{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.697755785475084,
  "eval_steps": 3000,
  "global_step": 57000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.791650009155274,
      "epoch": 0.00049982506122857,
      "grad_norm": 11.0,
      "learning_rate": 2e-06,
      "loss": 14.2639,
      "mean_token_accuracy": 0.0,
      "num_tokens": 8821.0,
      "step": 5
    },
    {
      "entropy": 4.82121729850769,
      "epoch": 0.00099965012245714,
      "grad_norm": 13.4375,
      "learning_rate": 4.5e-06,
      "loss": 14.2929,
      "mean_token_accuracy": 0.0,
      "num_tokens": 18288.0,
      "step": 10
    },
    {
      "entropy": 4.837881422042846,
      "epoch": 0.00149947518368571,
      "grad_norm": 16.5,
      "learning_rate": 7e-06,
      "loss": 14.1106,
      "mean_token_accuracy": 0.0001669449033215642,
      "num_tokens": 28664.0,
      "step": 15
    },
    {
      "entropy": 5.054859638214111,
      "epoch": 0.00199930024491428,
      "grad_norm": 25.125,
      "learning_rate": 9.5e-06,
      "loss": 13.5211,
      "mean_token_accuracy": 0.0,
      "num_tokens": 37807.0,
      "step": 20
    },
    {
      "entropy": 6.8463219165802,
      "epoch": 0.00249912530614285,
      "grad_norm": 21.375,
      "learning_rate": 1.2e-05,
      "loss": 12.1102,
      "mean_token_accuracy": 7.418397581204772e-05,
      "num_tokens": 48086.0,
      "step": 25
    },
    {
      "entropy": 10.3131178855896,
      "epoch": 0.00299895036737142,
      "grad_norm": 3.5,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.84,
      "mean_token_accuracy": 0.0007628825958818197,
      "num_tokens": 57427.0,
      "step": 30
    },
    {
      "entropy": 10.695811748504639,
      "epoch": 0.00349877542859999,
      "grad_norm": 3.125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.6279,
      "mean_token_accuracy": 0.013821808062493801,
      "num_tokens": 66662.0,
      "step": 35
    },
    {
      "entropy": 10.684320831298828,
      "epoch": 0.00399860048982856,
      "grad_norm": 2.3125,
      "learning_rate": 1.95e-05,
      "loss": 10.4081,
      "mean_token_accuracy": 0.020816447213292122,
      "num_tokens": 77388.0,
      "step": 40
    },
    {
      "entropy": 10.571028423309325,
      "epoch": 0.00449842555105713,
      "grad_norm": 1.984375,
      "learning_rate": 2.2e-05,
      "loss": 10.1091,
      "mean_token_accuracy": 0.028454202599823473,
      "num_tokens": 86093.0,
      "step": 45
    },
    {
      "entropy": 10.40514316558838,
      "epoch": 0.0049982506122857,
      "grad_norm": 1.4921875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.0138,
      "mean_token_accuracy": 0.04152352567762137,
      "num_tokens": 96097.0,
      "step": 50
    },
    {
      "entropy": 10.368546867370606,
      "epoch": 0.00549807567351427,
      "grad_norm": 1.5234375,
      "learning_rate": 2.7e-05,
      "loss": 9.9286,
      "mean_token_accuracy": 0.03463506633415818,
      "num_tokens": 106328.0,
      "step": 55
    },
    {
      "entropy": 10.427862358093261,
      "epoch": 0.00599790073474284,
      "grad_norm": 1.5625,
      "learning_rate": 2.95e-05,
      "loss": 9.7561,
      "mean_token_accuracy": 0.03965897653251886,
      "num_tokens": 114822.0,
      "step": 60
    },
    {
      "entropy": 10.424885177612305,
      "epoch": 0.00649772579597141,
      "grad_norm": 1.5859375,
      "learning_rate": 3.2e-05,
      "loss": 9.7727,
      "mean_token_accuracy": 0.04571409597992897,
      "num_tokens": 124984.0,
      "step": 65
    },
    {
      "entropy": 10.377210521697998,
      "epoch": 0.00699755085719998,
      "grad_norm": 1.3359375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.6547,
      "mean_token_accuracy": 0.042986732348799704,
      "num_tokens": 134054.0,
      "step": 70
    },
    {
      "entropy": 10.345882892608643,
      "epoch": 0.00749737591842855,
      "grad_norm": 1.2890625,
      "learning_rate": 3.7e-05,
      "loss": 9.6148,
      "mean_token_accuracy": 0.04352582097053528,
      "num_tokens": 144561.0,
      "step": 75
    },
    {
      "entropy": 10.32333002090454,
      "epoch": 0.00799720097965712,
      "grad_norm": 1.3671875,
      "learning_rate": 3.95e-05,
      "loss": 9.6092,
      "mean_token_accuracy": 0.034140726551413536,
      "num_tokens": 153701.0,
      "step": 80
    },
    {
      "entropy": 10.304623126983643,
      "epoch": 0.00849702604088569,
      "grad_norm": 1.390625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.463,
      "mean_token_accuracy": 0.0478254497051239,
      "num_tokens": 164291.0,
      "step": 85
    },
    {
      "entropy": 10.249341011047363,
      "epoch": 0.00899685110211426,
      "grad_norm": 1.3046875,
      "learning_rate": 4.45e-05,
      "loss": 9.4218,
      "mean_token_accuracy": 0.047897301986813544,
      "num_tokens": 174136.0,
      "step": 90
    },
    {
      "entropy": 10.243558979034423,
      "epoch": 0.00949667616334283,
      "grad_norm": 1.2109375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.3489,
      "mean_token_accuracy": 0.055490773916244504,
      "num_tokens": 184368.0,
      "step": 95
    },
    {
      "entropy": 10.187316131591796,
      "epoch": 0.0099965012245714,
      "grad_norm": 1.1484375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.3329,
      "mean_token_accuracy": 0.053723962232470515,
      "num_tokens": 194659.0,
      "step": 100
    },
    {
      "entropy": 10.174760437011718,
      "epoch": 0.01049632628579997,
      "grad_norm": 1.3671875,
      "learning_rate": 5.2e-05,
      "loss": 9.2164,
      "mean_token_accuracy": 0.05385222285985947,
      "num_tokens": 204875.0,
      "step": 105
    },
    {
      "entropy": 10.10236120223999,
      "epoch": 0.01099615134702854,
      "grad_norm": 1.65625,
      "learning_rate": 5.45e-05,
      "loss": 9.2078,
      "mean_token_accuracy": 0.05042871832847595,
      "num_tokens": 217033.0,
      "step": 110
    },
    {
      "entropy": 9.943318462371826,
      "epoch": 0.01149597640825711,
      "grad_norm": 1.078125,
      "learning_rate": 5.7e-05,
      "loss": 9.0442,
      "mean_token_accuracy": 0.059573306888341906,
      "num_tokens": 225946.0,
      "step": 115
    },
    {
      "entropy": 9.942243766784667,
      "epoch": 0.01199580146948568,
      "grad_norm": 1.21875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.9898,
      "mean_token_accuracy": 0.05044227540493011,
      "num_tokens": 235413.0,
      "step": 120
    },
    {
      "entropy": 9.90278606414795,
      "epoch": 0.01249562653071425,
      "grad_norm": 0.90234375,
      "learning_rate": 6.2e-05,
      "loss": 9.0153,
      "mean_token_accuracy": 0.049924901127815245,
      "num_tokens": 247913.0,
      "step": 125
    },
    {
      "entropy": 9.790752792358399,
      "epoch": 0.01299545159194282,
      "grad_norm": 0.99609375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.8984,
      "mean_token_accuracy": 0.055387107655406,
      "num_tokens": 258486.0,
      "step": 130
    },
    {
      "entropy": 9.678752231597901,
      "epoch": 0.01349527665317139,
      "grad_norm": 0.859375,
      "learning_rate": 6.7e-05,
      "loss": 8.8218,
      "mean_token_accuracy": 0.05492280051112175,
      "num_tokens": 268027.0,
      "step": 135
    },
    {
      "entropy": 9.582042121887207,
      "epoch": 0.01399510171439996,
      "grad_norm": 0.9375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.7463,
      "mean_token_accuracy": 0.05840916000306606,
      "num_tokens": 277470.0,
      "step": 140
    },
    {
      "entropy": 9.53208656311035,
      "epoch": 0.01449492677562853,
      "grad_norm": 0.9375,
      "learning_rate": 7.2e-05,
      "loss": 8.7007,
      "mean_token_accuracy": 0.05844452828168869,
      "num_tokens": 286891.0,
      "step": 145
    },
    {
      "entropy": 9.274843978881837,
      "epoch": 0.0149947518368571,
      "grad_norm": 0.94921875,
      "learning_rate": 7.45e-05,
      "loss": 8.6317,
      "mean_token_accuracy": 0.06578897573053837,
      "num_tokens": 295636.0,
      "step": 150
    },
    {
      "entropy": 9.232738018035889,
      "epoch": 0.01549457689808567,
      "grad_norm": 0.93359375,
      "learning_rate": 7.7e-05,
      "loss": 8.6046,
      "mean_token_accuracy": 0.06044535823166371,
      "num_tokens": 306542.0,
      "step": 155
    },
    {
      "entropy": 9.169378280639648,
      "epoch": 0.01599440195931424,
      "grad_norm": 1.015625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.5522,
      "mean_token_accuracy": 0.05850973688066006,
      "num_tokens": 315956.0,
      "step": 160
    },
    {
      "entropy": 9.077482509613038,
      "epoch": 0.01649422702054281,
      "grad_norm": 0.78125,
      "learning_rate": 8.2e-05,
      "loss": 8.556,
      "mean_token_accuracy": 0.05588611029088497,
      "num_tokens": 325500.0,
      "step": 165
    },
    {
      "entropy": 9.019215297698974,
      "epoch": 0.01699405208177138,
      "grad_norm": 1.109375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 8.5765,
      "mean_token_accuracy": 0.061797235161066055,
      "num_tokens": 335194.0,
      "step": 170
    },
    {
      "entropy": 8.91619348526001,
      "epoch": 0.01749387714299995,
      "grad_norm": 0.88671875,
      "learning_rate": 8.7e-05,
      "loss": 8.5278,
      "mean_token_accuracy": 0.06026073396205902,
      "num_tokens": 344717.0,
      "step": 175
    },
    {
      "entropy": 8.963479804992676,
      "epoch": 0.01799370220422852,
      "grad_norm": 0.97265625,
      "learning_rate": 8.95e-05,
      "loss": 8.4478,
      "mean_token_accuracy": 0.06314925663173199,
      "num_tokens": 354476.0,
      "step": 180
    },
    {
      "entropy": 8.849452114105224,
      "epoch": 0.01849352726545709,
      "grad_norm": 1.328125,
      "learning_rate": 9.2e-05,
      "loss": 8.4974,
      "mean_token_accuracy": 0.06487920247018338,
      "num_tokens": 363939.0,
      "step": 185
    },
    {
      "entropy": 8.840077209472657,
      "epoch": 0.01899335232668566,
      "grad_norm": 1.0546875,
      "learning_rate": 9.45e-05,
      "loss": 8.5395,
      "mean_token_accuracy": 0.059846711531281474,
      "num_tokens": 373187.0,
      "step": 190
    },
    {
      "entropy": 8.856325626373291,
      "epoch": 0.01949317738791423,
      "grad_norm": 0.94921875,
      "learning_rate": 9.7e-05,
      "loss": 8.4639,
      "mean_token_accuracy": 0.06266250982880592,
      "num_tokens": 382999.0,
      "step": 195
    },
    {
      "entropy": 8.76255006790161,
      "epoch": 0.0199930024491428,
      "grad_norm": 0.83984375,
      "learning_rate": 9.95e-05,
      "loss": 8.5246,
      "mean_token_accuracy": 0.055932244285941124,
      "num_tokens": 392639.0,
      "step": 200
    },
    {
      "entropy": 8.83744945526123,
      "epoch": 0.02049282751037137,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000102,
      "loss": 8.5051,
      "mean_token_accuracy": 0.060099894925951955,
      "num_tokens": 403089.0,
      "step": 205
    },
    {
      "entropy": 8.76892318725586,
      "epoch": 0.02099265257159994,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00010449999999999999,
      "loss": 8.4578,
      "mean_token_accuracy": 0.06044381149113178,
      "num_tokens": 412321.0,
      "step": 210
    },
    {
      "entropy": 8.67372522354126,
      "epoch": 0.02149247763282851,
      "grad_norm": 1.140625,
      "learning_rate": 0.000107,
      "loss": 8.4975,
      "mean_token_accuracy": 0.05988954678177834,
      "num_tokens": 421599.0,
      "step": 215
    },
    {
      "entropy": 8.718119621276855,
      "epoch": 0.02199230269405708,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0001095,
      "loss": 8.4825,
      "mean_token_accuracy": 0.06165708601474762,
      "num_tokens": 432012.0,
      "step": 220
    },
    {
      "entropy": 8.79115982055664,
      "epoch": 0.02249212775528565,
      "grad_norm": 0.921875,
      "learning_rate": 0.000112,
      "loss": 8.4559,
      "mean_token_accuracy": 0.0672556784003973,
      "num_tokens": 441989.0,
      "step": 225
    },
    {
      "entropy": 8.681735038757324,
      "epoch": 0.02299195281651422,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0001145,
      "loss": 8.4371,
      "mean_token_accuracy": 0.06710335984826088,
      "num_tokens": 451532.0,
      "step": 230
    },
    {
      "entropy": 8.714591789245606,
      "epoch": 0.02349177787774279,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00011700000000000001,
      "loss": 8.4554,
      "mean_token_accuracy": 0.06041393578052521,
      "num_tokens": 461815.0,
      "step": 235
    },
    {
      "entropy": 8.641080474853515,
      "epoch": 0.02399160293897136,
      "grad_norm": 1.140625,
      "learning_rate": 0.00011949999999999999,
      "loss": 8.5187,
      "mean_token_accuracy": 0.0578727588057518,
      "num_tokens": 470860.0,
      "step": 240
    },
    {
      "entropy": 8.66689691543579,
      "epoch": 0.02449142800019993,
      "grad_norm": 0.7578125,
      "learning_rate": 0.000122,
      "loss": 8.4869,
      "mean_token_accuracy": 0.0585276797413826,
      "num_tokens": 480649.0,
      "step": 245
    },
    {
      "entropy": 8.777419471740723,
      "epoch": 0.0249912530614285,
      "grad_norm": 0.875,
      "learning_rate": 0.0001245,
      "loss": 8.4179,
      "mean_token_accuracy": 0.06519291624426841,
      "num_tokens": 489969.0,
      "step": 250
    },
    {
      "entropy": 8.547350120544433,
      "epoch": 0.02549107812265707,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000127,
      "loss": 8.4388,
      "mean_token_accuracy": 0.061872397735714915,
      "num_tokens": 500000.0,
      "step": 255
    },
    {
      "entropy": 8.71965675354004,
      "epoch": 0.02599090318388564,
      "grad_norm": 0.890625,
      "learning_rate": 0.0001295,
      "loss": 8.4147,
      "mean_token_accuracy": 0.06390333585441113,
      "num_tokens": 510013.0,
      "step": 260
    },
    {
      "entropy": 8.639085960388183,
      "epoch": 0.026490728245114212,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000132,
      "loss": 8.4423,
      "mean_token_accuracy": 0.06236038282513619,
      "num_tokens": 519698.0,
      "step": 265
    },
    {
      "entropy": 8.607205486297607,
      "epoch": 0.02699055330634278,
      "grad_norm": 0.9375,
      "learning_rate": 0.00013450000000000002,
      "loss": 8.4579,
      "mean_token_accuracy": 0.06288496181368827,
      "num_tokens": 531631.0,
      "step": 270
    },
    {
      "entropy": 8.685933876037598,
      "epoch": 0.02749037836757135,
      "grad_norm": 1.09375,
      "learning_rate": 0.00013700000000000002,
      "loss": 8.4181,
      "mean_token_accuracy": 0.06637280248105526,
      "num_tokens": 542204.0,
      "step": 275
    },
    {
      "entropy": 8.591467189788819,
      "epoch": 0.02799020342879992,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0001395,
      "loss": 8.3423,
      "mean_token_accuracy": 0.06390709988772869,
      "num_tokens": 552269.0,
      "step": 280
    },
    {
      "entropy": 8.608537673950195,
      "epoch": 0.02849002849002849,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00014199999999999998,
      "loss": 8.3414,
      "mean_token_accuracy": 0.06816986836493015,
      "num_tokens": 561907.0,
      "step": 285
    },
    {
      "entropy": 8.648595142364503,
      "epoch": 0.02898985355125706,
      "grad_norm": 1.875,
      "learning_rate": 0.0001445,
      "loss": 8.4558,
      "mean_token_accuracy": 0.06445144601166249,
      "num_tokens": 571283.0,
      "step": 290
    },
    {
      "entropy": 8.61089735031128,
      "epoch": 0.02948967861248563,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000147,
      "loss": 8.3458,
      "mean_token_accuracy": 0.06920457892119884,
      "num_tokens": 579896.0,
      "step": 295
    },
    {
      "entropy": 8.51350793838501,
      "epoch": 0.0299895036737142,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001495,
      "loss": 8.3036,
      "mean_token_accuracy": 0.07353892996907234,
      "num_tokens": 588856.0,
      "step": 300
    },
    {
      "entropy": 8.551695823669434,
      "epoch": 0.03048932873494277,
      "grad_norm": 1.09375,
      "learning_rate": 0.000152,
      "loss": 8.3491,
      "mean_token_accuracy": 0.0707152359187603,
      "num_tokens": 598910.0,
      "step": 305
    },
    {
      "entropy": 8.606128025054932,
      "epoch": 0.03098915379617134,
      "grad_norm": 1.046875,
      "learning_rate": 0.00015450000000000001,
      "loss": 8.4008,
      "mean_token_accuracy": 0.06663076654076576,
      "num_tokens": 609248.0,
      "step": 310
    },
    {
      "entropy": 8.602969741821289,
      "epoch": 0.03148897885739991,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000157,
      "loss": 8.4004,
      "mean_token_accuracy": 0.07128242403268814,
      "num_tokens": 619352.0,
      "step": 315
    },
    {
      "entropy": 8.579054260253907,
      "epoch": 0.03198880391862848,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0001595,
      "loss": 8.4079,
      "mean_token_accuracy": 0.06752461530268192,
      "num_tokens": 628300.0,
      "step": 320
    },
    {
      "entropy": 8.591523170471191,
      "epoch": 0.03248862897985705,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000162,
      "loss": 8.3416,
      "mean_token_accuracy": 0.07400070279836654,
      "num_tokens": 637969.0,
      "step": 325
    },
    {
      "entropy": 8.576322555541992,
      "epoch": 0.03298845404108562,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00016450000000000001,
      "loss": 8.2732,
      "mean_token_accuracy": 0.07148236110806465,
      "num_tokens": 647518.0,
      "step": 330
    },
    {
      "entropy": 8.6574538230896,
      "epoch": 0.03348827910231419,
      "grad_norm": 2.1875,
      "learning_rate": 0.00016700000000000002,
      "loss": 8.4367,
      "mean_token_accuracy": 0.06586230397224427,
      "num_tokens": 658292.0,
      "step": 335
    },
    {
      "entropy": 8.541346549987793,
      "epoch": 0.03398810416354276,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016950000000000003,
      "loss": 8.3214,
      "mean_token_accuracy": 0.07294419221580029,
      "num_tokens": 667872.0,
      "step": 340
    },
    {
      "entropy": 8.493266773223876,
      "epoch": 0.03448792922477133,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017199999999999998,
      "loss": 8.2263,
      "mean_token_accuracy": 0.07715460062026977,
      "num_tokens": 676976.0,
      "step": 345
    },
    {
      "entropy": 8.568483257293702,
      "epoch": 0.0349877542859999,
      "grad_norm": 1.609375,
      "learning_rate": 0.00017449999999999999,
      "loss": 8.2417,
      "mean_token_accuracy": 0.07718590386211872,
      "num_tokens": 686149.0,
      "step": 350
    },
    {
      "entropy": 8.499206447601319,
      "epoch": 0.03548757934722847,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000177,
      "loss": 8.1559,
      "mean_token_accuracy": 0.08460776209831238,
      "num_tokens": 696439.0,
      "step": 355
    },
    {
      "entropy": 8.419091320037841,
      "epoch": 0.03598740440845704,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001795,
      "loss": 8.191,
      "mean_token_accuracy": 0.07839456871151924,
      "num_tokens": 705902.0,
      "step": 360
    },
    {
      "entropy": 8.40989990234375,
      "epoch": 0.03648722946968561,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000182,
      "loss": 8.3448,
      "mean_token_accuracy": 0.07528226897120475,
      "num_tokens": 715973.0,
      "step": 365
    },
    {
      "entropy": 8.515093135833741,
      "epoch": 0.03698705453091418,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001845,
      "loss": 8.2424,
      "mean_token_accuracy": 0.07501133531332016,
      "num_tokens": 725463.0,
      "step": 370
    },
    {
      "entropy": 8.500295448303223,
      "epoch": 0.03748687959214275,
      "grad_norm": 1.15625,
      "learning_rate": 0.000187,
      "loss": 8.2611,
      "mean_token_accuracy": 0.07273164205253124,
      "num_tokens": 734941.0,
      "step": 375
    },
    {
      "entropy": 8.510288524627686,
      "epoch": 0.03798670465337132,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001895,
      "loss": 8.2799,
      "mean_token_accuracy": 0.06865583658218384,
      "num_tokens": 744302.0,
      "step": 380
    },
    {
      "entropy": 8.47841272354126,
      "epoch": 0.03848652971459989,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000192,
      "loss": 8.2038,
      "mean_token_accuracy": 0.07888807356357574,
      "num_tokens": 754341.0,
      "step": 385
    },
    {
      "entropy": 8.518750953674317,
      "epoch": 0.03898635477582846,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001945,
      "loss": 8.336,
      "mean_token_accuracy": 0.06818274445831776,
      "num_tokens": 764087.0,
      "step": 390
    },
    {
      "entropy": 8.479927158355713,
      "epoch": 0.03948617983705703,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00019700000000000002,
      "loss": 8.269,
      "mean_token_accuracy": 0.07431223466992379,
      "num_tokens": 773728.0,
      "step": 395
    },
    {
      "entropy": 8.466327667236328,
      "epoch": 0.0399860048982856,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00019950000000000002,
      "loss": 8.2881,
      "mean_token_accuracy": 0.0715954463928938,
      "num_tokens": 784839.0,
      "step": 400
    },
    {
      "entropy": 8.545995807647705,
      "epoch": 0.04048582995951417,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000202,
      "loss": 8.2697,
      "mean_token_accuracy": 0.07756833955645562,
      "num_tokens": 794514.0,
      "step": 405
    },
    {
      "entropy": 8.35866994857788,
      "epoch": 0.04098565502074274,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00020449999999999998,
      "loss": 8.2412,
      "mean_token_accuracy": 0.08144036792218685,
      "num_tokens": 805254.0,
      "step": 410
    },
    {
      "entropy": 8.43459873199463,
      "epoch": 0.04148548008197131,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000207,
      "loss": 8.1496,
      "mean_token_accuracy": 0.07946307957172394,
      "num_tokens": 814135.0,
      "step": 415
    },
    {
      "entropy": 8.418154335021972,
      "epoch": 0.04198530514319988,
      "grad_norm": 0.921875,
      "learning_rate": 0.0002095,
      "loss": 8.1839,
      "mean_token_accuracy": 0.07811970189213753,
      "num_tokens": 823531.0,
      "step": 420
    },
    {
      "entropy": 8.321841144561768,
      "epoch": 0.04248513020442845,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000212,
      "loss": 8.1146,
      "mean_token_accuracy": 0.07991140782833099,
      "num_tokens": 832833.0,
      "step": 425
    },
    {
      "entropy": 8.342794036865234,
      "epoch": 0.04298495526565702,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002145,
      "loss": 8.1112,
      "mean_token_accuracy": 0.07684088721871377,
      "num_tokens": 841541.0,
      "step": 430
    },
    {
      "entropy": 8.320394039154053,
      "epoch": 0.04348478032688559,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00021700000000000002,
      "loss": 8.1394,
      "mean_token_accuracy": 0.0770353578031063,
      "num_tokens": 850689.0,
      "step": 435
    },
    {
      "entropy": 8.390166759490967,
      "epoch": 0.04398460538811416,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0002195,
      "loss": 8.0773,
      "mean_token_accuracy": 0.08421313390135765,
      "num_tokens": 860573.0,
      "step": 440
    },
    {
      "entropy": 8.347050094604493,
      "epoch": 0.04448443044934273,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000222,
      "loss": 8.1889,
      "mean_token_accuracy": 0.07550015076994895,
      "num_tokens": 869784.0,
      "step": 445
    },
    {
      "entropy": 8.39032440185547,
      "epoch": 0.0449842555105713,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002245,
      "loss": 8.1487,
      "mean_token_accuracy": 0.0786550410091877,
      "num_tokens": 878716.0,
      "step": 450
    },
    {
      "entropy": 8.328646183013916,
      "epoch": 0.04548408057179987,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00022700000000000002,
      "loss": 8.2197,
      "mean_token_accuracy": 0.077447821944952,
      "num_tokens": 888334.0,
      "step": 455
    },
    {
      "entropy": 8.372361946105958,
      "epoch": 0.04598390563302844,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022950000000000002,
      "loss": 8.1543,
      "mean_token_accuracy": 0.07392257861793042,
      "num_tokens": 897701.0,
      "step": 460
    },
    {
      "entropy": 8.399530696868897,
      "epoch": 0.04648373069425701,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023200000000000003,
      "loss": 8.1317,
      "mean_token_accuracy": 0.07822134792804718,
      "num_tokens": 907278.0,
      "step": 465
    },
    {
      "entropy": 8.294167709350585,
      "epoch": 0.04698355575548558,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00023449999999999998,
      "loss": 8.1437,
      "mean_token_accuracy": 0.07629175707697869,
      "num_tokens": 916194.0,
      "step": 470
    },
    {
      "entropy": 8.306899166107177,
      "epoch": 0.04748338081671415,
      "grad_norm": 1.21875,
      "learning_rate": 0.000237,
      "loss": 8.1509,
      "mean_token_accuracy": 0.08040244802832604,
      "num_tokens": 926743.0,
      "step": 475
    },
    {
      "entropy": 8.370134830474854,
      "epoch": 0.04798320587794272,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002395,
      "loss": 8.1868,
      "mean_token_accuracy": 0.0721151065081358,
      "num_tokens": 936059.0,
      "step": 480
    },
    {
      "entropy": 8.221036243438721,
      "epoch": 0.04848303093917129,
      "grad_norm": 1.140625,
      "learning_rate": 0.000242,
      "loss": 8.1031,
      "mean_token_accuracy": 0.08096854016184807,
      "num_tokens": 945990.0,
      "step": 485
    },
    {
      "entropy": 8.352540111541748,
      "epoch": 0.04898285600039986,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002445,
      "loss": 8.1913,
      "mean_token_accuracy": 0.07854288667440415,
      "num_tokens": 956626.0,
      "step": 490
    },
    {
      "entropy": 8.371198749542236,
      "epoch": 0.04948268106162843,
      "grad_norm": 1.234375,
      "learning_rate": 0.000247,
      "loss": 8.1251,
      "mean_token_accuracy": 0.0782523088157177,
      "num_tokens": 966200.0,
      "step": 495
    },
    {
      "entropy": 8.303024864196777,
      "epoch": 0.049982506122857,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0002495,
      "loss": 8.0814,
      "mean_token_accuracy": 0.08018661290407181,
      "num_tokens": 975651.0,
      "step": 500
    },
    {
      "entropy": 8.221511459350586,
      "epoch": 0.05048233118408557,
      "grad_norm": 1.015625,
      "learning_rate": 0.000252,
      "loss": 8.0613,
      "mean_token_accuracy": 0.086454738676548,
      "num_tokens": 985264.0,
      "step": 505
    },
    {
      "entropy": 8.337020969390869,
      "epoch": 0.05098215624531414,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002545,
      "loss": 8.1059,
      "mean_token_accuracy": 0.08457382321357727,
      "num_tokens": 996723.0,
      "step": 510
    },
    {
      "entropy": 8.217254400253296,
      "epoch": 0.05148198130654271,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000257,
      "loss": 8.0926,
      "mean_token_accuracy": 0.0818971797823906,
      "num_tokens": 1006098.0,
      "step": 515
    },
    {
      "entropy": 8.273353767395019,
      "epoch": 0.05198180636777128,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002595,
      "loss": 8.1417,
      "mean_token_accuracy": 0.07896760366857052,
      "num_tokens": 1016595.0,
      "step": 520
    },
    {
      "entropy": 8.265801811218262,
      "epoch": 0.05248163142899985,
      "grad_norm": 1.296875,
      "learning_rate": 0.000262,
      "loss": 8.0003,
      "mean_token_accuracy": 0.0856276884675026,
      "num_tokens": 1026281.0,
      "step": 525
    },
    {
      "entropy": 8.255826377868653,
      "epoch": 0.052981456490228424,
      "grad_norm": 0.953125,
      "learning_rate": 0.00026450000000000003,
      "loss": 8.1116,
      "mean_token_accuracy": 0.08526691868901252,
      "num_tokens": 1036697.0,
      "step": 530
    },
    {
      "entropy": 8.168128395080567,
      "epoch": 0.05348128155145699,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00026700000000000004,
      "loss": 8.0472,
      "mean_token_accuracy": 0.08230854123830796,
      "num_tokens": 1047433.0,
      "step": 535
    },
    {
      "entropy": 8.279230976104737,
      "epoch": 0.05398110661268556,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026950000000000005,
      "loss": 8.1059,
      "mean_token_accuracy": 0.07863970063626766,
      "num_tokens": 1056774.0,
      "step": 540
    },
    {
      "entropy": 8.211492776870728,
      "epoch": 0.05448093167391413,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00027200000000000005,
      "loss": 8.0195,
      "mean_token_accuracy": 0.08664621040225029,
      "num_tokens": 1068056.0,
      "step": 545
    },
    {
      "entropy": 8.15999402999878,
      "epoch": 0.0549807567351427,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002745,
      "loss": 8.0081,
      "mean_token_accuracy": 0.0843143068253994,
      "num_tokens": 1077698.0,
      "step": 550
    },
    {
      "entropy": 8.228297233581543,
      "epoch": 0.05548058179637127,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000277,
      "loss": 7.948,
      "mean_token_accuracy": 0.08828233480453491,
      "num_tokens": 1086485.0,
      "step": 555
    },
    {
      "entropy": 8.20526762008667,
      "epoch": 0.05598040685759984,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002795,
      "loss": 8.0437,
      "mean_token_accuracy": 0.07942705936729907,
      "num_tokens": 1097329.0,
      "step": 560
    },
    {
      "entropy": 8.111399030685424,
      "epoch": 0.05648023191882841,
      "grad_norm": 0.90625,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.9965,
      "mean_token_accuracy": 0.08688130974769592,
      "num_tokens": 1107990.0,
      "step": 565
    },
    {
      "entropy": 8.083545064926147,
      "epoch": 0.05698005698005698,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002845,
      "loss": 7.9398,
      "mean_token_accuracy": 0.09414497017860413,
      "num_tokens": 1117798.0,
      "step": 570
    },
    {
      "entropy": 8.12392086982727,
      "epoch": 0.05747988204128555,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000287,
      "loss": 8.0728,
      "mean_token_accuracy": 0.08350903168320656,
      "num_tokens": 1126976.0,
      "step": 575
    },
    {
      "entropy": 8.238962268829345,
      "epoch": 0.05797970710251412,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002895,
      "loss": 8.0056,
      "mean_token_accuracy": 0.08829480558633804,
      "num_tokens": 1136410.0,
      "step": 580
    },
    {
      "entropy": 8.136268615722656,
      "epoch": 0.05847953216374269,
      "grad_norm": 1.125,
      "learning_rate": 0.000292,
      "loss": 8.0448,
      "mean_token_accuracy": 0.08559292852878571,
      "num_tokens": 1147162.0,
      "step": 585
    },
    {
      "entropy": 8.142546319961548,
      "epoch": 0.05897935722497126,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002945,
      "loss": 7.9966,
      "mean_token_accuracy": 0.0838573083281517,
      "num_tokens": 1157578.0,
      "step": 590
    },
    {
      "entropy": 8.119126796722412,
      "epoch": 0.05947918228619983,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000297,
      "loss": 7.9013,
      "mean_token_accuracy": 0.09387225136160851,
      "num_tokens": 1166914.0,
      "step": 595
    },
    {
      "entropy": 8.238367509841918,
      "epoch": 0.0599790073474284,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002995,
      "loss": 8.0695,
      "mean_token_accuracy": 0.07503110878169536,
      "num_tokens": 1176740.0,
      "step": 600
    },
    {
      "entropy": 8.108829545974732,
      "epoch": 0.06047883240865697,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000302,
      "loss": 7.9774,
      "mean_token_accuracy": 0.07893616035580635,
      "num_tokens": 1185497.0,
      "step": 605
    },
    {
      "entropy": 8.179529571533203,
      "epoch": 0.06097865746988554,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003045,
      "loss": 8.0667,
      "mean_token_accuracy": 0.08129067718982697,
      "num_tokens": 1197390.0,
      "step": 610
    },
    {
      "entropy": 8.068883323669434,
      "epoch": 0.06147848253111411,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000307,
      "loss": 7.9543,
      "mean_token_accuracy": 0.08870518133044243,
      "num_tokens": 1208293.0,
      "step": 615
    },
    {
      "entropy": 8.15115303993225,
      "epoch": 0.06197830759234268,
      "grad_norm": 1.125,
      "learning_rate": 0.0003095,
      "loss": 7.9508,
      "mean_token_accuracy": 0.08918194845318794,
      "num_tokens": 1217860.0,
      "step": 620
    },
    {
      "entropy": 8.05653681755066,
      "epoch": 0.06247813265357125,
      "grad_norm": 0.96875,
      "learning_rate": 0.000312,
      "loss": 7.9353,
      "mean_token_accuracy": 0.08667743131518364,
      "num_tokens": 1227259.0,
      "step": 625
    },
    {
      "entropy": 8.01396565437317,
      "epoch": 0.06297795771479982,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003145,
      "loss": 7.8609,
      "mean_token_accuracy": 0.09146164059638977,
      "num_tokens": 1236775.0,
      "step": 630
    },
    {
      "entropy": 8.13114972114563,
      "epoch": 0.06347778277602839,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000317,
      "loss": 7.9588,
      "mean_token_accuracy": 0.0877492293715477,
      "num_tokens": 1246307.0,
      "step": 635
    },
    {
      "entropy": 8.100008535385133,
      "epoch": 0.06397760783725696,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003195,
      "loss": 7.9329,
      "mean_token_accuracy": 0.08653593435883522,
      "num_tokens": 1255197.0,
      "step": 640
    },
    {
      "entropy": 8.022264909744262,
      "epoch": 0.06447743289848552,
      "grad_norm": 1.0,
      "learning_rate": 0.000322,
      "loss": 7.9269,
      "mean_token_accuracy": 0.08907717168331146,
      "num_tokens": 1265056.0,
      "step": 645
    },
    {
      "entropy": 8.095293426513672,
      "epoch": 0.0649772579597141,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.8923,
      "mean_token_accuracy": 0.0903382994234562,
      "num_tokens": 1274718.0,
      "step": 650
    },
    {
      "entropy": 8.11915626525879,
      "epoch": 0.06547708302094267,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.9806,
      "mean_token_accuracy": 0.08302460163831711,
      "num_tokens": 1284364.0,
      "step": 655
    },
    {
      "entropy": 8.03192071914673,
      "epoch": 0.06597690808217124,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.9635,
      "mean_token_accuracy": 0.08643308579921723,
      "num_tokens": 1294186.0,
      "step": 660
    },
    {
      "entropy": 8.06210789680481,
      "epoch": 0.06647673314339982,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.8788,
      "mean_token_accuracy": 0.09071523323655128,
      "num_tokens": 1303393.0,
      "step": 665
    },
    {
      "entropy": 7.958633279800415,
      "epoch": 0.06697655820462838,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.869,
      "mean_token_accuracy": 0.08798071406781674,
      "num_tokens": 1312819.0,
      "step": 670
    },
    {
      "entropy": 8.046775722503662,
      "epoch": 0.06747638326585695,
      "grad_norm": 1.1875,
      "learning_rate": 0.000337,
      "loss": 7.9151,
      "mean_token_accuracy": 0.08215179368853569,
      "num_tokens": 1323001.0,
      "step": 675
    },
    {
      "entropy": 8.055660486221313,
      "epoch": 0.06797620832708552,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003395,
      "loss": 7.8848,
      "mean_token_accuracy": 0.08770225271582603,
      "num_tokens": 1333260.0,
      "step": 680
    },
    {
      "entropy": 7.934985589981079,
      "epoch": 0.0684760333883141,
      "grad_norm": 1.046875,
      "learning_rate": 0.000342,
      "loss": 7.8751,
      "mean_token_accuracy": 0.09213166236877442,
      "num_tokens": 1342696.0,
      "step": 685
    },
    {
      "entropy": 7.97315731048584,
      "epoch": 0.06897585844954265,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.8557,
      "mean_token_accuracy": 0.09062512218952179,
      "num_tokens": 1353104.0,
      "step": 690
    },
    {
      "entropy": 7.987550115585327,
      "epoch": 0.06947568351077123,
      "grad_norm": 1.15625,
      "learning_rate": 0.000347,
      "loss": 7.919,
      "mean_token_accuracy": 0.08597041219472885,
      "num_tokens": 1362871.0,
      "step": 695
    },
    {
      "entropy": 8.106160926818848,
      "epoch": 0.0699755085719998,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003495,
      "loss": 7.9071,
      "mean_token_accuracy": 0.08379044011235237,
      "num_tokens": 1373622.0,
      "step": 700
    },
    {
      "entropy": 7.853943729400635,
      "epoch": 0.07047533363322837,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000352,
      "loss": 7.7799,
      "mean_token_accuracy": 0.09153937473893166,
      "num_tokens": 1382707.0,
      "step": 705
    },
    {
      "entropy": 7.994217348098755,
      "epoch": 0.07097515869445695,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003545,
      "loss": 7.8429,
      "mean_token_accuracy": 0.08916826024651528,
      "num_tokens": 1392048.0,
      "step": 710
    },
    {
      "entropy": 7.870188045501709,
      "epoch": 0.0714749837556855,
      "grad_norm": 1.140625,
      "learning_rate": 0.000357,
      "loss": 7.7336,
      "mean_token_accuracy": 0.09760514721274376,
      "num_tokens": 1401387.0,
      "step": 715
    },
    {
      "entropy": 7.9647351741790775,
      "epoch": 0.07197480881691408,
      "grad_norm": 0.921875,
      "learning_rate": 0.0003595,
      "loss": 7.8467,
      "mean_token_accuracy": 0.0882681168615818,
      "num_tokens": 1411404.0,
      "step": 720
    },
    {
      "entropy": 7.9368860721588135,
      "epoch": 0.07247463387814265,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000362,
      "loss": 7.8307,
      "mean_token_accuracy": 0.09069122523069381,
      "num_tokens": 1420292.0,
      "step": 725
    },
    {
      "entropy": 7.9168719291687015,
      "epoch": 0.07297445893937123,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003645,
      "loss": 7.8032,
      "mean_token_accuracy": 0.09315077662467956,
      "num_tokens": 1429990.0,
      "step": 730
    },
    {
      "entropy": 7.91649284362793,
      "epoch": 0.07347428400059978,
      "grad_norm": 1.140625,
      "learning_rate": 0.000367,
      "loss": 7.8377,
      "mean_token_accuracy": 0.08601107746362686,
      "num_tokens": 1440148.0,
      "step": 735
    },
    {
      "entropy": 7.9139933586120605,
      "epoch": 0.07397410906182836,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003695,
      "loss": 7.7484,
      "mean_token_accuracy": 0.09189097732305526,
      "num_tokens": 1450015.0,
      "step": 740
    },
    {
      "entropy": 7.917560863494873,
      "epoch": 0.07447393412305693,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000372,
      "loss": 7.7691,
      "mean_token_accuracy": 0.08974878862500191,
      "num_tokens": 1459582.0,
      "step": 745
    },
    {
      "entropy": 7.895784044265747,
      "epoch": 0.0749737591842855,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003745,
      "loss": 7.8411,
      "mean_token_accuracy": 0.08744788095355034,
      "num_tokens": 1468151.0,
      "step": 750
    },
    {
      "entropy": 7.8630622863769535,
      "epoch": 0.07547358424551406,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000377,
      "loss": 7.726,
      "mean_token_accuracy": 0.09304906949400901,
      "num_tokens": 1479099.0,
      "step": 755
    },
    {
      "entropy": 7.873302745819092,
      "epoch": 0.07597340930674264,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003795,
      "loss": 7.8318,
      "mean_token_accuracy": 0.09302263855934143,
      "num_tokens": 1488060.0,
      "step": 760
    },
    {
      "entropy": 7.948469638824463,
      "epoch": 0.07647323436797121,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000382,
      "loss": 7.8108,
      "mean_token_accuracy": 0.0858851782977581,
      "num_tokens": 1498230.0,
      "step": 765
    },
    {
      "entropy": 7.869532060623169,
      "epoch": 0.07697305942919978,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003845,
      "loss": 7.7938,
      "mean_token_accuracy": 0.08861268982291222,
      "num_tokens": 1507962.0,
      "step": 770
    },
    {
      "entropy": 8.029801321029662,
      "epoch": 0.07747288449042836,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.9144,
      "mean_token_accuracy": 0.08749665692448616,
      "num_tokens": 1518607.0,
      "step": 775
    },
    {
      "entropy": 7.902935981750488,
      "epoch": 0.07797270955165692,
      "grad_norm": 1.125,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.7459,
      "mean_token_accuracy": 0.08882014006376267,
      "num_tokens": 1527166.0,
      "step": 780
    },
    {
      "entropy": 7.913458681106567,
      "epoch": 0.07847253461288549,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.8457,
      "mean_token_accuracy": 0.08774015679955482,
      "num_tokens": 1536683.0,
      "step": 785
    },
    {
      "entropy": 7.799791765213013,
      "epoch": 0.07897235967411406,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.7383,
      "mean_token_accuracy": 0.0890597477555275,
      "num_tokens": 1546243.0,
      "step": 790
    },
    {
      "entropy": 7.86862621307373,
      "epoch": 0.07947218473534264,
      "grad_norm": 1.125,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.7593,
      "mean_token_accuracy": 0.08828465566039086,
      "num_tokens": 1555916.0,
      "step": 795
    },
    {
      "entropy": 7.798974323272705,
      "epoch": 0.0799720097965712,
      "grad_norm": 0.90625,
      "learning_rate": 0.0003995,
      "loss": 7.6721,
      "mean_token_accuracy": 0.09845857620239258,
      "num_tokens": 1566617.0,
      "step": 800
    },
    {
      "entropy": 7.832266473770142,
      "epoch": 0.08047183485779977,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000402,
      "loss": 7.7156,
      "mean_token_accuracy": 0.0884881392121315,
      "num_tokens": 1575413.0,
      "step": 805
    },
    {
      "entropy": 7.771887922286988,
      "epoch": 0.08097165991902834,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004045,
      "loss": 7.7332,
      "mean_token_accuracy": 0.09316859021782875,
      "num_tokens": 1586863.0,
      "step": 810
    },
    {
      "entropy": 7.834986591339112,
      "epoch": 0.08147148498025691,
      "grad_norm": 0.984375,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.8049,
      "mean_token_accuracy": 0.09231851473450661,
      "num_tokens": 1597685.0,
      "step": 815
    },
    {
      "entropy": 7.867625284194946,
      "epoch": 0.08197131004148547,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004095,
      "loss": 7.6986,
      "mean_token_accuracy": 0.09077956452965737,
      "num_tokens": 1607103.0,
      "step": 820
    },
    {
      "entropy": 7.832145357131958,
      "epoch": 0.08247113510271405,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000412,
      "loss": 7.7485,
      "mean_token_accuracy": 0.08986216709017754,
      "num_tokens": 1617919.0,
      "step": 825
    },
    {
      "entropy": 7.794708204269409,
      "epoch": 0.08297096016394262,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004145,
      "loss": 7.768,
      "mean_token_accuracy": 0.09404098391532897,
      "num_tokens": 1627293.0,
      "step": 830
    },
    {
      "entropy": 7.90134825706482,
      "epoch": 0.08347078522517119,
      "grad_norm": 1.046875,
      "learning_rate": 0.000417,
      "loss": 7.7922,
      "mean_token_accuracy": 0.0896671935915947,
      "num_tokens": 1638440.0,
      "step": 835
    },
    {
      "entropy": 7.743810081481934,
      "epoch": 0.08397061028639977,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004195,
      "loss": 7.7189,
      "mean_token_accuracy": 0.09339917823672295,
      "num_tokens": 1648148.0,
      "step": 840
    },
    {
      "entropy": 7.798413705825806,
      "epoch": 0.08447043534762833,
      "grad_norm": 1.09375,
      "learning_rate": 0.000422,
      "loss": 7.7344,
      "mean_token_accuracy": 0.09364142566919327,
      "num_tokens": 1658334.0,
      "step": 845
    },
    {
      "entropy": 7.8422918796539305,
      "epoch": 0.0849702604088569,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004245,
      "loss": 7.7699,
      "mean_token_accuracy": 0.08807238787412644,
      "num_tokens": 1667963.0,
      "step": 850
    },
    {
      "entropy": 7.815905141830444,
      "epoch": 0.08547008547008547,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000427,
      "loss": 7.6715,
      "mean_token_accuracy": 0.09518391564488411,
      "num_tokens": 1677270.0,
      "step": 855
    },
    {
      "entropy": 7.7168646335601805,
      "epoch": 0.08596991053131404,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004295,
      "loss": 7.6433,
      "mean_token_accuracy": 0.09483171701431274,
      "num_tokens": 1686770.0,
      "step": 860
    },
    {
      "entropy": 7.760430765151978,
      "epoch": 0.0864697355925426,
      "grad_norm": 1.015625,
      "learning_rate": 0.000432,
      "loss": 7.7126,
      "mean_token_accuracy": 0.09171252027153969,
      "num_tokens": 1697507.0,
      "step": 865
    },
    {
      "entropy": 7.743275165557861,
      "epoch": 0.08696956065377118,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004345,
      "loss": 7.6634,
      "mean_token_accuracy": 0.09470361396670342,
      "num_tokens": 1708253.0,
      "step": 870
    },
    {
      "entropy": 7.803777408599854,
      "epoch": 0.08746938571499975,
      "grad_norm": 0.96875,
      "learning_rate": 0.000437,
      "loss": 7.7314,
      "mean_token_accuracy": 0.08744200840592384,
      "num_tokens": 1717827.0,
      "step": 875
    },
    {
      "entropy": 7.641020631790161,
      "epoch": 0.08796921077622832,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004395,
      "loss": 7.5264,
      "mean_token_accuracy": 0.10062878951430321,
      "num_tokens": 1726678.0,
      "step": 880
    },
    {
      "entropy": 7.729559707641601,
      "epoch": 0.0884690358374569,
      "grad_norm": 0.953125,
      "learning_rate": 0.000442,
      "loss": 7.6837,
      "mean_token_accuracy": 0.09284055829048157,
      "num_tokens": 1737518.0,
      "step": 885
    },
    {
      "entropy": 7.701491022109986,
      "epoch": 0.08896886089868546,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004445,
      "loss": 7.5668,
      "mean_token_accuracy": 0.09772049933671952,
      "num_tokens": 1746338.0,
      "step": 890
    },
    {
      "entropy": 7.655733203887939,
      "epoch": 0.08946868595991403,
      "grad_norm": 1.015625,
      "learning_rate": 0.000447,
      "loss": 7.667,
      "mean_token_accuracy": 0.09409659132361411,
      "num_tokens": 1756538.0,
      "step": 895
    },
    {
      "entropy": 7.753999948501587,
      "epoch": 0.0899685110211426,
      "grad_norm": 0.9375,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.654,
      "mean_token_accuracy": 0.09215902574360371,
      "num_tokens": 1766722.0,
      "step": 900
    },
    {
      "entropy": 7.662614202499389,
      "epoch": 0.09046833608237118,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045200000000000004,
      "loss": 7.5758,
      "mean_token_accuracy": 0.09804621636867523,
      "num_tokens": 1777039.0,
      "step": 905
    },
    {
      "entropy": 7.686301565170288,
      "epoch": 0.09096816114359974,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.69,
      "mean_token_accuracy": 0.08747479245066643,
      "num_tokens": 1787621.0,
      "step": 910
    },
    {
      "entropy": 7.739987802505493,
      "epoch": 0.09146798620482831,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045700000000000005,
      "loss": 7.6941,
      "mean_token_accuracy": 0.0946861956268549,
      "num_tokens": 1798345.0,
      "step": 915
    },
    {
      "entropy": 7.674608373641968,
      "epoch": 0.09196781126605688,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.6303,
      "mean_token_accuracy": 0.10006568729877471,
      "num_tokens": 1807994.0,
      "step": 920
    },
    {
      "entropy": 7.651217222213745,
      "epoch": 0.09246763632728545,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000462,
      "loss": 7.5652,
      "mean_token_accuracy": 0.08964619114995002,
      "num_tokens": 1817965.0,
      "step": 925
    },
    {
      "entropy": 7.639501237869263,
      "epoch": 0.09296746138851401,
      "grad_norm": 1.125,
      "learning_rate": 0.0004645,
      "loss": 7.62,
      "mean_token_accuracy": 0.09154486805200576,
      "num_tokens": 1827094.0,
      "step": 930
    },
    {
      "entropy": 7.7448944568634035,
      "epoch": 0.09346728644974259,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000467,
      "loss": 7.5907,
      "mean_token_accuracy": 0.09276118874549866,
      "num_tokens": 1837754.0,
      "step": 935
    },
    {
      "entropy": 7.587901210784912,
      "epoch": 0.09396711151097116,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004695,
      "loss": 7.5858,
      "mean_token_accuracy": 0.09547252953052521,
      "num_tokens": 1847500.0,
      "step": 940
    },
    {
      "entropy": 7.606810808181763,
      "epoch": 0.09446693657219973,
      "grad_norm": 1.03125,
      "learning_rate": 0.000472,
      "loss": 7.6324,
      "mean_token_accuracy": 0.08944343850016594,
      "num_tokens": 1857663.0,
      "step": 945
    },
    {
      "entropy": 7.784541130065918,
      "epoch": 0.0949667616334283,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004745,
      "loss": 7.6947,
      "mean_token_accuracy": 0.0932641200721264,
      "num_tokens": 1867752.0,
      "step": 950
    },
    {
      "entropy": 7.556209373474121,
      "epoch": 0.09546658669465687,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000477,
      "loss": 7.626,
      "mean_token_accuracy": 0.09295311495661736,
      "num_tokens": 1877479.0,
      "step": 955
    },
    {
      "entropy": 7.730465078353882,
      "epoch": 0.09596641175588544,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004795,
      "loss": 7.6473,
      "mean_token_accuracy": 0.09373209923505783,
      "num_tokens": 1887944.0,
      "step": 960
    },
    {
      "entropy": 7.654263734817505,
      "epoch": 0.09646623681711401,
      "grad_norm": 1.125,
      "learning_rate": 0.000482,
      "loss": 7.5671,
      "mean_token_accuracy": 0.09599127843976021,
      "num_tokens": 1896991.0,
      "step": 965
    },
    {
      "entropy": 7.621188306808472,
      "epoch": 0.09696606187834259,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004845,
      "loss": 7.5894,
      "mean_token_accuracy": 0.1009646438062191,
      "num_tokens": 1907216.0,
      "step": 970
    },
    {
      "entropy": 7.716098117828369,
      "epoch": 0.09746588693957114,
      "grad_norm": 1.09375,
      "learning_rate": 0.000487,
      "loss": 7.5806,
      "mean_token_accuracy": 0.09529191181063652,
      "num_tokens": 1917206.0,
      "step": 975
    },
    {
      "entropy": 7.511505031585694,
      "epoch": 0.09796571200079972,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004895,
      "loss": 7.672,
      "mean_token_accuracy": 0.09074099287390709,
      "num_tokens": 1929669.0,
      "step": 980
    },
    {
      "entropy": 7.717717981338501,
      "epoch": 0.09846553706202829,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000492,
      "loss": 7.5271,
      "mean_token_accuracy": 0.09860590994358062,
      "num_tokens": 1939881.0,
      "step": 985
    },
    {
      "entropy": 7.50510540008545,
      "epoch": 0.09896536212325686,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004945,
      "loss": 7.5609,
      "mean_token_accuracy": 0.09686254858970642,
      "num_tokens": 1949423.0,
      "step": 990
    },
    {
      "entropy": 7.61955680847168,
      "epoch": 0.09946518718448542,
      "grad_norm": 0.96875,
      "learning_rate": 0.000497,
      "loss": 7.5959,
      "mean_token_accuracy": 0.09370956495404244,
      "num_tokens": 1959214.0,
      "step": 995
    },
    {
      "entropy": 7.723949670791626,
      "epoch": 0.099965012245714,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995,
      "loss": 7.5537,
      "mean_token_accuracy": 0.0956396333873272,
      "num_tokens": 1968963.0,
      "step": 1000
    },
    {
      "entropy": 7.549405145645141,
      "epoch": 0.10046483730694257,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999999981884985,
      "loss": 7.5633,
      "mean_token_accuracy": 0.09498978704214096,
      "num_tokens": 1979551.0,
      "step": 1005
    },
    {
      "entropy": 7.5673760890960695,
      "epoch": 0.10096466236817114,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999999908292739,
      "loss": 7.5157,
      "mean_token_accuracy": 0.09872248247265816,
      "num_tokens": 1988432.0,
      "step": 1010
    },
    {
      "entropy": 7.55355863571167,
      "epoch": 0.10146448742939972,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999999778091072,
      "loss": 7.5591,
      "mean_token_accuracy": 0.09609471410512924,
      "num_tokens": 1997509.0,
      "step": 1015
    },
    {
      "entropy": 7.737374544143677,
      "epoch": 0.10196431249062828,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999999591279991,
      "loss": 7.7221,
      "mean_token_accuracy": 0.09012902900576591,
      "num_tokens": 2007406.0,
      "step": 1020
    },
    {
      "entropy": 7.63672366142273,
      "epoch": 0.10246413755185685,
      "grad_norm": 1.109375,
      "learning_rate": 0.00049999993478595,
      "loss": 7.5417,
      "mean_token_accuracy": 0.09073366522789002,
      "num_tokens": 2016760.0,
      "step": 1025
    },
    {
      "entropy": 7.532858657836914,
      "epoch": 0.10296396261308542,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999999047829604,
      "loss": 7.5635,
      "mean_token_accuracy": 0.09805454760789871,
      "num_tokens": 2027776.0,
      "step": 1030
    },
    {
      "entropy": 7.5644487857818605,
      "epoch": 0.103463787674314,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499999869119031,
      "loss": 7.4998,
      "mean_token_accuracy": 0.10216570869088173,
      "num_tokens": 2036955.0,
      "step": 1035
    },
    {
      "entropy": 7.550212717056274,
      "epoch": 0.10396361273554255,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999998277941629,
      "loss": 7.4273,
      "mean_token_accuracy": 0.10139072239398957,
      "num_tokens": 2046118.0,
      "step": 1040
    },
    {
      "entropy": 7.455883264541626,
      "epoch": 0.10446343779677113,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999997808083569,
      "loss": 7.5015,
      "mean_token_accuracy": 0.09121370837092399,
      "num_tokens": 2056266.0,
      "step": 1045
    },
    {
      "entropy": 7.592077207565308,
      "epoch": 0.1049632628579997,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999997281616144,
      "loss": 7.5511,
      "mean_token_accuracy": 0.0956337384879589,
      "num_tokens": 2067416.0,
      "step": 1050
    },
    {
      "entropy": 7.572665500640869,
      "epoch": 0.10546308791922827,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999996698539366,
      "loss": 7.5287,
      "mean_token_accuracy": 0.09303966537117958,
      "num_tokens": 2076926.0,
      "step": 1055
    },
    {
      "entropy": 7.479141426086426,
      "epoch": 0.10596291298045685,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499999605885325,
      "loss": 7.5214,
      "mean_token_accuracy": 0.09900636374950408,
      "num_tokens": 2088017.0,
      "step": 1060
    },
    {
      "entropy": 7.522082901000976,
      "epoch": 0.1064627380416854,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999995362557813,
      "loss": 7.5462,
      "mean_token_accuracy": 0.09880418479442596,
      "num_tokens": 2097249.0,
      "step": 1065
    },
    {
      "entropy": 7.551411294937134,
      "epoch": 0.10696256310291398,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499999460965307,
      "loss": 7.4761,
      "mean_token_accuracy": 0.0976270966231823,
      "num_tokens": 2105778.0,
      "step": 1070
    },
    {
      "entropy": 7.56266303062439,
      "epoch": 0.10746238816414255,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999993800139043,
      "loss": 7.5119,
      "mean_token_accuracy": 0.10032028555870057,
      "num_tokens": 2114710.0,
      "step": 1075
    },
    {
      "entropy": 7.5173601627349855,
      "epoch": 0.10796221322537113,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999992934015752,
      "loss": 7.4467,
      "mean_token_accuracy": 0.09914555251598359,
      "num_tokens": 2124002.0,
      "step": 1080
    },
    {
      "entropy": 7.5395721912384035,
      "epoch": 0.10846203828659969,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999992011283215,
      "loss": 7.5423,
      "mean_token_accuracy": 0.0994409941136837,
      "num_tokens": 2134592.0,
      "step": 1085
    },
    {
      "entropy": 7.472764301300049,
      "epoch": 0.10896186334782826,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999991031941459,
      "loss": 7.3796,
      "mean_token_accuracy": 0.10381273031234742,
      "num_tokens": 2143641.0,
      "step": 1090
    },
    {
      "entropy": 7.512874698638916,
      "epoch": 0.10946168840905683,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004999989995990509,
      "loss": 7.5118,
      "mean_token_accuracy": 0.09514869302511215,
      "num_tokens": 2154738.0,
      "step": 1095
    },
    {
      "entropy": 7.4650367259979244,
      "epoch": 0.1099615134702854,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999988903430389,
      "loss": 7.464,
      "mean_token_accuracy": 0.10286911129951477,
      "num_tokens": 2164365.0,
      "step": 1100
    },
    {
      "entropy": 7.42442216873169,
      "epoch": 0.11046133853151396,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999987754261126,
      "loss": 7.3251,
      "mean_token_accuracy": 0.10485819950699807,
      "num_tokens": 2173185.0,
      "step": 1105
    },
    {
      "entropy": 7.488422584533692,
      "epoch": 0.11096116359274254,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999986548482751,
      "loss": 7.4557,
      "mean_token_accuracy": 0.10024777427315712,
      "num_tokens": 2184717.0,
      "step": 1110
    },
    {
      "entropy": 7.4683778285980225,
      "epoch": 0.11146098865397111,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999985286095293,
      "loss": 7.4244,
      "mean_token_accuracy": 0.10170031487941741,
      "num_tokens": 2193681.0,
      "step": 1115
    },
    {
      "entropy": 7.480268287658691,
      "epoch": 0.11196081371519968,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999983967098785,
      "loss": 7.4427,
      "mean_token_accuracy": 0.09846253246068955,
      "num_tokens": 2203366.0,
      "step": 1120
    },
    {
      "entropy": 7.510249376296997,
      "epoch": 0.11246063877642826,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999982591493258,
      "loss": 7.437,
      "mean_token_accuracy": 0.10007026195526122,
      "num_tokens": 2213982.0,
      "step": 1125
    },
    {
      "entropy": 7.470330047607422,
      "epoch": 0.11296046383765682,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999981159278749,
      "loss": 7.4809,
      "mean_token_accuracy": 0.09837306439876556,
      "num_tokens": 2222897.0,
      "step": 1130
    },
    {
      "entropy": 7.428879880905152,
      "epoch": 0.11346028889888539,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999979670455292,
      "loss": 7.4428,
      "mean_token_accuracy": 0.10468190163373947,
      "num_tokens": 2232472.0,
      "step": 1135
    },
    {
      "entropy": 7.399091100692749,
      "epoch": 0.11396011396011396,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004999978125022926,
      "loss": 7.4736,
      "mean_token_accuracy": 0.10416980311274529,
      "num_tokens": 2242666.0,
      "step": 1140
    },
    {
      "entropy": 7.615888071060181,
      "epoch": 0.11445993902134254,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499997652298169,
      "loss": 7.5362,
      "mean_token_accuracy": 0.09456123262643815,
      "num_tokens": 2254586.0,
      "step": 1145
    },
    {
      "entropy": 7.44240026473999,
      "epoch": 0.1149597640825711,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999974864331624,
      "loss": 7.3675,
      "mean_token_accuracy": 0.10122138187289238,
      "num_tokens": 2264619.0,
      "step": 1150
    },
    {
      "entropy": 7.638830471038818,
      "epoch": 0.11545958914379967,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999973149072768,
      "loss": 7.4664,
      "mean_token_accuracy": 0.0914570763707161,
      "num_tokens": 2274114.0,
      "step": 1155
    },
    {
      "entropy": 7.202870082855225,
      "epoch": 0.11595941420502824,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999971377205167,
      "loss": 7.3752,
      "mean_token_accuracy": 0.10422463156282902,
      "num_tokens": 2283887.0,
      "step": 1160
    },
    {
      "entropy": 7.499454975128174,
      "epoch": 0.11645923926625681,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999969548728864,
      "loss": 7.3798,
      "mean_token_accuracy": 0.10416734889149666,
      "num_tokens": 2293090.0,
      "step": 1165
    },
    {
      "entropy": 7.405326890945434,
      "epoch": 0.11695906432748537,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999967663643908,
      "loss": 7.4941,
      "mean_token_accuracy": 0.1007530689239502,
      "num_tokens": 2302828.0,
      "step": 1170
    },
    {
      "entropy": 7.508965539932251,
      "epoch": 0.11745888938871395,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999965721950345,
      "loss": 7.3741,
      "mean_token_accuracy": 0.10420065671205521,
      "num_tokens": 2311415.0,
      "step": 1175
    },
    {
      "entropy": 7.418776798248291,
      "epoch": 0.11795871444994252,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999963723648222,
      "loss": 7.4587,
      "mean_token_accuracy": 0.09595207944512367,
      "num_tokens": 2321512.0,
      "step": 1180
    },
    {
      "entropy": 7.462272262573242,
      "epoch": 0.1184585395111711,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499996166873759,
      "loss": 7.3597,
      "mean_token_accuracy": 0.09774425327777862,
      "num_tokens": 2331398.0,
      "step": 1185
    },
    {
      "entropy": 7.390771961212158,
      "epoch": 0.11895836457239967,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999959557218502,
      "loss": 7.3398,
      "mean_token_accuracy": 0.10234742686152458,
      "num_tokens": 2341751.0,
      "step": 1190
    },
    {
      "entropy": 7.501025581359864,
      "epoch": 0.11945818963362823,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999957389091012,
      "loss": 7.3795,
      "mean_token_accuracy": 0.09952760934829712,
      "num_tokens": 2352538.0,
      "step": 1195
    },
    {
      "entropy": 7.375369358062744,
      "epoch": 0.1199580146948568,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999955164355171,
      "loss": 7.45,
      "mean_token_accuracy": 0.09802018627524375,
      "num_tokens": 2362433.0,
      "step": 1200
    },
    {
      "entropy": 7.532867479324341,
      "epoch": 0.12045783975608537,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999952883011038,
      "loss": 7.4279,
      "mean_token_accuracy": 0.09662150442600251,
      "num_tokens": 2372431.0,
      "step": 1205
    },
    {
      "entropy": 7.313345861434937,
      "epoch": 0.12095766481731395,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499995054505867,
      "loss": 7.4233,
      "mean_token_accuracy": 0.10795046091079712,
      "num_tokens": 2382483.0,
      "step": 1210
    },
    {
      "entropy": 7.574924325942993,
      "epoch": 0.1214574898785425,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999948150498124,
      "loss": 7.4059,
      "mean_token_accuracy": 0.09452533945441247,
      "num_tokens": 2392835.0,
      "step": 1215
    },
    {
      "entropy": 7.347304105758667,
      "epoch": 0.12195731493977108,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999945699329462,
      "loss": 7.3737,
      "mean_token_accuracy": 0.09439274296164513,
      "num_tokens": 2402305.0,
      "step": 1220
    },
    {
      "entropy": 7.343445539474487,
      "epoch": 0.12245714000099965,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999943191552745,
      "loss": 7.3078,
      "mean_token_accuracy": 0.10918881967663766,
      "num_tokens": 2412093.0,
      "step": 1225
    },
    {
      "entropy": 7.356309986114502,
      "epoch": 0.12295696506222822,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999940627168037,
      "loss": 7.2729,
      "mean_token_accuracy": 0.11116204932332038,
      "num_tokens": 2421349.0,
      "step": 1230
    },
    {
      "entropy": 7.334961366653443,
      "epoch": 0.12345679012345678,
      "grad_norm": 1.109375,
      "learning_rate": 0.00049999380061754,
      "loss": 7.2729,
      "mean_token_accuracy": 0.11540432274341583,
      "num_tokens": 2430672.0,
      "step": 1235
    },
    {
      "entropy": 7.312632274627686,
      "epoch": 0.12395661518468536,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999935328574903,
      "loss": 7.3496,
      "mean_token_accuracy": 0.1008877657353878,
      "num_tokens": 2439355.0,
      "step": 1240
    },
    {
      "entropy": 7.394206523895264,
      "epoch": 0.12445644024591393,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999932594366613,
      "loss": 7.3439,
      "mean_token_accuracy": 0.10909157991409302,
      "num_tokens": 2448860.0,
      "step": 1245
    },
    {
      "entropy": 7.350288391113281,
      "epoch": 0.1249562653071425,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999929803550597,
      "loss": 7.3731,
      "mean_token_accuracy": 0.10190236046910287,
      "num_tokens": 2458276.0,
      "step": 1250
    },
    {
      "entropy": 7.332216119766235,
      "epoch": 0.12545609036837108,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999926956126927,
      "loss": 7.2885,
      "mean_token_accuracy": 0.10416633412241935,
      "num_tokens": 2467791.0,
      "step": 1255
    },
    {
      "entropy": 7.432009840011597,
      "epoch": 0.12595591542959964,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999924052095672,
      "loss": 7.4218,
      "mean_token_accuracy": 0.10164169669151306,
      "num_tokens": 2477549.0,
      "step": 1260
    },
    {
      "entropy": 7.333640623092651,
      "epoch": 0.12645574049082822,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999921091456907,
      "loss": 7.323,
      "mean_token_accuracy": 0.09840168058872223,
      "num_tokens": 2487549.0,
      "step": 1265
    },
    {
      "entropy": 7.464362382888794,
      "epoch": 0.12695556555205678,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999918074210707,
      "loss": 7.3226,
      "mean_token_accuracy": 0.10223804861307144,
      "num_tokens": 2496799.0,
      "step": 1270
    },
    {
      "entropy": 7.4028078556060795,
      "epoch": 0.12745539061328534,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999915000357147,
      "loss": 7.4079,
      "mean_token_accuracy": 0.09940458238124847,
      "num_tokens": 2505795.0,
      "step": 1275
    },
    {
      "entropy": 7.363927745819092,
      "epoch": 0.12795521567451393,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999911869896305,
      "loss": 7.229,
      "mean_token_accuracy": 0.10232904329895973,
      "num_tokens": 2515289.0,
      "step": 1280
    },
    {
      "entropy": 7.393394041061401,
      "epoch": 0.1284550407357425,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999908682828258,
      "loss": 7.4109,
      "mean_token_accuracy": 0.09670610949397088,
      "num_tokens": 2524563.0,
      "step": 1285
    },
    {
      "entropy": 7.354452133178711,
      "epoch": 0.12895486579697105,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999905439153089,
      "loss": 7.3183,
      "mean_token_accuracy": 0.10640332400798798,
      "num_tokens": 2534057.0,
      "step": 1290
    },
    {
      "entropy": 7.3064008235931395,
      "epoch": 0.12945469085819963,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999902138870878,
      "loss": 7.2834,
      "mean_token_accuracy": 0.10942733064293861,
      "num_tokens": 2544646.0,
      "step": 1295
    },
    {
      "entropy": 7.4173472881317135,
      "epoch": 0.1299545159194282,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999898781981707,
      "loss": 7.3696,
      "mean_token_accuracy": 0.1016182117164135,
      "num_tokens": 2553843.0,
      "step": 1300
    },
    {
      "entropy": 7.349134922027588,
      "epoch": 0.13045434098065678,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999895368485662,
      "loss": 7.4124,
      "mean_token_accuracy": 0.09806189686059952,
      "num_tokens": 2564061.0,
      "step": 1305
    },
    {
      "entropy": 7.351945590972901,
      "epoch": 0.13095416604188534,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499989189838283,
      "loss": 7.2677,
      "mean_token_accuracy": 0.10726078897714615,
      "num_tokens": 2574758.0,
      "step": 1310
    },
    {
      "entropy": 7.341266250610351,
      "epoch": 0.1314539911031139,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999888371673295,
      "loss": 7.3301,
      "mean_token_accuracy": 0.10427657440304756,
      "num_tokens": 2585213.0,
      "step": 1315
    },
    {
      "entropy": 7.415141153335571,
      "epoch": 0.13195381616434249,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999884788357147,
      "loss": 7.3863,
      "mean_token_accuracy": 0.10240926444530488,
      "num_tokens": 2595511.0,
      "step": 1320
    },
    {
      "entropy": 7.322946357727051,
      "epoch": 0.13245364122557104,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999881148434478,
      "loss": 7.3065,
      "mean_token_accuracy": 0.1043054111301899,
      "num_tokens": 2605930.0,
      "step": 1325
    },
    {
      "entropy": 7.352214527130127,
      "epoch": 0.13295346628679963,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999877451905378,
      "loss": 7.1786,
      "mean_token_accuracy": 0.11697915494441986,
      "num_tokens": 2614224.0,
      "step": 1330
    },
    {
      "entropy": 7.341785049438476,
      "epoch": 0.1334532913480282,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999873698769941,
      "loss": 7.4133,
      "mean_token_accuracy": 0.09787786826491356,
      "num_tokens": 2624879.0,
      "step": 1335
    },
    {
      "entropy": 7.313656330108643,
      "epoch": 0.13395311640925675,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499986988902826,
      "loss": 7.2542,
      "mean_token_accuracy": 0.10557752177119255,
      "num_tokens": 2633836.0,
      "step": 1340
    },
    {
      "entropy": 7.303465509414673,
      "epoch": 0.13445294147048534,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999866022680431,
      "loss": 7.3265,
      "mean_token_accuracy": 0.10393745005130768,
      "num_tokens": 2643754.0,
      "step": 1345
    },
    {
      "entropy": 7.485411548614502,
      "epoch": 0.1349527665317139,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999862099726552,
      "loss": 7.3691,
      "mean_token_accuracy": 0.10118599981069565,
      "num_tokens": 2652708.0,
      "step": 1350
    },
    {
      "entropy": 7.293326139450073,
      "epoch": 0.13545259159294248,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999858120166721,
      "loss": 7.3332,
      "mean_token_accuracy": 0.10101174637675285,
      "num_tokens": 2662353.0,
      "step": 1355
    },
    {
      "entropy": 7.395275354385376,
      "epoch": 0.13595241665417104,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999854084001039,
      "loss": 7.2455,
      "mean_token_accuracy": 0.10808999240398406,
      "num_tokens": 2672376.0,
      "step": 1360
    },
    {
      "entropy": 7.28541202545166,
      "epoch": 0.1364522417153996,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999849991229608,
      "loss": 7.343,
      "mean_token_accuracy": 0.10266417935490608,
      "num_tokens": 2682115.0,
      "step": 1365
    },
    {
      "entropy": 7.361982917785644,
      "epoch": 0.1369520667766282,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499984584185253,
      "loss": 7.3249,
      "mean_token_accuracy": 0.10273946896195411,
      "num_tokens": 2692429.0,
      "step": 1370
    },
    {
      "entropy": 7.356666135787964,
      "epoch": 0.13745189183785675,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999841635869909,
      "loss": 7.2624,
      "mean_token_accuracy": 0.10483143851161003,
      "num_tokens": 2701373.0,
      "step": 1375
    },
    {
      "entropy": 7.448349714279175,
      "epoch": 0.1379517168990853,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004999837373281852,
      "loss": 7.4456,
      "mean_token_accuracy": 0.10275454893708229,
      "num_tokens": 2710717.0,
      "step": 1380
    },
    {
      "entropy": 7.243832445144653,
      "epoch": 0.1384515419603139,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999833054088465,
      "loss": 7.3053,
      "mean_token_accuracy": 0.1035049244761467,
      "num_tokens": 2720936.0,
      "step": 1385
    },
    {
      "entropy": 7.444944524765015,
      "epoch": 0.13895136702154245,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999828678289858,
      "loss": 7.2552,
      "mean_token_accuracy": 0.11174436882138253,
      "num_tokens": 2730504.0,
      "step": 1390
    },
    {
      "entropy": 7.234856271743775,
      "epoch": 0.13945119208277104,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499982424588614,
      "loss": 7.2567,
      "mean_token_accuracy": 0.10671055763959884,
      "num_tokens": 2739069.0,
      "step": 1395
    },
    {
      "entropy": 7.39029221534729,
      "epoch": 0.1399510171439996,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999819756877422,
      "loss": 7.3385,
      "mean_token_accuracy": 0.09800549671053886,
      "num_tokens": 2748941.0,
      "step": 1400
    },
    {
      "entropy": 7.275935411453247,
      "epoch": 0.14045084220522816,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999815211263819,
      "loss": 7.2932,
      "mean_token_accuracy": 0.11003827452659606,
      "num_tokens": 2758915.0,
      "step": 1405
    },
    {
      "entropy": 7.405569314956665,
      "epoch": 0.14095066726645675,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999810609045444,
      "loss": 7.3226,
      "mean_token_accuracy": 0.10157396346330642,
      "num_tokens": 2768972.0,
      "step": 1410
    },
    {
      "entropy": 7.269433259963989,
      "epoch": 0.1414504923276853,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999805950222412,
      "loss": 7.1646,
      "mean_token_accuracy": 0.11019177362322807,
      "num_tokens": 2777989.0,
      "step": 1415
    },
    {
      "entropy": 7.236843538284302,
      "epoch": 0.1419503173889139,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999801234794843,
      "loss": 7.2713,
      "mean_token_accuracy": 0.10585586354136467,
      "num_tokens": 2788227.0,
      "step": 1420
    },
    {
      "entropy": 7.350889062881469,
      "epoch": 0.14245014245014245,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999796462762853,
      "loss": 7.2754,
      "mean_token_accuracy": 0.10650887861847877,
      "num_tokens": 2797729.0,
      "step": 1425
    },
    {
      "entropy": 7.336771488189697,
      "epoch": 0.142949967511371,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999791634126562,
      "loss": 7.2716,
      "mean_token_accuracy": 0.1051708072423935,
      "num_tokens": 2807215.0,
      "step": 1430
    },
    {
      "entropy": 7.2520225048065186,
      "epoch": 0.1434497925725996,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999786748886094,
      "loss": 7.1883,
      "mean_token_accuracy": 0.10783864781260491,
      "num_tokens": 2816945.0,
      "step": 1435
    },
    {
      "entropy": 7.216451740264892,
      "epoch": 0.14394961763382816,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999781807041569,
      "loss": 7.2712,
      "mean_token_accuracy": 0.10556349977850914,
      "num_tokens": 2827072.0,
      "step": 1440
    },
    {
      "entropy": 7.333676815032959,
      "epoch": 0.14444944269505672,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999776808593113,
      "loss": 7.1848,
      "mean_token_accuracy": 0.10760891959071159,
      "num_tokens": 2837928.0,
      "step": 1445
    },
    {
      "entropy": 7.182110834121704,
      "epoch": 0.1449492677562853,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999771753540852,
      "loss": 7.224,
      "mean_token_accuracy": 0.11059348955750466,
      "num_tokens": 2847861.0,
      "step": 1450
    },
    {
      "entropy": 7.324659442901611,
      "epoch": 0.14544909281751386,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999766641884912,
      "loss": 7.2266,
      "mean_token_accuracy": 0.10654641091823577,
      "num_tokens": 2857885.0,
      "step": 1455
    },
    {
      "entropy": 7.342340612411499,
      "epoch": 0.14594891787874245,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999761473625422,
      "loss": 7.2576,
      "mean_token_accuracy": 0.10456847101449966,
      "num_tokens": 2867004.0,
      "step": 1460
    },
    {
      "entropy": 7.183152198791504,
      "epoch": 0.146448742939971,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999756248762513,
      "loss": 7.1181,
      "mean_token_accuracy": 0.11030580475926399,
      "num_tokens": 2876285.0,
      "step": 1465
    },
    {
      "entropy": 7.275135135650634,
      "epoch": 0.14694856800119957,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999750967296315,
      "loss": 7.2797,
      "mean_token_accuracy": 0.10354529283940792,
      "num_tokens": 2886967.0,
      "step": 1470
    },
    {
      "entropy": 7.288818311691284,
      "epoch": 0.14744839306242816,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999745629226962,
      "loss": 7.2214,
      "mean_token_accuracy": 0.10634861737489701,
      "num_tokens": 2897137.0,
      "step": 1475
    },
    {
      "entropy": 7.203562879562378,
      "epoch": 0.14794821812365672,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999740234554589,
      "loss": 7.189,
      "mean_token_accuracy": 0.11473996862769127,
      "num_tokens": 2907658.0,
      "step": 1480
    },
    {
      "entropy": 7.289450311660767,
      "epoch": 0.1484480431848853,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999734783279327,
      "loss": 7.1948,
      "mean_token_accuracy": 0.10778787955641747,
      "num_tokens": 2917801.0,
      "step": 1485
    },
    {
      "entropy": 7.242615985870361,
      "epoch": 0.14894786824611386,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499972927540132,
      "loss": 7.2853,
      "mean_token_accuracy": 0.10557879060506821,
      "num_tokens": 2927188.0,
      "step": 1490
    },
    {
      "entropy": 7.2714708805084225,
      "epoch": 0.14944769330734242,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999723710920703,
      "loss": 7.1637,
      "mean_token_accuracy": 0.10618824735283852,
      "num_tokens": 2936781.0,
      "step": 1495
    },
    {
      "entropy": 7.207219743728638,
      "epoch": 0.149947518368571,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999718089837616,
      "loss": 7.146,
      "mean_token_accuracy": 0.10595403760671615,
      "num_tokens": 2946426.0,
      "step": 1500
    },
    {
      "entropy": 7.163693475723266,
      "epoch": 0.15044734342979957,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00049997124121522,
      "loss": 7.1859,
      "mean_token_accuracy": 0.1059610590338707,
      "num_tokens": 2956246.0,
      "step": 1505
    },
    {
      "entropy": 7.299579715728759,
      "epoch": 0.15094716849102813,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999706677864599,
      "loss": 7.1782,
      "mean_token_accuracy": 0.10952525734901428,
      "num_tokens": 2965930.0,
      "step": 1510
    },
    {
      "entropy": 7.17916841506958,
      "epoch": 0.15144699355225671,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999700886974958,
      "loss": 7.1203,
      "mean_token_accuracy": 0.11073079928755761,
      "num_tokens": 2975975.0,
      "step": 1515
    },
    {
      "entropy": 7.203798532485962,
      "epoch": 0.15194681861348527,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499969503948342,
      "loss": 7.1483,
      "mean_token_accuracy": 0.10983792692422867,
      "num_tokens": 2985343.0,
      "step": 1520
    },
    {
      "entropy": 7.188997840881347,
      "epoch": 0.15244664367471386,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999689135390134,
      "loss": 7.1451,
      "mean_token_accuracy": 0.1086390919983387,
      "num_tokens": 2994420.0,
      "step": 1525
    },
    {
      "entropy": 7.235520648956299,
      "epoch": 0.15294646873594242,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999683174695249,
      "loss": 7.2723,
      "mean_token_accuracy": 0.10853548273444176,
      "num_tokens": 3004447.0,
      "step": 1530
    },
    {
      "entropy": 7.342757797241211,
      "epoch": 0.15344629379717098,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999677157398914,
      "loss": 7.1605,
      "mean_token_accuracy": 0.10739956200122833,
      "num_tokens": 3014596.0,
      "step": 1535
    },
    {
      "entropy": 7.138177442550659,
      "epoch": 0.15394611885839957,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999671083501281,
      "loss": 7.0981,
      "mean_token_accuracy": 0.10866019204258918,
      "num_tokens": 3024528.0,
      "step": 1540
    },
    {
      "entropy": 7.240619468688965,
      "epoch": 0.15444594391962813,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999664953002502,
      "loss": 7.2099,
      "mean_token_accuracy": 0.10421332791447639,
      "num_tokens": 3035233.0,
      "step": 1545
    },
    {
      "entropy": 7.229916572570801,
      "epoch": 0.1549457689808567,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499965876590273,
      "loss": 7.2099,
      "mean_token_accuracy": 0.10832755044102668,
      "num_tokens": 3045890.0,
      "step": 1550
    },
    {
      "entropy": 7.2049596309661865,
      "epoch": 0.15544559404208527,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999652522202125,
      "loss": 7.106,
      "mean_token_accuracy": 0.11494272649288177,
      "num_tokens": 3056640.0,
      "step": 1555
    },
    {
      "entropy": 7.144407796859741,
      "epoch": 0.15594541910331383,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999646221900839,
      "loss": 7.1441,
      "mean_token_accuracy": 0.11316709145903588,
      "num_tokens": 3066938.0,
      "step": 1560
    },
    {
      "entropy": 7.216827011108398,
      "epoch": 0.15644524416454242,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999639864999034,
      "loss": 7.2439,
      "mean_token_accuracy": 0.10790825858712197,
      "num_tokens": 3076791.0,
      "step": 1565
    },
    {
      "entropy": 7.263591861724853,
      "epoch": 0.15694506922577098,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499963345149687,
      "loss": 7.0932,
      "mean_token_accuracy": 0.10904844701290131,
      "num_tokens": 3087532.0,
      "step": 1570
    },
    {
      "entropy": 7.110728025436401,
      "epoch": 0.15744489428699954,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999626981394506,
      "loss": 7.2289,
      "mean_token_accuracy": 0.10976007133722306,
      "num_tokens": 3097465.0,
      "step": 1575
    },
    {
      "entropy": 7.32912278175354,
      "epoch": 0.15794471934822812,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999620454692106,
      "loss": 7.1224,
      "mean_token_accuracy": 0.1091330036520958,
      "num_tokens": 3107162.0,
      "step": 1580
    },
    {
      "entropy": 7.141360473632813,
      "epoch": 0.15844454440945668,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999613871389836,
      "loss": 7.1238,
      "mean_token_accuracy": 0.10997193083167076,
      "num_tokens": 3117154.0,
      "step": 1585
    },
    {
      "entropy": 7.207948064804077,
      "epoch": 0.15894436947068527,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999607231487858,
      "loss": 7.1408,
      "mean_token_accuracy": 0.10777789875864982,
      "num_tokens": 3127635.0,
      "step": 1590
    },
    {
      "entropy": 7.305689430236816,
      "epoch": 0.15944419453191383,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499960053498634,
      "loss": 7.176,
      "mean_token_accuracy": 0.10682187527418137,
      "num_tokens": 3137806.0,
      "step": 1595
    },
    {
      "entropy": 7.146860361099243,
      "epoch": 0.1599440195931424,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999593781885454,
      "loss": 7.2055,
      "mean_token_accuracy": 0.10350432991981506,
      "num_tokens": 3147704.0,
      "step": 1600
    },
    {
      "entropy": 7.246462678909301,
      "epoch": 0.16044384465437098,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999586972185366,
      "loss": 7.1665,
      "mean_token_accuracy": 0.10513406619429588,
      "num_tokens": 3158486.0,
      "step": 1605
    },
    {
      "entropy": 7.2225347518920895,
      "epoch": 0.16094366971559954,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999580105886248,
      "loss": 7.1637,
      "mean_token_accuracy": 0.1100760817527771,
      "num_tokens": 3168518.0,
      "step": 1610
    },
    {
      "entropy": 7.136386346817017,
      "epoch": 0.16144349477682812,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999573182988275,
      "loss": 7.0505,
      "mean_token_accuracy": 0.11440540626645088,
      "num_tokens": 3177502.0,
      "step": 1615
    },
    {
      "entropy": 7.223326826095581,
      "epoch": 0.16194331983805668,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999566203491619,
      "loss": 7.1259,
      "mean_token_accuracy": 0.11024409383535386,
      "num_tokens": 3187408.0,
      "step": 1620
    },
    {
      "entropy": 7.209620571136474,
      "epoch": 0.16244314489928524,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999559167396455,
      "loss": 7.1748,
      "mean_token_accuracy": 0.11251615285873413,
      "num_tokens": 3196812.0,
      "step": 1625
    },
    {
      "entropy": 7.105884313583374,
      "epoch": 0.16294296996051383,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999552074702963,
      "loss": 7.0282,
      "mean_token_accuracy": 0.11647915989160537,
      "num_tokens": 3206159.0,
      "step": 1630
    },
    {
      "entropy": 7.154392719268799,
      "epoch": 0.1634427950217424,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999544925411318,
      "loss": 7.0193,
      "mean_token_accuracy": 0.11799696087837219,
      "num_tokens": 3217329.0,
      "step": 1635
    },
    {
      "entropy": 7.113078355789185,
      "epoch": 0.16394262008297095,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999537719521703,
      "loss": 7.1432,
      "mean_token_accuracy": 0.1093616746366024,
      "num_tokens": 3227558.0,
      "step": 1640
    },
    {
      "entropy": 7.271105813980102,
      "epoch": 0.16444244514419953,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999530457034297,
      "loss": 7.2335,
      "mean_token_accuracy": 0.10456183776259423,
      "num_tokens": 3237719.0,
      "step": 1645
    },
    {
      "entropy": 7.20871205329895,
      "epoch": 0.1649422702054281,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999523137949283,
      "loss": 7.1569,
      "mean_token_accuracy": 0.11061653196811676,
      "num_tokens": 3246483.0,
      "step": 1650
    },
    {
      "entropy": 7.111011981964111,
      "epoch": 0.16544209526665668,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999515762266846,
      "loss": 7.0442,
      "mean_token_accuracy": 0.11448038592934609,
      "num_tokens": 3256955.0,
      "step": 1655
    },
    {
      "entropy": 7.170991039276123,
      "epoch": 0.16594192032788524,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999508329987173,
      "loss": 7.1342,
      "mean_token_accuracy": 0.1075492449104786,
      "num_tokens": 3266799.0,
      "step": 1660
    },
    {
      "entropy": 7.175664854049683,
      "epoch": 0.1664417453891138,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999500841110447,
      "loss": 7.1038,
      "mean_token_accuracy": 0.10848242715001107,
      "num_tokens": 3276595.0,
      "step": 1665
    },
    {
      "entropy": 7.121126556396485,
      "epoch": 0.16694157045034239,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499949329563686,
      "loss": 7.0484,
      "mean_token_accuracy": 0.11261371076107025,
      "num_tokens": 3286047.0,
      "step": 1670
    },
    {
      "entropy": 7.220827436447143,
      "epoch": 0.16744139551157095,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049994856935666,
      "loss": 7.1485,
      "mean_token_accuracy": 0.10454296097159385,
      "num_tokens": 3295364.0,
      "step": 1675
    },
    {
      "entropy": 7.113693380355835,
      "epoch": 0.16794122057279953,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499947803489986,
      "loss": 7.0472,
      "mean_token_accuracy": 0.11581237241625786,
      "num_tokens": 3305319.0,
      "step": 1680
    },
    {
      "entropy": 7.098750734329224,
      "epoch": 0.1684410456340281,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999470319636829,
      "loss": 7.0623,
      "mean_token_accuracy": 0.11507118344306946,
      "num_tokens": 3315349.0,
      "step": 1685
    },
    {
      "entropy": 7.151982069015503,
      "epoch": 0.16894087069525665,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999462547777705,
      "loss": 7.1436,
      "mean_token_accuracy": 0.10506928116083145,
      "num_tokens": 3325603.0,
      "step": 1690
    },
    {
      "entropy": 7.133093595504761,
      "epoch": 0.16944069575648524,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999454719322682,
      "loss": 7.0563,
      "mean_token_accuracy": 0.11085285693407058,
      "num_tokens": 3334693.0,
      "step": 1695
    },
    {
      "entropy": 7.1613341808319095,
      "epoch": 0.1699405208177138,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999446834271958,
      "loss": 7.0656,
      "mean_token_accuracy": 0.10509323179721833,
      "num_tokens": 3345259.0,
      "step": 1700
    },
    {
      "entropy": 7.052835845947266,
      "epoch": 0.17044034587894236,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999438892625728,
      "loss": 7.0619,
      "mean_token_accuracy": 0.108848088234663,
      "num_tokens": 3354072.0,
      "step": 1705
    },
    {
      "entropy": 7.273435640335083,
      "epoch": 0.17094017094017094,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999430894384198,
      "loss": 7.0622,
      "mean_token_accuracy": 0.10689497664570809,
      "num_tokens": 3364881.0,
      "step": 1710
    },
    {
      "entropy": 7.050932312011719,
      "epoch": 0.1714399960013995,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999422839547564,
      "loss": 7.0324,
      "mean_token_accuracy": 0.1095457673072815,
      "num_tokens": 3374030.0,
      "step": 1715
    },
    {
      "entropy": 7.092038965225219,
      "epoch": 0.1719398210626281,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999414728116029,
      "loss": 7.0593,
      "mean_token_accuracy": 0.10611691325902939,
      "num_tokens": 3384009.0,
      "step": 1720
    },
    {
      "entropy": 7.284945869445801,
      "epoch": 0.17243964612385665,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999406560089798,
      "loss": 7.1744,
      "mean_token_accuracy": 0.10681680515408516,
      "num_tokens": 3393660.0,
      "step": 1725
    },
    {
      "entropy": 7.17227029800415,
      "epoch": 0.1729394711850852,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999398335469078,
      "loss": 7.1897,
      "mean_token_accuracy": 0.1081137239933014,
      "num_tokens": 3404364.0,
      "step": 1730
    },
    {
      "entropy": 7.234745454788208,
      "epoch": 0.1734392962463138,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999390054254074,
      "loss": 7.0754,
      "mean_token_accuracy": 0.10843142941594124,
      "num_tokens": 3412990.0,
      "step": 1735
    },
    {
      "entropy": 6.947238445281982,
      "epoch": 0.17393912130754235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999381716444996,
      "loss": 7.0372,
      "mean_token_accuracy": 0.11110711246728897,
      "num_tokens": 3422339.0,
      "step": 1740
    },
    {
      "entropy": 7.149367761611939,
      "epoch": 0.17443894636877094,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999373322042052,
      "loss": 7.0285,
      "mean_token_accuracy": 0.11542280092835426,
      "num_tokens": 3433093.0,
      "step": 1745
    },
    {
      "entropy": 7.15626335144043,
      "epoch": 0.1749387714299995,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999364871045455,
      "loss": 7.1192,
      "mean_token_accuracy": 0.10659995451569557,
      "num_tokens": 3443464.0,
      "step": 1750
    },
    {
      "entropy": 7.114335584640503,
      "epoch": 0.17543859649122806,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999356363455415,
      "loss": 7.0495,
      "mean_token_accuracy": 0.10678325593471527,
      "num_tokens": 3453380.0,
      "step": 1755
    },
    {
      "entropy": 7.190001773834228,
      "epoch": 0.17593842155245665,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499934779927215,
      "loss": 7.1148,
      "mean_token_accuracy": 0.11223250329494476,
      "num_tokens": 3463583.0,
      "step": 1760
    },
    {
      "entropy": 7.0208066463470455,
      "epoch": 0.1764382466136852,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999339178495872,
      "loss": 7.0105,
      "mean_token_accuracy": 0.11592019125819206,
      "num_tokens": 3473640.0,
      "step": 1765
    },
    {
      "entropy": 7.113274097442627,
      "epoch": 0.1769380716749138,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999330501126798,
      "loss": 7.0769,
      "mean_token_accuracy": 0.11162511184811592,
      "num_tokens": 3483827.0,
      "step": 1770
    },
    {
      "entropy": 7.152011203765869,
      "epoch": 0.17743789673614235,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999321767165148,
      "loss": 7.0245,
      "mean_token_accuracy": 0.11725858971476555,
      "num_tokens": 3493763.0,
      "step": 1775
    },
    {
      "entropy": 7.105683755874634,
      "epoch": 0.1779377217973709,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999312976611142,
      "loss": 7.1331,
      "mean_token_accuracy": 0.10835669562220573,
      "num_tokens": 3503119.0,
      "step": 1780
    },
    {
      "entropy": 7.186274528503418,
      "epoch": 0.1784375468585995,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999304129465,
      "loss": 7.0585,
      "mean_token_accuracy": 0.10770363509654998,
      "num_tokens": 3513235.0,
      "step": 1785
    },
    {
      "entropy": 7.211988544464111,
      "epoch": 0.17893737191982806,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999295225726945,
      "loss": 7.1643,
      "mean_token_accuracy": 0.10430835708975791,
      "num_tokens": 3523633.0,
      "step": 1790
    },
    {
      "entropy": 7.082215309143066,
      "epoch": 0.17943719698105662,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00049992862653972,
      "loss": 7.0093,
      "mean_token_accuracy": 0.11031374111771583,
      "num_tokens": 3532888.0,
      "step": 1795
    },
    {
      "entropy": 7.140625905990601,
      "epoch": 0.1799370220422852,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999277248475992,
      "loss": 7.0527,
      "mean_token_accuracy": 0.11224330216646194,
      "num_tokens": 3541942.0,
      "step": 1800
    },
    {
      "entropy": 7.046121978759766,
      "epoch": 0.18043684710351376,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999268174963547,
      "loss": 7.0113,
      "mean_token_accuracy": 0.10969708785414696,
      "num_tokens": 3551544.0,
      "step": 1805
    },
    {
      "entropy": 7.076622200012207,
      "epoch": 0.18093667216474235,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999259044860093,
      "loss": 7.0036,
      "mean_token_accuracy": 0.11504804566502572,
      "num_tokens": 3561024.0,
      "step": 1810
    },
    {
      "entropy": 7.142001152038574,
      "epoch": 0.1814364972259709,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999249858165861,
      "loss": 6.9931,
      "mean_token_accuracy": 0.11953055635094642,
      "num_tokens": 3570962.0,
      "step": 1815
    },
    {
      "entropy": 6.922073125839233,
      "epoch": 0.18193632228719947,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999240614881082,
      "loss": 6.9879,
      "mean_token_accuracy": 0.11443007588386536,
      "num_tokens": 3580669.0,
      "step": 1820
    },
    {
      "entropy": 7.153632020950317,
      "epoch": 0.18243614734842806,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999231315005987,
      "loss": 7.0082,
      "mean_token_accuracy": 0.11323033571243286,
      "num_tokens": 3590627.0,
      "step": 1825
    },
    {
      "entropy": 7.0444121837615965,
      "epoch": 0.18293597240965662,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999221958540811,
      "loss": 7.0866,
      "mean_token_accuracy": 0.10459914803504944,
      "num_tokens": 3599872.0,
      "step": 1830
    },
    {
      "entropy": 7.135852479934693,
      "epoch": 0.1834357974708852,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999212545485789,
      "loss": 7.0671,
      "mean_token_accuracy": 0.11241727098822593,
      "num_tokens": 3609809.0,
      "step": 1835
    },
    {
      "entropy": 7.2499583721160885,
      "epoch": 0.18393562253211376,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999203075841159,
      "loss": 7.0773,
      "mean_token_accuracy": 0.11519879400730133,
      "num_tokens": 3619385.0,
      "step": 1840
    },
    {
      "entropy": 6.968499565124512,
      "epoch": 0.18443544759334232,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999193549607157,
      "loss": 6.9203,
      "mean_token_accuracy": 0.12208465933799743,
      "num_tokens": 3628867.0,
      "step": 1845
    },
    {
      "entropy": 7.032136726379394,
      "epoch": 0.1849352726545709,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999183966784026,
      "loss": 6.9271,
      "mean_token_accuracy": 0.11781778261065483,
      "num_tokens": 3637528.0,
      "step": 1850
    },
    {
      "entropy": 7.131372499465942,
      "epoch": 0.18543509771579947,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999174327372004,
      "loss": 7.0523,
      "mean_token_accuracy": 0.10981304794549943,
      "num_tokens": 3648011.0,
      "step": 1855
    },
    {
      "entropy": 7.092602205276489,
      "epoch": 0.18593492277702803,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999164631371335,
      "loss": 7.0339,
      "mean_token_accuracy": 0.11578032821416855,
      "num_tokens": 3657648.0,
      "step": 1860
    },
    {
      "entropy": 7.072668981552124,
      "epoch": 0.18643474783825661,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999154878782262,
      "loss": 7.0437,
      "mean_token_accuracy": 0.11318779215216637,
      "num_tokens": 3666965.0,
      "step": 1865
    },
    {
      "entropy": 7.153492784500122,
      "epoch": 0.18693457289948517,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499914506960503,
      "loss": 7.0735,
      "mean_token_accuracy": 0.10511444881558418,
      "num_tokens": 3676609.0,
      "step": 1870
    },
    {
      "entropy": 7.012405109405518,
      "epoch": 0.18743439796071376,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999135203839889,
      "loss": 6.9304,
      "mean_token_accuracy": 0.11578108370304108,
      "num_tokens": 3685884.0,
      "step": 1875
    },
    {
      "entropy": 7.043309020996094,
      "epoch": 0.18793422302194232,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999125281487084,
      "loss": 7.0879,
      "mean_token_accuracy": 0.11302101239562035,
      "num_tokens": 3695436.0,
      "step": 1880
    },
    {
      "entropy": 7.1143311023712155,
      "epoch": 0.18843404808317088,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999115302546866,
      "loss": 6.9076,
      "mean_token_accuracy": 0.12109773308038711,
      "num_tokens": 3704806.0,
      "step": 1885
    },
    {
      "entropy": 6.979010534286499,
      "epoch": 0.18893387314439947,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004999105267019486,
      "loss": 6.9918,
      "mean_token_accuracy": 0.11695808097720146,
      "num_tokens": 3713927.0,
      "step": 1890
    },
    {
      "entropy": 7.1983567237854,
      "epoch": 0.18943369820562803,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004999095174905195,
      "loss": 7.0666,
      "mean_token_accuracy": 0.10940969362854958,
      "num_tokens": 3723685.0,
      "step": 1895
    },
    {
      "entropy": 6.993364381790161,
      "epoch": 0.1899335232668566,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999085026204249,
      "loss": 7.0327,
      "mean_token_accuracy": 0.11247843354940415,
      "num_tokens": 3733918.0,
      "step": 1900
    },
    {
      "entropy": 7.085520219802857,
      "epoch": 0.19043334832808517,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999074820916903,
      "loss": 6.9868,
      "mean_token_accuracy": 0.11154783442616463,
      "num_tokens": 3744466.0,
      "step": 1905
    },
    {
      "entropy": 7.1121289253234865,
      "epoch": 0.19093317338931373,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999064559043412,
      "loss": 6.9372,
      "mean_token_accuracy": 0.11101839020848274,
      "num_tokens": 3754082.0,
      "step": 1910
    },
    {
      "entropy": 7.008188533782959,
      "epoch": 0.19143299845054232,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999054240584037,
      "loss": 6.9694,
      "mean_token_accuracy": 0.11511853858828544,
      "num_tokens": 3763111.0,
      "step": 1915
    },
    {
      "entropy": 7.1465226173400875,
      "epoch": 0.19193282351177088,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999043865539035,
      "loss": 7.0187,
      "mean_token_accuracy": 0.10961114466190339,
      "num_tokens": 3772555.0,
      "step": 1920
    },
    {
      "entropy": 7.155064296722412,
      "epoch": 0.19243264857299944,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999033433908667,
      "loss": 7.0649,
      "mean_token_accuracy": 0.1120420217514038,
      "num_tokens": 3783073.0,
      "step": 1925
    },
    {
      "entropy": 6.897677850723267,
      "epoch": 0.19293247363422802,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999022945693198,
      "loss": 7.0001,
      "mean_token_accuracy": 0.1147444948554039,
      "num_tokens": 3792277.0,
      "step": 1930
    },
    {
      "entropy": 7.110321664810181,
      "epoch": 0.19343229869545658,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499901240089289,
      "loss": 6.9678,
      "mean_token_accuracy": 0.11609976217150689,
      "num_tokens": 3801098.0,
      "step": 1935
    },
    {
      "entropy": 7.104807186126709,
      "epoch": 0.19393212375668517,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999001799508009,
      "loss": 6.9855,
      "mean_token_accuracy": 0.11322086304426193,
      "num_tokens": 3810614.0,
      "step": 1940
    },
    {
      "entropy": 7.023803997039795,
      "epoch": 0.19443194881791373,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998991141538821,
      "loss": 6.9879,
      "mean_token_accuracy": 0.11285669654607773,
      "num_tokens": 3818989.0,
      "step": 1945
    },
    {
      "entropy": 7.011999940872192,
      "epoch": 0.1949317738791423,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998980426985596,
      "loss": 7.0185,
      "mean_token_accuracy": 0.11454365253448487,
      "num_tokens": 3829319.0,
      "step": 1950
    },
    {
      "entropy": 7.093687629699707,
      "epoch": 0.19543159894037088,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00049989696558486,
      "loss": 7.052,
      "mean_token_accuracy": 0.10922376289963723,
      "num_tokens": 3839545.0,
      "step": 1955
    },
    {
      "entropy": 7.0437132835388185,
      "epoch": 0.19593142400159944,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998958828128108,
      "loss": 6.982,
      "mean_token_accuracy": 0.11502454653382302,
      "num_tokens": 3849029.0,
      "step": 1960
    },
    {
      "entropy": 7.036665391921997,
      "epoch": 0.19643124906282802,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998947943824389,
      "loss": 7.0307,
      "mean_token_accuracy": 0.11234677508473397,
      "num_tokens": 3859463.0,
      "step": 1965
    },
    {
      "entropy": 7.178880786895752,
      "epoch": 0.19693107412405658,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998937002937719,
      "loss": 7.0665,
      "mean_token_accuracy": 0.1138896718621254,
      "num_tokens": 3869695.0,
      "step": 1970
    },
    {
      "entropy": 7.052155542373657,
      "epoch": 0.19743089918528514,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998926005468373,
      "loss": 7.0182,
      "mean_token_accuracy": 0.11386001110076904,
      "num_tokens": 3879608.0,
      "step": 1975
    },
    {
      "entropy": 7.002656650543213,
      "epoch": 0.19793072424651373,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998914951416627,
      "loss": 6.9328,
      "mean_token_accuracy": 0.11322358548641205,
      "num_tokens": 3888709.0,
      "step": 1980
    },
    {
      "entropy": 6.9356893539428714,
      "epoch": 0.1984305493077423,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499890384078276,
      "loss": 6.8784,
      "mean_token_accuracy": 0.11385609433054925,
      "num_tokens": 3898392.0,
      "step": 1985
    },
    {
      "entropy": 7.109602928161621,
      "epoch": 0.19893037436897085,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998892673567051,
      "loss": 7.0962,
      "mean_token_accuracy": 0.10173079818487167,
      "num_tokens": 3909508.0,
      "step": 1990
    },
    {
      "entropy": 7.016517782211304,
      "epoch": 0.19943019943019943,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004998881449769781,
      "loss": 6.9499,
      "mean_token_accuracy": 0.11535055190324783,
      "num_tokens": 3918155.0,
      "step": 1995
    },
    {
      "entropy": 7.097119283676148,
      "epoch": 0.199930024491428,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998870169391232,
      "loss": 6.9733,
      "mean_token_accuracy": 0.11269012987613677,
      "num_tokens": 3928478.0,
      "step": 2000
    },
    {
      "entropy": 7.00243239402771,
      "epoch": 0.20042984955265658,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499885883243169,
      "loss": 7.0207,
      "mean_token_accuracy": 0.12159764543175697,
      "num_tokens": 3938126.0,
      "step": 2005
    },
    {
      "entropy": 6.912168693542481,
      "epoch": 0.20092967461388514,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998847438891437,
      "loss": 6.8656,
      "mean_token_accuracy": 0.11626059859991074,
      "num_tokens": 3948176.0,
      "step": 2010
    },
    {
      "entropy": 7.11488447189331,
      "epoch": 0.2014294996751137,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998835988770761,
      "loss": 6.9832,
      "mean_token_accuracy": 0.11340783983469009,
      "num_tokens": 3958983.0,
      "step": 2015
    },
    {
      "entropy": 7.037908840179443,
      "epoch": 0.20192932473634229,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499882448206995,
      "loss": 6.9383,
      "mean_token_accuracy": 0.11477451324462891,
      "num_tokens": 3968766.0,
      "step": 2020
    },
    {
      "entropy": 6.995463895797729,
      "epoch": 0.20242914979757085,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998812918789293,
      "loss": 6.9589,
      "mean_token_accuracy": 0.11554277166724206,
      "num_tokens": 3979487.0,
      "step": 2025
    },
    {
      "entropy": 7.027395153045655,
      "epoch": 0.20292897485879943,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998801298929084,
      "loss": 6.9399,
      "mean_token_accuracy": 0.11524380594491959,
      "num_tokens": 3988450.0,
      "step": 2030
    },
    {
      "entropy": 7.013701009750366,
      "epoch": 0.203428799920028,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998789622489611,
      "loss": 6.9426,
      "mean_token_accuracy": 0.11618846654891968,
      "num_tokens": 3997492.0,
      "step": 2035
    },
    {
      "entropy": 7.018197727203369,
      "epoch": 0.20392862498125655,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998777889471171,
      "loss": 6.9379,
      "mean_token_accuracy": 0.11913173943758011,
      "num_tokens": 4007298.0,
      "step": 2040
    },
    {
      "entropy": 7.063361120223999,
      "epoch": 0.20442845004248514,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998766099874056,
      "loss": 6.9694,
      "mean_token_accuracy": 0.11549432277679443,
      "num_tokens": 4017727.0,
      "step": 2045
    },
    {
      "entropy": 6.956744623184204,
      "epoch": 0.2049282751037137,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998754253698566,
      "loss": 6.9688,
      "mean_token_accuracy": 0.11394633799791336,
      "num_tokens": 4027618.0,
      "step": 2050
    },
    {
      "entropy": 7.018502855300904,
      "epoch": 0.20542810016494226,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998742350944998,
      "loss": 6.988,
      "mean_token_accuracy": 0.1113269604742527,
      "num_tokens": 4038831.0,
      "step": 2055
    },
    {
      "entropy": 7.088349199295044,
      "epoch": 0.20592792522617084,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998730391613651,
      "loss": 6.9536,
      "mean_token_accuracy": 0.11713754236698151,
      "num_tokens": 4048736.0,
      "step": 2060
    },
    {
      "entropy": 6.927730417251587,
      "epoch": 0.2064277502873994,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998718375704825,
      "loss": 6.8668,
      "mean_token_accuracy": 0.11648354679346085,
      "num_tokens": 4057633.0,
      "step": 2065
    },
    {
      "entropy": 7.0192708492279055,
      "epoch": 0.206927575348628,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998706303218825,
      "loss": 6.988,
      "mean_token_accuracy": 0.1224412478506565,
      "num_tokens": 4067883.0,
      "step": 2070
    },
    {
      "entropy": 7.048581552505493,
      "epoch": 0.20742740040985655,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998694174155952,
      "loss": 6.8595,
      "mean_token_accuracy": 0.12078917920589446,
      "num_tokens": 4078147.0,
      "step": 2075
    },
    {
      "entropy": 6.922640562057495,
      "epoch": 0.2079272254710851,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998681988516512,
      "loss": 6.8328,
      "mean_token_accuracy": 0.12206020206212997,
      "num_tokens": 4087130.0,
      "step": 2080
    },
    {
      "entropy": 6.915320634841919,
      "epoch": 0.2084270505323137,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998669746300811,
      "loss": 6.9515,
      "mean_token_accuracy": 0.12251704409718514,
      "num_tokens": 4096840.0,
      "step": 2085
    },
    {
      "entropy": 7.133029270172119,
      "epoch": 0.20892687559354226,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998657447509159,
      "loss": 6.9766,
      "mean_token_accuracy": 0.11555029898881912,
      "num_tokens": 4105842.0,
      "step": 2090
    },
    {
      "entropy": 6.949610185623169,
      "epoch": 0.20942670065477084,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998645092141864,
      "loss": 6.8914,
      "mean_token_accuracy": 0.11520621851086617,
      "num_tokens": 4114865.0,
      "step": 2095
    },
    {
      "entropy": 7.083245754241943,
      "epoch": 0.2099265257159994,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998632680199238,
      "loss": 7.0161,
      "mean_token_accuracy": 0.10906583741307259,
      "num_tokens": 4124954.0,
      "step": 2100
    },
    {
      "entropy": 6.9515073776245115,
      "epoch": 0.21042635077722796,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998620211681591,
      "loss": 6.8914,
      "mean_token_accuracy": 0.12042388617992401,
      "num_tokens": 4134091.0,
      "step": 2105
    },
    {
      "entropy": 7.048494529724121,
      "epoch": 0.21092617583845655,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998607686589239,
      "loss": 6.9573,
      "mean_token_accuracy": 0.11529561281204223,
      "num_tokens": 4143566.0,
      "step": 2110
    },
    {
      "entropy": 7.024812936782837,
      "epoch": 0.2114260008996851,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004998595104922496,
      "loss": 6.9208,
      "mean_token_accuracy": 0.11166013404726982,
      "num_tokens": 4153053.0,
      "step": 2115
    },
    {
      "entropy": 6.9540647029876705,
      "epoch": 0.2119258259609137,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998582466681678,
      "loss": 6.896,
      "mean_token_accuracy": 0.12232667654752731,
      "num_tokens": 4162925.0,
      "step": 2120
    },
    {
      "entropy": 6.9011767387390135,
      "epoch": 0.21242565102214225,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998569771867105,
      "loss": 6.8292,
      "mean_token_accuracy": 0.1230908177793026,
      "num_tokens": 4172077.0,
      "step": 2125
    },
    {
      "entropy": 6.9595249652862545,
      "epoch": 0.2129254760833708,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998557020479095,
      "loss": 6.9244,
      "mean_token_accuracy": 0.11470091789960861,
      "num_tokens": 4182134.0,
      "step": 2130
    },
    {
      "entropy": 7.040814352035523,
      "epoch": 0.2134253011445994,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998544212517968,
      "loss": 6.9441,
      "mean_token_accuracy": 0.11132864132523537,
      "num_tokens": 4191963.0,
      "step": 2135
    },
    {
      "entropy": 6.906561183929443,
      "epoch": 0.21392512620582796,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998531347984047,
      "loss": 6.8927,
      "mean_token_accuracy": 0.11642092615365982,
      "num_tokens": 4202180.0,
      "step": 2140
    },
    {
      "entropy": 7.05922155380249,
      "epoch": 0.21442495126705652,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998518426877656,
      "loss": 6.924,
      "mean_token_accuracy": 0.11686432883143424,
      "num_tokens": 4211874.0,
      "step": 2145
    },
    {
      "entropy": 6.941175365447998,
      "epoch": 0.2149247763282851,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998505449199122,
      "loss": 6.8725,
      "mean_token_accuracy": 0.11368393003940583,
      "num_tokens": 4221610.0,
      "step": 2150
    },
    {
      "entropy": 6.828761768341065,
      "epoch": 0.21542460138951366,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004998492414948768,
      "loss": 6.831,
      "mean_token_accuracy": 0.12876630201935768,
      "num_tokens": 4230610.0,
      "step": 2155
    },
    {
      "entropy": 6.962120294570923,
      "epoch": 0.21592442645074225,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004998479324126924,
      "loss": 6.9461,
      "mean_token_accuracy": 0.11206507235765457,
      "num_tokens": 4241389.0,
      "step": 2160
    },
    {
      "entropy": 7.020627546310425,
      "epoch": 0.2164242515119708,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998466176733918,
      "loss": 6.8901,
      "mean_token_accuracy": 0.11771368458867074,
      "num_tokens": 4250617.0,
      "step": 2165
    },
    {
      "entropy": 6.930497503280639,
      "epoch": 0.21692407657319937,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998452972770082,
      "loss": 6.8774,
      "mean_token_accuracy": 0.12080463096499443,
      "num_tokens": 4260000.0,
      "step": 2170
    },
    {
      "entropy": 6.881647682189941,
      "epoch": 0.21742390163442796,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998439712235747,
      "loss": 6.8233,
      "mean_token_accuracy": 0.11715791299939156,
      "num_tokens": 4269751.0,
      "step": 2175
    },
    {
      "entropy": 6.977957820892334,
      "epoch": 0.21792372669565652,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998426395131248,
      "loss": 6.8898,
      "mean_token_accuracy": 0.11791151389479637,
      "num_tokens": 4279210.0,
      "step": 2180
    },
    {
      "entropy": 7.000884628295898,
      "epoch": 0.2184235517568851,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998413021456919,
      "loss": 6.84,
      "mean_token_accuracy": 0.1202585093677044,
      "num_tokens": 4288573.0,
      "step": 2185
    },
    {
      "entropy": 6.901773357391358,
      "epoch": 0.21892337681811366,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998399591213098,
      "loss": 6.8756,
      "mean_token_accuracy": 0.11356504261493683,
      "num_tokens": 4298438.0,
      "step": 2190
    },
    {
      "entropy": 6.925487279891968,
      "epoch": 0.21942320187934222,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998386104400122,
      "loss": 6.7819,
      "mean_token_accuracy": 0.12836189493536948,
      "num_tokens": 4307174.0,
      "step": 2195
    },
    {
      "entropy": 7.067652225494385,
      "epoch": 0.2199230269405708,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998372561018329,
      "loss": 6.9745,
      "mean_token_accuracy": 0.1136770062148571,
      "num_tokens": 4317220.0,
      "step": 2200
    },
    {
      "entropy": 6.984744215011597,
      "epoch": 0.22042285200179937,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998358961068062,
      "loss": 6.8837,
      "mean_token_accuracy": 0.12194864749908448,
      "num_tokens": 4326496.0,
      "step": 2205
    },
    {
      "entropy": 6.905305337905884,
      "epoch": 0.22092267706302793,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998345304549661,
      "loss": 6.8797,
      "mean_token_accuracy": 0.11901782974600791,
      "num_tokens": 4335705.0,
      "step": 2210
    },
    {
      "entropy": 7.03793044090271,
      "epoch": 0.22142250212425652,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998331591463471,
      "loss": 6.9112,
      "mean_token_accuracy": 0.1200287975370884,
      "num_tokens": 4344711.0,
      "step": 2215
    },
    {
      "entropy": 6.928197145462036,
      "epoch": 0.22192232718548507,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998317821809837,
      "loss": 6.9239,
      "mean_token_accuracy": 0.11762578636407853,
      "num_tokens": 4355448.0,
      "step": 2220
    },
    {
      "entropy": 6.843606281280517,
      "epoch": 0.22242215224671366,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998303995589104,
      "loss": 6.7873,
      "mean_token_accuracy": 0.12028947919607162,
      "num_tokens": 4365104.0,
      "step": 2225
    },
    {
      "entropy": 6.9312232494354244,
      "epoch": 0.22292197730794222,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998290112801623,
      "loss": 6.7701,
      "mean_token_accuracy": 0.12458315268158912,
      "num_tokens": 4375219.0,
      "step": 2230
    },
    {
      "entropy": 6.943724775314331,
      "epoch": 0.22342180236917078,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499827617344774,
      "loss": 6.9059,
      "mean_token_accuracy": 0.11273334845900536,
      "num_tokens": 4385294.0,
      "step": 2235
    },
    {
      "entropy": 7.021584129333496,
      "epoch": 0.22392162743039937,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998262177527807,
      "loss": 6.8766,
      "mean_token_accuracy": 0.11276980042457581,
      "num_tokens": 4394305.0,
      "step": 2240
    },
    {
      "entropy": 6.873253679275512,
      "epoch": 0.22442145249162793,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998248125042177,
      "loss": 6.7415,
      "mean_token_accuracy": 0.12254593893885612,
      "num_tokens": 4403732.0,
      "step": 2245
    },
    {
      "entropy": 6.929276704788208,
      "epoch": 0.2249212775528565,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998234015991202,
      "loss": 6.9641,
      "mean_token_accuracy": 0.11173295974731445,
      "num_tokens": 4413730.0,
      "step": 2250
    },
    {
      "entropy": 7.033787107467651,
      "epoch": 0.22542110261408507,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998219850375238,
      "loss": 6.9852,
      "mean_token_accuracy": 0.11005716025829315,
      "num_tokens": 4423051.0,
      "step": 2255
    },
    {
      "entropy": 7.026523208618164,
      "epoch": 0.22592092767531363,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998205628194641,
      "loss": 6.9691,
      "mean_token_accuracy": 0.11586882919073105,
      "num_tokens": 4433363.0,
      "step": 2260
    },
    {
      "entropy": 6.950811290740967,
      "epoch": 0.22642075273654222,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004998191349449768,
      "loss": 6.9189,
      "mean_token_accuracy": 0.11593685448169708,
      "num_tokens": 4442984.0,
      "step": 2265
    },
    {
      "entropy": 6.989126873016358,
      "epoch": 0.22692057779777078,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998177014140981,
      "loss": 6.865,
      "mean_token_accuracy": 0.11428796350955964,
      "num_tokens": 4452364.0,
      "step": 2270
    },
    {
      "entropy": 7.007836103439331,
      "epoch": 0.22742040285899934,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998162622268637,
      "loss": 6.8918,
      "mean_token_accuracy": 0.11280409172177315,
      "num_tokens": 4461904.0,
      "step": 2275
    },
    {
      "entropy": 6.916662788391113,
      "epoch": 0.22792022792022792,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998148173833101,
      "loss": 6.838,
      "mean_token_accuracy": 0.11601078361272812,
      "num_tokens": 4470633.0,
      "step": 2280
    },
    {
      "entropy": 6.945051670074463,
      "epoch": 0.22842005298145648,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998133668834735,
      "loss": 6.8637,
      "mean_token_accuracy": 0.11577488556504249,
      "num_tokens": 4480673.0,
      "step": 2285
    },
    {
      "entropy": 6.933673238754272,
      "epoch": 0.22891987804268507,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998119107273905,
      "loss": 6.8411,
      "mean_token_accuracy": 0.11377951726317406,
      "num_tokens": 4489923.0,
      "step": 2290
    },
    {
      "entropy": 6.885636615753174,
      "epoch": 0.22941970310391363,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998104489150975,
      "loss": 6.8218,
      "mean_token_accuracy": 0.11800332292914391,
      "num_tokens": 4501007.0,
      "step": 2295
    },
    {
      "entropy": 6.897295141220093,
      "epoch": 0.2299195281651422,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998089814466318,
      "loss": 6.8368,
      "mean_token_accuracy": 0.11717171370983123,
      "num_tokens": 4510460.0,
      "step": 2300
    },
    {
      "entropy": 6.949653673171997,
      "epoch": 0.23041935322637078,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998075083220295,
      "loss": 6.9083,
      "mean_token_accuracy": 0.11965522542595863,
      "num_tokens": 4521026.0,
      "step": 2305
    },
    {
      "entropy": 6.94604001045227,
      "epoch": 0.23091917828759934,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998060295413286,
      "loss": 6.8298,
      "mean_token_accuracy": 0.12275644540786743,
      "num_tokens": 4530818.0,
      "step": 2310
    },
    {
      "entropy": 6.833601999282837,
      "epoch": 0.23141900334882792,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998045451045656,
      "loss": 6.7673,
      "mean_token_accuracy": 0.129067625105381,
      "num_tokens": 4539512.0,
      "step": 2315
    },
    {
      "entropy": 6.991441965103149,
      "epoch": 0.23191882841005648,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998030550117782,
      "loss": 6.9527,
      "mean_token_accuracy": 0.10756491348147393,
      "num_tokens": 4549030.0,
      "step": 2320
    },
    {
      "entropy": 6.9421521663665775,
      "epoch": 0.23241865347128504,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998015592630037,
      "loss": 6.8029,
      "mean_token_accuracy": 0.12596326917409897,
      "num_tokens": 4559347.0,
      "step": 2325
    },
    {
      "entropy": 6.901503849029541,
      "epoch": 0.23291847853251363,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004998000578582798,
      "loss": 6.8526,
      "mean_token_accuracy": 0.11550995856523513,
      "num_tokens": 4568721.0,
      "step": 2330
    },
    {
      "entropy": 6.883442735671997,
      "epoch": 0.2334183035937422,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997985507976443,
      "loss": 6.8582,
      "mean_token_accuracy": 0.11971477419137955,
      "num_tokens": 4578265.0,
      "step": 2335
    },
    {
      "entropy": 6.883966016769409,
      "epoch": 0.23391812865497075,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997970380811351,
      "loss": 6.8159,
      "mean_token_accuracy": 0.12598463222384454,
      "num_tokens": 4588694.0,
      "step": 2340
    },
    {
      "entropy": 6.959873485565185,
      "epoch": 0.23441795371619933,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997955197087904,
      "loss": 6.819,
      "mean_token_accuracy": 0.1255710557103157,
      "num_tokens": 4597754.0,
      "step": 2345
    },
    {
      "entropy": 6.884046936035157,
      "epoch": 0.2349177787774279,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997939956806481,
      "loss": 6.8304,
      "mean_token_accuracy": 0.12266772165894509,
      "num_tokens": 4606874.0,
      "step": 2350
    },
    {
      "entropy": 6.87273941040039,
      "epoch": 0.23541760383865648,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997924659967467,
      "loss": 6.7216,
      "mean_token_accuracy": 0.12119160741567611,
      "num_tokens": 4615588.0,
      "step": 2355
    },
    {
      "entropy": 6.8096753597259525,
      "epoch": 0.23591742889988504,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997909306571248,
      "loss": 6.8858,
      "mean_token_accuracy": 0.12267793342471123,
      "num_tokens": 4625919.0,
      "step": 2360
    },
    {
      "entropy": 6.987924003601075,
      "epoch": 0.2364172539611136,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997893896618209,
      "loss": 6.8238,
      "mean_token_accuracy": 0.1198611631989479,
      "num_tokens": 4635539.0,
      "step": 2365
    },
    {
      "entropy": 6.882870531082153,
      "epoch": 0.2369170790223422,
      "grad_norm": 0.875,
      "learning_rate": 0.0004997878430108738,
      "loss": 6.8214,
      "mean_token_accuracy": 0.11646069437265397,
      "num_tokens": 4645833.0,
      "step": 2370
    },
    {
      "entropy": 6.859278011322021,
      "epoch": 0.23741690408357075,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004997862907043224,
      "loss": 6.9272,
      "mean_token_accuracy": 0.12224952504038811,
      "num_tokens": 4655166.0,
      "step": 2375
    },
    {
      "entropy": 6.9933693408966064,
      "epoch": 0.23791672914479933,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997847327422059,
      "loss": 6.8107,
      "mean_token_accuracy": 0.1250378169119358,
      "num_tokens": 4664884.0,
      "step": 2380
    },
    {
      "entropy": 6.858077096939087,
      "epoch": 0.2384165542060279,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997831691245632,
      "loss": 6.9125,
      "mean_token_accuracy": 0.11868099793791771,
      "num_tokens": 4675172.0,
      "step": 2385
    },
    {
      "entropy": 6.997228765487671,
      "epoch": 0.23891637926725645,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997815998514337,
      "loss": 6.8008,
      "mean_token_accuracy": 0.1240590751171112,
      "num_tokens": 4684588.0,
      "step": 2390
    },
    {
      "entropy": 6.890038633346558,
      "epoch": 0.23941620432848504,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997800249228572,
      "loss": 6.9168,
      "mean_token_accuracy": 0.12128965482115746,
      "num_tokens": 4694434.0,
      "step": 2395
    },
    {
      "entropy": 6.885320329666138,
      "epoch": 0.2399160293897136,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997784443388732,
      "loss": 6.8599,
      "mean_token_accuracy": 0.11919219568371772,
      "num_tokens": 4705080.0,
      "step": 2400
    },
    {
      "entropy": 7.015338325500489,
      "epoch": 0.24041585445094216,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004997768580995212,
      "loss": 6.9445,
      "mean_token_accuracy": 0.115231192111969,
      "num_tokens": 4714466.0,
      "step": 2405
    },
    {
      "entropy": 6.8764424324035645,
      "epoch": 0.24091567951217074,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997752662048413,
      "loss": 6.8364,
      "mean_token_accuracy": 0.1243220515549183,
      "num_tokens": 4724047.0,
      "step": 2410
    },
    {
      "entropy": 6.968001174926758,
      "epoch": 0.2414155045733993,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997736686548736,
      "loss": 6.8619,
      "mean_token_accuracy": 0.11686329171061516,
      "num_tokens": 4733153.0,
      "step": 2415
    },
    {
      "entropy": 6.93342227935791,
      "epoch": 0.2419153296346279,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004997720654496582,
      "loss": 6.7318,
      "mean_token_accuracy": 0.1258167415857315,
      "num_tokens": 4743239.0,
      "step": 2420
    },
    {
      "entropy": 6.811439895629883,
      "epoch": 0.24241515469585645,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997704565892355,
      "loss": 6.7928,
      "mean_token_accuracy": 0.11975357532501221,
      "num_tokens": 4753128.0,
      "step": 2425
    },
    {
      "entropy": 7.061700773239136,
      "epoch": 0.242914979757085,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004997688420736459,
      "loss": 7.041,
      "mean_token_accuracy": 0.10650341138243676,
      "num_tokens": 4764056.0,
      "step": 2430
    },
    {
      "entropy": 6.925562858581543,
      "epoch": 0.2434148048183136,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997672219029301,
      "loss": 6.8635,
      "mean_token_accuracy": 0.1168606698513031,
      "num_tokens": 4773354.0,
      "step": 2435
    },
    {
      "entropy": 6.838783550262451,
      "epoch": 0.24391462987954216,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997655960771288,
      "loss": 6.7888,
      "mean_token_accuracy": 0.11904187425971031,
      "num_tokens": 4783367.0,
      "step": 2440
    },
    {
      "entropy": 6.842602968215942,
      "epoch": 0.24441445494077074,
      "grad_norm": 1.59375,
      "learning_rate": 0.000499763964596283,
      "loss": 6.6707,
      "mean_token_accuracy": 0.13226605579257011,
      "num_tokens": 4793028.0,
      "step": 2445
    },
    {
      "entropy": 6.91138973236084,
      "epoch": 0.2449142800019993,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997623274604336,
      "loss": 6.8631,
      "mean_token_accuracy": 0.11812737509608269,
      "num_tokens": 4804032.0,
      "step": 2450
    },
    {
      "entropy": 6.952124071121216,
      "epoch": 0.24541410506322786,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499760684669622,
      "loss": 6.8866,
      "mean_token_accuracy": 0.12024495378136635,
      "num_tokens": 4813294.0,
      "step": 2455
    },
    {
      "entropy": 6.905491209030151,
      "epoch": 0.24591393012445645,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997590362238894,
      "loss": 6.7871,
      "mean_token_accuracy": 0.12213733345270157,
      "num_tokens": 4823461.0,
      "step": 2460
    },
    {
      "entropy": 6.783637142181396,
      "epoch": 0.246413755185685,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997573821232772,
      "loss": 6.8421,
      "mean_token_accuracy": 0.12206422239542007,
      "num_tokens": 4832464.0,
      "step": 2465
    },
    {
      "entropy": 6.889177703857422,
      "epoch": 0.24691358024691357,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997557223678272,
      "loss": 6.8056,
      "mean_token_accuracy": 0.11893631145358086,
      "num_tokens": 4842697.0,
      "step": 2470
    },
    {
      "entropy": 6.928496265411377,
      "epoch": 0.24741340530814215,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997540569575809,
      "loss": 6.8574,
      "mean_token_accuracy": 0.120989241451025,
      "num_tokens": 4853129.0,
      "step": 2475
    },
    {
      "entropy": 6.909238862991333,
      "epoch": 0.2479132303693707,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997523858925805,
      "loss": 6.903,
      "mean_token_accuracy": 0.11157827898859977,
      "num_tokens": 4864048.0,
      "step": 2480
    },
    {
      "entropy": 6.845878267288208,
      "epoch": 0.2484130554305993,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997507091728678,
      "loss": 6.8355,
      "mean_token_accuracy": 0.1195888414978981,
      "num_tokens": 4874368.0,
      "step": 2485
    },
    {
      "entropy": 6.974607992172241,
      "epoch": 0.24891288049182786,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997490267984853,
      "loss": 6.8057,
      "mean_token_accuracy": 0.12127717137336731,
      "num_tokens": 4884813.0,
      "step": 2490
    },
    {
      "entropy": 6.844524717330932,
      "epoch": 0.24941270555305642,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997473387694749,
      "loss": 6.882,
      "mean_token_accuracy": 0.11751366555690765,
      "num_tokens": 4895594.0,
      "step": 2495
    },
    {
      "entropy": 6.964616298675537,
      "epoch": 0.249912530614285,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997456450858793,
      "loss": 6.774,
      "mean_token_accuracy": 0.12162590026855469,
      "num_tokens": 4904922.0,
      "step": 2500
    },
    {
      "entropy": 6.8929506778717045,
      "epoch": 0.25041235567551356,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997439457477412,
      "loss": 6.7701,
      "mean_token_accuracy": 0.11948562785983086,
      "num_tokens": 4915009.0,
      "step": 2505
    },
    {
      "entropy": 6.884179592132568,
      "epoch": 0.25091218073674215,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997422407551032,
      "loss": 6.914,
      "mean_token_accuracy": 0.1151105597615242,
      "num_tokens": 4924916.0,
      "step": 2510
    },
    {
      "entropy": 6.864490032196045,
      "epoch": 0.2514120057979707,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997405301080082,
      "loss": 6.8192,
      "mean_token_accuracy": 0.11743849441409111,
      "num_tokens": 4935127.0,
      "step": 2515
    },
    {
      "entropy": 6.89161958694458,
      "epoch": 0.25191183085919927,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997388138064994,
      "loss": 6.8121,
      "mean_token_accuracy": 0.12027946040034294,
      "num_tokens": 4944749.0,
      "step": 2520
    },
    {
      "entropy": 6.839969301223755,
      "epoch": 0.25241165592042786,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997370918506198,
      "loss": 6.6873,
      "mean_token_accuracy": 0.13174954876303674,
      "num_tokens": 4953868.0,
      "step": 2525
    },
    {
      "entropy": 6.85071587562561,
      "epoch": 0.25291148098165644,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997353642404129,
      "loss": 6.781,
      "mean_token_accuracy": 0.1238086350262165,
      "num_tokens": 4963282.0,
      "step": 2530
    },
    {
      "entropy": 6.700375747680664,
      "epoch": 0.253411306042885,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997336309759219,
      "loss": 6.6188,
      "mean_token_accuracy": 0.13516242504119874,
      "num_tokens": 4973099.0,
      "step": 2535
    },
    {
      "entropy": 6.858307123184204,
      "epoch": 0.25391113110411356,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997318920571907,
      "loss": 6.8016,
      "mean_token_accuracy": 0.11923333331942558,
      "num_tokens": 4983771.0,
      "step": 2540
    },
    {
      "entropy": 6.813778829574585,
      "epoch": 0.25441095616534215,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499730147484263,
      "loss": 6.6333,
      "mean_token_accuracy": 0.13007574677467346,
      "num_tokens": 4993648.0,
      "step": 2545
    },
    {
      "entropy": 6.8086004734039305,
      "epoch": 0.2549107812265707,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997283972571824,
      "loss": 6.8059,
      "mean_token_accuracy": 0.12374277338385582,
      "num_tokens": 5002821.0,
      "step": 2550
    },
    {
      "entropy": 6.9158666133880615,
      "epoch": 0.25541060628779927,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997266413759934,
      "loss": 6.8179,
      "mean_token_accuracy": 0.11914320066571235,
      "num_tokens": 5012655.0,
      "step": 2555
    },
    {
      "entropy": 6.749571657180786,
      "epoch": 0.25591043134902786,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997248798407397,
      "loss": 6.8112,
      "mean_token_accuracy": 0.11989297941327096,
      "num_tokens": 5023649.0,
      "step": 2560
    },
    {
      "entropy": 6.877628374099731,
      "epoch": 0.2564102564102564,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997231126514659,
      "loss": 6.7411,
      "mean_token_accuracy": 0.1278075806796551,
      "num_tokens": 5031945.0,
      "step": 2565
    },
    {
      "entropy": 6.862940835952759,
      "epoch": 0.256910081471485,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004997213398082165,
      "loss": 6.815,
      "mean_token_accuracy": 0.11514976248145103,
      "num_tokens": 5042036.0,
      "step": 2570
    },
    {
      "entropy": 6.813757658004761,
      "epoch": 0.25740990653271356,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499719561311036,
      "loss": 6.7905,
      "mean_token_accuracy": 0.12084953859448433,
      "num_tokens": 5050986.0,
      "step": 2575
    },
    {
      "entropy": 6.886504650115967,
      "epoch": 0.2579097315939421,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997177771599691,
      "loss": 6.7903,
      "mean_token_accuracy": 0.1246580310165882,
      "num_tokens": 5061408.0,
      "step": 2580
    },
    {
      "entropy": 6.785292768478394,
      "epoch": 0.2584095566551707,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997159873550607,
      "loss": 6.6987,
      "mean_token_accuracy": 0.12291771844029427,
      "num_tokens": 5071646.0,
      "step": 2585
    },
    {
      "entropy": 6.817554712295532,
      "epoch": 0.25890938171639927,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004997141918963561,
      "loss": 6.8184,
      "mean_token_accuracy": 0.12439500764012337,
      "num_tokens": 5083185.0,
      "step": 2590
    },
    {
      "entropy": 6.925652647018433,
      "epoch": 0.25940920677762785,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997123907839001,
      "loss": 6.7735,
      "mean_token_accuracy": 0.12057823091745376,
      "num_tokens": 5092993.0,
      "step": 2595
    },
    {
      "entropy": 6.828229284286499,
      "epoch": 0.2599090318388564,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997105840177381,
      "loss": 6.8237,
      "mean_token_accuracy": 0.11826542019844055,
      "num_tokens": 5102709.0,
      "step": 2600
    },
    {
      "entropy": 6.896103000640869,
      "epoch": 0.260408856900085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997087715979157,
      "loss": 6.7654,
      "mean_token_accuracy": 0.11819369643926621,
      "num_tokens": 5112858.0,
      "step": 2605
    },
    {
      "entropy": 6.856741428375244,
      "epoch": 0.26090868196131356,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997069535244785,
      "loss": 6.6749,
      "mean_token_accuracy": 0.1281345531344414,
      "num_tokens": 5123573.0,
      "step": 2610
    },
    {
      "entropy": 6.764648580551148,
      "epoch": 0.2614085070225421,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997051297974722,
      "loss": 6.8517,
      "mean_token_accuracy": 0.11786950081586837,
      "num_tokens": 5132908.0,
      "step": 2615
    },
    {
      "entropy": 6.912869548797607,
      "epoch": 0.2619083320837707,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997033004169425,
      "loss": 6.7485,
      "mean_token_accuracy": 0.1265025995671749,
      "num_tokens": 5142331.0,
      "step": 2620
    },
    {
      "entropy": 6.9357281684875485,
      "epoch": 0.26240815714499927,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997014653829357,
      "loss": 6.8393,
      "mean_token_accuracy": 0.11436363682150841,
      "num_tokens": 5152517.0,
      "step": 2625
    },
    {
      "entropy": 6.8380382537841795,
      "epoch": 0.2629079822062278,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996996246954977,
      "loss": 6.7778,
      "mean_token_accuracy": 0.12275150492787361,
      "num_tokens": 5162418.0,
      "step": 2630
    },
    {
      "entropy": 6.820027017593384,
      "epoch": 0.2634078072674564,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996977783546752,
      "loss": 6.7797,
      "mean_token_accuracy": 0.11911472529172898,
      "num_tokens": 5171948.0,
      "step": 2635
    },
    {
      "entropy": 6.898890399932862,
      "epoch": 0.26390763232868497,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996959263605143,
      "loss": 6.7266,
      "mean_token_accuracy": 0.12626254335045814,
      "num_tokens": 5180963.0,
      "step": 2640
    },
    {
      "entropy": 6.807502412796021,
      "epoch": 0.2644074573899135,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996940687130617,
      "loss": 6.8283,
      "mean_token_accuracy": 0.12082352638244628,
      "num_tokens": 5192305.0,
      "step": 2645
    },
    {
      "entropy": 6.86788969039917,
      "epoch": 0.2649072824511421,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996922054123641,
      "loss": 6.7126,
      "mean_token_accuracy": 0.12578068524599076,
      "num_tokens": 5201503.0,
      "step": 2650
    },
    {
      "entropy": 6.793289661407471,
      "epoch": 0.2654071075123707,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996903364584685,
      "loss": 6.6664,
      "mean_token_accuracy": 0.12352387979626656,
      "num_tokens": 5210206.0,
      "step": 2655
    },
    {
      "entropy": 6.692011928558349,
      "epoch": 0.26590693257359926,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996884618514219,
      "loss": 6.6962,
      "mean_token_accuracy": 0.1274379923939705,
      "num_tokens": 5220256.0,
      "step": 2660
    },
    {
      "entropy": 6.840415620803833,
      "epoch": 0.2664067576348278,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996865815912713,
      "loss": 6.7336,
      "mean_token_accuracy": 0.1253110870718956,
      "num_tokens": 5229128.0,
      "step": 2665
    },
    {
      "entropy": 6.794884872436524,
      "epoch": 0.2669065826960564,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996846956780642,
      "loss": 6.6659,
      "mean_token_accuracy": 0.12167781367897987,
      "num_tokens": 5239468.0,
      "step": 2670
    },
    {
      "entropy": 6.794555521011352,
      "epoch": 0.26740640775728497,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499682804111848,
      "loss": 6.768,
      "mean_token_accuracy": 0.12150566056370735,
      "num_tokens": 5247957.0,
      "step": 2675
    },
    {
      "entropy": 6.927483987808228,
      "epoch": 0.2679062328185135,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996809068926702,
      "loss": 6.7862,
      "mean_token_accuracy": 0.1188110314309597,
      "num_tokens": 5257960.0,
      "step": 2680
    },
    {
      "entropy": 6.722643184661865,
      "epoch": 0.2684060578797421,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996790040205786,
      "loss": 6.6354,
      "mean_token_accuracy": 0.1271999478340149,
      "num_tokens": 5266696.0,
      "step": 2685
    },
    {
      "entropy": 6.765709590911865,
      "epoch": 0.2689058829409707,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996770954956211,
      "loss": 6.6257,
      "mean_token_accuracy": 0.13072306290268898,
      "num_tokens": 5275619.0,
      "step": 2690
    },
    {
      "entropy": 6.817221784591675,
      "epoch": 0.2694057080021992,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996751813178456,
      "loss": 6.7757,
      "mean_token_accuracy": 0.1238712303340435,
      "num_tokens": 5284884.0,
      "step": 2695
    },
    {
      "entropy": 6.799127197265625,
      "epoch": 0.2699055330634278,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996732614873005,
      "loss": 6.6684,
      "mean_token_accuracy": 0.1343806691467762,
      "num_tokens": 5293909.0,
      "step": 2700
    },
    {
      "entropy": 6.712032699584961,
      "epoch": 0.2704053581246564,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996713360040339,
      "loss": 6.7552,
      "mean_token_accuracy": 0.1175712764263153,
      "num_tokens": 5303142.0,
      "step": 2705
    },
    {
      "entropy": 6.842455434799194,
      "epoch": 0.27090518318588497,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996694048680942,
      "loss": 6.6928,
      "mean_token_accuracy": 0.11897132322192192,
      "num_tokens": 5313055.0,
      "step": 2710
    },
    {
      "entropy": 6.790426254272461,
      "epoch": 0.2714050082471135,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996674680795302,
      "loss": 6.7946,
      "mean_token_accuracy": 0.11414426565170288,
      "num_tokens": 5323773.0,
      "step": 2715
    },
    {
      "entropy": 6.721665477752685,
      "epoch": 0.2719048333083421,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996655256383905,
      "loss": 6.6013,
      "mean_token_accuracy": 0.13403543531894685,
      "num_tokens": 5334613.0,
      "step": 2720
    },
    {
      "entropy": 6.777459335327149,
      "epoch": 0.2724046583695707,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499663577544724,
      "loss": 6.664,
      "mean_token_accuracy": 0.12990061491727828,
      "num_tokens": 5344151.0,
      "step": 2725
    },
    {
      "entropy": 6.7387457370758055,
      "epoch": 0.2729044834307992,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996616237985796,
      "loss": 6.6493,
      "mean_token_accuracy": 0.12913709804415702,
      "num_tokens": 5352425.0,
      "step": 2730
    },
    {
      "entropy": 6.82868595123291,
      "epoch": 0.2734043084920278,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996596644000066,
      "loss": 6.8319,
      "mean_token_accuracy": 0.11914294064044953,
      "num_tokens": 5361811.0,
      "step": 2735
    },
    {
      "entropy": 6.832614994049072,
      "epoch": 0.2739041335532564,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996576993490543,
      "loss": 6.7792,
      "mean_token_accuracy": 0.11754570752382279,
      "num_tokens": 5371463.0,
      "step": 2740
    },
    {
      "entropy": 6.818097829818726,
      "epoch": 0.2744039586144849,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996557286457721,
      "loss": 6.6401,
      "mean_token_accuracy": 0.1298965960741043,
      "num_tokens": 5381464.0,
      "step": 2745
    },
    {
      "entropy": 6.674585819244385,
      "epoch": 0.2749037836757135,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996537522902096,
      "loss": 6.7192,
      "mean_token_accuracy": 0.1203085571527481,
      "num_tokens": 5391012.0,
      "step": 2750
    },
    {
      "entropy": 6.939772033691407,
      "epoch": 0.2754036087369421,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996517702824163,
      "loss": 6.8237,
      "mean_token_accuracy": 0.12208779081702233,
      "num_tokens": 5401198.0,
      "step": 2755
    },
    {
      "entropy": 6.8920979499816895,
      "epoch": 0.2759034337981706,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996497826224425,
      "loss": 6.7498,
      "mean_token_accuracy": 0.12084006369113923,
      "num_tokens": 5410592.0,
      "step": 2760
    },
    {
      "entropy": 6.6667835235595705,
      "epoch": 0.2764032588593992,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996477893103379,
      "loss": 6.7307,
      "mean_token_accuracy": 0.12609246745705605,
      "num_tokens": 5420878.0,
      "step": 2765
    },
    {
      "entropy": 6.835362100601197,
      "epoch": 0.2769030839206278,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996457903461527,
      "loss": 6.7409,
      "mean_token_accuracy": 0.12492544427514077,
      "num_tokens": 5431199.0,
      "step": 2770
    },
    {
      "entropy": 6.808431673049927,
      "epoch": 0.2774029089818564,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996437857299373,
      "loss": 6.68,
      "mean_token_accuracy": 0.1254745155572891,
      "num_tokens": 5441661.0,
      "step": 2775
    },
    {
      "entropy": 6.681165075302124,
      "epoch": 0.2779027340430849,
      "grad_norm": 1.0,
      "learning_rate": 0.000499641775461742,
      "loss": 6.6732,
      "mean_token_accuracy": 0.12912385687232017,
      "num_tokens": 5451863.0,
      "step": 2780
    },
    {
      "entropy": 6.828213024139404,
      "epoch": 0.2784025591043135,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996397595416173,
      "loss": 6.7308,
      "mean_token_accuracy": 0.12578752413392066,
      "num_tokens": 5461600.0,
      "step": 2785
    },
    {
      "entropy": 6.906319808959961,
      "epoch": 0.2789023841655421,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996377379696143,
      "loss": 6.7552,
      "mean_token_accuracy": 0.11796140745282173,
      "num_tokens": 5470097.0,
      "step": 2790
    },
    {
      "entropy": 6.886836528778076,
      "epoch": 0.2794022092267706,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996357107457834,
      "loss": 6.8941,
      "mean_token_accuracy": 0.11296176090836525,
      "num_tokens": 5480906.0,
      "step": 2795
    },
    {
      "entropy": 6.949836778640747,
      "epoch": 0.2799020342879992,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499633677870176,
      "loss": 6.7854,
      "mean_token_accuracy": 0.11996870562434196,
      "num_tokens": 5489732.0,
      "step": 2800
    },
    {
      "entropy": 6.6947015762329105,
      "epoch": 0.2804018593492278,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996316393428429,
      "loss": 6.5782,
      "mean_token_accuracy": 0.13047064021229743,
      "num_tokens": 5498422.0,
      "step": 2805
    },
    {
      "entropy": 6.7099450588226315,
      "epoch": 0.2809016844104563,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996295951638357,
      "loss": 6.7244,
      "mean_token_accuracy": 0.12029694393277168,
      "num_tokens": 5507958.0,
      "step": 2810
    },
    {
      "entropy": 6.8781578540802,
      "epoch": 0.2814015094716849,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996275453332057,
      "loss": 6.7506,
      "mean_token_accuracy": 0.1246604599058628,
      "num_tokens": 5516752.0,
      "step": 2815
    },
    {
      "entropy": 6.751307106018066,
      "epoch": 0.2819013345329135,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996254898510043,
      "loss": 6.7763,
      "mean_token_accuracy": 0.12477909997105599,
      "num_tokens": 5527211.0,
      "step": 2820
    },
    {
      "entropy": 6.780982398986817,
      "epoch": 0.282401159594142,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996234287172835,
      "loss": 6.666,
      "mean_token_accuracy": 0.12461251020431519,
      "num_tokens": 5536925.0,
      "step": 2825
    },
    {
      "entropy": 6.845695829391479,
      "epoch": 0.2829009846553706,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499621361932095,
      "loss": 6.8396,
      "mean_token_accuracy": 0.1158826120197773,
      "num_tokens": 5546800.0,
      "step": 2830
    },
    {
      "entropy": 6.741491508483887,
      "epoch": 0.2834008097165992,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996192894954909,
      "loss": 6.6801,
      "mean_token_accuracy": 0.1240173600614071,
      "num_tokens": 5557266.0,
      "step": 2835
    },
    {
      "entropy": 6.809528160095215,
      "epoch": 0.2839006347778278,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996172114075232,
      "loss": 6.6583,
      "mean_token_accuracy": 0.12244971096515656,
      "num_tokens": 5566169.0,
      "step": 2840
    },
    {
      "entropy": 6.791384696960449,
      "epoch": 0.2844004598390563,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996151276682443,
      "loss": 6.7552,
      "mean_token_accuracy": 0.12420285567641258,
      "num_tokens": 5575566.0,
      "step": 2845
    },
    {
      "entropy": 6.755648517608643,
      "epoch": 0.2849002849002849,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996130382777065,
      "loss": 6.6381,
      "mean_token_accuracy": 0.12499300390481949,
      "num_tokens": 5584804.0,
      "step": 2850
    },
    {
      "entropy": 6.7822061538696286,
      "epoch": 0.2854001099615135,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996109432359625,
      "loss": 6.7054,
      "mean_token_accuracy": 0.1191313810646534,
      "num_tokens": 5594361.0,
      "step": 2855
    },
    {
      "entropy": 6.848558330535889,
      "epoch": 0.285899935022742,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004996088425430651,
      "loss": 6.7122,
      "mean_token_accuracy": 0.12073459699749947,
      "num_tokens": 5605682.0,
      "step": 2860
    },
    {
      "entropy": 6.7933878898620605,
      "epoch": 0.2863997600839706,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996067361990669,
      "loss": 6.768,
      "mean_token_accuracy": 0.118593979626894,
      "num_tokens": 5617251.0,
      "step": 2865
    },
    {
      "entropy": 6.756092596054077,
      "epoch": 0.2868995851451992,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004996046242040209,
      "loss": 6.6517,
      "mean_token_accuracy": 0.12677795737981795,
      "num_tokens": 5627140.0,
      "step": 2870
    },
    {
      "entropy": 6.708715581893921,
      "epoch": 0.28739941020642773,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996025065579805,
      "loss": 6.5964,
      "mean_token_accuracy": 0.12997669354081154,
      "num_tokens": 5635486.0,
      "step": 2875
    },
    {
      "entropy": 6.793992185592652,
      "epoch": 0.2878992352676563,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996003832609988,
      "loss": 6.7669,
      "mean_token_accuracy": 0.11563235074281693,
      "num_tokens": 5645621.0,
      "step": 2880
    },
    {
      "entropy": 6.9069499492645265,
      "epoch": 0.2883990603288849,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995982543131292,
      "loss": 6.8248,
      "mean_token_accuracy": 0.11520569920539855,
      "num_tokens": 5655626.0,
      "step": 2885
    },
    {
      "entropy": 6.817208242416382,
      "epoch": 0.28889888539011344,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995961197144254,
      "loss": 6.7666,
      "mean_token_accuracy": 0.12128069922327996,
      "num_tokens": 5664547.0,
      "step": 2890
    },
    {
      "entropy": 6.791419172286988,
      "epoch": 0.289398710451342,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499593979464941,
      "loss": 6.747,
      "mean_token_accuracy": 0.12001895159482956,
      "num_tokens": 5674861.0,
      "step": 2895
    },
    {
      "entropy": 6.762214231491089,
      "epoch": 0.2898985355125706,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995918335647298,
      "loss": 6.6782,
      "mean_token_accuracy": 0.11858229264616967,
      "num_tokens": 5684143.0,
      "step": 2900
    },
    {
      "entropy": 6.764250040054321,
      "epoch": 0.2903983605737992,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995896820138459,
      "loss": 6.6529,
      "mean_token_accuracy": 0.12445368096232415,
      "num_tokens": 5693837.0,
      "step": 2905
    },
    {
      "entropy": 6.869545316696167,
      "epoch": 0.29089818563502773,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995875248123434,
      "loss": 6.7437,
      "mean_token_accuracy": 0.11787635460495949,
      "num_tokens": 5704827.0,
      "step": 2910
    },
    {
      "entropy": 6.8025154113769535,
      "epoch": 0.2913980106962563,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995853619602766,
      "loss": 6.7719,
      "mean_token_accuracy": 0.12075797915458679,
      "num_tokens": 5714419.0,
      "step": 2915
    },
    {
      "entropy": 6.739893627166748,
      "epoch": 0.2918978357574849,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004995831934576998,
      "loss": 6.7054,
      "mean_token_accuracy": 0.12215565443038941,
      "num_tokens": 5724983.0,
      "step": 2920
    },
    {
      "entropy": 6.787188720703125,
      "epoch": 0.29239766081871343,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995810193046677,
      "loss": 6.6291,
      "mean_token_accuracy": 0.12742045596241952,
      "num_tokens": 5734471.0,
      "step": 2925
    },
    {
      "entropy": 6.803781652450562,
      "epoch": 0.292897485879942,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499578839501235,
      "loss": 6.7518,
      "mean_token_accuracy": 0.12121198177337647,
      "num_tokens": 5744801.0,
      "step": 2930
    },
    {
      "entropy": 6.755331993103027,
      "epoch": 0.2933973109411706,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995766540474564,
      "loss": 6.7167,
      "mean_token_accuracy": 0.11747701466083527,
      "num_tokens": 5754267.0,
      "step": 2935
    },
    {
      "entropy": 6.691815185546875,
      "epoch": 0.29389713600239914,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499574462943387,
      "loss": 6.6434,
      "mean_token_accuracy": 0.1278532288968563,
      "num_tokens": 5764466.0,
      "step": 2940
    },
    {
      "entropy": 6.797782039642334,
      "epoch": 0.2943969610636277,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499572266189082,
      "loss": 6.7755,
      "mean_token_accuracy": 0.12012682259082794,
      "num_tokens": 5775158.0,
      "step": 2945
    },
    {
      "entropy": 6.804054927825928,
      "epoch": 0.2948967861248563,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995700637845964,
      "loss": 6.5639,
      "mean_token_accuracy": 0.1323493793606758,
      "num_tokens": 5783320.0,
      "step": 2950
    },
    {
      "entropy": 6.675914669036866,
      "epoch": 0.29539661118608485,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004995678557299859,
      "loss": 6.7086,
      "mean_token_accuracy": 0.12223785743117332,
      "num_tokens": 5792980.0,
      "step": 2955
    },
    {
      "entropy": 6.874987697601318,
      "epoch": 0.29589643624731343,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995656420253058,
      "loss": 6.7467,
      "mean_token_accuracy": 0.12244661524891853,
      "num_tokens": 5802338.0,
      "step": 2960
    },
    {
      "entropy": 6.739057731628418,
      "epoch": 0.296396261308542,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499563422670612,
      "loss": 6.6364,
      "mean_token_accuracy": 0.12737288773059846,
      "num_tokens": 5811388.0,
      "step": 2965
    },
    {
      "entropy": 6.72182264328003,
      "epoch": 0.2968960863697706,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995611976659604,
      "loss": 6.6834,
      "mean_token_accuracy": 0.12516121342778205,
      "num_tokens": 5821108.0,
      "step": 2970
    },
    {
      "entropy": 6.731955909729004,
      "epoch": 0.29739591143099914,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995589670114067,
      "loss": 6.7281,
      "mean_token_accuracy": 0.1192234419286251,
      "num_tokens": 5831389.0,
      "step": 2975
    },
    {
      "entropy": 6.72219705581665,
      "epoch": 0.2978957364922277,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995567307070072,
      "loss": 6.5871,
      "mean_token_accuracy": 0.13089271262288094,
      "num_tokens": 5840392.0,
      "step": 2980
    },
    {
      "entropy": 6.748080110549926,
      "epoch": 0.2983955615534563,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995544887528182,
      "loss": 6.6116,
      "mean_token_accuracy": 0.12855779454112054,
      "num_tokens": 5850032.0,
      "step": 2985
    },
    {
      "entropy": 6.770363426208496,
      "epoch": 0.29889538661468484,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995522411488959,
      "loss": 6.6699,
      "mean_token_accuracy": 0.12747157365083694,
      "num_tokens": 5859204.0,
      "step": 2990
    },
    {
      "entropy": 6.715180587768555,
      "epoch": 0.29939521167591343,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995499878952971,
      "loss": 6.7008,
      "mean_token_accuracy": 0.12443819269537926,
      "num_tokens": 5868582.0,
      "step": 2995
    },
    {
      "entropy": 6.790849494934082,
      "epoch": 0.299895036737142,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995477289920784,
      "loss": 6.6241,
      "mean_token_accuracy": 0.1298421509563923,
      "num_tokens": 5879601.0,
      "step": 3000
    },
    {
      "epoch": 0.299895036737142,
      "eval_entropy": 6.615208382644594,
      "eval_loss": 6.721776008605957,
      "eval_mean_token_accuracy": 0.1284465000584829,
      "eval_num_tokens": 5879601.0,
      "eval_runtime": 23.8533,
      "eval_samples_per_second": 1301.497,
      "eval_steps_per_second": 162.703,
      "step": 3000
    },
    {
      "entropy": 6.722339534759522,
      "epoch": 0.30039486179837055,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995454644392967,
      "loss": 6.6878,
      "mean_token_accuracy": 0.11981307566165925,
      "num_tokens": 5890255.0,
      "step": 3005
    },
    {
      "entropy": 6.71357798576355,
      "epoch": 0.30089468685959914,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995431942370087,
      "loss": 6.7052,
      "mean_token_accuracy": 0.1337470643222332,
      "num_tokens": 5899965.0,
      "step": 3010
    },
    {
      "entropy": 6.788109397888183,
      "epoch": 0.3013945119208277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995409183852718,
      "loss": 6.5939,
      "mean_token_accuracy": 0.1301611252129078,
      "num_tokens": 5910082.0,
      "step": 3015
    },
    {
      "entropy": 6.672738313674927,
      "epoch": 0.30189433698205626,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995386368841432,
      "loss": 6.7355,
      "mean_token_accuracy": 0.12291843146085739,
      "num_tokens": 5920267.0,
      "step": 3020
    },
    {
      "entropy": 6.816665887832642,
      "epoch": 0.30239416204328484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995363497336802,
      "loss": 6.7141,
      "mean_token_accuracy": 0.12424932271242142,
      "num_tokens": 5929275.0,
      "step": 3025
    },
    {
      "entropy": 6.827727603912353,
      "epoch": 0.30289398710451343,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995340569339405,
      "loss": 6.7413,
      "mean_token_accuracy": 0.11787217631936073,
      "num_tokens": 5939305.0,
      "step": 3030
    },
    {
      "entropy": 6.836606121063232,
      "epoch": 0.303393812165742,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004995317584849817,
      "loss": 6.8025,
      "mean_token_accuracy": 0.11330446675419807,
      "num_tokens": 5950816.0,
      "step": 3035
    },
    {
      "entropy": 6.678486394882202,
      "epoch": 0.30389363722697055,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995294543868615,
      "loss": 6.5774,
      "mean_token_accuracy": 0.13259122967720033,
      "num_tokens": 5959824.0,
      "step": 3040
    },
    {
      "entropy": 6.767902612686157,
      "epoch": 0.30439346228819913,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995271446396382,
      "loss": 6.6938,
      "mean_token_accuracy": 0.12707387283444405,
      "num_tokens": 5968752.0,
      "step": 3045
    },
    {
      "entropy": 6.74622220993042,
      "epoch": 0.3048932873494277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995248292433697,
      "loss": 6.6351,
      "mean_token_accuracy": 0.12331249415874482,
      "num_tokens": 5979085.0,
      "step": 3050
    },
    {
      "entropy": 6.713514804840088,
      "epoch": 0.30539311241065625,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004995225081981142,
      "loss": 6.6764,
      "mean_token_accuracy": 0.12420432046055793,
      "num_tokens": 5990378.0,
      "step": 3055
    },
    {
      "entropy": 6.778734540939331,
      "epoch": 0.30589293747188484,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995201815039301,
      "loss": 6.5685,
      "mean_token_accuracy": 0.13564710617065429,
      "num_tokens": 5999524.0,
      "step": 3060
    },
    {
      "entropy": 6.687535905838013,
      "epoch": 0.3063927625331134,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995178491608762,
      "loss": 6.6806,
      "mean_token_accuracy": 0.12482227534055709,
      "num_tokens": 6009928.0,
      "step": 3065
    },
    {
      "entropy": 6.640461874008179,
      "epoch": 0.30689258759434196,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995155111690107,
      "loss": 6.5585,
      "mean_token_accuracy": 0.1330404207110405,
      "num_tokens": 6019683.0,
      "step": 3070
    },
    {
      "entropy": 6.843991374969482,
      "epoch": 0.30739241265557055,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499513167528393,
      "loss": 6.6812,
      "mean_token_accuracy": 0.1237489826977253,
      "num_tokens": 6030506.0,
      "step": 3075
    },
    {
      "entropy": 6.77349967956543,
      "epoch": 0.30789223771679913,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995108182390815,
      "loss": 6.6639,
      "mean_token_accuracy": 0.1260622389614582,
      "num_tokens": 6039720.0,
      "step": 3080
    },
    {
      "entropy": 6.598236227035523,
      "epoch": 0.30839206277802766,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995084633011359,
      "loss": 6.5091,
      "mean_token_accuracy": 0.13624039590358733,
      "num_tokens": 6049336.0,
      "step": 3085
    },
    {
      "entropy": 6.623972368240357,
      "epoch": 0.30889188783925625,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499506102714615,
      "loss": 6.5245,
      "mean_token_accuracy": 0.12994642406702042,
      "num_tokens": 6059420.0,
      "step": 3090
    },
    {
      "entropy": 6.739403963088989,
      "epoch": 0.30939171290048484,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995037364795782,
      "loss": 6.6253,
      "mean_token_accuracy": 0.1298194333910942,
      "num_tokens": 6069035.0,
      "step": 3095
    },
    {
      "entropy": 6.70915584564209,
      "epoch": 0.3098915379617134,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995013645960853,
      "loss": 6.6141,
      "mean_token_accuracy": 0.123585245013237,
      "num_tokens": 6077863.0,
      "step": 3100
    },
    {
      "entropy": 6.67954273223877,
      "epoch": 0.31039136302294196,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994989870641958,
      "loss": 6.6708,
      "mean_token_accuracy": 0.12600963637232782,
      "num_tokens": 6088929.0,
      "step": 3105
    },
    {
      "entropy": 6.679830694198609,
      "epoch": 0.31089118808417054,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994966038839695,
      "loss": 6.6119,
      "mean_token_accuracy": 0.13017525151371956,
      "num_tokens": 6098673.0,
      "step": 3110
    },
    {
      "entropy": 6.706893873214722,
      "epoch": 0.31139101314539913,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004994942150554666,
      "loss": 6.4884,
      "mean_token_accuracy": 0.1320834055542946,
      "num_tokens": 6107289.0,
      "step": 3115
    },
    {
      "entropy": 6.5927753925323485,
      "epoch": 0.31189083820662766,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004994918205787468,
      "loss": 6.5466,
      "mean_token_accuracy": 0.13047196641564368,
      "num_tokens": 6117811.0,
      "step": 3120
    },
    {
      "entropy": 6.712067270278931,
      "epoch": 0.31239066326785625,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994894204538706,
      "loss": 6.6538,
      "mean_token_accuracy": 0.12287433221936225,
      "num_tokens": 6127740.0,
      "step": 3125
    },
    {
      "entropy": 6.733972358703613,
      "epoch": 0.31289048832908484,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994870146808984,
      "loss": 6.6332,
      "mean_token_accuracy": 0.12414981052279472,
      "num_tokens": 6137253.0,
      "step": 3130
    },
    {
      "entropy": 6.68615574836731,
      "epoch": 0.31339031339031337,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994846032598906,
      "loss": 6.6661,
      "mean_token_accuracy": 0.12876934111118316,
      "num_tokens": 6147420.0,
      "step": 3135
    },
    {
      "entropy": 6.702655982971192,
      "epoch": 0.31389013845154196,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994821861909082,
      "loss": 6.6069,
      "mean_token_accuracy": 0.1283666931092739,
      "num_tokens": 6156436.0,
      "step": 3140
    },
    {
      "entropy": 6.622131776809693,
      "epoch": 0.31438996351277054,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994797634740114,
      "loss": 6.5698,
      "mean_token_accuracy": 0.126331477612257,
      "num_tokens": 6165526.0,
      "step": 3145
    },
    {
      "entropy": 6.778597736358643,
      "epoch": 0.3148897885739991,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994773351092618,
      "loss": 6.7677,
      "mean_token_accuracy": 0.12162536010146141,
      "num_tokens": 6176093.0,
      "step": 3150
    },
    {
      "entropy": 6.703013753890991,
      "epoch": 0.31538961363522766,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00049947490109672,
      "loss": 6.5791,
      "mean_token_accuracy": 0.1345700092613697,
      "num_tokens": 6185373.0,
      "step": 3155
    },
    {
      "entropy": 6.736076927185058,
      "epoch": 0.31588943869645625,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994724614364476,
      "loss": 6.6927,
      "mean_token_accuracy": 0.12343047335743904,
      "num_tokens": 6196252.0,
      "step": 3160
    },
    {
      "entropy": 6.801714611053467,
      "epoch": 0.31638926375768484,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994700161285058,
      "loss": 6.6215,
      "mean_token_accuracy": 0.1300746403634548,
      "num_tokens": 6206463.0,
      "step": 3165
    },
    {
      "entropy": 6.67550048828125,
      "epoch": 0.31688908881891337,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499467565172956,
      "loss": 6.6001,
      "mean_token_accuracy": 0.1312672823667526,
      "num_tokens": 6216676.0,
      "step": 3170
    },
    {
      "entropy": 6.689395141601563,
      "epoch": 0.31738891388014195,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004994651085698602,
      "loss": 6.6236,
      "mean_token_accuracy": 0.1254762552678585,
      "num_tokens": 6227261.0,
      "step": 3175
    },
    {
      "entropy": 6.707279872894287,
      "epoch": 0.31788873894137054,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049946264631928,
      "loss": 6.5674,
      "mean_token_accuracy": 0.12468275651335717,
      "num_tokens": 6237468.0,
      "step": 3180
    },
    {
      "entropy": 6.642337942123413,
      "epoch": 0.3183885640025991,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994601784212774,
      "loss": 6.5122,
      "mean_token_accuracy": 0.136201723664999,
      "num_tokens": 6247164.0,
      "step": 3185
    },
    {
      "entropy": 6.649035453796387,
      "epoch": 0.31888838906382766,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004994577048759144,
      "loss": 6.6264,
      "mean_token_accuracy": 0.13031774014234543,
      "num_tokens": 6257635.0,
      "step": 3190
    },
    {
      "entropy": 6.798333024978637,
      "epoch": 0.31938821412505625,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994552256832533,
      "loss": 6.6984,
      "mean_token_accuracy": 0.1256155826151371,
      "num_tokens": 6266823.0,
      "step": 3195
    },
    {
      "entropy": 6.774302530288696,
      "epoch": 0.3198880391862848,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004994527408433566,
      "loss": 6.6072,
      "mean_token_accuracy": 0.12974709048867225,
      "num_tokens": 6276527.0,
      "step": 3200
    },
    {
      "entropy": 6.751862907409668,
      "epoch": 0.32038786424751337,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994502503562865,
      "loss": 6.6704,
      "mean_token_accuracy": 0.12288417518138886,
      "num_tokens": 6286499.0,
      "step": 3205
    },
    {
      "entropy": 6.685272693634033,
      "epoch": 0.32088768930874195,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499447754222106,
      "loss": 6.5856,
      "mean_token_accuracy": 0.130257061868906,
      "num_tokens": 6296550.0,
      "step": 3210
    },
    {
      "entropy": 6.667414760589599,
      "epoch": 0.3213875143699705,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994452524408777,
      "loss": 6.6489,
      "mean_token_accuracy": 0.1256569355726242,
      "num_tokens": 6306951.0,
      "step": 3215
    },
    {
      "entropy": 6.833637809753418,
      "epoch": 0.32188733943119907,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004994427450126646,
      "loss": 6.6768,
      "mean_token_accuracy": 0.11989659443497658,
      "num_tokens": 6316026.0,
      "step": 3220
    },
    {
      "entropy": 6.648589992523194,
      "epoch": 0.32238716449242766,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004994402319375298,
      "loss": 6.6104,
      "mean_token_accuracy": 0.12591632306575776,
      "num_tokens": 6325064.0,
      "step": 3225
    },
    {
      "entropy": 6.721154403686524,
      "epoch": 0.32288698955365625,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994377132155366,
      "loss": 6.6325,
      "mean_token_accuracy": 0.12740512788295746,
      "num_tokens": 6334443.0,
      "step": 3230
    },
    {
      "entropy": 6.727579784393311,
      "epoch": 0.3233868146148848,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994351888467482,
      "loss": 6.643,
      "mean_token_accuracy": 0.12921743839979172,
      "num_tokens": 6346103.0,
      "step": 3235
    },
    {
      "entropy": 6.644028043746948,
      "epoch": 0.32388663967611336,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994326588312282,
      "loss": 6.6215,
      "mean_token_accuracy": 0.12528072744607927,
      "num_tokens": 6356576.0,
      "step": 3240
    },
    {
      "entropy": 6.75742621421814,
      "epoch": 0.32438646473734195,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994301231690403,
      "loss": 6.6181,
      "mean_token_accuracy": 0.12732079029083251,
      "num_tokens": 6364796.0,
      "step": 3245
    },
    {
      "entropy": 6.732314491271973,
      "epoch": 0.3248862897985705,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994275818602482,
      "loss": 6.6953,
      "mean_token_accuracy": 0.12702387496829032,
      "num_tokens": 6374951.0,
      "step": 3250
    },
    {
      "entropy": 6.662586832046509,
      "epoch": 0.32538611485979907,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994250349049159,
      "loss": 6.5877,
      "mean_token_accuracy": 0.1256158784031868,
      "num_tokens": 6384230.0,
      "step": 3255
    },
    {
      "entropy": 6.721761417388916,
      "epoch": 0.32588593992102766,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994224823031075,
      "loss": 6.5777,
      "mean_token_accuracy": 0.12750042378902435,
      "num_tokens": 6394070.0,
      "step": 3260
    },
    {
      "entropy": 6.610982847213745,
      "epoch": 0.3263857649822562,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994199240548873,
      "loss": 6.5717,
      "mean_token_accuracy": 0.13098210096359253,
      "num_tokens": 6403968.0,
      "step": 3265
    },
    {
      "entropy": 6.757144117355347,
      "epoch": 0.3268855900434848,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994173601603194,
      "loss": 6.6887,
      "mean_token_accuracy": 0.1251855470240116,
      "num_tokens": 6413254.0,
      "step": 3270
    },
    {
      "entropy": 6.648221397399903,
      "epoch": 0.32738541510471336,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994147906194684,
      "loss": 6.5344,
      "mean_token_accuracy": 0.1325840137898922,
      "num_tokens": 6421266.0,
      "step": 3275
    },
    {
      "entropy": 6.694507598876953,
      "epoch": 0.3278852401659419,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994122154323992,
      "loss": 6.6342,
      "mean_token_accuracy": 0.1262752406299114,
      "num_tokens": 6429596.0,
      "step": 3280
    },
    {
      "entropy": 6.775129127502441,
      "epoch": 0.3283850652271705,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994096345991763,
      "loss": 6.6376,
      "mean_token_accuracy": 0.12423462271690369,
      "num_tokens": 6440825.0,
      "step": 3285
    },
    {
      "entropy": 6.608078670501709,
      "epoch": 0.32888489028839907,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994070481198649,
      "loss": 6.5928,
      "mean_token_accuracy": 0.12965668961405755,
      "num_tokens": 6450576.0,
      "step": 3290
    },
    {
      "entropy": 6.702484655380249,
      "epoch": 0.32938471534962765,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994044559945296,
      "loss": 6.5951,
      "mean_token_accuracy": 0.13519497066736222,
      "num_tokens": 6459673.0,
      "step": 3295
    },
    {
      "entropy": 6.664275407791138,
      "epoch": 0.3298845404108562,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004994018582232361,
      "loss": 6.5782,
      "mean_token_accuracy": 0.12735630720853805,
      "num_tokens": 6469353.0,
      "step": 3300
    },
    {
      "entropy": 6.654953193664551,
      "epoch": 0.3303843654720848,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993992548060497,
      "loss": 6.5833,
      "mean_token_accuracy": 0.1305153839290142,
      "num_tokens": 6479610.0,
      "step": 3305
    },
    {
      "entropy": 6.682379102706909,
      "epoch": 0.33088419053331336,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993966457430357,
      "loss": 6.6392,
      "mean_token_accuracy": 0.131649050116539,
      "num_tokens": 6488866.0,
      "step": 3310
    },
    {
      "entropy": 6.717644453048706,
      "epoch": 0.3313840155945419,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993940310342598,
      "loss": 6.518,
      "mean_token_accuracy": 0.13404856026172637,
      "num_tokens": 6498396.0,
      "step": 3315
    },
    {
      "entropy": 6.6136088371276855,
      "epoch": 0.3318838406557705,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993914106797879,
      "loss": 6.5414,
      "mean_token_accuracy": 0.1354752868413925,
      "num_tokens": 6508347.0,
      "step": 3320
    },
    {
      "entropy": 6.722569513320923,
      "epoch": 0.33238366571699907,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993887846796858,
      "loss": 6.6592,
      "mean_token_accuracy": 0.12250009402632714,
      "num_tokens": 6518369.0,
      "step": 3325
    },
    {
      "entropy": 6.662774610519409,
      "epoch": 0.3328834907782276,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993861530340196,
      "loss": 6.6214,
      "mean_token_accuracy": 0.13503153324127198,
      "num_tokens": 6527477.0,
      "step": 3330
    },
    {
      "entropy": 6.718849229812622,
      "epoch": 0.3333833158394562,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993835157428555,
      "loss": 6.5275,
      "mean_token_accuracy": 0.13233767300844193,
      "num_tokens": 6536271.0,
      "step": 3335
    },
    {
      "entropy": 6.716184806823731,
      "epoch": 0.33388314090068477,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993808728062599,
      "loss": 6.6382,
      "mean_token_accuracy": 0.13095614314079285,
      "num_tokens": 6546295.0,
      "step": 3340
    },
    {
      "entropy": 6.672792673110962,
      "epoch": 0.3343829659619133,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004993782242242994,
      "loss": 6.5532,
      "mean_token_accuracy": 0.13651400059461594,
      "num_tokens": 6557338.0,
      "step": 3345
    },
    {
      "entropy": 6.640097570419312,
      "epoch": 0.3348827910231419,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993755699970404,
      "loss": 6.57,
      "mean_token_accuracy": 0.13229951858520508,
      "num_tokens": 6567916.0,
      "step": 3350
    },
    {
      "entropy": 6.671002817153931,
      "epoch": 0.3353826160843705,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993729101245498,
      "loss": 6.6875,
      "mean_token_accuracy": 0.12816037237644196,
      "num_tokens": 6577821.0,
      "step": 3355
    },
    {
      "entropy": 6.742249488830566,
      "epoch": 0.33588244114559906,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993702446068945,
      "loss": 6.5681,
      "mean_token_accuracy": 0.1282327353954315,
      "num_tokens": 6587674.0,
      "step": 3360
    },
    {
      "entropy": 6.683386659622192,
      "epoch": 0.3363822662068276,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993675734441416,
      "loss": 6.6515,
      "mean_token_accuracy": 0.13280568197369574,
      "num_tokens": 6597292.0,
      "step": 3365
    },
    {
      "entropy": 6.594243669509888,
      "epoch": 0.3368820912680562,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993648966363583,
      "loss": 6.5701,
      "mean_token_accuracy": 0.13267690166831017,
      "num_tokens": 6608698.0,
      "step": 3370
    },
    {
      "entropy": 6.730088233947754,
      "epoch": 0.33738191632928477,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499362214183612,
      "loss": 6.5643,
      "mean_token_accuracy": 0.12787219882011414,
      "num_tokens": 6617914.0,
      "step": 3375
    },
    {
      "entropy": 6.590554046630859,
      "epoch": 0.3378817413905133,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049935952608597,
      "loss": 6.5302,
      "mean_token_accuracy": 0.13169657364487647,
      "num_tokens": 6627457.0,
      "step": 3380
    },
    {
      "entropy": 6.691078758239746,
      "epoch": 0.3383815664517419,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993568323435002,
      "loss": 6.5613,
      "mean_token_accuracy": 0.13399697616696357,
      "num_tokens": 6636511.0,
      "step": 3385
    },
    {
      "entropy": 6.647966480255127,
      "epoch": 0.3388813915129705,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993541329562702,
      "loss": 6.5478,
      "mean_token_accuracy": 0.12832800596952437,
      "num_tokens": 6647028.0,
      "step": 3390
    },
    {
      "entropy": 6.6662462711334225,
      "epoch": 0.339381216574199,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993514279243477,
      "loss": 6.616,
      "mean_token_accuracy": 0.1287503607571125,
      "num_tokens": 6656407.0,
      "step": 3395
    },
    {
      "entropy": 6.704301548004151,
      "epoch": 0.3398810416354276,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993487172478012,
      "loss": 6.5443,
      "mean_token_accuracy": 0.13133591488003732,
      "num_tokens": 6665554.0,
      "step": 3400
    },
    {
      "entropy": 6.704265069961548,
      "epoch": 0.3403808666966562,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993460009266988,
      "loss": 6.6183,
      "mean_token_accuracy": 0.12598370388150215,
      "num_tokens": 6675730.0,
      "step": 3405
    },
    {
      "entropy": 6.577371120452881,
      "epoch": 0.3408806917578847,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993432789611086,
      "loss": 6.5253,
      "mean_token_accuracy": 0.13635042160749436,
      "num_tokens": 6685871.0,
      "step": 3410
    },
    {
      "entropy": 6.715635204315186,
      "epoch": 0.3413805168191133,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993405513510992,
      "loss": 6.7433,
      "mean_token_accuracy": 0.1186644345521927,
      "num_tokens": 6697503.0,
      "step": 3415
    },
    {
      "entropy": 6.721425724029541,
      "epoch": 0.3418803418803419,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993378180967394,
      "loss": 6.5449,
      "mean_token_accuracy": 0.13019852563738823,
      "num_tokens": 6708022.0,
      "step": 3420
    },
    {
      "entropy": 6.631089878082276,
      "epoch": 0.3423801669415705,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004993350791980978,
      "loss": 6.6171,
      "mean_token_accuracy": 0.13050465658307076,
      "num_tokens": 6717510.0,
      "step": 3425
    },
    {
      "entropy": 6.6999125480651855,
      "epoch": 0.342879992002799,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004993323346552433,
      "loss": 6.6302,
      "mean_token_accuracy": 0.1296307161450386,
      "num_tokens": 6726538.0,
      "step": 3430
    },
    {
      "entropy": 6.747427368164063,
      "epoch": 0.3433798170640276,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993295844682449,
      "loss": 6.5439,
      "mean_token_accuracy": 0.13428208231925964,
      "num_tokens": 6736750.0,
      "step": 3435
    },
    {
      "entropy": 6.616900205612183,
      "epoch": 0.3438796421252562,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499326828637172,
      "loss": 6.576,
      "mean_token_accuracy": 0.1279594212770462,
      "num_tokens": 6746366.0,
      "step": 3440
    },
    {
      "entropy": 6.578363037109375,
      "epoch": 0.3443794671864847,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993240671620938,
      "loss": 6.5935,
      "mean_token_accuracy": 0.12818874344229697,
      "num_tokens": 6755887.0,
      "step": 3445
    },
    {
      "entropy": 6.709596347808838,
      "epoch": 0.3448792922477133,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993213000430797,
      "loss": 6.5994,
      "mean_token_accuracy": 0.13560046702623368,
      "num_tokens": 6764434.0,
      "step": 3450
    },
    {
      "entropy": 6.743025636672973,
      "epoch": 0.3453791173089419,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993185272801995,
      "loss": 6.7018,
      "mean_token_accuracy": 0.11916685327887536,
      "num_tokens": 6774237.0,
      "step": 3455
    },
    {
      "entropy": 6.760094881057739,
      "epoch": 0.3458789423701704,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993157488735228,
      "loss": 6.5579,
      "mean_token_accuracy": 0.12662035152316092,
      "num_tokens": 6784537.0,
      "step": 3460
    },
    {
      "entropy": 6.629041051864624,
      "epoch": 0.346378767431399,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993129648231197,
      "loss": 6.5326,
      "mean_token_accuracy": 0.13424717038869857,
      "num_tokens": 6794223.0,
      "step": 3465
    },
    {
      "entropy": 6.571995496749878,
      "epoch": 0.3468785924926276,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049931017512906,
      "loss": 6.5773,
      "mean_token_accuracy": 0.1289962500333786,
      "num_tokens": 6803680.0,
      "step": 3470
    },
    {
      "entropy": 6.819319772720337,
      "epoch": 0.3473784175538562,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993073797914141,
      "loss": 6.5711,
      "mean_token_accuracy": 0.1251891352236271,
      "num_tokens": 6812853.0,
      "step": 3475
    },
    {
      "entropy": 6.609120512008667,
      "epoch": 0.3478782426150847,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993045788102522,
      "loss": 6.6439,
      "mean_token_accuracy": 0.12747282832860946,
      "num_tokens": 6822104.0,
      "step": 3480
    },
    {
      "entropy": 6.676856327056885,
      "epoch": 0.3483780676763133,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993017721856449,
      "loss": 6.609,
      "mean_token_accuracy": 0.12209485694766045,
      "num_tokens": 6832683.0,
      "step": 3485
    },
    {
      "entropy": 6.785378742218017,
      "epoch": 0.3488778927375419,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992989599176626,
      "loss": 6.5971,
      "mean_token_accuracy": 0.12543685734272003,
      "num_tokens": 6843191.0,
      "step": 3490
    },
    {
      "entropy": 6.6238781929016115,
      "epoch": 0.3493777177987704,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992961420063763,
      "loss": 6.5765,
      "mean_token_accuracy": 0.128035569190979,
      "num_tokens": 6852205.0,
      "step": 3495
    },
    {
      "entropy": 6.5474427223205565,
      "epoch": 0.349877542859999,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992933184518567,
      "loss": 6.4989,
      "mean_token_accuracy": 0.1326678402721882,
      "num_tokens": 6863052.0,
      "step": 3500
    },
    {
      "entropy": 6.698447322845459,
      "epoch": 0.3503773679212276,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992904892541749,
      "loss": 6.5544,
      "mean_token_accuracy": 0.13316056430339812,
      "num_tokens": 6872717.0,
      "step": 3505
    },
    {
      "entropy": 6.612038040161133,
      "epoch": 0.3508771929824561,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004992876544134022,
      "loss": 6.4816,
      "mean_token_accuracy": 0.14064068496227264,
      "num_tokens": 6881351.0,
      "step": 3510
    },
    {
      "entropy": 6.6335334300994875,
      "epoch": 0.3513770180436847,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992848139296098,
      "loss": 6.5409,
      "mean_token_accuracy": 0.1337484709918499,
      "num_tokens": 6891353.0,
      "step": 3515
    },
    {
      "entropy": 6.704902029037475,
      "epoch": 0.3518768431049133,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992819678028692,
      "loss": 6.632,
      "mean_token_accuracy": 0.1213303491473198,
      "num_tokens": 6902352.0,
      "step": 3520
    },
    {
      "entropy": 6.70453896522522,
      "epoch": 0.3523766681661418,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499279116033252,
      "loss": 6.6033,
      "mean_token_accuracy": 0.12167268171906472,
      "num_tokens": 6911950.0,
      "step": 3525
    },
    {
      "entropy": 6.6038049221038815,
      "epoch": 0.3528764932273704,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992762586208299,
      "loss": 6.5089,
      "mean_token_accuracy": 0.13899552896618844,
      "num_tokens": 6921949.0,
      "step": 3530
    },
    {
      "entropy": 6.612030172348023,
      "epoch": 0.353376318288599,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992733955656749,
      "loss": 6.5462,
      "mean_token_accuracy": 0.12957046926021576,
      "num_tokens": 6932561.0,
      "step": 3535
    },
    {
      "entropy": 6.609406471252441,
      "epoch": 0.3538761433498276,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992705268678589,
      "loss": 6.5414,
      "mean_token_accuracy": 0.13685213252902031,
      "num_tokens": 6941663.0,
      "step": 3540
    },
    {
      "entropy": 6.713863706588745,
      "epoch": 0.3543759684110561,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992676525274542,
      "loss": 6.5588,
      "mean_token_accuracy": 0.12517699003219604,
      "num_tokens": 6952373.0,
      "step": 3545
    },
    {
      "entropy": 6.566033792495728,
      "epoch": 0.3548757934722847,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992647725445331,
      "loss": 6.5627,
      "mean_token_accuracy": 0.13661421239376068,
      "num_tokens": 6961337.0,
      "step": 3550
    },
    {
      "entropy": 6.611299800872803,
      "epoch": 0.3553756185335133,
      "grad_norm": 1.0,
      "learning_rate": 0.000499261886919168,
      "loss": 6.5146,
      "mean_token_accuracy": 0.13746540248394012,
      "num_tokens": 6971649.0,
      "step": 3555
    },
    {
      "entropy": 6.637835550308227,
      "epoch": 0.3558754435947418,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992589956514316,
      "loss": 6.5288,
      "mean_token_accuracy": 0.13472158834338188,
      "num_tokens": 6982261.0,
      "step": 3560
    },
    {
      "entropy": 6.612603998184204,
      "epoch": 0.3563752686559704,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992560987413965,
      "loss": 6.5257,
      "mean_token_accuracy": 0.13068577498197556,
      "num_tokens": 6991324.0,
      "step": 3565
    },
    {
      "entropy": 6.5739234447479244,
      "epoch": 0.356875093717199,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992531961891356,
      "loss": 6.5021,
      "mean_token_accuracy": 0.1353994868695736,
      "num_tokens": 7001396.0,
      "step": 3570
    },
    {
      "entropy": 6.6809157371521,
      "epoch": 0.35737491877842753,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992502879947222,
      "loss": 6.5211,
      "mean_token_accuracy": 0.13213353380560874,
      "num_tokens": 7010354.0,
      "step": 3575
    },
    {
      "entropy": 6.7170140743255615,
      "epoch": 0.3578747438396561,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992473741582291,
      "loss": 6.6138,
      "mean_token_accuracy": 0.1270657241344452,
      "num_tokens": 7019835.0,
      "step": 3580
    },
    {
      "entropy": 6.604812335968018,
      "epoch": 0.3583745689008847,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992444546797298,
      "loss": 6.4893,
      "mean_token_accuracy": 0.13503527417778968,
      "num_tokens": 7030515.0,
      "step": 3585
    },
    {
      "entropy": 6.482302570343018,
      "epoch": 0.35887439396211324,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992415295592977,
      "loss": 6.5475,
      "mean_token_accuracy": 0.1355128839612007,
      "num_tokens": 7041248.0,
      "step": 3590
    },
    {
      "entropy": 6.6671782493591305,
      "epoch": 0.3593742190233418,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004992385987970064,
      "loss": 6.5773,
      "mean_token_accuracy": 0.12790912985801697,
      "num_tokens": 7050578.0,
      "step": 3595
    },
    {
      "entropy": 6.6769624710083,
      "epoch": 0.3598740440845704,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992356623929297,
      "loss": 6.5618,
      "mean_token_accuracy": 0.12897216081619262,
      "num_tokens": 7060270.0,
      "step": 3600
    },
    {
      "entropy": 6.673832035064697,
      "epoch": 0.360373869145799,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992327203471414,
      "loss": 6.61,
      "mean_token_accuracy": 0.1256176747381687,
      "num_tokens": 7070259.0,
      "step": 3605
    },
    {
      "entropy": 6.6119787216186525,
      "epoch": 0.36087369420702753,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004992297726597156,
      "loss": 6.5286,
      "mean_token_accuracy": 0.13310132548213005,
      "num_tokens": 7078784.0,
      "step": 3610
    },
    {
      "entropy": 6.685658502578735,
      "epoch": 0.3613735192682561,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992268193307264,
      "loss": 6.576,
      "mean_token_accuracy": 0.12648143023252487,
      "num_tokens": 7088632.0,
      "step": 3615
    },
    {
      "entropy": 6.586120223999023,
      "epoch": 0.3618733443294847,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004992238603602481,
      "loss": 6.5741,
      "mean_token_accuracy": 0.13020266816020012,
      "num_tokens": 7097703.0,
      "step": 3620
    },
    {
      "entropy": 6.632631015777588,
      "epoch": 0.36237316939071323,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992208957483552,
      "loss": 6.4773,
      "mean_token_accuracy": 0.13478429317474366,
      "num_tokens": 7107183.0,
      "step": 3625
    },
    {
      "entropy": 6.700596857070923,
      "epoch": 0.3628729944519418,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992179254951223,
      "loss": 6.6018,
      "mean_token_accuracy": 0.12561058178544043,
      "num_tokens": 7118220.0,
      "step": 3630
    },
    {
      "entropy": 6.612575483322144,
      "epoch": 0.3633728195131704,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499214949600624,
      "loss": 6.5345,
      "mean_token_accuracy": 0.1306740276515484,
      "num_tokens": 7128322.0,
      "step": 3635
    },
    {
      "entropy": 6.569721794128418,
      "epoch": 0.36387264457439894,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004992119680649355,
      "loss": 6.5336,
      "mean_token_accuracy": 0.1307792566716671,
      "num_tokens": 7137284.0,
      "step": 3640
    },
    {
      "entropy": 6.63422269821167,
      "epoch": 0.3643724696356275,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992089808881313,
      "loss": 6.4659,
      "mean_token_accuracy": 0.13586053177714347,
      "num_tokens": 7146598.0,
      "step": 3645
    },
    {
      "entropy": 6.760889482498169,
      "epoch": 0.3648722946968561,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992059880702869,
      "loss": 6.6247,
      "mean_token_accuracy": 0.13818052485585214,
      "num_tokens": 7156816.0,
      "step": 3650
    },
    {
      "entropy": 6.649261569976806,
      "epoch": 0.36537211975808465,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992029896114775,
      "loss": 6.5811,
      "mean_token_accuracy": 0.12462318763136863,
      "num_tokens": 7166855.0,
      "step": 3655
    },
    {
      "entropy": 6.510309648513794,
      "epoch": 0.36587194481931323,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991999855117787,
      "loss": 6.4467,
      "mean_token_accuracy": 0.13755838945508003,
      "num_tokens": 7176477.0,
      "step": 3660
    },
    {
      "entropy": 6.62469482421875,
      "epoch": 0.3663717698805418,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004991969757712659,
      "loss": 6.494,
      "mean_token_accuracy": 0.1370445244014263,
      "num_tokens": 7186173.0,
      "step": 3665
    },
    {
      "entropy": 6.57085165977478,
      "epoch": 0.3668715949417704,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991939603900149,
      "loss": 6.4577,
      "mean_token_accuracy": 0.13398399949073792,
      "num_tokens": 7195403.0,
      "step": 3670
    },
    {
      "entropy": 6.644307947158813,
      "epoch": 0.36737142000299894,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991909393681015,
      "loss": 6.5439,
      "mean_token_accuracy": 0.13114903271198272,
      "num_tokens": 7204179.0,
      "step": 3675
    },
    {
      "entropy": 6.545143175125122,
      "epoch": 0.3678712450642275,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991879127056017,
      "loss": 6.5021,
      "mean_token_accuracy": 0.13197836726903917,
      "num_tokens": 7214159.0,
      "step": 3680
    },
    {
      "entropy": 6.579304599761963,
      "epoch": 0.3683710701254561,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991848804025918,
      "loss": 6.4746,
      "mean_token_accuracy": 0.1352105386555195,
      "num_tokens": 7224410.0,
      "step": 3685
    },
    {
      "entropy": 6.567760944366455,
      "epoch": 0.36887089518668464,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499181842459148,
      "loss": 6.5287,
      "mean_token_accuracy": 0.1385655641555786,
      "num_tokens": 7233740.0,
      "step": 3690
    },
    {
      "entropy": 6.60739917755127,
      "epoch": 0.36937072024791323,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991787988753467,
      "loss": 6.5076,
      "mean_token_accuracy": 0.13437757566571235,
      "num_tokens": 7242834.0,
      "step": 3695
    },
    {
      "entropy": 6.596049737930298,
      "epoch": 0.3698705453091418,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991757496512645,
      "loss": 6.4689,
      "mean_token_accuracy": 0.14437328651547432,
      "num_tokens": 7252098.0,
      "step": 3700
    },
    {
      "entropy": 6.69606122970581,
      "epoch": 0.37037037037037035,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991726947869782,
      "loss": 6.6471,
      "mean_token_accuracy": 0.12470240518450737,
      "num_tokens": 7261396.0,
      "step": 3705
    },
    {
      "entropy": 6.645996379852295,
      "epoch": 0.37087019543159894,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004991696342825645,
      "loss": 6.5657,
      "mean_token_accuracy": 0.12552324086427688,
      "num_tokens": 7271138.0,
      "step": 3710
    },
    {
      "entropy": 6.613135957717896,
      "epoch": 0.3713700204928275,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991665681381005,
      "loss": 6.4879,
      "mean_token_accuracy": 0.13709988445043564,
      "num_tokens": 7280164.0,
      "step": 3715
    },
    {
      "entropy": 6.61772780418396,
      "epoch": 0.37186984555405606,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991634963536633,
      "loss": 6.5355,
      "mean_token_accuracy": 0.12880035787820815,
      "num_tokens": 7288830.0,
      "step": 3720
    },
    {
      "entropy": 6.611929702758789,
      "epoch": 0.37236967061528464,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004991604189293303,
      "loss": 6.5986,
      "mean_token_accuracy": 0.12981483489274978,
      "num_tokens": 7298955.0,
      "step": 3725
    },
    {
      "entropy": 6.655997228622437,
      "epoch": 0.37286949567651323,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991573358651787,
      "loss": 6.4864,
      "mean_token_accuracy": 0.1318989910185337,
      "num_tokens": 7309829.0,
      "step": 3730
    },
    {
      "entropy": 6.608906221389771,
      "epoch": 0.3733693207377418,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991542471612864,
      "loss": 6.5707,
      "mean_token_accuracy": 0.1318919137120247,
      "num_tokens": 7319654.0,
      "step": 3735
    },
    {
      "entropy": 6.657841587066651,
      "epoch": 0.37386914579897035,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991511528177307,
      "loss": 6.5522,
      "mean_token_accuracy": 0.1277618497610092,
      "num_tokens": 7330415.0,
      "step": 3740
    },
    {
      "entropy": 6.699305057525635,
      "epoch": 0.37436897086019894,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991480528345898,
      "loss": 6.5558,
      "mean_token_accuracy": 0.13054054901003836,
      "num_tokens": 7339511.0,
      "step": 3745
    },
    {
      "entropy": 6.668748903274536,
      "epoch": 0.3748687959214275,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991449472119416,
      "loss": 6.5632,
      "mean_token_accuracy": 0.12433832362294198,
      "num_tokens": 7348488.0,
      "step": 3750
    },
    {
      "entropy": 6.558110761642456,
      "epoch": 0.37536862098265605,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991418359498642,
      "loss": 6.5041,
      "mean_token_accuracy": 0.13206134289503096,
      "num_tokens": 7357804.0,
      "step": 3755
    },
    {
      "entropy": 6.666992425918579,
      "epoch": 0.37586844604388464,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004991387190484358,
      "loss": 6.5618,
      "mean_token_accuracy": 0.1281794637441635,
      "num_tokens": 7367644.0,
      "step": 3760
    },
    {
      "entropy": 6.595050287246704,
      "epoch": 0.37636827110511323,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000499135596507735,
      "loss": 6.3975,
      "mean_token_accuracy": 0.13311903551220894,
      "num_tokens": 7377357.0,
      "step": 3765
    },
    {
      "entropy": 6.59745774269104,
      "epoch": 0.37686809616634176,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991324683278402,
      "loss": 6.5948,
      "mean_token_accuracy": 0.1272532433271408,
      "num_tokens": 7387306.0,
      "step": 3770
    },
    {
      "entropy": 6.632649183273315,
      "epoch": 0.37736792122757035,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991293345088301,
      "loss": 6.5351,
      "mean_token_accuracy": 0.13223535642027856,
      "num_tokens": 7397472.0,
      "step": 3775
    },
    {
      "entropy": 6.616477346420288,
      "epoch": 0.37786774628879893,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991261950507837,
      "loss": 6.5722,
      "mean_token_accuracy": 0.1321080319583416,
      "num_tokens": 7408786.0,
      "step": 3780
    },
    {
      "entropy": 6.612939405441284,
      "epoch": 0.37836757135002747,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049912304995378,
      "loss": 6.5466,
      "mean_token_accuracy": 0.12643783465027808,
      "num_tokens": 7417857.0,
      "step": 3785
    },
    {
      "entropy": 6.605091714859009,
      "epoch": 0.37886739641125605,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991198992178979,
      "loss": 6.5288,
      "mean_token_accuracy": 0.1272548370063305,
      "num_tokens": 7427007.0,
      "step": 3790
    },
    {
      "entropy": 6.648046636581421,
      "epoch": 0.37936722147248464,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991167428432168,
      "loss": 6.4306,
      "mean_token_accuracy": 0.13773693814873694,
      "num_tokens": 7437090.0,
      "step": 3795
    },
    {
      "entropy": 6.557371425628662,
      "epoch": 0.3798670465337132,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991135808298162,
      "loss": 6.4864,
      "mean_token_accuracy": 0.1327134557068348,
      "num_tokens": 7446249.0,
      "step": 3800
    },
    {
      "entropy": 6.659952926635742,
      "epoch": 0.38036687159494176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991104131777755,
      "loss": 6.5095,
      "mean_token_accuracy": 0.1275357760488987,
      "num_tokens": 7455838.0,
      "step": 3805
    },
    {
      "entropy": 6.599836826324463,
      "epoch": 0.38086669665617034,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991072398871746,
      "loss": 6.5768,
      "mean_token_accuracy": 0.12697244733572005,
      "num_tokens": 7466057.0,
      "step": 3810
    },
    {
      "entropy": 6.653259801864624,
      "epoch": 0.38136652171739893,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991040609580932,
      "loss": 6.5368,
      "mean_token_accuracy": 0.13301845714449884,
      "num_tokens": 7475174.0,
      "step": 3815
    },
    {
      "entropy": 6.621293783187866,
      "epoch": 0.38186634677862746,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004991008763906112,
      "loss": 6.4778,
      "mean_token_accuracy": 0.1384720526635647,
      "num_tokens": 7484635.0,
      "step": 3820
    },
    {
      "entropy": 6.563889026641846,
      "epoch": 0.38236617183985605,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004990976861848087,
      "loss": 6.5357,
      "mean_token_accuracy": 0.13021118119359015,
      "num_tokens": 7496428.0,
      "step": 3825
    },
    {
      "entropy": 6.613593816757202,
      "epoch": 0.38286599690108464,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990944903407663,
      "loss": 6.5627,
      "mean_token_accuracy": 0.1307319588959217,
      "num_tokens": 7506581.0,
      "step": 3830
    },
    {
      "entropy": 6.635296249389649,
      "epoch": 0.38336582196231317,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990912888585641,
      "loss": 6.4424,
      "mean_token_accuracy": 0.1384415827691555,
      "num_tokens": 7515809.0,
      "step": 3835
    },
    {
      "entropy": 6.487102079391479,
      "epoch": 0.38386564702354176,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990880817382828,
      "loss": 6.466,
      "mean_token_accuracy": 0.13676456585526467,
      "num_tokens": 7525800.0,
      "step": 3840
    },
    {
      "entropy": 6.626596069335937,
      "epoch": 0.38436547208477034,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499084868980003,
      "loss": 6.5738,
      "mean_token_accuracy": 0.13002618104219438,
      "num_tokens": 7536903.0,
      "step": 3845
    },
    {
      "entropy": 6.612464952468872,
      "epoch": 0.3848652971459989,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990816505838054,
      "loss": 6.5298,
      "mean_token_accuracy": 0.1257986307144165,
      "num_tokens": 7546004.0,
      "step": 3850
    },
    {
      "entropy": 6.652705144882202,
      "epoch": 0.38536512220722746,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990784265497713,
      "loss": 6.6585,
      "mean_token_accuracy": 0.1285960353910923,
      "num_tokens": 7555971.0,
      "step": 3855
    },
    {
      "entropy": 6.683477163314819,
      "epoch": 0.38586494726845605,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004990751968779817,
      "loss": 6.5536,
      "mean_token_accuracy": 0.13212867379188536,
      "num_tokens": 7564569.0,
      "step": 3860
    },
    {
      "entropy": 6.594289970397949,
      "epoch": 0.38636477232968464,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990719615685176,
      "loss": 6.5395,
      "mean_token_accuracy": 0.1284814417362213,
      "num_tokens": 7574134.0,
      "step": 3865
    },
    {
      "entropy": 6.67630033493042,
      "epoch": 0.38686459739091317,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004990687206214608,
      "loss": 6.4237,
      "mean_token_accuracy": 0.13904297202825547,
      "num_tokens": 7583631.0,
      "step": 3870
    },
    {
      "entropy": 6.487108564376831,
      "epoch": 0.38736442245214175,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990654740368925,
      "loss": 6.3683,
      "mean_token_accuracy": 0.13795554339885713,
      "num_tokens": 7593281.0,
      "step": 3875
    },
    {
      "entropy": 6.448543834686279,
      "epoch": 0.38786424751337034,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004990622218148947,
      "loss": 6.5306,
      "mean_token_accuracy": 0.13075224310159683,
      "num_tokens": 7603735.0,
      "step": 3880
    },
    {
      "entropy": 6.709584045410156,
      "epoch": 0.3883640725745989,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990589639555489,
      "loss": 6.4985,
      "mean_token_accuracy": 0.1292933240532875,
      "num_tokens": 7613126.0,
      "step": 3885
    },
    {
      "entropy": 6.645783376693726,
      "epoch": 0.38886389763582746,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990557004589373,
      "loss": 6.5278,
      "mean_token_accuracy": 0.12787676602602005,
      "num_tokens": 7623191.0,
      "step": 3890
    },
    {
      "entropy": 6.572176170349121,
      "epoch": 0.38936372269705605,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004990524313251419,
      "loss": 6.5025,
      "mean_token_accuracy": 0.13649313077330588,
      "num_tokens": 7634118.0,
      "step": 3895
    },
    {
      "entropy": 6.572763109207154,
      "epoch": 0.3898635477582846,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004990491565542449,
      "loss": 6.393,
      "mean_token_accuracy": 0.1431651771068573,
      "num_tokens": 7643228.0,
      "step": 3900
    },
    {
      "entropy": 6.490762662887573,
      "epoch": 0.39036337281951317,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990458761463289,
      "loss": 6.4961,
      "mean_token_accuracy": 0.1269106924533844,
      "num_tokens": 7652578.0,
      "step": 3905
    },
    {
      "entropy": 6.612861061096192,
      "epoch": 0.39086319788074175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004990425901014763,
      "loss": 6.4509,
      "mean_token_accuracy": 0.13192059248685836,
      "num_tokens": 7661617.0,
      "step": 3910
    },
    {
      "entropy": 6.5336158752441404,
      "epoch": 0.3913630229419703,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990392984197697,
      "loss": 6.4563,
      "mean_token_accuracy": 0.1394348494708538,
      "num_tokens": 7670864.0,
      "step": 3915
    },
    {
      "entropy": 6.572642993927002,
      "epoch": 0.39186284800319887,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990360011012921,
      "loss": 6.5172,
      "mean_token_accuracy": 0.12711714655160905,
      "num_tokens": 7681528.0,
      "step": 3920
    },
    {
      "entropy": 6.668974161148071,
      "epoch": 0.39236267306442746,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990326981461264,
      "loss": 6.4429,
      "mean_token_accuracy": 0.13255584686994554,
      "num_tokens": 7690398.0,
      "step": 3925
    },
    {
      "entropy": 6.461314916610718,
      "epoch": 0.39286249812565605,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990293895543555,
      "loss": 6.4765,
      "mean_token_accuracy": 0.13599226996302605,
      "num_tokens": 7699843.0,
      "step": 3930
    },
    {
      "entropy": 6.546249485015869,
      "epoch": 0.3933623231868846,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499026075326063,
      "loss": 6.5107,
      "mean_token_accuracy": 0.1358805276453495,
      "num_tokens": 7710181.0,
      "step": 3935
    },
    {
      "entropy": 6.648956441879273,
      "epoch": 0.39386214824811316,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499022755461332,
      "loss": 6.4443,
      "mean_token_accuracy": 0.13313849717378617,
      "num_tokens": 7719148.0,
      "step": 3940
    },
    {
      "entropy": 6.438974761962891,
      "epoch": 0.39436197330934175,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499019429960246,
      "loss": 6.3295,
      "mean_token_accuracy": 0.14517943039536477,
      "num_tokens": 7728551.0,
      "step": 3945
    },
    {
      "entropy": 6.55150351524353,
      "epoch": 0.3948617983705703,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990160988228888,
      "loss": 6.5259,
      "mean_token_accuracy": 0.12912257239222527,
      "num_tokens": 7739083.0,
      "step": 3950
    },
    {
      "entropy": 6.653018426895142,
      "epoch": 0.39536162343179887,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990127620493444,
      "loss": 6.5178,
      "mean_token_accuracy": 0.13356704488396645,
      "num_tokens": 7749684.0,
      "step": 3955
    },
    {
      "entropy": 6.540141296386719,
      "epoch": 0.39586144849302746,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004990094196396964,
      "loss": 6.3969,
      "mean_token_accuracy": 0.13822795003652572,
      "num_tokens": 7759281.0,
      "step": 3960
    },
    {
      "entropy": 6.52960433959961,
      "epoch": 0.396361273554256,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499006071594029,
      "loss": 6.4888,
      "mean_token_accuracy": 0.13620271161198616,
      "num_tokens": 7768612.0,
      "step": 3965
    },
    {
      "entropy": 6.552225589752197,
      "epoch": 0.3968610986154846,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990027179124266,
      "loss": 6.424,
      "mean_token_accuracy": 0.13467555791139602,
      "num_tokens": 7778333.0,
      "step": 3970
    },
    {
      "entropy": 6.65633454322815,
      "epoch": 0.39736092367671316,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989993585949733,
      "loss": 6.5541,
      "mean_token_accuracy": 0.12997083067893983,
      "num_tokens": 7787547.0,
      "step": 3975
    },
    {
      "entropy": 6.531152963638306,
      "epoch": 0.3978607487379417,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989959936417541,
      "loss": 6.5018,
      "mean_token_accuracy": 0.13278501704335213,
      "num_tokens": 7798407.0,
      "step": 3980
    },
    {
      "entropy": 6.558915996551514,
      "epoch": 0.3983605737991703,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000498992623052853,
      "loss": 6.4154,
      "mean_token_accuracy": 0.14237021654844284,
      "num_tokens": 7807879.0,
      "step": 3985
    },
    {
      "entropy": 6.485930013656616,
      "epoch": 0.39886039886039887,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004989892468283553,
      "loss": 6.3977,
      "mean_token_accuracy": 0.13894546329975127,
      "num_tokens": 7817210.0,
      "step": 3990
    },
    {
      "entropy": 6.660465812683105,
      "epoch": 0.39936022392162746,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004989858649683457,
      "loss": 6.6211,
      "mean_token_accuracy": 0.12636715099215506,
      "num_tokens": 7827845.0,
      "step": 3995
    },
    {
      "entropy": 6.727644062042236,
      "epoch": 0.399860048982856,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989824774729094,
      "loss": 6.5035,
      "mean_token_accuracy": 0.13661194518208503,
      "num_tokens": 7836994.0,
      "step": 4000
    },
    {
      "entropy": 6.53096227645874,
      "epoch": 0.4003598740440846,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989790843421317,
      "loss": 6.5293,
      "mean_token_accuracy": 0.13115536719560622,
      "num_tokens": 7846733.0,
      "step": 4005
    },
    {
      "entropy": 6.515123510360718,
      "epoch": 0.40085969910531316,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004989756855760979,
      "loss": 6.4429,
      "mean_token_accuracy": 0.13563456535339355,
      "num_tokens": 7856393.0,
      "step": 4010
    },
    {
      "entropy": 6.645159435272217,
      "epoch": 0.4013595241665417,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989722811748934,
      "loss": 6.5328,
      "mean_token_accuracy": 0.13331457674503328,
      "num_tokens": 7866394.0,
      "step": 4015
    },
    {
      "entropy": 6.595888614654541,
      "epoch": 0.4018593492277703,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989688711386039,
      "loss": 6.525,
      "mean_token_accuracy": 0.1277965322136879,
      "num_tokens": 7877084.0,
      "step": 4020
    },
    {
      "entropy": 6.656990814208984,
      "epoch": 0.40235917428899887,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989654554673152,
      "loss": 6.5519,
      "mean_token_accuracy": 0.1284612499177456,
      "num_tokens": 7887289.0,
      "step": 4025
    },
    {
      "entropy": 6.553845167160034,
      "epoch": 0.4028589993502274,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989620341611133,
      "loss": 6.4928,
      "mean_token_accuracy": 0.1331907719373703,
      "num_tokens": 7897315.0,
      "step": 4030
    },
    {
      "entropy": 6.681213474273681,
      "epoch": 0.403358824411456,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989586072200843,
      "loss": 6.5918,
      "mean_token_accuracy": 0.12706032618880272,
      "num_tokens": 7905913.0,
      "step": 4035
    },
    {
      "entropy": 6.57252607345581,
      "epoch": 0.40385864947268457,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989551746443143,
      "loss": 6.3878,
      "mean_token_accuracy": 0.1398092433810234,
      "num_tokens": 7914879.0,
      "step": 4040
    },
    {
      "entropy": 6.54695086479187,
      "epoch": 0.4043584745339131,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989517364338899,
      "loss": 6.5407,
      "mean_token_accuracy": 0.13144304752349853,
      "num_tokens": 7925163.0,
      "step": 4045
    },
    {
      "entropy": 6.656905651092529,
      "epoch": 0.4048582995951417,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989482925888974,
      "loss": 6.4442,
      "mean_token_accuracy": 0.13604791313409806,
      "num_tokens": 7934380.0,
      "step": 4050
    },
    {
      "entropy": 6.5723072528839115,
      "epoch": 0.4053581246563703,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004989448431094233,
      "loss": 6.505,
      "mean_token_accuracy": 0.12940779998898505,
      "num_tokens": 7944600.0,
      "step": 4055
    },
    {
      "entropy": 6.492002820968628,
      "epoch": 0.40585794971759886,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989413879955548,
      "loss": 6.4116,
      "mean_token_accuracy": 0.1383023105561733,
      "num_tokens": 7953545.0,
      "step": 4060
    },
    {
      "entropy": 6.512488126754761,
      "epoch": 0.4063577747788274,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989379272473785,
      "loss": 6.4916,
      "mean_token_accuracy": 0.13396764397621155,
      "num_tokens": 7962931.0,
      "step": 4065
    },
    {
      "entropy": 6.619950819015503,
      "epoch": 0.406857599840056,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004989344608649815,
      "loss": 6.442,
      "mean_token_accuracy": 0.13626711219549179,
      "num_tokens": 7973090.0,
      "step": 4070
    },
    {
      "entropy": 6.57343430519104,
      "epoch": 0.40735742490128457,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004989309888484511,
      "loss": 6.5865,
      "mean_token_accuracy": 0.12389916554093361,
      "num_tokens": 7982208.0,
      "step": 4075
    },
    {
      "entropy": 6.533903121948242,
      "epoch": 0.4078572499625131,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989275111978748,
      "loss": 6.5283,
      "mean_token_accuracy": 0.1353437304496765,
      "num_tokens": 7993389.0,
      "step": 4080
    },
    {
      "entropy": 6.577138662338257,
      "epoch": 0.4083570750237417,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004989240279133398,
      "loss": 6.4166,
      "mean_token_accuracy": 0.13624165430665017,
      "num_tokens": 8002749.0,
      "step": 4085
    },
    {
      "entropy": 6.607124853134155,
      "epoch": 0.4088569000849703,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004989205389949339,
      "loss": 6.4177,
      "mean_token_accuracy": 0.13583638072013854,
      "num_tokens": 8011552.0,
      "step": 4090
    },
    {
      "entropy": 6.465263938903808,
      "epoch": 0.4093567251461988,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989170444427448,
      "loss": 6.4401,
      "mean_token_accuracy": 0.13448849990963935,
      "num_tokens": 8020622.0,
      "step": 4095
    },
    {
      "entropy": 6.486138868331909,
      "epoch": 0.4098565502074274,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989135442568606,
      "loss": 6.3737,
      "mean_token_accuracy": 0.13940291702747346,
      "num_tokens": 8030356.0,
      "step": 4100
    },
    {
      "entropy": 6.590310192108154,
      "epoch": 0.410356375268656,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004989100384373693,
      "loss": 6.4842,
      "mean_token_accuracy": 0.12944861948490144,
      "num_tokens": 8039903.0,
      "step": 4105
    },
    {
      "entropy": 6.558611154556274,
      "epoch": 0.4108562003298845,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989065269843588,
      "loss": 6.4327,
      "mean_token_accuracy": 0.1317477710545063,
      "num_tokens": 8050488.0,
      "step": 4110
    },
    {
      "entropy": 6.529536294937134,
      "epoch": 0.4113560253911131,
      "grad_norm": 1.09375,
      "learning_rate": 0.000498903009897918,
      "loss": 6.451,
      "mean_token_accuracy": 0.13413968533277512,
      "num_tokens": 8060179.0,
      "step": 4115
    },
    {
      "entropy": 6.548681688308716,
      "epoch": 0.4118558504523417,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988994871781349,
      "loss": 6.5179,
      "mean_token_accuracy": 0.13198244497179984,
      "num_tokens": 8070915.0,
      "step": 4120
    },
    {
      "entropy": 6.566651725769043,
      "epoch": 0.4123556755135703,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004988959588250983,
      "loss": 6.4498,
      "mean_token_accuracy": 0.13837631642818451,
      "num_tokens": 8080203.0,
      "step": 4125
    },
    {
      "entropy": 6.558019542694092,
      "epoch": 0.4128555005747988,
      "grad_norm": 1.078125,
      "learning_rate": 0.000498892424838897,
      "loss": 6.4488,
      "mean_token_accuracy": 0.13629123866558074,
      "num_tokens": 8090552.0,
      "step": 4130
    },
    {
      "entropy": 6.532246923446655,
      "epoch": 0.4133553256360274,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004988888852196199,
      "loss": 6.4183,
      "mean_token_accuracy": 0.13859262466430664,
      "num_tokens": 8099947.0,
      "step": 4135
    },
    {
      "entropy": 6.519168424606323,
      "epoch": 0.413855150697256,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004988853399673562,
      "loss": 6.5004,
      "mean_token_accuracy": 0.12965430393815042,
      "num_tokens": 8110502.0,
      "step": 4140
    },
    {
      "entropy": 6.593302869796753,
      "epoch": 0.4143549757584845,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004988817890821948,
      "loss": 6.5276,
      "mean_token_accuracy": 0.12973300293087958,
      "num_tokens": 8118762.0,
      "step": 4145
    },
    {
      "entropy": 6.579014635086059,
      "epoch": 0.4148548008197131,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004988782325642252,
      "loss": 6.4409,
      "mean_token_accuracy": 0.12978375628590583,
      "num_tokens": 8129072.0,
      "step": 4150
    },
    {
      "entropy": 6.577251672744751,
      "epoch": 0.4153546258809417,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000498874670413537,
      "loss": 6.4643,
      "mean_token_accuracy": 0.12749237492680549,
      "num_tokens": 8138504.0,
      "step": 4155
    },
    {
      "entropy": 6.687809801101684,
      "epoch": 0.4158544509421702,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004988711026302197,
      "loss": 6.5391,
      "mean_token_accuracy": 0.1325111597776413,
      "num_tokens": 8148534.0,
      "step": 4160
    },
    {
      "entropy": 6.560407733917236,
      "epoch": 0.4163542760033988,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004988675292143629,
      "loss": 6.5791,
      "mean_token_accuracy": 0.12417998611927032,
      "num_tokens": 8160272.0,
      "step": 4165
    },
    {
      "entropy": 6.616597700119018,
      "epoch": 0.4168541010646274,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004988639501660568,
      "loss": 6.4497,
      "mean_token_accuracy": 0.13744847923517228,
      "num_tokens": 8170053.0,
      "step": 4170
    },
    {
      "entropy": 6.641278076171875,
      "epoch": 0.4173539261258559,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004988603654853913,
      "loss": 6.4686,
      "mean_token_accuracy": 0.1311160683631897,
      "num_tokens": 8179425.0,
      "step": 4175
    },
    {
      "entropy": 6.508156824111938,
      "epoch": 0.4178537511870845,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004988567751724565,
      "loss": 6.4887,
      "mean_token_accuracy": 0.13093385472893715,
      "num_tokens": 8189846.0,
      "step": 4180
    },
    {
      "entropy": 6.579239797592163,
      "epoch": 0.4183535762483131,
      "grad_norm": 1.078125,
      "learning_rate": 0.000498853179227343,
      "loss": 6.5015,
      "mean_token_accuracy": 0.13485166803002357,
      "num_tokens": 8198596.0,
      "step": 4185
    },
    {
      "entropy": 6.628087949752808,
      "epoch": 0.4188534013095417,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000498849577650141,
      "loss": 6.4227,
      "mean_token_accuracy": 0.13375448137521745,
      "num_tokens": 8206849.0,
      "step": 4190
    },
    {
      "entropy": 6.547364807128906,
      "epoch": 0.4193532263707702,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004988459704409413,
      "loss": 6.4421,
      "mean_token_accuracy": 0.13590684682130813,
      "num_tokens": 8216260.0,
      "step": 4195
    },
    {
      "entropy": 6.561947917938232,
      "epoch": 0.4198530514319988,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004988423575998345,
      "loss": 6.4716,
      "mean_token_accuracy": 0.13637885823845863,
      "num_tokens": 8224650.0,
      "step": 4200
    },
    {
      "entropy": 6.416500473022461,
      "epoch": 0.4203528764932274,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004988387391269117,
      "loss": 6.3113,
      "mean_token_accuracy": 0.1439281329512596,
      "num_tokens": 8235276.0,
      "step": 4205
    },
    {
      "entropy": 6.57451548576355,
      "epoch": 0.4208527015544559,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004988351150222638,
      "loss": 6.4129,
      "mean_token_accuracy": 0.13464779406785965,
      "num_tokens": 8244687.0,
      "step": 4210
    },
    {
      "entropy": 6.576171207427978,
      "epoch": 0.4213525266156845,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988314852859819,
      "loss": 6.5495,
      "mean_token_accuracy": 0.1311263471841812,
      "num_tokens": 8255091.0,
      "step": 4215
    },
    {
      "entropy": 6.430039262771606,
      "epoch": 0.4218523516769131,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004988278499181575,
      "loss": 6.3527,
      "mean_token_accuracy": 0.14942151606082915,
      "num_tokens": 8265172.0,
      "step": 4220
    },
    {
      "entropy": 6.5439403533935545,
      "epoch": 0.4223521767381416,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000498824208918882,
      "loss": 6.433,
      "mean_token_accuracy": 0.13363671451807022,
      "num_tokens": 8275625.0,
      "step": 4225
    },
    {
      "entropy": 6.441909456253052,
      "epoch": 0.4228520017993702,
      "grad_norm": 1.140625,
      "learning_rate": 0.000498820562288247,
      "loss": 6.4127,
      "mean_token_accuracy": 0.1346836991608143,
      "num_tokens": 8284343.0,
      "step": 4230
    },
    {
      "entropy": 6.592450761795044,
      "epoch": 0.4233518268605988,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004988169100263442,
      "loss": 6.4988,
      "mean_token_accuracy": 0.12972462475299834,
      "num_tokens": 8293430.0,
      "step": 4235
    },
    {
      "entropy": 6.562317848205566,
      "epoch": 0.4238516519218274,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004988132521332656,
      "loss": 6.4753,
      "mean_token_accuracy": 0.12977709472179413,
      "num_tokens": 8302925.0,
      "step": 4240
    },
    {
      "entropy": 6.502162170410156,
      "epoch": 0.4243514769830559,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004988095886091031,
      "loss": 6.4335,
      "mean_token_accuracy": 0.1341563068330288,
      "num_tokens": 8312134.0,
      "step": 4245
    },
    {
      "entropy": 6.54831829071045,
      "epoch": 0.4248513020442845,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988059194539489,
      "loss": 6.4208,
      "mean_token_accuracy": 0.1345089428126812,
      "num_tokens": 8320942.0,
      "step": 4250
    },
    {
      "entropy": 6.595415735244751,
      "epoch": 0.4253511271055131,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004988022446678954,
      "loss": 6.4512,
      "mean_token_accuracy": 0.13361355662345886,
      "num_tokens": 8331898.0,
      "step": 4255
    },
    {
      "entropy": 6.467858219146729,
      "epoch": 0.4258509521667416,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987985642510352,
      "loss": 6.4475,
      "mean_token_accuracy": 0.1376283772289753,
      "num_tokens": 8341728.0,
      "step": 4260
    },
    {
      "entropy": 6.537617444992065,
      "epoch": 0.4263507772279702,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987948782034605,
      "loss": 6.3813,
      "mean_token_accuracy": 0.13970415219664573,
      "num_tokens": 8351611.0,
      "step": 4265
    },
    {
      "entropy": 6.519968032836914,
      "epoch": 0.4268506022891988,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004987911865252644,
      "loss": 6.4076,
      "mean_token_accuracy": 0.1417715758085251,
      "num_tokens": 8360528.0,
      "step": 4270
    },
    {
      "entropy": 6.440678501129151,
      "epoch": 0.42735042735042733,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004987874892165395,
      "loss": 6.4679,
      "mean_token_accuracy": 0.1354478821158409,
      "num_tokens": 8369898.0,
      "step": 4275
    },
    {
      "entropy": 6.54773736000061,
      "epoch": 0.4278502524116559,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000498783786277379,
      "loss": 6.4142,
      "mean_token_accuracy": 0.1384521782398224,
      "num_tokens": 8379928.0,
      "step": 4280
    },
    {
      "entropy": 6.575797080993652,
      "epoch": 0.4283500774728845,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004987800777078759,
      "loss": 6.4527,
      "mean_token_accuracy": 0.1369698688387871,
      "num_tokens": 8389208.0,
      "step": 4285
    },
    {
      "entropy": 6.507793331146241,
      "epoch": 0.42884990253411304,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004987763635081238,
      "loss": 6.4523,
      "mean_token_accuracy": 0.13886258453130723,
      "num_tokens": 8398364.0,
      "step": 4290
    },
    {
      "entropy": 6.527761459350586,
      "epoch": 0.4293497275953416,
      "grad_norm": 1.03125,
      "learning_rate": 0.000498772643678216,
      "loss": 6.4492,
      "mean_token_accuracy": 0.13071525171399118,
      "num_tokens": 8408756.0,
      "step": 4295
    },
    {
      "entropy": 6.588776969909668,
      "epoch": 0.4298495526565702,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987689182182459,
      "loss": 6.4157,
      "mean_token_accuracy": 0.1393563099205494,
      "num_tokens": 8418318.0,
      "step": 4300
    },
    {
      "entropy": 6.485627889633179,
      "epoch": 0.4303493777177988,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987651871283075,
      "loss": 6.4512,
      "mean_token_accuracy": 0.129922616481781,
      "num_tokens": 8427709.0,
      "step": 4305
    },
    {
      "entropy": 6.570569610595703,
      "epoch": 0.43084920277902733,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004987614504084946,
      "loss": 6.4601,
      "mean_token_accuracy": 0.12916277721524239,
      "num_tokens": 8437848.0,
      "step": 4310
    },
    {
      "entropy": 6.584947824478149,
      "epoch": 0.4313490278402559,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987577080589012,
      "loss": 6.4075,
      "mean_token_accuracy": 0.1359340153634548,
      "num_tokens": 8447466.0,
      "step": 4315
    },
    {
      "entropy": 6.501195859909058,
      "epoch": 0.4318488529014845,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987539600796213,
      "loss": 6.411,
      "mean_token_accuracy": 0.14214334413409233,
      "num_tokens": 8456651.0,
      "step": 4320
    },
    {
      "entropy": 6.558067512512207,
      "epoch": 0.43234867796271303,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987502064707494,
      "loss": 6.4643,
      "mean_token_accuracy": 0.1354033648967743,
      "num_tokens": 8466421.0,
      "step": 4325
    },
    {
      "entropy": 6.496439695358276,
