666666668e-05,
      "loss": 1.1226,
      "mean_token_accuracy": 0.695407111570239,
      "num_tokens": 12699566.0,
      "step": 170
    },
    {
      "entropy": 1.1608664955943824,
      "epoch": 0.053880865557618854,
      "grad_norm": 0.6953125,
      "learning_rate": 1.16e-05,
      "loss": 1.156,
      "mean_token_accuracy": 0.6909525521099568,
      "num_tokens": 13076955.0,
      "step": 175
    },
    {
      "entropy": 1.1388528019189834,
      "epoch": 0.05542031885926511,
      "grad_norm": 0.7265625,
      "learning_rate": 1.1933333333333335e-05,
      "loss": 1.1475,
      "mean_token_accuracy": 0.6935607939958572,
      "num_tokens": 13457841.0,
      "step": 180
    },
    {
      "entropy": 1.1540386551991104,
      "epoch": 0.05695977216091136,
      "grad_norm": 0.71484375,
      "learning_rate": 1.2266666666666667e-05,
      "loss": 1.1501,
      "mean_token_accuracy": 0.690580427646637,
      "num_tokens": 13823331.0,
      "step": 185
    },
    {
      "entropy": 1.1634995128959418,
      "epoch": 0.05849922546255761,
      "grad_norm": 0.74609375,
      "learning_rate": 1.2600000000000001e-05,
      "loss": 1.1689,
      "mean_token_accuracy": 0.6883407317101955,
      "num_tokens": 14189829.0,
      "step": 190
    },
    {
      "entropy": 1.1286564406007529,
      "epoch": 0.060038678764203865,
      "grad_norm": 0.68359375,
      "learning_rate": 1.2933333333333334e-05,
      "loss": 1.1091,
      "mean_token_accuracy": 0.6968111153692007,
      "num_tokens": 14568687.0,
      "step": 195
    },
    {
      "entropy": 1.0947596225887537,
      "epoch": 0.06157813206585012,
      "grad_norm": 0.73828125,
      "learning_rate": 1.3266666666666668e-05,
      "loss": 1.0865,
      "mean_token_accuracy": 0.7018399801105261,
      "num_tokens": 14956883.0,
      "step": 200
    },
    {
      "entropy": 1.105372793599963,
      "epoch": 0.06311758536749637,
      "grad_norm": 0.76171875,
      "learning_rate": 1.3600000000000002e-05,
      "loss": 1.098,
      "mean_token_accuracy": 0.7011617802083492,
      "num_tokens": 15329120.0,
      "step": 205
    },
    {
      "entropy": 1.1237883031368257,
      "epoch": 0.06465703866914262,
      "grad_norm": 0.73828125,
      "learning_rate": 1.3933333333333334e-05,
      "loss": 1.1159,
      "mean_token_accuracy": 0.6975264508277178,
      "num_tokens": 15706521.0,
      "step": 210
    },
    {
      "entropy": 1.1330916548147798,
      "epoch": 0.06619649197078888,
      "grad_norm": 0.76953125,
      "learning_rate": 1.4266666666666668e-05,
      "loss": 1.1323,
      "mean_token_accuracy": 0.6972741197794676,
      "num_tokens": 16066978.0,
      "step": 215
    },
    {
      "entropy": 1.1772997351363301,
      "epoch": 0.06773594527243512,
      "grad_norm": 0.7421875,
      "learning_rate": 1.46e-05,
      "loss": 1.1734,
      "mean_token_accuracy": 0.6860150098800659,
      "num_tokens": 16449339.0,
      "step": 220
    },
    {
      "entropy": 1.158057621307671,
      "epoch": 0.06927539857408138,
      "grad_norm": 0.73828125,
      "learning_rate": 1.4933333333333335e-05,
      "loss": 1.1629,
      "mean_token_accuracy": 0.6880886014550924,
      "num_tokens": 16831396.0,
      "step": 225
    },
    {
      "entropy": 1.139371989108622,
      "epoch": 0.07081485187572763,
      "grad_norm": 0.75390625,
      "learning_rate": 1.5266666666666667e-05,
      "loss": 1.1223,
      "mean_token_accuracy": 0.6936335910111666,
      "num_tokens": 17210779.0,
      "step": 230
    },
    {
      "entropy": 1.1907519888132811,
      "epoch": 0.07235430517737389,
      "grad_norm": 0.78515625,
      "learning_rate": 1.5600000000000003e-05,
      "loss": 1.187,
      "mean_token_accuracy": 0.6834761939942837,
      "num_tokens": 17582881.0,
      "step": 235
    },
    {
      "entropy": 1.1354382883757352,
      "epoch": 0.07389375847902013,
      "grad_norm": 0.765625,
      "learning_rate": 1.5933333333333336e-05,
      "loss": 1.1418,
      "mean_token_accuracy": 0.6936424177139997,
      "num_tokens": 17952831.0,
      "step": 240
    },
    {
      "entropy": 1.131275094114244,
      "epoch": 0.07543321178066639,
      "grad_norm": 0.68359375,
      "learning_rate": 1.6266666666666668e-05,
      "loss": 1.1324,
      "mean_token_accuracy": 0.6951741587370635,
      "num_tokens": 18341435.0,
      "step": 245
    },
    {
      "entropy": 1.1359387850388885,
      "epoch": 0.07697266508231264,
      "grad_norm": 0.7109375,
      "learning_rate": 1.66e-05,
      "loss": 1.1273,
      "mean_token_accuracy": 0.6938830468803644,
      "num_tokens": 18710959.0,
      "step": 250
    },
    {
      "entropy": 1.1200769424438477,
      "epoch": 0.0785121183839589,
      "grad_norm": 0.75390625,
      "learning_rate": 1.6933333333333336e-05,
      "loss": 1.1116,
      "mean_token_accuracy": 0.698444551974535,
      "num_tokens": 19077026.0,
      "step": 255
    },
    {
      "entropy": 1.144037862494588,
      "epoch": 0.08005157168560514,
      "grad_norm": 0.6953125,
      "learning_rate": 1.726666666666667e-05,
      "loss": 1.1399,
      "mean_token_accuracy": 0.694709181971848,
      "num_tokens": 19433930.0,
      "step": 260
    },
    {
      "entropy": 1.1314732745289802,
      "epoch": 0.0815910249872514,
      "grad_norm": 0.7421875,
      "learning_rate": 1.76e-05,
      "loss": 1.1265,
      "mean_token_accuracy": 0.6972466479986906,
      "num_tokens": 19801472.0,
      "step": 265
    },
    {
      "entropy": 1.1006763726472855,
      "epoch": 0.08313047828889765,
      "grad_norm": 0.734375,
      "learning_rate": 1.7933333333333333e-05,
      "loss": 1.0907,
      "mean_token_accuracy": 0.7027702957391739,
      "num_tokens": 20178765.0,
      "step": 270
    },
    {
      "entropy": 1.1250910840928554,
      "epoch": 0.08466993159054391,
      "grad_norm": 0.6640625,
      "learning_rate": 1.826666666666667e-05,
      "loss": 1.1259,
      "mean_token_accuracy": 0.6969286557286978,
      "num_tokens": 20540840.0,
      "step": 275
    },
    {
      "entropy": 1.1310079213231803,
      "epoch": 0.08620938489219016,
      "grad_norm": 0.734375,
      "learning_rate": 1.86e-05,
      "loss": 1.1182,
      "mean_token_accuracy": 0.6943910989910365,
      "num_tokens": 20917703.0,
      "step": 280
    },
    {
      "entropy": 1.1366374537348747,
      "epoch": 0.08774883819383641,
      "grad_norm": 0.76953125,
      "learning_rate": 1.8933333333333334e-05,
      "loss": 1.1538,
      "mean_token_accuracy": 0.692281323671341,
      "num_tokens": 21286716.0,
      "step": 285
    },
    {
      "entropy": 1.1471243999898433,
      "epoch": 0.08928829149548266,
      "grad_norm": 0.73828125,
      "learning_rate": 1.926666666666667e-05,
      "loss": 1.1422,
      "mean_token_accuracy": 0.6914283595979214,
      "num_tokens": 21672957.0,
      "step": 290
    },
    {
      "entropy": 1.0941655661910772,
      "epoch": 0.09082774479712892,
      "grad_norm": 0.734375,
      "learning_rate": 1.9600000000000002e-05,
      "loss": 1.0918,
      "mean_token_accuracy": 0.703205331414938,
      "num_tokens": 22057533.0,
      "step": 295
    },
    {
      "entropy": 1.1071835961192846,
      "epoch": 0.09236719809877517,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9933333333333334e-05,
      "loss": 1.1111,
      "mean_token_accuracy": 0.6998405870050192,
      "num_tokens": 22435546.0,
      "step": 300
    },
    {
      "entropy": 1.1055552622303366,
      "epoch": 0.09390665140042143,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9999991608372392e-05,
      "loss": 1.0997,
      "mean_token_accuracy": 0.7011382140219211,
      "num_tokens": 22797277.0,
      "step": 305
    },
    {
      "entropy": 1.1513552486896514,
      "epoch": 0.09544610470206767,
      "grad_norm": 0.76953125,
      "learning_rate": 1.9999957517409375e-05,
      "loss": 1.1525,
      "mean_token_accuracy": 0.6906207267194986,
      "num_tokens": 23157199.0,
      "step": 310
    },
    {
      "entropy": 1.1360066479071975,
      "epoch": 0.09698555800371393,
      "grad_norm": 0.74609375,
      "learning_rate": 1.9999897202723546e-05,
      "loss": 1.1269,
      "mean_token_accuracy": 0.6963287502527237,
      "num_tokens": 23542710.0,
      "step": 315
    },
    {
      "entropy": 1.1110562330111862,
      "epoch": 0.09852501130536018,
      "grad_norm": 0.734375,
      "learning_rate": 1.999981066447308e-05,
      "loss": 1.103,
      "mean_token_accuracy": 0.700805151462555,
      "num_tokens": 23929848.0,
      "step": 320
    },
    {
      "entropy": 1.0776942696422338,
      "epoch": 0.10006446460700644,
      "grad_norm": 0.70703125,
      "learning_rate": 1.999969790288491e-05,
      "loss": 1.0754,
      "mean_token_accuracy": 0.706406794860959,
      "num_tokens": 24320781.0,
      "step": 325
    },
    {
      "entropy": 1.1285138919949531,
      "epoch": 0.10160391790865268,
      "grad_norm": 0.69140625,
      "learning_rate": 1.9999558918254746e-05,
      "loss": 1.1159,
      "mean_token_accuracy": 0.6957792080938816,
      "num_tokens": 24718182.0,
      "step": 330
    },
    {
      "entropy": 1.1143408741801977,
      "epoch": 0.10314337121029894,
      "grad_norm": 0.71875,
      "learning_rate": 1.999939371094705e-05,
      "loss": 1.1132,
      "mean_token_accuracy": 0.6973476313054562,
      "num_tokens": 25078136.0,
      "step": 335
    },
    {
      "entropy": 1.1544642113149166,
      "epoch": 0.1046828245119452,
      "grad_norm": 0.71484375,
      "learning_rate": 1.9999202281395064e-05,
      "loss": 1.1505,
      "mean_token_accuracy": 0.6898461397737264,
      "num_tokens": 25462544.0,
      "step": 340
    },
    {
      "entropy": 1.110057471320033,
      "epoch": 0.10622227781359145,
      "grad_norm": 0.69921875,
      "learning_rate": 1.999898463010079e-05,
      "loss": 1.1093,
      "mean_token_accuracy": 0.696648533269763,
      "num_tokens": 25832507.0,
      "step": 345
    },
    {
      "entropy": 1.121138433739543,
      "epoch": 0.10776173111523771,
      "grad_norm": 0.765625,
      "learning_rate": 1.9998740757634998e-05,
      "loss": 1.1206,
      "mean_token_accuracy": 0.695626575127244,
      "num_tokens": 26208168.0,
      "step": 350
    },
    {
      "entropy": 1.1249326327815652,
      "epoch": 0.10930118441688395,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9998470664637205e-05,
      "loss": 1.1066,
      "mean_token_accuracy": 0.6976452205330134,
      "num_tokens": 26567141.0,
      "step": 355
    },
    {
      "entropy": 1.1190939696505666,
      "epoch": 0.11084063771853021,
      "grad_norm": 0.74609375,
      "learning_rate": 1.9998174351815704e-05,
      "loss": 1.109,
      "mean_token_accuracy": 0.696621885150671,
      "num_tokens": 26946409.0,
      "step": 360
    },
    {
      "entropy": 1.097430343925953,
      "epoch": 0.11238009102017646,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9997851819947537e-05,
      "loss": 1.0981,
      "mean_token_accuracy": 0.6984185025095939,
      "num_tokens": 27323310.0,
      "step": 365
    },
    {
      "entropy": 1.1255185015499591,
      "epoch": 0.11391954432182272,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9997503069878515e-05,
      "loss": 1.1021,
      "mean_token_accuracy": 0.6977558217942714,
      "num_tokens": 27728681.0,
      "step": 370
    },
    {
      "entropy": 1.1182758403941988,
      "epoch": 0.11545899762346896,
      "grad_norm": 0.7421875,
      "learning_rate": 1.9997128102523186e-05,
      "loss": 1.1144,
      "mean_token_accuracy": 0.6975483104586602,
      "num_tokens": 28099689.0,
      "step": 375
    },
    {
      "entropy": 1.0848099140450358,
      "epoch": 0.11699845092511522,
      "grad_norm": 0.69921875,
      "learning_rate": 1.9996726918864857e-05,
      "loss": 1.0807,
      "mean_token_accuracy": 0.7057063952088356,
      "num_tokens": 28470608.0,
      "step": 380
    },
    {
      "entropy": 1.147358151897788,
      "epoch": 0.11853790422676147,
      "grad_norm": 0.72265625,
      "learning_rate": 1.999629951995559e-05,
      "loss": 1.1361,
      "mean_token_accuracy": 0.6940991956740618,
      "num_tokens": 28850518.0,
      "step": 385
    },
    {
      "entropy": 1.158296991325915,
      "epoch": 0.12007735752840773,
      "grad_norm": 0.7109375,
      "learning_rate": 1.999584590691619e-05,
      "loss": 1.1613,
      "mean_token_accuracy": 0.6874197501689195,
      "num_tokens": 29223644.0,
      "step": 390
    },
    {
      "entropy": 1.1296958446502685,
      "epoch": 0.12161681083005398,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9995366080936206e-05,
      "loss": 1.1062,
      "mean_token_accuracy": 0.6964040424674749,
      "num_tokens": 29597922.0,
      "step": 395
    },
    {
      "entropy": 1.1090093098580838,
      "epoch": 0.12315626413170024,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9994860043273915e-05,
      "loss": 1.1119,