      "epoch": 0.4328485030239416,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987464472323799,
      "loss": 6.4057,
      "mean_token_accuracy": 0.13391259014606477,
      "num_tokens": 8476419.0,
      "step": 4330
    },
    {
      "entropy": 6.551824522018433,
      "epoch": 0.4333483280851702,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004987426823646074,
      "loss": 6.4302,
      "mean_token_accuracy": 0.13297546803951263,
      "num_tokens": 8486652.0,
      "step": 4335
    },
    {
      "entropy": 6.498467206954956,
      "epoch": 0.43384815314639874,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004987389118675266,
      "loss": 6.3827,
      "mean_token_accuracy": 0.13717494830489158,
      "num_tokens": 8495965.0,
      "step": 4340
    },
    {
      "entropy": 6.535243654251099,
      "epoch": 0.4343479782076273,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987351357412322,
      "loss": 6.4479,
      "mean_token_accuracy": 0.1338575914502144,
      "num_tokens": 8506018.0,
      "step": 4345
    },
    {
      "entropy": 6.497014379501342,
      "epoch": 0.4348478032688559,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004987313539858196,
      "loss": 6.4454,
      "mean_token_accuracy": 0.13558314517140388,
      "num_tokens": 8516178.0,
      "step": 4350
    },
    {
      "entropy": 6.585453891754151,
      "epoch": 0.43534762833008445,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004987275666013836,
      "loss": 6.4918,
      "mean_token_accuracy": 0.13078081533312796,
      "num_tokens": 8525548.0,
      "step": 4355
    },
    {
      "entropy": 6.610603332519531,
      "epoch": 0.43584745339131303,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987237735880195,
      "loss": 6.524,
      "mean_token_accuracy": 0.1313353419303894,
      "num_tokens": 8535941.0,
      "step": 4360
    },
    {
      "entropy": 6.449689245223999,
      "epoch": 0.4363472784525416,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004987199749458229,
      "loss": 6.3392,
      "mean_token_accuracy": 0.1431761458516121,
      "num_tokens": 8545381.0,
      "step": 4365
    },
    {
      "entropy": 6.507636165618896,
      "epoch": 0.4368471035137702,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004987161706748893,
      "loss": 6.4532,
      "mean_token_accuracy": 0.13456403464078903,
      "num_tokens": 8555827.0,
      "step": 4370
    },
    {
      "entropy": 6.456200647354126,
      "epoch": 0.43734692857499874,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987123607753144,
      "loss": 6.3278,
      "mean_token_accuracy": 0.14644061625003815,
      "num_tokens": 8565508.0,
      "step": 4375
    },
    {
      "entropy": 6.585708093643189,
      "epoch": 0.4378467536362273,
      "grad_norm": 1.125,
      "learning_rate": 0.000498708545247194,
      "loss": 6.5418,
      "mean_token_accuracy": 0.12930210530757905,
      "num_tokens": 8576273.0,
      "step": 4380
    },
    {
      "entropy": 6.59512243270874,
      "epoch": 0.4383465786974559,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004987047240906243,
      "loss": 6.4783,
      "mean_token_accuracy": 0.13295421227812768,
      "num_tokens": 8585488.0,
      "step": 4385
    },
    {
      "entropy": 6.563721609115601,
      "epoch": 0.43884640375868444,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004987008973057012,
      "loss": 6.5241,
      "mean_token_accuracy": 0.1319626584649086,
      "num_tokens": 8595299.0,
      "step": 4390
    },
    {
      "entropy": 6.517096900939942,
      "epoch": 0.43934622881991303,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986970648925211,
      "loss": 6.3914,
      "mean_token_accuracy": 0.13585217520594597,
      "num_tokens": 8604910.0,
      "step": 4395
    },
    {
      "entropy": 6.542190885543823,
      "epoch": 0.4398460538811416,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004986932268511804,
      "loss": 6.4173,
      "mean_token_accuracy": 0.1373402938246727,
      "num_tokens": 8614966.0,
      "step": 4400
    },
    {
      "entropy": 6.479302310943604,
      "epoch": 0.44034587894237015,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004986893831817758,
      "loss": 6.4699,
      "mean_token_accuracy": 0.1341715231537819,
      "num_tokens": 8624539.0,
      "step": 4405
    },
    {
      "entropy": 6.482782411575317,
      "epoch": 0.44084570400359874,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986855338844038,
      "loss": 6.3063,
      "mean_token_accuracy": 0.14404137432575226,
      "num_tokens": 8633775.0,
      "step": 4410
    },
    {
      "entropy": 6.537249374389648,
      "epoch": 0.4413455290648273,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004986816789591613,
      "loss": 6.4477,
      "mean_token_accuracy": 0.13106013759970664,
      "num_tokens": 8643798.0,
      "step": 4415
    },
    {
      "entropy": 6.402342319488525,
      "epoch": 0.44184535412605586,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004986778184061453,
      "loss": 6.3681,
      "mean_token_accuracy": 0.14136381298303605,
      "num_tokens": 8653695.0,
      "step": 4420
    },
    {
      "entropy": 6.498728609085083,
      "epoch": 0.44234517918728444,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986739522254529,
      "loss": 6.3935,
      "mean_token_accuracy": 0.133646971732378,
      "num_tokens": 8663521.0,
      "step": 4425
    },
    {
      "entropy": 6.602633380889893,
      "epoch": 0.44284500424851303,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986700804171816,
      "loss": 6.5281,
      "mean_token_accuracy": 0.13027760684490203,
      "num_tokens": 8673108.0,
      "step": 4430
    },
    {
      "entropy": 6.553408813476563,
      "epoch": 0.4433448293097416,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004986662029814285,
      "loss": 6.3673,
      "mean_token_accuracy": 0.13585576713085173,
      "num_tokens": 8682917.0,
      "step": 4435
    },
    {
      "entropy": 6.493464040756225,
      "epoch": 0.44384465437097015,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004986623199182913,
      "loss": 6.5117,
      "mean_token_accuracy": 0.12620678618550302,
      "num_tokens": 8692436.0,
      "step": 4440
    },
    {
      "entropy": 6.477138090133667,
      "epoch": 0.44434447943219874,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004986584312278678,
      "loss": 6.4052,
      "mean_token_accuracy": 0.13831330388784407,
      "num_tokens": 8702627.0,
      "step": 4445
    },
    {
      "entropy": 6.529372692108154,
      "epoch": 0.4448443044934273,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004986545369102556,
      "loss": 6.4209,
      "mean_token_accuracy": 0.12833660021424292,
      "num_tokens": 8711697.0,
      "step": 4450
    },
    {
      "entropy": 6.534554958343506,
      "epoch": 0.44534412955465585,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004986506369655528,
      "loss": 6.4369,
      "mean_token_accuracy": 0.12715906277298927,
      "num_tokens": 8721449.0,
      "step": 4455
    },
    {
      "entropy": 6.482955360412598,
      "epoch": 0.44584395461588444,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986467313938575,
      "loss": 6.4338,
      "mean_token_accuracy": 0.13206081166863443,
      "num_tokens": 8731829.0,
      "step": 4460
    },
    {
      "entropy": 6.509197664260864,
      "epoch": 0.44634377967711303,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000498642820195268,
      "loss": 6.2239,
      "mean_token_accuracy": 0.15030839443206787,
      "num_tokens": 8741768.0,
      "step": 4465
    },
    {
      "entropy": 6.541743755340576,
      "epoch": 0.44684360473834156,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986389033698827,
      "loss": 6.5257,
      "mean_token_accuracy": 0.12767333239316941,
      "num_tokens": 8751473.0,
      "step": 4470
    },
    {
      "entropy": 6.594388389587403,
      "epoch": 0.44734342979957015,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986349809178002,
      "loss": 6.4339,
      "mean_token_accuracy": 0.12825369611382484,
      "num_tokens": 8761229.0,
      "step": 4475
    },
    {
      "entropy": 6.460408163070679,
      "epoch": 0.44784325486079873,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004986310528391192,
      "loss": 6.3992,
      "mean_token_accuracy": 0.13727328032255173,
      "num_tokens": 8770670.0,
      "step": 4480
    },
    {
      "entropy": 6.5151279926300045,
      "epoch": 0.44834307992202727,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004986271191339383,
      "loss": 6.424,
      "mean_token_accuracy": 0.13391717299818992,
      "num_tokens": 8780816.0,
      "step": 4485
    },
    {
      "entropy": 6.590042543411255,
      "epoch": 0.44884290498325585,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004986231798023566,
      "loss": 6.4903,
      "mean_token_accuracy": 0.13050450384616852,
      "num_tokens": 8790304.0,
      "step": 4490
    },
    {
      "entropy": 6.51934061050415,
      "epoch": 0.44934273004448444,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004986192348444734,
      "loss": 6.4289,
      "mean_token_accuracy": 0.13259972184896468,
      "num_tokens": 8799188.0,
      "step": 4495
    },
    {
      "entropy": 6.5724410057067875,
      "epoch": 0.449842555105713,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986152842603877,
      "loss": 6.4841,
      "mean_token_accuracy": 0.1367173857986927,
      "num_tokens": 8810194.0,
      "step": 4500
    },
    {
      "entropy": 6.550623226165771,
      "epoch": 0.45034238016694156,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004986113280501989,
      "loss": 6.4936,
      "mean_token_accuracy": 0.13392266556620597,
      "num_tokens": 8820394.0,
      "step": 4505
    },
    {
      "entropy": 6.518058443069458,
      "epoch": 0.45084220522817015,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004986073662140068,
      "loss": 6.4358,
      "mean_token_accuracy": 0.1375273697078228,
      "num_tokens": 8831308.0,
      "step": 4510
    },
    {
      "entropy": 6.5074177265167235,
      "epoch": 0.45134203028939873,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004986033987519109,
      "loss": 6.3587,
      "mean_token_accuracy": 0.13824616745114326,
      "num_tokens": 8841461.0,
      "step": 4515
    },
    {
      "entropy": 6.522912502288818,
      "epoch": 0.45184185535062726,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985994256640108,
      "loss": 6.3972,
      "mean_token_accuracy": 0.13351252377033235,
      "num_tokens": 8850356.0,
      "step": 4520
    },
    {
      "entropy": 6.560211801528931,
      "epoch": 0.45234168041185585,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985954469504069,
      "loss": 6.4882,
      "mean_token_accuracy": 0.12924721613526344,
      "num_tokens": 8859863.0,
      "step": 4525
    },
    {
      "entropy": 6.542526531219482,
      "epoch": 0.45284150547308444,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004985914626111991,
      "loss": 6.3715,
      "mean_token_accuracy": 0.1316630944609642,
      "num_tokens": 8869887.0,
      "step": 4530
    },
    {
      "entropy": 6.534689950942993,
      "epoch": 0.45334133053431297,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985874726464875,
      "loss": 6.5185,
      "mean_token_accuracy": 0.1328124076128006,
      "num_tokens": 8879950.0,
      "step": 4535
    },
    {
      "entropy": 6.506281328201294,
      "epoch": 0.45384115559554156,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985834770563727,
      "loss": 6.2745,
      "mean_token_accuracy": 0.14639786183834075,
      "num_tokens": 8888304.0,
      "step": 4540
    },
    {
      "entropy": 6.448099851608276,
      "epoch": 0.45434098065677014,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004985794758409552,
      "loss": 6.2773,
      "mean_token_accuracy": 0.1479883685708046,
      "num_tokens": 8898642.0,
      "step": 4545
    },
    {
      "entropy": 6.4671882629394535,
      "epoch": 0.4548408057179987,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004985754690003354,
      "loss": 6.5083,
      "mean_token_accuracy": 0.12737697884440422,
      "num_tokens": 8909202.0,
      "step": 4550
    },
    {
      "entropy": 6.428459501266479,
      "epoch": 0.45534063077922726,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004985714565346145,
      "loss": 6.3327,
      "mean_token_accuracy": 0.14508851915597915,
      "num_tokens": 8918844.0,
      "step": 4555
    },
    {
      "entropy": 6.563408041000367,
      "epoch": 0.45584045584045585,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004985674384438933,
      "loss": 6.4299,
      "mean_token_accuracy": 0.13017046675086022,
      "num_tokens": 8928706.0,
      "step": 4560
    },
    {
      "entropy": 6.520368385314941,
      "epoch": 0.45634028090168444,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004985634147282729,
      "loss": 6.3717,
      "mean_token_accuracy": 0.13715771734714508,
      "num_tokens": 8939293.0,
      "step": 4565
    },
    {
      "entropy": 6.4421862125396725,
      "epoch": 0.45684010596291297,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004985593853878544,
      "loss": 6.3647,
      "mean_token_accuracy": 0.13338134214282035,
      "num_tokens": 8947922.0,
      "step": 4570
    },
    {
      "entropy": 6.471314525604248,
      "epoch": 0.45733993102414155,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985553504227394,
      "loss": 6.3686,
      "mean_token_accuracy": 0.14011099338531494,
      "num_tokens": 8957745.0,
      "step": 4575
    },
    {
      "entropy": 6.5710876941680905,
      "epoch": 0.45783975608537014,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004985513098330293,
      "loss": 6.5076,
      "mean_token_accuracy": 0.128147342056036,
      "num_tokens": 8968331.0,
      "step": 4580
    },
    {
      "entropy": 6.486983776092529,
      "epoch": 0.4583395811465987,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004985472636188257,
      "loss": 6.357,
      "mean_token_accuracy": 0.1420034348964691,
      "num_tokens": 8977752.0,
      "step": 4585
    },
    {
      "entropy": 6.475470018386841,
      "epoch": 0.45883940620782726,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004985432117802305,
      "loss": 6.4062,
      "mean_token_accuracy": 0.13253813087940217,
      "num_tokens": 8986701.0,
      "step": 4590
    },
    {
      "entropy": 6.4574915885925295,
      "epoch": 0.45933923126905585,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004985391543173456,
      "loss": 6.4363,
      "mean_token_accuracy": 0.1383625775575638,
      "num_tokens": 8996192.0,
      "step": 4595
    },
    {
      "entropy": 6.53081111907959,
      "epoch": 0.4598390563302844,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498535091230273,
      "loss": 6.3961,
      "mean_token_accuracy": 0.13408192694187165,
      "num_tokens": 9007059.0,
      "step": 4600
    },
    {
      "entropy": 6.431706523895263,
      "epoch": 0.46033888139151297,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004985310225191152,
      "loss": 6.3542,
      "mean_token_accuracy": 0.13652804791927337,
      "num_tokens": 9016923.0,
      "step": 4605
    },
    {
      "entropy": 6.507685661315918,
      "epoch": 0.46083870645274155,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004985269481839744,
      "loss": 6.4386,
      "mean_token_accuracy": 0.13447274193167685,
      "num_tokens": 9026034.0,
      "step": 4610
    },
    {
      "entropy": 6.65610179901123,
      "epoch": 0.4613385315139701,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004985228682249529,
      "loss": 6.4619,
      "mean_token_accuracy": 0.13391346409916877,
      "num_tokens": 9036773.0,
      "step": 4615
    },
    {
      "entropy": 6.514581489562988,
      "epoch": 0.46183835657519867,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985187826421538,
      "loss": 6.4226,
      "mean_token_accuracy": 0.13582958281040192,
      "num_tokens": 9047825.0,
      "step": 4620
    },
    {
      "entropy": 6.560742998123169,
      "epoch": 0.46233818163642726,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004985146914356795,
      "loss": 6.5195,
      "mean_token_accuracy": 0.1312625989317894,
      "num_tokens": 9056295.0,
      "step": 4625
    },
    {
      "entropy": 6.4778799533844,
      "epoch": 0.46283800669765585,
      "grad_norm": 1.0625,
      "learning_rate": 0.000498510594605633,
      "loss": 6.4434,
      "mean_token_accuracy": 0.1328299857676029,
      "num_tokens": 9067245.0,
      "step": 4630
    },
    {
      "entropy": 6.512774324417114,
      "epoch": 0.4633378317588844,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004985064921521177,
      "loss": 6.4092,
      "mean_token_accuracy": 0.13674387484788894,
      "num_tokens": 9077528.0,
      "step": 4635
    },
    {
      "entropy": 6.516362190246582,
      "epoch": 0.46383765682011296,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004985023840752364,
      "loss": 6.3513,
      "mean_token_accuracy": 0.1389579489827156,
      "num_tokens": 9086015.0,
      "step": 4640
    },
    {
      "entropy": 6.543647289276123,
      "epoch": 0.46433748188134155,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004984982703750926,
      "loss": 6.416,
      "mean_token_accuracy": 0.13164832666516305,
      "num_tokens": 9095963.0,
      "step": 4645
    },
    {
      "entropy": 6.420813703536988,
      "epoch": 0.4648373069425701,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004984941510517899,
      "loss": 6.2562,
      "mean_token_accuracy": 0.1416421115398407,
      "num_tokens": 9105819.0,
      "step": 4650
    },
    {
      "entropy": 6.399018716812134,
      "epoch": 0.46533713200379867,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004984900261054319,
      "loss": 6.2664,
      "mean_token_accuracy": 0.14381264969706536,
      "num_tokens": 9114574.0,
      "step": 4655
    },
    {
      "entropy": 6.508393812179565,
      "epoch": 0.46583695706502726,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004984858955361224,
      "loss": 6.4682,
      "mean_token_accuracy": 0.12810177654027938,
      "num_tokens": 9125490.0,
      "step": 4660
    },
    {
      "entropy": 6.613577079772949,
      "epoch": 0.4663367821262558,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004984817593439651,
      "loss": 6.4272,
      "mean_token_accuracy": 0.13605340495705603,
      "num_tokens": 9134886.0,
      "step": 4665
    },
    {
      "entropy": 6.437576580047607,
      "epoch": 0.4668366071874844,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004984776175290643,
      "loss": 6.3636,
      "mean_token_accuracy": 0.1394120089709759,
      "num_tokens": 9143598.0,
      "step": 4670
    },
    {
      "entropy": 6.389187288284302,
      "epoch": 0.46733643224871296,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004984734700915242,
      "loss": 6.5052,
      "mean_token_accuracy": 0.12344528511166572,
      "num_tokens": 9153051.0,
      "step": 4675
    },
    {
      "entropy": 6.58816876411438,
      "epoch": 0.4678362573099415,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000498469317031449,
      "loss": 6.3858,
      "mean_token_accuracy": 0.13521411269903183,
      "num_tokens": 9162472.0,
      "step": 4680
    },
    {
      "entropy": 6.503498077392578,
      "epoch": 0.4683360823711701,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004984651583489434,
      "loss": 6.4418,
      "mean_token_accuracy": 0.13402297496795654,
      "num_tokens": 9171453.0,
      "step": 4685
    },
    {
      "entropy": 6.438160753250122,
      "epoch": 0.46883590743239867,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984609940441119,
      "loss": 6.3562,
      "mean_token_accuracy": 0.13773744702339172,
      "num_tokens": 9181104.0,
      "step": 4690
    },
    {
      "entropy": 6.49449200630188,
      "epoch": 0.46933573249362726,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004984568241170593,
      "loss": 6.4403,
      "mean_token_accuracy": 0.14121029451489447,
      "num_tokens": 9190403.0,
      "step": 4695
    },
    {
      "entropy": 6.460302019119263,
      "epoch": 0.4698355575548558,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004984526485678905,
      "loss": 6.3077,
      "mean_token_accuracy": 0.14223534017801284,
      "num_tokens": 9199132.0,
      "step": 4700
    },
    {
      "entropy": 6.528104639053344,
      "epoch": 0.4703353826160844,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004984484673967105,
      "loss": 6.4731,
      "mean_token_accuracy": 0.12893762961030006,
      "num_tokens": 9208274.0,
      "step": 4705
    },
    {
      "entropy": 6.529750728607178,
      "epoch": 0.47083520767731296,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004984442806036246,
      "loss": 6.4343,
      "mean_token_accuracy": 0.1426217757165432,
      "num_tokens": 9218513.0,
      "step": 4710
    },
    {
      "entropy": 6.514943647384643,
      "epoch": 0.4713350327385415,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004984400881887381,
      "loss": 6.4473,
      "mean_token_accuracy": 0.12802284434437752,
      "num_tokens": 9229197.0,
      "step": 4715
    },
    {
      "entropy": 6.48367395401001,
      "epoch": 0.4718348577997701,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004984358901521565,
      "loss": 6.31,
      "mean_token_accuracy": 0.1381663367152214,
      "num_tokens": 9239465.0,
      "step": 4720
    },
    {
      "entropy": 6.483723926544189,
      "epoch": 0.47233468286099867,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004984316864939853,
      "loss": 6.4142,
      "mean_token_accuracy": 0.1365165390074253,
      "num_tokens": 9250112.0,
      "step": 4725
    },
    {
      "entropy": 6.5218719959259035,
      "epoch": 0.4728345079222272,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004984274772143304,
      "loss": 6.4217,
      "mean_token_accuracy": 0.13553764224052428,
      "num_tokens": 9259406.0,
      "step": 4730
    },
    {
      "entropy": 6.527883625030517,
      "epoch": 0.4733343329834558,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004984232623132977,
      "loss": 6.3466,
      "mean_token_accuracy": 0.1345366470515728,
      "num_tokens": 9269462.0,
      "step": 4735
    },
    {
      "entropy": 6.471977996826172,
      "epoch": 0.4738341580446844,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004984190417909932,
      "loss": 6.4256,
      "mean_token_accuracy": 0.13124653920531273,
      "num_tokens": 9280291.0,
      "step": 4740
    },
    {
      "entropy": 6.5116735935211185,
      "epoch": 0.4743339831059129,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004984148156475231,
      "loss": 6.3778,
      "mean_token_accuracy": 0.12980158179998397,
      "num_tokens": 9289680.0,
      "step": 4745
    },
    {
      "entropy": 6.504691743850708,
      "epoch": 0.4748338081671415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984105838829937,
      "loss": 6.3949,
      "mean_token_accuracy": 0.13800340965390206,
      "num_tokens": 9301043.0,
      "step": 4750
    },
    {
      "entropy": 6.359691429138183,
      "epoch": 0.4753336332283701,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004984063464975116,
      "loss": 6.4217,
      "mean_token_accuracy": 0.1337859511375427,
      "num_tokens": 9311851.0,
      "step": 4755
    },
    {
      "entropy": 6.592877006530761,
      "epoch": 0.47583345828959867,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004984021034911833,
      "loss": 6.3696,
      "mean_token_accuracy": 0.13857403621077538,
      "num_tokens": 9320886.0,
      "step": 4760
    },
    {
      "entropy": 6.487786293029785,
      "epoch": 0.4763332833508272,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983978548641154,
      "loss": 6.4401,
      "mean_token_accuracy": 0.1338131070137024,
      "num_tokens": 9330518.0,
      "step": 4765
    },
    {
      "entropy": 6.443555736541748,
      "epoch": 0.4768331084120558,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004983936006164151,
      "loss": 6.2931,
      "mean_token_accuracy": 0.13514435514807702,
      "num_tokens": 9339703.0,
      "step": 4770
    },
    {
      "entropy": 6.66024603843689,
      "epoch": 0.47733293347328437,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983893407481892,
      "loss": 6.5267,
      "mean_token_accuracy": 0.1282417118549347,
      "num_tokens": 9349931.0,
      "step": 4775
    },
    {
      "entropy": 6.463759994506836,
      "epoch": 0.4778327585345129,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004983850752595451,
      "loss": 6.3831,
      "mean_token_accuracy": 0.13968004956841468,
      "num_tokens": 9358984.0,
      "step": 4780
    },
    {
      "entropy": 6.527974271774292,
      "epoch": 0.4783325835957415,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004983808041505898,
      "loss": 6.3281,
      "mean_token_accuracy": 0.13808016926050187,
      "num_tokens": 9368081.0,
      "step": 4785
    },
    {
      "entropy": 6.543414449691772,
      "epoch": 0.4788324086569701,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004983765274214311,
      "loss": 6.4715,
      "mean_token_accuracy": 0.12913113012909888,
      "num_tokens": 9377346.0,
      "step": 4790
    },
    {
      "entropy": 6.38665771484375,
      "epoch": 0.4793322337181986,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983722450721764,
      "loss": 6.4229,
      "mean_token_accuracy": 0.13916267231106758,
      "num_tokens": 9387493.0,
      "step": 4795
    },
    {
      "entropy": 6.544322204589844,
      "epoch": 0.4798320587794272,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004983679571029336,
      "loss": 6.382,
      "mean_token_accuracy": 0.13306548595428466,
      "num_tokens": 9397154.0,
      "step": 4800
    },
    {
      "entropy": 6.542267990112305,
      "epoch": 0.4803318838406558,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983636635138103,
      "loss": 6.5058,
      "mean_token_accuracy": 0.13155010417103769,
      "num_tokens": 9406705.0,
      "step": 4805
    },
    {
      "entropy": 6.495486164093018,
      "epoch": 0.4808317089018843,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983593643049148,
      "loss": 6.355,
      "mean_token_accuracy": 0.14006763696670532,
      "num_tokens": 9416850.0,
      "step": 4810
    },
    {
      "entropy": 6.415793943405151,
      "epoch": 0.4813315339631129,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004983550594763552,
      "loss": 6.3267,
      "mean_token_accuracy": 0.13997766971588135,
      "num_tokens": 9425841.0,
      "step": 4815
    },
    {
      "entropy": 6.418641853332519,
      "epoch": 0.4818313590243415,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004983507490282397,
      "loss": 6.2565,
      "mean_token_accuracy": 0.1414778672158718,
      "num_tokens": 9436748.0,
      "step": 4820
    },
    {
      "entropy": 6.478595304489136,
      "epoch": 0.4823311840855701,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004983464329606768,
      "loss": 6.4341,
      "mean_token_accuracy": 0.1369209259748459,
      "num_tokens": 9446043.0,
      "step": 4825
    },
    {
      "entropy": 6.4545375347137455,
      "epoch": 0.4828310091467986,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004983421112737752,
      "loss": 6.4416,
      "mean_token_accuracy": 0.132239031791687,
      "num_tokens": 9455480.0,
      "step": 4830
    },
    {
      "entropy": 6.503011035919189,
      "epoch": 0.4833308342080272,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983377839676435,
      "loss": 6.4336,
      "mean_token_accuracy": 0.13139901012182237,
      "num_tokens": 9464412.0,
      "step": 4835
    },
    {
      "entropy": 6.522000694274903,
      "epoch": 0.4838306592692558,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983334510423906,
      "loss": 6.3695,
      "mean_token_accuracy": 0.13747578635811805,
      "num_tokens": 9475639.0,
      "step": 4840
    },
    {
      "entropy": 6.4971763610839846,
      "epoch": 0.4843304843304843,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004983291124981257,
      "loss": 6.3307,
      "mean_token_accuracy": 0.1367343321442604,
      "num_tokens": 9484721.0,
      "step": 4845
    },
    {
      "entropy": 6.388935947418213,
      "epoch": 0.4848303093917129,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004983247683349576,
      "loss": 6.2831,
      "mean_token_accuracy": 0.14191571101546288,
      "num_tokens": 9494594.0,
      "step": 4850
    },
    {
      "entropy": 6.374627542495728,
      "epoch": 0.4853301344529415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004983204185529959,
      "loss": 6.3107,
      "mean_token_accuracy": 0.14378543868660926,
      "num_tokens": 9503707.0,
      "step": 4855
    },
    {
      "entropy": 6.455869340896607,
      "epoch": 0.48582995951417,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004983160631523499,
      "loss": 6.2809,
      "mean_token_accuracy": 0.13805698603391647,
      "num_tokens": 9512473.0,
      "step": 4860
    },
    {
      "entropy": 6.418364858627319,
      "epoch": 0.4863297845753986,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004983117021331293,
      "loss": 6.4083,
      "mean_token_accuracy": 0.13700531646609307,
      "num_tokens": 9523253.0,
      "step": 4865
    },
    {
      "entropy": 6.521939849853515,
      "epoch": 0.4868296096366272,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004983073354954436,
      "loss": 6.3777,
      "mean_token_accuracy": 0.13949814885854722,
      "num_tokens": 9531938.0,
      "step": 4870
    },
    {
      "entropy": 6.59939751625061,
      "epoch": 0.4873294346978557,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004983029632394029,
      "loss": 6.3754,
      "mean_token_accuracy": 0.13508883193135263,
      "num_tokens": 9541542.0,
      "step": 4875
    },
    {
      "entropy": 6.371125173568726,
      "epoch": 0.4878292597590843,
      "grad_norm": 1.03125,
      "learning_rate": 0.000498298585365117,
      "loss": 6.3222,
      "mean_token_accuracy": 0.14197706952691078,
      "num_tokens": 9551521.0,
      "step": 4880
    },
    {
      "entropy": 6.467463541030884,
      "epoch": 0.4883290848203129,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004982942018726963,
      "loss": 6.443,
      "mean_token_accuracy": 0.141734978556633,
      "num_tokens": 9561605.0,
      "step": 4885
    },
    {
      "entropy": 6.471908760070801,
      "epoch": 0.4888289098815415,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004982898127622508,
      "loss": 6.3501,
      "mean_token_accuracy": 0.13869891613721846,
      "num_tokens": 9571591.0,
      "step": 4890
    },
    {
      "entropy": 6.385383081436157,
      "epoch": 0.48932873494277,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004982854180338911,
      "loss": 6.362,
      "mean_token_accuracy": 0.1401084266602993,
      "num_tokens": 9580994.0,
      "step": 4895
    },
    {
      "entropy": 6.559015560150146,
      "epoch": 0.4898285600039986,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004982810176877279,
      "loss": 6.3444,
      "mean_token_accuracy": 0.1354878470301628,
      "num_tokens": 9589268.0,
      "step": 4900
    },
    {
      "entropy": 6.46404128074646,
      "epoch": 0.4903283850652272,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982766117238716,
      "loss": 6.4162,
      "mean_token_accuracy": 0.13586771190166474,
      "num_tokens": 9599461.0,
      "step": 4905
    },
    {
      "entropy": 6.503689765930176,
      "epoch": 0.4908282101264557,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004982722001424331,
      "loss": 6.3698,
      "mean_token_accuracy": 0.134616681933403,
      "num_tokens": 9609629.0,
      "step": 4910
    },
    {
      "entropy": 6.419895076751709,
      "epoch": 0.4913280351876843,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004982677829435236,
      "loss": 6.2921,
      "mean_token_accuracy": 0.1450149841606617,
      "num_tokens": 9619618.0,
      "step": 4915
    },
    {
      "entropy": 6.437411975860596,
      "epoch": 0.4918278602489129,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004982633601272543,
      "loss": 6.355,
      "mean_token_accuracy": 0.13925076201558112,
      "num_tokens": 9629042.0,
      "step": 4920
    },
    {
      "entropy": 6.463222169876099,
      "epoch": 0.4923276853101414,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004982589316937361,
      "loss": 6.3413,
      "mean_token_accuracy": 0.14059789702296258,
      "num_tokens": 9638596.0,
      "step": 4925
    },
    {
      "entropy": 6.488814878463745,
      "epoch": 0.49282751037137,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004982544976430808,
      "loss": 6.4204,
      "mean_token_accuracy": 0.13408228531479835,
      "num_tokens": 9648505.0,
      "step": 4930
    },
    {
      "entropy": 6.38788857460022,
      "epoch": 0.4933273354325986,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004982500579753995,
      "loss": 6.3899,
      "mean_token_accuracy": 0.1382318913936615,
      "num_tokens": 9658113.0,
      "step": 4935
    },
    {
      "entropy": 6.476542186737061,
      "epoch": 0.49382716049382713,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004982456126908044,
      "loss": 6.407,
      "mean_token_accuracy": 0.13840780779719353,
      "num_tokens": 9668075.0,
      "step": 4940
    },
    {
      "entropy": 6.5929343700408936,
      "epoch": 0.4943269855550557,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004982411617894071,
      "loss": 6.3737,
      "mean_token_accuracy": 0.13169767037034036,
      "num_tokens": 9678139.0,
      "step": 4945
    },
    {
      "entropy": 6.444349527359009,
      "epoch": 0.4948268106162843,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004982367052713196,
      "loss": 6.4219,
      "mean_token_accuracy": 0.12752186879515648,
      "num_tokens": 9687328.0,
      "step": 4950
    },
    {
      "entropy": 6.430422306060791,
      "epoch": 0.4953266356775129,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982322431366541,
      "loss": 6.3659,
      "mean_token_accuracy": 0.1412187784910202,
      "num_tokens": 9697536.0,
      "step": 4955
    },
    {
      "entropy": 6.502705144882202,
      "epoch": 0.4958264607387414,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004982277753855227,
      "loss": 6.4243,
      "mean_token_accuracy": 0.1386316880583763,
      "num_tokens": 9706804.0,
      "step": 4960
    },
    {
      "entropy": 6.534018230438233,
      "epoch": 0.49632628579997,
      "grad_norm": 1.109375,
      "learning_rate": 0.000498223302018038,
      "loss": 6.3119,
      "mean_token_accuracy": 0.14683042243123054,
      "num_tokens": 9716868.0,
      "step": 4965
    },
    {
      "entropy": 6.403007745742798,
      "epoch": 0.4968261108611986,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004982188230343124,
      "loss": 6.4068,
      "mean_token_accuracy": 0.13918600156903266,
      "num_tokens": 9728790.0,
      "step": 4970
    },
    {
      "entropy": 6.54256911277771,
      "epoch": 0.49732593592242713,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004982143384344587,
      "loss": 6.4124,
      "mean_token_accuracy": 0.13571546748280525,
      "num_tokens": 9739118.0,
      "step": 4975
    },
    {
      "entropy": 6.522776079177857,
      "epoch": 0.4978257609836557,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004982098482185897,
      "loss": 6.38,
      "mean_token_accuracy": 0.1403428755700588,
      "num_tokens": 9749347.0,
      "step": 4980
    },
    {
      "entropy": 6.51669135093689,
      "epoch": 0.4983255860448843,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004982053523868182,
      "loss": 6.4333,
      "mean_token_accuracy": 0.13202799037098883,
      "num_tokens": 9759432.0,
      "step": 4985
    },
    {
      "entropy": 6.41639199256897,
      "epoch": 0.49882541110611284,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004982008509392576,
      "loss": 6.3441,
      "mean_token_accuracy": 0.13148636296391486,
      "num_tokens": 9768666.0,
      "step": 4990
    },
    {
      "entropy": 6.398564863204956,
      "epoch": 0.4993252361673414,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000498196343876021,
      "loss": 6.2805,
      "mean_token_accuracy": 0.14879042357206346,
      "num_tokens": 9778699.0,
      "step": 4995
    },
    {
      "entropy": 6.455457496643066,
      "epoch": 0.49982506122857,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981918311972219,
      "loss": 6.3361,
      "mean_token_accuracy": 0.13913736939430238,
      "num_tokens": 9788156.0,
      "step": 5000
    },
    {
      "entropy": 6.448054504394531,
      "epoch": 0.5003248862897985,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004981873129029737,
      "loss": 6.2712,
      "mean_token_accuracy": 0.14729730635881425,
      "num_tokens": 9797293.0,
      "step": 5005
    },
    {
      "entropy": 6.3916370391845705,
      "epoch": 0.5008247113510271,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049818278899339,
      "loss": 6.2946,
      "mean_token_accuracy": 0.14214758723974227,
      "num_tokens": 9806962.0,
      "step": 5010
    },
    {
      "entropy": 6.443386697769165,
      "epoch": 0.5013245364122557,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000498178259468585,
      "loss": 6.4267,
      "mean_token_accuracy": 0.13721726089715958,
      "num_tokens": 9816689.0,
      "step": 5015
    },
    {
      "entropy": 6.508077478408813,
      "epoch": 0.5018243614734843,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004981737243286722,
      "loss": 6.3153,
      "mean_token_accuracy": 0.1371029019355774,
      "num_tokens": 9826321.0,
      "step": 5020
    },
    {
      "entropy": 6.540329885482788,
      "epoch": 0.5023241865347129,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004981691835737661,
      "loss": 6.4547,
      "mean_token_accuracy": 0.12951614782214166,
      "num_tokens": 9836054.0,
      "step": 5025
    },
    {
      "entropy": 6.554930019378662,
      "epoch": 0.5028240115959414,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004981646372039807,
      "loss": 6.4394,
      "mean_token_accuracy": 0.13512118458747863,
      "num_tokens": 9846516.0,
      "step": 5030
    },
    {
      "entropy": 6.424113798141479,
      "epoch": 0.50332383665717,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004981600852194305,
      "loss": 6.227,
      "mean_token_accuracy": 0.14665252566337586,
      "num_tokens": 9856599.0,
      "step": 5035
    },
    {
      "entropy": 6.4051251888275145,
      "epoch": 0.5038236617183985,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049815552762023,
      "loss": 6.3299,
      "mean_token_accuracy": 0.1417568065226078,
      "num_tokens": 9866922.0,
      "step": 5040
    },
    {
      "entropy": 6.43504490852356,
      "epoch": 0.5043234867796271,
      "grad_norm": 1.125,
      "learning_rate": 0.0004981509644064937,
      "loss": 6.3299,
      "mean_token_accuracy": 0.13958919197320938,
      "num_tokens": 9876174.0,
      "step": 5045
    },
    {
      "entropy": 6.415535640716553,
      "epoch": 0.5048233118408557,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004981463955783367,
      "loss": 6.277,
      "mean_token_accuracy": 0.14316560924053193,
      "num_tokens": 9886208.0,
      "step": 5050
    },
    {
      "entropy": 6.410051488876343,
      "epoch": 0.5053231369020843,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004981418211358736,
      "loss": 6.3386,
      "mean_token_accuracy": 0.1408532105386257,
      "num_tokens": 9896522.0,
      "step": 5055
    },
    {
      "entropy": 6.4151115894317625,
      "epoch": 0.5058229619633129,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004981372410792199,
      "loss": 6.3417,
      "mean_token_accuracy": 0.14682593569159508,
      "num_tokens": 9905753.0,
      "step": 5060
    },
    {
      "entropy": 6.4666502475738525,
      "epoch": 0.5063227870245414,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004981326554084906,
      "loss": 6.4301,
      "mean_token_accuracy": 0.13811509534716607,
      "num_tokens": 9916566.0,
      "step": 5065
    },
    {
      "entropy": 6.468195056915283,
      "epoch": 0.50682261208577,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004981280641238011,
      "loss": 6.415,
      "mean_token_accuracy": 0.1303804561495781,
      "num_tokens": 9926126.0,
      "step": 5070
    },
    {
      "entropy": 6.513017320632935,
      "epoch": 0.5073224371469985,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004981234672252669,
      "loss": 6.417,
      "mean_token_accuracy": 0.13405416831374167,
      "num_tokens": 9937666.0,
      "step": 5075
    },
    {
      "entropy": 6.552185869216919,
      "epoch": 0.5078222622082271,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004981188647130037,
      "loss": 6.4457,
      "mean_token_accuracy": 0.13834275752305986,
      "num_tokens": 9947107.0,
      "step": 5080
    },
    {
      "entropy": 6.480519962310791,
      "epoch": 0.5083220872694557,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004981142565871272,
      "loss": 6.3068,
      "mean_token_accuracy": 0.1375648148357868,
      "num_tokens": 9956484.0,
      "step": 5085
    },
    {
      "entropy": 6.3329003810882565,
      "epoch": 0.5088219123306843,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981096428477535,
      "loss": 6.3494,
      "mean_token_accuracy": 0.13904502466320992,
      "num_tokens": 9966731.0,
      "step": 5090
    },
    {
      "entropy": 6.461158180236817,
      "epoch": 0.5093217373919128,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004981050234949986,
      "loss": 6.3298,
      "mean_token_accuracy": 0.141248919069767,
      "num_tokens": 9976491.0,
      "step": 5095
    },
    {
      "entropy": 6.507654333114624,
      "epoch": 0.5098215624531414,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004981003985289786,
      "loss": 6.2748,
      "mean_token_accuracy": 0.1455710768699646,
      "num_tokens": 9985248.0,
      "step": 5100
    },
    {
      "entropy": 6.403945112228394,
      "epoch": 0.51032138751437,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004980957679498101,
      "loss": 6.3729,
      "mean_token_accuracy": 0.14013429135084152,
      "num_tokens": 9995578.0,
      "step": 5105
    },
    {
      "entropy": 6.518911504745484,
      "epoch": 0.5108212125755985,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004980911317576095,
      "loss": 6.4478,
      "mean_token_accuracy": 0.131076168268919,
      "num_tokens": 10006078.0,
      "step": 5110
    },
    {
      "entropy": 6.529238319396972,
      "epoch": 0.5113210376368271,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004980864899524933,
      "loss": 6.4132,
      "mean_token_accuracy": 0.12879174426198006,
      "num_tokens": 10015943.0,
      "step": 5115
    },
    {
      "entropy": 6.472807931900024,
      "epoch": 0.5118208626980557,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004980818425345786,
      "loss": 6.4037,
      "mean_token_accuracy": 0.13248924538493156,
      "num_tokens": 10025738.0,
      "step": 5120
    },
    {
      "entropy": 6.562513828277588,
      "epoch": 0.5123206877592843,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000498077189503982,
      "loss": 6.3645,
      "mean_token_accuracy": 0.13115599527955055,
      "num_tokens": 10035032.0,
      "step": 5125
    },
    {
      "entropy": 6.44153413772583,
      "epoch": 0.5128205128205128,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980725308608208,
      "loss": 6.3326,
      "mean_token_accuracy": 0.14097934886813163,
      "num_tokens": 10044201.0,
      "step": 5130
    },
    {
      "entropy": 6.356708717346192,
      "epoch": 0.5133203378817414,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004980678666052121,
      "loss": 6.3701,
      "mean_token_accuracy": 0.1402401879429817,
      "num_tokens": 10054163.0,
      "step": 5135
    },
    {
      "entropy": 6.438456249237061,
      "epoch": 0.51382016294297,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004980631967372733,
      "loss": 6.3646,
      "mean_token_accuracy": 0.14144863486289977,
      "num_tokens": 10065225.0,
      "step": 5140
    },
    {
      "entropy": 6.506520509719849,
      "epoch": 0.5143199880041985,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004980585212571218,
      "loss": 6.3526,
      "mean_token_accuracy": 0.13947251960635185,
      "num_tokens": 10074776.0,
      "step": 5145
    },
    {
      "entropy": 6.506457757949829,
      "epoch": 0.5148198130654271,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980538401648753,
      "loss": 6.4989,
      "mean_token_accuracy": 0.1341721385717392,
      "num_tokens": 10085917.0,
      "step": 5150
    },
    {
      "entropy": 6.491835975646973,
      "epoch": 0.5153196381266557,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004980491534606517,
      "loss": 6.3067,
      "mean_token_accuracy": 0.14042604863643646,
      "num_tokens": 10094727.0,
      "step": 5155
    },
    {
      "entropy": 6.414020919799805,
      "epoch": 0.5158194631878842,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004980444611445686,
      "loss": 6.2994,
      "mean_token_accuracy": 0.14486273229122162,
      "num_tokens": 10104839.0,
      "step": 5160
    },
    {
      "entropy": 6.405583190917969,
      "epoch": 0.5163192882491128,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004980397632167444,
      "loss": 6.283,
      "mean_token_accuracy": 0.14061587750911714,
      "num_tokens": 10114501.0,
      "step": 5165
    },
    {
      "entropy": 6.421072101593017,
      "epoch": 0.5168191133103414,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004980350596772971,
      "loss": 6.3908,
      "mean_token_accuracy": 0.1360914647579193,
      "num_tokens": 10124597.0,
      "step": 5170
    },
    {
      "entropy": 6.322241735458374,
      "epoch": 0.51731893837157,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004980303505263451,
      "loss": 6.267,
      "mean_token_accuracy": 0.1398468352854252,
      "num_tokens": 10135695.0,
      "step": 5175
    },
    {
      "entropy": 6.337439060211182,
      "epoch": 0.5178187634327985,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004980256357640069,
      "loss": 6.33,
      "mean_token_accuracy": 0.1377136953175068,
      "num_tokens": 10146107.0,
      "step": 5180
    },
    {
      "entropy": 6.509029197692871,
      "epoch": 0.5183185884940271,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004980209153904011,
      "loss": 6.3958,
      "mean_token_accuracy": 0.14012961685657502,
      "num_tokens": 10155983.0,
      "step": 5185
    },
    {
      "entropy": 6.484198379516601,
      "epoch": 0.5188184135552557,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004980161894056464,
      "loss": 6.3634,
      "mean_token_accuracy": 0.13489748686552047,
      "num_tokens": 10166183.0,
      "step": 5190
    },
    {
      "entropy": 6.425005865097046,
      "epoch": 0.5193182386164842,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004980114578098618,
      "loss": 6.2387,
      "mean_token_accuracy": 0.14645392522215844,
      "num_tokens": 10175524.0,
      "step": 5195
    },
    {
      "entropy": 6.425123023986816,
      "epoch": 0.5198180636777128,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004980067206031663,
      "loss": 6.4059,
      "mean_token_accuracy": 0.1262257769703865,
      "num_tokens": 10186235.0,
      "step": 5200
    },
    {
      "entropy": 6.554209804534912,
      "epoch": 0.5203178887389414,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004980019777856791,
      "loss": 6.4135,
      "mean_token_accuracy": 0.13365318030118942,
      "num_tokens": 10196166.0,
      "step": 5205
    },
    {
      "entropy": 6.4457080364227295,
      "epoch": 0.52081771380017,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979972293575195,
      "loss": 6.3336,
      "mean_token_accuracy": 0.1418666884303093,
      "num_tokens": 10205691.0,
      "step": 5210
    },
    {
      "entropy": 6.373795413970948,
      "epoch": 0.5213175388613985,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979924753188069,
      "loss": 6.2839,
      "mean_token_accuracy": 0.14232254922389984,
      "num_tokens": 10216224.0,
      "step": 5215
    },
    {
      "entropy": 6.42751636505127,
      "epoch": 0.5218173639226271,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004979877156696612,
      "loss": 6.3408,
      "mean_token_accuracy": 0.13963486701250077,
      "num_tokens": 10226468.0,
      "step": 5220
    },
    {
      "entropy": 6.572753286361694,
      "epoch": 0.5223171889838556,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004979829504102018,
      "loss": 6.3602,
      "mean_token_accuracy": 0.13437848538160324,
      "num_tokens": 10235932.0,
      "step": 5225
    },
    {
      "entropy": 6.501198482513428,
      "epoch": 0.5228170140450842,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004979781795405488,
      "loss": 6.4778,
      "mean_token_accuracy": 0.13081177026033403,
      "num_tokens": 10246052.0,
      "step": 5230
    },
    {
      "entropy": 6.514893627166748,
      "epoch": 0.5233168391063128,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004979734030608222,
      "loss": 6.3769,
      "mean_token_accuracy": 0.13877685070037843,
      "num_tokens": 10255056.0,
      "step": 5235
    },
    {
      "entropy": 6.41351842880249,
      "epoch": 0.5238166641675414,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004979686209711422,
      "loss": 6.3236,
      "mean_token_accuracy": 0.14136184826493264,
      "num_tokens": 10265403.0,
      "step": 5240
    },
    {
      "entropy": 6.4000389099121096,
      "epoch": 0.5243164892287699,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497963833271629,
      "loss": 6.3106,
      "mean_token_accuracy": 0.14661918953061104,
      "num_tokens": 10275148.0,
      "step": 5245
    },
    {
      "entropy": 6.497857189178466,
      "epoch": 0.5248163142899985,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004979590399624031,
      "loss": 6.3258,
      "mean_token_accuracy": 0.1367676131427288,
      "num_tokens": 10285083.0,
      "step": 5250
    },
    {
      "entropy": 6.351488733291626,
      "epoch": 0.5253161393512271,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004979542410435852,
      "loss": 6.2959,
      "mean_token_accuracy": 0.1327349342405796,
      "num_tokens": 10294466.0,
      "step": 5255
    },
    {
      "entropy": 6.473018169403076,
      "epoch": 0.5258159644124556,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004979494365152959,
      "loss": 6.4041,
      "mean_token_accuracy": 0.13443218246102334,
      "num_tokens": 10304333.0,
      "step": 5260
    },
    {
      "entropy": 6.546953201293945,
      "epoch": 0.5263157894736842,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004979446263776562,
      "loss": 6.3379,
      "mean_token_accuracy": 0.13749662265181542,
      "num_tokens": 10314880.0,
      "step": 5265
    },
    {
      "entropy": 6.468399381637573,
      "epoch": 0.5268156145349128,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000497939810630787,
      "loss": 6.3068,
      "mean_token_accuracy": 0.13549398854374886,
      "num_tokens": 10324060.0,
      "step": 5270
    },
    {
      "entropy": 6.399018430709839,
      "epoch": 0.5273154395961414,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004979349892748096,
      "loss": 6.4183,
      "mean_token_accuracy": 0.13540468364953995,
      "num_tokens": 10335248.0,
      "step": 5275
    },
    {
      "entropy": 6.429918575286865,
      "epoch": 0.5278152646573699,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004979301623098452,
      "loss": 6.2308,
      "mean_token_accuracy": 0.14516004547476768,
      "num_tokens": 10343974.0,
      "step": 5280
    },
    {
      "entropy": 6.467012786865235,
      "epoch": 0.5283150897185985,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004979253297360153,
      "loss": 6.3335,
      "mean_token_accuracy": 0.132409206032753,
      "num_tokens": 10354037.0,
      "step": 5285
    },
    {
      "entropy": 6.463950347900391,
      "epoch": 0.528814914779827,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979204915534415,
      "loss": 6.3708,
      "mean_token_accuracy": 0.1383878469467163,
      "num_tokens": 10364714.0,
      "step": 5290
    },
    {
      "entropy": 6.403480291366577,
      "epoch": 0.5293147398410556,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004979156477622454,
      "loss": 6.3683,
      "mean_token_accuracy": 0.13828956931829453,
      "num_tokens": 10373972.0,
      "step": 5295
    },
    {
      "entropy": 6.431853246688843,
      "epoch": 0.5298145649022842,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004979107983625491,
      "loss": 6.2434,
      "mean_token_accuracy": 0.14456052631139754,
      "num_tokens": 10382882.0,
      "step": 5300
    },
    {
      "entropy": 6.368200063705444,
      "epoch": 0.5303143899635128,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004979059433544744,
      "loss": 6.3292,
      "mean_token_accuracy": 0.14171645417809486,
      "num_tokens": 10392257.0,
      "step": 5305
    },
    {
      "entropy": 6.441417694091797,
      "epoch": 0.5308142150247414,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004979010827381435,
      "loss": 6.3507,
      "mean_token_accuracy": 0.13942453265190125,
      "num_tokens": 10401055.0,
      "step": 5310
    },
    {
      "entropy": 6.545503997802735,
      "epoch": 0.5313140400859699,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004978962165136787,
      "loss": 6.4602,
      "mean_token_accuracy": 0.12886520549654962,
      "num_tokens": 10411443.0,
      "step": 5315
    },
    {
      "entropy": 6.514301300048828,
      "epoch": 0.5318138651471985,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004978913446812025,
      "loss": 6.4339,
      "mean_token_accuracy": 0.13929549753665924,
      "num_tokens": 10422142.0,
      "step": 5320
    },
    {
      "entropy": 6.498802709579468,
      "epoch": 0.532313690208427,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004978864672408374,
      "loss": 6.3601,
      "mean_token_accuracy": 0.1346695177257061,
      "num_tokens": 10432454.0,
      "step": 5325
    },
    {
      "entropy": 6.387129497528076,
      "epoch": 0.5328135152696556,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004978815841927061,
      "loss": 6.3211,
      "mean_token_accuracy": 0.14032170325517654,
      "num_tokens": 10443625.0,
      "step": 5330
    },
    {
      "entropy": 6.516609001159668,
      "epoch": 0.5333133403308842,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004978766955369315,
      "loss": 6.3943,
      "mean_token_accuracy": 0.14243654310703277,
      "num_tokens": 10454055.0,
      "step": 5335
    },
    {
      "entropy": 6.39882082939148,
      "epoch": 0.5338131653921128,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004978718012736367,
      "loss": 6.2307,
      "mean_token_accuracy": 0.1444314181804657,
      "num_tokens": 10462683.0,
      "step": 5340
    },
    {
      "entropy": 6.375852918624878,
      "epoch": 0.5343129904533414,
      "grad_norm": 1.125,
      "learning_rate": 0.0004978669014029446,
      "loss": 6.2677,
      "mean_token_accuracy": 0.14716348573565483,
      "num_tokens": 10471699.0,
      "step": 5345
    },
    {
      "entropy": 6.343105220794678,
      "epoch": 0.5348128155145699,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004978619959249787,
      "loss": 6.1928,
      "mean_token_accuracy": 0.14977386742830276,
      "num_tokens": 10480790.0,
      "step": 5350
    },
    {
      "entropy": 6.384857606887818,
      "epoch": 0.5353126405757984,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004978570848398623,
      "loss": 6.3182,
      "mean_token_accuracy": 0.1346427671611309,
      "num_tokens": 10490202.0,
      "step": 5355
    },
    {
      "entropy": 6.409596252441406,
      "epoch": 0.535812465637027,
      "grad_norm": 1.078125,
      "learning_rate": 0.000497852168147719,
      "loss": 6.3715,
      "mean_token_accuracy": 0.135125882178545,
      "num_tokens": 10499690.0,
      "step": 5360
    },
    {
      "entropy": 6.506616926193237,
      "epoch": 0.5363122906982556,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004978472458486724,
      "loss": 6.3409,
      "mean_token_accuracy": 0.13681497648358346,
      "num_tokens": 10510120.0,
      "step": 5365
    },
    {
      "entropy": 6.451282787322998,
      "epoch": 0.5368121157594842,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004978423179428466,
      "loss": 6.4563,
      "mean_token_accuracy": 0.1353305995464325,
      "num_tokens": 10520521.0,
      "step": 5370
    },
    {
      "entropy": 6.345026111602783,
      "epoch": 0.5373119408207128,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004978373844303653,
      "loss": 6.1488,
      "mean_token_accuracy": 0.146633118391037,
      "num_tokens": 10530233.0,
      "step": 5375
    },
    {
      "entropy": 6.39206018447876,
      "epoch": 0.5378117658819414,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004978324453113528,
      "loss": 6.3042,
      "mean_token_accuracy": 0.1419048637151718,
      "num_tokens": 10539683.0,
      "step": 5380
    },
    {
      "entropy": 6.433778715133667,
      "epoch": 0.5383115909431699,
      "grad_norm": 1.125,
      "learning_rate": 0.0004978275005859333,
      "loss": 6.3174,
      "mean_token_accuracy": 0.14073469638824462,
      "num_tokens": 10549483.0,
      "step": 5385
    },
    {
      "entropy": 6.4962962627410885,
      "epoch": 0.5388114160043984,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004978225502542313,
      "loss": 6.4297,
      "mean_token_accuracy": 0.1314571052789688,
      "num_tokens": 10560464.0,
      "step": 5390
    },
    {
      "entropy": 6.474843263626099,
      "epoch": 0.539311241065627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004978175943163712,
      "loss": 6.3327,
      "mean_token_accuracy": 0.13529158234596253,
      "num_tokens": 10568660.0,
      "step": 5395
    },
    {
      "entropy": 6.471502637863159,
      "epoch": 0.5398110661268556,
      "grad_norm": 1.125,
      "learning_rate": 0.0004978126327724777,
      "loss": 6.3692,
      "mean_token_accuracy": 0.1399627573788166,
      "num_tokens": 10577304.0,
      "step": 5400
    },
    {
      "entropy": 6.408673906326294,
      "epoch": 0.5403108911880842,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004978076656226758,
      "loss": 6.2824,
      "mean_token_accuracy": 0.14970972910523414,
      "num_tokens": 10587591.0,
      "step": 5405
    },
    {
      "entropy": 6.422517013549805,
      "epoch": 0.5408107162493128,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004978026928670902,
      "loss": 6.2308,
      "mean_token_accuracy": 0.13947502374649048,
      "num_tokens": 10598591.0,
      "step": 5410
    },
    {
      "entropy": 6.347539710998535,
      "epoch": 0.5413105413105413,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004977977145058463,
      "loss": 6.2566,
      "mean_token_accuracy": 0.13943241089582442,
      "num_tokens": 10608842.0,
      "step": 5415
    },
    {
      "entropy": 6.373945426940918,
      "epoch": 0.5418103663717699,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004977927305390693,
      "loss": 6.3463,
      "mean_token_accuracy": 0.1439303234219551,
      "num_tokens": 10618812.0,
      "step": 5420
    },
    {
      "entropy": 6.496698093414307,
      "epoch": 0.5423101914329984,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004977877409668845,
      "loss": 6.4554,
      "mean_token_accuracy": 0.13219047337770462,
      "num_tokens": 10629244.0,
      "step": 5425
    },
    {
      "entropy": 6.5158854007720945,
      "epoch": 0.542810016494227,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004977827457894174,
      "loss": 6.3879,
      "mean_token_accuracy": 0.13357137739658356,
      "num_tokens": 10638931.0,
      "step": 5430
    },
    {
      "entropy": 6.399742889404297,
      "epoch": 0.5433098415554556,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004977777450067937,
      "loss": 6.2627,
      "mean_token_accuracy": 0.14259956628084183,
      "num_tokens": 10648754.0,
      "step": 5435
    },
    {
      "entropy": 6.37289662361145,
      "epoch": 0.5438096666166842,
      "grad_norm": 1.125,
      "learning_rate": 0.0004977727386191393,
      "loss": 6.2343,
      "mean_token_accuracy": 0.14566398561000823,
      "num_tokens": 10657716.0,
      "step": 5440
    },
    {
      "entropy": 6.3968058109283445,
      "epoch": 0.5443094916779128,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004977677266265802,
      "loss": 6.2695,
      "mean_token_accuracy": 0.14065478518605232,
      "num_tokens": 10667621.0,
      "step": 5445
    },
    {
      "entropy": 6.349292135238647,
      "epoch": 0.5448093167391413,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004977627090292424,
      "loss": 6.2465,
      "mean_token_accuracy": 0.13979329094290732,
      "num_tokens": 10676640.0,
      "step": 5450
    },
    {
      "entropy": 6.358928394317627,
      "epoch": 0.5453091418003698,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004977576858272521,
      "loss": 6.3605,
      "mean_token_accuracy": 0.13097573891282083,
      "num_tokens": 10687034.0,
      "step": 5455
    },
    {
      "entropy": 6.441195011138916,
      "epoch": 0.5458089668615984,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004977526570207358,
      "loss": 6.2382,
      "mean_token_accuracy": 0.1447528585791588,
      "num_tokens": 10696234.0,
      "step": 5460
    },
    {
      "entropy": 6.350192499160767,
      "epoch": 0.546308791922827,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004977476226098199,
      "loss": 6.2766,
      "mean_token_accuracy": 0.14773816764354705,
      "num_tokens": 10705733.0,
      "step": 5465
    },
    {
      "entropy": 6.444482421875,
      "epoch": 0.5468086169840556,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004977425825946312,
      "loss": 6.2987,
      "mean_token_accuracy": 0.14336602985858918,
      "num_tokens": 10714628.0,
      "step": 5470
    },
    {
      "entropy": 6.5299904346466064,
      "epoch": 0.5473084420452842,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004977375369752964,
      "loss": 6.4323,
      "mean_token_accuracy": 0.12882368937134742,
      "num_tokens": 10726120.0,
      "step": 5475
    },
    {
      "entropy": 6.414555168151855,
      "epoch": 0.5478082671065128,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004977324857519425,
      "loss": 6.2861,
      "mean_token_accuracy": 0.13879045769572257,
      "num_tokens": 10735976.0,
      "step": 5480
    },
    {
      "entropy": 6.464913558959961,
      "epoch": 0.5483080921677413,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004977274289246967,
      "loss": 6.377,
      "mean_token_accuracy": 0.13373840898275374,
      "num_tokens": 10745730.0,
      "step": 5485
    },
    {
      "entropy": 6.473123788833618,
      "epoch": 0.5488079172289698,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004977223664936859,
      "loss": 6.3013,
      "mean_token_accuracy": 0.1395411156117916,
      "num_tokens": 10755435.0,
      "step": 5490
    },
    {
      "entropy": 6.343825244903565,
      "epoch": 0.5493077422901984,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004977172984590378,
      "loss": 6.1954,
      "mean_token_accuracy": 0.14526784494519235,
      "num_tokens": 10765483.0,
      "step": 5495
    },
    {
      "entropy": 6.333004331588745,
      "epoch": 0.549807567351427,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004977122248208798,
      "loss": 6.2856,
      "mean_token_accuracy": 0.1436666414141655,
      "num_tokens": 10775034.0,
      "step": 5500
    },
    {
      "entropy": 6.344209480285644,
      "epoch": 0.5503073924126556,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004977071455793394,
      "loss": 6.259,
      "mean_token_accuracy": 0.13889251425862312,
      "num_tokens": 10784799.0,
      "step": 5505
    },
    {
      "entropy": 6.418334102630615,
      "epoch": 0.5508072174738842,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004977020607345445,
      "loss": 6.3333,
      "mean_token_accuracy": 0.1300968937575817,
      "num_tokens": 10794907.0,
      "step": 5510
    },
    {
      "entropy": 6.361036586761474,
      "epoch": 0.5513070425351128,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004976969702866231,
      "loss": 6.2557,
      "mean_token_accuracy": 0.1447628490626812,
      "num_tokens": 10804374.0,
      "step": 5515
    },
    {
      "entropy": 6.394205904006958,
      "epoch": 0.5518068675963412,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004976918742357033,
      "loss": 6.2754,
      "mean_token_accuracy": 0.14091094583272934,
      "num_tokens": 10814245.0,
      "step": 5520
    },
    {
      "entropy": 6.374346590042114,
      "epoch": 0.5523066926575698,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004976867725819133,
      "loss": 6.2919,
      "mean_token_accuracy": 0.14092635586857796,
      "num_tokens": 10824667.0,
      "step": 5525
    },
    {
      "entropy": 6.367292499542236,
      "epoch": 0.5528065177187984,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004976816653253813,
      "loss": 6.2534,
      "mean_token_accuracy": 0.14192148000001908,
      "num_tokens": 10834890.0,
      "step": 5530
    },
    {
      "entropy": 6.393146085739136,
      "epoch": 0.553306342780027,
      "grad_norm": 1.0,
      "learning_rate": 0.0004976765524662359,
      "loss": 6.2179,
      "mean_token_accuracy": 0.14673539698123933,
      "num_tokens": 10844751.0,
      "step": 5535
    },
    {
      "entropy": 6.462450361251831,
      "epoch": 0.5538061678412556,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004976714340046058,
      "loss": 6.3814,
      "mean_token_accuracy": 0.14202267900109292,
      "num_tokens": 10854113.0,
      "step": 5540
    },
    {
      "entropy": 6.349339580535888,
      "epoch": 0.5543059929024842,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004976663099406196,
      "loss": 6.213,
      "mean_token_accuracy": 0.14216559752821922,
      "num_tokens": 10863468.0,
      "step": 5545
    },
    {
      "entropy": 6.440054082870484,
      "epoch": 0.5548058179637128,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004976611802744066,
      "loss": 6.3256,
      "mean_token_accuracy": 0.14115004390478134,
      "num_tokens": 10873381.0,
      "step": 5550
    },
    {
      "entropy": 6.44648699760437,
      "epoch": 0.5553056430249412,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004976560450060954,
      "loss": 6.2524,
      "mean_token_accuracy": 0.14238317012786866,
      "num_tokens": 10883457.0,
      "step": 5555
    },
    {
      "entropy": 6.284746313095093,
      "epoch": 0.5558054680861698,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004976509041358154,
      "loss": 6.2178,
      "mean_token_accuracy": 0.14156111627817153,
      "num_tokens": 10892822.0,
      "step": 5560
    },
    {
      "entropy": 6.328335857391357,
      "epoch": 0.5563052931473984,
      "grad_norm": 1.140625,
      "learning_rate": 0.000497645757663696,
      "loss": 6.1757,
      "mean_token_accuracy": 0.1516851395368576,
      "num_tokens": 10901935.0,
      "step": 5565
    },
    {
      "entropy": 6.307885026931762,
      "epoch": 0.556805118208627,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004976406055898665,
      "loss": 6.2954,
      "mean_token_accuracy": 0.13936870023608208,
      "num_tokens": 10911863.0,
      "step": 5570
    },
    {
      "entropy": 6.404980182647705,
      "epoch": 0.5573049432698556,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004976354479144568,
      "loss": 6.2631,
      "mean_token_accuracy": 0.14423334300518037,
      "num_tokens": 10921196.0,
      "step": 5575
    },
    {
      "entropy": 6.397053003311157,
      "epoch": 0.5578047683310842,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004976302846375965,
      "loss": 6.2285,
      "mean_token_accuracy": 0.14099265187978743,
      "num_tokens": 10931374.0,
      "step": 5580
    },
    {
      "entropy": 6.453075885772705,
      "epoch": 0.5583045933923126,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004976251157594155,
      "loss": 6.3915,
      "mean_token_accuracy": 0.13292488008737563,
      "num_tokens": 10941002.0,
      "step": 5585
    },
    {
      "entropy": 6.4787900924682615,
      "epoch": 0.5588044184535412,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004976199412800437,
      "loss": 6.3112,
      "mean_token_accuracy": 0.1385272614657879,
      "num_tokens": 10950602.0,
      "step": 5590
    },
    {
      "entropy": 6.421299600601197,
      "epoch": 0.5593042435147698,
      "grad_norm": 1.125,
      "learning_rate": 0.0004976147611996118,
      "loss": 6.2837,
      "mean_token_accuracy": 0.14085360541939734,
      "num_tokens": 10960363.0,
      "step": 5595
    },
    {
      "entropy": 6.42499852180481,
      "epoch": 0.5598040685759984,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004976095755182496,
      "loss": 6.3683,
      "mean_token_accuracy": 0.1343800149857998,
      "num_tokens": 10970179.0,
      "step": 5600
    },
    {
      "entropy": 6.388416910171509,
      "epoch": 0.560303893637227,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004976043842360877,
      "loss": 6.2608,
      "mean_token_accuracy": 0.1408168628811836,
      "num_tokens": 10980446.0,
      "step": 5605
    },
    {
      "entropy": 6.361445617675781,
      "epoch": 0.5608037186984556,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497599187353257,
      "loss": 6.3193,
      "mean_token_accuracy": 0.13696268126368522,
      "num_tokens": 10990883.0,
      "step": 5610
    },
    {
      "entropy": 6.464084434509277,
      "epoch": 0.5613035437596842,
      "grad_norm": 1.0,
      "learning_rate": 0.0004975939848698879,
      "loss": 6.2137,
      "mean_token_accuracy": 0.14719685465097426,
      "num_tokens": 11000956.0,
      "step": 5615
    },
    {
      "entropy": 6.375065851211548,
      "epoch": 0.5618033688209126,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004975887767861113,
      "loss": 6.2761,
      "mean_token_accuracy": 0.14079812467098235,
      "num_tokens": 11010873.0,
      "step": 5620
    },
    {
      "entropy": 6.432920503616333,
      "epoch": 0.5623031938821412,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004975835631020585,
      "loss": 6.2392,
      "mean_token_accuracy": 0.14312080442905425,
      "num_tokens": 11020192.0,
      "step": 5625
    },
    {
      "entropy": 6.3562421798706055,
      "epoch": 0.5628030189433698,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004975783438178605,
      "loss": 6.2921,
      "mean_token_accuracy": 0.13569467142224312,
      "num_tokens": 11030466.0,
      "step": 5630
    },
    {
      "entropy": 6.361410617828369,
      "epoch": 0.5633028440045984,
      "grad_norm": 1.0,
      "learning_rate": 0.0004975731189336487,
      "loss": 6.2416,
      "mean_token_accuracy": 0.14767612665891647,
      "num_tokens": 11040309.0,
      "step": 5635
    },
    {
      "entropy": 6.403742837905884,
      "epoch": 0.563802669065827,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004975678884495544,
      "loss": 6.3234,
      "mean_token_accuracy": 0.13869719654321672,
      "num_tokens": 11049560.0,
      "step": 5640
    },
    {
      "entropy": 6.479496192932129,
      "epoch": 0.5643024941270556,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004975626523657093,
      "loss": 6.4108,
      "mean_token_accuracy": 0.13150647208094596,
      "num_tokens": 11059352.0,
      "step": 5645
    },
    {
      "entropy": 6.46918420791626,
      "epoch": 0.564802319188284,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004975574106822451,
      "loss": 6.3069,
      "mean_token_accuracy": 0.1438026748597622,
      "num_tokens": 11069159.0,
      "step": 5650
    },
    {
      "entropy": 6.422984552383423,
      "epoch": 0.5653021442495126,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004975521633992938,
      "loss": 6.3772,
      "mean_token_accuracy": 0.13444813266396521,
      "num_tokens": 11078960.0,
      "step": 5655
    },
    {
      "entropy": 6.363971948623657,
      "epoch": 0.5658019693107412,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975469105169873,
      "loss": 6.2074,
      "mean_token_accuracy": 0.1523880749940872,
      "num_tokens": 11089408.0,
      "step": 5660
    },
    {
      "entropy": 6.4301598072052,
      "epoch": 0.5663017943719698,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004975416520354579,
      "loss": 6.3813,
      "mean_token_accuracy": 0.13310238793492318,
      "num_tokens": 11099485.0,
      "step": 5665
    },
    {
      "entropy": 6.457961130142212,
      "epoch": 0.5668016194331984,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004975363879548377,
      "loss": 6.3826,
      "mean_token_accuracy": 0.14362878128886222,
      "num_tokens": 11109568.0,
      "step": 5670
    },
    {
      "entropy": 6.543348550796509,
      "epoch": 0.567301444494427,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004975311182752592,
      "loss": 6.4642,
      "mean_token_accuracy": 0.12970269322395325,
      "num_tokens": 11119433.0,
      "step": 5675
    },
    {
      "entropy": 6.518721628189087,
      "epoch": 0.5678012695556556,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004975258429968551,
      "loss": 6.3588,
      "mean_token_accuracy": 0.13601181283593178,
      "num_tokens": 11130060.0,
      "step": 5680
    },
    {
      "entropy": 6.4756677627563475,
      "epoch": 0.568301094616884,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004975205621197581,
      "loss": 6.4047,
      "mean_token_accuracy": 0.13375984281301498,
      "num_tokens": 11140417.0,
      "step": 5685
    },
    {
      "entropy": 6.5089599609375,
      "epoch": 0.5688009196781126,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004975152756441009,
      "loss": 6.3731,
      "mean_token_accuracy": 0.1424649626016617,
      "num_tokens": 11150830.0,
      "step": 5690
    },
    {
      "entropy": 6.41643648147583,
      "epoch": 0.5693007447393412,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004975099835700166,
      "loss": 6.1837,
      "mean_token_accuracy": 0.14934798330068588,
      "num_tokens": 11161057.0,
      "step": 5695
    },
    {
      "entropy": 6.330704832077027,
      "epoch": 0.5698005698005698,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004975046858976386,
      "loss": 6.252,
      "mean_token_accuracy": 0.14019235074520112,
      "num_tokens": 11170182.0,
      "step": 5700
    },
    {
      "entropy": 6.366567659378052,
      "epoch": 0.5703003948617984,
      "grad_norm": 1.125,
      "learning_rate": 0.0004974993826270997,
      "loss": 6.2072,
      "mean_token_accuracy": 0.14439581632614135,
      "num_tokens": 11179962.0,
      "step": 5705
    },
    {
      "entropy": 6.347713899612427,
      "epoch": 0.570800219923027,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004974940737585337,
      "loss": 6.1597,
      "mean_token_accuracy": 0.1533850759267807,
      "num_tokens": 11189341.0,
      "step": 5710
    },
    {
      "entropy": 6.198566246032715,
      "epoch": 0.5713000449842555,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000497488759292074,
      "loss": 6.3078,
      "mean_token_accuracy": 0.14763169661164283,
      "num_tokens": 11199264.0,
      "step": 5715
    },
    {
      "entropy": 6.301087284088135,
      "epoch": 0.571799870045484,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974834392278545,
      "loss": 6.2995,
      "mean_token_accuracy": 0.14792427718639373,
      "num_tokens": 11209257.0,
      "step": 5720
    },
    {
      "entropy": 6.50325756072998,
      "epoch": 0.5722996951067126,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004974781135660088,
      "loss": 6.3471,
      "mean_token_accuracy": 0.1401437282562256,
      "num_tokens": 11218294.0,
      "step": 5725
    },
    {
      "entropy": 6.329670333862305,
      "epoch": 0.5727995201679412,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004974727823066709,
      "loss": 6.1759,
      "mean_token_accuracy": 0.14815108031034468,
      "num_tokens": 11227684.0,
      "step": 5730
    },
    {
      "entropy": 6.42995138168335,
      "epoch": 0.5732993452291698,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004974674454499753,
      "loss": 6.3792,
      "mean_token_accuracy": 0.131838421523571,
      "num_tokens": 11237195.0,
      "step": 5735
    },
    {
      "entropy": 6.459194278717041,
      "epoch": 0.5737991702903984,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974621029960559,
      "loss": 6.3426,
      "mean_token_accuracy": 0.1389889284968376,
      "num_tokens": 11246871.0,
      "step": 5740
    },
    {
      "entropy": 6.434127759933472,
      "epoch": 0.574298995351627,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004974567549450472,
      "loss": 6.243,
      "mean_token_accuracy": 0.1382865436375141,
      "num_tokens": 11256298.0,
      "step": 5745
    },
    {
      "entropy": 6.52130823135376,
      "epoch": 0.5747988204128555,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004974514012970839,
      "loss": 6.391,
      "mean_token_accuracy": 0.1376474231481552,
      "num_tokens": 11265857.0,
      "step": 5750
    },
    {
      "entropy": 6.310161113739014,
      "epoch": 0.575298645474084,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004974460420523005,
      "loss": 6.1783,
      "mean_token_accuracy": 0.15020117908716202,
      "num_tokens": 11275406.0,
      "step": 5755
    },
    {
      "entropy": 6.391592979431152,
      "epoch": 0.5757984705353126,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004974406772108319,
      "loss": 6.1733,
      "mean_token_accuracy": 0.14891301542520524,
      "num_tokens": 11283910.0,
      "step": 5760
    },
    {
      "entropy": 6.411246824264526,
      "epoch": 0.5762982955965412,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974353067728132,
      "loss": 6.2447,
      "mean_token_accuracy": 0.13905480802059172,
      "num_tokens": 11293808.0,
      "step": 5765
    },
    {
      "entropy": 6.410305786132812,
      "epoch": 0.5767981206577698,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004974299307383794,
      "loss": 6.28,
      "mean_token_accuracy": 0.14077953696250917,
      "num_tokens": 11302519.0,
      "step": 5770
    },
    {
      "entropy": 6.4257300853729244,
      "epoch": 0.5772979457189984,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004974245491076656,
      "loss": 6.3865,
      "mean_token_accuracy": 0.131597201526165,
      "num_tokens": 11312609.0,
      "step": 5775
    },
    {
      "entropy": 6.4404510974884035,
      "epoch": 0.5777977707802269,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004974191618808075,
      "loss": 6.3195,
      "mean_token_accuracy": 0.14218593165278434,
      "num_tokens": 11322471.0,
      "step": 5780
    },
    {
      "entropy": 6.474221086502075,
      "epoch": 0.5782975958414555,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004974137690579405,
      "loss": 6.3291,
      "mean_token_accuracy": 0.14048926532268524,
      "num_tokens": 11333332.0,
      "step": 5785
    },
    {
      "entropy": 6.422118186950684,
      "epoch": 0.578797420902684,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004974083706392004,
      "loss": 6.3176,
      "mean_token_accuracy": 0.1382746011018753,
      "num_tokens": 11342952.0,
      "step": 5790
    },
    {
      "entropy": 6.428772211074829,
      "epoch": 0.5792972459639126,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004974029666247228,
      "loss": 6.3823,
      "mean_token_accuracy": 0.13532100170850753,
      "num_tokens": 11353509.0,
      "step": 5795
    },
    {
      "entropy": 6.558149433135986,
      "epoch": 0.5797970710251412,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004973975570146438,
      "loss": 6.3787,
      "mean_token_accuracy": 0.13695346638560296,
      "num_tokens": 11363065.0,
      "step": 5800
    },
    {
      "entropy": 6.312878847122192,
      "epoch": 0.5802968960863698,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004973921418090997,
      "loss": 6.2497,
      "mean_token_accuracy": 0.14104935005307198,
      "num_tokens": 11373720.0,
      "step": 5805
    },
    {
      "entropy": 6.303160190582275,
      "epoch": 0.5807967211475984,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004973867210082263,
      "loss": 6.2445,
      "mean_token_accuracy": 0.1410558134317398,
      "num_tokens": 11384432.0,
      "step": 5810
    },
    {
      "entropy": 6.469102764129639,
      "epoch": 0.5812965462088269,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004973812946121603,
      "loss": 6.417,
      "mean_token_accuracy": 0.13949493914842606,
      "num_tokens": 11394100.0,
      "step": 5815
    },
    {
      "entropy": 6.391165828704834,
      "epoch": 0.5817963712700555,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004973758626210382,
      "loss": 6.2787,
      "mean_token_accuracy": 0.13762608021497727,
      "num_tokens": 11403359.0,
      "step": 5820
    },
    {
      "entropy": 6.431511735916137,
      "epoch": 0.582296196331284,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004973704250349965,
      "loss": 6.3119,
      "mean_token_accuracy": 0.13376598954200744,
      "num_tokens": 11412921.0,
      "step": 5825
    },
    {
      "entropy": 6.368031454086304,
      "epoch": 0.5827960213925126,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004973649818541722,
      "loss": 6.2622,
      "mean_token_accuracy": 0.1414501264691353,
      "num_tokens": 11422459.0,
      "step": 5830
    },
    {
      "entropy": 6.334111738204956,
      "epoch": 0.5832958464537412,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004973595330787022,
      "loss": 6.2433,
      "mean_token_accuracy": 0.1441892385482788,
      "num_tokens": 11431517.0,
      "step": 5835
    },
    {
      "entropy": 6.4042764663696286,
      "epoch": 0.5837956715149698,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004973540787087235,
      "loss": 6.2785,
      "mean_token_accuracy": 0.1401561513543129,
      "num_tokens": 11440662.0,
      "step": 5840
    },
    {
      "entropy": 6.305119132995605,
      "epoch": 0.5842954965761983,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004973486187443734,
      "loss": 6.1169,
      "mean_token_accuracy": 0.152145853638649,
      "num_tokens": 11449762.0,
      "step": 5845
    },
    {
      "entropy": 6.258690071105957,
      "epoch": 0.5847953216374269,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004973431531857893,
      "loss": 6.127,
      "mean_token_accuracy": 0.15245335996150972,
      "num_tokens": 11459963.0,
      "step": 5850
    },
    {
      "entropy": 6.378319978713989,
      "epoch": 0.5852951466986555,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004973376820331087,
      "loss": 6.3085,
      "mean_token_accuracy": 0.1369134873151779,
      "num_tokens": 11469063.0,
      "step": 5855
    },
    {
      "entropy": 6.395351839065552,
      "epoch": 0.585794971759884,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004973322052864692,
      "loss": 6.1971,
      "mean_token_accuracy": 0.1487408384680748,
      "num_tokens": 11477651.0,
      "step": 5860
    },
    {
      "entropy": 6.449237442016601,
      "epoch": 0.5862947968211126,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004973267229460087,
      "loss": 6.4146,
      "mean_token_accuracy": 0.13495168089866638,
      "num_tokens": 11487248.0,
      "step": 5865
    },
    {
      "entropy": 6.412726211547851,
      "epoch": 0.5867946218823412,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000497321235011865,
      "loss": 6.3799,
      "mean_token_accuracy": 0.13810717761516572,
      "num_tokens": 11497066.0,
      "step": 5870
    },
    {
      "entropy": 6.479768466949463,
      "epoch": 0.5872944469435698,
      "grad_norm": 1.0,
      "learning_rate": 0.0004973157414841763,
      "loss": 6.3386,
      "mean_token_accuracy": 0.13051055520772933,
      "num_tokens": 11508490.0,
      "step": 5875
    },
    {
      "entropy": 6.466969633102417,
      "epoch": 0.5877942720047983,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004973102423630806,
      "loss": 6.3229,
      "mean_token_accuracy": 0.1406203828752041,
      "num_tokens": 11518851.0,
      "step": 5880
    },
    {
      "entropy": 6.424728870391846,
      "epoch": 0.5882940970660269,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004973047376487166,
      "loss": 6.2937,
      "mean_token_accuracy": 0.14128684774041175,
      "num_tokens": 11529102.0,
      "step": 5885
    },
    {
      "entropy": 6.388212013244629,
      "epoch": 0.5887939221272555,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004972992273412224,
      "loss": 6.3187,
      "mean_token_accuracy": 0.1358910955488682,
      "num_tokens": 11540563.0,
      "step": 5890
    },
    {
      "entropy": 6.391242265701294,
      "epoch": 0.589293747188484,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004972937114407369,
      "loss": 6.2476,
      "mean_token_accuracy": 0.14669288769364358,
      "num_tokens": 11549955.0,
      "step": 5895
    },
    {
      "entropy": 6.376825428009033,
      "epoch": 0.5897935722497126,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497288189947399,
      "loss": 6.2282,
      "mean_token_accuracy": 0.14328095093369483,
      "num_tokens": 11558932.0,
      "step": 5900
    },
    {
      "entropy": 6.270180940628052,
      "epoch": 0.5902933973109412,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004972826628613472,
      "loss": 6.2223,
      "mean_token_accuracy": 0.14396324902772903,
      "num_tokens": 11568901.0,
      "step": 5905
    },
    {
      "entropy": 6.410828733444214,
      "epoch": 0.5907932223721697,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004972771301827209,
      "loss": 6.2864,
      "mean_token_accuracy": 0.14194692373275758,
      "num_tokens": 11579738.0,
      "step": 5910
    },
    {
      "entropy": 6.410345602035522,
      "epoch": 0.5912930474333983,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004972715919116592,
      "loss": 6.2539,
      "mean_token_accuracy": 0.14249148815870286,
      "num_tokens": 11590293.0,
      "step": 5915
    },
    {
      "entropy": 6.413597536087036,
      "epoch": 0.5917928724946269,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004972660480483015,
      "loss": 6.3234,
      "mean_token_accuracy": 0.13725078403949736,
      "num_tokens": 11599745.0,
      "step": 5920
    },
    {
      "entropy": 6.4196656227111815,
      "epoch": 0.5922926975558555,
      "grad_norm": 1.0,
      "learning_rate": 0.0004972604985927871,
      "loss": 6.3158,
      "mean_token_accuracy": 0.13385162949562074,
      "num_tokens": 11609958.0,
      "step": 5925
    },
    {
      "entropy": 6.465918827056885,
      "epoch": 0.592792522617084,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004972549435452558,
      "loss": 6.3028,
      "mean_token_accuracy": 0.13586054369807243,
      "num_tokens": 11619941.0,
      "step": 5930
    },
    {
      "entropy": 6.357407808303833,
      "epoch": 0.5932923476783126,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004972493829058474,
      "loss": 6.1748,
      "mean_token_accuracy": 0.14977660328149794,
      "num_tokens": 11630287.0,
      "step": 5935
    },
    {
      "entropy": 6.351117706298828,
      "epoch": 0.5937921727395412,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004972438166747016,
      "loss": 6.2309,
      "mean_token_accuracy": 0.14242334812879562,
      "num_tokens": 11639806.0,
      "step": 5940
    },
    {
      "entropy": 6.377180862426758,
      "epoch": 0.5942919978007697,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004972382448519586,
      "loss": 6.333,
      "mean_token_accuracy": 0.1424495354294777,
      "num_tokens": 11650132.0,
      "step": 5945
    },
    {
      "entropy": 6.390507555007934,
      "epoch": 0.5947918228619983,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004972326674377586,
      "loss": 6.2394,
      "mean_token_accuracy": 0.14251381009817124,
      "num_tokens": 11659670.0,
      "step": 5950
    },
    {
      "entropy": 6.415858125686645,
      "epoch": 0.5952916479232269,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004972270844322418,
      "loss": 6.3152,
      "mean_token_accuracy": 0.14169998168945314,
      "num_tokens": 11669998.0,
      "step": 5955
    },
    {
      "entropy": 6.416223192214966,
      "epoch": 0.5957914729844555,
      "grad_norm": 1.125,
      "learning_rate": 0.0004972214958355488,
      "loss": 6.306,
      "mean_token_accuracy": 0.13907138854265214,
      "num_tokens": 11679052.0,
      "step": 5960
    },
    {
      "entropy": 6.397159767150879,
      "epoch": 0.596291298045684,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004972159016478201,
      "loss": 6.2894,
      "mean_token_accuracy": 0.14112116992473603,
      "num_tokens": 11687957.0,
      "step": 5965
    },
    {
      "entropy": 6.3890931606292725,
      "epoch": 0.5967911231069126,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004972103018691965,
      "loss": 6.2085,
      "mean_token_accuracy": 0.14139744713902475,
      "num_tokens": 11696922.0,
      "step": 5970
    },
    {
      "entropy": 6.377246761322022,
      "epoch": 0.5972909481681411,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004972046964998189,
      "loss": 6.2039,
      "mean_token_accuracy": 0.14591391384601593,
      "num_tokens": 11706600.0,
      "step": 5975
    },
    {
      "entropy": 6.311473655700683,
      "epoch": 0.5977907732293697,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004971990855398283,
      "loss": 6.2514,
      "mean_token_accuracy": 0.1409190982580185,
      "num_tokens": 11715933.0,
      "step": 5980
    },
    {
      "entropy": 6.391643905639649,
      "epoch": 0.5982905982905983,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004971934689893658,
      "loss": 6.1959,
      "mean_token_accuracy": 0.14715758115053176,
      "num_tokens": 11724807.0,
      "step": 5985
    },
    {
      "entropy": 6.410089302062988,
      "epoch": 0.5987904233518269,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004971878468485729,
      "loss": 6.2802,
      "mean_token_accuracy": 0.14309779852628707,
      "num_tokens": 11734985.0,
      "step": 5990
    },
    {
      "entropy": 6.356278133392334,
      "epoch": 0.5992902484130554,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000497182219117591,
      "loss": 6.2708,
      "mean_token_accuracy": 0.13901931121945382,
      "num_tokens": 11743418.0,
      "step": 5995
    },
    {
      "entropy": 6.3269764423370365,
      "epoch": 0.599790073474284,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004971765857965615,
      "loss": 6.2077,
      "mean_token_accuracy": 0.13885150700807572,
      "num_tokens": 11752953.0,
      "step": 6000
    },
    {
      "epoch": 0.599790073474284,
      "eval_entropy": 6.200126294468027,
      "eval_loss": 6.319167137145996,
      "eval_mean_token_accuracy": 0.14492665400250562,
      "eval_num_tokens": 11752953.0,
      "eval_runtime": 23.6915,
      "eval_samples_per_second": 1310.384,
      "eval_steps_per_second": 163.814,
      "step": 6000
    },
    {
      "entropy": 6.37937445640564,
      "epoch": 0.6002898985355126,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004971709468856263,
      "loss": 6.3404,
      "mean_token_accuracy": 0.13874611407518386,
      "num_tokens": 11763260.0,
      "step": 6005
    },
    {
      "entropy": 6.33965482711792,
      "epoch": 0.6007897235967411,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004971653023849274,
      "loss": 6.1228,
      "mean_token_accuracy": 0.16097717061638833,
      "num_tokens": 11773255.0,
      "step": 6010
    },
    {
      "entropy": 6.375664234161377,
      "epoch": 0.6012895486579697,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004971596522946065,
      "loss": 6.2665,
      "mean_token_accuracy": 0.1420398011803627,
      "num_tokens": 11783443.0,
      "step": 6015
    },
    {
      "entropy": 6.311613464355469,
      "epoch": 0.6017893737191983,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004971539966148059,
      "loss": 6.2321,
      "mean_token_accuracy": 0.14004560485482215,
      "num_tokens": 11791932.0,
      "step": 6020
    },
    {
      "entropy": 6.380346775054932,
      "epoch": 0.6022891987804269,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004971483353456681,
      "loss": 6.2134,
      "mean_token_accuracy": 0.14575803130865098,
      "num_tokens": 11801420.0,
      "step": 6025
    },
    {
      "entropy": 6.4092343807220455,
      "epoch": 0.6027890238416554,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004971426684873352,
      "loss": 6.3576,
      "mean_token_accuracy": 0.13633519932627677,
      "num_tokens": 11810977.0,
      "step": 6030
    },
    {
      "entropy": 6.291591024398803,
      "epoch": 0.603288848902884,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004971369960399499,
      "loss": 6.1514,
      "mean_token_accuracy": 0.14933094680309295,
      "num_tokens": 11821016.0,
      "step": 6035
    },
    {
      "entropy": 6.431398200988769,
      "epoch": 0.6037886739641125,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497131318003655,
      "loss": 6.3002,
      "mean_token_accuracy": 0.138646849244833,
      "num_tokens": 11831157.0,
      "step": 6040
    },
    {
      "entropy": 6.349044561386108,
      "epoch": 0.6042884990253411,
      "grad_norm": 1.125,
      "learning_rate": 0.0004971256343785932,
      "loss": 6.2127,
      "mean_token_accuracy": 0.14890757203102112,
      "num_tokens": 11840319.0,
      "step": 6045
    },
    {
      "entropy": 6.384908151626587,
      "epoch": 0.6047883240865697,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004971199451649078,
      "loss": 6.3386,
      "mean_token_accuracy": 0.13837055936455728,
      "num_tokens": 11850808.0,
      "step": 6050
    },
    {
      "entropy": 6.472949266433716,
      "epoch": 0.6052881491477983,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004971142503627415,
      "loss": 6.3843,
      "mean_token_accuracy": 0.14083569422364234,
      "num_tokens": 11861123.0,
      "step": 6055
    },
    {
      "entropy": 6.3204974174499515,
      "epoch": 0.6057879742090269,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497108549972238,
      "loss": 6.115,
      "mean_token_accuracy": 0.1516193240880966,
      "num_tokens": 11870308.0,
      "step": 6060
    },
    {
      "entropy": 6.424834442138672,
      "epoch": 0.6062877992702554,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004971028439935404,
      "loss": 6.2622,
      "mean_token_accuracy": 0.1355499692261219,
      "num_tokens": 11880254.0,
      "step": 6065
    },
    {
      "entropy": 6.38127818107605,
      "epoch": 0.606787624331484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004970971324267925,
      "loss": 6.2683,
      "mean_token_accuracy": 0.1444159232079983,
      "num_tokens": 11891272.0,
      "step": 6070
    },
    {
      "entropy": 6.350743150711059,
      "epoch": 0.6072874493927125,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004970914152721378,
      "loss": 6.2815,
      "mean_token_accuracy": 0.1390413299202919,
      "num_tokens": 11901176.0,
      "step": 6075
    },
    {
      "entropy": 6.418502950668335,
      "epoch": 0.6077872744539411,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004970856925297204,
      "loss": 6.3012,
      "mean_token_accuracy": 0.13518301621079445,
      "num_tokens": 11910820.0,
      "step": 6080
    },
    {
      "entropy": 6.369509553909301,
      "epoch": 0.6082870995151697,
      "grad_norm": 1.1875,
      "learning_rate": 0.000497079964199684,
      "loss": 6.2519,
      "mean_token_accuracy": 0.1334824465215206,
      "num_tokens": 11920458.0,
      "step": 6085
    },
    {
      "entropy": 6.248316907882691,
      "epoch": 0.6087869245763983,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004970742302821729,
      "loss": 6.1138,
      "mean_token_accuracy": 0.15163505971431732,
      "num_tokens": 11931150.0,
      "step": 6090
    },
    {
      "entropy": 6.3462635517120365,
      "epoch": 0.6092867496376269,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004970684907773312,
      "loss": 6.2372,
      "mean_token_accuracy": 0.14376992285251616,
      "num_tokens": 11939960.0,
      "step": 6095
    },
    {
      "entropy": 6.463783550262451,
      "epoch": 0.6097865746988554,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004970627456853036,
      "loss": 6.4047,
      "mean_token_accuracy": 0.13501383736729622,
      "num_tokens": 11950781.0,
      "step": 6100
    },
    {
      "entropy": 6.448822689056397,
      "epoch": 0.6102863997600839,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004970569950062343,
      "loss": 6.2275,
      "mean_token_accuracy": 0.1446210116147995,
      "num_tokens": 11960489.0,
      "step": 6105
    },
    {
      "entropy": 6.357200622558594,
      "epoch": 0.6107862248213125,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004970512387402682,
      "loss": 6.2712,
      "mean_token_accuracy": 0.14385456815361977,
      "num_tokens": 11970651.0,
      "step": 6110
    },
    {
      "entropy": 6.280639219284057,
      "epoch": 0.6112860498825411,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004970454768875501,
      "loss": 6.2431,
      "mean_token_accuracy": 0.1407626338303089,
      "num_tokens": 11980349.0,
      "step": 6115
    },
    {
      "entropy": 6.4302277088165285,
      "epoch": 0.6117858749437697,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004970397094482249,
      "loss": 6.2705,
      "mean_token_accuracy": 0.14543462321162223,
      "num_tokens": 11990076.0,
      "step": 6120
    },
    {
      "entropy": 6.32853364944458,
      "epoch": 0.6122857000049983,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004970339364224378,
      "loss": 6.1901,
      "mean_token_accuracy": 0.1470900222659111,
      "num_tokens": 11999515.0,
      "step": 6125
    },
    {
      "entropy": 6.370386362075806,
      "epoch": 0.6127855250662269,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004970281578103339,
      "loss": 6.3115,
      "mean_token_accuracy": 0.13805368021130562,
      "num_tokens": 12009224.0,
      "step": 6130
    },
    {
      "entropy": 6.329615545272827,
      "epoch": 0.6132853501274554,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004970223736120587,
      "loss": 6.2021,
      "mean_token_accuracy": 0.14378652796149255,
      "num_tokens": 12019295.0,
      "step": 6135
    },
    {
      "entropy": 6.4103930473327635,
      "epoch": 0.6137851751886839,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004970165838277578,
      "loss": 6.2859,
      "mean_token_accuracy": 0.13757352903485298,
      "num_tokens": 12029130.0,
      "step": 6140
    },
    {
      "entropy": 6.372373342514038,
      "epoch": 0.6142850002499125,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004970107884575767,
      "loss": 6.2483,
      "mean_token_accuracy": 0.1443093605339527,
      "num_tokens": 12038676.0,
      "step": 6145
    },
    {
      "entropy": 6.379946947097778,
      "epoch": 0.6147848253111411,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004970049875016613,
      "loss": 6.1924,
      "mean_token_accuracy": 0.15777153447270392,
      "num_tokens": 12048089.0,
      "step": 6150
    },
    {
      "entropy": 6.3757730484008786,
      "epoch": 0.6152846503723697,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004969991809601576,
      "loss": 6.2735,
      "mean_token_accuracy": 0.14039193764328955,
      "num_tokens": 12059531.0,
      "step": 6155
    },
    {
      "entropy": 6.424350309371948,
      "epoch": 0.6157844754335983,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004969933688332115,
      "loss": 6.3018,
      "mean_token_accuracy": 0.13488076776266097,
      "num_tokens": 12070103.0,
      "step": 6160
    },
    {
      "entropy": 6.312880611419677,
      "epoch": 0.6162843004948269,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004969875511209694,
      "loss": 6.2722,
      "mean_token_accuracy": 0.14213053286075591,
      "num_tokens": 12078481.0,
      "step": 6165
    },
    {
      "entropy": 6.4138304710388185,
      "epoch": 0.6167841255560553,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004969817278235777,
      "loss": 6.3421,
      "mean_token_accuracy": 0.1397414065897465,
      "num_tokens": 12088875.0,
      "step": 6170
    },
    {
      "entropy": 6.44084415435791,
      "epoch": 0.6172839506172839,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004969758989411829,
      "loss": 6.263,
      "mean_token_accuracy": 0.14128744527697562,
      "num_tokens": 12099769.0,
      "step": 6175
    },
    {
      "entropy": 6.434930467605591,
      "epoch": 0.6177837756785125,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004969700644739315,
      "loss": 6.2425,
      "mean_token_accuracy": 0.1408712610602379,
      "num_tokens": 12109334.0,
      "step": 6180
    },
    {
      "entropy": 6.373007678985596,
      "epoch": 0.6182836007397411,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004969642244219705,
      "loss": 6.2379,
      "mean_token_accuracy": 0.1429442673921585,
      "num_tokens": 12119017.0,
      "step": 6185
    },
    {
      "entropy": 6.304239225387573,
      "epoch": 0.6187834258009697,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004969583787854467,
      "loss": 6.1542,
      "mean_token_accuracy": 0.1520055688917637,
      "num_tokens": 12128002.0,
      "step": 6190
    },
    {
      "entropy": 6.354333686828613,
      "epoch": 0.6192832508621983,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004969525275645071,
      "loss": 6.2409,
      "mean_token_accuracy": 0.1376312144100666,
      "num_tokens": 12137802.0,
      "step": 6195
    },
    {
      "entropy": 6.2719940662384035,
      "epoch": 0.6197830759234269,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004969466707592991,
      "loss": 6.2029,
      "mean_token_accuracy": 0.1463504582643509,
      "num_tokens": 12147270.0,
      "step": 6200
    },
    {
      "entropy": 6.343167400360107,
      "epoch": 0.6202829009846553,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00049694080836997,
      "loss": 6.249,
      "mean_token_accuracy": 0.141249780356884,
      "num_tokens": 12157358.0,
      "step": 6205
    },
    {
      "entropy": 6.504302453994751,
      "epoch": 0.6207827260458839,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004969349403966672,
      "loss": 6.3327,
      "mean_token_accuracy": 0.1395859971642494,
      "num_tokens": 12166159.0,
      "step": 6210
    },
    {
      "entropy": 6.388160991668701,
      "epoch": 0.6212825511071125,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004969290668395385,
      "loss": 6.2893,
      "mean_token_accuracy": 0.14317970126867294,
      "num_tokens": 12176405.0,
      "step": 6215
    },
    {
      "entropy": 6.4428479194641115,
      "epoch": 0.6217823761683411,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004969231876987315,
      "loss": 6.3357,
      "mean_token_accuracy": 0.13834265395998954,
      "num_tokens": 12186021.0,
      "step": 6220
    },
    {
      "entropy": 6.315100622177124,
      "epoch": 0.6222822012295697,
      "grad_norm": 1.125,
      "learning_rate": 0.0004969173029743942,
      "loss": 6.2045,
      "mean_token_accuracy": 0.14326648339629172,
      "num_tokens": 12195222.0,
      "step": 6225
    },
    {
      "entropy": 6.334034490585327,
      "epoch": 0.6227820262907983,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004969114126666746,
      "loss": 6.1856,
      "mean_token_accuracy": 0.1475788578391075,
      "num_tokens": 12205443.0,
      "step": 6230
    },
    {
      "entropy": 6.451016187667847,
      "epoch": 0.6232818513520267,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004969055167757211,
      "loss": 6.3885,
      "mean_token_accuracy": 0.14134080708026886,
      "num_tokens": 12215181.0,
      "step": 6235
    },
    {
      "entropy": 6.375768947601318,
      "epoch": 0.6237816764132553,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004968996153016819,
      "loss": 6.2351,
      "mean_token_accuracy": 0.1424890138208866,
      "num_tokens": 12224327.0,
      "step": 6240
    },
    {
      "entropy": 6.3680023670196535,
      "epoch": 0.6242815014744839,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004968937082447054,
      "loss": 6.2831,
      "mean_token_accuracy": 0.13955292478203773,
      "num_tokens": 12234962.0,
      "step": 6245
    },
    {
      "entropy": 6.420923805236816,
      "epoch": 0.6247813265357125,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004968877956049403,
      "loss": 6.3197,
      "mean_token_accuracy": 0.14237649738788605,
      "num_tokens": 12245883.0,
      "step": 6250
    },
    {
      "entropy": 6.450377702713013,
      "epoch": 0.6252811515969411,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004968818773825355,
      "loss": 6.317,
      "mean_token_accuracy": 0.1347426824271679,
      "num_tokens": 12254412.0,
      "step": 6255
    },
    {
      "entropy": 6.2472062587738035,
      "epoch": 0.6257809766581697,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004968759535776396,
      "loss": 6.1837,
      "mean_token_accuracy": 0.14387175366282462,
      "num_tokens": 12264894.0,
      "step": 6260
    },
    {
      "entropy": 6.369059276580811,
      "epoch": 0.6262808017193983,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004968700241904018,
      "loss": 6.2849,
      "mean_token_accuracy": 0.1456960268318653,
      "num_tokens": 12274557.0,
      "step": 6265
    },
    {
      "entropy": 6.345845031738281,
      "epoch": 0.6267806267806267,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004968640892209714,
      "loss": 6.1219,
      "mean_token_accuracy": 0.14875716716051102,
      "num_tokens": 12283696.0,
      "step": 6270
    },
    {
      "entropy": 6.295801639556885,
      "epoch": 0.6272804518418553,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004968581486694975,
      "loss": 6.2145,
      "mean_token_accuracy": 0.14838205724954606,
      "num_tokens": 12292981.0,
      "step": 6275
    },
    {
      "entropy": 6.311672401428223,
      "epoch": 0.6277802769030839,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004968522025361297,
      "loss": 6.0905,
      "mean_token_accuracy": 0.15002752989530563,
      "num_tokens": 12302865.0,
      "step": 6280
    },
    {
      "entropy": 6.229003810882569,
      "epoch": 0.6282801019643125,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004968462508210176,
      "loss": 6.2575,
      "mean_token_accuracy": 0.14212906286120414,
      "num_tokens": 12313099.0,
      "step": 6285
    },
    {
      "entropy": 6.334918403625489,
      "epoch": 0.6287799270255411,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004968402935243109,
      "loss": 6.3027,
      "mean_token_accuracy": 0.14179675355553628,
      "num_tokens": 12324155.0,
      "step": 6290
    },
    {
      "entropy": 6.336399221420288,
      "epoch": 0.6292797520867697,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004968343306461595,
      "loss": 6.1329,
      "mean_token_accuracy": 0.14738946855068208,
      "num_tokens": 12334082.0,
      "step": 6295
    },
    {
      "entropy": 6.284357261657715,
      "epoch": 0.6297795771479981,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004968283621867134,
      "loss": 6.2325,
      "mean_token_accuracy": 0.14591046348214148,
      "num_tokens": 12344673.0,
      "step": 6300
    },
    {
      "entropy": 6.355352258682251,
      "epoch": 0.6302794022092267,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004968223881461227,
      "loss": 6.2413,
      "mean_token_accuracy": 0.14010884910821914,
      "num_tokens": 12355149.0,
      "step": 6305
    },
    {
      "entropy": 6.332917261123657,
      "epoch": 0.6307792272704553,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000496816408524538,
      "loss": 6.1279,
      "mean_token_accuracy": 0.15396927446126937,
      "num_tokens": 12364043.0,
      "step": 6310
    },
    {
      "entropy": 6.326907682418823,
      "epoch": 0.6312790523316839,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004968104233221094,
      "loss": 6.3338,
      "mean_token_accuracy": 0.13666420578956603,
      "num_tokens": 12373591.0,
      "step": 6315
    },
    {
      "entropy": 6.306742525100708,
      "epoch": 0.6317788773929125,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004968044325389877,
      "loss": 6.2709,
      "mean_token_accuracy": 0.14339900016784668,
      "num_tokens": 12383570.0,
      "step": 6320
    },
    {
      "entropy": 6.469053125381469,
      "epoch": 0.6322787024541411,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004967984361753235,
      "loss": 6.2534,
      "mean_token_accuracy": 0.1384721130132675,
      "num_tokens": 12393682.0,
      "step": 6325
    },
    {
      "entropy": 6.399180221557617,
      "epoch": 0.6327785275153697,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004967924342312677,
      "loss": 6.1955,
      "mean_token_accuracy": 0.14260394275188445,
      "num_tokens": 12403478.0,
      "step": 6330
    },
    {
      "entropy": 6.348455286026001,
      "epoch": 0.6332783525765981,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004967864267069713,
      "loss": 6.2833,
      "mean_token_accuracy": 0.13903689235448838,
      "num_tokens": 12412810.0,
      "step": 6335
    },
    {
      "entropy": 6.3152111053466795,
      "epoch": 0.6337781776378267,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004967804136025856,
      "loss": 6.1627,
      "mean_token_accuracy": 0.14873817563056946,
      "num_tokens": 12422209.0,
      "step": 6340
    },
    {
      "entropy": 6.329520893096924,
      "epoch": 0.6342780026990553,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004967743949182617,
      "loss": 6.224,
      "mean_token_accuracy": 0.1447262704372406,
      "num_tokens": 12431569.0,
      "step": 6345
    },
    {
      "entropy": 6.266342306137085,
      "epoch": 0.6347778277602839,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004967683706541511,
      "loss": 6.0976,
      "mean_token_accuracy": 0.15906884968280793,
      "num_tokens": 12441607.0,
      "step": 6350
    },
    {
      "entropy": 6.311867952346802,
      "epoch": 0.6352776528215125,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004967623408104054,
      "loss": 6.2333,
      "mean_token_accuracy": 0.1393691383302212,
      "num_tokens": 12451072.0,
      "step": 6355
    },
    {
      "entropy": 6.4193377017974855,
      "epoch": 0.6357774778827411,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004967563053871762,
      "loss": 6.2671,
      "mean_token_accuracy": 0.14046888425946236,
      "num_tokens": 12460798.0,
      "step": 6360
    },
    {
      "entropy": 6.228210544586181,
      "epoch": 0.6362773029439696,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004967502643846155,
      "loss": 6.1613,
      "mean_token_accuracy": 0.14846557453274728,
      "num_tokens": 12471891.0,
      "step": 6365
    },
    {
      "entropy": 6.238931131362915,
      "epoch": 0.6367771280051981,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000496744217802875,
      "loss": 6.2195,
      "mean_token_accuracy": 0.14480772614479065,
      "num_tokens": 12482212.0,
      "step": 6370
    },
    {
      "entropy": 6.417540884017944,
      "epoch": 0.6372769530664267,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004967381656421073,
      "loss": 6.2337,
      "mean_token_accuracy": 0.1450836405158043,
      "num_tokens": 12492517.0,
      "step": 6375
    },
    {
      "entropy": 6.340852069854736,
      "epoch": 0.6377767781276553,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004967321079024643,
      "loss": 6.1674,
      "mean_token_accuracy": 0.14534047693014146,
      "num_tokens": 12502521.0,
      "step": 6380
    },
    {
      "entropy": 6.328003883361816,
      "epoch": 0.6382766031888839,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004967260445840985,
      "loss": 6.29,
      "mean_token_accuracy": 0.13856207579374313,
      "num_tokens": 12513195.0,
      "step": 6385
    },
    {
      "entropy": 6.3490440368652346,
      "epoch": 0.6387764282501125,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004967199756871624,
      "loss": 6.0913,
      "mean_token_accuracy": 0.15365727618336678,
      "num_tokens": 12521640.0,
      "step": 6390
    },
    {
      "entropy": 6.291057920455932,
      "epoch": 0.6392762533113411,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004967139012118088,
      "loss": 6.2473,
      "mean_token_accuracy": 0.14836501628160476,
      "num_tokens": 12530807.0,
      "step": 6395
    },
    {
      "entropy": 6.318085193634033,
      "epoch": 0.6397760783725696,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004967078211581907,
      "loss": 6.317,
      "mean_token_accuracy": 0.14261332005262375,
      "num_tokens": 12542049.0,
      "step": 6400
    },
    {
      "entropy": 6.431751871109009,
      "epoch": 0.6402759034337981,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004967017355264606,
      "loss": 6.259,
      "mean_token_accuracy": 0.1470594309270382,
      "num_tokens": 12551699.0,
      "step": 6405
    },
    {
      "entropy": 6.371747446060181,
      "epoch": 0.6407757284950267,
      "grad_norm": 1.15625,
      "learning_rate": 0.000496695644316772,
      "loss": 6.3154,
      "mean_token_accuracy": 0.13668317273259162,
      "num_tokens": 12561414.0,
      "step": 6410
    },
    {
      "entropy": 6.3026398658752445,
      "epoch": 0.6412755535562553,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004966895475292781,
      "loss": 6.3125,
      "mean_token_accuracy": 0.14197110161185264,
      "num_tokens": 12571950.0,
      "step": 6415
    },
    {
      "entropy": 6.475308132171631,
      "epoch": 0.6417753786174839,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004966834451641322,
      "loss": 6.3271,
      "mean_token_accuracy": 0.14135714769363403,
      "num_tokens": 12581264.0,
      "step": 6420
    },
    {
      "entropy": 6.416059350967407,
      "epoch": 0.6422752036787125,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004966773372214878,
      "loss": 6.2793,
      "mean_token_accuracy": 0.1413417190313339,
      "num_tokens": 12590950.0,
      "step": 6425
    },
    {
      "entropy": 6.3199985980987545,
      "epoch": 0.642775028739941,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004966712237014987,
      "loss": 6.212,
      "mean_token_accuracy": 0.14124323055148125,
      "num_tokens": 12599766.0,
      "step": 6430
    },
    {
      "entropy": 6.396810102462768,
      "epoch": 0.6432748538011696,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004966651046043186,
      "loss": 6.2447,
      "mean_token_accuracy": 0.14712451994419098,
      "num_tokens": 12610074.0,
      "step": 6435
    },
    {
      "entropy": 6.232436180114746,
      "epoch": 0.6437746788623981,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004966589799301016,
      "loss": 6.1178,
      "mean_token_accuracy": 0.1550958961248398,
      "num_tokens": 12619117.0,
      "step": 6440
    },
    {
      "entropy": 6.3377618312835695,
      "epoch": 0.6442745039236267,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004966528496790015,
      "loss": 6.2605,
      "mean_token_accuracy": 0.14155772924423218,
      "num_tokens": 12627993.0,
      "step": 6445
    },
    {
      "entropy": 6.378690242767334,
      "epoch": 0.6447743289848553,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004966467138511729,
      "loss": 6.2347,
      "mean_token_accuracy": 0.14662675112485885,
      "num_tokens": 12637787.0,
      "step": 6450
    },
    {
      "entropy": 6.389209461212158,
      "epoch": 0.6452741540460839,
      "grad_norm": 1.34375,
      "learning_rate": 0.00049664057244677,
      "loss": 6.3871,
      "mean_token_accuracy": 0.137041387706995,
      "num_tokens": 12648409.0,
      "step": 6455
    },
    {
      "entropy": 6.307158374786377,
      "epoch": 0.6457739791073125,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004966344254659473,
      "loss": 6.1724,
      "mean_token_accuracy": 0.14861769527196883,
      "num_tokens": 12658391.0,
      "step": 6460
    },
    {
      "entropy": 6.295201539993286,
      "epoch": 0.646273804168541,
      "grad_norm": 1.125,
      "learning_rate": 0.0004966282729088595,
      "loss": 6.2123,
      "mean_token_accuracy": 0.14254108741879462,
      "num_tokens": 12669926.0,
      "step": 6465
    },
    {
      "entropy": 6.331935834884644,
      "epoch": 0.6467736292297696,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004966221147756615,
      "loss": 6.2948,
      "mean_token_accuracy": 0.14155235886573792,
      "num_tokens": 12679176.0,
      "step": 6470
    },
    {
      "entropy": 6.369681787490845,
      "epoch": 0.6472734542909981,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004966159510665079,
      "loss": 6.1881,
      "mean_token_accuracy": 0.14433651566505432,
      "num_tokens": 12688954.0,
      "step": 6475
    },
    {
      "entropy": 6.31721830368042,
      "epoch": 0.6477732793522267,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004966097817815542,
      "loss": 6.1922,
      "mean_token_accuracy": 0.14558025598526,
      "num_tokens": 12697577.0,
      "step": 6480
    },
    {
      "entropy": 6.326429939270019,
      "epoch": 0.6482731044134553,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004966036069209552,
      "loss": 6.3122,
      "mean_token_accuracy": 0.13705639615654946,
      "num_tokens": 12708153.0,
      "step": 6485
    },
    {
      "entropy": 6.403918218612671,
      "epoch": 0.6487729294746839,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004965974264848666,
      "loss": 6.2758,
      "mean_token_accuracy": 0.14280798509716988,
      "num_tokens": 12717329.0,
      "step": 6490
    },
    {
      "entropy": 6.436331701278687,
      "epoch": 0.6492727545359124,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004965912404734438,
      "loss": 6.2678,
      "mean_token_accuracy": 0.14818769693374634,
      "num_tokens": 12726528.0,
      "step": 6495
    },
    {
      "entropy": 6.41108021736145,
      "epoch": 0.649772579597141,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004965850488868424,
      "loss": 6.2742,
      "mean_token_accuracy": 0.1369329161942005,
      "num_tokens": 12737341.0,
      "step": 6500
    },
    {
      "entropy": 6.319310188293457,
      "epoch": 0.6502724046583696,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004965788517252181,
      "loss": 6.1651,
      "mean_token_accuracy": 0.14523185566067695,
      "num_tokens": 12746973.0,
      "step": 6505
    },
    {
      "entropy": 6.405824851989746,
      "epoch": 0.6507722297195981,
      "grad_norm": 1.125,
      "learning_rate": 0.000496572648988727,
      "loss": 6.2012,
      "mean_token_accuracy": 0.1456592336297035,
      "num_tokens": 12756176.0,
      "step": 6510
    },
    {
      "entropy": 6.3502613544464115,
      "epoch": 0.6512720547808267,
      "grad_norm": 1.0625,
      "learning_rate": 0.000496566440677525,
      "loss": 6.2199,
      "mean_token_accuracy": 0.14301720783114433,
      "num_tokens": 12766349.0,
      "step": 6515
    },
    {
      "entropy": 6.284042453765869,
      "epoch": 0.6517718798420553,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004965602267917685,
      "loss": 6.1296,
      "mean_token_accuracy": 0.14756693169474602,
      "num_tokens": 12775857.0,
      "step": 6520
    },
    {
      "entropy": 6.318739175796509,
      "epoch": 0.6522717049032839,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004965540073316136,
      "loss": 6.2317,
      "mean_token_accuracy": 0.14006442874670028,
      "num_tokens": 12786376.0,
      "step": 6525
    },
    {
      "entropy": 6.362497472763062,
      "epoch": 0.6527715299645124,
      "grad_norm": 1.125,
      "learning_rate": 0.0004965477822972169,
      "loss": 6.2404,
      "mean_token_accuracy": 0.13885729312896727,
      "num_tokens": 12795719.0,
      "step": 6530
    },
    {
      "entropy": 6.3323705196380615,
      "epoch": 0.653271355025741,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004965415516887349,
      "loss": 6.25,
      "mean_token_accuracy": 0.1417411744594574,
      "num_tokens": 12806742.0,
      "step": 6535
    },
    {
      "entropy": 6.241915225982666,
      "epoch": 0.6537711800869695,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004965353155063246,
      "loss": 6.0981,
      "mean_token_accuracy": 0.14228224009275436,
      "num_tokens": 12817362.0,
      "step": 6540
    },
    {
      "entropy": 6.218412923812866,
      "epoch": 0.6542710051481981,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004965290737501427,
      "loss": 6.0674,
      "mean_token_accuracy": 0.15282209515571593,
      "num_tokens": 12826516.0,
      "step": 6545
    },
    {
      "entropy": 6.321288681030273,
      "epoch": 0.6547708302094267,
      "grad_norm": 1.125,
      "learning_rate": 0.0004965228264203464,
      "loss": 6.2913,
      "mean_token_accuracy": 0.13362951949238777,
      "num_tokens": 12836321.0,
      "step": 6550
    },
    {
      "entropy": 6.325507545471192,
      "epoch": 0.6552706552706553,
      "grad_norm": 1.125,
      "learning_rate": 0.0004965165735170928,
      "loss": 6.1465,
      "mean_token_accuracy": 0.14639638587832451,
      "num_tokens": 12845955.0,
      "step": 6555
    },
    {
      "entropy": 6.358684158325195,
      "epoch": 0.6557704803318838,
      "grad_norm": 1.171875,
      "learning_rate": 0.000496510315040539,
      "loss": 6.2988,
      "mean_token_accuracy": 0.13919584825634956,
      "num_tokens": 12854749.0,
      "step": 6560
    },
    {
      "entropy": 6.38391375541687,
      "epoch": 0.6562703053931124,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004965040509908429,
      "loss": 6.2844,
      "mean_token_accuracy": 0.13873137757182122,
      "num_tokens": 12865145.0,
      "step": 6565
    },
    {
      "entropy": 6.349937152862549,
      "epoch": 0.656770130454341,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004964977813681618,
      "loss": 6.2061,
      "mean_token_accuracy": 0.150709667801857,
      "num_tokens": 12876109.0,
      "step": 6570
    },
    {
      "entropy": 6.326040172576905,
      "epoch": 0.6572699555155695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004964915061726534,
      "loss": 6.2527,
      "mean_token_accuracy": 0.13785523623228074,
      "num_tokens": 12886813.0,
      "step": 6575
    },
    {
      "entropy": 6.305559062957764,
      "epoch": 0.6577697805767981,
      "grad_norm": 1.0,
      "learning_rate": 0.0004964852254044759,
      "loss": 6.2307,
      "mean_token_accuracy": 0.1438157469034195,
      "num_tokens": 12897991.0,
      "step": 6580
    },
    {
      "entropy": 6.355275344848633,
      "epoch": 0.6582696056380267,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004964789390637871,
      "loss": 6.3035,
      "mean_token_accuracy": 0.13300548642873763,
      "num_tokens": 12907354.0,
      "step": 6585
    },
    {
      "entropy": 6.443806552886963,
      "epoch": 0.6587694306992553,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004964726471507451,
      "loss": 6.2004,
      "mean_token_accuracy": 0.14749388620257378,
      "num_tokens": 12917187.0,
      "step": 6590
    },
    {
      "entropy": 6.269068622589112,
      "epoch": 0.6592692557604838,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004964663496655083,
      "loss": 6.1659,
      "mean_token_accuracy": 0.15150395929813384,
      "num_tokens": 12926470.0,
      "step": 6595
    },
    {
      "entropy": 6.281428384780884,
      "epoch": 0.6597690808217124,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004964600466082351,
      "loss": 6.2101,
      "mean_token_accuracy": 0.15038524940609932,
      "num_tokens": 12936609.0,
      "step": 6600
    },
    {
      "entropy": 6.308905792236328,
      "epoch": 0.660268905882941,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004964537379790842,
      "loss": 6.1395,
      "mean_token_accuracy": 0.14830734729766845,
      "num_tokens": 12946286.0,
      "step": 6605
    },
    {
      "entropy": 6.3056737899780275,
      "epoch": 0.6607687309441695,
      "grad_norm": 1.125,
      "learning_rate": 0.0004964474237782141,
      "loss": 6.1696,
      "mean_token_accuracy": 0.14814531356096267,
      "num_tokens": 12955881.0,
      "step": 6610
    },
    {
      "entropy": 6.335570669174194,
      "epoch": 0.6612685560053981,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004964411040057839,
      "loss": 6.3239,
      "mean_token_accuracy": 0.1406143069267273,
      "num_tokens": 12966273.0,
      "step": 6615
    },
    {
      "entropy": 6.298713064193725,
      "epoch": 0.6617683810666267,
      "grad_norm": 1.25,
      "learning_rate": 0.0004964347786619524,
      "loss": 6.2865,
      "mean_token_accuracy": 0.1392020493745804,
      "num_tokens": 12975300.0,
      "step": 6620
    },
    {
      "entropy": 6.374906253814697,
      "epoch": 0.6622682061278552,
      "grad_norm": 1.125,
      "learning_rate": 0.0004964284477468791,
      "loss": 6.2201,
      "mean_token_accuracy": 0.1534460574388504,
      "num_tokens": 12985291.0,
      "step": 6625
    },
    {
      "entropy": 6.337492036819458,
      "epoch": 0.6627680311890838,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004964221112607228,
      "loss": 6.1718,
      "mean_token_accuracy": 0.15107849091291428,
      "num_tokens": 12995935.0,
      "step": 6630
    },
    {
      "entropy": 6.318701982498169,
      "epoch": 0.6632678562503124,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004964157692036432,
      "loss": 6.1557,
      "mean_token_accuracy": 0.15307274758815764,
      "num_tokens": 13006916.0,
      "step": 6635
    },
    {
      "entropy": 6.4351232051849365,
      "epoch": 0.663767681311541,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004964094215757998,
      "loss": 6.4381,
      "mean_token_accuracy": 0.1288245812058449,
      "num_tokens": 13017247.0,
      "step": 6640
    },
    {
      "entropy": 6.3791268825531,
      "epoch": 0.6642675063727695,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004964030683773525,
      "loss": 6.2488,
      "mean_token_accuracy": 0.14469776675105095,
      "num_tokens": 13026038.0,
      "step": 6645
    },
    {
      "entropy": 6.359122467041016,
      "epoch": 0.6647673314339981,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004963967096084608,
      "loss": 6.3057,
      "mean_token_accuracy": 0.13457366079092026,
      "num_tokens": 13036436.0,
      "step": 6650
    },
    {
      "entropy": 6.398467350006103,
      "epoch": 0.6652671564952267,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004963903452692849,
      "loss": 6.2582,
      "mean_token_accuracy": 0.14259015768766403,
      "num_tokens": 13045594.0,
      "step": 6655
    },
    {
      "entropy": 6.390234756469726,
      "epoch": 0.6657669815564552,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004963839753599848,
      "loss": 6.2389,
      "mean_token_accuracy": 0.1451947383582592,
      "num_tokens": 13056023.0,
      "step": 6660
    },
    {
      "entropy": 6.3786498546600345,
      "epoch": 0.6662668066176838,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000496377599880721,
      "loss": 6.2006,
      "mean_token_accuracy": 0.14860118329524993,
      "num_tokens": 13064230.0,
      "step": 6665
    },
    {
      "entropy": 6.281377363204956,
      "epoch": 0.6667666316789124,
      "grad_norm": 2.125,
      "learning_rate": 0.0004963712188316537,
      "loss": 6.1232,
      "mean_token_accuracy": 0.1533061981201172,
      "num_tokens": 13073767.0,
      "step": 6670
    },
    {
      "entropy": 6.250968360900879,
      "epoch": 0.667266456740141,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004963648322129435,
      "loss": 6.1522,
      "mean_token_accuracy": 0.1492617003619671,
      "num_tokens": 13083026.0,
      "step": 6675
    },
    {
      "entropy": 6.3436332702636715,
      "epoch": 0.6677662818013695,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000496358440024751,
      "loss": 6.1742,
      "mean_token_accuracy": 0.14968721717596054,
      "num_tokens": 13092029.0,
      "step": 6680
    },
    {
      "entropy": 6.4132812976837155,
      "epoch": 0.6682661068625981,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004963520422672372,
      "loss": 6.2994,
      "mean_token_accuracy": 0.13909538462758064,
      "num_tokens": 13101859.0,
      "step": 6685
    },
    {
      "entropy": 6.348860025405884,
      "epoch": 0.6687659319238266,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000496345638940563,
      "loss": 6.1589,
      "mean_token_accuracy": 0.15023377239704133,
      "num_tokens": 13111007.0,
      "step": 6690
    },
    {
      "entropy": 6.2601402759552,
      "epoch": 0.6692657569850552,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004963392300448894,
      "loss": 6.2012,
      "mean_token_accuracy": 0.14847960770130159,
      "num_tokens": 13120662.0,
      "step": 6695
    },
    {
      "entropy": 6.292029523849488,
      "epoch": 0.6697655820462838,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004963328155803778,
      "loss": 6.1106,
      "mean_token_accuracy": 0.14400700628757476,
      "num_tokens": 13130263.0,
      "step": 6700
    },
    {
      "entropy": 6.333777236938476,
      "epoch": 0.6702654071075124,
      "grad_norm": 1.125,
      "learning_rate": 0.0004963263955471895,
      "loss": 6.1601,
      "mean_token_accuracy": 0.14590639770030975,
      "num_tokens": 13139821.0,
      "step": 6705
    },
    {
      "entropy": 6.2604982376098635,
      "epoch": 0.670765232168741,
      "grad_norm": 1.0625,
      "learning_rate": 0.000496319969945486,
      "loss": 6.1623,
      "mean_token_accuracy": 0.14811343029141427,
      "num_tokens": 13149731.0,
      "step": 6710
    },
    {
      "entropy": 6.306476831436157,
      "epoch": 0.6712650572299695,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000496313538775429,
      "loss": 6.2384,
      "mean_token_accuracy": 0.14828030839562417,
      "num_tokens": 13160689.0,
      "step": 6715
    },
    {
      "entropy": 6.310291719436646,
      "epoch": 0.6717648822911981,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004963071020371803,
      "loss": 6.3524,
      "mean_token_accuracy": 0.14713722988963127,
      "num_tokens": 13170694.0,
      "step": 6720
    },
    {
      "entropy": 6.416176795959473,
      "epoch": 0.6722647073524266,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004963006597309019,
      "loss": 6.3185,
      "mean_token_accuracy": 0.14080383405089378,
      "num_tokens": 13180514.0,
      "step": 6725
    },
    {
      "entropy": 6.373733186721802,
      "epoch": 0.6727645324136552,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004962942118567559,
      "loss": 6.1337,
      "mean_token_accuracy": 0.14855736643075942,
      "num_tokens": 13190144.0,
      "step": 6730
    },
    {
      "entropy": 6.348651552200318,
      "epoch": 0.6732643574748838,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004962877584149043,
      "loss": 6.1883,
      "mean_token_accuracy": 0.14105819463729857,
      "num_tokens": 13199640.0,
      "step": 6735
    },
    {
      "entropy": 6.2989074230194095,
      "epoch": 0.6737641825361124,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004962812994055098,
      "loss": 6.1397,
      "mean_token_accuracy": 0.15071182548999787,
      "num_tokens": 13208824.0,
      "step": 6740
    },
    {
      "entropy": 6.242956066131592,
      "epoch": 0.674264007597341,
      "grad_norm": 1.125,
      "learning_rate": 0.0004962748348287347,
      "loss": 6.0822,
      "mean_token_accuracy": 0.15880692675709723,
      "num_tokens": 13219305.0,
      "step": 6745
    },
    {
      "entropy": 6.278830528259277,
      "epoch": 0.6747638326585695,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004962683646847416,
      "loss": 6.1099,
      "mean_token_accuracy": 0.15298435240983962,
      "num_tokens": 13227727.0,
      "step": 6750
    },
    {
      "entropy": 6.290111017227173,
      "epoch": 0.675263657719798,
      "grad_norm": 1.125,
      "learning_rate": 0.0004962618889736936,
      "loss": 6.1484,
      "mean_token_accuracy": 0.14888853952288628,
      "num_tokens": 13237522.0,
      "step": 6755
    },
    {
      "entropy": 6.292743110656739,
      "epoch": 0.6757634827810266,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004962554076957533,
      "loss": 6.2395,
      "mean_token_accuracy": 0.14217953085899354,
      "num_tokens": 13247155.0,
      "step": 6760
    },
    {
      "entropy": 6.317713928222656,
      "epoch": 0.6762633078422552,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004962489208510838,
      "loss": 6.277,
      "mean_token_accuracy": 0.1422378495335579,
      "num_tokens": 13257324.0,
      "step": 6765
    },
    {
      "entropy": 6.34179368019104,
      "epoch": 0.6767631329034838,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004962424284398484,
      "loss": 6.2366,
      "mean_token_accuracy": 0.1413424715399742,
      "num_tokens": 13267025.0,
      "step": 6770
    },
    {
      "entropy": 6.381427860260009,
      "epoch": 0.6772629579647124,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004962359304622105,
      "loss": 6.1584,
      "mean_token_accuracy": 0.1551748812198639,
      "num_tokens": 13276709.0,
      "step": 6775
    },
    {
      "entropy": 6.292602157592773,
      "epoch": 0.677762783025941,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004962294269183335,
      "loss": 6.176,
      "mean_token_accuracy": 0.1427864134311676,
      "num_tokens": 13286029.0,
      "step": 6780
    },
    {
      "entropy": 6.323148965835571,
      "epoch": 0.6782626080871695,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000496222917808381,
      "loss": 6.3315,
      "mean_token_accuracy": 0.1304969921708107,
      "num_tokens": 13296367.0,
      "step": 6785
    },
    {
      "entropy": 6.383178281784057,
      "epoch": 0.678762433148398,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004962164031325169,
      "loss": 6.2466,
      "mean_token_accuracy": 0.14183300584554673,
      "num_tokens": 13305186.0,
      "step": 6790
    },
    {
      "entropy": 6.325276565551758,
      "epoch": 0.6792622582096266,
      "grad_norm": 1.125,
      "learning_rate": 0.000496209882890905,
      "loss": 6.1724,
      "mean_token_accuracy": 0.15078270733356475,
      "num_tokens": 13314512.0,
      "step": 6795
    },
    {
      "entropy": 6.328080129623413,
      "epoch": 0.6797620832708552,
      "grad_norm": 1.125,
      "learning_rate": 0.0004962033570837094,
      "loss": 6.277,
      "mean_token_accuracy": 0.14436183124780655,
      "num_tokens": 13324423.0,
      "step": 6800
    },
    {
      "entropy": 6.381993198394776,
      "epoch": 0.6802619083320838,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004961968257110941,
      "loss": 6.2777,
      "mean_token_accuracy": 0.1394027754664421,
      "num_tokens": 13334133.0,
      "step": 6805
    },
    {
      "entropy": 6.434943199157715,
      "epoch": 0.6807617333933124,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004961902887732236,
      "loss": 6.2922,
      "mean_token_accuracy": 0.143890181183815,
      "num_tokens": 13344017.0,
      "step": 6810
    },
    {
      "entropy": 6.188187932968139,
      "epoch": 0.681261558454541,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004961837462702626,
      "loss": 6.0194,
      "mean_token_accuracy": 0.15905791223049165,
      "num_tokens": 13353922.0,
      "step": 6815
    },
    {
      "entropy": 6.2807999610900875,
      "epoch": 0.6817613835157694,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004961771982023752,
      "loss": 6.1935,
      "mean_token_accuracy": 0.1435384839773178,
      "num_tokens": 13363890.0,
      "step": 6820
    },
    {
      "entropy": 6.343155002593994,
      "epoch": 0.682261208576998,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004961706445697265,
      "loss": 6.1479,
      "mean_token_accuracy": 0.14757106751203536,
      "num_tokens": 13373726.0,
      "step": 6825
    },
    {
      "entropy": 6.237061548233032,
      "epoch": 0.6827610336382266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004961640853724813,
      "loss": 6.0538,
      "mean_token_accuracy": 0.15523901879787444,
      "num_tokens": 13382049.0,
      "step": 6830
    },
    {
      "entropy": 6.116995525360108,
      "epoch": 0.6832608586994552,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004961575206108046,
      "loss": 6.0182,
      "mean_token_accuracy": 0.16070071011781692,
      "num_tokens": 13390977.0,
      "step": 6835
    },
    {
      "entropy": 6.365022993087768,
      "epoch": 0.6837606837606838,
      "grad_norm": 1.125,
      "learning_rate": 0.0004961509502848616,
      "loss": 6.2325,
      "mean_token_accuracy": 0.1436351679265499,
      "num_tokens": 13400165.0,
      "step": 6840
    },
    {
      "entropy": 6.384309959411621,
      "epoch": 0.6842605088219124,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004961443743948177,
      "loss": 6.2429,
      "mean_token_accuracy": 0.13668796122074128,
      "num_tokens": 13410201.0,
      "step": 6845
    },
    {
      "entropy": 6.300755643844605,
      "epoch": 0.684760333883141,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000496137792940838,
      "loss": 6.2263,
      "mean_token_accuracy": 0.1500900663435459,
      "num_tokens": 13421074.0,
      "step": 6850
    },
    {
      "entropy": 6.235409450531006,
      "epoch": 0.6852601589443694,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004961312059230885,
      "loss": 6.0503,
      "mean_token_accuracy": 0.15168904811143874,
      "num_tokens": 13430387.0,
      "step": 6855
    },
    {
      "entropy": 6.212957859039307,
      "epoch": 0.685759984005598,
      "grad_norm": 1.125,
      "learning_rate": 0.0004961246133417348,
      "loss": 6.1236,
      "mean_token_accuracy": 0.14864686354994774,
      "num_tokens": 13439497.0,
      "step": 6860
    },
    {
      "entropy": 6.412577104568482,
      "epoch": 0.6862598090668266,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004961180151969427,
      "loss": 6.268,
      "mean_token_accuracy": 0.13986073359847068,
      "num_tokens": 13450543.0,
      "step": 6865
    },
    {
      "entropy": 6.327650022506714,
      "epoch": 0.6867596341280552,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004961114114888782,
      "loss": 6.1784,
      "mean_token_accuracy": 0.15228293985128402,
      "num_tokens": 13459996.0,
      "step": 6870
    },
    {
      "entropy": 6.24297571182251,
      "epoch": 0.6872594591892838,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004961048022177074,
      "loss": 6.1411,
      "mean_token_accuracy": 0.14923641160130502,
      "num_tokens": 13469141.0,
      "step": 6875
    },
    {
      "entropy": 6.25680832862854,
      "epoch": 0.6877592842505124,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004960981873835968,
      "loss": 6.1138,
      "mean_token_accuracy": 0.15531359612941742,
      "num_tokens": 13478349.0,
      "step": 6880
    },
    {
      "entropy": 6.263523292541504,
      "epoch": 0.6882591093117408,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004960915669867126,
      "loss": 6.0695,
      "mean_token_accuracy": 0.14980947524309157,
      "num_tokens": 13487417.0,
      "step": 6885
    },
    {
      "entropy": 6.428225326538086,
      "epoch": 0.6887589343729694,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004960849410272216,
      "loss": 6.3289,
      "mean_token_accuracy": 0.13821690827608107,
      "num_tokens": 13497607.0,
      "step": 6890
    },
    {
      "entropy": 6.345714426040649,
      "epoch": 0.689258759434198,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004960783095052903,
      "loss": 6.2842,
      "mean_token_accuracy": 0.13931539580225943,
      "num_tokens": 13508222.0,
      "step": 6895
    },
    {
      "entropy": 6.252889919281006,
      "epoch": 0.6897585844954266,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004960716724210856,
      "loss": 6.1002,
      "mean_token_accuracy": 0.15268631428480148,
      "num_tokens": 13519593.0,
      "step": 6900
    },
    {
      "entropy": 6.394395637512207,
      "epoch": 0.6902584095566552,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004960650297747746,
      "loss": 6.2848,
      "mean_token_accuracy": 0.13542054072022439,
      "num_tokens": 13531089.0,
      "step": 6905
    },
    {
      "entropy": 6.3698523998260494,
      "epoch": 0.6907582346178838,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004960583815665242,
      "loss": 6.1957,
      "mean_token_accuracy": 0.14838260486721994,
      "num_tokens": 13539897.0,
      "step": 6910
    },
    {
      "entropy": 6.326138734817505,
      "epoch": 0.6912580596791124,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004960517277965018,
      "loss": 6.2718,
      "mean_token_accuracy": 0.13728462755680085,
      "num_tokens": 13549986.0,
      "step": 6915
    },
    {
      "entropy": 6.34544038772583,
      "epoch": 0.6917578847403408,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004960450684648749,
      "loss": 6.2824,
      "mean_token_accuracy": 0.14203501790761947,
      "num_tokens": 13559866.0,
      "step": 6920
    },
    {
      "entropy": 6.441147565841675,
      "epoch": 0.6922577098015694,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000496038403571811,
      "loss": 6.2884,
      "mean_token_accuracy": 0.1372080184519291,
      "num_tokens": 13569789.0,
      "step": 6925
    },
    {
      "entropy": 6.36728253364563,
      "epoch": 0.692757534862798,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004960317331174777,
      "loss": 6.2729,
      "mean_token_accuracy": 0.14114239066839218,
      "num_tokens": 13579121.0,
      "step": 6930
    },
    {
      "entropy": 6.28985276222229,
      "epoch": 0.6932573599240266,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004960250571020428,
      "loss": 6.165,
      "mean_token_accuracy": 0.1482217974960804,
      "num_tokens": 13589281.0,
      "step": 6935
    },
    {
      "entropy": 6.409862375259399,
      "epoch": 0.6937571849852552,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004960183755256744,
      "loss": 6.176,
      "mean_token_accuracy": 0.1478048637509346,
      "num_tokens": 13598312.0,
      "step": 6940
    },
    {
      "entropy": 6.31600604057312,
      "epoch": 0.6942570100464838,
      "grad_norm": 1.125,
      "learning_rate": 0.0004960116883885406,
      "loss": 6.2094,
      "mean_token_accuracy": 0.14780552238225936,
      "num_tokens": 13608164.0,
      "step": 6945
    },
    {
      "entropy": 6.27331166267395,
      "epoch": 0.6947568351077124,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004960049956908095,
      "loss": 6.2346,
      "mean_token_accuracy": 0.14084442257881163,
      "num_tokens": 13618314.0,
      "step": 6950
    },
    {
      "entropy": 6.363490152359009,
      "epoch": 0.6952566601689408,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004959982974326496,
      "loss": 6.2063,
      "mean_token_accuracy": 0.14357577860355378,
      "num_tokens": 13628846.0,
      "step": 6955
    },
    {
      "entropy": 6.307374382019043,
      "epoch": 0.6957564852301694,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004959915936142294,
      "loss": 6.1006,
      "mean_token_accuracy": 0.15307686775922774,
      "num_tokens": 13637941.0,
      "step": 6960
    },
    {
      "entropy": 6.300097370147705,
      "epoch": 0.696256310291398,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004959848842357175,
      "loss": 6.1746,
      "mean_token_accuracy": 0.1434026300907135,
      "num_tokens": 13647734.0,
      "step": 6965
    },
    {
      "entropy": 6.243784666061401,
      "epoch": 0.6967561353526266,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004959781692972829,
      "loss": 6.1565,
      "mean_token_accuracy": 0.14974140226840973,
      "num_tokens": 13657929.0,
      "step": 6970
    },
    {
      "entropy": 6.329504489898682,
      "epoch": 0.6972559604138552,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004959714487990943,
      "loss": 6.232,
      "mean_token_accuracy": 0.14123856276273727,
      "num_tokens": 13668258.0,
      "step": 6975
    },
    {
      "entropy": 6.3556571960449215,
      "epoch": 0.6977557854750838,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000495964722741321,
      "loss": 6.1818,
      "mean_token_accuracy": 0.14996417313814164,
      "num_tokens": 13677515.0,
      "step": 6980
    },
    {
      "entropy": 6.263739490509034,
      "epoch": 0.6982556105363122,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004959579911241322,
      "loss": 6.1911,
      "mean_token_accuracy": 0.14708277881145476,
      "num_tokens": 13686865.0,
      "step": 6985
    },
    {
      "entropy": 6.330478239059448,
      "epoch": 0.6987554355975408,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004959512539476971,
      "loss": 6.2855,
      "mean_token_accuracy": 0.14340697452425957,
      "num_tokens": 13696683.0,
      "step": 6990
    },
    {
      "entropy": 6.360787630081177,
      "epoch": 0.6992552606587694,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004959445112121853,
      "loss": 6.2491,
      "mean_token_accuracy": 0.13689724430441857,
      "num_tokens": 13706452.0,
      "step": 6995
    },
    {
      "entropy": 6.313475131988525,
      "epoch": 0.699755085719998,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004959377629177665,
      "loss": 6.1317,
      "mean_token_accuracy": 0.14988905787467957,
      "num_tokens": 13717474.0,
      "step": 7000
    },
    {
      "entropy": 6.216355895996093,
      "epoch": 0.7002549107812266,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004959310090646104,
      "loss": 6.1312,
      "mean_token_accuracy": 0.15386501997709273,
      "num_tokens": 13726723.0,
      "step": 7005
    },
    {
      "entropy": 6.41074857711792,
      "epoch": 0.7007547358424552,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004959242496528869,
      "loss": 6.3625,
      "mean_token_accuracy": 0.13916202187538146,
      "num_tokens": 13736432.0,
      "step": 7010
    },
    {
      "entropy": 6.4100022315979,
      "epoch": 0.7012545609036838,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004959174846827663,
      "loss": 6.1895,
      "mean_token_accuracy": 0.14562004953622817,
      "num_tokens": 13745767.0,
      "step": 7015
    },
    {
      "entropy": 6.345577669143677,
      "epoch": 0.7017543859649122,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004959107141544186,
      "loss": 6.2045,
      "mean_token_accuracy": 0.1495068609714508,
      "num_tokens": 13756035.0,
      "step": 7020
    },
    {
      "entropy": 6.196120643615723,
      "epoch": 0.7022542110261408,
      "grad_norm": 1.203125,
      "learning_rate": 0.000495903938068014,
      "loss": 6.0331,
      "mean_token_accuracy": 0.15444229617714883,
      "num_tokens": 13765715.0,
      "step": 7025
    },
    {
      "entropy": 6.2470938205719,
      "epoch": 0.7027540360873694,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004958971564237233,
      "loss": 6.212,
      "mean_token_accuracy": 0.14291437342762947,
      "num_tokens": 13775987.0,
      "step": 7030
    },
    {
      "entropy": 6.340149593353272,
      "epoch": 0.703253861148598,
      "grad_norm": 1.109375,
      "learning_rate": 0.000495890369221717,
      "loss": 6.137,
      "mean_token_accuracy": 0.14881762415170668,
      "num_tokens": 13785095.0,
      "step": 7035
    },
    {
      "entropy": 6.318028736114502,
      "epoch": 0.7037536862098266,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004958835764621657,
      "loss": 6.1661,
      "mean_token_accuracy": 0.1419317230582237,
      "num_tokens": 13794777.0,
      "step": 7040
    },
    {
      "entropy": 6.163327312469482,
      "epoch": 0.7042535112710552,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004958767781452406,
      "loss": 6.1036,
      "mean_token_accuracy": 0.14152265340089798,
      "num_tokens": 13803534.0,
      "step": 7045
    },
    {
      "entropy": 6.291766405105591,
      "epoch": 0.7047533363322837,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004958699742711126,
      "loss": 6.185,
      "mean_token_accuracy": 0.14469611048698425,
      "num_tokens": 13813941.0,
      "step": 7050
    },
    {
      "entropy": 6.308145809173584,
      "epoch": 0.7052531613935122,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004958631648399528,
      "loss": 6.177,
      "mean_token_accuracy": 0.14617753773927689,
      "num_tokens": 13823611.0,
      "step": 7055
    },
    {
      "entropy": 6.379688358306884,
      "epoch": 0.7057529864547408,
      "grad_norm": 1.125,
      "learning_rate": 0.0004958563498519327,
      "loss": 6.3045,
      "mean_token_accuracy": 0.13714892491698266,
      "num_tokens": 13833047.0,
      "step": 7060
    },
    {
      "entropy": 6.30211386680603,
      "epoch": 0.7062528115159694,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004958495293072235,
      "loss": 6.2061,
      "mean_token_accuracy": 0.1473926693201065,
      "num_tokens": 13843524.0,
      "step": 7065
    },
    {
      "entropy": 6.266264200210571,
      "epoch": 0.706752636577198,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004958427032059971,
      "loss": 6.095,
      "mean_token_accuracy": 0.1520598918199539,
      "num_tokens": 13852188.0,
      "step": 7070
    },
    {
      "entropy": 6.252702331542968,
      "epoch": 0.7072524616384266,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004958358715484251,
      "loss": 6.221,
      "mean_token_accuracy": 0.1437772773206234,
      "num_tokens": 13863540.0,
      "step": 7075
    },
    {
      "entropy": 6.38595814704895,
      "epoch": 0.7077522866996552,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004958290343346795,
      "loss": 6.2027,
      "mean_token_accuracy": 0.14450544491410255,
      "num_tokens": 13871828.0,
      "step": 7080
    },
    {
      "entropy": 6.352229070663452,
      "epoch": 0.7082521117608837,
      "grad_norm": 1.09375,
      "learning_rate": 0.000495822191564932,
      "loss": 6.2398,
      "mean_token_accuracy": 0.145554868131876,
      "num_tokens": 13882222.0,
      "step": 7085
    },
    {
      "entropy": 6.304844808578491,
      "epoch": 0.7087519368221122,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004958153432393553,
      "loss": 6.2782,
      "mean_token_accuracy": 0.1416569881141186,
      "num_tokens": 13891689.0,
      "step": 7090
    },
    {
      "entropy": 6.311014413833618,
      "epoch": 0.7092517618833408,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000495808489358121,
      "loss": 6.1711,
      "mean_token_accuracy": 0.14667459279298783,
      "num_tokens": 13901310.0,
      "step": 7095
    },
    {
      "entropy": 6.30207781791687,
      "epoch": 0.7097515869445694,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004958016299214022,
      "loss": 6.1536,
      "mean_token_accuracy": 0.14314044937491416,
      "num_tokens": 13910350.0,
      "step": 7100
    },
    {
      "entropy": 6.296815967559814,
      "epoch": 0.710251412005798,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004957947649293712,
      "loss": 6.2033,
      "mean_token_accuracy": 0.14538471177220344,
      "num_tokens": 13920481.0,
      "step": 7105
    },
    {
      "entropy": 6.389341545104981,
      "epoch": 0.7107512370670266,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004957878943822006,
      "loss": 6.2034,
      "mean_token_accuracy": 0.14248755797743798,
      "num_tokens": 13931232.0,
      "step": 7110
    },
    {
      "entropy": 6.313736057281494,
      "epoch": 0.7112510621282551,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004957810182800634,
      "loss": 6.2767,
      "mean_token_accuracy": 0.13966492265462876,
      "num_tokens": 13940717.0,
      "step": 7115
    },
    {
      "entropy": 6.319150066375732,
      "epoch": 0.7117508871894836,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004957741366231327,
      "loss": 6.1102,
      "mean_token_accuracy": 0.1487189218401909,
      "num_tokens": 13949918.0,
      "step": 7120
    },
    {
      "entropy": 6.300774669647216,
      "epoch": 0.7122507122507122,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004957672494115815,
      "loss": 6.1799,
      "mean_token_accuracy": 0.1476925477385521,
      "num_tokens": 13959914.0,
      "step": 7125
    },
    {
      "entropy": 6.277954912185669,
      "epoch": 0.7127505373119408,
      "grad_norm": 1.171875,
      "learning_rate": 0.000495760356645583,
      "loss": 6.1635,
      "mean_token_accuracy": 0.1487579256296158,
      "num_tokens": 13970054.0,
      "step": 7130
    },
    {
      "entropy": 6.17656192779541,
      "epoch": 0.7132503623731694,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004957534583253108,
      "loss": 6.0769,
      "mean_token_accuracy": 0.15268507823348046,
      "num_tokens": 13980339.0,
      "step": 7135
    },
    {
      "entropy": 6.2607893466949465,
      "epoch": 0.713750187434398,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004957465544509384,
      "loss": 6.1643,
      "mean_token_accuracy": 0.14543780237436293,
      "num_tokens": 13990466.0,
      "step": 7140
    },
    {
      "entropy": 6.336538314819336,
      "epoch": 0.7142500124956266,
      "grad_norm": 1.125,
      "learning_rate": 0.0004957396450226395,
      "loss": 6.196,
      "mean_token_accuracy": 0.1411573812365532,
      "num_tokens": 13999822.0,
      "step": 7145
    },
    {
      "entropy": 6.399840545654297,
      "epoch": 0.7147498375568551,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004957327300405879,
      "loss": 6.2216,
      "mean_token_accuracy": 0.14402197003364564,
      "num_tokens": 14008939.0,
      "step": 7150
    },
    {
      "entropy": 6.2654749870300295,
      "epoch": 0.7152496626180836,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004957258095049577,
      "loss": 6.1821,
      "mean_token_accuracy": 0.13896486908197403,
      "num_tokens": 14019005.0,
      "step": 7155
    },
    {
      "entropy": 6.248404502868652,
      "epoch": 0.7157494876793122,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004957188834159229,
      "loss": 6.0821,
      "mean_token_accuracy": 0.14790764898061753,
      "num_tokens": 14029390.0,
      "step": 7160
    },
    {
      "entropy": 6.238894414901734,
      "epoch": 0.7162493127405408,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004957119517736577,
      "loss": 6.1537,
      "mean_token_accuracy": 0.14542849808931352,
      "num_tokens": 14038613.0,
      "step": 7165
    },
    {
      "entropy": 6.238741779327393,
      "epoch": 0.7167491378017694,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004957050145783366,
      "loss": 6.1529,
      "mean_token_accuracy": 0.14703854992985727,
      "num_tokens": 14048144.0,
      "step": 7170
    },
    {
      "entropy": 6.176491832733154,
      "epoch": 0.717248962862998,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004956980718301342,
      "loss": 6.0549,
      "mean_token_accuracy": 0.15655977725982667,
      "num_tokens": 14056894.0,
      "step": 7175
    },
    {
      "entropy": 6.252416753768921,
      "epoch": 0.7177487879242265,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004956911235292251,
      "loss": 6.1767,
      "mean_token_accuracy": 0.14157176911830902,
      "num_tokens": 14066024.0,
      "step": 7180
    },
    {
      "entropy": 6.253168964385987,
      "epoch": 0.7182486129854551,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004956841696757841,
      "loss": 6.1669,
      "mean_token_accuracy": 0.14734007716178893,
      "num_tokens": 14075805.0,
      "step": 7185
    },
    {
      "entropy": 6.3037331104278564,
      "epoch": 0.7187484380466836,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004956772102699862,
      "loss": 6.1814,
      "mean_token_accuracy": 0.14717554822564125,
      "num_tokens": 14085229.0,
      "step": 7190
    },
    {
      "entropy": 6.307119417190552,
      "epoch": 0.7192482631079122,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004956702453120064,
      "loss": 6.1483,
      "mean_token_accuracy": 0.14640737995505332,
      "num_tokens": 14094594.0,
      "step": 7195
    },
    {
      "entropy": 6.2317572116851805,
      "epoch": 0.7197480881691408,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004956632748020201,
      "loss": 6.1915,
      "mean_token_accuracy": 0.15127859115600586,
      "num_tokens": 14104373.0,
      "step": 7200
    },
    {
      "entropy": 6.241860342025757,
      "epoch": 0.7202479132303694,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004956562987402026,
      "loss": 6.0959,
      "mean_token_accuracy": 0.15275246724486352,
      "num_tokens": 14113918.0,
      "step": 7205
    },
    {
      "entropy": 6.3966890335083,
      "epoch": 0.720747738291598,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004956493171267293,
      "loss": 6.3062,
      "mean_token_accuracy": 0.13498108386993407,
      "num_tokens": 14123711.0,
      "step": 7210
    },
    {
      "entropy": 6.340817165374756,
      "epoch": 0.7212475633528265,
      "grad_norm": 1.21875,
      "learning_rate": 0.000495642329961776,
      "loss": 6.0952,
      "mean_token_accuracy": 0.15226860120892524,
      "num_tokens": 14132994.0,
      "step": 7215
    },
    {
      "entropy": 6.244050168991089,
      "epoch": 0.7217473884140551,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004956353372455185,
      "loss": 6.1247,
      "mean_token_accuracy": 0.15045270025730134,
      "num_tokens": 14143398.0,
      "step": 7220
    },
    {
      "entropy": 6.325813484191895,
      "epoch": 0.7222472134752836,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004956283389781327,
      "loss": 6.1523,
      "mean_token_accuracy": 0.1496276319026947,
      "num_tokens": 14152374.0,
      "step": 7225
    },
    {
      "entropy": 6.3094688892364506,
      "epoch": 0.7227470385365122,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004956213351597946,
      "loss": 6.1147,
      "mean_token_accuracy": 0.14797163009643555,
      "num_tokens": 14162123.0,
      "step": 7230
    },
    {
      "entropy": 6.266576051712036,
      "epoch": 0.7232468635977408,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004956143257906805,
      "loss": 6.3052,
      "mean_token_accuracy": 0.14230608344078063,
      "num_tokens": 14172004.0,
      "step": 7235
    },
    {
      "entropy": 6.235523748397827,
      "epoch": 0.7237466886589694,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004956073108709667,
      "loss": 6.1612,
      "mean_token_accuracy": 0.14577551931142807,
      "num_tokens": 14181659.0,
      "step": 7240
    },
    {
      "entropy": 6.2913923263549805,
      "epoch": 0.7242465137201979,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004956002904008298,
      "loss": 6.1493,
      "mean_token_accuracy": 0.14673805311322213,
      "num_tokens": 14191051.0,
      "step": 7245
    },
    {
      "entropy": 6.378592824935913,
      "epoch": 0.7247463387814265,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004955932643804463,
      "loss": 6.2795,
      "mean_token_accuracy": 0.13885239958763124,
      "num_tokens": 14201232.0,
      "step": 7250
    },
    {
      "entropy": 6.295748853683472,
      "epoch": 0.7252461638426551,
      "grad_norm": 1.171875,
      "learning_rate": 0.000495586232809993,
      "loss": 6.1879,
      "mean_token_accuracy": 0.1442564234137535,
      "num_tokens": 14210426.0,
      "step": 7255
    },
    {
      "entropy": 6.18954496383667,
      "epoch": 0.7257459889038836,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004955791956896469,
      "loss": 6.0712,
      "mean_token_accuracy": 0.15404979959130288,
      "num_tokens": 14220013.0,
      "step": 7260
    },
    {
      "entropy": 6.299191188812256,
      "epoch": 0.7262458139651122,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000495572153019585,
      "loss": 6.2068,
      "mean_token_accuracy": 0.14496504366397858,
      "num_tokens": 14229249.0,
      "step": 7265
    },
    {
      "entropy": 6.3032938003540036,
      "epoch": 0.7267456390263408,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004955651047999844,
      "loss": 6.0932,
      "mean_token_accuracy": 0.15484636276960373,
      "num_tokens": 14238370.0,
      "step": 7270
    },
    {
      "entropy": 6.157650375366211,
      "epoch": 0.7272454640875694,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004955580510310226,
      "loss": 6.1384,
      "mean_token_accuracy": 0.1594982549548149,
      "num_tokens": 14248037.0,
      "step": 7275
    },
    {
      "entropy": 6.254744529724121,
      "epoch": 0.7277452891487979,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004955509917128771,
      "loss": 6.2037,
      "mean_token_accuracy": 0.1447198711335659,
      "num_tokens": 14257480.0,
      "step": 7280
    },
    {
      "entropy": 6.266466569900513,
      "epoch": 0.7282451142100265,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004955439268457252,
      "loss": 6.1555,
      "mean_token_accuracy": 0.14776824414730072,
      "num_tokens": 14266114.0,
      "step": 7285
    },
    {
      "entropy": 6.238157272338867,
      "epoch": 0.728744939271255,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004955368564297449,
      "loss": 6.1656,
      "mean_token_accuracy": 0.14812055081129075,
      "num_tokens": 14276048.0,
      "step": 7290
    },
    {
      "entropy": 6.278201103210449,
      "epoch": 0.7292447643324836,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004955297804651142,
      "loss": 6.0549,
      "mean_token_accuracy": 0.15845825672149658,
      "num_tokens": 14285225.0,
      "step": 7295
    },
    {
      "entropy": 6.357158946990967,
      "epoch": 0.7297445893937122,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004955226989520107,
      "loss": 6.1476,
      "mean_token_accuracy": 0.14691286832094191,
      "num_tokens": 14295566.0,
      "step": 7300
    },
    {
      "entropy": 6.250790977478028,
      "epoch": 0.7302444144549408,
      "grad_norm": 1.203125,
      "learning_rate": 0.000495515611890613,
      "loss": 6.131,
      "mean_token_accuracy": 0.14438453167676926,
      "num_tokens": 14305072.0,
      "step": 7305
    },
    {
      "entropy": 6.176940059661865,
      "epoch": 0.7307442395161693,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004955085192810993,
      "loss": 5.9813,
      "mean_token_accuracy": 0.1541096895933151,
      "num_tokens": 14315273.0,
      "step": 7310
    },
    {
      "entropy": 6.30296893119812,
      "epoch": 0.7312440645773979,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004955014211236479,
      "loss": 6.1806,
      "mean_token_accuracy": 0.14431563317775725,
      "num_tokens": 14324818.0,
      "step": 7315
    },
    {
      "entropy": 6.284177780151367,
      "epoch": 0.7317438896386265,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004954943174184375,
      "loss": 6.2594,
      "mean_token_accuracy": 0.14106540605425835,
      "num_tokens": 14335129.0,
      "step": 7320
    },
    {
      "entropy": 6.308487319946289,
      "epoch": 0.732243714699855,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004954872081656469,
      "loss": 6.2176,
      "mean_token_accuracy": 0.14290185272693634,
      "num_tokens": 14344339.0,
      "step": 7325
    },
    {
      "entropy": 6.310902547836304,
      "epoch": 0.7327435397610836,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004954800933654548,
      "loss": 6.1731,
      "mean_token_accuracy": 0.14204417243599893,
      "num_tokens": 14353391.0,
      "step": 7330
    },
    {
      "entropy": 6.442323112487793,
      "epoch": 0.7332433648223122,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004954729730180403,
      "loss": 6.2926,
      "mean_token_accuracy": 0.14274587482213974,
      "num_tokens": 14363684.0,
      "step": 7335
    },
    {
      "entropy": 6.333347463607788,
      "epoch": 0.7337431898835408,
      "grad_norm": 1.125,
      "learning_rate": 0.0004954658471235825,
      "loss": 6.177,
      "mean_token_accuracy": 0.14908413961529732,
      "num_tokens": 14372802.0,
      "step": 7340
    },
    {
      "entropy": 6.272087430953979,
      "epoch": 0.7342430149447693,
      "grad_norm": 1.125,
      "learning_rate": 0.0004954587156822607,
      "loss": 6.2013,
      "mean_token_accuracy": 0.14051852598786355,
      "num_tokens": 14382665.0,
      "step": 7345
    },
    {
      "entropy": 6.338935327529907,
      "epoch": 0.7347428400059979,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004954515786942544,
      "loss": 6.1465,
      "mean_token_accuracy": 0.14679207950830458,
      "num_tokens": 14393631.0,
      "step": 7350
    },
    {
      "entropy": 6.3339227676391605,
      "epoch": 0.7352426650672265,
      "grad_norm": 1.09375,
      "learning_rate": 0.000495444436159743,
      "loss": 6.1514,
      "mean_token_accuracy": 0.14726692587137222,
      "num_tokens": 14402713.0,
      "step": 7355
    },
    {
      "entropy": 6.425673151016236,
      "epoch": 0.735742490128455,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004954372880789064,
      "loss": 6.3085,
      "mean_token_accuracy": 0.13866232559084893,
      "num_tokens": 14412777.0,
      "step": 7360
    },
    {
      "entropy": 6.377609586715698,
      "epoch": 0.7362423151896836,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004954301344519244,
      "loss": 6.2439,
      "mean_token_accuracy": 0.14006474986672401,
      "num_tokens": 14422839.0,
      "step": 7365
    },
    {
      "entropy": 6.196596765518189,
      "epoch": 0.7367421402509122,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004954229752789769,
      "loss": 6.1244,
      "mean_token_accuracy": 0.14910946041345596,
      "num_tokens": 14433239.0,
      "step": 7370
    },
    {
      "entropy": 6.325506258010864,
      "epoch": 0.7372419653121407,
      "grad_norm": 1.125,
      "learning_rate": 0.0004954158105602441,
      "loss": 6.2263,
      "mean_token_accuracy": 0.1439547397196293,
      "num_tokens": 14442907.0,
      "step": 7375
    },
    {
      "entropy": 6.256664752960205,
      "epoch": 0.7377417903733693,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004954086402959061,
      "loss": 6.0655,
      "mean_token_accuracy": 0.15007940530776978,
      "num_tokens": 14452492.0,
      "step": 7380
    },
    {
      "entropy": 6.2603981494903564,
      "epoch": 0.7382416154345979,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004954014644861435,
      "loss": 6.2237,
      "mean_token_accuracy": 0.14188649654388427,
      "num_tokens": 14461353.0,
      "step": 7385
    },
    {
      "entropy": 6.287715482711792,
      "epoch": 0.7387414404958265,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004953942831311367,
      "loss": 6.0779,
      "mean_token_accuracy": 0.16012443453073502,
      "num_tokens": 14470579.0,
      "step": 7390
    },
    {
      "entropy": 6.2085224151611325,
      "epoch": 0.739241265557055,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004953870962310666,
      "loss": 6.0465,
      "mean_token_accuracy": 0.15338122248649597,
      "num_tokens": 14481608.0,
      "step": 7395
    },
    {
      "entropy": 6.247384452819825,
      "epoch": 0.7397410906182836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004953799037861138,
      "loss": 6.2335,
      "mean_token_accuracy": 0.14395986795425414,
      "num_tokens": 14493074.0,
      "step": 7400
    },
    {
      "entropy": 6.269499349594116,
      "epoch": 0.7402409156795122,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004953727057964594,
      "loss": 6.0747,
      "mean_token_accuracy": 0.15390085726976394,
      "num_tokens": 14502469.0,
      "step": 7405
    },
    {
      "entropy": 6.288875389099121,
      "epoch": 0.7407407407407407,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004953655022622843,
      "loss": 6.2252,
      "mean_token_accuracy": 0.1448018729686737,
      "num_tokens": 14512939.0,
      "step": 7410
    },
    {
      "entropy": 6.324220848083496,
      "epoch": 0.7412405658019693,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004953582931837699,
      "loss": 6.2175,
      "mean_token_accuracy": 0.1429981306195259,
      "num_tokens": 14522681.0,
      "step": 7415
    },
    {
      "entropy": 6.351051282882691,
      "epoch": 0.7417403908631979,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004953510785610976,
      "loss": 6.0608,
      "mean_token_accuracy": 0.15530810207128526,
      "num_tokens": 14532115.0,
      "step": 7420
    },
    {
      "entropy": 6.258867645263672,
      "epoch": 0.7422402159244265,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004953438583944487,
      "loss": 6.1585,
      "mean_token_accuracy": 0.14376655369997024,
      "num_tokens": 14541147.0,
      "step": 7425
    },
    {
      "entropy": 6.2483128070831295,
      "epoch": 0.742740040985655,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004953366326840052,
      "loss": 6.1921,
      "mean_token_accuracy": 0.14653110057115554,
      "num_tokens": 14550373.0,
      "step": 7430
    },
    {
      "entropy": 6.378145313262939,
      "epoch": 0.7432398660468836,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004953294014299486,
      "loss": 6.2247,
      "mean_token_accuracy": 0.1514709249138832,
      "num_tokens": 14559396.0,
      "step": 7435
    },
    {
      "entropy": 6.356553649902343,
      "epoch": 0.7437396911081121,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004953221646324611,
      "loss": 6.2671,
      "mean_token_accuracy": 0.14016986638307571,
      "num_tokens": 14570292.0,
      "step": 7440
    },
    {
      "entropy": 6.259710168838501,
      "epoch": 0.7442395161693407,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004953149222917245,
      "loss": 6.13,
      "mean_token_accuracy": 0.14948854744434356,
      "num_tokens": 14579887.0,
      "step": 7445
    },
    {
      "entropy": 6.337480258941651,
      "epoch": 0.7447393412305693,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004953076744079213,
      "loss": 6.234,
      "mean_token_accuracy": 0.14049667716026307,
      "num_tokens": 14590260.0,
      "step": 7450
    },
    {
      "entropy": 6.306977987289429,
      "epoch": 0.7452391662917979,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004953004209812335,
      "loss": 6.1508,
      "mean_token_accuracy": 0.14136431217193604,
      "num_tokens": 14599812.0,
      "step": 7455
    },
    {
      "entropy": 6.242241239547729,
      "epoch": 0.7457389913530265,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004952931620118439,
      "loss": 6.0903,
      "mean_token_accuracy": 0.15077660828828812,
      "num_tokens": 14609147.0,
      "step": 7460
    },
    {
      "entropy": 6.332976388931274,
      "epoch": 0.746238816414255,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004952858974999351,
      "loss": 6.1965,
      "mean_token_accuracy": 0.13714736998081206,
      "num_tokens": 14618834.0,
      "step": 7465
    },
    {
      "entropy": 6.248338842391968,
      "epoch": 0.7467386414754836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004952786274456896,
      "loss": 6.0811,
      "mean_token_accuracy": 0.15912527292966844,
      "num_tokens": 14628556.0,
      "step": 7470
    },
    {
      "entropy": 6.213376665115357,
      "epoch": 0.7472384665367121,
      "grad_norm": 1.125,
      "learning_rate": 0.0004952713518492907,
      "loss": 6.17,
      "mean_token_accuracy": 0.14860969707369803,
      "num_tokens": 14637486.0,
      "step": 7475
    },
    {
      "entropy": 6.2777327537536625,
      "epoch": 0.7477382915979407,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004952640707109213,
      "loss": 6.1443,
      "mean_token_accuracy": 0.14600803405046464,
      "num_tokens": 14647599.0,
      "step": 7480
    },
    {
      "entropy": 6.33806471824646,
      "epoch": 0.7482381166591693,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004952567840307644,
      "loss": 6.0405,
      "mean_token_accuracy": 0.15203938633203506,
      "num_tokens": 14656246.0,
      "step": 7485
    },
    {
      "entropy": 6.2583588600158695,
      "epoch": 0.7487379417203979,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004952494918090036,
      "loss": 6.1838,
      "mean_token_accuracy": 0.14016395732760428,
      "num_tokens": 14665550.0,
      "step": 7490
    },
    {
      "entropy": 6.325530242919922,
      "epoch": 0.7492377667816265,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004952421940458222,
      "loss": 6.2016,
      "mean_token_accuracy": 0.14500749185681344,
      "num_tokens": 14676462.0,
      "step": 7495
    },
    {
      "entropy": 6.265401411056518,
      "epoch": 0.749737591842855,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004952348907414039,
      "loss": 6.0449,
      "mean_token_accuracy": 0.15199752897024155,
      "num_tokens": 14685963.0,
      "step": 7500
    },
    {
      "entropy": 6.213968420028687,
      "epoch": 0.7502374169040835,
      "grad_norm": 1.125,
      "learning_rate": 0.0004952275818959325,
      "loss": 6.1765,
      "mean_token_accuracy": 0.14756314903497697,
      "num_tokens": 14695764.0,
      "step": 7505
    },
    {
      "entropy": 6.251408433914184,
      "epoch": 0.7507372419653121,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004952202675095916,
      "loss": 6.0977,
      "mean_token_accuracy": 0.15183492004871368,
      "num_tokens": 14705839.0,
      "step": 7510
    },
    {
      "entropy": 6.326344299316406,
      "epoch": 0.7512370670265407,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004952129475825657,
      "loss": 6.1068,
      "mean_token_accuracy": 0.151962049305439,
      "num_tokens": 14714801.0,
      "step": 7515
    },
    {
      "entropy": 6.30691614151001,
      "epoch": 0.7517368920877693,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004952056221150385,
      "loss": 6.2341,
      "mean_token_accuracy": 0.14641458168625832,
      "num_tokens": 14723803.0,
      "step": 7520
    },
    {
      "entropy": 6.238247394561768,
      "epoch": 0.7522367171489979,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004951982911071945,
      "loss": 6.1239,
      "mean_token_accuracy": 0.15238103419542312,
      "num_tokens": 14732598.0,
      "step": 7525
    },
    {
      "entropy": 6.314418935775757,
      "epoch": 0.7527365422102265,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004951909545592183,
      "loss": 6.1855,
      "mean_token_accuracy": 0.15153319537639617,
      "num_tokens": 14742598.0,
      "step": 7530
    },
    {
      "entropy": 6.347425937652588,
      "epoch": 0.753236367271455,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004951836124712942,
      "loss": 6.2083,
      "mean_token_accuracy": 0.15113184005022048,
      "num_tokens": 14751434.0,
      "step": 7535
    },
    {
      "entropy": 6.3458695888519285,
      "epoch": 0.7537361923326835,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004951762648436073,
      "loss": 6.1848,
      "mean_token_accuracy": 0.14759642481803895,
      "num_tokens": 14760661.0,
      "step": 7540
    },
    {
      "entropy": 6.318964338302612,
      "epoch": 0.7542360173939121,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004951689116763419,
      "loss": 6.1807,
      "mean_token_accuracy": 0.14175293669104577,
      "num_tokens": 14770431.0,
      "step": 7545
    },
    {
      "entropy": 6.296151113510132,
      "epoch": 0.7547358424551407,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004951615529696835,
      "loss": 6.196,
      "mean_token_accuracy": 0.1463593527674675,
      "num_tokens": 14780066.0,
      "step": 7550
    },
    {
      "entropy": 6.335024356842041,
      "epoch": 0.7552356675163693,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004951541887238172,
      "loss": 6.1982,
      "mean_token_accuracy": 0.14942315295338632,
      "num_tokens": 14789531.0,
      "step": 7555
    },
    {
      "entropy": 6.292646598815918,
      "epoch": 0.7557354925775979,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004951468189389281,
      "loss": 6.1175,
      "mean_token_accuracy": 0.14930404722690582,
      "num_tokens": 14799000.0,
      "step": 7560
    },
    {
      "entropy": 6.280312585830688,
      "epoch": 0.7562353176388265,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004951394436152016,
      "loss": 6.1406,
      "mean_token_accuracy": 0.15315015166997908,
      "num_tokens": 14808991.0,
      "step": 7565
    },
    {
      "entropy": 6.271987867355347,
      "epoch": 0.7567351427000549,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004951320627528234,
      "loss": 6.1938,
      "mean_token_accuracy": 0.1499066837131977,
      "num_tokens": 14819351.0,
      "step": 7570
    },
    {
      "entropy": 6.237832498550415,
      "epoch": 0.7572349677612835,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004951246763519791,
      "loss": 6.2057,
      "mean_token_accuracy": 0.14232874810695648,
      "num_tokens": 14828924.0,
      "step": 7575
    },
    {
      "entropy": 6.476323652267456,
      "epoch": 0.7577347928225121,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004951172844128546,
      "loss": 6.3338,
      "mean_token_accuracy": 0.13136404305696486,
      "num_tokens": 14841077.0,
      "step": 7580
    },
    {
      "entropy": 6.446495771408081,
      "epoch": 0.7582346178837407,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004951098869356358,
      "loss": 6.2711,
      "mean_token_accuracy": 0.1362413927912712,
      "num_tokens": 14851223.0,
      "step": 7585
    },
    {
      "entropy": 6.319491720199585,
      "epoch": 0.7587344429449693,
      "grad_norm": 1.09375,
      "learning_rate": 0.000495102483920509,
      "loss": 6.1078,
      "mean_token_accuracy": 0.14920608401298524,
      "num_tokens": 14861003.0,
      "step": 7590
    },
    {
      "entropy": 6.2940943241119385,
      "epoch": 0.7592342680061979,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004950950753676603,
      "loss": 6.25,
      "mean_token_accuracy": 0.14146062582731248,
      "num_tokens": 14870207.0,
      "step": 7595
    },
    {
      "entropy": 6.228797149658203,
      "epoch": 0.7597340930674265,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004950876612772763,
      "loss": 6.1087,
      "mean_token_accuracy": 0.14785358682274818,
      "num_tokens": 14879886.0,
      "step": 7600
    },
    {
      "entropy": 6.254355764389038,
      "epoch": 0.7602339181286549,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004950802416495431,
      "loss": 6.0864,
      "mean_token_accuracy": 0.1529966726899147,
      "num_tokens": 14889275.0,
      "step": 7605
    },
    {
      "entropy": 6.269910621643066,
      "epoch": 0.7607337431898835,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004950728164846478,
      "loss": 6.2113,
      "mean_token_accuracy": 0.1473403736948967,
      "num_tokens": 14898317.0,
      "step": 7610
    },
    {
      "entropy": 6.259096956253051,
      "epoch": 0.7612335682511121,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004950653857827771,
      "loss": 6.1338,
      "mean_token_accuracy": 0.15169525146484375,
      "num_tokens": 14907183.0,
      "step": 7615
    },
    {
      "entropy": 6.282602405548095,
      "epoch": 0.7617333933123407,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004950579495441177,
      "loss": 6.2957,
      "mean_token_accuracy": 0.14033146500587462,
      "num_tokens": 14917564.0,
      "step": 7620
    },
    {
      "entropy": 6.333585977554321,
      "epoch": 0.7622332183735693,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004950505077688572,
      "loss": 6.0998,
      "mean_token_accuracy": 0.15521016865968704,
      "num_tokens": 14927102.0,
      "step": 7625
    },
    {
      "entropy": 6.123705720901489,
      "epoch": 0.7627330434347979,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004950430604571824,
      "loss": 5.9934,
      "mean_token_accuracy": 0.16457961052656173,
      "num_tokens": 14937934.0,
      "step": 7630
    },
    {
      "entropy": 6.236283016204834,
      "epoch": 0.7632328684960263,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004950356076092808,
      "loss": 6.17,
      "mean_token_accuracy": 0.14447291493415831,
      "num_tokens": 14948109.0,
      "step": 7635
    },
    {
      "entropy": 6.31685700416565,
      "epoch": 0.7637326935572549,
      "grad_norm": 1.15625,
      "learning_rate": 0.00049502814922534,
      "loss": 6.2032,
      "mean_token_accuracy": 0.14082704931497575,
      "num_tokens": 14956971.0,
      "step": 7640
    },
    {
      "entropy": 6.33694281578064,
      "epoch": 0.7642325186184835,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004950206853055475,
      "loss": 6.1833,
      "mean_token_accuracy": 0.1479064255952835,
      "num_tokens": 14967046.0,
      "step": 7645
    },
    {
      "entropy": 6.313223838806152,
      "epoch": 0.7647323436797121,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004950132158500913,
      "loss": 6.1489,
      "mean_token_accuracy": 0.14074713215231896,
      "num_tokens": 14976234.0,
      "step": 7650
    },
    {
      "entropy": 6.189839506149292,
      "epoch": 0.7652321687409407,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004950057408591591,
      "loss": 6.067,
      "mean_token_accuracy": 0.1599932625889778,
      "num_tokens": 14985205.0,
      "step": 7655
    },
    {
      "entropy": 6.201777648925781,
      "epoch": 0.7657319938021693,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004949982603329391,
      "loss": 6.1068,
      "mean_token_accuracy": 0.150619775056839,
      "num_tokens": 14994233.0,
      "step": 7660
    },
    {
      "entropy": 6.265900993347168,
      "epoch": 0.7662318188633979,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004949907742716195,
      "loss": 6.1457,
      "mean_token_accuracy": 0.1514419674873352,
      "num_tokens": 15004001.0,
      "step": 7665
    },
    {
      "entropy": 6.337007236480713,
      "epoch": 0.7667316439246263,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004949832826753886,
      "loss": 6.1708,
      "mean_token_accuracy": 0.13828388676047326,
      "num_tokens": 15015046.0,
      "step": 7670
    },
    {
      "entropy": 6.205091285705566,
      "epoch": 0.7672314689858549,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000494975785544435,
      "loss": 6.1044,
      "mean_token_accuracy": 0.1461218535900116,
      "num_tokens": 15025404.0,
      "step": 7675
    },
    {
      "entropy": 6.273244333267212,
      "epoch": 0.7677312940470835,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004949682828789472,
      "loss": 6.1968,
      "mean_token_accuracy": 0.14511665254831313,
      "num_tokens": 15036160.0,
      "step": 7680
    },
    {
      "entropy": 6.278796577453614,
      "epoch": 0.7682311191083121,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000494960774679114,
      "loss": 6.1166,
      "mean_token_accuracy": 0.1522207513451576,
      "num_tokens": 15046356.0,
      "step": 7685
    },
    {
      "entropy": 6.267112970352173,
      "epoch": 0.7687309441695407,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004949532609451242,
      "loss": 6.1243,
      "mean_token_accuracy": 0.1450452357530594,
      "num_tokens": 15056101.0,
      "step": 7690
    },
    {
      "entropy": 6.260371685028076,
      "epoch": 0.7692307692307693,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004949457416771671,
      "loss": 6.1122,
      "mean_token_accuracy": 0.15273965746164322,
      "num_tokens": 15065726.0,
      "step": 7695
    },
    {
      "entropy": 6.242496013641357,
      "epoch": 0.7697305942919977,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004949382168754316,
      "loss": 6.1174,
      "mean_token_accuracy": 0.15517046749591829,
      "num_tokens": 15075211.0,
      "step": 7700
    },
    {
      "entropy": 6.2484479427337645,
      "epoch": 0.7702304193532263,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004949306865401073,
      "loss": 6.0157,
      "mean_token_accuracy": 0.1555662900209427,
      "num_tokens": 15085525.0,
      "step": 7705
    },
    {
      "entropy": 6.308200979232788,
      "epoch": 0.7707302444144549,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004949231506713833,
      "loss": 6.2238,
      "mean_token_accuracy": 0.14330217540264129,
      "num_tokens": 15093851.0,
      "step": 7710
    },
    {
      "entropy": 6.424705839157104,
      "epoch": 0.7712300694756835,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004949156092694496,
      "loss": 6.2959,
      "mean_token_accuracy": 0.13891450315713882,
      "num_tokens": 15104066.0,
      "step": 7715
    },
    {
      "entropy": 6.3186499118804935,
      "epoch": 0.7717298945369121,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004949080623344958,
      "loss": 6.1305,
      "mean_token_accuracy": 0.15288395136594773,
      "num_tokens": 15114047.0,
      "step": 7720
    },
    {
      "entropy": 6.296549701690674,
      "epoch": 0.7722297195981407,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004949005098667117,
      "loss": 6.1403,
      "mean_token_accuracy": 0.15056335628032685,
      "num_tokens": 15123719.0,
      "step": 7725
    },
    {
      "entropy": 6.233033323287964,
      "epoch": 0.7727295446593693,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004948929518662873,
      "loss": 6.1803,
      "mean_token_accuracy": 0.15036975219845772,
      "num_tokens": 15133978.0,
      "step": 7730
    },
    {
      "entropy": 6.276612949371338,
      "epoch": 0.7732293697205977,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004948853883334129,
      "loss": 6.2206,
      "mean_token_accuracy": 0.1453232929110527,
      "num_tokens": 15145058.0,
      "step": 7735
    },
    {
      "entropy": 6.372824764251709,
      "epoch": 0.7737291947818263,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004948778192682787,
      "loss": 6.2245,
      "mean_token_accuracy": 0.14160481840372086,
      "num_tokens": 15155400.0,
      "step": 7740
    },
    {
      "entropy": 6.265633249282837,
      "epoch": 0.7742290198430549,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000494870244671075,
      "loss": 6.0623,
      "mean_token_accuracy": 0.1620455116033554,
      "num_tokens": 15164969.0,
      "step": 7745
    },
    {
      "entropy": 6.250748014450073,
      "epoch": 0.7747288449042835,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004948626645419926,
      "loss": 6.1178,
      "mean_token_accuracy": 0.15456647872924806,
      "num_tokens": 15173840.0,
      "step": 7750
    },
    {
      "entropy": 6.324423313140869,
      "epoch": 0.7752286699655121,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004948550788812222,
      "loss": 6.2388,
      "mean_token_accuracy": 0.1355576992034912,
      "num_tokens": 15184541.0,
      "step": 7755
    },
    {
      "entropy": 6.2660486698150635,
      "epoch": 0.7757284950267407,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004948474876889545,
      "loss": 6.1692,
      "mean_token_accuracy": 0.14640089869499207,
      "num_tokens": 15195060.0,
      "step": 7760
    },
    {
      "entropy": 6.3367759704589846,
      "epoch": 0.7762283200879692,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004948398909653806,
      "loss": 6.1908,
      "mean_token_accuracy": 0.145336289703846,
      "num_tokens": 15206093.0,
      "step": 7765
    },
    {
      "entropy": 6.32977523803711,
      "epoch": 0.7767281451491977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004948322887106917,
      "loss": 6.1567,
      "mean_token_accuracy": 0.14237820357084274,
      "num_tokens": 15215359.0,
      "step": 7770
    },
    {
      "entropy": 6.276315021514892,
      "epoch": 0.7772279702104263,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004948246809250788,
      "loss": 6.1944,
      "mean_token_accuracy": 0.14711200818419456,
      "num_tokens": 15225283.0,
      "step": 7775
    },
    {
      "entropy": 6.212570142745972,
      "epoch": 0.7777277952716549,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004948170676087338,
      "loss": 6.0592,
      "mean_token_accuracy": 0.15494513958692552,
      "num_tokens": 15234319.0,
      "step": 7780
    },
    {
      "entropy": 6.242264890670777,
      "epoch": 0.7782276203328835,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004948094487618476,
      "loss": 6.0387,
      "mean_token_accuracy": 0.15930456817150115,
      "num_tokens": 15244410.0,
      "step": 7785
    },
    {
      "entropy": 6.276339197158814,
      "epoch": 0.7787274453941121,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004948018243846126,
      "loss": 6.1081,
      "mean_token_accuracy": 0.15080363154411316,
      "num_tokens": 15253431.0,
      "step": 7790
    },
    {
      "entropy": 6.2182506084442135,
      "epoch": 0.7792272704553407,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004947941944772199,
      "loss": 6.1435,
      "mean_token_accuracy": 0.1517641946673393,
      "num_tokens": 15262566.0,
      "step": 7795
    },
    {
      "entropy": 6.3042236328125,
      "epoch": 0.7797270955165692,
      "grad_norm": 1.171875,
      "learning_rate": 0.000494786559039862,
      "loss": 6.223,
      "mean_token_accuracy": 0.1447600543498993,
      "num_tokens": 15272122.0,
      "step": 7800
    },
    {
      "entropy": 6.318529081344605,
      "epoch": 0.7802269205777977,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004947789180727308,
      "loss": 6.1755,
      "mean_token_accuracy": 0.14170423597097398,
      "num_tokens": 15281165.0,
      "step": 7805
    },
    {
      "entropy": 6.208543348312378,
      "epoch": 0.7807267456390263,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004947712715760186,
      "loss": 6.1644,
      "mean_token_accuracy": 0.14214149862527847,
      "num_tokens": 15290992.0,
      "step": 7810
    },
    {
      "entropy": 6.3388128757476805,
      "epoch": 0.7812265707002549,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004947636195499177,
      "loss": 6.1554,
      "mean_token_accuracy": 0.14541168957948686,
      "num_tokens": 15300454.0,
      "step": 7815
    },
    {
      "entropy": 6.2795654296875,
      "epoch": 0.7817263957614835,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004947559619946207,
      "loss": 6.1088,
      "mean_token_accuracy": 0.15477105379104614,
      "num_tokens": 15310471.0,
      "step": 7820
    },
    {
      "entropy": 6.29819278717041,
      "epoch": 0.7822262208227121,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004947482989103203,
      "loss": 6.2112,
      "mean_token_accuracy": 0.14262558296322822,
      "num_tokens": 15320554.0,
      "step": 7825
    },
    {
      "entropy": 6.270034122467041,
      "epoch": 0.7827260458839406,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004947406302972092,
      "loss": 6.2402,
      "mean_token_accuracy": 0.14409922137856485,
      "num_tokens": 15331623.0,
      "step": 7830
    },
    {
      "entropy": 6.364425611495972,
      "epoch": 0.7832258709451692,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004947329561554804,
      "loss": 6.0876,
      "mean_token_accuracy": 0.15001410096883774,
      "num_tokens": 15340954.0,
      "step": 7835
    },
    {
      "entropy": 6.228172159194946,
      "epoch": 0.7837256960063977,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004947252764853269,
      "loss": 6.1951,
      "mean_token_accuracy": 0.14378104358911514,
      "num_tokens": 15350858.0,
      "step": 7840
    },
    {
      "entropy": 6.2581991195678714,
      "epoch": 0.7842255210676263,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004947175912869421,
      "loss": 6.091,
      "mean_token_accuracy": 0.14584627002477646,
      "num_tokens": 15360382.0,
      "step": 7845
    },
    {
      "entropy": 6.271379089355468,
      "epoch": 0.7847253461288549,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004947099005605191,
      "loss": 6.1998,
      "mean_token_accuracy": 0.14111343398690224,
      "num_tokens": 15370626.0,
      "step": 7850
    },
    {
      "entropy": 6.301034307479858,
      "epoch": 0.7852251711900835,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004947022043062517,
      "loss": 6.1668,
      "mean_token_accuracy": 0.14560750499367714,
      "num_tokens": 15379582.0,
      "step": 7855
    },
    {
      "entropy": 6.278450632095337,
      "epoch": 0.7857249962513121,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004946945025243333,
      "loss": 6.0572,
      "mean_token_accuracy": 0.1529711127281189,
      "num_tokens": 15390168.0,
      "step": 7860
    },
    {
      "entropy": 6.088787937164307,
      "epoch": 0.7862248213125406,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004946867952149577,
      "loss": 6.0138,
      "mean_token_accuracy": 0.1608179047703743,
      "num_tokens": 15400979.0,
      "step": 7865
    },
    {
      "entropy": 6.303697967529297,
      "epoch": 0.7867246463737692,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004946790823783189,
      "loss": 6.1096,
      "mean_token_accuracy": 0.14962270855903625,
      "num_tokens": 15409954.0,
      "step": 7870
    },
    {
      "entropy": 6.244659280776977,
      "epoch": 0.7872244714349977,
      "grad_norm": 1.140625,
      "learning_rate": 0.000494671364014611,
      "loss": 6.1972,
      "mean_token_accuracy": 0.14646629244089127,
      "num_tokens": 15419539.0,
      "step": 7875
    },
    {
      "entropy": 6.186657810211182,
      "epoch": 0.7877242964962263,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000494663640124028,
      "loss": 6.1426,
      "mean_token_accuracy": 0.1448366276919842,
      "num_tokens": 15428751.0,
      "step": 7880
    },
    {
      "entropy": 6.3191516399383545,
      "epoch": 0.7882241215574549,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004946559107067643,
      "loss": 6.1518,
      "mean_token_accuracy": 0.15063509792089463,
      "num_tokens": 15438466.0,
      "step": 7885
    },
    {
      "entropy": 6.264244413375854,
      "epoch": 0.7887239466186835,
      "grad_norm": 1.125,
      "learning_rate": 0.0004946481757630146,
      "loss": 6.134,
      "mean_token_accuracy": 0.14569272696971894,
      "num_tokens": 15448395.0,
      "step": 7890
    },
    {
      "entropy": 6.254844999313354,
      "epoch": 0.789223771679912,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004946404352929732,
      "loss": 6.1666,
      "mean_token_accuracy": 0.14378464519977568,
      "num_tokens": 15457986.0,
      "step": 7895
    },
    {
      "entropy": 6.283915853500366,
      "epoch": 0.7897235967411406,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000494632689296835,
      "loss": 6.159,
      "mean_token_accuracy": 0.14958941340446472,
      "num_tokens": 15467989.0,
      "step": 7900
    },
    {
      "entropy": 6.220441246032715,
      "epoch": 0.7902234218023692,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004946249377747947,
      "loss": 6.0359,
      "mean_token_accuracy": 0.15385646671056746,
      "num_tokens": 15477537.0,
      "step": 7905
    },
    {
      "entropy": 6.278183364868164,
      "epoch": 0.7907232468635977,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004946171807270476,
      "loss": 6.1017,
      "mean_token_accuracy": 0.1515183076262474,
      "num_tokens": 15488007.0,
      "step": 7910
    },
    {
      "entropy": 6.334575653076172,
      "epoch": 0.7912230719248263,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004946094181537887,
      "loss": 6.1588,
      "mean_token_accuracy": 0.1481189861893654,
      "num_tokens": 15498122.0,
      "step": 7915
    },
    {
      "entropy": 6.297288322448731,
      "epoch": 0.7917228969860549,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004946016500552134,
      "loss": 6.1699,
      "mean_token_accuracy": 0.14629243016242982,
      "num_tokens": 15507585.0,
      "step": 7920
    },
    {
      "entropy": 6.202771615982056,
      "epoch": 0.7922227220472835,
      "grad_norm": 1.125,
      "learning_rate": 0.0004945938764315171,
      "loss": 6.0841,
      "mean_token_accuracy": 0.15092861652374268,
      "num_tokens": 15516260.0,
      "step": 7925
    },
    {
      "entropy": 6.316201543807983,
      "epoch": 0.792722547108512,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004945860972828954,
      "loss": 6.1904,
      "mean_token_accuracy": 0.1480329692363739,
      "num_tokens": 15527372.0,
      "step": 7930
    },
    {
      "entropy": 6.2938704013824465,
      "epoch": 0.7932223721697406,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004945783126095439,
      "loss": 6.1613,
      "mean_token_accuracy": 0.14766191840171813,
      "num_tokens": 15537244.0,
      "step": 7935
    },
    {
      "entropy": 6.3674561977386475,
      "epoch": 0.7937221972309692,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004945705224116586,
      "loss": 6.1701,
      "mean_token_accuracy": 0.14418644309043885,
      "num_tokens": 15546512.0,
      "step": 7940
    },
    {
      "entropy": 6.22338228225708,
      "epoch": 0.7942220222921977,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004945627266894354,
      "loss": 6.1318,
      "mean_token_accuracy": 0.15029769837856294,
      "num_tokens": 15557381.0,
      "step": 7945
    },
    {
      "entropy": 6.290680980682373,
      "epoch": 0.7947218473534263,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004945549254430707,
      "loss": 6.2096,
      "mean_token_accuracy": 0.1386479027569294,
      "num_tokens": 15565806.0,
      "step": 7950
    },
    {
      "entropy": 6.281616115570069,
      "epoch": 0.7952216724146549,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004945471186727603,
      "loss": 6.124,
      "mean_token_accuracy": 0.14674990102648736,
      "num_tokens": 15577059.0,
      "step": 7955
    },
    {
      "entropy": 6.384580373764038,
      "epoch": 0.7957214974758834,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004945393063787011,
      "loss": 6.3065,
      "mean_token_accuracy": 0.1349059171974659,
      "num_tokens": 15587827.0,
      "step": 7960
    },
    {
      "entropy": 6.363341283798218,
      "epoch": 0.796221322537112,
      "grad_norm": 1.125,
      "learning_rate": 0.0004945314885610894,
      "loss": 6.1073,
      "mean_token_accuracy": 0.14445040374994278,
      "num_tokens": 15598232.0,
      "step": 7965
    },
    {
      "entropy": 6.226699352264404,
      "epoch": 0.7967211475983406,
      "grad_norm": 1.125,
      "learning_rate": 0.0004945236652201219,
      "loss": 6.081,
      "mean_token_accuracy": 0.14846306145191193,
      "num_tokens": 15606896.0,
      "step": 7970
    },
    {
      "entropy": 6.173016786575317,
      "epoch": 0.7972209726595691,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004945158363559955,
      "loss": 6.0662,
      "mean_token_accuracy": 0.15879194885492326,
      "num_tokens": 15616868.0,
      "step": 7975
    },
    {
      "entropy": 6.241082954406738,
      "epoch": 0.7977207977207977,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004945080019689071,
      "loss": 6.0978,
      "mean_token_accuracy": 0.14706755429506302,
      "num_tokens": 15626521.0,
      "step": 7980
    },
    {
      "entropy": 6.236615943908691,
      "epoch": 0.7982206227820263,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004945001620590538,
      "loss": 6.0891,
      "mean_token_accuracy": 0.15682599395513536,
      "num_tokens": 15635901.0,
      "step": 7985
    },
    {
      "entropy": 6.288385629653931,
      "epoch": 0.7987204478432549,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000494492316626633,
      "loss": 6.1472,
      "mean_token_accuracy": 0.14089618399739265,
      "num_tokens": 15645672.0,
      "step": 7990
    },
    {
      "entropy": 6.342327880859375,
      "epoch": 0.7992202729044834,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004944844656718419,
      "loss": 6.1788,
      "mean_token_accuracy": 0.14746755734086037,
      "num_tokens": 15656668.0,
      "step": 7995
    },
    {
      "entropy": 6.297449254989624,
      "epoch": 0.799720097965712,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004944766091948783,
      "loss": 6.1333,
      "mean_token_accuracy": 0.1526561349630356,
      "num_tokens": 15666176.0,
      "step": 8000
    },
    {
      "entropy": 6.28197865486145,
      "epoch": 0.8002199230269406,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004944687471959395,
      "loss": 6.1469,
      "mean_token_accuracy": 0.14437325149774552,
      "num_tokens": 15676162.0,
      "step": 8005
    },
    {
      "entropy": 6.281189823150635,
      "epoch": 0.8007197480881691,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004944608796752237,
      "loss": 6.1531,
      "mean_token_accuracy": 0.15332329273223877,
      "num_tokens": 15685393.0,
      "step": 8010
    },
    {
      "entropy": 6.231124687194824,
      "epoch": 0.8012195731493977,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004944530066329284,
      "loss": 6.067,
      "mean_token_accuracy": 0.15621009916067125,
      "num_tokens": 15694944.0,
      "step": 8015
    },
    {
      "entropy": 6.323437261581421,
      "epoch": 0.8017193982106263,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004944451280692521,
      "loss": 6.2198,
      "mean_token_accuracy": 0.14411858469247818,
      "num_tokens": 15704410.0,
      "step": 8020
    },
    {
      "entropy": 6.288111019134521,
      "epoch": 0.8022192232718548,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004944372439843927,
      "loss": 6.1028,
      "mean_token_accuracy": 0.14813474118709563,
      "num_tokens": 15713663.0,
      "step": 8025
    },
    {
      "entropy": 6.267939758300781,
      "epoch": 0.8027190483330834,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004944293543785489,
      "loss": 6.0572,
      "mean_token_accuracy": 0.1585802286863327,
      "num_tokens": 15722662.0,
      "step": 8030
    },
    {
      "entropy": 6.311334657669067,
      "epoch": 0.803218873394312,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004944214592519188,
      "loss": 6.2515,
      "mean_token_accuracy": 0.14156176820397376,
      "num_tokens": 15732615.0,
      "step": 8035
    },
    {
      "entropy": 6.282354927062988,
      "epoch": 0.8037186984555406,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004944135586047014,
      "loss": 6.1716,
      "mean_token_accuracy": 0.15733135342597962,
      "num_tokens": 15741569.0,
      "step": 8040
    },
    {
      "entropy": 6.26232419013977,
      "epoch": 0.8042185235167691,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004944056524370953,
      "loss": 6.0548,
      "mean_token_accuracy": 0.1563683867454529,
      "num_tokens": 15750734.0,
      "step": 8045
    },
    {
      "entropy": 6.299233484268188,
      "epoch": 0.8047183485779977,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004943977407492994,
      "loss": 6.1241,
      "mean_token_accuracy": 0.1543577417731285,
      "num_tokens": 15761664.0,
      "step": 8050
    },
    {
      "entropy": 6.210677242279052,
      "epoch": 0.8052181736392263,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004943898235415128,
      "loss": 6.0691,
      "mean_token_accuracy": 0.15466163158416749,
      "num_tokens": 15771159.0,
      "step": 8055
    },
    {
      "entropy": 6.330757570266724,
      "epoch": 0.8057179987004548,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004943819008139347,
      "loss": 6.1916,
      "mean_token_accuracy": 0.1438830614089966,
      "num_tokens": 15781429.0,
      "step": 8060
    },
    {
      "entropy": 6.2990278720855715,
      "epoch": 0.8062178237616834,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004943739725667644,
      "loss": 6.0918,
      "mean_token_accuracy": 0.14749809503555297,
      "num_tokens": 15791361.0,
      "step": 8065
    },
    {
      "entropy": 6.311406469345092,
      "epoch": 0.806717648822912,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004943660388002014,
      "loss": 6.2054,
      "mean_token_accuracy": 0.14021724089980125,
      "num_tokens": 15802020.0,
      "step": 8070
    },
    {
      "entropy": 6.162099075317383,
      "epoch": 0.8072174738841406,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004943580995144454,
      "loss": 6.037,
      "mean_token_accuracy": 0.1582739382982254,
      "num_tokens": 15811319.0,
      "step": 8075
    },
    {
      "entropy": 6.274627780914306,
      "epoch": 0.8077172989453691,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004943501547096961,
      "loss": 6.1805,
      "mean_token_accuracy": 0.14531080797314644,
      "num_tokens": 15821337.0,
      "step": 8080
    },
    {
      "entropy": 6.297515726089477,
      "epoch": 0.8082171240065977,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004943422043861531,
      "loss": 6.1582,
      "mean_token_accuracy": 0.14785483479499817,
      "num_tokens": 15832234.0,
      "step": 8085
    },
    {
      "entropy": 6.339212894439697,
      "epoch": 0.8087169490678262,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004943342485440169,
      "loss": 6.1592,
      "mean_token_accuracy": 0.14933843910694122,
      "num_tokens": 15841797.0,
      "step": 8090
    },
    {
      "entropy": 6.260598945617676,
      "epoch": 0.8092167741290548,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004943262871834874,
      "loss": 6.1493,
      "mean_token_accuracy": 0.14893108755350112,
      "num_tokens": 15851031.0,
      "step": 8095
    },
    {
      "entropy": 6.26089038848877,
      "epoch": 0.8097165991902834,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004943183203047649,
      "loss": 6.1744,
      "mean_token_accuracy": 0.14350613057613373,
      "num_tokens": 15861297.0,
      "step": 8100
    },
    {
      "entropy": 6.316806793212891,
      "epoch": 0.810216424251512,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004943103479080499,
      "loss": 6.1886,
      "mean_token_accuracy": 0.14830220639705657,
      "num_tokens": 15871066.0,
      "step": 8105
    },
    {
      "entropy": 6.3423620700836185,
      "epoch": 0.8107162493127406,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000494302369993543,
      "loss": 6.1532,
      "mean_token_accuracy": 0.14845208823680878,
      "num_tokens": 15880395.0,
      "step": 8110
    },
    {
      "entropy": 6.260035562515259,
      "epoch": 0.8112160743739691,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004942943865614449,
      "loss": 6.1112,
      "mean_token_accuracy": 0.15113046616315842,
      "num_tokens": 15889937.0,
      "step": 8115
    },
    {
      "entropy": 6.3086097240448,
      "epoch": 0.8117158994351977,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004942863976119565,
      "loss": 6.1119,
      "mean_token_accuracy": 0.14828324913978577,
      "num_tokens": 15899894.0,
      "step": 8120
    },
    {
      "entropy": 6.181340312957763,
      "epoch": 0.8122157244964262,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004942784031452788,
      "loss": 6.0097,
      "mean_token_accuracy": 0.16124274879693984,
      "num_tokens": 15909038.0,
      "step": 8125
    },
    {
      "entropy": 6.208627939224243,
      "epoch": 0.8127155495576548,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004942704031616127,
      "loss": 6.1151,
      "mean_token_accuracy": 0.14759739190340043,
      "num_tokens": 15917891.0,
      "step": 8130
    },
    {
      "entropy": 6.351293087005615,
      "epoch": 0.8132153746188834,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004942623976611598,
      "loss": 6.2362,
      "mean_token_accuracy": 0.14013138711452483,
      "num_tokens": 15928079.0,
      "step": 8135
    },
    {
      "entropy": 6.2469260692596436,
      "epoch": 0.813715199680112,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004942543866441215,
      "loss": 6.0692,
      "mean_token_accuracy": 0.15209134966135024,
      "num_tokens": 15938065.0,
      "step": 8140
    },
    {
      "entropy": 6.1801941871643065,
      "epoch": 0.8142150247413406,
      "grad_norm": 1.125,
      "learning_rate": 0.0004942463701106992,
      "loss": 6.1019,
      "mean_token_accuracy": 0.14780305474996566,
      "num_tokens": 15947650.0,
      "step": 8145
    },
    {
      "entropy": 6.259891176223755,
      "epoch": 0.8147148498025691,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004942383480610946,
      "loss": 6.122,
      "mean_token_accuracy": 0.1508016139268875,
      "num_tokens": 15957194.0,
      "step": 8150
    },
    {
      "entropy": 6.28717622756958,
      "epoch": 0.8152146748637976,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004942303204955096,
      "loss": 6.0911,
      "mean_token_accuracy": 0.14687171950936317,
      "num_tokens": 15966068.0,
      "step": 8155
    },
    {
      "entropy": 6.215104389190674,
      "epoch": 0.8157144999250262,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004942222874141461,
      "loss": 6.1034,
      "mean_token_accuracy": 0.15094738081097603,
      "num_tokens": 15975672.0,
      "step": 8160
    },
    {
      "entropy": 6.213860702514649,
      "epoch": 0.8162143249862548,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004942142488172063,
      "loss": 6.1296,
      "mean_token_accuracy": 0.15043410956859588,
      "num_tokens": 15986216.0,
      "step": 8165
    },
    {
      "entropy": 6.2506379127502445,
      "epoch": 0.8167141500474834,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004942062047048924,
      "loss": 6.0541,
      "mean_token_accuracy": 0.15532840192317962,
      "num_tokens": 15994673.0,
      "step": 8170
    },
    {
      "entropy": 6.271219396591187,
      "epoch": 0.817213975108712,
      "grad_norm": 1.125,
      "learning_rate": 0.0004941981550774068,
      "loss": 6.0936,
      "mean_token_accuracy": 0.1537529855966568,
      "num_tokens": 16003576.0,
      "step": 8175
    },
    {
      "entropy": 6.151399707794189,
      "epoch": 0.8177138001699406,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004941900999349521,
      "loss": 6.0608,
      "mean_token_accuracy": 0.15823685377836227,
      "num_tokens": 16013334.0,
      "step": 8180
    },
    {
      "entropy": 6.291089963912964,
      "epoch": 0.8182136252311691,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004941820392777307,
      "loss": 6.0902,
      "mean_token_accuracy": 0.15198469161987305,
      "num_tokens": 16023101.0,
      "step": 8185
    },
    {
      "entropy": 6.217065000534058,
      "epoch": 0.8187134502923976,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004941739731059457,
      "loss": 6.1509,
      "mean_token_accuracy": 0.14738564118742942,
      "num_tokens": 16034189.0,
      "step": 8190
    },
    {
      "entropy": 6.281302785873413,
      "epoch": 0.8192132753536262,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004941659014198,
      "loss": 6.0633,
      "mean_token_accuracy": 0.151087449491024,
      "num_tokens": 16044008.0,
      "step": 8195
    },
    {
      "entropy": 6.322632122039795,
      "epoch": 0.8197131004148548,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004941578242194965,
      "loss": 6.1656,
      "mean_token_accuracy": 0.15138754844665528,
      "num_tokens": 16053557.0,
      "step": 8200
    },
    {
      "entropy": 6.239453458786011,
      "epoch": 0.8202129254760834,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004941497415052387,
      "loss": 6.0872,
      "mean_token_accuracy": 0.1491052821278572,
      "num_tokens": 16063191.0,
      "step": 8205
    },
    {
      "entropy": 6.231597185134888,
      "epoch": 0.820712750537312,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004941416532772296,
      "loss": 6.113,
      "mean_token_accuracy": 0.14665559381246568,
      "num_tokens": 16072482.0,
      "step": 8210
    },
    {
      "entropy": 6.2053595066070555,
      "epoch": 0.8212125755985406,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000494133559535673,
      "loss": 6.0861,
      "mean_token_accuracy": 0.15713008493185043,
      "num_tokens": 16082848.0,
      "step": 8215
    },
    {
      "entropy": 6.308273601531982,
      "epoch": 0.821712400659769,
      "grad_norm": 1.25,
      "learning_rate": 0.0004941254602807724,
      "loss": 6.234,
      "mean_token_accuracy": 0.14311141073703765,
      "num_tokens": 16091604.0,
      "step": 8220
    },
    {
      "entropy": 6.340981340408325,
      "epoch": 0.8222122257209976,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004941173555127315,
      "loss": 6.1313,
      "mean_token_accuracy": 0.14595386683940886,
      "num_tokens": 16101427.0,
      "step": 8225
    },
    {
      "entropy": 6.1844323635101315,
      "epoch": 0.8227120507822262,
      "grad_norm": 1.125,
      "learning_rate": 0.0004941092452317546,
      "loss": 6.0846,
      "mean_token_accuracy": 0.15165154933929442,
      "num_tokens": 16112206.0,
      "step": 8230
    },
    {
      "entropy": 6.269962501525879,
      "epoch": 0.8232118758434548,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004941011294380452,
      "loss": 6.1985,
      "mean_token_accuracy": 0.14949085861444472,
      "num_tokens": 16122652.0,
      "step": 8235
    },
    {
      "entropy": 6.2866960048675535,
      "epoch": 0.8237117009046834,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004940930081318079,
      "loss": 6.0846,
      "mean_token_accuracy": 0.14825305044651033,
      "num_tokens": 16132976.0,
      "step": 8240
    },
    {
      "entropy": 6.304342603683471,
      "epoch": 0.824211525965912,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004940848813132469,
      "loss": 6.171,
      "mean_token_accuracy": 0.14888230636715888,
      "num_tokens": 16143514.0,
      "step": 8245
    },
    {
      "entropy": 6.278770017623901,
      "epoch": 0.8247113510271405,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004940767489825666,
      "loss": 6.1577,
      "mean_token_accuracy": 0.1420615002512932,
      "num_tokens": 16154235.0,
      "step": 8250
    },
    {
      "entropy": 6.275551700592041,
      "epoch": 0.825211176088369,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004940686111399716,
      "loss": 6.0568,
      "mean_token_accuracy": 0.1485327109694481,
      "num_tokens": 16164251.0,
      "step": 8255
    },
    {
      "entropy": 6.203812122344971,
      "epoch": 0.8257110011495976,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004940604677856669,
      "loss": 6.002,
      "mean_token_accuracy": 0.16260287761688233,
      "num_tokens": 16172780.0,
      "step": 8260
    },
    {
      "entropy": 6.035352325439453,
      "epoch": 0.8262108262108262,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000494052318919857,
      "loss": 5.9861,
      "mean_token_accuracy": 0.16190174221992493,
      "num_tokens": 16182443.0,
      "step": 8265
    },
    {
      "entropy": 6.241731834411621,
      "epoch": 0.8267106512720548,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004940441645427473,
      "loss": 6.1005,
      "mean_token_accuracy": 0.15218378305435182,
      "num_tokens": 16192515.0,
      "step": 8270
    },
    {
      "entropy": 6.26733169555664,
      "epoch": 0.8272104763332834,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004940360046545427,
      "loss": 6.093,
      "mean_token_accuracy": 0.15167328268289565,
      "num_tokens": 16201615.0,
      "step": 8275
    },
    {
      "entropy": 6.255200719833374,
      "epoch": 0.827710301394512,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004940278392554486,
      "loss": 6.1287,
      "mean_token_accuracy": 0.14669278562068938,
      "num_tokens": 16210461.0,
      "step": 8280
    },
    {
      "entropy": 6.204719924926758,
      "epoch": 0.8282101264557404,
      "grad_norm": 1.125,
      "learning_rate": 0.0004940196683456704,
      "loss": 6.119,
      "mean_token_accuracy": 0.14313342422246933,
      "num_tokens": 16220293.0,
      "step": 8285
    },
    {
      "entropy": 6.171269035339355,
      "epoch": 0.828709951516969,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004940114919254137,
      "loss": 6.0536,
      "mean_token_accuracy": 0.1509528174996376,
      "num_tokens": 16230020.0,
      "step": 8290
    },
    {
      "entropy": 6.221949291229248,
      "epoch": 0.8292097765781976,
      "grad_norm": 1.25,
      "learning_rate": 0.0004940033099948843,
      "loss": 6.0731,
      "mean_token_accuracy": 0.15002416521310807,
      "num_tokens": 16240187.0,
      "step": 8295
    },
    {
      "entropy": 6.226723003387451,
      "epoch": 0.8297096016394262,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004939951225542879,
      "loss": 6.0781,
      "mean_token_accuracy": 0.14879120513796806,
      "num_tokens": 16250200.0,
      "step": 8300
    },
    {
      "entropy": 6.277824783325196,
      "epoch": 0.8302094267006548,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004939869296038306,
      "loss": 6.1774,
      "mean_token_accuracy": 0.15186067372560502,
      "num_tokens": 16259396.0,
      "step": 8305
    },
    {
      "entropy": 6.282472801208496,
      "epoch": 0.8307092517618834,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004939787311437185,
      "loss": 6.1815,
      "mean_token_accuracy": 0.13600158393383027,
      "num_tokens": 16270033.0,
      "step": 8310
    },
    {
      "entropy": 6.322247409820557,
      "epoch": 0.831209076823112,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004939705271741579,
      "loss": 6.129,
      "mean_token_accuracy": 0.148136106133461,
      "num_tokens": 16281464.0,
      "step": 8315
    },
    {
      "entropy": 6.338821887969971,
      "epoch": 0.8317089018843404,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004939623176953551,
      "loss": 6.193,
      "mean_token_accuracy": 0.14136315062642096,
      "num_tokens": 16292125.0,
      "step": 8320
    },
    {
      "entropy": 6.190872478485107,
      "epoch": 0.832208726945569,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004939541027075168,
      "loss": 6.1273,
      "mean_token_accuracy": 0.14964912235736846,
      "num_tokens": 16302993.0,
      "step": 8325
    },
    {
      "entropy": 6.201830673217773,
      "epoch": 0.8327085520067976,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004939458822108496,
      "loss": 6.0959,
      "mean_token_accuracy": 0.14907075166702272,
      "num_tokens": 16311287.0,
      "step": 8330
    },
    {
      "entropy": 6.28732123374939,
      "epoch": 0.8332083770680262,
      "grad_norm": 1.125,
      "learning_rate": 0.0004939376562055602,
      "loss": 6.0874,
      "mean_token_accuracy": 0.14835006594657899,
      "num_tokens": 16320701.0,
      "step": 8335
    },
    {
      "entropy": 6.214312887191772,
      "epoch": 0.8337082021292548,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004939294246918558,
      "loss": 6.1192,
      "mean_token_accuracy": 0.14457828402519227,
      "num_tokens": 16329834.0,
      "step": 8340
    },
    {
      "entropy": 6.191360807418823,
      "epoch": 0.8342080271904834,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004939211876699435,
      "loss": 6.1208,
      "mean_token_accuracy": 0.15147167667746544,
      "num_tokens": 16339043.0,
      "step": 8345
    },
    {
      "entropy": 6.313658237457275,
      "epoch": 0.8347078522517118,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004939129451400305,
      "loss": 6.2191,
      "mean_token_accuracy": 0.14118876904249192,
      "num_tokens": 16348774.0,
      "step": 8350
    },
    {
      "entropy": 6.301902389526367,
      "epoch": 0.8352076773129404,
      "grad_norm": 1.171875,
      "learning_rate": 0.000493904697102324,
      "loss": 6.0914,
      "mean_token_accuracy": 0.14720873683691024,
      "num_tokens": 16357999.0,
      "step": 8355
    },
    {
      "entropy": 6.295412302017212,
      "epoch": 0.835707502374169,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004938964435570317,
      "loss": 6.1426,
      "mean_token_accuracy": 0.14987237825989724,
      "num_tokens": 16369308.0,
      "step": 8360
    },
    {
      "entropy": 6.149048805236816,
      "epoch": 0.8362073274353976,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004938881845043612,
      "loss": 6.0392,
      "mean_token_accuracy": 0.14802421182394027,
      "num_tokens": 16378001.0,
      "step": 8365
    },
    {
      "entropy": 6.281769514083862,
      "epoch": 0.8367071524966262,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004938799199445202,
      "loss": 6.2459,
      "mean_token_accuracy": 0.14276251643896104,
      "num_tokens": 16387925.0,
      "step": 8370
    },
    {
      "entropy": 6.33515944480896,
      "epoch": 0.8372069775578548,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004938716498777169,
      "loss": 6.2031,
      "mean_token_accuracy": 0.14070867970585824,
      "num_tokens": 16398200.0,
      "step": 8375
    },
    {
      "entropy": 6.151538133621216,
      "epoch": 0.8377068026190834,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004938633743041592,
      "loss": 5.9746,
      "mean_token_accuracy": 0.15544663965702057,
      "num_tokens": 16407874.0,
      "step": 8380
    },
    {
      "entropy": 6.232445764541626,
      "epoch": 0.8382066276803118,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004938550932240552,
      "loss": 6.1137,
      "mean_token_accuracy": 0.15000495910644532,
      "num_tokens": 16417825.0,
      "step": 8385
    },
    {
      "entropy": 6.2535772800445555,
      "epoch": 0.8387064527415404,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004938468066376134,
      "loss": 6.0954,
      "mean_token_accuracy": 0.15377777218818664,
      "num_tokens": 16427265.0,
      "step": 8390
    },
    {
      "entropy": 6.149886178970337,
      "epoch": 0.839206277802769,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004938385145450423,
      "loss": 6.0286,
      "mean_token_accuracy": 0.15962575674057006,
      "num_tokens": 16436639.0,
      "step": 8395
    },
    {
      "entropy": 6.241818237304687,
      "epoch": 0.8397061028639976,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004938302169465505,
      "loss": 6.132,
      "mean_token_accuracy": 0.14945596307516099,
      "num_tokens": 16446411.0,
      "step": 8400
    },
    {
      "entropy": 6.307015085220337,
      "epoch": 0.8402059279252262,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004938219138423467,
      "loss": 6.1613,
      "mean_token_accuracy": 0.14830330908298492,
      "num_tokens": 16455784.0,
      "step": 8405
    },
    {
      "entropy": 6.215744590759277,
      "epoch": 0.8407057529864548,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004938136052326399,
      "loss": 5.9934,
      "mean_token_accuracy": 0.1575619548559189,
      "num_tokens": 16464411.0,
      "step": 8410
    },
    {
      "entropy": 6.254569149017334,
      "epoch": 0.8412055780476833,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004938052911176392,
      "loss": 6.186,
      "mean_token_accuracy": 0.1385000117123127,
      "num_tokens": 16473299.0,
      "step": 8415
    },
    {
      "entropy": 6.173201560974121,
      "epoch": 0.8417054031089118,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004937969714975534,
      "loss": 5.9853,
      "mean_token_accuracy": 0.15716107934713364,
      "num_tokens": 16483040.0,
      "step": 8420
    },
    {
      "entropy": 6.221319150924683,
      "epoch": 0.8422052281701404,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004937886463725923,
      "loss": 6.0709,
      "mean_token_accuracy": 0.15436940044164657,
      "num_tokens": 16492608.0,
      "step": 8425
    },
    {
      "entropy": 6.292224168777466,
      "epoch": 0.842705053231369,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004937803157429651,
      "loss": 6.2158,
      "mean_token_accuracy": 0.14316556081175805,
      "num_tokens": 16503139.0,
      "step": 8430
    },
    {
      "entropy": 6.251599836349487,
      "epoch": 0.8432048782925976,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004937719796088815,
      "loss": 6.0771,
      "mean_token_accuracy": 0.15507673770189284,
      "num_tokens": 16511584.0,
      "step": 8435
    },
    {
      "entropy": 6.192686986923218,
      "epoch": 0.8437047033538262,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004937636379705511,
      "loss": 6.0784,
      "mean_token_accuracy": 0.14900906831026078,
      "num_tokens": 16522407.0,
      "step": 8440
    },
    {
      "entropy": 6.188032102584839,
      "epoch": 0.8442045284150548,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004937552908281839,
      "loss": 6.0949,
      "mean_token_accuracy": 0.1458570197224617,
      "num_tokens": 16532351.0,
      "step": 8445
    },
    {
      "entropy": 6.308607816696167,
      "epoch": 0.8447043534762833,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004937469381819899,
      "loss": 6.1619,
      "mean_token_accuracy": 0.14209038987755776,
      "num_tokens": 16541673.0,
      "step": 8450
    },
    {
      "entropy": 6.244873189926148,
      "epoch": 0.8452041785375118,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004937385800321792,
      "loss": 6.1328,
      "mean_token_accuracy": 0.1541708528995514,
      "num_tokens": 16551622.0,
      "step": 8455
    },
    {
      "entropy": 6.224560928344727,
      "epoch": 0.8457040035987404,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004937302163789621,
      "loss": 6.075,
      "mean_token_accuracy": 0.153325604647398,
      "num_tokens": 16559074.0,
      "step": 8460
    },
    {
      "entropy": 6.219470643997193,
      "epoch": 0.846203828659969,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004937218472225491,
      "loss": 6.0717,
      "mean_token_accuracy": 0.15759089291095735,
      "num_tokens": 16568650.0,
      "step": 8465
    },
    {
      "entropy": 6.225276136398316,
      "epoch": 0.8467036537211976,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004937134725631506,
      "loss": 6.1012,
      "mean_token_accuracy": 0.1527090534567833,
      "num_tokens": 16578696.0,
      "step": 8470
    },
    {
      "entropy": 6.2704683303833,
      "epoch": 0.8472034787824262,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004937050924009775,
      "loss": 6.1575,
      "mean_token_accuracy": 0.1451670192182064,
      "num_tokens": 16589014.0,
      "step": 8475
    },
    {
      "entropy": 6.261503124237061,
      "epoch": 0.8477033038436548,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004936967067362406,
      "loss": 6.0498,
      "mean_token_accuracy": 0.1592542201280594,
      "num_tokens": 16598319.0,
      "step": 8480
    },
    {
      "entropy": 6.240740442276001,
      "epoch": 0.8482031289048833,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004936883155691507,
      "loss": 6.1739,
      "mean_token_accuracy": 0.14734973311424254,
      "num_tokens": 16607941.0,
      "step": 8485
    },
    {
      "entropy": 6.1987278938293455,
      "epoch": 0.8487029539661118,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004936799188999192,
      "loss": 6.1098,
      "mean_token_accuracy": 0.15264756456017495,
      "num_tokens": 16618398.0,
      "step": 8490
    },
    {
      "entropy": 6.274617099761963,
      "epoch": 0.8492027790273404,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004936715167287571,
      "loss": 6.149,
      "mean_token_accuracy": 0.1433829329907894,
      "num_tokens": 16628953.0,
      "step": 8495
    },
    {
      "entropy": 6.311986255645752,
      "epoch": 0.849702604088569,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000493663109055876,
      "loss": 6.0938,
      "mean_token_accuracy": 0.1479376584291458,
      "num_tokens": 16638783.0,
      "step": 8500
    },
    {
      "entropy": 6.274936532974243,
      "epoch": 0.8502024291497976,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004936546958814873,
      "loss": 6.123,
      "mean_token_accuracy": 0.1526026099920273,
      "num_tokens": 16647432.0,
      "step": 8505
    },
    {
      "entropy": 6.26911678314209,
      "epoch": 0.8507022542110262,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004936462772058026,
      "loss": 6.1031,
      "mean_token_accuracy": 0.14441432058811188,
      "num_tokens": 16656839.0,
      "step": 8510
    },
    {
      "entropy": 6.282738590240479,
      "epoch": 0.8512020792722547,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004936378530290339,
      "loss": 6.1784,
      "mean_token_accuracy": 0.14249846562743188,
      "num_tokens": 16667054.0,
      "step": 8515
    },
    {
      "entropy": 6.228442764282226,
      "epoch": 0.8517019043334833,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004936294233513931,
      "loss": 6.1053,
      "mean_token_accuracy": 0.15022315680980683,
      "num_tokens": 16676851.0,
      "step": 8520
    },
    {
      "entropy": 6.265562582015991,
      "epoch": 0.8522017293947118,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004936209881730922,
      "loss": 6.1188,
      "mean_token_accuracy": 0.14851039201021193,
      "num_tokens": 16686433.0,
      "step": 8525
    },
    {
      "entropy": 6.201057195663452,
      "epoch": 0.8527015544559404,
      "grad_norm": 1.125,
      "learning_rate": 0.0004936125474943436,
      "loss": 6.0847,
      "mean_token_accuracy": 0.15974876582622527,
      "num_tokens": 16695859.0,
      "step": 8530
    },
    {
      "entropy": 6.212331485748291,
      "epoch": 0.853201379517169,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004936041013153596,
      "loss": 6.1034,
      "mean_token_accuracy": 0.15397711396217345,
      "num_tokens": 16704391.0,
      "step": 8535
    },
    {
      "entropy": 6.24004054069519,
      "epoch": 0.8537012045783976,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004935956496363525,
      "loss": 6.1932,
      "mean_token_accuracy": 0.13731725215911866,
      "num_tokens": 16715316.0,
      "step": 8540
    },
    {
      "entropy": 6.4582024097442625,
      "epoch": 0.8542010296396262,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004935871924575352,
      "loss": 6.2656,
      "mean_token_accuracy": 0.13669400438666343,
      "num_tokens": 16726215.0,
      "step": 8545
    },
    {
      "entropy": 6.302094841003418,
      "epoch": 0.8547008547008547,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004935787297791204,
      "loss": 6.1141,
      "mean_token_accuracy": 0.14495947808027268,
      "num_tokens": 16735768.0,
      "step": 8550
    },
    {
      "entropy": 6.185169887542725,
      "epoch": 0.8552006797620832,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004935702616013209,
      "loss": 5.9287,
      "mean_token_accuracy": 0.1614491894841194,
      "num_tokens": 16744265.0,
      "step": 8555
    },
    {
      "entropy": 6.163142156600952,
      "epoch": 0.8557005048233118,
      "grad_norm": 1.1875,
      "learning_rate": 0.00049356178792435,
      "loss": 6.1718,
      "mean_token_accuracy": 0.14948089718818663,
      "num_tokens": 16753713.0,
      "step": 8560
    },
    {
      "entropy": 6.219892930984497,
      "epoch": 0.8562003298845404,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004935533087484207,
      "loss": 6.1,
      "mean_token_accuracy": 0.15011856108903884,
      "num_tokens": 16763670.0,
      "step": 8565
    },
    {
      "entropy": 6.179305934906006,
      "epoch": 0.856700154945769,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004935448240737464,
      "loss": 6.1215,
      "mean_token_accuracy": 0.150160750746727,
      "num_tokens": 16773504.0,
      "step": 8570
    },
    {
      "entropy": 6.253590393066406,
      "epoch": 0.8571999800069976,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004935363339005405,
      "loss": 6.1766,
      "mean_token_accuracy": 0.14550895914435386,
      "num_tokens": 16784181.0,
      "step": 8575
    },
    {
      "entropy": 6.308042669296265,
      "epoch": 0.8576998050682261,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004935278382290168,
      "loss": 6.1436,
      "mean_token_accuracy": 0.1467001810669899,
      "num_tokens": 16793610.0,
      "step": 8580
    },
    {
      "entropy": 6.327728319168091,