      "mean_token_accuracy": 0.6979490287601948,
      "num_tokens": 29967597.0,
      "step": 400
    },
    {
      "entropy": 1.1302901729941368,
      "epoch": 0.12469571743334648,
      "grad_norm": 0.7109375,
      "learning_rate": 1.999432779525635e-05,
      "loss": 1.113,
      "mean_token_accuracy": 0.6980737678706646,
      "num_tokens": 30351751.0,
      "step": 405
    },
    {
      "entropy": 1.0851911935955285,
      "epoch": 0.12623517073499274,
      "grad_norm": 0.7109375,
      "learning_rate": 1.999376933827927e-05,
      "loss": 1.0852,
      "mean_token_accuracy": 0.7041132722049952,
      "num_tokens": 30718524.0,
      "step": 410
    },
    {
      "entropy": 1.092500716075301,
      "epoch": 0.127774624036639,
      "grad_norm": 0.68359375,
      "learning_rate": 1.999318467380716e-05,
      "loss": 1.0757,
      "mean_token_accuracy": 0.7052585650235415,
      "num_tokens": 31099388.0,
      "step": 415
    },
    {
      "entropy": 1.1496953256428242,
      "epoch": 0.12931407733828523,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9992573803373242e-05,
      "loss": 1.1516,
      "mean_token_accuracy": 0.6907676491886378,
      "num_tokens": 31461359.0,
      "step": 420
    },
    {
      "entropy": 1.1136517949402331,
      "epoch": 0.1308535306399315,
      "grad_norm": 0.71875,
      "learning_rate": 1.9991936728579438e-05,
      "loss": 1.1098,
      "mean_token_accuracy": 0.6990172352641821,
      "num_tokens": 31837580.0,
      "step": 425
    },
    {
      "entropy": 1.1240890389308333,
      "epoch": 0.13239298394157775,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9991273451096414e-05,
      "loss": 1.1144,
      "mean_token_accuracy": 0.6938084073364734,
      "num_tokens": 32219381.0,
      "step": 430
    },
    {
      "entropy": 1.1162253782153129,
      "epoch": 0.133932437243224,
      "grad_norm": 0.73046875,
      "learning_rate": 1.9990583972663534e-05,
      "loss": 1.107,
      "mean_token_accuracy": 0.6976239930838346,
      "num_tokens": 32600761.0,
      "step": 435
    },
    {
      "entropy": 1.1294534251093864,
      "epoch": 0.13547189054487024,
      "grad_norm": 0.78515625,
      "learning_rate": 1.9989868295088876e-05,
      "loss": 1.1381,
      "mean_token_accuracy": 0.6945139471441507,
      "num_tokens": 32967708.0,
      "step": 440
    },
    {
      "entropy": 1.1337316358461975,
      "epoch": 0.1370113438465165,
      "grad_norm": 0.73828125,
      "learning_rate": 1.998912642024922e-05,
      "loss": 1.1376,
      "mean_token_accuracy": 0.6949200943112374,
      "num_tokens": 33348622.0,
      "step": 445
    },
    {
      "entropy": 1.1307091983035207,
      "epoch": 0.13855079714816276,
      "grad_norm": 0.74609375,
      "learning_rate": 1.9988358350090045e-05,
      "loss": 1.1169,
      "mean_token_accuracy": 0.6963778145611286,
      "num_tokens": 33718264.0,
      "step": 450
    },
    {
      "entropy": 1.1685293976217508,
      "epoch": 0.14009025044980902,
      "grad_norm": 0.71484375,
      "learning_rate": 1.998756408662553e-05,
      "loss": 1.1656,
      "mean_token_accuracy": 0.6891588238999248,
      "num_tokens": 34095351.0,
      "step": 455
    },
    {
      "entropy": 1.1300161950290204,
      "epoch": 0.14162970375145525,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9986743631938536e-05,
      "loss": 1.1278,
      "mean_token_accuracy": 0.6970746215432883,
      "num_tokens": 34470308.0,
      "step": 460
    },
    {
      "entropy": 1.1416201300919055,
      "epoch": 0.14316915705310151,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9985896988180607e-05,
      "loss": 1.1224,
      "mean_token_accuracy": 0.6962168168276548,
      "num_tokens": 34842571.0,
      "step": 465
    },
    {
      "entropy": 1.1438201934099197,
      "epoch": 0.14470861035474777,
      "grad_norm": 0.75,
      "learning_rate": 1.9985024157571972e-05,
      "loss": 1.146,
      "mean_token_accuracy": 0.6919929884374142,
      "num_tokens": 35212666.0,
      "step": 470
    },
    {
      "entropy": 1.1576870299875737,
      "epoch": 0.14624806365639403,
      "grad_norm": 0.69921875,
      "learning_rate": 1.998412514240152e-05,
      "loss": 1.1606,
      "mean_token_accuracy": 0.692124840989709,
      "num_tokens": 35593532.0,
      "step": 475
    },
    {
      "entropy": 1.1232300328090787,
      "epoch": 0.14778751695804027,
      "grad_norm": 0.74609375,
      "learning_rate": 1.9983199945026822e-05,
      "loss": 1.1198,
      "mean_token_accuracy": 0.6935272339731455,
      "num_tokens": 35980581.0,
      "step": 480
    },
    {
      "entropy": 1.1406380519270898,
      "epoch": 0.14932697025968653,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9982248567874098e-05,
      "loss": 1.1303,
      "mean_token_accuracy": 0.6920850377529859,
      "num_tokens": 36366818.0,
      "step": 485
    },
    {
      "entropy": 1.1526461832225323,
      "epoch": 0.15086642356133279,
      "grad_norm": 0.69921875,
      "learning_rate": 1.998127101343822e-05,
      "loss": 1.1452,
      "mean_token_accuracy": 0.6928542651236057,
      "num_tokens": 36739879.0,
      "step": 490
    },
    {
      "entropy": 1.127664201334119,
      "epoch": 0.15240587686297905,
      "grad_norm": 0.69921875,
      "learning_rate": 1.9980267284282718e-05,
      "loss": 1.1309,
      "mean_token_accuracy": 0.6948033161461353,
      "num_tokens": 37126967.0,
      "step": 495
    },
    {
      "entropy": 1.1573752466589213,
      "epoch": 0.15394533016462528,
      "grad_norm": 0.75,
      "learning_rate": 1.9979237383039745e-05,
      "loss": 1.1486,
      "mean_token_accuracy": 0.6922336863353848,
      "num_tokens": 37491480.0,
      "step": 500
    },
    {
      "entropy": 1.1195718679577111,
      "epoch": 0.15548478346627154,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9978181312410104e-05,
      "loss": 1.1097,
      "mean_token_accuracy": 0.6958762314170599,
      "num_tokens": 37855648.0,
      "step": 505
    },
    {
      "entropy": 1.1157226022332907,
      "epoch": 0.1570242367679178,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9977099075163216e-05,
      "loss": 1.1044,
      "mean_token_accuracy": 0.6988037750124931,
      "num_tokens": 38239757.0,
      "step": 510
    },
    {
      "entropy": 1.1437922086566688,
      "epoch": 0.15856369006956406,
      "grad_norm": 0.66796875,
      "learning_rate": 1.997599067413712e-05,
      "loss": 1.1402,
      "mean_token_accuracy": 0.6914523551240563,
      "num_tokens": 38626774.0,
      "step": 515
    },
    {
      "entropy": 1.1398277616128325,
      "epoch": 0.1601031433712103,
      "grad_norm": 0.75390625,
      "learning_rate": 1.997485611223847e-05,
      "loss": 1.1292,
      "mean_token_accuracy": 0.6958260778337717,
      "num_tokens": 38999361.0,
      "step": 520
    },
    {
      "entropy": 1.1146018091589212,
      "epoch": 0.16164259667285655,
      "grad_norm": 0.765625,
      "learning_rate": 1.997369539244252e-05,
      "loss": 1.1124,
      "mean_token_accuracy": 0.7005707703530788,
      "num_tokens": 39381117.0,
      "step": 525
    },
    {
      "entropy": 1.1075803402811288,
      "epoch": 0.1631820499745028,
      "grad_norm": 0.71875,
      "learning_rate": 1.9972508517793125e-05,
      "loss": 1.1011,
      "mean_token_accuracy": 0.6986994445323944,
      "num_tokens": 39748243.0,
      "step": 530
    },
    {
      "entropy": 1.1309567619115115,
      "epoch": 0.16472150327614907,
      "grad_norm": 0.796875,
      "learning_rate": 1.9971295491402725e-05,
      "loss": 1.1364,
      "mean_token_accuracy": 0.6940356496721506,
      "num_tokens": 40115293.0,
      "step": 535
    },
    {
      "entropy": 1.18396236859262,
      "epoch": 0.1662609565777953,
      "grad_norm": 0.71484375,
      "learning_rate": 1.997005631645234e-05,
      "loss": 1.1725,
      "mean_token_accuracy": 0.686182476207614,
      "num_tokens": 40508440.0,
      "step": 540
    },
    {
      "entropy": 1.1049391619861126,
      "epoch": 0.16780040987944156,
      "grad_norm": 0.71484375,
      "learning_rate": 1.996879099619156e-05,
      "loss": 1.1144,
      "mean_token_accuracy": 0.696911821886897,
      "num_tokens": 40896598.0,
      "step": 545
    },
    {
      "entropy": 1.1429337464272975,
      "epoch": 0.16933986318108782,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9967499533938544e-05,
      "loss": 1.136,
      "mean_token_accuracy": 0.6921508580446243,
      "num_tokens": 41271458.0,
      "step": 550
    },
    {
      "entropy": 1.1443469554185868,
      "epoch": 0.17087931648273408,
      "grad_norm": 0.71875,
      "learning_rate": 1.996618193308e-05,
      "loss": 1.133,
      "mean_token_accuracy": 0.6909565668553114,
      "num_tokens": 41652091.0,
      "step": 555
    },
    {
      "entropy": 1.1591787867248058,
      "epoch": 0.1724187697843803,
      "grad_norm": 0.6875,
      "learning_rate": 1.9964838197071173e-05,
      "loss": 1.1434,
      "mean_token_accuracy": 0.6915002010762692,
      "num_tokens": 42027815.0,
      "step": 560
    },
    {
      "entropy": 1.1327795442193747,
      "epoch": 0.17395822308602657,
      "grad_norm": 0.796875,
      "learning_rate": 1.9963468329435872e-05,
      "loss": 1.1364,
      "mean_token_accuracy": 0.6924004014581442,
      "num_tokens": 42390261.0,
      "step": 565
    },
    {
      "entropy": 1.1397675037384034,
      "epoch": 0.17549767638767283,
      "grad_norm": 0.734375,
      "learning_rate": 1.99620723337664e-05,
      "loss": 1.1406,
      "mean_token_accuracy": 0.6912539415061474,
      "num_tokens": 42759613.0,
      "step": 570
    },
    {
      "entropy": 1.107204508036375,
      "epoch": 0.1770371296893191,
      "grad_norm": 0.71875,
      "learning_rate": 1.9960650213723604e-05,
      "loss": 1.0852,
      "mean_token_accuracy": 0.6994996588677168,
      "num_tokens": 43132014.0,
      "step": 575
    },
    {
      "entropy": 1.099429708532989,
      "epoch": 0.17857658299096532,
      "grad_norm": 0.734375,
      "learning_rate": 1.9959201973036816e-05,
      "loss": 1.1041,
      "mean_token_accuracy": 0.7006071075797081,
      "num_tokens": 43518177.0,
      "step": 580
    },
    {
      "entropy": 1.1445161992684008,
      "epoch": 0.18011603629261158,
      "grad_norm": 0.71484375,
      "learning_rate": 1.995772761550389e-05,
      "loss": 1.1382,
      "mean_token_accuracy": 0.6925386656075716,
      "num_tokens": 43894430.0,
      "step": 585
    },
    {
      "entropy": 1.1534817535430193,
      "epoch": 0.18165548959425784,
      "grad_norm": 0.73046875,
      "learning_rate": 1.995622714499115e-05,
      "loss": 1.1493,
      "mean_token_accuracy": 0.6913147930055856,
      "num_tokens": 44258015.0,
      "step": 590
    },
    {
      "entropy": 1.1478193078190089,
      "epoch": 0.1831949428959041,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9954700565433406e-05,
      "loss": 1.1502,
      "mean_token_accuracy": 0.6935486130416393,
      "num_tokens": 44636368.0,
      "step": 595
    },
    {
      "entropy": 1.1420258667320013,
      "epoch": 0.18473439619755033,
      "grad_norm": 0.6796875,
      "learning_rate": 1.9953147880833935e-05,
      "loss": 1.1395,
      "mean_token_accuracy": 0.693219494074583,
      "num_tokens": 45009190.0,
      "step": 600
    },
    {
      "entropy": 1.1204937249422073,
      "epoch": 0.1862738494991966,
      "grad_norm": 0.76953125,
      "learning_rate": 1.9951569095264473e-05,
      "loss": 1.1237,
      "mean_token_accuracy": 0.6966589823365211,
      "num_tokens": 45369198.0,
      "step": 605
    },
    {
      "entropy": 1.1316735215485096,
      "epoch": 0.18781330280084285,
      "grad_norm": 0.7578125,
      "learning_rate": 1.99499642128652e-05,
      "loss": 1.1151,
      "mean_token_accuracy": 0.6951681729406118,
      "num_tokens": 45735650.0,
      "step": 610
    },
    {
      "entropy": 1.119157313928008,
      "epoch": 0.1893527561024891,
      "grad_norm": 0.73828125,
      "learning_rate": 1.994833323784473e-05,
      "loss": 1.1205,
      "mean_token_accuracy": 0.6974445167928934,
      "num_tokens": 46105045.0,
      "step": 615
    },
    {
      "entropy": 1.1227743715047835,
      "epoch": 0.19089220940413534,
      "grad_norm": 0.74609375,
      "learning_rate": 1.9946676174480115e-05,
      "loss": 1.1122,
      "mean_token_accuracy": 0.6958862528204918,
      "num_tokens": 46468345.0,
      "step": 620
    },
    {
      "entropy": 1.123009406030178,
      "epoch": 0.1924316627057816,
      "grad_norm": 0.71484375,
      "learning_rate": 1.9944993027116798e-05,
      "loss": 1.1064,
      "mean_token_accuracy": 0.6993745014071464,
      "num_tokens": 46839653.0,
      "step": 625
    },
    {
      "entropy": 1.13520105779171,
      "epoch": 0.19397111600742786,
      "grad_norm": 0.734375,
      "learning_rate": 1.9943283800168643e-05,
      "loss": 1.1432,