      "epoch": 0.8581996301294547,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004935193370593887,
      "loss": 6.1971,
      "mean_token_accuracy": 0.13698992505669594,
      "num_tokens": 16804068.0,
      "step": 8585
    },
    {
      "entropy": 6.263300371170044,
      "epoch": 0.8586994551906832,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004935108303918705,
      "loss": 6.133,
      "mean_token_accuracy": 0.14942301213741302,
      "num_tokens": 16813769.0,
      "step": 8590
    },
    {
      "entropy": 6.16823992729187,
      "epoch": 0.8591992802519118,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004935023182266761,
      "loss": 6.0653,
      "mean_token_accuracy": 0.15182611346244812,
      "num_tokens": 16822859.0,
      "step": 8595
    },
    {
      "entropy": 6.3461995124816895,
      "epoch": 0.8596991053131404,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004934938005640194,
      "loss": 6.2919,
      "mean_token_accuracy": 0.13834105506539346,
      "num_tokens": 16834839.0,
      "step": 8600
    },
    {
      "entropy": 6.37769079208374,
      "epoch": 0.860198930374369,
      "grad_norm": 1.046875,
      "learning_rate": 0.000493485277404115,
      "loss": 6.1245,
      "mean_token_accuracy": 0.1470717318356037,
      "num_tokens": 16845205.0,
      "step": 8605
    },
    {
      "entropy": 6.324769401550293,
      "epoch": 0.8606987554355976,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004934767487471772,
      "loss": 6.133,
      "mean_token_accuracy": 0.145134799182415,
      "num_tokens": 16854874.0,
      "step": 8610
    },
    {
      "entropy": 6.18726601600647,
      "epoch": 0.8611985804968261,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004934682145934207,
      "loss": 6.0147,
      "mean_token_accuracy": 0.15377903133630752,
      "num_tokens": 16864771.0,
      "step": 8615
    },
    {
      "entropy": 6.228726768493653,
      "epoch": 0.8616984055580547,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00049345967494306,
      "loss": 6.1072,
      "mean_token_accuracy": 0.14587363675236703,
      "num_tokens": 16874227.0,
      "step": 8620
    },
    {
      "entropy": 6.188831520080567,
      "epoch": 0.8621982306192832,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004934511297963102,
      "loss": 6.0402,
      "mean_token_accuracy": 0.15956453680992128,
      "num_tokens": 16883895.0,
      "step": 8625
    },
    {
      "entropy": 6.197385501861572,
      "epoch": 0.8626980556805118,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004934425791533861,
      "loss": 6.0477,
      "mean_token_accuracy": 0.15518876388669015,
      "num_tokens": 16893012.0,
      "step": 8630
    },
    {
      "entropy": 6.290072679519653,
      "epoch": 0.8631978807417404,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004934340230145031,
      "loss": 6.1468,
      "mean_token_accuracy": 0.15736810937523843,
      "num_tokens": 16903831.0,
      "step": 8635
    },
    {
      "entropy": 6.34267749786377,
      "epoch": 0.863697705802969,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004934254613798761,
      "loss": 6.1879,
      "mean_token_accuracy": 0.14001786932349206,
      "num_tokens": 16913649.0,
      "step": 8640
    },
    {
      "entropy": 6.193124485015869,
      "epoch": 0.8641975308641975,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004934168942497207,
      "loss": 6.0805,
      "mean_token_accuracy": 0.15410650297999381,
      "num_tokens": 16923063.0,
      "step": 8645
    },
    {
      "entropy": 6.259519910812378,
      "epoch": 0.8646973559254261,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004934083216242526,
      "loss": 6.075,
      "mean_token_accuracy": 0.15394753217697144,
      "num_tokens": 16932901.0,
      "step": 8650
    },
    {
      "entropy": 6.288390684127807,
      "epoch": 0.8651971809866547,
      "grad_norm": 1.125,
      "learning_rate": 0.0004933997435036873,
      "loss": 6.2018,
      "mean_token_accuracy": 0.14151864424347876,
      "num_tokens": 16942781.0,
      "step": 8655
    },
    {
      "entropy": 6.250697469711303,
      "epoch": 0.8656970060478832,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004933911598882405,
      "loss": 6.0799,
      "mean_token_accuracy": 0.15049338936805726,
      "num_tokens": 16952064.0,
      "step": 8660
    },
    {
      "entropy": 6.175657939910889,
      "epoch": 0.8661968311091118,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004933825707781284,
      "loss": 6.0407,
      "mean_token_accuracy": 0.14716741293668748,
      "num_tokens": 16961528.0,
      "step": 8665
    },
    {
      "entropy": 6.266824865341187,
      "epoch": 0.8666966561703404,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004933739761735671,
      "loss": 6.1279,
      "mean_token_accuracy": 0.14926101565361022,
      "num_tokens": 16971066.0,
      "step": 8670
    },
    {
      "entropy": 6.2522862434387205,
      "epoch": 0.867196481231569,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004933653760747726,
      "loss": 6.1591,
      "mean_token_accuracy": 0.14760765731334685,
      "num_tokens": 16981438.0,
      "step": 8675
    },
    {
      "entropy": 6.2069446563720705,
      "epoch": 0.8676963062927975,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004933567704819615,
      "loss": 6.0339,
      "mean_token_accuracy": 0.1574970930814743,
      "num_tokens": 16990836.0,
      "step": 8680
    },
    {
      "entropy": 6.164374589920044,
      "epoch": 0.8681961313540261,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004933481593953502,
      "loss": 6.0412,
      "mean_token_accuracy": 0.15426290333271026,
      "num_tokens": 17000254.0,
      "step": 8685
    },
    {
      "entropy": 6.204534578323364,
      "epoch": 0.8686959564152547,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004933395428151554,
      "loss": 6.0955,
      "mean_token_accuracy": 0.15164344236254693,
      "num_tokens": 17010081.0,
      "step": 8690
    },
    {
      "entropy": 6.176742935180664,
      "epoch": 0.8691957814764832,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000493330920741594,
      "loss": 5.9573,
      "mean_token_accuracy": 0.15853566378355027,
      "num_tokens": 17018967.0,
      "step": 8695
    },
    {
      "entropy": 6.1779749393463135,
      "epoch": 0.8696956065377118,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004933222931748827,
      "loss": 6.0957,
      "mean_token_accuracy": 0.15252818763256074,
      "num_tokens": 17027615.0,
      "step": 8700
    },
    {
      "entropy": 6.203618764877319,
      "epoch": 0.8701954315989404,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004933136601152387,
      "loss": 6.0629,
      "mean_token_accuracy": 0.15414001643657685,
      "num_tokens": 17038446.0,
      "step": 8705
    },
    {
      "entropy": 6.293272638320923,
      "epoch": 0.8706952566601689,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004933050215628793,
      "loss": 6.1061,
      "mean_token_accuracy": 0.1501946985721588,
      "num_tokens": 17047787.0,
      "step": 8710
    },
    {
      "entropy": 6.245404148101807,
      "epoch": 0.8711950817213975,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004932963775180216,
      "loss": 6.144,
      "mean_token_accuracy": 0.15104712694883346,
      "num_tokens": 17056842.0,
      "step": 8715
    },
    {
      "entropy": 6.170654678344727,
      "epoch": 0.8716949067826261,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004932877279808833,
      "loss": 6.1563,
      "mean_token_accuracy": 0.14996298551559448,
      "num_tokens": 17066217.0,
      "step": 8720
    },
    {
      "entropy": 6.328350019454956,
      "epoch": 0.8721947318438547,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004932790729516818,
      "loss": 6.2049,
      "mean_token_accuracy": 0.13945267349481583,
      "num_tokens": 17077760.0,
      "step": 8725
    },
    {
      "entropy": 6.240718603134155,
      "epoch": 0.8726945569050832,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004932704124306352,
      "loss": 6.0534,
      "mean_token_accuracy": 0.15234245210886002,
      "num_tokens": 17088251.0,
      "step": 8730
    },
    {
      "entropy": 6.25137882232666,
      "epoch": 0.8731943819663118,
      "grad_norm": 1.140625,
      "learning_rate": 0.000493261746417961,
      "loss": 6.1171,
      "mean_token_accuracy": 0.14927061945199965,
      "num_tokens": 17098192.0,
      "step": 8735
    },
    {
      "entropy": 6.215445947647095,
      "epoch": 0.8736942070275404,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004932530749138775,
      "loss": 6.0931,
      "mean_token_accuracy": 0.15051602572202682,
      "num_tokens": 17107787.0,
      "step": 8740
    },
    {
      "entropy": 6.280746603012085,
      "epoch": 0.8741940320887689,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004932443979186028,
      "loss": 6.1814,
      "mean_token_accuracy": 0.14869068935513496,
      "num_tokens": 17118125.0,
      "step": 8745
    },
    {
      "entropy": 6.217524528503418,
      "epoch": 0.8746938571499975,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004932357154323552,
      "loss": 6.0498,
      "mean_token_accuracy": 0.15469740480184554,
      "num_tokens": 17127909.0,
      "step": 8750
    },
    {
      "entropy": 6.201910734176636,
      "epoch": 0.8751936822112261,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004932270274553531,
      "loss": 6.1071,
      "mean_token_accuracy": 0.15080273672938346,
      "num_tokens": 17137672.0,
      "step": 8755
    },
    {
      "entropy": 6.215283918380737,
      "epoch": 0.8756935072724547,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004932183339878153,
      "loss": 6.0877,
      "mean_token_accuracy": 0.1489563003182411,
      "num_tokens": 17147151.0,
      "step": 8760
    },
    {
      "entropy": 6.201694583892822,
      "epoch": 0.8761933323336832,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004932096350299602,
      "loss": 5.9762,
      "mean_token_accuracy": 0.15250695794820784,
      "num_tokens": 17156615.0,
      "step": 8765
    },
    {
      "entropy": 6.283472967147827,
      "epoch": 0.8766931573949118,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004932009305820069,
      "loss": 6.1145,
      "mean_token_accuracy": 0.1526653289794922,
      "num_tokens": 17166089.0,
      "step": 8770
    },
    {
      "entropy": 6.254987001419067,
      "epoch": 0.8771929824561403,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004931922206441743,
      "loss": 6.1276,
      "mean_token_accuracy": 0.15160841271281242,
      "num_tokens": 17176565.0,
      "step": 8775
    },
    {
      "entropy": 6.155684232711792,
      "epoch": 0.8776928075173689,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004931835052166817,
      "loss": 6.055,
      "mean_token_accuracy": 0.15296048074960708,
      "num_tokens": 17186987.0,
      "step": 8780
    },
    {
      "entropy": 6.210090684890747,
      "epoch": 0.8781926325785975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004931747842997481,
      "loss": 6.0409,
      "mean_token_accuracy": 0.15653039515018463,
      "num_tokens": 17196325.0,
      "step": 8785
    },
    {
      "entropy": 6.294657230377197,
      "epoch": 0.8786924576398261,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004931660578935931,
      "loss": 6.1052,
      "mean_token_accuracy": 0.14774607717990876,
      "num_tokens": 17206210.0,
      "step": 8790
    },
    {
      "entropy": 6.245672225952148,
      "epoch": 0.8791922827010547,
      "grad_norm": 1.125,
      "learning_rate": 0.0004931573259984363,
      "loss": 6.0683,
      "mean_token_accuracy": 0.15434812158346176,
      "num_tokens": 17216521.0,
      "step": 8795
    },
    {
      "entropy": 6.135601663589478,
      "epoch": 0.8796921077622832,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004931485886144972,
      "loss": 6.0769,
      "mean_token_accuracy": 0.154274819791317,
      "num_tokens": 17225125.0,
      "step": 8800
    },
    {
      "entropy": 6.213069105148316,
      "epoch": 0.8801919328235118,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004931398457419959,
      "loss": 6.0793,
      "mean_token_accuracy": 0.15032802671194076,
      "num_tokens": 17234674.0,
      "step": 8805
    },
    {
      "entropy": 6.187903165817261,
      "epoch": 0.8806917578847403,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004931310973811521,
      "loss": 6.0477,
      "mean_token_accuracy": 0.15058831870555878,
      "num_tokens": 17244362.0,
      "step": 8810
    },
    {
      "entropy": 6.197688579559326,
      "epoch": 0.8811915829459689,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000493122343532186,
      "loss": 6.0203,
      "mean_token_accuracy": 0.15211349204182625,
      "num_tokens": 17253541.0,
      "step": 8815
    },
    {
      "entropy": 6.202161979675293,
      "epoch": 0.8816914080071975,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000493113584195318,
      "loss": 6.0431,
      "mean_token_accuracy": 0.15537039190530777,
      "num_tokens": 17263214.0,
      "step": 8820
    },
    {
      "entropy": 6.203364276885987,
      "epoch": 0.8821912330684261,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004931048193707681,
      "loss": 6.1152,
      "mean_token_accuracy": 0.14149047806859016,
      "num_tokens": 17273991.0,
      "step": 8825
    },
    {
      "entropy": 6.132715845108033,
      "epoch": 0.8826910581296546,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004930960490587573,
      "loss": 5.9629,
      "mean_token_accuracy": 0.16362378001213074,
      "num_tokens": 17283221.0,
      "step": 8830
    },
    {
      "entropy": 6.1574865818023685,
      "epoch": 0.8831908831908832,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000493087273259506,
      "loss": 6.0903,
      "mean_token_accuracy": 0.1537564665079117,
      "num_tokens": 17293520.0,
      "step": 8835
    },
    {
      "entropy": 6.153212070465088,
      "epoch": 0.8836907082521117,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004930784919732349,
      "loss": 6.0791,
      "mean_token_accuracy": 0.1504323735833168,
      "num_tokens": 17303881.0,
      "step": 8840
    },
    {
      "entropy": 6.209928417205811,
      "epoch": 0.8841905333133403,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004930697052001651,
      "loss": 5.9968,
      "mean_token_accuracy": 0.1615106701850891,
      "num_tokens": 17314115.0,
      "step": 8845
    },
    {
      "entropy": 6.294680595397949,
      "epoch": 0.8846903583745689,
      "grad_norm": 1.125,
      "learning_rate": 0.0004930609129405177,
      "loss": 6.148,
      "mean_token_accuracy": 0.14716485440731047,
      "num_tokens": 17324517.0,
      "step": 8850
    },
    {
      "entropy": 6.2288980960845945,
      "epoch": 0.8851901834357975,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004930521151945138,
      "loss": 6.1241,
      "mean_token_accuracy": 0.14943552017211914,
      "num_tokens": 17334109.0,
      "step": 8855
    },
    {
      "entropy": 6.189024114608765,
      "epoch": 0.8856900084970261,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004930433119623749,
      "loss": 6.1129,
      "mean_token_accuracy": 0.14660831913352013,
      "num_tokens": 17344602.0,
      "step": 8860
    },
    {
      "entropy": 6.218776845932007,
      "epoch": 0.8861898335582546,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004930345032443221,
      "loss": 6.0319,
      "mean_token_accuracy": 0.16084310710430144,
      "num_tokens": 17354476.0,
      "step": 8865
    },
    {
      "entropy": 6.248809337615967,
      "epoch": 0.8866896586194832,
      "grad_norm": 1.25,
      "learning_rate": 0.0004930256890405777,
      "loss": 6.0068,
      "mean_token_accuracy": 0.15692270845174788,
      "num_tokens": 17364748.0,
      "step": 8870
    },
    {
      "entropy": 6.184897327423096,
      "epoch": 0.8871894836807117,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004930168693513627,
      "loss": 6.0862,
      "mean_token_accuracy": 0.14990119487047196,
      "num_tokens": 17374162.0,
      "step": 8875
    },
    {
      "entropy": 6.204228067398072,
      "epoch": 0.8876893087419403,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004930080441768995,
      "loss": 6.1142,
      "mean_token_accuracy": 0.15213522017002107,
      "num_tokens": 17384042.0,
      "step": 8880
    },
    {
      "entropy": 6.193892002105713,
      "epoch": 0.8881891338031689,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00049299921351741,
      "loss": 6.0148,
      "mean_token_accuracy": 0.15456045866012574,
      "num_tokens": 17394181.0,
      "step": 8885
    },
    {
      "entropy": 6.256800889968872,
      "epoch": 0.8886889588643975,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004929903773731164,
      "loss": 6.1869,
      "mean_token_accuracy": 0.14307404756546022,
      "num_tokens": 17402983.0,
      "step": 8890
    },
    {
      "entropy": 6.26918044090271,
      "epoch": 0.8891887839256261,
      "grad_norm": 1.125,
      "learning_rate": 0.0004929815357442409,
      "loss": 6.244,
      "mean_token_accuracy": 0.1416206791996956,
      "num_tokens": 17413442.0,
      "step": 8895
    },
    {
      "entropy": 6.225600051879883,
      "epoch": 0.8896886089868546,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004929726886310062,
      "loss": 6.0618,
      "mean_token_accuracy": 0.14702547788619996,
      "num_tokens": 17423644.0,
      "step": 8900
    },
    {
      "entropy": 6.2391955852508545,
      "epoch": 0.8901884340480831,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004929638360336348,
      "loss": 6.105,
      "mean_token_accuracy": 0.14813730865716934,
      "num_tokens": 17433400.0,
      "step": 8905
    },
    {
      "entropy": 6.2598494529724125,
      "epoch": 0.8906882591093117,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004929549779523492,
      "loss": 6.0794,
      "mean_token_accuracy": 0.14878565967082977,
      "num_tokens": 17443146.0,
      "step": 8910
    },
    {
      "entropy": 6.135402822494507,
      "epoch": 0.8911880841705403,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004929461143873725,
      "loss": 5.9754,
      "mean_token_accuracy": 0.16229354739189147,
      "num_tokens": 17452880.0,
      "step": 8915
    },
    {
      "entropy": 6.118289136886597,
      "epoch": 0.8916879092317689,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004929372453389277,
      "loss": 5.9852,
      "mean_token_accuracy": 0.1584864929318428,
      "num_tokens": 17461633.0,
      "step": 8920
    },
    {
      "entropy": 6.252451086044312,
      "epoch": 0.8921877342929975,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004929283708072378,
      "loss": 6.197,
      "mean_token_accuracy": 0.14696545153856277,
      "num_tokens": 17470892.0,
      "step": 8925
    },
    {
      "entropy": 6.272406673431396,
      "epoch": 0.8926875593542261,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004929194907925262,
      "loss": 6.0799,
      "mean_token_accuracy": 0.14626012295484542,
      "num_tokens": 17481457.0,
      "step": 8930
    },
    {
      "entropy": 6.201140117645264,
      "epoch": 0.8931873844154546,
      "grad_norm": 1.25,
      "learning_rate": 0.0004929106052950163,
      "loss": 6.114,
      "mean_token_accuracy": 0.14438339397311212,
      "num_tokens": 17490387.0,
      "step": 8935
    },
    {
      "entropy": 6.232717752456665,
      "epoch": 0.8936872094766831,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004929017143149317,
      "loss": 6.0703,
      "mean_token_accuracy": 0.1581308998167515,
      "num_tokens": 17499478.0,
      "step": 8940
    },
    {
      "entropy": 6.328111362457276,
      "epoch": 0.8941870345379117,
      "grad_norm": 1.375,
      "learning_rate": 0.000492892817852496,
      "loss": 6.1726,
      "mean_token_accuracy": 0.14408332109451294,
      "num_tokens": 17509396.0,
      "step": 8945
    },
    {
      "entropy": 6.1822426319122314,
      "epoch": 0.8946868595991403,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000492883915907933,
      "loss": 6.1095,
      "mean_token_accuracy": 0.15396877378225327,
      "num_tokens": 17519773.0,
      "step": 8950
    },
    {
      "entropy": 6.2347657680511475,
      "epoch": 0.8951866846603689,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004928750084814668,
      "loss": 6.0907,
      "mean_token_accuracy": 0.15340853333473206,
      "num_tokens": 17529873.0,
      "step": 8955
    },
    {
      "entropy": 6.177518844604492,
      "epoch": 0.8956865097215975,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004928660955733214,
      "loss": 6.0196,
      "mean_token_accuracy": 0.1571214094758034,
      "num_tokens": 17538642.0,
      "step": 8960
    },
    {
      "entropy": 6.190900897979736,
      "epoch": 0.896186334782826,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004928571771837211,
      "loss": 6.0738,
      "mean_token_accuracy": 0.14259648099541664,
      "num_tokens": 17546892.0,
      "step": 8965
    },
    {
      "entropy": 6.308866930007935,
      "epoch": 0.8966861598440545,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004928482533128902,
      "loss": 6.1195,
      "mean_token_accuracy": 0.14720739051699638,
      "num_tokens": 17556283.0,
      "step": 8970
    },
    {
      "entropy": 6.162022542953491,
      "epoch": 0.8971859849052831,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004928393239610535,
      "loss": 6.049,
      "mean_token_accuracy": 0.1500636763870716,
      "num_tokens": 17567021.0,
      "step": 8975
    },
    {
      "entropy": 6.200020217895508,
      "epoch": 0.8976858099665117,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004928303891284354,
      "loss": 6.0609,
      "mean_token_accuracy": 0.15545229315757753,
      "num_tokens": 17575927.0,
      "step": 8980
    },
    {
      "entropy": 6.229860067367554,
      "epoch": 0.8981856350277403,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004928214488152607,
      "loss": 6.0909,
      "mean_token_accuracy": 0.15287247747182847,
      "num_tokens": 17585882.0,
      "step": 8985
    },
    {
      "entropy": 6.155755138397216,
      "epoch": 0.8986854600889689,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004928125030217545,
      "loss": 6.0445,
      "mean_token_accuracy": 0.15322572737932205,
      "num_tokens": 17595998.0,
      "step": 8990
    },
    {
      "entropy": 6.3029591083526615,
      "epoch": 0.8991852851501975,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004928035517481416,
      "loss": 6.0573,
      "mean_token_accuracy": 0.1525214947760105,
      "num_tokens": 17605177.0,
      "step": 8995
    },
    {
      "entropy": 6.2255286693573,
      "epoch": 0.899685110211426,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004927945949946476,
      "loss": 6.1737,
      "mean_token_accuracy": 0.14645436704158782,
      "num_tokens": 17615089.0,
      "step": 9000
    },
    {
      "epoch": 0.899685110211426,
      "eval_entropy": 6.062303036889045,
      "eval_loss": 6.13262939453125,
      "eval_mean_token_accuracy": 0.15480020759817759,
      "eval_num_tokens": 17615089.0,
      "eval_runtime": 23.7516,
      "eval_samples_per_second": 1307.072,
      "eval_steps_per_second": 163.4,
      "step": 9000
    },
    {
      "entropy": 6.227158498764038,
      "epoch": 0.9001849352726545,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004927856327614975,
      "loss": 6.0549,
      "mean_token_accuracy": 0.14798942357301711,
      "num_tokens": 17624708.0,
      "step": 9005
    },
    {
      "entropy": 6.266047477722168,
      "epoch": 0.9006847603338831,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000492776665048917,
      "loss": 6.091,
      "mean_token_accuracy": 0.14740900099277496,
      "num_tokens": 17633639.0,
      "step": 9010
    },
    {
      "entropy": 6.180922746658325,
      "epoch": 0.9011845853951117,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004927676918571316,
      "loss": 6.105,
      "mean_token_accuracy": 0.1465431958436966,
      "num_tokens": 17643092.0,
      "step": 9015
    },
    {
      "entropy": 6.20935320854187,
      "epoch": 0.9016844104563403,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004927587131863671,
      "loss": 6.1305,
      "mean_token_accuracy": 0.14210475981235504,
      "num_tokens": 17652698.0,
      "step": 9020
    },
    {
      "entropy": 6.187350702285767,
      "epoch": 0.9021842355175689,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004927497290368493,
      "loss": 5.9898,
      "mean_token_accuracy": 0.15789729505777358,
      "num_tokens": 17662654.0,
      "step": 9025
    },
    {
      "entropy": 6.208523464202881,
      "epoch": 0.9026840605787975,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004927407394088046,
      "loss": 6.0425,
      "mean_token_accuracy": 0.14684979766607284,
      "num_tokens": 17672476.0,
      "step": 9030
    },
    {
      "entropy": 6.205368137359619,
      "epoch": 0.9031838856400259,
      "grad_norm": 1.125,
      "learning_rate": 0.0004927317443024587,
      "loss": 6.1168,
      "mean_token_accuracy": 0.14842985495924949,
      "num_tokens": 17682612.0,
      "step": 9035
    },
    {
      "entropy": 6.302953386306763,
      "epoch": 0.9036837107012545,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004927227437180382,
      "loss": 6.1039,
      "mean_token_accuracy": 0.1407620817422867,
      "num_tokens": 17692020.0,
      "step": 9040
    },
    {
      "entropy": 6.222799825668335,
      "epoch": 0.9041835357624831,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004927137376557696,
      "loss": 6.0921,
      "mean_token_accuracy": 0.15118202418088914,
      "num_tokens": 17702677.0,
      "step": 9045
    },
    {
      "entropy": 6.146578073501587,
      "epoch": 0.9046833608237117,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004927047261158793,
      "loss": 5.9514,
      "mean_token_accuracy": 0.15856276080012321,
      "num_tokens": 17711753.0,
      "step": 9050
    },
    {
      "entropy": 6.212048864364624,
      "epoch": 0.9051831858849403,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004926957090985942,
      "loss": 6.0836,
      "mean_token_accuracy": 0.14732278734445572,
      "num_tokens": 17721293.0,
      "step": 9055
    },
    {
      "entropy": 6.29421591758728,
      "epoch": 0.9056830109461689,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004926866866041409,
      "loss": 6.1351,
      "mean_token_accuracy": 0.14667329788208008,
      "num_tokens": 17732311.0,
      "step": 9060
    },
    {
      "entropy": 6.222735643386841,
      "epoch": 0.9061828360073975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004926776586327467,
      "loss": 6.0504,
      "mean_token_accuracy": 0.15361883193254472,
      "num_tokens": 17743040.0,
      "step": 9065
    },
    {
      "entropy": 6.200742769241333,
      "epoch": 0.9066826610686259,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004926686251846386,
      "loss": 6.0054,
      "mean_token_accuracy": 0.16133202016353607,
      "num_tokens": 17752436.0,
      "step": 9070
    },
    {
      "entropy": 6.188657999038696,
      "epoch": 0.9071824861298545,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004926595862600439,
      "loss": 6.0473,
      "mean_token_accuracy": 0.14954303056001664,
      "num_tokens": 17762043.0,
      "step": 9075
    },
    {
      "entropy": 6.198422050476074,
      "epoch": 0.9076823111910831,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004926505418591901,
      "loss": 6.067,
      "mean_token_accuracy": 0.15320897549390794,
      "num_tokens": 17772065.0,
      "step": 9080
    },
    {
      "entropy": 6.1272553443908695,
      "epoch": 0.9081821362523117,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004926414919823045,
      "loss": 5.9823,
      "mean_token_accuracy": 0.15631068795919417,
      "num_tokens": 17781472.0,
      "step": 9085
    },
    {
      "entropy": 6.249554634094238,
      "epoch": 0.9086819613135403,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004926324366296151,
      "loss": 6.0598,
      "mean_token_accuracy": 0.15021432638168336,
      "num_tokens": 17789435.0,
      "step": 9090
    },
    {
      "entropy": 6.261883354187011,
      "epoch": 0.9091817863747689,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004926233758013494,
      "loss": 6.0553,
      "mean_token_accuracy": 0.15010529309511184,
      "num_tokens": 17798367.0,
      "step": 9095
    },
    {
      "entropy": 6.243432426452637,
      "epoch": 0.9096816114359974,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004926143094977359,
      "loss": 6.0652,
      "mean_token_accuracy": 0.155719193816185,
      "num_tokens": 17807660.0,
      "step": 9100
    },
    {
      "entropy": 6.132892608642578,
      "epoch": 0.9101814364972259,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004926052377190022,
      "loss": 6.0739,
      "mean_token_accuracy": 0.15484517440199852,
      "num_tokens": 17816993.0,
      "step": 9105
    },
    {
      "entropy": 6.231070613861084,
      "epoch": 0.9106812615584545,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004925961604653765,
      "loss": 6.044,
      "mean_token_accuracy": 0.1521976485848427,
      "num_tokens": 17826570.0,
      "step": 9110
    },
    {
      "entropy": 6.2752138614654545,
      "epoch": 0.9111810866196831,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004925870777370877,
      "loss": 6.0502,
      "mean_token_accuracy": 0.15322708189487458,
      "num_tokens": 17835921.0,
      "step": 9115
    },
    {
      "entropy": 6.181320285797119,
      "epoch": 0.9116809116809117,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004925779895343639,
      "loss": 6.0751,
      "mean_token_accuracy": 0.15409771651029586,
      "num_tokens": 17845713.0,
      "step": 9120
    },
    {
      "entropy": 6.2090047836303714,
      "epoch": 0.9121807367421403,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004925688958574339,
      "loss": 6.0939,
      "mean_token_accuracy": 0.14878985062241554,
      "num_tokens": 17855808.0,
      "step": 9125
    },
    {
      "entropy": 6.284414863586425,
      "epoch": 0.9126805618033689,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004925597967065264,
      "loss": 6.0727,
      "mean_token_accuracy": 0.1510001540184021,
      "num_tokens": 17865427.0,
      "step": 9130
    },
    {
      "entropy": 6.255964517593384,
      "epoch": 0.9131803868645973,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004925506920818705,
      "loss": 6.1278,
      "mean_token_accuracy": 0.14633646607398987,
      "num_tokens": 17876371.0,
      "step": 9135
    },
    {
      "entropy": 6.2210314750671385,
      "epoch": 0.9136802119258259,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004925415819836951,
      "loss": 6.033,
      "mean_token_accuracy": 0.15429428219795227,
      "num_tokens": 17886050.0,
      "step": 9140
    },
    {
      "entropy": 6.271884965896606,
      "epoch": 0.9141800369870545,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004925324664122294,
      "loss": 6.0949,
      "mean_token_accuracy": 0.15774204432964326,
      "num_tokens": 17896509.0,
      "step": 9145
    },
    {
      "entropy": 6.31046462059021,
      "epoch": 0.9146798620482831,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004925233453677029,
      "loss": 6.1222,
      "mean_token_accuracy": 0.14721079021692277,
      "num_tokens": 17906242.0,
      "step": 9150
    },
    {
      "entropy": 6.20795841217041,
      "epoch": 0.9151796871095117,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004925142188503451,
      "loss": 6.0742,
      "mean_token_accuracy": 0.15292910113930702,
      "num_tokens": 17916261.0,
      "step": 9155
    },
    {
      "entropy": 6.224349546432495,
      "epoch": 0.9156795121707403,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004925050868603854,
      "loss": 6.0512,
      "mean_token_accuracy": 0.15091956555843353,
      "num_tokens": 17926486.0,
      "step": 9160
    },
    {
      "entropy": 6.202247762680054,
      "epoch": 0.9161793372319688,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004924959493980537,
      "loss": 6.0908,
      "mean_token_accuracy": 0.14863710701465607,
      "num_tokens": 17936374.0,
      "step": 9165
    },
    {
      "entropy": 6.360231590270996,
      "epoch": 0.9166791622931973,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004924868064635798,
      "loss": 6.326,
      "mean_token_accuracy": 0.14041147902607917,
      "num_tokens": 17946930.0,
      "step": 9170
    },
    {
      "entropy": 6.400591468811035,
      "epoch": 0.9171789873544259,
      "grad_norm": 1.21875,
      "learning_rate": 0.000492477658057194,
      "loss": 6.0971,
      "mean_token_accuracy": 0.1520356610417366,
      "num_tokens": 17955760.0,
      "step": 9175
    },
    {
      "entropy": 6.214747476577759,
      "epoch": 0.9176788124156545,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004924685041791262,
      "loss": 6.0254,
      "mean_token_accuracy": 0.15658829361200333,
      "num_tokens": 17965871.0,
      "step": 9180
    },
    {
      "entropy": 6.073985385894775,
      "epoch": 0.9181786374768831,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004924593448296068,
      "loss": 5.9763,
      "mean_token_accuracy": 0.15908344835042953,
      "num_tokens": 17974551.0,
      "step": 9185
    },
    {
      "entropy": 6.224821901321411,
      "epoch": 0.9186784625381117,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004924501800088662,
      "loss": 6.0994,
      "mean_token_accuracy": 0.14992304444313048,
      "num_tokens": 17983675.0,
      "step": 9190
    },
    {
      "entropy": 6.225045442581177,
      "epoch": 0.9191782875993403,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004924410097171351,
      "loss": 6.0533,
      "mean_token_accuracy": 0.1490116223692894,
      "num_tokens": 17993299.0,
      "step": 9195
    },
    {
      "entropy": 6.238255834579467,
      "epoch": 0.9196781126605688,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000492431833954644,
      "loss": 6.0972,
      "mean_token_accuracy": 0.15273767784237863,
      "num_tokens": 18004755.0,
      "step": 9200
    },
    {
      "entropy": 6.265366840362549,
      "epoch": 0.9201779377217973,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004924226527216241,
      "loss": 6.1686,
      "mean_token_accuracy": 0.14311764389276505,
      "num_tokens": 18014908.0,
      "step": 9205
    },
    {
      "entropy": 6.221875190734863,
      "epoch": 0.9206777627830259,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004924134660183061,
      "loss": 6.0423,
      "mean_token_accuracy": 0.1489653095602989,
      "num_tokens": 18023919.0,
      "step": 9210
    },
    {
      "entropy": 6.185988426208496,
      "epoch": 0.9211775878442545,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004924042738449212,
      "loss": 6.0064,
      "mean_token_accuracy": 0.15201392471790315,
      "num_tokens": 18033706.0,
      "step": 9215
    },
    {
      "entropy": 6.238229036331177,
      "epoch": 0.9216774129054831,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004923950762017008,
      "loss": 6.1105,
      "mean_token_accuracy": 0.1419001430273056,
      "num_tokens": 18043171.0,
      "step": 9220
    },
    {
      "entropy": 6.148072671890259,
      "epoch": 0.9221772379667117,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004923858730888762,
      "loss": 6.0257,
      "mean_token_accuracy": 0.16217175126075745,
      "num_tokens": 18052567.0,
      "step": 9225
    },
    {
      "entropy": 6.206769943237305,
      "epoch": 0.9226770630279402,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000492376664506679,
      "loss": 6.0421,
      "mean_token_accuracy": 0.15789373070001603,
      "num_tokens": 18061648.0,
      "step": 9230
    },
    {
      "entropy": 6.173633670806884,
      "epoch": 0.9231768880891688,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004923674504553408,
      "loss": 6.082,
      "mean_token_accuracy": 0.14862751811742783,
      "num_tokens": 18071079.0,
      "step": 9235
    },
    {
      "entropy": 6.180024766921997,
      "epoch": 0.9236767131503973,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004923582309350935,
      "loss": 6.1157,
      "mean_token_accuracy": 0.1496698834002018,
      "num_tokens": 18081162.0,
      "step": 9240
    },
    {
      "entropy": 6.295086765289307,
      "epoch": 0.9241765382116259,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000492349005946169,
      "loss": 6.0739,
      "mean_token_accuracy": 0.1535596191883087,
      "num_tokens": 18090072.0,
      "step": 9245
    },
    {
      "entropy": 6.276718282699585,
      "epoch": 0.9246763632728545,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004923397754887994,
      "loss": 6.1262,
      "mean_token_accuracy": 0.1434614598751068,
      "num_tokens": 18100565.0,
      "step": 9250
    },
    {
      "entropy": 6.18240213394165,
      "epoch": 0.9251761883340831,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004923305395632172,
      "loss": 6.064,
      "mean_token_accuracy": 0.1479712188243866,
      "num_tokens": 18109795.0,
      "step": 9255
    },
    {
      "entropy": 6.251630878448486,
      "epoch": 0.9256760133953117,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004923212981696543,
      "loss": 6.0983,
      "mean_token_accuracy": 0.14276359751820564,
      "num_tokens": 18118821.0,
      "step": 9260
    },
    {
      "entropy": 6.240859127044677,
      "epoch": 0.9261758384565402,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004923120513083436,
      "loss": 6.1139,
      "mean_token_accuracy": 0.15060005635023116,
      "num_tokens": 18128709.0,
      "step": 9265
    },
    {
      "entropy": 6.211163425445557,
      "epoch": 0.9266756635177688,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004923027989795175,
      "loss": 6.1506,
      "mean_token_accuracy": 0.1423851877450943,
      "num_tokens": 18138134.0,
      "step": 9270
    },
    {
      "entropy": 6.215956735610962,
      "epoch": 0.9271754885789973,
      "grad_norm": 1.25,
      "learning_rate": 0.000492293541183409,
      "loss": 6.0039,
      "mean_token_accuracy": 0.1499661237001419,
      "num_tokens": 18147455.0,
      "step": 9275
    },
    {
      "entropy": 6.1663487434387205,
      "epoch": 0.9276753136402259,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004922842779202509,
      "loss": 6.0654,
      "mean_token_accuracy": 0.15445779711008073,
      "num_tokens": 18157787.0,
      "step": 9280
    },
    {
      "entropy": 6.101349925994873,
      "epoch": 0.9281751387014545,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004922750091902763,
      "loss": 6.0371,
      "mean_token_accuracy": 0.15597591251134874,
      "num_tokens": 18167607.0,
      "step": 9285
    },
    {
      "entropy": 6.154468965530396,
      "epoch": 0.9286749637626831,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004922657349937183,
      "loss": 6.0907,
      "mean_token_accuracy": 0.14670682772994043,
      "num_tokens": 18178927.0,
      "step": 9290
    },
    {
      "entropy": 6.198839139938355,
      "epoch": 0.9291747888239116,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004922564553308103,
      "loss": 5.9932,
      "mean_token_accuracy": 0.15368764400482177,
      "num_tokens": 18188907.0,
      "step": 9295
    },
    {
      "entropy": 6.236937189102173,
      "epoch": 0.9296746138851402,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000492247170201786,
      "loss": 5.9557,
      "mean_token_accuracy": 0.15755588859319686,
      "num_tokens": 18198937.0,
      "step": 9300
    },
    {
      "entropy": 6.173686170578003,
      "epoch": 0.9301744389463688,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004922378796068785,
      "loss": 6.0204,
      "mean_token_accuracy": 0.1534600168466568,
      "num_tokens": 18209306.0,
      "step": 9305
    },
    {
      "entropy": 6.2713686466217045,
      "epoch": 0.9306742640075973,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000492228583546322,
      "loss": 6.1766,
      "mean_token_accuracy": 0.13778482079505922,
      "num_tokens": 18219665.0,
      "step": 9310
    },
    {
      "entropy": 6.227418327331543,
      "epoch": 0.9311740890688259,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004922192820203502,
      "loss": 6.0532,
      "mean_token_accuracy": 0.15062994211912156,
      "num_tokens": 18229044.0,
      "step": 9315
    },
    {
      "entropy": 6.218758392333984,
      "epoch": 0.9316739141300545,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004922099750291972,
      "loss": 6.0126,
      "mean_token_accuracy": 0.1531136766076088,
      "num_tokens": 18239373.0,
      "step": 9320
    },
    {
      "entropy": 6.202882862091064,
      "epoch": 0.9321737391912831,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004922006625730971,
      "loss": 6.0141,
      "mean_token_accuracy": 0.15455719977617263,
      "num_tokens": 18249195.0,
      "step": 9325
    },
    {
      "entropy": 6.2043242931365965,
      "epoch": 0.9326735642525116,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004921913446522843,
      "loss": 6.0332,
      "mean_token_accuracy": 0.15760938823223114,
      "num_tokens": 18258399.0,
      "step": 9330
    },
    {
      "entropy": 6.181161642074585,
      "epoch": 0.9331733893137402,
      "grad_norm": 1.796875,
      "learning_rate": 0.000492182021266993,
      "loss": 6.0658,
      "mean_token_accuracy": 0.15996170789003372,
      "num_tokens": 18268952.0,
      "step": 9335
    },
    {
      "entropy": 6.232255935668945,
      "epoch": 0.9336732143749688,
      "grad_norm": 1.109375,
      "learning_rate": 0.000492172692417458,
      "loss": 6.0109,
      "mean_token_accuracy": 0.15202466472983361,
      "num_tokens": 18279482.0,
      "step": 9340
    },
    {
      "entropy": 6.171581363677978,
      "epoch": 0.9341730394361973,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000492163358103914,
      "loss": 6.048,
      "mean_token_accuracy": 0.1556439533829689,
      "num_tokens": 18289069.0,
      "step": 9345
    },
    {
      "entropy": 6.1612170219421385,
      "epoch": 0.9346728644974259,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004921540183265957,
      "loss": 6.0083,
      "mean_token_accuracy": 0.15825594663619996,
      "num_tokens": 18298333.0,
      "step": 9350
    },
    {
      "entropy": 6.219698524475097,
      "epoch": 0.9351726895586545,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004921446730857383,
      "loss": 6.0842,
      "mean_token_accuracy": 0.1495981402695179,
      "num_tokens": 18307661.0,
      "step": 9355
    },
    {
      "entropy": 6.243803644180298,
      "epoch": 0.935672514619883,
      "grad_norm": 1.125,
      "learning_rate": 0.0004921353223815767,
      "loss": 6.1315,
      "mean_token_accuracy": 0.1451036177575588,
      "num_tokens": 18317786.0,
      "step": 9360
    },
    {
      "entropy": 6.273637390136718,
      "epoch": 0.9361723396811116,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004921259662143463,
      "loss": 6.1416,
      "mean_token_accuracy": 0.14351293295621873,
      "num_tokens": 18328430.0,
      "step": 9365
    },
    {
      "entropy": 6.272023296356201,
      "epoch": 0.9366721647423402,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004921166045842825,
      "loss": 6.1309,
      "mean_token_accuracy": 0.14674389064311982,
      "num_tokens": 18339800.0,
      "step": 9370
    },
    {
      "entropy": 6.393236303329468,
      "epoch": 0.9371719898035688,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004921072374916208,
      "loss": 6.1332,
      "mean_token_accuracy": 0.140720172226429,
      "num_tokens": 18348819.0,
      "step": 9375
    },
    {
      "entropy": 6.189776086807251,
      "epoch": 0.9376718148647973,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004920978649365968,
      "loss": 5.9813,
      "mean_token_accuracy": 0.15711806640028952,
      "num_tokens": 18358189.0,
      "step": 9380
    },
    {
      "entropy": 6.152691268920899,
      "epoch": 0.9381716399260259,
      "grad_norm": 1.25,
      "learning_rate": 0.0004920884869194463,
      "loss": 5.9937,
      "mean_token_accuracy": 0.15361985564231873,
      "num_tokens": 18367571.0,
      "step": 9385
    },
    {
      "entropy": 6.173201227188111,
      "epoch": 0.9386714649872545,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004920791034404054,
      "loss": 5.9982,
      "mean_token_accuracy": 0.1540670596063137,
      "num_tokens": 18376699.0,
      "step": 9390
    },
    {
      "entropy": 6.160214853286743,
      "epoch": 0.939171290048483,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004920697144997102,
      "loss": 5.8852,
      "mean_token_accuracy": 0.15974242091178895,
      "num_tokens": 18386036.0,
      "step": 9395
    },
    {
      "entropy": 6.142145013809204,
      "epoch": 0.9396711151097116,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004920603200975968,
      "loss": 6.0859,
      "mean_token_accuracy": 0.16132250279188157,
      "num_tokens": 18395363.0,
      "step": 9400
    },
    {
      "entropy": 6.201982164382935,
      "epoch": 0.9401709401709402,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004920509202343016,
      "loss": 6.0561,
      "mean_token_accuracy": 0.14631023555994033,
      "num_tokens": 18404919.0,
      "step": 9405
    },
    {
      "entropy": 6.22893500328064,
      "epoch": 0.9406707652321687,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000492041514910061,
      "loss": 6.0945,
      "mean_token_accuracy": 0.1527450531721115,
      "num_tokens": 18413520.0,
      "step": 9410
    },
    {
      "entropy": 6.265496873855591,
      "epoch": 0.9411705902933973,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004920321041251118,
      "loss": 6.1186,
      "mean_token_accuracy": 0.14995884895324707,
      "num_tokens": 18422942.0,
      "step": 9415
    },
    {
      "entropy": 6.255853652954102,
      "epoch": 0.9416704153546259,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004920226878796908,
      "loss": 6.0998,
      "mean_token_accuracy": 0.1469413846731186,
      "num_tokens": 18432782.0,
      "step": 9420
    },
    {
      "entropy": 6.2463826656341555,
      "epoch": 0.9421702404158544,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004920132661740348,
      "loss": 6.0929,
      "mean_token_accuracy": 0.1527676612138748,
      "num_tokens": 18443058.0,
      "step": 9425
    },
    {
      "entropy": 6.251214075088501,
      "epoch": 0.942670065477083,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004920038390083808,
      "loss": 5.9952,
      "mean_token_accuracy": 0.1557613119482994,
      "num_tokens": 18453138.0,
      "step": 9430
    },
    {
      "entropy": 6.175073432922363,
      "epoch": 0.9431698905383116,
      "grad_norm": 1.1875,
      "learning_rate": 0.000491994406382966,
      "loss": 6.125,
      "mean_token_accuracy": 0.1522134616971016,
      "num_tokens": 18463763.0,
      "step": 9435
    },
    {
      "entropy": 6.223813581466675,
      "epoch": 0.9436697155995402,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004919849682980278,
      "loss": 6.0259,
      "mean_token_accuracy": 0.15249929428100586,
      "num_tokens": 18473391.0,
      "step": 9440
    },
    {
      "entropy": 6.205363845825195,
      "epoch": 0.9441695406607687,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004919755247538037,
      "loss": 6.0567,
      "mean_token_accuracy": 0.1535480260848999,
      "num_tokens": 18483002.0,
      "step": 9445
    },
    {
      "entropy": 6.262763023376465,
      "epoch": 0.9446693657219973,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004919660757505312,
      "loss": 6.0644,
      "mean_token_accuracy": 0.1508279949426651,
      "num_tokens": 18493428.0,
      "step": 9450
    },
    {
      "entropy": 6.301561117172241,
      "epoch": 0.9451691907832259,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004919566212884481,
      "loss": 6.0371,
      "mean_token_accuracy": 0.1488076627254486,
      "num_tokens": 18502756.0,
      "step": 9455
    },
    {
      "entropy": 6.226949214935303,
      "epoch": 0.9456690158444544,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004919471613677921,
      "loss": 6.1132,
      "mean_token_accuracy": 0.15251423120498658,
      "num_tokens": 18512576.0,
      "step": 9460
    },
    {
      "entropy": 6.174415445327758,
      "epoch": 0.946168840905683,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004919376959888015,
      "loss": 5.9776,
      "mean_token_accuracy": 0.15120429545640945,
      "num_tokens": 18522271.0,
      "step": 9465
    },
    {
      "entropy": 6.140515375137329,
      "epoch": 0.9466686659669116,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004919282251517142,
      "loss": 5.9853,
      "mean_token_accuracy": 0.15275261849164962,
      "num_tokens": 18532738.0,
      "step": 9470
    },
    {
      "entropy": 6.235359191894531,
      "epoch": 0.9471684910281402,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004919187488567686,
      "loss": 6.1879,
      "mean_token_accuracy": 0.15060618668794631,
      "num_tokens": 18543132.0,
      "step": 9475
    },
    {
      "entropy": 6.179693937301636,
      "epoch": 0.9476683160893687,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004919092671042031,
      "loss": 5.9717,
      "mean_token_accuracy": 0.15885238349437714,
      "num_tokens": 18551777.0,
      "step": 9480
    },
    {
      "entropy": 6.148458194732666,
      "epoch": 0.9481681411505973,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004918997798942562,
      "loss": 6.0357,
      "mean_token_accuracy": 0.15350543707609177,
      "num_tokens": 18561137.0,
      "step": 9485
    },
    {
      "entropy": 6.186297750473022,
      "epoch": 0.9486679662118258,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004918902872271668,
      "loss": 6.079,
      "mean_token_accuracy": 0.14984983950853348,
      "num_tokens": 18571917.0,
      "step": 9490
    },
    {
      "entropy": 6.294495487213135,
      "epoch": 0.9491677912730544,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004918807891031734,
      "loss": 6.0552,
      "mean_token_accuracy": 0.1536863252520561,
      "num_tokens": 18581670.0,
      "step": 9495
    },
    {
      "entropy": 6.1549969673156735,
      "epoch": 0.949667616334283,
      "grad_norm": 1.125,
      "learning_rate": 0.0004918712855225154,
      "loss": 5.988,
      "mean_token_accuracy": 0.1552504613995552,
      "num_tokens": 18590140.0,
      "step": 9500
    },
    {
      "entropy": 6.08001298904419,
      "epoch": 0.9501674413955116,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004918617764854314,
      "loss": 5.8939,
      "mean_token_accuracy": 0.15390661656856536,
      "num_tokens": 18599496.0,
      "step": 9505
    },
    {
      "entropy": 6.166283464431762,
      "epoch": 0.9506672664567402,
      "grad_norm": 1.15625,
      "learning_rate": 0.000491852261992161,
      "loss": 6.0718,
      "mean_token_accuracy": 0.15903998166322708,
      "num_tokens": 18609415.0,
      "step": 9510
    },
    {
      "entropy": 6.233508920669555,
      "epoch": 0.9511670915179687,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004918427420429435,
      "loss": 6.0676,
      "mean_token_accuracy": 0.15113556012511253,
      "num_tokens": 18619215.0,
      "step": 9515
    },
    {
      "entropy": 6.294767999649048,
      "epoch": 0.9516669165791973,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004918332166380185,
      "loss": 6.1344,
      "mean_token_accuracy": 0.15008866935968398,
      "num_tokens": 18629381.0,
      "step": 9520
    },
    {
      "entropy": 6.202641391754151,
      "epoch": 0.9521667416404258,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004918236857776254,
      "loss": 6.064,
      "mean_token_accuracy": 0.1529494121670723,
      "num_tokens": 18639049.0,
      "step": 9525
    },
    {
      "entropy": 6.23467264175415,
      "epoch": 0.9526665667016544,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004918141494620042,
      "loss": 6.0697,
      "mean_token_accuracy": 0.14923292100429536,
      "num_tokens": 18649096.0,
      "step": 9530
    },
    {
      "entropy": 6.217487096786499,
      "epoch": 0.953166391762883,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004918046076913949,
      "loss": 6.0411,
      "mean_token_accuracy": 0.1568130984902382,
      "num_tokens": 18658747.0,
      "step": 9535
    },
    {
      "entropy": 6.2200665950775145,
      "epoch": 0.9536662168241116,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004917950604660374,
      "loss": 6.112,
      "mean_token_accuracy": 0.14543188661336898,
      "num_tokens": 18668538.0,
      "step": 9540
    },
    {
      "entropy": 6.2033202171325685,
      "epoch": 0.9541660418853402,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000491785507786172,
      "loss": 6.0757,
      "mean_token_accuracy": 0.14450264424085618,
      "num_tokens": 18679241.0,
      "step": 9545
    },
    {
      "entropy": 6.2209583759307865,
      "epoch": 0.9546658669465687,
      "grad_norm": 1.125,
      "learning_rate": 0.0004917759496520388,
      "loss": 6.07,
      "mean_token_accuracy": 0.15148010551929475,
      "num_tokens": 18689405.0,
      "step": 9550
    },
    {
      "entropy": 6.269361257553101,
      "epoch": 0.9551656920077972,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004917663860638787,
      "loss": 6.1374,
      "mean_token_accuracy": 0.14381617680191994,
      "num_tokens": 18697989.0,
      "step": 9555
    },
    {
      "entropy": 6.156712055206299,
      "epoch": 0.9556655170690258,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004917568170219321,
      "loss": 6.0157,
      "mean_token_accuracy": 0.15233486592769624,
      "num_tokens": 18706992.0,
      "step": 9560
    },
    {
      "entropy": 6.08630747795105,
      "epoch": 0.9561653421302544,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004917472425264397,
      "loss": 6.0519,
      "mean_token_accuracy": 0.15060753449797631,
      "num_tokens": 18716742.0,
      "step": 9565
    },
    {
      "entropy": 6.22516131401062,
      "epoch": 0.956665167191483,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004917376625776425,
      "loss": 6.1766,
      "mean_token_accuracy": 0.14741776138544083,
      "num_tokens": 18725955.0,
      "step": 9570
    },
    {
      "entropy": 6.30703387260437,
      "epoch": 0.9571649922527116,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004917280771757814,
      "loss": 6.0752,
      "mean_token_accuracy": 0.15511379688978194,
      "num_tokens": 18735199.0,
      "step": 9575
    },
    {
      "entropy": 6.286007452011108,
      "epoch": 0.9576648173139402,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004917184863210977,
      "loss": 6.1053,
      "mean_token_accuracy": 0.14804236739873886,
      "num_tokens": 18745457.0,
      "step": 9580
    },
    {
      "entropy": 6.25141167640686,
      "epoch": 0.9581646423751687,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004917088900138326,
      "loss": 6.0956,
      "mean_token_accuracy": 0.151437346637249,
      "num_tokens": 18753869.0,
      "step": 9585
    },
    {
      "entropy": 6.160632944107055,
      "epoch": 0.9586644674363972,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004916992882542276,
      "loss": 6.0647,
      "mean_token_accuracy": 0.15526636093854904,
      "num_tokens": 18763022.0,
      "step": 9590
    },
    {
      "entropy": 6.199096584320069,
      "epoch": 0.9591642924976258,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004916896810425244,
      "loss": 5.9786,
      "mean_token_accuracy": 0.158399361371994,
      "num_tokens": 18772827.0,
      "step": 9595
    },
    {
      "entropy": 6.1958331108093265,
      "epoch": 0.9596641175588544,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004916800683789645,
      "loss": 6.0261,
      "mean_token_accuracy": 0.15501539707183837,
      "num_tokens": 18782272.0,
      "step": 9600
    },
    {
      "entropy": 6.15570764541626,
      "epoch": 0.960163942620083,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004916704502637897,
      "loss": 6.0338,
      "mean_token_accuracy": 0.15215525776147842,
      "num_tokens": 18792125.0,
      "step": 9605
    },
    {
      "entropy": 6.159744930267334,
      "epoch": 0.9606637676813116,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004916608266972422,
      "loss": 6.0088,
      "mean_token_accuracy": 0.16090773344039916,
      "num_tokens": 18801498.0,
      "step": 9610
    },
    {
      "entropy": 6.249885988235474,
      "epoch": 0.9611635927425402,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004916511976795642,
      "loss": 6.1224,
      "mean_token_accuracy": 0.14701514989137648,
      "num_tokens": 18811904.0,
      "step": 9615
    },
    {
      "entropy": 6.234589338302612,
      "epoch": 0.9616634178037686,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004916415632109977,
      "loss": 6.0764,
      "mean_token_accuracy": 0.15597172379493712,
      "num_tokens": 18821388.0,
      "step": 9620
    },
    {
      "entropy": 6.225022077560425,
      "epoch": 0.9621632428649972,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004916319232917852,
      "loss": 6.1118,
      "mean_token_accuracy": 0.15659579783678054,
      "num_tokens": 18831392.0,
      "step": 9625
    },
    {
      "entropy": 6.231958103179932,
      "epoch": 0.9626630679262258,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004916222779221694,
      "loss": 6.0312,
      "mean_token_accuracy": 0.14850951507687568,
      "num_tokens": 18841211.0,
      "step": 9630
    },
    {
      "entropy": 6.281595897674561,
      "epoch": 0.9631628929874544,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004916126271023926,
      "loss": 6.0651,
      "mean_token_accuracy": 0.1544529989361763,
      "num_tokens": 18851033.0,
      "step": 9635
    },
    {
      "entropy": 6.20293436050415,
      "epoch": 0.963662718048683,
      "grad_norm": 1.171875,
      "learning_rate": 0.000491602970832698,
      "loss": 6.1014,
      "mean_token_accuracy": 0.15091072767972946,
      "num_tokens": 18861439.0,
      "step": 9640
    },
    {
      "entropy": 6.242417621612549,
      "epoch": 0.9641625431099116,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004915933091133284,
      "loss": 6.0832,
      "mean_token_accuracy": 0.14760126173496246,
      "num_tokens": 18872056.0,
      "step": 9645
    },
    {
      "entropy": 6.249822044372559,
      "epoch": 0.9646623681711401,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004915836419445268,
      "loss": 5.9917,
      "mean_token_accuracy": 0.15385040789842605,
      "num_tokens": 18881540.0,
      "step": 9650
    },
    {
      "entropy": 6.152493667602539,
      "epoch": 0.9651621932323686,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004915739693265365,
      "loss": 6.0268,
      "mean_token_accuracy": 0.15216706693172455,
      "num_tokens": 18890466.0,
      "step": 9655
    },
    {
      "entropy": 6.184817171096801,
      "epoch": 0.9656620182935972,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004915642912596009,
      "loss": 6.0889,
      "mean_token_accuracy": 0.1468469649553299,
      "num_tokens": 18901397.0,
      "step": 9660
    },
    {
      "entropy": 6.217543697357177,
      "epoch": 0.9661618433548258,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004915546077439634,
      "loss": 6.0151,
      "mean_token_accuracy": 0.15533899068832396,
      "num_tokens": 18911357.0,
      "step": 9665
    },
    {
      "entropy": 6.122984743118286,
      "epoch": 0.9666616684160544,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004915449187798677,
      "loss": 5.966,
      "mean_token_accuracy": 0.1643661677837372,
      "num_tokens": 18920901.0,
      "step": 9670
    },
    {
      "entropy": 6.232167100906372,
      "epoch": 0.967161493477283,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004915352243675576,
      "loss": 6.0807,
      "mean_token_accuracy": 0.14711482524871827,
      "num_tokens": 18931012.0,
      "step": 9675
    },
    {
      "entropy": 6.274874305725097,
      "epoch": 0.9676613185385116,
      "grad_norm": 1.203125,
      "learning_rate": 0.000491525524507277,
      "loss": 6.1233,
      "mean_token_accuracy": 0.14937701672315598,
      "num_tokens": 18941222.0,
      "step": 9680
    },
    {
      "entropy": 6.276605844497681,
      "epoch": 0.96816114359974,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004915158191992699,
      "loss": 6.111,
      "mean_token_accuracy": 0.15091582164168357,
      "num_tokens": 18951073.0,
      "step": 9685
    },
    {
      "entropy": 6.237700653076172,
      "epoch": 0.9686609686609686,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004915061084437805,
      "loss": 6.1216,
      "mean_token_accuracy": 0.148659910261631,
      "num_tokens": 18960996.0,
      "step": 9690
    },
    {
      "entropy": 6.234710264205932,
      "epoch": 0.9691607937221972,
      "grad_norm": 1.125,
      "learning_rate": 0.0004914963922410531,
      "loss": 6.1415,
      "mean_token_accuracy": 0.1443525329232216,
      "num_tokens": 18970451.0,
      "step": 9695
    },
    {
      "entropy": 6.184514808654785,
      "epoch": 0.9696606187834258,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004914866705913322,
      "loss": 6.0205,
      "mean_token_accuracy": 0.1525132969021797,
      "num_tokens": 18979781.0,
      "step": 9700
    },
    {
      "entropy": 6.2285130500793455,
      "epoch": 0.9701604438446544,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004914769434948623,
      "loss": 6.0659,
      "mean_token_accuracy": 0.15246652364730834,
      "num_tokens": 18989663.0,
      "step": 9705
    },
    {
      "entropy": 6.217658233642578,
      "epoch": 0.970660268905883,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004914672109518883,
      "loss": 6.18,
      "mean_token_accuracy": 0.14951386153697968,
      "num_tokens": 18999613.0,
      "step": 9710
    },
    {
      "entropy": 6.102007055282593,
      "epoch": 0.9711600939671116,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000491457472962655,
      "loss": 5.9772,
      "mean_token_accuracy": 0.16518737375736237,
      "num_tokens": 19008976.0,
      "step": 9715
    },
    {
      "entropy": 6.11379222869873,
      "epoch": 0.97165991902834,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004914477295274073,
      "loss": 5.9911,
      "mean_token_accuracy": 0.1585332378745079,
      "num_tokens": 19019013.0,
      "step": 9720
    },
    {
      "entropy": 6.270997953414917,
      "epoch": 0.9721597440895686,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004914379806463905,
      "loss": 6.0998,
      "mean_token_accuracy": 0.1508036345243454,
      "num_tokens": 19029252.0,
      "step": 9725
    },
    {
      "entropy": 6.2492962837219235,
      "epoch": 0.9726595691507972,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004914282263198498,
      "loss": 6.0597,
      "mean_token_accuracy": 0.15353047251701354,
      "num_tokens": 19038563.0,
      "step": 9730
    },
    {
      "entropy": 6.213043689727783,
      "epoch": 0.9731593942120258,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004914184665480305,
      "loss": 6.0711,
      "mean_token_accuracy": 0.15182880610227584,
      "num_tokens": 19048568.0,
      "step": 9735
    },
    {
      "entropy": 6.123023939132691,
      "epoch": 0.9736592192732544,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004914087013311784,
      "loss": 5.8562,
      "mean_token_accuracy": 0.1695294477045536,
      "num_tokens": 19057527.0,
      "step": 9740
    },
    {
      "entropy": 6.17437891960144,
      "epoch": 0.974159044334483,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004913989306695389,
      "loss": 6.0059,
      "mean_token_accuracy": 0.15471534356474875,
      "num_tokens": 19067372.0,
      "step": 9745
    },
    {
      "entropy": 6.154889535903931,
      "epoch": 0.9746588693957114,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004913891545633582,
      "loss": 6.0005,
      "mean_token_accuracy": 0.16327692270278932,
      "num_tokens": 19077521.0,
      "step": 9750
    },
    {
      "entropy": 6.2207399845123295,
      "epoch": 0.97515869445694,
      "grad_norm": 1.109375,
      "learning_rate": 0.000491379373012882,
      "loss": 5.9832,
      "mean_token_accuracy": 0.15415498465299607,
      "num_tokens": 19086674.0,
      "step": 9755
    },
    {
      "entropy": 6.263242626190186,
      "epoch": 0.9756585195181686,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004913695860183563,
      "loss": 6.2018,
      "mean_token_accuracy": 0.1406306244432926,
      "num_tokens": 19097586.0,
      "step": 9760
    },
    {
      "entropy": 6.220029878616333,
      "epoch": 0.9761583445793972,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004913597935800276,
      "loss": 6.1831,
      "mean_token_accuracy": 0.14061613753437996,
      "num_tokens": 19107763.0,
      "step": 9765
    },
    {
      "entropy": 6.215551233291626,
      "epoch": 0.9766581696406258,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004913499956981422,
      "loss": 5.9729,
      "mean_token_accuracy": 0.16298652589321136,
      "num_tokens": 19117931.0,
      "step": 9770
    },
    {
      "entropy": 6.223272132873535,
      "epoch": 0.9771579947018544,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004913401923729465,
      "loss": 6.0652,
      "mean_token_accuracy": 0.15182324200868608,
      "num_tokens": 19127707.0,
      "step": 9775
    },
    {
      "entropy": 6.191423463821411,
      "epoch": 0.977657819763083,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004913303836046873,
      "loss": 5.9762,
      "mean_token_accuracy": 0.15523964017629624,
      "num_tokens": 19137633.0,
      "step": 9780
    },
    {
      "entropy": 6.203907489776611,
      "epoch": 0.9781576448243114,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004913205693936112,
      "loss": 6.0987,
      "mean_token_accuracy": 0.15070056766271592,
      "num_tokens": 19146247.0,
      "step": 9785
    },
    {
      "entropy": 6.122197055816651,
      "epoch": 0.97865746988554,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004913107497399653,
      "loss": 6.0028,
      "mean_token_accuracy": 0.1688719540834427,
      "num_tokens": 19156225.0,
      "step": 9790
    },
    {
      "entropy": 6.1869800090789795,
      "epoch": 0.9791572949467686,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004913009246439966,
      "loss": 5.9647,
      "mean_token_accuracy": 0.15260530561208724,
      "num_tokens": 19165763.0,
      "step": 9795
    },
    {
      "entropy": 6.186647891998291,
      "epoch": 0.9796571200079972,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004912910941059525,
      "loss": 5.9944,
      "mean_token_accuracy": 0.154931004345417,
      "num_tokens": 19174116.0,
      "step": 9800
    },
    {
      "entropy": 6.197454452514648,
      "epoch": 0.9801569450692258,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004912812581260799,
      "loss": 6.1747,
      "mean_token_accuracy": 0.1421088807284832,
      "num_tokens": 19184416.0,
      "step": 9805
    },
    {
      "entropy": 6.228710746765136,
      "epoch": 0.9806567701304544,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004912714167046265,
      "loss": 6.1353,
      "mean_token_accuracy": 0.14206735342741011,
      "num_tokens": 19193356.0,
      "step": 9810
    },
    {
      "entropy": 6.17477822303772,
      "epoch": 0.9811565951916829,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004912615698418399,
      "loss": 5.9507,
      "mean_token_accuracy": 0.1712735638022423,
      "num_tokens": 19204257.0,
      "step": 9815
    },
    {
      "entropy": 6.314099073410034,
      "epoch": 0.9816564202529114,
      "grad_norm": 1.125,
      "learning_rate": 0.0004912517175379679,
      "loss": 6.0706,
      "mean_token_accuracy": 0.15417387336492538,
      "num_tokens": 19213889.0,
      "step": 9820
    },
    {
      "entropy": 6.3021900177001955,
      "epoch": 0.98215624531414,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004912418597932584,
      "loss": 6.146,
      "mean_token_accuracy": 0.1453337252140045,
      "num_tokens": 19224008.0,
      "step": 9825
    },
    {
      "entropy": 6.227414608001709,
      "epoch": 0.9826560703753686,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004912319966079592,
      "loss": 6.132,
      "mean_token_accuracy": 0.1454176314175129,
      "num_tokens": 19234247.0,
      "step": 9830
    },
    {
      "entropy": 6.281443500518799,
      "epoch": 0.9831558954365972,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004912221279823187,
      "loss": 6.1399,
      "mean_token_accuracy": 0.1494423806667328,
      "num_tokens": 19244838.0,
      "step": 9835
    },
    {
      "entropy": 6.195976591110229,
      "epoch": 0.9836557204978258,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004912122539165851,
      "loss": 6.0495,
      "mean_token_accuracy": 0.15361892133951188,
      "num_tokens": 19254989.0,
      "step": 9840
    },
    {
      "entropy": 6.184459018707275,
      "epoch": 0.9841555455590544,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004912023744110068,
      "loss": 6.0975,
      "mean_token_accuracy": 0.150004356354475,
      "num_tokens": 19264519.0,
      "step": 9845
    },
    {
      "entropy": 6.193889904022217,
      "epoch": 0.9846553706202829,
      "grad_norm": 1.125,
      "learning_rate": 0.0004911924894658324,
      "loss": 6.0573,
      "mean_token_accuracy": 0.1562734618782997,
      "num_tokens": 19274153.0,
      "step": 9850
    },
    {
      "entropy": 6.149482679367066,
      "epoch": 0.9851551956815114,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004911825990813107,
      "loss": 5.9865,
      "mean_token_accuracy": 0.15565495043992997,
      "num_tokens": 19283383.0,
      "step": 9855
    },
    {
      "entropy": 6.162270641326904,
      "epoch": 0.98565502074274,
      "grad_norm": 1.25,
      "learning_rate": 0.0004911727032576903,
      "loss": 5.977,
      "mean_token_accuracy": 0.15174852609634398,
      "num_tokens": 19292758.0,
      "step": 9860
    },
    {
      "entropy": 6.229521226882935,
      "epoch": 0.9861548458039686,
      "grad_norm": 1.375,
      "learning_rate": 0.0004911628019952203,
      "loss": 6.0711,
      "mean_token_accuracy": 0.14723279997706412,
      "num_tokens": 19302339.0,
      "step": 9865
    },
    {
      "entropy": 6.185194873809815,
      "epoch": 0.9866546708651972,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004911528952941499,
      "loss": 5.9037,
      "mean_token_accuracy": 0.15146453082561492,
      "num_tokens": 19312560.0,
      "step": 9870
    },
    {
      "entropy": 6.168459892272949,
      "epoch": 0.9871544959264258,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004911429831547283,
      "loss": 5.977,
      "mean_token_accuracy": 0.15555412620306014,
      "num_tokens": 19322796.0,
      "step": 9875
    },
    {
      "entropy": 6.174897766113281,
      "epoch": 0.9876543209876543,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004911330655772047,
      "loss": 6.0666,
      "mean_token_accuracy": 0.15046636313199996,
      "num_tokens": 19332719.0,
      "step": 9880
    },
    {
      "entropy": 6.214740037918091,
      "epoch": 0.9881541460488829,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000491123142561829,
      "loss": 6.0601,
      "mean_token_accuracy": 0.14733994752168655,
      "num_tokens": 19342226.0,
      "step": 9885
    },
    {
      "entropy": 6.19035234451294,
      "epoch": 0.9886539711101114,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004911132141088505,
      "loss": 6.0603,
      "mean_token_accuracy": 0.14697122424840928,
      "num_tokens": 19351933.0,
      "step": 9890
    },
    {
      "entropy": 6.242229032516479,
      "epoch": 0.98915379617134,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004911032802185192,
      "loss": 5.9574,
      "mean_token_accuracy": 0.1515513241291046,
      "num_tokens": 19359900.0,
      "step": 9895
    },
    {
      "entropy": 6.215701341629028,
      "epoch": 0.9896536212325686,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000491093340891085,
      "loss": 6.0698,
      "mean_token_accuracy": 0.14505403339862824,
      "num_tokens": 19369139.0,
      "step": 9900
    },
    {
      "entropy": 6.1298071384429935,
      "epoch": 0.9901534462937972,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004910833961267979,
      "loss": 5.9914,
      "mean_token_accuracy": 0.15241192728281022,
      "num_tokens": 19378765.0,
      "step": 9905
    },
    {
      "entropy": 6.223695945739746,
      "epoch": 0.9906532713550258,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004910734459259081,
      "loss": 6.0874,
      "mean_token_accuracy": 0.14713988900184632,
      "num_tokens": 19389001.0,
      "step": 9910
    },
    {
      "entropy": 6.230083370208741,
      "epoch": 0.9911530964162543,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004910634902886662,
      "loss": 6.0345,
      "mean_token_accuracy": 0.1559482455253601,
      "num_tokens": 19397990.0,
      "step": 9915
    },
    {
      "entropy": 6.009120273590088,
      "epoch": 0.9916529214774829,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004910535292153224,
      "loss": 5.9345,
      "mean_token_accuracy": 0.15913379788398743,
      "num_tokens": 19408060.0,
      "step": 9920
    },
    {
      "entropy": 6.060460042953491,
      "epoch": 0.9921527465387114,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004910435627061273,
      "loss": 5.9693,
      "mean_token_accuracy": 0.15843064039945604,
      "num_tokens": 19417675.0,
      "step": 9925
    },
    {
      "entropy": 6.219476318359375,
      "epoch": 0.99265257159994,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004910335907613318,
      "loss": 6.004,
      "mean_token_accuracy": 0.15458038300275803,
      "num_tokens": 19426433.0,
      "step": 9930
    },
    {
      "entropy": 6.215752458572387,
      "epoch": 0.9931523966611686,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004910236133811867,
      "loss": 5.8934,
      "mean_token_accuracy": 0.15964245200157165,
      "num_tokens": 19435459.0,
      "step": 9935
    },
    {
      "entropy": 6.147295999526977,
      "epoch": 0.9936522217223972,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004910136305659432,
      "loss": 6.0462,
      "mean_token_accuracy": 0.15278443843126296,
      "num_tokens": 19445329.0,
      "step": 9940
    },
    {
      "entropy": 6.174296951293945,
      "epoch": 0.9941520467836257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004910036423158523,
      "loss": 5.9927,
      "mean_token_accuracy": 0.15378375798463823,
      "num_tokens": 19454417.0,
      "step": 9945
    },
    {
      "entropy": 6.21248722076416,
      "epoch": 0.9946518718448543,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004909936486311653,
      "loss": 6.0709,
      "mean_token_accuracy": 0.15004773586988449,
      "num_tokens": 19464274.0,
      "step": 9950
    },
    {
      "entropy": 6.232455062866211,
      "epoch": 0.9951516969060828,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004909836495121336,
      "loss": 6.0523,
      "mean_token_accuracy": 0.15589189529418945,
      "num_tokens": 19472455.0,
      "step": 9955
    },
    {
      "entropy": 6.245503330230713,
      "epoch": 0.9956515219673114,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004909736449590091,
      "loss": 6.142,
      "mean_token_accuracy": 0.14530933052301406,
      "num_tokens": 19483320.0,
      "step": 9960
    },
    {
      "entropy": 6.2394956111907955,
      "epoch": 0.99615134702854,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000490963634972043,
      "loss": 6.02,
      "mean_token_accuracy": 0.15101177245378494,
      "num_tokens": 19493644.0,
      "step": 9965
    },
    {
      "entropy": 6.162610578536987,
      "epoch": 0.9966511720897686,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004909536195514875,
      "loss": 5.9538,
      "mean_token_accuracy": 0.15775168910622597,
      "num_tokens": 19503035.0,
      "step": 9970
    },
    {
      "entropy": 6.15058217048645,
      "epoch": 0.9971509971509972,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004909435986975946,
      "loss": 6.0813,
      "mean_token_accuracy": 0.15467339605093003,
      "num_tokens": 19512254.0,
      "step": 9975
    },
    {
      "entropy": 6.182337093353271,
      "epoch": 0.9976508222122257,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004909335724106162,
      "loss": 5.9589,
      "mean_token_accuracy": 0.15946487635374068,
      "num_tokens": 19523114.0,
      "step": 9980
    },
    {
      "entropy": 6.227681398391724,
      "epoch": 0.9981506472734543,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004909235406908047,
      "loss": 6.0556,
      "mean_token_accuracy": 0.14987050294876098,
      "num_tokens": 19532940.0,
      "step": 9985
    },
    {
      "entropy": 6.13013186454773,
      "epoch": 0.9986504723346828,
      "grad_norm": 1.125,
      "learning_rate": 0.0004909135035384126,
      "loss": 6.0405,
      "mean_token_accuracy": 0.15940339714288712,
      "num_tokens": 19542807.0,
      "step": 9990
    },
    {
      "entropy": 6.136002779006958,
      "epoch": 0.9991502973959114,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004909034609536922,
      "loss": 5.945,
      "mean_token_accuracy": 0.15171423330903053,
      "num_tokens": 19552099.0,
      "step": 9995
    },
    {
      "entropy": 6.177865791320801,
      "epoch": 0.99965012245714,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004908934129368963,
      "loss": 5.9984,
      "mean_token_accuracy": 0.1580812469124794,
      "num_tokens": 19561709.0,
      "step": 10000
    },
    {
      "entropy": 6.1811365021599665,
      "epoch": 1.0000999650122457,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004908833594882777,
      "loss": 6.1297,
      "mean_token_accuracy": 0.15531002978483835,
      "num_tokens": 19571739.0,
      "step": 10005
    },
    {
      "entropy": 6.260761308670044,
      "epoch": 1.0005997900734742,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004908733006080893,
      "loss": 5.9689,
      "mean_token_accuracy": 0.15648650676012038,
      "num_tokens": 19582084.0,
      "step": 10010
    },
    {
      "entropy": 6.157464075088501,
      "epoch": 1.001099615134703,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004908632362965842,
      "loss": 5.9135,
      "mean_token_accuracy": 0.1570557102560997,
      "num_tokens": 19591889.0,
      "step": 10015
    },
    {
      "entropy": 6.215280628204345,
      "epoch": 1.0015994401959314,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004908531665540158,
      "loss": 6.0454,
      "mean_token_accuracy": 0.1517791785299778,
      "num_tokens": 19601998.0,
      "step": 10020
    },
    {
      "entropy": 6.160162878036499,
      "epoch": 1.00209926525716,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000490843091380637,
      "loss": 5.8343,
      "mean_token_accuracy": 0.16489042490720748,
      "num_tokens": 19611347.0,
      "step": 10025
    },
    {
      "entropy": 6.1917649269104,
      "epoch": 1.0025990903183886,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004908330107767018,
      "loss": 5.9954,
      "mean_token_accuracy": 0.15234703421592713,
      "num_tokens": 19621522.0,
      "step": 10030
    },
    {
      "entropy": 6.129675722122192,
      "epoch": 1.003098915379617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004908229247424633,
      "loss": 5.8501,
      "mean_token_accuracy": 0.1669887989759445,
      "num_tokens": 19630815.0,
      "step": 10035
    },
    {
      "entropy": 6.171536684036255,
      "epoch": 1.0035987404408457,
      "grad_norm": 1.125,
      "learning_rate": 0.0004908128332781758,
      "loss": 5.9378,
      "mean_token_accuracy": 0.16407083570957184,
      "num_tokens": 19640811.0,
      "step": 10040
    },
    {
      "entropy": 6.1419285297393795,
      "epoch": 1.0040985655020742,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004908027363840928,
      "loss": 5.9481,
      "mean_token_accuracy": 0.1554320827126503,
      "num_tokens": 19651335.0,
      "step": 10045
    },
    {
      "entropy": 6.110807657241821,
      "epoch": 1.004598390563303,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004907926340604686,
      "loss": 5.8841,
      "mean_token_accuracy": 0.16288528442382813,
      "num_tokens": 19660242.0,
      "step": 10050
    },
    {
      "entropy": 6.179576253890991,
      "epoch": 1.0050982156245314,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004907825263075573,
      "loss": 5.9827,
      "mean_token_accuracy": 0.14936700463294983,
      "num_tokens": 19670013.0,
      "step": 10055
    },
    {
      "entropy": 6.150372123718261,
      "epoch": 1.00559804068576,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004907724131256129,
      "loss": 5.8569,
      "mean_token_accuracy": 0.16397354900836944,
      "num_tokens": 19679479.0,
      "step": 10060
    },
    {
      "entropy": 6.13425931930542,
      "epoch": 1.0060978657469886,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004907622945148903,
      "loss": 5.9115,
      "mean_token_accuracy": 0.16079266518354415,
      "num_tokens": 19688603.0,
      "step": 10065
    },
    {
      "entropy": 6.119693088531494,
      "epoch": 1.006597690808217,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004907521704756438,
      "loss": 5.8875,
      "mean_token_accuracy": 0.15410733073949814,
      "num_tokens": 19698375.0,
      "step": 10070
    },
    {
      "entropy": 6.167830228805542,
      "epoch": 1.0070975158694457,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004907420410081284,
      "loss": 5.9589,
      "mean_token_accuracy": 0.15665326565504073,
      "num_tokens": 19708911.0,
      "step": 10075
    },
    {
      "entropy": 6.123469972610474,
      "epoch": 1.0075973409306742,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004907319061125986,
      "loss": 5.8998,
      "mean_token_accuracy": 0.15388508439064025,
      "num_tokens": 19717733.0,
      "step": 10080
    },
    {
      "entropy": 6.12524938583374,
      "epoch": 1.008097165991903,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004907217657893096,
      "loss": 5.9169,
      "mean_token_accuracy": 0.15649490505456926,
      "num_tokens": 19727192.0,
      "step": 10085
    },
    {
      "entropy": 6.14108715057373,
      "epoch": 1.0085969910531314,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004907116200385165,
      "loss": 5.8944,
      "mean_token_accuracy": 0.157904751598835,
      "num_tokens": 19737937.0,
      "step": 10090
    },
    {
      "entropy": 6.140891695022583,
      "epoch": 1.00909681611436,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004907014688604745,
      "loss": 5.8741,
      "mean_token_accuracy": 0.15400379598140718,
      "num_tokens": 19748270.0,
      "step": 10095
    },
    {
      "entropy": 6.059481287002564,
      "epoch": 1.0095966411755886,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004906913122554391,
      "loss": 5.7606,
      "mean_token_accuracy": 0.1737472251057625,
      "num_tokens": 19758328.0,
      "step": 10100
    },
    {
      "entropy": 6.073921823501587,
      "epoch": 1.010096466236817,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004906811502236657,
      "loss": 5.8636,
      "mean_token_accuracy": 0.1621103286743164,
      "num_tokens": 19767780.0,
      "step": 10105
    },
    {
      "entropy": 6.177298498153687,
      "epoch": 1.0105962912980457,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004906709827654101,
      "loss": 6.0175,
      "mean_token_accuracy": 0.15365107655525206,
      "num_tokens": 19777202.0,
      "step": 10110
    },
    {
      "entropy": 6.068292284011841,
      "epoch": 1.0110961163592742,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004906608098809281,
      "loss": 5.8887,
      "mean_token_accuracy": 0.15936959087848662,
      "num_tokens": 19788341.0,
      "step": 10115
    },
    {
      "entropy": 6.217078018188476,
      "epoch": 1.011595941420503,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004906506315704756,
      "loss": 5.9909,
      "mean_token_accuracy": 0.1571861431002617,
      "num_tokens": 19797511.0,
      "step": 10120
    },
    {
      "entropy": 6.227612018585205,
      "epoch": 1.0120957664817314,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004906404478343087,
      "loss": 5.9715,
      "mean_token_accuracy": 0.15643113106489182,
      "num_tokens": 19807639.0,
      "step": 10125
    },
    {
      "entropy": 6.247345018386841,
      "epoch": 1.0125955915429599,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004906302586726837,
      "loss": 6.0581,
      "mean_token_accuracy": 0.1466827869415283,
      "num_tokens": 19818379.0,
      "step": 10130
    },
    {
      "entropy": 6.147131681442261,
      "epoch": 1.0130954166041886,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004906200640858567,
      "loss": 5.9032,
      "mean_token_accuracy": 0.16134313195943834,
      "num_tokens": 19828706.0,
      "step": 10135
    },
    {
      "entropy": 6.119057512283325,
      "epoch": 1.013595241665417,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004906098640740846,
      "loss": 5.9226,
      "mean_token_accuracy": 0.16448748856782913,
      "num_tokens": 19838361.0,
      "step": 10140
    },
    {
      "entropy": 6.119967412948609,
      "epoch": 1.0140950667266457,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004905996586376236,
      "loss": 5.9649,
      "mean_token_accuracy": 0.159076264500618,
      "num_tokens": 19848205.0,
      "step": 10145
    },
    {
      "entropy": 6.176230192184448,
      "epoch": 1.0145948917878742,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004905894477767308,
      "loss": 6.0065,
      "mean_token_accuracy": 0.15325742438435555,
      "num_tokens": 19858490.0,
      "step": 10150
    },
    {
      "entropy": 6.15353102684021,
      "epoch": 1.015094716849103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004905792314916629,
      "loss": 5.9841,
      "mean_token_accuracy": 0.15750030130147935,
      "num_tokens": 19867551.0,
      "step": 10155
    },
    {
      "entropy": 6.186533355712891,
      "epoch": 1.0155945419103314,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004905690097826769,
      "loss": 5.9972,
      "mean_token_accuracy": 0.15465582609176637,
      "num_tokens": 19877581.0,
      "step": 10160
    },
    {
      "entropy": 6.305242013931275,
      "epoch": 1.0160943669715599,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004905587826500303,
      "loss": 6.0458,
      "mean_token_accuracy": 0.15035037845373153,
      "num_tokens": 19887827.0,
      "step": 10165
    },
    {
      "entropy": 6.173418807983398,
      "epoch": 1.0165941920327886,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004905485500939801,
      "loss": 5.9368,
      "mean_token_accuracy": 0.15681003779172897,
      "num_tokens": 19897395.0,
      "step": 10170
    },
    {
      "entropy": 6.139950847625732,
      "epoch": 1.017094017094017,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004905383121147838,
      "loss": 5.9348,
      "mean_token_accuracy": 0.15253501683473586,
      "num_tokens": 19906601.0,
      "step": 10175
    },
    {
      "entropy": 6.160625457763672,
      "epoch": 1.0175938421552457,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004905280687126991,
      "loss": 5.8822,
      "mean_token_accuracy": 0.1573880985379219,
      "num_tokens": 19916422.0,
      "step": 10180
    },
    {
      "entropy": 6.199394941329956,
      "epoch": 1.0180936672164742,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004905178198879837,
      "loss": 6.0295,
      "mean_token_accuracy": 0.1525586351752281,
      "num_tokens": 19927182.0,
      "step": 10185
    },
    {
      "entropy": 6.120484972000122,
      "epoch": 1.018593492277703,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004905075656408953,
      "loss": 5.9937,
      "mean_token_accuracy": 0.151900078356266,
      "num_tokens": 19936537.0,
      "step": 10190
    },
    {
      "entropy": 6.192416334152222,
      "epoch": 1.0190933173389314,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000490497305971692,
      "loss": 5.9854,
      "mean_token_accuracy": 0.15833306610584258,
      "num_tokens": 19946315.0,
      "step": 10195
    },
    {
      "entropy": 6.201804304122925,
      "epoch": 1.0195931424001599,
      "grad_norm": 1.15625,
      "learning_rate": 0.000490487040880632,
      "loss": 5.88,
      "mean_token_accuracy": 0.15713443011045455,
      "num_tokens": 19956145.0,
      "step": 10200
    },
    {
      "entropy": 6.158399486541748,
      "epoch": 1.0200929674613886,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004904767703679735,
      "loss": 6.0024,
      "mean_token_accuracy": 0.14811211526393891,
      "num_tokens": 19966890.0,
      "step": 10205
    },
    {
      "entropy": 6.153269290924072,
      "epoch": 1.020592792522617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004904664944339748,
      "loss": 5.9676,
      "mean_token_accuracy": 0.15149188041687012,
      "num_tokens": 19976156.0,
      "step": 10210
    },
    {
      "entropy": 6.162014102935791,
      "epoch": 1.0210926175838457,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004904562130788947,
      "loss": 5.8858,
      "mean_token_accuracy": 0.15853449553251267,
      "num_tokens": 19985851.0,
      "step": 10215
    },
    {
      "entropy": 6.121045017242432,
      "epoch": 1.0215924426450742,
      "grad_norm": 1.125,
      "learning_rate": 0.0004904459263029916,
      "loss": 5.9789,
      "mean_token_accuracy": 0.15916912108659745,
      "num_tokens": 19996531.0,
      "step": 10220
    },
    {
      "entropy": 6.16722526550293,
      "epoch": 1.0220922677063027,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004904356341065245,
      "loss": 5.8468,
      "mean_token_accuracy": 0.16380981653928756,
      "num_tokens": 20006050.0,
      "step": 10225
    },
    {
      "entropy": 6.192217588424683,
      "epoch": 1.0225920927675314,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004904253364897521,
      "loss": 5.9935,
      "mean_token_accuracy": 0.15024405121803283,
      "num_tokens": 20015997.0,
      "step": 10230
    },
    {
      "entropy": 6.202864313125611,
      "epoch": 1.0230919178287599,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004904150334529339,
      "loss": 6.1077,
      "mean_token_accuracy": 0.14537899941205978,
      "num_tokens": 20026682.0,
      "step": 10235
    },
    {
      "entropy": 6.16762375831604,
      "epoch": 1.0235917428899886,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004904047249963289,
      "loss": 5.9381,
      "mean_token_accuracy": 0.15313244313001634,
      "num_tokens": 20036166.0,
      "step": 10240
    },
    {
      "entropy": 6.1094144821167,
      "epoch": 1.024091567951217,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004903944111201964,
      "loss": 5.8897,
      "mean_token_accuracy": 0.16153058409690857,
      "num_tokens": 20045545.0,
      "step": 10245
    },
    {
      "entropy": 6.1941718578338625,
      "epoch": 1.0245913930124457,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004903840918247958,
      "loss": 5.8885,
      "mean_token_accuracy": 0.15716925859451295,
      "num_tokens": 20055072.0,
      "step": 10250
    },
    {
      "entropy": 6.233175373077392,
      "epoch": 1.0250912180736742,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000490373767110387,
      "loss": 6.0305,
      "mean_token_accuracy": 0.15243661850690843,
      "num_tokens": 20064227.0,
      "step": 10255
    },
    {
      "entropy": 6.176199865341187,
      "epoch": 1.0255910431349027,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004903634369772296,
      "loss": 5.9488,
      "mean_token_accuracy": 0.15869183838367462,
      "num_tokens": 20074833.0,
      "step": 10260
    },
    {
      "entropy": 6.167827224731445,
      "epoch": 1.0260908681961314,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004903531014255835,
      "loss": 5.9403,
      "mean_token_accuracy": 0.15362744331359862,
      "num_tokens": 20084340.0,
      "step": 10265
    },
    {
      "entropy": 6.183794641494751,
      "epoch": 1.0265906932573599,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004903427604557089,
      "loss": 5.9292,
      "mean_token_accuracy": 0.15183276534080506,
      "num_tokens": 20094372.0,
      "step": 10270
    },
    {
      "entropy": 6.154160690307617,
      "epoch": 1.0270905183185886,
      "grad_norm": 1.25,
      "learning_rate": 0.0004903324140678658,
      "loss": 5.891,
      "mean_token_accuracy": 0.15561358779668807,
      "num_tokens": 20104015.0,
      "step": 10275
    },
    {
      "entropy": 6.1259989738464355,
      "epoch": 1.027590343379817,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004903220622623146,
      "loss": 5.9372,
      "mean_token_accuracy": 0.15709602534770967,
      "num_tokens": 20113570.0,
      "step": 10280
    },
    {
      "entropy": 6.195018911361695,
      "epoch": 1.0280901684410457,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004903117050393157,
      "loss": 5.9817,
      "mean_token_accuracy": 0.1525372803211212,
      "num_tokens": 20124147.0,
      "step": 10285
    },
    {
      "entropy": 6.143146324157715,
      "epoch": 1.0285899935022742,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004903013423991296,
      "loss": 6.0077,
      "mean_token_accuracy": 0.15060942471027375,
      "num_tokens": 20134310.0,
      "step": 10290
    },
    {
      "entropy": 6.185633707046509,
      "epoch": 1.0290898185635027,
      "grad_norm": 1.125,
      "learning_rate": 0.0004902909743420173,
      "loss": 5.9219,
      "mean_token_accuracy": 0.16042520403861998,
      "num_tokens": 20143894.0,
      "step": 10295
    },
    {
      "entropy": 6.162026834487915,
      "epoch": 1.0295896436247314,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004902806008682394,
      "loss": 5.9644,
      "mean_token_accuracy": 0.1588928610086441,
      "num_tokens": 20154862.0,
      "step": 10300
    },
    {
      "entropy": 6.146246480941772,
      "epoch": 1.0300894686859599,
      "grad_norm": 1.375,
      "learning_rate": 0.000490270221978057,
      "loss": 5.8714,
      "mean_token_accuracy": 0.15972404181957245,
      "num_tokens": 20164109.0,
      "step": 10305
    },
    {
      "entropy": 6.2074470043182375,
      "epoch": 1.0305892937471886,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004902598376717312,
      "loss": 5.9571,
      "mean_token_accuracy": 0.1511505775153637,
      "num_tokens": 20173971.0,
      "step": 10310
    },
    {
      "entropy": 6.1420533657073975,
      "epoch": 1.031089118808417,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004902494479495233,
      "loss": 5.9752,
      "mean_token_accuracy": 0.1464315190911293,
      "num_tokens": 20183683.0,
      "step": 10315
    },
    {
      "entropy": 6.121149921417237,
      "epoch": 1.0315889438696457,
      "grad_norm": 1.125,
      "learning_rate": 0.0004902390528116946,
      "loss": 5.9056,
      "mean_token_accuracy": 0.15282565355300903,
      "num_tokens": 20192945.0,
      "step": 10320
    },
    {
      "entropy": 6.174913454055786,
      "epoch": 1.0320887689308742,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004902286522585068,
      "loss": 5.9586,
      "mean_token_accuracy": 0.15731764808297158,
      "num_tokens": 20202919.0,
      "step": 10325
    },
    {
      "entropy": 6.099916553497314,
      "epoch": 1.0325885939921027,
      "grad_norm": 1.25,
      "learning_rate": 0.0004902182462902215,
      "loss": 5.8587,
      "mean_token_accuracy": 0.16296564638614655,
      "num_tokens": 20212377.0,
      "step": 10330
    },
    {
      "entropy": 6.088365316390991,
      "epoch": 1.0330884190533314,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004902078349071004,
      "loss": 5.8681,
      "mean_token_accuracy": 0.15789700001478196,
      "num_tokens": 20221083.0,
      "step": 10335
    },
    {
      "entropy": 6.18060417175293,
      "epoch": 1.0335882441145599,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004901974181094057,
      "loss": 5.9884,
      "mean_token_accuracy": 0.1544646829366684,
      "num_tokens": 20229577.0,
      "step": 10340
    },
    {
      "entropy": 6.183709526062012,
      "epoch": 1.0340880691757885,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004901869958973991,
      "loss": 5.8463,
      "mean_token_accuracy": 0.16323690712451935,
      "num_tokens": 20238400.0,
      "step": 10345
    },
    {
      "entropy": 6.102051448822022,
      "epoch": 1.034587894237017,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004901765682713432,
      "loss": 6.0039,
      "mean_token_accuracy": 0.1537890076637268,
      "num_tokens": 20248479.0,
      "step": 10350
    },
    {
      "entropy": 6.131157112121582,
      "epoch": 1.0350877192982457,
      "grad_norm": 1.5,
      "learning_rate": 0.0004901661352315003,
      "loss": 5.9607,
      "mean_token_accuracy": 0.15368467047810555,
      "num_tokens": 20258268.0,
      "step": 10355
    },
    {
      "entropy": 6.172520589828491,
      "epoch": 1.0355875443594742,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004901556967781327,
      "loss": 5.8961,
      "mean_token_accuracy": 0.1650520756840706,
      "num_tokens": 20267924.0,
      "step": 10360
    },
    {
      "entropy": 6.228430891036988,
      "epoch": 1.0360873694207027,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004901452529115031,
      "loss": 5.9613,
      "mean_token_accuracy": 0.15430244505405427,
      "num_tokens": 20278173.0,
      "step": 10365
    },
    {
      "entropy": 6.099439811706543,
      "epoch": 1.0365871944819314,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004901348036318743,
      "loss": 5.8871,
      "mean_token_accuracy": 0.16255491375923156,
      "num_tokens": 20287557.0,
      "step": 10370
    },
    {
      "entropy": 6.203316783905029,
      "epoch": 1.0370870195431598,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004901243489395093,
      "loss": 6.0284,
      "mean_token_accuracy": 0.15116425603628159,
      "num_tokens": 20297517.0,
      "step": 10375
    },
    {
      "entropy": 6.159917306900025,
      "epoch": 1.0375868446043885,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000490113888834671,
      "loss": 5.8485,
      "mean_token_accuracy": 0.16409340500831604,
      "num_tokens": 20308029.0,
      "step": 10380
    },
    {
      "entropy": 6.016686058044433,
      "epoch": 1.038086669665617,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004901034233176227,
      "loss": 5.8328,
      "mean_token_accuracy": 0.16462446600198746,
      "num_tokens": 20318962.0,
      "step": 10385
    },
    {
      "entropy": 6.174820470809936,
      "epoch": 1.0385864947268455,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004900929523886274,
      "loss": 5.9536,
      "mean_token_accuracy": 0.15180271193385125,
      "num_tokens": 20328729.0,
      "step": 10390
    },
    {
      "entropy": 6.1586822986602785,
      "epoch": 1.0390863197880742,
      "grad_norm": 1.15625,
      "learning_rate": 0.000490082476047949,
      "loss": 5.9896,
      "mean_token_accuracy": 0.15001793652772905,
      "num_tokens": 20339002.0,
      "step": 10395
    },
    {
      "entropy": 6.190448379516601,
      "epoch": 1.0395861448493027,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004900719942958506,
      "loss": 5.9134,
      "mean_token_accuracy": 0.16036467850208283,
      "num_tokens": 20347869.0,
      "step": 10400
    },
    {
      "entropy": 6.0855059146881105,
      "epoch": 1.0400859699105314,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004900615071325963,
      "loss": 5.8055,
      "mean_token_accuracy": 0.16634249389171601,
      "num_tokens": 20356609.0,
      "step": 10405
    },
    {
      "entropy": 6.061008882522583,
      "epoch": 1.0405857949717598,
      "grad_norm": 1.140625,
      "learning_rate": 0.00049005101455845,
      "loss": 5.8852,
      "mean_token_accuracy": 0.16754458099603653,
      "num_tokens": 20368793.0,
      "step": 10410
    },
    {
      "entropy": 6.087546253204346,
      "epoch": 1.0410856200329885,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004900405165736753,
      "loss": 5.8443,
      "mean_token_accuracy": 0.1590973600745201,
      "num_tokens": 20379525.0,
      "step": 10415
    },
    {
      "entropy": 6.209928178787232,
      "epoch": 1.041585445094217,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004900300131785365,
      "loss": 5.9381,
      "mean_token_accuracy": 0.16295751333236694,
      "num_tokens": 20388253.0,
      "step": 10420
    },
    {
      "entropy": 6.085028076171875,
      "epoch": 1.0420852701554455,
      "grad_norm": 1.21875,
      "learning_rate": 0.000490019504373298,
      "loss": 5.9465,
      "mean_token_accuracy": 0.15388932675123215,
      "num_tokens": 20398086.0,
      "step": 10425
    },
    {
      "entropy": 6.122501611709595,
      "epoch": 1.0425850952166742,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004900089901582241,
      "loss": 6.0259,
      "mean_token_accuracy": 0.15358177125453948,
      "num_tokens": 20408113.0,
      "step": 10430
    },
    {
      "entropy": 6.2128089427947994,
      "epoch": 1.0430849202779027,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004899984705335792,
      "loss": 6.003,
      "mean_token_accuracy": 0.1570493310689926,
      "num_tokens": 20417717.0,
      "step": 10435
    },
    {
      "entropy": 6.1494463920593265,
      "epoch": 1.0435847453391314,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004899879454996282,
      "loss": 5.8711,
      "mean_token_accuracy": 0.16204603910446166,
      "num_tokens": 20428307.0,
      "step": 10440
    },
    {
      "entropy": 6.044814205169677,
      "epoch": 1.0440845704003598,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004899774150566358,
      "loss": 5.9094,
      "mean_token_accuracy": 0.1594282180070877,
      "num_tokens": 20437493.0,
      "step": 10445
    },
    {
      "entropy": 6.134442520141602,
      "epoch": 1.0445843954615885,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000489966879204867,
      "loss": 5.9309,
      "mean_token_accuracy": 0.15497382134199142,
      "num_tokens": 20448218.0,
      "step": 10450
    },
    {
      "entropy": 6.2330015182495115,
      "epoch": 1.045084220522817,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004899563379445869,
      "loss": 5.9662,
      "mean_token_accuracy": 0.16178002953529358,
      "num_tokens": 20457478.0,
      "step": 10455
    },
    {
      "entropy": 6.136241006851196,
      "epoch": 1.0455840455840455,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004899457912760605,
      "loss": 5.8447,
      "mean_token_accuracy": 0.15872877687215806,
      "num_tokens": 20466776.0,
      "step": 10460
    },
    {
      "entropy": 6.096967267990112,
      "epoch": 1.0460838706452742,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004899352391995533,
      "loss": 5.893,
      "mean_token_accuracy": 0.15971874445676804,
      "num_tokens": 20476461.0,
      "step": 10465
    },
    {
      "entropy": 6.094351720809937,
      "epoch": 1.0465836957065027,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004899246817153309,
      "loss": 5.9334,
      "mean_token_accuracy": 0.1564917892217636,
      "num_tokens": 20487022.0,
      "step": 10470
    },
    {
      "entropy": 6.130196809768677,
      "epoch": 1.0470835207677314,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004899141188236587,
      "loss": 5.887,
      "mean_token_accuracy": 0.16109856963157654,
      "num_tokens": 20496326.0,
      "step": 10475
    },
    {
      "entropy": 6.190846633911133,
      "epoch": 1.0475833458289598,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004899035505248027,
      "loss": 5.9052,
      "mean_token_accuracy": 0.15818253755569459,
      "num_tokens": 20504933.0,
      "step": 10480
    },
    {
      "entropy": 6.124947023391724,
      "epoch": 1.0480831708901883,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004898929768190286,
      "loss": 5.8878,
      "mean_token_accuracy": 0.1593479573726654,
      "num_tokens": 20513931.0,
      "step": 10485
    },
    {
      "entropy": 6.034820508956909,
      "epoch": 1.048582995951417,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004898823977066024,
      "loss": 5.8565,
      "mean_token_accuracy": 0.16673677563667297,
      "num_tokens": 20523176.0,
      "step": 10490
    },
    {
      "entropy": 6.133836078643799,
      "epoch": 1.0490828210126455,
      "grad_norm": 1.25,
      "learning_rate": 0.0004898718131877906,
      "loss": 5.9686,
      "mean_token_accuracy": 0.1552524112164974,
      "num_tokens": 20533049.0,
      "step": 10495
    },
    {
      "entropy": 6.127683496475219,
      "epoch": 1.0495826460738742,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004898612232628591,
      "loss": 5.8297,
      "mean_token_accuracy": 0.15948455035686493,
      "num_tokens": 20542132.0,
      "step": 10500
    },
    {
      "entropy": 6.12599892616272,
      "epoch": 1.0500824711351027,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004898506279320745,
      "loss": 5.9564,
      "mean_token_accuracy": 0.15409419685602188,
      "num_tokens": 20551779.0,
      "step": 10505
    },
    {
      "entropy": 6.138923788070679,
      "epoch": 1.0505822961963314,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004898400271957034,
      "loss": 5.9667,
      "mean_token_accuracy": 0.1574355900287628,
      "num_tokens": 20560548.0,
      "step": 10510
    },
    {
      "entropy": 6.185645484924317,
      "epoch": 1.0510821212575598,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004898294210540126,
      "loss": 6.0526,
      "mean_token_accuracy": 0.15029460936784744,
      "num_tokens": 20570909.0,
      "step": 10515
    },
    {
      "entropy": 6.249118423461914,
      "epoch": 1.0515819463187883,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004898188095072688,
      "loss": 5.9659,
      "mean_token_accuracy": 0.15334199666976928,
      "num_tokens": 20580367.0,
      "step": 10520
    },
    {
      "entropy": 6.149303531646728,
      "epoch": 1.052081771380017,
      "grad_norm": 1.171875,
      "learning_rate": 0.000489808192555739,
      "loss": 5.8749,
      "mean_token_accuracy": 0.1581062227487564,
      "num_tokens": 20589653.0,
      "step": 10525
    },
    {
      "entropy": 6.102756118774414,
      "epoch": 1.0525815964412455,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004897975701996903,
      "loss": 5.8967,
      "mean_token_accuracy": 0.15787848755717276,
      "num_tokens": 20599530.0,
      "step": 10530
    },
    {
      "entropy": 6.0699773788452145,
      "epoch": 1.0530814215024742,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004897869424393902,
      "loss": 5.8983,
      "mean_token_accuracy": 0.1571897730231285,
      "num_tokens": 20607989.0,
      "step": 10535
    },
    {
      "entropy": 6.1367888927459715,
      "epoch": 1.0535812465637027,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004897763092751057,
      "loss": 5.8689,
      "mean_token_accuracy": 0.15787062495946885,
      "num_tokens": 20618315.0,
      "step": 10540
    },
    {
      "entropy": 6.15116515159607,
      "epoch": 1.0540810716249314,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004897656707071045,
      "loss": 5.9495,
      "mean_token_accuracy": 0.15609804391860962,
      "num_tokens": 20627554.0,
      "step": 10545
    },
    {
      "entropy": 6.1289714813232425,
      "epoch": 1.0545808966861598,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004897550267356545,
      "loss": 5.9373,
      "mean_token_accuracy": 0.1588480606675148,
      "num_tokens": 20636841.0,
      "step": 10550
    },
    {
      "entropy": 6.1143111228942875,
      "epoch": 1.0550807217473883,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004897443773610231,
      "loss": 5.8662,
      "mean_token_accuracy": 0.15951626896858215,
      "num_tokens": 20647129.0,
      "step": 10555
    },
    {
      "entropy": 6.109253835678101,
      "epoch": 1.055580546808617,
      "grad_norm": 1.25,
      "learning_rate": 0.0004897337225834785,
      "loss": 5.9687,
      "mean_token_accuracy": 0.1503675773739815,
      "num_tokens": 20657971.0,
      "step": 10560
    },
    {
      "entropy": 6.124502277374267,
      "epoch": 1.0560803718698455,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004897230624032888,
      "loss": 5.9464,
      "mean_token_accuracy": 0.15286152958869934,
      "num_tokens": 20666266.0,
      "step": 10565
    },
    {
      "entropy": 6.220788240432739,
      "epoch": 1.0565801969310742,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004897123968207219,
      "loss": 5.906,
      "mean_token_accuracy": 0.1552642285823822,
      "num_tokens": 20676454.0,
      "step": 10570
    },
    {
      "entropy": 6.1276350021362305,
      "epoch": 1.0570800219923027,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004897017258360465,
      "loss": 6.0028,
      "mean_token_accuracy": 0.15198311060667039,
      "num_tokens": 20687333.0,
      "step": 10575
    },
    {
      "entropy": 6.062843608856201,
      "epoch": 1.0575798470535314,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004896910494495308,
      "loss": 5.827,
      "mean_token_accuracy": 0.1605209469795227,
      "num_tokens": 20696530.0,
      "step": 10580
    },
    {
      "entropy": 6.177244186401367,
      "epoch": 1.0580796721147598,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004896803676614436,
      "loss": 5.9972,
      "mean_token_accuracy": 0.15510690063238144,
      "num_tokens": 20706674.0,
      "step": 10585
    },
    {
      "entropy": 6.124896621704101,
      "epoch": 1.0585794971759883,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004896696804720536,
      "loss": 5.7829,
      "mean_token_accuracy": 0.17689590454101561,
      "num_tokens": 20715695.0,
      "step": 10590
    },
    {
      "entropy": 6.160253381729126,
      "epoch": 1.059079322237217,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004896589878816297,
      "loss": 5.9995,
      "mean_token_accuracy": 0.15657817721366882,
      "num_tokens": 20725763.0,
      "step": 10595
    },
    {
      "entropy": 6.1244189739227295,
      "epoch": 1.0595791472984455,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004896482898904409,
      "loss": 5.9234,
      "mean_token_accuracy": 0.15617608428001403,
      "num_tokens": 20734944.0,
      "step": 10600
    },
    {
      "entropy": 6.119585084915161,
      "epoch": 1.0600789723596742,
      "grad_norm": 1.25,
      "learning_rate": 0.0004896375864987563,
      "loss": 5.8784,
      "mean_token_accuracy": 0.164742611348629,
      "num_tokens": 20744951.0,
      "step": 10605
    },
    {
      "entropy": 6.0973857879638675,
      "epoch": 1.0605787974209027,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004896268777068453,
      "loss": 5.8458,
      "mean_token_accuracy": 0.16812437325716018,
      "num_tokens": 20754937.0,
      "step": 10610
    },
    {
      "entropy": 6.19054217338562,
      "epoch": 1.0610786224821314,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004896161635149773,
      "loss": 5.9479,
      "mean_token_accuracy": 0.1503271222114563,
      "num_tokens": 20767373.0,
      "step": 10615
    },
    {
      "entropy": 6.169440650939942,
      "epoch": 1.0615784475433598,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004896054439234217,
      "loss": 5.8892,
      "mean_token_accuracy": 0.15658601820468904,
      "num_tokens": 20777647.0,
      "step": 10620
    },
    {
      "entropy": 6.139704751968384,
      "epoch": 1.0620782726045883,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004895947189324486,
      "loss": 5.9026,
      "mean_token_accuracy": 0.16581550240516663,
      "num_tokens": 20788598.0,
      "step": 10625
    },
    {
      "entropy": 6.079235506057739,
      "epoch": 1.062578097665817,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004895839885423273,
      "loss": 5.8743,
      "mean_token_accuracy": 0.1627487763762474,
      "num_tokens": 20798586.0,
      "step": 10630
    },
    {
      "entropy": 6.063961410522461,
      "epoch": 1.0630779227270455,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004895732527533283,
      "loss": 5.8716,
      "mean_token_accuracy": 0.16372207030653954,
      "num_tokens": 20807447.0,
      "step": 10635
    },
    {
      "entropy": 6.074852800369262,
      "epoch": 1.0635777477882742,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004895625115657215,
      "loss": 5.8825,
      "mean_token_accuracy": 0.16121718436479568,
      "num_tokens": 20817889.0,
      "step": 10640
    },
    {
      "entropy": 6.096667098999023,
      "epoch": 1.0640775728495027,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000489551764979777,
      "loss": 5.905,
      "mean_token_accuracy": 0.15882671177387236,
      "num_tokens": 20827858.0,
      "step": 10645
    },
    {
      "entropy": 6.0945783138275145,
      "epoch": 1.0645773979107314,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004895410129957653,
      "loss": 5.8848,
      "mean_token_accuracy": 0.16504572331905365,
      "num_tokens": 20837114.0,
      "step": 10650
    },
    {
      "entropy": 6.169197988510132,
      "epoch": 1.0650772229719598,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004895302556139569,
      "loss": 5.9531,
      "mean_token_accuracy": 0.15619509667158127,
      "num_tokens": 20846979.0,
      "step": 10655
    },
    {
      "entropy": 6.153785133361817,
      "epoch": 1.0655770480331883,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004895194928346226,
      "loss": 5.9655,
      "mean_token_accuracy": 0.1587654784321785,
      "num_tokens": 20856831.0,
      "step": 10660
    },
    {
      "entropy": 6.088301420211792,
      "epoch": 1.066076873094417,
      "grad_norm": 2.140625,
      "learning_rate": 0.000489508724658033,
      "loss": 5.8626,
      "mean_token_accuracy": 0.16275615841150284,
      "num_tokens": 20866203.0,
      "step": 10665
    },
    {
      "entropy": 6.152517080307007,
      "epoch": 1.0665766981556455,
      "grad_norm": 1.1875,
      "learning_rate": 0.000489497951084459,
      "loss": 5.8928,
      "mean_token_accuracy": 0.15525520220398903,
      "num_tokens": 20875813.0,
      "step": 10670
    },
    {
      "entropy": 6.180606269836426,
      "epoch": 1.0670765232168742,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004894871721141718,
      "loss": 5.9439,
      "mean_token_accuracy": 0.15791741013526917,
      "num_tokens": 20884773.0,
      "step": 10675
    },
    {
      "entropy": 6.109880971908569,
      "epoch": 1.0675763482781027,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004894763877474426,
      "loss": 5.8854,
      "mean_token_accuracy": 0.16144793927669526,
      "num_tokens": 20894305.0,
      "step": 10680
    },
    {
      "entropy": 6.133791255950928,
      "epoch": 1.0680761733393311,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004894655979845427,
      "loss": 5.972,
      "mean_token_accuracy": 0.15857400894165039,
      "num_tokens": 20903849.0,
      "step": 10685
    },
    {
      "entropy": 6.1582294464111325,
      "epoch": 1.0685759984005598,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004894548028257434,
      "loss": 5.845,
      "mean_token_accuracy": 0.16525670737028123,
      "num_tokens": 20913380.0,
      "step": 10690
    },
    {
      "entropy": 6.106751203536987,
      "epoch": 1.0690758234617883,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004894440022713165,
      "loss": 5.9298,
      "mean_token_accuracy": 0.15463113337755202,
      "num_tokens": 20923921.0,
      "step": 10695
    },
    {
      "entropy": 6.114072608947754,
      "epoch": 1.069575648523017,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004894331963215337,
      "loss": 5.8616,
      "mean_token_accuracy": 0.16015164107084273,
      "num_tokens": 20934613.0,
      "step": 10700
    },
    {
      "entropy": 6.13280291557312,
      "epoch": 1.0700754735842455,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000489422384976667,
      "loss": 5.9081,
      "mean_token_accuracy": 0.16159875988960265,
      "num_tokens": 20944052.0,
      "step": 10705
    },
    {
      "entropy": 6.156523418426514,
      "epoch": 1.0705752986454742,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004894115682369881,
      "loss": 5.9186,
      "mean_token_accuracy": 0.15661503076553346,
      "num_tokens": 20953406.0,
      "step": 10710
    },
    {
      "entropy": 6.184299087524414,
      "epoch": 1.0710751237067027,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004894007461027693,
      "loss": 6.0393,
      "mean_token_accuracy": 0.15095090046525,
      "num_tokens": 20963425.0,
      "step": 10715
    },
    {
      "entropy": 6.178989505767822,
      "epoch": 1.0715749487679311,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004893899185742828,
      "loss": 5.9861,
      "mean_token_accuracy": 0.14714545086026193,
      "num_tokens": 20973777.0,
      "step": 10720
    },
    {
      "entropy": 6.116341304779053,
      "epoch": 1.0720747738291598,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004893790856518014,
      "loss": 5.8659,
      "mean_token_accuracy": 0.15920069739222525,
      "num_tokens": 20983418.0,
      "step": 10725
    },
    {
      "entropy": 6.107203769683838,
      "epoch": 1.0725745988903883,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004893682473355972,
      "loss": 5.8357,
      "mean_token_accuracy": 0.1609717234969139,
      "num_tokens": 20993051.0,
      "step": 10730
    },
    {
      "entropy": 6.085955858230591,
      "epoch": 1.073074423951617,
      "grad_norm": 1.25,
      "learning_rate": 0.0004893574036259432,
      "loss": 5.8632,
      "mean_token_accuracy": 0.1652638018131256,
      "num_tokens": 21003439.0,
      "step": 10735
    },
    {
      "entropy": 6.146237802505493,
      "epoch": 1.0735742490128455,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004893465545231119,
      "loss": 5.911,
      "mean_token_accuracy": 0.16203626990318298,
      "num_tokens": 21013242.0,
      "step": 10740
    },
    {
      "entropy": 6.148274946212768,
      "epoch": 1.074074074074074,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004893357000273765,
      "loss": 5.9025,
      "mean_token_accuracy": 0.16121488362550734,
      "num_tokens": 21022268.0,
      "step": 10745
    },
    {
      "entropy": 6.167896842956543,
      "epoch": 1.0745738991353027,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048932484013901,
      "loss": 5.9765,
      "mean_token_accuracy": 0.15618660897016526,
      "num_tokens": 21032130.0,
      "step": 10750
    },
    {
      "entropy": 6.05360107421875,
      "epoch": 1.0750737241965311,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004893139748582857,
      "loss": 5.8248,
      "mean_token_accuracy": 0.16739086508750917,
      "num_tokens": 21041966.0,
      "step": 10755
    },
    {
      "entropy": 6.069804382324219,
      "epoch": 1.0755735492577598,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000489303104185477,
      "loss": 5.8052,
      "mean_token_accuracy": 0.1692075937986374,
      "num_tokens": 21051270.0,
      "step": 10760
    },
    {
      "entropy": 6.165479373931885,
      "epoch": 1.0760733743189883,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004892922281208573,
      "loss": 5.9964,
      "mean_token_accuracy": 0.15572390109300613,
      "num_tokens": 21060926.0,
      "step": 10765
    },
    {
      "entropy": 6.2376594066619875,
      "epoch": 1.076573199380217,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004892813466647002,
      "loss": 6.021,
      "mean_token_accuracy": 0.14834572225809098,
      "num_tokens": 21071367.0,
      "step": 10770
    },
    {
      "entropy": 6.141106271743775,
      "epoch": 1.0770730244414455,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004892704598172797,
      "loss": 5.9047,
      "mean_token_accuracy": 0.15415329933166505,
      "num_tokens": 21082775.0,
      "step": 10775
    },
    {
      "entropy": 6.154003047943116,
      "epoch": 1.077572849502674,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004892595675788696,
      "loss": 5.9916,
      "mean_token_accuracy": 0.15317206382751464,
      "num_tokens": 21092018.0,
      "step": 10780
    },
    {
      "entropy": 6.130233573913574,
      "epoch": 1.0780726745639027,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004892486699497439,
      "loss": 5.8944,
      "mean_token_accuracy": 0.1591568723320961,
      "num_tokens": 21101246.0,
      "step": 10785
    },
    {
      "entropy": 6.152314043045044,
      "epoch": 1.0785724996251311,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004892377669301766,
      "loss": 5.9491,
      "mean_token_accuracy": 0.15479602441191673,
      "num_tokens": 21110816.0,
      "step": 10790
    },
    {
      "entropy": 6.135138654708863,
      "epoch": 1.0790723246863598,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004892268585204423,
      "loss": 5.9179,
      "mean_token_accuracy": 0.1555817946791649,
      "num_tokens": 21120228.0,
      "step": 10795
    },
    {
      "entropy": 6.057647657394409,
      "epoch": 1.0795721497475883,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004892159447208154,
      "loss": 5.8097,
      "mean_token_accuracy": 0.16528261601924896,
      "num_tokens": 21129921.0,
      "step": 10800
    },
    {
      "entropy": 6.104941368103027,
      "epoch": 1.080071974808817,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004892050255315705,
      "loss": 6.0062,
      "mean_token_accuracy": 0.14804640114307405,
      "num_tokens": 21139605.0,
      "step": 10805
    },
    {
      "entropy": 6.125179958343506,
      "epoch": 1.0805717998700455,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004891941009529822,
      "loss": 5.8619,
      "mean_token_accuracy": 0.16716196537017822,
      "num_tokens": 21149004.0,
      "step": 10810
    },
    {
      "entropy": 6.215397071838379,
      "epoch": 1.081071624931274,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004891831709853255,
      "loss": 5.9453,
      "mean_token_accuracy": 0.14972266405820847,
      "num_tokens": 21159013.0,
      "step": 10815
    },
    {
      "entropy": 6.074891567230225,
      "epoch": 1.0815714499925027,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004891722356288752,
      "loss": 5.8787,
      "mean_token_accuracy": 0.15840475261211395,
      "num_tokens": 21169432.0,
      "step": 10820
    },
    {
      "entropy": 6.019592905044556,
      "epoch": 1.0820712750537311,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004891612948839068,
      "loss": 5.8357,
      "mean_token_accuracy": 0.1658037394285202,
      "num_tokens": 21179672.0,
      "step": 10825
    },
    {
      "entropy": 6.143423938751221,
      "epoch": 1.0825711001149598,
      "grad_norm": 1.25,
      "learning_rate": 0.0004891503487506951,
      "loss": 5.9114,
      "mean_token_accuracy": 0.15828991383314134,
      "num_tokens": 21189562.0,
      "step": 10830
    },
    {
      "entropy": 6.225865459442138,
      "epoch": 1.0830709251761883,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004891393972295159,
      "loss": 6.0317,
      "mean_token_accuracy": 0.14400122910737992,
      "num_tokens": 21199538.0,
      "step": 10835
    },
    {
      "entropy": 6.163006734848023,
      "epoch": 1.083570750237417,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004891284403206444,
      "loss": 5.9798,
      "mean_token_accuracy": 0.15748743787407876,
      "num_tokens": 21209355.0,
      "step": 10840
    },
    {
      "entropy": 6.100597476959228,
      "epoch": 1.0840705752986455,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004891174780243566,
      "loss": 5.8561,
      "mean_token_accuracy": 0.1640974059700966,
      "num_tokens": 21218268.0,
      "step": 10845
    },
    {
      "entropy": 6.162593603134155,
      "epoch": 1.084570400359874,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000489106510340928,
      "loss": 5.9868,
      "mean_token_accuracy": 0.15502519980072976,
      "num_tokens": 21228376.0,
      "step": 10850
    },
    {
      "entropy": 6.16921443939209,
      "epoch": 1.0850702254211027,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004890955372706348,
      "loss": 5.9662,
      "mean_token_accuracy": 0.15376733839511872,
      "num_tokens": 21237781.0,
      "step": 10855
    },
    {
      "entropy": 6.097012329101562,
      "epoch": 1.0855700504823311,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004890845588137531,
      "loss": 5.8316,
      "mean_token_accuracy": 0.16100382506847383,
      "num_tokens": 21247096.0,
      "step": 10860
    },
    {
      "entropy": 6.128241634368896,
      "epoch": 1.0860698755435598,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004890735749705587,
      "loss": 5.9896,
      "mean_token_accuracy": 0.15674687623977662,
      "num_tokens": 21258645.0,
      "step": 10865
    },
    {
      "entropy": 6.18025393486023,
      "epoch": 1.0865697006047883,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004890625857413283,
      "loss": 5.9317,
      "mean_token_accuracy": 0.15905121415853501,
      "num_tokens": 21268634.0,
      "step": 10870
    },
    {
      "entropy": 6.1775041103363035,
      "epoch": 1.087069525666017,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004890515911263385,
      "loss": 5.8956,
      "mean_token_accuracy": 0.15525186508893968,
      "num_tokens": 21278996.0,
      "step": 10875
    },
    {
      "entropy": 6.027877283096314,
      "epoch": 1.0875693507272455,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004890405911258656,
      "loss": 5.7187,
      "mean_token_accuracy": 0.17386804074048995,
      "num_tokens": 21288689.0,
      "step": 10880
    },
    {
      "entropy": 6.134686708450317,
      "epoch": 1.088069175788474,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004890295857401866,
      "loss": 5.9392,
      "mean_token_accuracy": 0.15089482963085174,
      "num_tokens": 21298542.0,
      "step": 10885
    },
    {
      "entropy": 6.111986351013184,
      "epoch": 1.0885690008497027,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004890185749695783,
      "loss": 5.8824,
      "mean_token_accuracy": 0.1575816184282303,
      "num_tokens": 21308035.0,
      "step": 10890
    },
    {
      "entropy": 6.2039025783538815,
      "epoch": 1.0890688259109311,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004890075588143176,
      "loss": 6.015,
      "mean_token_accuracy": 0.15227721780538558,
      "num_tokens": 21317898.0,
      "step": 10895
    },
    {
      "entropy": 6.175704717636108,
      "epoch": 1.0895686509721598,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004889965372746819,
      "loss": 5.9133,
      "mean_token_accuracy": 0.15824043303728103,
      "num_tokens": 21327347.0,
      "step": 10900
    },
    {
      "entropy": 6.182752323150635,
      "epoch": 1.0900684760333883,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004889855103509484,
      "loss": 5.9238,
      "mean_token_accuracy": 0.1543071150779724,
      "num_tokens": 21336592.0,
      "step": 10905
    },
    {
      "entropy": 6.1160050392150875,
      "epoch": 1.090568301094617,
      "grad_norm": 1.125,
      "learning_rate": 0.0004889744780433944,
      "loss": 5.9862,
      "mean_token_accuracy": 0.1515984743833542,
      "num_tokens": 21345764.0,
      "step": 10910
    },
    {
      "entropy": 6.148985052108765,
      "epoch": 1.0910681261558455,
      "grad_norm": 1.25,
      "learning_rate": 0.0004889634403522976,
      "loss": 5.9247,
      "mean_token_accuracy": 0.1537087544798851,
      "num_tokens": 21354423.0,
      "step": 10915
    },
    {
      "entropy": 6.130626058578491,
      "epoch": 1.091567951217074,
      "grad_norm": 1.25,
      "learning_rate": 0.0004889523972779359,
      "loss": 5.9521,
      "mean_token_accuracy": 0.16638178676366805,
      "num_tokens": 21364407.0,
      "step": 10920
    },
    {
      "entropy": 6.2047937393188475,
      "epoch": 1.0920677762783026,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004889413488205868,
      "loss": 6.0418,
      "mean_token_accuracy": 0.15266222655773162,
      "num_tokens": 21374642.0,
      "step": 10925
    },
    {
      "entropy": 6.073461818695068,
      "epoch": 1.0925676013395311,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004889302949805283,
      "loss": 5.832,
      "mean_token_accuracy": 0.17156516760587692,
      "num_tokens": 21384437.0,
      "step": 10930
    },
    {
      "entropy": 6.071567583084106,
      "epoch": 1.0930674264007598,
      "grad_norm": 1.25,
      "learning_rate": 0.0004889192357580388,
      "loss": 5.8433,
      "mean_token_accuracy": 0.164827623963356,
      "num_tokens": 21393924.0,
      "step": 10935
    },
    {
      "entropy": 6.07019362449646,
      "epoch": 1.0935672514619883,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004889081711533964,
      "loss": 5.8559,
      "mean_token_accuracy": 0.1653650552034378,
      "num_tokens": 21402924.0,
      "step": 10940
    },
    {
      "entropy": 6.222117471694946,
      "epoch": 1.0940670765232168,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004888971011668793,
      "loss": 5.9846,
      "mean_token_accuracy": 0.1487400785088539,
      "num_tokens": 21412422.0,
      "step": 10945
    },
    {
      "entropy": 6.130621719360351,
      "epoch": 1.0945669015844455,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004888860257987662,
      "loss": 5.9929,
      "mean_token_accuracy": 0.15240899845957756,
      "num_tokens": 21421830.0,
      "step": 10950
    },
    {
      "entropy": 6.114597606658935,
      "epoch": 1.095066726645674,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004888749450493358,
      "loss": 5.904,
      "mean_token_accuracy": 0.1606943741440773,
      "num_tokens": 21431602.0,
      "step": 10955
    },
    {
      "entropy": 6.14792218208313,
      "epoch": 1.0955665517069026,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004888638589188669,
      "loss": 5.9434,
      "mean_token_accuracy": 0.15879127606749535,
      "num_tokens": 21441110.0,
      "step": 10960
    },
    {
      "entropy": 6.1988667964935305,
      "epoch": 1.0960663767681311,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004888527674076382,
      "loss": 5.992,
      "mean_token_accuracy": 0.1468866728246212,
      "num_tokens": 21450626.0,
      "step": 10965
    },
    {
      "entropy": 6.089318513870239,
      "epoch": 1.0965662018293598,
      "grad_norm": 1.15625,
      "learning_rate": 0.000488841670515929,
      "loss": 5.8301,
      "mean_token_accuracy": 0.16008712202310563,
      "num_tokens": 21461174.0,
      "step": 10970
    },
    {
      "entropy": 6.111391830444336,
      "epoch": 1.0970660268905883,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004888305682440183,
      "loss": 5.9057,
      "mean_token_accuracy": 0.1626032307744026,
      "num_tokens": 21470895.0,
      "step": 10975
    },
    {
      "entropy": 6.088943386077881,
      "epoch": 1.0975658519518168,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004888194605921855,
      "loss": 5.8386,
      "mean_token_accuracy": 0.16586087346076966,
      "num_tokens": 21480192.0,
      "step": 10980
    },
    {
      "entropy": 6.169686269760132,
      "epoch": 1.0980656770130455,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004888083475607102,
      "loss": 5.9445,
      "mean_token_accuracy": 0.15836086571216584,
      "num_tokens": 21489418.0,
      "step": 10985
    },
    {
      "entropy": 6.2061357498168945,
      "epoch": 1.098565502074274,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004887972291498718,
      "loss": 5.9977,
      "mean_token_accuracy": 0.15747082084417344,
      "num_tokens": 21499228.0,
      "step": 10990
    },
    {
      "entropy": 6.144369316101074,
      "epoch": 1.0990653271355026,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004887861053599503,
      "loss": 5.8727,
      "mean_token_accuracy": 0.15833927243947982,
      "num_tokens": 21509051.0,
      "step": 10995
    },
    {
      "entropy": 6.083475637435913,
      "epoch": 1.0995651521967311,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004887749761912252,
      "loss": 5.9259,
      "mean_token_accuracy": 0.15681913495063782,
      "num_tokens": 21518853.0,
      "step": 11000
    },
    {
      "entropy": 6.172072696685791,
      "epoch": 1.1000649772579596,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004887638416439768,
      "loss": 5.9767,
      "mean_token_accuracy": 0.15682023763656616,
      "num_tokens": 21528547.0,
      "step": 11005
    },
    {
      "entropy": 6.25327696800232,
      "epoch": 1.1005648023191883,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004887527017184851,
      "loss": 5.9812,
      "mean_token_accuracy": 0.14949675351381303,
      "num_tokens": 21538166.0,
      "step": 11010
    },
    {
      "entropy": 6.199994945526123,
      "epoch": 1.1010646273804168,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004887415564150305,
      "loss": 5.9261,
      "mean_token_accuracy": 0.15772126615047455,
      "num_tokens": 21548212.0,
      "step": 11015
    },
    {
      "entropy": 6.1589757919311525,
      "epoch": 1.1015644524416455,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004887304057338934,
      "loss": 5.9238,
      "mean_token_accuracy": 0.14836643487215043,
      "num_tokens": 21558809.0,
      "step": 11020
    },
    {
      "entropy": 6.0988829135894775,
      "epoch": 1.102064277502874,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000488719249675354,
      "loss": 5.8057,
      "mean_token_accuracy": 0.16677905023097991,
      "num_tokens": 21567923.0,
      "step": 11025
    },
    {
      "entropy": 6.0830385208129885,
      "epoch": 1.1025641025641026,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004887080882396936,
      "loss": 5.9367,
      "mean_token_accuracy": 0.16132512092590331,
      "num_tokens": 21578051.0,
      "step": 11030
    },
    {
      "entropy": 6.085067224502564,
      "epoch": 1.1030639276253311,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004886969214271925,
      "loss": 5.9643,
      "mean_token_accuracy": 0.16186195164918898,
      "num_tokens": 21586950.0,
      "step": 11035
    },
    {
      "entropy": 6.179152631759644,
      "epoch": 1.1035637526865596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004886857492381319,
      "loss": 5.9323,
      "mean_token_accuracy": 0.15611904263496398,
      "num_tokens": 21596577.0,
      "step": 11040
    },
    {
      "entropy": 6.165682363510132,
      "epoch": 1.1040635777477883,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004886745716727928,
      "loss": 5.9661,
      "mean_token_accuracy": 0.15750504583120345,
      "num_tokens": 21606474.0,
      "step": 11045
    },
    {
      "entropy": 6.046479940414429,
      "epoch": 1.1045634028090168,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004886633887314565,
      "loss": 5.7809,
      "mean_token_accuracy": 0.17183364927768707,
      "num_tokens": 21616277.0,
      "step": 11050
    },
    {
      "entropy": 6.1723658561706545,
      "epoch": 1.1050632278702455,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004886522004144044,
      "loss": 5.9841,
      "mean_token_accuracy": 0.14777544513344765,
      "num_tokens": 21626096.0,
      "step": 11055
    },
    {
      "entropy": 6.2020745277404785,
      "epoch": 1.105563052931474,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004886410067219178,
      "loss": 6.0089,
      "mean_token_accuracy": 0.15448441207408906,
      "num_tokens": 21635295.0,
      "step": 11060
    },
    {
      "entropy": 6.197398233413696,
      "epoch": 1.1060628779927026,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004886298076542785,
      "loss": 5.993,
      "mean_token_accuracy": 0.1461438611149788,
      "num_tokens": 21645900.0,
      "step": 11065
    },
    {
      "entropy": 6.127649211883545,
      "epoch": 1.1065627030539311,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004886186032117682,
      "loss": 5.9064,
      "mean_token_accuracy": 0.15611540228128434,
      "num_tokens": 21654757.0,
      "step": 11070
    },
    {
      "entropy": 6.0400580883026125,
      "epoch": 1.1070625281151596,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004886073933946688,
      "loss": 5.8512,
      "mean_token_accuracy": 0.16595082134008407,
      "num_tokens": 21665253.0,
      "step": 11075
    },
    {
      "entropy": 6.016236352920532,
      "epoch": 1.1075623531763883,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004885961782032624,
      "loss": 5.7616,
      "mean_token_accuracy": 0.16749607026576996,
      "num_tokens": 21674345.0,
      "step": 11080
    },
    {
      "entropy": 6.120702314376831,
      "epoch": 1.1080621782376168,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000488584957637831,
      "loss": 5.9737,
      "mean_token_accuracy": 0.15983041524887084,
      "num_tokens": 21683110.0,
      "step": 11085
    },
    {
      "entropy": 6.082137680053711,
      "epoch": 1.1085620032988455,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004885737316986571,
      "loss": 5.8224,
      "mean_token_accuracy": 0.1675446406006813,
      "num_tokens": 21692141.0,
      "step": 11090
    },
    {
      "entropy": 6.140713024139404,
      "epoch": 1.109061828360074,
      "grad_norm": 1.453125,
      "learning_rate": 0.000488562500386023,
      "loss": 5.9136,
      "mean_token_accuracy": 0.16146373450756074,
      "num_tokens": 21700945.0,
      "step": 11095
    },
    {
      "entropy": 6.022051763534546,
      "epoch": 1.1095616534213026,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004885512637002115,
      "loss": 5.8372,
      "mean_token_accuracy": 0.16943920999765397,
      "num_tokens": 21711630.0,
      "step": 11100
    },
    {
      "entropy": 6.172645521163941,
      "epoch": 1.1100614784825311,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004885400216415051,
      "loss": 6.0348,
      "mean_token_accuracy": 0.15020374804735184,
      "num_tokens": 21721673.0,
      "step": 11105
    },
    {
      "entropy": 6.2499021053314205,
      "epoch": 1.1105613035437596,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004885287742101865,
      "loss": 5.9522,
      "mean_token_accuracy": 0.14965983107686043,
      "num_tokens": 21731655.0,
      "step": 11110
    },
    {
      "entropy": 6.106457757949829,
      "epoch": 1.1110611286049883,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004885175214065392,
      "loss": 5.9538,
      "mean_token_accuracy": 0.16075284779071808,
      "num_tokens": 21741465.0,
      "step": 11115
    },
    {
      "entropy": 6.123031759262085,
      "epoch": 1.1115609536662168,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004885062632308457,
      "loss": 5.9039,
      "mean_token_accuracy": 0.16515492051839828,
      "num_tokens": 21750313.0,
      "step": 11120
    },
    {
      "entropy": 6.219555282592774,
      "epoch": 1.1120607787274455,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004884949996833898,
      "loss": 6.0114,
      "mean_token_accuracy": 0.15466771572828292,
      "num_tokens": 21759842.0,
      "step": 11125
    },
    {
      "entropy": 6.179280376434326,
      "epoch": 1.112560603788674,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004884837307644545,
      "loss": 5.8729,
      "mean_token_accuracy": 0.1651467725634575,
      "num_tokens": 21770126.0,
      "step": 11130
    },
    {
      "entropy": 6.037139272689819,
      "epoch": 1.1130604288499026,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004884724564743236,
      "loss": 5.8127,
      "mean_token_accuracy": 0.16151363849639894,
      "num_tokens": 21779878.0,
      "step": 11135
    },
    {
      "entropy": 6.115792751312256,
      "epoch": 1.1135602539111311,
      "grad_norm": 1.25,
      "learning_rate": 0.0004884611768132805,
      "loss": 5.9761,
      "mean_token_accuracy": 0.15403736084699632,
      "num_tokens": 21790971.0,
      "step": 11140
    },
    {
      "entropy": 6.177181959152222,
      "epoch": 1.1140600789723596,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004884498917816092,
      "loss": 5.9937,
      "mean_token_accuracy": 0.15131354853510856,
      "num_tokens": 21800733.0,
      "step": 11145
    },
    {
      "entropy": 6.178098964691162,
      "epoch": 1.1145599040335883,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004884386013795935,
      "loss": 5.8781,
      "mean_token_accuracy": 0.16289085745811463,
      "num_tokens": 21809840.0,
      "step": 11150
    },
    {
      "entropy": 6.029925727844239,
      "epoch": 1.1150597290948168,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004884273056075175,
      "loss": 5.8943,
      "mean_token_accuracy": 0.15777575373649597,
      "num_tokens": 21819539.0,
      "step": 11155
    },
    {
      "entropy": 6.120708036422729,
      "epoch": 1.1155595541560455,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004884160044656654,
      "loss": 5.9097,
      "mean_token_accuracy": 0.16547042578458787,
      "num_tokens": 21829488.0,
      "step": 11160
    },
    {
      "entropy": 6.275621318817139,
      "epoch": 1.116059379217274,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004884046979543217,
      "loss": 5.9753,
      "mean_token_accuracy": 0.1529346413910389,
      "num_tokens": 21839244.0,
      "step": 11165
    },
    {
      "entropy": 6.182572937011718,
      "epoch": 1.1165592042785026,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004883933860737706,
      "loss": 6.057,
      "mean_token_accuracy": 0.1522720903158188,
      "num_tokens": 21850102.0,
      "step": 11170
    },
    {
      "entropy": 6.238038778305054,
      "epoch": 1.1170590293397311,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004883820688242968,
      "loss": 5.9392,
      "mean_token_accuracy": 0.1606079176068306,
      "num_tokens": 21860409.0,
      "step": 11175
    },
    {
      "entropy": 6.072983646392823,
      "epoch": 1.1175588544009596,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004883707462061851,
      "loss": 5.8109,
      "mean_token_accuracy": 0.1597381204366684,
      "num_tokens": 21870055.0,
      "step": 11180
    },
    {
      "entropy": 6.127948570251465,
      "epoch": 1.1180586794621883,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004883594182197203,
      "loss": 5.9612,
      "mean_token_accuracy": 0.1562331050634384,
      "num_tokens": 21879103.0,
      "step": 11185
    },
    {
      "entropy": 6.109658193588257,
      "epoch": 1.1185585045234168,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004883480848651875,
      "loss": 5.9864,
      "mean_token_accuracy": 0.15493618398904802,
      "num_tokens": 21890140.0,
      "step": 11190
    },
    {
      "entropy": 6.273085403442383,
      "epoch": 1.1190583295846455,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004883367461428716,
      "loss": 6.1108,
      "mean_token_accuracy": 0.14526225477457047,
      "num_tokens": 21900917.0,
      "step": 11195
    },
    {
      "entropy": 6.212919092178344,
      "epoch": 1.119558154645874,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004883254020530582,
      "loss": 5.9383,
      "mean_token_accuracy": 0.14905427247285843,
      "num_tokens": 21910476.0,
      "step": 11200
    },
    {
      "entropy": 6.132585048675537,
      "epoch": 1.1200579797071024,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004883140525960327,
      "loss": 5.9157,
      "mean_token_accuracy": 0.1560958907008171,
      "num_tokens": 21920219.0,
      "step": 11205
    },
    {
      "entropy": 6.119680166244507,
      "epoch": 1.1205578047683311,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004883026977720803,
      "loss": 5.8815,
      "mean_token_accuracy": 0.1568511337041855,
      "num_tokens": 21930592.0,
      "step": 11210
    },
    {
      "entropy": 6.237420320510864,
      "epoch": 1.1210576298295596,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000488291337581487,
      "loss": 6.0996,
      "mean_token_accuracy": 0.14509026780724527,
      "num_tokens": 21939785.0,
      "step": 11215
    },
    {
      "entropy": 6.178774356842041,
      "epoch": 1.1215574548907883,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004882799720245385,
      "loss": 5.902,
      "mean_token_accuracy": 0.16120699644088746,
      "num_tokens": 21948243.0,
      "step": 11220
    },
    {
      "entropy": 6.152544736862183,
      "epoch": 1.1220572799520168,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004882686011015208,
      "loss": 5.9679,
      "mean_token_accuracy": 0.15412325263023377,
      "num_tokens": 21958523.0,
      "step": 11225
    },
    {
      "entropy": 6.1496899127960205,
      "epoch": 1.1225571050132455,
      "grad_norm": 1.125,
      "learning_rate": 0.00048825722481272006,
      "loss": 5.9506,
      "mean_token_accuracy": 0.15271241813898087,
      "num_tokens": 21968685.0,
      "step": 11230
    },
    {
      "entropy": 6.143970012664795,
      "epoch": 1.123056930074474,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004882458431584223,
      "loss": 5.9671,
      "mean_token_accuracy": 0.15277467668056488,
      "num_tokens": 21979896.0,
      "step": 11235
    },
    {
      "entropy": 6.1666834354400635,
      "epoch": 1.1235567551357024,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004882344561389141,
      "loss": 5.9488,
      "mean_token_accuracy": 0.15477074682712555,
      "num_tokens": 21989513.0,
      "step": 11240
    },
    {
      "entropy": 6.111846256256103,
      "epoch": 1.124056580196931,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048822306375448176,
      "loss": 5.8974,
      "mean_token_accuracy": 0.1582782819867134,
      "num_tokens": 21999238.0,
      "step": 11245
    },
    {
      "entropy": 6.09161343574524,
      "epoch": 1.1245564052581596,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004882116660054121,
      "loss": 5.9158,
      "mean_token_accuracy": 0.15433163493871688,
      "num_tokens": 22008992.0,
      "step": 11250
    },
    {
      "entropy": 6.1394720554351805,
      "epoch": 1.1250562303193883,
      "grad_norm": 1.125,
      "learning_rate": 0.0004882002628919917,
      "loss": 5.9071,
      "mean_token_accuracy": 0.1569565489888191,
      "num_tokens": 22018333.0,
      "step": 11255
    },
    {
      "entropy": 6.14697494506836,
      "epoch": 1.1255560553806168,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004881888544145076,
      "loss": 5.9505,
      "mean_token_accuracy": 0.16382328569889068,
      "num_tokens": 22028003.0,
      "step": 11260
    },
    {
      "entropy": 6.193151426315308,
      "epoch": 1.1260558804418452,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048817744057324673,
      "loss": 5.9688,
      "mean_token_accuracy": 0.15252194702625274,
      "num_tokens": 22038318.0,
      "step": 11265
    },
    {
      "entropy": 6.132825183868408,
      "epoch": 1.126555705503074,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048816602136849644,
      "loss": 5.9018,
      "mean_token_accuracy": 0.15703375786542892,
      "num_tokens": 22048180.0,
      "step": 11270
    },
    {
      "entropy": 6.119627952575684,
      "epoch": 1.1270555305643024,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048815459680054375,
      "loss": 5.9634,
      "mean_token_accuracy": 0.14813498929142951,
      "num_tokens": 22057831.0,
      "step": 11275
    },
    {
      "entropy": 6.133615207672119,
      "epoch": 1.127555355625531,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048814316686967635,
      "loss": 5.8524,
      "mean_token_accuracy": 0.16055646240711213,
      "num_tokens": 22068569.0,
      "step": 11280
    },
    {
      "entropy": 6.192531490325928,
      "epoch": 1.1280551806867596,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004881317315761817,
      "loss": 6.0089,
      "mean_token_accuracy": 0.14873782992362977,
      "num_tokens": 22079267.0,
      "step": 11285
    },
    {
      "entropy": 6.1397772312164305,
      "epoch": 1.1285550057479883,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048812029092034747,
      "loss": 5.8638,
      "mean_token_accuracy": 0.16101511120796203,
      "num_tokens": 22089469.0,
      "step": 11290
    },
    {
      "entropy": 6.140107011795044,
      "epoch": 1.1290548308092168,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048810884490246164,
      "loss": 5.9079,
      "mean_token_accuracy": 0.16269891932606698,
      "num_tokens": 22099148.0,
      "step": 11295
    },
    {
      "entropy": 6.107700967788697,
      "epoch": 1.1295546558704452,
      "grad_norm": 1.1875,
      "learning_rate": 0.000488097393522812,
      "loss": 5.8853,
      "mean_token_accuracy": 0.1590560108423233,
      "num_tokens": 22107750.0,
      "step": 11300
    },
    {
      "entropy": 6.1589192867279055,
      "epoch": 1.130054480931674,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004880859367816868,
      "loss": 6.0008,
      "mean_token_accuracy": 0.15062985941767693,
      "num_tokens": 22118336.0,
      "step": 11305
    },
    {
      "entropy": 6.146308469772339,
      "epoch": 1.1305543059929024,
      "grad_norm": 1.125,
      "learning_rate": 0.0004880744746793744,
      "loss": 5.9186,
      "mean_token_accuracy": 0.15588910430669783,
      "num_tokens": 22128166.0,
      "step": 11310
    },
    {
      "entropy": 6.196415710449219,
      "epoch": 1.131054131054131,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048806300721616295,
      "loss": 5.8985,
      "mean_token_accuracy": 0.15742285251617433,
      "num_tokens": 22136810.0,
      "step": 11315
    },
    {
      "entropy": 6.098136615753174,
      "epoch": 1.1315539561153596,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048805153439234107,
      "loss": 5.8614,
      "mean_token_accuracy": 0.1702558621764183,
      "num_tokens": 22146398.0,
      "step": 11320
    },
    {
      "entropy": 6.126552438735962,
      "epoch": 1.1320537811765883,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004880400562081974,
      "loss": 5.9179,
      "mean_token_accuracy": 0.1547420486807823,
      "num_tokens": 22155161.0,
      "step": 11325
    },
    {
      "entropy": 6.042434453964233,
      "epoch": 1.1325536062378168,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048802857266402075,
      "loss": 5.935,
      "mean_token_accuracy": 0.16181783974170685,
      "num_tokens": 22166486.0,
      "step": 11330
    },
    {
      "entropy": 6.094811487197876,
      "epoch": 1.1330534312990452,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004880170837601,
      "loss": 5.8448,
      "mean_token_accuracy": 0.1616156131029129,
      "num_tokens": 22176487.0,
      "step": 11335
    },
    {
      "entropy": 6.1747722148895265,
      "epoch": 1.133553256360274,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004880055894967242,
      "loss": 5.9282,
      "mean_token_accuracy": 0.15591204911470413,
      "num_tokens": 22186513.0,
      "step": 11340
    },
    {
      "entropy": 6.146732234954834,
      "epoch": 1.1340530814215024,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048799408987418264,
      "loss": 5.9422,
      "mean_token_accuracy": 0.1589522197842598,
      "num_tokens": 22197325.0,
      "step": 11345
    },
    {
      "entropy": 6.236143493652344,
      "epoch": 1.134552906482731,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048798258489276457,
      "loss": 5.9568,
      "mean_token_accuracy": 0.15502262860536575,
      "num_tokens": 22207160.0,
      "step": 11350
    },
    {
      "entropy": 6.183321142196656,
      "epoch": 1.1350527315439596,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004879710745527595,
      "loss": 5.9631,
      "mean_token_accuracy": 0.15324283689260482,
      "num_tokens": 22216533.0,
      "step": 11355
    },
    {
      "entropy": 6.128221702575684,
      "epoch": 1.1355525566051883,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048795955885445696,
      "loss": 5.9173,
      "mean_token_accuracy": 0.15494934916496278,
      "num_tokens": 22226186.0,
      "step": 11360
    },
    {
      "entropy": 6.11665530204773,
      "epoch": 1.1360523816664168,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004879480377981467,
      "loss": 5.8839,
      "mean_token_accuracy": 0.16089436113834382,
      "num_tokens": 22236315.0,
      "step": 11365
    },
    {
      "entropy": 6.156322622299195,
      "epoch": 1.1365522067276452,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048793651138411866,
      "loss": 5.862,
      "mean_token_accuracy": 0.16175806969404222,
      "num_tokens": 22245250.0,
      "step": 11370
    },
    {
      "entropy": 6.163503551483155,
      "epoch": 1.137052031788874,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048792497961266273,
      "loss": 5.991,
      "mean_token_accuracy": 0.15363508760929107,
      "num_tokens": 22255093.0,
      "step": 11375
    },
    {
      "entropy": 5.987509298324585,
      "epoch": 1.1375518568501024,
      "grad_norm": 1.125,
      "learning_rate": 0.00048791344248406916,
      "loss": 5.8033,
      "mean_token_accuracy": 0.16229974776506423,
      "num_tokens": 22264680.0,
      "step": 11380
    },
    {
      "entropy": 6.130703496932983,
      "epoch": 1.138051681911331,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048790189999862814,
      "loss": 5.9775,
      "mean_token_accuracy": 0.15135350674390793,
      "num_tokens": 22273804.0,
      "step": 11385
    },
    {
      "entropy": 6.1591321468353275,
      "epoch": 1.1385515069725596,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048789035215663013,
      "loss": 5.9133,
      "mean_token_accuracy": 0.156303970515728,
      "num_tokens": 22284302.0,
      "step": 11390
    },
    {
      "entropy": 6.0950844287872314,
      "epoch": 1.1390513320337883,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048787879895836563,
      "loss": 5.8578,
      "mean_token_accuracy": 0.16435280293226243,
      "num_tokens": 22293439.0,
      "step": 11395
    },
    {
      "entropy": 6.093597173690796,
      "epoch": 1.1395511570950168,
      "grad_norm": 1.125,
      "learning_rate": 0.0004878672404041253,
      "loss": 5.9155,
      "mean_token_accuracy": 0.15771330147981644,
      "num_tokens": 22303005.0,
      "step": 11400
    },
    {
      "entropy": 6.140190315246582,
      "epoch": 1.1400509821562452,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048785567649419994,
      "loss": 5.8068,
      "mean_token_accuracy": 0.16318179965019225,
      "num_tokens": 22311962.0,
      "step": 11405
    },
    {
      "entropy": 6.071467781066895,
      "epoch": 1.140550807217474,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004878441072288806,
      "loss": 5.9336,
      "mean_token_accuracy": 0.15063284039497377,
      "num_tokens": 22320998.0,
      "step": 11410
    },
    {
      "entropy": 6.134148693084716,
      "epoch": 1.1410506322787024,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048783253260845826,
      "loss": 5.9063,
      "mean_token_accuracy": 0.15667446330189705,
      "num_tokens": 22330977.0,
      "step": 11415
    },
    {
      "entropy": 6.208115005493164,
      "epoch": 1.141550457339931,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004878209526332242,
      "loss": 5.903,
      "mean_token_accuracy": 0.15968807637691498,
      "num_tokens": 22341287.0,
      "step": 11420
    },
    {
      "entropy": 6.138490200042725,
      "epoch": 1.1420502824011596,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048780936730346967,
      "loss": 5.8963,
      "mean_token_accuracy": 0.16170920431613922,
      "num_tokens": 22350507.0,
      "step": 11425
    },
    {
      "entropy": 6.131403684616089,
      "epoch": 1.1425501074623883,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048779777661948627,
      "loss": 5.9845,
      "mean_token_accuracy": 0.1480890154838562,
      "num_tokens": 22361202.0,
      "step": 11430
    },
    {
      "entropy": 6.217830324172974,
      "epoch": 1.1430499325236168,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004877861805815655,
      "loss": 6.0206,
      "mean_token_accuracy": 0.1587601199746132,
      "num_tokens": 22370824.0,
      "step": 11435
    },
    {
      "entropy": 6.069406032562256,
      "epoch": 1.1435497575848452,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004877745791899993,
      "loss": 5.8705,
      "mean_token_accuracy": 0.16209238469600679,
      "num_tokens": 22380857.0,
      "step": 11440
    },
    {
      "entropy": 6.207471036911011,
      "epoch": 1.144049582646074,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048776297244507936,
      "loss": 5.8881,
      "mean_token_accuracy": 0.15709212869405748,
      "num_tokens": 22391354.0,
      "step": 11445
    },
    {
      "entropy": 6.117812442779541,
      "epoch": 1.1445494077073024,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048775136034709784,
      "loss": 5.8476,
      "mean_token_accuracy": 0.16575112640857698,
      "num_tokens": 22400656.0,
      "step": 11450
    },
    {
      "entropy": 6.119547700881958,
      "epoch": 1.145049232768531,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048773974289634676,
      "loss": 5.9086,
      "mean_token_accuracy": 0.15170229226350784,
      "num_tokens": 22410887.0,
      "step": 11455
    },
    {
      "entropy": 6.116162824630737,
      "epoch": 1.1455490578297596,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004877281200931185,
      "loss": 5.9321,
      "mean_token_accuracy": 0.15286416560411453,
      "num_tokens": 22420370.0,
      "step": 11460
    },
    {
      "entropy": 6.059341049194336,
      "epoch": 1.1460488828909883,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048771649193770554,
      "loss": 5.8091,
      "mean_token_accuracy": 0.16730551272630692,
      "num_tokens": 22430746.0,
      "step": 11465
    },
    {
      "entropy": 6.133656215667725,
      "epoch": 1.1465487079522168,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004877048584304004,
      "loss": 5.8997,
      "mean_token_accuracy": 0.156261645257473,
      "num_tokens": 22440812.0,
      "step": 11470
    },
    {
      "entropy": 6.215521955490113,
      "epoch": 1.1470485330134452,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048769321957149557,
      "loss": 6.0483,
      "mean_token_accuracy": 0.14938197210431098,
      "num_tokens": 22450460.0,
      "step": 11475
    },
    {
      "entropy": 6.147781610488892,
      "epoch": 1.147548358074674,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004876815753612843,
      "loss": 5.9196,
      "mean_token_accuracy": 0.16128621250391006,
      "num_tokens": 22460660.0,
      "step": 11480
    },
    {
      "entropy": 6.0839306831359865,
      "epoch": 1.1480481831359024,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004876699258000592,
      "loss": 5.847,
      "mean_token_accuracy": 0.16610171794891357,
      "num_tokens": 22469238.0,
      "step": 11485
    },
    {
      "entropy": 6.066226816177368,
      "epoch": 1.148548008197131,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004876582708881135,
      "loss": 5.8804,
      "mean_token_accuracy": 0.15787602812051774,
      "num_tokens": 22479685.0,
      "step": 11490
    },
    {
      "entropy": 6.088213109970093,
      "epoch": 1.1490478332583596,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004876466106257404,
      "loss": 5.8275,
      "mean_token_accuracy": 0.1616256818175316,
      "num_tokens": 22489770.0,
      "step": 11495
    },
    {
      "entropy": 6.201134395599365,
      "epoch": 1.149547658319588,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048763494501323333,
      "loss": 6.0162,
      "mean_token_accuracy": 0.15101008266210555,
      "num_tokens": 22499957.0,
      "step": 11500
    },
    {
      "entropy": 6.0968852043151855,
      "epoch": 1.1500474833808167,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048762327405088573,
      "loss": 5.8255,
      "mean_token_accuracy": 0.1649577483534813,
      "num_tokens": 22508863.0,
      "step": 11505
    },
    {
      "entropy": 6.156015014648437,
      "epoch": 1.1505473084420452,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004876115977389913,
      "loss": 5.9184,
      "mean_token_accuracy": 0.154385244846344,
      "num_tokens": 22517563.0,
      "step": 11510
    },
    {
      "entropy": 6.006227588653564,
      "epoch": 1.151047133503274,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048759991607784375,
      "loss": 5.7718,
      "mean_token_accuracy": 0.17087078839540482,
      "num_tokens": 22527907.0,
      "step": 11515
    },
    {
      "entropy": 6.019870328903198,
      "epoch": 1.1515469585645024,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048758822906773706,
      "loss": 5.8312,
      "mean_token_accuracy": 0.16606579571962357,
      "num_tokens": 22537373.0,
      "step": 11520
    },
    {
      "entropy": 6.159598445892334,
      "epoch": 1.1520467836257309,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048757653670896523,
      "loss": 5.9806,
      "mean_token_accuracy": 0.15638059750199318,
      "num_tokens": 22547851.0,
      "step": 11525
    },
    {
      "entropy": 6.242302751541137,
      "epoch": 1.1525466086869596,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004875648390018224,
      "loss": 5.9529,
      "mean_token_accuracy": 0.15457541942596437,
      "num_tokens": 22558127.0,
      "step": 11530
    },
    {
      "entropy": 6.105423021316528,
      "epoch": 1.153046433748188,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000487553135946603,
      "loss": 5.9805,
      "mean_token_accuracy": 0.1586910977959633,
      "num_tokens": 22568151.0,
      "step": 11535
    },
    {
      "entropy": 6.138805294036866,
      "epoch": 1.1535462588094167,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048754142754360135,
      "loss": 6.047,
      "mean_token_accuracy": 0.15200379639863967,
      "num_tokens": 22579772.0,
      "step": 11540
    },
    {
      "entropy": 6.164836597442627,
      "epoch": 1.1540460838706452,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004875297137931121,
      "loss": 5.9291,
      "mean_token_accuracy": 0.15910463482141496,
      "num_tokens": 22589551.0,
      "step": 11545
    },
    {
      "entropy": 6.152883434295655,
      "epoch": 1.154545908931874,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004875179946954298,
      "loss": 5.9367,
      "mean_token_accuracy": 0.15863673985004426,
      "num_tokens": 22599139.0,
      "step": 11550
    },
    {
      "entropy": 6.184917736053467,
      "epoch": 1.1550457339931024,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048750627025084955,
      "loss": 6.0481,
      "mean_token_accuracy": 0.1498355060815811,
      "num_tokens": 22609413.0,
      "step": 11555
    },
    {
      "entropy": 6.269936180114746,
      "epoch": 1.1555455590543309,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004874945404596662,
      "loss": 6.0529,
      "mean_token_accuracy": 0.15218576043844223,
      "num_tokens": 22619395.0,
      "step": 11560
    },
    {
      "entropy": 6.111987924575805,
      "epoch": 1.1560453841155596,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048748280532217485,
      "loss": 5.8488,
      "mean_token_accuracy": 0.16580688953399658,
      "num_tokens": 22629623.0,
      "step": 11565
    },
    {
      "entropy": 6.02946138381958,
      "epoch": 1.156545209176788,
      "grad_norm": 1.25,
      "learning_rate": 0.0004874710648386709,
      "loss": 5.8968,
      "mean_token_accuracy": 0.15211973786354066,
      "num_tokens": 22639735.0,
      "step": 11570
    },
    {
      "entropy": 6.061814165115356,
      "epoch": 1.1570450342380167,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004874593190094496,
      "loss": 5.8089,
      "mean_token_accuracy": 0.16090735644102097,
      "num_tokens": 22648563.0,
      "step": 11575
    },
    {
      "entropy": 6.0428526401519775,
      "epoch": 1.1575448592992452,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004874475678348065,
      "loss": 5.8437,
      "mean_token_accuracy": 0.16174384504556655,
      "num_tokens": 22657137.0,
      "step": 11580
    },
    {
      "entropy": 6.076289796829224,
      "epoch": 1.158044684360474,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048743581131503716,
      "loss": 5.875,
      "mean_token_accuracy": 0.1554312564432621,
      "num_tokens": 22667987.0,
      "step": 11585
    },
    {
      "entropy": 6.111766004562378,
      "epoch": 1.1585445094217024,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004874240494504376,
      "loss": 5.9602,
      "mean_token_accuracy": 0.15176709666848182,
      "num_tokens": 22678830.0,
      "step": 11590
    },
    {
      "entropy": 6.240999174118042,
      "epoch": 1.1590443344829309,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048741228224130357,
      "loss": 5.9557,
      "mean_token_accuracy": 0.15433995127677919,
      "num_tokens": 22690568.0,
      "step": 11595
    },
    {
      "entropy": 6.16453218460083,
      "epoch": 1.1595441595441596,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048740050968793116,
      "loss": 5.9503,
      "mean_token_accuracy": 0.15487554669380188,
      "num_tokens": 22700150.0,
      "step": 11600
    },
    {
      "entropy": 6.012238264083862,
      "epoch": 1.160043984605388,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004873887317906166,
      "loss": 5.8552,
      "mean_token_accuracy": 0.15723595023155212,
      "num_tokens": 22709365.0,
      "step": 11605
    },
    {
      "entropy": 6.203862524032592,
      "epoch": 1.1605438096666167,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048737694854965617,
      "loss": 6.032,
      "mean_token_accuracy": 0.15099229887127877,
      "num_tokens": 22718636.0,
      "step": 11610
    },
    {
      "entropy": 6.210437631607055,
      "epoch": 1.1610436347278452,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048736515996534644,
      "loss": 6.015,
      "mean_token_accuracy": 0.15517956763505936,
      "num_tokens": 22728713.0,
      "step": 11615
    },
    {
      "entropy": 5.978393507003784,
      "epoch": 1.161543459789074,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004873533660379839,
      "loss": 5.7516,
      "mean_token_accuracy": 0.16850547343492508,
      "num_tokens": 22739061.0,
      "step": 11620
    },
    {
      "entropy": 6.201608324050904,
      "epoch": 1.1620432848503024,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048734156676786525,
      "loss": 5.9701,
      "mean_token_accuracy": 0.150502098351717,
      "num_tokens": 22748752.0,
      "step": 11625
    },
    {
      "entropy": 6.06671257019043,
      "epoch": 1.1625431099115309,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004873297621552875,
      "loss": 5.8308,
      "mean_token_accuracy": 0.16410483419895172,
      "num_tokens": 22758695.0,
      "step": 11630
    },
    {
      "entropy": 6.107732057571411,
      "epoch": 1.1630429349727596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048731795220054755,
      "loss": 5.9918,
      "mean_token_accuracy": 0.15247361809015275,
      "num_tokens": 22768948.0,
      "step": 11635
    },
    {
      "entropy": 6.134869766235352,
      "epoch": 1.163542760033988,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004873061369039425,
      "loss": 5.9393,
      "mean_token_accuracy": 0.15633041262626649,
      "num_tokens": 22779446.0,
      "step": 11640
    },
    {
      "entropy": 6.111254119873047,
      "epoch": 1.1640425850952167,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004872943162657697,
      "loss": 5.8606,
      "mean_token_accuracy": 0.16059275418519975,
      "num_tokens": 22789654.0,
      "step": 11645
    },
    {
      "entropy": 6.147224950790405,
      "epoch": 1.1645424101564452,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004872824902863265,
      "loss": 5.9725,
      "mean_token_accuracy": 0.1565178096294403,
      "num_tokens": 22798938.0,
      "step": 11650
    },
    {
      "entropy": 6.1495428562164305,
      "epoch": 1.165042235217674,
      "grad_norm": 1.125,
      "learning_rate": 0.0004872706589659106,
      "loss": 5.9359,
      "mean_token_accuracy": 0.1596113160252571,
      "num_tokens": 22808279.0,
      "step": 11655
    },
    {
      "entropy": 6.099793100357056,
      "epoch": 1.1655420602789024,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048725882230481947,
      "loss": 5.8893,
      "mean_token_accuracy": 0.15741243660449983,
      "num_tokens": 22817378.0,
      "step": 11660
    },
    {
      "entropy": 6.237301969528199,
      "epoch": 1.1660418853401309,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048724698030335096,
      "loss": 5.9817,
      "mean_token_accuracy": 0.14953510463237762,
      "num_tokens": 22828652.0,
      "step": 11665
    },
    {
      "entropy": 6.12910304069519,
      "epoch": 1.1665417104013596,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004872351329618031,
      "loss": 5.9291,
      "mean_token_accuracy": 0.16131235808134078,
      "num_tokens": 22839156.0,
      "step": 11670
    },
    {
      "entropy": 6.1852038383483885,
      "epoch": 1.167041535462588,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048722328028047386,
      "loss": 6.0058,
      "mean_token_accuracy": 0.15359752327203752,
      "num_tokens": 22849004.0,
      "step": 11675
    },
    {
      "entropy": 6.12124924659729,
      "epoch": 1.1675413605238167,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048721142225966146,
      "loss": 5.974,
      "mean_token_accuracy": 0.1512797400355339,
      "num_tokens": 22858984.0,
      "step": 11680
    },
    {
      "entropy": 6.116902112960815,
      "epoch": 1.1680411855850452,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004871995588996644,
      "loss": 5.8705,
      "mean_token_accuracy": 0.16128026694059372,
      "num_tokens": 22869149.0,
      "step": 11685
    },
    {
      "entropy": 6.19780125617981,
      "epoch": 1.168541010646274,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004871876902007809,
      "loss": 6.0255,
      "mean_token_accuracy": 0.1531459003686905,
      "num_tokens": 22878522.0,
      "step": 11690
    },
    {
      "entropy": 6.138104009628296,
      "epoch": 1.1690408357075024,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048717581616330985,
      "loss": 5.9154,
      "mean_token_accuracy": 0.1585894778370857,
      "num_tokens": 22888356.0,
      "step": 11695
    },
    {
      "entropy": 6.133490705490113,
      "epoch": 1.1695406607687309,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048716393678754975,
      "loss": 5.9145,
      "mean_token_accuracy": 0.15694701969623565,
      "num_tokens": 22897551.0,
      "step": 11700
    },
    {
      "entropy": 6.189616441726685,
      "epoch": 1.1700404858299596,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004871520520737996,
      "loss": 5.8953,
      "mean_token_accuracy": 0.1604890763759613,
      "num_tokens": 22906480.0,
      "step": 11705
    },
    {
      "entropy": 6.104088020324707,
      "epoch": 1.170540310891188,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004871401620223585,
      "loss": 5.7866,
      "mean_token_accuracy": 0.1621207520365715,
      "num_tokens": 22916544.0,
      "step": 11710
    },
    {
      "entropy": 6.041656923294068,
      "epoch": 1.1710401359524167,
      "grad_norm": 1.25,
      "learning_rate": 0.0004871282666335255,
      "loss": 5.8366,
      "mean_token_accuracy": 0.16841777116060258,
      "num_tokens": 22926083.0,
      "step": 11715
    },
    {
      "entropy": 6.19201078414917,
      "epoch": 1.1715399610136452,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048711636590759985,
      "loss": 5.9626,
      "mean_token_accuracy": 0.1518273413181305,
      "num_tokens": 22936127.0,
      "step": 11720
    },
    {
      "entropy": 6.159239006042481,
      "epoch": 1.172039786074874,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048710445984488106,
      "loss": 5.8885,
      "mean_token_accuracy": 0.1582811012864113,
      "num_tokens": 22946579.0,
      "step": 11725
    },
    {
      "entropy": 6.100362968444824,
      "epoch": 1.1725396111361024,
      "grad_norm": 1.125,
      "learning_rate": 0.0004870925484456686,
      "loss": 5.9144,
      "mean_token_accuracy": 0.16010892391204834,
      "num_tokens": 22956286.0,
      "step": 11730
    },
    {
      "entropy": 6.067992877960205,
      "epoch": 1.1730394361973309,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004870806317102622,
      "loss": 5.9536,
      "mean_token_accuracy": 0.15850478857755662,
      "num_tokens": 22965625.0,
      "step": 11735
    },
    {
      "entropy": 6.250034809112549,
      "epoch": 1.1735392612585596,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004870687096389617,
      "loss": 5.9942,
      "mean_token_accuracy": 0.1539678767323494,
      "num_tokens": 22975853.0,
      "step": 11740
    },
    {
      "entropy": 6.1940278053283695,
      "epoch": 1.174039086319788,
      "grad_norm": 1.3125,
      "learning_rate": 0.000487056782232067,
      "loss": 5.9607,
      "mean_token_accuracy": 0.1527936950325966,
      "num_tokens": 22985928.0,
      "step": 11745
    },
    {
      "entropy": 6.1184369087219235,
      "epoch": 1.1745389113810167,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048704484948987824,
      "loss": 5.9158,
      "mean_token_accuracy": 0.15915521383285522,
      "num_tokens": 22995222.0,
      "step": 11750
    },
    {
      "entropy": 6.12689437866211,
      "epoch": 1.1750387364422452,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004870329114126956,
      "loss": 5.9032,
      "mean_token_accuracy": 0.15619448125362395,
      "num_tokens": 23005735.0,
      "step": 11755
    },
    {
      "entropy": 6.107010126113892,
      "epoch": 1.1755385615034737,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004870209680008196,
      "loss": 5.869,
      "mean_token_accuracy": 0.1552758365869522,
      "num_tokens": 23016198.0,
      "step": 11760
    },
    {
      "entropy": 6.104145765304565,
      "epoch": 1.1760383865647024,
      "grad_norm": 1.25,
      "learning_rate": 0.0004870090192545505,
      "loss": 5.9575,
      "mean_token_accuracy": 0.15270556658506393,
      "num_tokens": 23025927.0,
      "step": 11765
    },
    {
      "entropy": 6.108724737167359,
      "epoch": 1.1765382116259309,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000486997065174189,
      "loss": 5.773,
      "mean_token_accuracy": 0.16741551756858825,
      "num_tokens": 23034626.0,
      "step": 11770
    },
    {
      "entropy": 6.1271192073822025,
      "epoch": 1.1770380366871596,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048698510576003593,
      "loss": 5.8617,
      "mean_token_accuracy": 0.1574167400598526,
      "num_tokens": 23045773.0,
      "step": 11775
    },
    {
      "entropy": 6.11742057800293,
      "epoch": 1.177537861748388,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048697314101239217,
      "loss": 5.8339,
      "mean_token_accuracy": 0.15908393412828445,
      "num_tokens": 23056803.0,
      "step": 11780
    },
    {
      "entropy": 6.088882350921631,
      "epoch": 1.1780376868096165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004869611709315587,
      "loss": 5.9098,
      "mean_token_accuracy": 0.15271433368325232,
      "num_tokens": 23066513.0,
      "step": 11785
    },
    {
      "entropy": 6.158081483840943,
      "epoch": 1.1785375118708452,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048694919551783665,
      "loss": 5.9397,
      "mean_token_accuracy": 0.1509637676179409,
      "num_tokens": 23077309.0,
      "step": 11790
    },
    {
      "entropy": 6.016227197647095,
      "epoch": 1.1790373369320737,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048693721477152755,
      "loss": 5.8939,
      "mean_token_accuracy": 0.16637452095746993,
      "num_tokens": 23088292.0,
      "step": 11795
    },
    {
      "entropy": 6.149525833129883,
      "epoch": 1.1795371619933024,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048692522869293246,
      "loss": 5.8811,
      "mean_token_accuracy": 0.15308161452412605,
      "num_tokens": 23098744.0,
      "step": 11800
    },
    {
      "entropy": 6.111117315292359,
      "epoch": 1.1800369870545309,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004869132372823533,
      "loss": 5.85,
      "mean_token_accuracy": 0.1664982721209526,
      "num_tokens": 23108223.0,
      "step": 11805
    },
    {
      "entropy": 6.076059246063233,
      "epoch": 1.1805368121157596,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048690124054009157,
      "loss": 5.941,
      "mean_token_accuracy": 0.15591681748628616,
      "num_tokens": 23118595.0,
      "step": 11810
    },
    {
      "entropy": 6.094328308105469,
      "epoch": 1.181036637176988,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004868892384664491,
      "loss": 5.7867,
      "mean_token_accuracy": 0.1678330510854721,
      "num_tokens": 23127640.0,
      "step": 11815
    },
    {
      "entropy": 6.163733100891113,
      "epoch": 1.1815364622382165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048687723106172805,
      "loss": 5.9295,
      "mean_token_accuracy": 0.15475355088710785,
      "num_tokens": 23138409.0,
      "step": 11820
    },
    {
      "entropy": 6.100816059112549,
      "epoch": 1.1820362872994452,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048686521832623026,
      "loss": 5.9014,
      "mean_token_accuracy": 0.16128589063882828,
      "num_tokens": 23148427.0,
      "step": 11825
    },
    {
      "entropy": 6.1230696678161625,
      "epoch": 1.1825361123606737,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004868532002602581,
      "loss": 5.9131,
      "mean_token_accuracy": 0.16012639552354813,
      "num_tokens": 23157873.0,
      "step": 11830
    },
    {
      "entropy": 6.071974611282348,
      "epoch": 1.1830359374219024,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000486841176864114,
      "loss": 5.8539,
      "mean_token_accuracy": 0.16170584559440612,
      "num_tokens": 23166488.0,
      "step": 11835
    },
    {
      "entropy": 6.074644899368286,
      "epoch": 1.1835357624831309,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048682914813810035,
      "loss": 5.7938,
      "mean_token_accuracy": 0.16883472353219986,
      "num_tokens": 23176548.0,
      "step": 11840
    },
    {
      "entropy": 6.0430079936981205,
      "epoch": 1.1840355875443596,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048681711408251986,
      "loss": 5.8759,
      "mean_token_accuracy": 0.15774462670087813,
      "num_tokens": 23186269.0,
      "step": 11845
    },
    {
      "entropy": 6.0795704364776615,
      "epoch": 1.184535412605588,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048680507469767533,
      "loss": 5.8351,
      "mean_token_accuracy": 0.16153195649385452,
      "num_tokens": 23197181.0,
      "step": 11850
    },
    {
      "entropy": 6.106466674804688,
      "epoch": 1.1850352376668165,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004867930299838696,
      "loss": 5.8232,
      "mean_token_accuracy": 0.16667625904083253,
      "num_tokens": 23207391.0,
      "step": 11855
    },
    {
      "entropy": 6.2232287406921385,
      "epoch": 1.1855350627280452,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048678097994140566,
      "loss": 6.0666,
      "mean_token_accuracy": 0.1549858883023262,
      "num_tokens": 23217510.0,
      "step": 11860
    },
    {
      "entropy": 6.036048746109008,
      "epoch": 1.1860348877892737,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048676892457058687,
      "loss": 5.7921,
      "mean_token_accuracy": 0.16512476801872253,
      "num_tokens": 23226746.0,
      "step": 11865
    },
    {
      "entropy": 6.129440546035767,
      "epoch": 1.1865347128505024,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004867568638717164,
      "loss": 5.9455,
      "mean_token_accuracy": 0.1580897569656372,
      "num_tokens": 23236333.0,
      "step": 11870
    },
    {
      "entropy": 6.068762254714966,
      "epoch": 1.1870345379117309,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048674479784509766,
      "loss": 5.7641,
      "mean_token_accuracy": 0.16918087899684905,
      "num_tokens": 23245071.0,
      "step": 11875
    },
    {
      "entropy": 6.061614179611206,
      "epoch": 1.1875343629729596,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048673272649103435,
      "loss": 5.8598,
      "mean_token_accuracy": 0.1623161867260933,
      "num_tokens": 23254225.0,
      "step": 11880
    },
    {
      "entropy": 6.113137531280517,
      "epoch": 1.188034188034188,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048672064980983013,
      "loss": 5.8513,
      "mean_token_accuracy": 0.1643107756972313,
      "num_tokens": 23264822.0,
      "step": 11885
    },
    {
      "entropy": 6.149943161010742,
      "epoch": 1.1885340130954165,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048670856780178874,
      "loss": 5.9981,
      "mean_token_accuracy": 0.15331889688968658,
      "num_tokens": 23274559.0,
      "step": 11890
    },
    {
      "entropy": 6.132507419586181,
      "epoch": 1.1890338381566452,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048669648046721436,
      "loss": 5.9032,
      "mean_token_accuracy": 0.16584499776363373,
      "num_tokens": 23284431.0,
      "step": 11895
    },
    {
      "entropy": 6.150027465820313,
      "epoch": 1.1895336632178737,
      "grad_norm": 1.328125,
      "learning_rate": 0.000486684387806411,
      "loss": 5.9769,
      "mean_token_accuracy": 0.15231848508119583,
      "num_tokens": 23294814.0,
      "step": 11900
    },
    {
      "entropy": 6.146082019805908,
      "epoch": 1.1900334882791024,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048667228981968285,
      "loss": 5.8894,
      "mean_token_accuracy": 0.1553812265396118,
      "num_tokens": 23304701.0,
      "step": 11905
    },
    {
      "entropy": 6.207196140289307,
      "epoch": 1.1905333133403309,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048666018650733446,
      "loss": 5.9519,
      "mean_token_accuracy": 0.15436269119381904,
      "num_tokens": 23314335.0,
      "step": 11910
    },
    {
      "entropy": 6.103688669204712,
      "epoch": 1.1910331384015596,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004866480778696702,
      "loss": 5.9891,
      "mean_token_accuracy": 0.15368614047765733,
      "num_tokens": 23324591.0,
      "step": 11915
    },
    {
      "entropy": 6.037664794921875,
      "epoch": 1.191532963462788,
      "grad_norm": 1.125,
      "learning_rate": 0.00048663596390699477,
      "loss": 5.8866,
      "mean_token_accuracy": 0.16131106168031692,
      "num_tokens": 23334682.0,
      "step": 11920
    },
    {
      "entropy": 6.242043495178223,
      "epoch": 1.1920327885240165,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004866238446196129,
      "loss": 5.912,
      "mean_token_accuracy": 0.16002513766288756,
      "num_tokens": 23343943.0,
      "step": 11925
    },
    {
      "entropy": 6.144790983200073,
      "epoch": 1.1925326135852452,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004866117200078296,
      "loss": 5.8736,
      "mean_token_accuracy": 0.15773849189281464,
      "num_tokens": 23354031.0,
      "step": 11930
    },
    {
      "entropy": 6.105603170394898,
      "epoch": 1.1930324386464737,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048659959007194985,
      "loss": 5.9887,
      "mean_token_accuracy": 0.1507854975759983,
      "num_tokens": 23363922.0,
      "step": 11935
    },
    {
      "entropy": 6.115167665481567,
      "epoch": 1.1935322637077024,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048658745481227884,
      "loss": 5.9447,
      "mean_token_accuracy": 0.15322858095169067,
      "num_tokens": 23372552.0,
      "step": 11940
    },
    {
      "entropy": 6.099942779541015,
      "epoch": 1.1940320887689309,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000486575314229122,
      "loss": 5.8804,
      "mean_token_accuracy": 0.16570656299591063,
      "num_tokens": 23383454.0,
      "step": 11945
    },
    {
      "entropy": 6.174167680740356,
      "epoch": 1.1945319138301596,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004865631683227846,
      "loss": 5.9262,
      "mean_token_accuracy": 0.15205035209655762,
      "num_tokens": 23393429.0,
      "step": 11950
    },
    {
      "entropy": 6.142827463150025,
      "epoch": 1.195031738891388,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048655101709357243,
      "loss": 5.8789,
      "mean_token_accuracy": 0.15747981071472167,
      "num_tokens": 23403742.0,
      "step": 11955
    },
    {
      "entropy": 6.04516224861145,
      "epoch": 1.1955315639526165,
      "grad_norm": 1.21875,
      "learning_rate": 0.000486538860541791,
      "loss": 5.821,
      "mean_token_accuracy": 0.1605915680527687,
      "num_tokens": 23412875.0,
      "step": 11960
    },
    {
      "entropy": 6.119252109527588,
      "epoch": 1.1960313890138452,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048652669866774636,
      "loss": 5.9387,
      "mean_token_accuracy": 0.15322231501340866,
      "num_tokens": 23422838.0,
      "step": 11965
    },
    {
      "entropy": 6.174497747421265,
      "epoch": 1.1965312140750737,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004865145314717443,
      "loss": 5.8841,
      "mean_token_accuracy": 0.15678165182471276,
      "num_tokens": 23431407.0,
      "step": 11970
    },
    {
      "entropy": 6.050436925888062,
      "epoch": 1.1970310391363024,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048650235895409115,
      "loss": 5.8767,
      "mean_token_accuracy": 0.16674434244632722,
      "num_tokens": 23441580.0,
      "step": 11975
    },
    {
      "entropy": 6.073388767242432,
      "epoch": 1.1975308641975309,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004864901811150931,
      "loss": 5.9552,
      "mean_token_accuracy": 0.15157094448804856,
      "num_tokens": 23451508.0,
      "step": 11980
    },
    {
      "entropy": 6.115357255935669,
      "epoch": 1.1980306892587596,
      "grad_norm": 1.25,
      "learning_rate": 0.0004864779979550564,
      "loss": 5.9323,
      "mean_token_accuracy": 0.15766377300024031,
      "num_tokens": 23461928.0,
      "step": 11985
    },
    {
      "entropy": 6.142044878005981,
      "epoch": 1.198530514319988,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004864658094742878,
      "loss": 5.854,
      "mean_token_accuracy": 0.1676865816116333,
      "num_tokens": 23471501.0,
      "step": 11990
    },
    {
      "entropy": 6.261177396774292,
      "epoch": 1.1990303393812165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048645361567309375,
      "loss": 6.0179,
      "mean_token_accuracy": 0.16207321882247924,
      "num_tokens": 23481478.0,
      "step": 11995
    },
    {
      "entropy": 6.192924118041992,
      "epoch": 1.1995301644424452,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004864414165517812,
      "loss": 5.9625,
      "mean_token_accuracy": 0.15490425378084183,
      "num_tokens": 23491253.0,
      "step": 12000
    },
    {
      "epoch": 1.1995301644424452,
      "eval_entropy": 5.9069157600157105,
      "eval_loss": 6.023349285125732,
      "eval_mean_token_accuracy": 0.16134706332116236,
      "eval_num_tokens": 23491253.0,
      "eval_runtime": 23.9588,
      "eval_samples_per_second": 1295.768,
      "eval_steps_per_second": 161.987,
      "step": 12000
    },
    {
      "entropy": 6.102289152145386,
      "epoch": 1.2000299895036737,
      "grad_norm": 1.140625,
      "learning_rate": 0.000486429212110657,
      "loss": 5.9568,
      "mean_token_accuracy": 0.15527821630239486,
      "num_tokens": 23500753.0,
      "step": 12005
    },
    {
      "entropy": 6.173083353042602,
      "epoch": 1.2005298145649024,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004864170023500282,
      "loss": 5.9897,
      "mean_token_accuracy": 0.1506602570414543,
      "num_tokens": 23509837.0,
      "step": 12010
    },
    {
      "entropy": 6.134479141235351,
      "epoch": 1.2010296396261309,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004864047872702021,
      "loss": 5.8929,
      "mean_token_accuracy": 0.1577495738863945,
      "num_tokens": 23519926.0,
      "step": 12015
    },
    {
      "entropy": 6.125346374511719,
      "epoch": 1.2015294646873593,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004863925668714859,
      "loss": 6.0607,
      "mean_token_accuracy": 0.14753381311893463,
      "num_tokens": 23531165.0,
      "step": 12020
    },
    {
      "entropy": 6.1413110256195065,
      "epoch": 1.202029289748588,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004863803411541871,
      "loss": 5.8785,
      "mean_token_accuracy": 0.1558782145380974,
      "num_tokens": 23541696.0,
      "step": 12025
    },
    {
      "entropy": 6.169222354888916,
      "epoch": 1.2025291148098165,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004863681101186134,
      "loss": 5.9093,
      "mean_token_accuracy": 0.15498466044664383,
      "num_tokens": 23552382.0,
      "step": 12030
    },
    {
      "entropy": 6.151580190658569,
      "epoch": 1.2030289398710452,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004863558737650723,
      "loss": 5.9788,
      "mean_token_accuracy": 0.15583425760269165,
      "num_tokens": 23561444.0,
      "step": 12035
    },
    {
      "entropy": 6.091583776473999,
      "epoch": 1.2035287649322737,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004863436320938719,
      "loss": 5.8919,
      "mean_token_accuracy": 0.1557927832007408,
      "num_tokens": 23570985.0,
      "step": 12040
    },
    {
      "entropy": 6.002496147155762,
      "epoch": 1.2040285899935022,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048633138510532003,
      "loss": 5.8271,
      "mean_token_accuracy": 0.16926287859678268,
      "num_tokens": 23581291.0,
      "step": 12045
    },
    {
      "entropy": 6.169947481155395,
      "epoch": 1.2045284150547308,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048631913279972497,
      "loss": 5.9889,
      "mean_token_accuracy": 0.1531054586172104,
      "num_tokens": 23591087.0,
      "step": 12050
    },
    {
      "entropy": 6.171611976623535,
      "epoch": 1.2050282401159593,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048630687517739484,
      "loss": 5.969,
      "mean_token_accuracy": 0.14974582716822624,
      "num_tokens": 23601635.0,
      "step": 12055
    },
    {
      "entropy": 6.217007780075074,
      "epoch": 1.205528065177188,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004862946122386381,
      "loss": 5.9387,
      "mean_token_accuracy": 0.15877343714237213,
      "num_tokens": 23612523.0,
      "step": 12060
    },
    {
      "entropy": 6.096794748306275,
      "epoch": 1.2060278902384165,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048628234398376334,
      "loss": 5.9321,
      "mean_token_accuracy": 0.15881869792938233,
      "num_tokens": 23622258.0,
      "step": 12065
    },
    {
      "entropy": 6.031498193740845,
      "epoch": 1.2065277152996452,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048627007041307914,
      "loss": 5.7994,
      "mean_token_accuracy": 0.16519799083471298,
      "num_tokens": 23631198.0,
      "step": 12070
    },
    {
      "entropy": 6.171346855163574,
      "epoch": 1.2070275403608737,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048625779152689435,
      "loss": 5.9358,
      "mean_token_accuracy": 0.1546371266245842,
      "num_tokens": 23641052.0,
      "step": 12075
    },
    {
      "entropy": 6.218582677841186,
      "epoch": 1.2075273654221022,
      "grad_norm": 1.25,
      "learning_rate": 0.0004862455073255179,
      "loss": 5.9046,
      "mean_token_accuracy": 0.16020112335681916,
      "num_tokens": 23650785.0,
      "step": 12080
    },
    {
      "entropy": 6.089391088485717,
      "epoch": 1.2080271904833308,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048623321780925883,
      "loss": 5.9241,
      "mean_token_accuracy": 0.15806041806936263,
      "num_tokens": 23660548.0,
      "step": 12085
    },
    {
      "entropy": 5.984379673004151,
      "epoch": 1.2085270155445593,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048622092297842637,
      "loss": 5.8027,
      "mean_token_accuracy": 0.16259512007236482,
      "num_tokens": 23669979.0,
      "step": 12090
    },
    {
      "entropy": 6.153009700775146,
      "epoch": 1.209026840605788,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004862086228333298,
      "loss": 5.9645,
      "mean_token_accuracy": 0.15104301422834396,
      "num_tokens": 23679740.0,
      "step": 12095
    },
    {
      "entropy": 6.17504096031189,
      "epoch": 1.2095266656670165,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048619631737427865,
      "loss": 5.958,
      "mean_token_accuracy": 0.16154096722602845,
      "num_tokens": 23689688.0,
      "step": 12100
    },
    {
      "entropy": 6.14042100906372,
      "epoch": 1.2100264907282452,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004861840066015825,
      "loss": 5.8693,
      "mean_token_accuracy": 0.16754159033298494,
      "num_tokens": 23699550.0,
      "step": 12105
    },
    {
      "entropy": 6.123619508743286,
      "epoch": 1.2105263157894737,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048617169051555113,
      "loss": 5.9513,
      "mean_token_accuracy": 0.15646493434906006,
      "num_tokens": 23709114.0,
      "step": 12110
    },
    {
      "entropy": 6.152635335922241,
      "epoch": 1.2110261408507021,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004861593691164942,
      "loss": 5.9195,
      "mean_token_accuracy": 0.15780582875013352,
      "num_tokens": 23718758.0,
      "step": 12115
    },
    {
      "entropy": 6.133005094528198,
      "epoch": 1.2115259659119308,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004861470424047221,
      "loss": 5.891,
      "mean_token_accuracy": 0.15964819341897965,
      "num_tokens": 23728175.0,
      "step": 12120
    },
    {
      "entropy": 5.939164066314698,
      "epoch": 1.2120257909731593,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004861347103805446,
      "loss": 5.7325,
      "mean_token_accuracy": 0.17180475145578383,
      "num_tokens": 23738308.0,
      "step": 12125
    },
    {
      "entropy": 6.128927230834961,
      "epoch": 1.212525616034388,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004861223730442722,
      "loss": 5.9004,
      "mean_token_accuracy": 0.156107597053051,
      "num_tokens": 23747867.0,
      "step": 12130
    },
    {
      "entropy": 6.045731258392334,
      "epoch": 1.2130254410956165,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048611003039621516,
      "loss": 5.8396,
      "mean_token_accuracy": 0.16276348382234573,
      "num_tokens": 23758347.0,
      "step": 12135
    },
    {
      "entropy": 6.127686023712158,
      "epoch": 1.2135252661568452,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004860976824366841,
      "loss": 5.9151,
      "mean_token_accuracy": 0.16771481186151505,
      "num_tokens": 23767847.0,
      "step": 12140
    },
    {
      "entropy": 6.178754711151123,
      "epoch": 1.2140250912180737,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048608532916598964,
      "loss": 5.9453,
      "mean_token_accuracy": 0.15886199995875358,
      "num_tokens": 23777098.0,
      "step": 12145
    },
    {
      "entropy": 6.111155605316162,
      "epoch": 1.2145249162793021,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048607297058444267,
      "loss": 5.9482,
      "mean_token_accuracy": 0.16033577546477318,
      "num_tokens": 23787332.0,
      "step": 12150
    },
    {
      "entropy": 6.0930695056915285,
      "epoch": 1.2150247413405308,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004860606066923541,
      "loss": 5.8547,
      "mean_token_accuracy": 0.16679299175739287,
      "num_tokens": 23796263.0,
      "step": 12155
    },
    {
      "entropy": 6.113790130615234,
      "epoch": 1.2155245664017593,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048604823749003485,
      "loss": 5.8949,
      "mean_token_accuracy": 0.1568261593580246,
      "num_tokens": 23806051.0,
      "step": 12160
    },
    {
      "entropy": 6.12924747467041,
      "epoch": 1.216024391462988,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004860358629777963,
      "loss": 5.8408,
      "mean_token_accuracy": 0.1663694128394127,
      "num_tokens": 23814925.0,
      "step": 12165
    },
    {
      "entropy": 6.058565664291382,
      "epoch": 1.2165242165242165,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004860234831559498,
      "loss": 5.7901,
      "mean_token_accuracy": 0.16382089704275132,
      "num_tokens": 23824785.0,
      "step": 12170
    },
    {
      "entropy": 6.1559943675994875,
      "epoch": 1.2170240415854452,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048601109802480676,
      "loss": 5.9654,
      "mean_token_accuracy": 0.1467147946357727,
      "num_tokens": 23835102.0,
      "step": 12175
    },
    {
      "entropy": 6.1578285694122314,
      "epoch": 1.2175238666466737,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048599870758467875,
      "loss": 5.9303,
      "mean_token_accuracy": 0.15632469058036805,
      "num_tokens": 23845718.0,
      "step": 12180
    },
    {
      "entropy": 6.0674844741821286,
      "epoch": 1.2180236917079021,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048598631183587753,
      "loss": 5.8646,
      "mean_token_accuracy": 0.16330512762069702,
      "num_tokens": 23855078.0,
      "step": 12185
    },
    {
      "entropy": 6.152185487747192,
      "epoch": 1.2185235167691308,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048597391077871504,
      "loss": 5.9405,
      "mean_token_accuracy": 0.15571191012859345,
      "num_tokens": 23864287.0,
      "step": 12190
    },
    {
      "entropy": 6.173195219039917,
      "epoch": 1.2190233418303593,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048596150441350324,
      "loss": 6.025,
      "mean_token_accuracy": 0.14913687482476234,
      "num_tokens": 23874477.0,
      "step": 12195
    },
    {
      "entropy": 6.052039432525635,
      "epoch": 1.219523166891588,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004859490927405543,
      "loss": 5.799,
      "mean_token_accuracy": 0.1491182841360569,
      "num_tokens": 23884208.0,
      "step": 12200
    },
    {
      "entropy": 6.248810911178589,
      "epoch": 1.2200229919528165,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004859366757601804,
      "loss": 6.0198,
      "mean_token_accuracy": 0.1539500281214714,
      "num_tokens": 23894226.0,
      "step": 12205
    },
    {
      "entropy": 6.079480409622192,
      "epoch": 1.2205228170140452,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048592425347269405,
      "loss": 5.8436,
      "mean_token_accuracy": 0.16406033486127852,
      "num_tokens": 23904238.0,
      "step": 12210
    },
    {
      "entropy": 6.078441953659057,
      "epoch": 1.2210226420752737,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048591182587840777,
      "loss": 5.8527,
      "mean_token_accuracy": 0.16123137325048448,
      "num_tokens": 23914611.0,
      "step": 12215
    },
    {
      "entropy": 6.154345941543579,
      "epoch": 1.2215224671365021,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004858993929776342,
      "loss": 5.983,
      "mean_token_accuracy": 0.15250156670808793,
      "num_tokens": 23924270.0,
      "step": 12220
    },
    {
      "entropy": 6.143490266799927,
      "epoch": 1.2220222921977308,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048588695477068617,
      "loss": 5.8659,
      "mean_token_accuracy": 0.15965988039970397,
      "num_tokens": 23932677.0,
      "step": 12225
    },
    {
      "entropy": 6.136343765258789,
      "epoch": 1.2225221172589593,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048587451125787665,
      "loss": 5.8967,
      "mean_token_accuracy": 0.1526498794555664,
      "num_tokens": 23943259.0,
      "step": 12230
    },
    {
      "entropy": 6.133910417556763,
      "epoch": 1.223021942320188,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004858620624395187,
      "loss": 5.9539,
      "mean_token_accuracy": 0.15788768827915192,
      "num_tokens": 23953162.0,
      "step": 12235
    },
    {
      "entropy": 6.192581748962402,
      "epoch": 1.2235217673814165,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048584960831592547,
      "loss": 5.9857,
      "mean_token_accuracy": 0.15631678700447083,
      "num_tokens": 23961577.0,
      "step": 12240
    },
    {
      "entropy": 6.049570560455322,
      "epoch": 1.2240215924426452,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048583714888741035,
      "loss": 5.801,
      "mean_token_accuracy": 0.16860109865665435,
      "num_tokens": 23969691.0,
      "step": 12245
    },
    {
      "entropy": 6.0210247993469235,
      "epoch": 1.2245214175038737,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048582468415428687,
      "loss": 5.8778,
      "mean_token_accuracy": 0.15848847180604936,
      "num_tokens": 23980753.0,
      "step": 12250
    },
    {
      "entropy": 6.167635726928711,
      "epoch": 1.2250212425651021,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048581221411686857,
      "loss": 5.9415,
      "mean_token_accuracy": 0.15743371695280076,
      "num_tokens": 23991841.0,
      "step": 12255
    },
    {
      "entropy": 6.131909275054932,
      "epoch": 1.2255210676263308,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048579973877546916,
      "loss": 5.8641,
      "mean_token_accuracy": 0.1634696289896965,
      "num_tokens": 24000749.0,
      "step": 12260
    },
    {
      "entropy": 6.20639214515686,
      "epoch": 1.2260208926875593,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048578725813040263,
      "loss": 6.0611,
      "mean_token_accuracy": 0.1524174228310585,
      "num_tokens": 24010055.0,
      "step": 12265
    },
    {
      "entropy": 6.133679533004761,
      "epoch": 1.226520717748788,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004857747721819829,
      "loss": 5.9112,
      "mean_token_accuracy": 0.15845078825950623,
      "num_tokens": 24019307.0,
      "step": 12270
    },
    {
      "entropy": 6.1019188404083256,
      "epoch": 1.2270205428100165,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004857622809305241,
      "loss": 5.7644,
      "mean_token_accuracy": 0.1679854840040207,
      "num_tokens": 24028913.0,
      "step": 12275
    },
    {
      "entropy": 6.112960147857666,
      "epoch": 1.2275203678712452,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048574978437634057,
      "loss": 5.8012,
      "mean_token_accuracy": 0.16816159933805466,
      "num_tokens": 24038357.0,
      "step": 12280
    },
    {
      "entropy": 6.095643758773804,
      "epoch": 1.2280201929324737,
      "grad_norm": 1.125,
      "learning_rate": 0.0004857372825197468,
      "loss": 5.8649,
      "mean_token_accuracy": 0.16309794336557387,
      "num_tokens": 24048199.0,
      "step": 12285
    },
    {
      "entropy": 6.105725908279419,
      "epoch": 1.2285200179937021,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004857247753610571,
      "loss": 5.9362,
      "mean_token_accuracy": 0.15908170491456985,
      "num_tokens": 24058311.0,
      "step": 12290
    },
    {
      "entropy": 6.142247867584229,
      "epoch": 1.2290198430549308,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048571226290058635,
      "loss": 5.9115,
      "mean_token_accuracy": 0.15605428814888,
      "num_tokens": 24068471.0,
      "step": 12295
    },
    {
      "entropy": 6.141541767120361,
      "epoch": 1.2295196681161593,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004856997451386492,
      "loss": 5.9019,
      "mean_token_accuracy": 0.16354478448629378,
      "num_tokens": 24078118.0,
      "step": 12300
    },
    {
      "entropy": 6.16800889968872,
      "epoch": 1.2300194931773878,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004856872220755607,
      "loss": 5.8831,
      "mean_token_accuracy": 0.16593519002199172,
      "num_tokens": 24088517.0,
      "step": 12305
    },
    {
      "entropy": 6.217232990264892,
      "epoch": 1.2305193182386165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000485674693711636,
      "loss": 5.9944,
      "mean_token_accuracy": 0.1572984978556633,
      "num_tokens": 24097421.0,
      "step": 12310
    },
    {
      "entropy": 6.034582424163818,
      "epoch": 1.231019143299845,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004856621600471901,
      "loss": 5.8004,
      "mean_token_accuracy": 0.16696758717298507,
      "num_tokens": 24106531.0,
      "step": 12315
    },
    {
      "entropy": 6.071604013442993,
      "epoch": 1.2315189683610737,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048564962108253853,
      "loss": 5.8725,
      "mean_token_accuracy": 0.15827251970767975,
      "num_tokens": 24115246.0,
      "step": 12320
    },
    {
      "entropy": 6.197845506668091,
      "epoch": 1.2320187934223021,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004856370768179967,
      "loss": 5.9603,
      "mean_token_accuracy": 0.1553197145462036,
      "num_tokens": 24126163.0,
      "step": 12325
    },
    {
      "entropy": 6.121444272994995,
      "epoch": 1.2325186184835308,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048562452725388024,
      "loss": 5.9226,
      "mean_token_accuracy": 0.1507203310728073,
      "num_tokens": 24135722.0,
      "step": 12330
    },
    {
      "entropy": 6.029284381866455,
      "epoch": 1.2330184435447593,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004856119723905048,
      "loss": 5.737,
      "mean_token_accuracy": 0.17486302703619003,
      "num_tokens": 24145278.0,
      "step": 12335
    },
    {
      "entropy": 6.117523336410523,
      "epoch": 1.2335182686059878,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004855994122281864,
      "loss": 5.9126,
      "mean_token_accuracy": 0.15642016530036926,
      "num_tokens": 24154710.0,
      "step": 12340
    },
    {
      "entropy": 6.164228248596191,
      "epoch": 1.2340180936672165,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048558684676724083,
      "loss": 5.932,
      "mean_token_accuracy": 0.15176677852869033,
      "num_tokens": 24164373.0,
      "step": 12345
    },
    {
      "entropy": 6.047241163253784,
      "epoch": 1.234517918728445,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048557427600798454,
      "loss": 5.7966,
      "mean_token_accuracy": 0.16844219714403152,
      "num_tokens": 24173085.0,
      "step": 12350
    },
    {
      "entropy": 6.108587789535522,
      "epoch": 1.2350177437896737,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004855616999507336,
      "loss": 5.9401,
      "mean_token_accuracy": 0.1597892791032791,
      "num_tokens": 24182936.0,
      "step": 12355
    },
    {
      "entropy": 6.145156192779541,
      "epoch": 1.2355175688509021,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048554911859580445,
      "loss": 5.8964,
      "mean_token_accuracy": 0.15547910034656526,
      "num_tokens": 24192407.0,
      "step": 12360
    },
    {
      "entropy": 6.23434534072876,
      "epoch": 1.2360173939121308,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004855365319435137,
      "loss": 5.8998,
      "mean_token_accuracy": 0.1609629660844803,
      "num_tokens": 24201770.0,
      "step": 12365
    },
    {
      "entropy": 6.091249561309814,
      "epoch": 1.2365172189733593,
      "grad_norm": 1.21875,
      "learning_rate": 0.000485523939994178,
      "loss": 5.854,
      "mean_token_accuracy": 0.1622677341103554,
      "num_tokens": 24211287.0,
      "step": 12370
    },
    {
      "entropy": 6.0098296165466305,
      "epoch": 1.2370170440345878,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004855113427481141,
      "loss": 5.8891,
      "mean_token_accuracy": 0.1664668694138527,
      "num_tokens": 24221277.0,
      "step": 12375
    },
    {
      "entropy": 6.130071687698364,
      "epoch": 1.2375168690958165,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000485498740205639,
      "loss": 5.7944,
      "mean_token_accuracy": 0.16343142986297607,
      "num_tokens": 24231272.0,
      "step": 12380
    },
    {
      "entropy": 6.149914741516113,
      "epoch": 1.238016694157045,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048548613236706976,
      "loss": 5.9331,
      "mean_token_accuracy": 0.15955800712108612,
      "num_tokens": 24242073.0,
      "step": 12385
    },
    {
      "entropy": 6.047646522521973,
      "epoch": 1.2385165192182737,
      "grad_norm": 1.25,
      "learning_rate": 0.0004854735192327236,
      "loss": 5.7592,
      "mean_token_accuracy": 0.1684726819396019,
      "num_tokens": 24251928.0,
      "step": 12390
    },
    {
      "entropy": 6.102841424942016,
      "epoch": 1.2390163442795021,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048546090080291783,
      "loss": 5.8454,
      "mean_token_accuracy": 0.16722441762685775,
      "num_tokens": 24261146.0,
      "step": 12395
    },
    {
      "entropy": 6.162425565719604,
      "epoch": 1.2395161693407308,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048544827707797,
      "loss": 5.9986,
      "mean_token_accuracy": 0.1550571322441101,
      "num_tokens": 24271675.0,
      "step": 12400
    },
    {
      "entropy": 6.1142419338226315,
      "epoch": 1.2400159944019593,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048543564805819764,
      "loss": 5.8509,
      "mean_token_accuracy": 0.16124819666147233,
      "num_tokens": 24280795.0,
      "step": 12405
    },
    {
      "entropy": 6.103174352645874,
      "epoch": 1.2405158194631878,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004854230137439185,
      "loss": 5.8712,
      "mean_token_accuracy": 0.16056080162525177,
      "num_tokens": 24291204.0,
      "step": 12410
    },
    {
      "entropy": 6.118503475189209,
      "epoch": 1.2410156445244165,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048541037413545046,
      "loss": 5.8433,
      "mean_token_accuracy": 0.16472341865301132,
      "num_tokens": 24300926.0,
      "step": 12415
    },
    {
      "entropy": 6.094638729095459,
      "epoch": 1.241515469585645,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004853977292331116,
      "loss": 5.7578,
      "mean_token_accuracy": 0.16175394207239152,
      "num_tokens": 24310319.0,
      "step": 12420
    },
    {
      "entropy": 6.111541604995727,
      "epoch": 1.2420152946468737,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004853850790372201,
      "loss": 5.9215,
      "mean_token_accuracy": 0.1596653565764427,
      "num_tokens": 24321671.0,
      "step": 12425
    },
    {
      "entropy": 6.114717721939087,
      "epoch": 1.2425151197081021,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048537242354809403,
      "loss": 5.9639,
      "mean_token_accuracy": 0.15145705938339232,
      "num_tokens": 24330275.0,
      "step": 12430
    },
    {
      "entropy": 6.09766993522644,
      "epoch": 1.2430149447693308,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048535976276605186,
      "loss": 5.9055,
      "mean_token_accuracy": 0.15735120475292205,
      "num_tokens": 24339753.0,
      "step": 12435
    },
    {
      "entropy": 6.111421823501587,
      "epoch": 1.2435147698305593,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004853470966914123,
      "loss": 5.865,
      "mean_token_accuracy": 0.15753898471593858,
      "num_tokens": 24349517.0,
      "step": 12440
    },
    {
      "entropy": 6.0802809715271,
      "epoch": 1.2440145948917878,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004853344253244939,
      "loss": 5.7915,
      "mean_token_accuracy": 0.1708078309893608,
      "num_tokens": 24358036.0,
      "step": 12445
    },
    {
      "entropy": 6.02224555015564,
      "epoch": 1.2445144199530165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048532174866561543,
      "loss": 5.7758,
      "mean_token_accuracy": 0.165852190554142,
      "num_tokens": 24367488.0,
      "step": 12450
    },
    {
      "entropy": 6.215190076828003,
      "epoch": 1.245014245014245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048530906671509597,
      "loss": 6.0514,
      "mean_token_accuracy": 0.14878157824277877,
      "num_tokens": 24377053.0,
      "step": 12455
    },
    {
      "entropy": 6.145028829574585,
      "epoch": 1.2455140700754737,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048529637947325444,
      "loss": 5.9461,
      "mean_token_accuracy": 0.15392180681228637,
      "num_tokens": 24387245.0,
      "step": 12460
    },
    {
      "entropy": 6.154189157485962,
      "epoch": 1.2460138951367021,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004852836869404101,
      "loss": 5.9563,
      "mean_token_accuracy": 0.157228747010231,
      "num_tokens": 24395901.0,
      "step": 12465
    },
    {
      "entropy": 6.115062046051025,
      "epoch": 1.2465137201979308,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004852709891168823,
      "loss": 5.8616,
      "mean_token_accuracy": 0.16213465481996536,
      "num_tokens": 24405148.0,
      "step": 12470
    },
    {
      "entropy": 6.057921600341797,
      "epoch": 1.2470135452591593,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004852582860029906,
      "loss": 5.7967,
      "mean_token_accuracy": 0.1651548072695732,
      "num_tokens": 24415406.0,
      "step": 12475
    },
    {
      "entropy": 6.037519741058349,
      "epoch": 1.2475133703203878,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048524557759905445,
      "loss": 5.7949,
      "mean_token_accuracy": 0.16894869655370712,
      "num_tokens": 24425702.0,
      "step": 12480
    },
    {
      "entropy": 6.106580781936645,
      "epoch": 1.2480131953816165,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004852328639053937,
      "loss": 5.8986,
      "mean_token_accuracy": 0.1577592670917511,
      "num_tokens": 24435702.0,
      "step": 12485
    },
    {
      "entropy": 6.136786317825317,
      "epoch": 1.248513020442845,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004852201449223282,
      "loss": 5.9381,
      "mean_token_accuracy": 0.15173982158303262,
      "num_tokens": 24445545.0,
      "step": 12490
    },
    {
      "entropy": 6.128707361221314,
      "epoch": 1.2490128455040737,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000485207420650178,
      "loss": 5.9711,
      "mean_token_accuracy": 0.16175037622451782,
      "num_tokens": 24455269.0,
      "step": 12495
    },
    {
      "entropy": 6.1359106540679935,
      "epoch": 1.2495126705653021,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004851946910892631,
      "loss": 5.9005,
      "mean_token_accuracy": 0.16655759066343306,
      "num_tokens": 24465563.0,
      "step": 12500
    },
    {
      "entropy": 6.073263120651245,
      "epoch": 1.2500124956265308,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004851819562399039,
      "loss": 5.9601,
      "mean_token_accuracy": 0.1586577132344246,
      "num_tokens": 24475837.0,
      "step": 12505
    },
    {
      "entropy": 6.2093264102935795,
      "epoch": 1.2505123206877593,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004851692161024207,
      "loss": 5.999,
      "mean_token_accuracy": 0.14727206751704217,
      "num_tokens": 24486384.0,
      "step": 12510
    },
    {
      "entropy": 6.177690076828003,
      "epoch": 1.2510121457489878,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048515647067713425,
      "loss": 6.0185,
      "mean_token_accuracy": 0.1562282308936119,
      "num_tokens": 24495840.0,
      "step": 12515
    },
    {
      "entropy": 6.142094516754151,
      "epoch": 1.2515119708102165,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000485143719964365,
      "loss": 5.878,
      "mean_token_accuracy": 0.15573504865169524,
      "num_tokens": 24505849.0,
      "step": 12520
    },
    {
      "entropy": 6.094946336746216,
      "epoch": 1.252011795871445,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004851309639644338,
      "loss": 5.9488,
      "mean_token_accuracy": 0.15808524340391159,
      "num_tokens": 24517440.0,
      "step": 12525
    },
    {
      "entropy": 6.10922384262085,
      "epoch": 1.2525116209326734,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004851182026776616,
      "loss": 5.8259,
      "mean_token_accuracy": 0.16584111750125885,
      "num_tokens": 24527602.0,
      "step": 12530
    },
    {
      "entropy": 6.077944660186768,
      "epoch": 1.2530114459939021,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004851054361043696,
      "loss": 5.7973,
      "mean_token_accuracy": 0.17179999500513077,
      "num_tokens": 24536770.0,
      "step": 12535
    },
    {
      "entropy": 6.074821710586548,
      "epoch": 1.2535112710551308,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048509266424487873,
      "loss": 5.9122,
      "mean_token_accuracy": 0.15946899875998496,
      "num_tokens": 24546230.0,
      "step": 12540
    },
    {
      "entropy": 6.04455976486206,
      "epoch": 1.2540110961163593,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004850798870995106,
      "loss": 5.8829,
      "mean_token_accuracy": 0.15580087900161743,
      "num_tokens": 24555130.0,
      "step": 12545
    },
    {
      "entropy": 6.197753047943115,
      "epoch": 1.2545109211775878,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048506710466858657,
      "loss": 5.9513,
      "mean_token_accuracy": 0.1588471055030823,
      "num_tokens": 24565167.0,
      "step": 12550
    },
    {
      "entropy": 6.130963611602783,
      "epoch": 1.2550107462388165,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048505431695242823,
      "loss": 5.939,
      "mean_token_accuracy": 0.16069782972335817,
      "num_tokens": 24574545.0,
      "step": 12555
    },
    {
      "entropy": 6.120665502548218,
      "epoch": 1.255510571300045,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048504152395135726,
      "loss": 5.9735,
      "mean_token_accuracy": 0.15429106950759888,
      "num_tokens": 24585050.0,
      "step": 12560
    },
    {
      "entropy": 6.220523738861084,
      "epoch": 1.2560103963612734,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048502872566569564,
      "loss": 5.9788,
      "mean_token_accuracy": 0.15256681814789771,
      "num_tokens": 24596334.0,
      "step": 12565
    },
    {
      "entropy": 6.153748893737793,
      "epoch": 1.2565102214225021,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048501592209576535,
      "loss": 5.8786,
      "mean_token_accuracy": 0.1588839754462242,
      "num_tokens": 24605947.0,
      "step": 12570
    },
    {
      "entropy": 6.096932220458984,
      "epoch": 1.2570100464837308,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004850031132418885,
      "loss": 5.7631,
      "mean_token_accuracy": 0.1687811627984047,
      "num_tokens": 24614336.0,
      "step": 12575
    },
    {
      "entropy": 6.087422561645508,
      "epoch": 1.2575098715449593,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048499029910438734,
      "loss": 5.8666,
      "mean_token_accuracy": 0.1568305090069771,
      "num_tokens": 24624906.0,
      "step": 12580
    },
    {
      "entropy": 6.226953172683716,
      "epoch": 1.2580096966061878,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048497747968358424,
      "loss": 5.9177,
      "mean_token_accuracy": 0.1590772822499275,
      "num_tokens": 24634127.0,
      "step": 12585
    },
    {
      "entropy": 6.0634064197540285,
      "epoch": 1.2585095216674165,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004849646549798019,
      "loss": 5.8017,
      "mean_token_accuracy": 0.17101142257452012,
      "num_tokens": 24642451.0,
      "step": 12590
    },
    {
      "entropy": 6.069763040542602,
      "epoch": 1.259009346728645,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048495182499336275,
      "loss": 5.8164,
      "mean_token_accuracy": 0.1646719440817833,
      "num_tokens": 24651229.0,
      "step": 12595
    },
    {
      "entropy": 6.110188436508179,
      "epoch": 1.2595091717898734,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048493898972458985,
      "loss": 6.0023,
      "mean_token_accuracy": 0.1581289738416672,
      "num_tokens": 24660435.0,
      "step": 12600
    },
    {
      "entropy": 6.149118041992187,
      "epoch": 1.2600089968511021,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004849261491738059,
      "loss": 5.9168,
      "mean_token_accuracy": 0.15610892921686173,
      "num_tokens": 24669254.0,
      "step": 12605
    },
    {
      "entropy": 6.1171670913696286,
      "epoch": 1.2605088219123306,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048491330334133405,
      "loss": 5.8181,
      "mean_token_accuracy": 0.16177233457565307,
      "num_tokens": 24677943.0,
      "step": 12610
    },
    {
      "entropy": 6.2122002124786375,
      "epoch": 1.2610086469735593,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048490045222749744,
      "loss": 5.9777,
      "mean_token_accuracy": 0.15704429596662522,
      "num_tokens": 24686678.0,
      "step": 12615
    },
    {
      "entropy": 6.107089853286743,
      "epoch": 1.2615084720347878,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048488759583261955,
      "loss": 5.7952,
      "mean_token_accuracy": 0.16736848801374435,
      "num_tokens": 24696417.0,
      "step": 12620
    },
    {
      "entropy": 6.10536527633667,
      "epoch": 1.2620082970960165,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004848747341570238,
      "loss": 5.9461,
      "mean_token_accuracy": 0.15602667704224588,
      "num_tokens": 24705624.0,
      "step": 12625
    },
    {
      "entropy": 6.095356225967407,
      "epoch": 1.262508122157245,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048486186720103364,
      "loss": 5.8673,
      "mean_token_accuracy": 0.15515717789530753,
      "num_tokens": 24715705.0,
      "step": 12630
    },
    {
      "entropy": 6.150159692764282,
      "epoch": 1.2630079472184734,
      "grad_norm": 1.078125,
      "learning_rate": 0.000484848994964973,
      "loss": 5.8286,
      "mean_token_accuracy": 0.16443679630756378,
      "num_tokens": 24725358.0,
      "step": 12635
    },
    {
      "entropy": 6.093785572052002,
      "epoch": 1.2635077722797021,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048483611744916556,
      "loss": 5.9283,
      "mean_token_accuracy": 0.15895747840404512,
      "num_tokens": 24735457.0,
      "step": 12640
    },
    {
      "entropy": 6.19423246383667,
      "epoch": 1.2640075973409306,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048482323465393545,
      "loss": 5.9513,
      "mean_token_accuracy": 0.15099166035652162,
      "num_tokens": 24745391.0,
      "step": 12645
    },
    {
      "entropy": 6.115315580368042,
      "epoch": 1.2645074224021593,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004848103465796067,
      "loss": 5.8682,
      "mean_token_accuracy": 0.1647888243198395,
      "num_tokens": 24754578.0,
      "step": 12650
    },
    {
      "entropy": 6.057281732559204,
      "epoch": 1.2650072474633878,
      "grad_norm": 1.125,
      "learning_rate": 0.0004847974532265037,
      "loss": 5.9011,
      "mean_token_accuracy": 0.1647224545478821,
      "num_tokens": 24764115.0,
      "step": 12655
    },
    {
      "entropy": 6.111315298080444,
      "epoch": 1.2655070725246165,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004847845545949507,
      "loss": 5.8793,
      "mean_token_accuracy": 0.16161119788885117,
      "num_tokens": 24773247.0,
      "step": 12660
    },
    {
      "entropy": 6.091278123855591,
      "epoch": 1.266006897585845,
      "grad_norm": 1.25,
      "learning_rate": 0.00048477165068527237,
      "loss": 5.8451,
      "mean_token_accuracy": 0.1651742048561573,
      "num_tokens": 24783139.0,
      "step": 12665
    },
    {
      "entropy": 6.135012865066528,
      "epoch": 1.2665067226470734,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048475874149779313,
      "loss": 5.9386,
      "mean_token_accuracy": 0.1567077651619911,
      "num_tokens": 24793591.0,
      "step": 12670
    },
    {
      "entropy": 6.108259296417236,
      "epoch": 1.2670065477083021,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000484745827032838,
      "loss": 5.8425,
      "mean_token_accuracy": 0.16511600762605666,
      "num_tokens": 24803340.0,
      "step": 12675
    },
    {
      "entropy": 6.064791870117188,
      "epoch": 1.2675063727695306,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048473290729073194,
      "loss": 5.7622,
      "mean_token_accuracy": 0.1648744449019432,
      "num_tokens": 24812133.0,
      "step": 12680
    },
    {
      "entropy": 6.072785425186157,
      "epoch": 1.2680061978307593,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004847199822717998,
      "loss": 5.8971,
      "mean_token_accuracy": 0.15774383693933486,
      "num_tokens": 24822443.0,
      "step": 12685
    },
    {
      "entropy": 6.011959457397461,
      "epoch": 1.2685060228919878,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004847070519763668,
      "loss": 5.8191,
      "mean_token_accuracy": 0.1752520576119423,
      "num_tokens": 24832390.0,
      "step": 12690
    },
    {
      "entropy": 6.109362840652466,
      "epoch": 1.2690058479532165,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004846941164047585,
      "loss": 5.8294,
      "mean_token_accuracy": 0.16760096400976182,
      "num_tokens": 24841704.0,
      "step": 12695
    },
    {
      "entropy": 6.227966594696045,
      "epoch": 1.269505673014445,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048468117555730013,
      "loss": 5.9703,
      "mean_token_accuracy": 0.1537506565451622,
      "num_tokens": 24852459.0,
      "step": 12700
    },
    {
      "entropy": 6.138503360748291,
      "epoch": 1.2700054980756734,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004846682294343173,
      "loss": 5.936,
      "mean_token_accuracy": 0.15599740147590638,
      "num_tokens": 24861758.0,
      "step": 12705
    },
    {
      "entropy": 6.2220557689666744,
      "epoch": 1.2705053231369021,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004846552780361358,
      "loss": 5.9671,
      "mean_token_accuracy": 0.15607512444257737,
      "num_tokens": 24871855.0,
      "step": 12710
    },
    {
      "entropy": 6.122804117202759,
      "epoch": 1.2710051481981306,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048464232136308146,
      "loss": 5.8797,
      "mean_token_accuracy": 0.15654121786355973,
      "num_tokens": 24880602.0,
      "step": 12715
    },
    {
      "entropy": 6.1020668029785154,
      "epoch": 1.2715049732593593,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004846293594154802,
      "loss": 5.9088,
      "mean_token_accuracy": 0.16348991990089418,
      "num_tokens": 24891032.0,
      "step": 12720
    },
    {
      "entropy": 6.132189559936523,
      "epoch": 1.2720047983205878,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004846163921936583,
      "loss": 5.8819,
      "mean_token_accuracy": 0.156574048101902,
      "num_tokens": 24900892.0,
      "step": 12725
    },
    {
      "entropy": 6.167391538619995,
      "epoch": 1.2725046233818165,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048460341969794183,
      "loss": 5.9514,
      "mean_token_accuracy": 0.1528493046760559,
      "num_tokens": 24910576.0,
      "step": 12730
    },
    {
      "entropy": 6.0532200813293455,
      "epoch": 1.273004448443045,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048459044192865726,
      "loss": 5.9395,
      "mean_token_accuracy": 0.15876043438911439,
      "num_tokens": 24920861.0,
      "step": 12735
    },
    {
      "entropy": 6.163606357574463,
      "epoch": 1.2735042735042734,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004845774588861311,
      "loss": 5.924,
      "mean_token_accuracy": 0.15188627541065217,
      "num_tokens": 24930539.0,
      "step": 12740
    },
    {
      "entropy": 6.117689561843872,
      "epoch": 1.2740040985655021,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048456447057069004,
      "loss": 5.9485,
      "mean_token_accuracy": 0.15381332933902742,
      "num_tokens": 24939737.0,
      "step": 12745
    },
    {
      "entropy": 6.149089002609253,
      "epoch": 1.2745039236267306,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048455147698266084,
      "loss": 5.9741,
      "mean_token_accuracy": 0.1551267221570015,
      "num_tokens": 24949316.0,
      "step": 12750
    },
    {
      "entropy": 6.126246213912964,
      "epoch": 1.2750037486879593,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004845384781223703,
      "loss": 5.836,
      "mean_token_accuracy": 0.16342331618070602,
      "num_tokens": 24959454.0,
      "step": 12755
    },
    {
      "entropy": 6.102388048171997,
      "epoch": 1.2755035737491878,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048452547399014566,
      "loss": 5.9263,
      "mean_token_accuracy": 0.15901632905006408,
      "num_tokens": 24969091.0,
      "step": 12760
    },
    {
      "entropy": 6.133287811279297,
      "epoch": 1.2760033988104165,
      "grad_norm": 1.21875,
      "learning_rate": 0.000484512464586314,
      "loss": 5.7675,
      "mean_token_accuracy": 0.16850318163633346,
      "num_tokens": 24979239.0,
      "step": 12765
    },
    {
      "entropy": 6.17395510673523,
      "epoch": 1.276503223871645,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048449944991120264,
      "loss": 5.9559,
      "mean_token_accuracy": 0.15632082968950273,
      "num_tokens": 24989274.0,
      "step": 12770
    },
    {
      "entropy": 6.160051250457764,
      "epoch": 1.2770030489328734,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004844864299651391,
      "loss": 5.8426,
      "mean_token_accuracy": 0.16378968060016633,
      "num_tokens": 24997873.0,
      "step": 12775
    },
    {
      "entropy": 6.0142723560333256,
      "epoch": 1.2775028739941021,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048447340474845076,
      "loss": 5.8548,
      "mean_token_accuracy": 0.16476618200540544,
      "num_tokens": 25007829.0,
      "step": 12780
    },
    {
      "entropy": 6.042388296127319,
      "epoch": 1.2780026990553306,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048446037426146555,
      "loss": 5.8246,
      "mean_token_accuracy": 0.1620969831943512,
      "num_tokens": 25016795.0,
      "step": 12785
    },
    {
      "entropy": 6.119889593124389,
      "epoch": 1.278502524116559,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004844473385045112,
      "loss": 5.8999,
      "mean_token_accuracy": 0.15906380116939545,
      "num_tokens": 25026012.0,
      "step": 12790
    },
    {
      "entropy": 6.110365581512451,
      "epoch": 1.2790023491777878,
      "grad_norm": 1.25,
      "learning_rate": 0.0004844342974779157,
      "loss": 5.7977,
      "mean_token_accuracy": 0.1697342351078987,
      "num_tokens": 25035353.0,
      "step": 12795
    },
    {
      "entropy": 6.112552881240845,
      "epoch": 1.2795021742390165,
      "grad_norm": 1.125,
      "learning_rate": 0.00048442125118200707,
      "loss": 5.911,
      "mean_token_accuracy": 0.15943946987390517,
      "num_tokens": 25045301.0,
      "step": 12800
    },
    {
      "entropy": 6.15975866317749,
      "epoch": 1.280001999300245,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004844081996171137,
      "loss": 5.9453,
      "mean_token_accuracy": 0.15830883160233497,
      "num_tokens": 25055576.0,
      "step": 12805
    },
    {
      "entropy": 6.083001804351807,
      "epoch": 1.2805018243614734,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000484395142783564,
      "loss": 5.804,
      "mean_token_accuracy": 0.16590606719255446,
      "num_tokens": 25064077.0,
      "step": 12810
    },
    {
      "entropy": 6.0278356075286865,
      "epoch": 1.2810016494227021,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004843820806816863,
      "loss": 5.7819,
      "mean_token_accuracy": 0.17272310853004455,
      "num_tokens": 25072908.0,
      "step": 12815
    },
    {
      "entropy": 6.076610565185547,