      "mean_token_accuracy": 0.69097990244627,
      "num_tokens": 47218180.0,
      "step": 630
    },
    {
      "entropy": 1.1393942264840007,
      "epoch": 0.19551056930907412,
      "grad_norm": 0.72265625,
      "learning_rate": 1.9941548498117894e-05,
      "loss": 1.1382,
      "mean_token_accuracy": 0.6938273806124926,
      "num_tokens": 47595740.0,
      "step": 635
    },
    {
      "entropy": 1.1325588449835777,
      "epoch": 0.19705002261072035,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9939787125515188e-05,
      "loss": 1.1298,
      "mean_token_accuracy": 0.6938525449484587,
      "num_tokens": 47982358.0,
      "step": 640
    },
    {
      "entropy": 1.121693492308259,
      "epoch": 0.19858947591236661,
      "grad_norm": 0.75,
      "learning_rate": 1.993799968697951e-05,
      "loss": 1.1191,
      "mean_token_accuracy": 0.6956870190799236,
      "num_tokens": 48372670.0,
      "step": 645
    },
    {
      "entropy": 1.1251621477305889,
      "epoch": 0.20012892921401287,
      "grad_norm": 0.76953125,
      "learning_rate": 1.9936186187198214e-05,
      "loss": 1.1236,
      "mean_token_accuracy": 0.6986733213067055,
      "num_tokens": 48737613.0,
      "step": 650
    },
    {
      "entropy": 1.139292366988957,
      "epoch": 0.20166838251565913,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9934346630926988e-05,
      "loss": 1.1304,
      "mean_token_accuracy": 0.6938830778002739,
      "num_tokens": 49100209.0,
      "step": 655
    },
    {
      "entropy": 1.1345834810286761,
      "epoch": 0.20320783581730537,
      "grad_norm": 0.66796875,
      "learning_rate": 1.9932481022989857e-05,
      "loss": 1.1181,
      "mean_token_accuracy": 0.6951233502477407,
      "num_tokens": 49467553.0,
      "step": 660
    },
    {
      "entropy": 1.094562499411404,
      "epoch": 0.20474728911895163,
      "grad_norm": 0.76171875,
      "learning_rate": 1.993058936827916e-05,
      "loss": 1.0977,
      "mean_token_accuracy": 0.7043172724545002,
      "num_tokens": 49849811.0,
      "step": 665
    },
    {
      "entropy": 1.139369383826852,
      "epoch": 0.20628674242059789,
      "grad_norm": 0.73828125,
      "learning_rate": 1.992867167175554e-05,
      "loss": 1.1425,
      "mean_token_accuracy": 0.6948151815682649,
      "num_tokens": 50225933.0,
      "step": 670
    },
    {
      "entropy": 1.079656113870442,
      "epoch": 0.20782619572224414,
      "grad_norm": 0.6875,
      "learning_rate": 1.9926727938447935e-05,
      "loss": 1.0692,
      "mean_token_accuracy": 0.7051354993134737,
      "num_tokens": 50610063.0,
      "step": 675
    },
    {
      "entropy": 1.1766547793522477,
      "epoch": 0.2093656490238904,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9924758173453555e-05,
      "loss": 1.1785,
      "mean_token_accuracy": 0.6878008231520653,
      "num_tokens": 50990235.0,
      "step": 680
    },
    {
      "entropy": 1.0988596007227898,
      "epoch": 0.21090510232553664,
      "grad_norm": 0.671875,
      "learning_rate": 1.992276238193788e-05,
      "loss": 1.0932,
      "mean_token_accuracy": 0.7026484467089176,
      "num_tokens": 51363981.0,
      "step": 685
    },
    {
      "entropy": 1.1409345027059317,
      "epoch": 0.2124445556271829,
      "grad_norm": 0.73046875,
      "learning_rate": 1.992074056913464e-05,
      "loss": 1.1352,
      "mean_token_accuracy": 0.6939107369631529,
      "num_tokens": 51737796.0,
      "step": 690
    },
    {
      "entropy": 1.120890161767602,
      "epoch": 0.21398400892882916,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9918692740345804e-05,
      "loss": 1.1338,
      "mean_token_accuracy": 0.6986452504992485,
      "num_tokens": 52114356.0,
      "step": 695
    },
    {
      "entropy": 1.119481140933931,
      "epoch": 0.21552346223047542,
      "grad_norm": 0.74609375,
      "learning_rate": 1.9916618900941567e-05,
      "loss": 1.1131,
      "mean_token_accuracy": 0.6989668853580951,
      "num_tokens": 52484836.0,
      "step": 700
    },
    {
      "entropy": 1.0964285958558322,
      "epoch": 0.21706291553212165,
      "grad_norm": 0.71484375,
      "learning_rate": 1.991451905636033e-05,
      "loss": 1.0937,
      "mean_token_accuracy": 0.7034055396914483,
      "num_tokens": 52859792.0,
      "step": 705
    },
    {
      "entropy": 1.1160279937088489,
      "epoch": 0.2186023688337679,
      "grad_norm": 0.75390625,
      "learning_rate": 1.9912393212108692e-05,
      "loss": 1.1109,
      "mean_token_accuracy": 0.6976350143551826,
      "num_tokens": 53212860.0,
      "step": 710
    },
    {
      "entropy": 1.14042426943779,
      "epoch": 0.22014182213541417,
      "grad_norm": 0.71484375,
      "learning_rate": 1.9910241373761426e-05,
      "loss": 1.1385,
      "mean_token_accuracy": 0.694380571320653,
      "num_tokens": 53583748.0,
      "step": 715
    },
    {
      "entropy": 1.1466562129557132,
      "epoch": 0.22168127543706043,
      "grad_norm": 0.74609375,
      "learning_rate": 1.9908063546961482e-05,
      "loss": 1.1338,
      "mean_token_accuracy": 0.6904741197824478,
      "num_tokens": 53940694.0,
      "step": 720
    },
    {
      "entropy": 1.121897478029132,
      "epoch": 0.22322072873870666,
      "grad_norm": 0.7265625,
      "learning_rate": 1.990585973741996e-05,
      "loss": 1.1211,
      "mean_token_accuracy": 0.6964031044393778,
      "num_tokens": 54316660.0,
      "step": 725
    },
    {
      "entropy": 1.1026945130899548,
      "epoch": 0.22476018204035292,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9903629950916083e-05,
      "loss": 1.0919,
      "mean_token_accuracy": 0.7022646889090538,
      "num_tokens": 54700093.0,
      "step": 730
    },
    {
      "entropy": 1.1913654580712318,
      "epoch": 0.22629963534199918,
      "grad_norm": 0.71875,
      "learning_rate": 1.9901374193297212e-05,
      "loss": 1.1867,
      "mean_token_accuracy": 0.6831524942070246,
      "num_tokens": 55072259.0,
      "step": 735
    },
    {
      "entropy": 1.0986135648563504,
      "epoch": 0.22783908864364544,
      "grad_norm": 0.69921875,
      "learning_rate": 1.989909247047881e-05,
      "loss": 1.0859,
      "mean_token_accuracy": 0.70220061019063,
      "num_tokens": 55431253.0,
      "step": 740
    },
    {
      "entropy": 1.1104688480496407,
      "epoch": 0.22937854194529167,
      "grad_norm": 0.70703125,
      "learning_rate": 1.989678478844443e-05,
      "loss": 1.1053,
      "mean_token_accuracy": 0.7009272977709771,
      "num_tokens": 55806238.0,
      "step": 745
    },
    {
      "entropy": 1.1378316521644591,
      "epoch": 0.23091799524693793,
      "grad_norm": 0.77734375,
      "learning_rate": 1.9894451153245695e-05,
      "loss": 1.1323,
      "mean_token_accuracy": 0.6941636923700572,
      "num_tokens": 56167420.0,
      "step": 750
    },
    {
      "entropy": 1.0988559927791357,
      "epoch": 0.2324574485485842,
      "grad_norm": 0.6875,
      "learning_rate": 1.9892091571002295e-05,
      "loss": 1.1008,
      "mean_token_accuracy": 0.7004305239766836,
      "num_tokens": 56550823.0,
      "step": 755
    },
    {
      "entropy": 1.1214569361880422,
      "epoch": 0.23399690185023045,
      "grad_norm": 0.75,
      "learning_rate": 1.9889706047901956e-05,
      "loss": 1.1044,
      "mean_token_accuracy": 0.6996850349009037,
      "num_tokens": 56920052.0,
      "step": 760
    },
    {
      "entropy": 1.1102859888225793,
      "epoch": 0.23553635515187668,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9887294590200437e-05,
      "loss": 1.11,
      "mean_token_accuracy": 0.6958253476768732,
      "num_tokens": 57288683.0,
      "step": 765
    },
    {
      "entropy": 1.1473000289872288,
      "epoch": 0.23707580845352294,
      "grad_norm": 0.734375,
      "learning_rate": 1.9884857204221503e-05,
      "loss": 1.1542,
      "mean_token_accuracy": 0.6928766690194607,
      "num_tokens": 57645195.0,
      "step": 770
    },
    {
      "entropy": 1.1104052532464266,
      "epoch": 0.2386152617551692,
      "grad_norm": 0.7421875,
      "learning_rate": 1.9882393896356915e-05,
      "loss": 1.096,
      "mean_token_accuracy": 0.6981043085455895,
      "num_tokens": 58010298.0,
      "step": 775
    },
    {
      "entropy": 1.1217896696180105,
      "epoch": 0.24015471505681546,
      "grad_norm": 0.69921875,
      "learning_rate": 1.987990467306641e-05,
      "loss": 1.1139,
      "mean_token_accuracy": 0.6969779424369336,
      "num_tokens": 58403545.0,
      "step": 780
    },
    {
      "entropy": 1.1406789550557732,
      "epoch": 0.2416941683584617,
      "grad_norm": 0.640625,
      "learning_rate": 1.9877389540877686e-05,
      "loss": 1.1472,
      "mean_token_accuracy": 0.6947149783372879,
      "num_tokens": 58774798.0,
      "step": 785
    },
    {
      "entropy": 1.124403426796198,
      "epoch": 0.24323362166010795,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9874848506386392e-05,
      "loss": 1.1214,
      "mean_token_accuracy": 0.6990401953458786,
      "num_tokens": 59159772.0,
      "step": 790
    },
    {
      "entropy": 1.0997486164793373,
      "epoch": 0.2447730749617542,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9872281576256078e-05,
      "loss": 1.0879,
      "mean_token_accuracy": 0.7038913916796445,
      "num_tokens": 59540395.0,
      "step": 795
    },
    {
      "entropy": 1.1355868607759476,
      "epoch": 0.24631252826340047,
      "grad_norm": 0.73046875,
      "learning_rate": 1.9869688757218233e-05,
      "loss": 1.1415,
      "mean_token_accuracy": 0.6912806447595358,
      "num_tokens": 59905249.0,
      "step": 800
    },
    {
      "entropy": 1.1410336146131157,
      "epoch": 0.2478519815650467,
      "grad_norm": 0.71875,
      "learning_rate": 1.9867070056072215e-05,
      "loss": 1.1433,
      "mean_token_accuracy": 0.695595920830965,
      "num_tokens": 60268255.0,
      "step": 805
    },
    {
      "entropy": 1.1349990352988244,
      "epoch": 0.24939143486669296,
      "grad_norm": 0.734375,
      "learning_rate": 1.986442547968527e-05,
      "loss": 1.1253,
      "mean_token_accuracy": 0.6944803945720196,
      "num_tokens": 60639389.0,
      "step": 810
    },
    {
      "entropy": 1.071821440383792,
      "epoch": 0.2509308881683392,
      "grad_norm": 0.75,
      "learning_rate": 1.9861755034992483e-05,
      "loss": 1.0666,
      "mean_token_accuracy": 0.7084377076476812,
      "num_tokens": 61024428.0,
      "step": 815
    },
    {
      "entropy": 1.1051497608423233,
      "epoch": 0.2524703414699855,
      "grad_norm": 0.7421875,
      "learning_rate": 1.9859058728996788e-05,
      "loss": 1.0965,
      "mean_token_accuracy": 0.7005310852080584,
      "num_tokens": 61402658.0,
      "step": 820
    },
    {
      "entropy": 1.1438620645552873,
      "epoch": 0.2540097947716317,
      "grad_norm": 0.71875,
      "learning_rate": 1.9856336568768936e-05,
      "loss": 1.1399,
      "mean_token_accuracy": 0.6912882044911385,
      "num_tokens": 61780493.0,
      "step": 825
    },
    {
      "entropy": 1.0913935292512178,
      "epoch": 0.255549248073278,
      "grad_norm": 0.76953125,
      "learning_rate": 1.985358856144747e-05,
      "loss": 1.0922,
      "mean_token_accuracy": 0.7028463281691074,
      "num_tokens": 62152861.0,
      "step": 830
    },
    {
      "entropy": 1.1374814191833138,
      "epoch": 0.25708870137492423,
      "grad_norm": 0.75,
      "learning_rate": 1.9850814714238718e-05,
      "loss": 1.13,
      "mean_token_accuracy": 0.6964493870735169,
      "num_tokens": 62514124.0,
      "step": 835
    },
    {
      "entropy": 1.1332073567435146,
      "epoch": 0.25862815467657047,
      "grad_norm": 0.69140625,
      "learning_rate": 1.9848015034416776e-05,
      "loss": 1.1199,
      "mean_token_accuracy": 0.6939284823834896,
      "num_tokens": 62901998.0,
      "step": 840
    },
    {
      "entropy": 1.1137794975191355,
      "epoch": 0.26016760797821675,
      "grad_norm": 0.7734375,
      "learning_rate": 1.9845189529323473e-05,
      "loss": 1.103,
      "mean_token_accuracy": 0.7006840953603387,
      "num_tokens": 63266839.0,
      "step": 845
    },
    {
      "entropy": 1.1099360350519418,
      "epoch": 0.261707061279863,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9842338206368375e-05,
      "loss": 1.0998,
      "mean_token_accuracy": 0.6972894985228777,
      "num_tokens": 63654737.0,
      "step": 850
    },
    {
      "entropy": 1.1284802999347447,
      "epoch": 0.2632465145815092,
      "grad_norm": 0.75390625,
      "learning_rate": 1.9839461073028733e-05,
      "loss": 1.1083,
      "mean_token_accuracy": 0.6968803893774748,
      "num_tokens": 64032094.0,
      "step": 855
    },
    {
      "entropy": 1.0841836363077164,
      "epoch": 0.2647859678831555,
      "grad_norm": 0.75390625,
      "learning_rate": 1.98365581368495e-05,
      "loss": 1.0933,
      "mean_token_accuracy": 0.7028985098004341,
      "num_tokens": 64418973.0,
      "step": 860
    },
    {
      "entropy": 1.1070842415094375,
      "epoch": 0.26632542118480174,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9833629405443283e-05,
      "loss": 1.1091,
      "mean_token_accuracy": 0.7023597385734319,
      "num_tokens": 64798969.0,
      "step": 865
    },
    {
      "entropy": 1.1159617979079486,
      "epoch": 0.267864874486448,
      "grad_norm": 0.734375,
      "learning_rate": 1.983067488649035e-05,
      "loss": 1.1119,
      "mean_token_accuracy": 0.6988450512290001,
      "num_tokens": 65171715.0,
      "step": 870
    },
    {
      "entropy": 1.1518226452171803,
      "epoch": 0.26940432778809426,
      "grad_norm": 0.72265625,
      "learning_rate": 1.982769458773857e-05,
      "loss": 1.1421,
      "mean_token_accuracy": 0.6895007479935884,
      "num_tokens": 65548454.0,
      "step": 875
    },
    {
      "entropy": 1.1310207761824131,
      "epoch": 0.2709437810897405,
      "grad_norm": 0.71875,
      "learning_rate": 1.9824688517003433e-05,
      "loss": 1.1165,
      "mean_token_accuracy": 0.6969778280705213,
      "num_tokens": 65912713.0,
      "step": 880
    },
    {
      "entropy": 1.0850966442376375,
      "epoch": 0.2724832343913868,
      "grad_norm": 0.78125,
      "learning_rate": 1.9821656682168013e-05,
      "loss": 1.0868,
      "mean_token_accuracy": 0.7045084740966558,
      "num_tokens": 66288389.0,
      "step": 885
    },
    {
      "entropy": 1.116120758280158,
      "epoch": 0.274022687693033,
      "grad_norm": 0.69921875,
      "learning_rate": 1.981859909118294e-05,
      "loss": 1.1183,
      "mean_token_accuracy": 0.6976446002721787,
      "num_tokens": 66673853.0,
      "step": 890
    },
    {
      "entropy": 1.1547842472791672,
      "epoch": 0.27556214099467924,
      "grad_norm": 0.76171875,
      "learning_rate": 1.9815515752066386e-05,
      "loss": 1.1512,
      "mean_token_accuracy": 0.6902260981500149,
      "num_tokens": 67054614.0,
      "step": 895
    },
    {
      "entropy": 1.0831589922308922,
      "epoch": 0.2771015942963255,
      "grad_norm": 0.6796875,
      "learning_rate": 1.9812406672904058e-05,
      "loss": 1.0824,
      "mean_token_accuracy": 0.7042932607233524,
      "num_tokens": 67422842.0,
      "step": 900
    },
    {
      "entropy": 1.1236521264538168,
      "epoch": 0.27864104759797176,
      "grad_norm": 0.80078125,
      "learning_rate": 1.9809271861849147e-05,
      "loss": 1.1257,
      "mean_token_accuracy": 0.6985704395920038,
      "num_tokens": 67791499.0,
      "step": 905
    },
    {
      "entropy": 1.1213534710928799,
      "epoch": 0.28018050089961805,
      "grad_norm": 0.72265625,
      "learning_rate": 1.9806111327122332e-05,
      "loss": 1.1056,
      "mean_token_accuracy": 0.6963206488639117,
      "num_tokens": 68168533.0,
      "step": 910
    },
    {
      "entropy": 1.0945915594696998,
      "epoch": 0.2817199542012643,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9802925077011742e-05,
      "loss": 1.0805,
      "mean_token_accuracy": 0.7033400624990463,
      "num_tokens": 68534200.0,
      "step": 915
    },
    {
      "entropy": 1.1363273495808244,
      "epoch": 0.2832594075029105,
      "grad_norm": 0.74609375,
      "learning_rate": 1.979971311987295e-05,
      "loss": 1.1331,
      "mean_token_accuracy": 0.6933297269046307,
      "num_tokens": 68904756.0,
      "step": 920
    },
    {
      "entropy": 1.113439468294382,
      "epoch": 0.2847988608045568,
      "grad_norm": 0.67578125,
      "learning_rate": 1.9796475464128943e-05,
      "loss": 1.1136,
      "mean_token_accuracy": 0.6984921019524336,
      "num_tokens": 69279269.0,
      "step": 925
    },
    {
      "entropy": 1.106918627768755,
      "epoch": 0.28633831410620303,
      "grad_norm": 0.7265625,
      "learning_rate": 1.979321211827009e-05,
      "loss": 1.0971,
      "mean_token_accuracy": 0.699981477856636,
      "num_tokens": 69652060.0,
      "step": 930
    },
    {
      "entropy": 1.1193274904042483,
      "epoch": 0.28787776740784926,
      "grad_norm": 0.6953125,
      "learning_rate": 1.9789923090854138e-05,
      "loss": 1.1141,
      "mean_token_accuracy": 0.6945442810654641,
      "num_tokens": 70031530.0,
      "step": 935
    },
    {
      "entropy": 1.1201645512133838,
      "epoch": 0.28941722070949555,
      "grad_norm": 0.69921875,
      "learning_rate": 1.9786608390506176e-05,
      "loss": 1.1248,
      "mean_token_accuracy": 0.6959635071456433,
      "num_tokens": 70405759.0,
      "step": 940
    },
    {
      "entropy": 1.158202064409852,
      "epoch": 0.2909566740111418,
      "grad_norm": 0.7578125,
      "learning_rate": 1.9783268025918622e-05,
      "loss": 1.1405,
      "mean_token_accuracy": 0.6916137792170047,
      "num_tokens": 70773411.0,
      "step": 945
    },
    {
      "entropy": 1.1136011434718966,
      "epoch": 0.29249612731278807,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9779902005851187e-05,
      "loss": 1.1032,
      "mean_token_accuracy": 0.6998088311403989,
      "num_tokens": 71138724.0,
      "step": 950
    },
    {
      "entropy": 1.1528440322726965,
      "epoch": 0.2940355806144343,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9776510339130874e-05,
      "loss": 1.1432,
      "mean_token_accuracy": 0.6916706405580044,
      "num_tokens": 71522642.0,
      "step": 955
    },
    {
      "entropy": 1.1416105089709163,
      "epoch": 0.29557503391608053,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9773093034651928e-05,
      "loss": 1.1315,
      "mean_token_accuracy": 0.6934664513915777,
      "num_tokens": 71885197.0,
      "step": 960
    },
    {
      "entropy": 1.1288053080439568,
      "epoch": 0.2971144872177268,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9769650101375835e-05,
      "loss": 1.1253,
      "mean_token_accuracy": 0.6961982771754265,
      "num_tokens": 72262607.0,
      "step": 965
    },
    {
      "entropy": 1.1250951839610934,
      "epoch": 0.29865394051937305,
      "grad_norm": 0.6796875,
      "learning_rate": 1.9766181548331283e-05,
      "loss": 1.1186,
      "mean_token_accuracy": 0.6970525443553924,
      "num_tokens": 72653402.0,
      "step": 970
    },
    {
      "entropy": 1.1368791069835424,
      "epoch": 0.3001933938210193,
      "grad_norm": 0.78125,
      "learning_rate": 1.9762687384614152e-05,
      "loss": 1.132,
      "mean_token_accuracy": 0.6947309102863073,
      "num_tokens": 73049063.0,
      "step": 975
    },
    {
      "entropy": 1.083188571408391,
      "epoch": 0.30173284712266557,
      "grad_norm": 0.69921875,
      "learning_rate": 1.9759167619387474e-05,
      "loss": 1.0838,
      "mean_token_accuracy": 0.7051486879587173,
      "num_tokens": 73435640.0,
      "step": 980
    },
    {
      "entropy": 1.1301437310874463,
      "epoch": 0.3032723004243118,
      "grad_norm": 0.71875,
      "learning_rate": 1.975562226188143e-05,
      "loss": 1.1135,
      "mean_token_accuracy": 0.6951402083039284,
      "num_tokens": 73804251.0,
      "step": 985
    },
    {
      "entropy": 1.1021712820976972,
      "epoch": 0.3048117537259581,
      "grad_norm": 0.72265625,
      "learning_rate": 1.97520513213933e-05,
      "loss": 1.1047,
      "mean_token_accuracy": 0.6995002727955579,
      "num_tokens": 74186734.0,
      "step": 990
    },
    {
      "entropy": 1.1268383231014014,
      "epoch": 0.3063512070276043,
      "grad_norm": 0.67578125,
      "learning_rate": 1.9748454807287458e-05,
      "loss": 1.1197,
      "mean_token_accuracy": 0.6963075909763574,
      "num_tokens": 74564695.0,
      "step": 995
    },
    {
      "entropy": 1.1207784935832024,
      "epoch": 0.30789066032925055,
      "grad_norm": 0.7421875,
      "learning_rate": 1.974483272899535e-05,
      "loss": 1.1225,
      "mean_token_accuracy": 0.6975628361105919,
      "num_tokens": 74954009.0,
      "step": 1000
    },
    {
      "entropy": 1.1320615615695715,
      "epoch": 0.30943011363089684,
      "grad_norm": 0.7109375,
      "learning_rate": 1.974118509601545e-05,
      "loss": 1.1245,
      "mean_token_accuracy": 0.6957272073253989,
      "num_tokens": 75329081.0,
      "step": 1005
    },
    {
      "entropy": 1.0997402749955654,
      "epoch": 0.3109695669325431,
      "grad_norm": 0.7265625,
      "learning_rate": 1.973751191791325e-05,
      "loss": 1.0889,
      "mean_token_accuracy": 0.7017048012465239,
      "num_tokens": 75696214.0,
      "step": 1010
    },
    {
      "entropy": 1.0914274197071792,
      "epoch": 0.31250902023418936,
      "grad_norm": 0.671875,
      "learning_rate": 1.9733813204321233e-05,
      "loss": 1.0761,
      "mean_token_accuracy": 0.7037324849516153,
      "num_tokens": 76068328.0,
      "step": 1015
    },
    {
      "entropy": 1.1191833563148976,
      "epoch": 0.3140484735358356,
      "grad_norm": 0.71484375,
      "learning_rate": 1.9730088964938842e-05,
      "loss": 1.1121,
      "mean_token_accuracy": 0.6989397961646319,
      "num_tokens": 76441289.0,
      "step": 1020
    },
    {
      "entropy": 1.1179291112348437,
      "epoch": 0.3155879268374818,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9726339209532462e-05,
      "loss": 1.0951,
      "mean_token_accuracy": 0.7015183545649052,
      "num_tokens": 76814090.0,
      "step": 1025
    },
    {
      "entropy": 1.1256714541465045,
      "epoch": 0.3171273801391281,
      "grad_norm": 0.72265625,
      "learning_rate": 1.972256394793539e-05,
      "loss": 1.1131,
      "mean_token_accuracy": 0.6952388241887093,
      "num_tokens": 77179591.0,
      "step": 1030
    },
    {
      "entropy": 1.136961457133293,
      "epoch": 0.31866683344077434,
      "grad_norm": 0.7421875,
      "learning_rate": 1.971876319004781e-05,
      "loss": 1.1182,
      "mean_token_accuracy": 0.6935086127370595,
      "num_tokens": 77550799.0,
      "step": 1035
    },
    {
      "entropy": 1.1292035300284624,
      "epoch": 0.3202062867424206,
      "grad_norm": 0.73828125,
      "learning_rate": 1.9714936945836764e-05,
      "loss": 1.1177,
      "mean_token_accuracy": 0.6930017314851284,
      "num_tokens": 77920454.0,
      "step": 1040
    },
    {
      "entropy": 1.0919932153075933,
      "epoch": 0.32174574004406686,
      "grad_norm": 0.75390625,
      "learning_rate": 1.971108522533613e-05,
      "loss": 1.078,
      "mean_token_accuracy": 0.7059255817905068,
      "num_tokens": 78285931.0,
      "step": 1045
    },
    {
      "entropy": 1.116225427389145,
      "epoch": 0.3232851933457131,
      "grad_norm": 0.73046875,
      "learning_rate": 1.9707208038646605e-05,
      "loss": 1.1197,
      "mean_token_accuracy": 0.6983599919825793,
      "num_tokens": 78657492.0,
      "step": 1050
    },
    {
      "entropy": 1.1196499440819025,
      "epoch": 0.3248246466473594,
      "grad_norm": 0.7109375,
      "learning_rate": 1.970330539593565e-05,
      "loss": 1.1043,
      "mean_token_accuracy": 0.6965556625276804,
      "num_tokens": 79026794.0,
      "step": 1055
    },
    {
      "entropy": 1.1315435644239187,
      "epoch": 0.3263640999490056,
      "grad_norm": 0.71875,
      "learning_rate": 1.969937730743749e-05,
      "loss": 1.1248,
      "mean_token_accuracy": 0.696228601038456,
      "num_tokens": 79400596.0,
      "step": 1060
    },
    {
      "entropy": 1.0820589877665043,
      "epoch": 0.32790355325065185,
      "grad_norm": 0.7421875,
      "learning_rate": 1.9695423783453086e-05,
      "loss": 1.0836,
      "mean_token_accuracy": 0.705596823990345,
      "num_tokens": 79767657.0,
      "step": 1065
    },
    {
      "entropy": 1.1282978903502225,
      "epoch": 0.32944300655229813,
      "grad_norm": 0.703125,
      "learning_rate": 1.969144483435009e-05,
      "loss": 1.1241,
      "mean_token_accuracy": 0.6953945115208626,
      "num_tokens": 80144261.0,
      "step": 1070
    },
    {
      "entropy": 1.1406402667984366,
      "epoch": 0.33098245985394437,
      "grad_norm": 0.734375,
      "learning_rate": 1.968744047056283e-05,
      "loss": 1.1297,
      "mean_token_accuracy": 0.6934267330914736,
      "num_tokens": 80525193.0,
      "step": 1075
    },
    {
      "entropy": 1.1295597156509758,
      "epoch": 0.3325219131555906,
      "grad_norm": 0.75,
      "learning_rate": 1.9683410702592284e-05,
      "loss": 1.1203,
      "mean_token_accuracy": 0.6963451337069273,
      "num_tokens": 80905579.0,
      "step": 1080
    },
    {
      "entropy": 1.1091342974454164,
      "epoch": 0.3340613664572369,
      "grad_norm": 0.72265625,
      "learning_rate": 1.9679355541006056e-05,
      "loss": 1.1029,
      "mean_token_accuracy": 0.6987722083926201,
      "num_tokens": 81285902.0,
      "step": 1085
    },
    {
      "entropy": 1.1366256965324282,
      "epoch": 0.3356008197588831,
      "grad_norm": 0.77734375,
      "learning_rate": 1.9675274996438324e-05,
      "loss": 1.1364,