      "epoch": 1.2815014744839306,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004843690133118094,
      "loss": 5.8958,
      "mean_token_accuracy": 0.16215440779924392,
      "num_tokens": 25082919.0,
      "step": 12820
    },
    {
      "entropy": 6.157906770706177,
      "epoch": 1.282001299545159,
      "grad_norm": 1.140625,
      "learning_rate": 0.000484355940674262,
      "loss": 5.9291,
      "mean_token_accuracy": 0.16057779639959335,
      "num_tokens": 25092520.0,
      "step": 12825
    },
    {
      "entropy": 6.119645404815674,
      "epoch": 1.2825011246063878,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048434286276937294,
      "loss": 5.9231,
      "mean_token_accuracy": 0.15133509933948516,
      "num_tokens": 25102638.0,
      "step": 12830
    },
    {
      "entropy": 6.1060038089752195,
      "epoch": 1.2830009496676165,
      "grad_norm": 1.25,
      "learning_rate": 0.0004843297795974713,
      "loss": 5.9308,
      "mean_token_accuracy": 0.15562995821237563,
      "num_tokens": 25112011.0,
      "step": 12835
    },
    {
      "entropy": 6.176648378372192,
      "epoch": 1.283500774728845,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004843166911588864,
      "loss": 5.9728,
      "mean_token_accuracy": 0.15475585907697678,
      "num_tokens": 25122051.0,
      "step": 12840
    },
    {
      "entropy": 6.187306022644043,
      "epoch": 1.2840005997900734,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004843035974539473,
      "loss": 5.9243,
      "mean_token_accuracy": 0.15360394045710563,
      "num_tokens": 25131928.0,
      "step": 12845
    },
    {
      "entropy": 6.005591773986817,
      "epoch": 1.2845004248513021,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048429049848298364,
      "loss": 5.7177,
      "mean_token_accuracy": 0.17313863039016725,
      "num_tokens": 25140204.0,
      "step": 12850
    },
    {
      "entropy": 6.069479322433471,
      "epoch": 1.2850002499125306,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048427739424632485,
      "loss": 5.7999,
      "mean_token_accuracy": 0.16081364154815675,
      "num_tokens": 25149445.0,
      "step": 12855
    },
    {
      "entropy": 6.081435012817383,
      "epoch": 1.285500074973759,
      "grad_norm": 1.125,
      "learning_rate": 0.00048426428474430055,
      "loss": 5.8391,
      "mean_token_accuracy": 0.16168939024209977,
      "num_tokens": 25159311.0,
      "step": 12860
    },
    {
      "entropy": 6.012763547897339,
      "epoch": 1.2859999000349878,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048425116997724083,
      "loss": 5.7792,
      "mean_token_accuracy": 0.17219433784484864,
      "num_tokens": 25168880.0,
      "step": 12865
    },
    {
      "entropy": 6.070995569229126,
      "epoch": 1.2864997250962165,
      "grad_norm": 1.25,
      "learning_rate": 0.00048423804994547555,
      "loss": 5.8319,
      "mean_token_accuracy": 0.16453014612197875,
      "num_tokens": 25177812.0,
      "step": 12870
    },
    {
      "entropy": 5.974812030792236,
      "epoch": 1.286999550157445,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048422492464933464,
      "loss": 5.7748,
      "mean_token_accuracy": 0.16490316689014434,
      "num_tokens": 25187831.0,
      "step": 12875
    },
    {
      "entropy": 6.039895820617676,
      "epoch": 1.2874993752186734,
      "grad_norm": 1.125,
      "learning_rate": 0.00048421179408914857,
      "loss": 5.7785,
      "mean_token_accuracy": 0.16835372596979142,
      "num_tokens": 25198164.0,
      "step": 12880
    },
    {
      "entropy": 6.157233953475952,
      "epoch": 1.2879992002799021,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048419865826524757,
      "loss": 5.9211,
      "mean_token_accuracy": 0.1602065756917,
      "num_tokens": 25208740.0,
      "step": 12885
    },
    {
      "entropy": 6.0837815284729,
      "epoch": 1.2884990253411306,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048418551717796217,
      "loss": 5.8224,
      "mean_token_accuracy": 0.1604737088084221,
      "num_tokens": 25218475.0,
      "step": 12890
    },
    {
      "entropy": 6.12854037284851,
      "epoch": 1.288998850402359,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048417237082762295,
      "loss": 5.9523,
      "mean_token_accuracy": 0.1595451518893242,
      "num_tokens": 25227943.0,
      "step": 12895
    },
    {
      "entropy": 5.9635148525238035,
      "epoch": 1.2894986754635878,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048415921921456074,
      "loss": 5.7443,
      "mean_token_accuracy": 0.17236646115779877,
      "num_tokens": 25237131.0,
      "step": 12900
    },
    {
      "entropy": 6.137505054473877,
      "epoch": 1.2899985005248162,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048414606233910634,
      "loss": 5.831,
      "mean_token_accuracy": 0.1645911455154419,
      "num_tokens": 25245983.0,
      "step": 12905
    },
    {
      "entropy": 6.02618145942688,
      "epoch": 1.290498325586045,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004841329002015909,
      "loss": 5.8397,
      "mean_token_accuracy": 0.16070908457040786,
      "num_tokens": 25255155.0,
      "step": 12910
    },
    {
      "entropy": 6.019999933242798,
      "epoch": 1.2909981506472734,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048411973280234553,
      "loss": 5.8449,
      "mean_token_accuracy": 0.16239690631628037,
      "num_tokens": 25265207.0,
      "step": 12915
    },
    {
      "entropy": 6.063660478591919,
      "epoch": 1.291497975708502,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048410656014170135,
      "loss": 5.8027,
      "mean_token_accuracy": 0.17182541638612747,
      "num_tokens": 25273995.0,
      "step": 12920
    },
    {
      "entropy": 6.052890253067017,
      "epoch": 1.2919978007697306,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048409338221999,
      "loss": 5.7918,
      "mean_token_accuracy": 0.1665608197450638,
      "num_tokens": 25284334.0,
      "step": 12925
    },
    {
      "entropy": 6.051181125640869,
      "epoch": 1.292497625830959,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048408019903754297,
      "loss": 5.8148,
      "mean_token_accuracy": 0.1697144940495491,
      "num_tokens": 25294360.0,
      "step": 12930
    },
    {
      "entropy": 6.045327138900757,
      "epoch": 1.2929974508921878,
      "grad_norm": 1.203125,
      "learning_rate": 0.000484067010594692,
      "loss": 5.8554,
      "mean_token_accuracy": 0.1658591017127037,
      "num_tokens": 25302869.0,
      "step": 12935
    },
    {
      "entropy": 6.030210351943969,
      "epoch": 1.2934972759534162,
      "grad_norm": 3.1875,
      "learning_rate": 0.00048405381689176877,
      "loss": 5.856,
      "mean_token_accuracy": 0.16682717055082322,
      "num_tokens": 25313825.0,
      "step": 12940
    },
    {
      "entropy": 6.120831871032715,
      "epoch": 1.293997101014645,
      "grad_norm": 1.25,
      "learning_rate": 0.00048404061792910527,
      "loss": 5.8835,
      "mean_token_accuracy": 0.16436718106269838,
      "num_tokens": 25323212.0,
      "step": 12945
    },
    {
      "entropy": 6.213246822357178,
      "epoch": 1.2944969260758734,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048402741370703363,
      "loss": 5.9789,
      "mean_token_accuracy": 0.16018767058849334,
      "num_tokens": 25333037.0,
      "step": 12950
    },
    {
      "entropy": 6.1134350299835205,
      "epoch": 1.294996751137102,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004840142042258861,
      "loss": 5.8821,
      "mean_token_accuracy": 0.16458263993263245,
      "num_tokens": 25342810.0,
      "step": 12955
    },
    {
      "entropy": 6.101698207855224,
      "epoch": 1.2954965761983306,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000484000989485995,
      "loss": 5.9643,
      "mean_token_accuracy": 0.15440210551023484,
      "num_tokens": 25353119.0,
      "step": 12960
    },
    {
      "entropy": 6.157714223861694,
      "epoch": 1.295996401259559,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048398776948769265,
      "loss": 5.8442,
      "mean_token_accuracy": 0.1670530080795288,
      "num_tokens": 25363450.0,
      "step": 12965
    },
    {
      "entropy": 6.073920345306396,
      "epoch": 1.2964962263207878,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004839745442313119,
      "loss": 5.8225,
      "mean_token_accuracy": 0.1631324976682663,
      "num_tokens": 25372830.0,
      "step": 12970
    },
    {
      "entropy": 6.062277793884277,
      "epoch": 1.2969960513820162,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048396131371718535,
      "loss": 5.8816,
      "mean_token_accuracy": 0.1564596563577652,
      "num_tokens": 25382102.0,
      "step": 12975
    },
    {
      "entropy": 6.093380784988403,
      "epoch": 1.297495876443245,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048394807794564586,
      "loss": 5.9336,
      "mean_token_accuracy": 0.16076814979314805,
      "num_tokens": 25390817.0,
      "step": 12980
    },
    {
      "entropy": 6.10476655960083,
      "epoch": 1.2979957015044734,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004839348369170266,
      "loss": 5.8938,
      "mean_token_accuracy": 0.16325128227472305,
      "num_tokens": 25399958.0,
      "step": 12985
    },
    {
      "entropy": 6.093589353561401,
      "epoch": 1.298495526565702,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048392159063166055,
      "loss": 5.8791,
      "mean_token_accuracy": 0.15971508175134658,
      "num_tokens": 25410870.0,
      "step": 12990
    },
    {
      "entropy": 6.166405296325683,
      "epoch": 1.2989953516269306,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004839083390898811,
      "loss": 5.9318,
      "mean_token_accuracy": 0.15347964465618133,
      "num_tokens": 25420906.0,
      "step": 12995
    },
    {
      "entropy": 6.128982210159302,
      "epoch": 1.299495176688159,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004838950822920215,
      "loss": 5.8513,
      "mean_token_accuracy": 0.1577014923095703,
      "num_tokens": 25430046.0,
      "step": 13000
    },
    {
      "entropy": 6.021689414978027,
      "epoch": 1.2999950017493878,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004838818202384155,
      "loss": 5.8313,
      "mean_token_accuracy": 0.16804277896881104,
      "num_tokens": 25438785.0,
      "step": 13005
    },
    {
      "entropy": 6.063944578170776,
      "epoch": 1.3004948268106162,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048386855292939664,
      "loss": 5.9112,
      "mean_token_accuracy": 0.16108431220054625,
      "num_tokens": 25448815.0,
      "step": 13010
    },
    {
      "entropy": 6.1099144458770756,
      "epoch": 1.300994651871845,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004838552803652986,
      "loss": 5.8425,
      "mean_token_accuracy": 0.160015070438385,
      "num_tokens": 25458528.0,
      "step": 13015
    },
    {
      "entropy": 6.052693033218384,
      "epoch": 1.3014944769330734,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048384200254645557,
      "loss": 5.7881,
      "mean_token_accuracy": 0.15969149768352509,
      "num_tokens": 25467678.0,
      "step": 13020
    },
    {
      "entropy": 5.973464775085449,
      "epoch": 1.301994301994302,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048382871947320145,
      "loss": 5.7811,
      "mean_token_accuracy": 0.17770419120788575,
      "num_tokens": 25478601.0,
      "step": 13025
    },
    {
      "entropy": 6.088258552551269,
      "epoch": 1.3024941270555306,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004838154311458705,
      "loss": 5.8523,
      "mean_token_accuracy": 0.15584911853075029,
      "num_tokens": 25489439.0,
      "step": 13030
    },
    {
      "entropy": 6.004664707183838,
      "epoch": 1.302993952116759,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048380213756479704,
      "loss": 5.8544,
      "mean_token_accuracy": 0.1620506152510643,
      "num_tokens": 25499976.0,
      "step": 13035
    },
    {
      "entropy": 6.098305654525757,
      "epoch": 1.3034937771779878,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048378883873031553,
      "loss": 5.9979,
      "mean_token_accuracy": 0.15408513620495795,
      "num_tokens": 25509737.0,
      "step": 13040
    },
    {
      "entropy": 6.167645645141602,
      "epoch": 1.3039936022392162,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004837755346427605,
      "loss": 5.9158,
      "mean_token_accuracy": 0.15471998751163482,
      "num_tokens": 25519130.0,
      "step": 13045
    },
    {
      "entropy": 6.14183931350708,
      "epoch": 1.3044934273004447,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048376222530246684,
      "loss": 5.9093,
      "mean_token_accuracy": 0.1622430056333542,
      "num_tokens": 25528625.0,
      "step": 13050
    },
    {
      "entropy": 6.098031902313233,
      "epoch": 1.3049932523616734,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048374891070976925,
      "loss": 5.9313,
      "mean_token_accuracy": 0.14882470890879632,
      "num_tokens": 25537912.0,
      "step": 13055
    },
    {
      "entropy": 6.13481502532959,
      "epoch": 1.305493077422902,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004837355908650028,
      "loss": 5.9127,
      "mean_token_accuracy": 0.15988366752862931,
      "num_tokens": 25548622.0,
      "step": 13060
    },
    {
      "entropy": 5.997596549987793,
      "epoch": 1.3059929024841306,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004837222657685026,
      "loss": 5.6984,
      "mean_token_accuracy": 0.1751744493842125,
      "num_tokens": 25558322.0,
      "step": 13065
    },
    {
      "entropy": 6.11558518409729,
      "epoch": 1.306492727545359,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048370893542060383,
      "loss": 5.8952,
      "mean_token_accuracy": 0.1556934669613838,
      "num_tokens": 25568586.0,
      "step": 13070
    },
    {
      "entropy": 6.139873600006103,
      "epoch": 1.3069925526065878,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004836955998216421,
      "loss": 5.9153,
      "mean_token_accuracy": 0.15343356803059577,
      "num_tokens": 25577405.0,
      "step": 13075
    },
    {
      "entropy": 5.998936605453491,
      "epoch": 1.3074923776678162,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048368225897195264,
      "loss": 5.6938,
      "mean_token_accuracy": 0.17574868500232696,
      "num_tokens": 25587719.0,
      "step": 13080
    },
    {
      "entropy": 6.053676414489746,
      "epoch": 1.3079922027290447,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048366891287187133,
      "loss": 5.8758,
      "mean_token_accuracy": 0.15833213403820992,
      "num_tokens": 25597334.0,
      "step": 13085
    },
    {
      "entropy": 6.207055902481079,
      "epoch": 1.3084920277902734,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004836555615217337,
      "loss": 6.0352,
      "mean_token_accuracy": 0.15187447071075438,
      "num_tokens": 25606684.0,
      "step": 13090
    },
    {
      "entropy": 6.028065538406372,
      "epoch": 1.308991852851502,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004836422049218761,
      "loss": 5.7837,
      "mean_token_accuracy": 0.17651990205049514,
      "num_tokens": 25615808.0,
      "step": 13095
    },
    {
      "entropy": 5.985558605194091,
      "epoch": 1.3094916779127306,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048362884307263416,
      "loss": 5.776,
      "mean_token_accuracy": 0.1688755065202713,
      "num_tokens": 25626444.0,
      "step": 13100
    },
    {
      "entropy": 6.1132653713226315,
      "epoch": 1.309991502973959,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048361547597434426,
      "loss": 5.8096,
      "mean_token_accuracy": 0.16361831426620482,
      "num_tokens": 25635730.0,
      "step": 13105
    },
    {
      "entropy": 6.142320966720581,
      "epoch": 1.3104913280351878,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048360210362734257,
      "loss": 5.8649,
      "mean_token_accuracy": 0.16621324121952058,
      "num_tokens": 25645938.0,
      "step": 13110
    },
    {
      "entropy": 6.086990785598755,
      "epoch": 1.3109911530964162,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004835887260319657,
      "loss": 5.9531,
      "mean_token_accuracy": 0.1555735871195793,
      "num_tokens": 25656135.0,
      "step": 13115
    },
    {
      "entropy": 6.11965765953064,
      "epoch": 1.3114909781576447,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004835753431885502,
      "loss": 5.9702,
      "mean_token_accuracy": 0.15565354377031326,
      "num_tokens": 25666134.0,
      "step": 13120
    },
    {
      "entropy": 6.166890954971313,
      "epoch": 1.3119908032188734,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048356195509743265,
      "loss": 6.0221,
      "mean_token_accuracy": 0.15108728185296058,
      "num_tokens": 25675716.0,
      "step": 13125
    },
    {
      "entropy": 6.131795883178711,
      "epoch": 1.312490628280102,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048354856175895,
      "loss": 5.8322,
      "mean_token_accuracy": 0.16076042801141738,
      "num_tokens": 25685814.0,
      "step": 13130
    },
    {
      "entropy": 6.115576505661011,
      "epoch": 1.3129904533413306,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004835351631734392,
      "loss": 5.8272,
      "mean_token_accuracy": 0.1630224421620369,
      "num_tokens": 25694995.0,
      "step": 13135
    },
    {
      "entropy": 6.13975157737732,
      "epoch": 1.313490278402559,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004835217593412374,
      "loss": 5.8955,
      "mean_token_accuracy": 0.1588338829576969,
      "num_tokens": 25705041.0,
      "step": 13140
    },
    {
      "entropy": 6.1766139507293705,
      "epoch": 1.3139901034637878,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004835083502626818,
      "loss": 5.9026,
      "mean_token_accuracy": 0.15458351820707322,
      "num_tokens": 25715830.0,
      "step": 13145
    },
    {
      "entropy": 6.04741849899292,
      "epoch": 1.3144899285250162,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048349493593810973,
      "loss": 5.8817,
      "mean_token_accuracy": 0.15484704226255416,
      "num_tokens": 25725357.0,
      "step": 13150
    },
    {
      "entropy": 6.110965442657471,
      "epoch": 1.3149897535862447,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004834815163678587,
      "loss": 5.9029,
      "mean_token_accuracy": 0.16191788911819457,
      "num_tokens": 25735720.0,
      "step": 13155
    },
    {
      "entropy": 6.185384845733642,
      "epoch": 1.3154895786474734,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004834680915522664,
      "loss": 5.8866,
      "mean_token_accuracy": 0.1561019390821457,
      "num_tokens": 25745728.0,
      "step": 13160
    },
    {
      "entropy": 6.148108196258545,
      "epoch": 1.3159894037087019,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004834546614916705,
      "loss": 5.8924,
      "mean_token_accuracy": 0.15815246850252151,
      "num_tokens": 25755917.0,
      "step": 13165
    },
    {
      "entropy": 6.099628353118897,
      "epoch": 1.3164892287699306,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048344122618640905,
      "loss": 5.9691,
      "mean_token_accuracy": 0.15749092102050782,
      "num_tokens": 25766195.0,
      "step": 13170
    },
    {
      "entropy": 6.114178037643432,
      "epoch": 1.316989053831159,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004834277856368199,
      "loss": 5.789,
      "mean_token_accuracy": 0.16717372238636016,
      "num_tokens": 25774134.0,
      "step": 13175
    },
    {
      "entropy": 5.991566514968872,
      "epoch": 1.3174888788923877,
      "grad_norm": 1.125,
      "learning_rate": 0.00048341433984324127,
      "loss": 5.8555,
      "mean_token_accuracy": 0.1657226175069809,
      "num_tokens": 25783377.0,
      "step": 13180
    },
    {
      "entropy": 5.966299962997437,
      "epoch": 1.3179887039536162,
      "grad_norm": 1.0,
      "learning_rate": 0.00048340088880601163,
      "loss": 5.7039,
      "mean_token_accuracy": 0.16939700841903688,
      "num_tokens": 25793562.0,
      "step": 13185
    },
    {
      "entropy": 6.110503101348877,
      "epoch": 1.3184885290148447,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004833874325254691,
      "loss": 5.9175,
      "mean_token_accuracy": 0.1645504742860794,
      "num_tokens": 25803299.0,
      "step": 13190
    },
    {
      "entropy": 6.161547517776489,
      "epoch": 1.3189883540760734,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048337397100195245,
      "loss": 5.9406,
      "mean_token_accuracy": 0.1575458213686943,
      "num_tokens": 25813053.0,
      "step": 13195
    },
    {
      "entropy": 6.211876440048218,
      "epoch": 1.3194881791373019,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004833605042358003,
      "loss": 5.924,
      "mean_token_accuracy": 0.15908022075891495,
      "num_tokens": 25822311.0,
      "step": 13200
    },
    {
      "entropy": 6.087135171890258,
      "epoch": 1.3199880041985306,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004833470322273515,
      "loss": 5.9553,
      "mean_token_accuracy": 0.14934322834014893,
      "num_tokens": 25832473.0,
      "step": 13205
    },
    {
      "entropy": 6.121328115463257,
      "epoch": 1.320487829259759,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048333355497694506,
      "loss": 5.8354,
      "mean_token_accuracy": 0.16701277047395707,
      "num_tokens": 25840919.0,
      "step": 13210
    },
    {
      "entropy": 6.088587236404419,
      "epoch": 1.3209876543209877,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004833200724849199,
      "loss": 5.8525,
      "mean_token_accuracy": 0.16251428723335265,
      "num_tokens": 25850517.0,
      "step": 13215
    },
    {
      "entropy": 6.178922128677368,
      "epoch": 1.3214874793822162,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048330658475161526,
      "loss": 6.0019,
      "mean_token_accuracy": 0.14719906747341155,
      "num_tokens": 25861580.0,
      "step": 13220
    },
    {
      "entropy": 6.155812740325928,
      "epoch": 1.3219873044434447,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004832930917773706,
      "loss": 5.9245,
      "mean_token_accuracy": 0.15770259797573088,
      "num_tokens": 25872383.0,
      "step": 13225
    },
    {
      "entropy": 6.114022922515869,
      "epoch": 1.3224871295046734,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048327959356252544,
      "loss": 5.9273,
      "mean_token_accuracy": 0.15749266967177392,
      "num_tokens": 25883016.0,
      "step": 13230
    },
    {
      "entropy": 6.1147233009338375,
      "epoch": 1.3229869545659019,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004832660901074192,
      "loss": 5.8827,
      "mean_token_accuracy": 0.16089068502187728,
      "num_tokens": 25892810.0,
      "step": 13235
    },
    {
      "entropy": 6.092725467681885,
      "epoch": 1.3234867796271306,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048325258141239174,
      "loss": 5.8624,
      "mean_token_accuracy": 0.16152692288160325,
      "num_tokens": 25902548.0,
      "step": 13240
    },
    {
      "entropy": 6.071629858016967,
      "epoch": 1.323986604688359,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004832390674777829,
      "loss": 5.8954,
      "mean_token_accuracy": 0.16885911524295807,
      "num_tokens": 25911572.0,
      "step": 13245
    },
    {
      "entropy": 6.037160968780517,
      "epoch": 1.3244864297495877,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048322554830393274,
      "loss": 5.7698,
      "mean_token_accuracy": 0.16488249152898787,
      "num_tokens": 25921226.0,
      "step": 13250
    },
    {
      "entropy": 6.080060958862305,
      "epoch": 1.3249862548108162,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004832120238911813,
      "loss": 5.8709,
      "mean_token_accuracy": 0.15589988455176354,
      "num_tokens": 25930920.0,
      "step": 13255
    },
    {
      "entropy": 6.1363664150238035,
      "epoch": 1.3254860798720447,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000483198494239869,
      "loss": 5.826,
      "mean_token_accuracy": 0.1663771763443947,
      "num_tokens": 25941042.0,
      "step": 13260
    },
    {
      "entropy": 6.189861631393432,
      "epoch": 1.3259859049332734,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048318495935033615,
      "loss": 5.9949,
      "mean_token_accuracy": 0.14939590245485307,
      "num_tokens": 25950696.0,
      "step": 13265
    },
    {
      "entropy": 6.094739627838135,
      "epoch": 1.3264857299945019,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048317141922292326,
      "loss": 5.8724,
      "mean_token_accuracy": 0.15801718533039094,
      "num_tokens": 25960034.0,
      "step": 13270
    },
    {
      "entropy": 6.060680150985718,
      "epoch": 1.3269855550557306,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004831578738579711,
      "loss": 5.8806,
      "mean_token_accuracy": 0.16799089610576629,
      "num_tokens": 25969264.0,
      "step": 13275
    },
    {
      "entropy": 6.141150760650635,
      "epoch": 1.327485380116959,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004831443232558203,
      "loss": 5.9965,
      "mean_token_accuracy": 0.15829773545265197,
      "num_tokens": 25979046.0,
      "step": 13280
    },
    {
      "entropy": 6.077375030517578,
      "epoch": 1.3279852051781877,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048313076741681186,
      "loss": 5.855,
      "mean_token_accuracy": 0.15565813779830934,
      "num_tokens": 25988797.0,
      "step": 13285
    },
    {
      "entropy": 6.057637357711792,
      "epoch": 1.3284850302394162,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048311720634128697,
      "loss": 5.767,
      "mean_token_accuracy": 0.1652786135673523,
      "num_tokens": 25997586.0,
      "step": 13290
    },
    {
      "entropy": 6.066136741638184,
      "epoch": 1.3289848553006447,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004831036400295867,
      "loss": 5.9241,
      "mean_token_accuracy": 0.1584106996655464,
      "num_tokens": 26007369.0,
      "step": 13295
    },
    {
      "entropy": 6.121802091598511,
      "epoch": 1.3294846803618734,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048309006848205237,
      "loss": 5.878,
      "mean_token_accuracy": 0.15919523686170578,
      "num_tokens": 26017063.0,
      "step": 13300
    },
    {
      "entropy": 6.1240394115448,
      "epoch": 1.3299845054231019,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004830764916990255,
      "loss": 5.8191,
      "mean_token_accuracy": 0.16593037843704223,
      "num_tokens": 26026555.0,
      "step": 13305
    },
    {
      "entropy": 6.139000034332275,
      "epoch": 1.3304843304843303,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004830629096808476,
      "loss": 5.9037,
      "mean_token_accuracy": 0.16369318068027497,
      "num_tokens": 26036698.0,
      "step": 13310
    },
    {
      "entropy": 6.05262598991394,
      "epoch": 1.330984155545559,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004830493224278605,
      "loss": 5.8639,
      "mean_token_accuracy": 0.15910121500492097,
      "num_tokens": 26046462.0,
      "step": 13315
    },
    {
      "entropy": 6.082375001907349,
      "epoch": 1.3314839806067877,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004830357299404059,
      "loss": 5.8968,
      "mean_token_accuracy": 0.1666272446513176,
      "num_tokens": 26056180.0,
      "step": 13320
    },
    {
      "entropy": 6.1733277320861815,
      "epoch": 1.3319838056680162,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004830221322188259,
      "loss": 5.8801,
      "mean_token_accuracy": 0.15627087652683258,
      "num_tokens": 26065621.0,
      "step": 13325
    },
    {
      "entropy": 6.145702171325683,
      "epoch": 1.3324836307292447,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004830085292634626,
      "loss": 5.8837,
      "mean_token_accuracy": 0.1610833778977394,
      "num_tokens": 26074640.0,
      "step": 13330
    },
    {
      "entropy": 6.023242568969726,
      "epoch": 1.3329834557904734,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048299492107465816,
      "loss": 5.8565,
      "mean_token_accuracy": 0.15822488218545913,
      "num_tokens": 26084221.0,
      "step": 13335
    },
    {
      "entropy": 6.0651928901672365,
      "epoch": 1.3334832808517019,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004829813076527551,
      "loss": 5.8908,
      "mean_token_accuracy": 0.1641850382089615,
      "num_tokens": 26095065.0,
      "step": 13340
    },
    {
      "entropy": 6.0986237049102785,
      "epoch": 1.3339831059129303,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048296768899809577,
      "loss": 5.8177,
      "mean_token_accuracy": 0.17071849703788758,
      "num_tokens": 26104978.0,
      "step": 13345
    },
    {
      "entropy": 6.015296745300293,
      "epoch": 1.334482930974159,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048295406511102303,
      "loss": 5.7986,
      "mean_token_accuracy": 0.16540125459432603,
      "num_tokens": 26115023.0,
      "step": 13350
    },
    {
      "entropy": 6.132810163497925,
      "epoch": 1.3349827560353877,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004829404359918795,
      "loss": 5.9304,
      "mean_token_accuracy": 0.160078065097332,
      "num_tokens": 26124533.0,
      "step": 13355
    },
    {
      "entropy": 6.112121057510376,
      "epoch": 1.3354825810966162,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000482926801641008,
      "loss": 5.8364,
      "mean_token_accuracy": 0.16775282770395278,
      "num_tokens": 26134466.0,
      "step": 13360
    },
    {
      "entropy": 6.171880531311035,
      "epoch": 1.3359824061578447,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048291316205875174,
      "loss": 5.8994,
      "mean_token_accuracy": 0.1598622605204582,
      "num_tokens": 26145710.0,
      "step": 13365
    },
    {
      "entropy": 6.11035647392273,
      "epoch": 1.3364822312190734,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048289951724545384,
      "loss": 5.8368,
      "mean_token_accuracy": 0.16917218267917633,
      "num_tokens": 26154393.0,
      "step": 13370
    },
    {
      "entropy": 5.989476203918457,
      "epoch": 1.3369820562803019,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048288586720145757,
      "loss": 5.7455,
      "mean_token_accuracy": 0.17496614158153534,
      "num_tokens": 26164230.0,
      "step": 13375
    },
    {
      "entropy": 6.0642862796783445,
      "epoch": 1.3374818813415303,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004828722119271064,
      "loss": 5.8739,
      "mean_token_accuracy": 0.16245689243078232,
      "num_tokens": 26174224.0,
      "step": 13380
    },
    {
      "entropy": 6.098923826217652,
      "epoch": 1.337981706402759,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048285855142274375,
      "loss": 5.8484,
      "mean_token_accuracy": 0.17275321781635283,
      "num_tokens": 26183447.0,
      "step": 13385
    },
    {
      "entropy": 6.202675676345825,
      "epoch": 1.3384815314639877,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004828448856887135,
      "loss": 5.9611,
      "mean_token_accuracy": 0.15889402180910112,
      "num_tokens": 26193479.0,
      "step": 13390
    },
    {
      "entropy": 6.182955884933472,
      "epoch": 1.3389813565252162,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004828312147253595,
      "loss": 5.9224,
      "mean_token_accuracy": 0.15838345885276794,
      "num_tokens": 26202493.0,
      "step": 13395
    },
    {
      "entropy": 6.120592403411865,
      "epoch": 1.3394811815864447,
      "grad_norm": 1.125,
      "learning_rate": 0.0004828175385330255,
      "loss": 5.9109,
      "mean_token_accuracy": 0.15726855099201204,
      "num_tokens": 26212133.0,
      "step": 13400
    },
    {
      "entropy": 6.0121198177337645,
      "epoch": 1.3399810066476734,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004828038571120558,
      "loss": 5.895,
      "mean_token_accuracy": 0.1660205155611038,
      "num_tokens": 26222905.0,
      "step": 13405
    },
    {
      "entropy": 6.1286204814910885,
      "epoch": 1.3404808317089019,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048279017046279446,
      "loss": 5.8954,
      "mean_token_accuracy": 0.15772564262151717,
      "num_tokens": 26234273.0,
      "step": 13410
    },
    {
      "entropy": 6.111844444274903,
      "epoch": 1.3409806567701303,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048277647858558596,
      "loss": 5.8159,
      "mean_token_accuracy": 0.1688517689704895,
      "num_tokens": 26243965.0,
      "step": 13415
    },
    {
      "entropy": 6.066837120056152,
      "epoch": 1.341480481831359,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048276278148077465,
      "loss": 5.872,
      "mean_token_accuracy": 0.16296130269765854,
      "num_tokens": 26253401.0,
      "step": 13420
    },
    {
      "entropy": 6.026414918899536,
      "epoch": 1.3419803068925875,
      "grad_norm": 1.25,
      "learning_rate": 0.00048274907914870526,
      "loss": 5.8171,
      "mean_token_accuracy": 0.16280947178602218,
      "num_tokens": 26262481.0,
      "step": 13425
    },
    {
      "entropy": 6.14510178565979,
      "epoch": 1.3424801319538162,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048273537158972246,
      "loss": 5.9681,
      "mean_token_accuracy": 0.1572621986269951,
      "num_tokens": 26271153.0,
      "step": 13430
    },
    {
      "entropy": 6.157395362854004,
      "epoch": 1.3429799570150447,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048272165880417124,
      "loss": 6.007,
      "mean_token_accuracy": 0.151905670017004,
      "num_tokens": 26281047.0,
      "step": 13435
    },
    {
      "entropy": 6.1846349239349365,
      "epoch": 1.3434797820762734,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048270794079239643,
      "loss": 5.9232,
      "mean_token_accuracy": 0.15429447665810586,
      "num_tokens": 26291029.0,
      "step": 13440
    },
    {
      "entropy": 6.081436109542847,
      "epoch": 1.3439796071375019,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048269421755474336,
      "loss": 5.8665,
      "mean_token_accuracy": 0.16000353321433067,
      "num_tokens": 26300637.0,
      "step": 13445
    },
    {
      "entropy": 6.072849369049072,
      "epoch": 1.3444794321987303,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004826804890915572,
      "loss": 5.9247,
      "mean_token_accuracy": 0.1556345596909523,
      "num_tokens": 26309867.0,
      "step": 13450
    },
    {
      "entropy": 6.135583400726318,
      "epoch": 1.344979257259959,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004826667554031834,
      "loss": 5.8199,
      "mean_token_accuracy": 0.1681687280535698,
      "num_tokens": 26318691.0,
      "step": 13455
    },
    {
      "entropy": 6.180652713775634,
      "epoch": 1.3454790823211875,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004826530164899674,
      "loss": 5.9427,
      "mean_token_accuracy": 0.15495157092809678,
      "num_tokens": 26327294.0,
      "step": 13460
    },
    {
      "entropy": 6.084870338439941,
      "epoch": 1.3459789073824162,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048263927235225505,
      "loss": 5.9262,
      "mean_token_accuracy": 0.15840095430612564,
      "num_tokens": 26339965.0,
      "step": 13465
    },
    {
      "entropy": 6.16666989326477,
      "epoch": 1.3464787324436447,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004826255229903919,
      "loss": 5.8408,
      "mean_token_accuracy": 0.1650846540927887,
      "num_tokens": 26350447.0,
      "step": 13470
    },
    {
      "entropy": 6.178971767425537,
      "epoch": 1.3469785575048734,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048261176840472416,
      "loss": 5.9079,
      "mean_token_accuracy": 0.1546420693397522,
      "num_tokens": 26359641.0,
      "step": 13475
    },
    {
      "entropy": 6.0952620029449465,
      "epoch": 1.3474783825661019,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048259800859559764,
      "loss": 6.0085,
      "mean_token_accuracy": 0.15207088813185693,
      "num_tokens": 26369161.0,
      "step": 13480
    },
    {
      "entropy": 6.098331499099731,
      "epoch": 1.3479782076273303,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004825842435633587,
      "loss": 5.9048,
      "mean_token_accuracy": 0.16599240005016327,
      "num_tokens": 26379592.0,
      "step": 13485
    },
    {
      "entropy": 6.109163522720337,
      "epoch": 1.348478032688559,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048257047330835354,
      "loss": 5.8455,
      "mean_token_accuracy": 0.1652609810233116,
      "num_tokens": 26389164.0,
      "step": 13490
    },
    {
      "entropy": 6.121085834503174,
      "epoch": 1.3489778577497875,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048255669783092876,
      "loss": 5.9274,
      "mean_token_accuracy": 0.15638701617717743,
      "num_tokens": 26399574.0,
      "step": 13495
    },
    {
      "entropy": 6.122523498535156,
      "epoch": 1.3494776828110162,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048254291713143083,
      "loss": 5.8585,
      "mean_token_accuracy": 0.1649501845240593,
      "num_tokens": 26410240.0,
      "step": 13500
    },
    {
      "entropy": 6.129127550125122,
      "epoch": 1.3499775078722447,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004825291312102066,
      "loss": 5.8089,
      "mean_token_accuracy": 0.1612767443060875,
      "num_tokens": 26419763.0,
      "step": 13505
    },
    {
      "entropy": 6.07573447227478,
      "epoch": 1.3504773329334734,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004825153400676027,
      "loss": 5.9219,
      "mean_token_accuracy": 0.15108790770173072,
      "num_tokens": 26429370.0,
      "step": 13510
    },
    {
      "entropy": 6.102692413330078,
      "epoch": 1.3509771579947019,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004825015437039663,
      "loss": 5.8416,
      "mean_token_accuracy": 0.1667666658759117,
      "num_tokens": 26438425.0,
      "step": 13515
    },
    {
      "entropy": 6.114971494674682,
      "epoch": 1.3514769830559303,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004824877421196444,
      "loss": 5.9002,
      "mean_token_accuracy": 0.15935111194849014,
      "num_tokens": 26448039.0,
      "step": 13520
    },
    {
      "entropy": 6.059904003143311,
      "epoch": 1.351976808117159,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004824739353149844,
      "loss": 5.8403,
      "mean_token_accuracy": 0.1667902022600174,
      "num_tokens": 26457682.0,
      "step": 13525
    },
    {
      "entropy": 5.9882104873657225,
      "epoch": 1.3524766331783875,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048246012329033355,
      "loss": 5.7706,
      "mean_token_accuracy": 0.16449393182992936,
      "num_tokens": 26466292.0,
      "step": 13530
    },
    {
      "entropy": 6.148700618743897,
      "epoch": 1.3529764582396162,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048244630604603935,
      "loss": 6.0376,
      "mean_token_accuracy": 0.1509573884308338,
      "num_tokens": 26476880.0,
      "step": 13535
    },
    {
      "entropy": 6.183209657669067,
      "epoch": 1.3534762833008447,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048243248358244945,
      "loss": 5.8992,
      "mean_token_accuracy": 0.1588621497154236,
      "num_tokens": 26486132.0,
      "step": 13540
    },
    {
      "entropy": 6.139745378494263,
      "epoch": 1.3539761083620734,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004824186558999117,
      "loss": 5.8177,
      "mean_token_accuracy": 0.15554581433534623,
      "num_tokens": 26495647.0,
      "step": 13545
    },
    {
      "entropy": 5.986030340194702,
      "epoch": 1.3544759334233019,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048240482299877386,
      "loss": 5.792,
      "mean_token_accuracy": 0.1691782146692276,
      "num_tokens": 26506048.0,
      "step": 13550
    },
    {
      "entropy": 6.169332027435303,
      "epoch": 1.3549757584845303,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004823909848793841,
      "loss": 5.9453,
      "mean_token_accuracy": 0.16499796360731125,
      "num_tokens": 26515460.0,
      "step": 13555
    },
    {
      "entropy": 6.152423620223999,
      "epoch": 1.355475583545759,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048237714154209053,
      "loss": 5.9092,
      "mean_token_accuracy": 0.15594387650489808,
      "num_tokens": 26526106.0,
      "step": 13560
    },
    {
      "entropy": 6.022279930114746,
      "epoch": 1.3559754086069875,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004823632929872415,
      "loss": 5.792,
      "mean_token_accuracy": 0.15970796942710877,
      "num_tokens": 26535582.0,
      "step": 13565
    },
    {
      "entropy": 6.079316854476929,
      "epoch": 1.3564752336682162,
      "grad_norm": 1.125,
      "learning_rate": 0.00048234943921518524,
      "loss": 5.7894,
      "mean_token_accuracy": 0.16604193449020385,
      "num_tokens": 26544811.0,
      "step": 13570
    },
    {
      "entropy": 6.16995062828064,
      "epoch": 1.3569750587294447,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004823355802262705,
      "loss": 5.9467,
      "mean_token_accuracy": 0.1589004561305046,
      "num_tokens": 26555340.0,
      "step": 13575
    },
    {
      "entropy": 6.1405116558074955,
      "epoch": 1.3574748837906734,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004823217160208459,
      "loss": 5.8506,
      "mean_token_accuracy": 0.15805254131555557,
      "num_tokens": 26564876.0,
      "step": 13580
    },
    {
      "entropy": 6.023207187652588,
      "epoch": 1.3579747088519019,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004823078465992604,
      "loss": 5.8249,
      "mean_token_accuracy": 0.1656326726078987,
      "num_tokens": 26573967.0,
      "step": 13585
    },
    {
      "entropy": 6.0553630828857425,
      "epoch": 1.3584745339131303,
      "grad_norm": 1.25,
      "learning_rate": 0.0004822939719618627,
      "loss": 5.8492,
      "mean_token_accuracy": 0.16512156277894974,
      "num_tokens": 26583753.0,
      "step": 13590
    },
    {
      "entropy": 6.060771703720093,
      "epoch": 1.358974358974359,
      "grad_norm": 1.1875,
      "learning_rate": 0.000482280092109002,
      "loss": 5.8743,
      "mean_token_accuracy": 0.1625478371977806,
      "num_tokens": 26593214.0,
      "step": 13595
    },
    {
      "entropy": 6.118198394775391,
      "epoch": 1.3594741840355875,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048226620704102757,
      "loss": 5.9136,
      "mean_token_accuracy": 0.16295095831155776,
      "num_tokens": 26603257.0,
      "step": 13600
    },
    {
      "entropy": 6.076246976852417,
      "epoch": 1.359974009096816,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004822523167582886,
      "loss": 5.8168,
      "mean_token_accuracy": 0.1665407493710518,
      "num_tokens": 26612483.0,
      "step": 13605
    },
    {
      "entropy": 6.144712448120117,
      "epoch": 1.3604738341580447,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004822384212611348,
      "loss": 5.9205,
      "mean_token_accuracy": 0.16189026683568955,
      "num_tokens": 26623116.0,
      "step": 13610
    },
    {
      "entropy": 6.115827655792236,
      "epoch": 1.3609736592192734,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048222452054991563,
      "loss": 5.7793,
      "mean_token_accuracy": 0.1682182490825653,
      "num_tokens": 26632648.0,
      "step": 13615
    },
    {
      "entropy": 6.017352294921875,
      "epoch": 1.3614734842805019,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004822106146249808,
      "loss": 5.8118,
      "mean_token_accuracy": 0.17011586874723433,
      "num_tokens": 26642208.0,
      "step": 13620
    },
    {
      "entropy": 6.116962623596192,
      "epoch": 1.3619733093417303,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048219670348668026,
      "loss": 5.8929,
      "mean_token_accuracy": 0.15915140360593796,
      "num_tokens": 26651967.0,
      "step": 13625
    },
    {
      "entropy": 6.059648942947388,
      "epoch": 1.362473134402959,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048218278713536394,
      "loss": 5.7648,
      "mean_token_accuracy": 0.1689758986234665,
      "num_tokens": 26661138.0,
      "step": 13630
    },
    {
      "entropy": 6.073150587081909,
      "epoch": 1.3629729594641875,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004821688655713821,
      "loss": 5.8773,
      "mean_token_accuracy": 0.15725380629301072,
      "num_tokens": 26671779.0,
      "step": 13635
    },
    {
      "entropy": 5.914076852798462,
      "epoch": 1.363472784525416,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048215493879508474,
      "loss": 5.5913,
      "mean_token_accuracy": 0.18325228691101075,
      "num_tokens": 26681030.0,
      "step": 13640
    },
    {
      "entropy": 6.10153603553772,
      "epoch": 1.3639726095866447,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004821410068068226,
      "loss": 5.9543,
      "mean_token_accuracy": 0.15527025312185289,
      "num_tokens": 26691609.0,
      "step": 13645
    },
    {
      "entropy": 6.234211063385009,
      "epoch": 1.3644724346478734,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004821270696069459,
      "loss": 5.9852,
      "mean_token_accuracy": 0.14705763682723044,
      "num_tokens": 26700874.0,
      "step": 13650
    },
    {
      "entropy": 6.087811374664307,
      "epoch": 1.3649722597091019,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048211312719580553,
      "loss": 5.8268,
      "mean_token_accuracy": 0.16636868566274643,
      "num_tokens": 26711234.0,
      "step": 13655
    },
    {
      "entropy": 6.051929664611817,
      "epoch": 1.3654720847703303,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004820991795737521,
      "loss": 5.7647,
      "mean_token_accuracy": 0.17183658331632615,
      "num_tokens": 26720348.0,
      "step": 13660
    },
    {
      "entropy": 6.026276922225952,
      "epoch": 1.365971909831559,
      "grad_norm": 1.125,
      "learning_rate": 0.00048208522674113665,
      "loss": 5.781,
      "mean_token_accuracy": 0.17158832997083664,
      "num_tokens": 26730868.0,
      "step": 13665
    },
    {
      "entropy": 6.118976831436157,
      "epoch": 1.3664717348927875,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004820712686983101,
      "loss": 5.9137,
      "mean_token_accuracy": 0.1611762300133705,
      "num_tokens": 26740093.0,
      "step": 13670
    },
    {
      "entropy": 6.073562526702881,
      "epoch": 1.366971559954016,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048205730544562385,
      "loss": 5.8704,
      "mean_token_accuracy": 0.15991122722625734,
      "num_tokens": 26749550.0,
      "step": 13675
    },
    {
      "entropy": 6.140735530853272,
      "epoch": 1.3674713850152447,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000482043336983429,
      "loss": 5.9089,
      "mean_token_accuracy": 0.16120171993970872,
      "num_tokens": 26759416.0,
      "step": 13680
    },
    {
      "entropy": 6.027294063568116,
      "epoch": 1.3679712100764732,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048202936331207707,
      "loss": 5.8271,
      "mean_token_accuracy": 0.16028245836496352,
      "num_tokens": 26769782.0,
      "step": 13685
    },
    {
      "entropy": 6.131178283691407,
      "epoch": 1.3684710351377019,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004820153844319196,
      "loss": 5.8269,
      "mean_token_accuracy": 0.16766310781240462,
      "num_tokens": 26779007.0,
      "step": 13690
    },
    {
      "entropy": 6.048009538650513,
      "epoch": 1.3689708601989303,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048200140034330844,
      "loss": 5.815,
      "mean_token_accuracy": 0.16395278126001359,
      "num_tokens": 26787828.0,
      "step": 13695
    },
    {
      "entropy": 6.066766691207886,
      "epoch": 1.369470685260159,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048198741104659516,
      "loss": 5.7749,
      "mean_token_accuracy": 0.16377027928829194,
      "num_tokens": 26798190.0,
      "step": 13700
    },
    {
      "entropy": 6.067559480667114,
      "epoch": 1.3699705103213875,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000481973416542132,
      "loss": 5.8324,
      "mean_token_accuracy": 0.16061000376939774,
      "num_tokens": 26807312.0,
      "step": 13705
    },
    {
      "entropy": 6.0398722171783445,
      "epoch": 1.370470335382616,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004819594168302709,
      "loss": 5.8399,
      "mean_token_accuracy": 0.16815754771232605,
      "num_tokens": 26816477.0,
      "step": 13710
    },
    {
      "entropy": 5.977291536331177,
      "epoch": 1.3709701604438447,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004819454119113641,
      "loss": 5.722,
      "mean_token_accuracy": 0.16968167424201966,
      "num_tokens": 26825826.0,
      "step": 13715
    },
    {
      "entropy": 6.099328851699829,
      "epoch": 1.3714699855050732,
      "grad_norm": 1.171875,
      "learning_rate": 0.000481931401785764,
      "loss": 5.8256,
      "mean_token_accuracy": 0.16401616334915162,
      "num_tokens": 26834809.0,
      "step": 13720
    },
    {
      "entropy": 6.012691164016724,
      "epoch": 1.3719698105663019,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004819173864538231,
      "loss": 5.7735,
      "mean_token_accuracy": 0.16702558994293212,
      "num_tokens": 26845322.0,
      "step": 13725
    },
    {
      "entropy": 6.110891580581665,
      "epoch": 1.3724696356275303,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048190336591589394,
      "loss": 5.9521,
      "mean_token_accuracy": 0.14909159541130065,
      "num_tokens": 26854328.0,
      "step": 13730
    },
    {
      "entropy": 6.10371994972229,
      "epoch": 1.372969460688759,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004818893401723294,
      "loss": 5.9295,
      "mean_token_accuracy": 0.15479914620518684,
      "num_tokens": 26863299.0,
      "step": 13735
    },
    {
      "entropy": 5.998346757888794,
      "epoch": 1.3734692857499875,
      "grad_norm": 1.125,
      "learning_rate": 0.00048187530922348226,
      "loss": 5.8088,
      "mean_token_accuracy": 0.16938377171754837,
      "num_tokens": 26873461.0,
      "step": 13740
    },
    {
      "entropy": 6.069341659545898,
      "epoch": 1.373969110811216,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004818612730697056,
      "loss": 5.8343,
      "mean_token_accuracy": 0.1666663959622383,
      "num_tokens": 26883310.0,
      "step": 13745
    },
    {
      "entropy": 6.105596685409546,
      "epoch": 1.3744689358724447,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048184723171135256,
      "loss": 5.8273,
      "mean_token_accuracy": 0.16938856095075608,
      "num_tokens": 26892009.0,
      "step": 13750
    },
    {
      "entropy": 6.064331579208374,
      "epoch": 1.3749687609336732,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048183318514877634,
      "loss": 5.8416,
      "mean_token_accuracy": 0.15985509157180786,
      "num_tokens": 26902246.0,
      "step": 13755
    },
    {
      "entropy": 6.12076849937439,
      "epoch": 1.3754685859949018,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004818191333823305,
      "loss": 5.8907,
      "mean_token_accuracy": 0.1584865540266037,
      "num_tokens": 26912912.0,
      "step": 13760
    },
    {
      "entropy": 6.1428138256073,
      "epoch": 1.3759684110561303,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004818050764123684,
      "loss": 5.8251,
      "mean_token_accuracy": 0.16657163500785827,
      "num_tokens": 26921431.0,
      "step": 13765
    },
    {
      "entropy": 6.08453893661499,
      "epoch": 1.376468236117359,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004817910142392438,
      "loss": 5.8421,
      "mean_token_accuracy": 0.16551962196826936,
      "num_tokens": 26931990.0,
      "step": 13770
    },
    {
      "entropy": 6.110919046401977,
      "epoch": 1.3769680611785875,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048177694686331047,
      "loss": 5.8968,
      "mean_token_accuracy": 0.16200568079948424,
      "num_tokens": 26942801.0,
      "step": 13775
    },
    {
      "entropy": 6.10925178527832,
      "epoch": 1.377467886239816,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004817628742849224,
      "loss": 5.9153,
      "mean_token_accuracy": 0.16292500644922256,
      "num_tokens": 26952896.0,
      "step": 13780
    },
    {
      "entropy": 6.12700719833374,
      "epoch": 1.3779677113010447,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048174879650443364,
      "loss": 5.9416,
      "mean_token_accuracy": 0.15835729688405992,
      "num_tokens": 26963170.0,
      "step": 13785
    },
    {
      "entropy": 6.122665691375732,
      "epoch": 1.3784675363622731,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048173471352219835,
      "loss": 5.9208,
      "mean_token_accuracy": 0.16176882907748222,
      "num_tokens": 26972937.0,
      "step": 13790
    },
    {
      "entropy": 6.169432163238525,
      "epoch": 1.3789673614235018,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004817206253385708,
      "loss": 5.9648,
      "mean_token_accuracy": 0.16034233272075654,
      "num_tokens": 26985115.0,
      "step": 13795
    },
    {
      "entropy": 6.106958341598511,
      "epoch": 1.3794671864847303,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048170653195390555,
      "loss": 5.9038,
      "mean_token_accuracy": 0.16717433631420137,
      "num_tokens": 26993658.0,
      "step": 13800
    },
    {
      "entropy": 6.083568906784057,
      "epoch": 1.379967011545959,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004816924333685572,
      "loss": 5.8894,
      "mean_token_accuracy": 0.1588427171111107,
      "num_tokens": 27003877.0,
      "step": 13805
    },
    {
      "entropy": 6.07158522605896,
      "epoch": 1.3804668366071875,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004816783295828804,
      "loss": 5.7496,
      "mean_token_accuracy": 0.17428085654973985,
      "num_tokens": 27012125.0,
      "step": 13810
    },
    {
      "entropy": 6.1037986278533936,
      "epoch": 1.380966661668416,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048166422059722996,
      "loss": 5.9578,
      "mean_token_accuracy": 0.1585586652159691,
      "num_tokens": 27022158.0,
      "step": 13815
    },
    {
      "entropy": 6.012720346450806,
      "epoch": 1.3814664867296447,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004816501064119609,
      "loss": 5.8637,
      "mean_token_accuracy": 0.16490198373794557,
      "num_tokens": 27031966.0,
      "step": 13820
    },
    {
      "entropy": 6.114004993438721,
      "epoch": 1.3819663117908731,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004816359870274284,
      "loss": 5.8251,
      "mean_token_accuracy": 0.1633490338921547,
      "num_tokens": 27040805.0,
      "step": 13825
    },
    {
      "entropy": 6.106872272491455,
      "epoch": 1.3824661368521018,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004816218624439876,
      "loss": 5.915,
      "mean_token_accuracy": 0.1572038322687149,
      "num_tokens": 27049210.0,
      "step": 13830
    },
    {
      "entropy": 6.022852516174316,
      "epoch": 1.3829659619133303,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048160773266199384,
      "loss": 5.7769,
      "mean_token_accuracy": 0.1669204279780388,
      "num_tokens": 27058980.0,
      "step": 13835
    },
    {
      "entropy": 6.132980012893677,
      "epoch": 1.383465786974559,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004815935976818028,
      "loss": 5.8973,
      "mean_token_accuracy": 0.1630334660410881,
      "num_tokens": 27068539.0,
      "step": 13840
    },
    {
      "entropy": 6.105209875106811,
      "epoch": 1.3839656120357875,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048157945750377005,
      "loss": 5.8197,
      "mean_token_accuracy": 0.16305186599493027,
      "num_tokens": 27077500.0,
      "step": 13845
    },
    {
      "entropy": 6.005625772476196,
      "epoch": 1.384465437097016,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004815653121282512,
      "loss": 5.7692,
      "mean_token_accuracy": 0.17083501666784287,
      "num_tokens": 27087039.0,
      "step": 13850
    },
    {
      "entropy": 6.054522895812989,
      "epoch": 1.3849652621582447,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048155116155560227,
      "loss": 5.8297,
      "mean_token_accuracy": 0.16544469743967055,
      "num_tokens": 27096575.0,
      "step": 13855
    },
    {
      "entropy": 6.1430864334106445,
      "epoch": 1.3854650872194731,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048153700578617924,
      "loss": 5.919,
      "mean_token_accuracy": 0.15717772021889687,
      "num_tokens": 27106246.0,
      "step": 13860
    },
    {
      "entropy": 6.1079833030700685,
      "epoch": 1.3859649122807016,
      "grad_norm": 1.125,
      "learning_rate": 0.0004815228448203384,
      "loss": 5.9112,
      "mean_token_accuracy": 0.16186932623386383,
      "num_tokens": 27115529.0,
      "step": 13865
    },
    {
      "entropy": 6.094822692871094,
      "epoch": 1.3864647373419303,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004815086786584358,
      "loss": 5.782,
      "mean_token_accuracy": 0.16238153874874114,
      "num_tokens": 27125204.0,
      "step": 13870
    },
    {
      "entropy": 6.011035537719726,
      "epoch": 1.386964562403159,
      "grad_norm": 1.8125,
      "learning_rate": 0.000481494507300828,
      "loss": 5.8288,
      "mean_token_accuracy": 0.15971094518899917,
      "num_tokens": 27135053.0,
      "step": 13875
    },
    {
      "entropy": 6.157387542724609,
      "epoch": 1.3874643874643875,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004814803307478716,
      "loss": 5.8869,
      "mean_token_accuracy": 0.15456154495477675,
      "num_tokens": 27144863.0,
      "step": 13880
    },
    {
      "entropy": 6.043198680877685,
      "epoch": 1.387964212525616,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048146614899992313,
      "loss": 5.8041,
      "mean_token_accuracy": 0.16863977611064912,
      "num_tokens": 27154178.0,
      "step": 13885
    },
    {
      "entropy": 6.063104486465454,
      "epoch": 1.3884640375868447,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048145196205733954,
      "loss": 5.7875,
      "mean_token_accuracy": 0.1654209777712822,
      "num_tokens": 27164218.0,
      "step": 13890
    },
    {
      "entropy": 6.128119945526123,
      "epoch": 1.3889638626480731,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048143776992047767,
      "loss": 5.815,
      "mean_token_accuracy": 0.1623504117131233,
      "num_tokens": 27174255.0,
      "step": 13895
    },
    {
      "entropy": 6.110506963729859,
      "epoch": 1.3894636877093016,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048142357258969465,
      "loss": 5.9186,
      "mean_token_accuracy": 0.15519357919692994,
      "num_tokens": 27184989.0,
      "step": 13900
    },
    {
      "entropy": 6.027588367462158,
      "epoch": 1.3899635127705303,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004814093700653477,
      "loss": 5.8286,
      "mean_token_accuracy": 0.1661423146724701,
      "num_tokens": 27196249.0,
      "step": 13905
    },
    {
      "entropy": 6.04661374092102,
      "epoch": 1.390463337831759,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048139516234779403,
      "loss": 5.7624,
      "mean_token_accuracy": 0.17108414322137833,
      "num_tokens": 27204997.0,
      "step": 13910
    },
    {
      "entropy": 6.124519491195679,
      "epoch": 1.3909631628929875,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048138094943739123,
      "loss": 5.837,
      "mean_token_accuracy": 0.1620077520608902,
      "num_tokens": 27214344.0,
      "step": 13915
    },
    {
      "entropy": 6.161216735839844,
      "epoch": 1.391462987954216,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048136673133449687,
      "loss": 5.8473,
      "mean_token_accuracy": 0.16305551528930665,
      "num_tokens": 27223950.0,
      "step": 13920
    },
    {
      "entropy": 6.014966917037964,
      "epoch": 1.3919628130154447,
      "grad_norm": 1.125,
      "learning_rate": 0.0004813525080394686,
      "loss": 5.6787,
      "mean_token_accuracy": 0.16915944069623948,
      "num_tokens": 27233371.0,
      "step": 13925
    },
    {
      "entropy": 6.074912214279175,
      "epoch": 1.3924626380766731,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004813382795526644,
      "loss": 5.9219,
      "mean_token_accuracy": 0.16234522312879562,
      "num_tokens": 27243413.0,
      "step": 13930
    },
    {
      "entropy": 6.110294437408447,
      "epoch": 1.3929624631379016,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004813240458744421,
      "loss": 5.8296,
      "mean_token_accuracy": 0.16170283406972885,
      "num_tokens": 27254117.0,
      "step": 13935
    },
    {
      "entropy": 6.0584851741790775,
      "epoch": 1.3934622881991303,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048130980700516,
      "loss": 5.8906,
      "mean_token_accuracy": 0.16052232086658477,
      "num_tokens": 27263202.0,
      "step": 13940
    },
    {
      "entropy": 6.045048618316651,
      "epoch": 1.3939621132603588,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004812955629451761,
      "loss": 5.7905,
      "mean_token_accuracy": 0.16111986041069032,
      "num_tokens": 27272830.0,
      "step": 13945
    },
    {
      "entropy": 6.083675050735474,
      "epoch": 1.3944619383215875,
      "grad_norm": 1.125,
      "learning_rate": 0.0004812813136948491,
      "loss": 5.8363,
      "mean_token_accuracy": 0.165091772377491,
      "num_tokens": 27283567.0,
      "step": 13950
    },
    {
      "entropy": 6.074646615982056,
      "epoch": 1.394961763382816,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048126705925453716,
      "loss": 5.7532,
      "mean_token_accuracy": 0.16984589248895646,
      "num_tokens": 27292113.0,
      "step": 13955
    },
    {
      "entropy": 5.9883486270904545,
      "epoch": 1.3954615884440447,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004812527996245992,
      "loss": 5.7897,
      "mean_token_accuracy": 0.16766416430473327,
      "num_tokens": 27301605.0,
      "step": 13960
    },
    {
      "entropy": 6.067078065872193,
      "epoch": 1.3959614135052731,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004812385348053938,
      "loss": 5.9054,
      "mean_token_accuracy": 0.15933808237314223,
      "num_tokens": 27312453.0,
      "step": 13965
    },
    {
      "entropy": 6.102055311203003,
      "epoch": 1.3964612385665016,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048122426479728,
      "loss": 5.8907,
      "mean_token_accuracy": 0.15531259179115295,
      "num_tokens": 27322641.0,
      "step": 13970
    },
    {
      "entropy": 5.999157857894898,
      "epoch": 1.3969610636277303,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004812099896006167,
      "loss": 5.675,
      "mean_token_accuracy": 0.1708381339907646,
      "num_tokens": 27332283.0,
      "step": 13975
    },
    {
      "entropy": 6.0551563739776615,
      "epoch": 1.3974608886889588,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004811957092157631,
      "loss": 5.7986,
      "mean_token_accuracy": 0.16840740442276,
      "num_tokens": 27341203.0,
      "step": 13980
    },
    {
      "entropy": 6.021498584747315,
      "epoch": 1.3979607137501875,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048118142364307864,
      "loss": 5.8925,
      "mean_token_accuracy": 0.1633613184094429,
      "num_tokens": 27351263.0,
      "step": 13985
    },
    {
      "entropy": 6.0399330139160154,
      "epoch": 1.398460538811416,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004811671328829225,
      "loss": 5.8064,
      "mean_token_accuracy": 0.16331521719694136,
      "num_tokens": 27361001.0,
      "step": 13990
    },
    {
      "entropy": 6.117093324661255,
      "epoch": 1.3989603638726447,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048115283693565445,
      "loss": 5.8366,
      "mean_token_accuracy": 0.16248753368854524,
      "num_tokens": 27370097.0,
      "step": 13995
    },
    {
      "entropy": 6.015833187103271,
      "epoch": 1.3994601889338731,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000481138535801634,
      "loss": 5.8543,
      "mean_token_accuracy": 0.16704343855381013,
      "num_tokens": 27379795.0,
      "step": 14000
    },
    {
      "entropy": 5.9998438358306885,
      "epoch": 1.3999600139951016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000481124229481221,
      "loss": 5.8348,
      "mean_token_accuracy": 0.16843980103731154,
      "num_tokens": 27389515.0,
      "step": 14005
    },
    {
      "entropy": 6.084005069732666,
      "epoch": 1.4004598390563303,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004811099179747755,
      "loss": 5.7364,
      "mean_token_accuracy": 0.18125393688678743,
      "num_tokens": 27400177.0,
      "step": 14010
    },
    {
      "entropy": 6.139904975891113,
      "epoch": 1.4009596641175588,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048109560128265747,
      "loss": 5.8249,
      "mean_token_accuracy": 0.16502975821495056,
      "num_tokens": 27409381.0,
      "step": 14015
    },
    {
      "entropy": 6.110792541503907,
      "epoch": 1.4014594891787875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004810812794052271,
      "loss": 5.9326,
      "mean_token_accuracy": 0.1579738199710846,
      "num_tokens": 27419897.0,
      "step": 14020
    },
    {
      "entropy": 6.204196166992188,
      "epoch": 1.401959314240016,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004810669523428449,
      "loss": 6.028,
      "mean_token_accuracy": 0.15139959156513214,
      "num_tokens": 27430693.0,
      "step": 14025
    },
    {
      "entropy": 6.188246774673462,
      "epoch": 1.4024591393012447,
      "grad_norm": 1.15625,
      "learning_rate": 0.000481052620095871,
      "loss": 5.9607,
      "mean_token_accuracy": 0.15715833902359008,
      "num_tokens": 27440834.0,
      "step": 14030
    },
    {
      "entropy": 6.016510915756226,
      "epoch": 1.4029589643624731,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004810382826646664,
      "loss": 5.7634,
      "mean_token_accuracy": 0.1724162995815277,
      "num_tokens": 27450166.0,
      "step": 14035
    },
    {
      "entropy": 5.986206960678101,
      "epoch": 1.4034587894237016,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004810239400495915,
      "loss": 5.8533,
      "mean_token_accuracy": 0.15899006724357606,
      "num_tokens": 27459863.0,
      "step": 14040
    },
    {
      "entropy": 6.0454038143157955,
      "epoch": 1.4039586144849303,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004810095922510073,
      "loss": 5.7622,
      "mean_token_accuracy": 0.16292489916086197,
      "num_tokens": 27469118.0,
      "step": 14045
    },
    {
      "entropy": 6.0967450618743895,
      "epoch": 1.4044584395461588,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048099523926927487,
      "loss": 5.855,
      "mean_token_accuracy": 0.16465488001704215,
      "num_tokens": 27479550.0,
      "step": 14050
    },
    {
      "entropy": 6.094939184188843,
      "epoch": 1.4049582646073875,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004809808811047552,
      "loss": 5.8956,
      "mean_token_accuracy": 0.15917786732316017,
      "num_tokens": 27489625.0,
      "step": 14055
    },
    {
      "entropy": 6.181684446334839,
      "epoch": 1.405458089668616,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048096651775780943,
      "loss": 5.9311,
      "mean_token_accuracy": 0.15699216201901436,
      "num_tokens": 27501459.0,
      "step": 14060
    },
    {
      "entropy": 6.102576398849488,
      "epoch": 1.4059579147298447,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004809521492287992,
      "loss": 5.7676,
      "mean_token_accuracy": 0.16084038093686104,
      "num_tokens": 27511450.0,
      "step": 14065
    },
    {
      "entropy": 5.981971597671508,
      "epoch": 1.4064577397910731,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004809377755180858,
      "loss": 5.7459,
      "mean_token_accuracy": 0.1672506034374237,
      "num_tokens": 27520783.0,
      "step": 14070
    },
    {
      "entropy": 6.143625211715698,
      "epoch": 1.4069575648523016,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004809233966260311,
      "loss": 5.887,
      "mean_token_accuracy": 0.15933382585644723,
      "num_tokens": 27529433.0,
      "step": 14075
    },
    {
      "entropy": 6.020356845855713,
      "epoch": 1.4074573899135303,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004809090125529966,
      "loss": 5.8195,
      "mean_token_accuracy": 0.16606845408678056,
      "num_tokens": 27539278.0,
      "step": 14080
    },
    {
      "entropy": 6.048391246795655,
      "epoch": 1.4079572149747588,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048089462329934437,
      "loss": 5.8888,
      "mean_token_accuracy": 0.1603079229593277,
      "num_tokens": 27548474.0,
      "step": 14085
    },
    {
      "entropy": 6.146444272994995,
      "epoch": 1.4084570400359875,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048088022886543636,
      "loss": 5.8794,
      "mean_token_accuracy": 0.16731360256671907,
      "num_tokens": 27557757.0,
      "step": 14090
    },
    {
      "entropy": 6.138942289352417,
      "epoch": 1.408956865097216,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004808658292516348,
      "loss": 5.9481,
      "mean_token_accuracy": 0.15234176963567733,
      "num_tokens": 27567971.0,
      "step": 14095
    },
    {
      "entropy": 6.183567190170288,
      "epoch": 1.4094566901584447,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048085142445830197,
      "loss": 5.9257,
      "mean_token_accuracy": 0.1484348312020302,
      "num_tokens": 27578135.0,
      "step": 14100
    },
    {
      "entropy": 6.120506620407104,
      "epoch": 1.4099565152196731,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048083701448580027,
      "loss": 5.801,
      "mean_token_accuracy": 0.16539654284715652,
      "num_tokens": 27587459.0,
      "step": 14105
    },
    {
      "entropy": 6.102673196792603,
      "epoch": 1.4104563402809016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048082259933449224,
      "loss": 5.9114,
      "mean_token_accuracy": 0.1612028121948242,
      "num_tokens": 27596494.0,
      "step": 14110
    },
    {
      "entropy": 5.967799139022827,
      "epoch": 1.4109561653421303,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004808081790047406,
      "loss": 5.7725,
      "mean_token_accuracy": 0.17277775406837464,
      "num_tokens": 27605482.0,
      "step": 14115
    },
    {
      "entropy": 6.071335506439209,
      "epoch": 1.4114559904033588,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048079375349690807,
      "loss": 5.8042,
      "mean_token_accuracy": 0.16155136823654176,
      "num_tokens": 27615386.0,
      "step": 14120
    },
    {
      "entropy": 6.127578449249268,
      "epoch": 1.4119558154645873,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048077932281135775,
      "loss": 5.8869,
      "mean_token_accuracy": 0.1642364516854286,
      "num_tokens": 27624834.0,
      "step": 14125
    },
    {
      "entropy": 6.076176595687866,
      "epoch": 1.412455640525816,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048076488694845257,
      "loss": 5.8026,
      "mean_token_accuracy": 0.1637851119041443,
      "num_tokens": 27634087.0,
      "step": 14130
    },
    {
      "entropy": 6.084689331054688,
      "epoch": 1.4129554655870447,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048075044590855574,
      "loss": 5.7456,
      "mean_token_accuracy": 0.16912320405244827,
      "num_tokens": 27643739.0,
      "step": 14135
    },
    {
      "entropy": 6.040442895889282,
      "epoch": 1.4134552906482731,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048073599969203064,
      "loss": 5.8762,
      "mean_token_accuracy": 0.16216918379068374,
      "num_tokens": 27652594.0,
      "step": 14140
    },
    {
      "entropy": 6.1067084789276125,
      "epoch": 1.4139551157095016,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004807215482992408,
      "loss": 5.8743,
      "mean_token_accuracy": 0.16033419370651245,
      "num_tokens": 27662766.0,
      "step": 14145
    },
    {
      "entropy": 6.143559837341309,
      "epoch": 1.4144549407707303,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004807070917305497,
      "loss": 5.8932,
      "mean_token_accuracy": 0.16287220418453216,
      "num_tokens": 27673547.0,
      "step": 14150
    },
    {
      "entropy": 6.148661422729492,
      "epoch": 1.4149547658319588,
      "grad_norm": 1.25,
      "learning_rate": 0.00048069262998632115,
      "loss": 5.8668,
      "mean_token_accuracy": 0.16413661539554597,
      "num_tokens": 27683056.0,
      "step": 14155
    },
    {
      "entropy": 6.0662841796875,
      "epoch": 1.4154545908931873,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004806781630669189,
      "loss": 5.8698,
      "mean_token_accuracy": 0.15854871571063994,
      "num_tokens": 27693049.0,
      "step": 14160
    },
    {
      "entropy": 6.072657918930053,
      "epoch": 1.415954415954416,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048066369097270694,
      "loss": 5.8374,
      "mean_token_accuracy": 0.1523838944733143,
      "num_tokens": 27703712.0,
      "step": 14165
    },
    {
      "entropy": 6.082448244094849,
      "epoch": 1.4164542410156447,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004806492137040495,
      "loss": 5.8442,
      "mean_token_accuracy": 0.16726205199956895,
      "num_tokens": 27713460.0,
      "step": 14170
    },
    {
      "entropy": 6.139132928848267,
      "epoch": 1.4169540660768731,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048063473126131086,
      "loss": 5.897,
      "mean_token_accuracy": 0.1601972609758377,
      "num_tokens": 27722942.0,
      "step": 14175
    },
    {
      "entropy": 6.078747367858886,
      "epoch": 1.4174538911381016,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004806202436448551,
      "loss": 5.8622,
      "mean_token_accuracy": 0.15828557014465333,
      "num_tokens": 27732738.0,
      "step": 14180
    },
    {
      "entropy": 6.115740966796875,
      "epoch": 1.4179537161993303,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004806057508550471,
      "loss": 5.9138,
      "mean_token_accuracy": 0.15778427794575692,
      "num_tokens": 27742169.0,
      "step": 14185
    },
    {
      "entropy": 6.113169622421265,
      "epoch": 1.4184535412605588,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004805912528922512,
      "loss": 5.8525,
      "mean_token_accuracy": 0.1645225390791893,
      "num_tokens": 27752539.0,
      "step": 14190
    },
    {
      "entropy": 6.121021795272827,
      "epoch": 1.4189533663217873,
      "grad_norm": 1.25,
      "learning_rate": 0.0004805767497568323,
      "loss": 5.9132,
      "mean_token_accuracy": 0.1571896880865097,
      "num_tokens": 27763494.0,
      "step": 14195
    },
    {
      "entropy": 6.06669602394104,
      "epoch": 1.419453191383016,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048056224144915534,
      "loss": 5.7597,
      "mean_token_accuracy": 0.17333806604146956,
      "num_tokens": 27772884.0,
      "step": 14200
    },
    {
      "entropy": 5.989477825164795,
      "epoch": 1.4199530164442444,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048054772796958517,
      "loss": 5.746,
      "mean_token_accuracy": 0.17673812955617904,
      "num_tokens": 27782447.0,
      "step": 14205
    },
    {
      "entropy": 6.089370441436768,
      "epoch": 1.4204528415054731,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048053320931848713,
      "loss": 5.8345,
      "mean_token_accuracy": 0.16472698599100113,
      "num_tokens": 27790781.0,
      "step": 14210
    },
    {
      "entropy": 6.065466785430909,
      "epoch": 1.4209526665667016,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048051868549622644,
      "loss": 5.7531,
      "mean_token_accuracy": 0.17438847124576567,
      "num_tokens": 27799781.0,
      "step": 14215
    },
    {
      "entropy": 5.984084177017212,
      "epoch": 1.4214524916279303,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048050415650316853,
      "loss": 5.8108,
      "mean_token_accuracy": 0.15955475345253944,
      "num_tokens": 27809067.0,
      "step": 14220
    },
    {
      "entropy": 6.135573387145996,
      "epoch": 1.4219523166891588,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004804896223396789,
      "loss": 5.8818,
      "mean_token_accuracy": 0.15889531522989273,
      "num_tokens": 27819194.0,
      "step": 14225
    },
    {
      "entropy": 6.121607446670533,
      "epoch": 1.4224521417503873,
      "grad_norm": 1.125,
      "learning_rate": 0.0004804750830061232,
      "loss": 5.8629,
      "mean_token_accuracy": 0.16082900166511535,
      "num_tokens": 27830179.0,
      "step": 14230
    },
    {
      "entropy": 6.106305074691773,
      "epoch": 1.422951966811616,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004804605385028673,
      "loss": 5.8321,
      "mean_token_accuracy": 0.16052630990743638,
      "num_tokens": 27840476.0,
      "step": 14235
    },
    {
      "entropy": 6.035150194168091,
      "epoch": 1.4234517918728444,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004804459888302772,
      "loss": 5.7418,
      "mean_token_accuracy": 0.16598664224147797,
      "num_tokens": 27850247.0,
      "step": 14240
    },
    {
      "entropy": 6.087751293182373,
      "epoch": 1.4239516169340731,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048043143398871874,
      "loss": 5.8844,
      "mean_token_accuracy": 0.15817656591534615,
      "num_tokens": 27860290.0,
      "step": 14245
    },
    {
      "entropy": 6.124210834503174,
      "epoch": 1.4244514419953016,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004804168739785583,
      "loss": 5.8759,
      "mean_token_accuracy": 0.15635038912296295,
      "num_tokens": 27870506.0,
      "step": 14250
    },
    {
      "entropy": 6.142401456832886,
      "epoch": 1.4249512670565303,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004804023088001623,
      "loss": 5.9314,
      "mean_token_accuracy": 0.15704911947250366,
      "num_tokens": 27880679.0,
      "step": 14255
    },
    {
      "entropy": 6.036970615386963,
      "epoch": 1.4254510921177588,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004803877384538969,
      "loss": 5.812,
      "mean_token_accuracy": 0.16879701018333435,
      "num_tokens": 27891532.0,
      "step": 14260
    },
    {
      "entropy": 5.859154462814331,
      "epoch": 1.4259509171789873,
      "grad_norm": 1.5,
      "learning_rate": 0.000480373162940129,
      "loss": 5.624,
      "mean_token_accuracy": 0.1901191547513008,
      "num_tokens": 27902659.0,
      "step": 14265
    },
    {
      "entropy": 6.087169218063354,
      "epoch": 1.426450742240216,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048035858225922506,
      "loss": 5.8174,
      "mean_token_accuracy": 0.1671767368912697,
      "num_tokens": 27912422.0,
      "step": 14270
    },
    {
      "entropy": 6.037039279937744,
      "epoch": 1.4269505673014444,
      "grad_norm": 1.125,
      "learning_rate": 0.00048034399641155206,
      "loss": 5.8204,
      "mean_token_accuracy": 0.16071702688932418,
      "num_tokens": 27922898.0,
      "step": 14275
    },
    {
      "entropy": 6.109409284591675,
      "epoch": 1.4274503923626731,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048032940539747697,
      "loss": 5.8281,
      "mean_token_accuracy": 0.1614475056529045,
      "num_tokens": 27932915.0,
      "step": 14280
    },
    {
      "entropy": 6.128507137298584,
      "epoch": 1.4279502174239016,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048031480921736686,
      "loss": 5.8096,
      "mean_token_accuracy": 0.16437938809394836,
      "num_tokens": 27942760.0,
      "step": 14285
    },
    {
      "entropy": 6.086759042739868,
      "epoch": 1.4284500424851303,
      "grad_norm": 1.171875,
      "learning_rate": 0.000480300207871589,
      "loss": 5.8901,
      "mean_token_accuracy": 0.1562187224626541,
      "num_tokens": 27952898.0,
      "step": 14290
    },
    {
      "entropy": 6.035344934463501,
      "epoch": 1.4289498675463588,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048028560136051076,
      "loss": 5.8014,
      "mean_token_accuracy": 0.16789491921663285,
      "num_tokens": 27962292.0,
      "step": 14295
    },
    {
      "entropy": 6.106490612030029,
      "epoch": 1.4294496926075873,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048027098968449954,
      "loss": 5.8592,
      "mean_token_accuracy": 0.16489725559949875,
      "num_tokens": 27972480.0,
      "step": 14300
    },
    {
      "entropy": 6.154496145248413,
      "epoch": 1.429949517668816,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004802563728439231,
      "loss": 5.9378,
      "mean_token_accuracy": 0.15596820116043092,
      "num_tokens": 27982224.0,
      "step": 14305
    },
    {
      "entropy": 6.115724802017212,
      "epoch": 1.4304493427300444,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004802417508391491,
      "loss": 5.9203,
      "mean_token_accuracy": 0.1634940892457962,
      "num_tokens": 27991512.0,
      "step": 14310
    },
    {
      "entropy": 6.028168106079102,
      "epoch": 1.4309491677912731,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004802271236705455,
      "loss": 5.7271,
      "mean_token_accuracy": 0.1677955284714699,
      "num_tokens": 28001392.0,
      "step": 14315
    },
    {
      "entropy": 5.971262359619141,
      "epoch": 1.4314489928525016,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048021249133848025,
      "loss": 5.7071,
      "mean_token_accuracy": 0.17243615984916688,
      "num_tokens": 28009934.0,
      "step": 14320
    },
    {
      "entropy": 6.061548328399658,
      "epoch": 1.4319488179137303,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048019785384332154,
      "loss": 5.8552,
      "mean_token_accuracy": 0.16028969138860702,
      "num_tokens": 28019485.0,
      "step": 14325
    },
    {
      "entropy": 6.065999460220337,
      "epoch": 1.4324486429749588,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004801832111854376,
      "loss": 5.8367,
      "mean_token_accuracy": 0.16402855813503264,
      "num_tokens": 28028666.0,
      "step": 14330
    },
    {
      "entropy": 6.088905668258667,
      "epoch": 1.4329484680361873,
      "grad_norm": 1.125,
      "learning_rate": 0.0004801685633651969,
      "loss": 5.8446,
      "mean_token_accuracy": 0.15929820388555527,
      "num_tokens": 28038677.0,
      "step": 14335
    },
    {
      "entropy": 6.1781820297241214,
      "epoch": 1.433448293097416,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004801539103829679,
      "loss": 5.9406,
      "mean_token_accuracy": 0.15137656927108764,
      "num_tokens": 28048580.0,
      "step": 14340
    },
    {
      "entropy": 6.135907602310181,
      "epoch": 1.4339481181586444,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004801392522391193,
      "loss": 5.8462,
      "mean_token_accuracy": 0.15881284028291703,
      "num_tokens": 28058958.0,
      "step": 14345
    },
    {
      "entropy": 6.090902853012085,
      "epoch": 1.4344479432198731,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048012458893401995,
      "loss": 5.8529,
      "mean_token_accuracy": 0.16032872945070267,
      "num_tokens": 28068861.0,
      "step": 14350
    },
    {
      "entropy": 6.0439637184143065,
      "epoch": 1.4349477682811016,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004801099204680387,
      "loss": 5.8592,
      "mean_token_accuracy": 0.16746776849031447,
      "num_tokens": 28078638.0,
      "step": 14355
    },
    {
      "entropy": 6.103947019577026,
      "epoch": 1.4354475933423303,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048009524684154464,
      "loss": 5.8306,
      "mean_token_accuracy": 0.16375937983393668,
      "num_tokens": 28088549.0,
      "step": 14360
    },
    {
      "entropy": 6.160644578933716,
      "epoch": 1.4359474184035588,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004800805680549069,
      "loss": 5.8821,
      "mean_token_accuracy": 0.160674025118351,
      "num_tokens": 28097743.0,
      "step": 14365
    },
    {
      "entropy": 6.047574043273926,
      "epoch": 1.4364472434647872,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048006588410849497,
      "loss": 5.8006,
      "mean_token_accuracy": 0.16531443744897842,
      "num_tokens": 28107187.0,
      "step": 14370
    },
    {
      "entropy": 6.014564847946167,
      "epoch": 1.436947068526016,
      "grad_norm": 1.0625,
      "learning_rate": 0.000480051195002678,
      "loss": 5.7789,
      "mean_token_accuracy": 0.16393042504787445,
      "num_tokens": 28117730.0,
      "step": 14375
    },
    {
      "entropy": 6.134649467468262,
      "epoch": 1.4374468935872444,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048003650073782586,
      "loss": 5.8958,
      "mean_token_accuracy": 0.15972108542919158,
      "num_tokens": 28127803.0,
      "step": 14380
    },
    {
      "entropy": 6.132762908935547,
      "epoch": 1.437946718648473,
      "grad_norm": 1.15625,
      "learning_rate": 0.000480021801314308,
      "loss": 5.8287,
      "mean_token_accuracy": 0.16596219837665557,
      "num_tokens": 28137946.0,
      "step": 14385
    },
    {
      "entropy": 6.0943175792694095,
      "epoch": 1.4384465437097016,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004800070967324944,
      "loss": 5.9163,
      "mean_token_accuracy": 0.1601300448179245,
      "num_tokens": 28147895.0,
      "step": 14390
    },
    {
      "entropy": 6.005673170089722,
      "epoch": 1.4389463687709303,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047999238699275506,
      "loss": 5.8217,
      "mean_token_accuracy": 0.16754652857780455,
      "num_tokens": 28157641.0,
      "step": 14395
    },
    {
      "entropy": 6.060574865341186,
      "epoch": 1.4394461938321588,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047997767209546,
      "loss": 5.8761,
      "mean_token_accuracy": 0.156871934235096,
      "num_tokens": 28167320.0,
      "step": 14400
    },
    {
      "entropy": 6.071553421020508,
      "epoch": 1.4399460188933872,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004799629520409795,
      "loss": 5.8162,
      "mean_token_accuracy": 0.17021039873361588,
      "num_tokens": 28176697.0,
      "step": 14405
    },
    {
      "entropy": 6.088327360153198,
      "epoch": 1.440445843954616,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047994822682968373,
      "loss": 5.8731,
      "mean_token_accuracy": 0.16499809324741363,
      "num_tokens": 28186688.0,
      "step": 14410
    },
    {
      "entropy": 6.130187749862671,
      "epoch": 1.4409456690158444,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047993349646194343,
      "loss": 5.9747,
      "mean_token_accuracy": 0.1570012740790844,
      "num_tokens": 28197293.0,
      "step": 14415
    },
    {
      "entropy": 6.183364534378052,
      "epoch": 1.441445494077073,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047991876093812904,
      "loss": 5.9722,
      "mean_token_accuracy": 0.1556063175201416,
      "num_tokens": 28207413.0,
      "step": 14420
    },
    {
      "entropy": 6.075418519973755,
      "epoch": 1.4419453191383016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047990402025861143,
      "loss": 5.861,
      "mean_token_accuracy": 0.1621878668665886,
      "num_tokens": 28217575.0,
      "step": 14425
    },
    {
      "entropy": 5.957625484466552,
      "epoch": 1.4424451441995303,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004798892744237613,
      "loss": 5.6909,
      "mean_token_accuracy": 0.1722387731075287,
      "num_tokens": 28226460.0,
      "step": 14430
    },
    {
      "entropy": 6.03192105293274,
      "epoch": 1.4429449692607588,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047987452343394976,
      "loss": 5.836,
      "mean_token_accuracy": 0.1613019436597824,
      "num_tokens": 28236689.0,
      "step": 14435
    },
    {
      "entropy": 6.0852562427520756,
      "epoch": 1.4434447943219872,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047985976728954794,
      "loss": 5.8366,
      "mean_token_accuracy": 0.16860390305519105,
      "num_tokens": 28247057.0,
      "step": 14440
    },
    {
      "entropy": 6.155788469314575,
      "epoch": 1.443944619383216,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004798450059909271,
      "loss": 5.8471,
      "mean_token_accuracy": 0.16668982207775115,
      "num_tokens": 28256475.0,
      "step": 14445
    },
    {
      "entropy": 6.03723111152649,
      "epoch": 1.4444444444444444,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004798302395384586,
      "loss": 5.8134,
      "mean_token_accuracy": 0.16186632215976715,
      "num_tokens": 28266120.0,
      "step": 14450
    },
    {
      "entropy": 6.033184432983399,
      "epoch": 1.444944269505673,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047981546793251405,
      "loss": 5.8287,
      "mean_token_accuracy": 0.16238273084163665,
      "num_tokens": 28276325.0,
      "step": 14455
    },
    {
      "entropy": 6.082345056533813,
      "epoch": 1.4454440945669016,
      "grad_norm": 1.109375,
      "learning_rate": 0.000479800691173465,
      "loss": 5.8054,
      "mean_token_accuracy": 0.16285579428076744,
      "num_tokens": 28286007.0,
      "step": 14460
    },
    {
      "entropy": 6.105082607269287,
      "epoch": 1.4459439196281303,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004797859092616832,
      "loss": 5.8194,
      "mean_token_accuracy": 0.16023024320602416,
      "num_tokens": 28296346.0,
      "step": 14465
    },
    {
      "entropy": 5.9656054973602295,
      "epoch": 1.4464437446893588,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047977112219754064,
      "loss": 5.7049,
      "mean_token_accuracy": 0.17640091329813004,
      "num_tokens": 28306137.0,
      "step": 14470
    },
    {
      "entropy": 6.076648759841919,
      "epoch": 1.4469435697505872,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004797563299814094,
      "loss": 5.9378,
      "mean_token_accuracy": 0.15814715921878814,
      "num_tokens": 28316369.0,
      "step": 14475
    },
    {
      "entropy": 6.133443975448609,
      "epoch": 1.447443394811816,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047974153261366156,
      "loss": 5.8182,
      "mean_token_accuracy": 0.16561497300863265,
      "num_tokens": 28325338.0,
      "step": 14480
    },
    {
      "entropy": 6.018494510650635,
      "epoch": 1.4479432198730444,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004797267300946694,
      "loss": 5.8146,
      "mean_token_accuracy": 0.1627163290977478,
      "num_tokens": 28334929.0,
      "step": 14485
    },
    {
      "entropy": 6.114715576171875,
      "epoch": 1.448443044934273,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047971192242480544,
      "loss": 5.841,
      "mean_token_accuracy": 0.1671803429722786,
      "num_tokens": 28345157.0,
      "step": 14490
    },
    {
      "entropy": 6.0402219772338865,
      "epoch": 1.4489428699955016,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004796971096044422,
      "loss": 5.7632,
      "mean_token_accuracy": 0.16323568522930146,
      "num_tokens": 28354277.0,
      "step": 14495
    },
    {
      "entropy": 6.030753087997437,
      "epoch": 1.44944269505673,
      "grad_norm": 1.125,
      "learning_rate": 0.0004796822916339523,
      "loss": 5.7785,
      "mean_token_accuracy": 0.1692432627081871,
      "num_tokens": 28364154.0,
      "step": 14500
    },
    {
      "entropy": 6.104008150100708,
      "epoch": 1.4499425201179588,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047966746851370865,
      "loss": 5.842,
      "mean_token_accuracy": 0.16231500059366227,
      "num_tokens": 28375260.0,
      "step": 14505
    },
    {
      "entropy": 6.077221918106079,
      "epoch": 1.4504423451791872,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004796526402440841,
      "loss": 5.8377,
      "mean_token_accuracy": 0.16398291438817977,
      "num_tokens": 28384359.0,
      "step": 14510
    },
    {
      "entropy": 6.043411159515381,
      "epoch": 1.450942170240416,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004796378068254518,
      "loss": 5.8512,
      "mean_token_accuracy": 0.16902666836977004,
      "num_tokens": 28394389.0,
      "step": 14515
    },
    {
      "entropy": 6.015130996704102,
      "epoch": 1.4514419953016444,
      "grad_norm": 1.171875,
      "learning_rate": 0.000479622968258185,
      "loss": 5.8363,
      "mean_token_accuracy": 0.15997121930122377,
      "num_tokens": 28403495.0,
      "step": 14520
    },
    {
      "entropy": 6.099828577041626,
      "epoch": 1.451941820362873,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004796081245426569,
      "loss": 5.9063,
      "mean_token_accuracy": 0.15974282026290892,
      "num_tokens": 28413132.0,
      "step": 14525
    },
    {
      "entropy": 6.136829900741577,
      "epoch": 1.4524416454241016,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047959327567924115,
      "loss": 5.8273,
      "mean_token_accuracy": 0.16639683842658998,
      "num_tokens": 28423815.0,
      "step": 14530
    },
    {
      "entropy": 6.116727638244629,
      "epoch": 1.45294147048533,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047957842166831103,
      "loss": 5.7958,
      "mean_token_accuracy": 0.16064665466547012,
      "num_tokens": 28433682.0,
      "step": 14535
    },
    {
      "entropy": 6.112469005584717,
      "epoch": 1.4534412955465588,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047956356251024066,
      "loss": 5.792,
      "mean_token_accuracy": 0.16930551528930665,
      "num_tokens": 28442388.0,
      "step": 14540
    },
    {
      "entropy": 6.075103712081909,
      "epoch": 1.4539411206077872,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004795486982054036,
      "loss": 5.8216,
      "mean_token_accuracy": 0.16418371349573135,
      "num_tokens": 28452753.0,
      "step": 14545
    },
    {
      "entropy": 6.0989004611969,
      "epoch": 1.454440945669016,
      "grad_norm": 1.578125,
      "learning_rate": 0.000479533828754174,
      "loss": 5.9087,
      "mean_token_accuracy": 0.16191526055335997,
      "num_tokens": 28462818.0,
      "step": 14550
    },
    {
      "entropy": 6.098900413513183,
      "epoch": 1.4549407707302444,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004795189541569258,
      "loss": 5.836,
      "mean_token_accuracy": 0.1634120061993599,
      "num_tokens": 28472208.0,
      "step": 14555
    },
    {
      "entropy": 6.113014507293701,
      "epoch": 1.455440595791473,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047950407441403343,
      "loss": 5.8404,
      "mean_token_accuracy": 0.16555726230144502,
      "num_tokens": 28480887.0,
      "step": 14560
    },
    {
      "entropy": 6.125952577590942,
      "epoch": 1.4559404208527016,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004794891895258711,
      "loss": 5.9162,
      "mean_token_accuracy": 0.16009568721055983,
      "num_tokens": 28491805.0,
      "step": 14565
    },
    {
      "entropy": 6.130037117004394,
      "epoch": 1.45644024591393,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004794742994928134,
      "loss": 5.8933,
      "mean_token_accuracy": 0.1629488378763199,
      "num_tokens": 28502913.0,
      "step": 14570
    },
    {
      "entropy": 6.124187803268432,
      "epoch": 1.4569400709751588,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000479459404315235,
      "loss": 5.7683,
      "mean_token_accuracy": 0.1633821338415146,
      "num_tokens": 28513143.0,
      "step": 14575
    },
    {
      "entropy": 6.084576034545899,
      "epoch": 1.4574398960363872,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047944450399351054,
      "loss": 5.8618,
      "mean_token_accuracy": 0.161972676217556,
      "num_tokens": 28522211.0,
      "step": 14580
    },
    {
      "entropy": 6.047715377807617,
      "epoch": 1.457939721097616,
      "grad_norm": 1.28125,
      "learning_rate": 0.000479429598528015,
      "loss": 5.8886,
      "mean_token_accuracy": 0.1630793385207653,
      "num_tokens": 28531672.0,
      "step": 14585
    },
    {
      "entropy": 6.065981101989746,
      "epoch": 1.4584395461588444,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004794146879191233,
      "loss": 5.798,
      "mean_token_accuracy": 0.16894703656435012,
      "num_tokens": 28541614.0,
      "step": 14590
    },
    {
      "entropy": 6.119306182861328,
      "epoch": 1.458939371220073,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004793997721672107,
      "loss": 5.9094,
      "mean_token_accuracy": 0.15380261540412904,
      "num_tokens": 28550353.0,
      "step": 14595
    },
    {
      "entropy": 6.090170955657959,
      "epoch": 1.4594391962813016,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047938485127265243,
      "loss": 5.8164,
      "mean_token_accuracy": 0.1655651330947876,
      "num_tokens": 28559772.0,
      "step": 14600
    },
    {
      "entropy": 6.101992273330689,
      "epoch": 1.45993902134253,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004793699252358239,
      "loss": 5.8441,
      "mean_token_accuracy": 0.16067686751484872,
      "num_tokens": 28569564.0,
      "step": 14605
    },
    {
      "entropy": 6.077731037139893,
      "epoch": 1.4604388464037588,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004793549940571006,
      "loss": 5.8409,
      "mean_token_accuracy": 0.162192203104496,
      "num_tokens": 28579473.0,
      "step": 14610
    },
    {
      "entropy": 6.153349590301514,
      "epoch": 1.4609386714649872,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047934005773685824,
      "loss": 5.8555,
      "mean_token_accuracy": 0.1618000313639641,
      "num_tokens": 28588407.0,
      "step": 14615
    },
    {
      "entropy": 6.096490049362183,
      "epoch": 1.461438496526216,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047932511627547266,
      "loss": 5.9476,
      "mean_token_accuracy": 0.15777512192726134,
      "num_tokens": 28599931.0,
      "step": 14620
    },
    {
      "entropy": 5.927955627441406,
      "epoch": 1.4619383215874444,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004793101696733197,
      "loss": 5.6296,
      "mean_token_accuracy": 0.1869066148996353,
      "num_tokens": 28609317.0,
      "step": 14625
    },
    {
      "entropy": 6.111376714706421,
      "epoch": 1.4624381466486729,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047929521793077545,
      "loss": 5.9149,
      "mean_token_accuracy": 0.15493391454219818,
      "num_tokens": 28618248.0,
      "step": 14630
    },
    {
      "entropy": 6.11438364982605,
      "epoch": 1.4629379717099016,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004792802610482161,
      "loss": 5.8976,
      "mean_token_accuracy": 0.16167005449533461,
      "num_tokens": 28629046.0,
      "step": 14635
    },
    {
      "entropy": 6.232850027084351,
      "epoch": 1.46343779677113,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004792652990260179,
      "loss": 5.9841,
      "mean_token_accuracy": 0.1493583142757416,
      "num_tokens": 28640829.0,
      "step": 14640
    },
    {
      "entropy": 6.177036142349243,
      "epoch": 1.4639376218323585,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004792503318645574,
      "loss": 5.8863,
      "mean_token_accuracy": 0.15729128271341325,
      "num_tokens": 28650812.0,
      "step": 14645
    },
    {
      "entropy": 6.191830444335937,
      "epoch": 1.4644374468935872,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047923535956421104,
      "loss": 5.9143,
      "mean_token_accuracy": 0.15560978800058364,
      "num_tokens": 28661795.0,
      "step": 14650
    },
    {
      "entropy": 6.040817451477051,
      "epoch": 1.464937271954816,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047922038212535565,
      "loss": 5.7319,
      "mean_token_accuracy": 0.16909156888723373,
      "num_tokens": 28670729.0,
      "step": 14655
    },
    {
      "entropy": 6.018751621246338,
      "epoch": 1.4654370970160444,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047920539954836793,
      "loss": 5.7531,
      "mean_token_accuracy": 0.17287863045930862,
      "num_tokens": 28680469.0,
      "step": 14660
    },
    {
      "entropy": 6.059257507324219,
      "epoch": 1.4659369220772729,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000479190411833625,
      "loss": 5.8624,
      "mean_token_accuracy": 0.1585902139544487,
      "num_tokens": 28689698.0,
      "step": 14665
    },
    {
      "entropy": 6.122041893005371,
      "epoch": 1.4664367471385016,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047917541898150377,
      "loss": 5.8037,
      "mean_token_accuracy": 0.16409275084733962,
      "num_tokens": 28699861.0,
      "step": 14670
    },
    {
      "entropy": 6.091724586486817,
      "epoch": 1.46693657219973,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047916042099238154,
      "loss": 5.8223,
      "mean_token_accuracy": 0.1578425295650959,
      "num_tokens": 28709370.0,
      "step": 14675
    },
    {
      "entropy": 6.094666862487793,
      "epoch": 1.4674363972609585,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047914541786663566,
      "loss": 6.0542,
      "mean_token_accuracy": 0.16446294039487838,
      "num_tokens": 28718123.0,
      "step": 14680
    },
    {
      "entropy": 6.166715049743653,
      "epoch": 1.4679362223221872,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047913040960464363,
      "loss": 5.9539,
      "mean_token_accuracy": 0.15581630170345306,
      "num_tokens": 28727972.0,
      "step": 14685
    },
    {
      "entropy": 6.125916767120361,
      "epoch": 1.468436047383416,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047911539620678305,
      "loss": 5.8334,
      "mean_token_accuracy": 0.15576455742120743,
      "num_tokens": 28737988.0,
      "step": 14690
    },
    {
      "entropy": 6.131151103973389,
      "epoch": 1.4689358724446444,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004791003776734316,
      "loss": 5.8854,
      "mean_token_accuracy": 0.165060593187809,
      "num_tokens": 28747852.0,
      "step": 14695
    },
    {
      "entropy": 6.043006706237793,
      "epoch": 1.4694356975058729,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047908535400496713,
      "loss": 5.7762,
      "mean_token_accuracy": 0.17144134342670442,
      "num_tokens": 28758652.0,
      "step": 14700
    },
    {
      "entropy": 6.016968011856079,
      "epoch": 1.4699355225671016,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047907032520176765,
      "loss": 5.7634,
      "mean_token_accuracy": 0.16589006334543227,
      "num_tokens": 28767603.0,
      "step": 14705
    },
    {
      "entropy": 6.079040670394898,
      "epoch": 1.47043534762833,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004790552912642113,
      "loss": 5.8198,
      "mean_token_accuracy": 0.16492786556482314,
      "num_tokens": 28777984.0,
      "step": 14710
    },
    {
      "entropy": 6.18289794921875,
      "epoch": 1.4709351726895585,
      "grad_norm": 1.25,
      "learning_rate": 0.00047904025219267643,
      "loss": 5.8753,
      "mean_token_accuracy": 0.16434124559164048,
      "num_tokens": 28789033.0,
      "step": 14715
    },
    {
      "entropy": 6.0184930801391605,
      "epoch": 1.4714349977507872,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047902520798754125,
      "loss": 5.7639,
      "mean_token_accuracy": 0.17215773463249207,
      "num_tokens": 28797728.0,
      "step": 14720
    },
    {
      "entropy": 5.951673078536987,
      "epoch": 1.471934822812016,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004790101586491843,
      "loss": 5.7875,
      "mean_token_accuracy": 0.16234516501426696,
      "num_tokens": 28807505.0,
      "step": 14725
    },
    {
      "entropy": 6.141392040252685,
      "epoch": 1.4724346478732444,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004789951041779843,
      "loss": 5.87,
      "mean_token_accuracy": 0.16494024097919463,
      "num_tokens": 28817289.0,
      "step": 14730
    },
    {
      "entropy": 6.101599025726318,
      "epoch": 1.4729344729344729,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047898004457432,
      "loss": 5.8394,
      "mean_token_accuracy": 0.16408341974020005,
      "num_tokens": 28827697.0,
      "step": 14735
    },
    {
      "entropy": 6.167039203643799,
      "epoch": 1.4734342979957016,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004789649798385702,
      "loss": 5.8965,
      "mean_token_accuracy": 0.16139206439256668,
      "num_tokens": 28837427.0,
      "step": 14740
    },
    {
      "entropy": 6.123134279251099,
      "epoch": 1.47393412305693,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047894990997111403,
      "loss": 5.83,
      "mean_token_accuracy": 0.15883630365133286,
      "num_tokens": 28845692.0,
      "step": 14745
    },
    {
      "entropy": 6.0505928039550785,
      "epoch": 1.4744339481181585,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047893483497233057,
      "loss": 5.7577,
      "mean_token_accuracy": 0.16481212377548218,
      "num_tokens": 28854854.0,
      "step": 14750
    },
    {
      "entropy": 6.04986891746521,
      "epoch": 1.4749337731793872,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004789197548425992,
      "loss": 5.8544,
      "mean_token_accuracy": 0.15819251239299775,
      "num_tokens": 28865711.0,
      "step": 14755
    },
    {
      "entropy": 6.027111148834228,
      "epoch": 1.4754335982406157,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004789046695822992,
      "loss": 5.7183,
      "mean_token_accuracy": 0.17759053707122802,
      "num_tokens": 28874193.0,
      "step": 14760
    },
    {
      "entropy": 5.999184942245483,
      "epoch": 1.4759334233018444,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047888957919181026,
      "loss": 5.8089,
      "mean_token_accuracy": 0.16792616695165635,
      "num_tokens": 28883652.0,
      "step": 14765
    },
    {
      "entropy": 6.08264741897583,
      "epoch": 1.4764332483630729,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004788744836715119,
      "loss": 5.8466,
      "mean_token_accuracy": 0.1634826108813286,
      "num_tokens": 28893040.0,
      "step": 14770
    },
    {
      "entropy": 6.142033052444458,
      "epoch": 1.4769330734243016,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004788593830217841,
      "loss": 5.7968,
      "mean_token_accuracy": 0.1674753502011299,
      "num_tokens": 28903330.0,
      "step": 14775
    },
    {
      "entropy": 6.090061044692993,
      "epoch": 1.47743289848553,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047884427724300657,
      "loss": 5.8324,
      "mean_token_accuracy": 0.16506770253181458,
      "num_tokens": 28913032.0,
      "step": 14780
    },
    {
      "entropy": 6.05846791267395,
      "epoch": 1.4779327235467585,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004788291663355595,
      "loss": 5.8115,
      "mean_token_accuracy": 0.16614584773778915,
      "num_tokens": 28921298.0,
      "step": 14785
    },
    {
      "entropy": 6.034140110015869,
      "epoch": 1.4784325486079872,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004788140502998231,
      "loss": 5.762,
      "mean_token_accuracy": 0.16833754777908325,
      "num_tokens": 28930724.0,
      "step": 14790
    },
    {
      "entropy": 6.191229248046875,
      "epoch": 1.4789323736692157,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004787989291361776,
      "loss": 5.9507,
      "mean_token_accuracy": 0.15255059897899628,
      "num_tokens": 28941670.0,
      "step": 14795
    },
    {
      "entropy": 6.02094612121582,
      "epoch": 1.4794321987304444,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004787838028450035,
      "loss": 5.743,
      "mean_token_accuracy": 0.16965675055980683,
      "num_tokens": 28950175.0,
      "step": 14800
    },
    {
      "entropy": 6.06300539970398,
      "epoch": 1.4799320237916729,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047876867142668136,
      "loss": 5.7278,
      "mean_token_accuracy": 0.16908481419086457,
      "num_tokens": 28958410.0,
      "step": 14805
    },
    {
      "entropy": 6.062980794906617,
      "epoch": 1.4804318488529016,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004787535348815918,
      "loss": 5.6774,
      "mean_token_accuracy": 0.17355235666036606,
      "num_tokens": 28967911.0,
      "step": 14810
    },
    {
      "entropy": 5.983389616012573,
      "epoch": 1.48093167391413,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047873839321011587,
      "loss": 5.7873,
      "mean_token_accuracy": 0.17146263569593428,
      "num_tokens": 28978112.0,
      "step": 14815
    },
    {
      "entropy": 6.011178970336914,
      "epoch": 1.4814314989753585,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004787232464126343,
      "loss": 5.7892,
      "mean_token_accuracy": 0.16614456921815873,
      "num_tokens": 28987960.0,
      "step": 14820
    },
    {
      "entropy": 6.074428749084473,
      "epoch": 1.4819313240365872,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004787080944895283,
      "loss": 5.8511,
      "mean_token_accuracy": 0.16128759533166886,
      "num_tokens": 28997000.0,
      "step": 14825
    },
    {
      "entropy": 6.1852795600891115,
      "epoch": 1.4824311490978157,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047869293744117906,
      "loss": 5.8499,
      "mean_token_accuracy": 0.15864676237106323,
      "num_tokens": 29005725.0,
      "step": 14830
    },
    {
      "entropy": 6.013558006286621,
      "epoch": 1.4829309741590444,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047867777526796793,
      "loss": 5.7855,
      "mean_token_accuracy": 0.1734295055270195,
      "num_tokens": 29015168.0,
      "step": 14835
    },
    {
      "entropy": 6.09476752281189,
      "epoch": 1.4834307992202729,
      "grad_norm": 1.125,
      "learning_rate": 0.0004786626079702764,
      "loss": 5.9309,
      "mean_token_accuracy": 0.15995499342679978,
      "num_tokens": 29025170.0,
      "step": 14840
    },
    {
      "entropy": 6.14103627204895,
      "epoch": 1.4839306242815016,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004786474355484861,
      "loss": 5.7905,
      "mean_token_accuracy": 0.1702966123819351,
      "num_tokens": 29034262.0,
      "step": 14845
    },
    {
      "entropy": 6.114272689819336,
      "epoch": 1.48443044934273,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004786322580029786,
      "loss": 5.8401,
      "mean_token_accuracy": 0.16252973228693007,
      "num_tokens": 29043847.0,
      "step": 14850
    },
    {
      "entropy": 6.052633285522461,
      "epoch": 1.4849302744039585,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047861707533413593,
      "loss": 5.8241,
      "mean_token_accuracy": 0.16816114038228988,
      "num_tokens": 29053378.0,
      "step": 14855
    },
    {
      "entropy": 6.094302797317505,
      "epoch": 1.4854300994651872,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004786018875423401,
      "loss": 5.8221,
      "mean_token_accuracy": 0.16711684912443162,
      "num_tokens": 29063335.0,
      "step": 14860
    },
    {
      "entropy": 6.03166127204895,
      "epoch": 1.4859299245264157,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047858669462797317,
      "loss": 5.8064,
      "mean_token_accuracy": 0.17063071429729462,
      "num_tokens": 29073322.0,
      "step": 14865
    },
    {
      "entropy": 6.097191333770752,
      "epoch": 1.4864297495876444,
      "grad_norm": 1.125,
      "learning_rate": 0.00047857149659141736,
      "loss": 5.8324,
      "mean_token_accuracy": 0.1619692862033844,
      "num_tokens": 29082618.0,
      "step": 14870
    },
    {
      "entropy": 6.08108491897583,
      "epoch": 1.4869295746488729,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004785562934330551,
      "loss": 5.6864,
      "mean_token_accuracy": 0.16814938932657242,
      "num_tokens": 29091854.0,
      "step": 14875
    },
    {
      "entropy": 6.030675935745239,
      "epoch": 1.4874293997101016,
      "grad_norm": 1.25,
      "learning_rate": 0.00047854108515326885,
      "loss": 5.7382,
      "mean_token_accuracy": 0.17307704836130142,
      "num_tokens": 29100816.0,
      "step": 14880
    },
    {
      "entropy": 6.071699762344361,
      "epoch": 1.48792922477133,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004785258717524412,
      "loss": 5.8207,
      "mean_token_accuracy": 0.16693780422210694,
      "num_tokens": 29110216.0,
      "step": 14885
    },
    {
      "entropy": 6.119169044494629,
      "epoch": 1.4884290498325585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047851065323095514,
      "loss": 5.8438,
      "mean_token_accuracy": 0.16381413191556932,
      "num_tokens": 29120910.0,
      "step": 14890
    },
    {
      "entropy": 6.108213138580322,
      "epoch": 1.4889288748937872,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004784954295891933,
      "loss": 5.8723,
      "mean_token_accuracy": 0.16313107758760453,
      "num_tokens": 29131163.0,
      "step": 14895
    },
    {
      "entropy": 6.035593032836914,
      "epoch": 1.4894286999550157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047848020082753883,
      "loss": 5.7706,
      "mean_token_accuracy": 0.17240241318941116,
      "num_tokens": 29140549.0,
      "step": 14900
    },
    {
      "entropy": 6.057381820678711,
      "epoch": 1.4899285250162442,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004784649669463749,
      "loss": 5.7369,
      "mean_token_accuracy": 0.16420047730207443,
      "num_tokens": 29149884.0,
      "step": 14905
    },
    {
      "entropy": 6.067973613739014,
      "epoch": 1.4904283500774729,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047844972794608474,
      "loss": 5.9068,
      "mean_token_accuracy": 0.16146813333034515,
      "num_tokens": 29158913.0,
      "step": 14910
    },
    {
      "entropy": 6.113250637054444,
      "epoch": 1.4909281751387016,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047843448382705185,
      "loss": 5.8457,
      "mean_token_accuracy": 0.1574658840894699,
      "num_tokens": 29167548.0,
      "step": 14915
    },
    {
      "entropy": 6.061174058914185,
      "epoch": 1.49142800019993,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047841923458965966,
      "loss": 5.8187,
      "mean_token_accuracy": 0.1668119326233864,
      "num_tokens": 29177155.0,
      "step": 14920
    },
    {
      "entropy": 6.059987735748291,
      "epoch": 1.4919278252611585,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047840398023429176,
      "loss": 5.845,
      "mean_token_accuracy": 0.16502925902605056,
      "num_tokens": 29185920.0,
      "step": 14925
    },
    {
      "entropy": 6.157622194290161,
      "epoch": 1.4924276503223872,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047838872076133213,
      "loss": 5.8263,
      "mean_token_accuracy": 0.164167882502079,
      "num_tokens": 29195707.0,
      "step": 14930
    },
    {
      "entropy": 6.1057202339172365,
      "epoch": 1.4929274753836157,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004783734561711647,
      "loss": 5.8782,
      "mean_token_accuracy": 0.1587187245488167,
      "num_tokens": 29206395.0,
      "step": 14935
    },
    {
      "entropy": 6.052242231369019,
      "epoch": 1.4934273004448442,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047835818646417333,
      "loss": 5.8711,
      "mean_token_accuracy": 0.16700909435749053,
      "num_tokens": 29215903.0,
      "step": 14940
    },
    {
      "entropy": 6.114767217636109,
      "epoch": 1.4939271255060729,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004783429116407423,
      "loss": 5.8852,
      "mean_token_accuracy": 0.16300320327281953,
      "num_tokens": 29225768.0,
      "step": 14945
    },
    {
      "entropy": 6.123958444595337,
      "epoch": 1.4944269505673016,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047832763170125605,
      "loss": 5.8556,
      "mean_token_accuracy": 0.15959609746932985,
      "num_tokens": 29235612.0,
      "step": 14950
    },
    {
      "entropy": 6.057261848449707,
      "epoch": 1.49492677562853,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047831234664609887,
      "loss": 5.7355,
      "mean_token_accuracy": 0.16945449262857437,
      "num_tokens": 29244556.0,
      "step": 14955
    },
    {
      "entropy": 6.0706713676452635,
      "epoch": 1.4954266006897585,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047829705647565535,
      "loss": 5.9359,
      "mean_token_accuracy": 0.1581117555499077,
      "num_tokens": 29254978.0,
      "step": 14960
    },
    {
      "entropy": 6.006599712371826,
      "epoch": 1.4959264257509872,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004782817611903102,
      "loss": 5.7283,
      "mean_token_accuracy": 0.16540569365024566,
      "num_tokens": 29263339.0,
      "step": 14965
    },
    {
      "entropy": 6.041868686676025,
      "epoch": 1.4964262508122157,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047826646079044826,
      "loss": 5.6895,
      "mean_token_accuracy": 0.16997961848974227,
      "num_tokens": 29272707.0,
      "step": 14970
    },
    {
      "entropy": 6.043829011917114,
      "epoch": 1.4969260758734442,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047825115527645447,
      "loss": 5.8248,
      "mean_token_accuracy": 0.16586797684431076,
      "num_tokens": 29282292.0,
      "step": 14975
    },
    {
      "entropy": 6.074822473526001,
      "epoch": 1.4974259009346729,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004782358446487139,
      "loss": 5.9126,
      "mean_token_accuracy": 0.17390037477016448,
      "num_tokens": 29291289.0,
      "step": 14980
    },
    {
      "entropy": 6.03560152053833,
      "epoch": 1.4979257259959016,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047822052890761173,
      "loss": 5.7177,
      "mean_token_accuracy": 0.168414843082428,
      "num_tokens": 29300166.0,
      "step": 14985
    },
    {
      "entropy": 5.980249261856079,
      "epoch": 1.49842555105713,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004782052080535334,
      "loss": 5.7406,
      "mean_token_accuracy": 0.16387926787137985,
      "num_tokens": 29308422.0,
      "step": 14990
    },
    {
      "entropy": 6.106278276443481,
      "epoch": 1.4989253761183585,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004781898820868644,
      "loss": 5.8723,
      "mean_token_accuracy": 0.16848377883434296,
      "num_tokens": 29317699.0,
      "step": 14995
    },
    {
      "entropy": 6.142735528945923,
      "epoch": 1.4994252011795872,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004781745510079901,
      "loss": 5.8839,
      "mean_token_accuracy": 0.15503703206777572,
      "num_tokens": 29327711.0,
      "step": 15000
    },
    {
      "epoch": 1.4994252011795872,
      "eval_entropy": 5.952691121680071,
      "eval_loss": 5.930527210235596,
      "eval_mean_token_accuracy": 0.16697731941911523,
      "eval_num_tokens": 29327711.0,
      "eval_runtime": 24.0106,
      "eval_samples_per_second": 1292.971,
      "eval_steps_per_second": 161.637,
      "step": 15000
    },
    {
      "entropy": 6.024403715133667,
      "epoch": 1.4999250262408157,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047815921481729647,
      "loss": 5.7482,
      "mean_token_accuracy": 0.16747903376817702,
      "num_tokens": 29336540.0,
      "step": 15005
    },
    {
      "entropy": 6.093067359924317,
      "epoch": 1.5004248513020442,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004781438735151693,
      "loss": 5.973,
      "mean_token_accuracy": 0.15314984768629075,
      "num_tokens": 29345512.0,
      "step": 15010
    },
    {
      "entropy": 6.116947889328003,
      "epoch": 1.5009246763632729,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047812852710199446,
      "loss": 5.93,
      "mean_token_accuracy": 0.1548838958144188,
      "num_tokens": 29355726.0,
      "step": 15015
    },
    {
      "entropy": 6.117451238632202,
      "epoch": 1.5014245014245016,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004781131755781582,
      "loss": 5.7688,
      "mean_token_accuracy": 0.1708882048726082,
      "num_tokens": 29366570.0,
      "step": 15020
    },
    {
      "entropy": 6.088492965698242,
      "epoch": 1.50192432648573,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004780978189440467,
      "loss": 5.8899,
      "mean_token_accuracy": 0.16271362602710723,
      "num_tokens": 29376356.0,
      "step": 15025
    },
    {
      "entropy": 6.176832056045532,
      "epoch": 1.5024241515469585,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047808245720004633,
      "loss": 5.8883,
      "mean_token_accuracy": 0.15761558562517167,
      "num_tokens": 29385246.0,
      "step": 15030
    },
    {
      "entropy": 6.073028326034546,
      "epoch": 1.5029239766081872,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047806709034654355,
      "loss": 5.7407,
      "mean_token_accuracy": 0.170181705057621,
      "num_tokens": 29395352.0,
      "step": 15035
    },
    {
      "entropy": 6.101519584655762,
      "epoch": 1.5034238016694157,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004780517183839251,
      "loss": 5.8705,
      "mean_token_accuracy": 0.15990639477968216,
      "num_tokens": 29404771.0,
      "step": 15040
    },
    {
      "entropy": 6.054186820983887,
      "epoch": 1.5039236267306442,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004780363413125777,
      "loss": 5.776,
      "mean_token_accuracy": 0.1657913774251938,
      "num_tokens": 29413378.0,
      "step": 15045
    },
    {
      "entropy": 6.114541053771973,
      "epoch": 1.5044234517918729,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047802095913288814,
      "loss": 5.8993,
      "mean_token_accuracy": 0.16488889753818511,
      "num_tokens": 29424125.0,
      "step": 15050
    },
    {
      "entropy": 6.0674902439117435,
      "epoch": 1.5049232768531016,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047800557184524345,
      "loss": 5.8357,
      "mean_token_accuracy": 0.15939660370349884,
      "num_tokens": 29435411.0,
      "step": 15055
    },
    {
      "entropy": 6.076913452148437,
      "epoch": 1.5054231019143298,
      "grad_norm": 1.125,
      "learning_rate": 0.0004779901794500308,
      "loss": 5.8565,
      "mean_token_accuracy": 0.16051626205444336,
      "num_tokens": 29445515.0,
      "step": 15060
    },
    {
      "entropy": 6.176061916351318,
      "epoch": 1.5059229269755585,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047797478194763754,
      "loss": 5.8574,
      "mean_token_accuracy": 0.16085462272167206,
      "num_tokens": 29454864.0,
      "step": 15065
    },
    {
      "entropy": 6.136804056167603,
      "epoch": 1.5064227520367872,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047795937933845103,
      "loss": 5.894,
      "mean_token_accuracy": 0.15710541382431983,
      "num_tokens": 29464443.0,
      "step": 15070
    },
    {
      "entropy": 6.139208555221558,
      "epoch": 1.5069225770980157,
      "grad_norm": 1.25,
      "learning_rate": 0.0004779439716228587,
      "loss": 5.8878,
      "mean_token_accuracy": 0.15734200328588485,
      "num_tokens": 29474954.0,
      "step": 15075
    },
    {
      "entropy": 6.146669912338257,
      "epoch": 1.5074224021592442,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004779285588012482,
      "loss": 5.8547,
      "mean_token_accuracy": 0.15914391577243805,
      "num_tokens": 29484647.0,
      "step": 15080
    },
    {
      "entropy": 6.122530317306518,
      "epoch": 1.5079222272204729,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047791314087400745,
      "loss": 5.859,
      "mean_token_accuracy": 0.1646413743495941,
      "num_tokens": 29494782.0,
      "step": 15085
    },
    {
      "entropy": 6.0605841159820555,
      "epoch": 1.5084220522817016,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004778977178415243,
      "loss": 5.8066,
      "mean_token_accuracy": 0.16665096282958985,
      "num_tokens": 29505311.0,
      "step": 15090
    },
    {
      "entropy": 6.043413066864014,
      "epoch": 1.5089218773429298,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004778822897041868,
      "loss": 5.7887,
      "mean_token_accuracy": 0.1602967068552971,
      "num_tokens": 29515192.0,
      "step": 15095
    },
    {
      "entropy": 6.141283369064331,
      "epoch": 1.5094217024041585,
      "grad_norm": 1.3125,
      "learning_rate": 0.000477866856462383,
      "loss": 5.8343,
      "mean_token_accuracy": 0.1636444039642811,
      "num_tokens": 29524429.0,
      "step": 15100
    },
    {
      "entropy": 6.128295755386352,
      "epoch": 1.5099215274653872,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004778514181165014,
      "loss": 5.7867,
      "mean_token_accuracy": 0.1705568552017212,
      "num_tokens": 29532926.0,
      "step": 15105
    },
    {
      "entropy": 6.1278472423553465,
      "epoch": 1.5104213525266157,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047783597466693025,
      "loss": 5.8711,
      "mean_token_accuracy": 0.1631425589323044,
      "num_tokens": 29542588.0,
      "step": 15110
    },
    {
      "entropy": 6.076506042480469,
      "epoch": 1.5109211775878442,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004778205261140582,
      "loss": 5.8055,
      "mean_token_accuracy": 0.16449604332447051,
      "num_tokens": 29551710.0,
      "step": 15115
    },
    {
      "entropy": 6.074311590194702,
      "epoch": 1.5114210026490729,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004778050724582739,
      "loss": 5.891,
      "mean_token_accuracy": 0.16317658871412277,
      "num_tokens": 29561357.0,
      "step": 15120
    },
    {
      "entropy": 6.08810248374939,
      "epoch": 1.5119208277103013,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047778961369996623,
      "loss": 5.7995,
      "mean_token_accuracy": 0.16664882749319077,
      "num_tokens": 29570060.0,
      "step": 15125
    },
    {
      "entropy": 6.134357118606568,
      "epoch": 1.5124206527715298,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000477774149839524,
      "loss": 5.8389,
      "mean_token_accuracy": 0.16309096664190292,
      "num_tokens": 29580097.0,
      "step": 15130
    },
    {
      "entropy": 6.188675737380981,
      "epoch": 1.5129204778327585,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047775868087733634,
      "loss": 6.0162,
      "mean_token_accuracy": 0.14989048093557358,
      "num_tokens": 29590290.0,
      "step": 15135
    },
    {
      "entropy": 6.09381422996521,
      "epoch": 1.5134203028939872,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047774320681379237,
      "loss": 5.8114,
      "mean_token_accuracy": 0.16330965757369995,
      "num_tokens": 29598839.0,
      "step": 15140
    },
    {
      "entropy": 6.163496732711792,
      "epoch": 1.5139201279552157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004777277276492816,
      "loss": 5.8767,
      "mean_token_accuracy": 0.16937931925058364,
      "num_tokens": 29609550.0,
      "step": 15145
    },
    {
      "entropy": 6.1531449317932125,
      "epoch": 1.5144199530164442,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047771224338419326,
      "loss": 5.9745,
      "mean_token_accuracy": 0.14925648421049117,
      "num_tokens": 29619786.0,
      "step": 15150
    },
    {
      "entropy": 6.048215007781982,
      "epoch": 1.5149197780776729,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004776967540189171,
      "loss": 5.7662,
      "mean_token_accuracy": 0.16797539442777634,
      "num_tokens": 29630492.0,
      "step": 15155
    },
    {
      "entropy": 6.162397146224976,
      "epoch": 1.5154196031389013,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004776812595538428,
      "loss": 5.8541,
      "mean_token_accuracy": 0.15777112990617753,
      "num_tokens": 29641965.0,
      "step": 15160
    },
    {
      "entropy": 6.134107780456543,
      "epoch": 1.5159194282001298,
      "grad_norm": 1.125,
      "learning_rate": 0.00047766575998936013,
      "loss": 5.8899,
      "mean_token_accuracy": 0.16055671721696854,
      "num_tokens": 29651692.0,
      "step": 15165
    },
    {
      "entropy": 6.108399152755737,
      "epoch": 1.5164192532613585,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047765025532585907,
      "loss": 5.8525,
      "mean_token_accuracy": 0.16720761805772782,
      "num_tokens": 29660541.0,
      "step": 15170
    },
    {
      "entropy": 6.03242712020874,
      "epoch": 1.5169190783225872,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004776347455637298,
      "loss": 5.8384,
      "mean_token_accuracy": 0.1635917142033577,
      "num_tokens": 29670498.0,
      "step": 15175
    },
    {
      "entropy": 6.016808700561524,
      "epoch": 1.5174189033838157,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004776192307033624,
      "loss": 5.7082,
      "mean_token_accuracy": 0.17321868538856505,
      "num_tokens": 29678898.0,
      "step": 15180
    },
    {
      "entropy": 6.127698564529419,
      "epoch": 1.5179187284450442,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004776037107451473,
      "loss": 5.8714,
      "mean_token_accuracy": 0.1614261820912361,
      "num_tokens": 29689723.0,
      "step": 15185
    },
    {
      "entropy": 6.0613795757293705,
      "epoch": 1.5184185535062729,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000477588185689475,
      "loss": 5.7724,
      "mean_token_accuracy": 0.17052101641893386,
      "num_tokens": 29699246.0,
      "step": 15190
    },
    {
      "entropy": 6.009236669540405,
      "epoch": 1.5189183785675013,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047757265553673606,
      "loss": 5.7358,
      "mean_token_accuracy": 0.17132902890443802,
      "num_tokens": 29709125.0,
      "step": 15195
    },
    {
      "entropy": 6.04731969833374,
      "epoch": 1.5194182036287298,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047755712028732126,
      "loss": 5.7676,
      "mean_token_accuracy": 0.16394769847393037,
      "num_tokens": 29718953.0,
      "step": 15200
    },
    {
      "entropy": 6.103776121139527,
      "epoch": 1.5199180286899585,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004775415799416214,
      "loss": 5.8231,
      "mean_token_accuracy": 0.15949769914150239,
      "num_tokens": 29728764.0,
      "step": 15205
    },
    {
      "entropy": 6.031689405441284,
      "epoch": 1.5204178537511872,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004775260345000275,
      "loss": 5.842,
      "mean_token_accuracy": 0.15985133945941926,
      "num_tokens": 29738033.0,
      "step": 15210
    },
    {
      "entropy": 5.974576616287232,
      "epoch": 1.5209176788124157,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047751048396293077,
      "loss": 5.7511,
      "mean_token_accuracy": 0.17815377488732337,
      "num_tokens": 29748450.0,
      "step": 15215
    },
    {
      "entropy": 6.124851417541504,
      "epoch": 1.5214175038736442,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047749492833072226,
      "loss": 5.8931,
      "mean_token_accuracy": 0.15868438631296158,
      "num_tokens": 29757984.0,
      "step": 15220
    },
    {
      "entropy": 6.085717487335205,
      "epoch": 1.5219173289348729,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047747936760379354,
      "loss": 5.8519,
      "mean_token_accuracy": 0.15937797874212264,
      "num_tokens": 29768200.0,
      "step": 15225
    },
    {
      "entropy": 6.02923583984375,
      "epoch": 1.5224171539961013,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000477463801782536,
      "loss": 5.8033,
      "mean_token_accuracy": 0.16860561966896057,
      "num_tokens": 29778242.0,
      "step": 15230
    },
    {
      "entropy": 6.175228691101074,
      "epoch": 1.5229169790573298,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047744823086734136,
      "loss": 5.9409,
      "mean_token_accuracy": 0.15405061691999436,
      "num_tokens": 29787636.0,
      "step": 15235
    },
    {
      "entropy": 6.13984899520874,
      "epoch": 1.5234168041185585,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047743265485860123,
      "loss": 5.7714,
      "mean_token_accuracy": 0.16734917014837264,
      "num_tokens": 29796537.0,
      "step": 15240
    },
    {
      "entropy": 6.047903776168823,
      "epoch": 1.5239166291797872,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047741707375670766,
      "loss": 5.8189,
      "mean_token_accuracy": 0.16314048916101456,
      "num_tokens": 29805495.0,
      "step": 15245
    },
    {
      "entropy": 6.045965909957886,
      "epoch": 1.5244164542410157,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004774014875620526,
      "loss": 5.7431,
      "mean_token_accuracy": 0.16881490349769593,
      "num_tokens": 29815193.0,
      "step": 15250
    },
    {
      "entropy": 6.086957311630249,
      "epoch": 1.5249162793022442,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047738589627502814,
      "loss": 5.835,
      "mean_token_accuracy": 0.1658613622188568,
      "num_tokens": 29825004.0,
      "step": 15255
    },
    {
      "entropy": 6.057898998260498,
      "epoch": 1.5254161043634729,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047737029989602665,
      "loss": 5.865,
      "mean_token_accuracy": 0.16699073761701583,
      "num_tokens": 29834507.0,
      "step": 15260
    },
    {
      "entropy": 6.00806679725647,
      "epoch": 1.5259159294247013,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004773546984254405,
      "loss": 5.7158,
      "mean_token_accuracy": 0.16878599524497986,
      "num_tokens": 29844397.0,
      "step": 15265
    },
    {
      "entropy": 6.113274812698364,
      "epoch": 1.5264157544859298,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004773390918636622,
      "loss": 5.8327,
      "mean_token_accuracy": 0.16741204857826233,
      "num_tokens": 29855702.0,
      "step": 15270
    },
    {
      "entropy": 6.105196475982666,
      "epoch": 1.5269155795471585,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004773234802110844,
      "loss": 5.7397,
      "mean_token_accuracy": 0.17111674845218658,
      "num_tokens": 29865806.0,
      "step": 15275
    },
    {
      "entropy": 5.995190811157227,
      "epoch": 1.5274154046083872,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004773078634681,
      "loss": 5.7412,
      "mean_token_accuracy": 0.17543636560440062,
      "num_tokens": 29874827.0,
      "step": 15280
    },
    {
      "entropy": 5.96733980178833,
      "epoch": 1.5279152296696157,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047729224163510167,
      "loss": 5.8197,
      "mean_token_accuracy": 0.16055009365081788,
      "num_tokens": 29883517.0,
      "step": 15285
    },
    {
      "entropy": 5.9974609375,
      "epoch": 1.5284150547308442,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047727661471248263,
      "loss": 5.7472,
      "mean_token_accuracy": 0.16385783180594443,
      "num_tokens": 29893303.0,
      "step": 15290
    },
    {
      "entropy": 6.154442071914673,
      "epoch": 1.5289148797920729,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000477260982700636,
      "loss": 5.9679,
      "mean_token_accuracy": 0.15659339651465415,
      "num_tokens": 29904719.0,
      "step": 15295
    },
    {
      "entropy": 6.168337774276734,
      "epoch": 1.5294147048533013,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004772453455999552,
      "loss": 5.8473,
      "mean_token_accuracy": 0.15929200649261474,
      "num_tokens": 29914852.0,
      "step": 15300
    },
    {
      "entropy": 6.035875701904297,
      "epoch": 1.5299145299145298,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004772297034108334,
      "loss": 5.805,
      "mean_token_accuracy": 0.16422408819198608,
      "num_tokens": 29924479.0,
      "step": 15305
    },
    {
      "entropy": 6.066788625717163,
      "epoch": 1.5304143549757585,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047721405613366434,
      "loss": 5.8301,
      "mean_token_accuracy": 0.16059157848358155,
      "num_tokens": 29933162.0,
      "step": 15310
    },
    {
      "entropy": 6.000638246536255,
      "epoch": 1.5309141800369872,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047719840376884173,
      "loss": 5.7552,
      "mean_token_accuracy": 0.1683478832244873,
      "num_tokens": 29942468.0,
      "step": 15315
    },
    {
      "entropy": 6.080678081512451,
      "epoch": 1.5314140050982155,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004771827463167593,
      "loss": 5.8518,
      "mean_token_accuracy": 0.16608970314264299,
      "num_tokens": 29951716.0,
      "step": 15320
    },
    {
      "entropy": 6.086277103424072,
      "epoch": 1.5319138301594442,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004771670837778109,
      "loss": 5.922,
      "mean_token_accuracy": 0.15985568165779113,
      "num_tokens": 29960233.0,
      "step": 15325
    },
    {
      "entropy": 6.121091508865357,
      "epoch": 1.5324136552206729,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047715141615239075,
      "loss": 5.8483,
      "mean_token_accuracy": 0.16505551636219024,
      "num_tokens": 29968979.0,
      "step": 15330
    },
    {
      "entropy": 6.077244091033935,
      "epoch": 1.5329134802819013,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047713574344089305,
      "loss": 5.7813,
      "mean_token_accuracy": 0.1632857456803322,
      "num_tokens": 29978830.0,
      "step": 15335
    },
    {
      "entropy": 6.092856502532959,
      "epoch": 1.5334133053431298,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000477120065643712,
      "loss": 5.8099,
      "mean_token_accuracy": 0.16594863981008529,
      "num_tokens": 29987708.0,
      "step": 15340
    },
    {
      "entropy": 6.150106716156006,
      "epoch": 1.5339131304043585,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004771043827612421,
      "loss": 5.8264,
      "mean_token_accuracy": 0.16569228023290633,
      "num_tokens": 29997351.0,
      "step": 15345
    },
    {
      "entropy": 6.059939432144165,
      "epoch": 1.5344129554655872,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047708869479387795,
      "loss": 5.8319,
      "mean_token_accuracy": 0.16368046179413795,
      "num_tokens": 30006412.0,
      "step": 15350
    },
    {
      "entropy": 6.070898103713989,
      "epoch": 1.5349127805268155,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004770730017420143,
      "loss": 5.8203,
      "mean_token_accuracy": 0.15753526687622071,
      "num_tokens": 30016629.0,
      "step": 15355
    },
    {
      "entropy": 6.043881845474243,
      "epoch": 1.5354126055880442,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004770573036060458,
      "loss": 5.6825,
      "mean_token_accuracy": 0.17183737009763717,
      "num_tokens": 30026395.0,
      "step": 15360
    },
    {
      "entropy": 6.080069255828858,
      "epoch": 1.5359124306492729,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004770416003863676,
      "loss": 5.8696,
      "mean_token_accuracy": 0.1627057522535324,
      "num_tokens": 30037178.0,
      "step": 15365
    },
    {
      "entropy": 6.05632963180542,
      "epoch": 1.5364122557105013,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004770258920833748,
      "loss": 5.7501,
      "mean_token_accuracy": 0.16758398413658143,
      "num_tokens": 30046335.0,
      "step": 15370
    },
    {
      "entropy": 6.078595495223999,
      "epoch": 1.5369120807717298,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047701017869746236,
      "loss": 5.819,
      "mean_token_accuracy": 0.1656502291560173,
      "num_tokens": 30056574.0,
      "step": 15375
    },
    {
      "entropy": 6.015065097808838,
      "epoch": 1.5374119058329585,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047699446022902596,
      "loss": 5.7322,
      "mean_token_accuracy": 0.1704367682337761,
      "num_tokens": 30067048.0,
      "step": 15380
    },
    {
      "entropy": 6.001787996292114,
      "epoch": 1.537911730894187,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047697873667846094,
      "loss": 5.7632,
      "mean_token_accuracy": 0.1628413528203964,
      "num_tokens": 30077177.0,
      "step": 15385
    },
    {
      "entropy": 6.109357929229736,
      "epoch": 1.5384115559554155,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047696300804616276,
      "loss": 5.8725,
      "mean_token_accuracy": 0.15948958247900008,
      "num_tokens": 30086348.0,
      "step": 15390
    },
    {
      "entropy": 6.060103034973144,
      "epoch": 1.5389113810166442,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004769472743325273,
      "loss": 5.7634,
      "mean_token_accuracy": 0.1669648602604866,
      "num_tokens": 30096721.0,
      "step": 15395
    },
    {
      "entropy": 6.011430740356445,
      "epoch": 1.5394112060778729,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047693153553795046,
      "loss": 5.7626,
      "mean_token_accuracy": 0.16805630326271057,
      "num_tokens": 30106797.0,
      "step": 15400
    },
    {
      "entropy": 6.03499927520752,
      "epoch": 1.5399110311391013,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004769157916628281,
      "loss": 5.7953,
      "mean_token_accuracy": 0.1597955420613289,
      "num_tokens": 30116643.0,
      "step": 15405
    },
    {
      "entropy": 6.058139705657959,
      "epoch": 1.5404108562003298,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047690004270755634,
      "loss": 5.7525,
      "mean_token_accuracy": 0.1702285423874855,
      "num_tokens": 30125336.0,
      "step": 15410
    },
    {
      "entropy": 5.949384689331055,
      "epoch": 1.5409106812615585,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004768842886725314,
      "loss": 5.763,
      "mean_token_accuracy": 0.17791938930749893,
      "num_tokens": 30135238.0,
      "step": 15415
    },
    {
      "entropy": 6.142117309570312,
      "epoch": 1.541410506322787,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047686852955814986,
      "loss": 5.7927,
      "mean_token_accuracy": 0.16054288893938065,
      "num_tokens": 30144469.0,
      "step": 15420
    },
    {
      "entropy": 6.098428201675415,
      "epoch": 1.5419103313840155,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000476852765364808,
      "loss": 5.8081,
      "mean_token_accuracy": 0.16158611625432967,
      "num_tokens": 30154484.0,
      "step": 15425
    },
    {
      "entropy": 6.023132848739624,
      "epoch": 1.5424101564452442,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004768369960929025,
      "loss": 5.809,
      "mean_token_accuracy": 0.16695237159729004,
      "num_tokens": 30163832.0,
      "step": 15430
    },
    {
      "entropy": 5.997958755493164,
      "epoch": 1.5429099815064728,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004768212217428301,
      "loss": 5.7357,
      "mean_token_accuracy": 0.16630097925662995,
      "num_tokens": 30173503.0,
      "step": 15435
    },
    {
      "entropy": 6.09139609336853,
      "epoch": 1.5434098065677013,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047680544231498766,
      "loss": 5.8388,
      "mean_token_accuracy": 0.16930627077817917,
      "num_tokens": 30183476.0,
      "step": 15440
    },
    {
      "entropy": 6.109343671798706,
      "epoch": 1.5439096316289298,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004767896578097723,
      "loss": 5.8304,
      "mean_token_accuracy": 0.16964736729860305,
      "num_tokens": 30192552.0,
      "step": 15445
    },
    {
      "entropy": 6.144826936721802,
      "epoch": 1.5444094566901585,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004767738682275811,
      "loss": 5.8793,
      "mean_token_accuracy": 0.15962909758090973,
      "num_tokens": 30202796.0,
      "step": 15450
    },
    {
      "entropy": 6.0616480827331545,
      "epoch": 1.544909281751387,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004767580735688112,
      "loss": 5.8017,
      "mean_token_accuracy": 0.16594992578029633,
      "num_tokens": 30211878.0,
      "step": 15455
    },
    {
      "entropy": 6.077480411529541,
      "epoch": 1.5454091068126155,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004767422738338601,
      "loss": 5.8427,
      "mean_token_accuracy": 0.16039080992341043,
      "num_tokens": 30221572.0,
      "step": 15460
    },
    {
      "entropy": 6.083427238464355,
      "epoch": 1.5459089318738441,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004767264690231253,
      "loss": 5.7819,
      "mean_token_accuracy": 0.16881288290023805,
      "num_tokens": 30230770.0,
      "step": 15465
    },
    {
      "entropy": 6.07643756866455,
      "epoch": 1.5464087569350728,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004767106591370045,
      "loss": 5.7496,
      "mean_token_accuracy": 0.17124352604150772,
      "num_tokens": 30241404.0,
      "step": 15470
    },
    {
      "entropy": 6.079020023345947,
      "epoch": 1.5469085819963013,
      "grad_norm": 1.125,
      "learning_rate": 0.00047669484417589537,
      "loss": 5.8705,
      "mean_token_accuracy": 0.16001143157482148,
      "num_tokens": 30251471.0,
      "step": 15475
    },
    {
      "entropy": 6.106610488891602,
      "epoch": 1.5474084070575298,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047667902414019585,
      "loss": 5.8357,
      "mean_token_accuracy": 0.1609158605337143,
      "num_tokens": 30260962.0,
      "step": 15480
    },
    {
      "entropy": 6.13154034614563,
      "epoch": 1.5479082321187585,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000476663199030304,
      "loss": 5.8552,
      "mean_token_accuracy": 0.16160022616386413,
      "num_tokens": 30271959.0,
      "step": 15485
    },
    {
      "entropy": 6.015404748916626,
      "epoch": 1.548408057179987,
      "grad_norm": 1.125,
      "learning_rate": 0.00047664736884661804,
      "loss": 5.7402,
      "mean_token_accuracy": 0.16993009597063063,
      "num_tokens": 30281134.0,
      "step": 15490
    },
    {
      "entropy": 5.983353519439698,
      "epoch": 1.5489078822412155,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047663153358953604,
      "loss": 5.7206,
      "mean_token_accuracy": 0.17315862625837325,
      "num_tokens": 30289989.0,
      "step": 15495
    },
    {
      "entropy": 6.11238431930542,
      "epoch": 1.5494077073024441,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004766156932594566,
      "loss": 5.8537,
      "mean_token_accuracy": 0.1621530085802078,
      "num_tokens": 30300416.0,
      "step": 15500
    },
    {
      "entropy": 6.132857370376587,
      "epoch": 1.5499075323636728,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047659984785677824,
      "loss": 5.8099,
      "mean_token_accuracy": 0.16617406606674195,
      "num_tokens": 30310056.0,
      "step": 15505
    },
    {
      "entropy": 6.097787714004516,
      "epoch": 1.5504073574249013,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047658399738189955,
      "loss": 5.8445,
      "mean_token_accuracy": 0.16412217319011688,
      "num_tokens": 30319489.0,
      "step": 15510
    },
    {
      "entropy": 6.10493836402893,
      "epoch": 1.5509071824861298,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047656814183521937,
      "loss": 5.9051,
      "mean_token_accuracy": 0.15890609622001647,
      "num_tokens": 30328814.0,
      "step": 15515
    },
    {
      "entropy": 6.0553563117980955,
      "epoch": 1.5514070075473585,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004765522812171366,
      "loss": 5.7833,
      "mean_token_accuracy": 0.16718293428421022,
      "num_tokens": 30339219.0,
      "step": 15520
    },
    {
      "entropy": 6.001720523834228,
      "epoch": 1.551906832608587,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047653641552805023,
      "loss": 5.7929,
      "mean_token_accuracy": 0.16303002685308457,
      "num_tokens": 30349553.0,
      "step": 15525
    },
    {
      "entropy": 6.113082408905029,
      "epoch": 1.5524066576698154,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004765205447683595,
      "loss": 5.8588,
      "mean_token_accuracy": 0.15497085303068162,
      "num_tokens": 30359782.0,
      "step": 15530
    },
    {
      "entropy": 6.091501522064209,
      "epoch": 1.5529064827310441,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004765046689384638,
      "loss": 5.7846,
      "mean_token_accuracy": 0.16711372137069702,
      "num_tokens": 30369635.0,
      "step": 15535
    },
    {
      "entropy": 6.006160831451416,
      "epoch": 1.5534063077922728,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004764887880387625,
      "loss": 5.7311,
      "mean_token_accuracy": 0.17074052095413209,
      "num_tokens": 30379453.0,
      "step": 15540
    },
    {
      "entropy": 6.102718448638916,
      "epoch": 1.5539061328535013,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047647290206965506,
      "loss": 5.8065,
      "mean_token_accuracy": 0.16007761806249618,
      "num_tokens": 30388989.0,
      "step": 15545
    },
    {
      "entropy": 6.099312877655029,
      "epoch": 1.5544059579147298,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047645701103154125,
      "loss": 5.8348,
      "mean_token_accuracy": 0.16998312771320342,
      "num_tokens": 30399401.0,
      "step": 15550
    },
    {
      "entropy": 6.1167608261108395,
      "epoch": 1.5549057829759585,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047644111492482094,
      "loss": 5.87,
      "mean_token_accuracy": 0.15713770762085916,
      "num_tokens": 30409795.0,
      "step": 15555
    },
    {
      "entropy": 6.114370822906494,
      "epoch": 1.555405608037187,
      "grad_norm": 1.1875,
      "learning_rate": 0.000476425213749894,
      "loss": 5.9035,
      "mean_token_accuracy": 0.15847662836313248,
      "num_tokens": 30419243.0,
      "step": 15560
    },
    {
      "entropy": 6.068156576156616,
      "epoch": 1.5559054330984154,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004764093075071605,
      "loss": 5.857,
      "mean_token_accuracy": 0.16599513292312623,
      "num_tokens": 30429278.0,
      "step": 15565
    },
    {
      "entropy": 6.004425191879273,
      "epoch": 1.5564052581596441,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004763933961970206,
      "loss": 5.8165,
      "mean_token_accuracy": 0.1713077962398529,
      "num_tokens": 30438442.0,
      "step": 15570
    },
    {
      "entropy": 6.055127239227295,
      "epoch": 1.5569050832208728,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004763774798198748,
      "loss": 5.7954,
      "mean_token_accuracy": 0.16136319935321808,
      "num_tokens": 30448779.0,
      "step": 15575
    },
    {
      "entropy": 6.083128166198731,
      "epoch": 1.5574049082821013,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004763615583761234,
      "loss": 5.779,
      "mean_token_accuracy": 0.16810402274131775,
      "num_tokens": 30458219.0,
      "step": 15580
    },
    {
      "entropy": 6.0836718559265135,
      "epoch": 1.5579047333433298,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004763456318661669,
      "loss": 5.8155,
      "mean_token_accuracy": 0.16702513098716737,
      "num_tokens": 30467772.0,
      "step": 15585
    },
    {
      "entropy": 6.0435584545135494,
      "epoch": 1.5584045584045585,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004763297002904062,
      "loss": 5.7752,
      "mean_token_accuracy": 0.16313598155975342,
      "num_tokens": 30477025.0,
      "step": 15590
    },
    {
      "entropy": 6.08624415397644,
      "epoch": 1.558904383465787,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047631376364924195,
      "loss": 5.8522,
      "mean_token_accuracy": 0.16718666553497313,
      "num_tokens": 30486691.0,
      "step": 15595
    },
    {
      "entropy": 6.045085716247558,
      "epoch": 1.5594042085270154,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004762978219430753,
      "loss": 5.7778,
      "mean_token_accuracy": 0.15837837308645247,
      "num_tokens": 30497138.0,
      "step": 15600
    },
    {
      "entropy": 6.1284572124481205,
      "epoch": 1.5599040335882441,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004762818751723072,
      "loss": 5.9062,
      "mean_token_accuracy": 0.15814828872680664,
      "num_tokens": 30507526.0,
      "step": 15605
    },
    {
      "entropy": 6.072921085357666,
      "epoch": 1.5604038586494728,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047626592333733895,
      "loss": 5.747,
      "mean_token_accuracy": 0.1643367275595665,
      "num_tokens": 30517823.0,
      "step": 15610
    },
    {
      "entropy": 6.030939435958862,
      "epoch": 1.560903683710701,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004762499664385719,
      "loss": 5.7683,
      "mean_token_accuracy": 0.16682183593511582,
      "num_tokens": 30527892.0,
      "step": 15615
    },
    {
      "entropy": 6.0847431182861325,
      "epoch": 1.5614035087719298,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047623400447640733,
      "loss": 5.8019,
      "mean_token_accuracy": 0.1685531735420227,
      "num_tokens": 30537329.0,
      "step": 15620
    },
    {
      "entropy": 6.057531929016113,
      "epoch": 1.5619033338331585,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047621803745124717,
      "loss": 5.7916,
      "mean_token_accuracy": 0.16621876806020736,
      "num_tokens": 30546406.0,
      "step": 15625
    },
    {
      "entropy": 6.095958948135376,
      "epoch": 1.562403158894387,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047620206536349294,
      "loss": 5.8672,
      "mean_token_accuracy": 0.1640881672501564,
      "num_tokens": 30556646.0,
      "step": 15630
    },
    {
      "entropy": 6.052614212036133,
      "epoch": 1.5629029839556154,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047618608821354636,
      "loss": 5.7454,
      "mean_token_accuracy": 0.174944868683815,
      "num_tokens": 30565984.0,
      "step": 15635
    },
    {
      "entropy": 6.071623802185059,
      "epoch": 1.5634028090168441,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004761701060018097,
      "loss": 5.8383,
      "mean_token_accuracy": 0.16620049774646758,
      "num_tokens": 30575257.0,
      "step": 15640
    },
    {
      "entropy": 6.05193133354187,
      "epoch": 1.5639026340780728,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000476154118728685,
      "loss": 5.7775,
      "mean_token_accuracy": 0.16141719371080399,
      "num_tokens": 30584248.0,
      "step": 15645
    },
    {
      "entropy": 6.111474657058716,
      "epoch": 1.564402459139301,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047613812639457425,
      "loss": 5.8216,
      "mean_token_accuracy": 0.16151486784219743,
      "num_tokens": 30593875.0,
      "step": 15650
    },
    {
      "entropy": 6.035809421539307,
      "epoch": 1.5649022842005298,
      "grad_norm": 1.25,
      "learning_rate": 0.0004761221289998801,
      "loss": 5.7222,
      "mean_token_accuracy": 0.16866254359483718,
      "num_tokens": 30603696.0,
      "step": 15655
    },
    {
      "entropy": 6.057669019699096,
      "epoch": 1.5654021092617585,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004761061265450049,
      "loss": 5.7601,
      "mean_token_accuracy": 0.1659014865756035,
      "num_tokens": 30613451.0,
      "step": 15660
    },
    {
      "entropy": 6.065277147293091,
      "epoch": 1.565901934322987,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047609011903035136,
      "loss": 5.8425,
      "mean_token_accuracy": 0.16389566361904145,
      "num_tokens": 30623567.0,
      "step": 15665
    },
    {
      "entropy": 6.102051210403443,
      "epoch": 1.5664017593842154,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047607410645632216,
      "loss": 5.7858,
      "mean_token_accuracy": 0.16584625095129013,
      "num_tokens": 30634828.0,
      "step": 15670
    },
    {
      "entropy": 6.175020980834961,
      "epoch": 1.5669015844454441,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047605808882332013,
      "loss": 5.9406,
      "mean_token_accuracy": 0.1588262103497982,
      "num_tokens": 30644552.0,
      "step": 15675
    },
    {
      "entropy": 6.094100856781006,
      "epoch": 1.5674014095066726,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004760420661317484,
      "loss": 5.9405,
      "mean_token_accuracy": 0.1593879133462906,
      "num_tokens": 30654874.0,
      "step": 15680
    },
    {
      "entropy": 6.145671606063843,
      "epoch": 1.567901234567901,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004760260383820099,
      "loss": 5.794,
      "mean_token_accuracy": 0.1623311907052994,
      "num_tokens": 30665619.0,
      "step": 15685
    },
    {
      "entropy": 6.033340072631836,
      "epoch": 1.5684010596291298,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047601000557450816,
      "loss": 5.7472,
      "mean_token_accuracy": 0.168060502409935,
      "num_tokens": 30675072.0,
      "step": 15690
    },
    {
      "entropy": 6.17101821899414,
      "epoch": 1.5689008846903585,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004759939677096463,
      "loss": 5.8814,
      "mean_token_accuracy": 0.15875439047813417,
      "num_tokens": 30684567.0,
      "step": 15695
    },
    {
      "entropy": 6.078986120223999,
      "epoch": 1.569400709751587,
      "grad_norm": 1.125,
      "learning_rate": 0.0004759779247878279,
      "loss": 5.8647,
      "mean_token_accuracy": 0.16377130895853043,
      "num_tokens": 30693808.0,
      "step": 15700
    },
    {
      "entropy": 5.93624472618103,
      "epoch": 1.5699005348128154,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047596187680945675,
      "loss": 5.7424,
      "mean_token_accuracy": 0.18065049052238463,
      "num_tokens": 30702895.0,
      "step": 15705
    },
    {
      "entropy": 6.1225310325622555,
      "epoch": 1.5704003598740441,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004759458237749364,
      "loss": 5.9019,
      "mean_token_accuracy": 0.16245635896921157,
      "num_tokens": 30714582.0,
      "step": 15710
    },
    {
      "entropy": 6.133759212493897,
      "epoch": 1.5709001849352726,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004759297656846708,
      "loss": 5.8471,
      "mean_token_accuracy": 0.1591715469956398,
      "num_tokens": 30724415.0,
      "step": 15715
    },
    {
      "entropy": 6.067207098007202,
      "epoch": 1.571400009996501,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047591370253906415,
      "loss": 5.7611,
      "mean_token_accuracy": 0.1668723538517952,
      "num_tokens": 30733040.0,
      "step": 15720
    },
    {
      "entropy": 6.023939990997315,
      "epoch": 1.5718998350577298,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047589763433852027,
      "loss": 5.7573,
      "mean_token_accuracy": 0.1695743829011917,
      "num_tokens": 30743022.0,
      "step": 15725
    },
    {
      "entropy": 6.023278331756591,
      "epoch": 1.5723996601189585,
      "grad_norm": 1.125,
      "learning_rate": 0.0004758815610834437,
      "loss": 5.7732,
      "mean_token_accuracy": 0.16710205674171447,
      "num_tokens": 30753335.0,
      "step": 15730
    },
    {
      "entropy": 6.0726155757904055,
      "epoch": 1.572899485180187,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047586548277423873,
      "loss": 5.8115,
      "mean_token_accuracy": 0.16010416597127913,
      "num_tokens": 30762863.0,
      "step": 15735
    },
    {
      "entropy": 6.109581041336059,
      "epoch": 1.5733993102414154,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047584939941130994,
      "loss": 5.8376,
      "mean_token_accuracy": 0.1687723569571972,
      "num_tokens": 30773090.0,
      "step": 15740
    },
    {
      "entropy": 6.128092527389526,
      "epoch": 1.5738991353026441,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004758333109950619,
      "loss": 5.8396,
      "mean_token_accuracy": 0.1625595822930336,
      "num_tokens": 30782589.0,
      "step": 15745
    },
    {
      "entropy": 6.086670684814453,
      "epoch": 1.5743989603638726,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047581721752589943,
      "loss": 5.7393,
      "mean_token_accuracy": 0.171626016497612,
      "num_tokens": 30791508.0,
      "step": 15750
    },
    {
      "entropy": 6.031594944000244,
      "epoch": 1.574898785425101,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004758011190042274,
      "loss": 5.7721,
      "mean_token_accuracy": 0.1667802155017853,
      "num_tokens": 30801133.0,
      "step": 15755
    },
    {
      "entropy": 6.049183034896851,
      "epoch": 1.5753986104863298,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004757850154304509,
      "loss": 5.6746,
      "mean_token_accuracy": 0.17032696753740312,
      "num_tokens": 30810100.0,
      "step": 15760
    },
    {
      "entropy": 6.025739240646362,
      "epoch": 1.5758984355475585,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004757689068049752,
      "loss": 5.7635,
      "mean_token_accuracy": 0.17247294634580612,
      "num_tokens": 30820118.0,
      "step": 15765
    },
    {
      "entropy": 6.030754232406617,
      "epoch": 1.576398260608787,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047575279312820533,
      "loss": 5.8042,
      "mean_token_accuracy": 0.1603006437420845,
      "num_tokens": 30830480.0,
      "step": 15770
    },
    {
      "entropy": 6.08657660484314,
      "epoch": 1.5768980856700154,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004757366744005469,
      "loss": 5.9192,
      "mean_token_accuracy": 0.15584653317928315,
      "num_tokens": 30840421.0,
      "step": 15775
    },
    {
      "entropy": 6.124019050598145,
      "epoch": 1.5773979107312441,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047572055062240537,
      "loss": 5.779,
      "mean_token_accuracy": 0.16486170142889023,
      "num_tokens": 30849247.0,
      "step": 15780
    },
    {
      "entropy": 6.0592742443084715,
      "epoch": 1.5778977357924726,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004757044217941865,
      "loss": 5.7664,
      "mean_token_accuracy": 0.16957200318574905,
      "num_tokens": 30858804.0,
      "step": 15785
    },
    {
      "entropy": 6.063622570037841,
      "epoch": 1.578397560853701,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047568828791629605,
      "loss": 5.8216,
      "mean_token_accuracy": 0.1645832508802414,
      "num_tokens": 30869150.0,
      "step": 15790
    },
    {
      "entropy": 5.927814531326294,
      "epoch": 1.5788973859149298,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004756721489891398,
      "loss": 5.7223,
      "mean_token_accuracy": 0.17677051872014998,
      "num_tokens": 30878671.0,
      "step": 15795
    },
    {
      "entropy": 6.149916362762451,
      "epoch": 1.5793972109761585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004756560050131239,
      "loss": 5.8716,
      "mean_token_accuracy": 0.15524349063634874,
      "num_tokens": 30888064.0,
      "step": 15800
    },
    {
      "entropy": 6.143588018417359,
      "epoch": 1.579897036037387,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047563985598865466,
      "loss": 5.8694,
      "mean_token_accuracy": 0.16245703995227814,
      "num_tokens": 30898205.0,
      "step": 15805
    },
    {
      "entropy": 6.1065630435943605,
      "epoch": 1.5803968610986154,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047562370191613816,
      "loss": 5.8381,
      "mean_token_accuracy": 0.16250723600387573,
      "num_tokens": 30907371.0,
      "step": 15810
    },
    {
      "entropy": 6.181057167053223,
      "epoch": 1.5808966861598441,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004756075427959811,
      "loss": 5.9382,
      "mean_token_accuracy": 0.1563638284802437,
      "num_tokens": 30917481.0,
      "step": 15815
    },
    {
      "entropy": 6.108288669586182,
      "epoch": 1.5813965112210726,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004755913786285898,
      "loss": 5.8761,
      "mean_token_accuracy": 0.15825229287147521,
      "num_tokens": 30927375.0,
      "step": 15820
    },
    {
      "entropy": 6.047292947769165,
      "epoch": 1.581896336282301,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047557520941437093,
      "loss": 5.7666,
      "mean_token_accuracy": 0.16892989724874496,
      "num_tokens": 30935888.0,
      "step": 15825
    },
    {
      "entropy": 6.117914867401123,
      "epoch": 1.5823961613435298,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047555903515373156,
      "loss": 5.8131,
      "mean_token_accuracy": 0.17164721190929413,
      "num_tokens": 30946433.0,
      "step": 15830
    },
    {
      "entropy": 6.12684268951416,
      "epoch": 1.5828959864047585,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047554285584707834,
      "loss": 5.8164,
      "mean_token_accuracy": 0.16427652314305305,
      "num_tokens": 30956951.0,
      "step": 15835
    },
    {
      "entropy": 5.945752668380737,
      "epoch": 1.583395811465987,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004755266714948186,
      "loss": 5.6973,
      "mean_token_accuracy": 0.17365053296089172,
      "num_tokens": 30967332.0,
      "step": 15840
    },
    {
      "entropy": 6.049661779403687,
      "epoch": 1.5838956365272154,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004755104820973593,
      "loss": 5.8935,
      "mean_token_accuracy": 0.15945337563753129,
      "num_tokens": 30977704.0,
      "step": 15845
    },
    {
      "entropy": 6.2506794929504395,
      "epoch": 1.5843954615884441,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047549428765510793,
      "loss": 5.9737,
      "mean_token_accuracy": 0.15435580909252167,
      "num_tokens": 30987916.0,
      "step": 15850
    },
    {
      "entropy": 6.073114061355591,
      "epoch": 1.5848952866496726,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047547808816847186,
      "loss": 5.7454,
      "mean_token_accuracy": 0.18016756027936937,
      "num_tokens": 30997357.0,
      "step": 15855
    },
    {
      "entropy": 6.096533107757568,
      "epoch": 1.585395111710901,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004754618836378587,
      "loss": 5.8831,
      "mean_token_accuracy": 0.16084128767251968,
      "num_tokens": 31006113.0,
      "step": 15860
    },
    {
      "entropy": 6.026092100143432,
      "epoch": 1.5858949367721298,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004754456740636761,
      "loss": 5.7387,
      "mean_token_accuracy": 0.17089740186929703,
      "num_tokens": 31014952.0,
      "step": 15865
    },
    {
      "entropy": 6.194831848144531,
      "epoch": 1.5863947618333585,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047542945944633187,
      "loss": 5.9025,
      "mean_token_accuracy": 0.15056465864181517,
      "num_tokens": 31024934.0,
      "step": 15870
    },
    {
      "entropy": 6.07628607749939,
      "epoch": 1.5868945868945867,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047541323978623405,
      "loss": 5.7195,
      "mean_token_accuracy": 0.17368005067110062,
      "num_tokens": 31033934.0,
      "step": 15875
    },
    {
      "entropy": 6.096920490264893,
      "epoch": 1.5873944119558154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047539701508379073,
      "loss": 5.8868,
      "mean_token_accuracy": 0.16522404551506042,
      "num_tokens": 31043896.0,
      "step": 15880
    },
    {
      "entropy": 6.0546918392181395,
      "epoch": 1.5878942370170441,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004753807853394101,
      "loss": 5.8223,
      "mean_token_accuracy": 0.1652487426996231,
      "num_tokens": 31055103.0,
      "step": 15885
    },
    {
      "entropy": 6.1153843879699705,
      "epoch": 1.5883940620782726,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047536455055350034,
      "loss": 5.8778,
      "mean_token_accuracy": 0.16184850335121154,
      "num_tokens": 31065478.0,
      "step": 15890
    },
    {
      "entropy": 6.079290246963501,
      "epoch": 1.588893887139501,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047534831072647016,
      "loss": 5.8157,
      "mean_token_accuracy": 0.1633896693587303,
      "num_tokens": 31074408.0,
      "step": 15895
    },
    {
      "entropy": 6.100226497650146,
      "epoch": 1.5893937122007298,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047533206585872805,
      "loss": 5.8275,
      "mean_token_accuracy": 0.16440559476613997,
      "num_tokens": 31084223.0,
      "step": 15900
    },
    {
      "entropy": 6.030626916885376,
      "epoch": 1.5898935372619585,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004753158159506827,
      "loss": 5.7126,
      "mean_token_accuracy": 0.1752822741866112,
      "num_tokens": 31094133.0,
      "step": 15905
    },
    {
      "entropy": 6.0829778671264645,
      "epoch": 1.5903933623231867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004752995610027429,
      "loss": 5.846,
      "mean_token_accuracy": 0.16437599509954454,
      "num_tokens": 31104283.0,
      "step": 15910
    },
    {
      "entropy": 6.0048740863800045,
      "epoch": 1.5908931873844154,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047528330101531777,
      "loss": 5.8139,
      "mean_token_accuracy": 0.16067801117897035,
      "num_tokens": 31114274.0,
      "step": 15915
    },
    {
      "entropy": 6.113825750350952,
      "epoch": 1.5913930124456441,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047526703598881623,
      "loss": 5.8077,
      "mean_token_accuracy": 0.16401219516992568,
      "num_tokens": 31123680.0,
      "step": 15920
    },
    {
      "entropy": 6.049138832092285,
      "epoch": 1.5918928375068726,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004752507659236476,
      "loss": 5.7598,
      "mean_token_accuracy": 0.17305042296648027,
      "num_tokens": 31133520.0,
      "step": 15925
    },
    {
      "entropy": 5.98088812828064,
      "epoch": 1.592392662568101,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004752344908202213,
      "loss": 5.6943,
      "mean_token_accuracy": 0.1785082459449768,
      "num_tokens": 31142915.0,
      "step": 15930
    },
    {
      "entropy": 6.002624988555908,
      "epoch": 1.5928924876293298,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004752182106789467,
      "loss": 5.746,
      "mean_token_accuracy": 0.16845571547746657,
      "num_tokens": 31152133.0,
      "step": 15935
    },
    {
      "entropy": 6.011069393157959,
      "epoch": 1.5933923126905583,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047520192550023344,
      "loss": 5.7466,
      "mean_token_accuracy": 0.1734068885445595,
      "num_tokens": 31161666.0,
      "step": 15940
    },
    {
      "entropy": 6.007725524902344,
      "epoch": 1.5938921377517867,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004751856352844913,
      "loss": 5.7859,
      "mean_token_accuracy": 0.16638416945934295,
      "num_tokens": 31171591.0,
      "step": 15945
    },
    {
      "entropy": 6.184588146209717,
      "epoch": 1.5943919628130154,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047516934003213,
      "loss": 5.9031,
      "mean_token_accuracy": 0.16140573024749755,
      "num_tokens": 31181396.0,
      "step": 15950
    },
    {
      "entropy": 6.130023717880249,
      "epoch": 1.5948917878742441,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047515303974355965,
      "loss": 5.8386,
      "mean_token_accuracy": 0.16169044971466065,
      "num_tokens": 31191926.0,
      "step": 15955
    },
    {
      "entropy": 6.056381893157959,
      "epoch": 1.5953916129354726,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047513673441919036,
      "loss": 5.8309,
      "mean_token_accuracy": 0.16076339930295944,
      "num_tokens": 31202809.0,
      "step": 15960
    },
    {
      "entropy": 6.097253465652466,
      "epoch": 1.595891437996701,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004751204240594323,
      "loss": 5.7834,
      "mean_token_accuracy": 0.1686539277434349,
      "num_tokens": 31211763.0,
      "step": 15965
    },
    {
      "entropy": 6.095336818695069,
      "epoch": 1.5963912630579298,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004751041086646959,
      "loss": 5.8239,
      "mean_token_accuracy": 0.1589602641761303,
      "num_tokens": 31220881.0,
      "step": 15970
    },
    {
      "entropy": 6.050861740112305,
      "epoch": 1.5968910881191583,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004750877882353915,
      "loss": 5.7647,
      "mean_token_accuracy": 0.16889928430318832,
      "num_tokens": 31230740.0,
      "step": 15975
    },
    {
      "entropy": 5.974371147155762,
      "epoch": 1.5973909131803867,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047507146277192997,
      "loss": 5.6339,
      "mean_token_accuracy": 0.179343119263649,
      "num_tokens": 31239777.0,
      "step": 15980
    },
    {
      "entropy": 6.02530608177185,
      "epoch": 1.5978907382416154,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004750551322747219,
      "loss": 5.8188,
      "mean_token_accuracy": 0.16145926266908645,
      "num_tokens": 31248574.0,
      "step": 15985
    },
    {
      "entropy": 6.055920219421386,
      "epoch": 1.5983905633028441,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004750387967441781,
      "loss": 5.8612,
      "mean_token_accuracy": 0.1629121333360672,
      "num_tokens": 31258888.0,
      "step": 15990
    },
    {
      "entropy": 6.032140016555786,
      "epoch": 1.5988903883640726,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047502245618070976,
      "loss": 5.7528,
      "mean_token_accuracy": 0.1715491533279419,
      "num_tokens": 31269966.0,
      "step": 15995
    },
    {
      "entropy": 6.04396800994873,
      "epoch": 1.599390213425301,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047500611058472783,
      "loss": 5.8137,
      "mean_token_accuracy": 0.16593980491161348,
      "num_tokens": 31279746.0,
      "step": 16000
    },
    {
      "entropy": 6.072383117675781,
      "epoch": 1.5998900384865298,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047498975995664356,
      "loss": 5.8327,
      "mean_token_accuracy": 0.16279804855585098,
      "num_tokens": 31289677.0,
      "step": 16005
    },
    {
      "entropy": 6.090796041488647,
      "epoch": 1.6003898635477583,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004749734042968685,
      "loss": 5.8324,
      "mean_token_accuracy": 0.17088265866041183,
      "num_tokens": 31299752.0,
      "step": 16010
    },
    {
      "entropy": 6.060184383392334,
      "epoch": 1.6008896886089867,
      "grad_norm": 1.203125,
      "learning_rate": 0.000474957043605814,
      "loss": 5.7622,
      "mean_token_accuracy": 0.16598546057939528,
      "num_tokens": 31308986.0,
      "step": 16015
    },
    {
      "entropy": 5.992848539352417,
      "epoch": 1.6013895136702154,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047494067788389174,
      "loss": 5.7256,
      "mean_token_accuracy": 0.1770410045981407,
      "num_tokens": 31318848.0,
      "step": 16020
    },
    {
      "entropy": 6.0903974056243895,
      "epoch": 1.6018893387314441,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047492430713151347,
      "loss": 5.8634,
      "mean_token_accuracy": 0.15820887684822083,
      "num_tokens": 31329397.0,
      "step": 16025
    },
    {
      "entropy": 6.1950085163116455,
      "epoch": 1.6023891637926726,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004749079313490911,
      "loss": 5.9108,
      "mean_token_accuracy": 0.15854822397232055,
      "num_tokens": 31340440.0,
      "step": 16030
    },
    {
      "entropy": 6.1701795101165775,
      "epoch": 1.602888988853901,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047489155053703655,
      "loss": 5.8451,
      "mean_token_accuracy": 0.16444195806980133,
      "num_tokens": 31350418.0,
      "step": 16035
    },
    {
      "entropy": 6.019715690612793,
      "epoch": 1.6033888139151298,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004748751646957621,
      "loss": 5.7829,
      "mean_token_accuracy": 0.16820483654737473,
      "num_tokens": 31360714.0,
      "step": 16040
    },
    {
      "entropy": 6.0980815410614015,
      "epoch": 1.6038886389763582,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004748587738256799,
      "loss": 5.8086,
      "mean_token_accuracy": 0.16902743130922318,
      "num_tokens": 31369693.0,
      "step": 16045
    },
    {
      "entropy": 6.045738077163696,
      "epoch": 1.6043884640375867,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004748423779272024,
      "loss": 5.7627,
      "mean_token_accuracy": 0.16247181892395018,
      "num_tokens": 31380068.0,
      "step": 16050
    },
    {
      "entropy": 6.084030246734619,
      "epoch": 1.6048882890988154,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047482597700074204,
      "loss": 5.8337,
      "mean_token_accuracy": 0.1687791258096695,
      "num_tokens": 31389151.0,
      "step": 16055
    },
    {
      "entropy": 6.170691347122192,
      "epoch": 1.6053881141600441,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047480957104671157,
      "loss": 5.9453,
      "mean_token_accuracy": 0.15546009689569473,
      "num_tokens": 31400260.0,
      "step": 16060
    },
    {
      "entropy": 6.199683332443238,
      "epoch": 1.6058879392212726,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004747931600655237,
      "loss": 5.8504,
      "mean_token_accuracy": 0.1612503744661808,
      "num_tokens": 31410747.0,
      "step": 16065
    },
    {
      "entropy": 6.0829322814941404,
      "epoch": 1.606387764282501,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047477674405759136,
      "loss": 5.7916,
      "mean_token_accuracy": 0.17114344835281373,
      "num_tokens": 31419230.0,
      "step": 16070
    },
    {
      "entropy": 6.060317277908325,
      "epoch": 1.6068875893437298,
      "grad_norm": 1.125,
      "learning_rate": 0.0004747603230233275,
      "loss": 5.768,
      "mean_token_accuracy": 0.1681264117360115,
      "num_tokens": 31428314.0,
      "step": 16075
    },
    {
      "entropy": 5.992963075637817,
      "epoch": 1.6073874144049582,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004747438969631453,
      "loss": 5.7616,
      "mean_token_accuracy": 0.16702873706817628,
      "num_tokens": 31438141.0,
      "step": 16080
    },
    {
      "entropy": 6.054623317718506,
      "epoch": 1.6078872394661867,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004747274658774581,
      "loss": 5.8116,
      "mean_token_accuracy": 0.16641197949647904,
      "num_tokens": 31448609.0,
      "step": 16085
    },
    {
      "entropy": 6.125154590606689,
      "epoch": 1.6083870645274154,
      "grad_norm": 1.25,
      "learning_rate": 0.0004747110297666793,
      "loss": 5.8109,
      "mean_token_accuracy": 0.16668089032173156,
      "num_tokens": 31458603.0,
      "step": 16090
    },
    {
      "entropy": 6.059641695022583,
      "epoch": 1.6088868895886441,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004746945886312223,
      "loss": 5.7368,
      "mean_token_accuracy": 0.17442202270030976,
      "num_tokens": 31468917.0,
      "step": 16095
    },
    {
      "entropy": 6.024519348144532,
      "epoch": 1.6093867146498726,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004746781424715009,
      "loss": 5.8378,
      "mean_token_accuracy": 0.1632831797003746,
      "num_tokens": 31479061.0,
      "step": 16100
    },
    {
      "entropy": 6.157224321365357,
      "epoch": 1.609886539711101,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047466169128792873,
      "loss": 5.9092,
      "mean_token_accuracy": 0.15356053784489632,
      "num_tokens": 31488723.0,
      "step": 16105
    },
    {
      "entropy": 6.1015197277069095,
      "epoch": 1.6103863647723298,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004746452350809198,
      "loss": 5.8309,
      "mean_token_accuracy": 0.16502804309129715,
      "num_tokens": 31497715.0,
      "step": 16110
    },
    {
      "entropy": 6.097683620452881,
      "epoch": 1.6108861898335582,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047462877385088815,
      "loss": 5.793,
      "mean_token_accuracy": 0.16549659967422486,
      "num_tokens": 31507571.0,
      "step": 16115
    },
    {
      "entropy": 6.047437334060669,
      "epoch": 1.6113860148947867,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047461230759824794,
      "loss": 5.8083,
      "mean_token_accuracy": 0.16341807842254638,
      "num_tokens": 31517582.0,
      "step": 16120
    },
    {
      "entropy": 6.104514455795288,
      "epoch": 1.6118858399560154,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047459583632341343,
      "loss": 5.8532,
      "mean_token_accuracy": 0.16830697134137154,
      "num_tokens": 31528225.0,
      "step": 16125
    },
    {
      "entropy": 6.064204120635987,
      "epoch": 1.6123856650172441,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047457936002679905,
      "loss": 5.6659,
      "mean_token_accuracy": 0.17152563035488128,
      "num_tokens": 31537778.0,
      "step": 16130
    },
    {
      "entropy": 6.022012376785279,
      "epoch": 1.6128854900784724,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047456287870881924,
      "loss": 5.7474,
      "mean_token_accuracy": 0.16920860707759858,
      "num_tokens": 31548431.0,
      "step": 16135
    },
    {
      "entropy": 6.053183603286743,
      "epoch": 1.613385315139701,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004745463923698888,
      "loss": 5.7432,
      "mean_token_accuracy": 0.17522135972976685,
      "num_tokens": 31557333.0,
      "step": 16140
    },
    {
      "entropy": 6.063870668411255,
      "epoch": 1.6138851402009298,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004745299010104226,
      "loss": 5.7775,
      "mean_token_accuracy": 0.1661730080842972,
      "num_tokens": 31566333.0,
      "step": 16145
    },
    {
      "entropy": 5.994939947128296,
      "epoch": 1.6143849652621582,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047451340463083527,
      "loss": 5.7047,
      "mean_token_accuracy": 0.17882685214281083,
      "num_tokens": 31575971.0,
      "step": 16150
    },
    {
      "entropy": 5.9962362289428714,
      "epoch": 1.6148847903233867,
      "grad_norm": 1.25,
      "learning_rate": 0.00047449690323154206,
      "loss": 5.7618,
      "mean_token_accuracy": 0.1756792739033699,
      "num_tokens": 31585396.0,
      "step": 16155
    },
    {
      "entropy": 6.082142114639282,
      "epoch": 1.6153846153846154,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047448039681295814,
      "loss": 5.8262,
      "mean_token_accuracy": 0.15942814499139785,
      "num_tokens": 31595800.0,
      "step": 16160
    },
    {
      "entropy": 6.065332508087158,
      "epoch": 1.6158844404458441,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047446388537549867,
      "loss": 5.8321,
      "mean_token_accuracy": 0.1608894243836403,
      "num_tokens": 31604792.0,
      "step": 16165
    },
    {
      "entropy": 6.089527177810669,
      "epoch": 1.6163842655070724,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004744473689195793,
      "loss": 5.8411,
      "mean_token_accuracy": 0.16218537837266922,
      "num_tokens": 31614952.0,
      "step": 16170
    },
    {
      "entropy": 6.10189208984375,
      "epoch": 1.616884090568301,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004744308474456154,
      "loss": 5.7683,
      "mean_token_accuracy": 0.17142685502767563,
      "num_tokens": 31624635.0,
      "step": 16175
    },
    {
      "entropy": 5.992224073410034,
      "epoch": 1.6173839156295298,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004744143209540227,
      "loss": 5.7376,
      "mean_token_accuracy": 0.17339555993676187,
      "num_tokens": 31634710.0,
      "step": 16180
    },
    {
      "entropy": 6.183680629730224,
      "epoch": 1.6178837406907582,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004743977894452169,
      "loss": 5.8853,
      "mean_token_accuracy": 0.16218243390321732,
      "num_tokens": 31644371.0,
      "step": 16185
    },
    {
      "entropy": 6.118359756469727,
      "epoch": 1.6183835657519867,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004743812529196142,
      "loss": 5.779,
      "mean_token_accuracy": 0.16732556670904158,
      "num_tokens": 31654262.0,
      "step": 16190
    },
    {
      "entropy": 6.122082948684692,
      "epoch": 1.6188833908132154,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004743647113776304,
      "loss": 5.8402,
      "mean_token_accuracy": 0.15755100995302201,
      "num_tokens": 31663209.0,
      "step": 16195
    },
    {
      "entropy": 5.986993980407715,
      "epoch": 1.619383215874444,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047434816481968175,
      "loss": 5.7752,
      "mean_token_accuracy": 0.1663312792778015,
      "num_tokens": 31672370.0,
      "step": 16200
    },
    {
      "entropy": 6.064879703521728,
      "epoch": 1.6198830409356724,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047433161324618456,
      "loss": 5.8299,
      "mean_token_accuracy": 0.16975468397140503,
      "num_tokens": 31681886.0,
      "step": 16205
    },
    {
      "entropy": 6.108088397979737,
      "epoch": 1.620382865996901,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004743150566575553,
      "loss": 5.7862,
      "mean_token_accuracy": 0.17354244142770767,
      "num_tokens": 31691098.0,
      "step": 16210
    },
    {
      "entropy": 6.0279247760772705,
      "epoch": 1.6208826910581298,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004742984950542105,
      "loss": 5.755,
      "mean_token_accuracy": 0.16234150379896164,
      "num_tokens": 31700638.0,
      "step": 16215
    },
    {
      "entropy": 6.0445161819458,
      "epoch": 1.6213825161193582,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047428192843656684,
      "loss": 5.805,
      "mean_token_accuracy": 0.16646316945552825,
      "num_tokens": 31711326.0,
      "step": 16220
    },
    {
      "entropy": 6.0514280796051025,
      "epoch": 1.6218823411805867,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047426535680504105,
      "loss": 5.7551,
      "mean_token_accuracy": 0.1717141643166542,
      "num_tokens": 31721220.0,
      "step": 16225
    },
    {
      "entropy": 6.058555555343628,
      "epoch": 1.6223821662418154,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047424878016005036,
      "loss": 5.8226,
      "mean_token_accuracy": 0.16051426082849501,
      "num_tokens": 31731190.0,
      "step": 16230
    },
    {
      "entropy": 6.138337802886963,
      "epoch": 1.622881991303044,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004742321985020115,
      "loss": 5.8246,
      "mean_token_accuracy": 0.17179190069437028,
      "num_tokens": 31740761.0,
      "step": 16235
    },
    {
      "entropy": 6.036164855957031,
      "epoch": 1.6233818163642724,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004742156118313418,
      "loss": 5.8335,
      "mean_token_accuracy": 0.1640630766749382,
      "num_tokens": 31750038.0,
      "step": 16240
    },
    {
      "entropy": 6.080115509033203,
      "epoch": 1.623881641425501,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047419902014845855,
      "loss": 5.7771,
      "mean_token_accuracy": 0.16978770345449448,
      "num_tokens": 31761574.0,
      "step": 16245
    },
    {
      "entropy": 6.062795639038086,
      "epoch": 1.6243814664867298,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004741824234537793,
      "loss": 5.7883,
      "mean_token_accuracy": 0.16592679768800736,
      "num_tokens": 31771980.0,
      "step": 16250
    },
    {
      "entropy": 6.0683835506439205,
      "epoch": 1.6248812915479582,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047416582174772143,
      "loss": 5.7937,
      "mean_token_accuracy": 0.1663748323917389,
      "num_tokens": 31781242.0,
      "step": 16255
    },
    {
      "entropy": 6.1791040897369385,
      "epoch": 1.6253811166091867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047414921503070276,
      "loss": 5.8623,
      "mean_token_accuracy": 0.16528930366039277,
      "num_tokens": 31792250.0,
      "step": 16260
    },
    {
      "entropy": 6.038084316253662,
      "epoch": 1.6258809416704154,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047413260330314116,
      "loss": 5.7266,
      "mean_token_accuracy": 0.1697758764028549,
      "num_tokens": 31802512.0,
      "step": 16265
    },
    {
      "entropy": 5.989192819595337,
      "epoch": 1.626380766731644,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004741159865654544,
      "loss": 5.6882,
      "mean_token_accuracy": 0.1764667436480522,
      "num_tokens": 31811957.0,
      "step": 16270
    },
    {
      "entropy": 6.040925550460815,
      "epoch": 1.6268805917928724,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047409936481806076,
      "loss": 5.7866,
      "mean_token_accuracy": 0.1701054409146309,
      "num_tokens": 31821101.0,
      "step": 16275
    },
    {
      "entropy": 6.061092329025269,
      "epoch": 1.627380416854101,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004740827380613782,
      "loss": 5.6575,
      "mean_token_accuracy": 0.16977206468582154,
      "num_tokens": 31831085.0,
      "step": 16280
    },
    {
      "entropy": 6.023536443710327,
      "epoch": 1.6278802419153298,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047406610629582526,
      "loss": 5.8265,
      "mean_token_accuracy": 0.16301012188196182,
      "num_tokens": 31840354.0,
      "step": 16285
    },
    {
      "entropy": 5.986581945419312,
      "epoch": 1.6283800669765582,
      "grad_norm": 1.125,
      "learning_rate": 0.00047404946952182034,
      "loss": 5.7479,
      "mean_token_accuracy": 0.17222623527050018,
      "num_tokens": 31849250.0,
      "step": 16290
    },
    {
      "entropy": 6.104855394363403,
      "epoch": 1.6288798920377867,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004740328277397819,
      "loss": 5.7953,
      "mean_token_accuracy": 0.1642753764986992,
      "num_tokens": 31859533.0,
      "step": 16295
    },
    {
      "entropy": 6.156737804412842,
      "epoch": 1.6293797170990154,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004740161809501288,
      "loss": 5.8588,
      "mean_token_accuracy": 0.15760941207408904,
      "num_tokens": 31868821.0,
      "step": 16300
    },
    {
      "entropy": 6.070944929122925,
      "epoch": 1.6298795421602439,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047399952915327975,
      "loss": 5.7931,
      "mean_token_accuracy": 0.1604434460401535,
      "num_tokens": 31878084.0,
      "step": 16305
    },
    {
      "entropy": 6.051668930053711,
      "epoch": 1.6303793672214724,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004739828723496538,
      "loss": 5.7418,
      "mean_token_accuracy": 0.1659568652510643,
      "num_tokens": 31887634.0,
      "step": 16310
    },
    {
      "entropy": 6.009443712234497,
      "epoch": 1.630879192282701,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047396621053967,
      "loss": 5.7521,
      "mean_token_accuracy": 0.16874603778123856,
      "num_tokens": 31897871.0,
      "step": 16315
    },
    {
      "entropy": 6.109589767456055,
      "epoch": 1.6313790173439298,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047394954372374743,
      "loss": 5.8207,
      "mean_token_accuracy": 0.16483808308839798,
      "num_tokens": 31907960.0,
      "step": 16320
    },
    {
      "entropy": 6.0910162925720215,
      "epoch": 1.6318788424051582,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047393287190230566,
      "loss": 5.7768,
      "mean_token_accuracy": 0.16673094183206558,
      "num_tokens": 31918550.0,
      "step": 16325
    },
    {
      "entropy": 6.0775810241699215,
      "epoch": 1.6323786674663867,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004739161950757639,
      "loss": 5.9078,
      "mean_token_accuracy": 0.1608056604862213,
      "num_tokens": 31927941.0,
      "step": 16330
    },
    {
      "entropy": 6.119914436340332,
      "epoch": 1.6328784925276154,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000473899513244542,
      "loss": 5.8558,
      "mean_token_accuracy": 0.1636231243610382,
      "num_tokens": 31938179.0,
      "step": 16335
    },
    {
      "entropy": 6.045485019683838,
      "epoch": 1.6333783175888439,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004738828264090594,
      "loss": 5.728,
      "mean_token_accuracy": 0.17136624902486802,
      "num_tokens": 31947287.0,
      "step": 16340
    },
    {
      "entropy": 5.9772820472717285,
      "epoch": 1.6338781426500724,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047386613456973614,
      "loss": 5.7385,
      "mean_token_accuracy": 0.17193353176116943,
      "num_tokens": 31955953.0,
      "step": 16345
    },
    {
      "entropy": 5.999509620666504,
      "epoch": 1.634377967711301,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004738494377269921,
      "loss": 5.7729,
      "mean_token_accuracy": 0.16572466343641282,
      "num_tokens": 31964874.0,
      "step": 16350
    },
    {
      "entropy": 6.128712320327759,
      "epoch": 1.6348777927725298,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047383273588124736,
      "loss": 5.8532,
      "mean_token_accuracy": 0.16079513132572174,
      "num_tokens": 31975865.0,
      "step": 16355
    },
    {
      "entropy": 6.11778621673584,
      "epoch": 1.6353776178337582,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004738160290329222,
      "loss": 5.7482,
      "mean_token_accuracy": 0.16990725696086884,
      "num_tokens": 31985229.0,
      "step": 16360
    },
    {
      "entropy": 6.028687191009522,
      "epoch": 1.6358774428949867,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047379931718243684,
      "loss": 5.8505,
      "mean_token_accuracy": 0.16991683691740037,
      "num_tokens": 31995195.0,
      "step": 16365
    },
    {
      "entropy": 6.077043533325195,
      "epoch": 1.6363772679562154,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004737826003302118,
      "loss": 5.8403,
      "mean_token_accuracy": 0.16641165018081666,
      "num_tokens": 32006053.0,
      "step": 16370
    },
    {
      "entropy": 6.152651596069336,
      "epoch": 1.6368770930174439,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047376587847666767,
      "loss": 5.8716,
      "mean_token_accuracy": 0.1639844998717308,
      "num_tokens": 32015025.0,
      "step": 16375
    },
    {
      "entropy": 5.978629732131958,
      "epoch": 1.6373769180786724,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004737491516222253,
      "loss": 5.718,
      "mean_token_accuracy": 0.17198902368545532,
      "num_tokens": 32025116.0,
      "step": 16380
    },
    {
      "entropy": 5.994442701339722,
      "epoch": 1.637876743139901,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047373241976730537,
      "loss": 5.782,
      "mean_token_accuracy": 0.16659699827432634,
      "num_tokens": 32034271.0,
      "step": 16385
    },
    {
      "entropy": 6.054227066040039,
      "epoch": 1.6383765682011298,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004737156829123289,
      "loss": 5.6832,
      "mean_token_accuracy": 0.18116614818572999,
      "num_tokens": 32043139.0,
      "step": 16390
    },
    {
      "entropy": 6.055826854705811,
      "epoch": 1.638876393262358,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047369894105771693,
      "loss": 5.7361,
      "mean_token_accuracy": 0.1673491671681404,
      "num_tokens": 32052719.0,
      "step": 16395
    },
    {
      "entropy": 6.1131446838378904,
      "epoch": 1.6393762183235867,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047368219420389077,
      "loss": 5.8456,
      "mean_token_accuracy": 0.1595255210995674,
      "num_tokens": 32062430.0,
      "step": 16400
    },
    {
      "entropy": 6.014576435089111,
      "epoch": 1.6398760433848154,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004736654423512717,
      "loss": 5.7564,
      "mean_token_accuracy": 0.1703423023223877,
      "num_tokens": 32072396.0,
      "step": 16405
    },
    {
      "entropy": 6.0701933860778805,
      "epoch": 1.6403758684460439,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004736486855002813,
      "loss": 5.8471,
      "mean_token_accuracy": 0.16008925586938857,
      "num_tokens": 32082200.0,
      "step": 16410
    },
    {
      "entropy": 6.055090856552124,
      "epoch": 1.6408756935072724,
      "grad_norm": 1.3125,
      "learning_rate": 0.000473631923651341,
      "loss": 5.7318,
      "mean_token_accuracy": 0.16984516829252244,
      "num_tokens": 32091670.0,
      "step": 16415
    },
    {
      "entropy": 6.110691070556641,
      "epoch": 1.641375518568501,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004736151568048726,
      "loss": 5.892,
      "mean_token_accuracy": 0.15376087948679923,
      "num_tokens": 32102072.0,
      "step": 16420
    },
    {
      "entropy": 6.044223165512085,
      "epoch": 1.6418753436297298,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000473598384961298,
      "loss": 5.703,
      "mean_token_accuracy": 0.1686980292201042,
      "num_tokens": 32111367.0,
      "step": 16425
    },
    {
      "entropy": 6.110084915161133,
      "epoch": 1.642375168690958,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047358160812103925,
      "loss": 5.7822,
      "mean_token_accuracy": 0.1648818626999855,
      "num_tokens": 32121065.0,
      "step": 16430
    },
    {
      "entropy": 6.109892749786377,
      "epoch": 1.6428749937521867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004735648262845182,
      "loss": 5.8038,
      "mean_token_accuracy": 0.16128632724285125,
      "num_tokens": 32130142.0,
      "step": 16435
    },
    {
      "entropy": 6.0779807567596436,
      "epoch": 1.6433748188134154,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047354803945215727,
      "loss": 5.896,
      "mean_token_accuracy": 0.16313230469822884,
      "num_tokens": 32140689.0,
      "step": 16440
    },
    {
      "entropy": 6.047557830810547,
      "epoch": 1.6438746438746439,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047353124762437865,
      "loss": 5.6772,
      "mean_token_accuracy": 0.18151141852140426,
      "num_tokens": 32150888.0,
      "step": 16445
    },
    {
      "entropy": 6.132944631576538,
      "epoch": 1.6443744689358724,
      "grad_norm": 1.25,
      "learning_rate": 0.0004735144508016051,
      "loss": 5.8956,
      "mean_token_accuracy": 0.16031043380498886,
      "num_tokens": 32161188.0,
      "step": 16450
    },
    {
      "entropy": 6.073975992202759,
      "epoch": 1.644874293997101,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004734976489842589,
      "loss": 5.7677,
      "mean_token_accuracy": 0.16933270394802094,
      "num_tokens": 32171684.0,
      "step": 16455
    },
    {
      "entropy": 6.109004116058349,
      "epoch": 1.6453741190583295,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000473480842172763,
      "loss": 5.8601,
      "mean_token_accuracy": 0.16053954511880875,
      "num_tokens": 32181497.0,
      "step": 16460
    },
    {
      "entropy": 6.073763608932495,
      "epoch": 1.645873944119558,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047346403036754023,
      "loss": 5.7374,
      "mean_token_accuracy": 0.17107273489236832,
      "num_tokens": 32190653.0,
      "step": 16465
    },
    {
      "entropy": 6.063727951049804,
      "epoch": 1.6463737691807867,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047344721356901344,
      "loss": 5.8176,
      "mean_token_accuracy": 0.16836004108190536,
      "num_tokens": 32200367.0,
      "step": 16470
    },
    {
      "entropy": 6.09568772315979,
      "epoch": 1.6468735942420154,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004734303917776059,
      "loss": 5.8606,
      "mean_token_accuracy": 0.1601605385541916,
      "num_tokens": 32210187.0,
      "step": 16475
    },
    {
      "entropy": 6.122977018356323,
      "epoch": 1.6473734193032439,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047341356499374065,
      "loss": 5.7508,
      "mean_token_accuracy": 0.16829127073287964,
      "num_tokens": 32218754.0,
      "step": 16480
    },
    {
      "entropy": 6.057099294662476,
      "epoch": 1.6478732443644724,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004733967332178412,
      "loss": 5.7749,
      "mean_token_accuracy": 0.16628727316856384,
      "num_tokens": 32228105.0,
      "step": 16485
    },
    {
      "entropy": 6.079679489135742,
      "epoch": 1.648373069425701,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004733798964503311,
      "loss": 5.8441,
      "mean_token_accuracy": 0.1639799177646637,
      "num_tokens": 32239303.0,
      "step": 16490
    },
    {
      "entropy": 6.023457813262939,
      "epoch": 1.6488728944869295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004733630546916338,
      "loss": 5.6812,
      "mean_token_accuracy": 0.1760495275259018,
      "num_tokens": 32248614.0,
      "step": 16495
    },
    {
      "entropy": 6.040826797485352,
      "epoch": 1.649372719548158,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047334620794217306,
      "loss": 5.8372,
      "mean_token_accuracy": 0.16869492679834366,
      "num_tokens": 32257012.0,
      "step": 16500
    },
    {
      "entropy": 6.084132862091065,
      "epoch": 1.6498725446093867,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004733293562023727,
      "loss": 5.7642,
      "mean_token_accuracy": 0.17101866006851196,
      "num_tokens": 32267990.0,
      "step": 16505
    },
    {
      "entropy": 6.063880920410156,
      "epoch": 1.6503723696706154,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047331249947265685,
      "loss": 5.7289,
      "mean_token_accuracy": 0.1686910420656204,
      "num_tokens": 32277519.0,
      "step": 16510
    },
    {
      "entropy": 6.060239601135254,
      "epoch": 1.6508721947318439,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004732956377534496,
      "loss": 5.7731,
      "mean_token_accuracy": 0.16132729947566987,
      "num_tokens": 32286749.0,
      "step": 16515
    },
    {
      "entropy": 6.124370288848877,
      "epoch": 1.6513720197930724,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000473278771045175,
      "loss": 5.9312,
      "mean_token_accuracy": 0.16101037263870238,
      "num_tokens": 32296344.0,
      "step": 16520
    },
    {
      "entropy": 6.0963194370269775,
      "epoch": 1.651871844854301,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047326189934825764,
      "loss": 5.752,
      "mean_token_accuracy": 0.1679437443614006,
      "num_tokens": 32305857.0,
      "step": 16525
    },
    {
      "entropy": 6.059949779510498,
      "epoch": 1.6523716699155295,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004732450226631219,
      "loss": 5.7836,
      "mean_token_accuracy": 0.17067249417304992,
      "num_tokens": 32315265.0,
      "step": 16530
    },
    {
      "entropy": 6.181730890274048,
      "epoch": 1.652871494976758,
      "grad_norm": 1.125,
      "learning_rate": 0.00047322814099019236,
      "loss": 5.9343,
      "mean_token_accuracy": 0.15583020597696304,
      "num_tokens": 32324724.0,
      "step": 16535
    },
    {
      "entropy": 6.1046144485473635,
      "epoch": 1.6533713200379867,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004732112543298938,
      "loss": 5.8064,
      "mean_token_accuracy": 0.16409349888563157,
      "num_tokens": 32334958.0,
      "step": 16540
    },
    {
      "entropy": 6.079411411285401,
      "epoch": 1.6538711450992154,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004731943626826511,
      "loss": 5.8589,
      "mean_token_accuracy": 0.1644941344857216,
      "num_tokens": 32345153.0,
      "step": 16545
    },
    {
      "entropy": 6.074976968765259,
      "epoch": 1.6543709701604439,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047317746604888923,
      "loss": 5.7493,
      "mean_token_accuracy": 0.16529782712459565,
      "num_tokens": 32354692.0,
      "step": 16550
    },
    {
      "entropy": 6.081219720840454,
      "epoch": 1.6548707952216724,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047316056442903333,
      "loss": 5.8441,
      "mean_token_accuracy": 0.15971016734838486,
      "num_tokens": 32363614.0,
      "step": 16555
    },
    {
      "entropy": 6.104169702529907,
      "epoch": 1.655370620282901,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004731436578235086,
      "loss": 5.8003,
      "mean_token_accuracy": 0.16548737436532973,
      "num_tokens": 32373139.0,
      "step": 16560
    },
    {
      "entropy": 6.099363470077515,
      "epoch": 1.6558704453441295,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047312674623274046,
      "loss": 5.8371,
      "mean_token_accuracy": 0.1591850131750107,
      "num_tokens": 32383624.0,
      "step": 16565
    },
    {
      "entropy": 6.033934736251831,
      "epoch": 1.656370270405358,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004731098296571543,
      "loss": 5.6483,
      "mean_token_accuracy": 0.17401236444711685,
      "num_tokens": 32393221.0,
      "step": 16570
    },
    {
      "entropy": 6.087524890899658,
      "epoch": 1.6568700954665867,
      "grad_norm": 1.125,
      "learning_rate": 0.00047309290809717584,
      "loss": 5.8424,
      "mean_token_accuracy": 0.16399402916431427,
      "num_tokens": 32402908.0,
      "step": 16575
    },
    {
      "entropy": 6.040187788009644,
      "epoch": 1.6573699205278154,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047307598155323083,
      "loss": 5.7939,
      "mean_token_accuracy": 0.16343969702720643,
      "num_tokens": 32413597.0,
      "step": 16580
    },
    {
      "entropy": 6.086679744720459,
      "epoch": 1.6578697455890439,
      "grad_norm": 1.21875,
      "learning_rate": 0.000473059050025745,
      "loss": 5.8139,
      "mean_token_accuracy": 0.16954085677862168,
      "num_tokens": 32423948.0,
      "step": 16585
    },
    {
      "entropy": 6.079872989654541,
      "epoch": 1.6583695706502724,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004730421135151445,
      "loss": 5.7743,
      "mean_token_accuracy": 0.1648908570408821,
      "num_tokens": 32433176.0,
      "step": 16590
    },
    {
      "entropy": 6.007696342468262,
      "epoch": 1.658869395711501,
      "grad_norm": 1.125,
      "learning_rate": 0.0004730251720218554,
      "loss": 5.7518,
      "mean_token_accuracy": 0.16767329722642899,
      "num_tokens": 32442868.0,
      "step": 16595
    },
    {
      "entropy": 6.045664548873901,
      "epoch": 1.6593692207727295,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047300822554630393,
      "loss": 5.7935,
      "mean_token_accuracy": 0.16907393634319307,
      "num_tokens": 32451966.0,
      "step": 16600
    },
    {
      "entropy": 6.093768119812012,
      "epoch": 1.659869045833958,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004729912740889164,
      "loss": 5.7763,
      "mean_token_accuracy": 0.17338257879018784,
      "num_tokens": 32462442.0,
      "step": 16605
    },
    {
      "entropy": 6.060785579681396,
      "epoch": 1.6603688708951867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047297431765011945,
      "loss": 5.7254,
      "mean_token_accuracy": 0.17285429388284684,
      "num_tokens": 32471709.0,
      "step": 16610
    },
    {
      "entropy": 6.0325689792633055,
      "epoch": 1.6608686959564154,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047295735623033953,
      "loss": 5.7975,
      "mean_token_accuracy": 0.1717447131872177,
      "num_tokens": 32482147.0,
      "step": 16615
    },
    {
      "entropy": 6.048233127593994,
      "epoch": 1.6613685210176439,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004729403898300035,
      "loss": 5.7063,
      "mean_token_accuracy": 0.17334574609994888,
      "num_tokens": 32492626.0,
      "step": 16620
    },
    {
      "entropy": 6.026914834976196,
      "epoch": 1.6618683460788723,
      "grad_norm": 1.25,
      "learning_rate": 0.00047292341844953813,
      "loss": 5.7539,
      "mean_token_accuracy": 0.17628212571144103,
      "num_tokens": 32502887.0,
      "step": 16625
    },
    {
      "entropy": 6.0966778755187985,
      "epoch": 1.662368171140101,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004729064420893706,
      "loss": 5.8656,
      "mean_token_accuracy": 0.1628693923354149,
      "num_tokens": 32512860.0,
      "step": 16630
    },
    {
      "entropy": 6.1373497486114506,
      "epoch": 1.6628679962013295,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004728894607499278,
      "loss": 5.9234,
      "mean_token_accuracy": 0.15375887602567673,
      "num_tokens": 32523103.0,
      "step": 16635
    },
    {
      "entropy": 5.990721082687378,
      "epoch": 1.663367821262558,
      "grad_norm": 1.125,
      "learning_rate": 0.0004728724744316371,
      "loss": 5.7049,
      "mean_token_accuracy": 0.17155631482601166,
      "num_tokens": 32532940.0,
      "step": 16640
    },
    {
      "entropy": 6.1484301567077635,
      "epoch": 1.6638676463237867,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047285548313492594,
      "loss": 5.8505,
      "mean_token_accuracy": 0.16416150480508804,
      "num_tokens": 32543503.0,
      "step": 16645
    },
    {
      "entropy": 6.125384998321533,
      "epoch": 1.6643674713850154,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004728384868602217,
      "loss": 5.7194,
      "mean_token_accuracy": 0.17565450370311736,
      "num_tokens": 32552025.0,
      "step": 16650
    },
    {
      "entropy": 6.002880191802978,
      "epoch": 1.6648672964462436,
      "grad_norm": 1.234375,
      "learning_rate": 0.000472821485607952,
      "loss": 5.7146,
      "mean_token_accuracy": 0.1669606864452362,
      "num_tokens": 32561262.0,
      "step": 16655
    },
    {
      "entropy": 6.0140159130096436,
      "epoch": 1.6653671215074723,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004728044793785446,
      "loss": 5.8067,
      "mean_token_accuracy": 0.16245571225881578,
      "num_tokens": 32572067.0,
      "step": 16660
    },
    {
      "entropy": 6.116350555419922,
      "epoch": 1.665866946568701,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047278746817242754,
      "loss": 5.7985,
      "mean_token_accuracy": 0.17317099124193192,
      "num_tokens": 32582033.0,
      "step": 16665
    },
    {
      "entropy": 6.056389284133911,
      "epoch": 1.6663667716299295,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004727704519900286,
      "loss": 5.7696,
      "mean_token_accuracy": 0.16686231046915054,
      "num_tokens": 32592338.0,
      "step": 16670
    },
    {
      "entropy": 6.013908910751343,
      "epoch": 1.666866596691158,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000472753430831776,
      "loss": 5.7567,
      "mean_token_accuracy": 0.1658183217048645,
      "num_tokens": 32602310.0,
      "step": 16675
    },
    {
      "entropy": 6.210496091842652,
      "epoch": 1.6673664217523867,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047273640469809786,
      "loss": 5.8666,
      "mean_token_accuracy": 0.16141892373561859,
      "num_tokens": 32612632.0,
      "step": 16680
    },
    {
      "entropy": 6.073301887512207,
      "epoch": 1.6678662468136154,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004727193735894227,
      "loss": 5.7711,
      "mean_token_accuracy": 0.16707826256752015,
      "num_tokens": 32622455.0,
      "step": 16685
    },
    {
      "entropy": 5.96614146232605,
      "epoch": 1.6683660718748436,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047270233750617907,
      "loss": 5.7559,
      "mean_token_accuracy": 0.1708046868443489,
      "num_tokens": 32631854.0,
      "step": 16690
    },
    {
      "entropy": 6.020282363891601,
      "epoch": 1.6688658969360723,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047268529644879545,
      "loss": 5.824,
      "mean_token_accuracy": 0.16722732186317443,
      "num_tokens": 32641532.0,
      "step": 16695
    },
    {
      "entropy": 6.079303836822509,
      "epoch": 1.669365721997301,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047266825041770066,
      "loss": 5.7923,
      "mean_token_accuracy": 0.16729707270860672,
      "num_tokens": 32650748.0,
      "step": 16700
    },
    {
      "entropy": 6.054289197921753,
      "epoch": 1.6698655470585295,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004726511994133235,
      "loss": 5.7996,
      "mean_token_accuracy": 0.16691591292619706,
      "num_tokens": 32660828.0,
      "step": 16705
    },
    {
      "entropy": 6.028136491775513,
      "epoch": 1.670365372119758,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047263414343609307,
      "loss": 5.8115,
      "mean_token_accuracy": 0.16619486659765242,
      "num_tokens": 32670951.0,
      "step": 16710
    },
    {
      "entropy": 6.075101041793824,
      "epoch": 1.6708651971809867,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047261708248643846,
      "loss": 5.7942,
      "mean_token_accuracy": 0.1668395772576332,
      "num_tokens": 32680977.0,
      "step": 16715
    },
    {
      "entropy": 6.16903076171875,
      "epoch": 1.6713650222422152,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004726000165647888,
      "loss": 5.8139,
      "mean_token_accuracy": 0.17002842128276824,
      "num_tokens": 32690004.0,
      "step": 16720
    },
    {
      "entropy": 6.104844617843628,
      "epoch": 1.6718648473034436,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004725829456715737,
      "loss": 5.7956,
      "mean_token_accuracy": 0.1605338543653488,
      "num_tokens": 32699862.0,
      "step": 16725
    },
    {
      "entropy": 6.06551308631897,
      "epoch": 1.6723646723646723,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004725658698072225,
      "loss": 5.7834,
      "mean_token_accuracy": 0.1644750103354454,
      "num_tokens": 32708732.0,
      "step": 16730
    },
    {
      "entropy": 6.106829977035522,
      "epoch": 1.672864497425901,
      "grad_norm": 1.125,
      "learning_rate": 0.0004725487889721648,
      "loss": 5.9078,
      "mean_token_accuracy": 0.1552678629755974,
      "num_tokens": 32719117.0,
      "step": 16735
    },
    {
      "entropy": 6.067719173431397,
      "epoch": 1.6733643224871295,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004725317031668304,
      "loss": 5.7648,
      "mean_token_accuracy": 0.16359293907880784,
      "num_tokens": 32728357.0,
      "step": 16740
    },
    {
      "entropy": 6.102327442169189,
      "epoch": 1.673864147548358,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047251461239164916,
      "loss": 5.8002,
      "mean_token_accuracy": 0.1623322769999504,
      "num_tokens": 32738249.0,
      "step": 16745
    },
    {
      "entropy": 5.9564512252807615,
      "epoch": 1.6743639726095867,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004724975166470512,
      "loss": 5.7154,
      "mean_token_accuracy": 0.1735158696770668,
      "num_tokens": 32747880.0,
      "step": 16750
    },
    {
      "entropy": 6.029469299316406,
      "epoch": 1.6748637976708152,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004724804159334665,
      "loss": 5.7107,
      "mean_token_accuracy": 0.16456150710582734,
      "num_tokens": 32758215.0,
      "step": 16755
    },
    {
      "entropy": 6.110155439376831,
      "epoch": 1.6753636227320436,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004724633102513253,
      "loss": 5.8578,
      "mean_token_accuracy": 0.15659331083297728,
      "num_tokens": 32767426.0,
      "step": 16760
    },
    {
      "entropy": 6.015844202041626,
      "epoch": 1.6758634477932723,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004724461996010581,
      "loss": 5.7477,
      "mean_token_accuracy": 0.1673281028866768,
      "num_tokens": 32776507.0,
      "step": 16765
    },
    {
      "entropy": 6.000391149520874,
      "epoch": 1.676363272854501,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004724290839830953,
      "loss": 5.7566,
      "mean_token_accuracy": 0.17151677906513213,
      "num_tokens": 32787386.0,
      "step": 16770
    },
    {
      "entropy": 6.0641608238220215,
      "epoch": 1.6768630979157295,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004724119633978675,
      "loss": 5.7927,
      "mean_token_accuracy": 0.16742006987333297,
      "num_tokens": 32797245.0,
      "step": 16775
    },
    {
      "entropy": 6.081221055984497,
      "epoch": 1.677362922976958,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047239483784580554,
      "loss": 5.8132,
      "mean_token_accuracy": 0.1690059259533882,
      "num_tokens": 32807256.0,
      "step": 16780
    },
    {
      "entropy": 6.085030603408813,
      "epoch": 1.6778627480381867,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004723777073273403,
      "loss": 5.823,
      "mean_token_accuracy": 0.1652229130268097,
      "num_tokens": 32816818.0,
      "step": 16785
    },
    {
      "entropy": 6.046489906311035,
      "epoch": 1.6783625730994152,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004723605718429026,
      "loss": 5.8382,
      "mean_token_accuracy": 0.16240782886743546,
      "num_tokens": 32826275.0,
      "step": 16790
    },
    {
      "entropy": 6.007450819015503,
      "epoch": 1.6788623981606436,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004723434313929238,
      "loss": 5.7816,
      "mean_token_accuracy": 0.16952600628137587,
      "num_tokens": 32836703.0,
      "step": 16795
    },
    {
      "entropy": 6.010953569412232,
      "epoch": 1.6793622232218723,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047232628597783507,
      "loss": 5.6892,
      "mean_token_accuracy": 0.17593244314193726,
      "num_tokens": 32847745.0,
      "step": 16800
    },
    {
      "entropy": 6.120220947265625,
      "epoch": 1.679862048283101,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047230913559806773,
      "loss": 5.7623,
      "mean_token_accuracy": 0.16712561398744583,
      "num_tokens": 32857123.0,
      "step": 16805
    },
    {
      "entropy": 6.014214086532593,
      "epoch": 1.6803618733443295,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004722919802540533,
      "loss": 5.722,
      "mean_token_accuracy": 0.167767234146595,
      "num_tokens": 32866337.0,
      "step": 16810
    },
    {
      "entropy": 6.019205236434937,
      "epoch": 1.680861698405558,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004722748199462234,
      "loss": 5.8397,
      "mean_token_accuracy": 0.1698933720588684,
      "num_tokens": 32877525.0,
      "step": 16815
    },
    {
      "entropy": 6.138436794281006,
      "epoch": 1.6813615234667867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004722576546750098,
      "loss": 5.8176,
      "mean_token_accuracy": 0.1701180726289749,
      "num_tokens": 32888165.0,
      "step": 16820
    },
    {
      "entropy": 6.132360506057739,
      "epoch": 1.6818613485280152,
      "grad_norm": 1.703125,
      "learning_rate": 0.00047224048444084434,
      "loss": 5.8356,
      "mean_token_accuracy": 0.16238732039928436,
      "num_tokens": 32898541.0,
      "step": 16825
    },
    {
      "entropy": 5.936971712112427,
      "epoch": 1.6823611735892436,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047222330924415906,
      "loss": 5.6761,
      "mean_token_accuracy": 0.18108348399400712,
      "num_tokens": 32907796.0,
      "step": 16830
    },
    {
      "entropy": 6.016633701324463,
      "epoch": 1.6828609986504723,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004722061290853862,
      "loss": 5.7186,
      "mean_token_accuracy": 0.17361532747745514,
      "num_tokens": 32917067.0,
      "step": 16835
    },
    {
      "entropy": 6.076180171966553,
      "epoch": 1.683360823711701,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047218894396495765,
      "loss": 5.7513,
      "mean_token_accuracy": 0.17127334475517272,
      "num_tokens": 32927039.0,
      "step": 16840
    },
    {
      "entropy": 6.071084690093994,
      "epoch": 1.6838606487729295,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004721717538833061,
      "loss": 5.7725,
      "mean_token_accuracy": 0.16609492599964143,
      "num_tokens": 32937574.0,
      "step": 16845
    },
    {
      "entropy": 6.117663097381592,
      "epoch": 1.684360473834158,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000472154558840864,
      "loss": 5.8578,
      "mean_token_accuracy": 0.16354204565286637,
      "num_tokens": 32948401.0,
      "step": 16850
    },
    {
      "entropy": 6.048561429977417,
      "epoch": 1.6848602988953867,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000472137358838064,
      "loss": 5.7474,
      "mean_token_accuracy": 0.17421502768993377,
      "num_tokens": 32957359.0,
      "step": 16855
    },
    {
      "entropy": 5.986212348937988,
      "epoch": 1.6853601239566152,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004721201538753387,
      "loss": 5.6883,
      "mean_token_accuracy": 0.1782438263297081,
      "num_tokens": 32966340.0,
      "step": 16860
    },
    {
      "entropy": 6.016357278823852,
      "epoch": 1.6858599490178436,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047210294395312103,
      "loss": 5.7404,
      "mean_token_accuracy": 0.1712093695998192,
      "num_tokens": 32976376.0,
      "step": 16865
    },
    {
      "entropy": 5.998928880691528,
      "epoch": 1.6863597740790723,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000472085729071844,
      "loss": 5.7247,
      "mean_token_accuracy": 0.17807436734437943,
      "num_tokens": 32985680.0,
      "step": 16870
    },
    {
      "entropy": 6.00435094833374,
      "epoch": 1.686859599140301,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047206850923194076,
      "loss": 5.7444,
      "mean_token_accuracy": 0.16983980387449266,
      "num_tokens": 32995557.0,
      "step": 16875
    },
    {
      "entropy": 6.092774534225464,
      "epoch": 1.6873594242015295,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047205128443384456,
      "loss": 5.7413,
      "mean_token_accuracy": 0.16904614567756654,
      "num_tokens": 33004655.0,
      "step": 16880
    },
    {
      "entropy": 6.030259227752685,
      "epoch": 1.687859249262758,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004720340546779887,
      "loss": 5.7896,
      "mean_token_accuracy": 0.16942677795886993,
      "num_tokens": 33014621.0,
      "step": 16885
    },
    {
      "entropy": 5.9918159484863285,
      "epoch": 1.6883590743239867,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004720168199648068,
      "loss": 5.7271,
      "mean_token_accuracy": 0.17137011289596557,
      "num_tokens": 33024390.0,
      "step": 16890
    },
    {
      "entropy": 6.0706353187561035,
      "epoch": 1.6888588993852152,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047199958029473237,
      "loss": 5.8874,
      "mean_token_accuracy": 0.16772480905056,
      "num_tokens": 33035351.0,
      "step": 16895
    },
    {
      "entropy": 6.01846489906311,
      "epoch": 1.6893587244464436,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047198233566819915,
      "loss": 5.6619,
      "mean_token_accuracy": 0.1789632558822632,
      "num_tokens": 33044148.0,
      "step": 16900
    },
    {
      "entropy": 5.970454692840576,
      "epoch": 1.6898585495076723,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004719650860856411,
      "loss": 5.7611,
      "mean_token_accuracy": 0.16868236511945725,
      "num_tokens": 33053772.0,
      "step": 16905
    },
    {
      "entropy": 6.070581436157227,
      "epoch": 1.690358374568901,
      "grad_norm": 1.25,
      "learning_rate": 0.00047194783154749217,
      "loss": 5.7689,
      "mean_token_accuracy": 0.17051683515310287,
      "num_tokens": 33064422.0,
      "step": 16910
    },
    {
      "entropy": 6.1852940082550045,
      "epoch": 1.6908581996301293,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004719305720541865,
      "loss": 5.9224,
      "mean_token_accuracy": 0.15259888768196106,
      "num_tokens": 33075235.0,
      "step": 16915
    },
    {
      "entropy": 6.037226247787475,
      "epoch": 1.691358024691358,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047191330760615823,
      "loss": 5.7478,
      "mean_token_accuracy": 0.1711474597454071,
      "num_tokens": 33085489.0,
      "step": 16920
    },
    {
      "entropy": 5.994906425476074,
      "epoch": 1.6918578497525867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047189603820384186,
      "loss": 5.8141,
      "mean_token_accuracy": 0.16649553775787354,
      "num_tokens": 33094932.0,
      "step": 16925
    },
    {
      "entropy": 5.993769025802612,
      "epoch": 1.6923576748138152,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047187876384767187,
      "loss": 5.7472,
      "mean_token_accuracy": 0.17101514637470244,
      "num_tokens": 33104224.0,
      "step": 16930
    },
    {
      "entropy": 6.028055095672608,
      "epoch": 1.6928574998750436,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047186148453808274,
      "loss": 5.7553,
      "mean_token_accuracy": 0.17007557451725006,
      "num_tokens": 33114039.0,
      "step": 16935
    },
    {
      "entropy": 6.064574098587036,
      "epoch": 1.6933573249362723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004718442002755094,
      "loss": 5.7899,
      "mean_token_accuracy": 0.16874147206544876,
      "num_tokens": 33123274.0,
      "step": 16940
    },
    {
      "entropy": 6.078436803817749,
      "epoch": 1.693857149997501,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004718269110603866,
      "loss": 5.8122,
      "mean_token_accuracy": 0.16872053295373918,
      "num_tokens": 33134259.0,
      "step": 16945
    },
    {
      "entropy": 6.1039896488189695,
      "epoch": 1.6943569750587293,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047180961689314937,
      "loss": 5.7678,
      "mean_token_accuracy": 0.17350187301635742,
      "num_tokens": 33144662.0,
      "step": 16950
    },
    {
      "entropy": 6.070421648025513,
      "epoch": 1.694856800119958,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004717923177742329,
      "loss": 5.7654,
      "mean_token_accuracy": 0.16974784582853317,
      "num_tokens": 33154074.0,
      "step": 16955
    },
    {
      "entropy": 6.049105930328369,
      "epoch": 1.6953566251811867,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047177501370407223,
      "loss": 5.7586,
      "mean_token_accuracy": 0.1676055058836937,
      "num_tokens": 33163973.0,
      "step": 16960
    },
    {
      "entropy": 6.085944747924804,
      "epoch": 1.6958564502424152,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047175770468310294,
      "loss": 5.8368,
      "mean_token_accuracy": 0.1623539462685585,
      "num_tokens": 33173607.0,
      "step": 16965
    },
    {
      "entropy": 6.054363012313843,
      "epoch": 1.6963562753036436,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047174039071176037,
      "loss": 5.7116,
      "mean_token_accuracy": 0.17007369101047515,
      "num_tokens": 33183104.0,
      "step": 16970
    },
    {
      "entropy": 6.047815227508545,
      "epoch": 1.6968561003648723,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047172307179048025,
      "loss": 5.7136,
      "mean_token_accuracy": 0.1668689116835594,
      "num_tokens": 33191066.0,
      "step": 16975
    },
    {
      "entropy": 6.029032850265503,
      "epoch": 1.697355925426101,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004717057479196982,
      "loss": 5.7841,
      "mean_token_accuracy": 0.1696281313896179,
      "num_tokens": 33201499.0,
      "step": 16980
    },
    {
      "entropy": 5.971209383010864,
      "epoch": 1.6978557504873293,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004716884190998502,
      "loss": 5.7079,
      "mean_token_accuracy": 0.1707802727818489,
      "num_tokens": 33211989.0,
      "step": 16985
    },
    {
      "entropy": 6.100582838058472,
      "epoch": 1.698355575548558,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004716710853313721,
      "loss": 5.8325,
      "mean_token_accuracy": 0.1669354945421219,
      "num_tokens": 33221715.0,
      "step": 16990
    },
    {
      "entropy": 6.078948879241944,
      "epoch": 1.6988554006097867,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047165374661470023,
      "loss": 5.8004,
      "mean_token_accuracy": 0.15835767090320588,
      "num_tokens": 33232812.0,
      "step": 16995
    },
    {
      "entropy": 6.151984310150146,
      "epoch": 1.6993552256710152,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047163640295027064,
      "loss": 5.7138,
      "mean_token_accuracy": 0.17472400665283203,
      "num_tokens": 33242147.0,
      "step": 17000
    },
    {
      "entropy": 6.029457473754883,
      "epoch": 1.6998550507322436,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047161905433851974,
      "loss": 5.7549,
      "mean_token_accuracy": 0.17216732650995253,
      "num_tokens": 33251433.0,
      "step": 17005
    },
    {
      "entropy": 5.9943153858184814,
      "epoch": 1.7003548757934723,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047160170077988403,
      "loss": 5.6964,
      "mean_token_accuracy": 0.16899217069149017,
      "num_tokens": 33261338.0,
      "step": 17010
    },
    {
      "entropy": 6.025051069259644,
      "epoch": 1.7008547008547008,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004715843422748001,
      "loss": 5.8063,
      "mean_token_accuracy": 0.17069874703884125,
      "num_tokens": 33271312.0,
      "step": 17015
    },
    {
      "entropy": 6.103582096099854,
      "epoch": 1.7013545259159293,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047156697882370474,
      "loss": 5.7923,
      "mean_token_accuracy": 0.1615240290760994,
      "num_tokens": 33280525.0,
      "step": 17020
    },
    {
      "entropy": 6.067063999176026,
      "epoch": 1.701854350977158,
      "grad_norm": 1.375,
      "learning_rate": 0.0004715496104270348,
      "loss": 5.7704,
      "mean_token_accuracy": 0.16922464743256568,
      "num_tokens": 33289968.0,
      "step": 17025
    },
    {
      "entropy": 5.990065813064575,
      "epoch": 1.7023541760383867,
      "grad_norm": 1.125,
      "learning_rate": 0.0004715322370852272,
      "loss": 5.7588,
      "mean_token_accuracy": 0.17030611336231233,
      "num_tokens": 33299262.0,
      "step": 17030
    },
    {
      "entropy": 6.035293817520142,
      "epoch": 1.7028540010996152,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047151485879871907,
      "loss": 5.7579,
      "mean_token_accuracy": 0.16702912598848343,
      "num_tokens": 33309433.0,
      "step": 17035
    },
    {
      "entropy": 6.011049556732178,
      "epoch": 1.7033538261608436,
      "grad_norm": 1.25,
      "learning_rate": 0.00047149747556794763,
      "loss": 5.7181,
      "mean_token_accuracy": 0.1692178040742874,
      "num_tokens": 33318587.0,
      "step": 17040
    },
    {
      "entropy": 6.0463173389434814,
      "epoch": 1.7038536512220723,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047148008739335035,
      "loss": 5.777,
      "mean_token_accuracy": 0.16443521976470948,
      "num_tokens": 33327487.0,
      "step": 17045
    },
    {
      "entropy": 6.0977424621582035,
      "epoch": 1.7043534762833008,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004714626942753646,
      "loss": 5.8491,
      "mean_token_accuracy": 0.16116262078285218,
      "num_tokens": 33337151.0,
      "step": 17050
    },
    {
      "entropy": 6.1172511100769045,
      "epoch": 1.7048533013445293,
      "grad_norm": 1.171875,
      "learning_rate": 0.000471445296214428,