      "mean_token_accuracy": 0.6947485759854317,
      "num_tokens": 81653385.0,
      "step": 1090
    },
    {
      "entropy": 1.1193459507077932,
      "epoch": 0.3371402730605294,
      "grad_norm": 0.74609375,
      "learning_rate": 1.967116907958985e-05,
      "loss": 1.1022,
      "mean_token_accuracy": 0.7003271400928497,
      "num_tokens": 82021575.0,
      "step": 1095
    },
    {
      "entropy": 1.1462248302996159,
      "epoch": 0.33867972636217564,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9667037801227914e-05,
      "loss": 1.1351,
      "mean_token_accuracy": 0.6924805622547865,
      "num_tokens": 82399380.0,
      "step": 1100
    },
    {
      "entropy": 1.107696833834052,
      "epoch": 0.34021917966382187,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9662881172186313e-05,
      "loss": 1.1058,
      "mean_token_accuracy": 0.7026475485414266,
      "num_tokens": 82771918.0,
      "step": 1105
    },
    {
      "entropy": 1.1295810148119927,
      "epoch": 0.34175863296546816,
      "grad_norm": 0.75390625,
      "learning_rate": 1.965869920336533e-05,
      "loss": 1.1236,
      "mean_token_accuracy": 0.6973136257380247,
      "num_tokens": 83137444.0,
      "step": 1110
    },
    {
      "entropy": 1.1330510720610618,
      "epoch": 0.3432980862671144,
      "grad_norm": 0.80859375,
      "learning_rate": 1.965449190573168e-05,
      "loss": 1.1317,
      "mean_token_accuracy": 0.6966150533407927,
      "num_tokens": 83500864.0,
      "step": 1115
    },
    {
      "entropy": 1.1691056948155165,
      "epoch": 0.3448375395687606,
      "grad_norm": 0.7109375,
      "learning_rate": 1.965025929031852e-05,
      "loss": 1.1643,
      "mean_token_accuracy": 0.6873539566993714,
      "num_tokens": 83868277.0,
      "step": 1120
    },
    {
      "entropy": 1.0945931367576123,
      "epoch": 0.3463769928704069,
      "grad_norm": 0.67578125,
      "learning_rate": 1.9646001368225382e-05,
      "loss": 1.1016,
      "mean_token_accuracy": 0.7012989364564419,
      "num_tokens": 84252198.0,
      "step": 1125
    },
    {
      "entropy": 1.165550697594881,
      "epoch": 0.34791644617205314,
      "grad_norm": 0.74609375,
      "learning_rate": 1.9641718150618177e-05,
      "loss": 1.1467,
      "mean_token_accuracy": 0.6915617097169161,
      "num_tokens": 84617359.0,
      "step": 1130
    },
    {
      "entropy": 1.1262395735830069,
      "epoch": 0.3494558994736994,
      "grad_norm": 0.71875,
      "learning_rate": 1.9637409648729142e-05,
      "loss": 1.1208,
      "mean_token_accuracy": 0.6933946672827005,
      "num_tokens": 85000015.0,
      "step": 1135
    },
    {
      "entropy": 1.129942279495299,
      "epoch": 0.35099535277534566,
      "grad_norm": 0.7109375,
      "learning_rate": 1.963307587385682e-05,
      "loss": 1.1206,
      "mean_token_accuracy": 0.6961002223193645,
      "num_tokens": 85384245.0,
      "step": 1140
    },
    {
      "entropy": 1.1344817027449607,
      "epoch": 0.3525348060769919,
      "grad_norm": 0.703125,
      "learning_rate": 1.962871683736603e-05,
      "loss": 1.1255,
      "mean_token_accuracy": 0.6936954416334629,
      "num_tokens": 85754567.0,
      "step": 1145
    },
    {
      "entropy": 1.1344742052257062,
      "epoch": 0.3540742593786382,
      "grad_norm": 0.69921875,
      "learning_rate": 1.9624332550687834e-05,
      "loss": 1.1209,
      "mean_token_accuracy": 0.695965689048171,
      "num_tokens": 86130538.0,
      "step": 1150
    },
    {
      "entropy": 1.1286322576925159,
      "epoch": 0.3556137126802844,
      "grad_norm": 0.703125,
      "learning_rate": 1.961992302531952e-05,
      "loss": 1.1213,
      "mean_token_accuracy": 0.6975085325539112,
      "num_tokens": 86505523.0,
      "step": 1155
    },
    {
      "entropy": 1.161212246119976,
      "epoch": 0.35715316598193064,
      "grad_norm": 0.69140625,
      "learning_rate": 1.961548827282455e-05,
      "loss": 1.1607,
      "mean_token_accuracy": 0.6912364710122347,
      "num_tokens": 86889854.0,
      "step": 1160
    },
    {
      "entropy": 1.1156503956764936,
      "epoch": 0.35869261928357693,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9611028304832547e-05,
      "loss": 1.1063,
      "mean_token_accuracy": 0.697390603646636,
      "num_tokens": 87275027.0,
      "step": 1165
    },
    {
      "entropy": 1.117064966261387,
      "epoch": 0.36023207258522316,
      "grad_norm": 0.69140625,
      "learning_rate": 1.9606543133039254e-05,
      "loss": 1.1028,
      "mean_token_accuracy": 0.70021205060184,
      "num_tokens": 87649310.0,
      "step": 1170
    },
    {
      "entropy": 1.1142473477870225,
      "epoch": 0.36177152588686945,
      "grad_norm": 0.71484375,
      "learning_rate": 1.9602032769206517e-05,
      "loss": 1.1113,
      "mean_token_accuracy": 0.6982534524053335,
      "num_tokens": 88021284.0,
      "step": 1175
    },
    {
      "entropy": 1.1146583622321486,
      "epoch": 0.3633109791885157,
      "grad_norm": 0.734375,
      "learning_rate": 1.9597497225162233e-05,
      "loss": 1.1098,
      "mean_token_accuracy": 0.6998208686709404,
      "num_tokens": 88397256.0,
      "step": 1180
    },
    {
      "entropy": 1.1496895281597972,
      "epoch": 0.3648504324901619,
      "grad_norm": 0.76171875,
      "learning_rate": 1.959293651280034e-05,
      "loss": 1.1573,
      "mean_token_accuracy": 0.6919562425464392,
      "num_tokens": 88763426.0,
      "step": 1185
    },
    {
      "entropy": 1.1269241459667683,
      "epoch": 0.3663898857918082,
      "grad_norm": 0.73828125,
      "learning_rate": 1.9588350644080777e-05,
      "loss": 1.1105,
      "mean_token_accuracy": 0.6988791462033987,
      "num_tokens": 89118435.0,
      "step": 1190
    },
    {
      "entropy": 1.0833050038665533,
      "epoch": 0.36792933909345443,
      "grad_norm": 0.6796875,
      "learning_rate": 1.9583739631029446e-05,
      "loss": 1.0802,
      "mean_token_accuracy": 0.7041361082345248,
      "num_tokens": 89508601.0,
      "step": 1195
    },
    {
      "entropy": 1.1451345276087523,
      "epoch": 0.36946879239510066,
      "grad_norm": 0.734375,
      "learning_rate": 1.957910348573819e-05,
      "loss": 1.1369,
      "mean_token_accuracy": 0.6917477056384087,
      "num_tokens": 89893267.0,
      "step": 1200
    },
    {
      "entropy": 1.0983753545209765,
      "epoch": 0.37100824569674695,
      "grad_norm": 0.6953125,
      "learning_rate": 1.9574442220364768e-05,
      "loss": 1.0951,
      "mean_token_accuracy": 0.7006513494998217,
      "num_tokens": 90284037.0,
      "step": 1205
    },
    {
      "entropy": 1.116368711926043,
      "epoch": 0.3725476989983932,
      "grad_norm": 0.71484375,
      "learning_rate": 1.9569755847132796e-05,
      "loss": 1.1144,
      "mean_token_accuracy": 0.6977820225059986,
      "num_tokens": 90670311.0,
      "step": 1210
    },
    {
      "entropy": 1.094654480740428,
      "epoch": 0.37408715230003947,
      "grad_norm": 0.734375,
      "learning_rate": 1.9565044378331745e-05,
      "loss": 1.0829,
      "mean_token_accuracy": 0.7023903023451566,
      "num_tokens": 91055385.0,
      "step": 1215
    },
    {
      "entropy": 1.115629038028419,
      "epoch": 0.3756266056016857,
      "grad_norm": 0.71484375,
      "learning_rate": 1.9560307826316892e-05,
      "loss": 1.1184,
      "mean_token_accuracy": 0.6971225813031197,
      "num_tokens": 91424223.0,
      "step": 1220
    },
    {
      "entropy": 1.1635486509650945,
      "epoch": 0.37716605890333194,
      "grad_norm": 0.75390625,
      "learning_rate": 1.9555546203509297e-05,
      "loss": 1.1615,
      "mean_token_accuracy": 0.6906606066972018,
      "num_tokens": 91796012.0,
      "step": 1225
    },
    {
      "entropy": 1.120249069854617,
      "epoch": 0.3787055122049782,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9550759522395753e-05,
      "loss": 1.1022,
      "mean_token_accuracy": 0.6964717313647271,
      "num_tokens": 92162375.0,
      "step": 1230
    },
    {
      "entropy": 1.1134164540097118,
      "epoch": 0.38024496550662445,
      "grad_norm": 0.734375,
      "learning_rate": 1.9545947795528777e-05,
      "loss": 1.1205,
      "mean_token_accuracy": 0.6987472154200077,
      "num_tokens": 92545936.0,
      "step": 1235
    },
    {
      "entropy": 1.086888193897903,
      "epoch": 0.3817844188082707,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9541111035526563e-05,
      "loss": 1.0868,
      "mean_token_accuracy": 0.7030756056308747,
      "num_tokens": 92928503.0,
      "step": 1240
    },
    {
      "entropy": 1.1007358327507972,
      "epoch": 0.383323872109917,
      "grad_norm": 0.6953125,
      "learning_rate": 1.953624925507295e-05,
      "loss": 1.0914,
      "mean_token_accuracy": 0.7022611912339926,
      "num_tokens": 93313261.0,
      "step": 1245
    },
    {
      "entropy": 1.135701997205615,
      "epoch": 0.3848633254115632,
      "grad_norm": 0.71875,
      "learning_rate": 1.9531362466917388e-05,
      "loss": 1.1208,
      "mean_token_accuracy": 0.6953465500846505,
      "num_tokens": 93694867.0,
      "step": 1250
    },
    {
      "entropy": 1.141565052047372,
      "epoch": 0.3864027787132095,
      "grad_norm": 0.74609375,
      "learning_rate": 1.952645068387491e-05,
      "loss": 1.1328,
      "mean_token_accuracy": 0.69326724819839,
      "num_tokens": 94082631.0,
      "step": 1255
    },
    {
      "entropy": 1.1221220299601555,
      "epoch": 0.3879422320148557,
      "grad_norm": 0.72265625,
      "learning_rate": 1.95215139188261e-05,
      "loss": 1.1079,
      "mean_token_accuracy": 0.6988440815359354,
      "num_tokens": 94457533.0,
      "step": 1260
    },
    {
      "entropy": 1.1194452840834856,
      "epoch": 0.38948168531650196,
      "grad_norm": 0.73046875,
      "learning_rate": 1.9516552184717036e-05,
      "loss": 1.1215,
      "mean_token_accuracy": 0.696359645575285,
      "num_tokens": 94829571.0,
      "step": 1265
    },
    {
      "entropy": 1.095098103582859,
      "epoch": 0.39102113861814825,
      "grad_norm": 0.69140625,
      "learning_rate": 1.9511565494559298e-05,
      "loss": 1.0958,
      "mean_token_accuracy": 0.7043366067111492,
      "num_tokens": 95200118.0,
      "step": 1270
    },
    {
      "entropy": 1.1012009687721729,
      "epoch": 0.3925605919197945,
      "grad_norm": 0.73828125,
      "learning_rate": 1.95065538614299e-05,
      "loss": 1.0911,
      "mean_token_accuracy": 0.7037279218435287,
      "num_tokens": 95570272.0,
      "step": 1275
    },
    {
      "entropy": 1.1257656961679459,
      "epoch": 0.3941000452214407,
      "grad_norm": 0.77734375,
      "learning_rate": 1.950151729847126e-05,
      "loss": 1.1224,
      "mean_token_accuracy": 0.6986918987706303,
      "num_tokens": 95942085.0,
      "step": 1280
    },
    {
      "entropy": 1.1478566963225603,
      "epoch": 0.395639498523087,
      "grad_norm": 0.68359375,
      "learning_rate": 1.949645581889118e-05,
      "loss": 1.1455,
      "mean_token_accuracy": 0.6915257848799229,
      "num_tokens": 96317314.0,
      "step": 1285
    },
    {
      "entropy": 1.1262834113091231,
      "epoch": 0.39717895182473323,
      "grad_norm": 0.68359375,
      "learning_rate": 1.9491369435962802e-05,
      "loss": 1.1219,
      "mean_token_accuracy": 0.6945701058954,
      "num_tokens": 96698109.0,
      "step": 1290
    },
    {
      "entropy": 1.0999675188213587,
      "epoch": 0.3987184051263795,
      "grad_norm": 0.71875,
      "learning_rate": 1.9486258163024567e-05,
      "loss": 1.0967,
      "mean_token_accuracy": 0.7008256938308477,
      "num_tokens": 97074251.0,
      "step": 1295
    },
    {
      "entropy": 1.127872526086867,
      "epoch": 0.40025785842802575,
      "grad_norm": 0.76171875,
      "learning_rate": 1.94811220134802e-05,
      "loss": 1.1225,
      "mean_token_accuracy": 0.6996113903820514,
      "num_tokens": 97446524.0,
      "step": 1300
    },
    {
      "entropy": 1.1096723936498165,
      "epoch": 0.401797311729672,
      "grad_norm": 0.73046875,
      "learning_rate": 1.9475961000798645e-05,
      "loss": 1.1048,
      "mean_token_accuracy": 0.6985050681978464,
      "num_tokens": 97831999.0,
      "step": 1305
    },
    {
      "entropy": 1.1193984607234597,
      "epoch": 0.40333676503131827,
      "grad_norm": 0.734375,
      "learning_rate": 1.9470775138514063e-05,
      "loss": 1.111,
      "mean_token_accuracy": 0.6988893166184426,
      "num_tokens": 98203695.0,
      "step": 1310
    },
    {
      "entropy": 1.1497751019895077,
      "epoch": 0.4048762183329645,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9465564440225768e-05,
      "loss": 1.1275,
      "mean_token_accuracy": 0.6922019004821778,
      "num_tokens": 98579150.0,
      "step": 1315
    },
    {
      "entropy": 1.146084163337946,
      "epoch": 0.40641567163461073,
      "grad_norm": 0.7421875,
      "learning_rate": 1.9460328919598216e-05,
      "loss": 1.1382,
      "mean_token_accuracy": 0.6955520674586296,
      "num_tokens": 98943956.0,
      "step": 1320
    },
    {
      "entropy": 1.1438047185540199,
      "epoch": 0.407955124936257,
      "grad_norm": 0.7421875,
      "learning_rate": 1.9455068590360943e-05,
      "loss": 1.1408,
      "mean_token_accuracy": 0.6917512208223343,
      "num_tokens": 99326339.0,
      "step": 1325
    },
    {
      "entropy": 1.1105543179437518,
      "epoch": 0.40949457823790325,
      "grad_norm": 0.765625,
      "learning_rate": 1.9449783466308553e-05,
      "loss": 1.1093,
      "mean_token_accuracy": 0.6983879100531339,
      "num_tokens": 99693304.0,
      "step": 1330
    },
    {
      "entropy": 1.1154874304309488,
      "epoch": 0.41103403153954954,
      "grad_norm": 0.67578125,
      "learning_rate": 1.9444473561300666e-05,
      "loss": 1.1124,
      "mean_token_accuracy": 0.6970164556056261,
      "num_tokens": 100076953.0,
      "step": 1335
    },
    {
      "entropy": 1.1657130055129528,
      "epoch": 0.41257348484119577,
      "grad_norm": 0.7109375,
      "learning_rate": 1.943913888926189e-05,
      "loss": 1.1697,
      "mean_token_accuracy": 0.6894268091768027,
      "num_tokens": 100447813.0,
      "step": 1340
    },
    {
      "entropy": 1.0991169594228267,
      "epoch": 0.414112938142842,
      "grad_norm": 0.7109375,
      "learning_rate": 1.943377946418178e-05,
      "loss": 1.0923,
      "mean_token_accuracy": 0.7011879209429026,
      "num_tokens": 100835066.0,
      "step": 1345
    },
    {
      "entropy": 1.0682245310395957,
      "epoch": 0.4156523914444883,
      "grad_norm": 0.7734375,
      "learning_rate": 1.9428395300114803e-05,
      "loss": 1.0678,
      "mean_token_accuracy": 0.709788928553462,
      "num_tokens": 101195353.0,
      "step": 1350
    },
    {
      "entropy": 1.1283558243885636,
      "epoch": 0.4171918447461345,
      "grad_norm": 0.66796875,
      "learning_rate": 1.94229864111803e-05,
      "loss": 1.1191,
      "mean_token_accuracy": 0.6948264576494694,
      "num_tokens": 101571638.0,
      "step": 1355
    },
    {
      "entropy": 1.114413278736174,
      "epoch": 0.4187312980477808,
      "grad_norm": 0.6953125,
      "learning_rate": 1.9417552811562457e-05,
      "loss": 1.1174,
      "mean_token_accuracy": 0.6991838045418263,
      "num_tokens": 101941042.0,
      "step": 1360
    },
    {
      "entropy": 1.1175668630748987,
      "epoch": 0.42027075134942704,
      "grad_norm": 0.74609375,
      "learning_rate": 1.941209451551025e-05,
      "loss": 1.1069,
      "mean_token_accuracy": 0.6971942469477653,
      "num_tokens": 102314017.0,
      "step": 1365
    },
    {
      "entropy": 1.0761339336633682,
      "epoch": 0.4218102046510733,
      "grad_norm": 0.69921875,
      "learning_rate": 1.9406611537337425e-05,
      "loss": 1.0622,
      "mean_token_accuracy": 0.7066352400928736,
      "num_tokens": 102691186.0,
      "step": 1370
    },
    {
      "entropy": 1.127281517535448,
      "epoch": 0.42334965795271956,
      "grad_norm": 0.7890625,
      "learning_rate": 1.9401103891422455e-05,
      "loss": 1.1202,
      "mean_token_accuracy": 0.6942048665136099,
      "num_tokens": 103071698.0,
      "step": 1375
    },
    {
      "entropy": 1.086047711968422,
      "epoch": 0.4248891112543658,
      "grad_norm": 0.71484375,
      "learning_rate": 1.93955715922085e-05,
      "loss": 1.0819,
      "mean_token_accuracy": 0.7019645646214485,
      "num_tokens": 103457908.0,
      "step": 1380
    },
    {
      "entropy": 1.1055951166898013,
      "epoch": 0.426428564556012,
      "grad_norm": 0.71875,
      "learning_rate": 1.939001465420337e-05,
      "loss": 1.1062,
      "mean_token_accuracy": 0.6994627211242914,
      "num_tokens": 103836470.0,
      "step": 1385
    },
    {
      "entropy": 1.1144292425364255,
      "epoch": 0.4279680178576583,
      "grad_norm": 0.71875,
      "learning_rate": 1.938443309197948e-05,
      "loss": 1.1048,
      "mean_token_accuracy": 0.6957655198872089,
      "num_tokens": 104216895.0,
      "step": 1390
    },
    {
      "entropy": 1.1153745524585248,
      "epoch": 0.42950747115930454,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9378826920173835e-05,
      "loss": 1.1234,
      "mean_token_accuracy": 0.6969841800630092,
      "num_tokens": 104601319.0,
      "step": 1395
    },
    {
      "entropy": 1.1316595112904906,
      "epoch": 0.43104692446095083,
      "grad_norm": 0.71875,
      "learning_rate": 1.9373196153487962e-05,
      "loss": 1.129,
      "mean_token_accuracy": 0.6969845864921809,
      "num_tokens": 104967171.0,
      "step": 1400
    },
    {
      "entropy": 1.1544791884720325,
      "epoch": 0.43258637776259706,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9367540806687894e-05,
      "loss": 1.15,
      "mean_token_accuracy": 0.6929264325648546,
      "num_tokens": 105333438.0,
      "step": 1405
    },
    {
      "entropy": 1.1271388109773397,
      "epoch": 0.4341258310642433,
      "grad_norm": 0.796875,
      "learning_rate": 1.9361860894604116e-05,
      "loss": 1.1161,
      "mean_token_accuracy": 0.6955098442733287,
      "num_tokens": 105691785.0,
      "step": 1410
    },
    {
      "entropy": 1.1479596089571715,
      "epoch": 0.4356652843658896,
      "grad_norm": 0.69140625,
      "learning_rate": 1.9356156432131533e-05,
      "loss": 1.1482,
      "mean_token_accuracy": 0.6920965306460858,
      "num_tokens": 106078624.0,
      "step": 1415
    },
    {
      "entropy": 1.1149195641279221,
      "epoch": 0.4372047376675358,
      "grad_norm": 0.72265625,
      "learning_rate": 1.935042743422944e-05,
      "loss": 1.1028,
      "mean_token_accuracy": 0.6970803182572126,
      "num_tokens": 106454246.0,
      "step": 1420
    },
    {
      "entropy": 1.0929339196532966,
      "epoch": 0.43874419096918205,
      "grad_norm": 0.75390625,
      "learning_rate": 1.934467391592146e-05,
      "loss": 1.1047,
      "mean_token_accuracy": 0.7016421973705291,
      "num_tokens": 106810481.0,
      "step": 1425
    },
    {
      "entropy": 1.0979965701699257,
      "epoch": 0.44028364427082833,
      "grad_norm": 0.765625,
      "learning_rate": 1.9338895892295527e-05,
      "loss": 1.0956,
      "mean_token_accuracy": 0.7029366530478001,
      "num_tokens": 107196667.0,
      "step": 1430
    },
    {
      "entropy": 1.1433403573930263,
      "epoch": 0.44182309757247457,
      "grad_norm": 0.78515625,
      "learning_rate": 1.9333093378503832e-05,
      "loss": 1.1375,
      "mean_token_accuracy": 0.6927531309425831,
      "num_tokens": 107575900.0,
      "step": 1435
    },
    {
      "entropy": 1.1472560165449976,
      "epoch": 0.44336255087412085,
      "grad_norm": 0.74609375,
      "learning_rate": 1.9327266389762787e-05,
      "loss": 1.1371,
      "mean_token_accuracy": 0.6947022203356028,
      "num_tokens": 107944294.0,
      "step": 1440
    },
    {
      "entropy": 1.101127756945789,
      "epoch": 0.4449020041757671,
      "grad_norm": 0.75390625,
      "learning_rate": 1.9321414941353006e-05,
      "loss": 1.0928,
      "mean_token_accuracy": 0.7010299023240805,
      "num_tokens": 108312798.0,
      "step": 1445
    },
    {
      "entropy": 1.1180545002222062,
      "epoch": 0.4464414574774133,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9315539048619212e-05,
      "loss": 1.1234,
      "mean_token_accuracy": 0.6960787653923035,
      "num_tokens": 108690533.0,
      "step": 1450
    },
    {
      "entropy": 1.1227679682895542,
      "epoch": 0.4479809107790596,
      "grad_norm": 0.76953125,
      "learning_rate": 1.930963872697026e-05,
      "loss": 1.1046,
      "mean_token_accuracy": 0.6982706602662802,
      "num_tokens": 109068398.0,
      "step": 1455
    },
    {
      "entropy": 1.107317195646465,
      "epoch": 0.44952036408070584,
      "grad_norm": 0.71875,
      "learning_rate": 1.9303713991879052e-05,
      "loss": 1.1001,
      "mean_token_accuracy": 0.6980501331388951,
      "num_tokens": 109446077.0,
      "step": 1460
    },
    {
      "entropy": 1.1386326614767313,
      "epoch": 0.45105981738235207,
      "grad_norm": 0.71875,
      "learning_rate": 1.9297764858882516e-05,
      "loss": 1.1239,
      "mean_token_accuracy": 0.6934384491294623,
      "num_tokens": 109826830.0,
      "step": 1465
    },
    {
      "entropy": 1.1245597016066313,
      "epoch": 0.45259927068399836,
      "grad_norm": 0.765625,
      "learning_rate": 1.9291791343581557e-05,
      "loss": 1.1149,
      "mean_token_accuracy": 0.6970586907118559,
      "num_tokens": 110206021.0,
      "step": 1470
    },
    {
      "entropy": 1.112450549006462,
      "epoch": 0.4541387239856446,
      "grad_norm": 0.7578125,
      "learning_rate": 1.928579346164103e-05,
      "loss": 1.1129,
      "mean_token_accuracy": 0.6991185300052166,
      "num_tokens": 110575825.0,
      "step": 1475
    },
    {
      "entropy": 1.1001921689137817,
      "epoch": 0.4556781772872909,
      "grad_norm": 0.734375,
      "learning_rate": 1.927977122878967e-05,
      "loss": 1.1028,
      "mean_token_accuracy": 0.7005803253501653,
      "num_tokens": 110933606.0,
      "step": 1480
    },
    {
      "entropy": 1.136134173721075,
      "epoch": 0.4572176305889371,
      "grad_norm": 0.69921875,
      "learning_rate": 1.9273724660820086e-05,
      "loss": 1.1169,
      "mean_token_accuracy": 0.6969247065484524,
      "num_tokens": 111297173.0,
      "step": 1485
    },
    {
      "entropy": 1.1197788801044226,
      "epoch": 0.45875708389058334,
      "grad_norm": 0.671875,
      "learning_rate": 1.9267653773588702e-05,
      "loss": 1.1111,
      "mean_token_accuracy": 0.6978486493229866,
      "num_tokens": 111679485.0,
      "step": 1490
    },
    {
      "entropy": 1.1003839764744043,
      "epoch": 0.4602965371922296,
      "grad_norm": 0.66796875,
      "learning_rate": 1.926155858301571e-05,
      "loss": 1.0893,
      "mean_token_accuracy": 0.7018561966717243,
      "num_tokens": 112067380.0,
      "step": 1495
    },
    {
      "entropy": 1.0989170737564564,
      "epoch": 0.46183599049387586,
      "grad_norm": 0.71484375,
      "learning_rate": 1.925543910508503e-05,
      "loss": 1.1127,
      "mean_token_accuracy": 0.699932586029172,
      "num_tokens": 112448405.0,
      "step": 1500
    },
    {
      "entropy": 1.0723091229796409,
      "epoch": 0.4633754437955221,
      "grad_norm": 0.66796875,
      "learning_rate": 1.9249295355844286e-05,
      "loss": 1.058,
      "mean_token_accuracy": 0.7080515351146459,
      "num_tokens": 112821434.0,
      "step": 1505
    },
    {
      "entropy": 1.1173736985772849,
      "epoch": 0.4649148970971684,
      "grad_norm": 0.71484375,
      "learning_rate": 1.9243127351404743e-05,
      "loss": 1.1042,
      "mean_token_accuracy": 0.6990774724632501,
      "num_tokens": 113207766.0,
      "step": 1510
    },
    {
      "entropy": 1.136980064958334,
      "epoch": 0.4664543503988146,
      "grad_norm": 0.765625,
      "learning_rate": 1.9236935107941272e-05,
      "loss": 1.1307,
      "mean_token_accuracy": 0.6950142856687307,
      "num_tokens": 113564658.0,
      "step": 1515
    },
    {
      "entropy": 1.1760700676590203,
      "epoch": 0.4679938037004609,
      "grad_norm": 0.79296875,
      "learning_rate": 1.923071864169231e-05,
      "loss": 1.1714,
      "mean_token_accuracy": 0.6890874560922384,
      "num_tokens": 113935260.0,
      "step": 1520
    },
    {
      "entropy": 1.1633779585361481,
      "epoch": 0.46953325700210713,
      "grad_norm": 0.72265625,
      "learning_rate": 1.922447796895982e-05,
      "loss": 1.1552,
      "mean_token_accuracy": 0.6897667136043311,
      "num_tokens": 114300528.0,
      "step": 1525
    },
    {
      "entropy": 1.096606163494289,
      "epoch": 0.47107271030375336,
      "grad_norm": 0.6796875,
      "learning_rate": 1.9218213106109234e-05,
      "loss": 1.0923,
      "mean_token_accuracy": 0.7029213454574347,
      "num_tokens": 114670533.0,
      "step": 1530
    },
    {
      "entropy": 1.0995285920798779,
      "epoch": 0.47261216360539965,
      "grad_norm": 0.7265625,
      "learning_rate": 1.9211924069569422e-05,
      "loss": 1.1045,
      "mean_token_accuracy": 0.7023968610912561,
      "num_tokens": 115033065.0,
      "step": 1535
    },
    {
      "entropy": 1.1373300217092037,
      "epoch": 0.4741516169070459,
      "grad_norm": 0.734375,
      "learning_rate": 1.920561087583265e-05,
      "loss": 1.1375,
      "mean_token_accuracy": 0.6914490062743426,
      "num_tokens": 115395504.0,
      "step": 1540
    },
    {
      "entropy": 1.1257287595421075,
      "epoch": 0.4756910702086921,
      "grad_norm": 0.74609375,
      "learning_rate": 1.919927354145454e-05,
      "loss": 1.1123,
      "mean_token_accuracy": 0.6976802740246058,
      "num_tokens": 115768072.0,
      "step": 1545
    },
    {
      "entropy": 1.1214987369254232,