      "loss": 5.7184,
      "mean_token_accuracy": 0.16329462081193924,
      "num_tokens": 33346547.0,
      "step": 17055
    },
    {
      "entropy": 6.030295324325562,
      "epoch": 1.705353126405758,
      "grad_norm": 1.125,
      "learning_rate": 0.00047142789321097834,
      "loss": 5.7223,
      "mean_token_accuracy": 0.17046571522951126,
      "num_tokens": 33355277.0,
      "step": 17060
    },
    {
      "entropy": 6.109853363037109,
      "epoch": 1.7058529514669867,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004714104852654534,
      "loss": 5.7312,
      "mean_token_accuracy": 0.17379533052444457,
      "num_tokens": 33364340.0,
      "step": 17065
    },
    {
      "entropy": 6.084222650527954,
      "epoch": 1.7063527765282152,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004713930723782912,
      "loss": 5.7542,
      "mean_token_accuracy": 0.17032799273729324,
      "num_tokens": 33374302.0,
      "step": 17070
    },
    {
      "entropy": 6.0598396301269535,
      "epoch": 1.7068526015894436,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047137565454992983,
      "loss": 5.8228,
      "mean_token_accuracy": 0.16679605096578598,
      "num_tokens": 33384098.0,
      "step": 17075
    },
    {
      "entropy": 6.032701110839843,
      "epoch": 1.7073524266506723,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047135823178080745,
      "loss": 5.7244,
      "mean_token_accuracy": 0.1726317323744297,
      "num_tokens": 33393032.0,
      "step": 17080
    },
    {
      "entropy": 5.969227075576782,
      "epoch": 1.7078522517119008,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004713408040713626,
      "loss": 5.6946,
      "mean_token_accuracy": 0.16789161115884782,
      "num_tokens": 33402533.0,
      "step": 17085
    },
    {
      "entropy": 6.035128927230835,
      "epoch": 1.7083520767731293,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047132337142203365,
      "loss": 5.7767,
      "mean_token_accuracy": 0.16690108329057693,
      "num_tokens": 33412826.0,
      "step": 17090
    },
    {
      "entropy": 6.09186053276062,
      "epoch": 1.708851901834358,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004713059338332591,
      "loss": 5.7681,
      "mean_token_accuracy": 0.16933013647794723,
      "num_tokens": 33422485.0,
      "step": 17095
    },
    {
      "entropy": 6.02112340927124,
      "epoch": 1.7093517268955867,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047128849130547774,
      "loss": 5.6581,
      "mean_token_accuracy": 0.17699847966432572,
      "num_tokens": 33432575.0,
      "step": 17100
    },
    {
      "entropy": 6.005268573760986,
      "epoch": 1.7098515519568152,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047127104383912856,
      "loss": 5.793,
      "mean_token_accuracy": 0.16773949414491654,
      "num_tokens": 33442893.0,
      "step": 17105
    },
    {
      "entropy": 6.042659139633178,
      "epoch": 1.7103513770180436,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004712535914346504,
      "loss": 5.859,
      "mean_token_accuracy": 0.16437726616859435,
      "num_tokens": 33452080.0,
      "step": 17110
    },
    {
      "entropy": 6.054730176925659,
      "epoch": 1.7108512020792723,
      "grad_norm": 1.125,
      "learning_rate": 0.00047123613409248233,
      "loss": 5.7104,
      "mean_token_accuracy": 0.17712970972061157,
      "num_tokens": 33461721.0,
      "step": 17115
    },
    {
      "entropy": 6.128001880645752,
      "epoch": 1.7113510271405008,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047121867181306365,
      "loss": 5.8035,
      "mean_token_accuracy": 0.15717067793011666,
      "num_tokens": 33470731.0,
      "step": 17120
    },
    {
      "entropy": 6.033674812316894,
      "epoch": 1.7118508522017293,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004712012045968337,
      "loss": 5.7965,
      "mean_token_accuracy": 0.16744285672903061,
      "num_tokens": 33481000.0,
      "step": 17125
    },
    {
      "entropy": 6.015569686889648,
      "epoch": 1.712350677262958,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047118373244423197,
      "loss": 5.7513,
      "mean_token_accuracy": 0.1691114589571953,
      "num_tokens": 33489535.0,
      "step": 17130
    },
    {
      "entropy": 6.050276517868042,
      "epoch": 1.7128505023241867,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047116625535569787,
      "loss": 5.6932,
      "mean_token_accuracy": 0.1658480152487755,
      "num_tokens": 33499746.0,
      "step": 17135
    },
    {
      "entropy": 6.04907193183899,
      "epoch": 1.7133503273854152,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047114877333167133,
      "loss": 5.7645,
      "mean_token_accuracy": 0.1713144972920418,
      "num_tokens": 33508526.0,
      "step": 17140
    },
    {
      "entropy": 6.013023567199707,
      "epoch": 1.7138501524466436,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047113128637259216,
      "loss": 5.748,
      "mean_token_accuracy": 0.16712633669376373,
      "num_tokens": 33518675.0,
      "step": 17145
    },
    {
      "entropy": 6.053994035720825,
      "epoch": 1.7143499775078723,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004711137944789003,
      "loss": 5.8317,
      "mean_token_accuracy": 0.1627560317516327,
      "num_tokens": 33528132.0,
      "step": 17150
    },
    {
      "entropy": 6.005673122406006,
      "epoch": 1.7148498025691008,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047109629765103577,
      "loss": 5.733,
      "mean_token_accuracy": 0.17153792530298234,
      "num_tokens": 33537259.0,
      "step": 17155
    },
    {
      "entropy": 5.950057601928711,
      "epoch": 1.7153496276303293,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047107879588943887,
      "loss": 5.7366,
      "mean_token_accuracy": 0.17360775470733641,
      "num_tokens": 33546189.0,
      "step": 17160
    },
    {
      "entropy": 6.1067835807800295,
      "epoch": 1.715849452691558,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047106128919455,
      "loss": 5.8476,
      "mean_token_accuracy": 0.16031165048480034,
      "num_tokens": 33555392.0,
      "step": 17165
    },
    {
      "entropy": 6.108438110351562,
      "epoch": 1.7163492777527867,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004710437775668094,
      "loss": 5.8573,
      "mean_token_accuracy": 0.16745765209198,
      "num_tokens": 33564766.0,
      "step": 17170
    },
    {
      "entropy": 6.043480062484742,
      "epoch": 1.7168491028140151,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004710262610066579,
      "loss": 5.83,
      "mean_token_accuracy": 0.16587002575397491,
      "num_tokens": 33574680.0,
      "step": 17175
    },
    {
      "entropy": 6.082858467102051,
      "epoch": 1.7173489278752436,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047100873951453617,
      "loss": 5.7818,
      "mean_token_accuracy": 0.16405347287654876,
      "num_tokens": 33583068.0,
      "step": 17180
    },
    {
      "entropy": 6.02089638710022,
      "epoch": 1.7178487529364723,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047099121309088487,
      "loss": 5.7621,
      "mean_token_accuracy": 0.1724606141448021,
      "num_tokens": 33592838.0,
      "step": 17185
    },
    {
      "entropy": 6.059350872039795,
      "epoch": 1.7183485779977008,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047097368173614515,
      "loss": 5.7785,
      "mean_token_accuracy": 0.16714874505996705,
      "num_tokens": 33603627.0,
      "step": 17190
    },
    {
      "entropy": 6.048834133148193,
      "epoch": 1.7188484030589293,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047095614545075807,
      "loss": 5.799,
      "mean_token_accuracy": 0.1661535993218422,
      "num_tokens": 33614604.0,
      "step": 17195
    },
    {
      "entropy": 6.020707845687866,
      "epoch": 1.719348228120158,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004709386042351647,
      "loss": 5.7243,
      "mean_token_accuracy": 0.17270652204751968,
      "num_tokens": 33624604.0,
      "step": 17200
    },
    {
      "entropy": 6.156314849853516,
      "epoch": 1.7198480531813867,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004709210580898065,
      "loss": 5.826,
      "mean_token_accuracy": 0.1645928591489792,
      "num_tokens": 33634842.0,
      "step": 17205
    },
    {
      "entropy": 6.172335290908814,
      "epoch": 1.720347878242615,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004709035070151249,
      "loss": 5.8382,
      "mean_token_accuracy": 0.16492917388677597,
      "num_tokens": 33644829.0,
      "step": 17210
    },
    {
      "entropy": 6.118891191482544,
      "epoch": 1.7208477033038436,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004708859510115614,
      "loss": 5.8737,
      "mean_token_accuracy": 0.15634865537285805,
      "num_tokens": 33653952.0,
      "step": 17215
    },
    {
      "entropy": 6.064244031906128,
      "epoch": 1.7213475283650723,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004708683900795579,
      "loss": 5.8786,
      "mean_token_accuracy": 0.1656796231865883,
      "num_tokens": 33664014.0,
      "step": 17220
    },
    {
      "entropy": 6.067992782592773,
      "epoch": 1.7218473534263008,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047085082421955603,
      "loss": 5.7555,
      "mean_token_accuracy": 0.1697011947631836,
      "num_tokens": 33674575.0,
      "step": 17225
    },
    {
      "entropy": 6.076864910125733,
      "epoch": 1.7223471784875293,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047083325343199786,
      "loss": 5.8142,
      "mean_token_accuracy": 0.17024437934160233,
      "num_tokens": 33684431.0,
      "step": 17230
    },
    {
      "entropy": 6.151990985870361,
      "epoch": 1.722847003548758,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004708156777173254,
      "loss": 5.8554,
      "mean_token_accuracy": 0.16116755306720734,
      "num_tokens": 33693502.0,
      "step": 17235
    },
    {
      "entropy": 6.093337678909302,
      "epoch": 1.7233468286099867,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004707980970759809,
      "loss": 5.8051,
      "mean_token_accuracy": 0.16634645611047744,
      "num_tokens": 33703034.0,
      "step": 17240
    },
    {
      "entropy": 6.047815752029419,
      "epoch": 1.723846653671215,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047078051150840666,
      "loss": 5.7662,
      "mean_token_accuracy": 0.17159664034843444,
      "num_tokens": 33712429.0,
      "step": 17245
    },
    {
      "entropy": 6.065186548233032,
      "epoch": 1.7243464787324436,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004707629210150451,
      "loss": 5.818,
      "mean_token_accuracy": 0.16698980182409287,
      "num_tokens": 33724254.0,
      "step": 17250
    },
    {
      "entropy": 5.914949512481689,
      "epoch": 1.7248463037936723,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004707453255963388,
      "loss": 5.576,
      "mean_token_accuracy": 0.18104408234357833,
      "num_tokens": 33735706.0,
      "step": 17255
    },
    {
      "entropy": 6.04404845237732,
      "epoch": 1.7253461288549008,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047072772525273053,
      "loss": 5.7801,
      "mean_token_accuracy": 0.166812764108181,
      "num_tokens": 33745683.0,
      "step": 17260
    },
    {
      "entropy": 6.093956327438354,
      "epoch": 1.7258459539161293,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047071011998466304,
      "loss": 5.804,
      "mean_token_accuracy": 0.1622059166431427,
      "num_tokens": 33756776.0,
      "step": 17265
    },
    {
      "entropy": 6.044362115859985,
      "epoch": 1.726345778977358,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047069250979257934,
      "loss": 5.7062,
      "mean_token_accuracy": 0.17195480912923813,
      "num_tokens": 33766004.0,
      "step": 17270
    },
    {
      "entropy": 6.085702037811279,
      "epoch": 1.7268456040385864,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047067489467692236,
      "loss": 5.8503,
      "mean_token_accuracy": 0.1626116991043091,
      "num_tokens": 33774709.0,
      "step": 17275
    },
    {
      "entropy": 6.094661855697632,
      "epoch": 1.727345429099815,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004706572746381354,
      "loss": 5.8282,
      "mean_token_accuracy": 0.1695658564567566,
      "num_tokens": 33785032.0,
      "step": 17280
    },
    {
      "entropy": 6.093116092681885,
      "epoch": 1.7278452541610436,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047063964967666177,
      "loss": 5.7469,
      "mean_token_accuracy": 0.1665853038430214,
      "num_tokens": 33795320.0,
      "step": 17285
    },
    {
      "entropy": 6.098775959014892,
      "epoch": 1.7283450792222723,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004706220197929449,
      "loss": 5.8951,
      "mean_token_accuracy": 0.1652766689658165,
      "num_tokens": 33805764.0,
      "step": 17290
    },
    {
      "entropy": 6.072508859634399,
      "epoch": 1.7288449042835008,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004706043849874284,
      "loss": 5.8256,
      "mean_token_accuracy": 0.1613133430480957,
      "num_tokens": 33816045.0,
      "step": 17295
    },
    {
      "entropy": 6.085056495666504,
      "epoch": 1.7293447293447293,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047058674526055585,
      "loss": 5.7874,
      "mean_token_accuracy": 0.16730739623308183,
      "num_tokens": 33825192.0,
      "step": 17300
    },
    {
      "entropy": 6.071237182617187,
      "epoch": 1.729844554405958,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047056910061277115,
      "loss": 5.7844,
      "mean_token_accuracy": 0.1705868661403656,
      "num_tokens": 33835034.0,
      "step": 17305
    },
    {
      "entropy": 6.049752521514892,
      "epoch": 1.7303443794671864,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004705514510445182,
      "loss": 5.8142,
      "mean_token_accuracy": 0.1638649359345436,
      "num_tokens": 33845225.0,
      "step": 17310
    },
    {
      "entropy": 6.063369083404541,
      "epoch": 1.730844204528415,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000470533796556241,
      "loss": 5.7888,
      "mean_token_accuracy": 0.1710229881107807,
      "num_tokens": 33855303.0,
      "step": 17315
    },
    {
      "entropy": 6.051058578491211,
      "epoch": 1.7313440295896436,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004705161371483839,
      "loss": 5.8072,
      "mean_token_accuracy": 0.16436761915683745,
      "num_tokens": 33865336.0,
      "step": 17320
    },
    {
      "entropy": 6.1124499320983885,
      "epoch": 1.7318438546508723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047049847282139095,
      "loss": 5.8492,
      "mean_token_accuracy": 0.16112802922725677,
      "num_tokens": 33875707.0,
      "step": 17325
    },
    {
      "entropy": 6.1614219665527346,
      "epoch": 1.7323436797121008,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047048080357570685,
      "loss": 5.8972,
      "mean_token_accuracy": 0.1599901147186756,
      "num_tokens": 33886649.0,
      "step": 17330
    },
    {
      "entropy": 6.097561073303223,
      "epoch": 1.7328435047733293,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000470463129411776,
      "loss": 5.8328,
      "mean_token_accuracy": 0.16422399431467055,
      "num_tokens": 33896815.0,
      "step": 17335
    },
    {
      "entropy": 6.100197124481201,
      "epoch": 1.733343329834558,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004704454503300431,
      "loss": 5.809,
      "mean_token_accuracy": 0.16327809989452363,
      "num_tokens": 33907481.0,
      "step": 17340
    },
    {
      "entropy": 6.051394271850586,
      "epoch": 1.7338431548957864,
      "grad_norm": 1.15625,
      "learning_rate": 0.000470427766330953,
      "loss": 5.7746,
      "mean_token_accuracy": 0.16979339718818665,
      "num_tokens": 33917011.0,
      "step": 17345
    },
    {
      "entropy": 5.87927393913269,
      "epoch": 1.734342979957015,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004704100774149506,
      "loss": 5.5887,
      "mean_token_accuracy": 0.19167857617139816,
      "num_tokens": 33926528.0,
      "step": 17350
    },
    {
      "entropy": 6.100354623794556,
      "epoch": 1.7348428050182436,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047039238358248086,
      "loss": 5.8563,
      "mean_token_accuracy": 0.1596961185336113,
      "num_tokens": 33936207.0,
      "step": 17355
    },
    {
      "entropy": 6.019531393051148,
      "epoch": 1.7353426300794723,
      "grad_norm": 1.125,
      "learning_rate": 0.000470374684833989,
      "loss": 5.8273,
      "mean_token_accuracy": 0.1669295385479927,
      "num_tokens": 33946632.0,
      "step": 17360
    },
    {
      "entropy": 6.083300352096558,
      "epoch": 1.7358424551407008,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004703569811699205,
      "loss": 5.8307,
      "mean_token_accuracy": 0.16439008265733718,
      "num_tokens": 33956469.0,
      "step": 17365
    },
    {
      "entropy": 6.097750234603882,
      "epoch": 1.7363422802019293,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047033927259072046,
      "loss": 5.7768,
      "mean_token_accuracy": 0.17001710683107377,
      "num_tokens": 33965699.0,
      "step": 17370
    },
    {
      "entropy": 5.989881181716919,
      "epoch": 1.736842105263158,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047032155909683464,
      "loss": 5.7161,
      "mean_token_accuracy": 0.168760222196579,
      "num_tokens": 33975108.0,
      "step": 17375
    },
    {
      "entropy": 6.11153655052185,
      "epoch": 1.7373419303243864,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004703038406887087,
      "loss": 5.86,
      "mean_token_accuracy": 0.15491053834557533,
      "num_tokens": 33986321.0,
      "step": 17380
    },
    {
      "entropy": 6.155897998809815,
      "epoch": 1.737841755385615,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047028611736678837,
      "loss": 5.8678,
      "mean_token_accuracy": 0.15644377022981643,
      "num_tokens": 33996038.0,
      "step": 17385
    },
    {
      "entropy": 6.023533773422241,
      "epoch": 1.7383415804468436,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004702683891315196,
      "loss": 5.7529,
      "mean_token_accuracy": 0.17738939821720123,
      "num_tokens": 34004773.0,
      "step": 17390
    },
    {
      "entropy": 6.044296169281006,
      "epoch": 1.7388414055080723,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047025065598334834,
      "loss": 5.8051,
      "mean_token_accuracy": 0.16762003302574158,
      "num_tokens": 34014249.0,
      "step": 17395
    },
    {
      "entropy": 6.104206323623657,
      "epoch": 1.7393412305693008,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004702329179227208,
      "loss": 5.8901,
      "mean_token_accuracy": 0.15904048979282379,
      "num_tokens": 34025553.0,
      "step": 17400
    },
    {
      "entropy": 6.090550184249878,
      "epoch": 1.7398410556305293,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047021517495008335,
      "loss": 5.8108,
      "mean_token_accuracy": 0.16349001973867416,
      "num_tokens": 34036536.0,
      "step": 17405
    },
    {
      "entropy": 6.096603441238403,
      "epoch": 1.740340880691758,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047019742706588233,
      "loss": 5.8164,
      "mean_token_accuracy": 0.1632094129920006,
      "num_tokens": 34046353.0,
      "step": 17410
    },
    {
      "entropy": 6.106586170196533,
      "epoch": 1.7408407057529864,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004701796742705642,
      "loss": 5.7168,
      "mean_token_accuracy": 0.17502001523971558,
      "num_tokens": 34056172.0,
      "step": 17415
    },
    {
      "entropy": 6.040268898010254,
      "epoch": 1.741340530814215,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047016191656457573,
      "loss": 5.7277,
      "mean_token_accuracy": 0.16902779638767243,
      "num_tokens": 34066313.0,
      "step": 17420
    },
    {
      "entropy": 6.011306953430176,
      "epoch": 1.7418403558754436,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047014415394836374,
      "loss": 5.7367,
      "mean_token_accuracy": 0.17500469088554382,
      "num_tokens": 34075594.0,
      "step": 17425
    },
    {
      "entropy": 6.05975980758667,
      "epoch": 1.7423401809366723,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004701263864223749,
      "loss": 5.7678,
      "mean_token_accuracy": 0.16924577951431274,
      "num_tokens": 34085619.0,
      "step": 17430
    },
    {
      "entropy": 6.063728713989258,
      "epoch": 1.7428400059979008,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004701086139870565,
      "loss": 5.7953,
      "mean_token_accuracy": 0.16432467252016067,
      "num_tokens": 34094626.0,
      "step": 17435
    },
    {
      "entropy": 6.084858560562134,
      "epoch": 1.7433398310591293,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004700908366428556,
      "loss": 5.7575,
      "mean_token_accuracy": 0.1671794682741165,
      "num_tokens": 34104950.0,
      "step": 17440
    },
    {
      "entropy": 6.150067901611328,
      "epoch": 1.743839656120358,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004700730543902194,
      "loss": 5.9178,
      "mean_token_accuracy": 0.15507936477661133,
      "num_tokens": 34114407.0,
      "step": 17445
    },
    {
      "entropy": 6.125154304504394,
      "epoch": 1.7443394811815864,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047005526722959537,
      "loss": 5.8363,
      "mean_token_accuracy": 0.15927685648202897,
      "num_tokens": 34124842.0,
      "step": 17450
    },
    {
      "entropy": 6.0680320262908936,
      "epoch": 1.744839306242815,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047003747516143095,
      "loss": 5.7109,
      "mean_token_accuracy": 0.17666914016008378,
      "num_tokens": 34134265.0,
      "step": 17455
    },
    {
      "entropy": 6.095453310012817,
      "epoch": 1.7453391313040436,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047001967818617385,
      "loss": 5.842,
      "mean_token_accuracy": 0.15618855953216554,
      "num_tokens": 34143915.0,
      "step": 17460
    },
    {
      "entropy": 6.06427149772644,
      "epoch": 1.7458389563652723,
      "grad_norm": 1.25,
      "learning_rate": 0.0004700018763042719,
      "loss": 5.8084,
      "mean_token_accuracy": 0.17402468025684356,
      "num_tokens": 34154869.0,
      "step": 17465
    },
    {
      "entropy": 6.015555953979492,
      "epoch": 1.7463387814265006,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004699840695161728,
      "loss": 5.698,
      "mean_token_accuracy": 0.1753888890147209,
      "num_tokens": 34164494.0,
      "step": 17470
    },
    {
      "entropy": 6.071428632736206,
      "epoch": 1.7468386064877293,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004699662578223248,
      "loss": 5.7763,
      "mean_token_accuracy": 0.1656183734536171,
      "num_tokens": 34173986.0,
      "step": 17475
    },
    {
      "entropy": 6.1805713176727295,
      "epoch": 1.747338431548958,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046994844122317587,
      "loss": 5.9655,
      "mean_token_accuracy": 0.15836663469672202,
      "num_tokens": 34183129.0,
      "step": 17480
    },
    {
      "entropy": 5.997074604034424,
      "epoch": 1.7478382566101864,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046993061971917433,
      "loss": 5.7237,
      "mean_token_accuracy": 0.17336013317108154,
      "num_tokens": 34192559.0,
      "step": 17485
    },
    {
      "entropy": 6.061177730560303,
      "epoch": 1.748338081671415,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046991279331076865,
      "loss": 5.8403,
      "mean_token_accuracy": 0.16499746963381767,
      "num_tokens": 34202599.0,
      "step": 17490
    },
    {
      "entropy": 6.0803831100463865,
      "epoch": 1.7488379067326436,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004698949619984071,
      "loss": 5.8647,
      "mean_token_accuracy": 0.15468012318015098,
      "num_tokens": 34212883.0,
      "step": 17495
    },
    {
      "entropy": 6.0326793670654295,
      "epoch": 1.7493377317938723,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004698771257825385,
      "loss": 5.7569,
      "mean_token_accuracy": 0.17627610862255097,
      "num_tokens": 34222019.0,
      "step": 17500
    },
    {
      "entropy": 5.998229551315307,
      "epoch": 1.7498375568551006,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004698592846636116,
      "loss": 5.7018,
      "mean_token_accuracy": 0.17559410631656647,
      "num_tokens": 34231817.0,
      "step": 17505
    },
    {
      "entropy": 6.096578979492188,
      "epoch": 1.7503373819163293,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004698414386420752,
      "loss": 5.8296,
      "mean_token_accuracy": 0.1591603636741638,
      "num_tokens": 34242035.0,
      "step": 17510
    },
    {
      "entropy": 6.06802887916565,
      "epoch": 1.750837206977558,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004698235877183784,
      "loss": 5.761,
      "mean_token_accuracy": 0.17255861759185792,
      "num_tokens": 34251728.0,
      "step": 17515
    },
    {
      "entropy": 5.998467206954956,
      "epoch": 1.7513370320387864,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046980573189297017,
      "loss": 5.8176,
      "mean_token_accuracy": 0.1632327914237976,
      "num_tokens": 34262036.0,
      "step": 17520
    },
    {
      "entropy": 6.037278079986573,
      "epoch": 1.751836857100015,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004697878711662999,
      "loss": 5.8145,
      "mean_token_accuracy": 0.16549225747585297,
      "num_tokens": 34272335.0,
      "step": 17525
    },
    {
      "entropy": 6.097747659683227,
      "epoch": 1.7523366821612436,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046977000553881696,
      "loss": 5.7743,
      "mean_token_accuracy": 0.16070619523525237,
      "num_tokens": 34282931.0,
      "step": 17530
    },
    {
      "entropy": 6.089384317398071,
      "epoch": 1.752836507222472,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046975213501097076,
      "loss": 5.766,
      "mean_token_accuracy": 0.16849926710128785,
      "num_tokens": 34293716.0,
      "step": 17535
    },
    {
      "entropy": 6.085835695266724,
      "epoch": 1.7533363322837006,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004697342595832109,
      "loss": 5.8682,
      "mean_token_accuracy": 0.15758422166109085,
      "num_tokens": 34303411.0,
      "step": 17540
    },
    {
      "entropy": 6.126341819763184,
      "epoch": 1.7538361573449293,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046971637925598725,
      "loss": 5.7777,
      "mean_token_accuracy": 0.16717239022254943,
      "num_tokens": 34313092.0,
      "step": 17545
    },
    {
      "entropy": 6.082841157913208,
      "epoch": 1.754335982406158,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004696984940297496,
      "loss": 5.7248,
      "mean_token_accuracy": 0.17098990082740784,
      "num_tokens": 34321909.0,
      "step": 17550
    },
    {
      "entropy": 6.166890096664429,
      "epoch": 1.7548358074673864,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000469680603904948,
      "loss": 5.8677,
      "mean_token_accuracy": 0.15985916703939437,
      "num_tokens": 34330213.0,
      "step": 17555
    },
    {
      "entropy": 6.115846729278564,
      "epoch": 1.755335632528615,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046966270888203237,
      "loss": 5.7946,
      "mean_token_accuracy": 0.16652001738548278,
      "num_tokens": 34339731.0,
      "step": 17560
    },
    {
      "entropy": 6.032681226730347,
      "epoch": 1.7558354575898436,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004696448089614532,
      "loss": 5.7184,
      "mean_token_accuracy": 0.17316577285528184,
      "num_tokens": 34348620.0,
      "step": 17565
    },
    {
      "entropy": 6.019125270843506,
      "epoch": 1.756335282651072,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004696269041436606,
      "loss": 5.7138,
      "mean_token_accuracy": 0.17194076925516127,
      "num_tokens": 34358292.0,
      "step": 17570
    },
    {
      "entropy": 6.044404554367065,
      "epoch": 1.7568351077123006,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004696089944291052,
      "loss": 5.8067,
      "mean_token_accuracy": 0.17161548733711243,
      "num_tokens": 34367379.0,
      "step": 17575
    },
    {
      "entropy": 6.126109504699707,
      "epoch": 1.7573349327735293,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004695910798182377,
      "loss": 5.8388,
      "mean_token_accuracy": 0.15645819753408433,
      "num_tokens": 34376422.0,
      "step": 17580
    },
    {
      "entropy": 6.114535999298096,
      "epoch": 1.757834757834758,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004695731603115086,
      "loss": 5.8711,
      "mean_token_accuracy": 0.16654087007045745,
      "num_tokens": 34386509.0,
      "step": 17585
    },
    {
      "entropy": 6.101844692230225,
      "epoch": 1.7583345828959864,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046955523590936884,
      "loss": 5.8445,
      "mean_token_accuracy": 0.16042127460241318,
      "num_tokens": 34396070.0,
      "step": 17590
    },
    {
      "entropy": 6.109809398651123,
      "epoch": 1.758834407957215,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004695373066122694,
      "loss": 5.7713,
      "mean_token_accuracy": 0.16556950509548188,
      "num_tokens": 34405707.0,
      "step": 17595
    },
    {
      "entropy": 6.066991376876831,
      "epoch": 1.7593342330184436,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004695193724206615,
      "loss": 5.7777,
      "mean_token_accuracy": 0.16707913875579833,
      "num_tokens": 34415621.0,
      "step": 17600
    },
    {
      "entropy": 6.1149860382080075,
      "epoch": 1.759834058079672,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046950143333499613,
      "loss": 5.8395,
      "mean_token_accuracy": 0.1616729035973549,
      "num_tokens": 34425115.0,
      "step": 17605
    },
    {
      "entropy": 6.104553842544556,
      "epoch": 1.7603338831409006,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046948348935572476,
      "loss": 5.81,
      "mean_token_accuracy": 0.16118673086166382,
      "num_tokens": 34434164.0,
      "step": 17610
    },
    {
      "entropy": 6.019186019897461,
      "epoch": 1.7608337082021293,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004694655404832988,
      "loss": 5.6586,
      "mean_token_accuracy": 0.17374128252267837,
      "num_tokens": 34443620.0,
      "step": 17615
    },
    {
      "entropy": 6.117373514175415,
      "epoch": 1.761333533263358,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046944758671817,
      "loss": 5.791,
      "mean_token_accuracy": 0.16575043350458146,
      "num_tokens": 34452831.0,
      "step": 17620
    },
    {
      "entropy": 6.000422716140747,
      "epoch": 1.7618333583245864,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004694296280607899,
      "loss": 5.6372,
      "mean_token_accuracy": 0.17604150772094726,
      "num_tokens": 34462484.0,
      "step": 17625
    },
    {
      "entropy": 5.970747566223144,
      "epoch": 1.762333183385815,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004694116645116104,
      "loss": 5.709,
      "mean_token_accuracy": 0.1764972612261772,
      "num_tokens": 34471546.0,
      "step": 17630
    },
    {
      "entropy": 6.142988824844361,
      "epoch": 1.7628330084470436,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004693936960710834,
      "loss": 5.8209,
      "mean_token_accuracy": 0.16325144618749618,
      "num_tokens": 34482384.0,
      "step": 17635
    },
    {
      "entropy": 6.06432957649231,
      "epoch": 1.763332833508272,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046937572273966104,
      "loss": 5.6845,
      "mean_token_accuracy": 0.18174973577260972,
      "num_tokens": 34491897.0,
      "step": 17640
    },
    {
      "entropy": 6.022173023223877,
      "epoch": 1.7638326585695006,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046935774451779556,
      "loss": 5.7709,
      "mean_token_accuracy": 0.16397448182106017,
      "num_tokens": 34501709.0,
      "step": 17645
    },
    {
      "entropy": 6.043772792816162,
      "epoch": 1.7643324836307293,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004693397614059392,
      "loss": 5.6973,
      "mean_token_accuracy": 0.17369380295276643,
      "num_tokens": 34510560.0,
      "step": 17650
    },
    {
      "entropy": 6.021186351776123,
      "epoch": 1.764832308691958,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004693217734045445,
      "loss": 5.7761,
      "mean_token_accuracy": 0.1667092427611351,
      "num_tokens": 34519363.0,
      "step": 17655
    },
    {
      "entropy": 5.985272169113159,
      "epoch": 1.7653321337531864,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046930378051406393,
      "loss": 5.7569,
      "mean_token_accuracy": 0.17549827098846435,
      "num_tokens": 34530335.0,
      "step": 17660
    },
    {
      "entropy": 5.9891924381256105,
      "epoch": 1.765831958814415,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046928578273495026,
      "loss": 5.686,
      "mean_token_accuracy": 0.17259572297334672,
      "num_tokens": 34540603.0,
      "step": 17665
    },
    {
      "entropy": 6.0426530838012695,
      "epoch": 1.7663317838756436,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004692677800676563,
      "loss": 5.7123,
      "mean_token_accuracy": 0.17691874206066133,
      "num_tokens": 34550765.0,
      "step": 17670
    },
    {
      "entropy": 6.118461656570434,
      "epoch": 1.766831608936872,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046924977251263495,
      "loss": 5.7441,
      "mean_token_accuracy": 0.17046940177679062,
      "num_tokens": 34560330.0,
      "step": 17675
    },
    {
      "entropy": 6.023738861083984,
      "epoch": 1.7673314339981006,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046923176007033937,
      "loss": 5.7807,
      "mean_token_accuracy": 0.17085849791765212,
      "num_tokens": 34569356.0,
      "step": 17680
    },
    {
      "entropy": 6.0324944972991945,
      "epoch": 1.7678312590593293,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046921374274122265,
      "loss": 5.7286,
      "mean_token_accuracy": 0.17466444969177247,
      "num_tokens": 34578613.0,
      "step": 17685
    },
    {
      "entropy": 5.967509937286377,
      "epoch": 1.768331084120558,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046919572052573814,
      "loss": 5.6394,
      "mean_token_accuracy": 0.1790156990289688,
      "num_tokens": 34587394.0,
      "step": 17690
    },
    {
      "entropy": 5.989468240737915,
      "epoch": 1.7688309091817864,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046917769342433925,
      "loss": 5.761,
      "mean_token_accuracy": 0.17278807163238524,
      "num_tokens": 34598707.0,
      "step": 17695
    },
    {
      "entropy": 6.087161588668823,
      "epoch": 1.769330734243015,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004691596614374796,
      "loss": 5.783,
      "mean_token_accuracy": 0.1656171515583992,
      "num_tokens": 34608270.0,
      "step": 17700
    },
    {
      "entropy": 6.091655683517456,
      "epoch": 1.7698305593042436,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004691416245656128,
      "loss": 5.7994,
      "mean_token_accuracy": 0.169183861464262,
      "num_tokens": 34618533.0,
      "step": 17705
    },
    {
      "entropy": 6.141868019104004,
      "epoch": 1.770330384365472,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046912358280919267,
      "loss": 5.8481,
      "mean_token_accuracy": 0.16292693614959716,
      "num_tokens": 34628158.0,
      "step": 17710
    },
    {
      "entropy": 6.083349227905273,
      "epoch": 1.7708302094267006,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046910553616867314,
      "loss": 5.7075,
      "mean_token_accuracy": 0.1750742405653,
      "num_tokens": 34638046.0,
      "step": 17715
    },
    {
      "entropy": 6.091694927215576,
      "epoch": 1.7713300344879293,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004690874846445083,
      "loss": 5.8279,
      "mean_token_accuracy": 0.16213375478982925,
      "num_tokens": 34647847.0,
      "step": 17720
    },
    {
      "entropy": 6.057362699508667,
      "epoch": 1.771829859549158,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004690694282371523,
      "loss": 5.7296,
      "mean_token_accuracy": 0.16664013415575027,
      "num_tokens": 34658830.0,
      "step": 17725
    },
    {
      "entropy": 6.026039886474609,
      "epoch": 1.7723296846103862,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004690513669470594,
      "loss": 5.6975,
      "mean_token_accuracy": 0.17722171545028687,
      "num_tokens": 34668679.0,
      "step": 17730
    },
    {
      "entropy": 6.072541856765747,
      "epoch": 1.772829509671615,
      "grad_norm": 1.171875,
      "learning_rate": 0.000469033300774684,
      "loss": 5.7615,
      "mean_token_accuracy": 0.1721448391675949,
      "num_tokens": 34678443.0,
      "step": 17735
    },
    {
      "entropy": 6.134245729446411,
      "epoch": 1.7733293347328436,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004690152297204808,
      "loss": 5.7955,
      "mean_token_accuracy": 0.1678522601723671,
      "num_tokens": 34688348.0,
      "step": 17740
    },
    {
      "entropy": 5.844039964675903,
      "epoch": 1.773829159794072,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046899715378490427,
      "loss": 5.4968,
      "mean_token_accuracy": 0.1941676989197731,
      "num_tokens": 34698402.0,
      "step": 17745
    },
    {
      "entropy": 5.988340663909912,
      "epoch": 1.7743289848553006,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046897907296840927,
      "loss": 5.726,
      "mean_token_accuracy": 0.17130696028470993,
      "num_tokens": 34707761.0,
      "step": 17750
    },
    {
      "entropy": 6.078943490982056,
      "epoch": 1.7748288099165292,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004689609872714507,
      "loss": 5.8412,
      "mean_token_accuracy": 0.16632397770881652,
      "num_tokens": 34717934.0,
      "step": 17755
    },
    {
      "entropy": 6.16288194656372,
      "epoch": 1.775328634977758,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046894289669448366,
      "loss": 5.8245,
      "mean_token_accuracy": 0.1678251400589943,
      "num_tokens": 34727486.0,
      "step": 17760
    },
    {
      "entropy": 6.078250980377197,
      "epoch": 1.7758284600389862,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046892480123796314,
      "loss": 5.7412,
      "mean_token_accuracy": 0.17410959899425507,
      "num_tokens": 34738080.0,
      "step": 17765
    },
    {
      "entropy": 6.022730302810669,
      "epoch": 1.776328285100215,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004689067009023446,
      "loss": 5.7943,
      "mean_token_accuracy": 0.1639723464846611,
      "num_tokens": 34747595.0,
      "step": 17770
    },
    {
      "entropy": 5.994482517242432,
      "epoch": 1.7768281101614436,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004688885956880834,
      "loss": 5.6111,
      "mean_token_accuracy": 0.17979071736335756,
      "num_tokens": 34756818.0,
      "step": 17775
    },
    {
      "entropy": 6.00158543586731,
      "epoch": 1.777327935222672,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004688704855956349,
      "loss": 5.7231,
      "mean_token_accuracy": 0.16988227218389512,
      "num_tokens": 34767263.0,
      "step": 17780
    },
    {
      "entropy": 6.135664701461792,
      "epoch": 1.7778277602839005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046885237062545493,
      "loss": 5.8118,
      "mean_token_accuracy": 0.1637120723724365,
      "num_tokens": 34776767.0,
      "step": 17785
    },
    {
      "entropy": 6.068994617462158,
      "epoch": 1.7783275853451292,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046883425077799923,
      "loss": 5.7458,
      "mean_token_accuracy": 0.1670161560177803,
      "num_tokens": 34787587.0,
      "step": 17790
    },
    {
      "entropy": 6.066124820709229,
      "epoch": 1.7788274104063577,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004688161260537236,
      "loss": 5.8339,
      "mean_token_accuracy": 0.15526469349861144,
      "num_tokens": 34798586.0,
      "step": 17795
    },
    {
      "entropy": 6.082526397705078,
      "epoch": 1.7793272354675862,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004687979964530842,
      "loss": 5.783,
      "mean_token_accuracy": 0.16527727246284485,
      "num_tokens": 34808469.0,
      "step": 17800
    },
    {
      "entropy": 6.089687967300415,
      "epoch": 1.779827060528815,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000468779861976537,
      "loss": 5.7499,
      "mean_token_accuracy": 0.17043453454971313,
      "num_tokens": 34817933.0,
      "step": 17805
    },
    {
      "entropy": 6.016081142425537,
      "epoch": 1.7803268855900436,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046876172262453843,
      "loss": 5.7043,
      "mean_token_accuracy": 0.17293307185173035,
      "num_tokens": 34827370.0,
      "step": 17810
    },
    {
      "entropy": 6.07455620765686,
      "epoch": 1.780826710651272,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046874357839754463,
      "loss": 5.7718,
      "mean_token_accuracy": 0.16662538796663284,
      "num_tokens": 34835521.0,
      "step": 17815
    },
    {
      "entropy": 6.132340574264527,
      "epoch": 1.7813265357125005,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046872542929601234,
      "loss": 5.9699,
      "mean_token_accuracy": 0.15785237699747084,
      "num_tokens": 34846151.0,
      "step": 17820
    },
    {
      "entropy": 6.100719213485718,
      "epoch": 1.7818263607737292,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046870727532039816,
      "loss": 5.6839,
      "mean_token_accuracy": 0.16969969794154166,
      "num_tokens": 34855536.0,
      "step": 17825
    },
    {
      "entropy": 6.0820738792419435,
      "epoch": 1.7823261858349577,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004686891164711586,
      "loss": 5.7611,
      "mean_token_accuracy": 0.1658921167254448,
      "num_tokens": 34865428.0,
      "step": 17830
    },
    {
      "entropy": 6.030757331848145,
      "epoch": 1.7828260108961862,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046867095274875085,
      "loss": 5.6943,
      "mean_token_accuracy": 0.16604602783918382,
      "num_tokens": 34875273.0,
      "step": 17835
    },
    {
      "entropy": 6.02469220161438,
      "epoch": 1.783325835957415,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046865278415363174,
      "loss": 5.8061,
      "mean_token_accuracy": 0.16339254304766654,
      "num_tokens": 34885438.0,
      "step": 17840
    },
    {
      "entropy": 6.14726767539978,
      "epoch": 1.7838256610186436,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004686346106862584,
      "loss": 5.8167,
      "mean_token_accuracy": 0.16415626257658006,
      "num_tokens": 34895245.0,
      "step": 17845
    },
    {
      "entropy": 6.089934539794922,
      "epoch": 1.784325486079872,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004686164323470881,
      "loss": 5.7121,
      "mean_token_accuracy": 0.17301984429359435,
      "num_tokens": 34905028.0,
      "step": 17850
    },
    {
      "entropy": 6.104522037506103,
      "epoch": 1.7848253111411005,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004685982491365781,
      "loss": 5.8347,
      "mean_token_accuracy": 0.1657872185111046,
      "num_tokens": 34915756.0,
      "step": 17855
    },
    {
      "entropy": 6.092056035995483,
      "epoch": 1.7853251362023292,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046858006105518615,
      "loss": 5.7302,
      "mean_token_accuracy": 0.17417582422494887,
      "num_tokens": 34925369.0,
      "step": 17860
    },
    {
      "entropy": 6.076295566558838,
      "epoch": 1.7858249612635577,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004685618681033695,
      "loss": 5.808,
      "mean_token_accuracy": 0.1645830199122429,
      "num_tokens": 34935693.0,
      "step": 17865
    },
    {
      "entropy": 6.01708345413208,
      "epoch": 1.7863247863247862,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004685436702815861,
      "loss": 5.7961,
      "mean_token_accuracy": 0.16627729684114456,
      "num_tokens": 34944582.0,
      "step": 17870
    },
    {
      "entropy": 6.077019453048706,
      "epoch": 1.786824611386015,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004685254675902938,
      "loss": 5.7598,
      "mean_token_accuracy": 0.16877981573343276,
      "num_tokens": 34953762.0,
      "step": 17875
    },
    {
      "entropy": 6.1380480289459225,
      "epoch": 1.7873244364472436,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004685072600299505,
      "loss": 5.8279,
      "mean_token_accuracy": 0.16665087789297103,
      "num_tokens": 34963003.0,
      "step": 17880
    },
    {
      "entropy": 6.047939300537109,
      "epoch": 1.787824261508472,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004684890476010144,
      "loss": 5.792,
      "mean_token_accuracy": 0.16185449212789535,
      "num_tokens": 34972968.0,
      "step": 17885
    },
    {
      "entropy": 6.072402381896973,
      "epoch": 1.7883240865697005,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004684708303039435,
      "loss": 5.8542,
      "mean_token_accuracy": 0.1691744640469551,
      "num_tokens": 34983050.0,
      "step": 17890
    },
    {
      "entropy": 6.147964668273926,
      "epoch": 1.7888239116309292,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046845260813919645,
      "loss": 5.8026,
      "mean_token_accuracy": 0.1655648559331894,
      "num_tokens": 34992192.0,
      "step": 17895
    },
    {
      "entropy": 6.036663627624511,
      "epoch": 1.7893237366921577,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046843438110723143,
      "loss": 5.7292,
      "mean_token_accuracy": 0.17029812186956406,
      "num_tokens": 35001135.0,
      "step": 17900
    },
    {
      "entropy": 6.0346471786499025,
      "epoch": 1.7898235617533862,
      "grad_norm": 2.609375,
      "learning_rate": 0.00046841614920850724,
      "loss": 5.7895,
      "mean_token_accuracy": 0.1674162909388542,
      "num_tokens": 35011389.0,
      "step": 17905
    },
    {
      "entropy": 6.071837997436523,
      "epoch": 1.790323386814615,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046839791244348243,
      "loss": 5.825,
      "mean_token_accuracy": 0.17088944017887114,
      "num_tokens": 35021752.0,
      "step": 17910
    },
    {
      "entropy": 6.08894157409668,
      "epoch": 1.7908232118758436,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046837967081261606,
      "loss": 5.7498,
      "mean_token_accuracy": 0.16929379254579544,
      "num_tokens": 35031878.0,
      "step": 17915
    },
    {
      "entropy": 6.0362128734588625,
      "epoch": 1.791323036937072,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004683614243163668,
      "loss": 5.6877,
      "mean_token_accuracy": 0.17646117508411407,
      "num_tokens": 35040995.0,
      "step": 17920
    },
    {
      "entropy": 6.0441638946533205,
      "epoch": 1.7918228619983005,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004683431729551939,
      "loss": 5.7296,
      "mean_token_accuracy": 0.17516793459653854,
      "num_tokens": 35050680.0,
      "step": 17925
    },
    {
      "entropy": 5.9795958518981935,
      "epoch": 1.7923226870595292,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004683249167295565,
      "loss": 5.7408,
      "mean_token_accuracy": 0.1694035202264786,
      "num_tokens": 35059910.0,
      "step": 17930
    },
    {
      "entropy": 5.956493711471557,
      "epoch": 1.7928225121207577,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000468306655639914,
      "loss": 5.6951,
      "mean_token_accuracy": 0.17827134281396867,
      "num_tokens": 35069711.0,
      "step": 17935
    },
    {
      "entropy": 6.132191753387451,
      "epoch": 1.7933223371819862,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004682883896867257,
      "loss": 5.7941,
      "mean_token_accuracy": 0.17016424536705016,
      "num_tokens": 35078711.0,
      "step": 17940
    },
    {
      "entropy": 6.048142766952514,
      "epoch": 1.793822162243215,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004682701188704513,
      "loss": 5.7419,
      "mean_token_accuracy": 0.17507303804159163,
      "num_tokens": 35087632.0,
      "step": 17945
    },
    {
      "entropy": 6.001174259185791,
      "epoch": 1.7943219873044436,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004682518431915505,
      "loss": 5.7525,
      "mean_token_accuracy": 0.17240643352270127,
      "num_tokens": 35097257.0,
      "step": 17950
    },
    {
      "entropy": 6.056989765167236,
      "epoch": 1.794821812365672,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046823356265048293,
      "loss": 5.771,
      "mean_token_accuracy": 0.17271180748939513,
      "num_tokens": 35107734.0,
      "step": 17955
    },
    {
      "entropy": 6.070879697799683,
      "epoch": 1.7953216374269005,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004682152772477087,
      "loss": 5.7392,
      "mean_token_accuracy": 0.17258943170309066,
      "num_tokens": 35117052.0,
      "step": 17960
    },
    {
      "entropy": 6.055917024612427,
      "epoch": 1.7958214624881292,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046819698698368785,
      "loss": 5.7874,
      "mean_token_accuracy": 0.1623857356607914,
      "num_tokens": 35127345.0,
      "step": 17965
    },
    {
      "entropy": 6.0150604248046875,
      "epoch": 1.7963212875493577,
      "grad_norm": 1.125,
      "learning_rate": 0.00046817869185888044,
      "loss": 5.6898,
      "mean_token_accuracy": 0.16851614713668822,
      "num_tokens": 35137216.0,
      "step": 17970
    },
    {
      "entropy": 6.111589813232422,
      "epoch": 1.7968211126105862,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004681603918737469,
      "loss": 5.7682,
      "mean_token_accuracy": 0.1767457976937294,
      "num_tokens": 35146433.0,
      "step": 17975
    },
    {
      "entropy": 6.0560142993927,
      "epoch": 1.7973209376718149,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046814208702874766,
      "loss": 5.7625,
      "mean_token_accuracy": 0.16765585094690322,
      "num_tokens": 35155841.0,
      "step": 17980
    },
    {
      "entropy": 5.970711374282837,
      "epoch": 1.7978207627330436,
      "grad_norm": 1.25,
      "learning_rate": 0.00046812377732434316,
      "loss": 5.8262,
      "mean_token_accuracy": 0.16857350021600723,
      "num_tokens": 35165962.0,
      "step": 17985
    },
    {
      "entropy": 6.0751180171966555,
      "epoch": 1.7983205877942718,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046810546276099415,
      "loss": 5.8047,
      "mean_token_accuracy": 0.16453880667686463,
      "num_tokens": 35175887.0,
      "step": 17990
    },
    {
      "entropy": 6.153341960906983,
      "epoch": 1.7988204128555005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004680871433391613,
      "loss": 5.7815,
      "mean_token_accuracy": 0.17031702697277068,
      "num_tokens": 35185763.0,
      "step": 17995
    },
    {
      "entropy": 6.06359634399414,
      "epoch": 1.7993202379167292,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046806881905930565,
      "loss": 5.764,
      "mean_token_accuracy": 0.17303440272808074,
      "num_tokens": 35195452.0,
      "step": 18000
    },
    {
      "epoch": 1.7993202379167292,
      "eval_entropy": 5.861335724784185,
      "eval_loss": 5.857707500457764,
      "eval_mean_token_accuracy": 0.1717265115451395,
      "eval_num_tokens": 35195452.0,
      "eval_runtime": 23.9459,
      "eval_samples_per_second": 1296.462,
      "eval_steps_per_second": 162.073,
      "step": 18000
    },
    {
      "entropy": 6.060310173034668,
      "epoch": 1.7998200629779577,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046805048992188823,
      "loss": 5.8119,
      "mean_token_accuracy": 0.16683428436517717,
      "num_tokens": 35205261.0,
      "step": 18005
    },
    {
      "entropy": 6.056484842300415,
      "epoch": 1.8003198880391862,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046803215592737014,
      "loss": 5.7975,
      "mean_token_accuracy": 0.16813943386077881,
      "num_tokens": 35214965.0,
      "step": 18010
    },
    {
      "entropy": 6.054237508773804,
      "epoch": 1.8008197131004149,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046801381707621266,
      "loss": 5.7617,
      "mean_token_accuracy": 0.1705940693616867,
      "num_tokens": 35224371.0,
      "step": 18015
    },
    {
      "entropy": 6.025848531723023,
      "epoch": 1.8013195381616436,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046799547336887727,
      "loss": 5.6809,
      "mean_token_accuracy": 0.17710628658533095,
      "num_tokens": 35234282.0,
      "step": 18020
    },
    {
      "entropy": 6.05817551612854,
      "epoch": 1.8018193632228718,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004679771248058254,
      "loss": 5.7152,
      "mean_token_accuracy": 0.17837765365839003,
      "num_tokens": 35243112.0,
      "step": 18025
    },
    {
      "entropy": 6.0354231834411625,
      "epoch": 1.8023191882841005,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004679587713875187,
      "loss": 5.7514,
      "mean_token_accuracy": 0.1611502707004547,
      "num_tokens": 35253398.0,
      "step": 18030
    },
    {
      "entropy": 6.010533714294434,
      "epoch": 1.8028190133453292,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000467940413114419,
      "loss": 5.6686,
      "mean_token_accuracy": 0.1742311879992485,
      "num_tokens": 35263359.0,
      "step": 18035
    },
    {
      "entropy": 6.092684602737426,
      "epoch": 1.8033188384065577,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004679220499869881,
      "loss": 5.8463,
      "mean_token_accuracy": 0.1628038227558136,
      "num_tokens": 35272500.0,
      "step": 18040
    },
    {
      "entropy": 6.160508060455323,
      "epoch": 1.8038186634677862,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004679036820056881,
      "loss": 5.8293,
      "mean_token_accuracy": 0.1559961974620819,
      "num_tokens": 35282415.0,
      "step": 18045
    },
    {
      "entropy": 6.083537006378174,
      "epoch": 1.8043184885290149,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046788530917098105,
      "loss": 5.7592,
      "mean_token_accuracy": 0.1707555130124092,
      "num_tokens": 35291567.0,
      "step": 18050
    },
    {
      "entropy": 6.0545025825500485,
      "epoch": 1.8048183135902434,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004678669314833293,
      "loss": 5.8889,
      "mean_token_accuracy": 0.16160404682159424,
      "num_tokens": 35301397.0,
      "step": 18055
    },
    {
      "entropy": 6.030395126342773,
      "epoch": 1.8053181386514718,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046784854894319525,
      "loss": 5.7094,
      "mean_token_accuracy": 0.17616275250911712,
      "num_tokens": 35311296.0,
      "step": 18060
    },
    {
      "entropy": 6.077094459533692,
      "epoch": 1.8058179637127005,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046783016155104123,
      "loss": 5.8065,
      "mean_token_accuracy": 0.17171565741300582,
      "num_tokens": 35322819.0,
      "step": 18065
    },
    {
      "entropy": 6.074342107772827,
      "epoch": 1.8063177887739292,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046781176930732997,
      "loss": 5.6993,
      "mean_token_accuracy": 0.17211078852415085,
      "num_tokens": 35331374.0,
      "step": 18070
    },
    {
      "entropy": 6.030455112457275,
      "epoch": 1.8068176138351577,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004677933722125243,
      "loss": 5.7892,
      "mean_token_accuracy": 0.1724635146558285,
      "num_tokens": 35341040.0,
      "step": 18075
    },
    {
      "entropy": 6.10050401687622,
      "epoch": 1.8073174388963862,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004677749702670869,
      "loss": 5.8144,
      "mean_token_accuracy": 0.15908677875995636,
      "num_tokens": 35351387.0,
      "step": 18080
    },
    {
      "entropy": 6.101424646377564,
      "epoch": 1.8078172639576149,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004677565634714809,
      "loss": 5.8188,
      "mean_token_accuracy": 0.16395028978586196,
      "num_tokens": 35361864.0,
      "step": 18085
    },
    {
      "entropy": 5.993679237365723,
      "epoch": 1.8083170890188434,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004677381518261693,
      "loss": 5.717,
      "mean_token_accuracy": 0.1787855863571167,
      "num_tokens": 35370539.0,
      "step": 18090
    },
    {
      "entropy": 5.992382383346557,
      "epoch": 1.8088169140800718,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046771973533161543,
      "loss": 5.7086,
      "mean_token_accuracy": 0.17921897321939467,
      "num_tokens": 35381851.0,
      "step": 18095
    },
    {
      "entropy": 6.137729167938232,
      "epoch": 1.8093167391413005,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046770131398828264,
      "loss": 5.7528,
      "mean_token_accuracy": 0.1701904296875,
      "num_tokens": 35391440.0,
      "step": 18100
    },
    {
      "entropy": 6.099941301345825,
      "epoch": 1.8098165642025292,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046768288779663436,
      "loss": 5.7845,
      "mean_token_accuracy": 0.16568878069519996,
      "num_tokens": 35401370.0,
      "step": 18105
    },
    {
      "entropy": 6.050946140289307,
      "epoch": 1.8103163892637577,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004676644567571342,
      "loss": 5.751,
      "mean_token_accuracy": 0.1655309498310089,
      "num_tokens": 35411015.0,
      "step": 18110
    },
    {
      "entropy": 6.114871263504028,
      "epoch": 1.8108162143249862,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046764602087024596,
      "loss": 5.7666,
      "mean_token_accuracy": 0.16842869222164153,
      "num_tokens": 35421741.0,
      "step": 18115
    },
    {
      "entropy": 6.172026824951172,
      "epoch": 1.8113160393862149,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004676275801364332,
      "loss": 5.9015,
      "mean_token_accuracy": 0.15891308933496476,
      "num_tokens": 35432899.0,
      "step": 18120
    },
    {
      "entropy": 6.0086079120635985,
      "epoch": 1.8118158644474434,
      "grad_norm": 1.125,
      "learning_rate": 0.0004676091345561603,
      "loss": 5.6463,
      "mean_token_accuracy": 0.17850535213947297,
      "num_tokens": 35442220.0,
      "step": 18125
    },
    {
      "entropy": 5.964807796478271,
      "epoch": 1.8123156895086718,
      "grad_norm": 1.15625,
      "learning_rate": 0.000467590684129891,
      "loss": 5.6938,
      "mean_token_accuracy": 0.17979492694139482,
      "num_tokens": 35452193.0,
      "step": 18130
    },
    {
      "entropy": 6.03503909111023,
      "epoch": 1.8128155145699005,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046757222885808973,
      "loss": 5.7735,
      "mean_token_accuracy": 0.17158698588609694,
      "num_tokens": 35462152.0,
      "step": 18135
    },
    {
      "entropy": 6.028393077850342,
      "epoch": 1.8133153396311292,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046755376874122075,
      "loss": 5.7533,
      "mean_token_accuracy": 0.1702207326889038,
      "num_tokens": 35472210.0,
      "step": 18140
    },
    {
      "entropy": 6.068217849731445,
      "epoch": 1.8138151646923577,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004675353037797485,
      "loss": 5.8184,
      "mean_token_accuracy": 0.15764254629611968,
      "num_tokens": 35482238.0,
      "step": 18145
    },
    {
      "entropy": 6.037251949310303,
      "epoch": 1.8143149897535862,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004675168339741375,
      "loss": 5.8089,
      "mean_token_accuracy": 0.1639021784067154,
      "num_tokens": 35492042.0,
      "step": 18150
    },
    {
      "entropy": 6.116870641708374,
      "epoch": 1.8148148148148149,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004674983593248525,
      "loss": 5.8632,
      "mean_token_accuracy": 0.16245257556438447,
      "num_tokens": 35501892.0,
      "step": 18155
    },
    {
      "entropy": 6.131846189498901,
      "epoch": 1.8153146398760434,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046747987983235836,
      "loss": 5.8157,
      "mean_token_accuracy": 0.165689717233181,
      "num_tokens": 35512472.0,
      "step": 18160
    },
    {
      "entropy": 6.07648549079895,
      "epoch": 1.8158144649372718,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046746139549711993,
      "loss": 5.6827,
      "mean_token_accuracy": 0.17046304643154145,
      "num_tokens": 35521467.0,
      "step": 18165
    },
    {
      "entropy": 5.954675245285034,
      "epoch": 1.8163142899985005,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004674429063196023,
      "loss": 5.7815,
      "mean_token_accuracy": 0.16854082196950912,
      "num_tokens": 35531584.0,
      "step": 18170
    },
    {
      "entropy": 6.012395143508911,
      "epoch": 1.8168141150597292,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046742441230027067,
      "loss": 5.6199,
      "mean_token_accuracy": 0.18237232714891433,
      "num_tokens": 35541687.0,
      "step": 18175
    },
    {
      "entropy": 6.054369354248047,
      "epoch": 1.8173139401209577,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004674059134395904,
      "loss": 5.7358,
      "mean_token_accuracy": 0.1695303052663803,
      "num_tokens": 35551310.0,
      "step": 18180
    },
    {
      "entropy": 6.012951993942261,
      "epoch": 1.8178137651821862,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004673874097380268,
      "loss": 5.6992,
      "mean_token_accuracy": 0.175189071893692,
      "num_tokens": 35560356.0,
      "step": 18185
    },
    {
      "entropy": 6.017389392852783,
      "epoch": 1.8183135902434149,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004673689011960455,
      "loss": 5.6928,
      "mean_token_accuracy": 0.17251627147197723,
      "num_tokens": 35570633.0,
      "step": 18190
    },
    {
      "entropy": 6.104699897766113,
      "epoch": 1.8188134153046434,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046735038781411204,
      "loss": 5.8745,
      "mean_token_accuracy": 0.16037281304597856,
      "num_tokens": 35581341.0,
      "step": 18195
    },
    {
      "entropy": 6.004894876480103,
      "epoch": 1.8193132403658718,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004673318695926925,
      "loss": 5.6154,
      "mean_token_accuracy": 0.1856730103492737,
      "num_tokens": 35592301.0,
      "step": 18200
    },
    {
      "entropy": 6.01012225151062,
      "epoch": 1.8198130654271005,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004673133465322525,
      "loss": 5.6732,
      "mean_token_accuracy": 0.17967179864645005,
      "num_tokens": 35601782.0,
      "step": 18205
    },
    {
      "entropy": 5.924084138870239,
      "epoch": 1.8203128904883292,
      "grad_norm": 1.125,
      "learning_rate": 0.0004672948186332582,
      "loss": 5.6611,
      "mean_token_accuracy": 0.17057402282953263,
      "num_tokens": 35611307.0,
      "step": 18210
    },
    {
      "entropy": 6.072911024093628,
      "epoch": 1.8208127155495577,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046727628589617574,
      "loss": 5.8135,
      "mean_token_accuracy": 0.16275690197944642,
      "num_tokens": 35621358.0,
      "step": 18215
    },
    {
      "entropy": 6.032770490646362,
      "epoch": 1.8213125406107862,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046725774832147147,
      "loss": 5.7166,
      "mean_token_accuracy": 0.1703886643052101,
      "num_tokens": 35630976.0,
      "step": 18220
    },
    {
      "entropy": 5.989142274856567,
      "epoch": 1.8218123656720149,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046723920590961167,
      "loss": 5.6909,
      "mean_token_accuracy": 0.16934550404548646,
      "num_tokens": 35640126.0,
      "step": 18225
    },
    {
      "entropy": 6.0534656047821045,
      "epoch": 1.8223121907332434,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004672206586610629,
      "loss": 5.754,
      "mean_token_accuracy": 0.1670188382267952,
      "num_tokens": 35649282.0,
      "step": 18230
    },
    {
      "entropy": 5.99921612739563,
      "epoch": 1.8228120157944718,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046720210657629185,
      "loss": 5.6595,
      "mean_token_accuracy": 0.1768149182200432,
      "num_tokens": 35658284.0,
      "step": 18235
    },
    {
      "entropy": 6.036451148986816,
      "epoch": 1.8233118408557005,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046718354965576527,
      "loss": 5.8322,
      "mean_token_accuracy": 0.16478145271539688,
      "num_tokens": 35668355.0,
      "step": 18240
    },
    {
      "entropy": 6.093158626556397,
      "epoch": 1.8238116659169292,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046716498789995,
      "loss": 5.7874,
      "mean_token_accuracy": 0.16449143439531327,
      "num_tokens": 35677582.0,
      "step": 18245
    },
    {
      "entropy": 6.102607727050781,
      "epoch": 1.8243114909781575,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046714642130931306,
      "loss": 5.8101,
      "mean_token_accuracy": 0.16693980097770691,
      "num_tokens": 35687691.0,
      "step": 18250
    },
    {
      "entropy": 6.039275455474853,
      "epoch": 1.8248113160393862,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046712784988432173,
      "loss": 5.7966,
      "mean_token_accuracy": 0.1675327315926552,
      "num_tokens": 35697796.0,
      "step": 18255
    },
    {
      "entropy": 6.061287832260132,
      "epoch": 1.8253111411006149,
      "grad_norm": 1.1875,
      "learning_rate": 0.000467109273625443,
      "loss": 5.7016,
      "mean_token_accuracy": 0.17484399825334548,
      "num_tokens": 35706795.0,
      "step": 18260
    },
    {
      "entropy": 6.074637985229492,
      "epoch": 1.8258109661618434,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004670906925331444,
      "loss": 5.8165,
      "mean_token_accuracy": 0.1713494062423706,
      "num_tokens": 35717260.0,
      "step": 18265
    },
    {
      "entropy": 6.060304784774781,
      "epoch": 1.8263107912230718,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046707210660789343,
      "loss": 5.8372,
      "mean_token_accuracy": 0.16259816139936448,
      "num_tokens": 35726671.0,
      "step": 18270
    },
    {
      "entropy": 6.0898048877716064,
      "epoch": 1.8268106162843005,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004670535158501577,
      "loss": 5.8053,
      "mean_token_accuracy": 0.16204888075590135,
      "num_tokens": 35736672.0,
      "step": 18275
    },
    {
      "entropy": 6.112015962600708,
      "epoch": 1.8273104413455292,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046703492026040495,
      "loss": 5.7232,
      "mean_token_accuracy": 0.16783074885606766,
      "num_tokens": 35746291.0,
      "step": 18280
    },
    {
      "entropy": 6.001485443115234,
      "epoch": 1.8278102664067575,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046701631983910296,
      "loss": 5.708,
      "mean_token_accuracy": 0.16792598962783814,
      "num_tokens": 35755683.0,
      "step": 18285
    },
    {
      "entropy": 6.059262228012085,
      "epoch": 1.8283100914679862,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046699771458671974,
      "loss": 5.7709,
      "mean_token_accuracy": 0.1670281931757927,
      "num_tokens": 35765870.0,
      "step": 18290
    },
    {
      "entropy": 5.917414331436158,
      "epoch": 1.8288099165292149,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046697910450372356,
      "loss": 5.6094,
      "mean_token_accuracy": 0.18506603240966796,
      "num_tokens": 35777055.0,
      "step": 18295
    },
    {
      "entropy": 5.970369434356689,
      "epoch": 1.8293097415904434,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046696048959058236,
      "loss": 5.6534,
      "mean_token_accuracy": 0.1809844046831131,
      "num_tokens": 35786846.0,
      "step": 18300
    },
    {
      "entropy": 6.008553791046142,
      "epoch": 1.8298095666516718,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004669418698477647,
      "loss": 5.6673,
      "mean_token_accuracy": 0.17619094103574753,
      "num_tokens": 35795471.0,
      "step": 18305
    },
    {
      "entropy": 5.983393955230713,
      "epoch": 1.8303093917129005,
      "grad_norm": 1.125,
      "learning_rate": 0.00046692324527573907,
      "loss": 5.6228,
      "mean_token_accuracy": 0.1791856601834297,
      "num_tokens": 35804559.0,
      "step": 18310
    },
    {
      "entropy": 6.001714849472046,
      "epoch": 1.830809216774129,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004669046158749738,
      "loss": 5.7635,
      "mean_token_accuracy": 0.17234887629747392,
      "num_tokens": 35813545.0,
      "step": 18315
    },
    {
      "entropy": 6.0069163799285885,
      "epoch": 1.8313090418353575,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046688598164593793,
      "loss": 5.7716,
      "mean_token_accuracy": 0.17123837172985076,
      "num_tokens": 35822740.0,
      "step": 18320
    },
    {
      "entropy": 6.057091188430786,
      "epoch": 1.8318088668965862,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004668673425891001,
      "loss": 5.7981,
      "mean_token_accuracy": 0.16512157171964645,
      "num_tokens": 35832593.0,
      "step": 18325
    },
    {
      "entropy": 6.124195528030396,
      "epoch": 1.8323086919578149,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046684869870492927,
      "loss": 5.7417,
      "mean_token_accuracy": 0.16736694276332856,
      "num_tokens": 35841399.0,
      "step": 18330
    },
    {
      "entropy": 6.03677077293396,
      "epoch": 1.8328085170190433,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046683004999389455,
      "loss": 5.7744,
      "mean_token_accuracy": 0.16967875212430955,
      "num_tokens": 35850099.0,
      "step": 18335
    },
    {
      "entropy": 6.0577880382537845,
      "epoch": 1.8333083420802718,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004668113964564651,
      "loss": 5.8053,
      "mean_token_accuracy": 0.16554514616727828,
      "num_tokens": 35860398.0,
      "step": 18340
    },
    {
      "entropy": 6.042994165420533,
      "epoch": 1.8338081671415005,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004667927380931104,
      "loss": 5.722,
      "mean_token_accuracy": 0.17011613696813582,
      "num_tokens": 35870390.0,
      "step": 18345
    },
    {
      "entropy": 6.036911058425903,
      "epoch": 1.834307992202729,
      "grad_norm": 1.125,
      "learning_rate": 0.0004667740749042997,
      "loss": 5.6633,
      "mean_token_accuracy": 0.1768166333436966,
      "num_tokens": 35881633.0,
      "step": 18350
    },
    {
      "entropy": 6.009915781021118,
      "epoch": 1.8348078172639575,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046675540689050253,
      "loss": 5.7986,
      "mean_token_accuracy": 0.16395699381828308,
      "num_tokens": 35891904.0,
      "step": 18355
    },
    {
      "entropy": 6.177428245544434,
      "epoch": 1.8353076423251862,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046673673405218876,
      "loss": 6.0238,
      "mean_token_accuracy": 0.1517401859164238,
      "num_tokens": 35901712.0,
      "step": 18360
    },
    {
      "entropy": 6.139548206329346,
      "epoch": 1.8358074673864149,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004667180563898281,
      "loss": 5.9021,
      "mean_token_accuracy": 0.1591898187994957,
      "num_tokens": 35911314.0,
      "step": 18365
    },
    {
      "entropy": 6.143383359909057,
      "epoch": 1.8363072924476433,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046669937390389043,
      "loss": 5.8611,
      "mean_token_accuracy": 0.161187344789505,
      "num_tokens": 35921135.0,
      "step": 18370
    },
    {
      "entropy": 6.148154449462891,
      "epoch": 1.8368071175088718,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046668068659484584,
      "loss": 5.8089,
      "mean_token_accuracy": 0.1659947544336319,
      "num_tokens": 35930784.0,
      "step": 18375
    },
    {
      "entropy": 6.053774881362915,
      "epoch": 1.8373069425701005,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046666199446316454,
      "loss": 5.6834,
      "mean_token_accuracy": 0.1707798033952713,
      "num_tokens": 35940372.0,
      "step": 18380
    },
    {
      "entropy": 6.070195579528809,
      "epoch": 1.837806767631329,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046664329750931675,
      "loss": 5.7002,
      "mean_token_accuracy": 0.16984677612781524,
      "num_tokens": 35949507.0,
      "step": 18385
    },
    {
      "entropy": 6.02717432975769,
      "epoch": 1.8383065926925575,
      "grad_norm": 1.234375,
      "learning_rate": 0.000466624595733773,
      "loss": 5.6957,
      "mean_token_accuracy": 0.17170563489198684,
      "num_tokens": 35958981.0,
      "step": 18390
    },
    {
      "entropy": 6.080955505371094,
      "epoch": 1.8388064177537862,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004666058891370036,
      "loss": 5.7271,
      "mean_token_accuracy": 0.17183950245380403,
      "num_tokens": 35967842.0,
      "step": 18395
    },
    {
      "entropy": 6.022426271438599,
      "epoch": 1.8393062428150149,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004665871777194794,
      "loss": 5.7144,
      "mean_token_accuracy": 0.16507743299007416,
      "num_tokens": 35977669.0,
      "step": 18400
    },
    {
      "entropy": 6.029884290695191,
      "epoch": 1.8398060678762433,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046656846148167107,
      "loss": 5.7556,
      "mean_token_accuracy": 0.17298200726509094,
      "num_tokens": 35987815.0,
      "step": 18405
    },
    {
      "entropy": 6.13895058631897,
      "epoch": 1.8403058929374718,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046654974042404957,
      "loss": 5.8999,
      "mean_token_accuracy": 0.16053846776485442,
      "num_tokens": 35997831.0,
      "step": 18410
    },
    {
      "entropy": 6.097833251953125,
      "epoch": 1.8408057179987005,
      "grad_norm": 1.25,
      "learning_rate": 0.00046653101454708593,
      "loss": 5.7425,
      "mean_token_accuracy": 0.17255839407444,
      "num_tokens": 36006822.0,
      "step": 18415
    },
    {
      "entropy": 5.994505071640015,
      "epoch": 1.841305543059929,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004665122838512512,
      "loss": 5.6475,
      "mean_token_accuracy": 0.174712173640728,
      "num_tokens": 36015648.0,
      "step": 18420
    },
    {
      "entropy": 6.076385974884033,
      "epoch": 1.8418053681211575,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004664935483370167,
      "loss": 5.8831,
      "mean_token_accuracy": 0.16027942746877671,
      "num_tokens": 36025713.0,
      "step": 18425
    },
    {
      "entropy": 6.0798962116241455,
      "epoch": 1.8423051931823862,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004664748080048537,
      "loss": 5.7017,
      "mean_token_accuracy": 0.1726372569799423,
      "num_tokens": 36035054.0,
      "step": 18430
    },
    {
      "entropy": 6.044128656387329,
      "epoch": 1.8428050182436149,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046645606285523395,
      "loss": 5.7488,
      "mean_token_accuracy": 0.17287470698356627,
      "num_tokens": 36045449.0,
      "step": 18435
    },
    {
      "entropy": 6.068214654922485,
      "epoch": 1.8433048433048433,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046643731288862884,
      "loss": 5.7938,
      "mean_token_accuracy": 0.16925143003463744,
      "num_tokens": 36055269.0,
      "step": 18440
    },
    {
      "entropy": 6.1050478458404545,
      "epoch": 1.8438046683660718,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004664185581055103,
      "loss": 5.8979,
      "mean_token_accuracy": 0.1583724647760391,
      "num_tokens": 36064791.0,
      "step": 18445
    },
    {
      "entropy": 6.066727018356323,
      "epoch": 1.8443044934273005,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00046639979850634995,
      "loss": 5.7148,
      "mean_token_accuracy": 0.16446462273597717,
      "num_tokens": 36075583.0,
      "step": 18450
    },
    {
      "entropy": 6.089434051513672,
      "epoch": 1.844804318488529,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046638103409162,
      "loss": 5.8004,
      "mean_token_accuracy": 0.16398454010486602,
      "num_tokens": 36084664.0,
      "step": 18455
    },
    {
      "entropy": 6.151490020751953,
      "epoch": 1.8453041435497575,
      "grad_norm": 1.25,
      "learning_rate": 0.00046636226486179244,
      "loss": 5.7373,
      "mean_token_accuracy": 0.17530999183654786,
      "num_tokens": 36094184.0,
      "step": 18460
    },
    {
      "entropy": 6.101826858520508,
      "epoch": 1.8458039686109862,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004663434908173396,
      "loss": 5.7327,
      "mean_token_accuracy": 0.17384556084871292,
      "num_tokens": 36103148.0,
      "step": 18465
    },
    {
      "entropy": 6.049436759948731,
      "epoch": 1.8463037936722149,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004663247119587338,
      "loss": 5.755,
      "mean_token_accuracy": 0.1697106719017029,
      "num_tokens": 36114268.0,
      "step": 18470
    },
    {
      "entropy": 6.058325624465942,
      "epoch": 1.8468036187334433,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004663059282864474,
      "loss": 5.7422,
      "mean_token_accuracy": 0.1685776188969612,
      "num_tokens": 36124690.0,
      "step": 18475
    },
    {
      "entropy": 6.0745697021484375,
      "epoch": 1.8473034437946718,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004662871398009531,
      "loss": 5.7472,
      "mean_token_accuracy": 0.17005280405282974,
      "num_tokens": 36134628.0,
      "step": 18480
    },
    {
      "entropy": 6.062219047546387,
      "epoch": 1.8478032688559005,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004662683465027235,
      "loss": 5.8016,
      "mean_token_accuracy": 0.16646270900964738,
      "num_tokens": 36144919.0,
      "step": 18485
    },
    {
      "entropy": 6.014105796813965,
      "epoch": 1.848303093917129,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046624954839223165,
      "loss": 5.6959,
      "mean_token_accuracy": 0.1682801514863968,
      "num_tokens": 36154081.0,
      "step": 18490
    },
    {
      "entropy": 5.943833589553833,
      "epoch": 1.8488029189783575,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046623074546995024,
      "loss": 5.6796,
      "mean_token_accuracy": 0.17790587097406388,
      "num_tokens": 36164318.0,
      "step": 18495
    },
    {
      "entropy": 6.0817710876464846,
      "epoch": 1.8493027440395862,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046621193773635256,
      "loss": 5.7902,
      "mean_token_accuracy": 0.17122818827629088,
      "num_tokens": 36173361.0,
      "step": 18500
    },
    {
      "entropy": 5.971076107025146,
      "epoch": 1.8498025691008149,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004661931251919118,
      "loss": 5.646,
      "mean_token_accuracy": 0.17520214021205902,
      "num_tokens": 36183133.0,
      "step": 18505
    },
    {
      "entropy": 6.046332836151123,
      "epoch": 1.8503023941620433,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046617430783710105,
      "loss": 5.6746,
      "mean_token_accuracy": 0.1733727514743805,
      "num_tokens": 36193041.0,
      "step": 18510
    },
    {
      "entropy": 6.02428297996521,
      "epoch": 1.8508022192232718,
      "grad_norm": 1.140625,
      "learning_rate": 0.000466155485672394,
      "loss": 5.6623,
      "mean_token_accuracy": 0.1831003710627556,
      "num_tokens": 36202306.0,
      "step": 18515
    },
    {
      "entropy": 5.98944239616394,
      "epoch": 1.8513020442845005,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004661366586982641,
      "loss": 5.6648,
      "mean_token_accuracy": 0.17335156053304673,
      "num_tokens": 36212342.0,
      "step": 18520
    },
    {
      "entropy": 6.024898386001587,
      "epoch": 1.851801869345729,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004661178269151851,
      "loss": 5.7494,
      "mean_token_accuracy": 0.17023625671863557,
      "num_tokens": 36221181.0,
      "step": 18525
    },
    {
      "entropy": 6.085935306549072,
      "epoch": 1.8523016944069575,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046609899032363074,
      "loss": 5.7589,
      "mean_token_accuracy": 0.17154577225446702,
      "num_tokens": 36230813.0,
      "step": 18530
    },
    {
      "entropy": 6.077480173110962,
      "epoch": 1.8528015194681862,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004660801489240749,
      "loss": 5.826,
      "mean_token_accuracy": 0.1657283052802086,
      "num_tokens": 36242315.0,
      "step": 18535
    },
    {
      "entropy": 6.008757734298706,
      "epoch": 1.8533013445294149,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046606130271699175,
      "loss": 5.7054,
      "mean_token_accuracy": 0.17465660721063614,
      "num_tokens": 36253665.0,
      "step": 18540
    },
    {
      "entropy": 6.028655910491944,
      "epoch": 1.8538011695906431,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004660424517028553,
      "loss": 5.8872,
      "mean_token_accuracy": 0.16051968336105346,
      "num_tokens": 36263514.0,
      "step": 18545
    },
    {
      "entropy": 6.072268486022949,
      "epoch": 1.8543009946518718,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046602359588214,
      "loss": 5.7331,
      "mean_token_accuracy": 0.16603456139564515,
      "num_tokens": 36273103.0,
      "step": 18550
    },
    {
      "entropy": 6.129279041290284,
      "epoch": 1.8548008197131005,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004660047352553202,
      "loss": 5.7526,
      "mean_token_accuracy": 0.16985220313072205,
      "num_tokens": 36283079.0,
      "step": 18555
    },
    {
      "entropy": 6.075054073333741,
      "epoch": 1.855300644774329,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046598586982287034,
      "loss": 5.7523,
      "mean_token_accuracy": 0.1619232103228569,
      "num_tokens": 36292405.0,
      "step": 18560
    },
    {
      "entropy": 5.9956177234649655,
      "epoch": 1.8558004698355575,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046596699958526515,
      "loss": 5.6924,
      "mean_token_accuracy": 0.17434561401605606,
      "num_tokens": 36301516.0,
      "step": 18565
    },
    {
      "entropy": 6.077717924118042,
      "epoch": 1.8563002948967862,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046594812454297944,
      "loss": 5.8227,
      "mean_token_accuracy": 0.16303227990865707,
      "num_tokens": 36311414.0,
      "step": 18570
    },
    {
      "entropy": 6.20360255241394,
      "epoch": 1.8568001199580149,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000465929244696488,
      "loss": 5.8783,
      "mean_token_accuracy": 0.1568348303437233,
      "num_tokens": 36321521.0,
      "step": 18575
    },
    {
      "entropy": 6.132644748687744,
      "epoch": 1.8572999450192431,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004659103600462659,
      "loss": 5.811,
      "mean_token_accuracy": 0.16388128101825714,
      "num_tokens": 36331115.0,
      "step": 18580
    },
    {
      "entropy": 6.002081632614136,
      "epoch": 1.8577997700804718,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004658914705927883,
      "loss": 5.8233,
      "mean_token_accuracy": 0.17065856009721755,
      "num_tokens": 36341131.0,
      "step": 18585
    },
    {
      "entropy": 5.979644107818603,
      "epoch": 1.8582995951417005,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004658725763365304,
      "loss": 5.6652,
      "mean_token_accuracy": 0.17956755459308624,
      "num_tokens": 36350378.0,
      "step": 18590
    },
    {
      "entropy": 6.005631589889527,
      "epoch": 1.858799420202929,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004658536772779676,
      "loss": 5.7056,
      "mean_token_accuracy": 0.17340375185012818,
      "num_tokens": 36359706.0,
      "step": 18595
    },
    {
      "entropy": 6.027868127822876,
      "epoch": 1.8592992452641575,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004658347734175754,
      "loss": 5.7831,
      "mean_token_accuracy": 0.17165062576532364,
      "num_tokens": 36370068.0,
      "step": 18600
    },
    {
      "entropy": 6.0299193382263185,
      "epoch": 1.8597990703253862,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046581586475582946,
      "loss": 5.7283,
      "mean_token_accuracy": 0.1722511515021324,
      "num_tokens": 36380400.0,
      "step": 18605
    },
    {
      "entropy": 6.057851552963257,
      "epoch": 1.8602988953866146,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004657969512932054,
      "loss": 5.7366,
      "mean_token_accuracy": 0.16998562216758728,
      "num_tokens": 36390114.0,
      "step": 18610
    },
    {
      "entropy": 6.00553183555603,
      "epoch": 1.8607987204478431,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004657780330301791,
      "loss": 5.6863,
      "mean_token_accuracy": 0.1737401694059372,
      "num_tokens": 36399389.0,
      "step": 18615
    },
    {
      "entropy": 6.0273980617523195,
      "epoch": 1.8612985455090718,
      "grad_norm": 1.25,
      "learning_rate": 0.00046575910996722657,
      "loss": 5.6884,
      "mean_token_accuracy": 0.1707937315106392,
      "num_tokens": 36408512.0,
      "step": 18620
    },
    {
      "entropy": 6.120745325088501,
      "epoch": 1.8617983705703005,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000465740182104824,
      "loss": 5.8762,
      "mean_token_accuracy": 0.15837269872426987,
      "num_tokens": 36419049.0,
      "step": 18625
    },
    {
      "entropy": 6.048524665832519,
      "epoch": 1.862298195631529,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046572124944344745,
      "loss": 5.6155,
      "mean_token_accuracy": 0.17541998624801636,
      "num_tokens": 36428561.0,
      "step": 18630
    },
    {
      "entropy": 6.066813039779663,
      "epoch": 1.8627980206927575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046570231198357344,
      "loss": 5.8062,
      "mean_token_accuracy": 0.16524759083986282,
      "num_tokens": 36438346.0,
      "step": 18635
    },
    {
      "entropy": 6.062178611755371,
      "epoch": 1.8632978457539862,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004656833697256783,
      "loss": 5.7099,
      "mean_token_accuracy": 0.17902867794036864,
      "num_tokens": 36447756.0,
      "step": 18640
    },
    {
      "entropy": 6.040937042236328,
      "epoch": 1.8637976708152146,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004656644226702386,
      "loss": 5.6873,
      "mean_token_accuracy": 0.17907849401235582,
      "num_tokens": 36457784.0,
      "step": 18645
    },
    {
      "entropy": 5.939584827423095,
      "epoch": 1.8642974958764431,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046564547081773115,
      "loss": 5.6678,
      "mean_token_accuracy": 0.1775304302573204,
      "num_tokens": 36467952.0,
      "step": 18650
    },
    {
      "entropy": 6.16763916015625,
      "epoch": 1.8647973209376718,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004656265141686326,
      "loss": 5.8792,
      "mean_token_accuracy": 0.1557847812771797,
      "num_tokens": 36478595.0,
      "step": 18655
    },
    {
      "entropy": 6.0220221996307375,
      "epoch": 1.8652971459989005,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046560755272342014,
      "loss": 5.59,
      "mean_token_accuracy": 0.1818245142698288,
      "num_tokens": 36487866.0,
      "step": 18660
    },
    {
      "entropy": 6.066146278381348,
      "epoch": 1.865796971060129,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004655885864825706,
      "loss": 5.7216,
      "mean_token_accuracy": 0.16682749539613723,
      "num_tokens": 36497828.0,
      "step": 18665
    },
    {
      "entropy": 6.045554161071777,
      "epoch": 1.8662967961213575,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004655696154465613,
      "loss": 5.6996,
      "mean_token_accuracy": 0.16775216162204742,
      "num_tokens": 36508361.0,
      "step": 18670
    },
    {
      "entropy": 6.005676555633545,
      "epoch": 1.8667966211825862,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046555063961586956,
      "loss": 5.6265,
      "mean_token_accuracy": 0.17894013226032257,
      "num_tokens": 36518458.0,
      "step": 18675
    },
    {
      "entropy": 5.987021350860596,
      "epoch": 1.8672964462438146,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004655316589909727,
      "loss": 5.7095,
      "mean_token_accuracy": 0.1736234411597252,
      "num_tokens": 36527952.0,
      "step": 18680
    },
    {
      "entropy": 6.052872180938721,
      "epoch": 1.867796271305043,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004655126735723484,
      "loss": 5.7634,
      "mean_token_accuracy": 0.17058760523796082,
      "num_tokens": 36537231.0,
      "step": 18685
    },
    {
      "entropy": 6.145179891586304,
      "epoch": 1.8682960963662718,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046549368336047415,
      "loss": 5.8086,
      "mean_token_accuracy": 0.17157945781946182,
      "num_tokens": 36546524.0,
      "step": 18690
    },
    {
      "entropy": 6.052974033355713,
      "epoch": 1.8687959214275005,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004654746883558279,
      "loss": 5.6984,
      "mean_token_accuracy": 0.1744202643632889,
      "num_tokens": 36557064.0,
      "step": 18695
    },
    {
      "entropy": 5.950128269195557,
      "epoch": 1.869295746488729,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004654556885588875,
      "loss": 5.6162,
      "mean_token_accuracy": 0.1733273059129715,
      "num_tokens": 36565986.0,
      "step": 18700
    },
    {
      "entropy": 6.059814596176148,
      "epoch": 1.8697955715499575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046543668397013103,
      "loss": 5.782,
      "mean_token_accuracy": 0.16154301911592484,
      "num_tokens": 36575646.0,
      "step": 18705
    },
    {
      "entropy": 6.028028535842895,
      "epoch": 1.8702953966111862,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004654176745900365,
      "loss": 5.6931,
      "mean_token_accuracy": 0.17303369343280792,
      "num_tokens": 36585472.0,
      "step": 18710
    },
    {
      "entropy": 6.031279039382935,
      "epoch": 1.8707952216724146,
      "grad_norm": 1.25,
      "learning_rate": 0.00046539866041908233,
      "loss": 5.6811,
      "mean_token_accuracy": 0.179507714509964,
      "num_tokens": 36594309.0,
      "step": 18715
    },
    {
      "entropy": 6.122454118728638,
      "epoch": 1.871295046733643,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046537964145774687,
      "loss": 5.8174,
      "mean_token_accuracy": 0.1648060694336891,
      "num_tokens": 36604027.0,
      "step": 18720
    },
    {
      "entropy": 6.031936168670654,
      "epoch": 1.8717948717948718,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004653606177065086,
      "loss": 5.7215,
      "mean_token_accuracy": 0.17216662615537642,
      "num_tokens": 36613613.0,
      "step": 18725
    },
    {
      "entropy": 6.044622135162354,
      "epoch": 1.8722946968561005,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004653415891658462,
      "loss": 5.7611,
      "mean_token_accuracy": 0.17881008982658386,
      "num_tokens": 36624049.0,
      "step": 18730
    },
    {
      "entropy": 6.066958665847778,
      "epoch": 1.872794521917329,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046532255583623837,
      "loss": 5.7299,
      "mean_token_accuracy": 0.17104153633117675,
      "num_tokens": 36634815.0,
      "step": 18735
    },
    {
      "entropy": 6.12563271522522,
      "epoch": 1.8732943469785575,
      "grad_norm": 1.21875,
      "learning_rate": 0.000465303517718164,
      "loss": 5.786,
      "mean_token_accuracy": 0.1633099913597107,
      "num_tokens": 36644701.0,
      "step": 18740
    },
    {
      "entropy": 6.031326627731323,
      "epoch": 1.8737941720397862,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004652844748121021,
      "loss": 5.6948,
      "mean_token_accuracy": 0.17467292845249177,
      "num_tokens": 36653581.0,
      "step": 18745
    },
    {
      "entropy": 5.971631050109863,
      "epoch": 1.8742939971010146,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004652654271185318,
      "loss": 5.7511,
      "mean_token_accuracy": 0.16854411363601685,
      "num_tokens": 36662644.0,
      "step": 18750
    },
    {
      "entropy": 6.044879817962647,
      "epoch": 1.874793822162243,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046524637463793233,
      "loss": 5.7632,
      "mean_token_accuracy": 0.16743034273386,
      "num_tokens": 36673260.0,
      "step": 18755
    },
    {
      "entropy": 6.130960607528687,
      "epoch": 1.8752936472234718,
      "grad_norm": 1.125,
      "learning_rate": 0.000465227317370783,
      "loss": 5.8049,
      "mean_token_accuracy": 0.1701282262802124,
      "num_tokens": 36682899.0,
      "step": 18760
    },
    {
      "entropy": 6.05665774345398,
      "epoch": 1.8757934722847005,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046520825531756326,
      "loss": 5.7655,
      "mean_token_accuracy": 0.16566611975431442,
      "num_tokens": 36693190.0,
      "step": 18765
    },
    {
      "entropy": 5.975313758850097,
      "epoch": 1.876293297345929,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004651891884787528,
      "loss": 5.5955,
      "mean_token_accuracy": 0.17309701144695283,
      "num_tokens": 36701263.0,
      "step": 18770
    },
    {
      "entropy": 6.029411745071411,
      "epoch": 1.8767931224071575,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046517011685483136,
      "loss": 5.7386,
      "mean_token_accuracy": 0.16906986236572266,
      "num_tokens": 36710622.0,
      "step": 18775
    },
    {
      "entropy": 6.054358911514282,
      "epoch": 1.8772929474683862,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046515104044627867,
      "loss": 5.7344,
      "mean_token_accuracy": 0.16845835447311402,
      "num_tokens": 36720510.0,
      "step": 18780
    },
    {
      "entropy": 6.033756399154663,
      "epoch": 1.8777927725296146,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004651319592535747,
      "loss": 5.7853,
      "mean_token_accuracy": 0.16453448981046676,
      "num_tokens": 36730570.0,
      "step": 18785
    },
    {
      "entropy": 6.082548332214356,
      "epoch": 1.878292597590843,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004651128732771996,
      "loss": 5.7128,
      "mean_token_accuracy": 0.17275211364030837,
      "num_tokens": 36739834.0,
      "step": 18790
    },
    {
      "entropy": 6.031401348114014,
      "epoch": 1.8787924226520718,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046509378251763346,
      "loss": 5.6095,
      "mean_token_accuracy": 0.1756454661488533,
      "num_tokens": 36750193.0,
      "step": 18795
    },
    {
      "entropy": 6.093265438079834,
      "epoch": 1.8792922477133005,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004650746869753567,
      "loss": 5.734,
      "mean_token_accuracy": 0.16545181721448898,
      "num_tokens": 36759253.0,
      "step": 18800
    },
    {
      "entropy": 6.133418321609497,
      "epoch": 1.8797920727745288,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004650555866508498,
      "loss": 5.8111,
      "mean_token_accuracy": 0.16232355386018754,
      "num_tokens": 36768841.0,
      "step": 18805
    },
    {
      "entropy": 6.067687129974365,
      "epoch": 1.8802918978357575,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046503648154459315,
      "loss": 5.8527,
      "mean_token_accuracy": 0.1605804443359375,
      "num_tokens": 36780146.0,
      "step": 18810
    },
    {
      "entropy": 6.1102025508880615,
      "epoch": 1.8807917228969862,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004650173716570676,
      "loss": 5.7596,
      "mean_token_accuracy": 0.16325363963842393,
      "num_tokens": 36789948.0,
      "step": 18815
    },
    {
      "entropy": 6.021517896652222,
      "epoch": 1.8812915479582146,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046499825698875384,
      "loss": 5.6409,
      "mean_token_accuracy": 0.17799757570028304,
      "num_tokens": 36799496.0,
      "step": 18820
    },
    {
      "entropy": 6.024153709411621,
      "epoch": 1.881791373019443,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046497913754013284,
      "loss": 5.7601,
      "mean_token_accuracy": 0.1687973305583,
      "num_tokens": 36809905.0,
      "step": 18825
    },
    {
      "entropy": 6.05136022567749,
      "epoch": 1.8822911980806718,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004649600133116856,
      "loss": 5.7372,
      "mean_token_accuracy": 0.16457730084657668,
      "num_tokens": 36819855.0,
      "step": 18830
    },
    {
      "entropy": 6.055913782119751,
      "epoch": 1.8827910231419005,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046494088430389335,
      "loss": 5.7595,
      "mean_token_accuracy": 0.16741115525364875,
      "num_tokens": 36830456.0,
      "step": 18835
    },
    {
      "entropy": 5.954379034042359,
      "epoch": 1.8832908482031288,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004649217505172373,
      "loss": 5.6788,
      "mean_token_accuracy": 0.17773350924253464,
      "num_tokens": 36839914.0,
      "step": 18840
    },
    {
      "entropy": 6.105844449996948,
      "epoch": 1.8837906732643575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004649026119521989,
      "loss": 5.6238,
      "mean_token_accuracy": 0.17401022017002105,
      "num_tokens": 36848644.0,
      "step": 18845
    },
    {
      "entropy": 6.00680947303772,
      "epoch": 1.8842904983255861,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046488346860925955,
      "loss": 5.732,
      "mean_token_accuracy": 0.17057296335697175,
      "num_tokens": 36857490.0,
      "step": 18850
    },
    {
      "entropy": 6.159400939941406,
      "epoch": 1.8847903233868146,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046486432048890104,
      "loss": 5.9159,
      "mean_token_accuracy": 0.16038260310888292,
      "num_tokens": 36867617.0,
      "step": 18855
    },
    {
      "entropy": 6.083397293090821,
      "epoch": 1.885290148448043,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004648451675916051,
      "loss": 5.7447,
      "mean_token_accuracy": 0.16964511573314667,
      "num_tokens": 36876626.0,
      "step": 18860
    },
    {
      "entropy": 6.082736873626709,
      "epoch": 1.8857899735092718,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046482600991785365,
      "loss": 5.7866,
      "mean_token_accuracy": 0.16290743499994279,
      "num_tokens": 36887693.0,
      "step": 18865
    },
    {
      "entropy": 6.116111707687378,
      "epoch": 1.8862897985705003,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046480684746812856,
      "loss": 5.7955,
      "mean_token_accuracy": 0.17398798912763597,
      "num_tokens": 36897817.0,
      "step": 18870
    },
    {
      "entropy": 6.042200088500977,
      "epoch": 1.8867896236317288,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046478768024291205,
      "loss": 5.7183,
      "mean_token_accuracy": 0.17387909442186356,
      "num_tokens": 36908154.0,
      "step": 18875
    },
    {
      "entropy": 6.065615224838257,
      "epoch": 1.8872894486929574,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046476850824268633,
      "loss": 5.7211,
      "mean_token_accuracy": 0.16260092705488205,
      "num_tokens": 36917748.0,
      "step": 18880
    },
    {
      "entropy": 5.99603419303894,
      "epoch": 1.8877892737541861,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004647493314679338,
      "loss": 5.7381,
      "mean_token_accuracy": 0.17880412936210632,
      "num_tokens": 36927128.0,
      "step": 18885
    },
    {
      "entropy": 5.9762814998626705,
      "epoch": 1.8882890988154146,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000464730149919137,
      "loss": 5.6645,
      "mean_token_accuracy": 0.17610636055469514,
      "num_tokens": 36936822.0,
      "step": 18890
    },
    {
      "entropy": 6.131170034408569,
      "epoch": 1.888788923876643,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004647109635967783,
      "loss": 5.7854,
      "mean_token_accuracy": 0.16517676860094072,
      "num_tokens": 36946781.0,
      "step": 18895
    },
    {
      "entropy": 6.062129735946655,
      "epoch": 1.8892887489378718,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004646917725013406,
      "loss": 5.7085,
      "mean_token_accuracy": 0.17274270206689835,
      "num_tokens": 36956998.0,
      "step": 18900
    },
    {
      "entropy": 6.086064863204956,
      "epoch": 1.8897885739991003,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004646725766333068,
      "loss": 5.7591,
      "mean_token_accuracy": 0.16620344072580337,
      "num_tokens": 36966138.0,
      "step": 18905
    },
    {
      "entropy": 5.963523483276367,
      "epoch": 1.8902883990603288,
      "grad_norm": 1.25,
      "learning_rate": 0.00046465337599315966,
      "loss": 5.7102,
      "mean_token_accuracy": 0.1722027137875557,
      "num_tokens": 36976282.0,
      "step": 18910
    },
    {
      "entropy": 6.075245141983032,
      "epoch": 1.8907882241215574,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046463417058138247,
      "loss": 5.7732,
      "mean_token_accuracy": 0.1656029537320137,
      "num_tokens": 36986795.0,
      "step": 18915
    },
    {
      "entropy": 6.024499559402466,
      "epoch": 1.8912880491827861,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004646149603984583,
      "loss": 5.7229,
      "mean_token_accuracy": 0.1687234103679657,
      "num_tokens": 36997334.0,
      "step": 18920
    },
    {
      "entropy": 6.028505802154541,
      "epoch": 1.8917878742440146,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004645957454448706,
      "loss": 5.6779,
      "mean_token_accuracy": 0.17228348702192306,
      "num_tokens": 37006978.0,
      "step": 18925
    },
    {
      "entropy": 5.9859534740448,
      "epoch": 1.892287699305243,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004645765257211026,
      "loss": 5.6816,
      "mean_token_accuracy": 0.18252800554037094,
      "num_tokens": 37016821.0,
      "step": 18930
    },
    {
      "entropy": 5.981835079193115,
      "epoch": 1.8927875243664718,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046455730122763804,
      "loss": 5.6898,
      "mean_token_accuracy": 0.175089493393898,
      "num_tokens": 37026596.0,
      "step": 18935
    },
    {
      "entropy": 6.039088535308838,
      "epoch": 1.8932873494277003,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004645380719649606,
      "loss": 5.7265,
      "mean_token_accuracy": 0.1674495205283165,
      "num_tokens": 37035899.0,
      "step": 18940
    },
    {
      "entropy": 6.05708155632019,
      "epoch": 1.8937871744889287,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046451883793355406,
      "loss": 5.8127,
      "mean_token_accuracy": 0.16186247915029525,
      "num_tokens": 37045747.0,
      "step": 18945
    },
    {
      "entropy": 6.062295007705688,
      "epoch": 1.8942869995501574,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046449959913390227,
      "loss": 5.7747,
      "mean_token_accuracy": 0.16448410749435424,
      "num_tokens": 37056673.0,
      "step": 18950
    },
    {
      "entropy": 6.018760299682617,
      "epoch": 1.8947868246113861,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046448035556648936,
      "loss": 5.7509,
      "mean_token_accuracy": 0.17022194415330888,
      "num_tokens": 37065375.0,
      "step": 18955
    },
    {
      "entropy": 6.108520746231079,
      "epoch": 1.8952866496726146,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046446110723179945,
      "loss": 5.7074,
      "mean_token_accuracy": 0.17693106681108475,
      "num_tokens": 37074088.0,
      "step": 18960
    },
    {
      "entropy": 6.011275577545166,
      "epoch": 1.895786474733843,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004644418541303169,
      "loss": 5.6971,
      "mean_token_accuracy": 0.1663039281964302,
      "num_tokens": 37085418.0,
      "step": 18965
    },
    {
      "entropy": 6.0478047847747805,
      "epoch": 1.8962862997950718,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046442259626252593,
      "loss": 5.784,
      "mean_token_accuracy": 0.16710949093103408,
      "num_tokens": 37095029.0,
      "step": 18970
    },
    {
      "entropy": 6.097657299041748,
      "epoch": 1.8967861248563003,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046440333362891123,
      "loss": 5.6848,
      "mean_token_accuracy": 0.17944256365299224,
      "num_tokens": 37104398.0,
      "step": 18975
    },
    {
      "entropy": 6.181704330444336,
      "epoch": 1.8972859499175287,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004643840662299574,
      "loss": 5.7985,
      "mean_token_accuracy": 0.1668916568160057,
      "num_tokens": 37114546.0,
      "step": 18980
    },
    {
      "entropy": 6.008179426193237,
      "epoch": 1.8977857749787574,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004643647940661492,
      "loss": 5.6568,
      "mean_token_accuracy": 0.18543745279312135,
      "num_tokens": 37124113.0,
      "step": 18985
    },
    {
      "entropy": 6.064219999313354,
      "epoch": 1.8982856000399861,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046434551713797146,
      "loss": 5.7604,
      "mean_token_accuracy": 0.17150904685258866,
      "num_tokens": 37133661.0,
      "step": 18990
    },
    {
      "entropy": 6.065147829055786,
      "epoch": 1.8987854251012146,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004643262354459093,
      "loss": 5.7737,
      "mean_token_accuracy": 0.16883448660373687,
      "num_tokens": 37143987.0,
      "step": 18995
    },
    {
      "entropy": 6.1252306461334225,
      "epoch": 1.899285250162443,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046430694899044774,
      "loss": 5.8533,
      "mean_token_accuracy": 0.16827880293130876,
      "num_tokens": 37153227.0,
      "step": 19000
    },
    {
      "entropy": 6.117315244674683,
      "epoch": 1.8997850752236718,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000464287657772072,
      "loss": 5.7778,
      "mean_token_accuracy": 0.16745907366275786,
      "num_tokens": 37162524.0,
      "step": 19005
    },
    {
      "entropy": 6.06255612373352,
      "epoch": 1.9002849002849003,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004642683617912676,
      "loss": 5.7334,
      "mean_token_accuracy": 0.17320870608091354,
      "num_tokens": 37172729.0,
      "step": 19010
    },
    {
      "entropy": 6.019890689849854,
      "epoch": 1.9007847253461287,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046424906104851994,
      "loss": 5.6938,
      "mean_token_accuracy": 0.173066945374012,
      "num_tokens": 37182561.0,
      "step": 19015
    },
    {
      "entropy": 6.104333543777466,
      "epoch": 1.9012845504073574,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046422975554431453,
      "loss": 5.769,
      "mean_token_accuracy": 0.16884943395853041,
      "num_tokens": 37192089.0,
      "step": 19020
    },
    {
      "entropy": 6.046509647369385,
      "epoch": 1.9017843754685861,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046421044527913713,
      "loss": 5.7062,
      "mean_token_accuracy": 0.17405100613832475,
      "num_tokens": 37200739.0,
      "step": 19025
    },
    {
      "entropy": 6.063115167617798,
      "epoch": 1.9022842005298146,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004641911302534737,
      "loss": 5.7729,
      "mean_token_accuracy": 0.167390114068985,
      "num_tokens": 37209507.0,
      "step": 19030
    },
    {
      "entropy": 6.056872415542602,
      "epoch": 1.902784025591043,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046417181046781004,
      "loss": 5.7245,
      "mean_token_accuracy": 0.1785199373960495,
      "num_tokens": 37219252.0,
      "step": 19035
    },
    {
      "entropy": 6.0452436923980715,
      "epoch": 1.9032838506522718,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004641524859226324,
      "loss": 5.672,
      "mean_token_accuracy": 0.17199978083372117,
      "num_tokens": 37229925.0,
      "step": 19040
    },
    {
      "entropy": 5.9875589370727536,
      "epoch": 1.9037836757135003,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046413315661842676,
      "loss": 5.7331,
      "mean_token_accuracy": 0.17435779571533203,
      "num_tokens": 37239513.0,
      "step": 19045
    },
    {
      "entropy": 6.04871940612793,
      "epoch": 1.9042835007747287,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004641138225556797,
      "loss": 5.7022,
      "mean_token_accuracy": 0.177726349234581,
      "num_tokens": 37249143.0,
      "step": 19050
    },
    {
      "entropy": 6.079456090927124,
      "epoch": 1.9047833258359574,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004640944837348774,
      "loss": 5.7057,
      "mean_token_accuracy": 0.1697197064757347,
      "num_tokens": 37258834.0,
      "step": 19055
    },
    {
      "entropy": 6.022529268264771,
      "epoch": 1.9052831508971861,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046407514015650667,
      "loss": 5.7687,
      "mean_token_accuracy": 0.1648374080657959,
      "num_tokens": 37269554.0,
      "step": 19060
    },
    {
      "entropy": 6.077891540527344,
      "epoch": 1.9057829759584144,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046405579182105393,
      "loss": 5.7555,
      "mean_token_accuracy": 0.17224454879760742,
      "num_tokens": 37278842.0,
      "step": 19065
    },
    {
      "entropy": 6.006040477752686,
      "epoch": 1.906282801019643,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004640364387290062,
      "loss": 5.7014,
      "mean_token_accuracy": 0.17941470742225646,
      "num_tokens": 37289056.0,
      "step": 19070
    },
    {
      "entropy": 6.028718090057373,
      "epoch": 1.9067826260808718,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046401708088085036,
      "loss": 5.7357,
      "mean_token_accuracy": 0.1762111723423004,
      "num_tokens": 37298823.0,
      "step": 19075
    },
    {
      "entropy": 6.147389078140259,
      "epoch": 1.9072824511421003,
      "grad_norm": 1.125,
      "learning_rate": 0.00046399771827707335,
      "loss": 5.8576,
      "mean_token_accuracy": 0.16345098987221718,
      "num_tokens": 37310065.0,
      "step": 19080
    },
    {
      "entropy": 6.092795753479004,
      "epoch": 1.9077822762033287,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046397835091816235,
      "loss": 5.7381,
      "mean_token_accuracy": 0.16464521139860153,
      "num_tokens": 37320856.0,
      "step": 19085
    },
    {
      "entropy": 6.088702011108398,
      "epoch": 1.9082821012645574,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046395897880460474,
      "loss": 5.7468,
      "mean_token_accuracy": 0.17334351539611817,
      "num_tokens": 37330705.0,
      "step": 19090
    },
    {
      "entropy": 6.034607076644898,
      "epoch": 1.9087819263257861,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046393960193688783,
      "loss": 5.7504,
      "mean_token_accuracy": 0.17301979511976243,
      "num_tokens": 37339450.0,
      "step": 19095
    },
    {
      "entropy": 6.0350728034973145,
      "epoch": 1.9092817513870144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046392022031549907,
      "loss": 5.7151,
      "mean_token_accuracy": 0.17688481658697128,
      "num_tokens": 37349173.0,
      "step": 19100
    },
    {
      "entropy": 6.079658651351929,
      "epoch": 1.909781576448243,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004639008339409263,
      "loss": 5.8321,
      "mean_token_accuracy": 0.1599546790122986,
      "num_tokens": 37359246.0,
      "step": 19105
    },
    {
      "entropy": 6.119112634658814,
      "epoch": 1.9102814015094718,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004638814428136571,
      "loss": 5.8057,
      "mean_token_accuracy": 0.17009767442941665,
      "num_tokens": 37368056.0,
      "step": 19110
    },
    {
      "entropy": 6.201664209365845,
      "epoch": 1.9107812265707003,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004638620469341794,
      "loss": 5.996,
      "mean_token_accuracy": 0.15521800220012666,
      "num_tokens": 37377997.0,
      "step": 19115
    },
    {
      "entropy": 6.0499913692474365,
      "epoch": 1.9112810516319287,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004638426463029812,
      "loss": 5.6863,
      "mean_token_accuracy": 0.1710143655538559,
      "num_tokens": 37387365.0,
      "step": 19120
    },
    {
      "entropy": 6.025808238983155,
      "epoch": 1.9117808766931574,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004638232409205506,
      "loss": 5.7257,
      "mean_token_accuracy": 0.1694667488336563,
      "num_tokens": 37396649.0,
      "step": 19125
    },
    {
      "entropy": 6.080922079086304,
      "epoch": 1.912280701754386,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046380383078737584,
      "loss": 5.8436,
      "mean_token_accuracy": 0.16327402740716934,
      "num_tokens": 37407576.0,
      "step": 19130
    },
    {
      "entropy": 6.0935405731201175,
      "epoch": 1.9127805268156144,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046378441590394525,
      "loss": 5.7465,
      "mean_token_accuracy": 0.16785481572151184,
      "num_tokens": 37419270.0,
      "step": 19135
    },
    {
      "entropy": 6.080062437057495,
      "epoch": 1.913280351876843,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004637649962707474,
      "loss": 5.7911,
      "mean_token_accuracy": 0.16064177006483077,
      "num_tokens": 37429155.0,
      "step": 19140
    },
    {
      "entropy": 6.056681728363037,
      "epoch": 1.9137801769380718,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046374557188827084,
      "loss": 5.7866,
      "mean_token_accuracy": 0.1661999613046646,
      "num_tokens": 37439952.0,
      "step": 19145
    },
    {
      "entropy": 6.096636867523193,
      "epoch": 1.9142800019993003,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004637261427570042,
      "loss": 5.7392,
      "mean_token_accuracy": 0.16804916709661483,
      "num_tokens": 37449537.0,
      "step": 19150
    },
    {
      "entropy": 6.063786745071411,
      "epoch": 1.9147798270605287,
      "grad_norm": 1.125,
      "learning_rate": 0.0004637067088774364,
      "loss": 5.7554,
      "mean_token_accuracy": 0.17516845911741258,
      "num_tokens": 37460202.0,
      "step": 19155
    },
    {
      "entropy": 6.051062154769897,
      "epoch": 1.9152796521217574,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046368727025005634,
      "loss": 5.7039,
      "mean_token_accuracy": 0.17159729301929474,
      "num_tokens": 37469715.0,
      "step": 19160
    },
    {
      "entropy": 6.028839874267578,
      "epoch": 1.915779477182986,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046366782687535315,
      "loss": 5.7796,
      "mean_token_accuracy": 0.16818709969520568,
      "num_tokens": 37478495.0,
      "step": 19165
    },
    {
      "entropy": 6.104867649078369,
      "epoch": 1.9162793022442144,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000463648378753816,
      "loss": 5.8147,
      "mean_token_accuracy": 0.17088623344898224,
      "num_tokens": 37488125.0,
      "step": 19170
    },
    {
      "entropy": 6.054468441009521,
      "epoch": 1.916779127305443,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046362892588593406,
      "loss": 5.6472,
      "mean_token_accuracy": 0.17727805227041243,
      "num_tokens": 37498097.0,
      "step": 19175
    },
    {
      "entropy": 6.058595132827759,
      "epoch": 1.9172789523666718,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046360946827219703,
      "loss": 5.8043,
      "mean_token_accuracy": 0.17317695021629334,
      "num_tokens": 37507713.0,
      "step": 19180
    },
    {
      "entropy": 6.041218185424805,
      "epoch": 1.9177787774279003,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004635900059130942,
      "loss": 5.8302,
      "mean_token_accuracy": 0.16518302410840988,
      "num_tokens": 37517862.0,
      "step": 19185
    },
    {
      "entropy": 6.045876836776733,
      "epoch": 1.9182786024891287,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004635705388091154,
      "loss": 5.7512,
      "mean_token_accuracy": 0.16729546934366227,
      "num_tokens": 37528988.0,
      "step": 19190
    },
    {
      "entropy": 6.098604679107666,
      "epoch": 1.9187784275503574,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004635510669607504,
      "loss": 5.869,
      "mean_token_accuracy": 0.16229045242071152,
      "num_tokens": 37538467.0,
      "step": 19195
    },
    {
      "entropy": 6.11997127532959,
      "epoch": 1.919278252611586,
      "grad_norm": 1.171875,
      "learning_rate": 0.000463531590368489,
      "loss": 5.7043,
      "mean_token_accuracy": 0.1662604957818985,
      "num_tokens": 37547986.0,
      "step": 19200
    },
    {
      "entropy": 6.079444408416748,
      "epoch": 1.9197780776728144,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046351210903282136,
      "loss": 5.8143,
      "mean_token_accuracy": 0.17016706317663194,
      "num_tokens": 37557403.0,
      "step": 19205
    },
    {
      "entropy": 6.062028789520264,
      "epoch": 1.920277902734043,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046349262295423754,
      "loss": 5.8088,
      "mean_token_accuracy": 0.1655028536915779,
      "num_tokens": 37568504.0,
      "step": 19210
    },
    {
      "entropy": 6.061728668212891,
      "epoch": 1.9207777277952718,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046347313213322786,
      "loss": 5.7172,
      "mean_token_accuracy": 0.1766082227230072,
      "num_tokens": 37577816.0,
      "step": 19215
    },
    {
      "entropy": 6.020142650604248,
      "epoch": 1.9212775528565003,
      "grad_norm": 1.125,
      "learning_rate": 0.0004634536365702826,
      "loss": 5.6796,
      "mean_token_accuracy": 0.1727835163474083,
      "num_tokens": 37587315.0,
      "step": 19220
    },
    {
      "entropy": 6.140668535232544,
      "epoch": 1.9217773779177287,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046343413626589243,
      "loss": 5.816,
      "mean_token_accuracy": 0.15932586789131165,
      "num_tokens": 37597708.0,
      "step": 19225
    },
    {
      "entropy": 6.003057527542114,
      "epoch": 1.9222772029789574,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004634146312205478,
      "loss": 5.6572,
      "mean_token_accuracy": 0.1797218292951584,
      "num_tokens": 37608093.0,
      "step": 19230
    },
    {
      "entropy": 5.942911672592163,
      "epoch": 1.922777028040186,
      "grad_norm": 1.25,
      "learning_rate": 0.00046339512143473956,
      "loss": 5.6778,
      "mean_token_accuracy": 0.17225816398859023,
      "num_tokens": 37617742.0,
      "step": 19235
    },
    {
      "entropy": 6.015300273895264,
      "epoch": 1.9232768531014144,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046337560690895856,
      "loss": 5.7284,
      "mean_token_accuracy": 0.16880898922681808,
      "num_tokens": 37626999.0,
      "step": 19240
    },
    {
      "entropy": 6.090170574188233,
      "epoch": 1.923776678162643,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046335608764369574,
      "loss": 5.8446,
      "mean_token_accuracy": 0.16221433281898498,
      "num_tokens": 37638142.0,
      "step": 19245
    },
    {
      "entropy": 6.056753206253052,
      "epoch": 1.9242765032238718,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046333656363944226,
      "loss": 5.7471,
      "mean_token_accuracy": 0.1709660232067108,
      "num_tokens": 37647373.0,
      "step": 19250
    },
    {
      "entropy": 6.0005731105804445,
      "epoch": 1.9247763282851003,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046331703489668925,
      "loss": 5.6828,
      "mean_token_accuracy": 0.17413519769907,
      "num_tokens": 37658677.0,
      "step": 19255
    },
    {
      "entropy": 5.988557291030884,
      "epoch": 1.9252761533463287,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046329750141592814,
      "loss": 5.5985,
      "mean_token_accuracy": 0.17969269305467606,
      "num_tokens": 37668344.0,
      "step": 19260
    },
    {
      "entropy": 5.994919109344482,
      "epoch": 1.9257759784075574,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004632779631976503,
      "loss": 5.7036,
      "mean_token_accuracy": 0.16470025032758712,
      "num_tokens": 37678000.0,
      "step": 19265
    },
    {
      "entropy": 6.087340927124023,
      "epoch": 1.926275803468786,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046325842024234735,
      "loss": 5.8555,
      "mean_token_accuracy": 0.16163640171289445,
      "num_tokens": 37688404.0,
      "step": 19270
    },
    {
      "entropy": 6.144286251068115,
      "epoch": 1.9267756285300144,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046323887255051105,
      "loss": 5.7994,
      "mean_token_accuracy": 0.1674676701426506,
      "num_tokens": 37697699.0,
      "step": 19275
    },
    {
      "entropy": 6.052452278137207,
      "epoch": 1.927275453591243,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046321932012263317,
      "loss": 5.6573,
      "mean_token_accuracy": 0.17490581870079042,
      "num_tokens": 37707261.0,
      "step": 19280
    },
    {
      "entropy": 5.998844385147095,
      "epoch": 1.9277752786524718,
      "grad_norm": 1.125,
      "learning_rate": 0.0004631997629592055,
      "loss": 5.6787,
      "mean_token_accuracy": 0.17709397822618483,
      "num_tokens": 37716500.0,
      "step": 19285
    },
    {
      "entropy": 5.94435133934021,
      "epoch": 1.9282751037137003,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004631802010607204,
      "loss": 5.63,
      "mean_token_accuracy": 0.17945474684238433,
      "num_tokens": 37725881.0,
      "step": 19290
    },
    {
      "entropy": 5.955582189559936,
      "epoch": 1.9287749287749287,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046316063442766975,
      "loss": 5.7164,
      "mean_token_accuracy": 0.1716433897614479,
      "num_tokens": 37736233.0,
      "step": 19295
    },
    {
      "entropy": 6.089796257019043,
      "epoch": 1.9292747538361574,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046314106306054596,
      "loss": 5.7561,
      "mean_token_accuracy": 0.16808542907238005,
      "num_tokens": 37747230.0,
      "step": 19300
    },
    {
      "entropy": 6.070726680755615,
      "epoch": 1.929774578897386,
      "grad_norm": 1.125,
      "learning_rate": 0.00046312148695984147,
      "loss": 5.7363,
      "mean_token_accuracy": 0.17420753240585327,
      "num_tokens": 37758470.0,
      "step": 19305
    },
    {
      "entropy": 6.124831295013427,
      "epoch": 1.9302744039586144,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004631019061260488,
      "loss": 5.8732,
      "mean_token_accuracy": 0.16375025361776352,
      "num_tokens": 37767912.0,
      "step": 19310
    },
    {
      "entropy": 6.017858362197876,
      "epoch": 1.930774229019843,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004630823205596605,
      "loss": 5.6906,
      "mean_token_accuracy": 0.16743541657924652,
      "num_tokens": 37776847.0,
      "step": 19315
    },
    {
      "entropy": 6.082772493362427,
      "epoch": 1.9312740540810718,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046306273026116944,
      "loss": 5.7919,
      "mean_token_accuracy": 0.16594902276992798,
      "num_tokens": 37786960.0,
      "step": 19320
    },
    {
      "entropy": 6.052681636810303,
      "epoch": 1.9317738791423,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004630431352310685,
      "loss": 5.7609,
      "mean_token_accuracy": 0.16332867443561555,
      "num_tokens": 37797007.0,
      "step": 19325
    },
    {
      "entropy": 6.002829456329346,
      "epoch": 1.9322737042035287,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046302353546985066,
      "loss": 5.682,
      "mean_token_accuracy": 0.17297326475381852,
      "num_tokens": 37806877.0,
      "step": 19330
    },
    {
      "entropy": 6.0623753547668455,
      "epoch": 1.9327735292647574,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000463003930978009,
      "loss": 5.6772,
      "mean_token_accuracy": 0.17363169491291047,
      "num_tokens": 37815513.0,
      "step": 19335
    },
    {
      "entropy": 6.000029182434082,
      "epoch": 1.933273354325986,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004629843217560369,
      "loss": 5.6799,
      "mean_token_accuracy": 0.1733634039759636,
      "num_tokens": 37825240.0,
      "step": 19340
    },
    {
      "entropy": 6.065841722488403,
      "epoch": 1.9337731793872144,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004629647078044276,
      "loss": 5.748,
      "mean_token_accuracy": 0.16917895525693893,
      "num_tokens": 37834333.0,
      "step": 19345
    },
    {
      "entropy": 6.0676717281341555,
      "epoch": 1.934273004448443,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004629450891236746,
      "loss": 5.6952,
      "mean_token_accuracy": 0.17238609939813615,
      "num_tokens": 37844548.0,
      "step": 19350
    },
    {
      "entropy": 6.045228481292725,
      "epoch": 1.9347728295096718,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004629254657142715,
      "loss": 5.7106,
      "mean_token_accuracy": 0.16965472102165222,
      "num_tokens": 37853995.0,
      "step": 19355
    },
    {
      "entropy": 6.097453212738037,
      "epoch": 1.9352726545709,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004629058375767121,
      "loss": 5.7853,
      "mean_token_accuracy": 0.1664825588464737,
      "num_tokens": 37863472.0,
      "step": 19360
    },
    {
      "entropy": 6.125846815109253,
      "epoch": 1.9357724796321287,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046288620471149014,
      "loss": 5.744,
      "mean_token_accuracy": 0.17435512095689773,
      "num_tokens": 37872690.0,
      "step": 19365
    },
    {
      "entropy": 6.098662948608398,
      "epoch": 1.9362723046933574,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046286656711909966,
      "loss": 5.7807,
      "mean_token_accuracy": 0.16908457651734352,
      "num_tokens": 37882296.0,
      "step": 19370
    },
    {
      "entropy": 6.013405227661133,
      "epoch": 1.936772129754586,
      "grad_norm": 1.25,
      "learning_rate": 0.00046284692480003467,
      "loss": 5.7205,
      "mean_token_accuracy": 0.169911889731884,
      "num_tokens": 37891612.0,
      "step": 19375
    },
    {
      "entropy": 6.0535825252532955,
      "epoch": 1.9372719548158144,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046282727775478935,
      "loss": 5.7022,
      "mean_token_accuracy": 0.17639618515968322,
      "num_tokens": 37901107.0,
      "step": 19380
    },
    {
      "entropy": 6.078053569793701,
      "epoch": 1.937771779877043,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004628076259838581,
      "loss": 5.881,
      "mean_token_accuracy": 0.15789297819137574,
      "num_tokens": 37912020.0,
      "step": 19385
    },
    {
      "entropy": 6.043918466567993,
      "epoch": 1.9382716049382716,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046278796948773525,
      "loss": 5.714,
      "mean_token_accuracy": 0.17442672699689865,
      "num_tokens": 37922681.0,
      "step": 19390
    },
    {
      "entropy": 6.091444778442383,
      "epoch": 1.9387714299995,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046276830826691553,
      "loss": 5.7669,
      "mean_token_accuracy": 0.166643388569355,
      "num_tokens": 37934322.0,
      "step": 19395
    },
    {
      "entropy": 6.028872728347778,
      "epoch": 1.9392712550607287,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046274864232189334,
      "loss": 5.7381,
      "mean_token_accuracy": 0.1749526247382164,
      "num_tokens": 37945028.0,
      "step": 19400
    },
    {
      "entropy": 5.977642345428467,
      "epoch": 1.9397710801219574,
      "grad_norm": 1.25,
      "learning_rate": 0.00046272897165316367,
      "loss": 5.6352,
      "mean_token_accuracy": 0.17351069003343583,
      "num_tokens": 37955055.0,
      "step": 19405
    },
    {
      "entropy": 6.04843053817749,
      "epoch": 1.940270905183186,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004627092962612215,
      "loss": 5.8213,
      "mean_token_accuracy": 0.16416927576065063,
      "num_tokens": 37966018.0,
      "step": 19410
    },
    {
      "entropy": 6.025820016860962,
      "epoch": 1.9407707302444144,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004626896161465616,
      "loss": 5.6926,
      "mean_token_accuracy": 0.1762984961271286,
      "num_tokens": 37975541.0,
      "step": 19415
    },
    {
      "entropy": 6.071085786819458,
      "epoch": 1.941270555305643,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046266993130967925,
      "loss": 5.7573,
      "mean_token_accuracy": 0.16360429525375367,
      "num_tokens": 37986312.0,
      "step": 19420
    },
    {
      "entropy": 6.091105890274048,
      "epoch": 1.9417703803668716,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004626502417510698,
      "loss": 5.744,
      "mean_token_accuracy": 0.17030294835567475,
      "num_tokens": 37994717.0,
      "step": 19425
    },
    {
      "entropy": 6.0861732006073,
      "epoch": 1.9422702054281,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046263054747122857,
      "loss": 5.8158,
      "mean_token_accuracy": 0.16568248271942138,
      "num_tokens": 38004282.0,
      "step": 19430
    },
    {
      "entropy": 6.0631516456604,
      "epoch": 1.9427700304893287,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004626108484706509,
      "loss": 5.7485,
      "mean_token_accuracy": 0.17075673192739488,
      "num_tokens": 38013255.0,
      "step": 19435
    },
    {
      "entropy": 6.062076044082642,
      "epoch": 1.9432698555505574,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046259114474983266,
      "loss": 5.7457,
      "mean_token_accuracy": 0.1699928194284439,
      "num_tokens": 38023041.0,
      "step": 19440
    },
    {
      "entropy": 6.068369960784912,
      "epoch": 1.943769680611786,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046257143630926953,
      "loss": 5.7397,
      "mean_token_accuracy": 0.1745337128639221,
      "num_tokens": 38032193.0,
      "step": 19445
    },
    {
      "entropy": 5.986350727081299,
      "epoch": 1.9442695056730144,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004625517231494573,
      "loss": 5.7336,
      "mean_token_accuracy": 0.16847582459449767,
      "num_tokens": 38041470.0,
      "step": 19450
    },
    {
      "entropy": 5.936720752716065,
      "epoch": 1.944769330734243,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004625320052708919,
      "loss": 5.5907,
      "mean_token_accuracy": 0.18271080553531646,
      "num_tokens": 38051997.0,
      "step": 19455
    },
    {
      "entropy": 5.958968257904052,
      "epoch": 1.9452691557954715,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004625122826740696,
      "loss": 5.7597,
      "mean_token_accuracy": 0.17531616538763045,
      "num_tokens": 38061418.0,
      "step": 19460
    },
    {
      "entropy": 5.939451551437378,
      "epoch": 1.9457689808567,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004624925553594865,
      "loss": 5.6493,
      "mean_token_accuracy": 0.17442776560783385,
      "num_tokens": 38070853.0,
      "step": 19465
    },
    {
      "entropy": 6.083439731597901,
      "epoch": 1.9462688059179287,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004624728233276389,
      "loss": 5.7643,
      "mean_token_accuracy": 0.16667909920215607,
      "num_tokens": 38080565.0,
      "step": 19470
    },
    {
      "entropy": 6.080991935729981,
      "epoch": 1.9467686309791574,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046245308657902334,
      "loss": 5.6679,
      "mean_token_accuracy": 0.17863743752241135,
      "num_tokens": 38090899.0,
      "step": 19475
    },
    {
      "entropy": 5.945226383209229,
      "epoch": 1.947268456040386,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004624333451141364,
      "loss": 5.6835,
      "mean_token_accuracy": 0.1766862079501152,
      "num_tokens": 38100104.0,
      "step": 19480
    },
    {
      "entropy": 5.997041940689087,
      "epoch": 1.9477682811016144,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004624135989334746,
      "loss": 5.7543,
      "mean_token_accuracy": 0.17217225283384324,
      "num_tokens": 38109720.0,
      "step": 19485
    },
    {
      "entropy": 6.022518444061279,
      "epoch": 1.948268106162843,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046239384803753496,
      "loss": 5.7862,
      "mean_token_accuracy": 0.16664169132709503,
      "num_tokens": 38119703.0,
      "step": 19490
    },
    {
      "entropy": 6.052297639846802,
      "epoch": 1.9487679312240715,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046237409242681435,
      "loss": 5.6377,
      "mean_token_accuracy": 0.1721953868865967,
      "num_tokens": 38129329.0,
      "step": 19495
    },
    {
      "entropy": 6.086818599700928,
      "epoch": 1.9492677562853,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004623543321018097,
      "loss": 5.7152,
      "mean_token_accuracy": 0.16437384933233262,
      "num_tokens": 38138607.0,
      "step": 19500
    },
    {
      "entropy": 5.993689441680909,
      "epoch": 1.9497675813465287,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004623345670630184,
      "loss": 5.7079,
      "mean_token_accuracy": 0.17126376628875734,
      "num_tokens": 38148749.0,
      "step": 19505
    },
    {
      "entropy": 6.069580507278443,
      "epoch": 1.9502674064077574,
      "grad_norm": 1.125,
      "learning_rate": 0.0004623147973109375,
      "loss": 5.803,
      "mean_token_accuracy": 0.16660282164812087,
      "num_tokens": 38158929.0,
      "step": 19510
    },
    {
      "entropy": 6.090256786346435,
      "epoch": 1.950767231468986,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046229502284606455,
      "loss": 5.7393,
      "mean_token_accuracy": 0.1632416032254696,
      "num_tokens": 38169966.0,
      "step": 19515
    },
    {
      "entropy": 6.08269681930542,
      "epoch": 1.9512670565302144,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000462275243668897,
      "loss": 5.7732,
      "mean_token_accuracy": 0.16876624524593353,
      "num_tokens": 38180363.0,
      "step": 19520
    },
    {
      "entropy": 6.0360801219940186,
      "epoch": 1.951766881591443,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046225545977993245,
      "loss": 5.6859,
      "mean_token_accuracy": 0.17402878850698472,
      "num_tokens": 38190802.0,
      "step": 19525
    },
    {
      "entropy": 6.029698467254638,
      "epoch": 1.9522667066526715,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004622356711796688,
      "loss": 5.7321,
      "mean_token_accuracy": 0.1702787309885025,
      "num_tokens": 38201183.0,
      "step": 19530
    },
    {
      "entropy": 6.105321311950684,
      "epoch": 1.9527665317139,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004622158778686038,
      "loss": 5.8146,
      "mean_token_accuracy": 0.16219646632671356,
      "num_tokens": 38210953.0,
      "step": 19535
    },
    {
      "entropy": 6.007713508605957,
      "epoch": 1.9532663567751287,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046219607984723554,
      "loss": 5.6576,
      "mean_token_accuracy": 0.1770336225628853,
      "num_tokens": 38220335.0,
      "step": 19540
    },
    {
      "entropy": 6.025471019744873,
      "epoch": 1.9537661818363574,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046217627711606205,
      "loss": 5.7412,
      "mean_token_accuracy": 0.16494044810533523,
      "num_tokens": 38230530.0,
      "step": 19545
    },
    {
      "entropy": 6.088161754608154,
      "epoch": 1.954266006897586,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00046215646967558155,
      "loss": 5.6675,
      "mean_token_accuracy": 0.1747520849108696,
      "num_tokens": 38241275.0,
      "step": 19550
    },
    {
      "entropy": 6.163885927200317,
      "epoch": 1.9547658319588144,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046213665752629246,
      "loss": 5.6722,
      "mean_token_accuracy": 0.17954683303833008,
      "num_tokens": 38250158.0,
      "step": 19555
    },
    {
      "entropy": 6.110208797454834,
      "epoch": 1.955265657020043,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004621168406686933,
      "loss": 5.9093,
      "mean_token_accuracy": 0.16480902284383775,
      "num_tokens": 38259172.0,
      "step": 19560
    },
    {
      "entropy": 5.9689253807067875,
      "epoch": 1.9557654820812715,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046209701910328245,
      "loss": 5.7464,
      "mean_token_accuracy": 0.167494697868824,
      "num_tokens": 38268261.0,
      "step": 19565
    },
    {
      "entropy": 6.041550397872925,
      "epoch": 1.9562653071425,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004620771928305588,
      "loss": 5.7085,
      "mean_token_accuracy": 0.17461237758398057,
      "num_tokens": 38277843.0,
      "step": 19570
    },
    {
      "entropy": 5.982901382446289,
      "epoch": 1.9567651322037287,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004620573618510211,
      "loss": 5.5922,
      "mean_token_accuracy": 0.18088301420211791,
      "num_tokens": 38286878.0,
      "step": 19575
    },
    {
      "entropy": 6.013923978805542,
      "epoch": 1.9572649572649574,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046203752616516834,
      "loss": 5.6385,
      "mean_token_accuracy": 0.1778041660785675,
      "num_tokens": 38296289.0,
      "step": 19580
    },
    {
      "entropy": 6.0818225860595705,
      "epoch": 1.9577647823261857,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046201768577349957,
      "loss": 5.813,
      "mean_token_accuracy": 0.16244988292455673,
      "num_tokens": 38305238.0,
      "step": 19585
    },
    {
      "entropy": 6.003421831130981,
      "epoch": 1.9582646073874144,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004619978406765139,
      "loss": 5.7051,
      "mean_token_accuracy": 0.17093469500541686,
      "num_tokens": 38315079.0,
      "step": 19590
    },
    {
      "entropy": 6.088837814331055,
      "epoch": 1.958764432448643,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004619779908747107,
      "loss": 5.8168,
      "mean_token_accuracy": 0.16320439875125886,
      "num_tokens": 38324314.0,
      "step": 19595
    },
    {
      "entropy": 6.1452868461608885,
      "epoch": 1.9592642575098715,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004619581363685893,
      "loss": 5.7774,
      "mean_token_accuracy": 0.1674414485692978,
      "num_tokens": 38334708.0,
      "step": 19600
    },
    {
      "entropy": 6.134855031967163,
      "epoch": 1.9597640825711,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004619382771586494,
      "loss": 5.7574,
      "mean_token_accuracy": 0.1679583117365837,
      "num_tokens": 38344800.0,
      "step": 19605
    },
    {
      "entropy": 5.938862609863281,
      "epoch": 1.9602639076323287,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004619184132453905,
      "loss": 5.6769,
      "mean_token_accuracy": 0.17837360203266145,
      "num_tokens": 38354778.0,
      "step": 19610
    },
    {
      "entropy": 6.068877124786377,
      "epoch": 1.9607637326935574,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004618985446293125,
      "loss": 5.6678,
      "mean_token_accuracy": 0.1685323014855385,
      "num_tokens": 38365032.0,
      "step": 19615
    },
    {
      "entropy": 5.996339654922485,
      "epoch": 1.9612635577547857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046187867131091504,
      "loss": 5.6909,
      "mean_token_accuracy": 0.17214155793190003,
      "num_tokens": 38375051.0,
      "step": 19620
    },
    {
      "entropy": 5.9678973197937015,
      "epoch": 1.9617633828160144,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004618587932906984,
      "loss": 5.698,
      "mean_token_accuracy": 0.17447369545698166,
      "num_tokens": 38384107.0,
      "step": 19625
    },
    {
      "entropy": 5.956564569473267,
      "epoch": 1.962263207877243,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046183891056916256,
      "loss": 5.5727,
      "mean_token_accuracy": 0.18639166802167892,
      "num_tokens": 38394054.0,
      "step": 19630
    },
    {
      "entropy": 6.068665790557861,
      "epoch": 1.9627630329384715,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004618190231468079,
      "loss": 5.7546,
      "mean_token_accuracy": 0.166679222881794,
      "num_tokens": 38403001.0,
      "step": 19635
    },
    {
      "entropy": 5.929578828811645,
      "epoch": 1.9632628579997,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046179913102413456,
      "loss": 5.635,
      "mean_token_accuracy": 0.18139762580394744,
      "num_tokens": 38412752.0,
      "step": 19640
    },
    {
      "entropy": 6.104719781875611,
      "epoch": 1.9637626830609287,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004617792342016433,
      "loss": 5.8318,
      "mean_token_accuracy": 0.16574220061302186,
      "num_tokens": 38423523.0,
      "step": 19645
    },
    {
      "entropy": 6.188168287277222,
      "epoch": 1.9642625081221572,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004617593326798344,
      "loss": 5.7812,
      "mean_token_accuracy": 0.17137563228607178,
      "num_tokens": 38433062.0,
      "step": 19650
    },
    {
      "entropy": 6.082399892807007,
      "epoch": 1.9647623331833857,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004617394264592089,
      "loss": 5.8589,
      "mean_token_accuracy": 0.15725727528333663,
      "num_tokens": 38443008.0,
      "step": 19655
    },
    {
      "entropy": 6.011508274078369,
      "epoch": 1.9652621582446144,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004617195155402674,
      "loss": 5.6857,
      "mean_token_accuracy": 0.17865172326564788,
      "num_tokens": 38451805.0,
      "step": 19660
    },
    {
      "entropy": 6.103921890258789,
      "epoch": 1.965761983305843,
      "grad_norm": 1.25,
      "learning_rate": 0.00046169959992351103,
      "loss": 5.785,
      "mean_token_accuracy": 0.1630512684583664,
      "num_tokens": 38461826.0,
      "step": 19665
    },
    {
      "entropy": 6.043173789978027,
      "epoch": 1.9662618083670715,
      "grad_norm": 1.125,
      "learning_rate": 0.00046167967960944066,
      "loss": 5.7032,
      "mean_token_accuracy": 0.17060994505882263,
      "num_tokens": 38471746.0,
      "step": 19670
    },
    {
      "entropy": 6.050225400924683,
      "epoch": 1.9667616334283,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004616597545985577,
      "loss": 5.6853,
      "mean_token_accuracy": 0.17493851482868195,
      "num_tokens": 38480749.0,
      "step": 19675
    },
    {
      "entropy": 6.009523200988769,
      "epoch": 1.9672614584895287,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004616398248913632,
      "loss": 5.7415,
      "mean_token_accuracy": 0.16749725937843324,
      "num_tokens": 38491327.0,
      "step": 19680
    },
    {
      "entropy": 6.091417217254639,
      "epoch": 1.9677612835507572,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046161989048835886,
      "loss": 5.7841,
      "mean_token_accuracy": 0.16930991560220718,
      "num_tokens": 38500573.0,
      "step": 19685
    },
    {
      "entropy": 6.0420928478240965,
      "epoch": 1.9682611086119857,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046159995139004594,
      "loss": 5.7828,
      "mean_token_accuracy": 0.1686786577105522,
      "num_tokens": 38510156.0,
      "step": 19690
    },
    {
      "entropy": 6.068179178237915,
      "epoch": 1.9687609336732144,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004615800075969264,
      "loss": 5.7184,
      "mean_token_accuracy": 0.17326877415180206,
      "num_tokens": 38519804.0,
      "step": 19695
    },
    {
      "entropy": 6.113399505615234,
      "epoch": 1.969260758734443,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046156005910950187,
      "loss": 5.788,
      "mean_token_accuracy": 0.1713164657354355,
      "num_tokens": 38530332.0,
      "step": 19700
    },
    {
      "entropy": 6.137618637084961,
      "epoch": 1.9697605837956715,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004615401059282742,
      "loss": 5.7887,
      "mean_token_accuracy": 0.16423254758119582,
      "num_tokens": 38540480.0,
      "step": 19705
    },
    {
      "entropy": 6.049063062667846,
      "epoch": 1.9702604088569,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046152014805374546,
      "loss": 5.7137,
      "mean_token_accuracy": 0.17481796741485595,
      "num_tokens": 38549644.0,
      "step": 19710
    },
    {
      "entropy": 6.059892416000366,
      "epoch": 1.9707602339181287,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004615001854864179,
      "loss": 5.8107,
      "mean_token_accuracy": 0.16354365795850753,
      "num_tokens": 38560018.0,
      "step": 19715
    },
    {
      "entropy": 6.0678996562957765,
      "epoch": 1.9712600589793572,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046148021822679347,
      "loss": 5.6665,
      "mean_token_accuracy": 0.1753750815987587,
      "num_tokens": 38569584.0,
      "step": 19720
    },
    {
      "entropy": 6.090463352203369,
      "epoch": 1.9717598840405857,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004614602462753749,
      "loss": 5.7911,
      "mean_token_accuracy": 0.16732856184244155,
      "num_tokens": 38579358.0,
      "step": 19725
    },
    {
      "entropy": 5.960558366775513,
      "epoch": 1.9722597091018144,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004614402696326645,
      "loss": 5.6706,
      "mean_token_accuracy": 0.17523909509181976,
      "num_tokens": 38588850.0,
      "step": 19730
    },
    {
      "entropy": 6.029089879989624,
      "epoch": 1.972759534163043,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046142028829916485,
      "loss": 5.7736,
      "mean_token_accuracy": 0.1745072364807129,
      "num_tokens": 38598972.0,
      "step": 19735
    },
    {
      "entropy": 6.219658708572387,
      "epoch": 1.9732593592242715,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004614003022753787,
      "loss": 5.8762,
      "mean_token_accuracy": 0.16549083590507507,
      "num_tokens": 38609640.0,
      "step": 19740
    },
    {
      "entropy": 6.058277273178101,
      "epoch": 1.9737591842855,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004613803115618089,
      "loss": 5.7776,
      "mean_token_accuracy": 0.16859937310218812,
      "num_tokens": 38619487.0,
      "step": 19745
    },
    {
      "entropy": 6.1139151096344,
      "epoch": 1.9742590093467287,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046136031615895846,
      "loss": 5.8907,
      "mean_token_accuracy": 0.15986474901437758,
      "num_tokens": 38629681.0,
      "step": 19750
    },
    {
      "entropy": 6.108110857009888,
      "epoch": 1.9747588344079572,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004613403160673305,
      "loss": 5.8197,
      "mean_token_accuracy": 0.16743160039186478,
      "num_tokens": 38640003.0,
      "step": 19755
    },
    {
      "entropy": 6.073555183410645,
      "epoch": 1.9752586594691857,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000461320311287428,
      "loss": 5.6857,
      "mean_token_accuracy": 0.16956078112125397,
      "num_tokens": 38650691.0,
      "step": 19760
    },
    {
      "entropy": 6.035345983505249,
      "epoch": 1.9757584845304144,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046130030181975446,
      "loss": 5.7195,
      "mean_token_accuracy": 0.17016769647598268,
      "num_tokens": 38660144.0,
      "step": 19765
    },
    {
      "entropy": 6.034298706054687,
      "epoch": 1.976258309591643,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004612802876648133,
      "loss": 5.7773,
      "mean_token_accuracy": 0.16898135095834732,
      "num_tokens": 38670654.0,
      "step": 19770
    },
    {
      "entropy": 6.050202512741089,
      "epoch": 1.9767581346528715,
      "grad_norm": 1.0625,
      "learning_rate": 0.000461260268823108,
      "loss": 5.7235,
      "mean_token_accuracy": 0.16817380487918854,
      "num_tokens": 38681616.0,
      "step": 19775
    },
    {
      "entropy": 6.027742052078247,
      "epoch": 1.9772579597141,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046124024529514224,
      "loss": 5.6385,
      "mean_token_accuracy": 0.1836381435394287,
      "num_tokens": 38691586.0,
      "step": 19780
    },
    {
      "entropy": 6.134327173233032,
      "epoch": 1.9777577847753287,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004612202170814198,
      "loss": 5.833,
      "mean_token_accuracy": 0.16382364481687545,
      "num_tokens": 38703320.0,
      "step": 19785
    },
    {
      "entropy": 6.020958948135376,
      "epoch": 1.9782576098365572,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004612001841824447,
      "loss": 5.7432,
      "mean_token_accuracy": 0.17177781462669373,
      "num_tokens": 38713318.0,
      "step": 19790
    },
    {
      "entropy": 6.02319712638855,
      "epoch": 1.9787574348977857,
      "grad_norm": 1.25,
      "learning_rate": 0.00046118014659872085,
      "loss": 5.7553,
      "mean_token_accuracy": 0.1694430813193321,
      "num_tokens": 38723237.0,
      "step": 19795
    },
    {
      "entropy": 6.052574682235718,
      "epoch": 1.9792572599590144,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004611601043307525,
      "loss": 5.7055,
      "mean_token_accuracy": 0.17852598130702974,
      "num_tokens": 38732742.0,
      "step": 19800
    },
    {
      "entropy": 6.035595321655274,
      "epoch": 1.979757085020243,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046114005737904376,
      "loss": 5.7989,
      "mean_token_accuracy": 0.1625058263540268,
      "num_tokens": 38743259.0,
      "step": 19805
    },
    {
      "entropy": 6.0666259765625,
      "epoch": 1.9802569100814715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000461120005744099,
      "loss": 5.7699,
      "mean_token_accuracy": 0.17109783440828324,
      "num_tokens": 38753267.0,
      "step": 19810
    },
    {
      "entropy": 6.022266292572022,
      "epoch": 1.9807567351427,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046109994942642287,
      "loss": 5.6547,
      "mean_token_accuracy": 0.17694422453641892,
      "num_tokens": 38763897.0,
      "step": 19815
    },
    {
      "entropy": 6.04311237335205,
      "epoch": 1.9812565602039287,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004610798884265199,
      "loss": 5.7312,
      "mean_token_accuracy": 0.16701896488666534,
      "num_tokens": 38774864.0,
      "step": 19820
    },
    {
      "entropy": 6.0358551979064945,
      "epoch": 1.9817563852651572,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046105982274489483,
      "loss": 5.6839,
      "mean_token_accuracy": 0.1749960333108902,
      "num_tokens": 38784726.0,
      "step": 19825
    },
    {
      "entropy": 6.041944456100464,
      "epoch": 1.9822562103263857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046103975238205244,
      "loss": 5.7718,
      "mean_token_accuracy": 0.16601650267839432,
      "num_tokens": 38794649.0,
      "step": 19830
    },
    {
      "entropy": 6.121722936630249,
      "epoch": 1.9827560353876144,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004610196773384978,
      "loss": 5.7924,
      "mean_token_accuracy": 0.16713008880615235,
      "num_tokens": 38804567.0,
      "step": 19835
    },
    {
      "entropy": 6.106262254714966,
      "epoch": 1.983255860448843,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046099959761473593,
      "loss": 5.7596,
      "mean_token_accuracy": 0.17046350091695786,
      "num_tokens": 38813874.0,
      "step": 19840
    },
    {
      "entropy": 5.968999671936035,
      "epoch": 1.9837556855100713,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046097951321127207,
      "loss": 5.7499,
      "mean_token_accuracy": 0.1727908030152321,
      "num_tokens": 38822110.0,
      "step": 19845
    },
    {
      "entropy": 6.0343485355377195,
      "epoch": 1.9842555105713,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046095942412861145,
      "loss": 5.7554,
      "mean_token_accuracy": 0.17428786009550096,
      "num_tokens": 38832628.0,
      "step": 19850
    },
    {
      "entropy": 6.084467506408691,
      "epoch": 1.9847553356325287,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004609393303672596,
      "loss": 5.662,
      "mean_token_accuracy": 0.1792497992515564,
      "num_tokens": 38841742.0,
      "step": 19855
    },
    {
      "entropy": 6.095827388763428,
      "epoch": 1.9852551606937572,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046091923192772205,
      "loss": 5.7086,
      "mean_token_accuracy": 0.16906415671110153,
      "num_tokens": 38851338.0,
      "step": 19860
    },
    {
      "entropy": 5.993762350082397,
      "epoch": 1.9857549857549857,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004608991288105045,
      "loss": 5.6762,
      "mean_token_accuracy": 0.1692856565117836,
      "num_tokens": 38861308.0,
      "step": 19865
    },
    {
      "entropy": 6.011417961120605,
      "epoch": 1.9862548108162144,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046087902101611265,
      "loss": 5.7276,
      "mean_token_accuracy": 0.17411121279001235,
      "num_tokens": 38871911.0,
      "step": 19870
    },
    {
      "entropy": 5.950232219696045,
      "epoch": 1.986754635877443,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004608589085450525,
      "loss": 5.6016,
      "mean_token_accuracy": 0.17787590324878694,
      "num_tokens": 38882330.0,
      "step": 19875
    },
    {
      "entropy": 6.0173427104949955,
      "epoch": 1.9872544609386713,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046083879139783,
      "loss": 5.6636,
      "mean_token_accuracy": 0.17897972762584685,
      "num_tokens": 38892080.0,
      "step": 19880
    },
    {
      "entropy": 6.020077085494995,
      "epoch": 1.9877542859999,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004608186695749514,
      "loss": 5.669,
      "mean_token_accuracy": 0.17181855738162993,
      "num_tokens": 38901190.0,
      "step": 19885
    },
    {
      "entropy": 6.069392728805542,
      "epoch": 1.9882541110611287,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004607985430769228,
      "loss": 5.8987,
      "mean_token_accuracy": 0.162244513630867,
      "num_tokens": 38910900.0,
      "step": 19890
    },
    {
      "entropy": 5.990805721282959,
      "epoch": 1.9887539361223572,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004607784119042507,
      "loss": 5.6694,
      "mean_token_accuracy": 0.1741039350628853,
      "num_tokens": 38919988.0,
      "step": 19895
    },
    {
      "entropy": 6.010992860794067,
      "epoch": 1.9892537611835857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004607582760574416,
      "loss": 5.705,
      "mean_token_accuracy": 0.1741425484418869,
      "num_tokens": 38930288.0,
      "step": 19900
    },
    {
      "entropy": 5.979183053970337,
      "epoch": 1.9897535862448144,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000460738135537002,
      "loss": 5.541,
      "mean_token_accuracy": 0.18373922556638717,
      "num_tokens": 38938920.0,
      "step": 19905
    },
    {
      "entropy": 6.024484920501709,
      "epoch": 1.9902534113060428,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046071799034343874,
      "loss": 5.7448,
      "mean_token_accuracy": 0.17143324613571168,
      "num_tokens": 38948453.0,
      "step": 19910
    },
    {
      "entropy": 6.135925531387329,
      "epoch": 1.9907532363672713,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004606978404772587,
      "loss": 5.8633,
      "mean_token_accuracy": 0.1622862696647644,
      "num_tokens": 38958421.0,
      "step": 19915
    },
    {
      "entropy": 5.978642416000366,
      "epoch": 1.9912530614285,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046067768593896865,
      "loss": 5.6496,
      "mean_token_accuracy": 0.17501959055662156,
      "num_tokens": 38967641.0,
      "step": 19920
    },
    {
      "entropy": 6.078980445861816,
      "epoch": 1.9917528864897287,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004606575267290759,
      "loss": 5.7065,
      "mean_token_accuracy": 0.1794956460595131,
      "num_tokens": 38977601.0,
      "step": 19925
    },
    {
      "entropy": 5.950236225128174,
      "epoch": 1.9922527115509572,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004606373628480875,
      "loss": 5.7418,
      "mean_token_accuracy": 0.1728557288646698,
      "num_tokens": 38987398.0,
      "step": 19930
    },
    {
      "entropy": 6.027473449707031,
      "epoch": 1.9927525366121857,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004606171942965108,
      "loss": 5.7049,
      "mean_token_accuracy": 0.1745085224509239,
      "num_tokens": 38997831.0,
      "step": 19935
    },
    {
      "entropy": 6.09678897857666,
      "epoch": 1.9932523616734144,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004605970210748533,
      "loss": 5.8598,
      "mean_token_accuracy": 0.1658730298280716,
      "num_tokens": 39007667.0,
      "step": 19940
    },
    {
      "entropy": 5.990819597244263,
      "epoch": 1.9937521867346428,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004605768431836225,
      "loss": 5.6678,
      "mean_token_accuracy": 0.18160402923822402,
      "num_tokens": 39018192.0,
      "step": 19945
    },
    {
      "entropy": 5.987255954742432,
      "epoch": 1.9942520117958713,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046055666062332605,
      "loss": 5.6567,
      "mean_token_accuracy": 0.17882856279611586,
      "num_tokens": 39027440.0,
      "step": 19950
    },
    {
      "entropy": 5.909539270401001,
      "epoch": 1.9947518368571,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046053647339447184,
      "loss": 5.5885,
      "mean_token_accuracy": 0.18338546752929688,
      "num_tokens": 39037078.0,
      "step": 19955
    },
    {
      "entropy": 6.034127235412598,
      "epoch": 1.9952516619183287,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046051628149756767,
      "loss": 5.7461,
      "mean_token_accuracy": 0.16813134998083115,
      "num_tokens": 39046462.0,
      "step": 19960
    },
    {
      "entropy": 6.088680648803711,
      "epoch": 1.9957514869795572,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004604960849331216,
      "loss": 5.7223,
      "mean_token_accuracy": 0.1731522187590599,
      "num_tokens": 39057004.0,
      "step": 19965
    },
    {
      "entropy": 6.101143503189087,
      "epoch": 1.9962513120407857,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004604758837016417,
      "loss": 5.7955,
      "mean_token_accuracy": 0.17086956799030303,
      "num_tokens": 39066612.0,
      "step": 19970
    },
    {
      "entropy": 6.078153896331787,
      "epoch": 1.9967511371020144,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004604556778036363,
      "loss": 5.7715,
      "mean_token_accuracy": 0.16510728597640992,
      "num_tokens": 39076702.0,
      "step": 19975
    },
    {
      "entropy": 5.979477977752685,
      "epoch": 1.9972509621632428,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046043546723961384,
      "loss": 5.6031,
      "mean_token_accuracy": 0.1853533446788788,
      "num_tokens": 39087110.0,
      "step": 19980
    },
    {
      "entropy": 6.019652271270752,
      "epoch": 1.9977507872244713,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046041525201008266,
      "loss": 5.7526,
      "mean_token_accuracy": 0.16540322601795196,
      "num_tokens": 39097204.0,
      "step": 19985
    },
    {
      "entropy": 6.009469366073608,
      "epoch": 1.9982506122857,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046039503211555154,
      "loss": 5.7081,
      "mean_token_accuracy": 0.17297495454549788,
      "num_tokens": 39106469.0,
      "step": 19990
    },
    {
      "entropy": 6.040913867950439,
      "epoch": 1.9987504373469287,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046037480755652913,
      "loss": 5.7263,
      "mean_token_accuracy": 0.17220913618803024,
      "num_tokens": 39116304.0,
      "step": 19995
    },
    {
      "entropy": 6.098722219467163,
      "epoch": 1.9992502624081572,
      "grad_norm": 1.125,
      "learning_rate": 0.00046035457833352414,
      "loss": 5.8055,
      "mean_token_accuracy": 0.1611578345298767,
      "num_tokens": 39125573.0,
      "step": 20000
    },
    {
      "entropy": 6.070246410369873,
      "epoch": 1.9997500874693857,
      "grad_norm": 1.25,
      "learning_rate": 0.00046033434444704577,
      "loss": 5.7089,
      "mean_token_accuracy": 0.1736064076423645,
      "num_tokens": 39134876.0,
      "step": 20005
    },
    {
      "entropy": 5.974986924065484,
      "epoch": 2.0001999300244915,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046031410589760294,
      "loss": 5.6436,
      "mean_token_accuracy": 0.17571770151456198,
      "num_tokens": 39143851.0,
      "step": 20010
    },
    {
      "entropy": 5.881540966033936,
      "epoch": 2.00069975508572,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046029386268570485,
      "loss": 5.5222,
      "mean_token_accuracy": 0.18613556772470474,
      "num_tokens": 39152748.0,
      "step": 20015
    },
    {
      "entropy": 6.047613143920898,
      "epoch": 2.0011995801469484,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004602736148118609,
      "loss": 5.6438,
      "mean_token_accuracy": 0.17128264158964157,
      "num_tokens": 39162097.0,
      "step": 20020
    },
    {
      "entropy": 6.015650510787964,
      "epoch": 2.001699405208177,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046025336227658047,
      "loss": 5.664,
      "mean_token_accuracy": 0.18242388665676118,
      "num_tokens": 39171629.0,
      "step": 20025
    },
    {
      "entropy": 6.123390102386475,
      "epoch": 2.002199230269406,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046023310508037316,
      "loss": 5.7539,
      "mean_token_accuracy": 0.15988478511571885,
      "num_tokens": 39181368.0,
      "step": 20030
    },
    {
      "entropy": 5.996371841430664,
      "epoch": 2.002699055330634,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046021284322374847,
      "loss": 5.5736,
      "mean_token_accuracy": 0.18542833030223846,
      "num_tokens": 39190492.0,
      "step": 20035
    },
    {
      "entropy": 6.04601469039917,
      "epoch": 2.003198880391863,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004601925767072164,
      "loss": 5.6792,
      "mean_token_accuracy": 0.17055387049913406,
      "num_tokens": 39200915.0,
      "step": 20040
    },
    {
      "entropy": 6.047324085235596,
      "epoch": 2.0036987054530915,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046017230553128675,
      "loss": 5.618,
      "mean_token_accuracy": 0.1771472468972206,
      "num_tokens": 39210370.0,
      "step": 20045
    },
    {
      "entropy": 6.102301502227784,
      "epoch": 2.00419853051432,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046015202969646953,
      "loss": 5.6976,
      "mean_token_accuracy": 0.1714154452085495,
      "num_tokens": 39221731.0,
      "step": 20050
    },
    {
      "entropy": 6.119371318817139,
      "epoch": 2.0046983555755484,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004601317492032749,
      "loss": 5.6649,
      "mean_token_accuracy": 0.17392414361238479,
      "num_tokens": 39232951.0,
      "step": 20055
    },
    {
      "entropy": 6.102449798583985,
      "epoch": 2.005198180636777,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000460111464052213,
      "loss": 5.6933,
      "mean_token_accuracy": 0.17088469862937927,
      "num_tokens": 39242433.0,
      "step": 20060
    },
    {
      "entropy": 6.068913125991822,
      "epoch": 2.005698005698006,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046009117424379445,
      "loss": 5.6939,
      "mean_token_accuracy": 0.16950975060462953,
      "num_tokens": 39253496.0,
      "step": 20065
    },
    {
      "entropy": 6.000219011306763,
      "epoch": 2.006197830759234,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046007087977852947,
      "loss": 5.6466,
      "mean_token_accuracy": 0.18226450383663179,
      "num_tokens": 39263108.0,
      "step": 20070
    },
    {
      "entropy": 5.968654155731201,
      "epoch": 2.0066976558204628,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046005058065692883,
      "loss": 5.5931,
      "mean_token_accuracy": 0.18022265434265136,
      "num_tokens": 39272331.0,
      "step": 20075
    },
    {
      "entropy": 6.041233777999878,
      "epoch": 2.0071974808816915,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046003027687950316,
      "loss": 5.6567,
      "mean_token_accuracy": 0.17361815571784972,
      "num_tokens": 39282387.0,
      "step": 20080
    },
    {
      "entropy": 6.030293703079224,
      "epoch": 2.00769730594292,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004600099684467634,
      "loss": 5.641,
      "mean_token_accuracy": 0.18191157877445222,
      "num_tokens": 39293212.0,
      "step": 20085
    },
    {
      "entropy": 6.068905591964722,
      "epoch": 2.0081971310041484,
      "grad_norm": 1.125,
      "learning_rate": 0.0004599896553592204,
      "loss": 5.7355,
      "mean_token_accuracy": 0.1676284670829773,
      "num_tokens": 39302728.0,
      "step": 20090
    },
    {
      "entropy": 6.017626428604126,
      "epoch": 2.008696956065377,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004599693376173853,
      "loss": 5.599,
      "mean_token_accuracy": 0.17849691361188888,
      "num_tokens": 39313079.0,
      "step": 20095
    },
    {
      "entropy": 6.042074060440063,
      "epoch": 2.009196781126606,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004599490152217692,
      "loss": 5.7295,
      "mean_token_accuracy": 0.16312910616397858,
      "num_tokens": 39323563.0,
      "step": 20100
    },
    {
      "entropy": 5.996487045288086,
      "epoch": 2.009696606187834,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045992868817288355,
      "loss": 5.5392,
      "mean_token_accuracy": 0.18215611428022385,
      "num_tokens": 39334541.0,
      "step": 20105
    },
    {
      "entropy": 6.037035322189331,
      "epoch": 2.0101964312490628,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004599083564712396,
      "loss": 5.6505,
      "mean_token_accuracy": 0.17451459020376206,
      "num_tokens": 39344848.0,
      "step": 20110
    },
    {
      "entropy": 6.049079418182373,
      "epoch": 2.0106962563102915,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045988802011734896,
      "loss": 5.6598,
      "mean_token_accuracy": 0.16321829557418824,
      "num_tokens": 39354830.0,
      "step": 20115
    },
    {
      "entropy": 6.131053113937378,
      "epoch": 2.01119608137152,
      "grad_norm": 1.125,
      "learning_rate": 0.00045986767911172343,
      "loss": 5.7162,
      "mean_token_accuracy": 0.17056582272052764,
      "num_tokens": 39363838.0,
      "step": 20120
    },
    {
      "entropy": 6.011920928955078,
      "epoch": 2.0116959064327484,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045984733345487465,
      "loss": 5.6297,
      "mean_token_accuracy": 0.17376412749290465,
      "num_tokens": 39373549.0,
      "step": 20125
    },
    {
      "entropy": 6.042475271224975,
      "epoch": 2.012195731493977,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045982698314731447,
      "loss": 5.6197,
      "mean_token_accuracy": 0.17322793006896972,
      "num_tokens": 39383747.0,
      "step": 20130
    },
    {
      "entropy": 6.0274364948272705,
      "epoch": 2.012695556555206,
      "grad_norm": 1.125,
      "learning_rate": 0.00045980662818955495,
      "loss": 5.6741,
      "mean_token_accuracy": 0.1721989169716835,
      "num_tokens": 39394283.0,
      "step": 20135
    },
    {
      "entropy": 6.045872735977173,
      "epoch": 2.013195381616434,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045978626858210833,
      "loss": 5.544,
      "mean_token_accuracy": 0.18375907987356185,
      "num_tokens": 39404777.0,
      "step": 20140
    },
    {
      "entropy": 6.017810487747193,
      "epoch": 2.0136952066776628,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004597659043254866,
      "loss": 5.5604,
      "mean_token_accuracy": 0.17722770273685456,
      "num_tokens": 39414749.0,
      "step": 20145
    },
    {
      "entropy": 6.013143301010132,
      "epoch": 2.0141950317388915,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004597455354202023,
      "loss": 5.6295,
      "mean_token_accuracy": 0.17685010135173798,
      "num_tokens": 39424185.0,
      "step": 20150
    },
    {
      "entropy": 6.0343879699707035,
      "epoch": 2.01469485680012,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004597251618667679,
      "loss": 5.7005,
      "mean_token_accuracy": 0.17496962398290633,
      "num_tokens": 39433442.0,
      "step": 20155
    },
    {
      "entropy": 6.046120834350586,
      "epoch": 2.0151946818613484,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045970478366569594,
      "loss": 5.6524,
      "mean_token_accuracy": 0.18149299919605255,
      "num_tokens": 39443373.0,
      "step": 20160
    },
    {
      "entropy": 6.0496519088745115,
      "epoch": 2.015694506922577,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045968440081749917,
      "loss": 5.5815,
      "mean_token_accuracy": 0.18044686317443848,
      "num_tokens": 39452293.0,
      "step": 20165
    },
    {
      "entropy": 5.99799222946167,
      "epoch": 2.016194331983806,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004596640133226904,
      "loss": 5.6126,
      "mean_token_accuracy": 0.176039719581604,
      "num_tokens": 39463063.0,
      "step": 20170
    },
    {
      "entropy": 6.027917289733887,
      "epoch": 2.016694157045034,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045964362118178253,
      "loss": 5.6287,
      "mean_token_accuracy": 0.1771824464201927,
      "num_tokens": 39473349.0,
      "step": 20175
    },
    {
      "entropy": 6.045373439788818,
      "epoch": 2.0171939821062628,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004596232243952887,
      "loss": 5.6324,
      "mean_token_accuracy": 0.16936601847410201,
      "num_tokens": 39485106.0,
      "step": 20180
    },
    {
      "entropy": 6.036385631561279,
      "epoch": 2.0176938071674915,
      "grad_norm": 1.109375,
      "learning_rate": 0.000459602822963722,
      "loss": 5.7255,
      "mean_token_accuracy": 0.1705815538764,
      "num_tokens": 39494785.0,
      "step": 20185
    },
    {
      "entropy": 5.960717821121216,
      "epoch": 2.01819363222872,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045958241688759583,
      "loss": 5.5219,
      "mean_token_accuracy": 0.1844915583729744,
      "num_tokens": 39503846.0,
      "step": 20190
    },
    {
      "entropy": 6.008686256408692,
      "epoch": 2.0186934572899484,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045956200616742356,
      "loss": 5.6827,
      "mean_token_accuracy": 0.1719376564025879,
      "num_tokens": 39515207.0,
      "step": 20195
    },
    {
      "entropy": 6.042387056350708,
      "epoch": 2.019193282351177,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004595415908037187,
      "loss": 5.6286,
      "mean_token_accuracy": 0.17359090000391006,
      "num_tokens": 39524698.0,
      "step": 20200
    },
    {
      "entropy": 6.013799715042114,
      "epoch": 2.019693107412406,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004595211707969948,
      "loss": 5.56,
      "mean_token_accuracy": 0.18518953621387482,
      "num_tokens": 39534485.0,
      "step": 20205
    },
    {
      "entropy": 5.966216230392456,
      "epoch": 2.020192932473634,
      "grad_norm": 1.25,
      "learning_rate": 0.0004595007461477658,
      "loss": 5.569,
      "mean_token_accuracy": 0.19078414738178254,
      "num_tokens": 39544039.0,
      "step": 20210
    },
    {
      "entropy": 6.0406512260437015,
      "epoch": 2.0206927575348628,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045948031685654557,
      "loss": 5.6151,
      "mean_token_accuracy": 0.17938570082187652,
      "num_tokens": 39553139.0,
      "step": 20215
    },
    {
      "entropy": 6.0836262702941895,
      "epoch": 2.0211925825960915,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004594598829238479,
      "loss": 5.6807,
      "mean_token_accuracy": 0.16832185834646224,
      "num_tokens": 39562385.0,
      "step": 20220
    },
    {
      "entropy": 6.0134693622589115,
      "epoch": 2.0216924076573197,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004594394443501872,
      "loss": 5.5636,
      "mean_token_accuracy": 0.18132245987653733,
      "num_tokens": 39571855.0,
      "step": 20225
    },
    {
      "entropy": 5.98592848777771,
      "epoch": 2.0221922327185484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004594190011360775,
      "loss": 5.5726,
      "mean_token_accuracy": 0.17747498005628587,
      "num_tokens": 39581230.0,
      "step": 20230
    },
    {
      "entropy": 6.070922899246216,
      "epoch": 2.022692057779777,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004593985532820331,
      "loss": 5.6796,
      "mean_token_accuracy": 0.1657413974404335,
      "num_tokens": 39590730.0,
      "step": 20235
    },
    {
      "entropy": 6.0047204971313475,
      "epoch": 2.023191882841006,
      "grad_norm": 1.375,
      "learning_rate": 0.0004593781007885686,
      "loss": 5.6251,
      "mean_token_accuracy": 0.1835960254073143,
      "num_tokens": 39600110.0,
      "step": 20240
    },
    {
      "entropy": 6.024718713760376,
      "epoch": 2.023691707902234,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004593576436561985,
      "loss": 5.731,
      "mean_token_accuracy": 0.16379526108503342,
      "num_tokens": 39610379.0,
      "step": 20245
    },
    {
      "entropy": 6.032265281677246,
      "epoch": 2.0241915329634628,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045933718188543763,
      "loss": 5.6972,
      "mean_token_accuracy": 0.16987859457731247,
      "num_tokens": 39619857.0,
      "step": 20250
    },
    {
      "entropy": 6.027840995788575,
      "epoch": 2.0246913580246915,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004593167154768006,
      "loss": 5.6053,
      "mean_token_accuracy": 0.17985074669122697,
      "num_tokens": 39630049.0,
      "step": 20255
    },
    {
      "entropy": 6.028066253662109,
      "epoch": 2.0251911830859197,
      "grad_norm": 1.125,
      "learning_rate": 0.0004592962444308024,
      "loss": 5.6755,
      "mean_token_accuracy": 0.17270128726959227,
      "num_tokens": 39638794.0,
      "step": 20260
    },
    {
      "entropy": 5.987614488601684,
      "epoch": 2.0256910081471484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004592757687479582,
      "loss": 5.554,
      "mean_token_accuracy": 0.18088043183088304,
      "num_tokens": 39648261.0,
      "step": 20265
    },
    {
      "entropy": 5.9981049537658695,
      "epoch": 2.026190833208377,
      "grad_norm": 1.0625,
      "learning_rate": 0.000459255288428783,
      "loss": 5.5131,
      "mean_token_accuracy": 0.18661936223506928,
      "num_tokens": 39659217.0,
      "step": 20270
    },
    {
      "entropy": 5.84541597366333,
      "epoch": 2.026690658269606,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004592348034737922,
      "loss": 5.4005,
      "mean_token_accuracy": 0.20068044513463973,
      "num_tokens": 39669535.0,
      "step": 20275
    },
    {
      "entropy": 6.0245361328125,
      "epoch": 2.027190483330834,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004592143138835012,
      "loss": 5.6719,
      "mean_token_accuracy": 0.17002248764038086,
      "num_tokens": 39678640.0,
      "step": 20280
    },
    {
      "entropy": 5.985559463500977,
      "epoch": 2.0276903083920628,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004591938196584254,
      "loss": 5.6272,
      "mean_token_accuracy": 0.18103720247745514,
      "num_tokens": 39688000.0,
      "step": 20285
    },
    {
      "entropy": 6.010308694839478,
      "epoch": 2.0281901334532915,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045917332079908055,
      "loss": 5.6244,
      "mean_token_accuracy": 0.17726407051086426,
      "num_tokens": 39697778.0,
      "step": 20290
    },
    {
      "entropy": 5.925933122634888,
      "epoch": 2.0286899585145197,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004591528173059823,
      "loss": 5.5344,
      "mean_token_accuracy": 0.17996697425842284,
      "num_tokens": 39707842.0,
      "step": 20295
    },
    {
      "entropy": 6.038950872421265,
      "epoch": 2.0291897835757484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004591323091796465,
      "loss": 5.6736,
      "mean_token_accuracy": 0.16954731643199922,
      "num_tokens": 39716465.0,
      "step": 20300
    },
    {
      "entropy": 6.004002666473388,
      "epoch": 2.029689608636977,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004591117964205893,
      "loss": 5.5609,
      "mean_token_accuracy": 0.1808478996157646,
      "num_tokens": 39725862.0,
      "step": 20305
    },
    {
      "entropy": 5.929534196853638,
      "epoch": 2.030189433698206,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045909127902932657,
      "loss": 5.654,
      "mean_token_accuracy": 0.1676175057888031,
      "num_tokens": 39737042.0,
      "step": 20310
    },
    {
      "entropy": 6.047214221954346,
      "epoch": 2.030689258759434,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004590707570063747,
      "loss": 5.6303,
      "mean_token_accuracy": 0.1721533179283142,
      "num_tokens": 39747849.0,
      "step": 20315
    },
    {
      "entropy": 6.0375611782073975,
      "epoch": 2.0311890838206628,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045905023035224993,
      "loss": 5.5923,
      "mean_token_accuracy": 0.17275273352861403,
      "num_tokens": 39757544.0,
      "step": 20320
    },
    {
      "entropy": 5.968128204345703,
      "epoch": 2.0316889088818915,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045902969906746875,
      "loss": 5.5689,
      "mean_token_accuracy": 0.18160321265459062,
      "num_tokens": 39766127.0,
      "step": 20325
    },
    {
      "entropy": 6.029345560073852,
      "epoch": 2.0321887339431197,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045900916315254765,
      "loss": 5.718,
      "mean_token_accuracy": 0.17203131020069123,
      "num_tokens": 39775711.0,
      "step": 20330
    },
    {
      "entropy": 6.056663656234742,
      "epoch": 2.0326885590043484,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045898862260800337,
      "loss": 5.7277,
      "mean_token_accuracy": 0.17564996778964997,
      "num_tokens": 39786132.0,
      "step": 20335
    },
    {
      "entropy": 6.113587236404419,
      "epoch": 2.033188384065577,
      "grad_norm": 1.25,
      "learning_rate": 0.00045896807743435263,
      "loss": 5.7084,
      "mean_token_accuracy": 0.16110291630029677,
      "num_tokens": 39796196.0,
      "step": 20340
    },
    {
      "entropy": 6.047494125366211,
      "epoch": 2.033688209126806,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004589475276321125,
      "loss": 5.5784,
      "mean_token_accuracy": 0.1777245208621025,
      "num_tokens": 39805912.0,
      "step": 20345
    },
    {
      "entropy": 5.970480012893677,
      "epoch": 2.034188034188034,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045892697320179987,
      "loss": 5.5693,
      "mean_token_accuracy": 0.18729718625545502,
      "num_tokens": 39814990.0,
      "step": 20350
    },
    {
      "entropy": 5.926724290847778,
      "epoch": 2.0346878592492628,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004589064141439319,
      "loss": 5.6537,
      "mean_token_accuracy": 0.1782863110303879,
      "num_tokens": 39823877.0,
      "step": 20355
    },
    {
      "entropy": 5.899235630035401,
      "epoch": 2.0351876843104915,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004588858504590259,
      "loss": 5.4985,
      "mean_token_accuracy": 0.18788677752017974,
      "num_tokens": 39832784.0,
      "step": 20360
    },
    {
      "entropy": 6.069557476043701,
      "epoch": 2.0356875093717197,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004588652821475992,
      "loss": 5.6269,
      "mean_token_accuracy": 0.1732041209936142,
      "num_tokens": 39842388.0,
      "step": 20365
    },
    {
      "entropy": 6.002089023590088,
      "epoch": 2.0361873344329484,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045884470921016935,
      "loss": 5.5891,
      "mean_token_accuracy": 0.18284299969673157,
      "num_tokens": 39851657.0,
      "step": 20370
    },
    {
      "entropy": 6.006623125076294,
      "epoch": 2.036687159494177,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045882413164725395,
      "loss": 5.6604,
      "mean_token_accuracy": 0.17797710746526718,
      "num_tokens": 39861941.0,
      "step": 20375
    },
    {
      "entropy": 5.974663925170899,
      "epoch": 2.037186984555406,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045880354945937056,
      "loss": 5.5869,
      "mean_token_accuracy": 0.18550484627485275,
      "num_tokens": 39872000.0,
      "step": 20380
    },
    {
      "entropy": 5.996733522415161,
      "epoch": 2.037686809616634,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004587829626470373,
      "loss": 5.6292,
      "mean_token_accuracy": 0.17913335859775542,
      "num_tokens": 39882281.0,
      "step": 20385
    },
    {
      "entropy": 6.013575553894043,
      "epoch": 2.0381866346778628,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045876237121077196,
      "loss": 5.6409,
      "mean_token_accuracy": 0.17723834812641143,
      "num_tokens": 39892445.0,
      "step": 20390
    },
    {
      "entropy": 6.016788482666016,
      "epoch": 2.0386864597390915,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004587417751510926,
      "loss": 5.5862,
      "mean_token_accuracy": 0.18109215945005416,
      "num_tokens": 39902077.0,
      "step": 20395
    },
    {
      "entropy": 6.052391624450683,
      "epoch": 2.0391862848003197,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045872117446851753,
      "loss": 5.6647,
      "mean_token_accuracy": 0.17347491532564163,
      "num_tokens": 39912930.0,
      "step": 20400
    },
    {
      "entropy": 6.070194053649902,
      "epoch": 2.0396861098615484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004587005691635649,
      "loss": 5.6144,
      "mean_token_accuracy": 0.17673270255327225,
      "num_tokens": 39921884.0,
      "step": 20405
    },
    {
      "entropy": 6.06322751045227,
      "epoch": 2.040185934922777,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045867995923675326,
      "loss": 5.6894,
      "mean_token_accuracy": 0.17127406299114228,
      "num_tokens": 39931589.0,
      "step": 20410
    },
    {
      "entropy": 6.080901050567627,
      "epoch": 2.040685759984006,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004586593446886011,
      "loss": 5.7121,
      "mean_token_accuracy": 0.17057961374521255,
      "num_tokens": 39941624.0,
      "step": 20415
    },
    {
      "entropy": 6.03312726020813,
      "epoch": 2.041185585045234,
      "grad_norm": 1.125,
      "learning_rate": 0.0004586387255196271,
      "loss": 5.5896,
      "mean_token_accuracy": 0.18155712634325027,
      "num_tokens": 39951541.0,
      "step": 20420
    },
    {
      "entropy": 6.046192741394043,
      "epoch": 2.0416854101064628,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045861810173035,
      "loss": 5.6646,
      "mean_token_accuracy": 0.17373231202363967,
      "num_tokens": 39962729.0,
      "step": 20425
    },
    {
      "entropy": 6.027585554122925,
      "epoch": 2.0421852351676915,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045859747332128883,
      "loss": 5.6762,
      "mean_token_accuracy": 0.1783364936709404,
      "num_tokens": 39972193.0,
      "step": 20430
    },
    {
      "entropy": 6.0102760791778564,
      "epoch": 2.0426850602289197,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004585768402929623,
      "loss": 5.648,
      "mean_token_accuracy": 0.167934413254261,
      "num_tokens": 39980813.0,
      "step": 20435
    },
    {
      "entropy": 6.039582443237305,
      "epoch": 2.0431848852901484,
      "grad_norm": 1.125,
      "learning_rate": 0.0004585562026458898,
      "loss": 5.6175,
      "mean_token_accuracy": 0.1786620020866394,
      "num_tokens": 39989147.0,
      "step": 20440
    },
    {
      "entropy": 5.970725250244141,
      "epoch": 2.043684710351377,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045853556038059044,
      "loss": 5.6164,
      "mean_token_accuracy": 0.1835292547941208,
      "num_tokens": 40000091.0,
      "step": 20445
    },
    {
      "entropy": 6.112064218521118,
      "epoch": 2.0441845354126054,
      "grad_norm": 1.609375,
      "learning_rate": 0.00045851491349758366,
      "loss": 5.7117,
      "mean_token_accuracy": 0.176460100710392,
      "num_tokens": 40009443.0,
      "step": 20450
    },
    {
      "entropy": 6.027121353149414,
      "epoch": 2.044684360473834,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045849426199738884,
      "loss": 5.6489,
      "mean_token_accuracy": 0.17534223198890686,
      "num_tokens": 40020877.0,
      "step": 20455
    },
    {
      "entropy": 6.059490728378296,
      "epoch": 2.0451841855350628,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004584736058805256,
      "loss": 5.7179,
      "mean_token_accuracy": 0.17425222247838973,
      "num_tokens": 40030952.0,
      "step": 20460
    },
    {
      "entropy": 6.047718906402588,
      "epoch": 2.0456840105962915,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004584529451475137,
      "loss": 5.6116,
      "mean_token_accuracy": 0.17377109974622726,
      "num_tokens": 40039995.0,
      "step": 20465
    },
    {
      "entropy": 5.995414304733276,
      "epoch": 2.0461838356575197,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045843227979887286,
      "loss": 5.5482,
      "mean_token_accuracy": 0.1853907212615013,
      "num_tokens": 40049346.0,
      "step": 20470
    },
    {
      "entropy": 5.972534656524658,
      "epoch": 2.0466836607187484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045841160983512303,
      "loss": 5.6561,
      "mean_token_accuracy": 0.17808615267276764,
      "num_tokens": 40058371.0,
      "step": 20475
    },
    {
      "entropy": 6.006093692779541,
      "epoch": 2.047183485779977,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004583909352567843,
      "loss": 5.5736,
      "mean_token_accuracy": 0.18292628824710847,
      "num_tokens": 40067504.0,
      "step": 20480
    },
    {
      "entropy": 6.008750152587891,
      "epoch": 2.0476833108412054,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004583702560643769,
      "loss": 5.6252,
      "mean_token_accuracy": 0.18585519194602967,
      "num_tokens": 40077551.0,
      "step": 20485
    },
    {
      "entropy": 5.987696266174316,
      "epoch": 2.048183135902434,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000458349572258421,
      "loss": 5.6165,
      "mean_token_accuracy": 0.17000813335180281,
      "num_tokens": 40087896.0,
      "step": 20490
    },
    {
      "entropy": 5.951995420455932,
      "epoch": 2.0486829609636628,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045832888383943706,
      "loss": 5.6061,
      "mean_token_accuracy": 0.17366162687540054,
      "num_tokens": 40097197.0,
      "step": 20495
    },
    {
      "entropy": 6.100425148010254,
      "epoch": 2.0491827860248915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004583081908079455,
      "loss": 5.6918,
      "mean_token_accuracy": 0.17162279933691024,
      "num_tokens": 40108068.0,
      "step": 20500
    },
    {
      "entropy": 5.9735640525817875,
      "epoch": 2.0496826110861197,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004582874931644671,
      "loss": 5.6193,
      "mean_token_accuracy": 0.17349585592746736,
      "num_tokens": 40116905.0,
      "step": 20505
    },
    {
      "entropy": 5.961503028869629,
      "epoch": 2.0501824361473484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045826679090952253,
      "loss": 5.6113,
      "mean_token_accuracy": 0.1743573307991028,
      "num_tokens": 40126611.0,
      "step": 20510
    },
    {
      "entropy": 5.999034738540649,
      "epoch": 2.050682261208577,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045824608404363267,
      "loss": 5.509,
      "mean_token_accuracy": 0.18213323801755904,
      "num_tokens": 40136557.0,
      "step": 20515
    },
    {
      "entropy": 6.057348966598511,
      "epoch": 2.0511820862698054,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004582253725673184,
      "loss": 5.6294,
      "mean_token_accuracy": 0.17268659472465514,
      "num_tokens": 40146637.0,
      "step": 20520
    },
    {
      "entropy": 6.022529602050781,
      "epoch": 2.051681911331034,
      "grad_norm": 1.171875,
      "learning_rate": 0.000458204656481101,
      "loss": 5.5775,
      "mean_token_accuracy": 0.18299775421619416,
      "num_tokens": 40156763.0,
      "step": 20525
    },
    {
      "entropy": 5.964795780181885,
      "epoch": 2.0521817363922628,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004581839357855015,
      "loss": 5.6296,
      "mean_token_accuracy": 0.17716884464025498,
      "num_tokens": 40167024.0,
      "step": 20530
    },
    {
      "entropy": 6.037343502044678,
      "epoch": 2.0526815614534915,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045816321048104143,
      "loss": 5.5921,
      "mean_token_accuracy": 0.178517147898674,
      "num_tokens": 40176977.0,
      "step": 20535
    },
    {
      "entropy": 5.983216905593872,
      "epoch": 2.0531813865147197,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004581424805682421,
      "loss": 5.5345,
      "mean_token_accuracy": 0.17673323154449463,
      "num_tokens": 40187820.0,
      "step": 20540
    },
    {
      "entropy": 6.024430131912231,
      "epoch": 2.0536812115759484,
      "grad_norm": 1.328125,
      "learning_rate": 0.000458121746047625,
      "loss": 5.6119,
      "mean_token_accuracy": 0.17417713403701782,
      "num_tokens": 40197518.0,
      "step": 20545
    },
    {
      "entropy": 6.032049179077148,
      "epoch": 2.054181036637177,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004581010069197119,
      "loss": 5.7044,
      "mean_token_accuracy": 0.1724739484488964,
      "num_tokens": 40208360.0,
      "step": 20550
    },
    {
      "entropy": 6.063866233825683,
      "epoch": 2.0546808616984054,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045808026318502464,
      "loss": 5.7079,
      "mean_token_accuracy": 0.17834578156471254,
      "num_tokens": 40219526.0,
      "step": 20555
    },
    {
      "entropy": 6.069508123397827,
      "epoch": 2.055180686759634,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004580595148440851,
      "loss": 5.649,
      "mean_token_accuracy": 0.17105533182621002,
      "num_tokens": 40230661.0,
      "step": 20560
    },
    {
      "entropy": 6.035384464263916,
      "epoch": 2.0556805118208628,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004580387618974152,
      "loss": 5.7266,
      "mean_token_accuracy": 0.1757706359028816,
      "num_tokens": 40241675.0,
      "step": 20565
    },
    {
      "entropy": 6.017474603652954,
      "epoch": 2.0561803368820915,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045801800434553715,
      "loss": 5.5621,
      "mean_token_accuracy": 0.17550031691789628,
      "num_tokens": 40251275.0,
      "step": 20570
    },
    {
      "entropy": 6.057634496688843,
      "epoch": 2.0566801619433197,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004579972421889732,
      "loss": 5.651,
      "mean_token_accuracy": 0.17831296771764754,
      "num_tokens": 40260975.0,
      "step": 20575
    },
    {
      "entropy": 5.985488414764404,
      "epoch": 2.0571799870045484,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004579764754282458,
      "loss": 5.6149,
      "mean_token_accuracy": 0.17391274869441986,
      "num_tokens": 40271617.0,
      "step": 20580
    },
    {
      "entropy": 5.960715389251709,
      "epoch": 2.057679812065777,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045795570406387726,
      "loss": 5.5569,
      "mean_token_accuracy": 0.1783503398299217,
      "num_tokens": 40281365.0,
      "step": 20585
    },
    {
      "entropy": 6.0138078212738035,
      "epoch": 2.0581796371270054,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004579349280963904,
      "loss": 5.6248,
      "mean_token_accuracy": 0.17264046818017958,
      "num_tokens": 40291050.0,
      "step": 20590
    },
    {
      "entropy": 5.980755472183228,
      "epoch": 2.058679462188234,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045791414752630765,
      "loss": 5.498,
      "mean_token_accuracy": 0.18274455815553664,
      "num_tokens": 40299911.0,
      "step": 20595
    },
    {
      "entropy": 6.014840459823608,
      "epoch": 2.0591792872494628,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004578933623541522,
      "loss": 5.5991,
      "mean_token_accuracy": 0.17610823959112168,
      "num_tokens": 40310860.0,
      "step": 20600
    },
    {
      "entropy": 6.044261693954468,
      "epoch": 2.0596791123106915,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045787257258044666,
      "loss": 5.673,
      "mean_token_accuracy": 0.17811238914728164,
      "num_tokens": 40320045.0,
      "step": 20605
    },
    {
      "entropy": 5.924721622467041,
      "epoch": 2.0601789373719197,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045785177820571434,
      "loss": 5.6519,
      "mean_token_accuracy": 0.17377233356237412,
      "num_tokens": 40329082.0,
      "step": 20610
    },
    {
      "entropy": 5.876181364059448,
      "epoch": 2.0606787624331484,
      "grad_norm": 1.25,
      "learning_rate": 0.00045783097923047833,
      "loss": 5.5671,
      "mean_token_accuracy": 0.18217574954032897,
      "num_tokens": 40338161.0,
      "step": 20615
    },
    {
      "entropy": 6.044011545181275,
      "epoch": 2.061178587494377,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045781017565526194,
      "loss": 5.6978,
      "mean_token_accuracy": 0.17152872830629348,
      "num_tokens": 40348316.0,
      "step": 20620
    },
    {
      "entropy": 6.075443410873413,
      "epoch": 2.0616784125556054,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00045778936748058855,
      "loss": 5.6294,
      "mean_token_accuracy": 0.17674710750579833,
      "num_tokens": 40358112.0,
      "step": 20625
    },
    {
      "entropy": 5.991676092147827,
      "epoch": 2.062178237616834,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045776855470698163,
      "loss": 5.6019,
      "mean_token_accuracy": 0.1871199682354927,
      "num_tokens": 40368670.0,
      "step": 20630
    },
    {
      "entropy": 5.927839946746826,
      "epoch": 2.0626780626780628,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045774773733496505,
      "loss": 5.5767,
      "mean_token_accuracy": 0.18058069944381713,
      "num_tokens": 40378082.0,
      "step": 20635
    },
    {
      "entropy": 6.035091209411621,
      "epoch": 2.0631778877392914,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004577269153650624,
      "loss": 5.6603,
      "mean_token_accuracy": 0.17204196006059647,
      "num_tokens": 40387679.0,
      "step": 20640
    },
    {
      "entropy": 6.044397306442261,
      "epoch": 2.0636777128005197,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004577060887977975,
      "loss": 5.6021,
      "mean_token_accuracy": 0.17486146986484527,
      "num_tokens": 40397599.0,
      "step": 20645
    },
    {
      "entropy": 6.052708578109741,
      "epoch": 2.0641775378617484,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045768525763369443,
      "loss": 5.6139,
      "mean_token_accuracy": 0.17453809529542924,
      "num_tokens": 40407163.0,
      "step": 20650
    },
    {
      "entropy": 5.9275538444519045,
      "epoch": 2.064677362922977,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045766442187327735,
      "loss": 5.596,
      "mean_token_accuracy": 0.17580835670232772,
      "num_tokens": 40416489.0,
      "step": 20655
    },
    {
      "entropy": 6.007649946212768,
      "epoch": 2.0651771879842054,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045764358151707043,
      "loss": 5.6754,
      "mean_token_accuracy": 0.1731174036860466,
      "num_tokens": 40425193.0,
      "step": 20660
    },
    {
      "entropy": 6.052390289306641,
      "epoch": 2.065677013045434,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004576227365655979,
      "loss": 5.7104,
      "mean_token_accuracy": 0.17952973693609237,
      "num_tokens": 40434025.0,
      "step": 20665
    },
    {
      "entropy": 6.011010217666626,
      "epoch": 2.0661768381066627,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004576018870193844,
      "loss": 5.694,
      "mean_token_accuracy": 0.17954467236995697,
      "num_tokens": 40443010.0,
      "step": 20670
    },
    {
      "entropy": 5.991280937194825,
      "epoch": 2.0666766631678914,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004575810328789543,
      "loss": 5.6496,
      "mean_token_accuracy": 0.1753791630268097,
      "num_tokens": 40452673.0,
      "step": 20675
    },
    {
      "entropy": 5.957796478271485,
      "epoch": 2.0671764882291197,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004575601741448325,
      "loss": 5.6462,
      "mean_token_accuracy": 0.1758592024445534,
      "num_tokens": 40461747.0,
      "step": 20680
    },
    {
      "entropy": 6.034953212738037,
      "epoch": 2.0676763132903484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045753931081754366,
      "loss": 5.5941,
      "mean_token_accuracy": 0.1758040815591812,
      "num_tokens": 40470674.0,
      "step": 20685
    },
    {
      "entropy": 5.980107069015503,
      "epoch": 2.068176138351577,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004575184428976127,
      "loss": 5.5908,
      "mean_token_accuracy": 0.17901499718427658,
      "num_tokens": 40480287.0,
      "step": 20690
    },
    {
      "entropy": 5.9709772109985355,
      "epoch": 2.0686759634128054,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004574975703855647,
      "loss": 5.5096,
      "mean_token_accuracy": 0.19111085534095765,
      "num_tokens": 40489990.0,
      "step": 20695
    },
    {
      "entropy": 6.024137020111084,
      "epoch": 2.069175788474034,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045747669328192477,
      "loss": 5.654,
      "mean_token_accuracy": 0.18113009855151177,
      "num_tokens": 40500999.0,
      "step": 20700
    },
    {
      "entropy": 6.0332207679748535,
      "epoch": 2.0696756135352627,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004574558115872183,
      "loss": 5.6395,
      "mean_token_accuracy": 0.1764688104391098,
      "num_tokens": 40509607.0,
      "step": 20705
    },
    {
      "entropy": 5.912532567977905,
      "epoch": 2.0701754385964914,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045743492530197046,
      "loss": 5.5279,
      "mean_token_accuracy": 0.1809765577316284,
      "num_tokens": 40518312.0,
      "step": 20710
    },
    {
      "entropy": 5.996905660629272,
      "epoch": 2.0706752636577197,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004574140344267069,
      "loss": 5.6486,
      "mean_token_accuracy": 0.17445210963487626,
      "num_tokens": 40528015.0,
      "step": 20715
    },
    {
      "entropy": 6.020040273666382,
      "epoch": 2.0711750887189484,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004573931389619531,
      "loss": 5.6599,
      "mean_token_accuracy": 0.17036096304655074,
      "num_tokens": 40537557.0,
      "step": 20720
    },
    {
      "entropy": 6.040363550186157,
      "epoch": 2.071674913780177,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045737223890823494,
      "loss": 5.6037,
      "mean_token_accuracy": 0.17245301604270935,
      "num_tokens": 40547360.0,
      "step": 20725
    },
    {
      "entropy": 6.121060132980347,
      "epoch": 2.0721747388414054,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045735133426607813,
      "loss": 5.769,
      "mean_token_accuracy": 0.16877160519361495,
      "num_tokens": 40558156.0,
      "step": 20730
    },
    {
      "entropy": 6.060563468933106,
      "epoch": 2.072674563902634,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045733042503600865,
      "loss": 5.6603,
      "mean_token_accuracy": 0.1684238389134407,
      "num_tokens": 40568732.0,
      "step": 20735
    },
    {
      "entropy": 5.949757051467896,
      "epoch": 2.0731743889638627,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045730951121855255,
      "loss": 5.5299,
      "mean_token_accuracy": 0.18401274532079698,
      "num_tokens": 40579093.0,
      "step": 20740
    },
    {
      "entropy": 6.150512742996216,
      "epoch": 2.0736742140250914,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004572885928142362,
      "loss": 5.7597,
      "mean_token_accuracy": 0.16455851644277572,
      "num_tokens": 40589327.0,
      "step": 20745
    },
    {
      "entropy": 6.010919427871704,
      "epoch": 2.0741740390863197,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004572676698235857,
      "loss": 5.6037,
      "mean_token_accuracy": 0.17275384217500686,
      "num_tokens": 40600054.0,
      "step": 20750
    },
    {
      "entropy": 6.061483716964721,
      "epoch": 2.0746738641475484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004572467422471275,
      "loss": 5.648,
      "mean_token_accuracy": 0.18214573860168456,
      "num_tokens": 40609286.0,
      "step": 20755
    },
    {
      "entropy": 6.053404331207275,
      "epoch": 2.075173689208777,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045722581008538816,
      "loss": 5.7571,
      "mean_token_accuracy": 0.1749042361974716,
      "num_tokens": 40618150.0,
      "step": 20760
    },
    {
      "entropy": 5.95385046005249,
      "epoch": 2.0756735142700053,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004572048733388944,
      "loss": 5.5044,
      "mean_token_accuracy": 0.19453241676092148,
      "num_tokens": 40627842.0,
      "step": 20765
    },
    {
      "entropy": 6.049459981918335,
      "epoch": 2.076173339331234,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045718393200817284,
      "loss": 5.5927,
      "mean_token_accuracy": 0.18024535924196244,
      "num_tokens": 40637767.0,
      "step": 20770
    },
    {
      "entropy": 6.004562616348267,
      "epoch": 2.0766731643924627,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045716298609375044,
      "loss": 5.6955,
      "mean_token_accuracy": 0.1696650892496109,
      "num_tokens": 40648611.0,
      "step": 20775
    },
    {
      "entropy": 6.051205348968506,
      "epoch": 2.077172989453691,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045714203559615423,
      "loss": 5.6673,
      "mean_token_accuracy": 0.17758984565734864,
      "num_tokens": 40659664.0,
      "step": 20780
    },
    {
      "entropy": 6.044132041931152,
      "epoch": 2.0776728145149197,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004571210805159112,
      "loss": 5.579,
      "mean_token_accuracy": 0.18015628159046174,
      "num_tokens": 40669145.0,
      "step": 20785
    },
    {
      "entropy": 6.043071222305298,
      "epoch": 2.0781726395761484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004571001208535487,
      "loss": 5.5912,
      "mean_token_accuracy": 0.17696579694747924,
      "num_tokens": 40678250.0,
      "step": 20790
    },
    {
      "entropy": 6.014349842071534,
      "epoch": 2.078672464637377,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000457079156609594,
      "loss": 5.5969,
      "mean_token_accuracy": 0.17782391458749772,
      "num_tokens": 40687590.0,
      "step": 20795
    },
    {
      "entropy": 5.859641027450562,
      "epoch": 2.0791722896986053,
      "grad_norm": 1.25,
      "learning_rate": 0.0004570581877845746,
      "loss": 5.5385,
      "mean_token_accuracy": 0.18244275599718093,
      "num_tokens": 40697452.0,
      "step": 20800
    },
    {
      "entropy": 6.037401151657105,
      "epoch": 2.079672114759834,
      "grad_norm": 1.1875,
      "learning_rate": 0.000457037214379018,
      "loss": 5.6984,
      "mean_token_accuracy": 0.17691597044467927,
      "num_tokens": 40707608.0,
      "step": 20805
    },
    {
      "entropy": 6.103165483474731,
      "epoch": 2.0801719398210627,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045701623639345204,
      "loss": 5.5906,
      "mean_token_accuracy": 0.18032891154289246,
      "num_tokens": 40717369.0,
      "step": 20810
    },
    {
      "entropy": 5.98867712020874,
      "epoch": 2.080671764882291,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045699525382840423,
      "loss": 5.5441,
      "mean_token_accuracy": 0.18621441572904587,
      "num_tokens": 40725899.0,
      "step": 20815
    },
    {
      "entropy": 6.001989889144897,
      "epoch": 2.0811715899435197,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045697426668440277,
      "loss": 5.7007,
      "mean_token_accuracy": 0.16793270856142045,
      "num_tokens": 40735252.0,
      "step": 20820
    },
    {
      "entropy": 6.067838525772094,
      "epoch": 2.0816714150047484,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045695327496197557,
      "loss": 5.7104,
      "mean_token_accuracy": 0.17102858871221543,
      "num_tokens": 40745828.0,
      "step": 20825
    },
    {
      "entropy": 6.058746194839477,
      "epoch": 2.082171240065977,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004569322786616508,
      "loss": 5.6295,
      "mean_token_accuracy": 0.17295683920383453,
      "num_tokens": 40756388.0,
      "step": 20830
    },
    {
      "entropy": 5.977232360839844,
      "epoch": 2.0826710651272053,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004569112777839567,
      "loss": 5.5146,
      "mean_token_accuracy": 0.17851062268018722,
      "num_tokens": 40765914.0,
      "step": 20835
    },
    {
      "entropy": 5.980428075790405,
      "epoch": 2.083170890188434,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045689027232942165,
      "loss": 5.5503,
      "mean_token_accuracy": 0.1799285888671875,
      "num_tokens": 40775015.0,
      "step": 20840
    },
    {
      "entropy": 5.8982970237731935,
      "epoch": 2.0836707152496627,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004568692622985742,
      "loss": 5.5363,
      "mean_token_accuracy": 0.1818180650472641,
      "num_tokens": 40783743.0,
      "step": 20845
    },
    {
      "entropy": 5.9784235000610355,
      "epoch": 2.084170540310891,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045684824769194284,
      "loss": 5.6771,
      "mean_token_accuracy": 0.18209614157676696,
      "num_tokens": 40793973.0,
      "step": 20850
    },
    {
      "entropy": 6.052599859237671,
      "epoch": 2.0846703653721197,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004568272285100564,
      "loss": 5.619,
      "mean_token_accuracy": 0.18239326179027557,
      "num_tokens": 40803800.0,
      "step": 20855
    },
    {
      "entropy": 6.083937358856201,
      "epoch": 2.0851701904333484,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004568062047534437,
      "loss": 5.6706,
      "mean_token_accuracy": 0.17524036765098572,
      "num_tokens": 40812791.0,
      "step": 20860
    },
    {
      "entropy": 6.003423261642456,
      "epoch": 2.085670015494577,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004567851764226337,
      "loss": 5.6612,
      "mean_token_accuracy": 0.1792384058237076,
      "num_tokens": 40822498.0,
      "step": 20865
    },
    {
      "entropy": 5.981271457672119,
      "epoch": 2.0861698405558053,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045676414351815533,
      "loss": 5.5351,
      "mean_token_accuracy": 0.17806432098150254,
      "num_tokens": 40831965.0,
      "step": 20870
    },
    {
      "entropy": 6.012896156311035,
      "epoch": 2.086669665617034,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004567431060405379,
      "loss": 5.6224,
      "mean_token_accuracy": 0.17349086999893187,
      "num_tokens": 40840521.0,
      "step": 20875
    },
    {
      "entropy": 5.9899329662323,
      "epoch": 2.0871694906782627,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004567220639903108,
      "loss": 5.7179,
      "mean_token_accuracy": 0.17707693129777907,
      "num_tokens": 40850935.0,
      "step": 20880
    },
    {
      "entropy": 5.982895421981811,
      "epoch": 2.087669315739491,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045670101736800325,
      "loss": 5.6224,
      "mean_token_accuracy": 0.17913060039281845,
      "num_tokens": 40860662.0,
      "step": 20885
    },
    {
      "entropy": 5.883236598968506,
      "epoch": 2.0881691408007197,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045667996617414485,
      "loss": 5.409,
      "mean_token_accuracy": 0.19338874369859696,
      "num_tokens": 40870818.0,
      "step": 20890
    },
    {
      "entropy": 6.058947324752808,
      "epoch": 2.0886689658619484,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004566589104092653,
      "loss": 5.7605,
      "mean_token_accuracy": 0.16706625372171402,
      "num_tokens": 40880842.0,
      "step": 20895
    },
    {
      "entropy": 6.086929035186768,
      "epoch": 2.089168790923177,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004566378500738943,
      "loss": 5.7125,
      "mean_token_accuracy": 0.1739475756883621,
      "num_tokens": 40892070.0,
      "step": 20900
    },
    {
      "entropy": 5.9633701801300045,
      "epoch": 2.0896686159844053,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045661678516856164,
      "loss": 5.4742,
      "mean_token_accuracy": 0.1846854493021965,
      "num_tokens": 40901588.0,
      "step": 20905
    },
    {
      "entropy": 6.064084482192993,
      "epoch": 2.090168441045634,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004565957156937975,
      "loss": 5.6527,
      "mean_token_accuracy": 0.17848791033029557,
      "num_tokens": 40909961.0,
      "step": 20910
    },
    {
      "entropy": 6.030108404159546,
      "epoch": 2.0906682661068627,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004565746416501318,
      "loss": 5.6257,
      "mean_token_accuracy": 0.17793516516685487,
      "num_tokens": 40919528.0,
      "step": 20915
    },
    {
      "entropy": 5.992210626602173,
      "epoch": 2.091168091168091,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045655356303809494,
      "loss": 5.6316,
      "mean_token_accuracy": 0.1749677538871765,
      "num_tokens": 40929975.0,
      "step": 20920
    },
    {
      "entropy": 5.9766021251678465,
      "epoch": 2.0916679162293197,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045653247985821707,
      "loss": 5.562,
      "mean_token_accuracy": 0.17822787761688233,
      "num_tokens": 40939952.0,
      "step": 20925
    },
    {
      "entropy": 5.9854583740234375,
      "epoch": 2.0921677412905484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045651139211102875,
      "loss": 5.5336,
      "mean_token_accuracy": 0.18660901188850404,
      "num_tokens": 40948935.0,
      "step": 20930
    },
    {
      "entropy": 5.943440961837768,
      "epoch": 2.092667566351777,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045649029979706054,
      "loss": 5.5752,
      "mean_token_accuracy": 0.1826597809791565,
      "num_tokens": 40958169.0,
      "step": 20935
    },
    {
      "entropy": 5.987160730361938,
      "epoch": 2.0931673914130053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045646920291684303,
      "loss": 5.6025,
      "mean_token_accuracy": 0.18105494230985641,
      "num_tokens": 40967920.0,
      "step": 20940
    },
    {
      "entropy": 5.998324680328369,
      "epoch": 2.093667216474234,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004564481014709072,
      "loss": 5.5967,
      "mean_token_accuracy": 0.18069231361150742,
      "num_tokens": 40977494.0,
      "step": 20945
    },
    {
      "entropy": 6.025364971160888,
      "epoch": 2.0941670415354627,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004564269954597836,
      "loss": 5.7003,
      "mean_token_accuracy": 0.17019741237163544,
      "num_tokens": 40987838.0,
      "step": 20950
    },
    {
      "entropy": 5.949441242218017,
      "epoch": 2.094666866596691,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045640588488400364,
      "loss": 5.4989,
      "mean_token_accuracy": 0.1813320428133011,
      "num_tokens": 40997308.0,
      "step": 20955
    },
    {
      "entropy": 5.945567035675049,
      "epoch": 2.0951666916579197,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004563847697440983,
      "loss": 5.6274,
      "mean_token_accuracy": 0.17575997859239578,
      "num_tokens": 41008159.0,
      "step": 20960
    },
    {
      "entropy": 6.060703134536743,
      "epoch": 2.0956665167191484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045636365004059876,
      "loss": 5.6522,
      "mean_token_accuracy": 0.17111900448799133,
      "num_tokens": 41018224.0,
      "step": 20965
    },
    {
      "entropy": 5.995701599121094,
      "epoch": 2.0961663417803766,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045634252577403644,
      "loss": 5.5651,
      "mean_token_accuracy": 0.18307170122861863,
      "num_tokens": 41026995.0,
      "step": 20970
    },
    {
      "entropy": 5.95415244102478,
      "epoch": 2.0966661668416053,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045632139694494294,
      "loss": 5.551,
      "mean_token_accuracy": 0.18786284178495408,
      "num_tokens": 41036243.0,
      "step": 20975
    },
    {
      "entropy": 5.988971376419068,
      "epoch": 2.097165991902834,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004563002635538496,
      "loss": 5.6128,
      "mean_token_accuracy": 0.17600207924842834,
      "num_tokens": 41046196.0,
      "step": 20980
    },
    {
      "entropy": 5.978775215148926,
      "epoch": 2.0976658169640627,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004562791256012884,
      "loss": 5.5954,
      "mean_token_accuracy": 0.18908213973045349,
      "num_tokens": 41057053.0,
      "step": 20985
    },
    {
      "entropy": 6.1095947265625,
      "epoch": 2.098165642025291,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004562579830877911,
      "loss": 5.7669,
      "mean_token_accuracy": 0.1740739166736603,
      "num_tokens": 41067349.0,
      "step": 20990
    },
    {
      "entropy": 6.066168165206909,
      "epoch": 2.0986654670865197,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004562368360138894,
      "loss": 5.6686,
      "mean_token_accuracy": 0.17063404768705367,
      "num_tokens": 41077327.0,
      "step": 20995
    },
    {
      "entropy": 5.996958780288696,
      "epoch": 2.0991652921477484,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004562156843801157,
      "loss": 5.6011,
      "mean_token_accuracy": 0.18262342363595963,
      "num_tokens": 41086569.0,
      "step": 21000
    },
    {
      "epoch": 2.0991652921477484,
      "eval_entropy": 5.760151024302144,
      "eval_loss": 5.785000801086426,
      "eval_mean_token_accuracy": 0.1773026290281751,
      "eval_num_tokens": 41086569.0,
      "eval_runtime": 24.0538,
      "eval_samples_per_second": 1290.647,
      "eval_steps_per_second": 161.346,
      "step": 21000
    },
    {
      "entropy": 5.9187723159790036,
      "epoch": 2.0996651172089766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045619452818700194,
      "loss": 5.5993,
      "mean_token_accuracy": 0.17001171112060548,
      "num_tokens": 41095170.0,
      "step": 21005
    },
    {
      "entropy": 6.011911153793335,
      "epoch": 2.1001649422702053,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045617336743508056,
      "loss": 5.5752,
      "mean_token_accuracy": 0.18012987971305847,
      "num_tokens": 41104778.0,
      "step": 21010
    },
    {
      "entropy": 5.963919162750244,
      "epoch": 2.100664767331434,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004561522021248837,
      "loss": 5.4837,
      "mean_token_accuracy": 0.1846874848008156,
      "num_tokens": 41114133.0,
      "step": 21015
    },
    {
      "entropy": 5.999068927764893,
      "epoch": 2.1011645923926627,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004561310322569442,
      "loss": 5.6125,
      "mean_token_accuracy": 0.18475084006786346,
      "num_tokens": 41125052.0,
      "step": 21020
    },
    {
      "entropy": 5.960227298736572,
      "epoch": 2.101664417453891,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045610985783179454,
      "loss": 5.6251,
      "mean_token_accuracy": 0.1772955983877182,
      "num_tokens": 41134160.0,
      "step": 21025
    },
    {
      "entropy": 5.987832164764404,
      "epoch": 2.1021642425151197,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004560886788499674,
      "loss": 5.5517,
      "mean_token_accuracy": 0.186137917637825,
      "num_tokens": 41144729.0,
      "step": 21030
    },
    {
      "entropy": 6.040094184875488,
      "epoch": 2.1026640675763484,
      "grad_norm": 1.125,
      "learning_rate": 0.00045606749531199573,
      "loss": 5.6511,
      "mean_token_accuracy": 0.17713914066553116,
      "num_tokens": 41154935.0,
      "step": 21035
    },
    {
      "entropy": 5.949699878692627,
      "epoch": 2.1031638926375766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045604630721841243,
      "loss": 5.6,
      "mean_token_accuracy": 0.17525193840265274,
      "num_tokens": 41164644.0,
      "step": 21040
    },
    {
      "entropy": 5.928917741775512,
      "epoch": 2.1036637176988053,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045602511456975064,
      "loss": 5.5299,
      "mean_token_accuracy": 0.18661922663450242,
      "num_tokens": 41174656.0,
      "step": 21045
    },
    {
      "entropy": 5.996991348266602,
      "epoch": 2.104163542760034,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004560039173665436,
      "loss": 5.6218,
      "mean_token_accuracy": 0.17511794418096543,
      "num_tokens": 41184631.0,
      "step": 21050
    },
    {
      "entropy": 5.998484659194946,
      "epoch": 2.1046633678212627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004559827156093245,
      "loss": 5.6099,
      "mean_token_accuracy": 0.1787296399474144,
      "num_tokens": 41195157.0,
      "step": 21055
    },
    {
      "entropy": 6.0453284740447994,
      "epoch": 2.105163192882491,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004559615092986268,
      "loss": 5.6599,
      "mean_token_accuracy": 0.17475499361753463,
      "num_tokens": 41205566.0,
      "step": 21060
    },
    {
      "entropy": 5.920873641967773,
      "epoch": 2.1056630179437197,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045594029843498425,
      "loss": 5.5571,
      "mean_token_accuracy": 0.1808719143271446,
      "num_tokens": 41214194.0,
      "step": 21065
    },
    {
      "entropy": 6.0325620651245115,
      "epoch": 2.1061628430049484,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045591908301893027,
      "loss": 5.7346,
      "mean_token_accuracy": 0.17309881001710892,
      "num_tokens": 41223965.0,
      "step": 21070
    },
    {
      "entropy": 6.003263664245606,
      "epoch": 2.1066626680661766,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045589786305099875,
      "loss": 5.6345,
      "mean_token_accuracy": 0.18083638846874237,
      "num_tokens": 41233374.0,
      "step": 21075
    },
    {
      "entropy": 6.038600063323974,
      "epoch": 2.1071624931274053,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004558766385317235,
      "loss": 5.691,
      "mean_token_accuracy": 0.1730738863348961,
      "num_tokens": 41243365.0,
      "step": 21080
    },
    {
      "entropy": 6.014450645446777,
      "epoch": 2.107662318188634,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045585540946163855,
      "loss": 5.5867,
      "mean_token_accuracy": 0.18894302397966384,
      "num_tokens": 41253013.0,
      "step": 21085
    },
    {
      "entropy": 5.97392840385437,
      "epoch": 2.1081621432498627,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004558341758412781,
      "loss": 5.5863,
      "mean_token_accuracy": 0.18169220983982087,
      "num_tokens": 41261889.0,
      "step": 21090
    },
    {
      "entropy": 5.965933752059937,
      "epoch": 2.108661968311091,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004558129376711763,
      "loss": 5.5846,
      "mean_token_accuracy": 0.17774361968040467,
      "num_tokens": 41271777.0,
      "step": 21095
    },
    {
      "entropy": 6.030124473571777,
      "epoch": 2.1091617933723197,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004557916949518675,
      "loss": 5.7101,
      "mean_token_accuracy": 0.17239125072956085,
      "num_tokens": 41281840.0,
      "step": 21100
    },
    {
      "entropy": 6.114180040359497,
      "epoch": 2.1096616184335484,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004557704476838862,
      "loss": 5.6964,
      "mean_token_accuracy": 0.17352344840765,
      "num_tokens": 41292200.0,
      "step": 21105
    },
    {
      "entropy": 6.071682596206665,
      "epoch": 2.1101614434947766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045574919586776686,
      "loss": 5.7465,
      "mean_token_accuracy": 0.17166658639907836,
      "num_tokens": 41301898.0,
      "step": 21110
    },
    {
      "entropy": 6.009588575363159,
      "epoch": 2.1106612685560053,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045572793950404434,
      "loss": 5.6167,
      "mean_token_accuracy": 0.17646341323852538,
      "num_tokens": 41311519.0,
      "step": 21115
    },
    {
      "entropy": 6.046613597869873,
      "epoch": 2.111161093617234,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045570667859325337,
      "loss": 5.6539,
      "mean_token_accuracy": 0.17206665873527527,
      "num_tokens": 41321955.0,
      "step": 21120
    },
    {
      "entropy": 6.091145086288452,
      "epoch": 2.1116609186784627,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045568541313592884,
      "loss": 5.7089,
      "mean_token_accuracy": 0.1690020114183426,
      "num_tokens": 41332467.0,
      "step": 21125
    },
    {
      "entropy": 6.075399494171142,
      "epoch": 2.112160743739691,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045566414313260586,
      "loss": 5.7321,
      "mean_token_accuracy": 0.17271345406770705,
      "num_tokens": 41342451.0,
      "step": 21130
    },
    {
      "entropy": 5.9603008270263675,
      "epoch": 2.1126605688009197,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004556428685838194,
      "loss": 5.5868,
      "mean_token_accuracy": 0.1772937834262848,
      "num_tokens": 41352678.0,
      "step": 21135
    },
    {
      "entropy": 6.013601589202881,
      "epoch": 2.1131603938621484,
      "grad_norm": 1.09375,
      "learning_rate": 0.000455621589490105,
      "loss": 5.6801,
      "mean_token_accuracy": 0.17622145414352416,
      "num_tokens": 41362584.0,
      "step": 21140
    },
    {
      "entropy": 6.027262210845947,
      "epoch": 2.1136602189233766,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004556003058519979,
      "loss": 5.6644,
      "mean_token_accuracy": 0.1770380362868309,
      "num_tokens": 41371988.0,
      "step": 21145
    },
    {
      "entropy": 6.032013273239135,
      "epoch": 2.1141600439846053,
      "grad_norm": 1.125,
      "learning_rate": 0.00045557901767003347,
      "loss": 5.584,
      "mean_token_accuracy": 0.1716403305530548,
      "num_tokens": 41381281.0,
      "step": 21150
    },
    {
      "entropy": 5.850177240371704,
      "epoch": 2.114659869045834,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045555772494474745,
      "loss": 5.4901,
      "mean_token_accuracy": 0.1935480996966362,
      "num_tokens": 41389711.0,
      "step": 21155
    },
    {
      "entropy": 6.019340705871582,
      "epoch": 2.1151596941070627,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004555364276766756,
      "loss": 5.6444,
      "mean_token_accuracy": 0.1763286516070366,
      "num_tokens": 41398375.0,
      "step": 21160
    },
    {
      "entropy": 6.054143619537354,
      "epoch": 2.115659519168291,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004555151258663535,
      "loss": 5.6921,
      "mean_token_accuracy": 0.16986872702836991,
      "num_tokens": 41408372.0,
      "step": 21165
    },
    {
      "entropy": 5.974713325500488,
      "epoch": 2.1161593442295197,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004554938195143175,
      "loss": 5.5988,
      "mean_token_accuracy": 0.1779281750321388,
      "num_tokens": 41418231.0,
      "step": 21170
    },
    {
      "entropy": 5.992068338394165,
      "epoch": 2.1166591692907484,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004554725086211033,
      "loss": 5.5982,
      "mean_token_accuracy": 0.1860845535993576,
      "num_tokens": 41427646.0,
      "step": 21175
    },
    {
      "entropy": 6.044457626342774,
      "epoch": 2.1171589943519766,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045545119318724734,
      "loss": 5.6811,
      "mean_token_accuracy": 0.17501528561115265,
      "num_tokens": 41437352.0,
      "step": 21180
    },
    {
      "entropy": 6.088871479034424,
      "epoch": 2.1176588194132053,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004554298732132857,
      "loss": 5.6453,
      "mean_token_accuracy": 0.1754573330283165,
      "num_tokens": 41446769.0,
      "step": 21185
    },
    {
      "entropy": 6.0376787185668945,
      "epoch": 2.118158644474434,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004554085486997549,
      "loss": 5.6499,
      "mean_token_accuracy": 0.1707320898771286,
      "num_tokens": 41456252.0,
      "step": 21190
    },
    {
      "entropy": 5.955099582672119,
      "epoch": 2.1186584695356627,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004553872196471914,
      "loss": 5.5765,
      "mean_token_accuracy": 0.1807342976331711,
      "num_tokens": 41465388.0,
      "step": 21195
    },
    {
      "entropy": 6.063111162185669,
      "epoch": 2.119158294596891,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045536588605613195,
      "loss": 5.6573,
      "mean_token_accuracy": 0.1710526466369629,
      "num_tokens": 41474772.0,
      "step": 21200
    },
    {
      "entropy": 5.974788570404053,
      "epoch": 2.1196581196581197,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045534454792711324,
      "loss": 5.5522,
      "mean_token_accuracy": 0.17948205024003983,
      "num_tokens": 41484767.0,
      "step": 21205
    },
    {
      "entropy": 5.977580785751343,
      "epoch": 2.1201579447193484,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045532320526067206,
      "loss": 5.6381,
      "mean_token_accuracy": 0.18072713762521744,
      "num_tokens": 41495305.0,
      "step": 21210
    },
    {
      "entropy": 6.025302457809448,
      "epoch": 2.1206577697805766,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004553018580573454,
      "loss": 5.581,
      "mean_token_accuracy": 0.1789972648024559,
      "num_tokens": 41504427.0,
      "step": 21215
    },
    {
      "entropy": 5.9873504638671875,
      "epoch": 2.1211575948418053,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045528050631767037,
      "loss": 5.6394,
      "mean_token_accuracy": 0.1763248175382614,
      "num_tokens": 41514607.0,
      "step": 21220
    },
    {
      "entropy": 5.998300457000733,
      "epoch": 2.121657419903034,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045525915004218426,
      "loss": 5.6451,
      "mean_token_accuracy": 0.1711382418870926,
      "num_tokens": 41524280.0,
      "step": 21225
    },
    {
      "entropy": 6.041611671447754,
      "epoch": 2.1221572449642627,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045523778923142435,
      "loss": 5.6293,
      "mean_token_accuracy": 0.1712128385901451,
      "num_tokens": 41533059.0,
      "step": 21230
    },
    {
      "entropy": 6.0867103099822994,
      "epoch": 2.122657070025491,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004552164238859279,
      "loss": 5.6749,
      "mean_token_accuracy": 0.1739403262734413,
      "num_tokens": 41542876.0,
      "step": 21235
    },
    {
      "entropy": 6.020427465438843,
      "epoch": 2.1231568950867197,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045519505400623274,
      "loss": 5.6908,
      "mean_token_accuracy": 0.16513564735651015,
      "num_tokens": 41552337.0,
      "step": 21240
    },
    {
      "entropy": 6.0266131401062015,
      "epoch": 2.1236567201479484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045517367959287635,
      "loss": 5.7752,
      "mean_token_accuracy": 0.17050580233335494,
      "num_tokens": 41563190.0,
      "step": 21245
    },
    {
      "entropy": 5.9976318359375,
      "epoch": 2.1241565452091766,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045515230064639657,
      "loss": 5.5495,
      "mean_token_accuracy": 0.18703828155994415,
      "num_tokens": 41573927.0,
      "step": 21250
    },
    {
      "entropy": 6.024460411071777,
      "epoch": 2.1246563702704053,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045513091716733117,
      "loss": 5.4677,
      "mean_token_accuracy": 0.19672871828079225,
      "num_tokens": 41583828.0,
      "step": 21255
    },
    {
      "entropy": 5.933193159103394,
      "epoch": 2.125156195331634,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045510952915621834,
      "loss": 5.5891,
      "mean_token_accuracy": 0.1767452508211136,
      "num_tokens": 41593179.0,
      "step": 21260
    },
    {
      "entropy": 6.0088615894317625,
      "epoch": 2.1256560203928627,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004550881366135961,
      "loss": 5.7108,
      "mean_token_accuracy": 0.17286068797111512,
      "num_tokens": 41603366.0,
      "step": 21265
    },
    {
      "entropy": 6.0568013191223145,
      "epoch": 2.126155845454091,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045506673954000265,
      "loss": 5.6515,
      "mean_token_accuracy": 0.17433069199323653,
      "num_tokens": 41612916.0,
      "step": 21270
    },
    {
      "entropy": 6.089569711685181,
      "epoch": 2.1266556705153197,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004550453379359763,
      "loss": 5.6883,
      "mean_token_accuracy": 0.1722179025411606,
      "num_tokens": 41622770.0,
      "step": 21275
    },
    {
      "entropy": 6.081288242340088,
      "epoch": 2.1271554955765484,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045502393180205564,
      "loss": 5.6794,
      "mean_token_accuracy": 0.1744096979498863,
      "num_tokens": 41632590.0,
      "step": 21280
    },
    {
      "entropy": 6.024157524108887,
      "epoch": 2.1276553206377766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004550025211387792,
      "loss": 5.6939,
      "mean_token_accuracy": 0.1776106759905815,
      "num_tokens": 41641932.0,
      "step": 21285
    },
    {
      "entropy": 6.055584049224853,