      "epoch": 0.4772305235103384,
      "grad_norm": 0.73046875,
      "learning_rate": 1.9192912083054003e-05,
      "loss": 1.1191,
      "mean_token_accuracy": 0.6951532650738954,
      "num_tokens": 116143853.0,
      "step": 1550
    },
    {
      "entropy": 1.1636530596762895,
      "epoch": 0.47876997681198463,
      "grad_norm": 0.7421875,
      "learning_rate": 1.9186526517313227e-05,
      "loss": 1.1578,
      "mean_token_accuracy": 0.6905960509553551,
      "num_tokens": 116501275.0,
      "step": 1555
    },
    {
      "entropy": 1.1044294204562903,
      "epoch": 0.4803094301136309,
      "grad_norm": 0.7109375,
      "learning_rate": 1.9180116860977613e-05,
      "loss": 1.0837,
      "mean_token_accuracy": 0.703742691129446,
      "num_tokens": 116885903.0,
      "step": 1560
    },
    {
      "entropy": 1.1139429537579417,
      "epoch": 0.48184888341527715,
      "grad_norm": 0.6953125,
      "learning_rate": 1.9173683130855737e-05,
      "loss": 1.104,
      "mean_token_accuracy": 0.6986370030790567,
      "num_tokens": 117262674.0,
      "step": 1565
    },
    {
      "entropy": 1.0733769409358502,
      "epoch": 0.4833883367169234,
      "grad_norm": 0.70703125,
      "learning_rate": 1.916722534381931e-05,
      "loss": 1.0699,
      "mean_token_accuracy": 0.7072780448943377,
      "num_tokens": 117633261.0,
      "step": 1570
    },
    {
      "entropy": 1.0940893176943063,
      "epoch": 0.48492779001856967,
      "grad_norm": 0.69921875,
      "learning_rate": 1.916074351680312e-05,
      "loss": 1.0899,
      "mean_token_accuracy": 0.7040331065654755,
      "num_tokens": 117994519.0,
      "step": 1575
    },
    {
      "entropy": 1.1256133463233708,
      "epoch": 0.4864672433202159,
      "grad_norm": 0.70703125,
      "learning_rate": 1.9154237666805005e-05,
      "loss": 1.1178,
      "mean_token_accuracy": 0.6980132691562175,
      "num_tokens": 118357854.0,
      "step": 1580
    },
    {
      "entropy": 1.1461247742176055,
      "epoch": 0.48800669662186213,
      "grad_norm": 0.7421875,
      "learning_rate": 1.9147707810885798e-05,
      "loss": 1.1273,
      "mean_token_accuracy": 0.6937376245856285,
      "num_tokens": 118727990.0,
      "step": 1585
    },
    {
      "entropy": 1.105697209574282,
      "epoch": 0.4895461499235084,
      "grad_norm": 0.78125,
      "learning_rate": 1.9141153966169287e-05,
      "loss": 1.0988,
      "mean_token_accuracy": 0.6974029045552015,
      "num_tokens": 119104452.0,
      "step": 1590
    },
    {
      "entropy": 1.0973486991599202,
      "epoch": 0.49108560322515465,
      "grad_norm": 0.72265625,
      "learning_rate": 1.9134576149842164e-05,
      "loss": 1.0897,
      "mean_token_accuracy": 0.7024951457977295,
      "num_tokens": 119492541.0,
      "step": 1595
    },
    {
      "entropy": 1.1395431403070688,
      "epoch": 0.49262505652680094,
      "grad_norm": 0.71484375,
      "learning_rate": 1.9127974379153983e-05,
      "loss": 1.1284,
      "mean_token_accuracy": 0.6936916135251522,
      "num_tokens": 119861950.0,
      "step": 1600
    },
    {
      "entropy": 1.119615512341261,
      "epoch": 0.4941645098284472,
      "grad_norm": 0.73828125,
      "learning_rate": 1.912134867141712e-05,
      "loss": 1.1297,
      "mean_token_accuracy": 0.6973624024540186,
      "num_tokens": 120235402.0,
      "step": 1605
    },
    {
      "entropy": 1.127389458194375,
      "epoch": 0.4957039631300934,
      "grad_norm": 0.76171875,
      "learning_rate": 1.9114699044006725e-05,
      "loss": 1.1246,
      "mean_token_accuracy": 0.6948788855224848,
      "num_tokens": 120616115.0,
      "step": 1610
    },
    {
      "entropy": 1.1248438712209463,
      "epoch": 0.4972434164317397,
      "grad_norm": 0.71875,
      "learning_rate": 1.910802551436066e-05,
      "loss": 1.1298,
      "mean_token_accuracy": 0.6980609927326441,
      "num_tokens": 120995993.0,
      "step": 1615
    },
    {
      "entropy": 1.0943362485617398,
      "epoch": 0.4987828697333859,
      "grad_norm": 0.765625,
      "learning_rate": 1.9101328099979496e-05,
      "loss": 1.0998,
      "mean_token_accuracy": 0.7005326244980097,
      "num_tokens": 121357538.0,
      "step": 1620
    },
    {
      "entropy": 1.158146957680583,
      "epoch": 0.5003223230350322,
      "grad_norm": 0.7421875,
      "learning_rate": 1.9094606818426403e-05,
      "loss": 1.1539,
      "mean_token_accuracy": 0.6923132240772247,
      "num_tokens": 121710013.0,
      "step": 1625
    },
    {
      "entropy": 1.1276648858562113,
      "epoch": 0.5018617763366784,
      "grad_norm": 0.6796875,
      "learning_rate": 1.908786168732717e-05,
      "loss": 1.1182,
      "mean_token_accuracy": 0.6962565090507269,
      "num_tokens": 122085828.0,
      "step": 1630
    },
    {
      "entropy": 1.0802762266248465,
      "epoch": 0.5034012296383247,
      "grad_norm": 0.69921875,
      "learning_rate": 1.9081092724370112e-05,
      "loss": 1.0793,
      "mean_token_accuracy": 0.7047099947929383,
      "num_tokens": 122464366.0,
      "step": 1635
    },
    {
      "entropy": 1.136762074381113,
      "epoch": 0.504940682939971,
      "grad_norm": 0.68359375,
      "learning_rate": 1.907429994730605e-05,
      "loss": 1.1273,
      "mean_token_accuracy": 0.6950546491891145,
      "num_tokens": 122835610.0,
      "step": 1640
    },
    {
      "entropy": 1.0965771291404962,
      "epoch": 0.5064801362416171,
      "grad_norm": 0.73828125,
      "learning_rate": 1.9067483373948245e-05,
      "loss": 1.0801,
      "mean_token_accuracy": 0.7031420111656189,
      "num_tokens": 123198090.0,
      "step": 1645
    },
    {
      "entropy": 1.1391410026699305,
      "epoch": 0.5080195895432634,
      "grad_norm": 0.7421875,
      "learning_rate": 1.9060643022172364e-05,
      "loss": 1.138,
      "mean_token_accuracy": 0.6941679731011391,
      "num_tokens": 123574849.0,
      "step": 1650
    },
    {
      "entropy": 1.1137908024713397,
      "epoch": 0.5095590428449097,
      "grad_norm": 0.77734375,
      "learning_rate": 1.905377890991644e-05,
      "loss": 1.1011,
      "mean_token_accuracy": 0.7002455405890942,
      "num_tokens": 123958243.0,
      "step": 1655
    },
    {
      "entropy": 1.0767975924536586,
      "epoch": 0.511098496146556,
      "grad_norm": 0.65625,
      "learning_rate": 1.90468910551808e-05,
      "loss": 1.0628,
      "mean_token_accuracy": 0.7064414627850055,
      "num_tokens": 124354285.0,
      "step": 1660
    },
    {
      "entropy": 1.1056279137730598,
      "epoch": 0.5126379494482022,
      "grad_norm": 0.76953125,
      "learning_rate": 1.9039979476028044e-05,
      "loss": 1.1108,
      "mean_token_accuracy": 0.7005944430828095,
      "num_tokens": 124716477.0,
      "step": 1665
    },
    {
      "entropy": 1.121751406788826,
      "epoch": 0.5141774027498485,
      "grad_norm": 0.7734375,
      "learning_rate": 1.903304419058298e-05,
      "loss": 1.1054,
      "mean_token_accuracy": 0.6988046038895845,
      "num_tokens": 125084153.0,
      "step": 1670
    },
    {
      "entropy": 1.1113942619413137,
      "epoch": 0.5157168560514948,
      "grad_norm": 0.73828125,
      "learning_rate": 1.9026085217032592e-05,
      "loss": 1.1038,
      "mean_token_accuracy": 0.6969317603856325,
      "num_tokens": 125454921.0,
      "step": 1675
    },
    {
      "entropy": 1.1110334230586887,
      "epoch": 0.5172563093531409,
      "grad_norm": 0.78125,
      "learning_rate": 1.9019102573625966e-05,
      "loss": 1.112,
      "mean_token_accuracy": 0.6979748774319887,
      "num_tokens": 125814552.0,
      "step": 1680
    },
    {
      "entropy": 1.115121967718005,
      "epoch": 0.5187957626547872,
      "grad_norm": 0.74609375,
      "learning_rate": 1.9012096278674283e-05,
      "loss": 1.1022,
      "mean_token_accuracy": 0.6978176638484002,
      "num_tokens": 126180100.0,
      "step": 1685
    },
    {
      "entropy": 1.1214980090036988,
      "epoch": 0.5203352159564335,
      "grad_norm": 0.75,
      "learning_rate": 1.9005066350550724e-05,
      "loss": 1.1107,
      "mean_token_accuracy": 0.6994861856102943,
      "num_tokens": 126544212.0,
      "step": 1690
    },
    {
      "entropy": 1.1030220981687306,
      "epoch": 0.5218746692580797,
      "grad_norm": 0.7109375,
      "learning_rate": 1.899801280769046e-05,
      "loss": 1.1027,
      "mean_token_accuracy": 0.7016896925866604,
      "num_tokens": 126936632.0,
      "step": 1695
    },
    {
      "entropy": 1.0752276331186295,
      "epoch": 0.523414122559726,
      "grad_norm": 0.73828125,
      "learning_rate": 1.8990935668590583e-05,
      "loss": 1.0732,
      "mean_token_accuracy": 0.706041594222188,
      "num_tokens": 127334816.0,
      "step": 1700
    },
    {
      "entropy": 1.137972903251648,
      "epoch": 0.5249535758613723,
      "grad_norm": 0.7421875,
      "learning_rate": 1.8983834951810068e-05,
      "loss": 1.1276,
      "mean_token_accuracy": 0.6936356756836176,
      "num_tokens": 127710399.0,
      "step": 1705
    },
    {
      "entropy": 1.1167596075683832,
      "epoch": 0.5264930291630184,
      "grad_norm": 0.67578125,
      "learning_rate": 1.8976710675969715e-05,
      "loss": 1.1118,
      "mean_token_accuracy": 0.6976578302681447,
      "num_tokens": 128099437.0,
      "step": 1710
    },
    {
      "entropy": 1.0992528446018697,
      "epoch": 0.5280324824646647,
      "grad_norm": 0.765625,
      "learning_rate": 1.896956285975211e-05,
      "loss": 1.0875,
      "mean_token_accuracy": 0.7028979767113924,
      "num_tokens": 128467478.0,
      "step": 1715
    },
    {
      "entropy": 1.0609646894037723,
      "epoch": 0.529571935766311,
      "grad_norm": 0.71875,
      "learning_rate": 1.8962391521901565e-05,
      "loss": 1.0554,
      "mean_token_accuracy": 0.7095225866883993,
      "num_tokens": 128838188.0,
      "step": 1720
    },
    {
      "entropy": 1.1092106204479932,
      "epoch": 0.5311113890679573,
      "grad_norm": 0.7109375,
      "learning_rate": 1.895519668122408e-05,
      "loss": 1.0967,
      "mean_token_accuracy": 0.6991548746824264,
      "num_tokens": 129214099.0,
      "step": 1725
    },
    {
      "entropy": 1.0980365563184022,
      "epoch": 0.5326508423696035,
      "grad_norm": 0.6796875,
      "learning_rate": 1.8947978356587283e-05,
      "loss": 1.0886,
      "mean_token_accuracy": 0.702478701993823,
      "num_tokens": 129593189.0,
      "step": 1730
    },
    {
      "entropy": 1.1081396404653787,
      "epoch": 0.5341902956712498,
      "grad_norm": 0.703125,
      "learning_rate": 1.8940736566920386e-05,
      "loss": 1.0974,
      "mean_token_accuracy": 0.7000274512916803,
      "num_tokens": 129972696.0,
      "step": 1735
    },
    {
      "entropy": 1.0982466185465456,
      "epoch": 0.535729748972896,
      "grad_norm": 0.69921875,
      "learning_rate": 1.893347133121415e-05,
      "loss": 1.0822,
      "mean_token_accuracy": 0.7039383064955473,
      "num_tokens": 130350766.0,
      "step": 1740
    },
    {
      "entropy": 1.1378747086971999,
      "epoch": 0.5372692022745422,
      "grad_norm": 0.71484375,
      "learning_rate": 1.8926182668520794e-05,
      "loss": 1.1298,
      "mean_token_accuracy": 0.6929653998464346,
      "num_tokens": 130712736.0,
      "step": 1745
    },
    {
      "entropy": 1.1180994376540183,
      "epoch": 0.5388086555761885,
      "grad_norm": 0.6640625,
      "learning_rate": 1.8918870597953996e-05,
      "loss": 1.1166,
      "mean_token_accuracy": 0.697513160482049,
      "num_tokens": 131108827.0,
      "step": 1750
    },
    {
      "entropy": 1.1118130307644605,
      "epoch": 0.5403481088778348,
      "grad_norm": 0.7578125,
      "learning_rate": 1.89115351386888e-05,
      "loss": 1.1135,
      "mean_token_accuracy": 0.7016211155802011,
      "num_tokens": 131484362.0,
      "step": 1755
    },
    {
      "entropy": 1.154317499510944,
      "epoch": 0.541887562179481,
      "grad_norm": 0.74609375,
      "learning_rate": 1.8904176309961606e-05,
      "loss": 1.1623,
      "mean_token_accuracy": 0.6900074496865273,
      "num_tokens": 131854979.0,
      "step": 1760
    },
    {
      "entropy": 1.1418317448347808,
      "epoch": 0.5434270154811273,
      "grad_norm": 0.7578125,
      "learning_rate": 1.8896794131070073e-05,
      "loss": 1.1409,
      "mean_token_accuracy": 0.6961004845798016,
      "num_tokens": 132231075.0,
      "step": 1765
    },
    {
      "entropy": 1.1063750840723514,
      "epoch": 0.5449664687827735,
      "grad_norm": 0.72265625,
      "learning_rate": 1.8889388621373107e-05,
      "loss": 1.094,
      "mean_token_accuracy": 0.6988186117261648,
      "num_tokens": 132614827.0,
      "step": 1770
    },
    {
      "entropy": 1.0908017337322236,
      "epoch": 0.5465059220844197,
      "grad_norm": 0.6796875,
      "learning_rate": 1.88819598002908e-05,
      "loss": 1.0762,
      "mean_token_accuracy": 0.7036438055336476,
      "num_tokens": 133006606.0,
      "step": 1775
    },
    {
      "entropy": 1.1063918098807335,
      "epoch": 0.548045375386066,
      "grad_norm": 0.7109375,
      "learning_rate": 1.887450768730436e-05,
      "loss": 1.0865,
      "mean_token_accuracy": 0.7016949482262135,
      "num_tokens": 133391329.0,
      "step": 1780
    },
    {
      "entropy": 1.111767608858645,
      "epoch": 0.5495848286877123,
      "grad_norm": 0.70703125,
      "learning_rate": 1.886703230195609e-05,
      "loss": 1.1104,
      "mean_token_accuracy": 0.7000030245631933,
      "num_tokens": 133783664.0,
      "step": 1785
    },
    {
      "entropy": 1.1513528198003768,
      "epoch": 0.5511242819893585,
      "grad_norm": 0.7265625,
      "learning_rate": 1.8859533663849318e-05,
      "loss": 1.1441,
      "mean_token_accuracy": 0.6949969541281462,
      "num_tokens": 134154971.0,
      "step": 1790
    },
    {
      "entropy": 1.109261892735958,
      "epoch": 0.5526637352910048,
      "grad_norm": 0.6875,
      "learning_rate": 1.885201179264834e-05,
      "loss": 1.1072,
      "mean_token_accuracy": 0.7005063198506832,
      "num_tokens": 134541521.0,
      "step": 1795
    },
    {
      "entropy": 1.1210583060979844,
      "epoch": 0.554203188592651,
      "grad_norm": 0.73828125,
      "learning_rate": 1.8844466708078398e-05,
      "loss": 1.1179,
      "mean_token_accuracy": 0.6973602317273617,
      "num_tokens": 134909569.0,
      "step": 1800
    },
    {
      "entropy": 1.1100708212703467,
      "epoch": 0.5557426418942973,
      "grad_norm": 0.71875,
      "learning_rate": 1.8836898429925586e-05,
      "loss": 1.09,
      "mean_token_accuracy": 0.7016674302518368,
      "num_tokens": 135278194.0,
      "step": 1805
    },
    {
      "entropy": 1.133229278959334,
      "epoch": 0.5572820951959435,
      "grad_norm": 0.75390625,
      "learning_rate": 1.8829306978036837e-05,
      "loss": 1.1279,
      "mean_token_accuracy": 0.6966449148952961,
      "num_tokens": 135644126.0,
      "step": 1810
    },
    {
      "entropy": 1.094052490964532,
      "epoch": 0.5588215484975898,
      "grad_norm": 0.68359375,
      "learning_rate": 1.8821692372319852e-05,
      "loss": 1.0881,
      "mean_token_accuracy": 0.7018960192799568,
      "num_tokens": 136040093.0,
      "step": 1815
    },
    {
      "entropy": 1.102319024130702,
      "epoch": 0.5603610017992361,
      "grad_norm": 0.69140625,
      "learning_rate": 1.8814054632743038e-05,
      "loss": 1.1009,
      "mean_token_accuracy": 0.7015113096684218,
      "num_tokens": 136421328.0,
      "step": 1820
    },
    {
      "entropy": 1.1264159340411424,
      "epoch": 0.5619004551008823,
      "grad_norm": 0.68359375,
      "learning_rate": 1.8806393779335483e-05,
      "loss": 1.1106,
      "mean_token_accuracy": 0.6967556666582823,
      "num_tokens": 136796837.0,
      "step": 1825
    },
    {
      "entropy": 1.1152265276759863,
      "epoch": 0.5634399084025286,
      "grad_norm": 0.6796875,
      "learning_rate": 1.879870983218688e-05,
      "loss": 1.0994,
      "mean_token_accuracy": 0.6976462483406067,
      "num_tokens": 137174396.0,
      "step": 1830
    },
    {
      "entropy": 1.116674688272178,
      "epoch": 0.5649793617041748,
      "grad_norm": 0.71484375,
      "learning_rate": 1.8791002811447483e-05,
      "loss": 1.1104,
      "mean_token_accuracy": 0.7000731214880943,
      "num_tokens": 137556532.0,
      "step": 1835
    },
    {
      "entropy": 1.1110275972634553,
      "epoch": 0.566518815005821,
      "grad_norm": 0.70703125,
      "learning_rate": 1.878327273732806e-05,
      "loss": 1.1098,
      "mean_token_accuracy": 0.698979677259922,
      "num_tokens": 137949242.0,
      "step": 1840
    },
    {
      "entropy": 1.0584419948980213,
      "epoch": 0.5680582683074673,
      "grad_norm": 0.703125,
      "learning_rate": 1.8775519630099822e-05,
      "loss": 1.0544,
      "mean_token_accuracy": 0.7107167650014162,
      "num_tokens": 138325463.0,
      "step": 1845
    },
    {
      "entropy": 1.120271611586213,
      "epoch": 0.5695977216091136,
      "grad_norm": 0.76171875,
      "learning_rate": 1.8767743510094395e-05,
      "loss": 1.1202,
      "mean_token_accuracy": 0.6949244394898415,
      "num_tokens": 138692189.0,
      "step": 1850
    },
    {
      "entropy": 1.1016495576128364,
      "epoch": 0.5711371749107598,
      "grad_norm": 0.7578125,
      "learning_rate": 1.8759944397703748e-05,
      "loss": 1.1078,
      "mean_token_accuracy": 0.7005269210785627,
      "num_tokens": 139066457.0,
      "step": 1855
    },
    {
      "entropy": 1.1251827346161007,
      "epoch": 0.5726766282124061,
      "grad_norm": 0.71484375,
      "learning_rate": 1.8752122313380137e-05,
      "loss": 1.1007,
      "mean_token_accuracy": 0.6988253463059664,
      "num_tokens": 139426172.0,
      "step": 1860
    },
    {
      "entropy": 1.1180114712566138,
      "epoch": 0.5742160815140523,
      "grad_norm": 0.68359375,
      "learning_rate": 1.874427727763607e-05,
      "loss": 1.1112,
      "mean_token_accuracy": 0.69961117208004,
      "num_tokens": 139801332.0,
      "step": 1865
    },
    {
      "entropy": 1.129542938247323,
      "epoch": 0.5757555348156985,
      "grad_norm": 0.734375,
      "learning_rate": 1.8736409311044244e-05,
      "loss": 1.1289,
      "mean_token_accuracy": 0.695055254548788,
      "num_tokens": 140171056.0,
      "step": 1870
    },
    {
      "entropy": 1.0734907962381839,
      "epoch": 0.5772949881173448,
      "grad_norm": 0.6796875,
      "learning_rate": 1.8728518434237476e-05,
      "loss": 1.0626,
      "mean_token_accuracy": 0.7072399683296681,
      "num_tokens": 140566923.0,
      "step": 1875
    },
    {
      "entropy": 1.1328235883265734,
      "epoch": 0.5788344414189911,
      "grad_norm": 0.74609375,
      "learning_rate": 1.8720604667908673e-05,
      "loss": 1.1338,
      "mean_token_accuracy": 0.6954614106565714,
      "num_tokens": 140929032.0,
      "step": 1880
    },
    {
      "entropy": 1.1626869395375252,
      "epoch": 0.5803738947206374,
      "grad_norm": 0.6875,
      "learning_rate": 1.8712668032810767e-05,
      "loss": 1.153,
      "mean_token_accuracy": 0.6904126744717359,
      "num_tokens": 141306794.0,
      "step": 1885
    },
    {
      "entropy": 1.0925796177238225,
      "epoch": 0.5819133480222836,
      "grad_norm": 0.703125,
      "learning_rate": 1.8704708549756657e-05,
      "loss": 1.0896,
      "mean_token_accuracy": 0.7039735574275255,
      "num_tokens": 141679922.0,
      "step": 1890
    },
    {
      "entropy": 1.1325601134449244,
      "epoch": 0.5834528013239298,
      "grad_norm": 0.73828125,
      "learning_rate": 1.869672623961916e-05,
      "loss": 1.1236,
      "mean_token_accuracy": 0.695359307155013,
      "num_tokens": 142064042.0,
      "step": 1895
    },
    {
      "entropy": 1.1089297072961926,
      "epoch": 0.5849922546255761,
      "grad_norm": 0.73828125,
      "learning_rate": 1.8688721123330952e-05,
      "loss": 1.0961,
      "mean_token_accuracy": 0.7020201427862048,
      "num_tokens": 142450974.0,
      "step": 1900
    },
    {
      "entropy": 1.0782432477921247,
      "epoch": 0.5865317079272223,
      "grad_norm": 0.73828125,
      "learning_rate": 1.868069322188452e-05,
      "loss": 1.077,
      "mean_token_accuracy": 0.7056647684425116,
      "num_tokens": 142833583.0,
      "step": 1905
    },
    {
      "entropy": 1.090103941783309,
      "epoch": 0.5880711612288686,
      "grad_norm": 0.67578125,
      "learning_rate": 1.8672642556332093e-05,
      "loss": 1.0815,
      "mean_token_accuracy": 0.7041718065738678,
      "num_tokens": 143202881.0,
      "step": 1910
    },
    {
      "entropy": 1.1132526526227593,
      "epoch": 0.5896106145305149,
      "grad_norm": 0.7578125,
      "learning_rate": 1.8664569147785615e-05,
      "loss": 1.1143,
      "mean_token_accuracy": 0.7002504725009203,
      "num_tokens": 143575381.0,
      "step": 1915
    },
    {
      "entropy": 1.0923281263560056,
      "epoch": 0.5911500678321611,
      "grad_norm": 0.6953125,
      "learning_rate": 1.8656473017416644e-05,
      "loss": 1.0894,
      "mean_token_accuracy": 0.7025896862149239,
      "num_tokens": 143945584.0,
      "step": 1920
    },
    {
      "entropy": 1.1606475939974188,
      "epoch": 0.5926895211338074,
      "grad_norm": 0.6875,
      "learning_rate": 1.864835418645635e-05,
      "loss": 1.1518,
      "mean_token_accuracy": 0.69000857565552,
      "num_tokens": 144327615.0,
      "step": 1925
    },
    {
      "entropy": 1.076964562945068,
      "epoch": 0.5942289744354536,
      "grad_norm": 0.72265625,
      "learning_rate": 1.8640212676195415e-05,
      "loss": 1.0674,
      "mean_token_accuracy": 0.705032504722476,
      "num_tokens": 144702165.0,
      "step": 1930
    },
    {
      "entropy": 1.1039565896615386,
      "epoch": 0.5957684277370998,
      "grad_norm": 0.703125,
      "learning_rate": 1.8632048507984e-05,
      "loss": 1.0888,
      "mean_token_accuracy": 0.7024720892310142,
      "num_tokens": 145076417.0,
      "step": 1935
    },
    {
      "entropy": 1.093964084982872,
      "epoch": 0.5973078810387461,
      "grad_norm": 0.75,
      "learning_rate": 1.862386170323169e-05,
      "loss": 1.081,
      "mean_token_accuracy": 0.7047437366098166,
      "num_tokens": 145455366.0,
      "step": 1940
    },
    {
      "entropy": 1.1448205880820752,
      "epoch": 0.5988473343403924,
      "grad_norm": 0.6953125,
      "learning_rate": 1.861565228340742e-05,
      "loss": 1.1458,
      "mean_token_accuracy": 0.6915356520563364,
      "num_tokens": 145838812.0,
      "step": 1945
    },
    {
      "entropy": 1.1012870259582996,
      "epoch": 0.6003867876420386,
      "grad_norm": 0.71484375,
      "learning_rate": 1.860742027003944e-05,
      "loss": 1.0907,
      "mean_token_accuracy": 0.7037778791040182,
      "num_tokens": 146208431.0,
      "step": 1950
    },
    {
      "entropy": 1.1052306432276964,
      "epoch": 0.6019262409436849,
      "grad_norm": 0.7421875,
      "learning_rate": 1.8599165684715238e-05,
      "loss": 1.0995,
      "mean_token_accuracy": 0.7021346285939216,
      "num_tokens": 146587958.0,
      "step": 1955
    },
    {
      "entropy": 1.0908923922106624,
      "epoch": 0.6034656942453311,
      "grad_norm": 0.74609375,
      "learning_rate": 1.8590888549081514e-05,
      "loss": 1.0687,
      "mean_token_accuracy": 0.7040816470980644,
      "num_tokens": 146956196.0,
      "step": 1960
    },
    {
      "entropy": 1.1141000390052795,
      "epoch": 0.6050051475469774,
      "grad_norm": 0.74609375,
      "learning_rate": 1.8582588884844086e-05,
      "loss": 1.1103,
      "mean_token_accuracy": 0.6970112754032016,
      "num_tokens": 147320054.0,
      "step": 1965
    },
    {
      "entropy": 1.0900843566283585,
      "epoch": 0.6065446008486236,
      "grad_norm": 0.7265625,
      "learning_rate": 1.857426671376785e-05,
      "loss": 1.0764,
      "mean_token_accuracy": 0.7049407433718443,
      "num_tokens": 147703735.0,
      "step": 1970
    },
    {
      "entropy": 1.1240591725334526,
      "epoch": 0.6080840541502699,
      "grad_norm": 0.734375,
      "learning_rate": 1.8565922057676738e-05,
      "loss": 1.1196,
      "mean_token_accuracy": 0.6968345880508423,
      "num_tokens": 148073305.0,
      "step": 1975
    },
    {
      "entropy": 1.1471635062247514,
      "epoch": 0.6096235074519162,
      "grad_norm": 0.74609375,
      "learning_rate": 1.855755493845363e-05,
      "loss": 1.1427,
      "mean_token_accuracy": 0.6943592650815844,
      "num_tokens": 148444038.0,
      "step": 1980
    },
    {
      "entropy": 1.132018794864416,
      "epoch": 0.6111629607535624,
      "grad_norm": 0.74609375,
      "learning_rate": 1.8549165378040328e-05,
      "loss": 1.1394,
      "mean_token_accuracy": 0.6943659249693155,
      "num_tokens": 148822233.0,
      "step": 1985
    },
    {
      "entropy": 1.0830021239817142,
      "epoch": 0.6127024140552086,
      "grad_norm": 0.69140625,
      "learning_rate": 1.8540753398437473e-05,
      "loss": 1.0754,
      "mean_token_accuracy": 0.7037800218909979,
      "num_tokens": 149211582.0,
      "step": 1990
    },
    {
      "entropy": 1.1431629927828908,
      "epoch": 0.6142418673568549,
      "grad_norm": 0.73046875,
      "learning_rate": 1.8532319021704502e-05,
      "loss": 1.1372,
      "mean_token_accuracy": 0.6933602422475815,
      "num_tokens": 149586556.0,
      "step": 1995
    },
    {
      "entropy": 1.1282612700015306,
      "epoch": 0.6157813206585011,
      "grad_norm": 0.7734375,
      "learning_rate": 1.852386226995958e-05,
      "loss": 1.1324,
      "mean_token_accuracy": 0.6929793257266283,
      "num_tokens": 149951555.0,
      "step": 2000
    }
  ],
  "logging_steps": 5,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.49648984129903e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}