      "epoch": 2.1281551456990053,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004549811059466857,
      "loss": 5.6918,
      "mean_token_accuracy": 0.1697285771369934,
      "num_tokens": 41651915.0,
      "step": 21290
    },
    {
      "entropy": 5.950869703292847,
      "epoch": 2.128654970760234,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004549596862263138,
      "loss": 5.4891,
      "mean_token_accuracy": 0.1926019683480263,
      "num_tokens": 41661087.0,
      "step": 21295
    },
    {
      "entropy": 5.980437278747559,
      "epoch": 2.1291547958214627,
      "grad_norm": 1.125,
      "learning_rate": 0.0004549382619782025,
      "loss": 5.6422,
      "mean_token_accuracy": 0.1795808807015419,
      "num_tokens": 41670418.0,
      "step": 21300
    },
    {
      "entropy": 6.011434125900268,
      "epoch": 2.129654620882691,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004549168332028908,
      "loss": 5.5774,
      "mean_token_accuracy": 0.1788506716489792,
      "num_tokens": 41680209.0,
      "step": 21305
    },
    {
      "entropy": 6.081386423110962,
      "epoch": 2.1301544459439197,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045489539990091796,
      "loss": 5.6533,
      "mean_token_accuracy": 0.171646586060524,
      "num_tokens": 41689933.0,
      "step": 21310
    },
    {
      "entropy": 5.956134366989136,
      "epoch": 2.1306542710051484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045487396207282316,
      "loss": 5.5316,
      "mean_token_accuracy": 0.18600809574127197,
      "num_tokens": 41699017.0,
      "step": 21315
    },
    {
      "entropy": 5.950873804092407,
      "epoch": 2.1311540960663766,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045485251971914565,
      "loss": 5.6098,
      "mean_token_accuracy": 0.17384656369686127,
      "num_tokens": 41708375.0,
      "step": 21320
    },
    {
      "entropy": 5.979651975631714,
      "epoch": 2.1316539211276053,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004548310728404251,
      "loss": 5.5848,
      "mean_token_accuracy": 0.1793954685330391,
      "num_tokens": 41718269.0,
      "step": 21325
    },
    {
      "entropy": 6.007793378829956,
      "epoch": 2.132153746188834,
      "grad_norm": 1.125,
      "learning_rate": 0.000454809621437201,
      "loss": 5.6478,
      "mean_token_accuracy": 0.17182848304510118,
      "num_tokens": 41728426.0,
      "step": 21330
    },
    {
      "entropy": 5.98306975364685,
      "epoch": 2.1326535712500623,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004547881655100131,
      "loss": 5.6159,
      "mean_token_accuracy": 0.17133540213108062,
      "num_tokens": 41738255.0,
      "step": 21335
    },
    {
      "entropy": 6.111949110031128,
      "epoch": 2.133153396311291,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004547667050594013,
      "loss": 5.7104,
      "mean_token_accuracy": 0.1687174305319786,
      "num_tokens": 41747668.0,
      "step": 21340
    },
    {
      "entropy": 5.986235523223877,
      "epoch": 2.1336532213725197,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045474524008590536,
      "loss": 5.5728,
      "mean_token_accuracy": 0.1809978097677231,
      "num_tokens": 41757414.0,
      "step": 21345
    },
    {
      "entropy": 5.961528158187866,
      "epoch": 2.1341530464337484,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004547237705900655,
      "loss": 5.6535,
      "mean_token_accuracy": 0.18244727253913878,
      "num_tokens": 41765413.0,
      "step": 21350
    },
    {
      "entropy": 5.992355251312256,
      "epoch": 2.1346528714949766,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045470229657242176,
      "loss": 5.6162,
      "mean_token_accuracy": 0.1787830635905266,
      "num_tokens": 41775951.0,
      "step": 21355
    },
    {
      "entropy": 5.975080347061157,
      "epoch": 2.1351526965562053,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045468081803351447,
      "loss": 5.572,
      "mean_token_accuracy": 0.18126858472824098,
      "num_tokens": 41785440.0,
      "step": 21360
    },
    {
      "entropy": 6.014530754089355,
      "epoch": 2.135652521617434,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004546593349738841,
      "loss": 5.6395,
      "mean_token_accuracy": 0.174817056953907,
      "num_tokens": 41794757.0,
      "step": 21365
    },
    {
      "entropy": 6.023870897293091,
      "epoch": 2.1361523466786623,
      "grad_norm": 1.171875,
      "learning_rate": 0.000454637847394071,
      "loss": 5.7113,
      "mean_token_accuracy": 0.17652756720781326,
      "num_tokens": 41804491.0,
      "step": 21370
    },
    {
      "entropy": 6.02180118560791,
      "epoch": 2.136652171739891,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045461635529461597,
      "loss": 5.5641,
      "mean_token_accuracy": 0.18839337080717086,
      "num_tokens": 41813867.0,
      "step": 21375
    },
    {
      "entropy": 6.063761472702026,
      "epoch": 2.1371519968011197,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004545948586760596,
      "loss": 5.7425,
      "mean_token_accuracy": 0.16165613606572152,
      "num_tokens": 41823167.0,
      "step": 21380
    },
    {
      "entropy": 6.005662965774536,
      "epoch": 2.1376518218623484,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004545733575389428,
      "loss": 5.5913,
      "mean_token_accuracy": 0.17197030037641525,
      "num_tokens": 41831757.0,
      "step": 21385
    },
    {
      "entropy": 6.0247540950775145,
      "epoch": 2.1381516469235766,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004545518518838066,
      "loss": 5.5585,
      "mean_token_accuracy": 0.18300383388996125,
      "num_tokens": 41841741.0,
      "step": 21390
    },
    {
      "entropy": 5.95669732093811,
      "epoch": 2.1386514719848053,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000454530341711192,
      "loss": 5.5536,
      "mean_token_accuracy": 0.1817866548895836,
      "num_tokens": 41851940.0,
      "step": 21395
    },
    {
      "entropy": 6.005340719223023,
      "epoch": 2.139151297046034,
      "grad_norm": 1.125,
      "learning_rate": 0.0004545088270216401,
      "loss": 5.5869,
      "mean_token_accuracy": 0.1774577796459198,
      "num_tokens": 41861240.0,
      "step": 21400
    },
    {
      "entropy": 5.974078416824341,
      "epoch": 2.1396511221072623,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004544873078156924,
      "loss": 5.7132,
      "mean_token_accuracy": 0.17424826472997665,
      "num_tokens": 41871513.0,
      "step": 21405
    },
    {
      "entropy": 6.133807706832886,
      "epoch": 2.140150947168491,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045446578409389016,
      "loss": 5.7018,
      "mean_token_accuracy": 0.17338773757219314,
      "num_tokens": 41880346.0,
      "step": 21410
    },
    {
      "entropy": 6.02174859046936,
      "epoch": 2.1406507722297197,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045444425585677503,
      "loss": 5.5829,
      "mean_token_accuracy": 0.17812291979789735,
      "num_tokens": 41889182.0,
      "step": 21415
    },
    {
      "entropy": 5.952095699310303,
      "epoch": 2.1411505972909484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004544227231048886,
      "loss": 5.5787,
      "mean_token_accuracy": 0.17877241373062133,
      "num_tokens": 41899108.0,
      "step": 21420
    },
    {
      "entropy": 5.9841197490692135,
      "epoch": 2.1416504223521766,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004544011858387726,
      "loss": 5.6265,
      "mean_token_accuracy": 0.18045714646577835,
      "num_tokens": 41908930.0,
      "step": 21425
    },
    {
      "entropy": 6.016934967041015,
      "epoch": 2.1421502474134053,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045437964405896895,
      "loss": 5.6259,
      "mean_token_accuracy": 0.18225086033344268,
      "num_tokens": 41919043.0,
      "step": 21430
    },
    {
      "entropy": 5.9735290050506595,
      "epoch": 2.142650072474634,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045435809776601954,
      "loss": 5.5738,
      "mean_token_accuracy": 0.18683733493089677,
      "num_tokens": 41929241.0,
      "step": 21435
    },
    {
      "entropy": 6.02107458114624,
      "epoch": 2.1431498975358623,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004543365469604666,
      "loss": 5.6449,
      "mean_token_accuracy": 0.17753954827785492,
      "num_tokens": 41940829.0,
      "step": 21440
    },
    {
      "entropy": 6.045332574844361,
      "epoch": 2.143649722597091,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004543149916428523,
      "loss": 5.6974,
      "mean_token_accuracy": 0.17361164540052415,
      "num_tokens": 41950137.0,
      "step": 21445
    },
    {
      "entropy": 6.014636659622193,
      "epoch": 2.1441495476583197,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004542934318137189,
      "loss": 5.6306,
      "mean_token_accuracy": 0.17607682794332505,
      "num_tokens": 41960089.0,
      "step": 21450
    },
    {
      "entropy": 5.999350452423096,
      "epoch": 2.1446493727195484,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004542718674736089,
      "loss": 5.6094,
      "mean_token_accuracy": 0.1798594743013382,
      "num_tokens": 41970168.0,
      "step": 21455
    },
    {
      "entropy": 5.960920763015747,
      "epoch": 2.1451491977807766,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004542502986230649,
      "loss": 5.5972,
      "mean_token_accuracy": 0.17687120139598847,
      "num_tokens": 41979540.0,
      "step": 21460
    },
    {
      "entropy": 6.000763988494873,
      "epoch": 2.1456490228420053,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004542287252626295,
      "loss": 5.6663,
      "mean_token_accuracy": 0.18208467960357666,
      "num_tokens": 41990090.0,
      "step": 21465
    },
    {
      "entropy": 6.045498943328857,
      "epoch": 2.146148847903234,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004542071473928455,
      "loss": 5.7106,
      "mean_token_accuracy": 0.1689659148454666,
      "num_tokens": 42000455.0,
      "step": 21470
    },
    {
      "entropy": 6.005690336227417,
      "epoch": 2.1466486729644623,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004541855650142557,
      "loss": 5.6952,
      "mean_token_accuracy": 0.172666272521019,
      "num_tokens": 42010767.0,
      "step": 21475
    },
    {
      "entropy": 5.997534465789795,
      "epoch": 2.147148498025691,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045416397812740323,
      "loss": 5.591,
      "mean_token_accuracy": 0.18050322979688643,
      "num_tokens": 42020043.0,
      "step": 21480
    },
    {
      "entropy": 5.98363037109375,
      "epoch": 2.1476483230869197,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004541423867328312,
      "loss": 5.663,
      "mean_token_accuracy": 0.17419141829013823,
      "num_tokens": 42030511.0,
      "step": 21485
    },
    {
      "entropy": 6.056385040283203,
      "epoch": 2.148148148148148,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004541207908310829,
      "loss": 5.6605,
      "mean_token_accuracy": 0.18034241795539857,
      "num_tokens": 42040051.0,
      "step": 21490
    },
    {
      "entropy": 5.982808589935303,
      "epoch": 2.1486479732093766,
      "grad_norm": 1.125,
      "learning_rate": 0.0004540991904227015,
      "loss": 5.6664,
      "mean_token_accuracy": 0.17471336126327514,
      "num_tokens": 42050206.0,
      "step": 21495
    },
    {
      "entropy": 6.013383960723877,
      "epoch": 2.1491477982706053,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045407758550823055,
      "loss": 5.6331,
      "mean_token_accuracy": 0.17562078982591628,
      "num_tokens": 42059420.0,
      "step": 21500
    },
    {
      "entropy": 6.037286758422852,
      "epoch": 2.149647623331834,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045405597608821365,
      "loss": 5.5824,
      "mean_token_accuracy": 0.17691798210144044,
      "num_tokens": 42068527.0,
      "step": 21505
    },
    {
      "entropy": 6.032791042327881,
      "epoch": 2.1501474483930623,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004540343621631944,
      "loss": 5.6566,
      "mean_token_accuracy": 0.17065726965665817,
      "num_tokens": 42078006.0,
      "step": 21510
    },
    {
      "entropy": 6.030219841003418,
      "epoch": 2.150647273454291,
      "grad_norm": 1.25,
      "learning_rate": 0.0004540127437337167,
      "loss": 5.6871,
      "mean_token_accuracy": 0.16744364351034163,
      "num_tokens": 42087535.0,
      "step": 21515
    },
    {
      "entropy": 6.045618534088135,
      "epoch": 2.1511470985155197,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004539911208003245,
      "loss": 5.6103,
      "mean_token_accuracy": 0.1765738010406494,
      "num_tokens": 42097259.0,
      "step": 21520
    },
    {
      "entropy": 5.960878896713257,
      "epoch": 2.151646923576748,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004539694933635617,
      "loss": 5.5451,
      "mean_token_accuracy": 0.18344005346298217,
      "num_tokens": 42107216.0,
      "step": 21525
    },
    {
      "entropy": 5.978801441192627,
      "epoch": 2.1521467486379766,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045394786142397253,
      "loss": 5.7345,
      "mean_token_accuracy": 0.17084817588329315,
      "num_tokens": 42116014.0,
      "step": 21530
    },
    {
      "entropy": 6.0122997760772705,
      "epoch": 2.1526465736992053,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004539262249821012,
      "loss": 5.6117,
      "mean_token_accuracy": 0.17381250113248825,
      "num_tokens": 42125864.0,
      "step": 21535
    },
    {
      "entropy": 5.918548583984375,
      "epoch": 2.153146398760434,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045390458403849203,
      "loss": 5.4906,
      "mean_token_accuracy": 0.19116761088371276,
      "num_tokens": 42135048.0,
      "step": 21540
    },
    {
      "entropy": 5.945430564880371,
      "epoch": 2.1536462238216623,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004538829385936896,
      "loss": 5.5942,
      "mean_token_accuracy": 0.17425792068243026,
      "num_tokens": 42144692.0,
      "step": 21545
    },
    {
      "entropy": 6.070108890533447,
      "epoch": 2.154146048882891,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004538612886482385,
      "loss": 5.7622,
      "mean_token_accuracy": 0.1736515685915947,
      "num_tokens": 42155117.0,
      "step": 21550
    },
    {
      "entropy": 6.080785274505615,
      "epoch": 2.1546458739441197,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004538396342026834,
      "loss": 5.6841,
      "mean_token_accuracy": 0.17129656672477722,
      "num_tokens": 42165428.0,
      "step": 21555
    },
    {
      "entropy": 5.950024366378784,
      "epoch": 2.155145699005348,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004538179752575691,
      "loss": 5.5687,
      "mean_token_accuracy": 0.18229983747005463,
      "num_tokens": 42175841.0,
      "step": 21560
    },
    {
      "entropy": 6.034072542190552,
      "epoch": 2.1556455240665766,
      "grad_norm": 1.25,
      "learning_rate": 0.0004537963118134406,
      "loss": 5.6338,
      "mean_token_accuracy": 0.17222427278757096,
      "num_tokens": 42186450.0,
      "step": 21565
    },
    {
      "entropy": 5.999091386795044,
      "epoch": 2.1561453491278053,
      "grad_norm": 1.125,
      "learning_rate": 0.00045377464387084287,
      "loss": 5.6935,
      "mean_token_accuracy": 0.17185195982456208,
      "num_tokens": 42195994.0,
      "step": 21570
    },
    {
      "entropy": 6.018128728866577,
      "epoch": 2.156645174189034,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004537529714303211,
      "loss": 5.5568,
      "mean_token_accuracy": 0.18517017215490342,
      "num_tokens": 42206662.0,
      "step": 21575
    },
    {
      "entropy": 5.9269983768463135,
      "epoch": 2.1571449992502623,
      "grad_norm": 1.125,
      "learning_rate": 0.00045373129449242054,
      "loss": 5.4931,
      "mean_token_accuracy": 0.18555851876735688,
      "num_tokens": 42216598.0,
      "step": 21580
    },
    {
      "entropy": 5.942392015457154,
      "epoch": 2.157644824311491,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004537096130576866,
      "loss": 5.5788,
      "mean_token_accuracy": 0.1811902865767479,
      "num_tokens": 42225869.0,
      "step": 21585
    },
    {
      "entropy": 5.999599933624268,
      "epoch": 2.1581446493727197,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045368792712666486,
      "loss": 5.7551,
      "mean_token_accuracy": 0.16748702228069307,
      "num_tokens": 42234749.0,
      "step": 21590
    },
    {
      "entropy": 6.021525764465332,
      "epoch": 2.158644474433948,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004536662366999008,
      "loss": 5.6341,
      "mean_token_accuracy": 0.16918477714061736,
      "num_tokens": 42243503.0,
      "step": 21595
    },
    {
      "entropy": 6.056520128250122,
      "epoch": 2.1591442994951766,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004536445417779402,
      "loss": 5.7106,
      "mean_token_accuracy": 0.17206961959600447,
      "num_tokens": 42253888.0,
      "step": 21600
    },
    {
      "entropy": 6.032310914993286,
      "epoch": 2.1596441245564053,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045362284236132895,
      "loss": 5.7017,
      "mean_token_accuracy": 0.17758455574512483,
      "num_tokens": 42264052.0,
      "step": 21605
    },
    {
      "entropy": 6.021656513214111,
      "epoch": 2.160143949617634,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045360113845061283,
      "loss": 5.6585,
      "mean_token_accuracy": 0.17350329905748368,
      "num_tokens": 42273749.0,
      "step": 21610
    },
    {
      "entropy": 6.01168851852417,
      "epoch": 2.1606437746788623,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045357943004633813,
      "loss": 5.6327,
      "mean_token_accuracy": 0.1790459409356117,
      "num_tokens": 42282209.0,
      "step": 21615
    },
    {
      "entropy": 5.98282527923584,
      "epoch": 2.161143599740091,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045355771714905086,
      "loss": 5.6434,
      "mean_token_accuracy": 0.1801111102104187,
      "num_tokens": 42291941.0,
      "step": 21620
    },
    {
      "entropy": 6.04854006767273,
      "epoch": 2.1616434248013197,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045353599975929746,
      "loss": 5.7142,
      "mean_token_accuracy": 0.17333365380764007,
      "num_tokens": 42300630.0,
      "step": 21625
    },
    {
      "entropy": 5.965293121337891,
      "epoch": 2.162143249862548,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004535142778776241,
      "loss": 5.5423,
      "mean_token_accuracy": 0.18314653784036636,
      "num_tokens": 42309973.0,
      "step": 21630
    },
    {
      "entropy": 6.006385564804077,
      "epoch": 2.1626430749237766,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004534925515045776,
      "loss": 5.7449,
      "mean_token_accuracy": 0.1693398281931877,
      "num_tokens": 42319880.0,
      "step": 21635
    },
    {
      "entropy": 6.028097772598267,
      "epoch": 2.1631428999850053,
      "grad_norm": 1.125,
      "learning_rate": 0.00045347082064070435,
      "loss": 5.5061,
      "mean_token_accuracy": 0.18925315588712693,
      "num_tokens": 42329554.0,
      "step": 21640
    },
    {
      "entropy": 5.937973880767823,
      "epoch": 2.163642725046234,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004534490852865512,
      "loss": 5.5116,
      "mean_token_accuracy": 0.1832076147198677,
      "num_tokens": 42339118.0,
      "step": 21645
    },
    {
      "entropy": 5.953856611251831,
      "epoch": 2.1641425501074623,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045342734544266493,
      "loss": 5.563,
      "mean_token_accuracy": 0.1813263103365898,
      "num_tokens": 42349115.0,
      "step": 21650
    },
    {
      "entropy": 5.948868083953857,
      "epoch": 2.164642375168691,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004534056011095926,
      "loss": 5.5857,
      "mean_token_accuracy": 0.17753728181123735,
      "num_tokens": 42357772.0,
      "step": 21655
    },
    {
      "entropy": 5.975905227661133,
      "epoch": 2.1651422002299197,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045338385228788124,
      "loss": 5.6047,
      "mean_token_accuracy": 0.17833266109228135,
      "num_tokens": 42366908.0,
      "step": 21660
    },
    {
      "entropy": 5.861214923858642,
      "epoch": 2.165642025291148,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045336209897807813,
      "loss": 5.4865,
      "mean_token_accuracy": 0.197310371696949,
      "num_tokens": 42377143.0,
      "step": 21665
    },
    {
      "entropy": 6.034654569625855,
      "epoch": 2.1661418503523766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045334034118073047,
      "loss": 5.6138,
      "mean_token_accuracy": 0.17795279622077942,
      "num_tokens": 42386815.0,
      "step": 21670
    },
    {
      "entropy": 5.974959802627564,
      "epoch": 2.1666416754136053,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045331857889638566,
      "loss": 5.5733,
      "mean_token_accuracy": 0.18344664722681045,
      "num_tokens": 42395821.0,
      "step": 21675
    },
    {
      "entropy": 5.9136899471282955,
      "epoch": 2.167141500474834,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004532968121255914,
      "loss": 5.5242,
      "mean_token_accuracy": 0.18578352630138398,
      "num_tokens": 42406757.0,
      "step": 21680
    },
    {
      "entropy": 5.96075758934021,
      "epoch": 2.1676413255360623,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045327504086889513,
      "loss": 5.6289,
      "mean_token_accuracy": 0.17703679651021959,
      "num_tokens": 42417288.0,
      "step": 21685
    },
    {
      "entropy": 5.9087827682495115,
      "epoch": 2.168141150597291,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045325326512684466,
      "loss": 5.4383,
      "mean_token_accuracy": 0.18585755825042724,
      "num_tokens": 42427614.0,
      "step": 21690
    },
    {
      "entropy": 6.056416940689087,
      "epoch": 2.1686409756585197,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045323148489998807,
      "loss": 5.7286,
      "mean_token_accuracy": 0.17170295119285583,
      "num_tokens": 42437475.0,
      "step": 21695
    },
    {
      "entropy": 5.966514253616333,
      "epoch": 2.169140800719748,
      "grad_norm": 1.171875,
      "learning_rate": 0.000453209700188873,
      "loss": 5.5424,
      "mean_token_accuracy": 0.18049188405275346,
      "num_tokens": 42446654.0,
      "step": 21700
    },
    {
      "entropy": 5.9660560131073,
      "epoch": 2.1696406257809766,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045318791099404785,
      "loss": 5.6011,
      "mean_token_accuracy": 0.17636545598506928,
      "num_tokens": 42456952.0,
      "step": 21705
    },
    {
      "entropy": 5.991500663757324,
      "epoch": 2.1701404508422053,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045316611731606075,
      "loss": 5.6061,
      "mean_token_accuracy": 0.17630643248558045,
      "num_tokens": 42466133.0,
      "step": 21710
    },
    {
      "entropy": 6.025596237182617,
      "epoch": 2.170640275903434,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004531443191554598,
      "loss": 5.7066,
      "mean_token_accuracy": 0.172885662317276,
      "num_tokens": 42476068.0,
      "step": 21715
    },
    {
      "entropy": 6.040198040008545,
      "epoch": 2.1711401009646623,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004531225165127938,
      "loss": 5.6168,
      "mean_token_accuracy": 0.17786945402622223,
      "num_tokens": 42485914.0,
      "step": 21720
    },
    {
      "entropy": 5.946213340759277,
      "epoch": 2.171639926025891,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045310070938861107,
      "loss": 5.6211,
      "mean_token_accuracy": 0.17254974991083144,
      "num_tokens": 42495563.0,
      "step": 21725
    },
    {
      "entropy": 6.0051287651062015,
      "epoch": 2.1721397510871197,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045307889778346035,
      "loss": 5.6264,
      "mean_token_accuracy": 0.17353340685367585,
      "num_tokens": 42505441.0,
      "step": 21730
    },
    {
      "entropy": 6.077885007858276,
      "epoch": 2.172639576148348,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004530570816978903,
      "loss": 5.6904,
      "mean_token_accuracy": 0.17312463968992234,
      "num_tokens": 42515636.0,
      "step": 21735
    },
    {
      "entropy": 6.0325840473175045,
      "epoch": 2.1731394012095766,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045303526113245,
      "loss": 5.6816,
      "mean_token_accuracy": 0.1759546607732773,
      "num_tokens": 42525143.0,
      "step": 21740
    },
    {
      "entropy": 5.994698762893677,
      "epoch": 2.1736392262708053,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004530134360876882,
      "loss": 5.5807,
      "mean_token_accuracy": 0.18245095014572144,
      "num_tokens": 42534591.0,
      "step": 21745
    },
    {
      "entropy": 6.0801738739013675,
      "epoch": 2.174139051332034,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004529916065641543,
      "loss": 5.7399,
      "mean_token_accuracy": 0.17144397497177125,
      "num_tokens": 42544956.0,
      "step": 21750
    },
    {
      "entropy": 5.978320741653443,
      "epoch": 2.1746388763932623,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004529697725623973,
      "loss": 5.5704,
      "mean_token_accuracy": 0.18397926092147826,
      "num_tokens": 42554383.0,
      "step": 21755
    },
    {
      "entropy": 5.949664878845215,
      "epoch": 2.175138701454491,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045294793408296656,
      "loss": 5.5621,
      "mean_token_accuracy": 0.18971733301877974,
      "num_tokens": 42563703.0,
      "step": 21760
    },
    {
      "entropy": 5.880137300491333,
      "epoch": 2.1756385265157197,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004529260911264117,
      "loss": 5.5334,
      "mean_token_accuracy": 0.18114394843578338,
      "num_tokens": 42572225.0,
      "step": 21765
    },
    {
      "entropy": 5.9198392868042,
      "epoch": 2.176138351576948,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004529042436932822,
      "loss": 5.5193,
      "mean_token_accuracy": 0.18935615867376326,
      "num_tokens": 42582053.0,
      "step": 21770
    },
    {
      "entropy": 5.904947185516358,
      "epoch": 2.1766381766381766,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045288239178412754,
      "loss": 5.5934,
      "mean_token_accuracy": 0.17763956785202026,
      "num_tokens": 42590682.0,
      "step": 21775
    },
    {
      "entropy": 5.996451663970947,
      "epoch": 2.1771380016994053,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004528605353994978,
      "loss": 5.6597,
      "mean_token_accuracy": 0.17756882309913635,
      "num_tokens": 42598769.0,
      "step": 21780
    },
    {
      "entropy": 5.970040941238404,
      "epoch": 2.177637826760634,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004528386745399428,
      "loss": 5.539,
      "mean_token_accuracy": 0.181155826151371,
      "num_tokens": 42608126.0,
      "step": 21785
    },
    {
      "entropy": 6.027914381027221,
      "epoch": 2.1781376518218623,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045281680920601235,
      "loss": 5.7231,
      "mean_token_accuracy": 0.17415328174829484,
      "num_tokens": 42617602.0,
      "step": 21790
    },
    {
      "entropy": 5.957484245300293,
      "epoch": 2.178637476883091,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004527949393982569,
      "loss": 5.554,
      "mean_token_accuracy": 0.18232868164777755,
      "num_tokens": 42627316.0,
      "step": 21795
    },
    {
      "entropy": 6.005135726928711,
      "epoch": 2.1791373019443196,
      "grad_norm": 1.125,
      "learning_rate": 0.00045277306511722645,
      "loss": 5.5992,
      "mean_token_accuracy": 0.17002962827682494,
      "num_tokens": 42636872.0,
      "step": 21800
    },
    {
      "entropy": 6.006972312927246,
      "epoch": 2.179637127005548,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004527511863634714,
      "loss": 5.6091,
      "mean_token_accuracy": 0.18088026642799376,
      "num_tokens": 42645754.0,
      "step": 21805
    },
    {
      "entropy": 5.936882591247558,
      "epoch": 2.1801369520667766,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004527293031375422,
      "loss": 5.5875,
      "mean_token_accuracy": 0.1815231442451477,
      "num_tokens": 42655219.0,
      "step": 21810
    },
    {
      "entropy": 5.962015390396118,
      "epoch": 2.1806367771280053,
      "grad_norm": 1.125,
      "learning_rate": 0.0004527074154399896,
      "loss": 5.6047,
      "mean_token_accuracy": 0.1816238448023796,
      "num_tokens": 42664398.0,
      "step": 21815
    },
    {
      "entropy": 5.973158502578736,
      "epoch": 2.181136602189234,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045268552327136415,
      "loss": 5.6723,
      "mean_token_accuracy": 0.17738437354564668,
      "num_tokens": 42674513.0,
      "step": 21820
    },
    {
      "entropy": 6.047936248779297,
      "epoch": 2.1816364272504623,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045266362663221656,
      "loss": 5.6291,
      "mean_token_accuracy": 0.18032426834106446,
      "num_tokens": 42684379.0,
      "step": 21825
    },
    {
      "entropy": 5.897942686080933,
      "epoch": 2.182136252311691,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045264172552309786,
      "loss": 5.5652,
      "mean_token_accuracy": 0.1862701579928398,
      "num_tokens": 42694020.0,
      "step": 21830
    },
    {
      "entropy": 5.946180963516236,
      "epoch": 2.1826360773729196,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004526198199445591,
      "loss": 5.5624,
      "mean_token_accuracy": 0.17922947108745574,
      "num_tokens": 42703882.0,
      "step": 21835
    },
    {
      "entropy": 6.02995457649231,
      "epoch": 2.183135902434148,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045259790989715133,
      "loss": 5.5866,
      "mean_token_accuracy": 0.17858875989913942,
      "num_tokens": 42713148.0,
      "step": 21840
    },
    {
      "entropy": 6.111298561096191,
      "epoch": 2.1836357274953766,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045257599538142587,
      "loss": 5.7167,
      "mean_token_accuracy": 0.17840371131896973,
      "num_tokens": 42722177.0,
      "step": 21845
    },
    {
      "entropy": 6.009223794937133,
      "epoch": 2.1841355525566053,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045255407639793407,
      "loss": 5.6751,
      "mean_token_accuracy": 0.17203425914049147,
      "num_tokens": 42732454.0,
      "step": 21850
    },
    {
      "entropy": 6.00037579536438,
      "epoch": 2.1846353776178336,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004525321529472274,
      "loss": 5.7175,
      "mean_token_accuracy": 0.17536755353212358,
      "num_tokens": 42742357.0,
      "step": 21855
    },
    {
      "entropy": 5.984578704833984,
      "epoch": 2.1851352026790622,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045251022502985737,
      "loss": 5.5885,
      "mean_token_accuracy": 0.17779186218976975,
      "num_tokens": 42751703.0,
      "step": 21860
    },
    {
      "entropy": 6.009545707702637,
      "epoch": 2.185635027740291,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004524882926463758,
      "loss": 5.6053,
      "mean_token_accuracy": 0.17552864998579026,
      "num_tokens": 42762198.0,
      "step": 21865
    },
    {
      "entropy": 6.023397541046142,
      "epoch": 2.1861348528015196,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004524663557973345,
      "loss": 5.663,
      "mean_token_accuracy": 0.1692187562584877,
      "num_tokens": 42772348.0,
      "step": 21870
    },
    {
      "entropy": 5.955418062210083,
      "epoch": 2.186634677862748,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045244441448328535,
      "loss": 5.5921,
      "mean_token_accuracy": 0.17798011749982834,
      "num_tokens": 42782592.0,
      "step": 21875
    },
    {
      "entropy": 6.008326816558838,
      "epoch": 2.1871345029239766,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045242246870478035,
      "loss": 5.6289,
      "mean_token_accuracy": 0.17862791568040848,
      "num_tokens": 42792452.0,
      "step": 21880
    },
    {
      "entropy": 5.98037314414978,
      "epoch": 2.1876343279852053,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004524005184623717,
      "loss": 5.5795,
      "mean_token_accuracy": 0.18368964791297912,
      "num_tokens": 42801545.0,
      "step": 21885
    },
    {
      "entropy": 6.003242111206054,
      "epoch": 2.1881341530464335,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004523785637566117,
      "loss": 5.5781,
      "mean_token_accuracy": 0.18493310660123824,
      "num_tokens": 42810504.0,
      "step": 21890
    },
    {
      "entropy": 5.9646217823028564,
      "epoch": 2.1886339781076622,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004523566045880527,
      "loss": 5.5962,
      "mean_token_accuracy": 0.1793338879942894,
      "num_tokens": 42820226.0,
      "step": 21895
    },
    {
      "entropy": 6.047881698608398,
      "epoch": 2.189133803168891,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045233464095724706,
      "loss": 5.6396,
      "mean_token_accuracy": 0.1696893885731697,
      "num_tokens": 42829707.0,
      "step": 21900
    },
    {
      "entropy": 6.003972578048706,
      "epoch": 2.1896336282301196,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045231267286474755,
      "loss": 5.6088,
      "mean_token_accuracy": 0.17393270879983902,
      "num_tokens": 42840078.0,
      "step": 21905
    },
    {
      "entropy": 5.998701333999634,
      "epoch": 2.190133453291348,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004522907003111068,
      "loss": 5.5653,
      "mean_token_accuracy": 0.18817774802446366,
      "num_tokens": 42849998.0,
      "step": 21910
    },
    {
      "entropy": 5.931480932235718,
      "epoch": 2.1906332783525766,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004522687232968776,
      "loss": 5.5336,
      "mean_token_accuracy": 0.18959808796644212,
      "num_tokens": 42859290.0,
      "step": 21915
    },
    {
      "entropy": 5.9457040309906,
      "epoch": 2.1911331034138053,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045224674182261306,
      "loss": 5.5925,
      "mean_token_accuracy": 0.17998373359441758,
      "num_tokens": 42867945.0,
      "step": 21920
    },
    {
      "entropy": 5.951509809494018,
      "epoch": 2.1916329284750335,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045222475588886597,
      "loss": 5.4761,
      "mean_token_accuracy": 0.1914937511086464,
      "num_tokens": 42877527.0,
      "step": 21925
    },
    {
      "entropy": 5.964667844772339,
      "epoch": 2.1921327535362622,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045220276549618973,
      "loss": 5.5835,
      "mean_token_accuracy": 0.1891569584608078,
      "num_tokens": 42887049.0,
      "step": 21930
    },
    {
      "entropy": 5.9612202644348145,
      "epoch": 2.192632578597491,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045218077064513744,
      "loss": 5.5476,
      "mean_token_accuracy": 0.17878181636333465,
      "num_tokens": 42896177.0,
      "step": 21935
    },
    {
      "entropy": 6.021487760543823,
      "epoch": 2.1931324036587196,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004521587713362626,
      "loss": 5.6606,
      "mean_token_accuracy": 0.1710110656917095,
      "num_tokens": 42906244.0,
      "step": 21940
    },
    {
      "entropy": 5.933267068862915,
      "epoch": 2.193632228719948,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004521367675701187,
      "loss": 5.5908,
      "mean_token_accuracy": 0.18198734670877456,
      "num_tokens": 42915776.0,
      "step": 21945
    },
    {
      "entropy": 6.033950138092041,
      "epoch": 2.1941320537811766,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004521147593472592,
      "loss": 5.5841,
      "mean_token_accuracy": 0.1811728447675705,
      "num_tokens": 42926411.0,
      "step": 21950
    },
    {
      "entropy": 5.927692651748657,
      "epoch": 2.1946318788424053,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045209274666823795,
      "loss": 5.4618,
      "mean_token_accuracy": 0.1883011430501938,
      "num_tokens": 42935540.0,
      "step": 21955
    },
    {
      "entropy": 5.922246026992798,
      "epoch": 2.1951317039036335,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004520707295336088,
      "loss": 5.5871,
      "mean_token_accuracy": 0.17577701658010483,
      "num_tokens": 42944796.0,
      "step": 21960
    },
    {
      "entropy": 6.0709463596344,
      "epoch": 2.1956315289648622,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045204870794392574,
      "loss": 5.7054,
      "mean_token_accuracy": 0.16893539428710938,
      "num_tokens": 42954141.0,
      "step": 21965
    },
    {
      "entropy": 6.086889266967773,
      "epoch": 2.196131354026091,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045202668189974263,
      "loss": 5.6844,
      "mean_token_accuracy": 0.17005401998758315,
      "num_tokens": 42963201.0,
      "step": 21970
    },
    {
      "entropy": 5.929008293151855,
      "epoch": 2.1966311790873196,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004520046514016138,
      "loss": 5.5492,
      "mean_token_accuracy": 0.18233297765254974,
      "num_tokens": 42973798.0,
      "step": 21975
    },
    {
      "entropy": 5.967147302627564,
      "epoch": 2.197131004148548,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045198261645009347,
      "loss": 5.7231,
      "mean_token_accuracy": 0.16750065386295318,
      "num_tokens": 42983587.0,
      "step": 21980
    },
    {
      "entropy": 6.010059356689453,
      "epoch": 2.1976308292097766,
      "grad_norm": 1.0625,
      "learning_rate": 0.000451960577045736,
      "loss": 5.688,
      "mean_token_accuracy": 0.17497552931308746,
      "num_tokens": 42994802.0,
      "step": 21985
    },
    {
      "entropy": 6.113627529144287,
      "epoch": 2.1981306542710053,
      "grad_norm": 1.0625,
      "learning_rate": 0.000451938533189096,
      "loss": 5.6175,
      "mean_token_accuracy": 0.17705109119415283,
      "num_tokens": 43003427.0,
      "step": 21990
    },
    {
      "entropy": 5.95150580406189,
      "epoch": 2.1986304793322335,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004519164848807281,
      "loss": 5.5568,
      "mean_token_accuracy": 0.18113564997911452,
      "num_tokens": 43013268.0,
      "step": 21995
    },
    {
      "entropy": 5.995624208450318,
      "epoch": 2.1991303043934622,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045189443212118686,
      "loss": 5.574,
      "mean_token_accuracy": 0.1793281093239784,
      "num_tokens": 43022389.0,
      "step": 22000
    },
    {
      "entropy": 6.048040723800659,
      "epoch": 2.199630129454691,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045187237491102727,
      "loss": 5.5888,
      "mean_token_accuracy": 0.1848485067486763,
      "num_tokens": 43033168.0,
      "step": 22005
    },
    {
      "entropy": 6.004336547851563,
      "epoch": 2.200129954515919,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045185031325080426,
      "loss": 5.6678,
      "mean_token_accuracy": 0.17353209257125854,
      "num_tokens": 43043002.0,
      "step": 22010
    },
    {
      "entropy": 6.022681856155396,
      "epoch": 2.200629779577148,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045182824714107284,
      "loss": 5.6332,
      "mean_token_accuracy": 0.18004322052001953,
      "num_tokens": 43053474.0,
      "step": 22015
    },
    {
      "entropy": 6.057219552993774,
      "epoch": 2.2011296046383766,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045180617658238827,
      "loss": 5.6161,
      "mean_token_accuracy": 0.18079748153686523,
      "num_tokens": 43063821.0,
      "step": 22020
    },
    {
      "entropy": 6.003571796417236,
      "epoch": 2.2016294296996053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004517841015753057,
      "loss": 5.6096,
      "mean_token_accuracy": 0.17822324335575104,
      "num_tokens": 43072981.0,
      "step": 22025
    },
    {
      "entropy": 5.949115133285522,
      "epoch": 2.2021292547608335,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004517620221203807,
      "loss": 5.584,
      "mean_token_accuracy": 0.18815023601055145,
      "num_tokens": 43082992.0,
      "step": 22030
    },
    {
      "entropy": 5.993915271759033,
      "epoch": 2.2026290798220622,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045173993821816873,
      "loss": 5.6572,
      "mean_token_accuracy": 0.17934244722127915,
      "num_tokens": 43092678.0,
      "step": 22035
    },
    {
      "entropy": 5.931911420822144,
      "epoch": 2.203128904883291,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004517178498692253,
      "loss": 5.5365,
      "mean_token_accuracy": 0.19188494682312013,
      "num_tokens": 43102672.0,
      "step": 22040
    },
    {
      "entropy": 5.982793855667114,
      "epoch": 2.203628729944519,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045169575707410637,
      "loss": 5.6244,
      "mean_token_accuracy": 0.17562653720378876,
      "num_tokens": 43112747.0,
      "step": 22045
    },
    {
      "entropy": 5.989478635787964,
      "epoch": 2.204128555005748,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045167365983336756,
      "loss": 5.5817,
      "mean_token_accuracy": 0.18105142414569855,
      "num_tokens": 43122370.0,
      "step": 22050
    },
    {
      "entropy": 6.000267553329468,
      "epoch": 2.2046283800669766,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045165155814756497,
      "loss": 5.5923,
      "mean_token_accuracy": 0.18007077574729918,
      "num_tokens": 43131538.0,
      "step": 22055
    },
    {
      "entropy": 6.122753238677978,
      "epoch": 2.2051282051282053,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045162945201725457,
      "loss": 5.7419,
      "mean_token_accuracy": 0.16827971637248992,
      "num_tokens": 43141669.0,
      "step": 22060
    },
    {
      "entropy": 6.047640752792359,
      "epoch": 2.2056280301894335,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004516073414429927,
      "loss": 5.7123,
      "mean_token_accuracy": 0.1742069661617279,
      "num_tokens": 43151456.0,
      "step": 22065
    },
    {
      "entropy": 5.9971167087554935,
      "epoch": 2.2061278552506622,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004515852264253356,
      "loss": 5.6116,
      "mean_token_accuracy": 0.1741713896393776,
      "num_tokens": 43160491.0,
      "step": 22070
    },
    {
      "entropy": 5.978020906448364,
      "epoch": 2.206627680311891,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045156310696483953,
      "loss": 5.596,
      "mean_token_accuracy": 0.17722519338130951,
      "num_tokens": 43170695.0,
      "step": 22075
    },
    {
      "entropy": 6.035321426391602,
      "epoch": 2.207127505373119,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004515409830620612,
      "loss": 5.7017,
      "mean_token_accuracy": 0.17375599443912507,
      "num_tokens": 43180542.0,
      "step": 22080
    },
    {
      "entropy": 5.939443922042846,
      "epoch": 2.207627330434348,
      "grad_norm": 1.375,
      "learning_rate": 0.0004515188547175571,
      "loss": 5.6297,
      "mean_token_accuracy": 0.1724090173840523,
      "num_tokens": 43190739.0,
      "step": 22085
    },
    {
      "entropy": 6.006565237045288,
      "epoch": 2.2081271554955766,
      "grad_norm": 1.109375,
      "learning_rate": 0.000451496721931884,
      "loss": 5.7236,
      "mean_token_accuracy": 0.1728265702724457,
      "num_tokens": 43201808.0,
      "step": 22090
    },
    {
      "entropy": 6.082464790344238,
      "epoch": 2.2086269805568053,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004514745847055989,
      "loss": 5.6142,
      "mean_token_accuracy": 0.17515232861042024,
      "num_tokens": 43211061.0,
      "step": 22095
    },
    {
      "entropy": 5.969518089294434,
      "epoch": 2.2091268056180335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004514524430392586,
      "loss": 5.4907,
      "mean_token_accuracy": 0.18856747895479203,
      "num_tokens": 43220333.0,
      "step": 22100
    },
    {
      "entropy": 5.873654842376709,
      "epoch": 2.2096266306792622,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004514302969334202,
      "loss": 5.5835,
      "mean_token_accuracy": 0.18415372520685197,
      "num_tokens": 43230134.0,
      "step": 22105
    },
    {
      "entropy": 5.972807836532593,
      "epoch": 2.210126455740491,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045140814638864105,
      "loss": 5.5799,
      "mean_token_accuracy": 0.18083077371120454,
      "num_tokens": 43239216.0,
      "step": 22110
    },
    {
      "entropy": 5.9959776401519775,
      "epoch": 2.210626280801719,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004513859914054782,
      "loss": 5.7103,
      "mean_token_accuracy": 0.16568433195352555,
      "num_tokens": 43249051.0,
      "step": 22115
    },
    {
      "entropy": 6.153115320205688,
      "epoch": 2.211126105862948,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045136383198448926,
      "loss": 5.7514,
      "mean_token_accuracy": 0.17331899106502532,
      "num_tokens": 43259507.0,
      "step": 22120
    },
    {
      "entropy": 5.993607044219971,
      "epoch": 2.2116259309241766,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004513416681262317,
      "loss": 5.6858,
      "mean_token_accuracy": 0.17118339836597443,
      "num_tokens": 43269097.0,
      "step": 22125
    },
    {
      "entropy": 5.928920078277588,
      "epoch": 2.2121257559854053,
      "grad_norm": 1.125,
      "learning_rate": 0.0004513194998312631,
      "loss": 5.5804,
      "mean_token_accuracy": 0.17561299502849578,
      "num_tokens": 43278514.0,
      "step": 22130
    },
    {
      "entropy": 6.0273970603942875,
      "epoch": 2.2126255810466335,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004512973271001413,
      "loss": 5.7106,
      "mean_token_accuracy": 0.16925644427537917,
      "num_tokens": 43288563.0,
      "step": 22135
    },
    {
      "entropy": 6.0473387241363525,
      "epoch": 2.2131254061078622,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045127514993342413,
      "loss": 5.5545,
      "mean_token_accuracy": 0.1749691918492317,
      "num_tokens": 43298168.0,
      "step": 22140
    },
    {
      "entropy": 5.9695781707763675,
      "epoch": 2.213625231169091,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004512529683316695,
      "loss": 5.6183,
      "mean_token_accuracy": 0.17835736274719238,
      "num_tokens": 43308295.0,
      "step": 22145
    },
    {
      "entropy": 5.9713725566864015,
      "epoch": 2.214125056230319,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045123078229543557,
      "loss": 5.6504,
      "mean_token_accuracy": 0.1803991064429283,
      "num_tokens": 43319283.0,
      "step": 22150
    },
    {
      "entropy": 6.0329993724822994,
      "epoch": 2.214624881291548,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004512085918252805,
      "loss": 5.6749,
      "mean_token_accuracy": 0.1737919345498085,
      "num_tokens": 43329000.0,
      "step": 22155
    },
    {
      "entropy": 6.010866975784301,
      "epoch": 2.2151247063527766,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004511863969217625,
      "loss": 5.5186,
      "mean_token_accuracy": 0.18936234563589097,
      "num_tokens": 43338150.0,
      "step": 22160
    },
    {
      "entropy": 5.941356658935547,
      "epoch": 2.2156245314140053,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045116419758544024,
      "loss": 5.5686,
      "mean_token_accuracy": 0.18521998077630997,
      "num_tokens": 43348736.0,
      "step": 22165
    },
    {
      "entropy": 5.957248640060425,
      "epoch": 2.2161243564752335,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045114199381687207,
      "loss": 5.6019,
      "mean_token_accuracy": 0.1788535311818123,
      "num_tokens": 43359208.0,
      "step": 22170
    },
    {
      "entropy": 5.985256814956665,
      "epoch": 2.2166241815364622,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004511197856166166,
      "loss": 5.5742,
      "mean_token_accuracy": 0.18071734607219697,
      "num_tokens": 43368503.0,
      "step": 22175
    },
    {
      "entropy": 5.975113344192505,
      "epoch": 2.217124006597691,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004510975729852327,
      "loss": 5.6349,
      "mean_token_accuracy": 0.16904256641864776,
      "num_tokens": 43378778.0,
      "step": 22180
    },
    {
      "entropy": 6.031142854690552,
      "epoch": 2.217623831658919,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004510753559232792,
      "loss": 5.6428,
      "mean_token_accuracy": 0.1799350708723068,
      "num_tokens": 43389983.0,
      "step": 22185
    },
    {
      "entropy": 6.013162994384766,
      "epoch": 2.218123656720148,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045105313443131496,
      "loss": 5.5919,
      "mean_token_accuracy": 0.17897503674030305,
      "num_tokens": 43400008.0,
      "step": 22190
    },
    {
      "entropy": 6.046950101852417,
      "epoch": 2.2186234817813766,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004510309085098992,
      "loss": 5.6001,
      "mean_token_accuracy": 0.18039639741182328,
      "num_tokens": 43409626.0,
      "step": 22195
    },
    {
      "entropy": 6.064537715911865,
      "epoch": 2.2191233068426053,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045100867815959107,
      "loss": 5.6571,
      "mean_token_accuracy": 0.1758416473865509,
      "num_tokens": 43419277.0,
      "step": 22200
    },
    {
      "entropy": 5.930632400512695,
      "epoch": 2.2196231319038335,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004509864433809499,
      "loss": 5.6168,
      "mean_token_accuracy": 0.1784783884882927,
      "num_tokens": 43429226.0,
      "step": 22205
    },
    {
      "entropy": 6.074641942977905,
      "epoch": 2.2201229569650622,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004509642041745351,
      "loss": 5.6869,
      "mean_token_accuracy": 0.16964807063341142,
      "num_tokens": 43438921.0,
      "step": 22210
    },
    {
      "entropy": 5.990420532226563,
      "epoch": 2.220622782026291,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004509419605409062,
      "loss": 5.6297,
      "mean_token_accuracy": 0.18819321393966676,
      "num_tokens": 43449450.0,
      "step": 22215
    },
    {
      "entropy": 6.046097898483277,
      "epoch": 2.221122607087519,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004509197124806229,
      "loss": 5.6293,
      "mean_token_accuracy": 0.17801617234945297,
      "num_tokens": 43458587.0,
      "step": 22220
    },
    {
      "entropy": 5.997212934494018,
      "epoch": 2.221622432148748,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045089745999424486,
      "loss": 5.6205,
      "mean_token_accuracy": 0.18135042786598204,
      "num_tokens": 43467983.0,
      "step": 22225
    },
    {
      "entropy": 5.971654748916626,
      "epoch": 2.2221222572099766,
      "grad_norm": 1.171875,
      "learning_rate": 0.000450875203082332,
      "loss": 5.625,
      "mean_token_accuracy": 0.17071253955364227,
      "num_tokens": 43476887.0,
      "step": 22230
    },
    {
      "entropy": 6.00282998085022,
      "epoch": 2.2226220822712053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004508529417454442,
      "loss": 5.5806,
      "mean_token_accuracy": 0.17463336437940596,
      "num_tokens": 43485871.0,
      "step": 22235
    },
    {
      "entropy": 6.001150846481323,
      "epoch": 2.2231219073324335,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045083067598414176,
      "loss": 5.6653,
      "mean_token_accuracy": 0.1720937415957451,
      "num_tokens": 43496062.0,
      "step": 22240
    },
    {
      "entropy": 6.042199325561524,
      "epoch": 2.2236217323936622,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045080840579898466,
      "loss": 5.6821,
      "mean_token_accuracy": 0.17168441116809846,
      "num_tokens": 43505391.0,
      "step": 22245
    },
    {
      "entropy": 6.008621168136597,
      "epoch": 2.224121557454891,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004507861311905334,
      "loss": 5.646,
      "mean_token_accuracy": 0.17309713810682298,
      "num_tokens": 43515500.0,
      "step": 22250
    },
    {
      "entropy": 5.905796098709106,
      "epoch": 2.224621382516119,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004507638521593482,
      "loss": 5.5197,
      "mean_token_accuracy": 0.1861933007836342,
      "num_tokens": 43524493.0,
      "step": 22255
    },
    {
      "entropy": 6.039707660675049,
      "epoch": 2.225121207577348,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004507415687059898,
      "loss": 5.6371,
      "mean_token_accuracy": 0.17257660031318664,
      "num_tokens": 43534075.0,
      "step": 22260
    },
    {
      "entropy": 5.995427083969116,
      "epoch": 2.2256210326385766,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00045071928083101874,
      "loss": 5.6073,
      "mean_token_accuracy": 0.18103711307048798,
      "num_tokens": 43543288.0,
      "step": 22265
    },
    {
      "entropy": 5.981675720214843,
      "epoch": 2.2261208576998053,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004506969885349957,
      "loss": 5.6641,
      "mean_token_accuracy": 0.18353092223405837,
      "num_tokens": 43553491.0,
      "step": 22270
    },
    {
      "entropy": 6.013086891174316,
      "epoch": 2.2266206827610335,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004506746918184818,
      "loss": 5.6408,
      "mean_token_accuracy": 0.1814496710896492,
      "num_tokens": 43563904.0,
      "step": 22275
    },
    {
      "entropy": 6.008880853652954,
      "epoch": 2.2271205078222622,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004506523906820378,
      "loss": 5.6434,
      "mean_token_accuracy": 0.17236477136611938,
      "num_tokens": 43573231.0,
      "step": 22280
    },
    {
      "entropy": 6.0244526863098145,
      "epoch": 2.227620332883491,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004506300851262248,
      "loss": 5.6134,
      "mean_token_accuracy": 0.1743125930428505,
      "num_tokens": 43584538.0,
      "step": 22285
    },
    {
      "entropy": 6.005807304382325,
      "epoch": 2.228120157944719,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045060777515160404,
      "loss": 5.4847,
      "mean_token_accuracy": 0.19057592004537582,
      "num_tokens": 43594665.0,
      "step": 22290
    },
    {
      "entropy": 6.0461243152618405,
      "epoch": 2.228619983005948,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004505854607587369,
      "loss": 5.6533,
      "mean_token_accuracy": 0.17619028091430664,
      "num_tokens": 43604681.0,
      "step": 22295
    },
    {
      "entropy": 5.9372821807861325,
      "epoch": 2.2291198080671766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004505631419481847,
      "loss": 5.5708,
      "mean_token_accuracy": 0.18155483156442642,
      "num_tokens": 43614186.0,
      "step": 22300
    },
    {
      "entropy": 6.046668958663941,
      "epoch": 2.2296196331284053,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004505408187205091,
      "loss": 5.6408,
      "mean_token_accuracy": 0.17341457903385163,
      "num_tokens": 43624202.0,
      "step": 22305
    },
    {
      "entropy": 6.042264556884765,
      "epoch": 2.2301194581896335,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004505184910762716,
      "loss": 5.6591,
      "mean_token_accuracy": 0.17941745072603227,
      "num_tokens": 43634303.0,
      "step": 22310
    },
    {
      "entropy": 6.060351419448852,
      "epoch": 2.2306192832508622,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045049615901603406,
      "loss": 5.6762,
      "mean_token_accuracy": 0.17763068675994872,
      "num_tokens": 43644045.0,
      "step": 22315
    },
    {
      "entropy": 6.016443538665771,
      "epoch": 2.231119108312091,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004504738225403583,
      "loss": 5.6975,
      "mean_token_accuracy": 0.17584263533353806,
      "num_tokens": 43654220.0,
      "step": 22320
    },
    {
      "entropy": 6.017688226699829,
      "epoch": 2.231618933373319,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004504514816498064,
      "loss": 5.6453,
      "mean_token_accuracy": 0.17092597484588623,
      "num_tokens": 43665124.0,
      "step": 22325
    },
    {
      "entropy": 6.031825208663941,
      "epoch": 2.232118758434548,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045042913634494024,
      "loss": 5.5624,
      "mean_token_accuracy": 0.18077410906553268,
      "num_tokens": 43674029.0,
      "step": 22330
    },
    {
      "entropy": 5.976103973388672,
      "epoch": 2.2326185834957766,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004504067866263222,
      "loss": 5.6105,
      "mean_token_accuracy": 0.1768794387578964,
      "num_tokens": 43683019.0,
      "step": 22335
    },
    {
      "entropy": 5.937965154647827,
      "epoch": 2.2331184085570053,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004503844324945146,
      "loss": 5.5376,
      "mean_token_accuracy": 0.18145806342363358,
      "num_tokens": 43692762.0,
      "step": 22340
    },
    {
      "entropy": 5.942523765563965,
      "epoch": 2.2336182336182335,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004503620739500797,
      "loss": 5.5499,
      "mean_token_accuracy": 0.18547918647527695,
      "num_tokens": 43701899.0,
      "step": 22345
    },
    {
      "entropy": 5.982271385192871,
      "epoch": 2.2341180586794622,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004503397109935802,
      "loss": 5.5655,
      "mean_token_accuracy": 0.18105277717113494,
      "num_tokens": 43711233.0,
      "step": 22350
    },
    {
      "entropy": 5.835763549804687,
      "epoch": 2.234617883740691,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004503173436255787,
      "loss": 5.4396,
      "mean_token_accuracy": 0.19705995470285415,
      "num_tokens": 43721754.0,
      "step": 22355
    },
    {
      "entropy": 5.934717082977295,
      "epoch": 2.235117708801919,
      "grad_norm": 1.125,
      "learning_rate": 0.000450294971846638,
      "loss": 5.5728,
      "mean_token_accuracy": 0.18399944007396699,
      "num_tokens": 43731114.0,
      "step": 22360
    },
    {
      "entropy": 6.057019281387329,
      "epoch": 2.235617533863148,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004502725956573208,
      "loss": 5.6096,
      "mean_token_accuracy": 0.17836668491363525,
      "num_tokens": 43741011.0,
      "step": 22365
    },
    {
      "entropy": 6.043218946456909,
      "epoch": 2.2361173589243766,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004502502150581903,
      "loss": 5.6865,
      "mean_token_accuracy": 0.17576609551906586,
      "num_tokens": 43750075.0,
      "step": 22370
    },
    {
      "entropy": 5.933797216415405,
      "epoch": 2.2366171839856053,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004502278300498094,
      "loss": 5.6386,
      "mean_token_accuracy": 0.18043465912342072,
      "num_tokens": 43759826.0,
      "step": 22375
    },
    {
      "entropy": 6.003633642196656,
      "epoch": 2.2371170090468335,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045020544063274147,
      "loss": 5.5933,
      "mean_token_accuracy": 0.18045036047697066,
      "num_tokens": 43769326.0,
      "step": 22380
    },
    {
      "entropy": 5.992435455322266,
      "epoch": 2.2376168341080622,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045018304680754975,
      "loss": 5.5503,
      "mean_token_accuracy": 0.18044261783361434,
      "num_tokens": 43778549.0,
      "step": 22385
    },
    {
      "entropy": 6.037090253829956,
      "epoch": 2.238116659169291,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004501606485747975,
      "loss": 5.7342,
      "mean_token_accuracy": 0.16526909917593002,
      "num_tokens": 43789032.0,
      "step": 22390
    },
    {
      "entropy": 6.041986513137817,
      "epoch": 2.238616484230519,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004501382459350485,
      "loss": 5.7095,
      "mean_token_accuracy": 0.17710478305816652,
      "num_tokens": 43798246.0,
      "step": 22395
    },
    {
      "entropy": 6.0435583114624025,
      "epoch": 2.239116309291748,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004501158388888663,
      "loss": 5.6392,
      "mean_token_accuracy": 0.18287479877471924,
      "num_tokens": 43807572.0,
      "step": 22400
    },
    {
      "entropy": 5.942183923721314,
      "epoch": 2.2396161343529766,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004500934274368147,
      "loss": 5.6304,
      "mean_token_accuracy": 0.18047478348016738,
      "num_tokens": 43816912.0,
      "step": 22405
    },
    {
      "entropy": 5.929489707946777,
      "epoch": 2.240115959414205,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045007101157945747,
      "loss": 5.6101,
      "mean_token_accuracy": 0.17197883874177933,
      "num_tokens": 43826386.0,
      "step": 22410
    },
    {
      "entropy": 5.985610389709473,
      "epoch": 2.2406157844754335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045004859131735867,
      "loss": 5.6236,
      "mean_token_accuracy": 0.17410420775413513,
      "num_tokens": 43836456.0,
      "step": 22415
    },
    {
      "entropy": 5.9652478218078615,
      "epoch": 2.2411156095366622,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045002616665108233,
      "loss": 5.4861,
      "mean_token_accuracy": 0.1851659059524536,
      "num_tokens": 43846360.0,
      "step": 22420
    },
    {
      "entropy": 5.977870750427246,
      "epoch": 2.241615434597891,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004500037375811926,
      "loss": 5.603,
      "mean_token_accuracy": 0.18451430648565292,
      "num_tokens": 43855999.0,
      "step": 22425
    },
    {
      "entropy": 5.969978523254395,
      "epoch": 2.242115259659119,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044998130410825395,
      "loss": 5.5725,
      "mean_token_accuracy": 0.18193040937185287,
      "num_tokens": 43865497.0,
      "step": 22430
    },
    {
      "entropy": 5.93932523727417,
      "epoch": 2.242615084720348,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004499588662328307,
      "loss": 5.5699,
      "mean_token_accuracy": 0.17916802018880845,
      "num_tokens": 43874450.0,
      "step": 22435
    },
    {
      "entropy": 5.970940637588501,
      "epoch": 2.2431149097815766,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004499364239554874,
      "loss": 5.5492,
      "mean_token_accuracy": 0.18387677073478698,
      "num_tokens": 43885552.0,
      "step": 22440
    },
    {
      "entropy": 6.0680454730987545,
      "epoch": 2.243614734842805,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004499139772767887,
      "loss": 5.6738,
      "mean_token_accuracy": 0.1778697818517685,
      "num_tokens": 43894928.0,
      "step": 22445
    },
    {
      "entropy": 5.959530401229858,
      "epoch": 2.2441145599040335,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004498915261972993,
      "loss": 5.5186,
      "mean_token_accuracy": 0.18366504162549974,
      "num_tokens": 43904841.0,
      "step": 22450
    },
    {
      "entropy": 5.9617351531982425,
      "epoch": 2.244614384965262,
      "grad_norm": 1.125,
      "learning_rate": 0.00044986907071758417,
      "loss": 5.5317,
      "mean_token_accuracy": 0.19109016060829162,
      "num_tokens": 43914690.0,
      "step": 22455
    },
    {
      "entropy": 5.866055917739868,
      "epoch": 2.245114210026491,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004498466108382082,
      "loss": 5.5168,
      "mean_token_accuracy": 0.1868074968457222,
      "num_tokens": 43925237.0,
      "step": 22460
    },
    {
      "entropy": 6.026759624481201,
      "epoch": 2.245614035087719,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004498241465597365,
      "loss": 5.6597,
      "mean_token_accuracy": 0.1724162682890892,
      "num_tokens": 43934635.0,
      "step": 22465
    },
    {
      "entropy": 5.99723072052002,
      "epoch": 2.246113860148948,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004498016778827342,
      "loss": 5.6202,
      "mean_token_accuracy": 0.18356586247682571,
      "num_tokens": 43943705.0,
      "step": 22470
    },
    {
      "entropy": 5.9665180206298825,
      "epoch": 2.2466136852101766,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004497792048077667,
      "loss": 5.5752,
      "mean_token_accuracy": 0.18285235911607742,
      "num_tokens": 43953559.0,
      "step": 22475
    },
    {
      "entropy": 5.976256084442139,
      "epoch": 2.247113510271405,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004497567273353994,
      "loss": 5.6383,
      "mean_token_accuracy": 0.17584685385227203,
      "num_tokens": 43963940.0,
      "step": 22480
    },
    {
      "entropy": 5.9804661750793455,
      "epoch": 2.2476133353326335,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044973424546619774,
      "loss": 5.5199,
      "mean_token_accuracy": 0.18529729545116425,
      "num_tokens": 43973364.0,
      "step": 22485
    },
    {
      "entropy": 6.025920724868774,
      "epoch": 2.248113160393862,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004497117592007275,
      "loss": 5.5809,
      "mean_token_accuracy": 0.17461952567100525,
      "num_tokens": 43982531.0,
      "step": 22490
    },
    {
      "entropy": 6.064519453048706,
      "epoch": 2.248612985455091,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004496892685395543,
      "loss": 5.7482,
      "mean_token_accuracy": 0.16890592128038406,
      "num_tokens": 43994244.0,
      "step": 22495
    },
    {
      "entropy": 5.932012748718262,
      "epoch": 2.249112810516319,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004496667734832441,
      "loss": 5.5591,
      "mean_token_accuracy": 0.18038996160030366,
      "num_tokens": 44003809.0,
      "step": 22500
    },
    {
      "entropy": 6.0355969905853275,
      "epoch": 2.249612635577548,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044964427403236274,
      "loss": 5.6052,
      "mean_token_accuracy": 0.17239041328430177,
      "num_tokens": 44014850.0,
      "step": 22505
    },
    {
      "entropy": 5.96818265914917,
      "epoch": 2.2501124606387766,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004496217701874765,
      "loss": 5.4808,
      "mean_token_accuracy": 0.18882594257593155,
      "num_tokens": 44025594.0,
      "step": 22510
    },
    {
      "entropy": 6.0479820728302,
      "epoch": 2.250612285700005,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004495992619491514,
      "loss": 5.714,
      "mean_token_accuracy": 0.1735781580209732,
      "num_tokens": 44034681.0,
      "step": 22515
    },
    {
      "entropy": 6.020373010635376,
      "epoch": 2.2511121107612335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004495767493179538,
      "loss": 5.5777,
      "mean_token_accuracy": 0.18004503101110458,
      "num_tokens": 44044149.0,
      "step": 22520
    },
    {
      "entropy": 5.99486813545227,
      "epoch": 2.251611935822462,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004495542322944501,
      "loss": 5.637,
      "mean_token_accuracy": 0.17005500048398972,
      "num_tokens": 44054238.0,
      "step": 22525
    },
    {
      "entropy": 5.9627938747406,
      "epoch": 2.2521117608836905,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044953171087920693,
      "loss": 5.6923,
      "mean_token_accuracy": 0.1695142611861229,
      "num_tokens": 44063826.0,
      "step": 22530
    },
    {
      "entropy": 6.100530576705933,
      "epoch": 2.252611585944919,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004495091850727907,
      "loss": 5.6848,
      "mean_token_accuracy": 0.16915721893310548,
      "num_tokens": 44074588.0,
      "step": 22535
    },
    {
      "entropy": 5.971357202529907,
      "epoch": 2.253111411006148,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044948665487576833,
      "loss": 5.6061,
      "mean_token_accuracy": 0.17801449000835418,
      "num_tokens": 44084737.0,
      "step": 22540
    },
    {
      "entropy": 6.038979005813599,
      "epoch": 2.2536112360673766,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044946412028870657,
      "loss": 5.5738,
      "mean_token_accuracy": 0.17808584421873092,
      "num_tokens": 44093542.0,
      "step": 22545
    },
    {
      "entropy": 6.063486957550049,
      "epoch": 2.254111061128605,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004494415813121725,
      "loss": 5.6196,
      "mean_token_accuracy": 0.17143776565790175,
      "num_tokens": 44104681.0,
      "step": 22550
    },
    {
      "entropy": 6.0833230972290036,
      "epoch": 2.2546108861898335,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004494190379467331,
      "loss": 5.798,
      "mean_token_accuracy": 0.17070560157299042,
      "num_tokens": 44116567.0,
      "step": 22555
    },
    {
      "entropy": 5.9798112392425535,
      "epoch": 2.255110711251062,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044939649019295564,
      "loss": 5.6021,
      "mean_token_accuracy": 0.1784723997116089,
      "num_tokens": 44125176.0,
      "step": 22560
    },
    {
      "entropy": 6.011581945419311,
      "epoch": 2.2556105363122905,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004493739380514073,
      "loss": 5.6314,
      "mean_token_accuracy": 0.17475399672985076,
      "num_tokens": 44136371.0,
      "step": 22565
    },
    {
      "entropy": 6.011478519439697,
      "epoch": 2.256110361373519,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004493513815226556,
      "loss": 5.546,
      "mean_token_accuracy": 0.19088552743196488,
      "num_tokens": 44146456.0,
      "step": 22570
    },
    {
      "entropy": 5.937900400161743,
      "epoch": 2.256610186434748,
      "grad_norm": 1.140625,
      "learning_rate": 0.000449328820607268,
      "loss": 5.5806,
      "mean_token_accuracy": 0.18200667351484298,
      "num_tokens": 44155719.0,
      "step": 22575
    },
    {
      "entropy": 5.905630588531494,
      "epoch": 2.2571100114959766,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004493062553058121,
      "loss": 5.5823,
      "mean_token_accuracy": 0.18962600976228713,
      "num_tokens": 44165994.0,
      "step": 22580
    },
    {
      "entropy": 6.004604959487915,
      "epoch": 2.257609836557205,
      "grad_norm": 1.375,
      "learning_rate": 0.00044928368561885566,
      "loss": 5.7149,
      "mean_token_accuracy": 0.17209159284830094,
      "num_tokens": 44175726.0,
      "step": 22585
    },
    {
      "entropy": 6.095354461669922,
      "epoch": 2.2581096616184335,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044926111154696655,
      "loss": 5.6424,
      "mean_token_accuracy": 0.17910559326410294,
      "num_tokens": 44185824.0,
      "step": 22590
    },
    {
      "entropy": 6.0126558303833,
      "epoch": 2.258609486679662,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004492385330907127,
      "loss": 5.5644,
      "mean_token_accuracy": 0.183484748005867,
      "num_tokens": 44195817.0,
      "step": 22595
    },
    {
      "entropy": 5.966447877883911,
      "epoch": 2.2591093117408905,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004492159502506622,
      "loss": 5.5713,
      "mean_token_accuracy": 0.18423541635274887,
      "num_tokens": 44205885.0,
      "step": 22600
    },
    {
      "entropy": 6.064156341552734,
      "epoch": 2.259609136802119,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044919336302738324,
      "loss": 5.6288,
      "mean_token_accuracy": 0.1778467610478401,
      "num_tokens": 44215472.0,
      "step": 22605
    },
    {
      "entropy": 6.046846342086792,
      "epoch": 2.260108961863348,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044917077142144407,
      "loss": 5.6733,
      "mean_token_accuracy": 0.17611514627933503,
      "num_tokens": 44225833.0,
      "step": 22610
    },
    {
      "entropy": 5.986122083663941,
      "epoch": 2.2606087869245766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044914817543341315,
      "loss": 5.5606,
      "mean_token_accuracy": 0.186021026968956,
      "num_tokens": 44236252.0,
      "step": 22615
    },
    {
      "entropy": 6.026542377471924,
      "epoch": 2.261108611985805,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004491255750638589,
      "loss": 5.6452,
      "mean_token_accuracy": 0.17752775102853774,
      "num_tokens": 44245387.0,
      "step": 22620
    },
    {
      "entropy": 6.056422424316406,
      "epoch": 2.2616084370470335,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044910297031335006,
      "loss": 5.7041,
      "mean_token_accuracy": 0.17400155514478682,
      "num_tokens": 44254578.0,
      "step": 22625
    },
    {
      "entropy": 5.960979795455932,
      "epoch": 2.262108262108262,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044908036118245523,
      "loss": 5.5129,
      "mean_token_accuracy": 0.18405826687812804,
      "num_tokens": 44263042.0,
      "step": 22630
    },
    {
      "entropy": 6.053934144973755,
      "epoch": 2.2626080871694905,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004490577476717433,
      "loss": 5.6403,
      "mean_token_accuracy": 0.17049760296940802,
      "num_tokens": 44272068.0,
      "step": 22635
    },
    {
      "entropy": 5.962977123260498,
      "epoch": 2.263107912230719,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004490351297817832,
      "loss": 5.5511,
      "mean_token_accuracy": 0.18766342401504515,
      "num_tokens": 44281829.0,
      "step": 22640
    },
    {
      "entropy": 5.982432794570923,
      "epoch": 2.263607737291948,
      "grad_norm": 1.203125,
      "learning_rate": 0.000449012507513144,
      "loss": 5.5789,
      "mean_token_accuracy": 0.17788737565279006,
      "num_tokens": 44290097.0,
      "step": 22645
    },
    {
      "entropy": 5.9863873481750485,
      "epoch": 2.2641075623531766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044898988086639497,
      "loss": 5.5969,
      "mean_token_accuracy": 0.18069355487823485,
      "num_tokens": 44300611.0,
      "step": 22650
    },
    {
      "entropy": 6.015520143508911,
      "epoch": 2.264607387414405,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004489672498421053,
      "loss": 5.6163,
      "mean_token_accuracy": 0.1802450180053711,
      "num_tokens": 44310324.0,
      "step": 22655
    },
    {
      "entropy": 5.864447593688965,
      "epoch": 2.2651072124756335,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004489446144408444,
      "loss": 5.4355,
      "mean_token_accuracy": 0.1980360895395279,
      "num_tokens": 44319750.0,
      "step": 22660
    },
    {
      "entropy": 5.990536642074585,
      "epoch": 2.265607037536862,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004489219746631817,
      "loss": 5.6369,
      "mean_token_accuracy": 0.17260272204875945,
      "num_tokens": 44329092.0,
      "step": 22665
    },
    {
      "entropy": 5.936722946166992,
      "epoch": 2.2661068625980905,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044889933050968695,
      "loss": 5.5164,
      "mean_token_accuracy": 0.18702349811792374,
      "num_tokens": 44337608.0,
      "step": 22670
    },
    {
      "entropy": 6.001026058197022,
      "epoch": 2.266606687659319,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044887668198092975,
      "loss": 5.6605,
      "mean_token_accuracy": 0.17853147834539412,
      "num_tokens": 44347600.0,
      "step": 22675
    },
    {
      "entropy": 5.917940092086792,
      "epoch": 2.267106512720548,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044885402907748,
      "loss": 5.616,
      "mean_token_accuracy": 0.17879174947738646,
      "num_tokens": 44358187.0,
      "step": 22680
    },
    {
      "entropy": 6.059200382232666,
      "epoch": 2.2676063377817766,
      "grad_norm": 1.546875,
      "learning_rate": 0.00044883137179990755,
      "loss": 5.5964,
      "mean_token_accuracy": 0.18628198951482772,
      "num_tokens": 44367563.0,
      "step": 22685
    },
    {
      "entropy": 6.098533344268799,
      "epoch": 2.268106162843005,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044880871014878253,
      "loss": 5.6551,
      "mean_token_accuracy": 0.16972524374723436,
      "num_tokens": 44378123.0,
      "step": 22690
    },
    {
      "entropy": 5.94244909286499,
      "epoch": 2.2686059879042335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044878604412467517,
      "loss": 5.6472,
      "mean_token_accuracy": 0.17677621841430663,
      "num_tokens": 44387753.0,
      "step": 22695
    },
    {
      "entropy": 5.980865621566773,
      "epoch": 2.269105812965462,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044876337372815555,
      "loss": 5.5589,
      "mean_token_accuracy": 0.18207692801952363,
      "num_tokens": 44397917.0,
      "step": 22700
    },
    {
      "entropy": 5.990543413162231,
      "epoch": 2.2696056380266905,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044874069895979417,
      "loss": 5.6138,
      "mean_token_accuracy": 0.1883714333176613,
      "num_tokens": 44407135.0,
      "step": 22705
    },
    {
      "entropy": 6.010629892349243,
      "epoch": 2.270105463087919,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004487180198201616,
      "loss": 5.6636,
      "mean_token_accuracy": 0.1697762444615364,
      "num_tokens": 44416814.0,
      "step": 22710
    },
    {
      "entropy": 6.067760944366455,
      "epoch": 2.270605288149148,
      "grad_norm": 1.125,
      "learning_rate": 0.00044869533630982833,
      "loss": 5.7665,
      "mean_token_accuracy": 0.16152915954589844,
      "num_tokens": 44427304.0,
      "step": 22715
    },
    {
      "entropy": 6.018003225326538,
      "epoch": 2.2711051132103766,
      "grad_norm": 1.15625,
      "learning_rate": 0.000448672648429365,
      "loss": 5.55,
      "mean_token_accuracy": 0.18414865732192992,
      "num_tokens": 44436813.0,
      "step": 22720
    },
    {
      "entropy": 5.918941259384155,
      "epoch": 2.271604938271605,
      "grad_norm": 1.125,
      "learning_rate": 0.00044864995617934265,
      "loss": 5.5367,
      "mean_token_accuracy": 0.18631697446107864,
      "num_tokens": 44446603.0,
      "step": 22725
    },
    {
      "entropy": 5.979697895050049,
      "epoch": 2.2721047633328335,
      "grad_norm": 1.125,
      "learning_rate": 0.000448627259560332,
      "loss": 5.6415,
      "mean_token_accuracy": 0.1789831817150116,
      "num_tokens": 44456016.0,
      "step": 22730
    },
    {
      "entropy": 6.042467737197876,
      "epoch": 2.272604588394062,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044860455857290415,
      "loss": 5.6084,
      "mean_token_accuracy": 0.18131624460220336,
      "num_tokens": 44465384.0,
      "step": 22735
    },
    {
      "entropy": 5.971633195877075,
      "epoch": 2.2731044134552905,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004485818532176304,
      "loss": 5.6333,
      "mean_token_accuracy": 0.17415223717689515,
      "num_tokens": 44475709.0,
      "step": 22740
    },
    {
      "entropy": 5.998405885696411,
      "epoch": 2.273604238516519,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004485591434950818,
      "loss": 5.6729,
      "mean_token_accuracy": 0.17478574216365814,
      "num_tokens": 44485602.0,
      "step": 22745
    },
    {
      "entropy": 5.993406534194946,
      "epoch": 2.274104063577748,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004485364294058298,
      "loss": 5.6336,
      "mean_token_accuracy": 0.18092431426048278,
      "num_tokens": 44495121.0,
      "step": 22750
    },
    {
      "entropy": 6.0408977508544925,
      "epoch": 2.2746038886389766,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044851371095044597,
      "loss": 5.5446,
      "mean_token_accuracy": 0.17782264351844787,
      "num_tokens": 44504336.0,
      "step": 22755
    },
    {
      "entropy": 6.060680913925171,
      "epoch": 2.275103713700205,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044849098812950174,
      "loss": 5.6494,
      "mean_token_accuracy": 0.17725302278995514,
      "num_tokens": 44514076.0,
      "step": 22760
    },
    {
      "entropy": 5.995963668823242,
      "epoch": 2.2756035387614335,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000448468260943569,
      "loss": 5.7131,
      "mean_token_accuracy": 0.17287178635597228,
      "num_tokens": 44523659.0,
      "step": 22765
    },
    {
      "entropy": 5.962305498123169,
      "epoch": 2.276103363822662,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044844552939321937,
      "loss": 5.5332,
      "mean_token_accuracy": 0.19011068865656852,
      "num_tokens": 44534024.0,
      "step": 22770
    },
    {
      "entropy": 5.98012056350708,
      "epoch": 2.2766031888838905,
      "grad_norm": 1.125,
      "learning_rate": 0.0004484227934790248,
      "loss": 5.5175,
      "mean_token_accuracy": 0.18421839475631713,
      "num_tokens": 44543206.0,
      "step": 22775
    },
    {
      "entropy": 6.070514154434204,
      "epoch": 2.277103013945119,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004484000532015575,
      "loss": 5.7264,
      "mean_token_accuracy": 0.17339643985033035,
      "num_tokens": 44553670.0,
      "step": 22780
    },
    {
      "entropy": 5.9481690406799315,
      "epoch": 2.277602839006348,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044837730856138935,
      "loss": 5.6636,
      "mean_token_accuracy": 0.17227899879217148,
      "num_tokens": 44564040.0,
      "step": 22785
    },
    {
      "entropy": 5.936240863800049,
      "epoch": 2.2781026640675766,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004483545595590928,
      "loss": 5.5156,
      "mean_token_accuracy": 0.18823879063129426,
      "num_tokens": 44573605.0,
      "step": 22790
    },
    {
      "entropy": 6.009485244750977,
      "epoch": 2.278602489128805,
      "grad_norm": 1.125,
      "learning_rate": 0.0004483318061952401,
      "loss": 5.5642,
      "mean_token_accuracy": 0.17571812123060226,
      "num_tokens": 44584298.0,
      "step": 22795
    },
    {
      "entropy": 6.057311201095581,
      "epoch": 2.2791023141900335,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004483090484704038,
      "loss": 5.6939,
      "mean_token_accuracy": 0.17137784361839295,
      "num_tokens": 44594585.0,
      "step": 22800
    },
    {
      "entropy": 5.924565219879151,
      "epoch": 2.279602139251262,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044828628638515647,
      "loss": 5.4774,
      "mean_token_accuracy": 0.19211678802967072,
      "num_tokens": 44604788.0,
      "step": 22805
    },
    {
      "entropy": 5.974728012084961,
      "epoch": 2.2801019643124905,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004482635199400707,
      "loss": 5.6261,
      "mean_token_accuracy": 0.1780260518193245,
      "num_tokens": 44614259.0,
      "step": 22810
    },
    {
      "entropy": 5.9531004428863525,
      "epoch": 2.280601789373719,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044824074913571936,
      "loss": 5.5655,
      "mean_token_accuracy": 0.18519069254398346,
      "num_tokens": 44624827.0,
      "step": 22815
    },
    {
      "entropy": 5.939093351364136,
      "epoch": 2.281101614434948,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044821797397267535,
      "loss": 5.5167,
      "mean_token_accuracy": 0.1833449900150299,
      "num_tokens": 44635741.0,
      "step": 22820
    },
    {
      "entropy": 6.038282155990601,
      "epoch": 2.2816014394961766,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004481951944515117,
      "loss": 5.6941,
      "mean_token_accuracy": 0.17689169943332672,
      "num_tokens": 44644935.0,
      "step": 22825
    },
    {
      "entropy": 5.846624183654785,
      "epoch": 2.282101264557405,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00044817241057280155,
      "loss": 5.4728,
      "mean_token_accuracy": 0.19011913239955902,
      "num_tokens": 44654265.0,
      "step": 22830
    },
    {
      "entropy": 6.057186317443848,
      "epoch": 2.2826010896186335,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004481496223371181,
      "loss": 5.586,
      "mean_token_accuracy": 0.17747249752283095,
      "num_tokens": 44664461.0,
      "step": 22835
    },
    {
      "entropy": 5.946252918243408,
      "epoch": 2.283100914679862,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044812682974503464,
      "loss": 5.5228,
      "mean_token_accuracy": 0.1812433585524559,
      "num_tokens": 44673774.0,
      "step": 22840
    },
    {
      "entropy": 5.987095880508423,
      "epoch": 2.2836007397410905,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004481040327971248,
      "loss": 5.6185,
      "mean_token_accuracy": 0.18233342468738556,
      "num_tokens": 44683061.0,
      "step": 22845
    },
    {
      "entropy": 6.011546325683594,
      "epoch": 2.284100564802319,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044808123149396193,
      "loss": 5.6125,
      "mean_token_accuracy": 0.17484525591135025,
      "num_tokens": 44691874.0,
      "step": 22850
    },
    {
      "entropy": 6.02160062789917,
      "epoch": 2.284600389863548,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044805842583611993,
      "loss": 5.6609,
      "mean_token_accuracy": 0.17628844529390336,
      "num_tokens": 44701796.0,
      "step": 22855
    },
    {
      "entropy": 6.054059982299805,
      "epoch": 2.2851002149247766,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044803561582417237,
      "loss": 5.744,
      "mean_token_accuracy": 0.16702216267585754,
      "num_tokens": 44711749.0,
      "step": 22860
    },
    {
      "entropy": 5.962600374221802,
      "epoch": 2.285600039986005,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044801280145869327,
      "loss": 5.5344,
      "mean_token_accuracy": 0.18523398786783218,
      "num_tokens": 44721507.0,
      "step": 22865
    },
    {
      "entropy": 5.99410605430603,
      "epoch": 2.2860998650472335,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004479899827402566,
      "loss": 5.6776,
      "mean_token_accuracy": 0.17056447565555571,
      "num_tokens": 44731178.0,
      "step": 22870
    },
    {
      "entropy": 6.010069847106934,
      "epoch": 2.286599690108462,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044796715966943654,
      "loss": 5.5826,
      "mean_token_accuracy": 0.17768374383449553,
      "num_tokens": 44741385.0,
      "step": 22875
    },
    {
      "entropy": 6.062394285202027,
      "epoch": 2.2870995151696905,
      "grad_norm": 1.125,
      "learning_rate": 0.0004479443322468072,
      "loss": 5.6429,
      "mean_token_accuracy": 0.17363049685955048,
      "num_tokens": 44751316.0,
      "step": 22880
    },
    {
      "entropy": 5.948034858703613,
      "epoch": 2.287599340230919,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004479215004729429,
      "loss": 5.614,
      "mean_token_accuracy": 0.1779513791203499,
      "num_tokens": 44760786.0,
      "step": 22885
    },
    {
      "entropy": 5.948672485351563,
      "epoch": 2.288099165292148,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004478986643484183,
      "loss": 5.6329,
      "mean_token_accuracy": 0.1729804828763008,
      "num_tokens": 44771203.0,
      "step": 22890
    },
    {
      "entropy": 6.035411071777344,
      "epoch": 2.2885989903533765,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044787582387380764,
      "loss": 5.6793,
      "mean_token_accuracy": 0.17101696580648423,
      "num_tokens": 44780739.0,
      "step": 22895
    },
    {
      "entropy": 6.0667853355407715,
      "epoch": 2.289098815414605,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004478529790496858,
      "loss": 5.6318,
      "mean_token_accuracy": 0.18048324137926103,
      "num_tokens": 44790534.0,
      "step": 22900
    },
    {
      "entropy": 5.942134094238281,
      "epoch": 2.2895986404758335,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00044783012987662755,
      "loss": 5.6432,
      "mean_token_accuracy": 0.18118431717157363,
      "num_tokens": 44799767.0,
      "step": 22905
    },
    {
      "entropy": 5.995871353149414,
      "epoch": 2.290098465537062,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004478072763552076,
      "loss": 5.6702,
      "mean_token_accuracy": 0.1684749886393547,
      "num_tokens": 44810431.0,
      "step": 22910
    },
    {
      "entropy": 5.969799041748047,
      "epoch": 2.2905982905982905,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004477844184860011,
      "loss": 5.5549,
      "mean_token_accuracy": 0.1843617394566536,
      "num_tokens": 44820605.0,
      "step": 22915
    },
    {
      "entropy": 6.054704427719116,
      "epoch": 2.291098115659519,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044776155626958305,
      "loss": 5.6685,
      "mean_token_accuracy": 0.1725287541747093,
      "num_tokens": 44829895.0,
      "step": 22920
    },
    {
      "entropy": 5.990314197540283,
      "epoch": 2.291597940720748,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044773868970652875,
      "loss": 5.659,
      "mean_token_accuracy": 0.17674512267112732,
      "num_tokens": 44839162.0,
      "step": 22925
    },
    {
      "entropy": 5.975676536560059,
      "epoch": 2.2920977657819765,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044771581879741334,
      "loss": 5.4411,
      "mean_token_accuracy": 0.18865386992692948,
      "num_tokens": 44849395.0,
      "step": 22930
    },
    {
      "entropy": 5.966769218444824,
      "epoch": 2.292597590843205,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044769294354281253,
      "loss": 5.5664,
      "mean_token_accuracy": 0.1858442410826683,
      "num_tokens": 44858695.0,
      "step": 22935
    },
    {
      "entropy": 5.996801710128784,
      "epoch": 2.2930974159044335,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044767006394330166,
      "loss": 5.6891,
      "mean_token_accuracy": 0.17484198957681657,
      "num_tokens": 44869447.0,
      "step": 22940
    },
    {
      "entropy": 5.856332206726075,
      "epoch": 2.293597240965662,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044764717999945644,
      "loss": 5.4463,
      "mean_token_accuracy": 0.19036267399787904,
      "num_tokens": 44878155.0,
      "step": 22945
    },
    {
      "entropy": 5.966739845275879,
      "epoch": 2.2940970660268905,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044762429171185246,
      "loss": 5.638,
      "mean_token_accuracy": 0.1758459433913231,
      "num_tokens": 44888465.0,
      "step": 22950
    },
    {
      "entropy": 5.986728382110596,
      "epoch": 2.294596891088119,
      "grad_norm": 1.125,
      "learning_rate": 0.00044760139908106583,
      "loss": 5.5869,
      "mean_token_accuracy": 0.17878851145505906,
      "num_tokens": 44898915.0,
      "step": 22955
    },
    {
      "entropy": 6.0160737991333,
      "epoch": 2.295096716149348,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004475785021076724,
      "loss": 5.5387,
      "mean_token_accuracy": 0.1800871640443802,
      "num_tokens": 44908543.0,
      "step": 22960
    },
    {
      "entropy": 5.95029730796814,
      "epoch": 2.2955965412105765,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044755560079224834,
      "loss": 5.5624,
      "mean_token_accuracy": 0.18831642270088195,
      "num_tokens": 44918429.0,
      "step": 22965
    },
    {
      "entropy": 6.00732011795044,
      "epoch": 2.296096366271805,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004475326951353697,
      "loss": 5.5577,
      "mean_token_accuracy": 0.18171066790819168,
      "num_tokens": 44927479.0,
      "step": 22970
    },
    {
      "entropy": 5.897715425491333,
      "epoch": 2.2965961913330335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044750978513761276,
      "loss": 5.5157,
      "mean_token_accuracy": 0.18681750893592836,
      "num_tokens": 44937317.0,
      "step": 22975
    },
    {
      "entropy": 6.039028024673462,
      "epoch": 2.297096016394262,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004474868707995541,
      "loss": 5.7149,
      "mean_token_accuracy": 0.17330888509750367,
      "num_tokens": 44947587.0,
      "step": 22980
    },
    {
      "entropy": 5.954914188385009,
      "epoch": 2.2975958414554905,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044746395212177015,
      "loss": 5.5078,
      "mean_token_accuracy": 0.19833067655563355,
      "num_tokens": 44956850.0,
      "step": 22985
    },
    {
      "entropy": 6.006265497207641,
      "epoch": 2.298095666516719,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044744102910483755,
      "loss": 5.6143,
      "mean_token_accuracy": 0.17245338708162308,
      "num_tokens": 44966310.0,
      "step": 22990
    },
    {
      "entropy": 5.992692613601685,
      "epoch": 2.298595491577948,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000447418101749333,
      "loss": 5.6398,
      "mean_token_accuracy": 0.17812255322933196,
      "num_tokens": 44975361.0,
      "step": 22995
    },
    {
      "entropy": 5.996175003051758,
      "epoch": 2.299095316639176,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004473951700558334,
      "loss": 5.6709,
      "mean_token_accuracy": 0.16987384706735612,
      "num_tokens": 44985420.0,
      "step": 23000
    },
    {
      "entropy": 5.973693370819092,
      "epoch": 2.299595141700405,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044737223402491574,
      "loss": 5.5902,
      "mean_token_accuracy": 0.18152419179677964,
      "num_tokens": 44995622.0,
      "step": 23005
    },
    {
      "entropy": 6.021467113494873,
      "epoch": 2.3000949667616335,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044734929365715685,
      "loss": 5.6393,
      "mean_token_accuracy": 0.18286869972944259,
      "num_tokens": 45006017.0,
      "step": 23010
    },
    {
      "entropy": 6.018559694290161,
      "epoch": 2.3005947918228618,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044732634895313424,
      "loss": 5.5973,
      "mean_token_accuracy": 0.17395117431879042,
      "num_tokens": 45017024.0,
      "step": 23015
    },
    {
      "entropy": 6.008487224578857,
      "epoch": 2.3010946168840904,
      "grad_norm": 1.109375,
      "learning_rate": 0.000447303399913425,
      "loss": 5.5961,
      "mean_token_accuracy": 0.1811240240931511,
      "num_tokens": 45026384.0,
      "step": 23020
    },
    {
      "entropy": 5.8753503322601315,
      "epoch": 2.301594441945319,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044728044653860653,
      "loss": 5.4784,
      "mean_token_accuracy": 0.19479072391986846,
      "num_tokens": 45037144.0,
      "step": 23025
    },
    {
      "entropy": 5.9388933181762695,
      "epoch": 2.302094267006548,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004472574888292564,
      "loss": 5.6464,
      "mean_token_accuracy": 0.17518241852521896,
      "num_tokens": 45048272.0,
      "step": 23030
    },
    {
      "entropy": 6.0299335479736325,
      "epoch": 2.302594092067776,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004472345267859521,
      "loss": 5.6224,
      "mean_token_accuracy": 0.1794913187623024,
      "num_tokens": 45057799.0,
      "step": 23035
    },
    {
      "entropy": 5.980448532104492,
      "epoch": 2.303093917129005,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004472115604092715,
      "loss": 5.629,
      "mean_token_accuracy": 0.17743305265903472,
      "num_tokens": 45066786.0,
      "step": 23040
    },
    {
      "entropy": 5.977276420593261,
      "epoch": 2.3035937421902335,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044718858969979236,
      "loss": 5.642,
      "mean_token_accuracy": 0.1782527431845665,
      "num_tokens": 45075425.0,
      "step": 23045
    },
    {
      "entropy": 6.066010046005249,
      "epoch": 2.3040935672514617,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044716561465809256,
      "loss": 5.7081,
      "mean_token_accuracy": 0.17693153619766236,
      "num_tokens": 45085470.0,
      "step": 23050
    },
    {
      "entropy": 6.133647775650024,
      "epoch": 2.3045933923126904,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044714263528475024,
      "loss": 5.8587,
      "mean_token_accuracy": 0.16292986869812012,
      "num_tokens": 45096093.0,
      "step": 23055
    },
    {
      "entropy": 6.057939338684082,
      "epoch": 2.305093217373919,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044711965158034345,
      "loss": 5.6647,
      "mean_token_accuracy": 0.17594714760780333,
      "num_tokens": 45106031.0,
      "step": 23060
    },
    {
      "entropy": 6.029297542572022,
      "epoch": 2.305593042435148,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004470966635454506,
      "loss": 5.5699,
      "mean_token_accuracy": 0.18769281953573227,
      "num_tokens": 45116211.0,
      "step": 23065
    },
    {
      "entropy": 5.958017253875733,
      "epoch": 2.306092867496376,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004470736711806499,
      "loss": 5.6469,
      "mean_token_accuracy": 0.1753700539469719,
      "num_tokens": 45126232.0,
      "step": 23070
    },
    {
      "entropy": 5.999030828475952,
      "epoch": 2.306592692557605,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044705067448651996,
      "loss": 5.5129,
      "mean_token_accuracy": 0.18200110495090485,
      "num_tokens": 45136388.0,
      "step": 23075
    },
    {
      "entropy": 5.947226858139038,
      "epoch": 2.3070925176188335,
      "grad_norm": 1.125,
      "learning_rate": 0.0004470276734636393,
      "loss": 5.6249,
      "mean_token_accuracy": 0.18490759432315826,
      "num_tokens": 45145569.0,
      "step": 23080
    },
    {
      "entropy": 5.887145853042602,
      "epoch": 2.3075923426800617,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044700466811258666,
      "loss": 5.5729,
      "mean_token_accuracy": 0.1820469543337822,
      "num_tokens": 45155172.0,
      "step": 23085
    },
    {
      "entropy": 5.941488361358642,
      "epoch": 2.3080921677412904,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004469816584339408,
      "loss": 5.5015,
      "mean_token_accuracy": 0.19054541140794753,
      "num_tokens": 45164537.0,
      "step": 23090
    },
    {
      "entropy": 6.0449083805084225,
      "epoch": 2.308591992802519,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044695864442828066,
      "loss": 5.6815,
      "mean_token_accuracy": 0.17206234782934188,
      "num_tokens": 45174031.0,
      "step": 23095
    },
    {
      "entropy": 5.930567836761474,
      "epoch": 2.309091817863748,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044693562609618534,
      "loss": 5.5613,
      "mean_token_accuracy": 0.17572588175535203,
      "num_tokens": 45183211.0,
      "step": 23100
    },
    {
      "entropy": 6.010422325134277,
      "epoch": 2.309591642924976,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004469126034382339,
      "loss": 5.6595,
      "mean_token_accuracy": 0.17383021265268325,
      "num_tokens": 45193619.0,
      "step": 23105
    },
    {
      "entropy": 5.971634864807129,
      "epoch": 2.310091467986205,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004468895764550055,
      "loss": 5.5497,
      "mean_token_accuracy": 0.18364301025867463,
      "num_tokens": 45203541.0,
      "step": 23110
    },
    {
      "entropy": 6.009035205841064,
      "epoch": 2.3105912930474335,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004468665451470797,
      "loss": 5.6698,
      "mean_token_accuracy": 0.17297618538141252,
      "num_tokens": 45213231.0,
      "step": 23115
    },
    {
      "entropy": 6.030452156066895,
      "epoch": 2.3110911181086617,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004468435095150358,
      "loss": 5.6599,
      "mean_token_accuracy": 0.1695420816540718,
      "num_tokens": 45222585.0,
      "step": 23120
    },
    {
      "entropy": 6.008185958862304,
      "epoch": 2.3115909431698904,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044682046955945345,
      "loss": 5.6624,
      "mean_token_accuracy": 0.17794084846973418,
      "num_tokens": 45233068.0,
      "step": 23125
    },
    {
      "entropy": 5.984184837341308,
      "epoch": 2.312090768231119,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004467974252809123,
      "loss": 5.5785,
      "mean_token_accuracy": 0.180698524415493,
      "num_tokens": 45242643.0,
      "step": 23130
    },
    {
      "entropy": 6.019986343383789,
      "epoch": 2.312590593292348,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044677437667999204,
      "loss": 5.591,
      "mean_token_accuracy": 0.1797529712319374,
      "num_tokens": 45251870.0,
      "step": 23135
    },
    {
      "entropy": 5.9711589336395265,
      "epoch": 2.313090418353576,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044675132375727273,
      "loss": 5.5747,
      "mean_token_accuracy": 0.18160913288593292,
      "num_tokens": 45261209.0,
      "step": 23140
    },
    {
      "entropy": 6.003178882598877,
      "epoch": 2.313590243414805,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004467282665133343,
      "loss": 5.6616,
      "mean_token_accuracy": 0.17829559296369552,
      "num_tokens": 45270339.0,
      "step": 23145
    },
    {
      "entropy": 6.070397567749024,
      "epoch": 2.3140900684760335,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004467052049487569,
      "loss": 5.702,
      "mean_token_accuracy": 0.16889885067939758,
      "num_tokens": 45281107.0,
      "step": 23150
    },
    {
      "entropy": 5.990440464019775,
      "epoch": 2.3145898935372617,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044668213906412074,
      "loss": 5.5314,
      "mean_token_accuracy": 0.18736650347709655,
      "num_tokens": 45289615.0,
      "step": 23155
    },
    {
      "entropy": 5.919295167922973,
      "epoch": 2.3150897185984904,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044665906886000604,
      "loss": 5.5289,
      "mean_token_accuracy": 0.18982944190502166,
      "num_tokens": 45299768.0,
      "step": 23160
    },
    {
      "entropy": 5.897447013854981,
      "epoch": 2.315589543659719,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044663599433699346,
      "loss": 5.4808,
      "mean_token_accuracy": 0.19093730449676513,
      "num_tokens": 45310407.0,
      "step": 23165
    },
    {
      "entropy": 6.009524488449097,
      "epoch": 2.316089368720948,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044661291549566337,
      "loss": 5.5523,
      "mean_token_accuracy": 0.18220468908548354,
      "num_tokens": 45319993.0,
      "step": 23170
    },
    {
      "entropy": 5.920161533355713,
      "epoch": 2.316589193782176,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004465898323365965,
      "loss": 5.6538,
      "mean_token_accuracy": 0.17731260061264037,
      "num_tokens": 45329598.0,
      "step": 23175
    },
    {
      "entropy": 6.0381049633026125,
      "epoch": 2.317089018843405,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004465667448603736,
      "loss": 5.613,
      "mean_token_accuracy": 0.18015653491020203,
      "num_tokens": 45339630.0,
      "step": 23180
    },
    {
      "entropy": 5.946627950668335,
      "epoch": 2.3175888439046335,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044654365306757553,
      "loss": 5.5546,
      "mean_token_accuracy": 0.177405446767807,
      "num_tokens": 45349441.0,
      "step": 23185
    },
    {
      "entropy": 6.0102166652679445,
      "epoch": 2.3180886689658617,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044652055695878335,
      "loss": 5.7214,
      "mean_token_accuracy": 0.17170425802469252,
      "num_tokens": 45359858.0,
      "step": 23190
    },
    {
      "entropy": 6.021116781234741,
      "epoch": 2.3185884940270904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044649745653457806,
      "loss": 5.6173,
      "mean_token_accuracy": 0.1752406910061836,
      "num_tokens": 45369100.0,
      "step": 23195
    },
    {
      "entropy": 5.997559213638306,
      "epoch": 2.319088319088319,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044647435179554087,
      "loss": 5.5803,
      "mean_token_accuracy": 0.18057015389204026,
      "num_tokens": 45380514.0,
      "step": 23200
    },
    {
      "entropy": 5.998266744613647,
      "epoch": 2.319588144149548,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004464512427422531,
      "loss": 5.6591,
      "mean_token_accuracy": 0.17296882420778276,
      "num_tokens": 45390689.0,
      "step": 23205
    },
    {
      "entropy": 5.981978464126587,
      "epoch": 2.320087969210776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044642812937529624,
      "loss": 5.5989,
      "mean_token_accuracy": 0.1755844086408615,
      "num_tokens": 45400436.0,
      "step": 23210
    },
    {
      "entropy": 6.026718473434448,
      "epoch": 2.320587794272005,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004464050116952517,
      "loss": 5.6817,
      "mean_token_accuracy": 0.17870720475912094,
      "num_tokens": 45409906.0,
      "step": 23215
    },
    {
      "entropy": 5.9429347038269045,
      "epoch": 2.3210876193332335,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044638188970270114,
      "loss": 5.5896,
      "mean_token_accuracy": 0.17775266617536545,
      "num_tokens": 45419869.0,
      "step": 23220
    },
    {
      "entropy": 5.896588325500488,
      "epoch": 2.3215874443944617,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044635876339822647,
      "loss": 5.5477,
      "mean_token_accuracy": 0.1880938157439232,
      "num_tokens": 45430630.0,
      "step": 23225
    },
    {
      "entropy": 5.9853515625,
      "epoch": 2.3220872694556904,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044633563278240933,
      "loss": 5.4743,
      "mean_token_accuracy": 0.19221193045377732,
      "num_tokens": 45439765.0,
      "step": 23230
    },
    {
      "entropy": 6.026055192947387,
      "epoch": 2.322587094516919,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044631249785583173,
      "loss": 5.6104,
      "mean_token_accuracy": 0.18408116102218627,
      "num_tokens": 45448619.0,
      "step": 23235
    },
    {
      "entropy": 5.965742683410644,
      "epoch": 2.323086919578148,
      "grad_norm": 1.125,
      "learning_rate": 0.0004462893586190758,
      "loss": 5.6264,
      "mean_token_accuracy": 0.17982811182737352,
      "num_tokens": 45456874.0,
      "step": 23240
    },
    {
      "entropy": 5.980522775650025,
      "epoch": 2.323586744639376,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044626621507272364,
      "loss": 5.6852,
      "mean_token_accuracy": 0.17244212627410888,
      "num_tokens": 45465971.0,
      "step": 23245
    },
    {
      "entropy": 6.024433612823486,
      "epoch": 2.324086569700605,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044624306721735763,
      "loss": 5.626,
      "mean_token_accuracy": 0.16988710314035416,
      "num_tokens": 45475802.0,
      "step": 23250
    },
    {
      "entropy": 6.046125030517578,
      "epoch": 2.3245863947618335,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044621991505356014,
      "loss": 5.6163,
      "mean_token_accuracy": 0.18521066904067993,
      "num_tokens": 45486270.0,
      "step": 23255
    },
    {
      "entropy": 6.002754068374633,
      "epoch": 2.3250862198230617,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044619675858191353,
      "loss": 5.6516,
      "mean_token_accuracy": 0.17985492050647736,
      "num_tokens": 45495936.0,
      "step": 23260
    },
    {
      "entropy": 5.852035999298096,
      "epoch": 2.3255860448842904,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004461735978030006,
      "loss": 5.4411,
      "mean_token_accuracy": 0.1923342987895012,
      "num_tokens": 45505719.0,
      "step": 23265
    },
    {
      "entropy": 6.035499238967896,
      "epoch": 2.326085869945519,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044615043271740404,
      "loss": 5.6635,
      "mean_token_accuracy": 0.17641016840934753,
      "num_tokens": 45515322.0,
      "step": 23270
    },
    {
      "entropy": 5.970142078399658,
      "epoch": 2.326585695006748,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004461272633257065,
      "loss": 5.5578,
      "mean_token_accuracy": 0.18381035029888154,
      "num_tokens": 45524763.0,
      "step": 23275
    },
    {
      "entropy": 5.9663244724273685,
      "epoch": 2.327085520067976,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004461040896284912,
      "loss": 5.5734,
      "mean_token_accuracy": 0.18266777843236923,
      "num_tokens": 45533140.0,
      "step": 23280
    },
    {
      "entropy": 6.017339992523193,
      "epoch": 2.327585345129205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000446080911626341,
      "loss": 5.5936,
      "mean_token_accuracy": 0.1807784117758274,
      "num_tokens": 45543162.0,
      "step": 23285
    },
    {
      "entropy": 6.0431112289428714,
      "epoch": 2.3280851701904335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004460577293198391,
      "loss": 5.7458,
      "mean_token_accuracy": 0.1668761968612671,
      "num_tokens": 45552863.0,
      "step": 23290
    },
    {
      "entropy": 5.991557788848877,
      "epoch": 2.3285849952516617,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044603454270956863,
      "loss": 5.6051,
      "mean_token_accuracy": 0.1798216477036476,
      "num_tokens": 45562177.0,
      "step": 23295
    },
    {
      "entropy": 6.078386354446411,
      "epoch": 2.3290848203128904,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004460113517961132,
      "loss": 5.6694,
      "mean_token_accuracy": 0.17674380838871,
      "num_tokens": 45572864.0,
      "step": 23300
    },
    {
      "entropy": 6.0210058212280275,
      "epoch": 2.329584645374119,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004459881565800561,
      "loss": 5.6837,
      "mean_token_accuracy": 0.17749397158622743,
      "num_tokens": 45582778.0,
      "step": 23305
    },
    {
      "entropy": 5.941277933120728,
      "epoch": 2.330084470435348,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004459649570619809,
      "loss": 5.5187,
      "mean_token_accuracy": 0.18296999633312225,
      "num_tokens": 45591659.0,
      "step": 23310
    },
    {
      "entropy": 6.01179609298706,
      "epoch": 2.330584295496576,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004459417532424715,
      "loss": 5.7865,
      "mean_token_accuracy": 0.1679225891828537,
      "num_tokens": 45602215.0,
      "step": 23315
    },
    {
      "entropy": 5.961009454727173,
      "epoch": 2.331084120557805,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004459185451221116,
      "loss": 5.5962,
      "mean_token_accuracy": 0.17740359753370286,
      "num_tokens": 45611680.0,
      "step": 23320
    },
    {
      "entropy": 5.923630428314209,
      "epoch": 2.3315839456190335,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000445895332701485,
      "loss": 5.4826,
      "mean_token_accuracy": 0.19029170423746108,
      "num_tokens": 45621452.0,
      "step": 23325
    },
    {
      "entropy": 6.034183216094971,
      "epoch": 2.3320837706802617,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044587211598117577,
      "loss": 5.7084,
      "mean_token_accuracy": 0.17418909519910813,
      "num_tokens": 45631594.0,
      "step": 23330
    },
    {
      "entropy": 5.955716133117676,
      "epoch": 2.3325835957414904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044584889496176815,
      "loss": 5.5999,
      "mean_token_accuracy": 0.178103531897068,
      "num_tokens": 45642373.0,
      "step": 23335
    },
    {
      "entropy": 6.010469198226929,
      "epoch": 2.333083420802719,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004458256696438463,
      "loss": 5.5752,
      "mean_token_accuracy": 0.17788645923137664,
      "num_tokens": 45652851.0,
      "step": 23340
    },
    {
      "entropy": 5.809536075592041,
      "epoch": 2.333583245863948,
      "grad_norm": 1.125,
      "learning_rate": 0.00044580244002799454,
      "loss": 5.4255,
      "mean_token_accuracy": 0.1896498903632164,
      "num_tokens": 45662638.0,
      "step": 23345
    },
    {
      "entropy": 5.888195610046386,
      "epoch": 2.334083070925176,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004457792061147973,
      "loss": 5.5446,
      "mean_token_accuracy": 0.1824161484837532,
      "num_tokens": 45672252.0,
      "step": 23350
    },
    {
      "entropy": 6.052847051620484,
      "epoch": 2.334582895986405,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004457559679048392,
      "loss": 5.6257,
      "mean_token_accuracy": 0.1745440512895584,
      "num_tokens": 45682678.0,
      "step": 23355
    },
    {
      "entropy": 6.0050372123718265,
      "epoch": 2.3350827210476335,
      "grad_norm": 1.25,
      "learning_rate": 0.000445732725398705,
      "loss": 5.5924,
      "mean_token_accuracy": 0.17794700860977172,
      "num_tokens": 45692454.0,
      "step": 23360
    },
    {
      "entropy": 5.962378263473511,
      "epoch": 2.3355825461088617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004457094785969792,
      "loss": 5.6159,
      "mean_token_accuracy": 0.1762449860572815,
      "num_tokens": 45701925.0,
      "step": 23365
    },
    {
      "entropy": 5.957872009277343,
      "epoch": 2.3360823711700904,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044568622750024693,
      "loss": 5.637,
      "mean_token_accuracy": 0.17510392367839814,
      "num_tokens": 45711217.0,
      "step": 23370
    },
    {
      "entropy": 5.955209541320801,
      "epoch": 2.336582196231319,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004456629721090931,
      "loss": 5.5659,
      "mean_token_accuracy": 0.1837950751185417,
      "num_tokens": 45720651.0,
      "step": 23375
    },
    {
      "entropy": 6.020229911804199,
      "epoch": 2.337082021292548,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004456397124241028,
      "loss": 5.6998,
      "mean_token_accuracy": 0.1772367089986801,
      "num_tokens": 45730649.0,
      "step": 23380
    },
    {
      "entropy": 6.0219604015350345,
      "epoch": 2.337581846353776,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004456164484458612,
      "loss": 5.5486,
      "mean_token_accuracy": 0.1836159721016884,
      "num_tokens": 45739862.0,
      "step": 23385
    },
    {
      "entropy": 6.012317848205567,
      "epoch": 2.338081671415005,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004455931801749537,
      "loss": 5.6543,
      "mean_token_accuracy": 0.180059015750885,
      "num_tokens": 45750710.0,
      "step": 23390
    },
    {
      "entropy": 6.097760772705078,
      "epoch": 2.3385814964762335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044556990761196566,
      "loss": 5.714,
      "mean_token_accuracy": 0.1723231703042984,
      "num_tokens": 45759725.0,
      "step": 23395
    },
    {
      "entropy": 6.068141508102417,
      "epoch": 2.3390813215374617,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044554663075748266,
      "loss": 5.7251,
      "mean_token_accuracy": 0.1751093700528145,
      "num_tokens": 45769143.0,
      "step": 23400
    },
    {
      "entropy": 6.040675830841065,
      "epoch": 2.3395811465986904,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004455233496120903,
      "loss": 5.6696,
      "mean_token_accuracy": 0.1746293142437935,
      "num_tokens": 45779505.0,
      "step": 23405
    },
    {
      "entropy": 6.079249906539917,
      "epoch": 2.340080971659919,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004455000641763743,
      "loss": 5.6912,
      "mean_token_accuracy": 0.17492078244686127,
      "num_tokens": 45789337.0,
      "step": 23410
    },
    {
      "entropy": 5.978463077545166,
      "epoch": 2.340580796721148,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044547677445092064,
      "loss": 5.6514,
      "mean_token_accuracy": 0.1761349618434906,
      "num_tokens": 45799385.0,
      "step": 23415
    },
    {
      "entropy": 5.981095314025879,
      "epoch": 2.341080621782376,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044545348043631525,
      "loss": 5.6386,
      "mean_token_accuracy": 0.1692594602704048,
      "num_tokens": 45810918.0,
      "step": 23420
    },
    {
      "entropy": 6.028969812393188,
      "epoch": 2.341580446843605,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044543018213314404,
      "loss": 5.6474,
      "mean_token_accuracy": 0.17506189793348312,
      "num_tokens": 45820910.0,
      "step": 23425
    },
    {
      "entropy": 5.981546020507812,
      "epoch": 2.3420802719048335,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044540687954199335,
      "loss": 5.5332,
      "mean_token_accuracy": 0.17889311611652375,
      "num_tokens": 45830709.0,
      "step": 23430
    },
    {
      "entropy": 5.918363761901856,
      "epoch": 2.3425800969660617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044538357266344936,
      "loss": 5.5635,
      "mean_token_accuracy": 0.17890190035104753,
      "num_tokens": 45840244.0,
      "step": 23435
    },
    {
      "entropy": 5.938828229904175,
      "epoch": 2.3430799220272904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004453602614980986,
      "loss": 5.6318,
      "mean_token_accuracy": 0.1819308578968048,
      "num_tokens": 45849643.0,
      "step": 23440
    },
    {
      "entropy": 6.0704389095306395,
      "epoch": 2.343579747088519,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004453369460465274,
      "loss": 5.6423,
      "mean_token_accuracy": 0.1745069608092308,
      "num_tokens": 45858689.0,
      "step": 23445
    },
    {
      "entropy": 6.053126096725464,
      "epoch": 2.344079572149748,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004453136263093225,
      "loss": 5.6707,
      "mean_token_accuracy": 0.17474040389060974,
      "num_tokens": 45869538.0,
      "step": 23450
    },
    {
      "entropy": 6.026639032363891,
      "epoch": 2.344579397210976,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004452903022870707,
      "loss": 5.6869,
      "mean_token_accuracy": 0.17201502919197081,
      "num_tokens": 45878248.0,
      "step": 23455
    },
    {
      "entropy": 5.923809337615967,
      "epoch": 2.3450792222722048,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004452669739803586,
      "loss": 5.5736,
      "mean_token_accuracy": 0.17954163253307343,
      "num_tokens": 45888682.0,
      "step": 23460
    },
    {
      "entropy": 6.008859348297119,
      "epoch": 2.3455790473334335,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004452436413897733,
      "loss": 5.6259,
      "mean_token_accuracy": 0.18024303019046783,
      "num_tokens": 45898690.0,
      "step": 23465
    },
    {
      "entropy": 6.012032079696655,
      "epoch": 2.3460788723946617,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004452203045159018,
      "loss": 5.58,
      "mean_token_accuracy": 0.17659247368574144,
      "num_tokens": 45907579.0,
      "step": 23470
    },
    {
      "entropy": 5.904170846939087,
      "epoch": 2.3465786974558904,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004451969633593312,
      "loss": 5.5379,
      "mean_token_accuracy": 0.18853887915611267,
      "num_tokens": 45917065.0,
      "step": 23475
    },
    {
      "entropy": 5.9885858535766605,
      "epoch": 2.347078522517119,
      "grad_norm": 1.25,
      "learning_rate": 0.00044517361792064883,
      "loss": 5.7086,
      "mean_token_accuracy": 0.16866388469934462,
      "num_tokens": 45927427.0,
      "step": 23480
    },
    {
      "entropy": 6.022962141036987,
      "epoch": 2.347578347578348,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044515026820044195,
      "loss": 5.5058,
      "mean_token_accuracy": 0.1819187194108963,
      "num_tokens": 45937453.0,
      "step": 23485
    },
    {
      "entropy": 5.962077713012695,
      "epoch": 2.348078172639576,
      "grad_norm": 1.125,
      "learning_rate": 0.0004451269141992982,
      "loss": 5.5368,
      "mean_token_accuracy": 0.18698531836271287,
      "num_tokens": 45947285.0,
      "step": 23490
    },
    {
      "entropy": 6.0513711929321286,
      "epoch": 2.3485779977008048,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044510355591780503,
      "loss": 5.7114,
      "mean_token_accuracy": 0.17182620167732238,
      "num_tokens": 45956413.0,
      "step": 23495
    },
    {
      "entropy": 6.020291137695312,
      "epoch": 2.3490778227620335,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004450801933565502,
      "loss": 5.6361,
      "mean_token_accuracy": 0.1801711991429329,
      "num_tokens": 45967243.0,
      "step": 23500
    },
    {
      "entropy": 5.932665491104126,
      "epoch": 2.3495776478232617,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004450568265161214,
      "loss": 5.5835,
      "mean_token_accuracy": 0.1820272296667099,
      "num_tokens": 45976837.0,
      "step": 23505
    },
    {
      "entropy": 5.972005367279053,
      "epoch": 2.3500774728844904,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044503345539710664,
      "loss": 5.6237,
      "mean_token_accuracy": 0.17929652333259583,
      "num_tokens": 45986498.0,
      "step": 23510
    },
    {
      "entropy": 6.026886367797852,
      "epoch": 2.350577297945719,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004450100800000939,
      "loss": 5.6257,
      "mean_token_accuracy": 0.18278508335351945,
      "num_tokens": 45995490.0,
      "step": 23515
    },
    {
      "entropy": 5.977897787094117,
      "epoch": 2.3510771230069474,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004449867003256713,
      "loss": 5.5668,
      "mean_token_accuracy": 0.18352842181921006,
      "num_tokens": 46004765.0,
      "step": 23520
    },
    {
      "entropy": 5.950900316238403,
      "epoch": 2.351576948068176,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044496331637442697,
      "loss": 5.6421,
      "mean_token_accuracy": 0.17834141850471497,
      "num_tokens": 46014666.0,
      "step": 23525
    },
    {
      "entropy": 5.924844408035279,
      "epoch": 2.3520767731294048,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004449399281469495,
      "loss": 5.5263,
      "mean_token_accuracy": 0.17962506115436555,
      "num_tokens": 46024104.0,
      "step": 23530
    },
    {
      "entropy": 5.943951368331909,
      "epoch": 2.3525765981906335,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044491653564382704,
      "loss": 5.5213,
      "mean_token_accuracy": 0.18812463134527208,
      "num_tokens": 46034709.0,
      "step": 23535
    },
    {
      "entropy": 6.000288677215576,
      "epoch": 2.3530764232518617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004448931388656483,
      "loss": 5.6496,
      "mean_token_accuracy": 0.17739568203687667,
      "num_tokens": 46044404.0,
      "step": 23540
    },
    {
      "entropy": 5.922780466079712,
      "epoch": 2.3535762483130904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044486973781300196,
      "loss": 5.594,
      "mean_token_accuracy": 0.17940337508916854,
      "num_tokens": 46054223.0,
      "step": 23545
    },
    {
      "entropy": 5.980970048904419,
      "epoch": 2.354076073374319,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044484633248647665,
      "loss": 5.578,
      "mean_token_accuracy": 0.17344460934400557,
      "num_tokens": 46063491.0,
      "step": 23550
    },
    {
      "entropy": 5.990363645553589,
      "epoch": 2.3545758984355474,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044482292288666134,
      "loss": 5.609,
      "mean_token_accuracy": 0.1813024416565895,
      "num_tokens": 46072552.0,
      "step": 23555
    },
    {
      "entropy": 5.866642808914184,
      "epoch": 2.355075723496776,
      "grad_norm": 1.140625,
      "learning_rate": 0.000444799509014145,
      "loss": 5.4778,
      "mean_token_accuracy": 0.1917403146624565,
      "num_tokens": 46083249.0,
      "step": 23560
    },
    {
      "entropy": 6.059669399261475,
      "epoch": 2.3555755485580048,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004447760908695167,
      "loss": 5.6559,
      "mean_token_accuracy": 0.17237832248210908,
      "num_tokens": 46091842.0,
      "step": 23565
    },
    {
      "entropy": 5.885122537612915,
      "epoch": 2.356075373619233,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044475266845336566,
      "loss": 5.5057,
      "mean_token_accuracy": 0.1876189738512039,
      "num_tokens": 46101453.0,
      "step": 23570
    },
    {
      "entropy": 6.034854507446289,
      "epoch": 2.3565751986804617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004447292417662811,
      "loss": 5.7584,
      "mean_token_accuracy": 0.16922744512557983,
      "num_tokens": 46110969.0,
      "step": 23575
    },
    {
      "entropy": 5.955658102035523,
      "epoch": 2.3570750237416904,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004447058108088525,
      "loss": 5.4654,
      "mean_token_accuracy": 0.1966336190700531,
      "num_tokens": 46120846.0,
      "step": 23580
    },
    {
      "entropy": 5.932594299316406,
      "epoch": 2.357574848802919,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044468237558166945,
      "loss": 5.5031,
      "mean_token_accuracy": 0.1900033473968506,
      "num_tokens": 46130039.0,
      "step": 23585
    },
    {
      "entropy": 5.934714698791504,
      "epoch": 2.3580746738641474,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044465893608532146,
      "loss": 5.5359,
      "mean_token_accuracy": 0.1862268790602684,
      "num_tokens": 46139503.0,
      "step": 23590
    },
    {
      "entropy": 5.981346940994262,
      "epoch": 2.358574498925376,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044463549232039824,
      "loss": 5.5982,
      "mean_token_accuracy": 0.1780770853161812,
      "num_tokens": 46148247.0,
      "step": 23595
    },
    {
      "entropy": 5.9502708435058596,
      "epoch": 2.3590743239866048,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004446120442874898,
      "loss": 5.6211,
      "mean_token_accuracy": 0.1848107784986496,
      "num_tokens": 46157233.0,
      "step": 23600
    },
    {
      "entropy": 5.975366640090942,
      "epoch": 2.359574149047833,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004445885919871859,
      "loss": 5.5916,
      "mean_token_accuracy": 0.17811685800552368,
      "num_tokens": 46166621.0,
      "step": 23605
    },
    {
      "entropy": 5.997378301620484,
      "epoch": 2.3600739741090617,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004445651354200766,
      "loss": 5.6466,
      "mean_token_accuracy": 0.1759398803114891,
      "num_tokens": 46176142.0,
      "step": 23610
    },
    {
      "entropy": 5.9136443614959715,
      "epoch": 2.3605737991702904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044454167458675225,
      "loss": 5.5094,
      "mean_token_accuracy": 0.18230413645505905,
      "num_tokens": 46185506.0,
      "step": 23615
    },
    {
      "entropy": 5.950039386749268,
      "epoch": 2.361073624231519,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044451820948780295,
      "loss": 5.6292,
      "mean_token_accuracy": 0.18282263427972795,
      "num_tokens": 46195351.0,
      "step": 23620
    },
    {
      "entropy": 5.956896734237671,
      "epoch": 2.3615734492927474,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044449474012381903,
      "loss": 5.5945,
      "mean_token_accuracy": 0.1781872272491455,
      "num_tokens": 46205431.0,
      "step": 23625
    },
    {
      "entropy": 6.015918493270874,
      "epoch": 2.362073274353976,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044447126649539123,
      "loss": 5.5892,
      "mean_token_accuracy": 0.17809753715991974,
      "num_tokens": 46214731.0,
      "step": 23630
    },
    {
      "entropy": 5.94448447227478,
      "epoch": 2.3625730994152048,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004444477886031098,
      "loss": 5.5775,
      "mean_token_accuracy": 0.18672974705696105,
      "num_tokens": 46223951.0,
      "step": 23635
    },
    {
      "entropy": 6.014925289154053,
      "epoch": 2.363072924476433,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004444243064475657,
      "loss": 5.6708,
      "mean_token_accuracy": 0.1735333576798439,
      "num_tokens": 46234458.0,
      "step": 23640
    },
    {
      "entropy": 5.977278566360473,
      "epoch": 2.3635727495376617,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004444008200293496,
      "loss": 5.664,
      "mean_token_accuracy": 0.1766430765390396,
      "num_tokens": 46244549.0,
      "step": 23645
    },
    {
      "entropy": 5.96056900024414,
      "epoch": 2.3640725745988904,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044437732934905256,
      "loss": 5.6039,
      "mean_token_accuracy": 0.17859516739845277,
      "num_tokens": 46254360.0,
      "step": 23650
    },
    {
      "entropy": 5.994832944869995,
      "epoch": 2.364572399660119,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004443538344072654,
      "loss": 5.5599,
      "mean_token_accuracy": 0.18126943558454514,
      "num_tokens": 46264153.0,
      "step": 23655
    },
    {
      "entropy": 6.026865148544312,
      "epoch": 2.3650722247213474,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044433033520457937,
      "loss": 5.5443,
      "mean_token_accuracy": 0.18109805434942244,
      "num_tokens": 46272797.0,
      "step": 23660
    },
    {
      "entropy": 5.962622117996216,
      "epoch": 2.365572049782576,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004443068317415857,
      "loss": 5.5596,
      "mean_token_accuracy": 0.18231860995292665,
      "num_tokens": 46282515.0,
      "step": 23665
    },
    {
      "entropy": 5.9555823802948,
      "epoch": 2.3660718748438048,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044428332401887566,
      "loss": 5.6912,
      "mean_token_accuracy": 0.16980768293142318,
      "num_tokens": 46292682.0,
      "step": 23670
    },
    {
      "entropy": 5.9160768508911135,
      "epoch": 2.366571699905033,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004442598120370408,
      "loss": 5.526,
      "mean_token_accuracy": 0.19138217568397523,
      "num_tokens": 46302598.0,
      "step": 23675
    },
    {
      "entropy": 5.946627759933472,
      "epoch": 2.3670715249662617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004442362957966726,
      "loss": 5.5348,
      "mean_token_accuracy": 0.18850620687007905,
      "num_tokens": 46313015.0,
      "step": 23680
    },
    {
      "entropy": 5.907001924514771,
      "epoch": 2.3675713500274904,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044421277529836276,
      "loss": 5.5913,
      "mean_token_accuracy": 0.1796092614531517,
      "num_tokens": 46322450.0,
      "step": 23685
    },
    {
      "entropy": 6.008005857467651,
      "epoch": 2.368071175088719,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004441892505427031,
      "loss": 5.5845,
      "mean_token_accuracy": 0.17648232132196426,
      "num_tokens": 46332357.0,
      "step": 23690
    },
    {
      "entropy": 6.068871641159058,
      "epoch": 2.3685710001499474,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004441657215302853,
      "loss": 5.6462,
      "mean_token_accuracy": 0.17598066627979278,
      "num_tokens": 46341805.0,
      "step": 23695
    },
    {
      "entropy": 6.0014190673828125,
      "epoch": 2.369070825211176,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044414218826170155,
      "loss": 5.599,
      "mean_token_accuracy": 0.1790556162595749,
      "num_tokens": 46352814.0,
      "step": 23700
    },
    {
      "entropy": 6.022558116912842,
      "epoch": 2.3695706502724048,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004441186507375439,
      "loss": 5.6574,
      "mean_token_accuracy": 0.17674969136714935,
      "num_tokens": 46362968.0,
      "step": 23705
    },
    {
      "entropy": 5.974521732330322,
      "epoch": 2.370070475333633,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044409510895840446,
      "loss": 5.629,
      "mean_token_accuracy": 0.1785505458712578,
      "num_tokens": 46373379.0,
      "step": 23710
    },
    {
      "entropy": 5.931599378585815,
      "epoch": 2.3705703003948617,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044407156292487553,
      "loss": 5.5305,
      "mean_token_accuracy": 0.18883699476718901,
      "num_tokens": 46382788.0,
      "step": 23715
    },
    {
      "entropy": 6.032562732696533,
      "epoch": 2.3710701254560904,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004440480126375497,
      "loss": 5.6471,
      "mean_token_accuracy": 0.17998366206884384,
      "num_tokens": 46392984.0,
      "step": 23720
    },
    {
      "entropy": 5.978254127502441,
      "epoch": 2.371569950517319,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004440244580970193,
      "loss": 5.545,
      "mean_token_accuracy": 0.18394755572080612,
      "num_tokens": 46403387.0,
      "step": 23725
    },
    {
      "entropy": 5.925871133804321,
      "epoch": 2.3720697755785474,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044400089930387703,
      "loss": 5.5225,
      "mean_token_accuracy": 0.18091276437044143,
      "num_tokens": 46413010.0,
      "step": 23730
    },
    {
      "entropy": 5.970633268356323,
      "epoch": 2.372569600639776,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044397733625871566,
      "loss": 5.6166,
      "mean_token_accuracy": 0.17739286571741103,
      "num_tokens": 46423790.0,
      "step": 23735
    },
    {
      "entropy": 6.072221183776856,
      "epoch": 2.3730694257010048,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044395376896212805,
      "loss": 5.6097,
      "mean_token_accuracy": 0.17449352890253067,
      "num_tokens": 46434210.0,
      "step": 23740
    },
    {
      "entropy": 5.932456111907959,
      "epoch": 2.373569250762233,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044393019741470696,
      "loss": 5.4988,
      "mean_token_accuracy": 0.18396941125392913,
      "num_tokens": 46443747.0,
      "step": 23745
    },
    {
      "entropy": 5.904684448242188,
      "epoch": 2.3740690758234617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004439066216170456,
      "loss": 5.5072,
      "mean_token_accuracy": 0.18636028617620468,
      "num_tokens": 46453829.0,
      "step": 23750
    },
    {
      "entropy": 6.003620958328247,
      "epoch": 2.3745689008846904,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044388304156973713,
      "loss": 5.6512,
      "mean_token_accuracy": 0.18152665048837663,
      "num_tokens": 46463406.0,
      "step": 23755
    },
    {
      "entropy": 5.9940833568573,
      "epoch": 2.375068725945919,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044385945727337477,
      "loss": 5.5907,
      "mean_token_accuracy": 0.18261998444795607,
      "num_tokens": 46473374.0,
      "step": 23760
    },
    {
      "entropy": 5.973348903656006,
      "epoch": 2.3755685510071474,
      "grad_norm": 1.078125,
      "learning_rate": 0.000443835868728552,
      "loss": 5.5872,
      "mean_token_accuracy": 0.18620624244213105,
      "num_tokens": 46482881.0,
      "step": 23765
    },
    {
      "entropy": 5.951352977752686,
      "epoch": 2.376068376068376,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004438122759358621,
      "loss": 5.6157,
      "mean_token_accuracy": 0.17811293303966522,
      "num_tokens": 46493040.0,
      "step": 23770
    },
    {
      "entropy": 6.033521318435669,
      "epoch": 2.3765682011296048,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044378867889589884,
      "loss": 5.6076,
      "mean_token_accuracy": 0.17941468358039855,
      "num_tokens": 46503243.0,
      "step": 23775
    },
    {
      "entropy": 5.9622279644012455,
      "epoch": 2.377068026190833,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004437650776092558,
      "loss": 5.5236,
      "mean_token_accuracy": 0.18519055992364883,
      "num_tokens": 46512864.0,
      "step": 23780
    },
    {
      "entropy": 5.934012699127197,
      "epoch": 2.3775678512520617,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044374147207652695,
      "loss": 5.6272,
      "mean_token_accuracy": 0.18526759445667268,
      "num_tokens": 46522404.0,
      "step": 23785
    },
    {
      "entropy": 5.932480764389038,
      "epoch": 2.3780676763132904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000443717862298306,
      "loss": 5.5775,
      "mean_token_accuracy": 0.17699134349822998,
      "num_tokens": 46532235.0,
      "step": 23790
    },
    {
      "entropy": 5.991212177276611,
      "epoch": 2.378567501374519,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044369424827518704,
      "loss": 5.5644,
      "mean_token_accuracy": 0.17901394814252852,
      "num_tokens": 46542337.0,
      "step": 23795
    },
    {
      "entropy": 6.030649280548095,
      "epoch": 2.3790673264357474,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044367063000776423,
      "loss": 5.6415,
      "mean_token_accuracy": 0.18214326053857804,
      "num_tokens": 46551446.0,
      "step": 23800
    },
    {
      "entropy": 6.028851175308228,
      "epoch": 2.379567151496976,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004436470074966317,
      "loss": 5.6586,
      "mean_token_accuracy": 0.18050740659236908,
      "num_tokens": 46562289.0,
      "step": 23805
    },
    {
      "entropy": 5.912641477584839,
      "epoch": 2.3800669765582048,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000443623380742384,
      "loss": 5.5148,
      "mean_token_accuracy": 0.18278919607400895,
      "num_tokens": 46572135.0,
      "step": 23810
    },
    {
      "entropy": 5.996970701217651,
      "epoch": 2.380566801619433,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004435997497456153,
      "loss": 5.5968,
      "mean_token_accuracy": 0.1803893506526947,
      "num_tokens": 46581810.0,
      "step": 23815
    },
    {
      "entropy": 5.958755207061768,
      "epoch": 2.3810666266806617,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004435761145069203,
      "loss": 5.5486,
      "mean_token_accuracy": 0.18072778433561326,
      "num_tokens": 46591886.0,
      "step": 23820
    },
    {
      "entropy": 5.9792563915252686,
      "epoch": 2.3815664517418904,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044355247502689374,
      "loss": 5.694,
      "mean_token_accuracy": 0.16591572165489196,
      "num_tokens": 46601632.0,
      "step": 23825
    },
    {
      "entropy": 5.952917766571045,
      "epoch": 2.382066276803119,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004435288313061302,
      "loss": 5.5857,
      "mean_token_accuracy": 0.1844136521220207,
      "num_tokens": 46611802.0,
      "step": 23830
    },
    {
      "entropy": 5.952917528152466,
      "epoch": 2.3825661018643474,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004435051833452246,
      "loss": 5.5181,
      "mean_token_accuracy": 0.18462881445884705,
      "num_tokens": 46621965.0,
      "step": 23835
    },
    {
      "entropy": 5.976242780685425,
      "epoch": 2.383065926925576,
      "grad_norm": 1.125,
      "learning_rate": 0.000443481531144772,
      "loss": 5.5543,
      "mean_token_accuracy": 0.1828855112195015,
      "num_tokens": 46632518.0,
      "step": 23840
    },
    {
      "entropy": 5.887647819519043,
      "epoch": 2.3835657519868048,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004434578747053674,
      "loss": 5.5207,
      "mean_token_accuracy": 0.19145217835903167,
      "num_tokens": 46642393.0,
      "step": 23845
    },
    {
      "entropy": 5.893708992004394,
      "epoch": 2.384065577048033,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044343421402760606,
      "loss": 5.5412,
      "mean_token_accuracy": 0.18712605983018876,
      "num_tokens": 46653025.0,
      "step": 23850
    },
    {
      "entropy": 5.99381947517395,
      "epoch": 2.3845654021092617,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004434105491120832,
      "loss": 5.5927,
      "mean_token_accuracy": 0.1759718582034111,
      "num_tokens": 46661655.0,
      "step": 23855
    },
    {
      "entropy": 6.043883419036865,
      "epoch": 2.3850652271704904,
      "grad_norm": 1.125,
      "learning_rate": 0.00044338687995939424,
      "loss": 5.6819,
      "mean_token_accuracy": 0.17279390394687652,
      "num_tokens": 46672392.0,
      "step": 23860
    },
    {
      "entropy": 6.039701175689697,
      "epoch": 2.385565052231719,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004433632065701348,
      "loss": 5.6246,
      "mean_token_accuracy": 0.17670282274484633,
      "num_tokens": 46683099.0,
      "step": 23865
    },
    {
      "entropy": 5.959401321411133,
      "epoch": 2.3860648772929474,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004433395289449003,
      "loss": 5.5652,
      "mean_token_accuracy": 0.18606751561164855,
      "num_tokens": 46692807.0,
      "step": 23870
    },
    {
      "entropy": 5.908703422546386,
      "epoch": 2.386564702354176,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004433158470842867,
      "loss": 5.5426,
      "mean_token_accuracy": 0.18078771680593492,
      "num_tokens": 46703093.0,
      "step": 23875
    },
    {
      "entropy": 5.93207483291626,
      "epoch": 2.3870645274154048,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004432921609888896,
      "loss": 5.568,
      "mean_token_accuracy": 0.17712525576353072,
      "num_tokens": 46712487.0,
      "step": 23880
    },
    {
      "entropy": 5.971924829483032,
      "epoch": 2.387564352476633,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044326847065930507,
      "loss": 5.6221,
      "mean_token_accuracy": 0.17720161229372025,
      "num_tokens": 46723627.0,
      "step": 23885
    },
    {
      "entropy": 6.000041198730469,
      "epoch": 2.3880641775378617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044324477609612907,
      "loss": 5.4911,
      "mean_token_accuracy": 0.18672308772802354,
      "num_tokens": 46733130.0,
      "step": 23890
    },
    {
      "entropy": 6.0100616931915285,
      "epoch": 2.3885640025990904,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004432210772999578,
      "loss": 5.684,
      "mean_token_accuracy": 0.17437852770090104,
      "num_tokens": 46743380.0,
      "step": 23895
    },
    {
      "entropy": 5.983158302307129,
      "epoch": 2.389063827660319,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004431973742713876,
      "loss": 5.6184,
      "mean_token_accuracy": 0.17453592270612717,
      "num_tokens": 46753420.0,
      "step": 23900
    },
    {
      "entropy": 6.00272798538208,
      "epoch": 2.3895636527215474,
      "grad_norm": 1.125,
      "learning_rate": 0.0004431736670110146,
      "loss": 5.5693,
      "mean_token_accuracy": 0.1748160570859909,
      "num_tokens": 46763342.0,
      "step": 23905
    },
    {
      "entropy": 5.889030170440674,
      "epoch": 2.390063477782776,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044314995551943554,
      "loss": 5.5033,
      "mean_token_accuracy": 0.18750476986169815,
      "num_tokens": 46772300.0,
      "step": 23910
    },
    {
      "entropy": 6.011675882339477,
      "epoch": 2.3905633028440048,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004431262397972468,
      "loss": 5.5406,
      "mean_token_accuracy": 0.18521214425563812,
      "num_tokens": 46781778.0,
      "step": 23915
    },
    {
      "entropy": 5.874504804611206,
      "epoch": 2.391063127905233,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044310251984504517,
      "loss": 5.3985,
      "mean_token_accuracy": 0.19847344756126403,
      "num_tokens": 46792845.0,
      "step": 23920
    },
    {
      "entropy": 6.066433048248291,
      "epoch": 2.3915629529664617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044307879566342736,
      "loss": 5.7134,
      "mean_token_accuracy": 0.16575897634029388,
      "num_tokens": 46802940.0,
      "step": 23925
    },
    {
      "entropy": 5.990143251419068,
      "epoch": 2.3920627780276904,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004430550672529903,
      "loss": 5.5347,
      "mean_token_accuracy": 0.18613084852695466,
      "num_tokens": 46812493.0,
      "step": 23930
    },
    {
      "entropy": 6.009265565872193,
      "epoch": 2.392562603088919,
      "grad_norm": 1.203125,
      "learning_rate": 0.000443031334614331,
      "loss": 5.6179,
      "mean_token_accuracy": 0.17940812706947326,
      "num_tokens": 46821102.0,
      "step": 23935
    },
    {
      "entropy": 5.935640954971314,
      "epoch": 2.3930624281501474,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004430075977480466,
      "loss": 5.5928,
      "mean_token_accuracy": 0.18228798359632492,
      "num_tokens": 46832717.0,
      "step": 23940
    },
    {
      "entropy": 5.887172222137451,
      "epoch": 2.393562253211376,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004429838566547343,
      "loss": 5.5228,
      "mean_token_accuracy": 0.18294385224580764,
      "num_tokens": 46842756.0,
      "step": 23945
    },
    {
      "entropy": 6.001048278808594,
      "epoch": 2.3940620782726048,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004429601113349913,
      "loss": 5.6104,
      "mean_token_accuracy": 0.17648856490850448,
      "num_tokens": 46853114.0,
      "step": 23950
    },
    {
      "entropy": 6.04467363357544,
      "epoch": 2.394561903333833,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044293636178941516,
      "loss": 5.5744,
      "mean_token_accuracy": 0.17400801181793213,
      "num_tokens": 46862286.0,
      "step": 23955
    },
    {
      "entropy": 5.982899188995361,
      "epoch": 2.3950617283950617,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004429126080186034,
      "loss": 5.6153,
      "mean_token_accuracy": 0.17627927511930466,
      "num_tokens": 46872155.0,
      "step": 23960
    },
    {
      "entropy": 5.936293888092041,
      "epoch": 2.3955615534562904,
      "grad_norm": 1.125,
      "learning_rate": 0.00044288885002315354,
      "loss": 5.5565,
      "mean_token_accuracy": 0.18651289939880372,
      "num_tokens": 46881969.0,
      "step": 23965
    },
    {
      "entropy": 5.971389722824097,
      "epoch": 2.396061378517519,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044286508780366345,
      "loss": 5.5967,
      "mean_token_accuracy": 0.17786987423896788,
      "num_tokens": 46890855.0,
      "step": 23970
    },
    {
      "entropy": 5.96508469581604,
      "epoch": 2.3965612035787474,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044284132136073097,
      "loss": 5.5263,
      "mean_token_accuracy": 0.1852623149752617,
      "num_tokens": 46900803.0,
      "step": 23975
    },
    {
      "entropy": 5.920167732238769,
      "epoch": 2.397061028639976,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004428175506949541,
      "loss": 5.5862,
      "mean_token_accuracy": 0.18705372512340546,
      "num_tokens": 46910178.0,
      "step": 23980
    },
    {
      "entropy": 6.00419430732727,
      "epoch": 2.3975608537012048,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044279377580693077,
      "loss": 5.5641,
      "mean_token_accuracy": 0.1836267188191414,
      "num_tokens": 46919860.0,
      "step": 23985
    },
    {
      "entropy": 6.020566606521607,
      "epoch": 2.398060678762433,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004427699966972592,
      "loss": 5.5423,
      "mean_token_accuracy": 0.18594708442687988,
      "num_tokens": 46929570.0,
      "step": 23990
    },
    {
      "entropy": 5.924635601043701,
      "epoch": 2.3985605038236617,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044274621336653765,
      "loss": 5.5552,
      "mean_token_accuracy": 0.1800750896334648,
      "num_tokens": 46939685.0,
      "step": 23995
    },
    {
      "entropy": 5.893094396591186,
      "epoch": 2.3990603288848904,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044272242581536457,
      "loss": 5.4806,
      "mean_token_accuracy": 0.1960731104016304,
      "num_tokens": 46949720.0,
      "step": 24000
    },
    {
      "epoch": 2.3990603288848904,
      "eval_entropy": 5.771433534649223,
      "eval_loss": 5.72047233581543,
      "eval_mean_token_accuracy": 0.1826065264726708,
      "eval_num_tokens": 46949720.0,
      "eval_runtime": 23.9463,
      "eval_samples_per_second": 1296.443,
      "eval_steps_per_second": 162.071,
      "step": 24000
    },
    {
      "entropy": 5.974677658081054,
      "epoch": 2.399560153946119,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044269863404433846,
      "loss": 5.5646,
      "mean_token_accuracy": 0.18066530972719191,
      "num_tokens": 46958883.0,
      "step": 24005
    },
    {
      "entropy": 6.044573259353638,
      "epoch": 2.4000599790073474,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004426748380540578,
      "loss": 5.6502,
      "mean_token_accuracy": 0.17266529202461242,
      "num_tokens": 46969146.0,
      "step": 24010
    },
    {
      "entropy": 5.969880056381226,
      "epoch": 2.400559804068576,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004426510378451215,
      "loss": 5.634,
      "mean_token_accuracy": 0.17416854053735734,
      "num_tokens": 46978250.0,
      "step": 24015
    },
    {
      "entropy": 6.016442537307739,
      "epoch": 2.4010596291298048,
      "grad_norm": 1.25,
      "learning_rate": 0.00044262723341812805,
      "loss": 5.6324,
      "mean_token_accuracy": 0.1779349148273468,
      "num_tokens": 46988413.0,
      "step": 24020
    },
    {
      "entropy": 6.075217914581299,
      "epoch": 2.401559454191033,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004426034247736766,
      "loss": 5.7137,
      "mean_token_accuracy": 0.17341222614049911,
      "num_tokens": 46998092.0,
      "step": 24025
    },
    {
      "entropy": 5.998846960067749,
      "epoch": 2.4020592792522617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004425796119123662,
      "loss": 5.6497,
      "mean_token_accuracy": 0.1816287964582443,
      "num_tokens": 47006764.0,
      "step": 24030
    },
    {
      "entropy": 6.017466974258423,
      "epoch": 2.4025591043134904,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044255579483479577,
      "loss": 5.5825,
      "mean_token_accuracy": 0.17741208523511887,
      "num_tokens": 47016789.0,
      "step": 24035
    },
    {
      "entropy": 5.938273096084595,
      "epoch": 2.4030589293747187,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004425319735415647,
      "loss": 5.5294,
      "mean_token_accuracy": 0.18994368761777877,
      "num_tokens": 47026984.0,
      "step": 24040
    },
    {
      "entropy": 6.001030683517456,
      "epoch": 2.4035587544359474,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004425081480332723,
      "loss": 5.6362,
      "mean_token_accuracy": 0.17687994688749314,
      "num_tokens": 47035893.0,
      "step": 24045
    },
    {
      "entropy": 5.886037063598633,
      "epoch": 2.404058579497176,
      "grad_norm": 1.140625,
      "learning_rate": 0.000442484318310518,
      "loss": 5.51,
      "mean_token_accuracy": 0.1807472065091133,
      "num_tokens": 47045351.0,
      "step": 24050
    },
    {
      "entropy": 5.91890058517456,
      "epoch": 2.4045584045584047,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044246048437390133,
      "loss": 5.6093,
      "mean_token_accuracy": 0.17623412758111953,
      "num_tokens": 47055849.0,
      "step": 24055
    },
    {
      "entropy": 6.0040162086486815,
      "epoch": 2.405058229619633,
      "grad_norm": 1.1875,
      "learning_rate": 0.000442436646224022,
      "loss": 5.6048,
      "mean_token_accuracy": 0.17397351413965226,
      "num_tokens": 47065068.0,
      "step": 24060
    },
    {
      "entropy": 5.959394407272339,
      "epoch": 2.4055580546808617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044241280386147965,
      "loss": 5.5179,
      "mean_token_accuracy": 0.18708568066358566,
      "num_tokens": 47075000.0,
      "step": 24065
    },
    {
      "entropy": 5.946767663955688,
      "epoch": 2.4060578797420904,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044238895728687433,
      "loss": 5.5913,
      "mean_token_accuracy": 0.1775379076600075,
      "num_tokens": 47084711.0,
      "step": 24070
    },
    {
      "entropy": 5.938310718536377,
      "epoch": 2.4065577048033187,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004423651065008059,
      "loss": 5.4709,
      "mean_token_accuracy": 0.1856689229607582,
      "num_tokens": 47094466.0,
      "step": 24075
    },
    {
      "entropy": 6.000906705856323,
      "epoch": 2.4070575298645474,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004423412515038744,
      "loss": 5.5783,
      "mean_token_accuracy": 0.17778525352478028,
      "num_tokens": 47104388.0,
      "step": 24080
    },
    {
      "entropy": 5.985598516464234,
      "epoch": 2.407557354925776,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004423173922966802,
      "loss": 5.6519,
      "mean_token_accuracy": 0.17858876138925553,
      "num_tokens": 47115158.0,
      "step": 24085
    },
    {
      "entropy": 6.075100612640381,
      "epoch": 2.4080571799870043,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044229352887982325,
      "loss": 5.6742,
      "mean_token_accuracy": 0.17107024639844895,
      "num_tokens": 47126739.0,
      "step": 24090
    },
    {
      "entropy": 6.0216516017913815,
      "epoch": 2.408557005048233,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044226966125390435,
      "loss": 5.621,
      "mean_token_accuracy": 0.1759901836514473,
      "num_tokens": 47136586.0,
      "step": 24095
    },
    {
      "entropy": 5.876541042327881,
      "epoch": 2.4090568301094617,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004422457894195237,
      "loss": 5.5437,
      "mean_token_accuracy": 0.18790817111730576,
      "num_tokens": 47146099.0,
      "step": 24100
    },
    {
      "entropy": 5.939631080627441,
      "epoch": 2.4095566551706904,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004422219133772821,
      "loss": 5.6212,
      "mean_token_accuracy": 0.18023970425128938,
      "num_tokens": 47154165.0,
      "step": 24105
    },
    {
      "entropy": 5.987698125839233,
      "epoch": 2.4100564802319187,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004421980331277802,
      "loss": 5.607,
      "mean_token_accuracy": 0.18353760242462158,
      "num_tokens": 47164060.0,
      "step": 24110
    },
    {
      "entropy": 5.99893012046814,
      "epoch": 2.4105563052931473,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044217414867161873,
      "loss": 5.6886,
      "mean_token_accuracy": 0.17244175672531128,
      "num_tokens": 47174256.0,
      "step": 24115
    },
    {
      "entropy": 5.929987621307373,
      "epoch": 2.411056130354376,
      "grad_norm": 1.125,
      "learning_rate": 0.0004421502600093987,
      "loss": 5.5563,
      "mean_token_accuracy": 0.1810513988137245,
      "num_tokens": 47183553.0,
      "step": 24120
    },
    {
      "entropy": 6.010694265365601,
      "epoch": 2.4115559554156043,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044212636714172115,
      "loss": 5.6058,
      "mean_token_accuracy": 0.1791193813085556,
      "num_tokens": 47194542.0,
      "step": 24125
    },
    {
      "entropy": 6.017143058776855,
      "epoch": 2.412055780476833,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004421024700691873,
      "loss": 5.5702,
      "mean_token_accuracy": 0.18206753879785537,
      "num_tokens": 47204418.0,
      "step": 24130
    },
    {
      "entropy": 5.911156940460205,
      "epoch": 2.4125556055380617,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004420785687923982,
      "loss": 5.5022,
      "mean_token_accuracy": 0.1971539855003357,
      "num_tokens": 47212727.0,
      "step": 24135
    },
    {
      "entropy": 5.834893703460693,
      "epoch": 2.4130554305992904,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004420546633119554,
      "loss": 5.611,
      "mean_token_accuracy": 0.1871244192123413,
      "num_tokens": 47222907.0,
      "step": 24140
    },
    {
      "entropy": 5.908796215057373,
      "epoch": 2.4135552556605187,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004420307536284602,
      "loss": 5.5244,
      "mean_token_accuracy": 0.18608964532613753,
      "num_tokens": 47233170.0,
      "step": 24145
    },
    {
      "entropy": 6.033581686019898,
      "epoch": 2.4140550807217473,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044200683974251424,
      "loss": 5.5336,
      "mean_token_accuracy": 0.1885179564356804,
      "num_tokens": 47243747.0,
      "step": 24150
    },
    {
      "entropy": 5.9430234909057615,
      "epoch": 2.414554905782976,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004419829216547191,
      "loss": 5.5549,
      "mean_token_accuracy": 0.1833737701177597,
      "num_tokens": 47253372.0,
      "step": 24155
    },
    {
      "entropy": 5.898102045059204,
      "epoch": 2.4150547308442043,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004419589993656767,
      "loss": 5.5479,
      "mean_token_accuracy": 0.19080411940813063,
      "num_tokens": 47263190.0,
      "step": 24160
    },
    {
      "entropy": 6.04074649810791,
      "epoch": 2.415554555905433,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004419350728759888,
      "loss": 5.7157,
      "mean_token_accuracy": 0.1779330775141716,
      "num_tokens": 47273702.0,
      "step": 24165
    },
    {
      "entropy": 6.00181884765625,
      "epoch": 2.4160543809666617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044191114218625736,
      "loss": 5.5187,
      "mean_token_accuracy": 0.18606777340173722,
      "num_tokens": 47282613.0,
      "step": 24170
    },
    {
      "entropy": 5.884095287322998,
      "epoch": 2.4165542060278904,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004418872072970846,
      "loss": 5.4768,
      "mean_token_accuracy": 0.19383601397275924,
      "num_tokens": 47292515.0,
      "step": 24175
    },
    {
      "entropy": 6.020293712615967,
      "epoch": 2.4170540310891186,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004418632682090727,
      "loss": 5.5909,
      "mean_token_accuracy": 0.1769144520163536,
      "num_tokens": 47303258.0,
      "step": 24180
    },
    {
      "entropy": 5.942699432373047,
      "epoch": 2.4175538561503473,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004418393249228238,
      "loss": 5.656,
      "mean_token_accuracy": 0.17229108810424804,
      "num_tokens": 47313544.0,
      "step": 24185
    },
    {
      "entropy": 6.078108263015747,
      "epoch": 2.418053681211576,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004418153774389405,
      "loss": 5.69,
      "mean_token_accuracy": 0.17333139926195146,
      "num_tokens": 47325221.0,
      "step": 24190
    },
    {
      "entropy": 6.018610429763794,
      "epoch": 2.4185535062728043,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004417914257580252,
      "loss": 5.61,
      "mean_token_accuracy": 0.1812483236193657,
      "num_tokens": 47335043.0,
      "step": 24195
    },
    {
      "entropy": 5.986982965469361,
      "epoch": 2.419053331334033,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004417674698806805,
      "loss": 5.5923,
      "mean_token_accuracy": 0.18150073289871216,
      "num_tokens": 47343572.0,
      "step": 24200
    },
    {
      "entropy": 6.019541168212891,
      "epoch": 2.4195531563952617,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004417435098075093,
      "loss": 5.706,
      "mean_token_accuracy": 0.17509280741214753,
      "num_tokens": 47352944.0,
      "step": 24205
    },
    {
      "entropy": 5.971295785903931,
      "epoch": 2.4200529814564904,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004417195455391142,
      "loss": 5.5545,
      "mean_token_accuracy": 0.18024055361747743,
      "num_tokens": 47362614.0,
      "step": 24210
    },
    {
      "entropy": 5.973486661911011,
      "epoch": 2.4205528065177186,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044169557707609826,
      "loss": 5.5631,
      "mean_token_accuracy": 0.1808450251817703,
      "num_tokens": 47372712.0,
      "step": 24215
    },
    {
      "entropy": 5.906800603866577,
      "epoch": 2.4210526315789473,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004416716044190644,
      "loss": 5.6386,
      "mean_token_accuracy": 0.17991985082626344,
      "num_tokens": 47382570.0,
      "step": 24220
    },
    {
      "entropy": 5.899760437011719,
      "epoch": 2.421552456640176,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000441647627568616,
      "loss": 5.554,
      "mean_token_accuracy": 0.18063207268714904,
      "num_tokens": 47392537.0,
      "step": 24225
    },
    {
      "entropy": 5.880550336837769,
      "epoch": 2.4220522817014043,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044162364652535613,
      "loss": 5.5656,
      "mean_token_accuracy": 0.1856301248073578,
      "num_tokens": 47403213.0,
      "step": 24230
    },
    {
      "entropy": 5.990244197845459,
      "epoch": 2.422552106762633,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004415996612898882,
      "loss": 5.575,
      "mean_token_accuracy": 0.176855631172657,
      "num_tokens": 47413832.0,
      "step": 24235
    },
    {
      "entropy": 5.931942367553711,
      "epoch": 2.4230519318238617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044157567186281553,
      "loss": 5.5639,
      "mean_token_accuracy": 0.18609726130962373,
      "num_tokens": 47423051.0,
      "step": 24240
    },
    {
      "entropy": 5.98658332824707,
      "epoch": 2.4235517568850904,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000441551678244742,
      "loss": 5.6849,
      "mean_token_accuracy": 0.17377596497535705,
      "num_tokens": 47431755.0,
      "step": 24245
    },
    {
      "entropy": 6.045824432373047,
      "epoch": 2.4240515819463186,
      "grad_norm": 1.03125,
      "learning_rate": 0.000441527680436271,
      "loss": 5.6242,
      "mean_token_accuracy": 0.1825401172041893,
      "num_tokens": 47441510.0,
      "step": 24250
    },
    {
      "entropy": 5.988466835021972,
      "epoch": 2.4245514070075473,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044150367843800645,
      "loss": 5.5986,
      "mean_token_accuracy": 0.17665716707706453,
      "num_tokens": 47451559.0,
      "step": 24255
    },
    {
      "entropy": 5.936364459991455,
      "epoch": 2.425051232068776,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004414796722505522,
      "loss": 5.5465,
      "mean_token_accuracy": 0.17960110008716584,
      "num_tokens": 47462068.0,
      "step": 24260
    },
    {
      "entropy": 5.939495801925659,
      "epoch": 2.4255510571300043,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004414556618745122,
      "loss": 5.5086,
      "mean_token_accuracy": 0.1816311866044998,
      "num_tokens": 47470684.0,
      "step": 24265
    },
    {
      "entropy": 5.964582300186157,
      "epoch": 2.426050882191233,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044143164731049063,
      "loss": 5.5485,
      "mean_token_accuracy": 0.18553757667541504,
      "num_tokens": 47480147.0,
      "step": 24270
    },
    {
      "entropy": 5.964361572265625,
      "epoch": 2.4265507072524617,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004414076285590916,
      "loss": 5.6067,
      "mean_token_accuracy": 0.1758522182703018,
      "num_tokens": 47489499.0,
      "step": 24275
    },
    {
      "entropy": 5.924182176589966,
      "epoch": 2.4270505323136904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004413836056209195,
      "loss": 5.406,
      "mean_token_accuracy": 0.19716742187738417,
      "num_tokens": 47500155.0,
      "step": 24280
    },
    {
      "entropy": 5.944450569152832,
      "epoch": 2.4275503573749186,
      "grad_norm": 1.125,
      "learning_rate": 0.00044135957849657866,
      "loss": 5.5597,
      "mean_token_accuracy": 0.18460413068532944,
      "num_tokens": 47510302.0,
      "step": 24285
    },
    {
      "entropy": 5.9310221672058105,
      "epoch": 2.4280501824361473,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044133554718667363,
      "loss": 5.585,
      "mean_token_accuracy": 0.17664476931095124,
      "num_tokens": 47519997.0,
      "step": 24290
    },
    {
      "entropy": 6.023407316207885,
      "epoch": 2.428550007497376,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044131151169180905,
      "loss": 5.611,
      "mean_token_accuracy": 0.17236542254686354,
      "num_tokens": 47528417.0,
      "step": 24295
    },
    {
      "entropy": 5.946655511856079,
      "epoch": 2.4290498325586043,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044128747201258956,
      "loss": 5.4924,
      "mean_token_accuracy": 0.18504984974861144,
      "num_tokens": 47538461.0,
      "step": 24300
    },
    {
      "entropy": 5.963157796859742,
      "epoch": 2.429549657619833,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044126342814962015,
      "loss": 5.5409,
      "mean_token_accuracy": 0.1891081765294075,
      "num_tokens": 47547256.0,
      "step": 24305
    },
    {
      "entropy": 5.933267450332641,
      "epoch": 2.4300494826810617,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044123938010350575,
      "loss": 5.6172,
      "mean_token_accuracy": 0.1736145406961441,
      "num_tokens": 47556827.0,
      "step": 24310
    },
    {
      "entropy": 5.978769683837891,
      "epoch": 2.4305493077422904,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044121532787485114,
      "loss": 5.4913,
      "mean_token_accuracy": 0.18539615571498871,
      "num_tokens": 47566600.0,
      "step": 24315
    },
    {
      "entropy": 5.9069887638092045,
      "epoch": 2.4310491328035186,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044119127146426167,
      "loss": 5.4991,
      "mean_token_accuracy": 0.18645391911268233,
      "num_tokens": 47575849.0,
      "step": 24320
    },
    {
      "entropy": 5.998135328292847,
      "epoch": 2.4315489578647473,
      "grad_norm": 1.125,
      "learning_rate": 0.00044116721087234266,
      "loss": 5.671,
      "mean_token_accuracy": 0.1719936713576317,
      "num_tokens": 47586417.0,
      "step": 24325
    },
    {
      "entropy": 6.027742052078247,
      "epoch": 2.432048782925976,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004411431460996994,
      "loss": 5.6429,
      "mean_token_accuracy": 0.17431900277733803,
      "num_tokens": 47596917.0,
      "step": 24330
    },
    {
      "entropy": 6.054064846038818,
      "epoch": 2.4325486079872043,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004411190771469373,
      "loss": 5.6511,
      "mean_token_accuracy": 0.17475556582212448,
      "num_tokens": 47607231.0,
      "step": 24335
    },
    {
      "entropy": 6.00790581703186,
      "epoch": 2.433048433048433,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044109500401466194,
      "loss": 5.6319,
      "mean_token_accuracy": 0.17695571184158326,
      "num_tokens": 47616320.0,
      "step": 24340
    },
    {
      "entropy": 5.962774181365967,
      "epoch": 2.4335482581096617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004410709267034791,
      "loss": 5.5977,
      "mean_token_accuracy": 0.17931319773197174,
      "num_tokens": 47625391.0,
      "step": 24345
    },
    {
      "entropy": 5.970915794372559,
      "epoch": 2.4340480831708904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044104684521399434,
      "loss": 5.6048,
      "mean_token_accuracy": 0.17570944279432296,
      "num_tokens": 47635474.0,
      "step": 24350
    },
    {
      "entropy": 5.998676681518555,
      "epoch": 2.4345479082321186,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004410227595468138,
      "loss": 5.5482,
      "mean_token_accuracy": 0.18625936955213546,
      "num_tokens": 47644860.0,
      "step": 24355
    },
    {
      "entropy": 5.883777141571045,
      "epoch": 2.4350477332933473,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044099866970254327,
      "loss": 5.5766,
      "mean_token_accuracy": 0.1913227006793022,
      "num_tokens": 47654531.0,
      "step": 24360
    },
    {
      "entropy": 5.8606462478637695,
      "epoch": 2.435547558354576,
      "grad_norm": 1.25,
      "learning_rate": 0.00044097457568178903,
      "loss": 5.4611,
      "mean_token_accuracy": 0.19925656020641327,
      "num_tokens": 47663110.0,
      "step": 24365
    },
    {
      "entropy": 5.926569080352783,
      "epoch": 2.4360473834158043,
      "grad_norm": 1.125,
      "learning_rate": 0.00044095047748515706,
      "loss": 5.4846,
      "mean_token_accuracy": 0.19065145999193192,
      "num_tokens": 47672466.0,
      "step": 24370
    },
    {
      "entropy": 6.023855257034302,
      "epoch": 2.436547208477033,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044092637511325384,
      "loss": 5.618,
      "mean_token_accuracy": 0.1832654133439064,
      "num_tokens": 47681949.0,
      "step": 24375
    },
    {
      "entropy": 5.976857089996338,
      "epoch": 2.4370470335382617,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004409022685666857,
      "loss": 5.5972,
      "mean_token_accuracy": 0.18439228236675262,
      "num_tokens": 47692515.0,
      "step": 24380
    },
    {
      "entropy": 5.985017204284668,
      "epoch": 2.4375468585994904,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004408781578460592,
      "loss": 5.6997,
      "mean_token_accuracy": 0.1787578880786896,
      "num_tokens": 47701604.0,
      "step": 24385
    },
    {
      "entropy": 5.901141548156739,
      "epoch": 2.4380466836607186,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044085404295198097,
      "loss": 5.5216,
      "mean_token_accuracy": 0.18577437102794647,
      "num_tokens": 47711567.0,
      "step": 24390
    },
    {
      "entropy": 6.020835304260254,
      "epoch": 2.4385465087219473,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004408299238850576,
      "loss": 5.6471,
      "mean_token_accuracy": 0.17957912981510163,
      "num_tokens": 47721575.0,
      "step": 24395
    },
    {
      "entropy": 6.045673847198486,
      "epoch": 2.439046333783176,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004408058006458961,
      "loss": 5.663,
      "mean_token_accuracy": 0.17762078046798707,
      "num_tokens": 47731510.0,
      "step": 24400
    },
    {
      "entropy": 5.993967056274414,
      "epoch": 2.4395461588444043,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004407816732351032,
      "loss": 5.6833,
      "mean_token_accuracy": 0.17492327094078064,
      "num_tokens": 47741842.0,
      "step": 24405
    },
    {
      "entropy": 6.023004531860352,
      "epoch": 2.440045983905633,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044075754165328614,
      "loss": 5.6595,
      "mean_token_accuracy": 0.1737188935279846,
      "num_tokens": 47752074.0,
      "step": 24410
    },
    {
      "entropy": 6.020928859710693,
      "epoch": 2.4405458089668617,
      "grad_norm": 1.171875,
      "learning_rate": 0.000440733405901052,
      "loss": 5.6029,
      "mean_token_accuracy": 0.1801467701792717,
      "num_tokens": 47760584.0,
      "step": 24415
    },
    {
      "entropy": 5.997047376632691,
      "epoch": 2.4410456340280904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044070926597900804,
      "loss": 5.6567,
      "mean_token_accuracy": 0.17974725365638733,
      "num_tokens": 47770744.0,
      "step": 24420
    },
    {
      "entropy": 6.003288888931275,
      "epoch": 2.4415454590893186,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004406851218877615,
      "loss": 5.65,
      "mean_token_accuracy": 0.17780663222074508,
      "num_tokens": 47780098.0,
      "step": 24425
    },
    {
      "entropy": 5.998757410049438,
      "epoch": 2.4420452841505473,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044066097362792,
      "loss": 5.55,
      "mean_token_accuracy": 0.1854086250066757,
      "num_tokens": 47789807.0,
      "step": 24430
    },
    {
      "entropy": 5.985694169998169,
      "epoch": 2.442545109211776,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000440636821200091,
      "loss": 5.5372,
      "mean_token_accuracy": 0.18868640214204788,
      "num_tokens": 47798110.0,
      "step": 24435
    },
    {
      "entropy": 5.948933076858521,
      "epoch": 2.4430449342730043,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004406126646048822,
      "loss": 5.6213,
      "mean_token_accuracy": 0.17987048178911208,
      "num_tokens": 47809084.0,
      "step": 24440
    },
    {
      "entropy": 6.103757429122925,
      "epoch": 2.443544759334233,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004405885038429014,
      "loss": 5.6302,
      "mean_token_accuracy": 0.17724513113498688,
      "num_tokens": 47819694.0,
      "step": 24445
    },
    {
      "entropy": 6.00100564956665,
      "epoch": 2.4440445843954617,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044056433891475645,
      "loss": 5.6039,
      "mean_token_accuracy": 0.17862383276224136,
      "num_tokens": 47828924.0,
      "step": 24450
    },
    {
      "entropy": 5.861030721664429,
      "epoch": 2.4445444094566904,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004405401698210553,
      "loss": 5.4489,
      "mean_token_accuracy": 0.1946899875998497,
      "num_tokens": 47838992.0,
      "step": 24455
    },
    {
      "entropy": 5.992513179779053,
      "epoch": 2.4450442345179186,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004405159965624061,
      "loss": 5.5377,
      "mean_token_accuracy": 0.18362595438957213,
      "num_tokens": 47848682.0,
      "step": 24460
    },
    {
      "entropy": 6.0230738639831545,
      "epoch": 2.4455440595791473,
      "grad_norm": 1.1875,
      "learning_rate": 0.000440491819139417,
      "loss": 5.572,
      "mean_token_accuracy": 0.18503547310829163,
      "num_tokens": 47858574.0,
      "step": 24465
    },
    {
      "entropy": 5.9502716064453125,
      "epoch": 2.446043884640376,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004404676375526963,
      "loss": 5.6374,
      "mean_token_accuracy": 0.18467056900262832,
      "num_tokens": 47868455.0,
      "step": 24470
    },
    {
      "entropy": 5.891462278366089,
      "epoch": 2.4465437097016043,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044044345180285245,
      "loss": 5.5796,
      "mean_token_accuracy": 0.18325379490852356,
      "num_tokens": 47878718.0,
      "step": 24475
    },
    {
      "entropy": 5.9347984313964846,
      "epoch": 2.447043534762833,
      "grad_norm": 1.15625,
      "learning_rate": 0.000440419261890494,
      "loss": 5.5081,
      "mean_token_accuracy": 0.18618869930505752,
      "num_tokens": 47887444.0,
      "step": 24480
    },
    {
      "entropy": 5.961130809783936,
      "epoch": 2.4475433598240617,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044039506781622936,
      "loss": 5.4847,
      "mean_token_accuracy": 0.19089901447296143,
      "num_tokens": 47895935.0,
      "step": 24485
    },
    {
      "entropy": 5.947934293746949,
      "epoch": 2.4480431848852904,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004403708695806674,
      "loss": 5.5824,
      "mean_token_accuracy": 0.18147898316383362,
      "num_tokens": 47906552.0,
      "step": 24490
    },
    {
      "entropy": 5.904606056213379,
      "epoch": 2.4485430099465186,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004403466671844169,
      "loss": 5.5372,
      "mean_token_accuracy": 0.18731196820735932,
      "num_tokens": 47916691.0,
      "step": 24495
    },
    {
      "entropy": 6.011654615402222,
      "epoch": 2.4490428350077473,
      "grad_norm": 1.125,
      "learning_rate": 0.0004403224606280869,
      "loss": 5.5723,
      "mean_token_accuracy": 0.18366436362266542,
      "num_tokens": 47926579.0,
      "step": 24500
    },
    {
      "entropy": 5.937326717376709,
      "epoch": 2.449542660068976,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004402982499122862,
      "loss": 5.439,
      "mean_token_accuracy": 0.193674498796463,
      "num_tokens": 47935889.0,
      "step": 24505
    },
    {
      "entropy": 5.923959255218506,
      "epoch": 2.4500424851302043,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044027403503762416,
      "loss": 5.5378,
      "mean_token_accuracy": 0.18383606225252153,
      "num_tokens": 47944993.0,
      "step": 24510
    },
    {
      "entropy": 6.057095098495483,
      "epoch": 2.450542310191433,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004402498160047099,
      "loss": 5.6252,
      "mean_token_accuracy": 0.17752402126789094,
      "num_tokens": 47954984.0,
      "step": 24515
    },
    {
      "entropy": 5.965072107315064,
      "epoch": 2.4510421352526617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004402255928141528,
      "loss": 5.6017,
      "mean_token_accuracy": 0.18049367815256118,
      "num_tokens": 47964986.0,
      "step": 24520
    },
    {
      "entropy": 5.992988109588623,
      "epoch": 2.4515419603138904,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004402013654665623,
      "loss": 5.6183,
      "mean_token_accuracy": 0.1796846404671669,
      "num_tokens": 47975962.0,
      "step": 24525
    },
    {
      "entropy": 6.038974761962891,
      "epoch": 2.4520417853751186,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044017713396254796,
      "loss": 5.6726,
      "mean_token_accuracy": 0.17596176117658616,
      "num_tokens": 47985159.0,
      "step": 24530
    },
    {
      "entropy": 5.959385013580322,
      "epoch": 2.4525416104363473,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044015289830271945,
      "loss": 5.522,
      "mean_token_accuracy": 0.1869758576154709,
      "num_tokens": 47993634.0,
      "step": 24535
    },
    {
      "entropy": 5.95837140083313,
      "epoch": 2.453041435497576,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044012865848768644,
      "loss": 5.5613,
      "mean_token_accuracy": 0.18212526589632033,
      "num_tokens": 48003573.0,
      "step": 24540
    },
    {
      "entropy": 5.88453311920166,
      "epoch": 2.4535412605588043,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044010441451805896,
      "loss": 5.4487,
      "mean_token_accuracy": 0.19031767100095748,
      "num_tokens": 48013688.0,
      "step": 24545
    },
    {
      "entropy": 5.948508262634277,
      "epoch": 2.454041085620033,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004400801663944468,
      "loss": 5.4334,
      "mean_token_accuracy": 0.190226773917675,
      "num_tokens": 48022629.0,
      "step": 24550
    },
    {
      "entropy": 5.981168365478515,
      "epoch": 2.4545409106812617,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044005591411746023,
      "loss": 5.6046,
      "mean_token_accuracy": 0.1744113892316818,
      "num_tokens": 48033495.0,
      "step": 24555
    },
    {
      "entropy": 5.934166812896729,
      "epoch": 2.4550407357424904,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044003165768770927,
      "loss": 5.4948,
      "mean_token_accuracy": 0.1882937267422676,
      "num_tokens": 48043471.0,
      "step": 24560
    },
    {
      "entropy": 5.995825576782226,
      "epoch": 2.4555405608037186,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004400073971058043,
      "loss": 5.6209,
      "mean_token_accuracy": 0.18151511549949645,
      "num_tokens": 48052811.0,
      "step": 24565
    },
    {
      "entropy": 5.992266368865967,
      "epoch": 2.4560403858649473,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043998313237235566,
      "loss": 5.5783,
      "mean_token_accuracy": 0.1799368903040886,
      "num_tokens": 48062826.0,
      "step": 24570
    },
    {
      "entropy": 5.982863712310791,
      "epoch": 2.456540210926176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004399588634879739,
      "loss": 5.579,
      "mean_token_accuracy": 0.18128301054239274,
      "num_tokens": 48072867.0,
      "step": 24575
    },
    {
      "entropy": 5.962683439254761,
      "epoch": 2.4570400359874043,
      "grad_norm": 1.125,
      "learning_rate": 0.0004399345904532695,
      "loss": 5.5641,
      "mean_token_accuracy": 0.18200048059225082,
      "num_tokens": 48082816.0,
      "step": 24580
    },
    {
      "entropy": 5.947390699386597,
      "epoch": 2.457539861048633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004399103132688533,
      "loss": 5.5746,
      "mean_token_accuracy": 0.18869741261005402,
      "num_tokens": 48092715.0,
      "step": 24585
    },
    {
      "entropy": 6.006002807617188,
      "epoch": 2.4580396861098617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043988603193533606,
      "loss": 5.5537,
      "mean_token_accuracy": 0.17882937341928482,
      "num_tokens": 48101372.0,
      "step": 24590
    },
    {
      "entropy": 5.976867055892944,
      "epoch": 2.45853951117109,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004398617464533287,
      "loss": 5.5634,
      "mean_token_accuracy": 0.18811026215553284,
      "num_tokens": 48111894.0,
      "step": 24595
    },
    {
      "entropy": 6.04946460723877,
      "epoch": 2.4590393362323186,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004398374568234422,
      "loss": 5.7014,
      "mean_token_accuracy": 0.16387065947055818,
      "num_tokens": 48121480.0,
      "step": 24600
    },
    {
      "entropy": 5.998653554916382,
      "epoch": 2.4595391612935473,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004398131630462877,
      "loss": 5.5852,
      "mean_token_accuracy": 0.18026969879865645,
      "num_tokens": 48130404.0,
      "step": 24605
    },
    {
      "entropy": 6.013297128677368,
      "epoch": 2.4600389863547756,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043978886512247646,
      "loss": 5.566,
      "mean_token_accuracy": 0.17894702553749084,
      "num_tokens": 48141218.0,
      "step": 24610
    },
    {
      "entropy": 6.01448130607605,
      "epoch": 2.4605388114160043,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004397645630526197,
      "loss": 5.607,
      "mean_token_accuracy": 0.1805765524506569,
      "num_tokens": 48151817.0,
      "step": 24615
    },
    {
      "entropy": 6.006993770599365,
      "epoch": 2.461038636477233,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000439740256837329,
      "loss": 5.5181,
      "mean_token_accuracy": 0.1871412217617035,
      "num_tokens": 48161044.0,
      "step": 24620
    },
    {
      "entropy": 5.973826360702515,
      "epoch": 2.4615384615384617,
      "grad_norm": 1.125,
      "learning_rate": 0.0004397159464772158,
      "loss": 5.6091,
      "mean_token_accuracy": 0.17860744446516036,
      "num_tokens": 48170139.0,
      "step": 24625
    },
    {
      "entropy": 5.907918548583984,
      "epoch": 2.46203828659969,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043969163197289174,
      "loss": 5.6086,
      "mean_token_accuracy": 0.18204981535673143,
      "num_tokens": 48179418.0,
      "step": 24630
    },
    {
      "entropy": 6.023828411102295,
      "epoch": 2.4625381116609186,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004396673133249687,
      "loss": 5.5546,
      "mean_token_accuracy": 0.18156721144914628,
      "num_tokens": 48188877.0,
      "step": 24635
    },
    {
      "entropy": 6.0282446384429935,
      "epoch": 2.4630379367221473,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004396429905340584,
      "loss": 5.6489,
      "mean_token_accuracy": 0.17159997522830964,
      "num_tokens": 48198969.0,
      "step": 24640
    },
    {
      "entropy": 6.079281234741211,
      "epoch": 2.4635377617833756,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043961866360077266,
      "loss": 5.6874,
      "mean_token_accuracy": 0.17310007512569428,
      "num_tokens": 48209814.0,
      "step": 24645
    },
    {
      "entropy": 6.003655290603637,
      "epoch": 2.4640375868446043,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004395943325257239,
      "loss": 5.582,
      "mean_token_accuracy": 0.18631424605846406,
      "num_tokens": 48218810.0,
      "step": 24650
    },
    {
      "entropy": 5.953561496734619,
      "epoch": 2.464537411905833,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043956999730952403,
      "loss": 5.5521,
      "mean_token_accuracy": 0.18263006210327148,
      "num_tokens": 48228484.0,
      "step": 24655
    },
    {
      "entropy": 5.958230113983154,
      "epoch": 2.4650372369670617,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043954565795278536,
      "loss": 5.6017,
      "mean_token_accuracy": 0.18470973670482635,
      "num_tokens": 48239029.0,
      "step": 24660
    },
    {
      "entropy": 5.99581561088562,
      "epoch": 2.46553706202829,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004395213144561203,
      "loss": 5.5599,
      "mean_token_accuracy": 0.18455136120319365,
      "num_tokens": 48249448.0,
      "step": 24665
    },
    {
      "entropy": 6.011719846725464,
      "epoch": 2.4660368870895186,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004394969668201413,
      "loss": 5.5519,
      "mean_token_accuracy": 0.18491209596395491,
      "num_tokens": 48258631.0,
      "step": 24670
    },
    {
      "entropy": 5.914530563354492,
      "epoch": 2.4665367121507473,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004394726150454609,
      "loss": 5.4349,
      "mean_token_accuracy": 0.19050201177597045,
      "num_tokens": 48266800.0,
      "step": 24675
    },
    {
      "entropy": 6.003184700012207,
      "epoch": 2.4670365372119756,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004394482591326918,
      "loss": 5.6303,
      "mean_token_accuracy": 0.1884240210056305,
      "num_tokens": 48277566.0,
      "step": 24680
    },
    {
      "entropy": 5.965236949920654,
      "epoch": 2.4675363622732043,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043942389908244693,
      "loss": 5.5717,
      "mean_token_accuracy": 0.1795808717608452,
      "num_tokens": 48287033.0,
      "step": 24685
    },
    {
      "entropy": 5.966031789779663,
      "epoch": 2.468036187334433,
      "grad_norm": 1.078125,
      "learning_rate": 0.000439399534895339,
      "loss": 5.5583,
      "mean_token_accuracy": 0.18299792110919952,
      "num_tokens": 48296627.0,
      "step": 24690
    },
    {
      "entropy": 6.010075902938842,
      "epoch": 2.4685360123956617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004393751665719811,
      "loss": 5.5607,
      "mean_token_accuracy": 0.18146976232528686,
      "num_tokens": 48305940.0,
      "step": 24695
    },
    {
      "entropy": 5.926075220108032,
      "epoch": 2.46903583745689,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004393507941129863,
      "loss": 5.532,
      "mean_token_accuracy": 0.17899715006351472,
      "num_tokens": 48315785.0,
      "step": 24700
    },
    {
      "entropy": 5.942726278305054,
      "epoch": 2.4695356625181186,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004393264175189678,
      "loss": 5.5628,
      "mean_token_accuracy": 0.19021676927804948,
      "num_tokens": 48325803.0,
      "step": 24705
    },
    {
      "entropy": 6.042229318618775,
      "epoch": 2.4700354875793473,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043930203679053885,
      "loss": 5.5556,
      "mean_token_accuracy": 0.18571594059467317,
      "num_tokens": 48334860.0,
      "step": 24710
    },
    {
      "entropy": 5.9907143115997314,
      "epoch": 2.4705353126405756,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043927765192831305,
      "loss": 5.6095,
      "mean_token_accuracy": 0.17560015618801117,
      "num_tokens": 48344939.0,
      "step": 24715
    },
    {
      "entropy": 5.917420434951782,
      "epoch": 2.4710351377018043,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004392532629329038,
      "loss": 5.5799,
      "mean_token_accuracy": 0.18011901825666427,
      "num_tokens": 48354493.0,
      "step": 24720
    },
    {
      "entropy": 5.892473840713501,
      "epoch": 2.471534962763033,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043922886980492465,
      "loss": 5.564,
      "mean_token_accuracy": 0.18549802154302597,
      "num_tokens": 48364650.0,
      "step": 24725
    },
    {
      "entropy": 5.986869287490845,
      "epoch": 2.4720347878242617,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043920447254498947,
      "loss": 5.644,
      "mean_token_accuracy": 0.18027642816305162,
      "num_tokens": 48374373.0,
      "step": 24730
    },
    {
      "entropy": 5.967944049835205,
      "epoch": 2.47253461288549,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004391800711537119,
      "loss": 5.5151,
      "mean_token_accuracy": 0.1825350731611252,
      "num_tokens": 48384488.0,
      "step": 24735
    },
    {
      "entropy": 6.037070417404175,
      "epoch": 2.4730344379467186,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043915566563170605,
      "loss": 5.6234,
      "mean_token_accuracy": 0.1792533203959465,
      "num_tokens": 48395083.0,
      "step": 24740
    },
    {
      "entropy": 5.912010335922242,
      "epoch": 2.4735342630079473,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043913125597958594,
      "loss": 5.5135,
      "mean_token_accuracy": 0.18921898752450944,
      "num_tokens": 48404229.0,
      "step": 24745
    },
    {
      "entropy": 5.969352388381958,
      "epoch": 2.4740340880691756,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043910684219796564,
      "loss": 5.6222,
      "mean_token_accuracy": 0.17901528179645537,
      "num_tokens": 48413521.0,
      "step": 24750
    },
    {
      "entropy": 5.963640356063843,
      "epoch": 2.4745339131304043,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043908242428745937,
      "loss": 5.5488,
      "mean_token_accuracy": 0.19231082797050475,
      "num_tokens": 48423229.0,
      "step": 24755
    },
    {
      "entropy": 6.013575315475464,
      "epoch": 2.475033738191633,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043905800224868153,
      "loss": 5.6334,
      "mean_token_accuracy": 0.17898086756467818,
      "num_tokens": 48432261.0,
      "step": 24760
    },
    {
      "entropy": 6.027656745910645,
      "epoch": 2.4755335632528617,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004390335760822466,
      "loss": 5.5592,
      "mean_token_accuracy": 0.18168837428092957,
      "num_tokens": 48441484.0,
      "step": 24765
    },
    {
      "entropy": 5.965752649307251,
      "epoch": 2.47603338831409,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004390091457887691,
      "loss": 5.4951,
      "mean_token_accuracy": 0.19565867632627487,
      "num_tokens": 48452055.0,
      "step": 24770
    },
    {
      "entropy": 5.872251176834107,
      "epoch": 2.4765332133753186,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004389847113688637,
      "loss": 5.3982,
      "mean_token_accuracy": 0.19505329579114913,
      "num_tokens": 48460349.0,
      "step": 24775
    },
    {
      "entropy": 5.896726417541504,
      "epoch": 2.4770330384365473,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043896027282314516,
      "loss": 5.5737,
      "mean_token_accuracy": 0.1877967208623886,
      "num_tokens": 48470625.0,
      "step": 24780
    },
    {
      "entropy": 5.974989557266236,
      "epoch": 2.4775328634977756,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004389358301522283,
      "loss": 5.547,
      "mean_token_accuracy": 0.18224262446165085,
      "num_tokens": 48480866.0,
      "step": 24785
    },
    {
      "entropy": 5.89061918258667,
      "epoch": 2.4780326885590043,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043891138335672825,
      "loss": 5.5184,
      "mean_token_accuracy": 0.18754566460847855,
      "num_tokens": 48490484.0,
      "step": 24790
    },
    {
      "entropy": 5.9380346775054935,
      "epoch": 2.478532513620233,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004388869324372599,
      "loss": 5.5398,
      "mean_token_accuracy": 0.18390410989522935,
      "num_tokens": 48501139.0,
      "step": 24795
    },
    {
      "entropy": 6.045819187164307,
      "epoch": 2.4790323386814617,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004388624773944385,
      "loss": 5.6773,
      "mean_token_accuracy": 0.17936634570360183,
      "num_tokens": 48510793.0,
      "step": 24800
    },
    {
      "entropy": 6.102351903915405,
      "epoch": 2.47953216374269,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043883801822887933,
      "loss": 5.724,
      "mean_token_accuracy": 0.1730061411857605,
      "num_tokens": 48521193.0,
      "step": 24805
    },
    {
      "entropy": 5.992374467849731,
      "epoch": 2.4800319888039186,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043881355494119773,
      "loss": 5.5441,
      "mean_token_accuracy": 0.18996454030275345,
      "num_tokens": 48530303.0,
      "step": 24810
    },
    {
      "entropy": 6.007930564880371,
      "epoch": 2.4805318138651473,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004387890875320093,
      "loss": 5.6127,
      "mean_token_accuracy": 0.1800522767007351,
      "num_tokens": 48539633.0,
      "step": 24815
    },
    {
      "entropy": 6.026777124404907,
      "epoch": 2.4810316389263756,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043876461600192955,
      "loss": 5.5545,
      "mean_token_accuracy": 0.19397071450948716,
      "num_tokens": 48550097.0,
      "step": 24820
    },
    {
      "entropy": 5.907761907577514,
      "epoch": 2.4815314639876043,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004387401403515743,
      "loss": 5.5717,
      "mean_token_accuracy": 0.18925828486680984,
      "num_tokens": 48561202.0,
      "step": 24825
    },
    {
      "entropy": 6.020737648010254,
      "epoch": 2.482031289048833,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043871566058155913,
      "loss": 5.6496,
      "mean_token_accuracy": 0.1730610966682434,
      "num_tokens": 48571283.0,
      "step": 24830
    },
    {
      "entropy": 6.058292293548584,
      "epoch": 2.4825311141100617,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043869117669250015,
      "loss": 5.6442,
      "mean_token_accuracy": 0.1707686424255371,
      "num_tokens": 48581818.0,
      "step": 24835
    },
    {
      "entropy": 5.982869863510132,
      "epoch": 2.48303093917129,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004386666886850132,
      "loss": 5.5922,
      "mean_token_accuracy": 0.18088561296463013,
      "num_tokens": 48591227.0,
      "step": 24840
    },
    {
      "entropy": 6.0513814926147464,
      "epoch": 2.4835307642325186,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043864219655971457,
      "loss": 5.6046,
      "mean_token_accuracy": 0.1755698725581169,
      "num_tokens": 48600704.0,
      "step": 24845
    },
    {
      "entropy": 6.002162885665894,
      "epoch": 2.4840305892937473,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004386177003172203,
      "loss": 5.6878,
      "mean_token_accuracy": 0.18272668719291688,
      "num_tokens": 48611964.0,
      "step": 24850
    },
    {
      "entropy": 6.036138343811035,
      "epoch": 2.4845304143549756,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043859319995814686,
      "loss": 5.6147,
      "mean_token_accuracy": 0.18313733637332916,
      "num_tokens": 48621012.0,
      "step": 24855
    },
    {
      "entropy": 5.947992563247681,
      "epoch": 2.4850302394162043,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004385686954831105,
      "loss": 5.5373,
      "mean_token_accuracy": 0.18718994706869124,
      "num_tokens": 48630242.0,
      "step": 24860
    },
    {
      "entropy": 5.896310615539551,
      "epoch": 2.485530064477433,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043854418689272793,
      "loss": 5.5104,
      "mean_token_accuracy": 0.1856616646051407,
      "num_tokens": 48640535.0,
      "step": 24865
    },
    {
      "entropy": 6.07455415725708,
      "epoch": 2.4860298895386617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043851967418761574,
      "loss": 5.6275,
      "mean_token_accuracy": 0.18229883164167404,
      "num_tokens": 48650096.0,
      "step": 24870
    },
    {
      "entropy": 5.96376600265503,
      "epoch": 2.48652971459989,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043849515736839057,
      "loss": 5.5705,
      "mean_token_accuracy": 0.184955832362175,
      "num_tokens": 48660883.0,
      "step": 24875
    },
    {
      "entropy": 6.016132926940918,
      "epoch": 2.4870295396611186,
      "grad_norm": 1.125,
      "learning_rate": 0.00043847063643566935,
      "loss": 5.6403,
      "mean_token_accuracy": 0.17767163515090942,
      "num_tokens": 48671345.0,
      "step": 24880
    },
    {
      "entropy": 6.048795461654663,
      "epoch": 2.4875293647223473,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00043844611139006893,
      "loss": 5.6245,
      "mean_token_accuracy": 0.17782227396965028,
      "num_tokens": 48681296.0,
      "step": 24885
    },
    {
      "entropy": 5.948370504379272,
      "epoch": 2.4880291897835756,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043842158223220644,
      "loss": 5.5985,
      "mean_token_accuracy": 0.18132154792547225,
      "num_tokens": 48691885.0,
      "step": 24890
    },
    {
      "entropy": 6.016615915298462,
      "epoch": 2.4885290148448043,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000438397048962699,
      "loss": 5.5884,
      "mean_token_accuracy": 0.17282411754131316,
      "num_tokens": 48701325.0,
      "step": 24895
    },
    {
      "entropy": 6.016634511947632,
      "epoch": 2.489028839906033,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00043837251158216386,
      "loss": 5.6481,
      "mean_token_accuracy": 0.18129211366176606,
      "num_tokens": 48712212.0,
      "step": 24900
    },
    {
      "entropy": 5.995718717575073,
      "epoch": 2.4895286649672617,
      "grad_norm": 1.125,
      "learning_rate": 0.00043834797009121836,
      "loss": 5.5636,
      "mean_token_accuracy": 0.18633929640054703,
      "num_tokens": 48721482.0,
      "step": 24905
    },
    {
      "entropy": 5.994885587692261,
      "epoch": 2.49002849002849,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043832342449047994,
      "loss": 5.6157,
      "mean_token_accuracy": 0.1814814791083336,
      "num_tokens": 48731740.0,
      "step": 24910
    },
    {
      "entropy": 5.9626868724822994,
      "epoch": 2.4905283150897186,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004382988747805661,
      "loss": 5.571,
      "mean_token_accuracy": 0.18355715423822402,
      "num_tokens": 48740351.0,
      "step": 24915
    },
    {
      "entropy": 5.913273620605469,
      "epoch": 2.4910281401509473,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004382743209620948,
      "loss": 5.5434,
      "mean_token_accuracy": 0.1836872801184654,
      "num_tokens": 48750571.0,
      "step": 24920
    },
    {
      "entropy": 5.947555112838745,
      "epoch": 2.4915279652121756,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004382497630356835,
      "loss": 5.5215,
      "mean_token_accuracy": 0.19037104099988938,
      "num_tokens": 48758740.0,
      "step": 24925
    },
    {
      "entropy": 5.995624351501465,
      "epoch": 2.4920277902734043,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043822520100195015,
      "loss": 5.5417,
      "mean_token_accuracy": 0.18195019364356996,
      "num_tokens": 48768469.0,
      "step": 24930
    },
    {
      "entropy": 5.964212036132812,
      "epoch": 2.492527615334633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043820063486151267,
      "loss": 5.4823,
      "mean_token_accuracy": 0.1873995453119278,
      "num_tokens": 48778142.0,
      "step": 24935
    },
    {
      "entropy": 5.916059970855713,
      "epoch": 2.4930274403958617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043817606461498933,
      "loss": 5.5233,
      "mean_token_accuracy": 0.1851300746202469,
      "num_tokens": 48787224.0,
      "step": 24940
    },
    {
      "entropy": 6.067304611206055,
      "epoch": 2.49352726545709,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043815149026299823,
      "loss": 5.6845,
      "mean_token_accuracy": 0.17051089704036712,
      "num_tokens": 48797098.0,
      "step": 24945
    },
    {
      "entropy": 6.0099952697753904,
      "epoch": 2.4940270905183186,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004381269118061575,
      "loss": 5.5742,
      "mean_token_accuracy": 0.18314881175756453,
      "num_tokens": 48806764.0,
      "step": 24950
    },
    {
      "entropy": 6.00052056312561,
      "epoch": 2.4945269155795473,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004381023292450857,
      "loss": 5.6186,
      "mean_token_accuracy": 0.18379699438810349,
      "num_tokens": 48818586.0,
      "step": 24955
    },
    {
      "entropy": 6.025336408615113,
      "epoch": 2.4950267406407756,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004380777425804012,
      "loss": 5.6065,
      "mean_token_accuracy": 0.18233540952205657,
      "num_tokens": 48829229.0,
      "step": 24960
    },
    {
      "entropy": 6.031009006500244,
      "epoch": 2.4955265657020043,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043805315181272275,
      "loss": 5.5661,
      "mean_token_accuracy": 0.18447521030902864,
      "num_tokens": 48838118.0,
      "step": 24965
    },
    {
      "entropy": 6.052378606796265,
      "epoch": 2.496026390763233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004380285569426689,
      "loss": 5.6856,
      "mean_token_accuracy": 0.17620655745267869,
      "num_tokens": 48847607.0,
      "step": 24970
    },
    {
      "entropy": 5.9264367580413815,
      "epoch": 2.4965262158244617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004380039579708586,
      "loss": 5.5036,
      "mean_token_accuracy": 0.18211423456668854,
      "num_tokens": 48856857.0,
      "step": 24975
    },
    {
      "entropy": 6.006782531738281,
      "epoch": 2.49702604088569,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004379793548979105,
      "loss": 5.5883,
      "mean_token_accuracy": 0.17503691464662552,
      "num_tokens": 48866145.0,
      "step": 24980
    },
    {
      "entropy": 6.014829492568969,
      "epoch": 2.4975258659469186,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004379547477244439,
      "loss": 5.6441,
      "mean_token_accuracy": 0.17605670243501664,
      "num_tokens": 48875941.0,
      "step": 24985
    },
    {
      "entropy": 5.942105770111084,
      "epoch": 2.4980256910081473,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004379301364510777,
      "loss": 5.4887,
      "mean_token_accuracy": 0.19092095345258714,
      "num_tokens": 48885870.0,
      "step": 24990
    },
    {
      "entropy": 5.999629068374634,
      "epoch": 2.4985255160693756,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004379055210784313,
      "loss": 5.6498,
      "mean_token_accuracy": 0.1712847411632538,
      "num_tokens": 48896358.0,
      "step": 24995
    },
    {
      "entropy": 5.976758241653442,
      "epoch": 2.4990253411306043,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004378809016071238,
      "loss": 5.5868,
      "mean_token_accuracy": 0.1823691114783287,
      "num_tokens": 48905397.0,
      "step": 25000
    },
    {
      "entropy": 6.021623134613037,
      "epoch": 2.499525166191833,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004378562780377748,
      "loss": 5.5979,
      "mean_token_accuracy": 0.18163074404001237,
      "num_tokens": 48914385.0,
      "step": 25005
    },
    {
      "entropy": 5.962156152725219,
      "epoch": 2.5000249912530617,
      "grad_norm": 1.25,
      "learning_rate": 0.0004378316503710037,
      "loss": 5.517,
      "mean_token_accuracy": 0.18357904255390167,
      "num_tokens": 48923515.0,
      "step": 25010
    },
    {
      "entropy": 5.947617101669311,
      "epoch": 2.50052481631429,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004378070186074302,
      "loss": 5.5957,
      "mean_token_accuracy": 0.18134247809648513,
      "num_tokens": 48933210.0,
      "step": 25015
    },
    {
      "entropy": 5.985237550735474,
      "epoch": 2.5010246413755186,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043778238274767395,
      "loss": 5.6044,
      "mean_token_accuracy": 0.17628576159477233,
      "num_tokens": 48944292.0,
      "step": 25020
    },
    {
      "entropy": 5.9809599876403805,
      "epoch": 2.501524466436747,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004377577427923549,
      "loss": 5.4688,
      "mean_token_accuracy": 0.18583004623651506,
      "num_tokens": 48953883.0,
      "step": 25025
    },
    {
      "entropy": 5.9578112125396725,
      "epoch": 2.5020242914979756,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043773309874209294,
      "loss": 5.5154,
      "mean_token_accuracy": 0.17986889481544494,
      "num_tokens": 48963779.0,
      "step": 25030
    },
    {
      "entropy": 6.031283807754517,
      "epoch": 2.5025241165592043,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043770845059750807,
      "loss": 5.5973,
      "mean_token_accuracy": 0.18411757946014404,
      "num_tokens": 48972821.0,
      "step": 25035
    },
    {
      "entropy": 6.013525724411011,
      "epoch": 2.503023941620433,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043768379835922044,
      "loss": 5.6324,
      "mean_token_accuracy": 0.17605073004961014,
      "num_tokens": 48984288.0,
      "step": 25040
    },
    {
      "entropy": 6.0336730003356935,
      "epoch": 2.5035237666816617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004376591420278503,
      "loss": 5.6736,
      "mean_token_accuracy": 0.16901442855596543,
      "num_tokens": 48994225.0,
      "step": 25045
    },
    {
      "entropy": 6.0096947193145756,
      "epoch": 2.50402359174289,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000437634481604018,
      "loss": 5.5406,
      "mean_token_accuracy": 0.17933988571166992,
      "num_tokens": 49004203.0,
      "step": 25050
    },
    {
      "entropy": 5.9671532154083256,
      "epoch": 2.5045234168041186,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000437609817088344,
      "loss": 5.5688,
      "mean_token_accuracy": 0.17545448690652848,
      "num_tokens": 49014201.0,
      "step": 25055
    },
    {
      "entropy": 6.030357503890992,
      "epoch": 2.505023241865347,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043758514848144886,
      "loss": 5.667,
      "mean_token_accuracy": 0.18060954362154008,
      "num_tokens": 49022996.0,
      "step": 25060
    },
    {
      "entropy": 6.034258127212524,
      "epoch": 2.5055230669265756,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004375604757839532,
      "loss": 5.619,
      "mean_token_accuracy": 0.17562198638916016,
      "num_tokens": 49033040.0,
      "step": 25065
    },
    {
      "entropy": 5.945316028594971,
      "epoch": 2.5060228919878043,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043753579899647783,
      "loss": 5.6171,
      "mean_token_accuracy": 0.1802961438894272,
      "num_tokens": 49043389.0,
      "step": 25070
    },
    {
      "entropy": 5.90709228515625,
      "epoch": 2.506522717049033,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043751111811964353,
      "loss": 5.4788,
      "mean_token_accuracy": 0.18838610947132112,
      "num_tokens": 49052558.0,
      "step": 25075
    },
    {
      "entropy": 6.027747058868409,
      "epoch": 2.5070225421102617,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004374864331540713,
      "loss": 5.7289,
      "mean_token_accuracy": 0.16859797686338424,
      "num_tokens": 49062911.0,
      "step": 25080
    },
    {
      "entropy": 5.9846587657928465,
      "epoch": 2.50752236717149,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004374617441003823,
      "loss": 5.5922,
      "mean_token_accuracy": 0.17792488634586334,
      "num_tokens": 49072411.0,
      "step": 25085
    },
    {
      "entropy": 6.015269136428833,
      "epoch": 2.5080221922327186,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004374370509591976,
      "loss": 5.5554,
      "mean_token_accuracy": 0.18485482931137084,
      "num_tokens": 49082568.0,
      "step": 25090
    },
    {
      "entropy": 5.817141199111939,
      "epoch": 2.508522017293947,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043741235373113854,
      "loss": 5.4256,
      "mean_token_accuracy": 0.1984498992562294,
      "num_tokens": 49092015.0,
      "step": 25095
    },
    {
      "entropy": 5.938234043121338,
      "epoch": 2.5090218423551756,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043738765241682637,
      "loss": 5.5517,
      "mean_token_accuracy": 0.1843581184744835,
      "num_tokens": 49102877.0,
      "step": 25100
    },
    {
      "entropy": 6.006336641311646,
      "epoch": 2.5095216674164043,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004373629470168826,
      "loss": 5.5317,
      "mean_token_accuracy": 0.1821967050433159,
      "num_tokens": 49112227.0,
      "step": 25105
    },
    {
      "entropy": 5.913351154327392,
      "epoch": 2.510021492477633,
      "grad_norm": 1.109375,
      "learning_rate": 0.000437338237531929,
      "loss": 5.5224,
      "mean_token_accuracy": 0.18678680807352066,
      "num_tokens": 49122840.0,
      "step": 25110
    },
    {
      "entropy": 5.927005767822266,
      "epoch": 2.5105213175388617,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004373135239625871,
      "loss": 5.6587,
      "mean_token_accuracy": 0.17376581132411956,
      "num_tokens": 49134067.0,
      "step": 25115
    },
    {
      "entropy": 5.972874546051026,
      "epoch": 2.51102114260009,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043728880630947864,
      "loss": 5.5509,
      "mean_token_accuracy": 0.1855858951807022,
      "num_tokens": 49144176.0,
      "step": 25120
    },
    {
      "entropy": 6.006255197525024,
      "epoch": 2.5115209676613186,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043726408457322557,
      "loss": 5.5866,
      "mean_token_accuracy": 0.18029394149780273,
      "num_tokens": 49153559.0,
      "step": 25125
    },
    {
      "entropy": 6.0099982738494875,
      "epoch": 2.512020792722547,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043723935875445,
      "loss": 5.534,
      "mean_token_accuracy": 0.18336892426013945,
      "num_tokens": 49164604.0,
      "step": 25130
    },
    {
      "entropy": 6.006793642044068,
      "epoch": 2.5125206177837756,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004372146288537738,
      "loss": 5.56,
      "mean_token_accuracy": 0.18040040582418443,
      "num_tokens": 49173766.0,
      "step": 25135
    },
    {
      "entropy": 5.950758457183838,
      "epoch": 2.5130204428450043,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043718989487181927,
      "loss": 5.5708,
      "mean_token_accuracy": 0.18999089151620865,
      "num_tokens": 49183244.0,
      "step": 25140
    },
    {
      "entropy": 5.938399696350098,
      "epoch": 2.513520267906233,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004371651568092088,
      "loss": 5.6017,
      "mean_token_accuracy": 0.1820274829864502,
      "num_tokens": 49194256.0,
      "step": 25145
    },
    {
      "entropy": 5.990288209915161,
      "epoch": 2.5140200929674617,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004371404146665647,
      "loss": 5.618,
      "mean_token_accuracy": 0.18025673031806946,
      "num_tokens": 49204680.0,
      "step": 25150
    },
    {
      "entropy": 6.049222707748413,
      "epoch": 2.51451991802869,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043711566844450934,
      "loss": 5.5751,
      "mean_token_accuracy": 0.17624234557151794,
      "num_tokens": 49213918.0,
      "step": 25155
    },
    {
      "entropy": 5.967915630340576,
      "epoch": 2.5150197430899186,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043709091814366565,
      "loss": 5.4912,
      "mean_token_accuracy": 0.18530732840299607,
      "num_tokens": 49223087.0,
      "step": 25160
    },
    {
      "entropy": 5.896346521377564,
      "epoch": 2.515519568151147,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043706616376465614,
      "loss": 5.4366,
      "mean_token_accuracy": 0.1923763632774353,
      "num_tokens": 49232072.0,
      "step": 25165
    },
    {
      "entropy": 5.918233251571655,
      "epoch": 2.5160193932123756,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004370414053081036,
      "loss": 5.5153,
      "mean_token_accuracy": 0.18734763711690902,
      "num_tokens": 49241772.0,
      "step": 25170
    },
    {
      "entropy": 5.999624633789063,
      "epoch": 2.5165192182736043,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000437016642774631,
      "loss": 5.5638,
      "mean_token_accuracy": 0.17954811900854112,
      "num_tokens": 49250605.0,
      "step": 25175
    },
    {
      "entropy": 5.9773749828338625,
      "epoch": 2.517019043334833,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043699187616486144,
      "loss": 5.6253,
      "mean_token_accuracy": 0.1844974085688591,
      "num_tokens": 49261104.0,
      "step": 25180
    },
    {
      "entropy": 6.070811128616333,
      "epoch": 2.5175188683960616,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004369671054794178,
      "loss": 5.7893,
      "mean_token_accuracy": 0.17657170966267585,
      "num_tokens": 49271456.0,
      "step": 25185
    },
    {
      "entropy": 6.017111015319824,
      "epoch": 2.51801869345729,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043694233071892366,
      "loss": 5.5444,
      "mean_token_accuracy": 0.18431718051433563,
      "num_tokens": 49279917.0,
      "step": 25190
    },
    {
      "entropy": 5.9029152393341064,
      "epoch": 2.5185185185185186,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043691755188400203,
      "loss": 5.5824,
      "mean_token_accuracy": 0.180727456510067,
      "num_tokens": 49290961.0,
      "step": 25195
    },
    {
      "entropy": 5.95005693435669,
      "epoch": 2.519018343579747,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004368927689752765,
      "loss": 5.5886,
      "mean_token_accuracy": 0.1812599018216133,
      "num_tokens": 49299813.0,
      "step": 25200
    },
    {
      "entropy": 5.987655878067017,
      "epoch": 2.5195181686409756,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043686798199337053,
      "loss": 5.5784,
      "mean_token_accuracy": 0.18776730597019195,
      "num_tokens": 49309041.0,
      "step": 25205
    },
    {
      "entropy": 5.947041749954224,
      "epoch": 2.5200179937022043,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043684319093890777,
      "loss": 5.5048,
      "mean_token_accuracy": 0.18541211485862732,
      "num_tokens": 49318647.0,
      "step": 25210
    },
    {
      "entropy": 5.909185695648193,
      "epoch": 2.520517818763433,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043681839581251206,
      "loss": 5.5418,
      "mean_token_accuracy": 0.1945367068052292,
      "num_tokens": 49328410.0,
      "step": 25215
    },
    {
      "entropy": 6.008784532546997,
      "epoch": 2.521017643824661,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043679359661480703,
      "loss": 5.6338,
      "mean_token_accuracy": 0.17716891318559647,
      "num_tokens": 49337766.0,
      "step": 25220
    },
    {
      "entropy": 5.802869844436645,
      "epoch": 2.52151746888589,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043676879334641686,
      "loss": 5.3451,
      "mean_token_accuracy": 0.20307790637016296,
      "num_tokens": 49348512.0,
      "step": 25225
    },
    {
      "entropy": 5.927501964569092,
      "epoch": 2.5220172939471186,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004367439860079654,
      "loss": 5.5899,
      "mean_token_accuracy": 0.1830173760652542,
      "num_tokens": 49359149.0,
      "step": 25230
    },
    {
      "entropy": 5.9852362155914305,
      "epoch": 2.522517119008347,
      "grad_norm": 1.125,
      "learning_rate": 0.00043671917460007693,
      "loss": 5.5685,
      "mean_token_accuracy": 0.18459220975637436,
      "num_tokens": 49368083.0,
      "step": 25235
    },
    {
      "entropy": 5.972605466842651,
      "epoch": 2.5230169440695756,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043669435912337557,
      "loss": 5.5925,
      "mean_token_accuracy": 0.17848362624645234,
      "num_tokens": 49378205.0,
      "step": 25240
    },
    {
      "entropy": 5.94252462387085,
      "epoch": 2.5235167691308042,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004366695395784858,
      "loss": 5.6234,
      "mean_token_accuracy": 0.1821250542998314,
      "num_tokens": 49388405.0,
      "step": 25245
    },
    {
      "entropy": 5.95191798210144,
      "epoch": 2.524016594192033,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000436644715966032,
      "loss": 5.5026,
      "mean_token_accuracy": 0.19193872064352036,
      "num_tokens": 49397966.0,
      "step": 25250
    },
    {
      "entropy": 6.071300411224366,
      "epoch": 2.524516419253261,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004366198882866387,
      "loss": 5.6943,
      "mean_token_accuracy": 0.16983194798231124,
      "num_tokens": 49408542.0,
      "step": 25255
    },
    {
      "entropy": 5.945215272903442,
      "epoch": 2.52501624431449,
      "grad_norm": 1.125,
      "learning_rate": 0.00043659505654093065,
      "loss": 5.5787,
      "mean_token_accuracy": 0.1822728395462036,
      "num_tokens": 49417499.0,
      "step": 25260
    },
    {
      "entropy": 5.905768775939942,
      "epoch": 2.5255160693757186,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043657022072953253,
      "loss": 5.5986,
      "mean_token_accuracy": 0.18064597994089127,
      "num_tokens": 49426962.0,
      "step": 25265
    },
    {
      "entropy": 5.919689130783081,
      "epoch": 2.526015894436947,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004365453808530693,
      "loss": 5.5549,
      "mean_token_accuracy": 0.18363003432750702,
      "num_tokens": 49437188.0,
      "step": 25270
    },
    {
      "entropy": 5.989172840118409,
      "epoch": 2.5265157194981755,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043652053691216574,
      "loss": 5.5875,
      "mean_token_accuracy": 0.18223422169685363,
      "num_tokens": 49446939.0,
      "step": 25275
    },
    {
      "entropy": 5.967310905456543,
      "epoch": 2.5270155445594042,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004364956889074471,
      "loss": 5.5785,
      "mean_token_accuracy": 0.19183507114648818,
      "num_tokens": 49458160.0,
      "step": 25280
    },
    {
      "entropy": 5.90620002746582,
      "epoch": 2.527515369620633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004364708368395385,
      "loss": 5.6148,
      "mean_token_accuracy": 0.17624633610248566,
      "num_tokens": 49468284.0,
      "step": 25285
    },
    {
      "entropy": 6.043430233001709,
      "epoch": 2.528015194681861,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043644598070906516,
      "loss": 5.6135,
      "mean_token_accuracy": 0.18494329750537872,
      "num_tokens": 49479348.0,
      "step": 25290
    },
    {
      "entropy": 5.948676729202271,
      "epoch": 2.52851501974309,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004364211205166525,
      "loss": 5.4419,
      "mean_token_accuracy": 0.1906107857823372,
      "num_tokens": 49488802.0,
      "step": 25295
    },
    {
      "entropy": 5.94613561630249,
      "epoch": 2.5290148448043186,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043639625626292595,
      "loss": 5.5946,
      "mean_token_accuracy": 0.17913421541452407,
      "num_tokens": 49498682.0,
      "step": 25300
    },
    {
      "entropy": 5.956101322174073,
      "epoch": 2.529514669865547,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004363713879485112,
      "loss": 5.4872,
      "mean_token_accuracy": 0.1940541684627533,
      "num_tokens": 49507530.0,
      "step": 25305
    },
    {
      "entropy": 5.929623651504516,
      "epoch": 2.5300144949267755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004363465155740338,
      "loss": 5.4584,
      "mean_token_accuracy": 0.1856735423207283,
      "num_tokens": 49518435.0,
      "step": 25310
    },
    {
      "entropy": 5.919427394866943,
      "epoch": 2.5305143199880042,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004363216391401196,
      "loss": 5.5173,
      "mean_token_accuracy": 0.18665823191404343,
      "num_tokens": 49528251.0,
      "step": 25315
    },
    {
      "entropy": 5.910904502868652,
      "epoch": 2.531014145049233,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004362967586473945,
      "loss": 5.5196,
      "mean_token_accuracy": 0.1877917692065239,
      "num_tokens": 49536884.0,
      "step": 25320
    },
    {
      "entropy": 5.859858179092408,
      "epoch": 2.531513970110461,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004362718740964844,
      "loss": 5.4684,
      "mean_token_accuracy": 0.19075966328382493,
      "num_tokens": 49547665.0,
      "step": 25325
    },
    {
      "entropy": 5.971954441070556,
      "epoch": 2.53201379517169,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004362469854880155,
      "loss": 5.6623,
      "mean_token_accuracy": 0.17925975918769838,
      "num_tokens": 49557413.0,
      "step": 25330
    },
    {
      "entropy": 6.021203708648682,
      "epoch": 2.5325136202329186,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004362220928226139,
      "loss": 5.6311,
      "mean_token_accuracy": 0.17010490745306014,
      "num_tokens": 49566305.0,
      "step": 25335
    },
    {
      "entropy": 6.008874082565308,
      "epoch": 2.533013445294147,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043619719610090595,
      "loss": 5.5231,
      "mean_token_accuracy": 0.1809989646077156,
      "num_tokens": 49576020.0,
      "step": 25340
    },
    {
      "entropy": 5.9226280689239506,
      "epoch": 2.5335132703553755,
      "grad_norm": 1.0,
      "learning_rate": 0.000436172295323518,
      "loss": 5.5403,
      "mean_token_accuracy": 0.18133606016635895,
      "num_tokens": 49586564.0,
      "step": 25345
    },
    {
      "entropy": 5.94023175239563,
      "epoch": 2.5340130954166042,
      "grad_norm": 1.125,
      "learning_rate": 0.00043614739049107666,
      "loss": 5.6161,
      "mean_token_accuracy": 0.17701368629932404,
      "num_tokens": 49595874.0,
      "step": 25350
    },
    {
      "entropy": 5.9283336162567135,
      "epoch": 2.534512920477833,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043612248160420833,
      "loss": 5.5456,
      "mean_token_accuracy": 0.1833578735589981,
      "num_tokens": 49605975.0,
      "step": 25355
    },
    {
      "entropy": 5.91959056854248,
      "epoch": 2.535012745539061,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043609756866353985,
      "loss": 5.4614,
      "mean_token_accuracy": 0.19039904028177262,
      "num_tokens": 49615187.0,
      "step": 25360
    },
    {
      "entropy": 5.946052408218383,
      "epoch": 2.53551257060029,
      "grad_norm": 1.1875,
      "learning_rate": 0.000436072651669698,
      "loss": 5.594,
      "mean_token_accuracy": 0.17806714177131652,
      "num_tokens": 49624660.0,
      "step": 25365
    },
    {
      "entropy": 5.934181165695191,
      "epoch": 2.5360123956615186,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004360477306233097,
      "loss": 5.5024,
      "mean_token_accuracy": 0.18895899057388305,
      "num_tokens": 49635450.0,
      "step": 25370
    },
    {
      "entropy": 5.963568305969238,
      "epoch": 2.536512220722747,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043602280552500195,
      "loss": 5.6183,
      "mean_token_accuracy": 0.17574782520532609,
      "num_tokens": 49646466.0,
      "step": 25375
    },
    {
      "entropy": 6.016869115829468,
      "epoch": 2.5370120457839755,
      "grad_norm": 1.46875,
      "learning_rate": 0.00043599787637540186,
      "loss": 5.6457,
      "mean_token_accuracy": 0.1752337098121643,
      "num_tokens": 49655440.0,
      "step": 25380
    },
    {
      "entropy": 5.985523509979248,
      "epoch": 2.5375118708452042,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043597294317513654,
      "loss": 5.5097,
      "mean_token_accuracy": 0.1828925833106041,
      "num_tokens": 49664072.0,
      "step": 25385
    },
    {
      "entropy": 5.916598749160767,
      "epoch": 2.538011695906433,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043594800592483345,
      "loss": 5.5668,
      "mean_token_accuracy": 0.18082392066717148,
      "num_tokens": 49673439.0,
      "step": 25390
    },
    {
      "entropy": 5.9415051460266115,
      "epoch": 2.538511520967661,
      "grad_norm": 1.125,
      "learning_rate": 0.00043592306462511993,
      "loss": 5.593,
      "mean_token_accuracy": 0.18136870712041855,
      "num_tokens": 49682307.0,
      "step": 25395
    },
    {
      "entropy": 5.901618099212646,
      "epoch": 2.53901134602889,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043589811927662356,
      "loss": 5.3926,
      "mean_token_accuracy": 0.20943498313426973,
      "num_tokens": 49692146.0,
      "step": 25400
    },
    {
      "entropy": 5.984374189376831,
      "epoch": 2.5395111710901186,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043587316987997183,
      "loss": 5.5831,
      "mean_token_accuracy": 0.18177687674760817,
      "num_tokens": 49701465.0,
      "step": 25405
    },
    {
      "entropy": 5.903594207763672,
      "epoch": 2.540010996151347,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043584821643579265,
      "loss": 5.5758,
      "mean_token_accuracy": 0.18184472918510436,
      "num_tokens": 49711845.0,
      "step": 25410
    },
    {
      "entropy": 6.003462457656861,
      "epoch": 2.5405108212125755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043582325894471366,
      "loss": 5.5234,
      "mean_token_accuracy": 0.18077536374330522,
      "num_tokens": 49721078.0,
      "step": 25415
    },
    {
      "entropy": 5.941278886795044,
      "epoch": 2.5410106462738042,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004357982974073629,
      "loss": 5.5661,
      "mean_token_accuracy": 0.1855879917740822,
      "num_tokens": 49730987.0,
      "step": 25420
    },
    {
      "entropy": 5.895964765548706,
      "epoch": 2.541510471335033,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004357733318243683,
      "loss": 5.4085,
      "mean_token_accuracy": 0.19331088662147522,
      "num_tokens": 49741688.0,
      "step": 25425
    },
    {
      "entropy": 6.033232641220093,
      "epoch": 2.542010296396261,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043574836219635805,
      "loss": 5.6428,
      "mean_token_accuracy": 0.17969945818185806,
      "num_tokens": 49750835.0,
      "step": 25430
    },
    {
      "entropy": 5.948549461364746,
      "epoch": 2.54251012145749,
      "grad_norm": 1.125,
      "learning_rate": 0.00043572338852396035,
      "loss": 5.6013,
      "mean_token_accuracy": 0.17722235023975372,
      "num_tokens": 49760654.0,
      "step": 25435
    },
    {
      "entropy": 5.9771486759185795,
      "epoch": 2.5430099465187186,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043569841080780367,
      "loss": 5.5529,
      "mean_token_accuracy": 0.18712225258350373,
      "num_tokens": 49771367.0,
      "step": 25440
    },
    {
      "entropy": 5.947537803649903,
      "epoch": 2.543509771579947,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043567342904851625,
      "loss": 5.5477,
      "mean_token_accuracy": 0.18409523516893386,
      "num_tokens": 49781302.0,
      "step": 25445
    },
    {
      "entropy": 5.941720342636108,
      "epoch": 2.5440095966411755,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004356484432467267,
      "loss": 5.6355,
      "mean_token_accuracy": 0.18498838543891907,
      "num_tokens": 49790659.0,
      "step": 25450
    },
    {
      "entropy": 6.0111510276794435,
      "epoch": 2.5445094217024042,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043562345340306363,
      "loss": 5.633,
      "mean_token_accuracy": 0.18370601832866668,
      "num_tokens": 49800285.0,
      "step": 25455
    },
    {
      "entropy": 5.928597545623779,
      "epoch": 2.545009246763633,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004355984595181559,
      "loss": 5.5743,
      "mean_token_accuracy": 0.17514148056507112,
      "num_tokens": 49810810.0,
      "step": 25460
    },
    {
      "entropy": 5.950218915939331,
      "epoch": 2.545509071824861,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004355734615926321,
      "loss": 5.5793,
      "mean_token_accuracy": 0.17853649854660034,
      "num_tokens": 49820728.0,
      "step": 25465
    },
    {
      "entropy": 6.070777320861817,
      "epoch": 2.54600889688609,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004355484596271214,
      "loss": 5.6193,
      "mean_token_accuracy": 0.17320524752140046,
      "num_tokens": 49830477.0,
      "step": 25470
    },
    {
      "entropy": 5.993196439743042,
      "epoch": 2.5465087219473186,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004355234536222528,
      "loss": 5.6321,
      "mean_token_accuracy": 0.17637867331504822,
      "num_tokens": 49839154.0,
      "step": 25475
    },
    {
      "entropy": 5.9101409912109375,
      "epoch": 2.547008547008547,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004354984435786554,
      "loss": 5.5643,
      "mean_token_accuracy": 0.18982855677604676,
      "num_tokens": 49848894.0,
      "step": 25480
    },
    {
      "entropy": 5.96398138999939,
      "epoch": 2.5475083720697755,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004354734294969585,
      "loss": 5.5842,
      "mean_token_accuracy": 0.1824532836675644,
      "num_tokens": 49858730.0,
      "step": 25485
    },
    {
      "entropy": 5.968670749664307,
      "epoch": 2.5480081971310042,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043544841137779127,
      "loss": 5.613,
      "mean_token_accuracy": 0.17724207639694214,
      "num_tokens": 49868553.0,
      "step": 25490
    },
    {
      "entropy": 5.953317785263062,
      "epoch": 2.548508022192233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043542338922178343,
      "loss": 5.5518,
      "mean_token_accuracy": 0.19168639183044434,
      "num_tokens": 49878053.0,
      "step": 25495
    },
    {
      "entropy": 5.970700263977051,
      "epoch": 2.549007847253461,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043539836302956434,
      "loss": 5.5158,
      "mean_token_accuracy": 0.1874014690518379,
      "num_tokens": 49888693.0,
      "step": 25500
    },
    {
      "entropy": 6.0157355785369875,
      "epoch": 2.54950767231469,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004353733328017637,
      "loss": 5.5458,
      "mean_token_accuracy": 0.18369834572076799,
      "num_tokens": 49899094.0,
      "step": 25505
    },
    {
      "entropy": 6.034107446670532,
      "epoch": 2.5500074973759186,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004353482985390113,
      "loss": 5.6188,
      "mean_token_accuracy": 0.18363016098737717,
      "num_tokens": 49909341.0,
      "step": 25510
    },
    {
      "entropy": 5.897311115264893,
      "epoch": 2.550507322437147,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000435323260241937,
      "loss": 5.5515,
      "mean_token_accuracy": 0.18389289379119872,
      "num_tokens": 49919602.0,
      "step": 25515
    },
    {
      "entropy": 5.934476566314697,
      "epoch": 2.5510071474983755,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043529821791117064,
      "loss": 5.5608,
      "mean_token_accuracy": 0.1844294250011444,
      "num_tokens": 49928958.0,
      "step": 25520
    },
    {
      "entropy": 5.929257822036743,
      "epoch": 2.5515069725596042,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004352731715473425,
      "loss": 5.5427,
      "mean_token_accuracy": 0.1820814162492752,
      "num_tokens": 49938988.0,
      "step": 25525
    },
    {
      "entropy": 5.967476034164429,
      "epoch": 2.552006797620833,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043524812115108244,
      "loss": 5.6573,
      "mean_token_accuracy": 0.17672113478183746,
      "num_tokens": 49949399.0,
      "step": 25530
    },
    {
      "entropy": 5.974836015701294,
      "epoch": 2.552506622682061,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043522306672302103,
      "loss": 5.4778,
      "mean_token_accuracy": 0.18751568347215652,
      "num_tokens": 49958666.0,
      "step": 25535
    },
    {
      "entropy": 5.874354982376099,
      "epoch": 2.55300644774329,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004351980082637884,
      "loss": 5.5118,
      "mean_token_accuracy": 0.1857276126742363,
      "num_tokens": 49967739.0,
      "step": 25540
    },
    {
      "entropy": 5.925473403930664,
      "epoch": 2.5535062728045186,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043517294577401504,
      "loss": 5.5885,
      "mean_token_accuracy": 0.180202454328537,
      "num_tokens": 49976980.0,
      "step": 25545
    },
    {
      "entropy": 5.971933126449585,
      "epoch": 2.554006097865747,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043514787925433176,
      "loss": 5.5895,
      "mean_token_accuracy": 0.18293219357728957,
      "num_tokens": 49986963.0,
      "step": 25550
    },
    {
      "entropy": 5.9880352973937985,
      "epoch": 2.5545059229269755,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004351228087053689,
      "loss": 5.6483,
      "mean_token_accuracy": 0.16966913342475892,
      "num_tokens": 49996009.0,
      "step": 25555
    },
    {
      "entropy": 5.975197172164917,
      "epoch": 2.5550057479882042,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043509773412775746,
      "loss": 5.5748,
      "mean_token_accuracy": 0.18090624064207078,
      "num_tokens": 50006346.0,
      "step": 25560
    },
    {
      "entropy": 5.923209857940674,
      "epoch": 2.555505573049433,
      "grad_norm": 1.0,
      "learning_rate": 0.0004350726555221282,
      "loss": 5.5576,
      "mean_token_accuracy": 0.190318700671196,
      "num_tokens": 50016512.0,
      "step": 25565
    },
    {
      "entropy": 5.908349132537841,
      "epoch": 2.556005398110661,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004350475728891121,
      "loss": 5.5473,
      "mean_token_accuracy": 0.183578285574913,
      "num_tokens": 50027012.0,
      "step": 25570
    },
    {
      "entropy": 5.821095895767212,
      "epoch": 2.55650522317189,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004350224862293403,
      "loss": 5.3058,
      "mean_token_accuracy": 0.2079145133495331,
      "num_tokens": 50036869.0,
      "step": 25575
    },
    {
      "entropy": 5.9462409019470215,
      "epoch": 2.557005048233118,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043499739554344393,
      "loss": 5.5136,
      "mean_token_accuracy": 0.1909269168972969,
      "num_tokens": 50046186.0,
      "step": 25580
    },
    {
      "entropy": 5.901857614517212,
      "epoch": 2.557504873294347,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043497230083205425,
      "loss": 5.527,
      "mean_token_accuracy": 0.19368426352739335,
      "num_tokens": 50056915.0,
      "step": 25585
    },
    {
      "entropy": 5.8258013248443605,
      "epoch": 2.5580046983555755,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004349472020958026,
      "loss": 5.5119,
      "mean_token_accuracy": 0.1860867142677307,
      "num_tokens": 50066143.0,
      "step": 25590
    },
    {
      "entropy": 5.9448141098022464,
      "epoch": 2.5585045234168042,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043492209933532054,
      "loss": 5.5539,
      "mean_token_accuracy": 0.17750124335289003,
      "num_tokens": 50077031.0,
      "step": 25595
    },
    {
      "entropy": 6.045050382614136,
      "epoch": 2.559004348478033,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043489699255123954,
      "loss": 5.6886,
      "mean_token_accuracy": 0.16588341742753981,
      "num_tokens": 50087545.0,
      "step": 25600
    },
    {
      "entropy": 5.955065441131592,
      "epoch": 2.559504173539261,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004348718817441914,
      "loss": 5.5202,
      "mean_token_accuracy": 0.18882030993700027,
      "num_tokens": 50096427.0,
      "step": 25605
    },
    {
      "entropy": 5.924861669540405,
      "epoch": 2.56000399860049,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004348467669148078,
      "loss": 5.5168,
      "mean_token_accuracy": 0.18574264496564866,
      "num_tokens": 50105382.0,
      "step": 25610
    },
    {
      "entropy": 6.02698540687561,
      "epoch": 2.560503823661718,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043482164806372075,
      "loss": 5.6687,
      "mean_token_accuracy": 0.1709901824593544,
      "num_tokens": 50116348.0,
      "step": 25615
    },
    {
      "entropy": 5.987402486801147,
      "epoch": 2.561003648722947,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004347965251915622,
      "loss": 5.5484,
      "mean_token_accuracy": 0.1856226623058319,
      "num_tokens": 50126081.0,
      "step": 25620
    },
    {
      "entropy": 5.925216436386108,
      "epoch": 2.5615034737841755,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043477139829896415,
      "loss": 5.5205,
      "mean_token_accuracy": 0.17986690402030944,
      "num_tokens": 50135719.0,
      "step": 25625
    },
    {
      "entropy": 5.92338342666626,
      "epoch": 2.5620032988454042,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043474626738655876,
      "loss": 5.5354,
      "mean_token_accuracy": 0.18524643182754516,
      "num_tokens": 50146201.0,
      "step": 25630
    },
    {
      "entropy": 5.8906481742858885,
      "epoch": 2.562503123906633,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004347211324549785,
      "loss": 5.4177,
      "mean_token_accuracy": 0.1948825791478157,
      "num_tokens": 50156352.0,
      "step": 25635
    },
    {
      "entropy": 5.98774471282959,
      "epoch": 2.563002948967861,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043469599350485556,
      "loss": 5.5515,
      "mean_token_accuracy": 0.17818311750888824,
      "num_tokens": 50165518.0,
      "step": 25640
    },
    {
      "entropy": 5.9211897373199465,
      "epoch": 2.56350277402909,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004346708505368225,
      "loss": 5.4849,
      "mean_token_accuracy": 0.19920744448900224,
      "num_tokens": 50173656.0,
      "step": 25645
    },
    {
      "entropy": 5.951807832717895,
      "epoch": 2.564002599090318,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004346457035515121,
      "loss": 5.5514,
      "mean_token_accuracy": 0.18827321380376816,
      "num_tokens": 50183024.0,
      "step": 25650
    },
    {
      "entropy": 5.987016344070435,
      "epoch": 2.564502424151547,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043462055254955666,
      "loss": 5.6524,
      "mean_token_accuracy": 0.180021071434021,
      "num_tokens": 50194192.0,
      "step": 25655
    },
    {
      "entropy": 5.9796465873718265,
      "epoch": 2.5650022492127755,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004345953975315893,
      "loss": 5.631,
      "mean_token_accuracy": 0.18385594338178635,
      "num_tokens": 50203655.0,
      "step": 25660
    },
    {
      "entropy": 6.052949333190918,
      "epoch": 2.5655020742740042,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004345702384982429,
      "loss": 5.6498,
      "mean_token_accuracy": 0.17724403589963914,
      "num_tokens": 50213400.0,
      "step": 25665
    },
    {
      "entropy": 5.991884708404541,
      "epoch": 2.566001899335233,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004345450754501502,
      "loss": 5.5994,
      "mean_token_accuracy": 0.17804084867238998,
      "num_tokens": 50224113.0,
      "step": 25670
    },
    {
      "entropy": 5.960135078430175,
      "epoch": 2.566501724396461,
      "grad_norm": 1.125,
      "learning_rate": 0.00043451990838794453,
      "loss": 5.54,
      "mean_token_accuracy": 0.18424474596977233,
      "num_tokens": 50232990.0,
      "step": 25675
    },
    {
      "entropy": 5.973487138748169,
      "epoch": 2.56700154945769,
      "grad_norm": 1.109375,
      "learning_rate": 0.000434494737312259,
      "loss": 5.549,
      "mean_token_accuracy": 0.1767607480287552,
      "num_tokens": 50242573.0,
      "step": 25680
    },
    {
      "entropy": 5.995744705200195,
      "epoch": 2.567501374518918,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043446956222372693,
      "loss": 5.6098,
      "mean_token_accuracy": 0.17712354958057402,
      "num_tokens": 50252654.0,
      "step": 25685
    },
    {
      "entropy": 5.972771787643433,
      "epoch": 2.568001199580147,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004344443831229816,
      "loss": 5.5159,
      "mean_token_accuracy": 0.18420054167509078,
      "num_tokens": 50262654.0,
      "step": 25690
    },
    {
      "entropy": 5.995862579345703,
      "epoch": 2.5685010246413755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043441920001065676,
      "loss": 5.6207,
      "mean_token_accuracy": 0.17931232526898383,
      "num_tokens": 50272653.0,
      "step": 25695
    },
    {
      "entropy": 6.020559406280517,
      "epoch": 2.5690008497026042,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004343940128873858,
      "loss": 5.7019,
      "mean_token_accuracy": 0.17556266486644745,
      "num_tokens": 50283157.0,
      "step": 25700
    },
    {
      "entropy": 5.976172161102295,
      "epoch": 2.569500674763833,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043436882175380256,
      "loss": 5.5683,
      "mean_token_accuracy": 0.1863444209098816,
      "num_tokens": 50292224.0,
      "step": 25705
    },
    {
      "entropy": 6.032931566238403,
      "epoch": 2.570000499825061,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004343436266105407,
      "loss": 5.6203,
      "mean_token_accuracy": 0.17747710645198822,
      "num_tokens": 50302083.0,
      "step": 25710
    },
    {
      "entropy": 5.872206449508667,
      "epoch": 2.57050032488629,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004343184274582343,
      "loss": 5.5299,
      "mean_token_accuracy": 0.18860879689455032,
      "num_tokens": 50312001.0,
      "step": 25715
    },
    {
      "entropy": 5.907359457015991,
      "epoch": 2.571000149947518,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004342932242975172,
      "loss": 5.4749,
      "mean_token_accuracy": 0.19071522057056428,
      "num_tokens": 50322652.0,
      "step": 25720
    },
    {
      "entropy": 6.008044099807739,
      "epoch": 2.571499975008747,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004342680171290235,
      "loss": 5.5274,
      "mean_token_accuracy": 0.1880177840590477,
      "num_tokens": 50332200.0,
      "step": 25725
    },
    {
      "entropy": 5.902482175827027,
      "epoch": 2.5719998000699755,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004342428059533875,
      "loss": 5.5125,
      "mean_token_accuracy": 0.1908721536397934,
      "num_tokens": 50342567.0,
      "step": 25730
    },
    {
      "entropy": 5.922927570343018,
      "epoch": 2.5724996251312042,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004342175907712435,
      "loss": 5.5373,
      "mean_token_accuracy": 0.19476069062948226,
      "num_tokens": 50352009.0,
      "step": 25735
    },
    {
      "entropy": 5.886311340332031,
      "epoch": 2.572999450192433,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043419237158322596,
      "loss": 5.536,
      "mean_token_accuracy": 0.1879701018333435,
      "num_tokens": 50362412.0,
      "step": 25740
    },
    {
      "entropy": 5.997099924087524,
      "epoch": 2.573499275253661,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043416714838996915,
      "loss": 5.599,
      "mean_token_accuracy": 0.1864205002784729,
      "num_tokens": 50372336.0,
      "step": 25745
    },
    {
      "entropy": 5.89397611618042,
      "epoch": 2.57399910031489,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000434141921192108,
      "loss": 5.446,
      "mean_token_accuracy": 0.1960783377289772,
      "num_tokens": 50381612.0,
      "step": 25750
    },
    {
      "entropy": 5.89833288192749,
      "epoch": 2.574498925376118,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000434116689990277,
      "loss": 5.5258,
      "mean_token_accuracy": 0.1905748054385185,
      "num_tokens": 50392073.0,
      "step": 25755
    },
    {
      "entropy": 5.930168771743775,
      "epoch": 2.574998750437347,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043409145478511105,
      "loss": 5.4774,
      "mean_token_accuracy": 0.1890529900789261,
      "num_tokens": 50401306.0,
      "step": 25760
    },
    {
      "entropy": 5.921191215515137,
      "epoch": 2.5754985754985755,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043406621557724495,
      "loss": 5.5534,
      "mean_token_accuracy": 0.18058284372091293,
      "num_tokens": 50410592.0,
      "step": 25765
    },
    {
      "entropy": 5.934085893630981,
      "epoch": 2.5759984005598042,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043404097236731396,
      "loss": 5.5517,
      "mean_token_accuracy": 0.18572560101747512,
      "num_tokens": 50420863.0,
      "step": 25770
    },
    {
      "entropy": 5.960568141937256,
      "epoch": 2.5764982256210325,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004340157251559529,
      "loss": 5.5981,
      "mean_token_accuracy": 0.18491554707288743,
      "num_tokens": 50431321.0,
      "step": 25775
    },
    {
      "entropy": 5.948149013519287,
      "epoch": 2.576998050682261,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043399047394379717,
      "loss": 5.5347,
      "mean_token_accuracy": 0.18420010209083557,
      "num_tokens": 50442223.0,
      "step": 25780
    },
    {
      "entropy": 5.884848880767822,
      "epoch": 2.57749787574349,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043396521873148195,
      "loss": 5.4926,
      "mean_token_accuracy": 0.18830681294202806,
      "num_tokens": 50451082.0,
      "step": 25785
    },
    {
      "entropy": 5.9503655433654785,
      "epoch": 2.577997700804718,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043393995951964277,
      "loss": 5.5471,
      "mean_token_accuracy": 0.1906142443418503,
      "num_tokens": 50461061.0,
      "step": 25790
    },
    {
      "entropy": 5.924425363540649,
      "epoch": 2.578497525865947,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043391469630891513,
      "loss": 5.4775,
      "mean_token_accuracy": 0.19381013661623,
      "num_tokens": 50470098.0,
      "step": 25795
    },
    {
      "entropy": 5.962984466552735,
      "epoch": 2.5789973509271755,
      "grad_norm": 1.125,
      "learning_rate": 0.0004338894290999346,
      "loss": 5.5227,
      "mean_token_accuracy": 0.182538765668869,
      "num_tokens": 50479669.0,
      "step": 25800
    },
    {
      "entropy": 5.9027153015136715,
      "epoch": 2.5794971759884042,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043386415789333687,
      "loss": 5.4938,
      "mean_token_accuracy": 0.18995393216609954,
      "num_tokens": 50488608.0,
      "step": 25805
    },
    {
      "entropy": 5.92372465133667,
      "epoch": 2.5799970010496325,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004338388826897579,
      "loss": 5.6064,
      "mean_token_accuracy": 0.17950460761785508,
      "num_tokens": 50497486.0,
      "step": 25810
    },
    {
      "entropy": 5.976496791839599,
      "epoch": 2.580496826110861,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004338136034898335,
      "loss": 5.5752,
      "mean_token_accuracy": 0.1848434552550316,
      "num_tokens": 50506034.0,
      "step": 25815
    },
    {
      "entropy": 5.985818719863891,
      "epoch": 2.58099665117209,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004337883202941997,
      "loss": 5.6182,
      "mean_token_accuracy": 0.17681290358304977,
      "num_tokens": 50516617.0,
      "step": 25820
    },
    {
      "entropy": 6.019786596298218,
      "epoch": 2.581496476233318,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004337630331034926,
      "loss": 5.5699,
      "mean_token_accuracy": 0.17912903279066086,
      "num_tokens": 50526287.0,
      "step": 25825
    },
    {
      "entropy": 5.959505271911621,
      "epoch": 2.581996301294547,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043373774191834844,
      "loss": 5.5745,
      "mean_token_accuracy": 0.17578665912151337,
      "num_tokens": 50535803.0,
      "step": 25830
    },
    {
      "entropy": 6.038329982757569,
      "epoch": 2.5824961263557755,
      "grad_norm": 1.421875,
      "learning_rate": 0.00043371244673940355,
      "loss": 5.7439,
      "mean_token_accuracy": 0.16636991500854492,
      "num_tokens": 50545386.0,
      "step": 25835
    },
    {
      "entropy": 6.007061910629273,
      "epoch": 2.582995951417004,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004336871475672943,
      "loss": 5.582,
      "mean_token_accuracy": 0.18218912035226822,
      "num_tokens": 50555299.0,
      "step": 25840
    },
    {
      "entropy": 5.891808891296387,
      "epoch": 2.5834957764782325,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004336618444026573,
      "loss": 5.4735,
      "mean_token_accuracy": 0.18772495836019515,
      "num_tokens": 50564530.0,
      "step": 25845
    },
    {
      "entropy": 5.968916940689087,
      "epoch": 2.583995601539461,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043363653724612903,
      "loss": 5.5882,
      "mean_token_accuracy": 0.18055772334337233,
      "num_tokens": 50574783.0,
      "step": 25850
    },
    {
      "entropy": 5.861871767044067,
      "epoch": 2.58449542660069,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043361122609834646,
      "loss": 5.42,
      "mean_token_accuracy": 0.19435933828353882,
      "num_tokens": 50584764.0,
      "step": 25855
    },
    {
      "entropy": 5.887496042251587,
      "epoch": 2.584995251661918,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004335859109599462,
      "loss": 5.4702,
      "mean_token_accuracy": 0.19161894619464875,
      "num_tokens": 50593547.0,
      "step": 25860
    },
    {
      "entropy": 5.899431276321411,
      "epoch": 2.585495076723147,
      "grad_norm": 1.40625,
      "learning_rate": 0.00043356059183156514,
      "loss": 5.4849,
      "mean_token_accuracy": 0.19351288080215454,
      "num_tokens": 50602749.0,
      "step": 25865
    },
    {
      "entropy": 5.989253234863281,
      "epoch": 2.5859949017843755,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043353526871384054,
      "loss": 5.523,
      "mean_token_accuracy": 0.19035777002573012,
      "num_tokens": 50612900.0,
      "step": 25870
    },
    {
      "entropy": 6.021588563919067,
      "epoch": 2.586494726845604,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004335099416074092,
      "loss": 5.6485,
      "mean_token_accuracy": 0.17162780463695526,
      "num_tokens": 50623131.0,
      "step": 25875
    },
    {
      "entropy": 5.970678997039795,
      "epoch": 2.5869945519068325,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004334846105129086,
      "loss": 5.4906,
      "mean_token_accuracy": 0.18646990209817887,
      "num_tokens": 50632312.0,
      "step": 25880
    },
    {
      "entropy": 6.019754886627197,
      "epoch": 2.587494376968061,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000433459275430976,
      "loss": 5.71,
      "mean_token_accuracy": 0.168869049847126,
      "num_tokens": 50643264.0,
      "step": 25885
    },
    {
      "entropy": 5.965439414978027,
      "epoch": 2.58799420202929,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004334339363622488,
      "loss": 5.5849,
      "mean_token_accuracy": 0.18430329561233522,
      "num_tokens": 50652525.0,
      "step": 25890
    },
    {
      "entropy": 5.994283962249756,
      "epoch": 2.588494027090518,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043340859330736454,
      "loss": 5.5824,
      "mean_token_accuracy": 0.18330912441015243,
      "num_tokens": 50661509.0,
      "step": 25895
    },
    {
      "entropy": 5.971550703048706,
      "epoch": 2.588993852151747,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00043338324626696084,
      "loss": 5.5529,
      "mean_token_accuracy": 0.178478966653347,
      "num_tokens": 50672646.0,
      "step": 25900
    },
    {
      "entropy": 5.963054513931274,
      "epoch": 2.5894936772129755,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004333578952416753,
      "loss": 5.4463,
      "mean_token_accuracy": 0.20142245590686797,
      "num_tokens": 50681967.0,
      "step": 25905
    },
    {
      "entropy": 5.990488433837891,
      "epoch": 2.589993502274204,
      "grad_norm": 1.265625,
      "learning_rate": 0.000433332540232146,
      "loss": 5.5481,
      "mean_token_accuracy": 0.18831525295972823,
      "num_tokens": 50690630.0,
      "step": 25910
    },
    {
      "entropy": 5.946015071868897,
      "epoch": 2.5904933273354325,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00043330718123901074,
      "loss": 5.5649,
      "mean_token_accuracy": 0.18741095215082168,
      "num_tokens": 50700618.0,
      "step": 25915
    },
    {
      "entropy": 5.939438104629517,
      "epoch": 2.590993152396661,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043328181826290745,
      "loss": 5.5044,
      "mean_token_accuracy": 0.18786118924617767,
      "num_tokens": 50710825.0,
      "step": 25920
    },
    {
      "entropy": 5.944729328155518,
      "epoch": 2.59149297745789,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004332564513044744,
      "loss": 5.5207,
      "mean_token_accuracy": 0.19021453559398652,
      "num_tokens": 50721869.0,
      "step": 25925
    },
    {
      "entropy": 5.9503480911254885,
      "epoch": 2.591992802519118,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043323108036434974,
      "loss": 5.4701,
      "mean_token_accuracy": 0.19139671623706817,
      "num_tokens": 50732088.0,
      "step": 25930
    },
    {
      "entropy": 5.911106157302856,
      "epoch": 2.592492627580347,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004332057054431718,
      "loss": 5.5707,
      "mean_token_accuracy": 0.18361196517944336,
      "num_tokens": 50743419.0,
      "step": 25935
    },
    {
      "entropy": 5.956199407577515,
      "epoch": 2.5929924526415755,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000433180326541579,
      "loss": 5.5318,
      "mean_token_accuracy": 0.1851876348257065,
      "num_tokens": 50754371.0,
      "step": 25940
    },
    {
      "entropy": 5.979596757888794,
      "epoch": 2.593492277702804,
      "grad_norm": 1.125,
      "learning_rate": 0.0004331549436602099,
      "loss": 5.5258,
      "mean_token_accuracy": 0.18789494633674622,
      "num_tokens": 50763737.0,
      "step": 25945
    },
    {
      "entropy": 5.978868675231934,
      "epoch": 2.5939921027640325,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043312955679970317,
      "loss": 5.6313,
      "mean_token_accuracy": 0.1808006390929222,
      "num_tokens": 50774571.0,
      "step": 25950
    },
    {
      "entropy": 6.03427791595459,
      "epoch": 2.594491927825261,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043310416596069746,
      "loss": 5.5789,
      "mean_token_accuracy": 0.18235338777303695,
      "num_tokens": 50783232.0,
      "step": 25955
    },
    {
      "entropy": 5.9636483669281,
      "epoch": 2.59499175288649,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043307877114383167,
      "loss": 5.5902,
      "mean_token_accuracy": 0.1908901110291481,
      "num_tokens": 50792238.0,
      "step": 25960
    },
    {
      "entropy": 5.880419540405273,
      "epoch": 2.595491577947718,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004330533723497445,
      "loss": 5.432,
      "mean_token_accuracy": 0.19653174430131912,
      "num_tokens": 50802012.0,
      "step": 25965
    },
    {
      "entropy": 5.953871345520019,
      "epoch": 2.595991403008947,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043302796957907523,
      "loss": 5.5452,
      "mean_token_accuracy": 0.19031720161437987,
      "num_tokens": 50811938.0,
      "step": 25970
    },
    {
      "entropy": 5.948025846481324,
      "epoch": 2.5964912280701755,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043300256283246297,
      "loss": 5.5049,
      "mean_token_accuracy": 0.18816644251346587,
      "num_tokens": 50820702.0,
      "step": 25975
    },
    {
      "entropy": 6.013067102432251,
      "epoch": 2.596991053131404,
      "grad_norm": 1.125,
      "learning_rate": 0.00043297715211054684,
      "loss": 5.6568,
      "mean_token_accuracy": 0.17539109140634537,
      "num_tokens": 50830291.0,
      "step": 25980
    },
    {
      "entropy": 5.890847396850586,
      "epoch": 2.5974908781926325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043295173741396626,
      "loss": 5.5348,
      "mean_token_accuracy": 0.18127792924642563,
      "num_tokens": 50840428.0,
      "step": 25985
    },
    {
      "entropy": 5.990366268157959,
      "epoch": 2.597990703253861,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004329263187433605,
      "loss": 5.6172,
      "mean_token_accuracy": 0.1825279861688614,
      "num_tokens": 50849596.0,
      "step": 25990
    },
    {
      "entropy": 5.94045672416687,
      "epoch": 2.59849052831509,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004329008960993694,
      "loss": 5.513,
      "mean_token_accuracy": 0.18331074863672256,
      "num_tokens": 50858347.0,
      "step": 25995
    },
    {
      "entropy": 5.975393581390381,
      "epoch": 2.598990353376318,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004328754694826322,
      "loss": 5.5324,
      "mean_token_accuracy": 0.1859849810600281,
      "num_tokens": 50868979.0,
      "step": 26000
    },
    {
      "entropy": 5.965642261505127,
      "epoch": 2.599490178437547,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004328500388937889,
      "loss": 5.5516,
      "mean_token_accuracy": 0.18894191682338715,
      "num_tokens": 50878789.0,
      "step": 26005
    },
    {
      "entropy": 5.942809295654297,
      "epoch": 2.5999900034987755,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043282460433347926,
      "loss": 5.599,
      "mean_token_accuracy": 0.17941048890352249,
      "num_tokens": 50889381.0,
      "step": 26010
    },
    {
      "entropy": 6.069531393051148,
      "epoch": 2.600489828560004,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004327991658023431,
      "loss": 5.5747,
      "mean_token_accuracy": 0.1757403150200844,
      "num_tokens": 50898201.0,
      "step": 26015
    },
    {
      "entropy": 6.0016662120819095,
      "epoch": 2.6009896536212325,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004327737233010205,
      "loss": 5.5985,
      "mean_token_accuracy": 0.19037162959575654,
      "num_tokens": 50908185.0,
      "step": 26020
    },
    {
      "entropy": 5.9272197723388675,
      "epoch": 2.601489478682461,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043274827683015174,
      "loss": 5.473,
      "mean_token_accuracy": 0.19481582194566727,
      "num_tokens": 50918766.0,
      "step": 26025
    },
    {
      "entropy": 5.909774732589722,
      "epoch": 2.60198930374369,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043272282639037683,
      "loss": 5.5404,
      "mean_token_accuracy": 0.18918127417564393,
      "num_tokens": 50928588.0,
      "step": 26030
    },
    {
      "entropy": 5.951172351837158,
      "epoch": 2.602489128804918,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043269737198233627,
      "loss": 5.5463,
      "mean_token_accuracy": 0.1814563289284706,
      "num_tokens": 50938106.0,
      "step": 26035
    },
    {
      "entropy": 6.055937957763672,
      "epoch": 2.602988953866147,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004326719136066704,
      "loss": 5.6902,
      "mean_token_accuracy": 0.17746190577745438,
      "num_tokens": 50947723.0,
      "step": 26040
    },
    {
      "entropy": 5.971568489074707,
      "epoch": 2.6034887789273755,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043264645126401977,
      "loss": 5.6335,
      "mean_token_accuracy": 0.17532318383455275,
      "num_tokens": 50956744.0,
      "step": 26045
    },
    {
      "entropy": 6.03110671043396,
      "epoch": 2.603988603988604,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043262098495502497,
      "loss": 5.7275,
      "mean_token_accuracy": 0.17725395113229753,
      "num_tokens": 50967778.0,
      "step": 26050
    },
    {
      "entropy": 5.988253498077393,
      "epoch": 2.6044884290498325,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004325955146803267,
      "loss": 5.483,
      "mean_token_accuracy": 0.19238365143537522,
      "num_tokens": 50978635.0,
      "step": 26055
    },
    {
      "entropy": 6.05434021949768,
      "epoch": 2.604988254111061,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004325700404405659,
      "loss": 5.7033,
      "mean_token_accuracy": 0.1701671600341797,
      "num_tokens": 50988823.0,
      "step": 26060
    },
    {
      "entropy": 6.012236499786377,
      "epoch": 2.60548807917229,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004325445622363834,
      "loss": 5.6225,
      "mean_token_accuracy": 0.1775982767343521,
      "num_tokens": 50998576.0,
      "step": 26065
    },
    {
      "entropy": 6.008157730102539,
      "epoch": 2.605987904233518,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004325190800684203,
      "loss": 5.5523,
      "mean_token_accuracy": 0.19217870533466339,
      "num_tokens": 51008142.0,
      "step": 26070
    },
    {
      "entropy": 5.844831800460815,
      "epoch": 2.606487729294747,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004324935939373176,
      "loss": 5.3931,
      "mean_token_accuracy": 0.1964648559689522,
      "num_tokens": 51018892.0,
      "step": 26075
    },
    {
      "entropy": 5.996842908859253,
      "epoch": 2.6069875543559755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004324681038437166,
      "loss": 5.6454,
      "mean_token_accuracy": 0.17830507308244706,
      "num_tokens": 51029116.0,
      "step": 26080
    },
    {
      "entropy": 5.86073317527771,
      "epoch": 2.607487379417204,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004324426097882587,
      "loss": 5.4384,
      "mean_token_accuracy": 0.19776744693517684,
      "num_tokens": 51038392.0,
      "step": 26085
    },
    {
      "entropy": 5.914388799667359,
      "epoch": 2.6079872044784325,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004324171117715852,
      "loss": 5.559,
      "mean_token_accuracy": 0.1820254221558571,
      "num_tokens": 51048344.0,
      "step": 26090
    },
    {
      "entropy": 5.956924867630005,
      "epoch": 2.608487029539661,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043239160979433765,
      "loss": 5.6343,
      "mean_token_accuracy": 0.1793489634990692,
      "num_tokens": 51057879.0,
      "step": 26095
    },
    {
      "entropy": 5.968038082122803,
      "epoch": 2.6089868546008894,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004323661038571578,
      "loss": 5.5163,
      "mean_token_accuracy": 0.18804190307855606,
      "num_tokens": 51067365.0,
      "step": 26100
    },
    {
      "entropy": 5.913079929351807,
      "epoch": 2.609486679662118,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043234059396068725,
      "loss": 5.4338,
      "mean_token_accuracy": 0.1890369638800621,
      "num_tokens": 51076168.0,
      "step": 26105
    },
    {
      "entropy": 5.961193990707398,
      "epoch": 2.609986504723347,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004323150801055677,
      "loss": 5.6016,
      "mean_token_accuracy": 0.17898748070001602,
      "num_tokens": 51086438.0,
      "step": 26110
    },
    {
      "entropy": 5.980804824829102,
      "epoch": 2.6104863297845755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004322895622924414,
      "loss": 5.5821,
      "mean_token_accuracy": 0.18505082726478578,
      "num_tokens": 51095386.0,
      "step": 26115
    },
    {
      "entropy": 5.909484243392944,
      "epoch": 2.610986154845804,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004322640405219501,
      "loss": 5.4815,
      "mean_token_accuracy": 0.1907034695148468,
      "num_tokens": 51105224.0,
      "step": 26120
    },
    {
      "entropy": 5.9011822700500485,
      "epoch": 2.6114859799070325,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043223851479473596,
      "loss": 5.6102,
      "mean_token_accuracy": 0.18147823065519333,
      "num_tokens": 51115011.0,
      "step": 26125
    },
    {
      "entropy": 6.013249111175537,
      "epoch": 2.611985804968261,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043221298511144133,
      "loss": 5.6087,
      "mean_token_accuracy": 0.18081451654434205,
      "num_tokens": 51124173.0,
      "step": 26130
    },
    {
      "entropy": 5.982674312591553,
      "epoch": 2.6124856300294894,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043218745147270845,
      "loss": 5.453,
      "mean_token_accuracy": 0.19349285066127778,
      "num_tokens": 51133493.0,
      "step": 26135
    },
    {
      "entropy": 5.995878601074219,
      "epoch": 2.612985455090718,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043216191387917966,
      "loss": 5.6487,
      "mean_token_accuracy": 0.1784988597035408,
      "num_tokens": 51142748.0,
      "step": 26140
    },
    {
      "entropy": 5.919480323791504,
      "epoch": 2.613485280151947,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004321363723314976,
      "loss": 5.5208,
      "mean_token_accuracy": 0.1926370605826378,
      "num_tokens": 51152681.0,
      "step": 26145
    },
    {
      "entropy": 5.902671861648559,
      "epoch": 2.6139851052131755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043211082683030483,
      "loss": 5.474,
      "mean_token_accuracy": 0.18395867198705673,
      "num_tokens": 51161134.0,
      "step": 26150
    },
    {
      "entropy": 5.931668424606324,
      "epoch": 2.614484930274404,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043208527737624406,
      "loss": 5.475,
      "mean_token_accuracy": 0.19371799975633622,
      "num_tokens": 51170623.0,
      "step": 26155
    },
    {
      "entropy": 6.002186441421509,
      "epoch": 2.6149847553356325,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043205972396995825,
      "loss": 5.5997,
      "mean_token_accuracy": 0.17735427916049956,
      "num_tokens": 51180290.0,
      "step": 26160
    },
    {
      "entropy": 5.955378675460816,
      "epoch": 2.615484580396861,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004320341666120901,
      "loss": 5.514,
      "mean_token_accuracy": 0.1890467569231987,
      "num_tokens": 51188733.0,
      "step": 26165
    },
    {
      "entropy": 5.840445613861084,
      "epoch": 2.6159844054580894,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043200860530328287,
      "loss": 5.4785,
      "mean_token_accuracy": 0.19647748321294783,
      "num_tokens": 51198403.0,
      "step": 26170
    },
    {
      "entropy": 5.8718963146209715,
      "epoch": 2.616484230519318,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043198304004417933,
      "loss": 5.4998,
      "mean_token_accuracy": 0.18831671625375748,
      "num_tokens": 51207654.0,
      "step": 26175
    },
    {
      "entropy": 5.95097165107727,
      "epoch": 2.616984055580547,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043195747083542296,
      "loss": 5.54,
      "mean_token_accuracy": 0.18232659995555878,
      "num_tokens": 51216285.0,
      "step": 26180
    },
    {
      "entropy": 5.80172643661499,
      "epoch": 2.6174838806417755,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000431931897677657,
      "loss": 5.3171,
      "mean_token_accuracy": 0.2069930076599121,
      "num_tokens": 51225182.0,
      "step": 26185
    },
    {
      "entropy": 5.920802879333496,
      "epoch": 2.617983705703004,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043190632057152497,
      "loss": 5.6137,
      "mean_token_accuracy": 0.18548991829156875,
      "num_tokens": 51235547.0,
      "step": 26190
    },
    {
      "entropy": 6.00066990852356,
      "epoch": 2.6184835307642325,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004318807395176702,
      "loss": 5.5822,
      "mean_token_accuracy": 0.18196050971746444,
      "num_tokens": 51244897.0,
      "step": 26195
    },
    {
      "entropy": 5.902744007110596,
      "epoch": 2.618983355825461,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004318551545167364,
      "loss": 5.4906,
      "mean_token_accuracy": 0.190519018471241,
      "num_tokens": 51255343.0,
      "step": 26200
    },
    {
      "entropy": 5.907501363754273,
      "epoch": 2.6194831808866894,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004318295655693674,
      "loss": 5.5357,
      "mean_token_accuracy": 0.1853316992521286,
      "num_tokens": 51265448.0,
      "step": 26205
    },
    {
      "entropy": 5.880193042755127,
      "epoch": 2.619983005947918,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043180397267620675,
      "loss": 5.4244,
      "mean_token_accuracy": 0.1885877877473831,
      "num_tokens": 51273330.0,
      "step": 26210
    },
    {
      "entropy": 5.895281410217285,
      "epoch": 2.620482831009147,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004317783758378985,
      "loss": 5.5669,
      "mean_token_accuracy": 0.18482257276773453,
      "num_tokens": 51283564.0,
      "step": 26215
    },
    {
      "entropy": 5.966788911819458,
      "epoch": 2.6209826560703755,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043175277505508677,
      "loss": 5.5766,
      "mean_token_accuracy": 0.18023480474948883,
      "num_tokens": 51293386.0,
      "step": 26220
    },
    {
      "entropy": 5.961998987197876,
      "epoch": 2.621482481131604,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004317271703284154,
      "loss": 5.5401,
      "mean_token_accuracy": 0.18214180022478105,
      "num_tokens": 51301813.0,
      "step": 26225
    },
    {
      "entropy": 5.905159425735474,
      "epoch": 2.6219823061928325,
      "grad_norm": 1.125,
      "learning_rate": 0.00043170156165852887,
      "loss": 5.516,
      "mean_token_accuracy": 0.18439912796020508,
      "num_tokens": 51312378.0,
      "step": 26230
    },
    {
      "entropy": 5.923118257522583,
      "epoch": 2.622482131254061,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004316759490460713,
      "loss": 5.5386,
      "mean_token_accuracy": 0.1934283345937729,
      "num_tokens": 51322022.0,
      "step": 26235
    },
    {
      "entropy": 5.93846025466919,
      "epoch": 2.6229819563152894,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043165033249168723,
      "loss": 5.5169,
      "mean_token_accuracy": 0.18454489558935167,
      "num_tokens": 51330288.0,
      "step": 26240
    },
    {
      "entropy": 5.927955865859985,
      "epoch": 2.623481781376518,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043162471199602103,
      "loss": 5.5742,
      "mean_token_accuracy": 0.18369664996862411,
      "num_tokens": 51338893.0,
      "step": 26245
    },
    {
      "entropy": 5.927968597412109,
      "epoch": 2.623981606437747,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043159908755971746,
      "loss": 5.5248,
      "mean_token_accuracy": 0.18419827967882157,
      "num_tokens": 51347860.0,
      "step": 26250
    },
    {
      "entropy": 5.9468591690063475,
      "epoch": 2.6244814314989755,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043157345918342103,
      "loss": 5.5661,
      "mean_token_accuracy": 0.18438886255025863,
      "num_tokens": 51359240.0,
      "step": 26255
    },
    {
      "entropy": 6.031884813308716,
      "epoch": 2.624981256560204,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004315478268677768,
      "loss": 5.6574,
      "mean_token_accuracy": 0.17782581150531768,
      "num_tokens": 51368574.0,
      "step": 26260
    },
    {
      "entropy": 6.008184957504272,
      "epoch": 2.6254810816214325,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004315221906134294,
      "loss": 5.561,
      "mean_token_accuracy": 0.18194278031587602,
      "num_tokens": 51378895.0,
      "step": 26265
    },
    {
      "entropy": 5.946164321899414,
      "epoch": 2.625980906682661,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043149655042102404,
      "loss": 5.5046,
      "mean_token_accuracy": 0.1891786813735962,
      "num_tokens": 51388659.0,
      "step": 26270
    },
    {
      "entropy": 6.008917665481567,
      "epoch": 2.6264807317438894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043147090629120575,
      "loss": 5.6531,
      "mean_token_accuracy": 0.18018709272146224,
      "num_tokens": 51398444.0,
      "step": 26275
    },
    {
      "entropy": 5.966279411315918,
      "epoch": 2.626980556805118,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004314452582246197,
      "loss": 5.517,
      "mean_token_accuracy": 0.18748368322849274,
      "num_tokens": 51408534.0,
      "step": 26280
    },
    {
      "entropy": 5.966874313354492,
      "epoch": 2.627480381866347,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043141960622191116,
      "loss": 5.5486,
      "mean_token_accuracy": 0.18316874504089356,
      "num_tokens": 51418174.0,
      "step": 26285
    },
    {
      "entropy": 5.823508930206299,
      "epoch": 2.6279802069275755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043139395028372565,
      "loss": 5.5774,
      "mean_token_accuracy": 0.1876854658126831,
      "num_tokens": 51428249.0,
      "step": 26290
    },
    {
      "entropy": 5.934199810028076,
      "epoch": 2.6284800319888038,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043136829041070856,
      "loss": 5.5401,
      "mean_token_accuracy": 0.1835727274417877,
      "num_tokens": 51438042.0,
      "step": 26295
    },
    {
      "entropy": 6.0275318145751955,
      "epoch": 2.6289798570500325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043134262660350555,
      "loss": 5.6188,
      "mean_token_accuracy": 0.17861606925725937,
      "num_tokens": 51447745.0,
      "step": 26300
    },
    {
      "entropy": 6.030848741531372,
      "epoch": 2.629479682111261,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004313169588627622,
      "loss": 5.5037,
      "mean_token_accuracy": 0.1838701382279396,
      "num_tokens": 51458163.0,
      "step": 26305
    },
    {
      "entropy": 5.982316875457764,
      "epoch": 2.6299795071724894,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004312912871891245,
      "loss": 5.6008,
      "mean_token_accuracy": 0.18121372610330583,
      "num_tokens": 51468753.0,
      "step": 26310
    },
    {
      "entropy": 5.961738204956054,
      "epoch": 2.630479332233718,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004312656115832381,
      "loss": 5.6315,
      "mean_token_accuracy": 0.1766458585858345,
      "num_tokens": 51479411.0,
      "step": 26315
    },
    {
      "entropy": 6.007168865203857,
      "epoch": 2.630979157294947,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043123993204574926,
      "loss": 5.6579,
      "mean_token_accuracy": 0.17856726348400115,
      "num_tokens": 51489663.0,
      "step": 26320
    },
    {
      "entropy": 6.048077154159546,
      "epoch": 2.6314789823561755,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043121424857730384,
      "loss": 5.601,
      "mean_token_accuracy": 0.17898371815681458,
      "num_tokens": 51500455.0,
      "step": 26325
    },
    {
      "entropy": 5.952033758163452,
      "epoch": 2.6319788074174038,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004311885611785482,
      "loss": 5.5623,
      "mean_token_accuracy": 0.1798251062631607,
      "num_tokens": 51509536.0,
      "step": 26330
    },
    {
      "entropy": 5.913718938827515,
      "epoch": 2.6324786324786325,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043116286985012847,
      "loss": 5.4461,
      "mean_token_accuracy": 0.18317161202430726,
      "num_tokens": 51519016.0,
      "step": 26335
    },
    {
      "entropy": 5.9349018096923825,
      "epoch": 2.632978457539861,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043113717459269116,
      "loss": 5.5024,
      "mean_token_accuracy": 0.18790213614702225,
      "num_tokens": 51528165.0,
      "step": 26340
    },
    {
      "entropy": 5.991881942749023,
      "epoch": 2.6334782826010894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004311114754068827,
      "loss": 5.6001,
      "mean_token_accuracy": 0.18534514009952546,
      "num_tokens": 51538338.0,
      "step": 26345
    },
    {
      "entropy": 5.944768285751342,
      "epoch": 2.633978107662318,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004310857722933497,
      "loss": 5.6136,
      "mean_token_accuracy": 0.18205619603395462,
      "num_tokens": 51547397.0,
      "step": 26350
    },
    {
      "entropy": 5.907018709182739,
      "epoch": 2.634477932723547,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043106006525273896,
      "loss": 5.5102,
      "mean_token_accuracy": 0.18308191150426864,
      "num_tokens": 51558572.0,
      "step": 26355
    },
    {
      "entropy": 5.848873138427734,
      "epoch": 2.6349777577847755,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004310343542856969,
      "loss": 5.3821,
      "mean_token_accuracy": 0.20078802853822708,
      "num_tokens": 51567522.0,
      "step": 26360
    },
    {
      "entropy": 5.9344275951385494,
      "epoch": 2.6354775828460038,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043100863939287077,
      "loss": 5.5748,
      "mean_token_accuracy": 0.17958331108093262,
      "num_tokens": 51577717.0,
      "step": 26365
    },
    {
      "entropy": 5.968832302093506,
      "epoch": 2.6359774079072325,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004309829205749074,
      "loss": 5.6,
      "mean_token_accuracy": 0.1861987143754959,
      "num_tokens": 51587301.0,
      "step": 26370
    },
    {
      "entropy": 5.949429702758789,
      "epoch": 2.636477232968461,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043095719783245386,
      "loss": 5.5665,
      "mean_token_accuracy": 0.18817090094089509,
      "num_tokens": 51597538.0,
      "step": 26375
    },
    {
      "entropy": 5.9550577163696286,
      "epoch": 2.6369770580296894,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043093147116615737,
      "loss": 5.47,
      "mean_token_accuracy": 0.19732449650764466,
      "num_tokens": 51607698.0,
      "step": 26380
    },
    {
      "entropy": 5.943111228942871,
      "epoch": 2.637476883090918,
      "grad_norm": 1.125,
      "learning_rate": 0.0004309057405766652,
      "loss": 5.622,
      "mean_token_accuracy": 0.18244528621435166,
      "num_tokens": 51616784.0,
      "step": 26385
    },
    {
      "entropy": 5.961528253555298,
      "epoch": 2.637976708152147,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043088000606462474,
      "loss": 5.6317,
      "mean_token_accuracy": 0.1817688152194023,
      "num_tokens": 51627024.0,
      "step": 26390
    },
    {
      "entropy": 5.9824292182922365,
      "epoch": 2.6384765332133755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043085426763068334,
      "loss": 5.5535,
      "mean_token_accuracy": 0.18193780928850173,
      "num_tokens": 51636224.0,
      "step": 26395
    },
    {
      "entropy": 5.889959716796875,
      "epoch": 2.6389763582746038,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043082852527548876,
      "loss": 5.5679,
      "mean_token_accuracy": 0.18034346401691437,
      "num_tokens": 51645457.0,
      "step": 26400
    },
    {
      "entropy": 5.956007480621338,
      "epoch": 2.6394761833358324,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004308027789996886,
      "loss": 5.5428,
      "mean_token_accuracy": 0.18963379859924318,
      "num_tokens": 51655649.0,
      "step": 26405
    },
    {
      "entropy": 5.931284761428833,
      "epoch": 2.639976008397061,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004307770288039306,
      "loss": 5.5814,
      "mean_token_accuracy": 0.18020458221435548,
      "num_tokens": 51664937.0,
      "step": 26410
    },
    {
      "entropy": 5.967978000640869,
      "epoch": 2.6404758334582894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043075127468886264,
      "loss": 5.5017,
      "mean_token_accuracy": 0.18372277021408082,
      "num_tokens": 51674473.0,
      "step": 26415
    },
    {
      "entropy": 5.9666236400604244,
      "epoch": 2.640975658519518,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043072551665513265,
      "loss": 5.5048,
      "mean_token_accuracy": 0.18169274181127548,
      "num_tokens": 51684510.0,
      "step": 26420
    },
    {
      "entropy": 5.879351997375489,
      "epoch": 2.641475483580747,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043069975470338875,
      "loss": 5.5523,
      "mean_token_accuracy": 0.1872408300638199,
      "num_tokens": 51693220.0,
      "step": 26425
    },
    {
      "entropy": 6.0288002490997314,
      "epoch": 2.6419753086419755,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043067398883427907,
      "loss": 5.6094,
      "mean_token_accuracy": 0.18379888832569122,
      "num_tokens": 51702396.0,
      "step": 26430
    },
    {
      "entropy": 6.061272382736206,
      "epoch": 2.6424751337032037,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043064821904845185,
      "loss": 5.531,
      "mean_token_accuracy": 0.18812179267406465,
      "num_tokens": 51712651.0,
      "step": 26435
    },
    {
      "entropy": 5.940893507003784,
      "epoch": 2.6429749587644324,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004306224453465556,
      "loss": 5.6111,
      "mean_token_accuracy": 0.1784769967198372,
      "num_tokens": 51722247.0,
      "step": 26440
    },
    {
      "entropy": 5.985502624511719,
      "epoch": 2.643474783825661,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043059666772923854,
      "loss": 5.5872,
      "mean_token_accuracy": 0.18224239647388457,
      "num_tokens": 51731537.0,
      "step": 26445
    },
    {
      "entropy": 6.053400945663452,
      "epoch": 2.6439746088868894,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004305708861971495,
      "loss": 5.6401,
      "mean_token_accuracy": 0.17791666686534882,
      "num_tokens": 51741661.0,
      "step": 26450
    },
    {
      "entropy": 5.932127237319946,
      "epoch": 2.644474433948118,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004305451007509369,
      "loss": 5.4798,
      "mean_token_accuracy": 0.1884821653366089,
      "num_tokens": 51752813.0,
      "step": 26455
    },
    {
      "entropy": 5.995592832565308,
      "epoch": 2.644974259009347,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004305193113912496,
      "loss": 5.5527,
      "mean_token_accuracy": 0.1870158866047859,
      "num_tokens": 51761690.0,
      "step": 26460
    },
    {
      "entropy": 5.980676078796387,
      "epoch": 2.6454740840705755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043049351811873645,
      "loss": 5.4949,
      "mean_token_accuracy": 0.1846751719713211,
      "num_tokens": 51772002.0,
      "step": 26465
    },
    {
      "entropy": 5.979540300369263,
      "epoch": 2.6459739091318037,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004304677209340464,
      "loss": 5.6593,
      "mean_token_accuracy": 0.18002594709396363,
      "num_tokens": 51783076.0,
      "step": 26470
    },
    {
      "entropy": 5.918977212905884,
      "epoch": 2.6464737341930324,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043044191983782846,
      "loss": 5.5053,
      "mean_token_accuracy": 0.189929062128067,
      "num_tokens": 51792710.0,
      "step": 26475
    },
    {
      "entropy": 6.0233148574829105,
      "epoch": 2.646973559254261,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043041611483073187,
      "loss": 5.7023,
      "mean_token_accuracy": 0.17655133605003356,
      "num_tokens": 51802896.0,
      "step": 26480
    },
    {
      "entropy": 5.939725494384765,
      "epoch": 2.6474733843154894,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004303903059134059,
      "loss": 5.5133,
      "mean_token_accuracy": 0.18861608505249022,
      "num_tokens": 51812363.0,
      "step": 26485
    },
    {
      "entropy": 5.949069404602051,
      "epoch": 2.647973209376718,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043036449308649967,
      "loss": 5.5174,
      "mean_token_accuracy": 0.19252956360578538,
      "num_tokens": 51820970.0,
      "step": 26490
    },
    {
      "entropy": 6.0310906887054445,
      "epoch": 2.648473034437947,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043033867635066287,
      "loss": 5.5453,
      "mean_token_accuracy": 0.1818703830242157,
      "num_tokens": 51830115.0,
      "step": 26495
    },
    {
      "entropy": 5.944020843505859,
      "epoch": 2.6489728594991755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004303128557065448,
      "loss": 5.489,
      "mean_token_accuracy": 0.19189443439245224,
      "num_tokens": 51839779.0,
      "step": 26500
    },
    {
      "entropy": 5.924003458023071,
      "epoch": 2.6494726845604037,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004302870311547954,
      "loss": 5.575,
      "mean_token_accuracy": 0.1817442148923874,
      "num_tokens": 51850800.0,
      "step": 26505
    },
    {
      "entropy": 5.955606746673584,
      "epoch": 2.6499725096216324,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043026120269606423,
      "loss": 5.565,
      "mean_token_accuracy": 0.185788793861866,
      "num_tokens": 51860059.0,
      "step": 26510
    },
    {
      "entropy": 5.9952925682067875,
      "epoch": 2.650472334682861,
      "grad_norm": 1.0,
      "learning_rate": 0.00043023537033100106,
      "loss": 5.6423,
      "mean_token_accuracy": 0.17843436002731322,
      "num_tokens": 51870001.0,
      "step": 26515
    },
    {
      "entropy": 5.9974531650543215,
      "epoch": 2.6509721597440894,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043020953406025606,
      "loss": 5.4899,
      "mean_token_accuracy": 0.18765356987714768,
      "num_tokens": 51879004.0,
      "step": 26520
    },
    {
      "entropy": 5.950621318817139,
      "epoch": 2.651471984805318,
      "grad_norm": 1.109375,
      "learning_rate": 0.000430183693884479,
      "loss": 5.5635,
      "mean_token_accuracy": 0.18553271591663362,
      "num_tokens": 51887911.0,
      "step": 26525
    },
    {
      "entropy": 5.85861988067627,
      "epoch": 2.651971809866547,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004301578498043202,
      "loss": 5.4337,
      "mean_token_accuracy": 0.19449917376041412,
      "num_tokens": 51897444.0,
      "step": 26530
    },
    {
      "entropy": 5.868001556396484,
      "epoch": 2.6524716349277755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043013200182042983,
      "loss": 5.4607,
      "mean_token_accuracy": 0.19467050433158875,
      "num_tokens": 51907344.0,
      "step": 26535
    },
    {
      "entropy": 5.967486572265625,
      "epoch": 2.6529714599890037,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043010614993345815,
      "loss": 5.4826,
      "mean_token_accuracy": 0.18737678229808807,
      "num_tokens": 51916541.0,
      "step": 26540
    },
    {
      "entropy": 6.0161213874816895,
      "epoch": 2.6534712850502324,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004300802941440557,
      "loss": 5.6709,
      "mean_token_accuracy": 0.17724319100379943,
      "num_tokens": 51926160.0,
      "step": 26545
    },
    {
      "entropy": 5.962722110748291,
      "epoch": 2.653971110111461,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004300544344528729,
      "loss": 5.585,
      "mean_token_accuracy": 0.18213831931352614,
      "num_tokens": 51936244.0,
      "step": 26550
    },
    {
      "entropy": 6.022981739044189,
      "epoch": 2.6544709351726894,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043002857086056053,
      "loss": 5.5485,
      "mean_token_accuracy": 0.1860679194331169,
      "num_tokens": 51946490.0,
      "step": 26555
    },
    {
      "entropy": 6.00039324760437,
      "epoch": 2.654970760233918,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004300027033677691,
      "loss": 5.5469,
      "mean_token_accuracy": 0.18575784265995027,
      "num_tokens": 51956129.0,
      "step": 26560
    },
    {
      "entropy": 5.943972492218018,
      "epoch": 2.655470585295147,
      "grad_norm": 1.125,
      "learning_rate": 0.0004299768319751496,
      "loss": 5.5672,
      "mean_token_accuracy": 0.18115961402654648,
      "num_tokens": 51966274.0,
      "step": 26565
    },
    {
      "entropy": 5.893281269073486,
      "epoch": 2.6559704103563755,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004299509566833529,
      "loss": 5.4733,
      "mean_token_accuracy": 0.18993555158376693,
      "num_tokens": 51975261.0,
      "step": 26570
    },
    {
      "entropy": 5.96535816192627,
      "epoch": 2.6564702354176037,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042992507749302995,
      "loss": 5.5315,
      "mean_token_accuracy": 0.17715491354465485,
      "num_tokens": 51984562.0,
      "step": 26575
    },
    {
      "entropy": 5.939205551147461,
      "epoch": 2.6569700604788324,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042989919440483194,
      "loss": 5.509,
      "mean_token_accuracy": 0.18468733131885529,
      "num_tokens": 51994763.0,
      "step": 26580
    },
    {
      "entropy": 6.0091101169586185,
      "epoch": 2.657469885540061,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004298733074194101,
      "loss": 5.5889,
      "mean_token_accuracy": 0.18978189527988434,
      "num_tokens": 52004935.0,
      "step": 26585
    },
    {
      "entropy": 5.927107286453247,
      "epoch": 2.6579697106012894,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004298474165374156,
      "loss": 5.5169,
      "mean_token_accuracy": 0.18286526799201966,
      "num_tokens": 52013879.0,
      "step": 26590
    },
    {
      "entropy": 5.91258397102356,
      "epoch": 2.658469535662518,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004298215217595,
      "loss": 5.5264,
      "mean_token_accuracy": 0.19087222963571548,
      "num_tokens": 52023354.0,
      "step": 26595
    },
    {
      "entropy": 5.9937659740448,
      "epoch": 2.658969360723747,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042979562308631476,
      "loss": 5.6026,
      "mean_token_accuracy": 0.18009405434131623,
      "num_tokens": 52033058.0,
      "step": 26600
    },
    {
      "entropy": 6.02861909866333,
      "epoch": 2.6594691857849755,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004297697205185115,
      "loss": 5.5375,
      "mean_token_accuracy": 0.17394384294748305,
      "num_tokens": 52042544.0,
      "step": 26605
    },
    {
      "entropy": 5.961648035049438,
      "epoch": 2.6599690108462037,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004297438140567419,
      "loss": 5.5598,
      "mean_token_accuracy": 0.18225936591625214,
      "num_tokens": 52052396.0,
      "step": 26610
    },
    {
      "entropy": 6.007733106613159,
      "epoch": 2.6604688359074324,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004297179037016577,
      "loss": 5.5788,
      "mean_token_accuracy": 0.18162550330162047,
      "num_tokens": 52061950.0,
      "step": 26615
    },
    {
      "entropy": 5.935448074340821,
      "epoch": 2.6609686609686607,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042969198945391084,
      "loss": 5.5812,
      "mean_token_accuracy": 0.18699375987052919,
      "num_tokens": 52071588.0,
      "step": 26620
    },
    {
      "entropy": 5.914863967895508,
      "epoch": 2.6614684860298894,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004296660713141534,
      "loss": 5.463,
      "mean_token_accuracy": 0.18864685595035552,
      "num_tokens": 52080826.0,
      "step": 26625
    },
    {
      "entropy": 5.943517112731934,
      "epoch": 2.661968311091118,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042964014928303733,
      "loss": 5.6105,
      "mean_token_accuracy": 0.1797039568424225,
      "num_tokens": 52090844.0,
      "step": 26630
    },
    {
      "entropy": 6.038613271713257,
      "epoch": 2.662468136152347,
      "grad_norm": 1.15625,
      "learning_rate": 0.000429614223361215,
      "loss": 5.7038,
      "mean_token_accuracy": 0.18056536018848418,
      "num_tokens": 52100189.0,
      "step": 26635
    },
    {
      "entropy": 5.958257865905762,
      "epoch": 2.6629679612135755,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004295882935493385,
      "loss": 5.605,
      "mean_token_accuracy": 0.18312350064516067,
      "num_tokens": 52110323.0,
      "step": 26640
    },
    {
      "entropy": 5.911790561676026,
      "epoch": 2.6634677862748037,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042956235984806034,
      "loss": 5.4831,
      "mean_token_accuracy": 0.1885166049003601,
      "num_tokens": 52120816.0,
      "step": 26645
    },
    {
      "entropy": 6.005259275436401,
      "epoch": 2.6639676113360324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042953642225803296,
      "loss": 5.5878,
      "mean_token_accuracy": 0.18712491244077684,
      "num_tokens": 52130610.0,
      "step": 26650
    },
    {
      "entropy": 5.972778177261352,
      "epoch": 2.6644674363972607,
      "grad_norm": 1.078125,
      "learning_rate": 0.000429510480779909,
      "loss": 5.5565,
      "mean_token_accuracy": 0.18843211531639098,
      "num_tokens": 52140581.0,
      "step": 26655
    },
    {
      "entropy": 5.953852844238281,
      "epoch": 2.6649672614584894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004294845354143411,
      "loss": 5.5345,
      "mean_token_accuracy": 0.18971368968486785,
      "num_tokens": 52150971.0,
      "step": 26660
    },
    {
      "entropy": 5.93402681350708,
      "epoch": 2.665467086519718,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004294585861619821,
      "loss": 5.5586,
      "mean_token_accuracy": 0.1829776167869568,
      "num_tokens": 52160663.0,
      "step": 26665
    },
    {
      "entropy": 5.9209388256072994,
      "epoch": 2.665966911580947,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042943263302348475,
      "loss": 5.5537,
      "mean_token_accuracy": 0.186874121427536,
      "num_tokens": 52170570.0,
      "step": 26670
    },
    {
      "entropy": 6.041050815582276,
      "epoch": 2.6664667366421755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042940667599950205,
      "loss": 5.6429,
      "mean_token_accuracy": 0.17836940437555313,
      "num_tokens": 52179682.0,
      "step": 26675
    },
    {
      "entropy": 5.942004728317261,
      "epoch": 2.6669665617034037,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042938071509068717,
      "loss": 5.5903,
      "mean_token_accuracy": 0.18712148666381836,
      "num_tokens": 52189856.0,
      "step": 26680
    },
    {
      "entropy": 5.915938043594361,
      "epoch": 2.6674663867646324,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042935475029769323,
      "loss": 5.429,
      "mean_token_accuracy": 0.1909005180001259,
      "num_tokens": 52197956.0,
      "step": 26685
    },
    {
      "entropy": 5.848339700698853,
      "epoch": 2.6679662118258607,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042932878162117347,
      "loss": 5.4267,
      "mean_token_accuracy": 0.19939057826995848,
      "num_tokens": 52208091.0,
      "step": 26690
    },
    {
      "entropy": 5.969365501403809,
      "epoch": 2.6684660368870894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004293028090617813,
      "loss": 5.6316,
      "mean_token_accuracy": 0.17687974125146866,
      "num_tokens": 52218568.0,
      "step": 26695
    },
    {
      "entropy": 5.934926891326905,
      "epoch": 2.668965861948318,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042927683262017016,
      "loss": 5.4873,
      "mean_token_accuracy": 0.19037033915519713,
      "num_tokens": 52228048.0,
      "step": 26700
    },
    {
      "entropy": 5.959352016448975,
      "epoch": 2.669465687009547,
      "grad_norm": 1.125,
      "learning_rate": 0.0004292508522969935,
      "loss": 5.6123,
      "mean_token_accuracy": 0.17843787968158722,
      "num_tokens": 52239878.0,
      "step": 26705
    },
    {
      "entropy": 5.919369459152222,
      "epoch": 2.6699655120707755,
      "grad_norm": 1.125,
      "learning_rate": 0.0004292248680929052,
      "loss": 5.6032,
      "mean_token_accuracy": 0.18480545282363892,
      "num_tokens": 52250244.0,
      "step": 26710
    },
    {
      "entropy": 5.961813735961914,
      "epoch": 2.6704653371320037,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042919888000855885,
      "loss": 5.5773,
      "mean_token_accuracy": 0.18477069288492204,
      "num_tokens": 52261420.0,
      "step": 26715
    },
    {
      "entropy": 5.882527542114258,
      "epoch": 2.6709651621932324,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042917288804460835,
      "loss": 5.467,
      "mean_token_accuracy": 0.19145268052816392,
      "num_tokens": 52270956.0,
      "step": 26720
    },
    {
      "entropy": 5.970886182785034,
      "epoch": 2.6714649872544607,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004291468922017078,
      "loss": 5.5315,
      "mean_token_accuracy": 0.18390308916568757,
      "num_tokens": 52280467.0,
      "step": 26725
    },
    {
      "entropy": 5.970154333114624,
      "epoch": 2.6719648123156894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000429120892480511,
      "loss": 5.5505,
      "mean_token_accuracy": 0.18874867111444474,
      "num_tokens": 52290630.0,
      "step": 26730
    },
    {
      "entropy": 5.937599945068359,
      "epoch": 2.672464637376918,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004290948888816722,
      "loss": 5.5965,
      "mean_token_accuracy": 0.18071547150611877,
      "num_tokens": 52300956.0,
      "step": 26735
    },
    {
      "entropy": 5.949377536773682,
      "epoch": 2.672964462438147,
      "grad_norm": 1.125,
      "learning_rate": 0.0004290688814058456,
      "loss": 5.6079,
      "mean_token_accuracy": 0.17945620715618132,
      "num_tokens": 52310887.0,
      "step": 26740
    },
    {
      "entropy": 5.936476802825927,
      "epoch": 2.6734642874993755,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004290428700536857,
      "loss": 5.482,
      "mean_token_accuracy": 0.19204075187444686,
      "num_tokens": 52320490.0,
      "step": 26745
    },
    {
      "entropy": 6.003557872772217,
      "epoch": 2.6739641125606037,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004290168548258467,
      "loss": 5.5828,
      "mean_token_accuracy": 0.1793288215994835,
      "num_tokens": 52330477.0,
      "step": 26750
    },
    {
      "entropy": 5.983366441726685,
      "epoch": 2.6744639376218324,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042899083572298335,
      "loss": 5.5552,
      "mean_token_accuracy": 0.18888159841299057,
      "num_tokens": 52339909.0,
      "step": 26755
    },
    {
      "entropy": 5.934063720703125,
      "epoch": 2.6749637626830607,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042896481274575014,
      "loss": 5.4513,
      "mean_token_accuracy": 0.19535102993249892,
      "num_tokens": 52349554.0,
      "step": 26760
    },
    {
      "entropy": 5.931085634231567,
      "epoch": 2.6754635877442894,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042893878589480195,
      "loss": 5.5298,
      "mean_token_accuracy": 0.18608884066343306,
      "num_tokens": 52359561.0,
      "step": 26765
    },
    {
      "entropy": 5.823393058776856,
      "epoch": 2.675963412805518,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042891275517079345,
      "loss": 5.3965,
      "mean_token_accuracy": 0.1972106084227562,
      "num_tokens": 52369360.0,
      "step": 26770
    },
    {
      "entropy": 6.0458026885986325,
      "epoch": 2.676463237866747,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042888672057437964,
      "loss": 5.6233,
      "mean_token_accuracy": 0.1764642231166363,
      "num_tokens": 52379433.0,
      "step": 26775
    },
    {
      "entropy": 5.983569097518921,
      "epoch": 2.6769630629279755,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042886068210621557,
      "loss": 5.561,
      "mean_token_accuracy": 0.19007329344749452,
      "num_tokens": 52389640.0,
      "step": 26780
    },
    {
      "entropy": 5.849414587020874,
      "epoch": 2.6774628879892037,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004288346397669563,
      "loss": 5.4741,
      "mean_token_accuracy": 0.19602277427911757,
      "num_tokens": 52399542.0,
      "step": 26785
    },
    {
      "entropy": 5.978439044952393,
      "epoch": 2.6779627130504324,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004288085935572571,
      "loss": 5.5529,
      "mean_token_accuracy": 0.187151937186718,
      "num_tokens": 52408366.0,
      "step": 26790
    },
    {
      "entropy": 5.9489216804504395,
      "epoch": 2.6784625381116607,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004287825434777733,
      "loss": 5.5491,
      "mean_token_accuracy": 0.1811882048845291,
      "num_tokens": 52419047.0,
      "step": 26795
    },
    {
      "entropy": 5.9229436874389645,
      "epoch": 2.6789623631728894,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042875648952916024,
      "loss": 5.4407,
      "mean_token_accuracy": 0.19851385354995726,
      "num_tokens": 52428871.0,
      "step": 26800
    },
    {
      "entropy": 5.897235965728759,
      "epoch": 2.679462188234118,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004287304317120734,
      "loss": 5.5892,
      "mean_token_accuracy": 0.18642741739749907,
      "num_tokens": 52439385.0,
      "step": 26805
    },
    {
      "entropy": 5.958187007904053,
      "epoch": 2.679962013295347,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042870437002716854,
      "loss": 5.5342,
      "mean_token_accuracy": 0.1900055781006813,
      "num_tokens": 52449844.0,
      "step": 26810
    },
    {
      "entropy": 5.887377500534058,
      "epoch": 2.680461838356575,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004286783044751013,
      "loss": 5.5164,
      "mean_token_accuracy": 0.18819458782672882,
      "num_tokens": 52458853.0,
      "step": 26815
    },
    {
      "entropy": 5.963205480575562,
      "epoch": 2.6809616634178037,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004286522350565274,
      "loss": 5.5878,
      "mean_token_accuracy": 0.1803748995065689,
      "num_tokens": 52468321.0,
      "step": 26820
    },
    {
      "entropy": 5.96819806098938,
      "epoch": 2.6814614884790324,
      "grad_norm": 1.125,
      "learning_rate": 0.00042862616177210286,
      "loss": 5.5778,
      "mean_token_accuracy": 0.18172797113656997,
      "num_tokens": 52478015.0,
      "step": 26825
    },
    {
      "entropy": 5.920689916610717,
      "epoch": 2.6819613135402607,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004286000846224836,
      "loss": 5.4871,
      "mean_token_accuracy": 0.18538208305835724,
      "num_tokens": 52487589.0,
      "step": 26830
    },
    {
      "entropy": 5.94365291595459,
      "epoch": 2.6824611386014894,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042857400360832575,
      "loss": 5.5945,
      "mean_token_accuracy": 0.1871691957116127,
      "num_tokens": 52496375.0,
      "step": 26835
    },
    {
      "entropy": 5.9958020687103275,
      "epoch": 2.682960963662718,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004285479187302854,
      "loss": 5.6144,
      "mean_token_accuracy": 0.18173079937696457,
      "num_tokens": 52506469.0,
      "step": 26840
    },
    {
      "entropy": 6.07341046333313,
      "epoch": 2.6834607887239468,
      "grad_norm": 1.125,
      "learning_rate": 0.000428521829989019,
      "loss": 5.6019,
      "mean_token_accuracy": 0.1772075980901718,
      "num_tokens": 52515748.0,
      "step": 26845
    },
    {
      "entropy": 5.982581615447998,
      "epoch": 2.683960613785175,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004284957373851829,
      "loss": 5.5616,
      "mean_token_accuracy": 0.18111457526683808,
      "num_tokens": 52524742.0,
      "step": 26850
    },
    {
      "entropy": 5.985930919647217,
      "epoch": 2.6844604388464037,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042846964091943347,
      "loss": 5.5789,
      "mean_token_accuracy": 0.17728432267904282,
      "num_tokens": 52534502.0,
      "step": 26855
    },
    {
      "entropy": 5.972628402709961,
      "epoch": 2.6849602639076324,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042844354059242735,
      "loss": 5.4904,
      "mean_token_accuracy": 0.18239665478467942,
      "num_tokens": 52544217.0,
      "step": 26860
    },
    {
      "entropy": 5.993966913223266,
      "epoch": 2.6854600889688607,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004284174364048213,
      "loss": 5.5016,
      "mean_token_accuracy": 0.18658897131681443,
      "num_tokens": 52555066.0,
      "step": 26865
    },
    {
      "entropy": 5.933136653900147,
      "epoch": 2.6859599140300894,
      "grad_norm": 1.0625,
      "learning_rate": 0.000428391328357272,
      "loss": 5.5635,
      "mean_token_accuracy": 0.1815538451075554,
      "num_tokens": 52564957.0,
      "step": 26870
    },
    {
      "entropy": 5.94973349571228,
      "epoch": 2.686459739091318,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004283652164504364,
      "loss": 5.5141,
      "mean_token_accuracy": 0.18283136188983917,
      "num_tokens": 52574893.0,
      "step": 26875
    },
    {
      "entropy": 5.914629411697388,
      "epoch": 2.6869595641525468,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042833910068497134,
      "loss": 5.4282,
      "mean_token_accuracy": 0.19936186373233794,
      "num_tokens": 52583307.0,
      "step": 26880
    },
    {
      "entropy": 5.903750419616699,
      "epoch": 2.687459389213775,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000428312981061534,
      "loss": 5.4892,
      "mean_token_accuracy": 0.18889988213777542,
      "num_tokens": 52593055.0,
      "step": 26885
    },
    {
      "entropy": 5.943116140365601,
      "epoch": 2.6879592142750037,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004282868575807815,
      "loss": 5.4755,
      "mean_token_accuracy": 0.1866347849369049,
      "num_tokens": 52602068.0,
      "step": 26890
    },
    {
      "entropy": 5.992987251281738,
      "epoch": 2.6884590393362324,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042826073024337117,
      "loss": 5.5016,
      "mean_token_accuracy": 0.18742504864931106,
      "num_tokens": 52609964.0,
      "step": 26895
    },
    {
      "entropy": 6.023123931884766,
      "epoch": 2.6889588643974607,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004282345990499602,
      "loss": 5.559,
      "mean_token_accuracy": 0.18147613257169723,
      "num_tokens": 52619144.0,
      "step": 26900
    },
    {
      "entropy": 6.0446311950683596,
      "epoch": 2.6894586894586894,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004282084640012062,
      "loss": 5.6445,
      "mean_token_accuracy": 0.17465363889932634,
      "num_tokens": 52629508.0,
      "step": 26905
    },
    {
      "entropy": 5.970034790039063,
      "epoch": 2.689958514519918,
      "grad_norm": 1.125,
      "learning_rate": 0.0004281823250977667,
      "loss": 5.5551,
      "mean_token_accuracy": 0.17844363301992416,
      "num_tokens": 52640037.0,
      "step": 26910
    },
    {
      "entropy": 5.911353302001953,
      "epoch": 2.6904583395811468,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042815618234029923,
      "loss": 5.4831,
      "mean_token_accuracy": 0.1889389917254448,
      "num_tokens": 52649293.0,
      "step": 26915
    },
    {
      "entropy": 5.977318143844604,
      "epoch": 2.690958164642375,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042813003572946163,
      "loss": 5.5547,
      "mean_token_accuracy": 0.18165612667798997,
      "num_tokens": 52659560.0,
      "step": 26920
    },
    {
      "entropy": 6.001524877548218,
      "epoch": 2.6914579897036037,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042810388526591184,
      "loss": 5.6947,
      "mean_token_accuracy": 0.17837398797273635,
      "num_tokens": 52669105.0,
      "step": 26925
    },
    {
      "entropy": 5.993082571029663,
      "epoch": 2.6919578147648324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042807773095030757,
      "loss": 5.4618,
      "mean_token_accuracy": 0.20104843080043794,
      "num_tokens": 52679633.0,
      "step": 26930
    },
    {
      "entropy": 5.931420564651489,
      "epoch": 2.6924576398260607,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004280515727833071,
      "loss": 5.509,
      "mean_token_accuracy": 0.18811412006616593,
      "num_tokens": 52690037.0,
      "step": 26935
    },
    {
      "entropy": 6.004485034942627,
      "epoch": 2.6929574648872894,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004280254107655684,
      "loss": 5.6104,
      "mean_token_accuracy": 0.17599117010831833,
      "num_tokens": 52701796.0,
      "step": 26940
    },
    {
      "entropy": 5.88510913848877,
      "epoch": 2.693457289948518,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004279992448977497,
      "loss": 5.3567,
      "mean_token_accuracy": 0.19652786552906037,
      "num_tokens": 52710634.0,
      "step": 26945
    },
    {
      "entropy": 5.914271879196167,
      "epoch": 2.6939571150097468,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004279730751805094,
      "loss": 5.6132,
      "mean_token_accuracy": 0.1770920932292938,
      "num_tokens": 52721178.0,
      "step": 26950
    },
    {
      "entropy": 5.940487289428711,
      "epoch": 2.694456940070975,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042794690161450583,
      "loss": 5.4654,
      "mean_token_accuracy": 0.19203825145959855,
      "num_tokens": 52730635.0,
      "step": 26955
    },
    {
      "entropy": 5.968632459640503,
      "epoch": 2.6949567651322037,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004279207242003976,
      "loss": 5.4835,
      "mean_token_accuracy": 0.1940087616443634,
      "num_tokens": 52740711.0,
      "step": 26960
    },
    {
      "entropy": 5.879199695587158,
      "epoch": 2.6954565901934324,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042789454293884336,
      "loss": 5.5586,
      "mean_token_accuracy": 0.18821661323308944,
      "num_tokens": 52749430.0,
      "step": 26965
    },
    {
      "entropy": 5.939274883270263,
      "epoch": 2.6959564152546607,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004278683578305017,
      "loss": 5.481,
      "mean_token_accuracy": 0.18932652175426484,
      "num_tokens": 52758288.0,
      "step": 26970
    },
    {
      "entropy": 6.006446838378906,
      "epoch": 2.6964562403158894,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042784216887603154,
      "loss": 5.5723,
      "mean_token_accuracy": 0.18246660232543946,
      "num_tokens": 52768190.0,
      "step": 26975
    },
    {
      "entropy": 5.959181880950927,
      "epoch": 2.696956065377118,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004278159760760917,
      "loss": 5.557,
      "mean_token_accuracy": 0.18353258371353148,
      "num_tokens": 52779262.0,
      "step": 26980
    },
    {
      "entropy": 5.955266857147217,
      "epoch": 2.6974558904383468,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004277897794313412,
      "loss": 5.5644,
      "mean_token_accuracy": 0.18507759720087053,
      "num_tokens": 52788564.0,
      "step": 26985
    },
    {
      "entropy": 5.996228265762329,
      "epoch": 2.697955715499575,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004277635789424393,
      "loss": 5.5576,
      "mean_token_accuracy": 0.18403831720352173,
      "num_tokens": 52797293.0,
      "step": 26990
    },
    {
      "entropy": 5.910314655303955,
      "epoch": 2.6984555405608037,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004277373746100449,
      "loss": 5.4752,
      "mean_token_accuracy": 0.18501528203487397,
      "num_tokens": 52806815.0,
      "step": 26995
    },
    {
      "entropy": 5.895352125167847,
      "epoch": 2.6989553656220324,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042771116643481753,
      "loss": 5.4964,
      "mean_token_accuracy": 0.1871676489710808,
      "num_tokens": 52815496.0,
      "step": 27000
    },
    {
      "epoch": 2.6989553656220324,
      "eval_entropy": 5.764868384874595,
      "eval_loss": 5.6601972579956055,
      "eval_mean_token_accuracy": 0.18797387170522092,
      "eval_num_tokens": 52815496.0,
      "eval_runtime": 23.8061,
      "eval_samples_per_second": 1304.075,
      "eval_steps_per_second": 163.025,
      "step": 27000
    },
    {
      "entropy": 6.030625200271606,
      "epoch": 2.6994551906832607,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004276849544174166,
      "loss": 5.6302,
      "mean_token_accuracy": 0.17612940669059754,
      "num_tokens": 52825203.0,
      "step": 27005
    },
    {
      "entropy": 5.958075761795044,
      "epoch": 2.6999550157444894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042765873855850133,
      "loss": 5.5787,
      "mean_token_accuracy": 0.1831500694155693,
      "num_tokens": 52835956.0,
      "step": 27010
    },
    {
      "entropy": 5.873317241668701,
      "epoch": 2.700454840805718,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004276325188587316,
      "loss": 5.5419,
      "mean_token_accuracy": 0.19293804466724396,
      "num_tokens": 52845314.0,
      "step": 27015
    },
    {
      "entropy": 5.941677379608154,
      "epoch": 2.7009546658669468,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004276062953187669,
      "loss": 5.5277,
      "mean_token_accuracy": 0.1852804660797119,
      "num_tokens": 52855693.0,
      "step": 27020
    },
    {
      "entropy": 5.9989828109741214,
      "epoch": 2.701454490928175,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004275800679392671,
      "loss": 5.5594,
      "mean_token_accuracy": 0.18408535271883011,
      "num_tokens": 52865931.0,
      "step": 27025
    },
    {
      "entropy": 5.983813762664795,
      "epoch": 2.7019543159894037,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004275538367208921,
      "loss": 5.5404,
      "mean_token_accuracy": 0.1885357454419136,
      "num_tokens": 52876329.0,
      "step": 27030
    },
    {
      "entropy": 6.0213549613952635,
      "epoch": 2.7024541410506324,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042752760166430185,
      "loss": 5.5738,
      "mean_token_accuracy": 0.1776261016726494,
      "num_tokens": 52886876.0,
      "step": 27035
    },
    {
      "entropy": 5.963567686080933,
      "epoch": 2.7029539661118607,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042750136277015625,
      "loss": 5.5814,
      "mean_token_accuracy": 0.1824508234858513,
      "num_tokens": 52896584.0,
      "step": 27040
    },
    {
      "entropy": 5.994350528717041,
      "epoch": 2.7034537911730894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042747512003911577,
      "loss": 5.5197,
      "mean_token_accuracy": 0.18813224285840988,
      "num_tokens": 52906388.0,
      "step": 27045
    },
    {
      "entropy": 5.913507699966431,
      "epoch": 2.703953616234318,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042744887347184035,
      "loss": 5.4423,
      "mean_token_accuracy": 0.19358773529529572,
      "num_tokens": 52915732.0,
      "step": 27050
    },
    {
      "entropy": 5.936895561218262,
      "epoch": 2.7044534412955468,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004274226230689906,
      "loss": 5.6128,
      "mean_token_accuracy": 0.17616986334323884,
      "num_tokens": 52926066.0,
      "step": 27055
    },
    {
      "entropy": 5.962563323974609,
      "epoch": 2.704953266356775,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042739636883122687,
      "loss": 5.5485,
      "mean_token_accuracy": 0.18316370397806167,
      "num_tokens": 52936075.0,
      "step": 27060
    },
    {
      "entropy": 5.951058006286621,
      "epoch": 2.7054530914180037,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042737011075920974,
      "loss": 5.5508,
      "mean_token_accuracy": 0.18243214190006257,
      "num_tokens": 52944266.0,
      "step": 27065
    },
    {
      "entropy": 6.032708072662354,
      "epoch": 2.7059529164792324,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004273438488535997,
      "loss": 5.646,
      "mean_token_accuracy": 0.17957666516304016,
      "num_tokens": 52954999.0,
      "step": 27070
    },
    {
      "entropy": 6.0375950813293455,
      "epoch": 2.7064527415404607,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042731758311505773,
      "loss": 5.665,
      "mean_token_accuracy": 0.17092355042696,
      "num_tokens": 52965374.0,
      "step": 27075
    },
    {
      "entropy": 5.903464221954346,
      "epoch": 2.7069525666016894,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042729131354424455,
      "loss": 5.4594,
      "mean_token_accuracy": 0.19186852127313614,
      "num_tokens": 52975956.0,
      "step": 27080
    },
    {
      "entropy": 5.930834436416626,
      "epoch": 2.707452391662918,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004272650401418211,
      "loss": 5.465,
      "mean_token_accuracy": 0.18453575670719147,
      "num_tokens": 52985595.0,
      "step": 27085
    },
    {
      "entropy": 5.977646350860596,
      "epoch": 2.7079522167241468,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004272387629084484,
      "loss": 5.6342,
      "mean_token_accuracy": 0.17792598605155946,
      "num_tokens": 52994755.0,
      "step": 27090
    },
    {
      "entropy": 5.991314697265625,
      "epoch": 2.708452041785375,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042721248184478765,
      "loss": 5.5955,
      "mean_token_accuracy": 0.17647868841886521,
      "num_tokens": 53004529.0,
      "step": 27095
    },
    {
      "entropy": 6.0708318710327145,
      "epoch": 2.7089518668466037,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004271861969515,
      "loss": 5.7084,
      "mean_token_accuracy": 0.1772710084915161,
      "num_tokens": 53015223.0,
      "step": 27100
    },
    {
      "entropy": 6.053754568099976,
      "epoch": 2.7094516919078324,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042715990822924673,
      "loss": 5.6164,
      "mean_token_accuracy": 0.1825556717813015,
      "num_tokens": 53026014.0,
      "step": 27105
    },
    {
      "entropy": 6.000459957122803,
      "epoch": 2.7099515169690607,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042713361567868946,
      "loss": 5.5616,
      "mean_token_accuracy": 0.1819307878613472,
      "num_tokens": 53036463.0,
      "step": 27110
    },
    {
      "entropy": 5.99230728149414,
      "epoch": 2.7104513420302894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042710731930048945,
      "loss": 5.538,
      "mean_token_accuracy": 0.18691609799861908,
      "num_tokens": 53045933.0,
      "step": 27115
    },
    {
      "entropy": 5.878900384902954,
      "epoch": 2.710951167091518,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004270810190953085,
      "loss": 5.4491,
      "mean_token_accuracy": 0.1935974985361099,
      "num_tokens": 53056098.0,
      "step": 27120
    },
    {
      "entropy": 5.893324470520019,
      "epoch": 2.7114509921527468,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004270547150638082,
      "loss": 5.4577,
      "mean_token_accuracy": 0.18674020171165467,
      "num_tokens": 53065757.0,
      "step": 27125
    },
    {
      "entropy": 5.960865020751953,
      "epoch": 2.711950817213975,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042702840720665043,
      "loss": 5.4587,
      "mean_token_accuracy": 0.19416671395301818,
      "num_tokens": 53074616.0,
      "step": 27130
    },
    {
      "entropy": 5.952418994903565,
      "epoch": 2.7124506422752037,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042700209552449706,
      "loss": 5.6044,
      "mean_token_accuracy": 0.17725115567445754,
      "num_tokens": 53084861.0,
      "step": 27135
    },
    {
      "entropy": 6.022545862197876,
      "epoch": 2.7129504673364324,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042697578001801,
      "loss": 5.6009,
      "mean_token_accuracy": 0.17461971342563629,
      "num_tokens": 53095175.0,
      "step": 27140
    },
    {
      "entropy": 6.0543403148651125,
      "epoch": 2.7134502923976607,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004269494606878515,
      "loss": 5.5892,
      "mean_token_accuracy": 0.18205632865428925,
      "num_tokens": 53105094.0,
      "step": 27145
    },
    {
      "entropy": 6.021087503433227,
      "epoch": 2.7139501174588894,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004269231375346836,
      "loss": 5.587,
      "mean_token_accuracy": 0.1773891568183899,
      "num_tokens": 53115716.0,
      "step": 27150
    },
    {
      "entropy": 5.907338953018188,
      "epoch": 2.714449942520118,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004268968105591687,
      "loss": 5.4848,
      "mean_token_accuracy": 0.18863687813282012,
      "num_tokens": 53124375.0,
      "step": 27155
    },
    {
      "entropy": 5.8356822490692135,
      "epoch": 2.7149497675813468,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004268704797619692,
      "loss": 5.415,
      "mean_token_accuracy": 0.19506745040416718,
      "num_tokens": 53135220.0,
      "step": 27160
    },
    {
      "entropy": 5.916583681106568,
      "epoch": 2.715449592642575,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042684414514374734,
      "loss": 5.4592,
      "mean_token_accuracy": 0.1947772115468979,
      "num_tokens": 53144424.0,
      "step": 27165
    },
    {
      "entropy": 5.923218011856079,
      "epoch": 2.7159494177038037,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042681780670516595,
      "loss": 5.5015,
      "mean_token_accuracy": 0.18810458183288575,
      "num_tokens": 53154611.0,
      "step": 27170
    },
    {
      "entropy": 6.053368997573853,
      "epoch": 2.716449242765032,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004267914644468877,
      "loss": 5.688,
      "mean_token_accuracy": 0.17553775608539582,
      "num_tokens": 53163061.0,
      "step": 27175
    },
    {
      "entropy": 5.948930168151856,
      "epoch": 2.7169490678262607,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004267651183695752,
      "loss": 5.4905,
      "mean_token_accuracy": 0.1887852892279625,
      "num_tokens": 53173078.0,
      "step": 27180
    },
    {
      "entropy": 5.9081212997436525,
      "epoch": 2.7174488928874894,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004267387684738913,
      "loss": 5.4492,
      "mean_token_accuracy": 0.1945612147450447,
      "num_tokens": 53182957.0,
      "step": 27185
    },
    {
      "entropy": 6.003697109222412,
      "epoch": 2.717948717948718,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004267124147604991,
      "loss": 5.5762,
      "mean_token_accuracy": 0.18178656399250032,
      "num_tokens": 53193074.0,
      "step": 27190
    },
    {
      "entropy": 5.931169652938843,
      "epoch": 2.7184485430099468,
      "grad_norm": 1.125,
      "learning_rate": 0.0004266860572300615,
      "loss": 5.5328,
      "mean_token_accuracy": 0.19156585931777953,
      "num_tokens": 53202233.0,
      "step": 27195
    },
    {
      "entropy": 6.071871995925903,
      "epoch": 2.718948368071175,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004266596958832419,
      "loss": 5.6938,
      "mean_token_accuracy": 0.17733075171709062,
      "num_tokens": 53212332.0,
      "step": 27200
    },
    {
      "entropy": 5.9371696472167965,
      "epoch": 2.7194481931324037,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004266333307207032,
      "loss": 5.5324,
      "mean_token_accuracy": 0.19034706503152848,
      "num_tokens": 53222440.0,
      "step": 27205
    },
    {
      "entropy": 5.9033753871917725,
      "epoch": 2.719948018193632,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042660696174310904,
      "loss": 5.5043,
      "mean_token_accuracy": 0.1870041310787201,
      "num_tokens": 53232129.0,
      "step": 27210
    },
    {
      "entropy": 5.989480066299438,
      "epoch": 2.7204478432548607,
      "grad_norm": 1.125,
      "learning_rate": 0.0004265805889511227,
      "loss": 5.6196,
      "mean_token_accuracy": 0.1756633698940277,
      "num_tokens": 53241678.0,
      "step": 27215
    },
    {
      "entropy": 5.981545877456665,
      "epoch": 2.7209476683160894,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004265542123454077,
      "loss": 5.5713,
      "mean_token_accuracy": 0.18476658314466476,
      "num_tokens": 53250939.0,
      "step": 27220
    },
    {
      "entropy": 6.030445718765259,
      "epoch": 2.721447493377318,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042652783192662774,
      "loss": 5.6452,
      "mean_token_accuracy": 0.17630977034568787,
      "num_tokens": 53261058.0,
      "step": 27225
    },
    {
      "entropy": 5.957859849929809,
      "epoch": 2.7219473184385468,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004265014476954465,
      "loss": 5.5129,
      "mean_token_accuracy": 0.19268291741609572,
      "num_tokens": 53270934.0,
      "step": 27230
    },
    {
      "entropy": 5.920158338546753,
      "epoch": 2.722447143499775,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004264750596525279,
      "loss": 5.5455,
      "mean_token_accuracy": 0.18899033665657045,
      "num_tokens": 53280593.0,
      "step": 27235
    },
    {
      "entropy": 5.949115419387818,
      "epoch": 2.7229469685610037,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042644866779853577,
      "loss": 5.555,
      "mean_token_accuracy": 0.18824589997529984,
      "num_tokens": 53289381.0,
      "step": 27240
    },
    {
      "entropy": 6.019200611114502,
      "epoch": 2.723446793622232,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042642227213413405,
      "loss": 5.6958,
      "mean_token_accuracy": 0.18032012432813643,
      "num_tokens": 53298715.0,
      "step": 27245
    },
    {
      "entropy": 5.961607456207275,
      "epoch": 2.7239466186834607,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042639587265998696,
      "loss": 5.5332,
      "mean_token_accuracy": 0.18807339370250703,
      "num_tokens": 53309056.0,
      "step": 27250
    },
    {
      "entropy": 6.008716821670532,
      "epoch": 2.7244464437446894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004263694693767587,
      "loss": 5.5859,
      "mean_token_accuracy": 0.18163159042596816,
      "num_tokens": 53319028.0,
      "step": 27255
    },
    {
      "entropy": 6.004213094711304,
      "epoch": 2.724946268805918,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004263430622851135,
      "loss": 5.5717,
      "mean_token_accuracy": 0.17490442097187042,
      "num_tokens": 53329516.0,
      "step": 27260
    },
    {
      "entropy": 6.0382835388183596,
      "epoch": 2.7254460938671468,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004263166513857158,
      "loss": 5.5973,
      "mean_token_accuracy": 0.17582445591688156,
      "num_tokens": 53341163.0,
      "step": 27265
    },
    {
      "entropy": 5.934865665435791,
      "epoch": 2.725945918928375,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042629023667923016,
      "loss": 5.4899,
      "mean_token_accuracy": 0.19235507100820542,
      "num_tokens": 53350735.0,
      "step": 27270
    },
    {
      "entropy": 5.944275665283203,
      "epoch": 2.7264457439896037,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042626381816632096,
      "loss": 5.4762,
      "mean_token_accuracy": 0.19442903399467468,
      "num_tokens": 53361023.0,
      "step": 27275
    },
    {
      "entropy": 5.946090221405029,
      "epoch": 2.726945569050832,
      "grad_norm": 1.125,
      "learning_rate": 0.00042623739584765315,
      "loss": 5.5487,
      "mean_token_accuracy": 0.17960871756076813,
      "num_tokens": 53370859.0,
      "step": 27280
    },
    {
      "entropy": 5.959639167785644,
      "epoch": 2.7274453941120607,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042621096972389136,
      "loss": 5.4837,
      "mean_token_accuracy": 0.189203879237175,
      "num_tokens": 53380610.0,
      "step": 27285
    },
    {
      "entropy": 5.972375154495239,
      "epoch": 2.7279452191732894,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042618453979570035,
      "loss": 5.6771,
      "mean_token_accuracy": 0.17790041863918304,
      "num_tokens": 53392497.0,
      "step": 27290
    },
    {
      "entropy": 5.9237525939941404,
      "epoch": 2.728445044234518,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042615810606374535,
      "loss": 5.525,
      "mean_token_accuracy": 0.18849246203899384,
      "num_tokens": 53402234.0,
      "step": 27295
    },
    {
      "entropy": 5.921745729446411,
      "epoch": 2.7289448692957468,
      "grad_norm": 1.125,
      "learning_rate": 0.0004261316685286913,
      "loss": 5.578,
      "mean_token_accuracy": 0.18449158668518068,
      "num_tokens": 53412430.0,
      "step": 27300
    },
    {
      "entropy": 5.993644332885742,
      "epoch": 2.729444694356975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004261052271912033,
      "loss": 5.5757,
      "mean_token_accuracy": 0.18282590210437774,
      "num_tokens": 53422574.0,
      "step": 27305
    },
    {
      "entropy": 5.941882944107055,
      "epoch": 2.7299445194182037,
      "grad_norm": 2.0,
      "learning_rate": 0.0004260787820519467,
      "loss": 5.3747,
      "mean_token_accuracy": 0.21085282266139985,
      "num_tokens": 53431322.0,
      "step": 27310
    },
    {
      "entropy": 5.9471251487731935,
      "epoch": 2.730444344479432,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004260523331115868,
      "loss": 5.4079,
      "mean_token_accuracy": 0.19519371092319487,
      "num_tokens": 53440389.0,
      "step": 27315
    },
    {
      "entropy": 5.900234365463257,
      "epoch": 2.7309441695406607,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042602588037078903,
      "loss": 5.5779,
      "mean_token_accuracy": 0.18611588031053544,
      "num_tokens": 53449458.0,
      "step": 27320
    },
    {
      "entropy": 6.017315912246704,
      "epoch": 2.7314439946018894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042599942383021906,
      "loss": 5.5992,
      "mean_token_accuracy": 0.18195363581180574,
      "num_tokens": 53459075.0,
      "step": 27325
    },
    {
      "entropy": 6.057790327072143,
      "epoch": 2.731943819663118,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004259729634905424,
      "loss": 5.6109,
      "mean_token_accuracy": 0.18301188498735427,
      "num_tokens": 53468660.0,
      "step": 27330
    },
    {
      "entropy": 5.946406269073487,
      "epoch": 2.7324436447243468,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042594649935242493,
      "loss": 5.5388,
      "mean_token_accuracy": 0.18911102116107942,
      "num_tokens": 53477663.0,
      "step": 27335
    },
    {
      "entropy": 5.966468811035156,
      "epoch": 2.732943469785575,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004259200314165322,
      "loss": 5.5707,
      "mean_token_accuracy": 0.1835503935813904,
      "num_tokens": 53487226.0,
      "step": 27340
    },
    {
      "entropy": 5.947313356399536,
      "epoch": 2.7334432948468037,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004258935596835303,
      "loss": 5.4859,
      "mean_token_accuracy": 0.18738772124052047,
      "num_tokens": 53496483.0,
      "step": 27345
    },
    {
      "entropy": 5.9917075634002686,
      "epoch": 2.733943119908032,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042586708415408547,
      "loss": 5.6642,
      "mean_token_accuracy": 0.17066902667284012,
      "num_tokens": 53506042.0,
      "step": 27350
    },
    {
      "entropy": 5.896229887008667,
      "epoch": 2.7344429449692607,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004258406048288635,
      "loss": 5.471,
      "mean_token_accuracy": 0.19266043305397035,
      "num_tokens": 53516543.0,
      "step": 27355
    },
    {
      "entropy": 5.9921492576599125,
      "epoch": 2.7349427700304894,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042581412170853073,
      "loss": 5.6129,
      "mean_token_accuracy": 0.18763264566659926,
      "num_tokens": 53527432.0,
      "step": 27360
    },
    {
      "entropy": 5.983204460144043,
      "epoch": 2.735442595091718,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004257876347937535,
      "loss": 5.6229,
      "mean_token_accuracy": 0.18339701145887374,
      "num_tokens": 53535930.0,
      "step": 27365
    },
    {
      "entropy": 5.865609741210937,
      "epoch": 2.7359424201529463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004257611440851981,
      "loss": 5.4457,
      "mean_token_accuracy": 0.1969909816980362,
      "num_tokens": 53545123.0,
      "step": 27370
    },
    {
      "entropy": 5.9850568771362305,
      "epoch": 2.736442245214175,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042573464958353116,
      "loss": 5.5319,
      "mean_token_accuracy": 0.1879495084285736,
      "num_tokens": 53554892.0,
      "step": 27375
    },
    {
      "entropy": 5.932071781158447,
      "epoch": 2.7369420702754037,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004257081512894192,
      "loss": 5.4651,
      "mean_token_accuracy": 0.20015295296907426,
      "num_tokens": 53564858.0,
      "step": 27380
    },
    {
      "entropy": 5.925116491317749,
      "epoch": 2.737441895336632,
      "grad_norm": 1.125,
      "learning_rate": 0.00042568164920352895,
      "loss": 5.4173,
      "mean_token_accuracy": 0.20038523226976396,
      "num_tokens": 53573750.0,
      "step": 27385
    },
    {
      "entropy": 5.805664825439453,
      "epoch": 2.7379417203978607,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004256551433265273,
      "loss": 5.3728,
      "mean_token_accuracy": 0.21116613149642943,
      "num_tokens": 53584416.0,
      "step": 27390
    },
    {
      "entropy": 5.850586462020874,
      "epoch": 2.7384415454590894,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004256286336590808,
      "loss": 5.4206,
      "mean_token_accuracy": 0.19371370673179628,
      "num_tokens": 53593166.0,
      "step": 27395
    },
    {
      "entropy": 5.9256843566894535,
      "epoch": 2.738941370520318,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004256021202018568,
      "loss": 5.5877,
      "mean_token_accuracy": 0.18466860353946685,
      "num_tokens": 53603025.0,
      "step": 27400
    },
    {
      "entropy": 6.010952758789062,
      "epoch": 2.7394411955815463,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004255756029555222,
      "loss": 5.5297,
      "mean_token_accuracy": 0.18338567167520523,
      "num_tokens": 53613266.0,
      "step": 27405
    },
    {
      "entropy": 5.9191361427307125,
      "epoch": 2.739941020642775,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042554908192074406,
      "loss": 5.4918,
      "mean_token_accuracy": 0.19574539363384247,
      "num_tokens": 53622271.0,
      "step": 27410
    },
    {
      "entropy": 5.977170133590699,
      "epoch": 2.7404408457040037,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042552255709818975,
      "loss": 5.6012,
      "mean_token_accuracy": 0.18020713329315186,
      "num_tokens": 53632096.0,
      "step": 27415
    },
    {
      "entropy": 5.964522647857666,
      "epoch": 2.740940670765232,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042549602848852673,
      "loss": 5.5462,
      "mean_token_accuracy": 0.19010716378688813,
      "num_tokens": 53640875.0,
      "step": 27420
    },
    {
      "entropy": 5.918202304840088,
      "epoch": 2.7414404958264607,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004254694960924224,
      "loss": 5.4794,
      "mean_token_accuracy": 0.19518865644931793,
      "num_tokens": 53650186.0,
      "step": 27425
    },
    {
      "entropy": 5.922190713882446,
      "epoch": 2.7419403208876894,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004254429599105442,
      "loss": 5.4833,
      "mean_token_accuracy": 0.19316719621419906,
      "num_tokens": 53659608.0,
      "step": 27430
    },
    {
      "entropy": 6.0070311546325685,
      "epoch": 2.742440145948918,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004254164199435598,
      "loss": 5.532,
      "mean_token_accuracy": 0.18266330063343048,
      "num_tokens": 53668479.0,
      "step": 27435
    },
    {
      "entropy": 5.947125196456909,
      "epoch": 2.7429399710101463,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004253898761921371,
      "loss": 5.5274,
      "mean_token_accuracy": 0.18717991858720778,
      "num_tokens": 53677496.0,
      "step": 27440
    },
    {
      "entropy": 5.9957600116729735,
      "epoch": 2.743439796071375,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004253633286569437,
      "loss": 5.5868,
      "mean_token_accuracy": 0.1833234503865242,
      "num_tokens": 53686865.0,
      "step": 27445
    },
    {
      "entropy": 6.010181283950805,
      "epoch": 2.7439396211326037,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004253367773386476,
      "loss": 5.5879,
      "mean_token_accuracy": 0.18544304966926575,
      "num_tokens": 53696159.0,
      "step": 27450
    },
    {
      "entropy": 6.013127994537354,
      "epoch": 2.744439446193832,
      "grad_norm": 1.171875,
      "learning_rate": 0.000425310222237917,
      "loss": 5.5416,
      "mean_token_accuracy": 0.17433348298072815,
      "num_tokens": 53705512.0,
      "step": 27455
    },
    {
      "entropy": 5.955356073379517,
      "epoch": 2.7449392712550607,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042528366335541984,
      "loss": 5.5729,
      "mean_token_accuracy": 0.18000805675983428,
      "num_tokens": 53714361.0,
      "step": 27460
    },
    {
      "entropy": 5.958677816390991,
      "epoch": 2.7454390963162894,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004252571006918243,
      "loss": 5.554,
      "mean_token_accuracy": 0.18741171360015868,
      "num_tokens": 53723606.0,
      "step": 27465
    },
    {
      "entropy": 5.966777563095093,
      "epoch": 2.745938921377518,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004252305342477989,
      "loss": 5.5159,
      "mean_token_accuracy": 0.18730472177267074,
      "num_tokens": 53734041.0,
      "step": 27470
    },
    {
      "entropy": 5.964026880264282,
      "epoch": 2.7464387464387463,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042520396402401185,
      "loss": 5.4974,
      "mean_token_accuracy": 0.19026966840028764,
      "num_tokens": 53743735.0,
      "step": 27475
    },
    {
      "entropy": 5.960513591766357,
      "epoch": 2.746938571499975,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004251773900211317,
      "loss": 5.5755,
      "mean_token_accuracy": 0.1823553591966629,
      "num_tokens": 53753960.0,
      "step": 27480
    },
    {
      "entropy": 5.982395076751709,
      "epoch": 2.7474383965612037,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042515081223982716,
      "loss": 5.5867,
      "mean_token_accuracy": 0.18760258406400682,
      "num_tokens": 53764874.0,
      "step": 27485
    },
    {
      "entropy": 6.016818571090698,
      "epoch": 2.747938221622432,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042512423068076673,
      "loss": 5.6138,
      "mean_token_accuracy": 0.18408676832914353,
      "num_tokens": 53775153.0,
      "step": 27490
    },
    {
      "entropy": 5.955229711532593,
      "epoch": 2.7484380466836607,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004250976453446193,
      "loss": 5.4547,
      "mean_token_accuracy": 0.19327027052640916,
      "num_tokens": 53785091.0,
      "step": 27495
    },
    {
      "entropy": 5.934490823745728,
      "epoch": 2.7489378717448893,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004250710562320538,
      "loss": 5.4452,
      "mean_token_accuracy": 0.19286636114120484,
      "num_tokens": 53794154.0,
      "step": 27500
    },
    {
      "entropy": 5.912927150726318,
      "epoch": 2.749437696806118,
      "grad_norm": 1.125,
      "learning_rate": 0.00042504446334373905,
      "loss": 5.4908,
      "mean_token_accuracy": 0.18545408844947814,
      "num_tokens": 53804129.0,
      "step": 27505
    },
    {
      "entropy": 5.959204959869385,
      "epoch": 2.7499375218673463,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004250178666803442,
      "loss": 5.616,
      "mean_token_accuracy": 0.18254123777151107,
      "num_tokens": 53813081.0,
      "step": 27510
    },
    {
      "entropy": 5.990995645523071,
      "epoch": 2.750437346928575,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004249912662425385,
      "loss": 5.5098,
      "mean_token_accuracy": 0.18694372922182084,
      "num_tokens": 53821949.0,
      "step": 27515
    },
    {
      "entropy": 6.003176498413086,
      "epoch": 2.7509371719898037,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042496466203099107,
      "loss": 5.545,
      "mean_token_accuracy": 0.18296138793230057,
      "num_tokens": 53831232.0,
      "step": 27520
    },
    {
      "entropy": 5.959999752044678,
      "epoch": 2.751436997051032,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042493805404637136,
      "loss": 5.5396,
      "mean_token_accuracy": 0.1829437404870987,
      "num_tokens": 53841068.0,
      "step": 27525
    },
    {
      "entropy": 5.851357793807983,
      "epoch": 2.7519368221122606,
      "grad_norm": 1.125,
      "learning_rate": 0.00042491144228934885,
      "loss": 5.4217,
      "mean_token_accuracy": 0.19591573476791382,
      "num_tokens": 53849874.0,
      "step": 27530
    },
    {
      "entropy": 5.9077941417694095,
      "epoch": 2.7524366471734893,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000424884826760593,
      "loss": 5.4237,
      "mean_token_accuracy": 0.19474369585514067,
      "num_tokens": 53859173.0,
      "step": 27535
    },
    {
      "entropy": 5.976163339614868,
      "epoch": 2.752936472234718,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042485820746077346,
      "loss": 5.5196,
      "mean_token_accuracy": 0.18791692107915878,
      "num_tokens": 53869000.0,
      "step": 27540
    },
    {
      "entropy": 6.054228830337524,
      "epoch": 2.7534362972959463,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042483158439055995,
      "loss": 5.63,
      "mean_token_accuracy": 0.17683846950531007,
      "num_tokens": 53879941.0,
      "step": 27545
    },
    {
      "entropy": 5.883149909973144,
      "epoch": 2.753936122357175,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004248049575506224,
      "loss": 5.4631,
      "mean_token_accuracy": 0.19062309712171555,
      "num_tokens": 53889480.0,
      "step": 27550
    },
    {
      "entropy": 5.9472283840179445,
      "epoch": 2.7544359474184037,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004247783269416306,
      "loss": 5.528,
      "mean_token_accuracy": 0.1917477250099182,
      "num_tokens": 53898554.0,
      "step": 27555
    },
    {
      "entropy": 5.904239749908447,
      "epoch": 2.754935772479632,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042475169256425466,
      "loss": 5.4829,
      "mean_token_accuracy": 0.1864876627922058,
      "num_tokens": 53907898.0,
      "step": 27560
    },
    {
      "entropy": 5.967459487915039,
      "epoch": 2.7554355975408606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004247250544191647,
      "loss": 5.5986,
      "mean_token_accuracy": 0.17569601088762282,
      "num_tokens": 53917948.0,
      "step": 27565
    },
    {
      "entropy": 5.936914348602295,
      "epoch": 2.7559354226020893,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042469841250703086,
      "loss": 5.5332,
      "mean_token_accuracy": 0.1896333545446396,
      "num_tokens": 53927409.0,
      "step": 27570
    },
    {
      "entropy": 5.96373143196106,
      "epoch": 2.756435247663318,
      "grad_norm": 1.125,
      "learning_rate": 0.0004246717668285235,
      "loss": 5.5658,
      "mean_token_accuracy": 0.18321035355329512,
      "num_tokens": 53936600.0,
      "step": 27575
    },
    {
      "entropy": 5.977739667892456,
      "epoch": 2.7569350727245463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042464511738431296,
      "loss": 5.5269,
      "mean_token_accuracy": 0.19327752888202668,
      "num_tokens": 53946247.0,
      "step": 27580
    },
    {
      "entropy": 5.876918983459473,
      "epoch": 2.757434897785775,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042461846417506983,
      "loss": 5.3448,
      "mean_token_accuracy": 0.1986766576766968,
      "num_tokens": 53956009.0,
      "step": 27585
    },
    {
      "entropy": 5.982678699493408,
      "epoch": 2.7579347228470037,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004245918072014646,
      "loss": 5.5525,
      "mean_token_accuracy": 0.18282256573438643,
      "num_tokens": 53965549.0,
      "step": 27590
    },
    {
      "entropy": 5.9460532665252686,
      "epoch": 2.758434547908232,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000424565146464168,
      "loss": 5.5694,
      "mean_token_accuracy": 0.1816099092364311,
      "num_tokens": 53975901.0,
      "step": 27595
    },
    {
      "entropy": 6.047082090377808,
      "epoch": 2.7589343729694606,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042453848196385085,
      "loss": 5.6065,
      "mean_token_accuracy": 0.1806547924876213,
      "num_tokens": 53985334.0,
      "step": 27600
    },
    {
      "entropy": 5.85233736038208,
      "epoch": 2.7594341980306893,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042451181370118394,
      "loss": 5.4186,
      "mean_token_accuracy": 0.2041640505194664,
      "num_tokens": 53995445.0,
      "step": 27605
    },
    {
      "entropy": 5.911671543121338,
      "epoch": 2.759934023091918,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004244851416768383,
      "loss": 5.519,
      "mean_token_accuracy": 0.19168734848499297,
      "num_tokens": 54004845.0,
      "step": 27610
    },
    {
      "entropy": 5.974104642868042,
      "epoch": 2.7604338481531463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004244584658914849,
      "loss": 5.5157,
      "mean_token_accuracy": 0.19377415031194686,
      "num_tokens": 54015074.0,
      "step": 27615
    },
    {
      "entropy": 5.951715421676636,
      "epoch": 2.760933673214375,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042443178634579497,
      "loss": 5.5439,
      "mean_token_accuracy": 0.18184617012739182,
      "num_tokens": 54025022.0,
      "step": 27620
    },
    {
      "entropy": 5.920310068130493,
      "epoch": 2.7614334982756037,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004244051030404398,
      "loss": 5.5573,
      "mean_token_accuracy": 0.183640693128109,
      "num_tokens": 54034901.0,
      "step": 27625
    },
    {
      "entropy": 5.956652450561523,
      "epoch": 2.761933323336832,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042437841597609054,
      "loss": 5.5439,
      "mean_token_accuracy": 0.1823224440217018,
      "num_tokens": 54044032.0,
      "step": 27630
    },
    {
      "entropy": 5.958889961242676,
      "epoch": 2.7624331483980606,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004243517251534189,
      "loss": 5.5318,
      "mean_token_accuracy": 0.19185253083705903,
      "num_tokens": 54054635.0,
      "step": 27635
    },
    {
      "entropy": 5.961057519912719,
      "epoch": 2.7629329734592893,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004243250305730963,
      "loss": 5.4867,
      "mean_token_accuracy": 0.1974022164940834,
      "num_tokens": 54064658.0,
      "step": 27640
    },
    {
      "entropy": 5.963742446899414,
      "epoch": 2.763432798520518,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042429833223579436,
      "loss": 5.5976,
      "mean_token_accuracy": 0.18417628705501557,
      "num_tokens": 54074788.0,
      "step": 27645
    },
    {
      "entropy": 6.01288104057312,
      "epoch": 2.7639326235817463,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042427163014218473,
      "loss": 5.5233,
      "mean_token_accuracy": 0.1868588447570801,
      "num_tokens": 54084468.0,
      "step": 27650
    },
    {
      "entropy": 5.991619062423706,
      "epoch": 2.764432448642975,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042424492429293933,
      "loss": 5.5604,
      "mean_token_accuracy": 0.19087306559085845,
      "num_tokens": 54094223.0,
      "step": 27655
    },
    {
      "entropy": 5.937386894226075,
      "epoch": 2.7649322737042037,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042421821468873003,
      "loss": 5.4809,
      "mean_token_accuracy": 0.18202575147151948,
      "num_tokens": 54102721.0,
      "step": 27660
    },
    {
      "entropy": 5.926418924331665,
      "epoch": 2.765432098765432,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042419150133022883,
      "loss": 5.5929,
      "mean_token_accuracy": 0.17888176143169404,
      "num_tokens": 54111986.0,
      "step": 27665
    },
    {
      "entropy": 5.914083194732666,
      "epoch": 2.7659319238266606,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004241647842181078,
      "loss": 5.487,
      "mean_token_accuracy": 0.18706816136837007,
      "num_tokens": 54121074.0,
      "step": 27670
    },
    {
      "entropy": 5.966044616699219,
      "epoch": 2.7664317488878893,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042413806335303935,
      "loss": 5.5422,
      "mean_token_accuracy": 0.18380776047706604,
      "num_tokens": 54131518.0,
      "step": 27675
    },
    {
      "entropy": 5.968699645996094,
      "epoch": 2.766931573949118,
      "grad_norm": 1.125,
      "learning_rate": 0.00042411133873569544,
      "loss": 5.5973,
      "mean_token_accuracy": 0.18384103924036027,
      "num_tokens": 54141228.0,
      "step": 27680
    },
    {
      "entropy": 5.861841344833374,
      "epoch": 2.7674313990103463,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004240846103667487,
      "loss": 5.4581,
      "mean_token_accuracy": 0.191938379406929,
      "num_tokens": 54150365.0,
      "step": 27685
    },
    {
      "entropy": 5.9753964900970455,
      "epoch": 2.767931224071575,
      "grad_norm": 1.125,
      "learning_rate": 0.00042405787824687153,
      "loss": 5.5467,
      "mean_token_accuracy": 0.1829365611076355,
      "num_tokens": 54159941.0,
      "step": 27690
    },
    {
      "entropy": 5.976675748825073,
      "epoch": 2.7684310491328032,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004240311423767364,
      "loss": 5.5457,
      "mean_token_accuracy": 0.18464966863393784,
      "num_tokens": 54169613.0,
      "step": 27695
    },
    {
      "entropy": 5.954810953140258,
      "epoch": 2.768930874194032,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004240044027570162,
      "loss": 5.5488,
      "mean_token_accuracy": 0.18665747940540314,
      "num_tokens": 54179927.0,
      "step": 27700
    },
    {
      "entropy": 5.925431871414185,
      "epoch": 2.7694306992552606,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004239776593883835,
      "loss": 5.475,
      "mean_token_accuracy": 0.19160089045763015,
      "num_tokens": 54189572.0,
      "step": 27705
    },
    {
      "entropy": 5.973875522613525,
      "epoch": 2.7699305243164893,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042395091227151124,
      "loss": 5.5799,
      "mean_token_accuracy": 0.1824919566512108,
      "num_tokens": 54200561.0,
      "step": 27710
    },
    {
      "entropy": 6.045355939865113,
      "epoch": 2.770430349377718,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042392416140707233,
      "loss": 5.6508,
      "mean_token_accuracy": 0.18646831959486007,
      "num_tokens": 54210519.0,
      "step": 27715
    },
    {
      "entropy": 5.894741439819336,
      "epoch": 2.7709301744389463,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004238974067957399,
      "loss": 5.3912,
      "mean_token_accuracy": 0.19754815250635147,
      "num_tokens": 54219142.0,
      "step": 27720
    },
    {
      "entropy": 5.947816514968872,
      "epoch": 2.771429999500175,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000423870648438187,
      "loss": 5.5186,
      "mean_token_accuracy": 0.18109541684389113,
      "num_tokens": 54228728.0,
      "step": 27725
    },
    {
      "entropy": 6.001570558547973,
      "epoch": 2.7719298245614032,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004238438863350869,
      "loss": 5.5181,
      "mean_token_accuracy": 0.18609738945961,
      "num_tokens": 54237767.0,
      "step": 27730
    },
    {
      "entropy": 5.885164070129394,
      "epoch": 2.772429649622632,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004238171204871129,
      "loss": 5.4838,
      "mean_token_accuracy": 0.18991725146770477,
      "num_tokens": 54247453.0,
      "step": 27735
    },
    {
      "entropy": 5.96748719215393,
      "epoch": 2.7729294746838606,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042379035089493844,
      "loss": 5.5862,
      "mean_token_accuracy": 0.18683990836143494,
      "num_tokens": 54256398.0,
      "step": 27740
    },
    {
      "entropy": 5.914302206039428,
      "epoch": 2.7734292997450893,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004237635775592371,
      "loss": 5.5924,
      "mean_token_accuracy": 0.1868361711502075,
      "num_tokens": 54265981.0,
      "step": 27745
    },
    {
      "entropy": 5.86876859664917,
      "epoch": 2.773929124806318,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004237368004806823,
      "loss": 5.4741,
      "mean_token_accuracy": 0.19469580054283142,
      "num_tokens": 54277285.0,
      "step": 27750
    },
    {
      "entropy": 5.934812402725219,
      "epoch": 2.7744289498675463,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000423710019659948,
      "loss": 5.451,
      "mean_token_accuracy": 0.1908887192606926,
      "num_tokens": 54286211.0,
      "step": 27755
    },
    {
      "entropy": 5.946382284164429,
      "epoch": 2.774928774928775,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042368323509770785,
      "loss": 5.5353,
      "mean_token_accuracy": 0.1904023766517639,
      "num_tokens": 54296101.0,
      "step": 27760
    },
    {
      "entropy": 5.934570264816284,
      "epoch": 2.7754285999900032,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004236564467946357,
      "loss": 5.5309,
      "mean_token_accuracy": 0.19257076531648637,
      "num_tokens": 54306027.0,
      "step": 27765
    },
    {
      "entropy": 5.92736783027649,
      "epoch": 2.775928425051232,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004236296547514056,
      "loss": 5.4582,
      "mean_token_accuracy": 0.1936216801404953,
      "num_tokens": 54315551.0,
      "step": 27770
    },
    {
      "entropy": 5.99992561340332,
      "epoch": 2.7764282501124606,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004236028589686917,
      "loss": 5.601,
      "mean_token_accuracy": 0.18309661000967026,
      "num_tokens": 54325312.0,
      "step": 27775
    },
    {
      "entropy": 5.893679094314575,
      "epoch": 2.7769280751736893,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042357605944716816,
      "loss": 5.5662,
      "mean_token_accuracy": 0.1833363354206085,
      "num_tokens": 54335854.0,
      "step": 27780
    },
    {
      "entropy": 5.9820562362670895,
      "epoch": 2.777427900234918,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042354925618750904,
      "loss": 5.4748,
      "mean_token_accuracy": 0.18471334874629974,
      "num_tokens": 54345887.0,
      "step": 27785
    },
    {
      "entropy": 5.984033632278442,
      "epoch": 2.7779277252961463,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042352244919038895,
      "loss": 5.5512,
      "mean_token_accuracy": 0.18957429677248,
      "num_tokens": 54355367.0,
      "step": 27790
    },
    {
      "entropy": 5.918541383743286,
      "epoch": 2.778427550357375,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042349563845648226,
      "loss": 5.5211,
      "mean_token_accuracy": 0.1886720612645149,
      "num_tokens": 54365689.0,
      "step": 27795
    },
    {
      "entropy": 5.956513595581055,
      "epoch": 2.7789273754186032,
      "grad_norm": 1.125,
      "learning_rate": 0.0004234688239864634,
      "loss": 5.5998,
      "mean_token_accuracy": 0.18059480786323548,
      "num_tokens": 54375317.0,
      "step": 27800
    },
    {
      "entropy": 5.930055475234985,
      "epoch": 2.779427200479832,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004234420057810073,
      "loss": 5.5157,
      "mean_token_accuracy": 0.1863020360469818,
      "num_tokens": 54384926.0,
      "step": 27805
    },
    {
      "entropy": 6.006679916381836,
      "epoch": 2.7799270255410606,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042341518384078847,
      "loss": 5.6036,
      "mean_token_accuracy": 0.17986817210912703,
      "num_tokens": 54394691.0,
      "step": 27810
    },
    {
      "entropy": 5.997537231445312,
      "epoch": 2.7804268506022893,
      "grad_norm": 1.125,
      "learning_rate": 0.0004233883581664818,
      "loss": 5.6086,
      "mean_token_accuracy": 0.17971620261669158,
      "num_tokens": 54405264.0,
      "step": 27815
    },
    {
      "entropy": 5.97053484916687,
      "epoch": 2.780926675663518,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004233615287587622,
      "loss": 5.5494,
      "mean_token_accuracy": 0.18584130853414535,
      "num_tokens": 54415477.0,
      "step": 27820
    },
    {
      "entropy": 6.015782403945923,
      "epoch": 2.7814265007247463,
      "grad_norm": 1.125,
      "learning_rate": 0.00042333469561830475,
      "loss": 5.6486,
      "mean_token_accuracy": 0.1790587767958641,
      "num_tokens": 54425715.0,
      "step": 27825
    },
    {
      "entropy": 6.05889310836792,
      "epoch": 2.781926325785975,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004233078587457845,
      "loss": 5.5856,
      "mean_token_accuracy": 0.17984528094530106,
      "num_tokens": 54435097.0,
      "step": 27830
    },
    {
      "entropy": 5.9562725067138675,
      "epoch": 2.7824261508472032,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004232810181418767,
      "loss": 5.6116,
      "mean_token_accuracy": 0.18071163445711136,
      "num_tokens": 54445511.0,
      "step": 27835
    },
    {
      "entropy": 5.906667041778564,
      "epoch": 2.782925975908432,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004232541738072566,
      "loss": 5.5271,
      "mean_token_accuracy": 0.18660216331481932,
      "num_tokens": 54456463.0,
      "step": 27840
    },
    {
      "entropy": 5.823899030685425,
      "epoch": 2.7834258009696606,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042322732574259976,
      "loss": 5.3868,
      "mean_token_accuracy": 0.2034952387213707,
      "num_tokens": 54466262.0,
      "step": 27845
    },
    {
      "entropy": 5.991084289550781,
      "epoch": 2.7839256260308893,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042320047394858143,
      "loss": 5.5671,
      "mean_token_accuracy": 0.18196186870336534,
      "num_tokens": 54476270.0,
      "step": 27850
    },
    {
      "entropy": 5.8973464012146,
      "epoch": 2.784425451092118,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004231736184258773,
      "loss": 5.5586,
      "mean_token_accuracy": 0.18525839895009993,
      "num_tokens": 54486803.0,
      "step": 27855
    },
    {
      "entropy": 6.056579542160034,
      "epoch": 2.7849252761533463,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004231467591751631,
      "loss": 5.5835,
      "mean_token_accuracy": 0.1799938902258873,
      "num_tokens": 54495525.0,
      "step": 27860
    },
    {
      "entropy": 5.891634607315064,
      "epoch": 2.785425101214575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004231198961971145,
      "loss": 5.4359,
      "mean_token_accuracy": 0.192692768573761,
      "num_tokens": 54506335.0,
      "step": 27865
    },
    {
      "entropy": 5.9036070823669435,
      "epoch": 2.7859249262758032,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004230930294924075,
      "loss": 5.5337,
      "mean_token_accuracy": 0.19298264682292937,
      "num_tokens": 54516505.0,
      "step": 27870
    },
    {
      "entropy": 5.995357179641724,
      "epoch": 2.786424751337032,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042306615906171794,
      "loss": 5.5759,
      "mean_token_accuracy": 0.1841145172715187,
      "num_tokens": 54525407.0,
      "step": 27875
    },
    {
      "entropy": 5.917439413070679,
      "epoch": 2.7869245763982606,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000423039284905722,
      "loss": 5.5016,
      "mean_token_accuracy": 0.19094815999269485,
      "num_tokens": 54535750.0,
      "step": 27880
    },
    {
      "entropy": 5.9782970428466795,
      "epoch": 2.7874244014594893,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042301240702509576,
      "loss": 5.5215,
      "mean_token_accuracy": 0.18681561052799225,
      "num_tokens": 54544284.0,
      "step": 27885
    },
    {
      "entropy": 5.854727506637573,
      "epoch": 2.7879242265207176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004229855254205154,
      "loss": 5.4105,
      "mean_token_accuracy": 0.19969730228185653,
      "num_tokens": 54553168.0,
      "step": 27890
    },
    {
      "entropy": 5.96324200630188,
      "epoch": 2.7884240515819463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042295864009265723,
      "loss": 5.499,
      "mean_token_accuracy": 0.18774142861366272,
      "num_tokens": 54563553.0,
      "step": 27895
    },
    {
      "entropy": 5.9258088111877445,
      "epoch": 2.788923876643175,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042293175104219776,
      "loss": 5.4851,
      "mean_token_accuracy": 0.19114592671394348,
      "num_tokens": 54573195.0,
      "step": 27900
    },
    {
      "entropy": 5.8161070346832275,
      "epoch": 2.7894237017044032,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042290485826981355,
      "loss": 5.4166,
      "mean_token_accuracy": 0.20053781270980836,
      "num_tokens": 54583417.0,
      "step": 27905
    },
    {
      "entropy": 5.981676197052002,
      "epoch": 2.789923526765632,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042287796177618115,
      "loss": 5.5812,
      "mean_token_accuracy": 0.18112883418798448,
      "num_tokens": 54592961.0,
      "step": 27910
    },
    {
      "entropy": 5.951387786865235,
      "epoch": 2.7904233518268606,
      "grad_norm": 1.125,
      "learning_rate": 0.00042285106156197727,
      "loss": 5.5123,
      "mean_token_accuracy": 0.1824626550078392,
      "num_tokens": 54602161.0,
      "step": 27915
    },
    {
      "entropy": 6.018195390701294,
      "epoch": 2.7909231768880893,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004228241576278787,
      "loss": 5.6389,
      "mean_token_accuracy": 0.1813039869070053,
      "num_tokens": 54612816.0,
      "step": 27920
    },
    {
      "entropy": 5.946853923797607,
      "epoch": 2.7914230019493176,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042279724997456246,
      "loss": 5.4928,
      "mean_token_accuracy": 0.18978315591812134,
      "num_tokens": 54622808.0,
      "step": 27925
    },
    {
      "entropy": 5.964504623413086,
      "epoch": 2.7919228270105463,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042277033860270544,
      "loss": 5.5141,
      "mean_token_accuracy": 0.19354663640260697,
      "num_tokens": 54633316.0,
      "step": 27930
    },
    {
      "entropy": 5.879451894760132,
      "epoch": 2.792422652071775,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042274342351298467,
      "loss": 5.4867,
      "mean_token_accuracy": 0.19268444180488586,
      "num_tokens": 54642655.0,
      "step": 27935
    },
    {
      "entropy": 5.8756194591522215,
      "epoch": 2.7929224771330032,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004227165047060773,
      "loss": 5.5003,
      "mean_token_accuracy": 0.191695699095726,
      "num_tokens": 54652390.0,
      "step": 27940
    },
    {
      "entropy": 5.977358150482178,
      "epoch": 2.793422302194232,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004226895821826608,
      "loss": 5.5848,
      "mean_token_accuracy": 0.1778972029685974,
      "num_tokens": 54661525.0,
      "step": 27945
    },
    {
      "entropy": 6.010542488098144,
      "epoch": 2.7939221272554606,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004226626559434124,
      "loss": 5.5101,
      "mean_token_accuracy": 0.1884256973862648,
      "num_tokens": 54670416.0,
      "step": 27950
    },
    {
      "entropy": 5.882136297225952,
      "epoch": 2.7944219523166893,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042263572598900956,
      "loss": 5.5083,
      "mean_token_accuracy": 0.1844273492693901,
      "num_tokens": 54680025.0,
      "step": 27955
    },
    {
      "entropy": 5.9118266105651855,
      "epoch": 2.7949217773779176,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042260879232012985,
      "loss": 5.5601,
      "mean_token_accuracy": 0.1802893951535225,
      "num_tokens": 54689591.0,
      "step": 27960
    },
    {
      "entropy": 5.95735502243042,
      "epoch": 2.7954216024391463,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042258185493745086,
      "loss": 5.525,
      "mean_token_accuracy": 0.18675436824560165,
      "num_tokens": 54699521.0,
      "step": 27965
    },
    {
      "entropy": 5.928467655181885,
      "epoch": 2.795921427500375,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004225549138416504,
      "loss": 5.4178,
      "mean_token_accuracy": 0.19433500319719316,
      "num_tokens": 54710159.0,
      "step": 27970
    },
    {
      "entropy": 5.960637331008911,
      "epoch": 2.7964212525616032,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004225279690334063,
      "loss": 5.4964,
      "mean_token_accuracy": 0.19435327202081681,
      "num_tokens": 54719657.0,
      "step": 27975
    },
    {
      "entropy": 6.01815357208252,
      "epoch": 2.796921077622832,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042250102051339647,
      "loss": 5.6299,
      "mean_token_accuracy": 0.1838567152619362,
      "num_tokens": 54730681.0,
      "step": 27980
    },
    {
      "entropy": 6.004740047454834,
      "epoch": 2.7974209026840606,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042247406828229897,
      "loss": 5.5973,
      "mean_token_accuracy": 0.17795604169368745,
      "num_tokens": 54740606.0,
      "step": 27985
    },
    {
      "entropy": 5.829167747497559,
      "epoch": 2.7979207277452893,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004224471123407917,
      "loss": 5.3291,
      "mean_token_accuracy": 0.2097541242837906,
      "num_tokens": 54750233.0,
      "step": 27990
    },
    {
      "entropy": 6.0070923328399655,
      "epoch": 2.7984205528065176,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004224201526895531,
      "loss": 5.7118,
      "mean_token_accuracy": 0.17275363504886626,
      "num_tokens": 54760287.0,
      "step": 27995
    },
    {
      "entropy": 6.020049619674682,
      "epoch": 2.7989203778677463,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042239318932926144,
      "loss": 5.4979,
      "mean_token_accuracy": 0.1867847204208374,
      "num_tokens": 54770287.0,
      "step": 28000
    },
    {
      "entropy": 5.9400139331817625,
      "epoch": 2.799420202928975,
      "grad_norm": 1.140625,
      "learning_rate": 0.000422366222260595,
      "loss": 5.5701,
      "mean_token_accuracy": 0.18340066224336624,
      "num_tokens": 54779780.0,
      "step": 28005
    },
    {
      "entropy": 5.895179796218872,
      "epoch": 2.7999200279902032,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042233925148423235,
      "loss": 5.4489,
      "mean_token_accuracy": 0.19588758498430253,
      "num_tokens": 54789628.0,
      "step": 28010
    },
    {
      "entropy": 5.850614976882935,
      "epoch": 2.800419853051432,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004223122770008521,
      "loss": 5.4571,
      "mean_token_accuracy": 0.19770427346229552,
      "num_tokens": 54799228.0,
      "step": 28015
    },
    {
      "entropy": 5.933167505264282,
      "epoch": 2.8009196781126606,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042228529881113285,
      "loss": 5.4999,
      "mean_token_accuracy": 0.18749878704547882,
      "num_tokens": 54809666.0,
      "step": 28020
    },
    {
      "entropy": 5.9589948654174805,
      "epoch": 2.8014195031738893,
      "grad_norm": 1.125,
      "learning_rate": 0.00042225831691575336,
      "loss": 5.4663,
      "mean_token_accuracy": 0.1910161018371582,
      "num_tokens": 54818948.0,
      "step": 28025
    },
    {
      "entropy": 5.933778762817383,
      "epoch": 2.8019193282351176,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004222313313153925,
      "loss": 5.5643,
      "mean_token_accuracy": 0.18619430512189866,
      "num_tokens": 54830091.0,
      "step": 28030
    },
    {
      "entropy": 5.926619005203247,
      "epoch": 2.8024191532963463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042220434201072917,
      "loss": 5.5079,
      "mean_token_accuracy": 0.1903132304549217,
      "num_tokens": 54839385.0,
      "step": 28035
    },
    {
      "entropy": 5.995261144638062,
      "epoch": 2.802918978357575,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004221773490024426,
      "loss": 5.5565,
      "mean_token_accuracy": 0.1862626016139984,
      "num_tokens": 54850034.0,
      "step": 28040
    },
    {
      "entropy": 6.0203691005706785,
      "epoch": 2.8034188034188032,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004221503522912117,
      "loss": 5.6471,
      "mean_token_accuracy": 0.1784633383154869,
      "num_tokens": 54859522.0,
      "step": 28045
    },
    {
      "entropy": 5.9057153224945065,
      "epoch": 2.803918628480032,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042212335187771587,
      "loss": 5.4827,
      "mean_token_accuracy": 0.19233769923448563,
      "num_tokens": 54868938.0,
      "step": 28050
    },
    {
      "entropy": 5.898127317428589,
      "epoch": 2.8044184535412606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042209634776263434,
      "loss": 5.391,
      "mean_token_accuracy": 0.1956111580133438,
      "num_tokens": 54878125.0,
      "step": 28055
    },
    {
      "entropy": 5.990341281890869,
      "epoch": 2.8049182786024893,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004220693399466466,
      "loss": 5.5795,
      "mean_token_accuracy": 0.1812012553215027,
      "num_tokens": 54886615.0,
      "step": 28060
    },
    {
      "entropy": 5.988999223709106,
      "epoch": 2.8054181036637176,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042204232843043207,
      "loss": 5.6279,
      "mean_token_accuracy": 0.18190946578979492,
      "num_tokens": 54897316.0,
      "step": 28065
    },
    {
      "entropy": 5.987453460693359,
      "epoch": 2.8059179287249463,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042201531321467027,
      "loss": 5.4843,
      "mean_token_accuracy": 0.1892397940158844,
      "num_tokens": 54906672.0,
      "step": 28070
    },
    {
      "entropy": 6.008827161788941,
      "epoch": 2.806417753786175,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042198829430004114,
      "loss": 5.5702,
      "mean_token_accuracy": 0.17572498172521592,
      "num_tokens": 54916465.0,
      "step": 28075
    },
    {
      "entropy": 5.880013799667358,
      "epoch": 2.8069175788474032,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042196127168722436,
      "loss": 5.4862,
      "mean_token_accuracy": 0.18764934539794922,
      "num_tokens": 54926655.0,
      "step": 28080
    },
    {
      "entropy": 5.9582091808319095,
      "epoch": 2.807417403908632,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004219342453768998,
      "loss": 5.4996,
      "mean_token_accuracy": 0.18584200441837312,
      "num_tokens": 54936151.0,
      "step": 28085
    },
    {
      "entropy": 5.955983734130859,
      "epoch": 2.8079172289698606,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004219072153697474,
      "loss": 5.5008,
      "mean_token_accuracy": 0.1898718759417534,
      "num_tokens": 54946303.0,
      "step": 28090
    },
    {
      "entropy": 5.905334329605102,
      "epoch": 2.8084170540310893,
      "grad_norm": 1.125,
      "learning_rate": 0.0004218801816664473,
      "loss": 5.5405,
      "mean_token_accuracy": 0.18495769500732423,
      "num_tokens": 54956078.0,
      "step": 28095
    },
    {
      "entropy": 5.884579133987427,
      "epoch": 2.8089168790923176,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004218531442676796,
      "loss": 5.556,
      "mean_token_accuracy": 0.18757327646017075,
      "num_tokens": 54965851.0,
      "step": 28100
    },
    {
      "entropy": 6.028243827819824,
      "epoch": 2.8094167041535463,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004218261031741246,
      "loss": 5.612,
      "mean_token_accuracy": 0.17707803398370742,
      "num_tokens": 54975381.0,
      "step": 28105
    },
    {
      "entropy": 5.985556888580322,
      "epoch": 2.809916529214775,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042179905838646254,
      "loss": 5.4837,
      "mean_token_accuracy": 0.19603989124298096,
      "num_tokens": 54986192.0,
      "step": 28110
    },
    {
      "entropy": 5.998457765579223,
      "epoch": 2.8104163542760032,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000421772009905374,
      "loss": 5.5439,
      "mean_token_accuracy": 0.1885556772351265,
      "num_tokens": 54996093.0,
      "step": 28115
    },
    {
      "entropy": 5.931805229187011,
      "epoch": 2.810916179337232,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004217449577315395,
      "loss": 5.5071,
      "mean_token_accuracy": 0.1876009598374367,
      "num_tokens": 55005284.0,
      "step": 28120
    },
    {
      "entropy": 5.948022842407227,
      "epoch": 2.8114160043984606,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004217179018656396,
      "loss": 5.5794,
      "mean_token_accuracy": 0.1820567950606346,
      "num_tokens": 55014693.0,
      "step": 28125
    },
    {
      "entropy": 5.902456569671631,
      "epoch": 2.8119158294596893,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000421690842308355,
      "loss": 5.4512,
      "mean_token_accuracy": 0.19610712677240372,
      "num_tokens": 55024354.0,
      "step": 28130
    },
    {
      "entropy": 5.838148593902588,
      "epoch": 2.8124156545209176,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004216637790603667,
      "loss": 5.3659,
      "mean_token_accuracy": 0.20116177946329117,
      "num_tokens": 55033590.0,
      "step": 28135
    },
    {
      "entropy": 5.894740104675293,
      "epoch": 2.8129154795821463,
      "grad_norm": 1.125,
      "learning_rate": 0.0004216367121223553,
      "loss": 5.4315,
      "mean_token_accuracy": 0.1955675020813942,
      "num_tokens": 55042639.0,
      "step": 28140
    },
    {
      "entropy": 5.9409466743469235,
      "epoch": 2.813415304643375,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000421609641495002,
      "loss": 5.5285,
      "mean_token_accuracy": 0.1878102645277977,
      "num_tokens": 55053825.0,
      "step": 28145
    },
    {
      "entropy": 5.878577709197998,
      "epoch": 2.8139151297046032,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042158256717898785,
      "loss": 5.4784,
      "mean_token_accuracy": 0.19048918187618255,
      "num_tokens": 55063142.0,
      "step": 28150
    },
    {
      "entropy": 5.903825044631958,
      "epoch": 2.814414954765832,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000421555489174994,
      "loss": 5.4877,
      "mean_token_accuracy": 0.1942693695425987,
      "num_tokens": 55073122.0,
      "step": 28155
    },
    {
      "entropy": 5.98216757774353,
      "epoch": 2.8149147798270606,
      "grad_norm": 1.0,
      "learning_rate": 0.00042152840748370184,
      "loss": 5.5863,
      "mean_token_accuracy": 0.1796499162912369,
      "num_tokens": 55083364.0,
      "step": 28160
    },
    {
      "entropy": 6.001443529129029,
      "epoch": 2.8154146048882893,
      "grad_norm": 1.125,
      "learning_rate": 0.0004215013221057926,
      "loss": 5.554,
      "mean_token_accuracy": 0.1857842966914177,
      "num_tokens": 55092540.0,
      "step": 28165
    },
    {
      "entropy": 6.027275657653808,
      "epoch": 2.8159144299495176,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004214742330419478,
      "loss": 5.612,
      "mean_token_accuracy": 0.1838885009288788,
      "num_tokens": 55103085.0,
      "step": 28170
    },
    {
      "entropy": 5.9994950771331785,
      "epoch": 2.8164142550107463,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042144714029284893,
      "loss": 5.6149,
      "mean_token_accuracy": 0.1809030741453171,
      "num_tokens": 55112962.0,
      "step": 28175
    },
    {
      "entropy": 5.826338481903076,
      "epoch": 2.816914080071975,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042142004385917783,
      "loss": 5.4148,
      "mean_token_accuracy": 0.19715265482664107,
      "num_tokens": 55122795.0,
      "step": 28180
    },
    {
      "entropy": 5.778695154190063,
      "epoch": 2.8174139051332032,
      "grad_norm": 1.703125,
      "learning_rate": 0.000421392943741616,
      "loss": 5.4349,
      "mean_token_accuracy": 0.20230239480733872,
      "num_tokens": 55133082.0,
      "step": 28185
    },
    {
      "entropy": 5.9763130187988285,
      "epoch": 2.817913730194432,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004213658399408454,
      "loss": 5.5412,
      "mean_token_accuracy": 0.19042918235063552,
      "num_tokens": 55142940.0,
      "step": 28190
    },
    {
      "entropy": 6.12561731338501,
      "epoch": 2.8184135552556606,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004213387324575479,
      "loss": 5.705,
      "mean_token_accuracy": 0.17354455441236497,
      "num_tokens": 55152233.0,
      "step": 28195
    },
    {
      "entropy": 6.060876083374024,
      "epoch": 2.8189133803168893,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004213116212924056,
      "loss": 5.5956,
      "mean_token_accuracy": 0.1884746253490448,
      "num_tokens": 55163159.0,
      "step": 28200
    },
    {
      "entropy": 5.910449504852295,
      "epoch": 2.8194132053781176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042128450644610046,
      "loss": 5.4613,
      "mean_token_accuracy": 0.19370798468589784,
      "num_tokens": 55173902.0,
      "step": 28205
    },
    {
      "entropy": 5.895769357681274,
      "epoch": 2.8199130304393463,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042125738791931486,
      "loss": 5.5113,
      "mean_token_accuracy": 0.18985333293676376,
      "num_tokens": 55184698.0,
      "step": 28210
    },
    {
      "entropy": 5.92152681350708,
      "epoch": 2.8204128555005745,
      "grad_norm": 1.0,
      "learning_rate": 0.000421230265712731,
      "loss": 5.5809,
      "mean_token_accuracy": 0.17714362889528273,
      "num_tokens": 55194463.0,
      "step": 28215
    },
    {
      "entropy": 5.897674655914306,
      "epoch": 2.820912680561803,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004212031398270313,
      "loss": 5.569,
      "mean_token_accuracy": 0.1921137571334839,
      "num_tokens": 55205844.0,
      "step": 28220
    },
    {
      "entropy": 6.064357566833496,
      "epoch": 2.821412505623032,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042117601026289827,
      "loss": 5.5907,
      "mean_token_accuracy": 0.18573297411203385,
      "num_tokens": 55215760.0,
      "step": 28225
    },
    {
      "entropy": 5.988179302215576,
      "epoch": 2.8219123306842606,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042114887702101434,
      "loss": 5.5407,
      "mean_token_accuracy": 0.18579611629247667,
      "num_tokens": 55226182.0,
      "step": 28230
    },
    {
      "entropy": 5.9334169864654545,
      "epoch": 2.8224121557454893,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042112174010206226,
      "loss": 5.5586,
      "mean_token_accuracy": 0.1791626200079918,
      "num_tokens": 55236200.0,
      "step": 28235
    },
    {
      "entropy": 5.989549446105957,
      "epoch": 2.8229119808067176,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042109459950672487,
      "loss": 5.5581,
      "mean_token_accuracy": 0.18055050522089006,
      "num_tokens": 55245455.0,
      "step": 28240
    },
    {
      "entropy": 5.985310029983521,
      "epoch": 2.8234118058679463,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042106745523568494,
      "loss": 5.5579,
      "mean_token_accuracy": 0.18382130116224288,
      "num_tokens": 55254555.0,
      "step": 28245
    },
    {
      "entropy": 6.004694032669067,
      "epoch": 2.8239116309291745,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042104030728962535,
      "loss": 5.6168,
      "mean_token_accuracy": 0.18760237097740173,
      "num_tokens": 55262897.0,
      "step": 28250
    },
    {
      "entropy": 5.898211050033569,
      "epoch": 2.824411455990403,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042101315566922927,
      "loss": 5.544,
      "mean_token_accuracy": 0.19160042852163314,
      "num_tokens": 55272715.0,
      "step": 28255
    },
    {
      "entropy": 5.955985164642334,
      "epoch": 2.824911281051632,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042098600037517985,
      "loss": 5.4985,
      "mean_token_accuracy": 0.18480609804391862,
      "num_tokens": 55282281.0,
      "step": 28260
    },
    {
      "entropy": 5.933600282669067,
      "epoch": 2.8254111061128606,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004209588414081601,
      "loss": 5.5131,
      "mean_token_accuracy": 0.19045957922935486,
      "num_tokens": 55292708.0,
      "step": 28265
    },
    {
      "entropy": 5.828449487686157,
      "epoch": 2.8259109311740893,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042093167876885345,
      "loss": 5.4997,
      "mean_token_accuracy": 0.18900159299373626,
      "num_tokens": 55302750.0,
      "step": 28270
    },
    {
      "entropy": 5.94438853263855,
      "epoch": 2.8264107562353176,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004209045124579434,
      "loss": 5.5235,
      "mean_token_accuracy": 0.18935246169567108,
      "num_tokens": 55312136.0,
      "step": 28275
    },
    {
      "entropy": 6.01275749206543,
      "epoch": 2.8269105812965463,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042087734247611333,
      "loss": 5.5381,
      "mean_token_accuracy": 0.18515995889902115,
      "num_tokens": 55321770.0,
      "step": 28280
    },
    {
      "entropy": 5.88593111038208,
      "epoch": 2.8274104063577745,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004208501688240468,
      "loss": 5.4735,
      "mean_token_accuracy": 0.18592232465744019,
      "num_tokens": 55331758.0,
      "step": 28285
    },
    {
      "entropy": 5.943528938293457,
      "epoch": 2.827910231419003,
      "grad_norm": 1.0,
      "learning_rate": 0.00042082299150242764,
      "loss": 5.5315,
      "mean_token_accuracy": 0.18609433025121688,
      "num_tokens": 55341879.0,
      "step": 28290
    },
    {
      "entropy": 5.835591411590576,
      "epoch": 2.828410056480232,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004207958105119395,
      "loss": 5.3723,
      "mean_token_accuracy": 0.20733868777751924,
      "num_tokens": 55351313.0,
      "step": 28295
    },
    {
      "entropy": 5.912341547012329,
      "epoch": 2.8289098815414606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042076862585326626,
      "loss": 5.4787,
      "mean_token_accuracy": 0.199055977165699,
      "num_tokens": 55360838.0,
      "step": 28300
    },
    {
      "entropy": 5.982745790481568,
      "epoch": 2.8294097066026893,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042074143752709195,
      "loss": 5.5896,
      "mean_token_accuracy": 0.1891237676143646,
      "num_tokens": 55370698.0,
      "step": 28305
    },
    {
      "entropy": 5.997162961959839,
      "epoch": 2.8299095316639176,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004207142455341005,
      "loss": 5.5752,
      "mean_token_accuracy": 0.18167130947113036,
      "num_tokens": 55379897.0,
      "step": 28310
    },
    {
      "entropy": 5.913574600219727,
      "epoch": 2.8304093567251463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042068704987497627,
      "loss": 5.5122,
      "mean_token_accuracy": 0.18687023222446442,
      "num_tokens": 55390950.0,
      "step": 28315
    },
    {
      "entropy": 5.9341541767120365,
      "epoch": 2.8309091817863745,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042065985055040326,
      "loss": 5.4466,
      "mean_token_accuracy": 0.1972130134701729,
      "num_tokens": 55400400.0,
      "step": 28320
    },
    {
      "entropy": 5.894726037979126,
      "epoch": 2.831409006847603,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004206326475610658,
      "loss": 5.4925,
      "mean_token_accuracy": 0.18738164007663727,
      "num_tokens": 55409583.0,
      "step": 28325
    },
    {
      "entropy": 5.844868946075439,
      "epoch": 2.831908831908832,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042060544090764856,
      "loss": 5.4179,
      "mean_token_accuracy": 0.20002300143241883,
      "num_tokens": 55418142.0,
      "step": 28330
    },
    {
      "entropy": 5.9746294021606445,
      "epoch": 2.8324086569700606,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004205782305908358,
      "loss": 5.5777,
      "mean_token_accuracy": 0.18009734004735947,
      "num_tokens": 55428258.0,
      "step": 28335
    },
    {
      "entropy": 6.001496696472168,
      "epoch": 2.8329084820312893,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004205510166113122,
      "loss": 5.5586,
      "mean_token_accuracy": 0.18607941418886184,
      "num_tokens": 55438414.0,
      "step": 28340
    },
    {
      "entropy": 5.951366758346557,
      "epoch": 2.8334083070925176,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042052379896976254,
      "loss": 5.555,
      "mean_token_accuracy": 0.1796771243214607,
      "num_tokens": 55448820.0,
      "step": 28345
    },
    {
      "entropy": 5.880062484741211,
      "epoch": 2.8339081321537463,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004204965776668716,
      "loss": 5.4574,
      "mean_token_accuracy": 0.19552581161260604,
      "num_tokens": 55459031.0,
      "step": 28350
    },
    {
      "entropy": 5.90883584022522,
      "epoch": 2.8344079572149745,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004204693527033241,
      "loss": 5.504,
      "mean_token_accuracy": 0.19597085416316987,
      "num_tokens": 55468647.0,
      "step": 28355
    },
    {
      "entropy": 6.008048725128174,
      "epoch": 2.834907782276203,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042044212407980517,
      "loss": 5.5072,
      "mean_token_accuracy": 0.1864774614572525,
      "num_tokens": 55478727.0,
      "step": 28360
    },
    {
      "entropy": 5.96106972694397,
      "epoch": 2.835407607337432,
      "grad_norm": 1.125,
      "learning_rate": 0.00042041489179699983,
      "loss": 5.5417,
      "mean_token_accuracy": 0.19198026955127717,
      "num_tokens": 55489115.0,
      "step": 28365
    },
    {
      "entropy": 5.968136644363403,
      "epoch": 2.8359074323986606,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004203876558555932,
      "loss": 5.5358,
      "mean_token_accuracy": 0.18547382950782776,
      "num_tokens": 55498730.0,
      "step": 28370
    },
    {
      "entropy": 5.87737889289856,
      "epoch": 2.8364072574598893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042036041625627055,
      "loss": 5.5174,
      "mean_token_accuracy": 0.19109119474887848,
      "num_tokens": 55507659.0,
      "step": 28375
    },
    {
      "entropy": 5.852508974075318,
      "epoch": 2.8369070825211176,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042033317299971726,
      "loss": 5.4181,
      "mean_token_accuracy": 0.19331544786691665,
      "num_tokens": 55517077.0,
      "step": 28380
    },
    {
      "entropy": 5.913910913467407,
      "epoch": 2.8374069075823463,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004203059260866187,
      "loss": 5.4155,
      "mean_token_accuracy": 0.19694342315196992,
      "num_tokens": 55526349.0,
      "step": 28385
    },
    {
      "entropy": 5.883852195739746,
      "epoch": 2.8379067326435745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004202786755176604,
      "loss": 5.4867,
      "mean_token_accuracy": 0.19208816885948182,
      "num_tokens": 55537417.0,
      "step": 28390
    },
    {
      "entropy": 5.933791637420654,
      "epoch": 2.838406557704803,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004202514212935281,
      "loss": 5.5119,
      "mean_token_accuracy": 0.1899716630578041,
      "num_tokens": 55546123.0,
      "step": 28395
    },
    {
      "entropy": 5.860758113861084,
      "epoch": 2.838906382766032,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004202241634149074,
      "loss": 5.3961,
      "mean_token_accuracy": 0.19951796680688857,
      "num_tokens": 55555740.0,
      "step": 28400
    },
    {
      "entropy": 5.915865898132324,
      "epoch": 2.8394062078272606,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042019690188248407,
      "loss": 5.4608,
      "mean_token_accuracy": 0.19289458990097047,
      "num_tokens": 55564435.0,
      "step": 28405
    },
    {
      "entropy": 6.012536573410034,
      "epoch": 2.839906032888489,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004201696366969441,
      "loss": 5.543,
      "mean_token_accuracy": 0.19358502179384232,
      "num_tokens": 55573893.0,
      "step": 28410
    },
    {
      "entropy": 5.981504344940186,
      "epoch": 2.8404058579497176,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004201423678589734,
      "loss": 5.4682,
      "mean_token_accuracy": 0.19280693531036378,
      "num_tokens": 55583330.0,
      "step": 28415
    },
    {
      "entropy": 5.852391099929809,
      "epoch": 2.8409056830109463,
      "grad_norm": 1.203125,
      "learning_rate": 0.000420115095369258,
      "loss": 5.4545,
      "mean_token_accuracy": 0.19578118324279786,
      "num_tokens": 55593000.0,
      "step": 28420
    },
    {
      "entropy": 5.852564954757691,
      "epoch": 2.8414055080721745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004200878192284843,
      "loss": 5.5431,
      "mean_token_accuracy": 0.18255771547555924,
      "num_tokens": 55602835.0,
      "step": 28425
    },
    {
      "entropy": 5.896863222122192,
      "epoch": 2.841905333133403,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004200605394373383,
      "loss": 5.4368,
      "mean_token_accuracy": 0.19375575482845306,
      "num_tokens": 55612978.0,
      "step": 28430
    },
    {
      "entropy": 5.974686098098755,
      "epoch": 2.842405158194632,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004200332559965065,
      "loss": 5.6076,
      "mean_token_accuracy": 0.17851597517728807,
      "num_tokens": 55622386.0,
      "step": 28435
    },
    {
      "entropy": 5.964634609222412,
      "epoch": 2.8429049832558606,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004200059689066752,
      "loss": 5.5616,
      "mean_token_accuracy": 0.18346254527568817,
      "num_tokens": 55631340.0,
      "step": 28440
    },
    {
      "entropy": 6.024762582778931,
      "epoch": 2.843404808317089,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041997867816853115,
      "loss": 5.5699,
      "mean_token_accuracy": 0.18097881078720093,
      "num_tokens": 55641000.0,
      "step": 28445
    },
    {
      "entropy": 5.89292049407959,
      "epoch": 2.8439046333783176,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041995138378276084,
      "loss": 5.4677,
      "mean_token_accuracy": 0.19279708862304687,
      "num_tokens": 55650398.0,
      "step": 28450
    },
    {
      "entropy": 5.969854688644409,
      "epoch": 2.8444044584395463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041992408575005095,
      "loss": 5.508,
      "mean_token_accuracy": 0.1902477338910103,
      "num_tokens": 55659454.0,
      "step": 28455
    },
    {
      "entropy": 5.958832216262818,
      "epoch": 2.8449042835007745,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004198967840710884,
      "loss": 5.6432,
      "mean_token_accuracy": 0.18291402757167816,
      "num_tokens": 55669166.0,
      "step": 28460
    },
    {
      "entropy": 6.020224618911743,
      "epoch": 2.845404108562003,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041986947874656,
      "loss": 5.5666,
      "mean_token_accuracy": 0.18640873581171036,
      "num_tokens": 55678753.0,
      "step": 28465
    },
    {
      "entropy": 5.973300838470459,
      "epoch": 2.845903933623232,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004198421697771529,
      "loss": 5.4995,
      "mean_token_accuracy": 0.18819977045059205,
      "num_tokens": 55688291.0,
      "step": 28470
    },
    {
      "entropy": 5.949191999435425,
      "epoch": 2.8464037586844606,
      "grad_norm": 1.109375,
      "learning_rate": 0.000419814857163554,
      "loss": 5.5568,
      "mean_token_accuracy": 0.1873816058039665,
      "num_tokens": 55697526.0,
      "step": 28475
    },
    {
      "entropy": 5.949254179000855,
      "epoch": 2.846903583745689,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004197875409064506,
      "loss": 5.5313,
      "mean_token_accuracy": 0.18691585808992386,
      "num_tokens": 55707560.0,
      "step": 28480
    },
    {
      "entropy": 5.976622152328491,
      "epoch": 2.8474034088069176,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004197602210065299,
      "loss": 5.5857,
      "mean_token_accuracy": 0.1843176767230034,
      "num_tokens": 55716428.0,
      "step": 28485
    },
    {
      "entropy": 5.924442434310913,
      "epoch": 2.8479032338681463,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004197328974644793,
      "loss": 5.5772,
      "mean_token_accuracy": 0.18756356090307236,
      "num_tokens": 55726123.0,
      "step": 28490
    },
    {
      "entropy": 5.879773044586182,
      "epoch": 2.8484030589293745,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004197055702809862,
      "loss": 5.4507,
      "mean_token_accuracy": 0.18925990015268326,
      "num_tokens": 55735668.0,
      "step": 28495
    },
    {
      "entropy": 5.9037011623382565,
      "epoch": 2.848902883990603,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004196782394567382,
      "loss": 5.4484,
      "mean_token_accuracy": 0.1917483314871788,
      "num_tokens": 55745651.0,
      "step": 28500
    },
    {
      "entropy": 5.948295307159424,
      "epoch": 2.849402709051832,
      "grad_norm": 1.125,
      "learning_rate": 0.000419650904992423,
      "loss": 5.5018,
      "mean_token_accuracy": 0.18462096452713012,
      "num_tokens": 55755024.0,
      "step": 28505
    },
    {
      "entropy": 6.012776136398315,
      "epoch": 2.8499025341130606,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041962356688872817,
      "loss": 5.5648,
      "mean_token_accuracy": 0.18106954097747802,
      "num_tokens": 55765524.0,
      "step": 28510
    },
    {
      "entropy": 6.024839448928833,
      "epoch": 2.850402359174289,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004195962251463417,
      "loss": 5.5751,
      "mean_token_accuracy": 0.17912918627262114,
      "num_tokens": 55775630.0,
      "step": 28515
    },
    {
      "entropy": 5.967176818847657,
      "epoch": 2.8509021842355176,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004195688797659513,
      "loss": 5.5962,
      "mean_token_accuracy": 0.18796517252922057,
      "num_tokens": 55786375.0,
      "step": 28520
    },
    {
      "entropy": 5.937998867034912,
      "epoch": 2.8514020092967463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041954153074824516,
      "loss": 5.5343,
      "mean_token_accuracy": 0.1851406902074814,
      "num_tokens": 55795080.0,
      "step": 28525
    },
    {
      "entropy": 6.000917625427246,
      "epoch": 2.8519018343579745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004195141780939113,
      "loss": 5.5319,
      "mean_token_accuracy": 0.18365013003349304,
      "num_tokens": 55804737.0,
      "step": 28530
    },
    {
      "entropy": 5.887533140182495,
      "epoch": 2.852401659419203,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041948682180363784,
      "loss": 5.3924,
      "mean_token_accuracy": 0.1946636915206909,
      "num_tokens": 55814166.0,
      "step": 28535
    },
    {
      "entropy": 5.903865623474121,
      "epoch": 2.852901484480432,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004194594618781132,
      "loss": 5.4718,
      "mean_token_accuracy": 0.19422465711832046,
      "num_tokens": 55823758.0,
      "step": 28540
    },
    {
      "entropy": 6.01491003036499,
      "epoch": 2.8534013095416606,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041943209831802567,
      "loss": 5.5257,
      "mean_token_accuracy": 0.1864191249012947,
      "num_tokens": 55833798.0,
      "step": 28545
    },
    {
      "entropy": 5.927063131332398,
      "epoch": 2.853901134602889,
      "grad_norm": 1.125,
      "learning_rate": 0.00041940473112406365,
      "loss": 5.5359,
      "mean_token_accuracy": 0.18275245428085327,
      "num_tokens": 55843456.0,
      "step": 28550
    },
    {
      "entropy": 5.868661880493164,
      "epoch": 2.8544009596641176,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041937736029691575,
      "loss": 5.4282,
      "mean_token_accuracy": 0.1976074367761612,
      "num_tokens": 55853401.0,
      "step": 28555
    },
    {
      "entropy": 5.930231428146362,
      "epoch": 2.8549007847253463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004193499858372706,
      "loss": 5.4754,
      "mean_token_accuracy": 0.19771249294281007,
      "num_tokens": 55863549.0,
      "step": 28560
    },
    {
      "entropy": 6.016086292266846,
      "epoch": 2.8554006097865745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000419322607745817,
      "loss": 5.5536,
      "mean_token_accuracy": 0.1886284038424492,
      "num_tokens": 55873718.0,
      "step": 28565
    },
    {
      "entropy": 5.924648284912109,
      "epoch": 2.855900434847803,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041929522602324377,
      "loss": 5.4715,
      "mean_token_accuracy": 0.19427599012851715,
      "num_tokens": 55883909.0,
      "step": 28570
    },
    {
      "entropy": 5.929278421401977,
      "epoch": 2.856400259909032,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004192678406702398,
      "loss": 5.5554,
      "mean_token_accuracy": 0.17643021643161774,
      "num_tokens": 55893090.0,
      "step": 28575
    },
    {
      "entropy": 5.859233713150024,
      "epoch": 2.8569000849702606,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000419240451687494,
      "loss": 5.4382,
      "mean_token_accuracy": 0.19422834515571594,
      "num_tokens": 55903820.0,
      "step": 28580
    },
    {
      "entropy": 6.036321020126342,
      "epoch": 2.857399910031489,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041921305907569555,
      "loss": 5.5442,
      "mean_token_accuracy": 0.18646028637886047,
      "num_tokens": 55913540.0,
      "step": 28585
    },
    {
      "entropy": 5.961747121810913,
      "epoch": 2.8578997350927176,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004191856628355337,
      "loss": 5.5287,
      "mean_token_accuracy": 0.18365492522716523,
      "num_tokens": 55923766.0,
      "step": 28590
    },
    {
      "entropy": 5.959750938415527,
      "epoch": 2.8583995601539463,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004191582629676977,
      "loss": 5.5664,
      "mean_token_accuracy": 0.18894660472869873,
      "num_tokens": 55932855.0,
      "step": 28595
    },
    {
      "entropy": 5.914099264144897,
      "epoch": 2.8588993852151745,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041913085947287675,
      "loss": 5.5204,
      "mean_token_accuracy": 0.1867166593670845,
      "num_tokens": 55942163.0,
      "step": 28600
    },
    {
      "entropy": 5.92921895980835,
      "epoch": 2.859399210276403,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004191034523517606,
      "loss": 5.497,
      "mean_token_accuracy": 0.191119047999382,
      "num_tokens": 55952521.0,
      "step": 28605
    },
    {
      "entropy": 5.945659875869751,
      "epoch": 2.859899035337632,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041907604160503864,
      "loss": 5.5194,
      "mean_token_accuracy": 0.1888028249144554,
      "num_tokens": 55962071.0,
      "step": 28610
    },
    {
      "entropy": 5.820277166366577,
      "epoch": 2.8603988603988606,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041904862723340046,
      "loss": 5.4314,
      "mean_token_accuracy": 0.1908804729580879,
      "num_tokens": 55971387.0,
      "step": 28615
    },
    {
      "entropy": 5.94819450378418,
      "epoch": 2.860898685460089,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041902120923753593,
      "loss": 5.5135,
      "mean_token_accuracy": 0.18648241609334945,
      "num_tokens": 55980435.0,
      "step": 28620
    },
    {
      "entropy": 5.882032775878907,
      "epoch": 2.8613985105213176,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004189937876181349,
      "loss": 5.4233,
      "mean_token_accuracy": 0.19598147124052048,
      "num_tokens": 55989868.0,
      "step": 28625
    },
    {
      "entropy": 5.910932016372681,
      "epoch": 2.8618983355825462,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004189663623758872,
      "loss": 5.4557,
      "mean_token_accuracy": 0.1945648893713951,
      "num_tokens": 55999423.0,
      "step": 28630
    },
    {
      "entropy": 5.853791761398315,
      "epoch": 2.8623981606437745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041893893351148284,
      "loss": 5.4569,
      "mean_token_accuracy": 0.1896259978413582,
      "num_tokens": 56009429.0,
      "step": 28635
    },
    {
      "entropy": 5.974388980865479,
      "epoch": 2.862897985705003,
      "grad_norm": 1.046875,
      "learning_rate": 0.000418911501025612,
      "loss": 5.5106,
      "mean_token_accuracy": 0.18749415427446364,
      "num_tokens": 56019039.0,
      "step": 28640
    },
    {
      "entropy": 5.97268123626709,
      "epoch": 2.863397810766232,
      "grad_norm": 1.125,
      "learning_rate": 0.0004188840649189648,
      "loss": 5.5572,
      "mean_token_accuracy": 0.1837095484137535,
      "num_tokens": 56029465.0,
      "step": 28645
    },
    {
      "entropy": 5.972398376464843,
      "epoch": 2.8638976358274606,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041885662519223154,
      "loss": 5.6086,
      "mean_token_accuracy": 0.1870669737458229,
      "num_tokens": 56038882.0,
      "step": 28650
    },
    {
      "entropy": 6.04751672744751,
      "epoch": 2.864397460888689,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004188291818461026,
      "loss": 5.6094,
      "mean_token_accuracy": 0.17886933386325837,
      "num_tokens": 56048512.0,
      "step": 28655
    },
    {
      "entropy": 5.926358032226562,
      "epoch": 2.8648972859499175,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041880173488126847,
      "loss": 5.4687,
      "mean_token_accuracy": 0.19171295464038848,
      "num_tokens": 56057553.0,
      "step": 28660
    },
    {
      "entropy": 5.878042984008789,
      "epoch": 2.8653971110111462,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041877428429841967,
      "loss": 5.4971,
      "mean_token_accuracy": 0.1892929419875145,
      "num_tokens": 56066819.0,
      "step": 28665
    },
    {
      "entropy": 5.90529146194458,
      "epoch": 2.8658969360723745,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004187468300982469,
      "loss": 5.5566,
      "mean_token_accuracy": 0.18872860074043274,
      "num_tokens": 56077012.0,
      "step": 28670
    },
    {
      "entropy": 5.97804627418518,
      "epoch": 2.866396761133603,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004187193722814409,
      "loss": 5.6165,
      "mean_token_accuracy": 0.18125576972961427,
      "num_tokens": 56086405.0,
      "step": 28675
    },
    {
      "entropy": 5.974224042892456,
      "epoch": 2.866896586194832,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004186919108486925,
      "loss": 5.4437,
      "mean_token_accuracy": 0.19002047032117844,
      "num_tokens": 56097135.0,
      "step": 28680
    },
    {
      "entropy": 5.883247756958008,
      "epoch": 2.8673964112560606,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004186644458006925,
      "loss": 5.4236,
      "mean_token_accuracy": 0.1965599998831749,
      "num_tokens": 56106401.0,
      "step": 28685
    },
    {
      "entropy": 5.839110565185547,
      "epoch": 2.867896236317289,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000418636977138132,
      "loss": 5.3685,
      "mean_token_accuracy": 0.20477745085954666,
      "num_tokens": 56115763.0,
      "step": 28690
    },
    {
      "entropy": 5.872228288650513,
      "epoch": 2.8683960613785175,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041860950486170227,
      "loss": 5.5089,
      "mean_token_accuracy": 0.1915462777018547,
      "num_tokens": 56124873.0,
      "step": 28695
    },
    {
      "entropy": 5.857125663757325,
      "epoch": 2.8688958864397462,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004185820289720942,
      "loss": 5.3817,
      "mean_token_accuracy": 0.20273543000221253,
      "num_tokens": 56133801.0,
      "step": 28700
    },
    {
      "entropy": 5.991905403137207,
      "epoch": 2.8693957115009745,
      "grad_norm": 1.125,
      "learning_rate": 0.00041855454946999925,
      "loss": 5.5694,
      "mean_token_accuracy": 0.17962211072444917,
      "num_tokens": 56143827.0,
      "step": 28705
    },
    {
      "entropy": 5.9452550411224365,
      "epoch": 2.869895536562203,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041852706635610884,
      "loss": 5.5714,
      "mean_token_accuracy": 0.17577485144138336,
      "num_tokens": 56152780.0,
      "step": 28710
    },
    {
      "entropy": 5.885336112976074,
      "epoch": 2.870395361623432,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041849957963111434,
      "loss": 5.4472,
      "mean_token_accuracy": 0.1961646631360054,
      "num_tokens": 56162168.0,
      "step": 28715
    },
    {
      "entropy": 5.934825229644775,
      "epoch": 2.8708951866846606,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004184720892957073,
      "loss": 5.5158,
      "mean_token_accuracy": 0.19030641615390778,
      "num_tokens": 56172069.0,
      "step": 28720
    },
    {
      "entropy": 5.957781457901001,
      "epoch": 2.871395011745889,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004184445953505795,
      "loss": 5.5167,
      "mean_token_accuracy": 0.18860740065574647,
      "num_tokens": 56181904.0,
      "step": 28725
    },
    {
      "entropy": 5.952534246444702,
      "epoch": 2.8718948368071175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004184170977964225,
      "loss": 5.4921,
      "mean_token_accuracy": 0.1852252885699272,
      "num_tokens": 56191203.0,
      "step": 28730
    },
    {
      "entropy": 5.950167465209961,
      "epoch": 2.8723946618683462,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00041838959663392827,
      "loss": 5.5104,
      "mean_token_accuracy": 0.18761863559484482,
      "num_tokens": 56201714.0,
      "step": 28735
    },
    {
      "entropy": 5.940114831924438,
      "epoch": 2.8728944869295745,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004183620918637887,
      "loss": 5.5221,
      "mean_token_accuracy": 0.18875628411769868,
      "num_tokens": 56213287.0,
      "step": 28740
    },
    {
      "entropy": 5.926656770706177,
      "epoch": 2.873394311990803,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004183345834866957,
      "loss": 5.4343,
      "mean_token_accuracy": 0.20224920213222503,
      "num_tokens": 56222593.0,
      "step": 28745
    },
    {
      "entropy": 5.919744539260864,
      "epoch": 2.873894137052032,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041830707150334164,
      "loss": 5.4768,
      "mean_token_accuracy": 0.1865305557847023,
      "num_tokens": 56232292.0,
      "step": 28750
    },
    {
      "entropy": 5.88309907913208,
      "epoch": 2.8743939621132606,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004182795559144184,
      "loss": 5.4666,
      "mean_token_accuracy": 0.19358340799808502,
      "num_tokens": 56242389.0,
      "step": 28755
    },
    {
      "entropy": 5.909613418579101,
      "epoch": 2.874893787174489,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004182520367206184,
      "loss": 5.5873,
      "mean_token_accuracy": 0.18656654059886932,
      "num_tokens": 56252383.0,
      "step": 28760
    },
    {
      "entropy": 5.908715772628784,
      "epoch": 2.8753936122357175,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041822451392263413,
      "loss": 5.5238,
      "mean_token_accuracy": 0.18587066233158112,
      "num_tokens": 56261676.0,
      "step": 28765
    },
    {
      "entropy": 5.929866600036621,
      "epoch": 2.875893437296946,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004181969875211578,
      "loss": 5.449,
      "mean_token_accuracy": 0.19332660883665084,
      "num_tokens": 56271938.0,
      "step": 28770
    },
    {
      "entropy": 5.910695314407349,
      "epoch": 2.8763932623581745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041816945751688215,
      "loss": 5.5229,
      "mean_token_accuracy": 0.18334124386310577,
      "num_tokens": 56281370.0,
      "step": 28775
    },
    {
      "entropy": 5.910592794418335,
      "epoch": 2.876893087419403,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041814192391049983,
      "loss": 5.5056,
      "mean_token_accuracy": 0.1888559266924858,
      "num_tokens": 56292162.0,
      "step": 28780
    },
    {
      "entropy": 5.9632213592529295,
      "epoch": 2.877392912480632,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004181143867027036,
      "loss": 5.4476,
      "mean_token_accuracy": 0.1970451220870018,
      "num_tokens": 56301129.0,
      "step": 28785
    },
    {
      "entropy": 5.954134607315064,
      "epoch": 2.8778927375418606,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041808684589418615,
      "loss": 5.544,
      "mean_token_accuracy": 0.18624549210071564,
      "num_tokens": 56310781.0,
      "step": 28790
    },
    {
      "entropy": 5.8753375053405765,
      "epoch": 2.878392562603089,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004180593014856404,
      "loss": 5.4771,
      "mean_token_accuracy": 0.1975930690765381,
      "num_tokens": 56320401.0,
      "step": 28795
    },
    {
      "entropy": 5.854989385604858,
      "epoch": 2.8788923876643175,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004180317534777596,
      "loss": 5.5079,
      "mean_token_accuracy": 0.19499246627092362,
      "num_tokens": 56330606.0,
      "step": 28800
    },
    {
      "entropy": 5.9276433944702145,
      "epoch": 2.879392212725546,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041800420187123656,
      "loss": 5.4819,
      "mean_token_accuracy": 0.19215121418237685,
      "num_tokens": 56340832.0,
      "step": 28805
    },
    {
      "entropy": 5.837391185760498,
      "epoch": 2.8798920377867745,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004179766466667646,
      "loss": 5.4514,
      "mean_token_accuracy": 0.19818236380815507,
      "num_tokens": 56350228.0,
      "step": 28810
    },
    {
      "entropy": 5.940051031112671,
      "epoch": 2.880391862848003,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004179490878650371,
      "loss": 5.4712,
      "mean_token_accuracy": 0.18900782465934754,
      "num_tokens": 56360087.0,
      "step": 28815
    },
    {
      "entropy": 5.90391845703125,
      "epoch": 2.880891687909232,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004179215254667473,
      "loss": 5.4547,
      "mean_token_accuracy": 0.19301908314228058,
      "num_tokens": 56369214.0,
      "step": 28820
    },
    {
      "entropy": 5.871996450424194,
      "epoch": 2.8813915129704606,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004178939594725887,
      "loss": 5.5485,
      "mean_token_accuracy": 0.19055332988500595,
      "num_tokens": 56378331.0,
      "step": 28825
    },
    {
      "entropy": 5.964061450958252,
      "epoch": 2.881891338031689,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004178663898832548,
      "loss": 5.5736,
      "mean_token_accuracy": 0.1795365706086159,
      "num_tokens": 56387856.0,
      "step": 28830
    },
    {
      "entropy": 6.056238031387329,
      "epoch": 2.8823911630929175,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004178388166994392,
      "loss": 5.6343,
      "mean_token_accuracy": 0.1784266859292984,
      "num_tokens": 56398885.0,
      "step": 28835
    },
    {
      "entropy": 5.869563055038452,
      "epoch": 2.882890988154146,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004178112399218359,
      "loss": 5.438,
      "mean_token_accuracy": 0.19177918434143065,
      "num_tokens": 56408169.0,
      "step": 28840
    },
    {
      "entropy": 5.8104650497436525,
      "epoch": 2.8833908132153745,
      "grad_norm": 1.125,
      "learning_rate": 0.00041778365955113844,
      "loss": 5.4166,
      "mean_token_accuracy": 0.1886136054992676,
      "num_tokens": 56417221.0,
      "step": 28845
    },
    {
      "entropy": 5.904326486587524,
      "epoch": 2.883890638276603,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004177560755880409,
      "loss": 5.4693,
      "mean_token_accuracy": 0.19470079988241196,
      "num_tokens": 56426860.0,
      "step": 28850
    },
    {
      "entropy": 6.010273456573486,
      "epoch": 2.884390463337832,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004177284880332373,
      "loss": 5.6061,
      "mean_token_accuracy": 0.1809718906879425,
      "num_tokens": 56436264.0,
      "step": 28855
    },
    {
      "entropy": 5.9656494617462155,
      "epoch": 2.8848902883990606,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041770089688742155,
      "loss": 5.4933,
      "mean_token_accuracy": 0.19577471017837525,
      "num_tokens": 56445835.0,
      "step": 28860
    },
    {
      "entropy": 5.940614461898804,
      "epoch": 2.885390113460289,
      "grad_norm": 1.21875,
      "learning_rate": 0.000417673302151288,
      "loss": 5.5596,
      "mean_token_accuracy": 0.19180833399295807,
      "num_tokens": 56455090.0,
      "step": 28865
    },
    {
      "entropy": 5.924184417724609,
      "epoch": 2.8858899385215175,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004176457038255309,
      "loss": 5.532,
      "mean_token_accuracy": 0.19062764346599578,
      "num_tokens": 56464782.0,
      "step": 28870
    },
    {
      "entropy": 5.9530298709869385,
      "epoch": 2.886389763582746,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004176181019108446,
      "loss": 5.5277,
      "mean_token_accuracy": 0.1884121760725975,
      "num_tokens": 56475375.0,
      "step": 28875
    },
    {
      "entropy": 5.892106914520264,
      "epoch": 2.8868895886439745,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004175904964079236,
      "loss": 5.3864,
      "mean_token_accuracy": 0.19411096423864366,
      "num_tokens": 56485654.0,
      "step": 28880
    },
    {
      "entropy": 5.937690591812133,
      "epoch": 2.887389413705203,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004175628873174623,
      "loss": 5.5438,
      "mean_token_accuracy": 0.19309738427400588,
      "num_tokens": 56495554.0,
      "step": 28885
    },
    {
      "entropy": 5.919396162033081,
      "epoch": 2.887889238766432,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041753527464015547,
      "loss": 5.4295,
      "mean_token_accuracy": 0.19479530453681945,
      "num_tokens": 56504760.0,
      "step": 28890
    },
    {
      "entropy": 5.864511251449585,
      "epoch": 2.8883890638276606,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004175076583766978,
      "loss": 5.4689,
      "mean_token_accuracy": 0.19731661975383757,
      "num_tokens": 56514312.0,
      "step": 28895
    },
    {
      "entropy": 5.900850868225097,
      "epoch": 2.888888888888889,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004174800385277842,
      "loss": 5.5123,
      "mean_token_accuracy": 0.187963604927063,
      "num_tokens": 56524741.0,
      "step": 28900
    },
    {
      "entropy": 5.919299650192261,
      "epoch": 2.8893887139501175,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004174524150941094,
      "loss": 5.5019,
      "mean_token_accuracy": 0.18415771722793578,
      "num_tokens": 56534339.0,
      "step": 28905
    },
    {
      "entropy": 6.0139223575592045,
      "epoch": 2.889888539011346,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004174247880763685,
      "loss": 5.6122,
      "mean_token_accuracy": 0.1814890906214714,
      "num_tokens": 56544722.0,
      "step": 28910
    },
    {
      "entropy": 5.936463022232056,
      "epoch": 2.8903883640725745,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004173971574752566,
      "loss": 5.4202,
      "mean_token_accuracy": 0.1904875412583351,
      "num_tokens": 56553917.0,
      "step": 28915
    },
    {
      "entropy": 5.977611398696899,
      "epoch": 2.890888189133803,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004173695232914688,
      "loss": 5.5327,
      "mean_token_accuracy": 0.19348582476377488,
      "num_tokens": 56563876.0,
      "step": 28920
    },
    {
      "entropy": 5.909697437286377,
      "epoch": 2.891388014195032,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004173418855257005,
      "loss": 5.4623,
      "mean_token_accuracy": 0.1935131549835205,
      "num_tokens": 56574017.0,
      "step": 28925
    },
    {
      "entropy": 5.884473609924316,
      "epoch": 2.8918878392562606,
      "grad_norm": 1.046875,
      "learning_rate": 0.000417314244178647,
      "loss": 5.4866,
      "mean_token_accuracy": 0.1820148527622223,
      "num_tokens": 56583927.0,
      "step": 28930
    },
    {
      "entropy": 5.92265887260437,
      "epoch": 2.892387664317489,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004172865992510037,
      "loss": 5.4924,
      "mean_token_accuracy": 0.1948337122797966,
      "num_tokens": 56593421.0,
      "step": 28935
    },
    {
      "entropy": 5.94312162399292,
      "epoch": 2.8928874893787175,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041725895074346615,
      "loss": 5.4597,
      "mean_token_accuracy": 0.1910391315817833,
      "num_tokens": 56601820.0,
      "step": 28940
    },
    {
      "entropy": 5.895274877548218,
      "epoch": 2.893387314439946,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004172312986567301,
      "loss": 5.4511,
      "mean_token_accuracy": 0.19462193697690963,
      "num_tokens": 56611688.0,
      "step": 28945
    },
    {
      "entropy": 5.855628585815429,
      "epoch": 2.8938871395011745,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004172036429914911,
      "loss": 5.4124,
      "mean_token_accuracy": 0.1983240932226181,
      "num_tokens": 56620719.0,
      "step": 28950
    },
    {
      "entropy": 5.921615219116211,
      "epoch": 2.894386964562403,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041717598374844505,
      "loss": 5.5612,
      "mean_token_accuracy": 0.18852267265319825,
      "num_tokens": 56631045.0,
      "step": 28955
    },
    {
      "entropy": 6.038997793197632,
      "epoch": 2.894886789623632,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041714832092828785,
      "loss": 5.5138,
      "mean_token_accuracy": 0.18551777452230453,
      "num_tokens": 56640935.0,
      "step": 28960
    },
    {
      "entropy": 6.034854507446289,
      "epoch": 2.89538661468486,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041712065453171544,
      "loss": 5.5788,
      "mean_token_accuracy": 0.18050847351551055,
      "num_tokens": 56651023.0,
      "step": 28965
    },
    {
      "entropy": 5.911570358276367,
      "epoch": 2.895886439746089,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041709298455942394,
      "loss": 5.5214,
      "mean_token_accuracy": 0.18776421397924423,
      "num_tokens": 56660591.0,
      "step": 28970
    },
    {
      "entropy": 5.997826290130615,
      "epoch": 2.8963862648073175,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041706531101210956,
      "loss": 5.5969,
      "mean_token_accuracy": 0.18456126302480697,
      "num_tokens": 56671635.0,
      "step": 28975
    },
    {
      "entropy": 5.9238609790802,
      "epoch": 2.896886089868546,
      "grad_norm": 1.125,
      "learning_rate": 0.00041703763389046853,
      "loss": 5.4217,
      "mean_token_accuracy": 0.1937377318739891,
      "num_tokens": 56681309.0,
      "step": 28980
    },
    {
      "entropy": 5.8336570262908936,
      "epoch": 2.8973859149297745,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041700995319519707,
      "loss": 5.479,
      "mean_token_accuracy": 0.19726309329271316,
      "num_tokens": 56692594.0,
      "step": 28985
    },
    {
      "entropy": 5.983526802062988,
      "epoch": 2.897885739991003,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004169822689269918,
      "loss": 5.56,
      "mean_token_accuracy": 0.18486166447401048,
      "num_tokens": 56702165.0,
      "step": 28990
    },
    {
      "entropy": 5.973510599136352,
      "epoch": 2.898385565052232,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041695458108654914,
      "loss": 5.5111,
      "mean_token_accuracy": 0.19282027930021287,
      "num_tokens": 56711398.0,
      "step": 28995
    },
    {
      "entropy": 5.926396751403809,
      "epoch": 2.89888539011346,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041692688967456576,
      "loss": 5.4655,
      "mean_token_accuracy": 0.18956004232168197,
      "num_tokens": 56720984.0,
      "step": 29000
    },
    {
      "entropy": 5.860626077651977,
      "epoch": 2.899385215174689,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041689919469173835,
      "loss": 5.5242,
      "mean_token_accuracy": 0.19547150582075118,
      "num_tokens": 56732126.0,
      "step": 29005
    },
    {
      "entropy": 5.855756187438965,
      "epoch": 2.8998850402359175,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004168714961387637,
      "loss": 5.4242,
      "mean_token_accuracy": 0.19619501531124114,
      "num_tokens": 56741235.0,
      "step": 29010
    },
    {
      "entropy": 5.919070816040039,
      "epoch": 2.900384865297146,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004168437940163388,
      "loss": 5.6057,
      "mean_token_accuracy": 0.18830548226833344,
      "num_tokens": 56751695.0,
      "step": 29015
    },
    {
      "entropy": 5.980143404006958,
      "epoch": 2.9008846903583745,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004168160883251605,
      "loss": 5.5289,
      "mean_token_accuracy": 0.18910813331604004,
      "num_tokens": 56761067.0,
      "step": 29020
    },
    {
      "entropy": 5.984479331970215,
      "epoch": 2.901384515419603,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041678837906592586,
      "loss": 5.4983,
      "mean_token_accuracy": 0.19068045765161515,
      "num_tokens": 56770267.0,
      "step": 29025
    },
    {
      "entropy": 5.911980772018433,
      "epoch": 2.901884340480832,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041676066623933216,
      "loss": 5.5046,
      "mean_token_accuracy": 0.19440484791994095,
      "num_tokens": 56780263.0,
      "step": 29030
    },
    {
      "entropy": 5.871411657333374,
      "epoch": 2.90238416554206,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004167329498460766,
      "loss": 5.4692,
      "mean_token_accuracy": 0.1902375727891922,
      "num_tokens": 56789490.0,
      "step": 29035
    },
    {
      "entropy": 5.902395153045655,
      "epoch": 2.902883990603289,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041670522988685637,
      "loss": 5.4946,
      "mean_token_accuracy": 0.1905912458896637,
      "num_tokens": 56799299.0,
      "step": 29040
    },
    {
      "entropy": 5.8899304389953615,
      "epoch": 2.9033838156645175,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041667750636236914,
      "loss": 5.4594,
      "mean_token_accuracy": 0.19022136479616164,
      "num_tokens": 56808518.0,
      "step": 29045
    },
    {
      "entropy": 6.000769805908203,
      "epoch": 2.903883640725746,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004166497792733122,
      "loss": 5.6117,
      "mean_token_accuracy": 0.18429325819015502,
      "num_tokens": 56819458.0,
      "step": 29050
    },
    {
      "entropy": 5.995767402648926,
      "epoch": 2.9043834657869745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004166220486203834,
      "loss": 5.5279,
      "mean_token_accuracy": 0.19054403007030488,
      "num_tokens": 56828219.0,
      "step": 29055
    },
    {
      "entropy": 5.866279363632202,
      "epoch": 2.904883290848203,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004165943144042802,
      "loss": 5.4444,
      "mean_token_accuracy": 0.19641445875167846,
      "num_tokens": 56837998.0,
      "step": 29060
    },
    {
      "entropy": 5.986551189422608,
      "epoch": 2.905383115909432,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041656657662570047,
      "loss": 5.5814,
      "mean_token_accuracy": 0.18097781091928483,
      "num_tokens": 56848244.0,
      "step": 29065
    },
    {
      "entropy": 5.957861614227295,
      "epoch": 2.90588294097066,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004165388352853421,
      "loss": 5.4408,
      "mean_token_accuracy": 0.19358172863721848,
      "num_tokens": 56857809.0,
      "step": 29070
    },
    {
      "entropy": 5.84353609085083,
      "epoch": 2.906382766031889,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041651109038390305,
      "loss": 5.3741,
      "mean_token_accuracy": 0.20138475596904754,
      "num_tokens": 56866985.0,
      "step": 29075
    },
    {
      "entropy": 5.853859567642212,
      "epoch": 2.9068825910931175,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041648334192208146,
      "loss": 5.5202,
      "mean_token_accuracy": 0.1851687952876091,
      "num_tokens": 56877402.0,
      "step": 29080
    },
    {
      "entropy": 5.994930362701416,
      "epoch": 2.907382416154346,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041645558990057536,
      "loss": 5.4861,
      "mean_token_accuracy": 0.18521223217248917,
      "num_tokens": 56886322.0,
      "step": 29085
    },
    {
      "entropy": 5.983855390548706,
      "epoch": 2.9078822412155745,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041642783432008294,
      "loss": 5.5336,
      "mean_token_accuracy": 0.18630242198705674,
      "num_tokens": 56896067.0,
      "step": 29090
    },
    {
      "entropy": 5.979978322982788,
      "epoch": 2.908382066276803,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041640007518130274,
      "loss": 5.5707,
      "mean_token_accuracy": 0.18374239951372145,
      "num_tokens": 56904850.0,
      "step": 29095
    },
    {
      "entropy": 5.9207521915435795,
      "epoch": 2.908881891338032,
      "grad_norm": 1.03125,
      "learning_rate": 0.000416372312484933,
      "loss": 5.5259,
      "mean_token_accuracy": 0.18969581127166749,
      "num_tokens": 56915856.0,
      "step": 29100
    },
    {
      "entropy": 5.930894470214843,
      "epoch": 2.90938171639926,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041634454623167217,
      "loss": 5.5134,
      "mean_token_accuracy": 0.19364154934883118,
      "num_tokens": 56925817.0,
      "step": 29105
    },
    {
      "entropy": 5.8832823753356935,
      "epoch": 2.909881541460489,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041631677642221896,
      "loss": 5.4628,
      "mean_token_accuracy": 0.2001198261976242,
      "num_tokens": 56935838.0,
      "step": 29110
    },
    {
      "entropy": 5.934537410736084,
      "epoch": 2.9103813665217175,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000416289003057272,
      "loss": 5.4944,
      "mean_token_accuracy": 0.19467269778251647,
      "num_tokens": 56945926.0,
      "step": 29115
    },
    {
      "entropy": 5.82604022026062,
      "epoch": 2.910881191582946,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004162612261375302,
      "loss": 5.3559,
      "mean_token_accuracy": 0.20788659751415253,
      "num_tokens": 56955639.0,
      "step": 29120
    },
    {
      "entropy": 5.938838052749634,
      "epoch": 2.9113810166441745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041623344566369216,
      "loss": 5.4867,
      "mean_token_accuracy": 0.19258057624101638,
      "num_tokens": 56966653.0,
      "step": 29125
    },
    {
      "entropy": 5.880430078506469,
      "epoch": 2.911880841705403,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000416205661636457,
      "loss": 5.4129,
      "mean_token_accuracy": 0.20130442827939987,
      "num_tokens": 56975797.0,
      "step": 29130
    },
    {
      "entropy": 5.874386978149414,
      "epoch": 2.912380666766632,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004161778740565238,
      "loss": 5.4181,
      "mean_token_accuracy": 0.19588720351457595,
      "num_tokens": 56984994.0,
      "step": 29135
    },
    {
      "entropy": 5.929035425186157,
      "epoch": 2.91288049182786,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041615008292459157,
      "loss": 5.5215,
      "mean_token_accuracy": 0.19105580002069472,
      "num_tokens": 56995630.0,
      "step": 29140
    },
    {
      "entropy": 5.919026851654053,
      "epoch": 2.913380316889089,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041612228824135954,
      "loss": 5.5304,
      "mean_token_accuracy": 0.1867549628019333,
      "num_tokens": 57005864.0,
      "step": 29145
    },
    {
      "entropy": 5.9533792495727536,
      "epoch": 2.9138801419503175,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041609449000752706,
      "loss": 5.5065,
      "mean_token_accuracy": 0.19277382344007493,
      "num_tokens": 57015821.0,
      "step": 29150
    },
    {
      "entropy": 5.928775787353516,
      "epoch": 2.914379967011546,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004160666882237936,
      "loss": 5.4835,
      "mean_token_accuracy": 0.19839630126953126,
      "num_tokens": 57024849.0,
      "step": 29155
    },
    {
      "entropy": 5.862076616287231,
      "epoch": 2.9148797920727745,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004160388828908585,
      "loss": 5.4742,
      "mean_token_accuracy": 0.18989332020282745,
      "num_tokens": 57033754.0,
      "step": 29160
    },
    {
      "entropy": 5.892014694213867,
      "epoch": 2.915379617134003,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041601107400942146,
      "loss": 5.4483,
      "mean_token_accuracy": 0.19064161777496338,
      "num_tokens": 57042880.0,
      "step": 29165
    },
    {
      "entropy": 5.930617475509644,
      "epoch": 2.915879442195232,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041598326158018197,
      "loss": 5.5542,
      "mean_token_accuracy": 0.18401291072368622,
      "num_tokens": 57052717.0,
      "step": 29170
    },
    {
      "entropy": 5.918581247329712,
      "epoch": 2.91637926725646,
      "grad_norm": 1.125,
      "learning_rate": 0.00041595544560384004,
      "loss": 5.4961,
      "mean_token_accuracy": 0.1923505187034607,
      "num_tokens": 57062813.0,
      "step": 29175
    },
    {
      "entropy": 5.967801284790039,
      "epoch": 2.916879092317689,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004159276260810952,
      "loss": 5.4596,
      "mean_token_accuracy": 0.18878760635852815,
      "num_tokens": 57072798.0,
      "step": 29180
    },
    {
      "entropy": 5.932494306564331,
      "epoch": 2.9173789173789175,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041589980301264765,
      "loss": 5.4322,
      "mean_token_accuracy": 0.20074267834424972,
      "num_tokens": 57082114.0,
      "step": 29185
    },
    {
      "entropy": 5.867237377166748,
      "epoch": 2.917878742440146,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004158719763991973,
      "loss": 5.4366,
      "mean_token_accuracy": 0.1977151408791542,
      "num_tokens": 57091603.0,
      "step": 29190
    },
    {
      "entropy": 5.835084915161133,
      "epoch": 2.9183785675013745,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041584414624144426,
      "loss": 5.4752,
      "mean_token_accuracy": 0.18537120521068573,
      "num_tokens": 57101403.0,
      "step": 29195
    },
    {
      "entropy": 5.889000368118286,
      "epoch": 2.918878392562603,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004158163125400887,
      "loss": 5.4311,
      "mean_token_accuracy": 0.1938806504011154,
      "num_tokens": 57109732.0,
      "step": 29200
    },
    {
      "entropy": 5.965185308456421,
      "epoch": 2.919378217623832,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041578847529583106,
      "loss": 5.508,
      "mean_token_accuracy": 0.1915491983294487,
      "num_tokens": 57119289.0,
      "step": 29205
    },
    {
      "entropy": 5.883949565887451,
      "epoch": 2.91987804268506,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004157606345093715,
      "loss": 5.4368,
      "mean_token_accuracy": 0.19517534226179123,
      "num_tokens": 57130243.0,
      "step": 29210
    },
    {
      "entropy": 5.994118404388428,
      "epoch": 2.920377867746289,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004157327901814107,
      "loss": 5.5192,
      "mean_token_accuracy": 0.19056809097528457,
      "num_tokens": 57140885.0,
      "step": 29215
    },
    {
      "entropy": 5.811543703079224,
      "epoch": 2.9208776928075175,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041570494231264906,
      "loss": 5.3974,
      "mean_token_accuracy": 0.19783900380134584,
      "num_tokens": 57150479.0,
      "step": 29220
    },
    {
      "entropy": 5.856419849395752,
      "epoch": 2.9213775178687458,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00041567709090378725,
      "loss": 5.3614,
      "mean_token_accuracy": 0.20006942003965378,
      "num_tokens": 57161561.0,
      "step": 29225
    },
    {
      "entropy": 5.922722673416137,
      "epoch": 2.9218773429299745,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041564923595552604,
      "loss": 5.5571,
      "mean_token_accuracy": 0.19207110852003098,
      "num_tokens": 57171734.0,
      "step": 29230
    },
    {
      "entropy": 5.973996210098266,
      "epoch": 2.922377167991203,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041562137746856625,
      "loss": 5.5951,
      "mean_token_accuracy": 0.18027477860450744,
      "num_tokens": 57182448.0,
      "step": 29235
    },
    {
      "entropy": 5.9162548065185545,
      "epoch": 2.922876993052432,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041559351544360893,
      "loss": 5.5095,
      "mean_token_accuracy": 0.18890593349933624,
      "num_tokens": 57192331.0,
      "step": 29240
    },
    {
      "entropy": 5.766718482971191,
      "epoch": 2.92337681811366,
      "grad_norm": 1.125,
      "learning_rate": 0.00041556564988135474,
      "loss": 5.2784,
      "mean_token_accuracy": 0.20856692492961884,
      "num_tokens": 57201636.0,
      "step": 29245
    },
    {
      "entropy": 5.918631362915039,
      "epoch": 2.923876643174889,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004155377807825051,
      "loss": 5.5058,
      "mean_token_accuracy": 0.19134286791086197,
      "num_tokens": 57210186.0,
      "step": 29250
    },
    {
      "entropy": 5.876980304718018,
      "epoch": 2.9243764682361175,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000415509908147761,
      "loss": 5.4333,
      "mean_token_accuracy": 0.19510093629360198,
      "num_tokens": 57219212.0,
      "step": 29255
    },
    {
      "entropy": 5.913569736480713,
      "epoch": 2.9248762932973458,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041548203197782377,
      "loss": 5.5217,
      "mean_token_accuracy": 0.19214244335889816,
      "num_tokens": 57229728.0,
      "step": 29260
    },
    {
      "entropy": 5.925522947311402,
      "epoch": 2.9253761183585745,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041545415227339474,
      "loss": 5.531,
      "mean_token_accuracy": 0.1869698330760002,
      "num_tokens": 57240233.0,
      "step": 29265
    },
    {
      "entropy": 5.975419998168945,
      "epoch": 2.925875943419803,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004154262690351754,
      "loss": 5.5673,
      "mean_token_accuracy": 0.18594320714473725,
      "num_tokens": 57249502.0,
      "step": 29270
    },
    {
      "entropy": 5.94862208366394,
      "epoch": 2.926375768481032,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004153983822638673,
      "loss": 5.4697,
      "mean_token_accuracy": 0.19468795210123063,
      "num_tokens": 57258825.0,
      "step": 29275
    },
    {
      "entropy": 5.86722502708435,
      "epoch": 2.92687559354226,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041537049196017196,
      "loss": 5.547,
      "mean_token_accuracy": 0.185953226685524,
      "num_tokens": 57269657.0,
      "step": 29280
    },
    {
      "entropy": 5.842076826095581,
      "epoch": 2.927375418603489,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041534259812479125,
      "loss": 5.3652,
      "mean_token_accuracy": 0.20722994059324265,
      "num_tokens": 57280192.0,
      "step": 29285
    },
    {
      "entropy": 5.8129595756530765,
      "epoch": 2.927875243664717,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004153147007584269,
      "loss": 5.3121,
      "mean_token_accuracy": 0.1996684730052948,
      "num_tokens": 57290292.0,
      "step": 29290
    },
    {
      "entropy": 5.9011858940124515,
      "epoch": 2.9283750687259458,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00041528679986178066,
      "loss": 5.4981,
      "mean_token_accuracy": 0.19138263911008835,
      "num_tokens": 57300844.0,
      "step": 29295
    },
    {
      "entropy": 5.871495962142944,
      "epoch": 2.9288748937871745,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041525889543555475,
      "loss": 5.4614,
      "mean_token_accuracy": 0.19168163537979127,
      "num_tokens": 57310162.0,
      "step": 29300
    },
    {
      "entropy": 5.962149953842163,
      "epoch": 2.929374718848403,
      "grad_norm": 1.1875,
      "learning_rate": 0.000415230987480451,
      "loss": 5.5634,
      "mean_token_accuracy": 0.1896727755665779,
      "num_tokens": 57319599.0,
      "step": 29305
    },
    {
      "entropy": 5.946922969818115,
      "epoch": 2.929874543909632,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004152030759971718,
      "loss": 5.585,
      "mean_token_accuracy": 0.1869478553533554,
      "num_tokens": 57329901.0,
      "step": 29310
    },
    {
      "entropy": 5.927590465545654,
      "epoch": 2.93037436897086,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004151751609864193,
      "loss": 5.5649,
      "mean_token_accuracy": 0.1900954931974411,
      "num_tokens": 57339362.0,
      "step": 29315
    },
    {
      "entropy": 5.921946239471436,
      "epoch": 2.930874194032089,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004151472424488957,
      "loss": 5.508,
      "mean_token_accuracy": 0.18802565336227417,
      "num_tokens": 57349527.0,
      "step": 29320
    },
    {
      "entropy": 5.980012226104736,
      "epoch": 2.931374019093317,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004151193203853036,
      "loss": 5.5121,
      "mean_token_accuracy": 0.18293097168207167,
      "num_tokens": 57360021.0,
      "step": 29325
    },
    {
      "entropy": 5.9790778160095215,
      "epoch": 2.9318738441545458,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041509139479634544,
      "loss": 5.5484,
      "mean_token_accuracy": 0.1875520169734955,
      "num_tokens": 57369938.0,
      "step": 29330
    },
    {
      "entropy": 5.893396234512329,
      "epoch": 2.9323736692157745,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041506346568272385,
      "loss": 5.5284,
      "mean_token_accuracy": 0.19472156465053558,
      "num_tokens": 57380399.0,
      "step": 29335
    },
    {
      "entropy": 5.958390569686889,
      "epoch": 2.932873494277003,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041503553304514157,
      "loss": 5.5927,
      "mean_token_accuracy": 0.1838377222418785,
      "num_tokens": 57390639.0,
      "step": 29340
    },
    {
      "entropy": 5.9360757827758786,
      "epoch": 2.933373319338232,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004150075968843012,
      "loss": 5.4572,
      "mean_token_accuracy": 0.20186718702316284,
      "num_tokens": 57401567.0,
      "step": 29345
    },
    {
      "entropy": 5.888880443572998,
      "epoch": 2.93387314439946,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041497965720090573,
      "loss": 5.4438,
      "mean_token_accuracy": 0.1948694810271263,
      "num_tokens": 57410550.0,
      "step": 29350
    },
    {
      "entropy": 5.9306347370147705,
      "epoch": 2.934372969460689,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041495171399565816,
      "loss": 5.5608,
      "mean_token_accuracy": 0.18240692168474198,
      "num_tokens": 57420029.0,
      "step": 29355
    },
    {
      "entropy": 6.056478834152221,
      "epoch": 2.934872794521917,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004149237672692615,
      "loss": 5.6802,
      "mean_token_accuracy": 0.18112274408340454,
      "num_tokens": 57429732.0,
      "step": 29360
    },
    {
      "entropy": 5.960883903503418,
      "epoch": 2.9353726195831458,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004148958170224188,
      "loss": 5.4934,
      "mean_token_accuracy": 0.19202911406755446,
      "num_tokens": 57439469.0,
      "step": 29365
    },
    {
      "entropy": 5.919888305664062,
      "epoch": 2.9358724446443745,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041486786325583343,
      "loss": 5.4686,
      "mean_token_accuracy": 0.19934248626232148,
      "num_tokens": 57450218.0,
      "step": 29370
    },
    {
      "entropy": 6.033197689056396,
      "epoch": 2.936372269705603,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041483990597020854,
      "loss": 5.5795,
      "mean_token_accuracy": 0.18794264346361161,
      "num_tokens": 57459610.0,
      "step": 29375
    },
    {
      "entropy": 6.012611532211304,
      "epoch": 2.936872094766832,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041481194516624763,
      "loss": 5.5778,
      "mean_token_accuracy": 0.18083072900772096,
      "num_tokens": 57468849.0,
      "step": 29380
    },
    {
      "entropy": 5.917841672897339,
      "epoch": 2.93737191982806,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004147839808446541,
      "loss": 5.494,
      "mean_token_accuracy": 0.18760545551776886,
      "num_tokens": 57477827.0,
      "step": 29385
    },
    {
      "entropy": 5.868738126754761,
      "epoch": 2.937871744889289,
      "grad_norm": 1.125,
      "learning_rate": 0.00041475601300613166,
      "loss": 5.4758,
      "mean_token_accuracy": 0.1968924418091774,
      "num_tokens": 57487502.0,
      "step": 29390
    },
    {
      "entropy": 5.927325010299683,
      "epoch": 2.938371569950517,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041472804165138384,
      "loss": 5.4832,
      "mean_token_accuracy": 0.19037804156541824,
      "num_tokens": 57498281.0,
      "step": 29395
    },
    {
      "entropy": 5.917412853240966,
      "epoch": 2.9388713950117458,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004147000667811145,
      "loss": 5.5394,
      "mean_token_accuracy": 0.19106678068637847,
      "num_tokens": 57509612.0,
      "step": 29400
    },
    {
      "entropy": 5.9369011402130125,
      "epoch": 2.9393712200729745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004146720883960273,
      "loss": 5.526,
      "mean_token_accuracy": 0.18754567354917526,
      "num_tokens": 57519211.0,
      "step": 29405
    },
    {
      "entropy": 5.918987703323364,
      "epoch": 2.939871045134203,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004146441064968264,
      "loss": 5.4769,
      "mean_token_accuracy": 0.18867489099502563,
      "num_tokens": 57528756.0,
      "step": 29410
    },
    {
      "entropy": 5.92028489112854,
      "epoch": 2.940370870195432,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041461612108421576,
      "loss": 5.5399,
      "mean_token_accuracy": 0.1808752492070198,
      "num_tokens": 57537892.0,
      "step": 29415
    },
    {
      "entropy": 5.920268869400024,
      "epoch": 2.94087069525666,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041458813215889935,
      "loss": 5.566,
      "mean_token_accuracy": 0.18077134490013122,
      "num_tokens": 57547650.0,
      "step": 29420
    },
    {
      "entropy": 5.879950714111328,
      "epoch": 2.941370520317889,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004145601397215815,
      "loss": 5.4795,
      "mean_token_accuracy": 0.19414965361356734,
      "num_tokens": 57557077.0,
      "step": 29425
    },
    {
      "entropy": 5.996497058868409,
      "epoch": 2.941870345379117,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041453214377296636,
      "loss": 5.5767,
      "mean_token_accuracy": 0.1901779368519783,
      "num_tokens": 57566402.0,
      "step": 29430
    },
    {
      "entropy": 5.941397380828858,
      "epoch": 2.9423701704403458,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004145041443137585,
      "loss": 5.5247,
      "mean_token_accuracy": 0.18602881580591202,
      "num_tokens": 57577093.0,
      "step": 29435
    },
    {
      "entropy": 5.944294500350952,
      "epoch": 2.9428699955015745,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004144761413446622,
      "loss": 5.4658,
      "mean_token_accuracy": 0.19287138283252717,
      "num_tokens": 57586790.0,
      "step": 29440
    },
    {
      "entropy": 5.878828477859497,
      "epoch": 2.943369820562803,
      "grad_norm": 1.125,
      "learning_rate": 0.00041444813486638215,
      "loss": 5.4223,
      "mean_token_accuracy": 0.1949970081448555,
      "num_tokens": 57595615.0,
      "step": 29445
    },
    {
      "entropy": 5.908865308761596,
      "epoch": 2.943869645624032,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041442012487962284,
      "loss": 5.4853,
      "mean_token_accuracy": 0.19153160452842713,
      "num_tokens": 57605436.0,
      "step": 29450
    },
    {
      "entropy": 5.992946720123291,
      "epoch": 2.94436947068526,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004143921113850891,
      "loss": 5.5612,
      "mean_token_accuracy": 0.18348506838083267,
      "num_tokens": 57614286.0,
      "step": 29455
    },
    {
      "entropy": 5.816237735748291,
      "epoch": 2.944869295746489,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041436409438348564,
      "loss": 5.3981,
      "mean_token_accuracy": 0.19862426370382308,
      "num_tokens": 57624591.0,
      "step": 29460
    },
    {
      "entropy": 5.895674705505371,
      "epoch": 2.945369120807717,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004143360738755175,
      "loss": 5.4549,
      "mean_token_accuracy": 0.19689897894859315,
      "num_tokens": 57633987.0,
      "step": 29465
    },
    {
      "entropy": 5.961906814575196,
      "epoch": 2.9458689458689458,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004143080498618895,
      "loss": 5.5514,
      "mean_token_accuracy": 0.1843030646443367,
      "num_tokens": 57643989.0,
      "step": 29470
    },
    {
      "entropy": 5.890832710266113,
      "epoch": 2.9463687709301745,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041428002234330694,
      "loss": 5.4856,
      "mean_token_accuracy": 0.18770056068897248,
      "num_tokens": 57653860.0,
      "step": 29475
    },
    {
      "entropy": 5.861746406555175,
      "epoch": 2.946868595991403,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004142519913204748,
      "loss": 5.4211,
      "mean_token_accuracy": 0.20266856700181962,
      "num_tokens": 57663245.0,
      "step": 29480
    },
    {
      "entropy": 5.907408857345581,
      "epoch": 2.9473684210526314,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041422395679409834,
      "loss": 5.4188,
      "mean_token_accuracy": 0.1942606672644615,
      "num_tokens": 57672624.0,
      "step": 29485
    },
    {
      "entropy": 5.858806180953979,
      "epoch": 2.94786824611386,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041419591876488304,
      "loss": 5.4926,
      "mean_token_accuracy": 0.19106595665216447,
      "num_tokens": 57683278.0,
      "step": 29490
    },
    {
      "entropy": 5.927419424057007,
      "epoch": 2.948368071175089,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004141678772335342,
      "loss": 5.459,
      "mean_token_accuracy": 0.19249752908945084,
      "num_tokens": 57692252.0,
      "step": 29495
    },
    {
      "entropy": 5.9781928062438965,
      "epoch": 2.948867896236317,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041413983220075736,
      "loss": 5.4927,
      "mean_token_accuracy": 0.1861422434449196,
      "num_tokens": 57703154.0,
      "step": 29500
    },
    {
      "entropy": 6.037700891494751,
      "epoch": 2.9493677212975458,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041411178366725814,
      "loss": 5.6819,
      "mean_token_accuracy": 0.17298830300569534,
      "num_tokens": 57714526.0,
      "step": 29505
    },
    {
      "entropy": 5.915503931045532,
      "epoch": 2.9498675463587745,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004140837316337423,
      "loss": 5.4739,
      "mean_token_accuracy": 0.19330107271671296,
      "num_tokens": 57724143.0,
      "step": 29510
    },
    {
      "entropy": 5.996884775161743,
      "epoch": 2.950367371420003,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004140556761009155,
      "loss": 5.6437,
      "mean_token_accuracy": 0.17951433658599852,
      "num_tokens": 57732772.0,
      "step": 29515
    },
    {
      "entropy": 5.894446945190429,
      "epoch": 2.9508671964812314,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004140276170694837,
      "loss": 5.481,
      "mean_token_accuracy": 0.19050192087888718,
      "num_tokens": 57742200.0,
      "step": 29520
    },
    {
      "entropy": 5.965362930297852,
      "epoch": 2.95136702154246,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041399955454015286,
      "loss": 5.5606,
      "mean_token_accuracy": 0.18587136268615723,
      "num_tokens": 57752477.0,
      "step": 29525
    },
    {
      "entropy": 5.932660722732544,
      "epoch": 2.951866846603689,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041397148851362903,
      "loss": 5.4887,
      "mean_token_accuracy": 0.19284258484840394,
      "num_tokens": 57761204.0,
      "step": 29530
    },
    {
      "entropy": 5.815188980102539,
      "epoch": 2.952366671664917,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041394341899061814,
      "loss": 5.3556,
      "mean_token_accuracy": 0.20124818831682206,
      "num_tokens": 57771222.0,
      "step": 29535
    },
    {
      "entropy": 5.938890838623047,
      "epoch": 2.9528664967261458,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004139153459718267,
      "loss": 5.5998,
      "mean_token_accuracy": 0.18038953542709352,
      "num_tokens": 57781200.0,
      "step": 29540
    },
    {
      "entropy": 5.808305931091309,
      "epoch": 2.9533663217873745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000413887269457961,
      "loss": 5.3376,
      "mean_token_accuracy": 0.20516733974218368,
      "num_tokens": 57790139.0,
      "step": 29545
    },
    {
      "entropy": 5.901345443725586,
      "epoch": 2.953866146848603,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041385918944972716,
      "loss": 5.5083,
      "mean_token_accuracy": 0.18750488758087158,
      "num_tokens": 57800156.0,
      "step": 29550
    },
    {
      "entropy": 5.994382905960083,
      "epoch": 2.9543659719098314,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041383110594783197,
      "loss": 5.5475,
      "mean_token_accuracy": 0.1923950657248497,
      "num_tokens": 57810344.0,
      "step": 29555
    },
    {
      "entropy": 5.914741039276123,
      "epoch": 2.95486579697106,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041380301895298173,
      "loss": 5.4451,
      "mean_token_accuracy": 0.19258186966180801,
      "num_tokens": 57820382.0,
      "step": 29560
    },
    {
      "entropy": 5.8900731086730955,
      "epoch": 2.955365622032289,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041377492846588333,
      "loss": 5.5717,
      "mean_token_accuracy": 0.18620391488075255,
      "num_tokens": 57831250.0,
      "step": 29565
    },
    {
      "entropy": 5.870552492141724,
      "epoch": 2.955865447093517,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041374683448724346,
      "loss": 5.5004,
      "mean_token_accuracy": 0.19536670595407485,
      "num_tokens": 57840134.0,
      "step": 29570
    },
    {
      "entropy": 6.016626787185669,
      "epoch": 2.9563652721547458,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004137187370177689,
      "loss": 5.531,
      "mean_token_accuracy": 0.18726823031902312,
      "num_tokens": 57850147.0,
      "step": 29575
    },
    {
      "entropy": 5.897124290466309,
      "epoch": 2.9568650972159745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041369063605816667,
      "loss": 5.4381,
      "mean_token_accuracy": 0.2021230176091194,
      "num_tokens": 57859516.0,
      "step": 29580
    },
    {
      "entropy": 5.823148488998413,
      "epoch": 2.957364922277203,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004136625316091437,
      "loss": 5.5169,
      "mean_token_accuracy": 0.18574697226285936,
      "num_tokens": 57869858.0,
      "step": 29585
    },
    {
      "entropy": 5.942329406738281,
      "epoch": 2.9578647473384314,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004136344236714071,
      "loss": 5.5902,
      "mean_token_accuracy": 0.1825385719537735,
      "num_tokens": 57879804.0,
      "step": 29590
    },
    {
      "entropy": 6.010490417480469,
      "epoch": 2.95836457239966,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041360631224566407,
      "loss": 5.4749,
      "mean_token_accuracy": 0.203529654443264,
      "num_tokens": 57889936.0,
      "step": 29595
    },
    {
      "entropy": 5.911497783660889,
      "epoch": 2.958864397460889,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004135781973326219,
      "loss": 5.4253,
      "mean_token_accuracy": 0.19629135131835937,
      "num_tokens": 57899230.0,
      "step": 29600
    },
    {
      "entropy": 5.9179362773895265,
      "epoch": 2.959364222522117,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041355007893298796,
      "loss": 5.5628,
      "mean_token_accuracy": 0.18396355211734772,
      "num_tokens": 57909741.0,
      "step": 29605
    },
    {
      "entropy": 5.897365665435791,
      "epoch": 2.9598640475833458,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004135219570474697,
      "loss": 5.5009,
      "mean_token_accuracy": 0.19038580507040023,
      "num_tokens": 57918558.0,
      "step": 29610
    },
    {
      "entropy": 5.809850597381592,
      "epoch": 2.9603638726445745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041349383167677456,
      "loss": 5.3205,
      "mean_token_accuracy": 0.20552463978528976,
      "num_tokens": 57928868.0,
      "step": 29615
    },
    {
      "entropy": 5.91221604347229,
      "epoch": 2.960863697705803,
      "grad_norm": 1.125,
      "learning_rate": 0.00041346570282161033,
      "loss": 5.441,
      "mean_token_accuracy": 0.19446675777435302,
      "num_tokens": 57937890.0,
      "step": 29620
    },
    {
      "entropy": 5.900111246109009,
      "epoch": 2.9613635227670314,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041343757048268464,
      "loss": 5.5373,
      "mean_token_accuracy": 0.18581462651491165,
      "num_tokens": 57946340.0,
      "step": 29625
    },
    {
      "entropy": 6.007351446151733,
      "epoch": 2.96186334782826,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004134094346607053,
      "loss": 5.6036,
      "mean_token_accuracy": 0.18115442246198654,
      "num_tokens": 57955968.0,
      "step": 29630
    },
    {
      "entropy": 5.915469455718994,
      "epoch": 2.962363172889489,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004133812953563802,
      "loss": 5.4194,
      "mean_token_accuracy": 0.19831952899694444,
      "num_tokens": 57965345.0,
      "step": 29635
    },
    {
      "entropy": 5.901114082336425,
      "epoch": 2.962862997950717,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041335315257041733,
      "loss": 5.4687,
      "mean_token_accuracy": 0.19654759764671326,
      "num_tokens": 57974652.0,
      "step": 29640
    },
    {
      "entropy": 5.904061889648437,
      "epoch": 2.9633628230119458,
      "grad_norm": 1.25,
      "learning_rate": 0.00041332500630352463,
      "loss": 5.5409,
      "mean_token_accuracy": 0.18690589517354966,
      "num_tokens": 57983845.0,
      "step": 29645
    },
    {
      "entropy": 5.900082731246949,
      "epoch": 2.9638626480731745,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004132968565564105,
      "loss": 5.5253,
      "mean_token_accuracy": 0.19406110644340516,
      "num_tokens": 57993455.0,
      "step": 29650
    },
    {
      "entropy": 5.9633222103118895,
      "epoch": 2.964362473134403,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000413268703329783,
      "loss": 5.4956,
      "mean_token_accuracy": 0.18038306385278702,
      "num_tokens": 58003983.0,
      "step": 29655
    },
    {
      "entropy": 5.941501903533935,
      "epoch": 2.9648622981956314,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004132405466243505,
      "loss": 5.4849,
      "mean_token_accuracy": 0.19048597365617753,
      "num_tokens": 58014486.0,
      "step": 29660
    },
    {
      "entropy": 5.947859573364258,
      "epoch": 2.96536212325686,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041321238644082144,
      "loss": 5.4014,
      "mean_token_accuracy": 0.1976131170988083,
      "num_tokens": 58023610.0,
      "step": 29665
    },
    {
      "entropy": 5.963757467269898,
      "epoch": 2.965861948318089,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041318422277990426,
      "loss": 5.6201,
      "mean_token_accuracy": 0.18893399238586425,
      "num_tokens": 58033829.0,
      "step": 29670
    },
    {
      "entropy": 5.85537281036377,
      "epoch": 2.966361773379317,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004131560556423076,
      "loss": 5.4808,
      "mean_token_accuracy": 0.18447321951389312,
      "num_tokens": 58042716.0,
      "step": 29675
    },
    {
      "entropy": 5.892200756072998,
      "epoch": 2.9668615984405458,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004131278850287401,
      "loss": 5.464,
      "mean_token_accuracy": 0.19222662150859832,
      "num_tokens": 58052667.0,
      "step": 29680
    },
    {
      "entropy": 5.938839483261108,
      "epoch": 2.9673614235017745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041309971093991067,
      "loss": 5.5359,
      "mean_token_accuracy": 0.1821529433131218,
      "num_tokens": 58061690.0,
      "step": 29685
    },
    {
      "entropy": 5.969122457504272,
      "epoch": 2.967861248563003,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041307153337652805,
      "loss": 5.5171,
      "mean_token_accuracy": 0.18236326426267624,
      "num_tokens": 58071212.0,
      "step": 29690
    },
    {
      "entropy": 5.894996643066406,
      "epoch": 2.9683610736242314,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000413043352339301,
      "loss": 5.4657,
      "mean_token_accuracy": 0.18884163945913315,
      "num_tokens": 58081227.0,
      "step": 29695
    },
    {
      "entropy": 5.936381578445435,
      "epoch": 2.96886089868546,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004130151678289388,
      "loss": 5.5337,
      "mean_token_accuracy": 0.18484532833099365,
      "num_tokens": 58091303.0,
      "step": 29700
    },
    {
      "entropy": 5.882755851745605,
      "epoch": 2.969360723746689,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004129869798461506,
      "loss": 5.3998,
      "mean_token_accuracy": 0.20020896643400193,
      "num_tokens": 58100124.0,
      "step": 29705
    },
    {
      "entropy": 5.850013494491577,
      "epoch": 2.969860548807917,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004129587883916454,
      "loss": 5.533,
      "mean_token_accuracy": 0.19058459401130676,
      "num_tokens": 58110326.0,
      "step": 29710
    },
    {
      "entropy": 5.8762492656707765,
      "epoch": 2.9703603738691458,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004129305934661326,
      "loss": 5.5123,
      "mean_token_accuracy": 0.1908722013235092,
      "num_tokens": 58119465.0,
      "step": 29715
    },
    {
      "entropy": 5.911401319503784,
      "epoch": 2.9708601989303745,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041290239507032155,
      "loss": 5.5805,
      "mean_token_accuracy": 0.18580035269260406,
      "num_tokens": 58130136.0,
      "step": 29720
    },
    {
      "entropy": 5.9019781112670895,
      "epoch": 2.971360023991603,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004128741932049218,
      "loss": 5.4169,
      "mean_token_accuracy": 0.19447132647037507,
      "num_tokens": 58139725.0,
      "step": 29725
    },
    {
      "entropy": 5.831573486328125,
      "epoch": 2.9718598490528314,
      "grad_norm": 1.125,
      "learning_rate": 0.0004128459878706428,
      "loss": 5.3998,
      "mean_token_accuracy": 0.20414263904094695,
      "num_tokens": 58149060.0,
      "step": 29730
    },
    {
      "entropy": 5.858699369430542,
      "epoch": 2.97235967411406,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004128177790681942,
      "loss": 5.5037,
      "mean_token_accuracy": 0.19039321392774583,
      "num_tokens": 58159627.0,
      "step": 29735
    },
    {
      "entropy": 5.915188646316528,
      "epoch": 2.972859499175289,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041278956679828566,
      "loss": 5.4018,
      "mean_token_accuracy": 0.19814104586839676,
      "num_tokens": 58169546.0,
      "step": 29740
    },
    {
      "entropy": 5.867446041107177,
      "epoch": 2.973359324236517,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041276135106162713,
      "loss": 5.4366,
      "mean_token_accuracy": 0.19571750462055207,
      "num_tokens": 58179521.0,
      "step": 29745
    },
    {
      "entropy": 5.899530601501465,
      "epoch": 2.9738591492977458,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004127331318589285,
      "loss": 5.4873,
      "mean_token_accuracy": 0.18980502635240554,
      "num_tokens": 58188975.0,
      "step": 29750
    },
    {
      "entropy": 5.932988166809082,
      "epoch": 2.9743589743589745,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041270490919089966,
      "loss": 5.5242,
      "mean_token_accuracy": 0.1853761225938797,
      "num_tokens": 58198692.0,
      "step": 29755
    },
    {
      "entropy": 6.01269679069519,
      "epoch": 2.974858799420203,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004126766830582508,
      "loss": 5.6256,
      "mean_token_accuracy": 0.1778615638613701,
      "num_tokens": 58208441.0,
      "step": 29760
    },
    {
      "entropy": 5.934330129623413,
      "epoch": 2.9753586244814314,
      "grad_norm": 1.078125,
      "learning_rate": 0.000412648453461692,
      "loss": 5.5656,
      "mean_token_accuracy": 0.19120707213878632,
      "num_tokens": 58218750.0,
      "step": 29765
    },
    {
      "entropy": 5.83009934425354,
      "epoch": 2.97585844954266,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041262022040193346,
      "loss": 5.4141,
      "mean_token_accuracy": 0.1952348232269287,
      "num_tokens": 58229445.0,
      "step": 29770
    },
    {
      "entropy": 5.971219110488891,
      "epoch": 2.976358274603889,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041259198387968564,
      "loss": 5.5816,
      "mean_token_accuracy": 0.18914134800434113,
      "num_tokens": 58240012.0,
      "step": 29775
    },
    {
      "entropy": 6.0030169010162355,
      "epoch": 2.976858099665117,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041256374389565895,
      "loss": 5.6012,
      "mean_token_accuracy": 0.17911868095397948,
      "num_tokens": 58250303.0,
      "step": 29780
    },
    {
      "entropy": 5.85063533782959,
      "epoch": 2.9773579247263458,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041253550045056375,
      "loss": 5.3371,
      "mean_token_accuracy": 0.2091983050107956,
      "num_tokens": 58259376.0,
      "step": 29785
    },
    {
      "entropy": 5.9336751937866214,
      "epoch": 2.9778577497875744,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041250725354511085,
      "loss": 5.6028,
      "mean_token_accuracy": 0.1826466664671898,
      "num_tokens": 58269620.0,
      "step": 29790
    },
    {
      "entropy": 5.8695659160614015,
      "epoch": 2.978357574848803,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004124790031800109,
      "loss": 5.4248,
      "mean_token_accuracy": 0.20369562804698943,
      "num_tokens": 58280121.0,
      "step": 29795
    },
    {
      "entropy": 5.963361406326294,
      "epoch": 2.9788573999100314,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004124507493559745,
      "loss": 5.6112,
      "mean_token_accuracy": 0.1789127916097641,
      "num_tokens": 58289944.0,
      "step": 29800
    },
    {
      "entropy": 5.9962646484375,
      "epoch": 2.97935722497126,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004124224920737126,
      "loss": 5.5879,
      "mean_token_accuracy": 0.18017932772636414,
      "num_tokens": 58300091.0,
      "step": 29805
    },
    {
      "entropy": 5.82749490737915,
      "epoch": 2.9798570500324884,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004123942313339361,
      "loss": 5.3196,
      "mean_token_accuracy": 0.2065373957157135,
      "num_tokens": 58309249.0,
      "step": 29810
    },
    {
      "entropy": 5.952248287200928,
      "epoch": 2.980356875093717,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041236596713735624,
      "loss": 5.5958,
      "mean_token_accuracy": 0.1846255272626877,
      "num_tokens": 58319238.0,
      "step": 29815
    },
    {
      "entropy": 5.944937753677368,
      "epoch": 2.9808567001549457,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004123376994846839,
      "loss": 5.515,
      "mean_token_accuracy": 0.18536411076784134,
      "num_tokens": 58329091.0,
      "step": 29820
    },
    {
      "entropy": 5.9501207828521725,
      "epoch": 2.9813565252161744,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004123094283766304,
      "loss": 5.4584,
      "mean_token_accuracy": 0.19843049496412277,
      "num_tokens": 58338581.0,
      "step": 29825
    },
    {
      "entropy": 5.97106614112854,
      "epoch": 2.981856350277403,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041228115381390705,
      "loss": 5.5517,
      "mean_token_accuracy": 0.1902434155344963,
      "num_tokens": 58347890.0,
      "step": 29830
    },
    {
      "entropy": 5.944241571426391,
      "epoch": 2.9823561753386314,
      "grad_norm": 1.125,
      "learning_rate": 0.00041225287579722526,
      "loss": 5.5889,
      "mean_token_accuracy": 0.17888880670070648,
      "num_tokens": 58357910.0,
      "step": 29835
    },
    {
      "entropy": 5.986602544784546,
      "epoch": 2.98285600039986,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004122245943272963,
      "loss": 5.5655,
      "mean_token_accuracy": 0.18525317013263704,
      "num_tokens": 58367819.0,
      "step": 29840
    },
    {
      "entropy": 5.927380132675171,
      "epoch": 2.9833558254610884,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041219630940483187,
      "loss": 5.5088,
      "mean_token_accuracy": 0.18443989455699922,
      "num_tokens": 58377324.0,
      "step": 29845
    },
    {
      "entropy": 5.931207895278931,
      "epoch": 2.983855650522317,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004121680210305436,
      "loss": 5.4945,
      "mean_token_accuracy": 0.19041841626167297,
      "num_tokens": 58386018.0,
      "step": 29850
    },
    {
      "entropy": 5.94778003692627,
      "epoch": 2.9843554755835457,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004121397292051433,
      "loss": 5.4448,
      "mean_token_accuracy": 0.19560083895921707,
      "num_tokens": 58396049.0,
      "step": 29855
    },
    {
      "entropy": 5.999994897842408,
      "epoch": 2.9848553006447744,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004121114339293427,
      "loss": 5.5281,
      "mean_token_accuracy": 0.18600770682096482,
      "num_tokens": 58407443.0,
      "step": 29860
    },
    {
      "entropy": 5.741796588897705,
      "epoch": 2.985355125706003,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004120831352038537,
      "loss": 5.3108,
      "mean_token_accuracy": 0.21345626264810563,
      "num_tokens": 58417466.0,
      "step": 29865
    },
    {
      "entropy": 5.943235111236572,
      "epoch": 2.9858549507672314,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004120548330293882,
      "loss": 5.4726,
      "mean_token_accuracy": 0.1924295425415039,
      "num_tokens": 58426819.0,
      "step": 29870
    },
    {
      "entropy": 5.882769250869751,
      "epoch": 2.98635477582846,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041202652740665845,
      "loss": 5.4897,
      "mean_token_accuracy": 0.19163150191307068,
      "num_tokens": 58436470.0,
      "step": 29875
    },
    {
      "entropy": 5.960244274139404,
      "epoch": 2.9868546008896883,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004119982183363765,
      "loss": 5.4849,
      "mean_token_accuracy": 0.19153866916894913,
      "num_tokens": 58448095.0,
      "step": 29880
    },
    {
      "entropy": 5.91520094871521,
      "epoch": 2.987354425950917,
      "grad_norm": 1.125,
      "learning_rate": 0.00041196990581925466,
      "loss": 5.4859,
      "mean_token_accuracy": 0.19610942751169205,
      "num_tokens": 58457162.0,
      "step": 29885
    },
    {
      "entropy": 5.982629108428955,
      "epoch": 2.9878542510121457,
      "grad_norm": 1.125,
      "learning_rate": 0.0004119415898560052,
      "loss": 5.6215,
      "mean_token_accuracy": 0.18998941481113435,
      "num_tokens": 58467763.0,
      "step": 29890
    },
    {
      "entropy": 6.003442621231079,
      "epoch": 2.9883540760733744,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004119132704473405,
      "loss": 5.5198,
      "mean_token_accuracy": 0.18299174606800078,
      "num_tokens": 58477769.0,
      "step": 29895
    },
    {
      "entropy": 5.949984073638916,
      "epoch": 2.988853901134603,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004118849475939733,
      "loss": 5.5234,
      "mean_token_accuracy": 0.18283967524766923,
      "num_tokens": 58486656.0,
      "step": 29900
    },
    {
      "entropy": 5.954256153106689,
      "epoch": 2.9893537261958314,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041185662129661593,
      "loss": 5.5735,
      "mean_token_accuracy": 0.18470313400030136,
      "num_tokens": 58496979.0,
      "step": 29905
    },
    {
      "entropy": 5.967100000381469,
      "epoch": 2.98985355125706,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004118282915559813,
      "loss": 5.5701,
      "mean_token_accuracy": 0.19455881416797638,
      "num_tokens": 58507340.0,
      "step": 29910
    },
    {
      "entropy": 5.905768013000488,
      "epoch": 2.9903533763182883,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004117999583727819,
      "loss": 5.5076,
      "mean_token_accuracy": 0.18942248225212097,
      "num_tokens": 58517364.0,
      "step": 29915
    },
    {
      "entropy": 5.947090196609497,
      "epoch": 2.990853201379517,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041177162174773084,
      "loss": 5.4385,
      "mean_token_accuracy": 0.1960573524236679,
      "num_tokens": 58526435.0,
      "step": 29920
    },
    {
      "entropy": 5.860172843933105,
      "epoch": 2.9913530264407457,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041174328168154094,
      "loss": 5.4558,
      "mean_token_accuracy": 0.19799811989068986,
      "num_tokens": 58536694.0,
      "step": 29925
    },
    {
      "entropy": 5.899044799804687,
      "epoch": 2.9918528515019744,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004117149381749253,
      "loss": 5.483,
      "mean_token_accuracy": 0.19568471163511275,
      "num_tokens": 58546854.0,
      "step": 29930
    },
    {
      "entropy": 5.901590967178345,
      "epoch": 2.992352676563203,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004116865912285971,
      "loss": 5.4997,
      "mean_token_accuracy": 0.18962839543819426,
      "num_tokens": 58557936.0,
      "step": 29935
    },
    {
      "entropy": 5.9425664901733395,
      "epoch": 2.9928525016244314,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041165824084326923,
      "loss": 5.592,
      "mean_token_accuracy": 0.1833338811993599,
      "num_tokens": 58568107.0,
      "step": 29940
    },
    {
      "entropy": 5.916060304641723,
      "epoch": 2.99335232668566,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004116298870196553,
      "loss": 5.3867,
      "mean_token_accuracy": 0.19967970997095108,
      "num_tokens": 58576756.0,
      "step": 29945
    },
    {
      "entropy": 5.890544319152832,
      "epoch": 2.9938521517468883,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041160152975846853,
      "loss": 5.4854,
      "mean_token_accuracy": 0.19440787136554719,
      "num_tokens": 58586894.0,
      "step": 29950
    },
    {
      "entropy": 5.920233488082886,
      "epoch": 2.994351976808117,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004115731690604225,
      "loss": 5.4633,
      "mean_token_accuracy": 0.1871543914079666,
      "num_tokens": 58596720.0,
      "step": 29955
    },
    {
      "entropy": 5.939529895782471,
      "epoch": 2.9948518018693457,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004115448049262306,
      "loss": 5.5116,
      "mean_token_accuracy": 0.18434690833091735,
      "num_tokens": 58605705.0,
      "step": 29960
    },
    {
      "entropy": 6.009579420089722,
      "epoch": 2.9953516269305744,
      "grad_norm": 1.25,
      "learning_rate": 0.0004115164373566065,
      "loss": 5.6571,
      "mean_token_accuracy": 0.18188927173614503,
      "num_tokens": 58615669.0,
      "step": 29965
    },
    {
      "entropy": 5.9736244678497314,
      "epoch": 2.995851451991803,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041148806635226407,
      "loss": 5.5472,
      "mean_token_accuracy": 0.18812862634658814,
      "num_tokens": 58625900.0,
      "step": 29970
    },
    {
      "entropy": 5.860826444625855,
      "epoch": 2.9963512770530314,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000411459691913917,
      "loss": 5.5687,
      "mean_token_accuracy": 0.18707561492919922,
      "num_tokens": 58637323.0,
      "step": 29975
    },
    {
      "entropy": 5.952162408828736,
      "epoch": 2.99685110211426,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041143131404227913,
      "loss": 5.4616,
      "mean_token_accuracy": 0.19492146372795105,
      "num_tokens": 58646395.0,
      "step": 29980
    },
    {
      "entropy": 5.919927263259888,
      "epoch": 2.9973509271754883,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004114029327380645,
      "loss": 5.4192,
      "mean_token_accuracy": 0.19282662123441696,
      "num_tokens": 58655966.0,
      "step": 29985
    },
    {
      "entropy": 5.908976459503174,
      "epoch": 2.997850752236717,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004113745480019872,
      "loss": 5.3798,
      "mean_token_accuracy": 0.20162972509860994,
      "num_tokens": 58666030.0,
      "step": 29990
    },
    {
      "entropy": 5.886649942398071,
      "epoch": 2.9983505772979457,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004113461598347614,
      "loss": 5.4181,
      "mean_token_accuracy": 0.19397590607404708,
      "num_tokens": 58674702.0,
      "step": 29995
    },
    {
      "entropy": 5.935790061950684,
      "epoch": 2.9988504023591744,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041131776823710134,
      "loss": 5.5764,
      "mean_token_accuracy": 0.1833466485142708,
      "num_tokens": 58686103.0,
      "step": 30000
    },
    {
      "epoch": 2.9988504023591744,
      "eval_entropy": 5.761890372797497,
      "eval_loss": 5.597193717956543,
      "eval_mean_token_accuracy": 0.19447231534422135,
      "eval_num_tokens": 58686103.0,
      "eval_runtime": 23.8155,
      "eval_samples_per_second": 1303.564,
      "eval_steps_per_second": 162.961,
      "step": 30000
    },
    {
      "entropy": 5.951220417022705,
      "epoch": 2.9993502274204027,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004112893732097213,
      "loss": 5.4608,
      "mean_token_accuracy": 0.19434579759836196,
      "num_tokens": 58695173.0,
      "step": 30005
    },
    {
      "entropy": 5.961740112304687,
      "epoch": 2.9998500524816314,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004112609747533356,
      "loss": 5.5693,
      "mean_token_accuracy": 0.18830457627773284,
      "num_tokens": 58705014.0,
      "step": 30010
    },
    {
      "entropy": 5.928573502434625,
      "epoch": 3.000299895036737,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004112325728686589,
      "loss": 5.4349,
      "mean_token_accuracy": 0.20432726542154947,
      "num_tokens": 58714656.0,
      "step": 30015
    },
    {
      "entropy": 6.0063416957855225,
      "epoch": 3.000799720097966,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041120416755640574,
      "loss": 5.3526,
      "mean_token_accuracy": 0.19647039622068405,
      "num_tokens": 58725157.0,
      "step": 30020
    },
    {
      "entropy": 5.941908502578736,
      "epoch": 3.001299545159194,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041117575881729075,
      "loss": 5.4866,
      "mean_token_accuracy": 0.18696325421333312,
      "num_tokens": 58734229.0,
      "step": 30025
    },
    {
      "entropy": 5.825024938583374,
      "epoch": 3.001799370220423,
      "grad_norm": 1.125,
      "learning_rate": 0.00041114734665202875,
      "loss": 5.2916,
      "mean_token_accuracy": 0.20325307101011275,
      "num_tokens": 58743374.0,
      "step": 30030
    },
    {
      "entropy": 5.891180419921875,
      "epoch": 3.0022991952816516,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004111189310613344,
      "loss": 5.3131,
      "mean_token_accuracy": 0.20221786499023436,
      "num_tokens": 58752846.0,
      "step": 30035
    },
    {
      "entropy": 5.94706768989563,
      "epoch": 3.00279902034288,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041109051204592294,
      "loss": 5.4323,
      "mean_token_accuracy": 0.1927326127886772,
      "num_tokens": 58763407.0,
      "step": 30040
    },
    {
      "entropy": 5.92323751449585,
      "epoch": 3.0032988454041085,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041106208960650913,
      "loss": 5.379,
      "mean_token_accuracy": 0.19480303972959517,
      "num_tokens": 58774207.0,
      "step": 30045
    },
    {
      "entropy": 5.869116926193238,
      "epoch": 3.003798670465337,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041103366374380817,
      "loss": 5.4048,
      "mean_token_accuracy": 0.19529158025979995,
      "num_tokens": 58783270.0,
      "step": 30050
    },
    {
      "entropy": 5.869815492630005,
      "epoch": 3.004298495526566,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041100523445853525,
      "loss": 5.399,
      "mean_token_accuracy": 0.20184941440820695,
      "num_tokens": 58793007.0,
      "step": 30055
    },
    {
      "entropy": 5.9986954689025875,
      "epoch": 3.004798320587794,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004109768017514056,
      "loss": 5.5214,
      "mean_token_accuracy": 0.1882593020796776,
      "num_tokens": 58802604.0,
      "step": 30060
    },
    {
      "entropy": 5.94795069694519,
      "epoch": 3.005298145649023,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004109483656231346,
      "loss": 5.4309,
      "mean_token_accuracy": 0.19476745128631592,
      "num_tokens": 58811739.0,
      "step": 30065
    },
    {
      "entropy": 5.871524238586426,
      "epoch": 3.0057979707102516,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041091992607443777,
      "loss": 5.4439,
      "mean_token_accuracy": 0.19649945199489594,
      "num_tokens": 58821079.0,
      "step": 30070
    },
    {
      "entropy": 5.999233198165894,
      "epoch": 3.00629779577148,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041089148310603053,
      "loss": 5.5122,
      "mean_token_accuracy": 0.18455551117658614,
      "num_tokens": 58831030.0,
      "step": 30075
    },
    {
      "entropy": 5.991444301605225,
      "epoch": 3.0067976208327085,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004108630367186285,
      "loss": 5.3859,
      "mean_token_accuracy": 0.20266994088888168,
      "num_tokens": 58840241.0,
      "step": 30080
    },
    {
      "entropy": 5.937366628646851,
      "epoch": 3.007297445893937,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041083458691294755,
      "loss": 5.46,
      "mean_token_accuracy": 0.1963067203760147,
      "num_tokens": 58849411.0,
      "step": 30085
    },
    {
      "entropy": 5.875902366638184,
      "epoch": 3.007797270955166,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004108061336897032,
      "loss": 5.4004,
      "mean_token_accuracy": 0.19966518580913545,
      "num_tokens": 58860478.0,
      "step": 30090
    },
    {
      "entropy": 5.971120977401734,
      "epoch": 3.008297096016394,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004107776770496116,
      "loss": 5.4896,
      "mean_token_accuracy": 0.18873147666454315,
      "num_tokens": 58871700.0,
      "step": 30095
    },
    {
      "entropy": 6.031062507629395,
      "epoch": 3.008796921077623,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004107492169933885,
      "loss": 5.5038,
      "mean_token_accuracy": 0.18169477730989456,
      "num_tokens": 58882154.0,
      "step": 30100
    },
    {
      "entropy": 5.950187349319458,
      "epoch": 3.0092967461388516,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041072075352175004,
      "loss": 5.4159,
      "mean_token_accuracy": 0.19788803458213805,
      "num_tokens": 58892012.0,
      "step": 30105
    },
    {
      "entropy": 5.951016235351562,
      "epoch": 3.00979657120008,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041069228663541234,
      "loss": 5.4077,
      "mean_token_accuracy": 0.19793081730604173,
      "num_tokens": 58901196.0,
      "step": 30110
    },
    {
      "entropy": 5.881593179702759,
      "epoch": 3.0102963962613085,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004106638163350917,
      "loss": 5.3589,
      "mean_token_accuracy": 0.20171667337417604,
      "num_tokens": 58911474.0,
      "step": 30115
    },
    {
      "entropy": 5.89168643951416,
      "epoch": 3.010796221322537,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004106353426215043,
      "loss": 5.3355,
      "mean_token_accuracy": 0.19688117355108262,
      "num_tokens": 58920440.0,
      "step": 30120
    },
    {
      "entropy": 5.887256908416748,
      "epoch": 3.0112960463837655,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041060686549536656,
      "loss": 5.4609,
      "mean_token_accuracy": 0.19399170875549315,
      "num_tokens": 58929652.0,
      "step": 30125
    },
    {
      "entropy": 5.946654272079468,
      "epoch": 3.011795871444994,
      "grad_norm": 1.0625,
      "learning_rate": 0.000410578384957395,
      "loss": 5.3709,
      "mean_token_accuracy": 0.2027082085609436,
      "num_tokens": 58940216.0,
      "step": 30130
    },
    {
      "entropy": 5.905625867843628,
      "epoch": 3.012295696506223,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004105499010083062,
      "loss": 5.3809,
      "mean_token_accuracy": 0.20219298899173738,
      "num_tokens": 58949517.0,
      "step": 30135
    },
    {
      "entropy": 5.827149295806885,
      "epoch": 3.0127955215674516,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041052141364881665,
      "loss": 5.3717,
      "mean_token_accuracy": 0.20460802018642427,
      "num_tokens": 58959044.0,
      "step": 30140
    },
    {
      "entropy": 5.938308000564575,
      "epoch": 3.01329534662868,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041049292287964335,
      "loss": 5.4272,
      "mean_token_accuracy": 0.1987910598516464,
      "num_tokens": 58968352.0,
      "step": 30145
    },
    {
      "entropy": 5.8687139511108395,
      "epoch": 3.0137951716899085,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041046442870150293,
      "loss": 5.2832,
      "mean_token_accuracy": 0.2093192756175995,
      "num_tokens": 58979082.0,
      "step": 30150
    },
    {
      "entropy": 5.87050085067749,
      "epoch": 3.014294996751137,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041043593111511227,
      "loss": 5.3394,
      "mean_token_accuracy": 0.20372072756290435,
      "num_tokens": 58988181.0,
      "step": 30155
    },
    {
      "entropy": 5.918384265899658,
      "epoch": 3.0147948218123655,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004104074301211885,
      "loss": 5.4171,
      "mean_token_accuracy": 0.20157066881656646,
      "num_tokens": 58999907.0,
      "step": 30160
    },
    {
      "entropy": 5.926093721389771,
      "epoch": 3.015294646873594,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041037892572044863,
      "loss": 5.4755,
      "mean_token_accuracy": 0.19371137619018555,
      "num_tokens": 59009293.0,
      "step": 30165
    },
    {
      "entropy": 5.8979285717010494,
      "epoch": 3.015794471934823,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041035041791360983,
      "loss": 5.4207,
      "mean_token_accuracy": 0.19754521250724794,
      "num_tokens": 59017079.0,
      "step": 30170
    },
    {
      "entropy": 5.916187715530396,
      "epoch": 3.0162942969960516,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004103219067013893,
      "loss": 5.3773,
      "mean_token_accuracy": 0.20021371245384217,
      "num_tokens": 59027008.0,
      "step": 30175
    },
    {
      "entropy": 5.910969066619873,
      "epoch": 3.01679412205728,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041029339208450445,
      "loss": 5.4465,
      "mean_token_accuracy": 0.19134718775749207,
      "num_tokens": 59036971.0,
      "step": 30180
    },
    {
      "entropy": 5.893609046936035,
      "epoch": 3.0172939471185085,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004102648740636727,
      "loss": 5.4004,
      "mean_token_accuracy": 0.1945594757795334,
      "num_tokens": 59048264.0,
      "step": 30185
    },
    {
      "entropy": 5.925199747085571,
      "epoch": 3.017793772179737,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004102363526396115,
      "loss": 5.5026,
      "mean_token_accuracy": 0.19645173251628875,
      "num_tokens": 59057981.0,
      "step": 30190
    },
    {
      "entropy": 6.038819074630737,
      "epoch": 3.0182935972409655,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004102078278130384,
      "loss": 5.5156,
      "mean_token_accuracy": 0.18966483175754548,
      "num_tokens": 59068441.0,
      "step": 30195
    },
    {
      "entropy": 5.975192880630493,
      "epoch": 3.018793422302194,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041017929958467113,
      "loss": 5.4518,
      "mean_token_accuracy": 0.19706981778144836,
      "num_tokens": 59078544.0,
      "step": 30200
    },
    {
      "entropy": 5.8789369583129885,
      "epoch": 3.019293247363423,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004101507679552276,
      "loss": 5.4198,
      "mean_token_accuracy": 0.1977532386779785,
      "num_tokens": 59088063.0,
      "step": 30205
    },
    {
      "entropy": 5.930240154266357,
      "epoch": 3.0197930724246516,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004101222329254255,
      "loss": 5.458,
      "mean_token_accuracy": 0.1938480406999588,
      "num_tokens": 59098003.0,
      "step": 30210
    },
    {
      "entropy": 5.982387256622315,
      "epoch": 3.02029289748588,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004100936944959828,
      "loss": 5.444,
      "mean_token_accuracy": 0.19126382619142532,
      "num_tokens": 59108595.0,
      "step": 30215
    },
    {
      "entropy": 5.967422533035278,
      "epoch": 3.0207927225471085,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004100651526676174,
      "loss": 5.4983,
      "mean_token_accuracy": 0.18988959938287736,
      "num_tokens": 59118453.0,
      "step": 30220
    },
    {
      "entropy": 5.870214128494263,
      "epoch": 3.021292547608337,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004100366074410476,
      "loss": 5.3878,
      "mean_token_accuracy": 0.19745416045188904,
      "num_tokens": 59128005.0,
      "step": 30225
    },
    {
      "entropy": 5.96715931892395,
      "epoch": 3.0217923726695655,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041000805881699143,
      "loss": 5.4257,
      "mean_token_accuracy": 0.1934930667281151,
      "num_tokens": 59137694.0,
      "step": 30230
    },
    {
      "entropy": 5.945396995544433,
      "epoch": 3.022292197730794,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004099795067961673,
      "loss": 5.4573,
      "mean_token_accuracy": 0.1957847759127617,
      "num_tokens": 59147715.0,
      "step": 30235
    },
    {
      "entropy": 5.909523248672485,
      "epoch": 3.022792022792023,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004099509513792935,
      "loss": 5.4335,
      "mean_token_accuracy": 0.19809267073869705,
      "num_tokens": 59157151.0,
      "step": 30240
    },
    {
      "entropy": 6.039516353607178,
      "epoch": 3.0232918478532516,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040992239256708844,
      "loss": 5.5091,
      "mean_token_accuracy": 0.19086128920316697,
      "num_tokens": 59167349.0,
      "step": 30245
    },
    {
      "entropy": 5.881411695480347,
      "epoch": 3.02379167291448,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004098938303602708,
      "loss": 5.382,
      "mean_token_accuracy": 0.19930053651332855,
      "num_tokens": 59176412.0,
      "step": 30250
    },
    {
      "entropy": 5.79476146697998,
      "epoch": 3.0242914979757085,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040986526475955904,
      "loss": 5.3762,
      "mean_token_accuracy": 0.19557883888483046,
      "num_tokens": 59186050.0,
      "step": 30255
    },
    {
      "entropy": 5.881894207000732,
      "epoch": 3.024791323036937,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000409836695765672,
      "loss": 5.373,
      "mean_token_accuracy": 0.20106489062309266,
      "num_tokens": 59196185.0,
      "step": 30260
    },
    {
      "entropy": 5.950100898742676,
      "epoch": 3.0252911480981655,
      "grad_norm": 1.125,
      "learning_rate": 0.00040980812337932835,
      "loss": 5.4493,
      "mean_token_accuracy": 0.1937931552529335,
      "num_tokens": 59208079.0,
      "step": 30265
    },
    {
      "entropy": 5.954550695419312,
      "epoch": 3.025790973159394,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040977954760124693,
      "loss": 5.4077,
      "mean_token_accuracy": 0.1927926167845726,
      "num_tokens": 59216968.0,
      "step": 30270
    },
    {
      "entropy": 5.849770975112915,
      "epoch": 3.026290798220623,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040975096843214677,
      "loss": 5.3261,
      "mean_token_accuracy": 0.20787594914436341,
      "num_tokens": 59226481.0,
      "step": 30275
    },
    {
      "entropy": 5.840699529647827,
      "epoch": 3.0267906232818516,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040972238587274695,
      "loss": 5.3477,
      "mean_token_accuracy": 0.2022372379899025,
      "num_tokens": 59235794.0,
      "step": 30280
    },
    {
      "entropy": 5.921406316757202,
      "epoch": 3.02729044834308,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040969379992376664,
      "loss": 5.441,
      "mean_token_accuracy": 0.19311554580926896,
      "num_tokens": 59245691.0,
      "step": 30285
    },
    {
      "entropy": 5.984455871582031,
      "epoch": 3.0277902734043085,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004096652105859249,
      "loss": 5.4594,
      "mean_token_accuracy": 0.1918927922844887,
      "num_tokens": 59255591.0,
      "step": 30290
    },
    {
      "entropy": 5.8751311779022215,
      "epoch": 3.028290098465537,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004096366178599412,
      "loss": 5.3106,
      "mean_token_accuracy": 0.209735906124115,
      "num_tokens": 59266183.0,
      "step": 30295
    },
    {
      "entropy": 5.957777786254883,
      "epoch": 3.0287899235267655,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004096080217465347,
      "loss": 5.4765,
      "mean_token_accuracy": 0.18781511038541793,
      "num_tokens": 59276241.0,
      "step": 30300
    },
    {
      "entropy": 5.880269575119018,
      "epoch": 3.029289748587994,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004095794222464252,
      "loss": 5.3993,
      "mean_token_accuracy": 0.19791244715452194,
      "num_tokens": 59286089.0,
      "step": 30305
    },
    {
      "entropy": 5.879623985290527,
      "epoch": 3.029789573649223,
      "grad_norm": 1.125,
      "learning_rate": 0.0004095508193603319,
      "loss": 5.4635,
      "mean_token_accuracy": 0.1899726539850235,
      "num_tokens": 59295122.0,
      "step": 30310
    },
    {
      "entropy": 5.903068399429321,
      "epoch": 3.0302893987104516,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004095222130889747,
      "loss": 5.4937,
      "mean_token_accuracy": 0.18820362091064452,
      "num_tokens": 59304277.0,
      "step": 30315
    },
    {
      "entropy": 5.8627927780151365,
      "epoch": 3.03078922377168,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040949360343307324,
      "loss": 5.3491,
      "mean_token_accuracy": 0.2013239309191704,
      "num_tokens": 59315560.0,
      "step": 30320
    },
    {
      "entropy": 5.909248018264771,
      "epoch": 3.0312890488329085,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004094649903933472,
      "loss": 5.3651,
      "mean_token_accuracy": 0.19693086594343184,
      "num_tokens": 59325238.0,
      "step": 30325
    },
    {
      "entropy": 5.884165096282959,
      "epoch": 3.031788873894137,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040943637397051683,
      "loss": 5.3299,
      "mean_token_accuracy": 0.20577120035886765,
      "num_tokens": 59335800.0,
      "step": 30330
    },
    {
      "entropy": 5.887612628936767,
      "epoch": 3.0322886989553655,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040940775416530164,
      "loss": 5.4851,
      "mean_token_accuracy": 0.1901850461959839,
      "num_tokens": 59345854.0,
      "step": 30335
    },
    {
      "entropy": 5.911321210861206,
      "epoch": 3.032788524016594,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040937913097842204,
      "loss": 5.3378,
      "mean_token_accuracy": 0.20077372640371322,
      "num_tokens": 59354789.0,
      "step": 30340
    },
    {
      "entropy": 5.938931226730347,
      "epoch": 3.033288349077823,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004093505044105982,
      "loss": 5.4859,
      "mean_token_accuracy": 0.18695714622735976,
      "num_tokens": 59364205.0,
      "step": 30345
    },
    {
      "entropy": 5.918861961364746,
      "epoch": 3.033788174139051,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040932187446255004,
      "loss": 5.462,
      "mean_token_accuracy": 0.19151501953601838,
      "num_tokens": 59375037.0,
      "step": 30350
    },
    {
      "entropy": 5.829775953292847,
      "epoch": 3.03428799920028,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004092932411349982,
      "loss": 5.2966,
      "mean_token_accuracy": 0.2061312600970268,
      "num_tokens": 59384820.0,
      "step": 30355
    },
    {
      "entropy": 5.84583683013916,
      "epoch": 3.0347878242615085,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000409264604428663,
      "loss": 5.4032,
      "mean_token_accuracy": 0.20066970437765122,
      "num_tokens": 59394571.0,
      "step": 30360
    },
    {
      "entropy": 5.896320533752442,
      "epoch": 3.035287649322737,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004092359643442648,
      "loss": 5.4768,
      "mean_token_accuracy": 0.19079758673906327,
      "num_tokens": 59405758.0,
      "step": 30365
    },
    {
      "entropy": 5.940830945968628,
      "epoch": 3.0357874743839655,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004092073208825243,
      "loss": 5.3722,
      "mean_token_accuracy": 0.19921401143074036,
      "num_tokens": 59415091.0,
      "step": 30370
    },
    {
      "entropy": 5.921091890335083,
      "epoch": 3.036287299445194,
      "grad_norm": 1.125,
      "learning_rate": 0.0004091786740441622,
      "loss": 5.46,
      "mean_token_accuracy": 0.18978944569826126,
      "num_tokens": 59423867.0,
      "step": 30375
    },
    {
      "entropy": 5.9031679153442385,
      "epoch": 3.036787124506423,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040915002382989915,
      "loss": 5.4328,
      "mean_token_accuracy": 0.191286201775074,
      "num_tokens": 59433594.0,
      "step": 30380
    },
    {
      "entropy": 5.887979173660279,
      "epoch": 3.037286949567651,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000409121370240456,
      "loss": 5.4366,
      "mean_token_accuracy": 0.196049265563488,
      "num_tokens": 59442411.0,
      "step": 30385
    },
    {
      "entropy": 5.8541100978851315,
      "epoch": 3.03778677462888,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004090927132765537,
      "loss": 5.3633,
      "mean_token_accuracy": 0.19822054207324982,
      "num_tokens": 59451507.0,
      "step": 30390
    },
    {
      "entropy": 5.9107935428619385,
      "epoch": 3.0382865996901085,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040906405293891323,
      "loss": 5.4328,
      "mean_token_accuracy": 0.19782036542892456,
      "num_tokens": 59462497.0,
      "step": 30395
    },
    {
      "entropy": 5.910058450698853,
      "epoch": 3.038786424751337,
      "grad_norm": 1.125,
      "learning_rate": 0.0004090353892282557,
      "loss": 5.4632,
      "mean_token_accuracy": 0.186564864218235,
      "num_tokens": 59472574.0,
      "step": 30400
    },
    {
      "entropy": 5.908770322799683,
      "epoch": 3.0392862498125655,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004090067221453024,
      "loss": 5.307,
      "mean_token_accuracy": 0.20288966596126556,
      "num_tokens": 59482508.0,
      "step": 30405
    },
    {
      "entropy": 6.036912870407105,
      "epoch": 3.039786074873794,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004089780516907743,
      "loss": 5.5522,
      "mean_token_accuracy": 0.18505992442369462,
      "num_tokens": 59493032.0,
      "step": 30410
    },
    {
      "entropy": 5.956371164321899,
      "epoch": 3.040285899935023,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040894937786539287,
      "loss": 5.4926,
      "mean_token_accuracy": 0.19285075068473817,
      "num_tokens": 59502769.0,
      "step": 30415
    },
    {
      "entropy": 5.852999353408814,
      "epoch": 3.040785724996251,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040892070066987966,
      "loss": 5.4075,
      "mean_token_accuracy": 0.18966768234968184,
      "num_tokens": 59512392.0,
      "step": 30420
    },
    {
      "entropy": 5.894126844406128,
      "epoch": 3.04128555005748,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040889202010495595,
      "loss": 5.3682,
      "mean_token_accuracy": 0.2008896842598915,
      "num_tokens": 59521936.0,
      "step": 30425
    },
    {
      "entropy": 5.944360208511353,
      "epoch": 3.0417853751187085,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004088633361713436,
      "loss": 5.4897,
      "mean_token_accuracy": 0.18608387112617492,
      "num_tokens": 59531460.0,
      "step": 30430
    },
    {
      "entropy": 5.943208122253418,
      "epoch": 3.042285200179937,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040883464886976414,
      "loss": 5.4954,
      "mean_token_accuracy": 0.18802045434713363,
      "num_tokens": 59541898.0,
      "step": 30435
    },
    {
      "entropy": 5.9251409530639645,
      "epoch": 3.0427850252411655,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004088059582009393,
      "loss": 5.5179,
      "mean_token_accuracy": 0.18757059276103974,
      "num_tokens": 59552960.0,
      "step": 30440
    },
    {
      "entropy": 5.919236183166504,
      "epoch": 3.043284850302394,
      "grad_norm": 1.109375,
      "learning_rate": 0.000408777264165591,
      "loss": 5.4015,
      "mean_token_accuracy": 0.19346380233764648,
      "num_tokens": 59563655.0,
      "step": 30445
    },
    {
      "entropy": 5.907702684402466,
      "epoch": 3.043784675363623,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040874856676444116,
      "loss": 5.3946,
      "mean_token_accuracy": 0.21265247464179993,
      "num_tokens": 59573554.0,
      "step": 30450
    },
    {
      "entropy": 5.905211162567139,
      "epoch": 3.044284500424851,
      "grad_norm": 1.125,
      "learning_rate": 0.00040871986599821177,
      "loss": 5.4051,
      "mean_token_accuracy": 0.199905164539814,
      "num_tokens": 59582720.0,
      "step": 30455
    },
    {
      "entropy": 5.869992303848266,
      "epoch": 3.04478432548608,
      "grad_norm": 1.25,
      "learning_rate": 0.0004086911618676251,
      "loss": 5.416,
      "mean_token_accuracy": 0.19761626273393632,
      "num_tokens": 59591589.0,
      "step": 30460
    },
    {
      "entropy": 5.987070894241333,
      "epoch": 3.0452841505473085,
      "grad_norm": 1.15625,
      "learning_rate": 0.000408662454373403,
      "loss": 5.5074,
      "mean_token_accuracy": 0.18262936919927597,
      "num_tokens": 59601575.0,
      "step": 30465
    },
    {
      "entropy": 5.883894300460815,
      "epoch": 3.045783975608537,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000408633743516268,
      "loss": 5.3921,
      "mean_token_accuracy": 0.2047439619898796,
      "num_tokens": 59610842.0,
      "step": 30470
    },
    {
      "entropy": 5.9977282047271725,
      "epoch": 3.0462838006697655,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040860502929694246,
      "loss": 5.569,
      "mean_token_accuracy": 0.18438994586467744,
      "num_tokens": 59622054.0,
      "step": 30475
    },
    {
      "entropy": 5.968014192581177,
      "epoch": 3.046783625730994,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040857631171614873,
      "loss": 5.3785,
      "mean_token_accuracy": 0.20312799513339996,
      "num_tokens": 59631516.0,
      "step": 30480
    },
    {
      "entropy": 5.921487808227539,
      "epoch": 3.047283450792223,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040854759077460934,
      "loss": 5.3807,
      "mean_token_accuracy": 0.19624816328287126,
      "num_tokens": 59641999.0,
      "step": 30485
    },
    {
      "entropy": 5.866176128387451,
      "epoch": 3.047783275853451,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040851886647304703,
      "loss": 5.366,
      "mean_token_accuracy": 0.2007091224193573,
      "num_tokens": 59651107.0,
      "step": 30490
    },
    {
      "entropy": 5.927489757537842,
      "epoch": 3.04828310091468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004084901388121843,
      "loss": 5.4285,
      "mean_token_accuracy": 0.19602158963680266,
      "num_tokens": 59661075.0,
      "step": 30495
    },
    {
      "entropy": 5.8429969310760494,
      "epoch": 3.0487829259759085,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040846140779274406,
      "loss": 5.2827,
      "mean_token_accuracy": 0.20475355237722398,
      "num_tokens": 59671449.0,
      "step": 30500
    },
    {
      "entropy": 5.94308443069458,
      "epoch": 3.049282751037137,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040843267341544913,
      "loss": 5.5239,
      "mean_token_accuracy": 0.18615092933177949,
      "num_tokens": 59680824.0,
      "step": 30505
    },
    {
      "entropy": 5.963815402984619,
      "epoch": 3.0497825760983654,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040840393568102246,
      "loss": 5.4038,
      "mean_token_accuracy": 0.1932188719511032,
      "num_tokens": 59689293.0,
      "step": 30510
    },
    {
      "entropy": 5.941911315917968,
      "epoch": 3.050282401159594,
      "grad_norm": 1.125,
      "learning_rate": 0.0004083751945901871,
      "loss": 5.5573,
      "mean_token_accuracy": 0.18351823836565018,
      "num_tokens": 59698693.0,
      "step": 30515
    },
    {
      "entropy": 5.956985855102539,
      "epoch": 3.050782226220823,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040834645014366623,
      "loss": 5.4663,
      "mean_token_accuracy": 0.1853393390774727,
      "num_tokens": 59709047.0,
      "step": 30520
    },
    {
      "entropy": 6.017885208129883,
      "epoch": 3.051282051282051,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004083177023421829,
      "loss": 5.4563,
      "mean_token_accuracy": 0.18435919880867005,
      "num_tokens": 59719392.0,
      "step": 30525
    },
    {
      "entropy": 5.827832221984863,
      "epoch": 3.05178187634328,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004082889511864606,
      "loss": 5.3081,
      "mean_token_accuracy": 0.20252157896757125,
      "num_tokens": 59728736.0,
      "step": 30530
    },
    {
      "entropy": 5.940513181686401,
      "epoch": 3.0522817014045085,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040826019667722244,
      "loss": 5.4917,
      "mean_token_accuracy": 0.1906200185418129,
      "num_tokens": 59738816.0,
      "step": 30535
    },
    {
      "entropy": 5.932342433929444,
      "epoch": 3.052781526465737,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004082314388151921,
      "loss": 5.41,
      "mean_token_accuracy": 0.2004308894276619,
      "num_tokens": 59747785.0,
      "step": 30540
    },
    {
      "entropy": 5.956087827682495,
      "epoch": 3.0532813515269654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004082026776010931,
      "loss": 5.4219,
      "mean_token_accuracy": 0.19642828404903412,
      "num_tokens": 59757429.0,
      "step": 30545
    },
    {
      "entropy": 5.958818864822388,
      "epoch": 3.053781176588194,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004081739130356489,
      "loss": 5.4743,
      "mean_token_accuracy": 0.18843790292739868,
      "num_tokens": 59767996.0,
      "step": 30550
    },
    {
      "entropy": 5.903211641311645,
      "epoch": 3.054281001649423,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040814514511958335,
      "loss": 5.4528,
      "mean_token_accuracy": 0.19425930678844452,
      "num_tokens": 59778487.0,
      "step": 30555
    },
    {
      "entropy": 5.886093997955323,
      "epoch": 3.054780826710651,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040811637385362024,
      "loss": 5.3331,
      "mean_token_accuracy": 0.210261932015419,
      "num_tokens": 59788929.0,
      "step": 30560
    },
    {
      "entropy": 5.920989608764648,
      "epoch": 3.05528065177188,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004080875992384834,
      "loss": 5.4515,
      "mean_token_accuracy": 0.19996186047792436,
      "num_tokens": 59797548.0,
      "step": 30565
    },
    {
      "entropy": 5.957885074615478,
      "epoch": 3.0557804768331085,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004080588212748968,
      "loss": 5.4768,
      "mean_token_accuracy": 0.19356207698583602,
      "num_tokens": 59807811.0,
      "step": 30570
    },
    {
      "entropy": 5.848476696014404,
      "epoch": 3.056280301894337,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040803003996358445,
      "loss": 5.3052,
      "mean_token_accuracy": 0.20778602808713914,
      "num_tokens": 59818118.0,
      "step": 30575
    },
    {
      "entropy": 5.869503402709961,
      "epoch": 3.0567801269555654,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040800125530527064,
      "loss": 5.4494,
      "mean_token_accuracy": 0.19325990974903107,
      "num_tokens": 59827749.0,
      "step": 30580
    },
    {
      "entropy": 5.938238143920898,
      "epoch": 3.057279952016794,
      "grad_norm": 1.25,
      "learning_rate": 0.00040797246730067937,
      "loss": 5.3958,
      "mean_token_accuracy": 0.20325591266155243,
      "num_tokens": 59836519.0,
      "step": 30585
    },
    {
      "entropy": 5.874657487869262,
      "epoch": 3.057779777078023,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004079436759505351,
      "loss": 5.3708,
      "mean_token_accuracy": 0.20246343761682511,
      "num_tokens": 59847201.0,
      "step": 30590
    },
    {
      "entropy": 5.858806705474853,
      "epoch": 3.058279602139251,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004079148812555621,
      "loss": 5.3418,
      "mean_token_accuracy": 0.20860953778028488,
      "num_tokens": 59855366.0,
      "step": 30595
    },
    {
      "entropy": 5.79272837638855,
      "epoch": 3.05877942720048,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040788608321648496,
      "loss": 5.2981,
      "mean_token_accuracy": 0.2059684455394745,
      "num_tokens": 59864820.0,
      "step": 30600
    },
    {
      "entropy": 5.944950723648072,
      "epoch": 3.0592792522617085,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004078572818340281,
      "loss": 5.45,
      "mean_token_accuracy": 0.19407192319631578,
      "num_tokens": 59875757.0,
      "step": 30605
    },
    {
      "entropy": 6.041924238204956,
      "epoch": 3.059779077322937,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040782847710891623,
      "loss": 5.5688,
      "mean_token_accuracy": 0.18870031833648682,
      "num_tokens": 59885039.0,
      "step": 30610
    },
    {
      "entropy": 5.958221864700318,
      "epoch": 3.0602789023841654,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040779966904187407,
      "loss": 5.4381,
      "mean_token_accuracy": 0.18902792632579804,
      "num_tokens": 59894516.0,
      "step": 30615
    },
    {
      "entropy": 5.938729953765869,
      "epoch": 3.060778727445394,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040777085763362634,
      "loss": 5.5076,
      "mean_token_accuracy": 0.19346444308757782,
      "num_tokens": 59905004.0,
      "step": 30620
    },
    {
      "entropy": 5.956977415084839,
      "epoch": 3.061278552506623,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040774204288489805,
      "loss": 5.5246,
      "mean_token_accuracy": 0.18891979157924652,
      "num_tokens": 59914814.0,
      "step": 30625
    },
    {
      "entropy": 6.030937433242798,
      "epoch": 3.061778377567851,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004077132247964142,
      "loss": 5.4999,
      "mean_token_accuracy": 0.18869392722845077,
      "num_tokens": 59924451.0,
      "step": 30630
    },
    {
      "entropy": 5.894324731826782,
      "epoch": 3.06227820262908,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040768440336889965,
      "loss": 5.4084,
      "mean_token_accuracy": 0.20128006786108016,
      "num_tokens": 59934607.0,
      "step": 30635
    },
    {
      "entropy": 5.726700210571289,
      "epoch": 3.0627780276903085,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004076555786030797,
      "loss": 5.2577,
      "mean_token_accuracy": 0.21344312578439711,
      "num_tokens": 59945172.0,
      "step": 30640
    },
    {
      "entropy": 5.979023504257202,
      "epoch": 3.063277852751537,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004076267504996795,
      "loss": 5.3947,
      "mean_token_accuracy": 0.19814635217189788,
      "num_tokens": 59955022.0,
      "step": 30645
    },
    {
      "entropy": 5.937502336502075,
      "epoch": 3.0637776778127654,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004075979190594244,
      "loss": 5.4451,
      "mean_token_accuracy": 0.19665387272834778,
      "num_tokens": 59965666.0,
      "step": 30650
    },
    {
      "entropy": 5.9510002613067625,
      "epoch": 3.064277502873994,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040756908428303984,
      "loss": 5.4386,
      "mean_token_accuracy": 0.19404580146074296,
      "num_tokens": 59975069.0,
      "step": 30655
    },
    {
      "entropy": 5.984569072723389,
      "epoch": 3.064777327935223,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040754024617125115,
      "loss": 5.5019,
      "mean_token_accuracy": 0.19193803071975707,
      "num_tokens": 59986100.0,
      "step": 30660
    },
    {
      "entropy": 5.834917116165161,
      "epoch": 3.065277152996451,
      "grad_norm": 1.25,
      "learning_rate": 0.000407511404724784,
      "loss": 5.3894,
      "mean_token_accuracy": 0.20320395827293397,
      "num_tokens": 59995929.0,
      "step": 30665
    },
    {
      "entropy": 5.892732906341553,
      "epoch": 3.06577697805768,
      "grad_norm": 1.140625,
      "learning_rate": 0.000407482559944364,
      "loss": 5.4529,
      "mean_token_accuracy": 0.19533451944589614,
      "num_tokens": 60005526.0,
      "step": 30670
    },
    {
      "entropy": 5.880696249008179,
      "epoch": 3.0662768031189085,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004074537118307169,
      "loss": 5.2816,
      "mean_token_accuracy": 0.20564714521169664,
      "num_tokens": 60014833.0,
      "step": 30675
    },
    {
      "entropy": 5.927912092208862,
      "epoch": 3.0667766281801367,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004074248603845684,
      "loss": 5.5153,
      "mean_token_accuracy": 0.1852419838309288,
      "num_tokens": 60024579.0,
      "step": 30680
    },
    {
      "entropy": 5.951638793945312,
      "epoch": 3.0672764532413654,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040739600560664457,
      "loss": 5.504,
      "mean_token_accuracy": 0.1918417200446129,
      "num_tokens": 60035162.0,
      "step": 30685
    },
    {
      "entropy": 5.92322416305542,
      "epoch": 3.067776278302594,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040736714749767136,
      "loss": 5.3534,
      "mean_token_accuracy": 0.20071193426847458,
      "num_tokens": 60044894.0,
      "step": 30690
    },
    {
      "entropy": 5.912854623794556,
      "epoch": 3.068276103363823,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004073382860583747,
      "loss": 5.4242,
      "mean_token_accuracy": 0.19758083820343017,
      "num_tokens": 60054441.0,
      "step": 30695
    },
    {
      "entropy": 5.958798456192016,
      "epoch": 3.068775928425051,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004073094212894809,
      "loss": 5.454,
      "mean_token_accuracy": 0.19997714161872865,
      "num_tokens": 60063701.0,
      "step": 30700
    },
    {
      "entropy": 5.931525945663452,
      "epoch": 3.06927575348628,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000407280553191716,
      "loss": 5.4557,
      "mean_token_accuracy": 0.18705841153860092,
      "num_tokens": 60073455.0,
      "step": 30705
    },
    {
      "entropy": 5.817700052261353,
      "epoch": 3.0697755785475085,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004072516817658065,
      "loss": 5.3889,
      "mean_token_accuracy": 0.19704056680202484,
      "num_tokens": 60083898.0,
      "step": 30710
    },
    {
      "entropy": 5.924721145629883,
      "epoch": 3.0702754036087367,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040722280701247865,
      "loss": 5.4883,
      "mean_token_accuracy": 0.19018111377954483,
      "num_tokens": 60093605.0,
      "step": 30715
    },
    {
      "entropy": 6.041705322265625,
      "epoch": 3.0707752286699654,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040719392893245903,
      "loss": 5.533,
      "mean_token_accuracy": 0.18823168724775313,
      "num_tokens": 60103240.0,
      "step": 30720
    },
    {
      "entropy": 5.840026998519898,
      "epoch": 3.071275053731194,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004071650475264742,
      "loss": 5.3493,
      "mean_token_accuracy": 0.2019277259707451,
      "num_tokens": 60112850.0,
      "step": 30725
    },
    {
      "entropy": 5.89941554069519,
      "epoch": 3.071774878792423,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040713616279525075,
      "loss": 5.3517,
      "mean_token_accuracy": 0.203596892952919,
      "num_tokens": 60122565.0,
      "step": 30730
    },
    {
      "entropy": 5.940197706222534,
      "epoch": 3.072274703853651,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040710727473951546,
      "loss": 5.3777,
      "mean_token_accuracy": 0.19023175686597824,
      "num_tokens": 60132239.0,
      "step": 30735
    },
    {
      "entropy": 5.874248027801514,
      "epoch": 3.07277452891488,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004070783833599951,
      "loss": 5.4305,
      "mean_token_accuracy": 0.18919731974601744,
      "num_tokens": 60141993.0,
      "step": 30740
    },
    {
      "entropy": 5.897395086288452,
      "epoch": 3.0732743539761085,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004070494886574166,
      "loss": 5.44,
      "mean_token_accuracy": 0.19512563198804855,
      "num_tokens": 60151350.0,
      "step": 30745
    },
    {
      "entropy": 5.950310611724854,
      "epoch": 3.0737741790373367,
      "grad_norm": 1.0625,
      "learning_rate": 0.000407020590632507,
      "loss": 5.4354,
      "mean_token_accuracy": 0.19564239382743837,
      "num_tokens": 60161120.0,
      "step": 30750
    },
    {
      "entropy": 5.925402593612671,
      "epoch": 3.0742740040985654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004069916892859933,
      "loss": 5.4805,
      "mean_token_accuracy": 0.19782979488372804,
      "num_tokens": 60170762.0,
      "step": 30755
    },
    {
      "entropy": 5.873849201202392,
      "epoch": 3.074773829159794,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004069627846186026,
      "loss": 5.391,
      "mean_token_accuracy": 0.19979631453752517,
      "num_tokens": 60180729.0,
      "step": 30760
    },
    {
      "entropy": 5.866496276855469,
      "epoch": 3.075273654221023,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040693387663106223,
      "loss": 5.3878,
      "mean_token_accuracy": 0.20263220369815826,
      "num_tokens": 60190640.0,
      "step": 30765
    },
    {
      "entropy": 5.8604209423065186,
      "epoch": 3.075773479282251,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004069049653240996,
      "loss": 5.388,
      "mean_token_accuracy": 0.1953180968761444,
      "num_tokens": 60200163.0,
      "step": 30770
    },
    {
      "entropy": 5.983507299423218,
      "epoch": 3.07627330434348,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004068760506984418,
      "loss": 5.4937,
      "mean_token_accuracy": 0.1856559470295906,
      "num_tokens": 60209469.0,
      "step": 30775
    },
    {
      "entropy": 5.812021923065186,
      "epoch": 3.0767731294047085,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004068471327548166,
      "loss": 5.2854,
      "mean_token_accuracy": 0.20400647819042206,
      "num_tokens": 60218950.0,
      "step": 30780
    },
    {
      "entropy": 5.8921318531036375,
      "epoch": 3.0772729544659367,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040681821149395146,
      "loss": 5.4271,
      "mean_token_accuracy": 0.19591247886419297,
      "num_tokens": 60228428.0,
      "step": 30785
    },
    {
      "entropy": 5.85732626914978,
      "epoch": 3.0777727795271654,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040678928691657395,
      "loss": 5.4196,
      "mean_token_accuracy": 0.20289104729890822,
      "num_tokens": 60237558.0,
      "step": 30790
    },
    {
      "entropy": 5.886354446411133,
      "epoch": 3.078272604588394,
      "grad_norm": 1.125,
      "learning_rate": 0.00040676035902341195,
      "loss": 5.3632,
      "mean_token_accuracy": 0.19786642342805863,
      "num_tokens": 60246282.0,
      "step": 30795
    },
    {
      "entropy": 5.785243082046509,
      "epoch": 3.078772429649623,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004067314278151933,
      "loss": 5.2307,
      "mean_token_accuracy": 0.2106954723596573,
      "num_tokens": 60255678.0,
      "step": 30800
    },
    {
      "entropy": 5.867339038848877,
      "epoch": 3.079272254710851,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040670249329264577,
      "loss": 5.3308,
      "mean_token_accuracy": 0.21641576141119004,
      "num_tokens": 60266900.0,
      "step": 30805
    },
    {
      "entropy": 5.9861485958099365,
      "epoch": 3.07977207977208,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004066735554564973,
      "loss": 5.4859,
      "mean_token_accuracy": 0.19021835029125214,
      "num_tokens": 60276974.0,
      "step": 30810
    },
    {
      "entropy": 6.053391838073731,
      "epoch": 3.0802719048333085,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004066446143074762,
      "loss": 5.5075,
      "mean_token_accuracy": 0.19700342863798143,
      "num_tokens": 60285830.0,
      "step": 30815
    },
    {
      "entropy": 5.860975360870361,
      "epoch": 3.0807717298945367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004066156698463104,
      "loss": 5.3578,
      "mean_token_accuracy": 0.19812726825475693,
      "num_tokens": 60295178.0,
      "step": 30820
    },
    {
      "entropy": 5.898056364059448,
      "epoch": 3.0812715549557654,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004065867220737282,
      "loss": 5.4072,
      "mean_token_accuracy": 0.19813573211431504,
      "num_tokens": 60305514.0,
      "step": 30825
    },
    {
      "entropy": 5.833637809753418,
      "epoch": 3.081771380016994,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040655777099045794,
      "loss": 5.3411,
      "mean_token_accuracy": 0.20424936413764955,
      "num_tokens": 60314407.0,
      "step": 30830
    },
    {
      "entropy": 5.878670406341553,
      "epoch": 3.082271205078223,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004065288165972281,
      "loss": 5.3054,
      "mean_token_accuracy": 0.20941711813211442,
      "num_tokens": 60322711.0,
      "step": 30835
    },
    {
      "entropy": 5.893785762786865,
      "epoch": 3.082771030139451,
      "grad_norm": 1.140625,
      "learning_rate": 0.000406499858894767,
      "loss": 5.5035,
      "mean_token_accuracy": 0.19416205137968062,
      "num_tokens": 60332718.0,
      "step": 30840
    },
    {
      "entropy": 5.846434783935547,
      "epoch": 3.08327085520068,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004064708978838033,
      "loss": 5.3572,
      "mean_token_accuracy": 0.2018183335661888,
      "num_tokens": 60343513.0,
      "step": 30845
    },
    {
      "entropy": 5.8632715225219725,
      "epoch": 3.0837706802619085,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040644193356506565,
      "loss": 5.3412,
      "mean_token_accuracy": 0.20277045518159867,
      "num_tokens": 60353374.0,
      "step": 30850
    },
    {
      "entropy": 5.938614559173584,
      "epoch": 3.0842705053231367,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004064129659392828,
      "loss": 5.398,
      "mean_token_accuracy": 0.20186079144477845,
      "num_tokens": 60364303.0,
      "step": 30855
    },
    {
      "entropy": 5.8528820991516115,
      "epoch": 3.0847703303843654,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004063839950071835,
      "loss": 5.2893,
      "mean_token_accuracy": 0.2042083442211151,
      "num_tokens": 60374282.0,
      "step": 30860
    },
    {
      "entropy": 5.966144371032715,
      "epoch": 3.085270155445594,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040635502076949665,
      "loss": 5.4813,
      "mean_token_accuracy": 0.1902784988284111,
      "num_tokens": 60384711.0,
      "step": 30865
    },
    {
      "entropy": 5.921846961975097,
      "epoch": 3.0857699805068224,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040632604322695135,
      "loss": 5.4818,
      "mean_token_accuracy": 0.19118872433900833,
      "num_tokens": 60394681.0,
      "step": 30870
    },
    {
      "entropy": 5.901931238174439,
      "epoch": 3.086269805568051,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004062970623802766,
      "loss": 5.4172,
      "mean_token_accuracy": 0.19242774099111556,
      "num_tokens": 60404226.0,
      "step": 30875
    },
    {
      "entropy": 5.922964572906494,
      "epoch": 3.08676963062928,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004062680782302015,
      "loss": 5.4501,
      "mean_token_accuracy": 0.19166597872972488,
      "num_tokens": 60413657.0,
      "step": 30880
    },
    {
      "entropy": 5.9257221698760985,
      "epoch": 3.0872694556905085,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040623909077745535,
      "loss": 5.3615,
      "mean_token_accuracy": 0.19614530205726624,
      "num_tokens": 60423574.0,
      "step": 30885
    },
    {
      "entropy": 5.863152170181275,
      "epoch": 3.0877692807517367,
      "grad_norm": 1.125,
      "learning_rate": 0.0004062101000227674,
      "loss": 5.3675,
      "mean_token_accuracy": 0.20142326056957244,
      "num_tokens": 60433349.0,
      "step": 30890
    },
    {
      "entropy": 5.8729626655578615,
      "epoch": 3.0882691058129654,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004061811059668672,
      "loss": 5.4911,
      "mean_token_accuracy": 0.1865743577480316,
      "num_tokens": 60442435.0,
      "step": 30895
    },
    {
      "entropy": 5.900747156143188,
      "epoch": 3.088768930874194,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004061521086104841,
      "loss": 5.3945,
      "mean_token_accuracy": 0.1979987546801567,
      "num_tokens": 60451386.0,
      "step": 30900
    },
    {
      "entropy": 5.939335870742798,
      "epoch": 3.0892687559354224,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040612310795434765,
      "loss": 5.4023,
      "mean_token_accuracy": 0.1940111130475998,
      "num_tokens": 60460780.0,
      "step": 30905
    },
    {
      "entropy": 5.936904382705689,
      "epoch": 3.089768580996651,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040609410399918754,
      "loss": 5.4615,
      "mean_token_accuracy": 0.19282951802015305,
      "num_tokens": 60470563.0,
      "step": 30910
    },
    {
      "entropy": 5.866451787948608,
      "epoch": 3.09026840605788,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040606509674573354,
      "loss": 5.3563,
      "mean_token_accuracy": 0.2072169840335846,
      "num_tokens": 60481170.0,
      "step": 30915
    },
    {
      "entropy": 6.002349853515625,
      "epoch": 3.0907682311191085,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004060360861947155,
      "loss": 5.5119,
      "mean_token_accuracy": 0.18652859479188919,
      "num_tokens": 60490700.0,
      "step": 30920
    },
    {
      "entropy": 5.90147647857666,
      "epoch": 3.0912680561803367,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040600707234686315,
      "loss": 5.3705,
      "mean_token_accuracy": 0.19808858633041382,
      "num_tokens": 60500262.0,
      "step": 30925
    },
    {
      "entropy": 5.908074617385864,
      "epoch": 3.0917678812415654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040597805520290657,
      "loss": 5.4997,
      "mean_token_accuracy": 0.1892884001135826,
      "num_tokens": 60510453.0,
      "step": 30930
    },
    {
      "entropy": 6.020986890792846,
      "epoch": 3.092267706302794,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004059490347635759,
      "loss": 5.4068,
      "mean_token_accuracy": 0.1937890589237213,
      "num_tokens": 60520069.0,
      "step": 30935
    },
    {
      "entropy": 5.96750807762146,
      "epoch": 3.0927675313640224,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040592001102960116,
      "loss": 5.4731,
      "mean_token_accuracy": 0.19519319087266923,
      "num_tokens": 60530193.0,
      "step": 30940
    },
    {
      "entropy": 5.8879632472991945,
      "epoch": 3.093267356425251,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004058909840017126,
      "loss": 5.4532,
      "mean_token_accuracy": 0.197843699157238,
      "num_tokens": 60539768.0,
      "step": 30945
    },
    {
      "entropy": 5.873777437210083,
      "epoch": 3.09376718148648,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004058619536806407,
      "loss": 5.4017,
      "mean_token_accuracy": 0.19760928601026534,
      "num_tokens": 60549488.0,
      "step": 30950
    },
    {
      "entropy": 5.87101001739502,
      "epoch": 3.0942670065477085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004058329200671155,
      "loss": 5.431,
      "mean_token_accuracy": 0.19835008829832076,
      "num_tokens": 60559611.0,
      "step": 30955
    },
    {
      "entropy": 5.917274045944214,
      "epoch": 3.0947668316089367,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040580388316186785,
      "loss": 5.4046,
      "mean_token_accuracy": 0.19984060376882554,
      "num_tokens": 60569091.0,
      "step": 30960
    },
    {
      "entropy": 5.993820142745972,
      "epoch": 3.0952666566701654,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004057748429656281,
      "loss": 5.5781,
      "mean_token_accuracy": 0.18739984780550004,
      "num_tokens": 60581173.0,
      "step": 30965
    },
    {
      "entropy": 5.868220376968384,
      "epoch": 3.095766481731394,
      "grad_norm": 1.203125,
      "learning_rate": 0.000405745799479127,
      "loss": 5.3076,
      "mean_token_accuracy": 0.20513591766357422,
      "num_tokens": 60590677.0,
      "step": 30970
    },
    {
      "entropy": 6.019811534881592,
      "epoch": 3.0962663067926224,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004057167527030952,
      "loss": 5.5168,
      "mean_token_accuracy": 0.18563170582056046,
      "num_tokens": 60599861.0,
      "step": 30975
    },
    {
      "entropy": 5.934838962554932,
      "epoch": 3.096766131853851,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040568770263826355,
      "loss": 5.5039,
      "mean_token_accuracy": 0.18692032545804976,
      "num_tokens": 60609061.0,
      "step": 30980
    },
    {
      "entropy": 5.9072051525115965,
      "epoch": 3.0972659569150798,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004056586492853629,
      "loss": 5.4205,
      "mean_token_accuracy": 0.19433056116104125,
      "num_tokens": 60618935.0,
      "step": 30985
    },
    {
      "entropy": 5.943292236328125,
      "epoch": 3.0977657819763085,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004056295926451243,
      "loss": 5.4097,
      "mean_token_accuracy": 0.2095871537923813,
      "num_tokens": 60628549.0,
      "step": 30990
    },
    {
      "entropy": 5.885662412643432,
      "epoch": 3.0982656070375367,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040560053271827866,
      "loss": 5.376,
      "mean_token_accuracy": 0.19594067931175232,
      "num_tokens": 60638253.0,
      "step": 30995
    },
    {
      "entropy": 5.881265258789062,
      "epoch": 3.0987654320987654,
      "grad_norm": 1.125,
      "learning_rate": 0.00040557146950555734,
      "loss": 5.3745,
      "mean_token_accuracy": 0.2018751695752144,
      "num_tokens": 60647575.0,
      "step": 31000
    },
    {
      "entropy": 5.864163732528686,
      "epoch": 3.099265257159994,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004055424030076914,
      "loss": 5.4385,
      "mean_token_accuracy": 0.19474892169237137,
      "num_tokens": 60658020.0,
      "step": 31005
    },
    {
      "entropy": 5.893328428268433,
      "epoch": 3.0997650822212224,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040551333322541225,
      "loss": 5.3438,
      "mean_token_accuracy": 0.19843878000974655,
      "num_tokens": 60668330.0,
      "step": 31010
    },
    {
      "entropy": 5.939054250717163,
      "epoch": 3.100264907282451,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004054842601594512,
      "loss": 5.4604,
      "mean_token_accuracy": 0.19381338059902192,
      "num_tokens": 60679573.0,
      "step": 31015
    },
    {
      "entropy": 5.888349533081055,
      "epoch": 3.1007647323436798,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040545518381053964,
      "loss": 5.3423,
      "mean_token_accuracy": 0.2048489734530449,
      "num_tokens": 60689363.0,
      "step": 31020
    },
    {
      "entropy": 5.819430589675903,
      "epoch": 3.1012645574049085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040542610417940926,
      "loss": 5.3566,
      "mean_token_accuracy": 0.20405761152505875,
      "num_tokens": 60699120.0,
      "step": 31025
    },
    {
      "entropy": 5.879909515380859,
      "epoch": 3.1017643824661367,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004053970212667917,
      "loss": 5.4484,
      "mean_token_accuracy": 0.19755512177944184,
      "num_tokens": 60709235.0,
      "step": 31030
    },
    {
      "entropy": 5.910928964614868,
      "epoch": 3.1022642075273654,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004053679350734186,
      "loss": 5.3583,
      "mean_token_accuracy": 0.20097439736127853,
      "num_tokens": 60718674.0,
      "step": 31035
    },
    {
      "entropy": 5.8752528667449955,
      "epoch": 3.102764032588594,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040533884560002196,
      "loss": 5.3041,
      "mean_token_accuracy": 0.21184711754322053,
      "num_tokens": 60727456.0,
      "step": 31040
    },
    {
      "entropy": 5.834132289886474,
      "epoch": 3.1032638576498224,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040530975284733327,
      "loss": 5.3106,
      "mean_token_accuracy": 0.2000842347741127,
      "num_tokens": 60736556.0,
      "step": 31045
    },
    {
      "entropy": 5.92428879737854,
      "epoch": 3.103763682711051,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004052806568160849,
      "loss": 5.4211,
      "mean_token_accuracy": 0.19231498688459397,
      "num_tokens": 60746868.0,
      "step": 31050
    },
    {
      "entropy": 5.973662424087524,
      "epoch": 3.1042635077722798,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040525155750700863,
      "loss": 5.507,
      "mean_token_accuracy": 0.18637456148862838,
      "num_tokens": 60757200.0,
      "step": 31055
    },
    {
      "entropy": 5.859785366058349,
      "epoch": 3.1047633328335085,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040522245492083674,
      "loss": 5.3555,
      "mean_token_accuracy": 0.20441421121358871,
      "num_tokens": 60767966.0,
      "step": 31060
    },
    {
      "entropy": 5.876324987411499,
      "epoch": 3.1052631578947367,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040519334905830134,
      "loss": 5.347,
      "mean_token_accuracy": 0.20151012092828752,
      "num_tokens": 60777556.0,
      "step": 31065
    },
    {
      "entropy": 5.878439903259277,
      "epoch": 3.1057629829559654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004051642399201349,
      "loss": 5.3781,
      "mean_token_accuracy": 0.20267528742551805,
      "num_tokens": 60787601.0,
      "step": 31070
    },
    {
      "entropy": 5.859705448150635,
      "epoch": 3.106262808017194,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040513512750706956,
      "loss": 5.3836,
      "mean_token_accuracy": 0.20009636580944062,
      "num_tokens": 60797023.0,
      "step": 31075
    },
    {
      "entropy": 5.960652589797974,
      "epoch": 3.1067626330784224,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000405106011819838,
      "loss": 5.4581,
      "mean_token_accuracy": 0.19083200693130492,
      "num_tokens": 60807030.0,
      "step": 31080
    },
    {
      "entropy": 5.887543869018555,
      "epoch": 3.107262458139651,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040507689285917254,
      "loss": 5.4018,
      "mean_token_accuracy": 0.19352059662342072,
      "num_tokens": 60816397.0,
      "step": 31085
    },
    {
      "entropy": 5.9415404319763185,
      "epoch": 3.1077622832008798,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040504777062580597,
      "loss": 5.4645,
      "mean_token_accuracy": 0.19124845713376998,
      "num_tokens": 60826859.0,
      "step": 31090
    },
    {
      "entropy": 5.784214019775391,
      "epoch": 3.1082621082621085,
      "grad_norm": 1.109375,
      "learning_rate": 0.000405018645120471,
      "loss": 5.2898,
      "mean_token_accuracy": 0.20706502497196197,
      "num_tokens": 60837010.0,
      "step": 31095
    },
    {
      "entropy": 5.916188478469849,
      "epoch": 3.1087619333233367,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004049895163439003,
      "loss": 5.4353,
      "mean_token_accuracy": 0.18891495168209077,
      "num_tokens": 60845502.0,
      "step": 31100
    },
    {
      "entropy": 5.864137792587281,
      "epoch": 3.1092617583845654,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040496038429682684,
      "loss": 5.4102,
      "mean_token_accuracy": 0.20184454023838044,
      "num_tokens": 60856783.0,
      "step": 31105
    },
    {
      "entropy": 5.872903299331665,
      "epoch": 3.109761583445794,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040493124897998363,
      "loss": 5.4069,
      "mean_token_accuracy": 0.19539555311203002,
      "num_tokens": 60866560.0,
      "step": 31110
    },
    {
      "entropy": 5.891687202453613,
      "epoch": 3.1102614085070224,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004049021103941036,
      "loss": 5.4391,
      "mean_token_accuracy": 0.19235603213310243,
      "num_tokens": 60876677.0,
      "step": 31115
    },
    {
      "entropy": 5.915797138214112,
      "epoch": 3.110761233568251,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040487296853991984,
      "loss": 5.4344,
      "mean_token_accuracy": 0.1916615381836891,
      "num_tokens": 60886943.0,
      "step": 31120
    },
    {
      "entropy": 5.977438592910767,
      "epoch": 3.1112610586294798,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004048438234181657,
      "loss": 5.4419,
      "mean_token_accuracy": 0.1964563548564911,
      "num_tokens": 60895740.0,
      "step": 31125
    },
    {
      "entropy": 5.896838903427124,
      "epoch": 3.1117608836907085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040481467502957426,
      "loss": 5.35,
      "mean_token_accuracy": 0.20216663777828217,
      "num_tokens": 60905354.0,
      "step": 31130
    },
    {
      "entropy": 5.944007301330567,
      "epoch": 3.1122607087519367,
      "grad_norm": 1.0,
      "learning_rate": 0.000404785523374879,
      "loss": 5.4292,
      "mean_token_accuracy": 0.20091671496629715,
      "num_tokens": 60915040.0,
      "step": 31135
    },
    {
      "entropy": 5.897307014465332,
      "epoch": 3.1127605338131654,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040475636845481347,
      "loss": 5.4633,
      "mean_token_accuracy": 0.1917279064655304,
      "num_tokens": 60925023.0,
      "step": 31140
    },
    {
      "entropy": 5.896753454208374,
      "epoch": 3.113260358874394,
      "grad_norm": 1.125,
      "learning_rate": 0.00040472721027011107,
      "loss": 5.4346,
      "mean_token_accuracy": 0.19712019264698027,
      "num_tokens": 60935058.0,
      "step": 31145
    },
    {
      "entropy": 5.905637741088867,
      "epoch": 3.1137601839356224,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004046980488215054,
      "loss": 5.4187,
      "mean_token_accuracy": 0.20168666690587997,
      "num_tokens": 60945241.0,
      "step": 31150
    },
    {
      "entropy": 5.91046667098999,
      "epoch": 3.114260008996851,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040466888410973026,
      "loss": 5.4499,
      "mean_token_accuracy": 0.19078803062438965,
      "num_tokens": 60955402.0,
      "step": 31155
    },
    {
      "entropy": 5.9318835735321045,
      "epoch": 3.1147598340580798,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004046397161355193,
      "loss": 5.4052,
      "mean_token_accuracy": 0.1995738312602043,
      "num_tokens": 60964837.0,
      "step": 31160
    },
    {
      "entropy": 5.937003517150879,
      "epoch": 3.1152596591193085,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004046105448996064,
      "loss": 5.5346,
      "mean_token_accuracy": 0.1821996420621872,
      "num_tokens": 60975441.0,
      "step": 31165
    },
    {
      "entropy": 5.9309093952178955,
      "epoch": 3.1157594841805367,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004045813704027257,
      "loss": 5.4911,
      "mean_token_accuracy": 0.19197438955307006,
      "num_tokens": 60985883.0,
      "step": 31170
    },
    {
      "entropy": 5.879035186767578,
      "epoch": 3.1162593092417654,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004045521926456109,
      "loss": 5.3941,
      "mean_token_accuracy": 0.20283464044332505,
      "num_tokens": 60996545.0,
      "step": 31175
    },
    {
      "entropy": 5.8673358917236325,
      "epoch": 3.116759134302994,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004045230116289964,
      "loss": 5.3876,
      "mean_token_accuracy": 0.20300195664167403,
      "num_tokens": 61006805.0,
      "step": 31180
    },
    {
      "entropy": 5.930665063858032,
      "epoch": 3.1172589593642224,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040449382735361614,
      "loss": 5.4646,
      "mean_token_accuracy": 0.19342875480651855,
      "num_tokens": 61017579.0,
      "step": 31185
    },
    {
      "entropy": 5.946778011322022,
      "epoch": 3.117758784425451,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004044646398202046,
      "loss": 5.4463,
      "mean_token_accuracy": 0.19993554651737214,
      "num_tokens": 61027939.0,
      "step": 31190
    },
    {
      "entropy": 5.959183073043823,
      "epoch": 3.1182586094866798,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000404435449029496,
      "loss": 5.3751,
      "mean_token_accuracy": 0.20237241983413695,
      "num_tokens": 61038445.0,
      "step": 31195
    },
    {
      "entropy": 5.907608556747436,
      "epoch": 3.118758434547908,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004044062549822248,
      "loss": 5.4698,
      "mean_token_accuracy": 0.19439577758312226,
      "num_tokens": 61049434.0,
      "step": 31200
    },
    {
      "entropy": 5.9285307884216305,
      "epoch": 3.1192582596091367,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004043770576791255,
      "loss": 5.4772,
      "mean_token_accuracy": 0.19216534197330476,
      "num_tokens": 61058405.0,
      "step": 31205
    },
    {
      "entropy": 5.8879156589508055,
      "epoch": 3.1197580846703654,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004043478571209328,
      "loss": 5.3812,
      "mean_token_accuracy": 0.19719208776950836,
      "num_tokens": 61068639.0,
      "step": 31210
    },
    {
      "entropy": 5.925715684890747,
      "epoch": 3.120257909731594,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004043186533083812,
      "loss": 5.3736,
      "mean_token_accuracy": 0.20471004247665406,
      "num_tokens": 61078336.0,
      "step": 31215
    },
    {
      "entropy": 5.936952209472656,
      "epoch": 3.1207577347928224,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040428944624220557,
      "loss": 5.4475,
      "mean_token_accuracy": 0.19400928169488907,
      "num_tokens": 61087940.0,
      "step": 31220
    },
    {
      "entropy": 5.9015923023223875,
      "epoch": 3.121257559854051,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004042602359231409,
      "loss": 5.4205,
      "mean_token_accuracy": 0.1927705615758896,
      "num_tokens": 61098688.0,
      "step": 31225
    },
    {
      "entropy": 5.887043046951294,
      "epoch": 3.1217573849152798,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004042310223519218,
      "loss": 5.3799,
      "mean_token_accuracy": 0.20215093940496445,
      "num_tokens": 61108066.0,
      "step": 31230
    },
    {
      "entropy": 5.845997905731201,
      "epoch": 3.122257209976508,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004042018055292835,
      "loss": 5.3322,
      "mean_token_accuracy": 0.2043377935886383,
      "num_tokens": 61117009.0,
      "step": 31235
    },
    {
      "entropy": 5.9510396957397464,
      "epoch": 3.1227570350377367,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004041725854559609,
      "loss": 5.4381,
      "mean_token_accuracy": 0.19598832130432128,
      "num_tokens": 61127285.0,
      "step": 31240
    },
    {
      "entropy": 5.920957803726196,
      "epoch": 3.1232568600989654,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040414336213268943,
      "loss": 5.3831,
      "mean_token_accuracy": 0.20411514937877656,
      "num_tokens": 61136894.0,
      "step": 31245
    },
    {
      "entropy": 5.9627927303314205,
      "epoch": 3.123756685160194,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004041141355602041,
      "loss": 5.5703,
      "mean_token_accuracy": 0.19165478348731996,
      "num_tokens": 61147308.0,
      "step": 31250
    },
    {
      "entropy": 5.905159425735474,
      "epoch": 3.1242565102214224,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004040849057392405,
      "loss": 5.3995,
      "mean_token_accuracy": 0.19240445494651795,
      "num_tokens": 61157021.0,
      "step": 31255
    },
    {
      "entropy": 5.980018520355225,
      "epoch": 3.124756335282651,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004040556726705337,
      "loss": 5.486,
      "mean_token_accuracy": 0.19316006153821946,
      "num_tokens": 61166741.0,
      "step": 31260
    },
    {
      "entropy": 5.905357456207275,
      "epoch": 3.1252561603438798,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004040264363548195,
      "loss": 5.3226,
      "mean_token_accuracy": 0.19956647157669066,
      "num_tokens": 61175951.0,
      "step": 31265
    },
    {
      "entropy": 5.8778613090515135,
      "epoch": 3.125755985405108,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004039971967928333,
      "loss": 5.4033,
      "mean_token_accuracy": 0.19797088652849198,
      "num_tokens": 61185950.0,
      "step": 31270
    },
    {
      "entropy": 5.796696519851684,
      "epoch": 3.1262558104663367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004039679539853108,
      "loss": 5.2773,
      "mean_token_accuracy": 0.20973587036132812,
      "num_tokens": 61195305.0,
      "step": 31275
    },
    {
      "entropy": 5.843999719619751,
      "epoch": 3.1267556355275654,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004039387079329878,
      "loss": 5.4463,
      "mean_token_accuracy": 0.20293673127889633,
      "num_tokens": 61204640.0,
      "step": 31280
    },
    {
      "entropy": 5.832121801376343,
      "epoch": 3.127255460588794,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004039094586366001,
      "loss": 5.3367,
      "mean_token_accuracy": 0.2067544639110565,
      "num_tokens": 61214644.0,
      "step": 31285
    },
    {
      "entropy": 5.889100503921509,
      "epoch": 3.1277552856500224,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040388020609688366,
      "loss": 5.3444,
      "mean_token_accuracy": 0.20403718054294587,
      "num_tokens": 61223970.0,
      "step": 31290
    },
    {
      "entropy": 5.889707708358765,
      "epoch": 3.128255110711251,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040385095031457435,
      "loss": 5.4684,
      "mean_token_accuracy": 0.1966148167848587,
      "num_tokens": 61233503.0,
      "step": 31295
    },
    {
      "entropy": 5.914538097381592,
      "epoch": 3.1287549357724798,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004038216912904082,
      "loss": 5.3251,
      "mean_token_accuracy": 0.20055494457483292,
      "num_tokens": 61243342.0,
      "step": 31300
    },
    {
      "entropy": 5.787724542617798,
      "epoch": 3.129254760833708,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004037924290251215,
      "loss": 5.2638,
      "mean_token_accuracy": 0.20524709671735764,
      "num_tokens": 61253456.0,
      "step": 31305
    },
    {
      "entropy": 5.874935626983643,
      "epoch": 3.1297545858949367,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004037631635194505,
      "loss": 5.4621,
      "mean_token_accuracy": 0.19205792397260665,
      "num_tokens": 61262604.0,
      "step": 31310
    },
    {
      "entropy": 5.881028842926026,
      "epoch": 3.1302544109561654,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004037338947741313,
      "loss": 5.2874,
      "mean_token_accuracy": 0.20755355656147004,
      "num_tokens": 61272127.0,
      "step": 31315
    },
    {
      "entropy": 5.889629077911377,
      "epoch": 3.130754236017394,
      "grad_norm": 1.125,
      "learning_rate": 0.00040370462278990055,
      "loss": 5.4576,
      "mean_token_accuracy": 0.18948882967233657,
      "num_tokens": 61280808.0,
      "step": 31320
    },
    {
      "entropy": 5.914097738265991,
      "epoch": 3.1312540610786224,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004036753475674945,
      "loss": 5.4849,
      "mean_token_accuracy": 0.19450539201498032,
      "num_tokens": 61290047.0,
      "step": 31325
    },
    {
      "entropy": 6.013272619247436,
      "epoch": 3.131753886139851,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040364606910764996,
      "loss": 5.5171,
      "mean_token_accuracy": 0.19166036397218705,
      "num_tokens": 61300140.0,
      "step": 31330
    },
    {
      "entropy": 6.002668714523315,
      "epoch": 3.1322537112010798,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004036167874111034,
      "loss": 5.5711,
      "mean_token_accuracy": 0.18777998834848403,
      "num_tokens": 61310790.0,
      "step": 31335
    },
    {
      "entropy": 5.986793899536133,
      "epoch": 3.132753536262308,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040358750247859146,
      "loss": 5.4616,
      "mean_token_accuracy": 0.19541421085596083,
      "num_tokens": 61320835.0,
      "step": 31340
    },
    {
      "entropy": 5.875973081588745,
      "epoch": 3.1332533613235367,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004035582143108511,
      "loss": 5.3455,
      "mean_token_accuracy": 0.2053150936961174,
      "num_tokens": 61330585.0,
      "step": 31345
    },
    {
      "entropy": 5.765958976745606,
      "epoch": 3.1337531863847654,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040352892290861916,
      "loss": 5.3142,
      "mean_token_accuracy": 0.21691997945308686,
      "num_tokens": 61341089.0,
      "step": 31350
    },
    {
      "entropy": 5.835565423965454,
      "epoch": 3.134253011445994,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004034996282726326,
      "loss": 5.3884,
      "mean_token_accuracy": 0.19609957784414292,
      "num_tokens": 61349796.0,
      "step": 31355
    },
    {
      "entropy": 5.980464363098145,
      "epoch": 3.1347528365072224,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040347033040362835,
      "loss": 5.4945,
      "mean_token_accuracy": 0.1967511847615242,
      "num_tokens": 61359345.0,
      "step": 31360
    },
    {
      "entropy": 5.966249990463257,
      "epoch": 3.135252661568451,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040344102930234385,
      "loss": 5.4786,
      "mean_token_accuracy": 0.18660276681184768,
      "num_tokens": 61369462.0,
      "step": 31365
    },
    {
      "entropy": 5.90588526725769,
      "epoch": 3.1357524866296798,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040341172496951596,
      "loss": 5.3866,
      "mean_token_accuracy": 0.19619600474834442,
      "num_tokens": 61378170.0,
      "step": 31370
    },
    {
      "entropy": 5.8847582817077635,
      "epoch": 3.136252311690908,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004033824174058821,
      "loss": 5.4618,
      "mean_token_accuracy": 0.20226778388023375,
      "num_tokens": 61388307.0,
      "step": 31375
    },
    {
      "entropy": 5.916901063919068,
      "epoch": 3.1367521367521367,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004033531066121797,
      "loss": 5.3753,
      "mean_token_accuracy": 0.19775373935699464,
      "num_tokens": 61398351.0,
      "step": 31380
    },
    {
      "entropy": 5.893314695358276,
      "epoch": 3.1372519618133654,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040332379258914614,
      "loss": 5.3704,
      "mean_token_accuracy": 0.20333703607320786,
      "num_tokens": 61406893.0,
      "step": 31385
    },
    {
      "entropy": 5.960612869262695,
      "epoch": 3.1377517868745937,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000403294475337519,
      "loss": 5.5067,
      "mean_token_accuracy": 0.1944995865225792,
      "num_tokens": 61416655.0,
      "step": 31390
    },
    {
      "entropy": 5.858287000656128,
      "epoch": 3.1382516119358224,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004032651548580359,
      "loss": 5.3624,
      "mean_token_accuracy": 0.20350971817970276,
      "num_tokens": 61425655.0,
      "step": 31395
    },
    {
      "entropy": 5.8316850662231445,
      "epoch": 3.138751436997051,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040323583115143446,
      "loss": 5.3014,
      "mean_token_accuracy": 0.2098345384001732,
      "num_tokens": 61435820.0,
      "step": 31400
    },
    {
      "entropy": 5.896674251556396,
      "epoch": 3.1392512620582798,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040320650421845256,
      "loss": 5.4029,
      "mean_token_accuracy": 0.2033734366297722,
      "num_tokens": 61445906.0,
      "step": 31405
    },
    {
      "entropy": 5.8794270038604735,
      "epoch": 3.139751087119508,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040317717405982797,
      "loss": 5.3549,
      "mean_token_accuracy": 0.19691593647003175,
      "num_tokens": 61455237.0,
      "step": 31410
    },
    {
      "entropy": 5.956326866149903,
      "epoch": 3.1402509121807367,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040314784067629867,
      "loss": 5.4632,
      "mean_token_accuracy": 0.19288019239902496,
      "num_tokens": 61464735.0,
      "step": 31415
    },
    {
      "entropy": 5.940885734558106,
      "epoch": 3.1407507372419654,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040311850406860267,
      "loss": 5.4551,
      "mean_token_accuracy": 0.19600463509559632,
      "num_tokens": 61475187.0,
      "step": 31420
    },
    {
      "entropy": 5.901913118362427,
      "epoch": 3.1412505623031937,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040308916423747803,
      "loss": 5.4292,
      "mean_token_accuracy": 0.20002886652946472,
      "num_tokens": 61485232.0,
      "step": 31425
    },
    {
      "entropy": 6.008521318435669,
      "epoch": 3.1417503873644224,
      "grad_norm": 1.125,
      "learning_rate": 0.000403059821183663,
      "loss": 5.6118,
      "mean_token_accuracy": 0.19140280336141585,
      "num_tokens": 61495601.0,
      "step": 31430
    },
    {
      "entropy": 5.907225179672241,
      "epoch": 3.142250212425651,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004030304749078959,
      "loss": 5.3745,
      "mean_token_accuracy": 0.1985194444656372,
      "num_tokens": 61506659.0,
      "step": 31435
    },
    {
      "entropy": 5.825550556182861,
      "epoch": 3.1427500374868798,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040300112541091494,
      "loss": 5.3626,
      "mean_token_accuracy": 0.20357151925563813,
      "num_tokens": 61517376.0,
      "step": 31440
    },
    {
      "entropy": 5.910529327392578,
      "epoch": 3.143249862548108,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004029717726934586,
      "loss": 5.3511,
      "mean_token_accuracy": 0.2006275862455368,
      "num_tokens": 61527001.0,
      "step": 31445
    },
    {
      "entropy": 5.960806894302368,
      "epoch": 3.1437496876093367,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040294241675626534,
      "loss": 5.4726,
      "mean_token_accuracy": 0.19348615705966948,
      "num_tokens": 61537554.0,
      "step": 31450
    },
    {
      "entropy": 5.956271505355835,
      "epoch": 3.1442495126705654,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004029130576000739,
      "loss": 5.4209,
      "mean_token_accuracy": 0.19936157315969466,
      "num_tokens": 61546457.0,
      "step": 31455
    },
    {
      "entropy": 5.889905977249145,
      "epoch": 3.1447493377317937,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004028836952256227,
      "loss": 5.4632,
      "mean_token_accuracy": 0.18930587023496628,
      "num_tokens": 61555872.0,
      "step": 31460
    },
    {
      "entropy": 5.926950550079345,
      "epoch": 3.1452491627930224,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004028543296336508,
      "loss": 5.378,
      "mean_token_accuracy": 0.20449561923742293,
      "num_tokens": 61565820.0,
      "step": 31465
    },
    {
      "entropy": 5.92706184387207,
      "epoch": 3.145748987854251,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040282496082489684,
      "loss": 5.439,
      "mean_token_accuracy": 0.20075054913759233,
      "num_tokens": 61575344.0,
      "step": 31470
    },
    {
      "entropy": 5.84251766204834,
      "epoch": 3.1462488129154798,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040279558880009967,
      "loss": 5.3338,
      "mean_token_accuracy": 0.21501011699438094,
      "num_tokens": 61584906.0,
      "step": 31475
    },
    {
      "entropy": 5.757793855667114,
      "epoch": 3.146748637976708,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004027662135599984,
      "loss": 5.2884,
      "mean_token_accuracy": 0.20900812447071077,
      "num_tokens": 61594490.0,
      "step": 31480
    },
    {
      "entropy": 5.8919414520263675,
      "epoch": 3.1472484630379367,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004027368351053321,
      "loss": 5.3661,
      "mean_token_accuracy": 0.20171024054288864,
      "num_tokens": 61603280.0,
      "step": 31485
    },
    {
      "entropy": 5.913060140609741,
      "epoch": 3.1477482880991654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004027074534368399,
      "loss": 5.4046,
      "mean_token_accuracy": 0.206545490026474,
      "num_tokens": 61612474.0,
      "step": 31490
    },
    {
      "entropy": 5.769423103332519,
      "epoch": 3.1482481131603937,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040267806855526103,
      "loss": 5.2231,
      "mean_token_accuracy": 0.2131379321217537,
      "num_tokens": 61622719.0,
      "step": 31495
    },
    {
      "entropy": 5.781323862075806,
      "epoch": 3.1487479382216224,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004026486804613347,
      "loss": 5.3705,
      "mean_token_accuracy": 0.20436352342367173,
      "num_tokens": 61633171.0,
      "step": 31500
    },
    {
      "entropy": 5.844699335098267,
      "epoch": 3.149247763282851,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040261928915580065,
      "loss": 5.3086,
      "mean_token_accuracy": 0.20366631150245668,
      "num_tokens": 61642437.0,
      "step": 31505
    },
    {
      "entropy": 5.920159244537354,
      "epoch": 3.1497475883440798,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004025898946393979,
      "loss": 5.5246,
      "mean_token_accuracy": 0.18515994101762773,
      "num_tokens": 61651884.0,
      "step": 31510
    },
    {
      "entropy": 5.921646308898926,
      "epoch": 3.150247413405308,
      "grad_norm": 1.0,
      "learning_rate": 0.0004025604969128663,
      "loss": 5.4555,
      "mean_token_accuracy": 0.19429163485765458,
      "num_tokens": 61661821.0,
      "step": 31515
    },
    {
      "entropy": 5.90688009262085,
      "epoch": 3.1507472384665367,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040253109597694544,
      "loss": 5.3769,
      "mean_token_accuracy": 0.2033492624759674,
      "num_tokens": 61672363.0,
      "step": 31520
    },
    {
      "entropy": 5.897107553482056,
      "epoch": 3.1512470635277654,
      "grad_norm": 1.125,
      "learning_rate": 0.00040250169183237507,
      "loss": 5.3771,
      "mean_token_accuracy": 0.20613158345222474,
      "num_tokens": 61681318.0,
      "step": 31525
    },
    {
      "entropy": 5.890831851959229,
      "epoch": 3.1517468885889937,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040247228447989485,
      "loss": 5.4345,
      "mean_token_accuracy": 0.1924078270792961,
      "num_tokens": 61690710.0,
      "step": 31530
    },
    {
      "entropy": 5.932021903991699,
      "epoch": 3.1522467136502224,
      "grad_norm": 1.484375,
      "learning_rate": 0.00040244287392024485,
      "loss": 5.4053,
      "mean_token_accuracy": 0.20523335337638854,
      "num_tokens": 61700913.0,
      "step": 31535
    },
    {
      "entropy": 5.93386869430542,
      "epoch": 3.152746538711451,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004024134601541649,
      "loss": 5.3604,
      "mean_token_accuracy": 0.20059905350208282,
      "num_tokens": 61710905.0,
      "step": 31540
    },
    {
      "entropy": 5.826361608505249,
      "epoch": 3.1532463637726798,
      "grad_norm": 1.125,
      "learning_rate": 0.0004023840431823951,
      "loss": 5.3401,
      "mean_token_accuracy": 0.2075880989432335,
      "num_tokens": 61719952.0,
      "step": 31545
    },
    {
      "entropy": 5.885459136962891,
      "epoch": 3.153746188833908,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004023546230056754,
      "loss": 5.3387,
      "mean_token_accuracy": 0.19872013330459595,
      "num_tokens": 61728948.0,
      "step": 31550
    },
    {
      "entropy": 5.9987101554870605,
      "epoch": 3.1542460138951367,
      "grad_norm": 1.25,
      "learning_rate": 0.00040232519962474635,
      "loss": 5.5318,
      "mean_token_accuracy": 0.18874025344848633,
      "num_tokens": 61738235.0,
      "step": 31555
    },
    {
      "entropy": 5.983381843566894,
      "epoch": 3.1547458389563654,
      "grad_norm": 1.25,
      "learning_rate": 0.00040229577304034793,
      "loss": 5.4472,
      "mean_token_accuracy": 0.19790114611387252,
      "num_tokens": 61746934.0,
      "step": 31560
    },
    {
      "entropy": 5.908118057250976,
      "epoch": 3.1552456640175937,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040226634325322063,
      "loss": 5.42,
      "mean_token_accuracy": 0.198232701420784,
      "num_tokens": 61756346.0,
      "step": 31565
    },
    {
      "entropy": 5.876107501983642,
      "epoch": 3.1557454890788224,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004022369102641049,
      "loss": 5.4712,
      "mean_token_accuracy": 0.19442981034517287,
      "num_tokens": 61765900.0,
      "step": 31570
    },
    {
      "entropy": 5.9813378810882565,
      "epoch": 3.156245314140051,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040220747407374133,
      "loss": 5.4643,
      "mean_token_accuracy": 0.19483010172843934,
      "num_tokens": 61775657.0,
      "step": 31575
    },
    {
      "entropy": 5.965229415893555,
      "epoch": 3.1567451392012797,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040217803468287034,
      "loss": 5.4187,
      "mean_token_accuracy": 0.20354881882667542,
      "num_tokens": 61784977.0,
      "step": 31580
    },
    {
      "entropy": 5.909129047393799,
      "epoch": 3.157244964262508,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040214859209223273,
      "loss": 5.3658,
      "mean_token_accuracy": 0.20080097764730453,
      "num_tokens": 61794086.0,
      "step": 31585
    },
    {
      "entropy": 5.837738513946533,
      "epoch": 3.1577447893237367,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004021191463025693,
      "loss": 5.3778,
      "mean_token_accuracy": 0.19607591331005098,
      "num_tokens": 61803737.0,
      "step": 31590
    },
    {
      "entropy": 5.863025569915772,
      "epoch": 3.1582446143849654,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004020896973146208,
      "loss": 5.4179,
      "mean_token_accuracy": 0.19348090142011642,
      "num_tokens": 61814078.0,
      "step": 31595
    },
    {
      "entropy": 5.937948274612427,
      "epoch": 3.1587444394461937,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040206024512912824,
      "loss": 5.3877,
      "mean_token_accuracy": 0.19722219407558442,
      "num_tokens": 61823680.0,
      "step": 31600
    },
    {
      "entropy": 5.911126518249512,
      "epoch": 3.1592442645074224,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040203078974683265,
      "loss": 5.4197,
      "mean_token_accuracy": 0.1953892469406128,
      "num_tokens": 61833704.0,
      "step": 31605
    },
    {
      "entropy": 5.929960918426514,
      "epoch": 3.159744089568651,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004020013311684751,
      "loss": 5.4392,
      "mean_token_accuracy": 0.20006785690784454,
      "num_tokens": 61842402.0,
      "step": 31610
    },
    {
      "entropy": 5.979666805267334,
      "epoch": 3.1602439146298797,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040197186939479665,
      "loss": 5.4354,
      "mean_token_accuracy": 0.1886878490447998,
      "num_tokens": 61852184.0,
      "step": 31615
    },
    {
      "entropy": 5.9974809169769285,
      "epoch": 3.160743739691108,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040194240442653867,
      "loss": 5.459,
      "mean_token_accuracy": 0.18850130289793016,
      "num_tokens": 61861260.0,
      "step": 31620
    },
    {
      "entropy": 5.900859785079956,
      "epoch": 3.1612435647523367,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004019129362644425,
      "loss": 5.3469,
      "mean_token_accuracy": 0.20431291610002517,
      "num_tokens": 61871131.0,
      "step": 31625
    },
    {
      "entropy": 5.8806251049041744,
      "epoch": 3.1617433898135654,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004018834649092494,
      "loss": 5.3912,
      "mean_token_accuracy": 0.19794175922870635,
      "num_tokens": 61881826.0,
      "step": 31630
    },
    {
      "entropy": 5.928707551956177,
      "epoch": 3.1622432148747937,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004018539903617011,
      "loss": 5.3733,
      "mean_token_accuracy": 0.1956825241446495,
      "num_tokens": 61891253.0,
      "step": 31635
    },
    {
      "entropy": 5.852584457397461,
      "epoch": 3.1627430399360223,
      "grad_norm": 1.125,
      "learning_rate": 0.00040182451262253904,
      "loss": 5.3452,
      "mean_token_accuracy": 0.20168899446725846,
      "num_tokens": 61901054.0,
      "step": 31640
    },
    {
      "entropy": 5.898493146896362,
      "epoch": 3.163242864997251,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004017950316925049,
      "loss": 5.316,
      "mean_token_accuracy": 0.2005171224474907,
      "num_tokens": 61912298.0,
      "step": 31645
    },
    {
      "entropy": 5.89299783706665,
      "epoch": 3.1637426900584797,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040176554757234017,
      "loss": 5.3993,
      "mean_token_accuracy": 0.19533049911260605,
      "num_tokens": 61921449.0,
      "step": 31650
    },
    {
      "entropy": 5.820459079742432,
      "epoch": 3.164242515119708,
      "grad_norm": 1.125,
      "learning_rate": 0.0004017360602627871,
      "loss": 5.3358,
      "mean_token_accuracy": 0.2041382297873497,
      "num_tokens": 61931220.0,
      "step": 31655
    },
    {
      "entropy": 6.012273216247559,
      "epoch": 3.1647423401809367,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040170656976458723,
      "loss": 5.5396,
      "mean_token_accuracy": 0.19422822147607804,
      "num_tokens": 61940981.0,
      "step": 31660
    },
    {
      "entropy": 6.015304231643677,
      "epoch": 3.1652421652421654,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040167707607848273,
      "loss": 5.487,
      "mean_token_accuracy": 0.19495080262422562,
      "num_tokens": 61950632.0,
      "step": 31665
    },
    {
      "entropy": 5.8706934452056885,
      "epoch": 3.1657419903033936,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004016475792052157,
      "loss": 5.3716,
      "mean_token_accuracy": 0.20254748016595842,
      "num_tokens": 61960652.0,
      "step": 31670
    },
    {
      "entropy": 5.948014593124389,
      "epoch": 3.1662418153646223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000401618079145528,
      "loss": 5.454,
      "mean_token_accuracy": 0.18853966146707535,
      "num_tokens": 61969480.0,
      "step": 31675
    },
    {
      "entropy": 5.857546997070313,
      "epoch": 3.166741640425851,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004015885759001621,
      "loss": 5.416,
      "mean_token_accuracy": 0.20330006033182144,
      "num_tokens": 61979164.0,
      "step": 31680
    },
    {
      "entropy": 5.853503799438476,
      "epoch": 3.1672414654870797,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040155906946986027,
      "loss": 5.3726,
      "mean_token_accuracy": 0.20030351728200912,
      "num_tokens": 61990513.0,
      "step": 31685
    },
    {
      "entropy": 5.951899528503418,
      "epoch": 3.167741290548308,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004015295598553647,
      "loss": 5.418,
      "mean_token_accuracy": 0.19239617139101028,
      "num_tokens": 61999599.0,
      "step": 31690
    },
    {
      "entropy": 5.889421272277832,
      "epoch": 3.1682411156095367,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004015000470574179,
      "loss": 5.4661,
      "mean_token_accuracy": 0.198386487364769,
      "num_tokens": 62009548.0,
      "step": 31695
    },
    {
      "entropy": 5.9249794483184814,
      "epoch": 3.1687409406707654,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040147053107676266,
      "loss": 5.4244,
      "mean_token_accuracy": 0.19121869653463364,
      "num_tokens": 62018536.0,
      "step": 31700
    },
    {
      "entropy": 5.9333662509918215,
      "epoch": 3.1692407657319936,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040144101191414127,
      "loss": 5.5074,
      "mean_token_accuracy": 0.19443974494934083,
      "num_tokens": 62027991.0,
      "step": 31705
    },
    {
      "entropy": 5.874882078170776,
      "epoch": 3.1697405907932223,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004014114895702966,
      "loss": 5.393,
      "mean_token_accuracy": 0.20078718513250352,
      "num_tokens": 62038554.0,
      "step": 31710
    },
    {
      "entropy": 5.934776544570923,
      "epoch": 3.170240415854451,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004013819640459714,
      "loss": 5.3621,
      "mean_token_accuracy": 0.2054046392440796,
      "num_tokens": 62048060.0,
      "step": 31715
    },
    {
      "entropy": 5.929137134552002,
      "epoch": 3.1707402409156797,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040135243534190847,
      "loss": 5.4212,
      "mean_token_accuracy": 0.19577299654483796,
      "num_tokens": 62057905.0,
      "step": 31720
    },
    {
      "entropy": 5.9197108268737795,
      "epoch": 3.171240065976908,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040132290345885077,
      "loss": 5.4109,
      "mean_token_accuracy": 0.19693997651338577,
      "num_tokens": 62067022.0,
      "step": 31725
    },
    {
      "entropy": 5.9490580558776855,
      "epoch": 3.1717398910381367,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004012933683975413,
      "loss": 5.43,
      "mean_token_accuracy": 0.19537671506404877,
      "num_tokens": 62075696.0,
      "step": 31730
    },
    {
      "entropy": 5.872826862335205,
      "epoch": 3.1722397160993654,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040126383015872324,
      "loss": 5.4275,
      "mean_token_accuracy": 0.19298358559608458,
      "num_tokens": 62085620.0,
      "step": 31735
    },
    {
      "entropy": 5.946121263504028,
      "epoch": 3.1727395411605936,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040123428874313974,
      "loss": 5.4172,
      "mean_token_accuracy": 0.19369582533836366,
      "num_tokens": 62094531.0,
      "step": 31740
    },
    {
      "entropy": 6.022593402862549,
      "epoch": 3.1732393662218223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040120474415153396,
      "loss": 5.5358,
      "mean_token_accuracy": 0.18564691841602327,
      "num_tokens": 62104036.0,
      "step": 31745
    },
    {
      "entropy": 5.9524585723876955,
      "epoch": 3.173739191283051,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040117519638464935,
      "loss": 5.3953,
      "mean_token_accuracy": 0.19598254114389418,
      "num_tokens": 62113772.0,
      "step": 31750
    },
    {
      "entropy": 5.847739362716675,
      "epoch": 3.1742390163442793,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040114564544322913,
      "loss": 5.3563,
      "mean_token_accuracy": 0.19879276305437088,
      "num_tokens": 62123237.0,
      "step": 31755
    },
    {
      "entropy": 5.861629247665405,
      "epoch": 3.174738841405508,
      "grad_norm": 1.140625,
      "learning_rate": 0.000401116091328017,
      "loss": 5.3792,
      "mean_token_accuracy": 0.2029305577278137,
      "num_tokens": 62132391.0,
      "step": 31760
    },
    {
      "entropy": 5.804901838302612,
      "epoch": 3.1752386664667367,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040108653403975647,
      "loss": 5.2679,
      "mean_token_accuracy": 0.215373133122921,
      "num_tokens": 62141808.0,
      "step": 31765
    },
    {
      "entropy": 5.7167243480682375,
      "epoch": 3.1757384915279654,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004010569735791912,
      "loss": 5.0955,
      "mean_token_accuracy": 0.22823552489280702,
      "num_tokens": 62151065.0,
      "step": 31770
    },
    {
      "entropy": 5.899328422546387,
      "epoch": 3.1762383165891936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004010274099470649,
      "loss": 5.3692,
      "mean_token_accuracy": 0.20170472264289857,
      "num_tokens": 62160885.0,
      "step": 31775
    },
    {
      "entropy": 5.7696123123168945,
      "epoch": 3.1767381416504223,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004009978431441214,
      "loss": 5.2745,
      "mean_token_accuracy": 0.21055861860513686,
      "num_tokens": 62169917.0,
      "step": 31780
    },
    {
      "entropy": 5.833308458328247,
      "epoch": 3.177237966711651,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040096827317110463,
      "loss": 5.3709,
      "mean_token_accuracy": 0.19654978215694427,
      "num_tokens": 62179896.0,
      "step": 31785
    },
    {
      "entropy": 5.855351972579956,
      "epoch": 3.1777377917728793,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004009387000287585,
      "loss": 5.3911,
      "mean_token_accuracy": 0.19610824435949326,
      "num_tokens": 62189378.0,
      "step": 31790
    },
    {
      "entropy": 5.929579019546509,
      "epoch": 3.178237616834108,
      "grad_norm": 1.0625,
      "learning_rate": 0.000400909123717827,
      "loss": 5.4457,
      "mean_token_accuracy": 0.1966703489422798,
      "num_tokens": 62198602.0,
      "step": 31795
    },
    {
      "entropy": 5.928867340087891,
      "epoch": 3.1787374418953367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004008795442390544,
      "loss": 5.4079,
      "mean_token_accuracy": 0.1969139650464058,
      "num_tokens": 62206989.0,
      "step": 31800
    },
    {
      "entropy": 5.902009057998657,
      "epoch": 3.1792372669565654,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004008499615931849,
      "loss": 5.4171,
      "mean_token_accuracy": 0.2004924461245537,
      "num_tokens": 62216922.0,
      "step": 31805
    },
    {
      "entropy": 6.013360404968262,
      "epoch": 3.1797370920177936,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004008203757809627,
      "loss": 5.441,
      "mean_token_accuracy": 0.1899332508444786,
      "num_tokens": 62225946.0,
      "step": 31810
    },
    {
      "entropy": 5.891789484024048,
      "epoch": 3.1802369170790223,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040079078680313224,
      "loss": 5.4084,
      "mean_token_accuracy": 0.19616751819849015,
      "num_tokens": 62236464.0,
      "step": 31815
    },
    {
      "entropy": 5.938845682144165,
      "epoch": 3.180736742140251,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000400761194660438,
      "loss": 5.4194,
      "mean_token_accuracy": 0.19428257495164872,
      "num_tokens": 62245789.0,
      "step": 31820
    },
    {
      "entropy": 5.846794319152832,
      "epoch": 3.1812365672014793,
      "grad_norm": 1.125,
      "learning_rate": 0.0004007315993536244,
      "loss": 5.3668,
      "mean_token_accuracy": 0.19534220099449157,
      "num_tokens": 62254843.0,
      "step": 31825
    },
    {
      "entropy": 5.861235094070435,
      "epoch": 3.181736392262708,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040070200088343613,
      "loss": 5.3937,
      "mean_token_accuracy": 0.19619150310754777,
      "num_tokens": 62265239.0,
      "step": 31830
    },
    {
      "entropy": 5.8682867050170895,
      "epoch": 3.1822362173239367,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040067239925061784,
      "loss": 5.3468,
      "mean_token_accuracy": 0.2050875887274742,
      "num_tokens": 62275496.0,
      "step": 31835
    },
    {
      "entropy": 5.918748617172241,
      "epoch": 3.1827360423851654,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004006427944559144,
      "loss": 5.4647,
      "mean_token_accuracy": 0.19526555985212327,
      "num_tokens": 62286071.0,
      "step": 31840
    },
    {
      "entropy": 5.870143699645996,
      "epoch": 3.1832358674463936,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040061318650007045,
      "loss": 5.3769,
      "mean_token_accuracy": 0.20099104195833206,
      "num_tokens": 62295242.0,
      "step": 31845
    },
    {
      "entropy": 5.859804773330689,
      "epoch": 3.1837356925076223,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040058357538383117,
      "loss": 5.3682,
      "mean_token_accuracy": 0.20648722648620604,
      "num_tokens": 62305546.0,
      "step": 31850
    },
    {
      "entropy": 5.895138168334961,
      "epoch": 3.184235517568851,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004005539611079414,
      "loss": 5.4307,
      "mean_token_accuracy": 0.20017442107200623,
      "num_tokens": 62315181.0,
      "step": 31855
    },
    {
      "entropy": 5.941577482223511,
      "epoch": 3.1847353426300793,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040052434367314634,
      "loss": 5.3846,
      "mean_token_accuracy": 0.19742350727319719,
      "num_tokens": 62324741.0,
      "step": 31860
    },
    {
      "entropy": 5.889970588684082,
      "epoch": 3.185235167691308,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040049472308019104,
      "loss": 5.3949,
      "mean_token_accuracy": 0.20524685829877853,
      "num_tokens": 62334637.0,
      "step": 31865
    },
    {
      "entropy": 6.058261919021606,
      "epoch": 3.1857349927525367,
      "grad_norm": 1.125,
      "learning_rate": 0.00040046509932982076,
      "loss": 5.5798,
      "mean_token_accuracy": 0.18533055186271669,
      "num_tokens": 62344637.0,
      "step": 31870
    },
    {
      "entropy": 5.889399909973145,
      "epoch": 3.1862348178137654,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004004354724227809,
      "loss": 5.3789,
      "mean_token_accuracy": 0.19738575220108032,
      "num_tokens": 62353867.0,
      "step": 31875
    },
    {
      "entropy": 5.76851224899292,
      "epoch": 3.1867346428749936,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004004058423598168,
      "loss": 5.2819,
      "mean_token_accuracy": 0.21200910806655884,
      "num_tokens": 62364729.0,
      "step": 31880
    },
    {
      "entropy": 5.91595983505249,
      "epoch": 3.1872344679362223,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040037620914167406,
      "loss": 5.4378,
      "mean_token_accuracy": 0.19334245920181276,
      "num_tokens": 62374295.0,
      "step": 31885
    },
    {
      "entropy": 5.924029541015625,
      "epoch": 3.187734292997451,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004003465727690981,
      "loss": 5.4484,
      "mean_token_accuracy": 0.19801906496286392,
      "num_tokens": 62383276.0,
      "step": 31890
    },
    {
      "entropy": 5.938668918609619,
      "epoch": 3.1882341180586793,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040031693324283464,
      "loss": 5.441,
      "mean_token_accuracy": 0.19811128824949265,
      "num_tokens": 62392473.0,
      "step": 31895
    },
    {
      "entropy": 5.9780967235565186,
      "epoch": 3.188733943119908,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004002872905636294,
      "loss": 5.5766,
      "mean_token_accuracy": 0.18453899174928665,
      "num_tokens": 62402127.0,
      "step": 31900
    },
    {
      "entropy": 5.992673635482788,
      "epoch": 3.1892337681811367,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040025764473222816,
      "loss": 5.4684,
      "mean_token_accuracy": 0.1915650948882103,
      "num_tokens": 62411757.0,
      "step": 31905
    },
    {
      "entropy": 5.735714673995972,
      "epoch": 3.189733593242365,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040022799574937674,
      "loss": 5.2221,
      "mean_token_accuracy": 0.21267260760068893,
      "num_tokens": 62422446.0,
      "step": 31910
    },
    {
      "entropy": 5.851648330688477,
      "epoch": 3.1902334183035936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040019834361582123,
      "loss": 5.3209,
      "mean_token_accuracy": 0.21431685984134674,
      "num_tokens": 62430970.0,
      "step": 31915
    },
    {
      "entropy": 5.8743876934051515,
      "epoch": 3.1907332433648223,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040016868833230766,
      "loss": 5.3752,
      "mean_token_accuracy": 0.20129471719264985,
      "num_tokens": 62440666.0,
      "step": 31920
    },
    {
      "entropy": 5.9637572288513185,
      "epoch": 3.191233068426051,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004001390298995821,
      "loss": 5.4727,
      "mean_token_accuracy": 0.1973235785961151,
      "num_tokens": 62450008.0,
      "step": 31925
    },
    {
      "entropy": 5.777582454681396,
      "epoch": 3.1917328934872793,
      "grad_norm": 1.25,
      "learning_rate": 0.00040010936831839065,
      "loss": 5.2499,
      "mean_token_accuracy": 0.21071867644786835,
      "num_tokens": 62459225.0,
      "step": 31930
    },
    {
      "entropy": 5.906385612487793,
      "epoch": 3.192232718548508,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040007970358947977,
      "loss": 5.4028,
      "mean_token_accuracy": 0.19360157400369643,
      "num_tokens": 62469094.0,
      "step": 31935
    },
    {
      "entropy": 5.861662340164185,
      "epoch": 3.1927325436097367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040005003571359565,
      "loss": 5.3696,
      "mean_token_accuracy": 0.19637318402528764,
      "num_tokens": 62479308.0,
      "step": 31940
    },
    {
      "entropy": 5.79908447265625,
      "epoch": 3.193232368670965,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040002036469148486,
      "loss": 5.2854,
      "mean_token_accuracy": 0.2120439440011978,
      "num_tokens": 62489758.0,
      "step": 31945
    },
    {
      "entropy": 5.9343267440795895,
      "epoch": 3.1937321937321936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039999069052389384,
      "loss": 5.4667,
      "mean_token_accuracy": 0.19174213856458663,
      "num_tokens": 62499084.0,
      "step": 31950
    },
    {
      "entropy": 5.905474615097046,
      "epoch": 3.1942320187934223,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039996101321156926,
      "loss": 5.4075,
      "mean_token_accuracy": 0.19589445143938064,
      "num_tokens": 62508054.0,
      "step": 31955
    },
    {
      "entropy": 5.923503446578979,
      "epoch": 3.194731843854651,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039993133275525777,
      "loss": 5.4103,
      "mean_token_accuracy": 0.19430425614118577,
      "num_tokens": 62517823.0,
      "step": 31960
    },
    {
      "entropy": 5.928025245666504,
      "epoch": 3.1952316689158793,
      "grad_norm": 1.046875,
      "learning_rate": 0.000399901649155706,
      "loss": 5.4544,
      "mean_token_accuracy": 0.1941968560218811,
      "num_tokens": 62528579.0,
      "step": 31965
    },
    {
      "entropy": 5.979582691192627,
      "epoch": 3.195731493977108,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039987196241366086,
      "loss": 5.4962,
      "mean_token_accuracy": 0.19058964401483536,
      "num_tokens": 62539682.0,
      "step": 31970
    },
    {
      "entropy": 6.006564664840698,
      "epoch": 3.1962313190383367,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039984227252986936,
      "loss": 5.5363,
      "mean_token_accuracy": 0.18488672524690627,
      "num_tokens": 62549431.0,
      "step": 31975
    },
    {
      "entropy": 5.944624996185302,
      "epoch": 3.196731144099565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039981257950507835,
      "loss": 5.3901,
      "mean_token_accuracy": 0.19986187070608138,
      "num_tokens": 62559372.0,
      "step": 31980
    },
    {
      "entropy": 5.859975147247314,
      "epoch": 3.1972309691607936,
      "grad_norm": 1.125,
      "learning_rate": 0.000399782883340035,
      "loss": 5.3571,
      "mean_token_accuracy": 0.20368506163358688,
      "num_tokens": 62568415.0,
      "step": 31985
    },
    {
      "entropy": 5.912347936630249,
      "epoch": 3.1977307942220223,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003997531840354865,
      "loss": 5.4805,
      "mean_token_accuracy": 0.18931994289159776,
      "num_tokens": 62578419.0,
      "step": 31990
    },
    {
      "entropy": 5.96358323097229,
      "epoch": 3.198230619283251,
      "grad_norm": 1.25,
      "learning_rate": 0.00039972348159217987,
      "loss": 5.3417,
      "mean_token_accuracy": 0.2065069019794464,
      "num_tokens": 62587408.0,
      "step": 31995
    },
    {
      "entropy": 5.872853994369507,
      "epoch": 3.1987304443444793,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039969377601086257,
      "loss": 5.3766,
      "mean_token_accuracy": 0.20528334081172944,
      "num_tokens": 62598753.0,
      "step": 32000
    },
    {
      "entropy": 5.867677974700928,
      "epoch": 3.199230269405708,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039966406729228187,
      "loss": 5.3512,
      "mean_token_accuracy": 0.20950085967779158,
      "num_tokens": 62608700.0,
      "step": 32005
    },
    {
      "entropy": 5.831635427474976,
      "epoch": 3.1997300944669367,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039963435543718544,
      "loss": 5.3432,
      "mean_token_accuracy": 0.20486099570989608,
      "num_tokens": 62618870.0,
      "step": 32010
    },
    {
      "entropy": 5.849548387527466,
      "epoch": 3.200229919528165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039960464044632067,
      "loss": 5.3614,
      "mean_token_accuracy": 0.21358303278684615,
      "num_tokens": 62629681.0,
      "step": 32015
    },
    {
      "entropy": 5.95084023475647,
      "epoch": 3.2007297445893936,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003995749223204351,
      "loss": 5.4749,
      "mean_token_accuracy": 0.19279395341873168,
      "num_tokens": 62638792.0,
      "step": 32020
    },
    {
      "entropy": 5.891513299942017,
      "epoch": 3.2012295696506223,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003995452010602767,
      "loss": 5.3639,
      "mean_token_accuracy": 0.21082978397607804,
      "num_tokens": 62648353.0,
      "step": 32025
    },
    {
      "entropy": 5.877854108810425,
      "epoch": 3.201729394711851,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039951547666659305,
      "loss": 5.3945,
      "mean_token_accuracy": 0.19529493451118468,
      "num_tokens": 62657585.0,
      "step": 32030
    },
    {
      "entropy": 5.857249975204468,
      "epoch": 3.2022292197730793,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000399485749140132,
      "loss": 5.317,
      "mean_token_accuracy": 0.19959746897220612,
      "num_tokens": 62668055.0,
      "step": 32035
    },
    {
      "entropy": 5.911676454544067,
      "epoch": 3.202729044834308,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039945601848164153,
      "loss": 5.4269,
      "mean_token_accuracy": 0.19884512424468995,
      "num_tokens": 62678279.0,
      "step": 32040
    },
    {
      "entropy": 5.859133768081665,
      "epoch": 3.2032288698955367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003994262846918697,
      "loss": 5.2765,
      "mean_token_accuracy": 0.2078260749578476,
      "num_tokens": 62687875.0,
      "step": 32045
    },
    {
      "entropy": 5.930257272720337,
      "epoch": 3.203728694956765,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039939654777156454,
      "loss": 5.4107,
      "mean_token_accuracy": 0.19824043959379195,
      "num_tokens": 62698345.0,
      "step": 32050
    },
    {
      "entropy": 5.941431522369385,
      "epoch": 3.2042285200179936,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003993668077214742,
      "loss": 5.4062,
      "mean_token_accuracy": 0.19415728896856307,
      "num_tokens": 62708176.0,
      "step": 32055
    },
    {
      "entropy": 5.864764595031739,
      "epoch": 3.2047283450792223,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003993370645423471,
      "loss": 5.4108,
      "mean_token_accuracy": 0.1992843210697174,
      "num_tokens": 62717194.0,
      "step": 32060
    },
    {
      "entropy": 5.843459510803223,
      "epoch": 3.205228170140451,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039930731823493136,
      "loss": 5.2862,
      "mean_token_accuracy": 0.20871593654155732,
      "num_tokens": 62727095.0,
      "step": 32065
    },
    {
      "entropy": 5.778787231445312,
      "epoch": 3.2057279952016793,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039927756879997556,
      "loss": 5.2863,
      "mean_token_accuracy": 0.20960606783628463,
      "num_tokens": 62737452.0,
      "step": 32070
    },
    {
      "entropy": 5.891500616073609,
      "epoch": 3.206227820262908,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039924781623822803,
      "loss": 5.3581,
      "mean_token_accuracy": 0.2061750277876854,
      "num_tokens": 62747345.0,
      "step": 32075
    },
    {
      "entropy": 5.896512365341186,
      "epoch": 3.2067276453241367,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003992180605504374,
      "loss": 5.48,
      "mean_token_accuracy": 0.1924220710992813,
      "num_tokens": 62757222.0,
      "step": 32080
    },
    {
      "entropy": 5.918145990371704,
      "epoch": 3.207227470385365,
      "grad_norm": 1.125,
      "learning_rate": 0.0003991883017373523,
      "loss": 5.4291,
      "mean_token_accuracy": 0.1963381752371788,
      "num_tokens": 62767279.0,
      "step": 32085
    },
    {
      "entropy": 5.94239239692688,
      "epoch": 3.2077272954465936,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003991585397997216,
      "loss": 5.3741,
      "mean_token_accuracy": 0.2067513570189476,
      "num_tokens": 62776547.0,
      "step": 32090
    },
    {
      "entropy": 5.9238659858703615,
      "epoch": 3.2082271205078223,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000399128774738294,
      "loss": 5.3831,
      "mean_token_accuracy": 0.19967788010835646,
      "num_tokens": 62786689.0,
      "step": 32095
    },
    {
      "entropy": 5.917508459091186,
      "epoch": 3.208726945569051,
      "grad_norm": 1.125,
      "learning_rate": 0.0003990990065538183,
      "loss": 5.5129,
      "mean_token_accuracy": 0.189319908618927,
      "num_tokens": 62795862.0,
      "step": 32100
    },
    {
      "entropy": 5.903531932830811,
      "epoch": 3.2092267706302793,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003990692352470435,
      "loss": 5.346,
      "mean_token_accuracy": 0.19709502458572387,
      "num_tokens": 62806065.0,
      "step": 32105
    },
    {
      "entropy": 5.921771764755249,
      "epoch": 3.209726595691508,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039903946081871865,
      "loss": 5.4154,
      "mean_token_accuracy": 0.1899395167827606,
      "num_tokens": 62815107.0,
      "step": 32110
    },
    {
      "entropy": 5.830198621749878,
      "epoch": 3.2102264207527367,
      "grad_norm": 1.421875,
      "learning_rate": 0.00039900968326959295,
      "loss": 5.2628,
      "mean_token_accuracy": 0.21574965566396714,
      "num_tokens": 62823734.0,
      "step": 32115
    },
    {
      "entropy": 5.8769432544708256,
      "epoch": 3.210726245813965,
      "grad_norm": 1.125,
      "learning_rate": 0.00039897990260041557,
      "loss": 5.4223,
      "mean_token_accuracy": 0.20269154608249665,
      "num_tokens": 62833804.0,
      "step": 32120
    },
    {
      "entropy": 5.863140344619751,
      "epoch": 3.2112260708751936,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003989501188119357,
      "loss": 5.3687,
      "mean_token_accuracy": 0.20164745450019836,
      "num_tokens": 62844414.0,
      "step": 32125
    },
    {
      "entropy": 5.965583944320679,
      "epoch": 3.2117258959364223,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039892033190490274,
      "loss": 5.3706,
      "mean_token_accuracy": 0.2062399670481682,
      "num_tokens": 62855315.0,
      "step": 32130
    },
    {
      "entropy": 5.8850969791412355,
      "epoch": 3.212225720997651,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039889054188006615,
      "loss": 5.3275,
      "mean_token_accuracy": 0.20361828655004502,
      "num_tokens": 62864526.0,
      "step": 32135
    },
    {
      "entropy": 5.815344524383545,
      "epoch": 3.2127255460588793,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003988607487381753,
      "loss": 5.4245,
      "mean_token_accuracy": 0.1928684413433075,
      "num_tokens": 62874599.0,
      "step": 32140
    },
    {
      "entropy": 5.90827431678772,
      "epoch": 3.213225371120108,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003988309524799799,
      "loss": 5.4156,
      "mean_token_accuracy": 0.19443698227405548,
      "num_tokens": 62885956.0,
      "step": 32145
    },
    {
      "entropy": 5.713649940490723,
      "epoch": 3.2137251961813367,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039880115310622974,
      "loss": 5.2077,
      "mean_token_accuracy": 0.2211028516292572,
      "num_tokens": 62896202.0,
      "step": 32150
    },
    {
      "entropy": 5.828578186035156,
      "epoch": 3.214225021242565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039877135061767424,
      "loss": 5.3054,
      "mean_token_accuracy": 0.2014404594898224,
      "num_tokens": 62905650.0,
      "step": 32155
    },
    {
      "entropy": 5.909622001647949,
      "epoch": 3.2147248463037936,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003987415450150635,
      "loss": 5.3961,
      "mean_token_accuracy": 0.2000793844461441,
      "num_tokens": 62916036.0,
      "step": 32160
    },
    {
      "entropy": 5.978132057189941,
      "epoch": 3.2152246713650223,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039871173629914736,
      "loss": 5.4656,
      "mean_token_accuracy": 0.1929624229669571,
      "num_tokens": 62925221.0,
      "step": 32165
    },
    {
      "entropy": 5.875661849975586,
      "epoch": 3.215724496426251,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039868192447067575,
      "loss": 5.3875,
      "mean_token_accuracy": 0.19375354051589966,
      "num_tokens": 62934917.0,
      "step": 32170
    },
    {
      "entropy": 5.7765686988830565,
      "epoch": 3.2162243214874793,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039865210953039863,
      "loss": 5.3417,
      "mean_token_accuracy": 0.20304033011198044,
      "num_tokens": 62944357.0,
      "step": 32175
    },
    {
      "entropy": 5.84767165184021,
      "epoch": 3.216724146548708,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003986222914790664,
      "loss": 5.4098,
      "mean_token_accuracy": 0.20423791706562042,
      "num_tokens": 62954149.0,
      "step": 32180
    },
    {
      "entropy": 5.924725961685181,
      "epoch": 3.2172239716099367,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039859247031742907,
      "loss": 5.4754,
      "mean_token_accuracy": 0.18813867419958114,
      "num_tokens": 62964721.0,
      "step": 32185
    },
    {
      "entropy": 5.891869688034058,
      "epoch": 3.217723796671165,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000398562646046237,
      "loss": 5.3639,
      "mean_token_accuracy": 0.20141468942165375,
      "num_tokens": 62974738.0,
      "step": 32190
    },
    {
      "entropy": 5.865064191818237,
      "epoch": 3.2182236217323936,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039853281866624054,
      "loss": 5.3895,
      "mean_token_accuracy": 0.20711638033390045,
      "num_tokens": 62984109.0,
      "step": 32195
    },
    {
      "entropy": 5.9468645572662355,
      "epoch": 3.2187234467936223,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039850298817819017,
      "loss": 5.4794,
      "mean_token_accuracy": 0.1881440684199333,
      "num_tokens": 62994443.0,
      "step": 32200
    },
    {
      "entropy": 5.984280014038086,
      "epoch": 3.219223271854851,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003984731545828364,
      "loss": 5.5437,
      "mean_token_accuracy": 0.17908304929733276,
      "num_tokens": 63003897.0,
      "step": 32205
    },
    {
      "entropy": 5.893624210357666,
      "epoch": 3.2197230969160793,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039844331788092986,
      "loss": 5.4413,
      "mean_token_accuracy": 0.1976270332932472,
      "num_tokens": 63013012.0,
      "step": 32210
    },
    {
      "entropy": 5.864507436752319,
      "epoch": 3.220222921977308,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003984134780732211,
      "loss": 5.4427,
      "mean_token_accuracy": 0.19510924071073532,
      "num_tokens": 63022756.0,
      "step": 32215
    },
    {
      "entropy": 5.85331449508667,
      "epoch": 3.2207227470385367,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003983836351604611,
      "loss": 5.3553,
      "mean_token_accuracy": 0.19888201355934143,
      "num_tokens": 63033109.0,
      "step": 32220
    },
    {
      "entropy": 5.916881084442139,
      "epoch": 3.221222572099765,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003983537891434005,
      "loss": 5.3947,
      "mean_token_accuracy": 0.19324279576539993,
      "num_tokens": 63042733.0,
      "step": 32225
    },
    {
      "entropy": 5.790520429611206,
      "epoch": 3.2217223971609936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003983239400227903,
      "loss": 5.2547,
      "mean_token_accuracy": 0.21936584413051605,
      "num_tokens": 63053654.0,
      "step": 32230
    },
    {
      "entropy": 5.85957522392273,
      "epoch": 3.2222222222222223,
      "grad_norm": 1.125,
      "learning_rate": 0.0003982940877993816,
      "loss": 5.3699,
      "mean_token_accuracy": 0.19776227474212646,
      "num_tokens": 63063285.0,
      "step": 32235
    },
    {
      "entropy": 6.026463174819947,
      "epoch": 3.222722047283451,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039826423247392537,
      "loss": 5.5564,
      "mean_token_accuracy": 0.18192590475082399,
      "num_tokens": 63072730.0,
      "step": 32240
    },
    {
      "entropy": 5.849705791473388,
      "epoch": 3.2232218723446793,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003982343740471727,
      "loss": 5.349,
      "mean_token_accuracy": 0.1974280908703804,
      "num_tokens": 63081703.0,
      "step": 32245
    },
    {
      "entropy": 5.862678480148316,
      "epoch": 3.223721697405908,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039820451251987494,
      "loss": 5.3598,
      "mean_token_accuracy": 0.2012451782822609,
      "num_tokens": 63091382.0,
      "step": 32250
    },
    {
      "entropy": 5.881317281723023,
      "epoch": 3.2242215224671367,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039817464789278326,
      "loss": 5.2999,
      "mean_token_accuracy": 0.21066614389419555,
      "num_tokens": 63101256.0,
      "step": 32255
    },
    {
      "entropy": 5.876102924346924,
      "epoch": 3.224721347528365,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039814478016664923,
      "loss": 5.3105,
      "mean_token_accuracy": 0.2098934605717659,
      "num_tokens": 63111001.0,
      "step": 32260
    },
    {
      "entropy": 5.828036451339722,
      "epoch": 3.2252211725895936,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039811490934222416,
      "loss": 5.4164,
      "mean_token_accuracy": 0.19627173542976378,
      "num_tokens": 63120524.0,
      "step": 32265
    },
    {
      "entropy": 5.906278467178344,
      "epoch": 3.2257209976508223,
      "grad_norm": 1.125,
      "learning_rate": 0.0003980850354202598,
      "loss": 5.4228,
      "mean_token_accuracy": 0.19798907339572908,
      "num_tokens": 63130673.0,
      "step": 32270
    },
    {
      "entropy": 5.946609592437744,
      "epoch": 3.226220822712051,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039805515840150743,
      "loss": 5.4297,
      "mean_token_accuracy": 0.19530714750289918,
      "num_tokens": 63140949.0,
      "step": 32275
    },
    {
      "entropy": 5.931145763397216,
      "epoch": 3.2267206477732793,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039802527828671896,
      "loss": 5.4289,
      "mean_token_accuracy": 0.1940406322479248,
      "num_tokens": 63151088.0,
      "step": 32280
    },
    {
      "entropy": 5.856358909606934,
      "epoch": 3.227220472834508,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003979953950766462,
      "loss": 5.2857,
      "mean_token_accuracy": 0.20541000366210938,
      "num_tokens": 63160739.0,
      "step": 32285
    },
    {
      "entropy": 5.816531944274902,
      "epoch": 3.2277202978957367,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039796550877204095,
      "loss": 5.3801,
      "mean_token_accuracy": 0.1967827409505844,
      "num_tokens": 63170657.0,
      "step": 32290
    },
    {
      "entropy": 5.883535289764405,
      "epoch": 3.228220122956965,
      "grad_norm": 1.125,
      "learning_rate": 0.00039793561937365515,
      "loss": 5.4011,
      "mean_token_accuracy": 0.20651754587888718,
      "num_tokens": 63180603.0,
      "step": 32295
    },
    {
      "entropy": 5.852366971969604,
      "epoch": 3.2287199480181936,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039790572688224075,
      "loss": 5.436,
      "mean_token_accuracy": 0.19746821671724318,
      "num_tokens": 63190589.0,
      "step": 32300
    },
    {
      "entropy": 5.939520835876465,
      "epoch": 3.2292197730794223,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00039787583129854997,
      "loss": 5.4279,
      "mean_token_accuracy": 0.19962105005979539,
      "num_tokens": 63200227.0,
      "step": 32305
    },
    {
      "entropy": 5.858824634552002,
      "epoch": 3.2297195981406506,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003978459326233349,
      "loss": 5.348,
      "mean_token_accuracy": 0.2020369827747345,
      "num_tokens": 63209273.0,
      "step": 32310
    },
    {
      "entropy": 5.781201934814453,
      "epoch": 3.2302194232018793,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0003978160308573477,
      "loss": 5.3005,
      "mean_token_accuracy": 0.2170563206076622,
      "num_tokens": 63218711.0,
      "step": 32315
    },
    {
      "entropy": 5.781380224227905,
      "epoch": 3.230719248263108,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003977861260013407,
      "loss": 5.2528,
      "mean_token_accuracy": 0.21938618570566176,
      "num_tokens": 63229982.0,
      "step": 32320
    },
    {
      "entropy": 6.053339624404908,
      "epoch": 3.2312190733243367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003977562180560665,
      "loss": 5.5613,
      "mean_token_accuracy": 0.18598955869674683,
      "num_tokens": 63239659.0,
      "step": 32325
    },
    {
      "entropy": 5.912880611419678,
      "epoch": 3.231718898385565,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003977263070222774,
      "loss": 5.3806,
      "mean_token_accuracy": 0.19768276065587997,
      "num_tokens": 63249379.0,
      "step": 32330
    },
    {
      "entropy": 5.858425664901733,
      "epoch": 3.2322187234467936,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000397696392900726,
      "loss": 5.2906,
      "mean_token_accuracy": 0.2053486093878746,
      "num_tokens": 63259857.0,
      "step": 32335
    },
    {
      "entropy": 5.921155595779419,
      "epoch": 3.2327185485080223,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003976664756921649,
      "loss": 5.375,
      "mean_token_accuracy": 0.19739738404750823,
      "num_tokens": 63268209.0,
      "step": 32340
    },
    {
      "entropy": 5.877609729766846,
      "epoch": 3.2332183735692506,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003976365553973469,
      "loss": 5.3775,
      "mean_token_accuracy": 0.19774318784475325,
      "num_tokens": 63278649.0,
      "step": 32345
    },
    {
      "entropy": 5.905547666549682,
      "epoch": 3.2337181986304793,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003976066320170247,
      "loss": 5.3965,
      "mean_token_accuracy": 0.19554246217012405,
      "num_tokens": 63288090.0,
      "step": 32350
    },
    {
      "entropy": 5.910655832290649,
      "epoch": 3.234218023691708,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039757670555195114,
      "loss": 5.4304,
      "mean_token_accuracy": 0.20099278539419174,
      "num_tokens": 63297443.0,
      "step": 32355
    },
    {
      "entropy": 5.8445648670196535,
      "epoch": 3.2347178487529367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039754677600287923,
      "loss": 5.4156,
      "mean_token_accuracy": 0.19332834780216218,
      "num_tokens": 63306485.0,
      "step": 32360
    },
    {
      "entropy": 5.861625003814697,
      "epoch": 3.235217673814165,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039751684337056197,
      "loss": 5.3641,
      "mean_token_accuracy": 0.2028803661465645,
      "num_tokens": 63315872.0,
      "step": 32365
    },
    {
      "entropy": 5.852970981597901,
      "epoch": 3.2357174988753936,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003974869076557525,
      "loss": 5.3207,
      "mean_token_accuracy": 0.19935210943222045,
      "num_tokens": 63325778.0,
      "step": 32370
    },
    {
      "entropy": 5.831177806854248,
      "epoch": 3.2362173239366223,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003974569688592038,
      "loss": 5.3192,
      "mean_token_accuracy": 0.20759278237819673,
      "num_tokens": 63335312.0,
      "step": 32375
    },
    {
      "entropy": 5.840210771560669,
      "epoch": 3.2367171489978506,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003974270269816695,
      "loss": 5.3998,
      "mean_token_accuracy": 0.19994597584009172,
      "num_tokens": 63344338.0,
      "step": 32380
    },
    {
      "entropy": 5.908917236328125,
      "epoch": 3.2372169740590793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039739708202390255,
      "loss": 5.4305,
      "mean_token_accuracy": 0.19301551431417466,
      "num_tokens": 63353775.0,
      "step": 32385
    },
    {
      "entropy": 5.926157140731812,
      "epoch": 3.237716799120308,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003973671339866566,
      "loss": 5.456,
      "mean_token_accuracy": 0.18599175959825515,
      "num_tokens": 63363474.0,
      "step": 32390
    },
    {
      "entropy": 5.89496488571167,
      "epoch": 3.2382166241815367,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039733718287068495,
      "loss": 5.4342,
      "mean_token_accuracy": 0.20021835267543792,
      "num_tokens": 63374120.0,
      "step": 32395
    },
    {
      "entropy": 5.887057256698609,
      "epoch": 3.238716449242765,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003973072286767413,
      "loss": 5.4053,
      "mean_token_accuracy": 0.19884207844734192,
      "num_tokens": 63384206.0,
      "step": 32400
    },
    {
      "entropy": 5.713797616958618,
      "epoch": 3.2392162743039936,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003972772714055793,
      "loss": 5.2102,
      "mean_token_accuracy": 0.21235571503639222,
      "num_tokens": 63394354.0,
      "step": 32405
    },
    {
      "entropy": 5.921784782409668,
      "epoch": 3.2397160993652223,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039724731105795255,
      "loss": 5.4163,
      "mean_token_accuracy": 0.2001093626022339,
      "num_tokens": 63403758.0,
      "step": 32410
    },
    {
      "entropy": 5.87131700515747,
      "epoch": 3.2402159244264506,
      "grad_norm": 1.15625,
      "learning_rate": 0.000397217347634615,
      "loss": 5.3874,
      "mean_token_accuracy": 0.19668328016996384,
      "num_tokens": 63413539.0,
      "step": 32415
    },
    {
      "entropy": 5.890050029754638,
      "epoch": 3.2407157494876793,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003971873811363203,
      "loss": 5.4575,
      "mean_token_accuracy": 0.1906759425997734,
      "num_tokens": 63423577.0,
      "step": 32420
    },
    {
      "entropy": 5.932536315917969,
      "epoch": 3.241215574548908,
      "grad_norm": 1.125,
      "learning_rate": 0.00039715741156382255,
      "loss": 5.428,
      "mean_token_accuracy": 0.19662974029779434,
      "num_tokens": 63433483.0,
      "step": 32425
    },
    {
      "entropy": 5.910219621658325,
      "epoch": 3.241715399610136,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003971274389178759,
      "loss": 5.4262,
      "mean_token_accuracy": 0.20129602104425431,
      "num_tokens": 63443341.0,
      "step": 32430
    },
    {
      "entropy": 5.814891338348389,
      "epoch": 3.242215224671365,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003970974631992342,
      "loss": 5.3112,
      "mean_token_accuracy": 0.20643816888332367,
      "num_tokens": 63452783.0,
      "step": 32435
    },
    {
      "entropy": 5.959586668014526,
      "epoch": 3.2427150497325936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003970674844086517,
      "loss": 5.4559,
      "mean_token_accuracy": 0.19249542951583862,
      "num_tokens": 63463420.0,
      "step": 32440
    },
    {
      "entropy": 5.908069753646851,
      "epoch": 3.2432148747938223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003970375025468829,
      "loss": 5.3377,
      "mean_token_accuracy": 0.2035536527633667,
      "num_tokens": 63472405.0,
      "step": 32445
    },
    {
      "entropy": 5.890574598312378,
      "epoch": 3.2437146998550506,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039700751761468175,
      "loss": 5.4484,
      "mean_token_accuracy": 0.1990016371011734,
      "num_tokens": 63481242.0,
      "step": 32450
    },
    {
      "entropy": 5.860441780090332,
      "epoch": 3.2442145249162793,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003969775296128029,
      "loss": 5.3797,
      "mean_token_accuracy": 0.20714791417121886,
      "num_tokens": 63492244.0,
      "step": 32455
    },
    {
      "entropy": 5.898698091506958,
      "epoch": 3.244714349977508,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003969475385420008,
      "loss": 5.3871,
      "mean_token_accuracy": 0.1972530961036682,
      "num_tokens": 63501117.0,
      "step": 32460
    },
    {
      "entropy": 5.990155982971191,
      "epoch": 3.245214175038736,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039691754440303003,
      "loss": 5.4937,
      "mean_token_accuracy": 0.18781276494264604,
      "num_tokens": 63509603.0,
      "step": 32465
    },
    {
      "entropy": 5.923660564422607,
      "epoch": 3.245714000099965,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039688754719664516,
      "loss": 5.4787,
      "mean_token_accuracy": 0.19542424529790878,
      "num_tokens": 63519600.0,
      "step": 32470
    },
    {
      "entropy": 5.8949672222137455,
      "epoch": 3.2462138251611936,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000396857546923601,
      "loss": 5.4133,
      "mean_token_accuracy": 0.19656375795602798,
      "num_tokens": 63529712.0,
      "step": 32475
    },
    {
      "entropy": 5.832987976074219,
      "epoch": 3.2467136502224223,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003968275435846523,
      "loss": 5.3716,
      "mean_token_accuracy": 0.1989708811044693,
      "num_tokens": 63540327.0,
      "step": 32480
    },
    {
      "entropy": 5.838025045394898,
      "epoch": 3.2472134752836506,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000396797537180554,
      "loss": 5.3429,
      "mean_token_accuracy": 0.2061411812901497,
      "num_tokens": 63549898.0,
      "step": 32485
    },
    {
      "entropy": 5.87333197593689,
      "epoch": 3.2477133003448793,
      "grad_norm": 1.09375,
      "learning_rate": 0.000396767527712061,
      "loss": 5.3323,
      "mean_token_accuracy": 0.20519259870052337,
      "num_tokens": 63559296.0,
      "step": 32490
    },
    {
      "entropy": 5.917629718780518,
      "epoch": 3.248213125406108,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039673751517992827,
      "loss": 5.4413,
      "mean_token_accuracy": 0.1950993597507477,
      "num_tokens": 63568464.0,
      "step": 32495
    },
    {
      "entropy": 5.868742847442627,
      "epoch": 3.248712950467336,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039670749958491103,
      "loss": 5.4436,
      "mean_token_accuracy": 0.19816715717315675,
      "num_tokens": 63577623.0,
      "step": 32500
    },
    {
      "entropy": 5.932502412796021,
      "epoch": 3.249212775528565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039667748092776453,
      "loss": 5.2966,
      "mean_token_accuracy": 0.19917395561933518,
      "num_tokens": 63586809.0,
      "step": 32505
    },
    {
      "entropy": 5.883761978149414,
      "epoch": 3.2497126005897936,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039664745920924374,
      "loss": 5.4168,
      "mean_token_accuracy": 0.20072222799062728,
      "num_tokens": 63597262.0,
      "step": 32510
    },
    {
      "entropy": 5.834929752349853,
      "epoch": 3.2502124256510223,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003966174344301043,
      "loss": 5.3861,
      "mean_token_accuracy": 0.2049715429544449,
      "num_tokens": 63608688.0,
      "step": 32515
    },
    {
      "entropy": 5.806754207611084,
      "epoch": 3.2507122507122506,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003965874065911014,
      "loss": 5.2548,
      "mean_token_accuracy": 0.21174240559339524,
      "num_tokens": 63619894.0,
      "step": 32520
    },
    {
      "entropy": 5.872159624099732,
      "epoch": 3.2512120757734793,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039655737569299073,
      "loss": 5.296,
      "mean_token_accuracy": 0.2015799343585968,
      "num_tokens": 63628607.0,
      "step": 32525
    },
    {
      "entropy": 5.905429124832153,
      "epoch": 3.251711900834708,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003965273417365277,
      "loss": 5.3814,
      "mean_token_accuracy": 0.19941970556974412,
      "num_tokens": 63639058.0,
      "step": 32530
    },
    {
      "entropy": 5.889187002182007,
      "epoch": 3.252211725895936,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039649730472246796,
      "loss": 5.3459,
      "mean_token_accuracy": 0.20453464537858962,
      "num_tokens": 63647338.0,
      "step": 32535
    },
    {
      "entropy": 5.805127573013306,
      "epoch": 3.252711550957165,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003964672646515674,
      "loss": 5.3086,
      "mean_token_accuracy": 0.20586963891983032,
      "num_tokens": 63657990.0,
      "step": 32540
    },
    {
      "entropy": 5.908466005325318,
      "epoch": 3.2532113760183936,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003964372215245817,
      "loss": 5.4905,
      "mean_token_accuracy": 0.19084559082984925,
      "num_tokens": 63667310.0,
      "step": 32545
    },
    {
      "entropy": 5.875511693954468,
      "epoch": 3.2537112010796223,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039640717534226684,
      "loss": 5.3502,
      "mean_token_accuracy": 0.20577007234096528,
      "num_tokens": 63677254.0,
      "step": 32550
    },
    {
      "entropy": 5.966557168960572,
      "epoch": 3.2542110261408506,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003963771261053786,
      "loss": 5.4325,
      "mean_token_accuracy": 0.19706568717956544,
      "num_tokens": 63686441.0,
      "step": 32555
    },
    {
      "entropy": 5.796702194213867,
      "epoch": 3.2547108512020793,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039634707381467303,
      "loss": 5.3173,
      "mean_token_accuracy": 0.21703273952007293,
      "num_tokens": 63696445.0,
      "step": 32560
    },
    {
      "entropy": 5.936091947555542,
      "epoch": 3.255210676263308,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003963170184709064,
      "loss": 5.4467,
      "mean_token_accuracy": 0.19821924567222596,
      "num_tokens": 63706536.0,
      "step": 32565
    },
    {
      "entropy": 5.918974685668945,
      "epoch": 3.255710501324536,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039628696007483477,
      "loss": 5.4263,
      "mean_token_accuracy": 0.1950952723622322,
      "num_tokens": 63716752.0,
      "step": 32570
    },
    {
      "entropy": 5.905401659011841,
      "epoch": 3.256210326385765,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039625689862721447,
      "loss": 5.4045,
      "mean_token_accuracy": 0.20037564933300017,
      "num_tokens": 63726995.0,
      "step": 32575
    },
    {
      "entropy": 5.879486513137818,
      "epoch": 3.2567101514469936,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039622683412880174,
      "loss": 5.3341,
      "mean_token_accuracy": 0.20883368700742722,
      "num_tokens": 63735261.0,
      "step": 32580
    },
    {
      "entropy": 5.898955202102661,
      "epoch": 3.2572099765082223,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003961967665803532,
      "loss": 5.4346,
      "mean_token_accuracy": 0.20397012531757355,
      "num_tokens": 63746307.0,
      "step": 32585
    },
    {
      "entropy": 5.9221758365631105,
      "epoch": 3.2577098015694506,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039616669598262504,
      "loss": 5.4547,
      "mean_token_accuracy": 0.18831576704978942,
      "num_tokens": 63755802.0,
      "step": 32590
    },
    {
      "entropy": 5.822965002059936,
      "epoch": 3.2582096266306793,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039613662233637406,
      "loss": 5.343,
      "mean_token_accuracy": 0.20232110917568208,
      "num_tokens": 63765759.0,
      "step": 32595
    },
    {
      "entropy": 5.970396995544434,
      "epoch": 3.258709451691908,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039610654564235684,
      "loss": 5.4588,
      "mean_token_accuracy": 0.19943392127752305,
      "num_tokens": 63775528.0,
      "step": 32600
    },
    {
      "entropy": 5.854634189605713,
      "epoch": 3.259209276753136,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003960764659013301,
      "loss": 5.3173,
      "mean_token_accuracy": 0.20206755101680757,
      "num_tokens": 63785480.0,
      "step": 32605
    },
    {
      "entropy": 5.825887060165405,
      "epoch": 3.259709101814365,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003960463831140505,
      "loss": 5.2494,
      "mean_token_accuracy": 0.21593891829252243,
      "num_tokens": 63793710.0,
      "step": 32610
    },
    {
      "entropy": 5.899285364151001,
      "epoch": 3.2602089268755936,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039601629728127524,
      "loss": 5.4896,
      "mean_token_accuracy": 0.19616891145706178,
      "num_tokens": 63802664.0,
      "step": 32615
    },
    {
      "entropy": 5.711821794509888,
      "epoch": 3.2607087519368223,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039598620840376096,
      "loss": 5.1227,
      "mean_token_accuracy": 0.22372929453849794,
      "num_tokens": 63813146.0,
      "step": 32620
    },
    {
      "entropy": 5.901413011550903,
      "epoch": 3.2612085769980506,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039595611648226485,
      "loss": 5.4509,
      "mean_token_accuracy": 0.19603076577186584,
      "num_tokens": 63823182.0,
      "step": 32625
    },
    {
      "entropy": 5.845896768569946,
      "epoch": 3.2617084020592793,
      "grad_norm": 1.171875,
      "learning_rate": 0.000395926021517544,
      "loss": 5.3793,
      "mean_token_accuracy": 0.20049204230308532,
      "num_tokens": 63832370.0,
      "step": 32630
    },
    {
      "entropy": 5.831495046615601,
      "epoch": 3.262208227120508,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039589592351035554,
      "loss": 5.3774,
      "mean_token_accuracy": 0.2051725834608078,
      "num_tokens": 63843156.0,
      "step": 32635
    },
    {
      "entropy": 5.981682300567627,
      "epoch": 3.262708052181736,
      "grad_norm": 1.125,
      "learning_rate": 0.0003958658224614567,
      "loss": 5.486,
      "mean_token_accuracy": 0.19754331707954406,
      "num_tokens": 63852744.0,
      "step": 32640
    },
    {
      "entropy": 5.8587864398956295,
      "epoch": 3.263207877242965,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039583571837160497,
      "loss": 5.3205,
      "mean_token_accuracy": 0.20866528898477554,
      "num_tokens": 63862694.0,
      "step": 32645
    },
    {
      "entropy": 5.82322211265564,
      "epoch": 3.2637077023041936,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003958056112415576,
      "loss": 5.3756,
      "mean_token_accuracy": 0.19828789234161376,
      "num_tokens": 63871943.0,
      "step": 32650
    },
    {
      "entropy": 5.897679615020752,
      "epoch": 3.2642075273654223,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003957755010720721,
      "loss": 5.4044,
      "mean_token_accuracy": 0.20406021922826767,
      "num_tokens": 63881354.0,
      "step": 32655
    },
    {
      "entropy": 5.887338304519654,
      "epoch": 3.2647073524266506,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039574538786390617,
      "loss": 5.411,
      "mean_token_accuracy": 0.19474628418684006,
      "num_tokens": 63890750.0,
      "step": 32660
    },
    {
      "entropy": 5.905306196212768,
      "epoch": 3.2652071774878793,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003957152716178173,
      "loss": 5.4198,
      "mean_token_accuracy": 0.19409598559141159,
      "num_tokens": 63900730.0,
      "step": 32665
    },
    {
      "entropy": 5.821209144592285,
      "epoch": 3.265707002549108,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039568515233456326,
      "loss": 5.4568,
      "mean_token_accuracy": 0.1932338148355484,
      "num_tokens": 63911812.0,
      "step": 32670
    },
    {
      "entropy": 5.900972032546997,
      "epoch": 3.266206827610336,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003956550300149018,
      "loss": 5.3705,
      "mean_token_accuracy": 0.19688964635133743,
      "num_tokens": 63920785.0,
      "step": 32675
    },
    {
      "entropy": 5.902582502365112,
      "epoch": 3.266706652671565,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003956249046595909,
      "loss": 5.3995,
      "mean_token_accuracy": 0.19979496747255326,
      "num_tokens": 63931167.0,
      "step": 32680
    },
    {
      "entropy": 5.928242778778076,
      "epoch": 3.2672064777327936,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003955947762693884,
      "loss": 5.4666,
      "mean_token_accuracy": 0.18765970170497895,
      "num_tokens": 63941448.0,
      "step": 32685
    },
    {
      "entropy": 5.968032932281494,
      "epoch": 3.2677063027940223,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039556464484505235,
      "loss": 5.4336,
      "mean_token_accuracy": 0.19465062767267227,
      "num_tokens": 63951412.0,
      "step": 32690
    },
    {
      "entropy": 5.937184476852417,
      "epoch": 3.2682061278552506,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003955345103873409,
      "loss": 5.4998,
      "mean_token_accuracy": 0.19013982266187668,
      "num_tokens": 63960957.0,
      "step": 32695
    },
    {
      "entropy": 5.878360271453857,
      "epoch": 3.2687059529164793,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003955043728970121,
      "loss": 5.4375,
      "mean_token_accuracy": 0.1970347911119461,
      "num_tokens": 63970229.0,
      "step": 32700
    },
    {
      "entropy": 5.856795883178711,
      "epoch": 3.269205777977708,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003954742323748244,
      "loss": 5.3781,
      "mean_token_accuracy": 0.20491316616535188,
      "num_tokens": 63979816.0,
      "step": 32705
    },
    {
      "entropy": 5.839673376083374,
      "epoch": 3.269705603038936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039544408882153596,
      "loss": 5.2495,
      "mean_token_accuracy": 0.21270727962255478,
      "num_tokens": 63989612.0,
      "step": 32710
    },
    {
      "entropy": 5.9214259624481205,
      "epoch": 3.270205428100165,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003954139422379052,
      "loss": 5.4246,
      "mean_token_accuracy": 0.19252515435218812,
      "num_tokens": 64001254.0,
      "step": 32715
    },
    {
      "entropy": 5.850560712814331,
      "epoch": 3.2707052531613936,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003953837926246907,
      "loss": 5.4114,
      "mean_token_accuracy": 0.20128345042467116,
      "num_tokens": 64011235.0,
      "step": 32720
    },
    {
      "entropy": 5.8185474395751955,
      "epoch": 3.2712050782226223,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039535363998265095,
      "loss": 5.3712,
      "mean_token_accuracy": 0.19794362485408784,
      "num_tokens": 64020437.0,
      "step": 32725
    },
    {
      "entropy": 5.978401708602905,
      "epoch": 3.2717049032838506,
      "grad_norm": 1.125,
      "learning_rate": 0.00039532348431254455,
      "loss": 5.5538,
      "mean_token_accuracy": 0.18772882223129272,
      "num_tokens": 64030930.0,
      "step": 32730
    },
    {
      "entropy": 5.901514720916748,
      "epoch": 3.2722047283450792,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003952933256151303,
      "loss": 5.3719,
      "mean_token_accuracy": 0.20151576548814773,
      "num_tokens": 64041175.0,
      "step": 32735
    },
    {
      "entropy": 5.9009984016418455,
      "epoch": 3.272704553406308,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003952631638911669,
      "loss": 5.3456,
      "mean_token_accuracy": 0.20310965180397034,
      "num_tokens": 64050208.0,
      "step": 32740
    },
    {
      "entropy": 5.880223035812378,
      "epoch": 3.273204378467536,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003952329991414132,
      "loss": 5.4221,
      "mean_token_accuracy": 0.19377750158309937,
      "num_tokens": 64060303.0,
      "step": 32745
    },
    {
      "entropy": 5.896939420700074,
      "epoch": 3.273704203528765,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039520283136662825,
      "loss": 5.3693,
      "mean_token_accuracy": 0.199366495013237,
      "num_tokens": 64070240.0,
      "step": 32750
    },
    {
      "entropy": 6.0026308536529545,
      "epoch": 3.2742040285899936,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039517266056757097,
      "loss": 5.63,
      "mean_token_accuracy": 0.17797797173261642,
      "num_tokens": 64080836.0,
      "step": 32755
    },
    {
      "entropy": 5.92116346359253,
      "epoch": 3.2747038536512223,
      "grad_norm": 1.125,
      "learning_rate": 0.00039514248674500055,
      "loss": 5.4211,
      "mean_token_accuracy": 0.1981012061238289,
      "num_tokens": 64090533.0,
      "step": 32760
    },
    {
      "entropy": 5.879924726486206,
      "epoch": 3.2752036787124506,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039511230989967594,
      "loss": 5.3398,
      "mean_token_accuracy": 0.20135882049798964,
      "num_tokens": 64100156.0,
      "step": 32765
    },
    {
      "entropy": 5.813510704040527,
      "epoch": 3.2757035037736792,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003950821300323566,
      "loss": 5.331,
      "mean_token_accuracy": 0.2045191302895546,
      "num_tokens": 64108906.0,
      "step": 32770
    },
    {
      "entropy": 5.858032178878784,
      "epoch": 3.276203328834908,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003950519471438017,
      "loss": 5.3872,
      "mean_token_accuracy": 0.20201565772295,
      "num_tokens": 64118361.0,
      "step": 32775
    },
    {
      "entropy": 6.004518938064575,
      "epoch": 3.276703153896136,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039502176123477076,
      "loss": 5.5456,
      "mean_token_accuracy": 0.18442662209272384,
      "num_tokens": 64128978.0,
      "step": 32780
    },
    {
      "entropy": 5.908996677398681,
      "epoch": 3.277202978957365,
      "grad_norm": 0.953125,
      "learning_rate": 0.00039499157230602313,
      "loss": 5.3774,
      "mean_token_accuracy": 0.2033190682530403,
      "num_tokens": 64139424.0,
      "step": 32785
    },
    {
      "entropy": 5.871485424041748,
      "epoch": 3.2777028040185936,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039496138035831845,
      "loss": 5.4371,
      "mean_token_accuracy": 0.19213270545005798,
      "num_tokens": 64149814.0,
      "step": 32790
    },
    {
      "entropy": 5.8975835800170895,
      "epoch": 3.2782026290798223,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039493118539241643,
      "loss": 5.4456,
      "mean_token_accuracy": 0.1928626298904419,
      "num_tokens": 64159777.0,
      "step": 32795
    },
    {
      "entropy": 5.849367713928222,
      "epoch": 3.2787024541410505,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039490098740907654,
      "loss": 5.3146,
      "mean_token_accuracy": 0.2069511368870735,
      "num_tokens": 64170241.0,
      "step": 32800
    },
    {
      "entropy": 5.829895877838135,
      "epoch": 3.2792022792022792,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039487078640905866,
      "loss": 5.3343,
      "mean_token_accuracy": 0.2017352983355522,
      "num_tokens": 64180514.0,
      "step": 32805
    },
    {
      "entropy": 5.880047369003296,
      "epoch": 3.279702104263508,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039484058239312264,
      "loss": 5.4371,
      "mean_token_accuracy": 0.19748082458972932,
      "num_tokens": 64191184.0,
      "step": 32810
    },
    {
      "entropy": 5.9288112163543705,
      "epoch": 3.280201929324736,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039481037536202835,
      "loss": 5.4072,
      "mean_token_accuracy": 0.20191001147031784,
      "num_tokens": 64201134.0,
      "step": 32815
    },
    {
      "entropy": 5.901302719116211,
      "epoch": 3.280701754385965,
      "grad_norm": 1.25,
      "learning_rate": 0.00039478016531653586,
      "loss": 5.344,
      "mean_token_accuracy": 0.20191725194454194,
      "num_tokens": 64210578.0,
      "step": 32820
    },
    {
      "entropy": 5.9722571849823,
      "epoch": 3.2812015794471936,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039474995225740526,
      "loss": 5.5464,
      "mean_token_accuracy": 0.18960326611995698,
      "num_tokens": 64219762.0,
      "step": 32825
    },
    {
      "entropy": 5.870886611938476,
      "epoch": 3.2817014045084223,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003947197361853968,
      "loss": 5.3589,
      "mean_token_accuracy": 0.1998378664255142,
      "num_tokens": 64230371.0,
      "step": 32830
    },
    {
      "entropy": 5.845559740066529,
      "epoch": 3.2822012295696505,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039468951710127043,
      "loss": 5.4383,
      "mean_token_accuracy": 0.20443026274442672,
      "num_tokens": 64239154.0,
      "step": 32835
    },
    {
      "entropy": 5.818621253967285,
      "epoch": 3.2827010546308792,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003946592950057867,
      "loss": 5.3597,
      "mean_token_accuracy": 0.2062579095363617,
      "num_tokens": 64249257.0,
      "step": 32840
    },
    {
      "entropy": 5.956381845474243,
      "epoch": 3.283200879692108,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003946290698997058,
      "loss": 5.4469,
      "mean_token_accuracy": 0.19813692569732666,
      "num_tokens": 64259171.0,
      "step": 32845
    },
    {
      "entropy": 5.840672874450684,
      "epoch": 3.283700704753336,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003945988417837883,
      "loss": 5.2498,
      "mean_token_accuracy": 0.21168768554925918,
      "num_tokens": 64268178.0,
      "step": 32850
    },
    {
      "entropy": 5.888388347625733,
      "epoch": 3.284200529814565,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003945686106587948,
      "loss": 5.4492,
      "mean_token_accuracy": 0.19472338259220123,
      "num_tokens": 64278136.0,
      "step": 32855
    },
    {
      "entropy": 5.7849438190460205,
      "epoch": 3.2847003548757936,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039453837652548575,
      "loss": 5.3399,
      "mean_token_accuracy": 0.20240785479545592,
      "num_tokens": 64288305.0,
      "step": 32860
    },
    {
      "entropy": 5.843150520324707,
      "epoch": 3.2852001799370223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039450813938462197,
      "loss": 5.3562,
      "mean_token_accuracy": 0.20545895397663116,
      "num_tokens": 64297636.0,
      "step": 32865
    },
    {
      "entropy": 5.920308208465576,
      "epoch": 3.2857000049982505,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039447789923696415,
      "loss": 5.3784,
      "mean_token_accuracy": 0.20937199592590333,
      "num_tokens": 64306706.0,
      "step": 32870
    },
    {
      "entropy": 5.8369879722595215,
      "epoch": 3.2861998300594792,
      "grad_norm": 1.296875,
      "learning_rate": 0.000394447656083273,
      "loss": 5.3539,
      "mean_token_accuracy": 0.20218289494514466,
      "num_tokens": 64316909.0,
      "step": 32875
    },
    {
      "entropy": 5.7661473751068115,
      "epoch": 3.286699655120708,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039441740992430965,
      "loss": 5.3273,
      "mean_token_accuracy": 0.20253474414348602,
      "num_tokens": 64326103.0,
      "step": 32880
    },
    {
      "entropy": 5.866425609588623,
      "epoch": 3.287199480181936,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039438716076083504,
      "loss": 5.371,
      "mean_token_accuracy": 0.20653457194566727,
      "num_tokens": 64334915.0,
      "step": 32885
    },
    {
      "entropy": 5.700768184661865,
      "epoch": 3.287699305243165,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039435690859361013,
      "loss": 5.3444,
      "mean_token_accuracy": 0.2104848250746727,
      "num_tokens": 64344102.0,
      "step": 32890
    },
    {
      "entropy": 5.827236938476562,
      "epoch": 3.2881991303043936,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039432665342339614,
      "loss": 5.3793,
      "mean_token_accuracy": 0.2021569639444351,
      "num_tokens": 64353011.0,
      "step": 32895
    },
    {
      "entropy": 5.865096950531006,
      "epoch": 3.288698955365622,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003942963952509543,
      "loss": 5.3415,
      "mean_token_accuracy": 0.2013437807559967,
      "num_tokens": 64361658.0,
      "step": 32900
    },
    {
      "entropy": 5.951446294784546,
      "epoch": 3.2891987804268505,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039426613407704574,
      "loss": 5.5463,
      "mean_token_accuracy": 0.1886441305279732,
      "num_tokens": 64371878.0,
      "step": 32905
    },
    {
      "entropy": 5.877272891998291,
      "epoch": 3.2896986054880792,
      "grad_norm": 1.015625,
      "learning_rate": 0.000394235869902432,
      "loss": 5.3743,
      "mean_token_accuracy": 0.19960424154996873,
      "num_tokens": 64381896.0,
      "step": 32910
    },
    {
      "entropy": 5.91466760635376,
      "epoch": 3.2901984305493075,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039420560272787446,
      "loss": 5.4582,
      "mean_token_accuracy": 0.19041621387004853,
      "num_tokens": 64391618.0,
      "step": 32915
    },
    {
      "entropy": 5.897331285476684,
      "epoch": 3.290698255610536,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039417533255413465,
      "loss": 5.4046,
      "mean_token_accuracy": 0.19577753990888597,
      "num_tokens": 64402165.0,
      "step": 32920
    },
    {
      "entropy": 5.951656675338745,
      "epoch": 3.291198080671765,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039414505938197404,
      "loss": 5.5075,
      "mean_token_accuracy": 0.19012962132692338,
      "num_tokens": 64411334.0,
      "step": 32925
    },
    {
      "entropy": 5.956559944152832,
      "epoch": 3.2916979057329936,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003941147832121545,
      "loss": 5.417,
      "mean_token_accuracy": 0.19492007195949554,
      "num_tokens": 64420985.0,
      "step": 32930
    },
    {
      "entropy": 5.909087038040161,
      "epoch": 3.292197730794222,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003940845040454376,
      "loss": 5.462,
      "mean_token_accuracy": 0.19866268783807756,
      "num_tokens": 64430141.0,
      "step": 32935
    },
    {
      "entropy": 5.8415868282318115,
      "epoch": 3.2926975558554505,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003940542218825853,
      "loss": 5.387,
      "mean_token_accuracy": 0.19714101999998093,
      "num_tokens": 64440321.0,
      "step": 32940
    },
    {
      "entropy": 5.855879163742065,
      "epoch": 3.2931973809166792,
      "grad_norm": 1.125,
      "learning_rate": 0.00039402393672435934,
      "loss": 5.4111,
      "mean_token_accuracy": 0.20289693623781205,
      "num_tokens": 64449422.0,
      "step": 32945
    },
    {
      "entropy": 5.9547083377838135,
      "epoch": 3.2936972059779075,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039399364857152177,
      "loss": 5.5236,
      "mean_token_accuracy": 0.18625193685293198,
      "num_tokens": 64459179.0,
      "step": 32950
    },
    {
      "entropy": 5.9021519184112545,
      "epoch": 3.294197031039136,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003939633574248346,
      "loss": 5.3604,
      "mean_token_accuracy": 0.2030947118997574,
      "num_tokens": 64468454.0,
      "step": 32955
    },
    {
      "entropy": 5.803720617294312,
      "epoch": 3.294696856100365,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039393306328506004,
      "loss": 5.3601,
      "mean_token_accuracy": 0.21318666785955429,
      "num_tokens": 64478374.0,
      "step": 32960
    },
    {
      "entropy": 5.9079657077789305,
      "epoch": 3.2951966811615936,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039390276615296015,
      "loss": 5.4455,
      "mean_token_accuracy": 0.20014672577381135,
      "num_tokens": 64489692.0,
      "step": 32965
    },
    {
      "entropy": 5.925157117843628,
      "epoch": 3.295696506222822,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039387246602929726,
      "loss": 5.3682,
      "mean_token_accuracy": 0.20555726140737535,
      "num_tokens": 64500167.0,
      "step": 32970
    },
    {
      "entropy": 5.919354486465454,
      "epoch": 3.2961963312840505,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003938421629148338,
      "loss": 5.4205,
      "mean_token_accuracy": 0.19167758971452714,
      "num_tokens": 64509973.0,
      "step": 32975
    },
    {
      "entropy": 5.884012222290039,
      "epoch": 3.2966961563452792,
      "grad_norm": 1.140625,
      "learning_rate": 0.000393811856810332,
      "loss": 5.4325,
      "mean_token_accuracy": 0.19555241912603377,
      "num_tokens": 64519926.0,
      "step": 32980
    },
    {
      "entropy": 5.928401899337769,
      "epoch": 3.2971959814065075,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003937815477165545,
      "loss": 5.5,
      "mean_token_accuracy": 0.19720748513936998,
      "num_tokens": 64529770.0,
      "step": 32985
    },
    {
      "entropy": 5.871786737442017,
      "epoch": 3.297695806467736,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003937512356342638,
      "loss": 5.2592,
      "mean_token_accuracy": 0.2137589544057846,
      "num_tokens": 64539867.0,
      "step": 32990
    },
    {
      "entropy": 5.886129426956177,
      "epoch": 3.298195631528965,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003937209205642226,
      "loss": 5.364,
      "mean_token_accuracy": 0.21010173261165618,
      "num_tokens": 64549276.0,
      "step": 32995
    },
    {
      "entropy": 5.911748695373535,
      "epoch": 3.2986954565901936,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003936906025071936,
      "loss": 5.473,
      "mean_token_accuracy": 0.19653134942054748,
      "num_tokens": 64559294.0,
      "step": 33000
    },
    {
      "epoch": 3.2986954565901936,
      "eval_entropy": 5.6880131013551285,
      "eval_loss": 5.556079864501953,
      "eval_mean_token_accuracy": 0.19994534608634193,
      "eval_num_tokens": 64559294.0,
      "eval_runtime": 23.8428,
      "eval_samples_per_second": 1302.069,
      "eval_steps_per_second": 162.774,
      "step": 33000
    },
    {
      "entropy": 5.894896697998047,
      "epoch": 3.299195281651422,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039366028146393955,
      "loss": 5.364,
      "mean_token_accuracy": 0.2062472105026245,
      "num_tokens": 64569000.0,
      "step": 33005
    },
    {
      "entropy": 5.919348335266113,
      "epoch": 3.2996951067126505,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003936299574352233,
      "loss": 5.3649,
      "mean_token_accuracy": 0.20420179963111879,
      "num_tokens": 64578984.0,
      "step": 33010
    },
    {
      "entropy": 5.801018524169922,
      "epoch": 3.3001949317738792,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000393599630421808,
      "loss": 5.3206,
      "mean_token_accuracy": 0.20306559950113295,
      "num_tokens": 64588248.0,
      "step": 33015
    },
    {
      "entropy": 5.793853330612182,
      "epoch": 3.3006947568351075,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003935693004244564,
      "loss": 5.2977,
      "mean_token_accuracy": 0.2066078707575798,
      "num_tokens": 64597719.0,
      "step": 33020
    },
    {
      "entropy": 5.8809113025665285,
      "epoch": 3.301194581896336,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039353896744393163,
      "loss": 5.4403,
      "mean_token_accuracy": 0.20885936468839644,
      "num_tokens": 64608085.0,
      "step": 33025
    },
    {
      "entropy": 5.924952220916748,
      "epoch": 3.301694406957565,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003935086314809971,
      "loss": 5.3667,
      "mean_token_accuracy": 0.1979884222149849,
      "num_tokens": 64617473.0,
      "step": 33030
    },
    {
      "entropy": 5.7905418395996096,
      "epoch": 3.3021942320187936,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039347829253641583,
      "loss": 5.305,
      "mean_token_accuracy": 0.2040557846426964,
      "num_tokens": 64627083.0,
      "step": 33035
    },
    {
      "entropy": 5.914600563049317,
      "epoch": 3.302694057080022,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039344795061095123,
      "loss": 5.4364,
      "mean_token_accuracy": 0.19287603944540024,
      "num_tokens": 64636382.0,
      "step": 33040
    },
    {
      "entropy": 5.912721347808838,
      "epoch": 3.3031938821412505,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003934176057053667,
      "loss": 5.4256,
      "mean_token_accuracy": 0.19594080597162247,
      "num_tokens": 64646670.0,
      "step": 33045
    },
    {
      "entropy": 5.878040409088134,
      "epoch": 3.3036937072024792,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039338725782042563,
      "loss": 5.4829,
      "mean_token_accuracy": 0.19160767644643784,
      "num_tokens": 64656420.0,
      "step": 33050
    },
    {
      "entropy": 5.889844799041748,
      "epoch": 3.3041935322637075,
      "grad_norm": 1.125,
      "learning_rate": 0.00039335690695689157,
      "loss": 5.4745,
      "mean_token_accuracy": 0.19166013449430466,
      "num_tokens": 64667219.0,
      "step": 33055
    },
    {
      "entropy": 5.900331020355225,
      "epoch": 3.304693357324936,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003933265531155283,
      "loss": 5.3998,
      "mean_token_accuracy": 0.19832913130521773,
      "num_tokens": 64677502.0,
      "step": 33060
    },
    {
      "entropy": 5.875587606430054,
      "epoch": 3.305193182386165,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039329619629709935,
      "loss": 5.3507,
      "mean_token_accuracy": 0.20170503109693527,
      "num_tokens": 64686913.0,
      "step": 33065
    },
    {
      "entropy": 5.871400976181031,
      "epoch": 3.3056930074473936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039326583650236847,
      "loss": 5.4653,
      "mean_token_accuracy": 0.1888732373714447,
      "num_tokens": 64696611.0,
      "step": 33070
    },
    {
      "entropy": 5.900642776489258,
      "epoch": 3.306192832508622,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039323547373209974,
      "loss": 5.4011,
      "mean_token_accuracy": 0.1958785593509674,
      "num_tokens": 64706830.0,
      "step": 33075
    },
    {
      "entropy": 5.96183214187622,
      "epoch": 3.3066926575698505,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039320510798705684,
      "loss": 5.5048,
      "mean_token_accuracy": 0.18889326006174087,
      "num_tokens": 64717008.0,
      "step": 33080
    },
    {
      "entropy": 5.95891170501709,
      "epoch": 3.3071924826310792,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003931747392680038,
      "loss": 5.4257,
      "mean_token_accuracy": 0.19289007186889648,
      "num_tokens": 64726140.0,
      "step": 33085
    },
    {
      "entropy": 5.9014589309692385,
      "epoch": 3.3076923076923075,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003931443675757047,
      "loss": 5.427,
      "mean_token_accuracy": 0.1987533152103424,
      "num_tokens": 64735814.0,
      "step": 33090
    },
    {
      "entropy": 5.90257043838501,
      "epoch": 3.308192132753536,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039311399291092384,
      "loss": 5.4532,
      "mean_token_accuracy": 0.1929035171866417,
      "num_tokens": 64746437.0,
      "step": 33095
    },
    {
      "entropy": 6.017256879806519,
      "epoch": 3.308691957814765,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003930836152744251,
      "loss": 5.5387,
      "mean_token_accuracy": 0.19082180112600328,
      "num_tokens": 64756398.0,
      "step": 33100
    },
    {
      "entropy": 5.842862749099732,
      "epoch": 3.3091917828759936,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039305323466697317,
      "loss": 5.3146,
      "mean_token_accuracy": 0.20508363842964172,
      "num_tokens": 64766538.0,
      "step": 33105
    },
    {
      "entropy": 5.959404325485229,
      "epoch": 3.309691607937222,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039302285108933224,
      "loss": 5.5136,
      "mean_token_accuracy": 0.19487412869930268,
      "num_tokens": 64777889.0,
      "step": 33110
    },
    {
      "entropy": 5.892955493927002,
      "epoch": 3.3101914329984505,
      "grad_norm": 1.125,
      "learning_rate": 0.0003929924645422667,
      "loss": 5.4345,
      "mean_token_accuracy": 0.1956675097346306,
      "num_tokens": 64787410.0,
      "step": 33115
    },
    {
      "entropy": 5.978410100936889,
      "epoch": 3.3106912580596792,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039296207502654103,
      "loss": 5.5111,
      "mean_token_accuracy": 0.1870902106165886,
      "num_tokens": 64796507.0,
      "step": 33120
    },
    {
      "entropy": 5.926166582107544,
      "epoch": 3.3111910831209075,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039293168254292,
      "loss": 5.4122,
      "mean_token_accuracy": 0.20234548449516296,
      "num_tokens": 64805818.0,
      "step": 33125
    },
    {
      "entropy": 5.919566106796265,
      "epoch": 3.311690908182136,
      "grad_norm": 1.125,
      "learning_rate": 0.0003929012870921681,
      "loss": 5.4955,
      "mean_token_accuracy": 0.19403509497642518,
      "num_tokens": 64815579.0,
      "step": 33130
    },
    {
      "entropy": 5.877822828292847,
      "epoch": 3.312190733243365,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003928708886750503,
      "loss": 5.4217,
      "mean_token_accuracy": 0.1947478786110878,
      "num_tokens": 64824770.0,
      "step": 33135
    },
    {
      "entropy": 5.937855529785156,
      "epoch": 3.3126905583045936,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039284048729233115,
      "loss": 5.4164,
      "mean_token_accuracy": 0.1906494155526161,
      "num_tokens": 64833258.0,
      "step": 33140
    },
    {
      "entropy": 5.931019449234009,
      "epoch": 3.313190383365822,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003928100829447757,
      "loss": 5.4042,
      "mean_token_accuracy": 0.19858056604862212,
      "num_tokens": 64842660.0,
      "step": 33145
    },
    {
      "entropy": 5.795791721343994,
      "epoch": 3.3136902084270505,
      "grad_norm": 1.125,
      "learning_rate": 0.0003927796756331489,
      "loss": 5.3335,
      "mean_token_accuracy": 0.2077307879924774,
      "num_tokens": 64852235.0,
      "step": 33150
    },
    {
      "entropy": 5.820771217346191,
      "epoch": 3.3141900334882792,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003927492653582157,
      "loss": 5.3159,
      "mean_token_accuracy": 0.2047523722052574,
      "num_tokens": 64863092.0,
      "step": 33155
    },
    {
      "entropy": 5.9198017597198485,
      "epoch": 3.3146898585495075,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039271885212074134,
      "loss": 5.4786,
      "mean_token_accuracy": 0.19344079345464707,
      "num_tokens": 64872015.0,
      "step": 33160
    },
    {
      "entropy": 5.852486896514892,
      "epoch": 3.315189683610736,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039268843592149094,
      "loss": 5.2664,
      "mean_token_accuracy": 0.2114405393600464,
      "num_tokens": 64881497.0,
      "step": 33165
    },
    {
      "entropy": 5.921627044677734,
      "epoch": 3.315689508671965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039265801676122977,
      "loss": 5.3792,
      "mean_token_accuracy": 0.19851885735988617,
      "num_tokens": 64890729.0,
      "step": 33170
    },
    {
      "entropy": 5.955086088180542,
      "epoch": 3.3161893337331936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039262759464072316,
      "loss": 5.4994,
      "mean_token_accuracy": 0.19219046980142593,
      "num_tokens": 64900868.0,
      "step": 33175
    },
    {
      "entropy": 5.88408145904541,
      "epoch": 3.316689158794422,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003925971695607366,
      "loss": 5.4974,
      "mean_token_accuracy": 0.18621675074100494,
      "num_tokens": 64910660.0,
      "step": 33180
    },
    {
      "entropy": 5.950798225402832,
      "epoch": 3.3171889838556505,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039256674152203546,
      "loss": 5.5053,
      "mean_token_accuracy": 0.1853724241256714,
      "num_tokens": 64920897.0,
      "step": 33185
    },
    {
      "entropy": 5.953738069534301,
      "epoch": 3.3176888089168792,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003925363105253854,
      "loss": 5.3942,
      "mean_token_accuracy": 0.19560682028532028,
      "num_tokens": 64930716.0,
      "step": 33190
    },
    {
      "entropy": 5.848955249786377,
      "epoch": 3.3181886339781075,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039250587657155197,
      "loss": 5.3742,
      "mean_token_accuracy": 0.19880760759115218,
      "num_tokens": 64940270.0,
      "step": 33195
    },
    {
      "entropy": 5.888826179504394,
      "epoch": 3.318688459039336,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003924754396613009,
      "loss": 5.4253,
      "mean_token_accuracy": 0.19558695554733277,
      "num_tokens": 64949937.0,
      "step": 33200
    },
    {
      "entropy": 5.88878378868103,
      "epoch": 3.319188284100565,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003924449997953981,
      "loss": 5.3853,
      "mean_token_accuracy": 0.19884882420301436,
      "num_tokens": 64959830.0,
      "step": 33205
    },
    {
      "entropy": 5.946761560440064,
      "epoch": 3.3196881091617936,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039241455697460923,
      "loss": 5.3871,
      "mean_token_accuracy": 0.1994066432118416,
      "num_tokens": 64970391.0,
      "step": 33210
    },
    {
      "entropy": 5.835961818695068,
      "epoch": 3.320187934223022,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039238411119970034,
      "loss": 5.3497,
      "mean_token_accuracy": 0.21257767230272293,
      "num_tokens": 64980267.0,
      "step": 33215
    },
    {
      "entropy": 5.784049129486084,
      "epoch": 3.3206877592842505,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003923536624714374,
      "loss": 5.313,
      "mean_token_accuracy": 0.20547708868980408,
      "num_tokens": 64988967.0,
      "step": 33220
    },
    {
      "entropy": 5.842833375930786,
      "epoch": 3.3211875843454792,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039232321079058657,
      "loss": 5.4662,
      "mean_token_accuracy": 0.19230620712041854,
      "num_tokens": 64998810.0,
      "step": 33225
    },
    {
      "entropy": 5.871535921096802,
      "epoch": 3.3216874094067075,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039229275615791393,
      "loss": 5.4027,
      "mean_token_accuracy": 0.19558112174272538,
      "num_tokens": 65007959.0,
      "step": 33230
    },
    {
      "entropy": 5.9114830493927,
      "epoch": 3.322187234467936,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003922622985741857,
      "loss": 5.4165,
      "mean_token_accuracy": 0.19945787489414216,
      "num_tokens": 65017482.0,
      "step": 33235
    },
    {
      "entropy": 5.90962061882019,
      "epoch": 3.322687059529165,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039223183804016823,
      "loss": 5.4338,
      "mean_token_accuracy": 0.20306237936019897,
      "num_tokens": 65027906.0,
      "step": 33240
    },
    {
      "entropy": 5.895747184753418,
      "epoch": 3.3231868845903936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039220137455662785,
      "loss": 5.4453,
      "mean_token_accuracy": 0.19500376135110856,
      "num_tokens": 65037154.0,
      "step": 33245
    },
    {
      "entropy": 5.84501166343689,
      "epoch": 3.323686709651622,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039217090812433106,
      "loss": 5.3437,
      "mean_token_accuracy": 0.2058848112821579,
      "num_tokens": 65046981.0,
      "step": 33250
    },
    {
      "entropy": 5.854577922821045,
      "epoch": 3.3241865347128505,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003921404387440444,
      "loss": 5.4276,
      "mean_token_accuracy": 0.19295886009931565,
      "num_tokens": 65055643.0,
      "step": 33255
    },
    {
      "entropy": 5.8571388721466064,
      "epoch": 3.3246863597740792,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003921099664165344,
      "loss": 5.3937,
      "mean_token_accuracy": 0.2054797738790512,
      "num_tokens": 65064997.0,
      "step": 33260
    },
    {
      "entropy": 5.857975959777832,
      "epoch": 3.3251861848353075,
      "grad_norm": 1.125,
      "learning_rate": 0.0003920794911425678,
      "loss": 5.3513,
      "mean_token_accuracy": 0.20159472078084945,
      "num_tokens": 65074862.0,
      "step": 33265
    },
    {
      "entropy": 5.864541912078858,
      "epoch": 3.325686009896536,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039204901292291133,
      "loss": 5.4049,
      "mean_token_accuracy": 0.20194824039936066,
      "num_tokens": 65084633.0,
      "step": 33270
    },
    {
      "entropy": 5.854910564422608,
      "epoch": 3.326185834957765,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003920185317583318,
      "loss": 5.3945,
      "mean_token_accuracy": 0.20105578750371933,
      "num_tokens": 65093914.0,
      "step": 33275
    },
    {
      "entropy": 5.865666723251342,
      "epoch": 3.3266856600189936,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003919880476495962,
      "loss": 5.3953,
      "mean_token_accuracy": 0.20346188247203828,
      "num_tokens": 65103176.0,
      "step": 33280
    },
    {
      "entropy": 5.891297006607056,
      "epoch": 3.327185485080222,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003919575605974713,
      "loss": 5.3424,
      "mean_token_accuracy": 0.2140410289168358,
      "num_tokens": 65111744.0,
      "step": 33285
    },
    {
      "entropy": 5.884328889846802,
      "epoch": 3.3276853101414505,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003919270706027244,
      "loss": 5.4434,
      "mean_token_accuracy": 0.19730449318885804,
      "num_tokens": 65121204.0,
      "step": 33290
    },
    {
      "entropy": 5.777101612091064,
      "epoch": 3.3281851352026792,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003918965776661224,
      "loss": 5.2764,
      "mean_token_accuracy": 0.20913276076316833,
      "num_tokens": 65131741.0,
      "step": 33295
    },
    {
      "entropy": 5.883254623413086,
      "epoch": 3.3286849602639075,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039186608178843257,
      "loss": 5.4083,
      "mean_token_accuracy": 0.20159930735826492,
      "num_tokens": 65141513.0,
      "step": 33300
    },
    {
      "entropy": 5.837399959564209,
      "epoch": 3.329184785325136,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039183558297042225,
      "loss": 5.3228,
      "mean_token_accuracy": 0.20727811604738236,
      "num_tokens": 65151353.0,
      "step": 33305
    },
    {
      "entropy": 5.880212497711182,
      "epoch": 3.329684610386365,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039180508121285866,
      "loss": 5.4349,
      "mean_token_accuracy": 0.1995631769299507,
      "num_tokens": 65161669.0,
      "step": 33310
    },
    {
      "entropy": 5.8479310989379885,
      "epoch": 3.3301844354475936,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003917745765165094,
      "loss": 5.4013,
      "mean_token_accuracy": 0.19588216692209243,
      "num_tokens": 65171287.0,
      "step": 33315
    },
    {
      "entropy": 5.850385141372681,
      "epoch": 3.330684260508822,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003917440688821417,
      "loss": 5.3971,
      "mean_token_accuracy": 0.19719241857528685,
      "num_tokens": 65181232.0,
      "step": 33320
    },
    {
      "entropy": 5.919908428192139,
      "epoch": 3.3311840855700505,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003917135583105235,
      "loss": 5.4546,
      "mean_token_accuracy": 0.19102376252412795,
      "num_tokens": 65191130.0,
      "step": 33325
    },
    {
      "entropy": 5.895980453491211,
      "epoch": 3.331683910631279,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039168304480242196,
      "loss": 5.3381,
      "mean_token_accuracy": 0.20706142783164977,
      "num_tokens": 65201941.0,
      "step": 33330
    },
    {
      "entropy": 5.872101640701294,
      "epoch": 3.3321837356925075,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003916525283586051,
      "loss": 5.3618,
      "mean_token_accuracy": 0.20189237743616104,
      "num_tokens": 65212837.0,
      "step": 33335
    },
    {
      "entropy": 5.917457580566406,
      "epoch": 3.332683560753736,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003916220089798407,
      "loss": 5.4082,
      "mean_token_accuracy": 0.19773049056529998,
      "num_tokens": 65223085.0,
      "step": 33340
    },
    {
      "entropy": 5.844998407363891,
      "epoch": 3.333183385814965,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039159148666689656,
      "loss": 5.3831,
      "mean_token_accuracy": 0.20073454231023788,
      "num_tokens": 65231921.0,
      "step": 33345
    },
    {
      "entropy": 5.871171569824218,
      "epoch": 3.3336832108761936,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003915609614205406,
      "loss": 5.3913,
      "mean_token_accuracy": 0.2013797342777252,
      "num_tokens": 65242025.0,
      "step": 33350
    },
    {
      "entropy": 5.850034761428833,
      "epoch": 3.334183035937422,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003915304332415408,
      "loss": 5.3614,
      "mean_token_accuracy": 0.20876762121915818,
      "num_tokens": 65251496.0,
      "step": 33355
    },
    {
      "entropy": 5.9034263610839846,
      "epoch": 3.3346828609986505,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003914999021306653,
      "loss": 5.43,
      "mean_token_accuracy": 0.19536297768354416,
      "num_tokens": 65261657.0,
      "step": 33360
    },
    {
      "entropy": 5.8448773384094235,
      "epoch": 3.335182686059879,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003914693680886823,
      "loss": 5.4024,
      "mean_token_accuracy": 0.19745106995105743,
      "num_tokens": 65270620.0,
      "step": 33365
    },
    {
      "entropy": 5.814071130752564,
      "epoch": 3.3356825111211075,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039143883111635995,
      "loss": 5.3396,
      "mean_token_accuracy": 0.20377412140369416,
      "num_tokens": 65279306.0,
      "step": 33370
    },
    {
      "entropy": 5.877602386474609,
      "epoch": 3.336182336182336,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003914082912144666,
      "loss": 5.4583,
      "mean_token_accuracy": 0.1984570786356926,
      "num_tokens": 65288470.0,
      "step": 33375
    },
    {
      "entropy": 5.864353132247925,
      "epoch": 3.336682161243565,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039137774838377047,
      "loss": 5.2976,
      "mean_token_accuracy": 0.2029828056693077,
      "num_tokens": 65299019.0,
      "step": 33380
    },
    {
      "entropy": 5.895981645584106,
      "epoch": 3.3371819863047936,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003913472026250402,
      "loss": 5.3724,
      "mean_token_accuracy": 0.1989423468708992,
      "num_tokens": 65307821.0,
      "step": 33385
    },
    {
      "entropy": 5.911171770095825,
      "epoch": 3.337681811366022,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003913166539390443,
      "loss": 5.4983,
      "mean_token_accuracy": 0.19113627076148987,
      "num_tokens": 65317684.0,
      "step": 33390
    },
    {
      "entropy": 5.87345871925354,
      "epoch": 3.3381816364272505,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003912861023265513,
      "loss": 5.3538,
      "mean_token_accuracy": 0.20019665360450745,
      "num_tokens": 65327327.0,
      "step": 33395
    },
    {
      "entropy": 5.846497821807861,
      "epoch": 3.338681461488479,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003912555477883299,
      "loss": 5.3597,
      "mean_token_accuracy": 0.20355502516031265,
      "num_tokens": 65336236.0,
      "step": 33400
    },
    {
      "entropy": 5.723798179626465,
      "epoch": 3.3391812865497075,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003912249903251488,
      "loss": 5.2581,
      "mean_token_accuracy": 0.2183630123734474,
      "num_tokens": 65346780.0,
      "step": 33405
    },
    {
      "entropy": 5.95919246673584,
      "epoch": 3.339681111610936,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003911944299377768,
      "loss": 5.4934,
      "mean_token_accuracy": 0.18526242822408676,
      "num_tokens": 65357758.0,
      "step": 33410
    },
    {
      "entropy": 5.944570589065552,
      "epoch": 3.340180936672165,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003911638666269829,
      "loss": 5.4068,
      "mean_token_accuracy": 0.1979237377643585,
      "num_tokens": 65367959.0,
      "step": 33415
    },
    {
      "entropy": 5.7550046920776365,
      "epoch": 3.340680761733393,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039113330039353605,
      "loss": 5.1727,
      "mean_token_accuracy": 0.21442694514989852,
      "num_tokens": 65376880.0,
      "step": 33420
    },
    {
      "entropy": 5.803403425216675,
      "epoch": 3.341180586794622,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039110273123820524,
      "loss": 5.3665,
      "mean_token_accuracy": 0.20287421345710754,
      "num_tokens": 65386034.0,
      "step": 33425
    },
    {
      "entropy": 5.862145471572876,
      "epoch": 3.3416804118558505,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039107215916175947,
      "loss": 5.3941,
      "mean_token_accuracy": 0.2010548710823059,
      "num_tokens": 65396034.0,
      "step": 33430
    },
    {
      "entropy": 5.937851476669311,
      "epoch": 3.3421802369170788,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003910415841649681,
      "loss": 5.5116,
      "mean_token_accuracy": 0.1869705393910408,
      "num_tokens": 65406071.0,
      "step": 33435
    },
    {
      "entropy": 5.835371398925782,
      "epoch": 3.3426800619783075,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003910110062486004,
      "loss": 5.2536,
      "mean_token_accuracy": 0.20866034775972367,
      "num_tokens": 65415845.0,
      "step": 33440
    },
    {
      "entropy": 5.903462648391724,
      "epoch": 3.343179887039536,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039098042541342554,
      "loss": 5.4176,
      "mean_token_accuracy": 0.1982878714799881,
      "num_tokens": 65425644.0,
      "step": 33445
    },
    {
      "entropy": 5.947914266586304,
      "epoch": 3.343679712100765,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039094984166021307,
      "loss": 5.473,
      "mean_token_accuracy": 0.1898080289363861,
      "num_tokens": 65436889.0,
      "step": 33450
    },
    {
      "entropy": 5.821611070632935,
      "epoch": 3.344179537161993,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003909192549897325,
      "loss": 5.3266,
      "mean_token_accuracy": 0.21002336144447326,
      "num_tokens": 65445922.0,
      "step": 33455
    },
    {
      "entropy": 5.858205509185791,
      "epoch": 3.344679362223222,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003908886654027532,
      "loss": 5.3342,
      "mean_token_accuracy": 0.20200039446353912,
      "num_tokens": 65455021.0,
      "step": 33460
    },
    {
      "entropy": 5.8101073741912845,
      "epoch": 3.3451791872844505,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039085807290004495,
      "loss": 5.2867,
      "mean_token_accuracy": 0.21007461249828338,
      "num_tokens": 65464806.0,
      "step": 33465
    },
    {
      "entropy": 5.839506578445435,
      "epoch": 3.3456790123456788,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039082747748237737,
      "loss": 5.3789,
      "mean_token_accuracy": 0.1934867173433304,
      "num_tokens": 65474911.0,
      "step": 33470
    },
    {
      "entropy": 5.76474027633667,
      "epoch": 3.3461788374069075,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003907968791505202,
      "loss": 5.3172,
      "mean_token_accuracy": 0.21098895817995073,
      "num_tokens": 65485334.0,
      "step": 33475
    },
    {
      "entropy": 5.830994033813477,
      "epoch": 3.346678662468136,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039076627790524343,
      "loss": 5.3907,
      "mean_token_accuracy": 0.2023337349295616,
      "num_tokens": 65495327.0,
      "step": 33480
    },
    {
      "entropy": 5.903634023666382,
      "epoch": 3.347178487529365,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039073567374731685,
      "loss": 5.4194,
      "mean_token_accuracy": 0.19314197897911073,
      "num_tokens": 65506224.0,
      "step": 33485
    },
    {
      "entropy": 5.945837354660034,
      "epoch": 3.347678312590593,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039070506667751066,
      "loss": 5.4951,
      "mean_token_accuracy": 0.1903420567512512,
      "num_tokens": 65515875.0,
      "step": 33490
    },
    {
      "entropy": 5.898107242584229,
      "epoch": 3.348178137651822,
      "grad_norm": 1.125,
      "learning_rate": 0.0003906744566965946,
      "loss": 5.3771,
      "mean_token_accuracy": 0.2026997834444046,
      "num_tokens": 65524936.0,
      "step": 33495
    },
    {
      "entropy": 5.8895470142364506,
      "epoch": 3.3486779627130505,
      "grad_norm": 1.1875,
      "learning_rate": 0.000390643843805339,
      "loss": 5.3633,
      "mean_token_accuracy": 0.2015493780374527,
      "num_tokens": 65534249.0,
      "step": 33500
    },
    {
      "entropy": 5.814746427536011,
      "epoch": 3.3491777877742788,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000390613228004514,
      "loss": 5.3297,
      "mean_token_accuracy": 0.20280729830265046,
      "num_tokens": 65543832.0,
      "step": 33505
    },
    {
      "entropy": 5.895641279220581,
      "epoch": 3.3496776128355075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039058260929489,
      "loss": 5.3515,
      "mean_token_accuracy": 0.2042974501848221,
      "num_tokens": 65553671.0,
      "step": 33510
    },
    {
      "entropy": 5.836749267578125,
      "epoch": 3.350177437896736,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039055198767723725,
      "loss": 5.3918,
      "mean_token_accuracy": 0.19599844366312028,
      "num_tokens": 65563111.0,
      "step": 33515
    },
    {
      "entropy": 5.947873258590699,
      "epoch": 3.350677262957965,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039052136315232625,
      "loss": 5.5603,
      "mean_token_accuracy": 0.1863631621003151,
      "num_tokens": 65575043.0,
      "step": 33520
    },
    {
      "entropy": 5.8511041641235355,
      "epoch": 3.351177088019193,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039049073572092755,
      "loss": 5.316,
      "mean_token_accuracy": 0.20631748288869858,
      "num_tokens": 65585152.0,
      "step": 33525
    },
    {
      "entropy": 5.85969820022583,
      "epoch": 3.351676913080422,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003904601053838116,
      "loss": 5.39,
      "mean_token_accuracy": 0.1993123173713684,
      "num_tokens": 65595220.0,
      "step": 33530
    },
    {
      "entropy": 5.924767541885376,
      "epoch": 3.3521767381416505,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039042947214174906,
      "loss": 5.4296,
      "mean_token_accuracy": 0.19818013459444045,
      "num_tokens": 65604213.0,
      "step": 33535
    },
    {
      "entropy": 5.947137546539307,
      "epoch": 3.3526765632028788,
      "grad_norm": 1.125,
      "learning_rate": 0.00039039883599551074,
      "loss": 5.4519,
      "mean_token_accuracy": 0.19431852102279662,
      "num_tokens": 65613399.0,
      "step": 33540
    },
    {
      "entropy": 5.948207044601441,
      "epoch": 3.3531763882641075,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003903681969458674,
      "loss": 5.5801,
      "mean_token_accuracy": 0.18804522454738617,
      "num_tokens": 65623191.0,
      "step": 33545
    },
    {
      "entropy": 5.856431531906128,
      "epoch": 3.353676213325336,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003903375549935899,
      "loss": 5.3293,
      "mean_token_accuracy": 0.20765096992254256,
      "num_tokens": 65633645.0,
      "step": 33550
    },
    {
      "entropy": 5.807925176620484,
      "epoch": 3.354176038386565,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003903069101394492,
      "loss": 5.2422,
      "mean_token_accuracy": 0.2166396513581276,
      "num_tokens": 65643854.0,
      "step": 33555
    },
    {
      "entropy": 5.778473806381226,
      "epoch": 3.354675863447793,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003902762623842164,
      "loss": 5.2583,
      "mean_token_accuracy": 0.20984948575496673,
      "num_tokens": 65652458.0,
      "step": 33560
    },
    {
      "entropy": 5.893372869491577,
      "epoch": 3.355175688509022,
      "grad_norm": 1.125,
      "learning_rate": 0.00039024561172866236,
      "loss": 5.4695,
      "mean_token_accuracy": 0.18967238962650299,
      "num_tokens": 65661235.0,
      "step": 33565
    },
    {
      "entropy": 5.919101619720459,
      "epoch": 3.3556755135702505,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003902149581735585,
      "loss": 5.4227,
      "mean_token_accuracy": 0.20229121595621108,
      "num_tokens": 65671315.0,
      "step": 33570
    },
    {
      "entropy": 5.867697143554688,
      "epoch": 3.3561753386314788,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039018430171967584,
      "loss": 5.337,
      "mean_token_accuracy": 0.202940534055233,
      "num_tokens": 65679892.0,
      "step": 33575
    },
    {
      "entropy": 5.7917876720428465,
      "epoch": 3.3566751636927075,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003901536423677859,
      "loss": 5.4153,
      "mean_token_accuracy": 0.19835300594568253,
      "num_tokens": 65689812.0,
      "step": 33580
    },
    {
      "entropy": 5.858279705047607,
      "epoch": 3.357174988753936,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003901229801186598,
      "loss": 5.3316,
      "mean_token_accuracy": 0.20161626636981964,
      "num_tokens": 65699172.0,
      "step": 33585
    },
    {
      "entropy": 5.9077178001403805,
      "epoch": 3.357674813815165,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003900923149730693,
      "loss": 5.4042,
      "mean_token_accuracy": 0.1958802670240402,
      "num_tokens": 65708595.0,
      "step": 33590
    },
    {
      "entropy": 5.844076919555664,
      "epoch": 3.358174638876393,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039006164693178573,
      "loss": 5.3502,
      "mean_token_accuracy": 0.20941908061504363,
      "num_tokens": 65717994.0,
      "step": 33595
    },
    {
      "entropy": 5.854034948348999,
      "epoch": 3.358674463937622,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003900309759955806,
      "loss": 5.3266,
      "mean_token_accuracy": 0.20648419559001924,
      "num_tokens": 65727796.0,
      "step": 33600
    },
    {
      "entropy": 5.941073179244995,
      "epoch": 3.3591742889988505,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003900003021652259,
      "loss": 5.4908,
      "mean_token_accuracy": 0.19405413866043092,
      "num_tokens": 65737338.0,
      "step": 33605
    },
    {
      "entropy": 5.839534282684326,
      "epoch": 3.3596741140600788,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038996962544149313,
      "loss": 5.3749,
      "mean_token_accuracy": 0.2063373804092407,
      "num_tokens": 65747507.0,
      "step": 33610
    },
    {
      "entropy": 5.873959732055664,
      "epoch": 3.3601739391213075,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003899389458251541,
      "loss": 5.3296,
      "mean_token_accuracy": 0.19963929802179337,
      "num_tokens": 65756698.0,
      "step": 33615
    },
    {
      "entropy": 5.8869281768798825,
      "epoch": 3.360673764182536,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003899082633169808,
      "loss": 5.397,
      "mean_token_accuracy": 0.2067707136273384,
      "num_tokens": 65767595.0,
      "step": 33620
    },
    {
      "entropy": 5.848145818710327,
      "epoch": 3.361173589243765,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038987757791774513,
      "loss": 5.3509,
      "mean_token_accuracy": 0.1999099001288414,
      "num_tokens": 65777268.0,
      "step": 33625
    },
    {
      "entropy": 5.852261352539062,
      "epoch": 3.361673414304993,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003898468896282193,
      "loss": 5.3572,
      "mean_token_accuracy": 0.2038755387067795,
      "num_tokens": 65788137.0,
      "step": 33630
    },
    {
      "entropy": 5.8372279644012455,
      "epoch": 3.362173239366222,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003898161984491752,
      "loss": 5.342,
      "mean_token_accuracy": 0.20099092423915862,
      "num_tokens": 65797849.0,
      "step": 33635
    },
    {
      "entropy": 5.914824962615967,
      "epoch": 3.3626730644274505,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038978550438138504,
      "loss": 5.518,
      "mean_token_accuracy": 0.19086427837610245,
      "num_tokens": 65808108.0,
      "step": 33640
    },
    {
      "entropy": 5.955737829208374,
      "epoch": 3.3631728894886788,
      "grad_norm": 1.0,
      "learning_rate": 0.0003897548074256211,
      "loss": 5.4712,
      "mean_token_accuracy": 0.19081235826015472,
      "num_tokens": 65819469.0,
      "step": 33645
    },
    {
      "entropy": 5.997887086868286,
      "epoch": 3.3636727145499075,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003897241075826559,
      "loss": 5.4813,
      "mean_token_accuracy": 0.194147989153862,
      "num_tokens": 65829546.0,
      "step": 33650
    },
    {
      "entropy": 5.83107476234436,
      "epoch": 3.364172539611136,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038969340485326147,
      "loss": 5.3793,
      "mean_token_accuracy": 0.20076631158590316,
      "num_tokens": 65839556.0,
      "step": 33655
    },
    {
      "entropy": 5.774729299545288,
      "epoch": 3.364672364672365,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003896626992382106,
      "loss": 5.2627,
      "mean_token_accuracy": 0.21876406967639922,
      "num_tokens": 65849098.0,
      "step": 33660
    },
    {
      "entropy": 5.830813646316528,
      "epoch": 3.365172189733593,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038963199073827574,
      "loss": 5.3225,
      "mean_token_accuracy": 0.21232474148273467,
      "num_tokens": 65859161.0,
      "step": 33665
    },
    {
      "entropy": 5.779143381118774,
      "epoch": 3.365672014794822,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038960127935422947,
      "loss": 5.3414,
      "mean_token_accuracy": 0.20645136535167694,
      "num_tokens": 65869246.0,
      "step": 33670
    },
    {
      "entropy": 5.865658283233643,
      "epoch": 3.3661718398560505,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003895705650868445,
      "loss": 5.3829,
      "mean_token_accuracy": 0.2053663283586502,
      "num_tokens": 65879191.0,
      "step": 33675
    },
    {
      "entropy": 5.901481056213379,
      "epoch": 3.3666716649172788,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038953984793689355,
      "loss": 5.4506,
      "mean_token_accuracy": 0.19725974947214125,
      "num_tokens": 65890026.0,
      "step": 33680
    },
    {
      "entropy": 5.922315454483032,
      "epoch": 3.3671714899785075,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038950912790514954,
      "loss": 5.3879,
      "mean_token_accuracy": 0.19939182102680206,
      "num_tokens": 65898742.0,
      "step": 33685
    },
    {
      "entropy": 5.8876372337341305,
      "epoch": 3.367671315039736,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003894784049923854,
      "loss": 5.3452,
      "mean_token_accuracy": 0.20766426473855973,
      "num_tokens": 65907952.0,
      "step": 33690
    },
    {
      "entropy": 5.9209027767181395,
      "epoch": 3.368171140100965,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003894476791993739,
      "loss": 5.4545,
      "mean_token_accuracy": 0.19517692178487778,
      "num_tokens": 65918247.0,
      "step": 33695
    },
    {
      "entropy": 5.8359509944915775,
      "epoch": 3.368670965162193,
      "grad_norm": 1.484375,
      "learning_rate": 0.00038941695052688833,
      "loss": 5.3153,
      "mean_token_accuracy": 0.20859525203704835,
      "num_tokens": 65928986.0,
      "step": 33700
    },
    {
      "entropy": 5.780025482177734,
      "epoch": 3.369170790223422,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038938621897570175,
      "loss": 5.3062,
      "mean_token_accuracy": 0.21068748831748962,
      "num_tokens": 65939784.0,
      "step": 33705
    },
    {
      "entropy": 5.909755516052246,
      "epoch": 3.3696706152846505,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003893554845465873,
      "loss": 5.4431,
      "mean_token_accuracy": 0.19495835453271865,
      "num_tokens": 65949413.0,
      "step": 33710
    },
    {
      "entropy": 5.915617036819458,
      "epoch": 3.3701704403458788,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003893247472403183,
      "loss": 5.4076,
      "mean_token_accuracy": 0.20130571871995925,
      "num_tokens": 65958594.0,
      "step": 33715
    },
    {
      "entropy": 5.905443906784058,
      "epoch": 3.3706702654071075,
      "grad_norm": 1.21875,
      "learning_rate": 0.000389294007057668,
      "loss": 5.4319,
      "mean_token_accuracy": 0.19363591223955154,
      "num_tokens": 65967988.0,
      "step": 33720
    },
    {
      "entropy": 5.9559250831604,
      "epoch": 3.371170090468336,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038926326399941,
      "loss": 5.3779,
      "mean_token_accuracy": 0.20568354576826095,
      "num_tokens": 65977809.0,
      "step": 33725
    },
    {
      "entropy": 5.845578527450561,
      "epoch": 3.371669915529565,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003892325180663176,
      "loss": 5.3827,
      "mean_token_accuracy": 0.19696158766746522,
      "num_tokens": 65987365.0,
      "step": 33730
    },
    {
      "entropy": 5.85096435546875,
      "epoch": 3.372169740590793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003892017692591645,
      "loss": 5.3514,
      "mean_token_accuracy": 0.20832259356975555,
      "num_tokens": 65996952.0,
      "step": 33735
    },
    {
      "entropy": 5.9755655288696286,
      "epoch": 3.372669565652022,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038917101757872434,
      "loss": 5.4693,
      "mean_token_accuracy": 0.1908153399825096,
      "num_tokens": 66005975.0,
      "step": 33740
    },
    {
      "entropy": 5.878107357025146,
      "epoch": 3.3731693907132505,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003891402630257706,
      "loss": 5.4071,
      "mean_token_accuracy": 0.1950754463672638,
      "num_tokens": 66015175.0,
      "step": 33745
    },
    {
      "entropy": 5.828190088272095,
      "epoch": 3.3736692157744788,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038910950560107734,
      "loss": 5.4306,
      "mean_token_accuracy": 0.1948390483856201,
      "num_tokens": 66025339.0,
      "step": 33750
    },
    {
      "entropy": 5.851226949691773,
      "epoch": 3.3741690408357075,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038907874530541824,
      "loss": 5.3095,
      "mean_token_accuracy": 0.19986644089221955,
      "num_tokens": 66034545.0,
      "step": 33755
    },
    {
      "entropy": 5.828026008605957,
      "epoch": 3.374668865896936,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038904798213956716,
      "loss": 5.3663,
      "mean_token_accuracy": 0.20380838364362716,
      "num_tokens": 66044664.0,
      "step": 33760
    },
    {
      "entropy": 5.915069341659546,
      "epoch": 3.375168690958165,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003890172161042983,
      "loss": 5.3532,
      "mean_token_accuracy": 0.20409768670797349,
      "num_tokens": 66053965.0,
      "step": 33765
    },
    {
      "entropy": 6.005077266693116,
      "epoch": 3.375668516019393,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038898644720038563,
      "loss": 5.4668,
      "mean_token_accuracy": 0.19499168246984483,
      "num_tokens": 66063249.0,
      "step": 33770
    },
    {
      "entropy": 5.798362731933594,
      "epoch": 3.376168341080622,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003889556754286032,
      "loss": 5.2873,
      "mean_token_accuracy": 0.20651095658540725,
      "num_tokens": 66073236.0,
      "step": 33775
    },
    {
      "entropy": 5.853227996826172,
      "epoch": 3.3766681661418505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038892490078972544,
      "loss": 5.3137,
      "mean_token_accuracy": 0.20768993347883224,
      "num_tokens": 66082431.0,
      "step": 33780
    },
    {
      "entropy": 5.896413660049438,
      "epoch": 3.3771679912030788,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003888941232845264,
      "loss": 5.3815,
      "mean_token_accuracy": 0.20501134544610977,
      "num_tokens": 66091789.0,
      "step": 33785
    },
    {
      "entropy": 5.875684213638306,
      "epoch": 3.3776678162643075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038886334291378047,
      "loss": 5.4438,
      "mean_token_accuracy": 0.19924987852573395,
      "num_tokens": 66100677.0,
      "step": 33790
    },
    {
      "entropy": 5.849861669540405,
      "epoch": 3.378167641325536,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003888325596782622,
      "loss": 5.3825,
      "mean_token_accuracy": 0.20564268082380294,
      "num_tokens": 66109609.0,
      "step": 33795
    },
    {
      "entropy": 5.949950075149536,
      "epoch": 3.378667466386765,
      "grad_norm": 1.28125,
      "learning_rate": 0.000388801773578746,
      "loss": 5.4565,
      "mean_token_accuracy": 0.19364732950925828,
      "num_tokens": 66119614.0,
      "step": 33800
    },
    {
      "entropy": 5.924844455718994,
      "epoch": 3.379167291447993,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003887709846160065,
      "loss": 5.441,
      "mean_token_accuracy": 0.1969011530280113,
      "num_tokens": 66129176.0,
      "step": 33805
    },
    {
      "entropy": 5.84581151008606,
      "epoch": 3.379667116509222,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003887401927908183,
      "loss": 5.3315,
      "mean_token_accuracy": 0.2019270494580269,
      "num_tokens": 66138547.0,
      "step": 33810
    },
    {
      "entropy": 5.800771045684814,
      "epoch": 3.3801669415704505,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038870939810395605,
      "loss": 5.2802,
      "mean_token_accuracy": 0.20536525249481202,
      "num_tokens": 66147450.0,
      "step": 33815
    },
    {
      "entropy": 5.877648878097534,
      "epoch": 3.3806667666316788,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038867860055619456,
      "loss": 5.349,
      "mean_token_accuracy": 0.200718554854393,
      "num_tokens": 66156758.0,
      "step": 33820
    },
    {
      "entropy": 5.869091415405274,
      "epoch": 3.3811665916929075,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003886478001483089,
      "loss": 5.3903,
      "mean_token_accuracy": 0.19583482295274734,
      "num_tokens": 66166542.0,
      "step": 33825
    },
    {
      "entropy": 5.930099868774414,
      "epoch": 3.381666416754136,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003886169968810737,
      "loss": 5.4131,
      "mean_token_accuracy": 0.19401058405637742,
      "num_tokens": 66175785.0,
      "step": 33830
    },
    {
      "entropy": 5.870937728881836,
      "epoch": 3.382166241815365,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003885861907552642,
      "loss": 5.3897,
      "mean_token_accuracy": 0.2040860801935196,
      "num_tokens": 66185012.0,
      "step": 33835
    },
    {
      "entropy": 5.853761959075928,
      "epoch": 3.382666066876593,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003885553817716553,
      "loss": 5.3217,
      "mean_token_accuracy": 0.19836530685424805,
      "num_tokens": 66194580.0,
      "step": 33840
    },
    {
      "entropy": 5.946949863433838,
      "epoch": 3.383165891937822,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003885245699310223,
      "loss": 5.4271,
      "mean_token_accuracy": 0.20294545441865922,
      "num_tokens": 66204049.0,
      "step": 33845
    },
    {
      "entropy": 5.88011794090271,
      "epoch": 3.3836657169990505,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003884937552341402,
      "loss": 5.371,
      "mean_token_accuracy": 0.19757407456636428,
      "num_tokens": 66214246.0,
      "step": 33850
    },
    {
      "entropy": 5.868472862243652,
      "epoch": 3.3841655420602788,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038846293768178444,
      "loss": 5.3628,
      "mean_token_accuracy": 0.20176077783107757,
      "num_tokens": 66224910.0,
      "step": 33855
    },
    {
      "entropy": 5.854574918746948,
      "epoch": 3.3846653671215075,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003884321172747304,
      "loss": 5.3281,
      "mean_token_accuracy": 0.20659685283899307,
      "num_tokens": 66233431.0,
      "step": 33860
    },
    {
      "entropy": 5.7967134475708,
      "epoch": 3.385165192182736,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003884012940137536,
      "loss": 5.2886,
      "mean_token_accuracy": 0.20399229377508163,
      "num_tokens": 66243376.0,
      "step": 33865
    },
    {
      "entropy": 5.845632743835449,
      "epoch": 3.385665017243965,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003883704678996293,
      "loss": 5.3597,
      "mean_token_accuracy": 0.20603939294815063,
      "num_tokens": 66252628.0,
      "step": 33870
    },
    {
      "entropy": 5.876152563095093,
      "epoch": 3.386164842305193,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038833963893313326,
      "loss": 5.3384,
      "mean_token_accuracy": 0.20149426013231278,
      "num_tokens": 66263609.0,
      "step": 33875
    },
    {
      "entropy": 5.8796929836273195,
      "epoch": 3.386664667366422,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003883088071150409,
      "loss": 5.465,
      "mean_token_accuracy": 0.19287297874689102,
      "num_tokens": 66274038.0,
      "step": 33880
    },
    {
      "entropy": 5.928383445739746,
      "epoch": 3.3871644924276505,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003882779724461282,
      "loss": 5.4878,
      "mean_token_accuracy": 0.19144389033317566,
      "num_tokens": 66282852.0,
      "step": 33885
    },
    {
      "entropy": 5.901441240310669,
      "epoch": 3.3876643174888788,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038824713492717093,
      "loss": 5.4637,
      "mean_token_accuracy": 0.19764654338359833,
      "num_tokens": 66292216.0,
      "step": 33890
    },
    {
      "entropy": 5.841231918334961,
      "epoch": 3.3881641425501074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003882162945589449,
      "loss": 5.2965,
      "mean_token_accuracy": 0.20032411813735962,
      "num_tokens": 66301140.0,
      "step": 33895
    },
    {
      "entropy": 5.911230373382568,
      "epoch": 3.388663967611336,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003881854513422259,
      "loss": 5.4475,
      "mean_token_accuracy": 0.19727034866809845,
      "num_tokens": 66311338.0,
      "step": 33900
    },
    {
      "entropy": 6.00305495262146,
      "epoch": 3.389163792672565,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038815460527779017,
      "loss": 5.5173,
      "mean_token_accuracy": 0.18564466387033463,
      "num_tokens": 66320751.0,
      "step": 33905
    },
    {
      "entropy": 5.928631973266602,
      "epoch": 3.389663617733793,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00038812375636641366,
      "loss": 5.4754,
      "mean_token_accuracy": 0.1899748459458351,
      "num_tokens": 66331048.0,
      "step": 33910
    },
    {
      "entropy": 5.92561297416687,
      "epoch": 3.390163442795022,
      "grad_norm": 1.125,
      "learning_rate": 0.00038809290460887255,
      "loss": 5.4081,
      "mean_token_accuracy": 0.1985364943742752,
      "num_tokens": 66340248.0,
      "step": 33915
    },
    {
      "entropy": 5.8624883651733395,
      "epoch": 3.3906632678562505,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038806205000594307,
      "loss": 5.363,
      "mean_token_accuracy": 0.19841270744800568,
      "num_tokens": 66350863.0,
      "step": 33920
    },
    {
      "entropy": 5.885634088516236,
      "epoch": 3.3911630929174787,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003880311925584015,
      "loss": 5.3729,
      "mean_token_accuracy": 0.2045864924788475,
      "num_tokens": 66361134.0,
      "step": 33925
    },
    {
      "entropy": 5.899930143356324,
      "epoch": 3.3916629179787074,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038800033226702417,
      "loss": 5.4247,
      "mean_token_accuracy": 0.1963112995028496,
      "num_tokens": 66370482.0,
      "step": 33930
    },
    {
      "entropy": 5.961932229995727,
      "epoch": 3.392162743039936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003879694691325877,
      "loss": 5.5527,
      "mean_token_accuracy": 0.19577737003564835,
      "num_tokens": 66380502.0,
      "step": 33935
    },
    {
      "entropy": 5.867996072769165,
      "epoch": 3.3926625681011644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038793860315586824,
      "loss": 5.3026,
      "mean_token_accuracy": 0.20588842630386353,
      "num_tokens": 66389600.0,
      "step": 33940
    },
    {
      "entropy": 5.8469874382019045,
      "epoch": 3.393162393162393,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003879077343376428,
      "loss": 5.3477,
      "mean_token_accuracy": 0.20558262318372728,
      "num_tokens": 66399029.0,
      "step": 33945
    },
    {
      "entropy": 5.860774564743042,
      "epoch": 3.393662218223622,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038787686267868765,
      "loss": 5.3421,
      "mean_token_accuracy": 0.20503697395324708,
      "num_tokens": 66409320.0,
      "step": 33950
    },
    {
      "entropy": 6.003642177581787,
      "epoch": 3.3941620432848505,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003878459881797798,
      "loss": 5.5146,
      "mean_token_accuracy": 0.1919151872396469,
      "num_tokens": 66419421.0,
      "step": 33955
    },
    {
      "entropy": 5.905330753326416,
      "epoch": 3.3946618683460787,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038781511084169586,
      "loss": 5.3901,
      "mean_token_accuracy": 0.19850335270166397,
      "num_tokens": 66429227.0,
      "step": 33960
    },
    {
      "entropy": 5.8613838195800785,
      "epoch": 3.3951616934073074,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038778423066521275,
      "loss": 5.4047,
      "mean_token_accuracy": 0.2013365298509598,
      "num_tokens": 66438945.0,
      "step": 33965
    },
    {
      "entropy": 5.91549243927002,
      "epoch": 3.395661518468536,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038775334765110746,
      "loss": 5.4512,
      "mean_token_accuracy": 0.19211975932121278,
      "num_tokens": 66449205.0,
      "step": 33970
    },
    {
      "entropy": 5.8156325340271,
      "epoch": 3.3961613435297644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038772246180015694,
      "loss": 5.2782,
      "mean_token_accuracy": 0.2098570391535759,
      "num_tokens": 66459475.0,
      "step": 33975
    },
    {
      "entropy": 5.845006132125855,
      "epoch": 3.396661168590993,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003876915731131383,
      "loss": 5.4088,
      "mean_token_accuracy": 0.20234861075878144,
      "num_tokens": 66469654.0,
      "step": 33980
    },
    {
      "entropy": 5.855386447906494,
      "epoch": 3.397160993652222,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003876606815908286,
      "loss": 5.3763,
      "mean_token_accuracy": 0.20375943630933763,
      "num_tokens": 66479590.0,
      "step": 33985
    },
    {
      "entropy": 5.886158895492554,
      "epoch": 3.39766081871345,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003876297872340052,
      "loss": 5.4231,
      "mean_token_accuracy": 0.20080761313438417,
      "num_tokens": 66490918.0,
      "step": 33990
    },
    {
      "entropy": 5.92637677192688,
      "epoch": 3.3981606437746787,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038759889004344537,
      "loss": 5.4606,
      "mean_token_accuracy": 0.19450632184743882,
      "num_tokens": 66500223.0,
      "step": 33995
    },
    {
      "entropy": 5.921319198608399,
      "epoch": 3.3986604688359074,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038756799001992646,
      "loss": 5.387,
      "mean_token_accuracy": 0.2003726363182068,
      "num_tokens": 66511194.0,
      "step": 34000
    },
    {
      "entropy": 5.8168316841125485,
      "epoch": 3.399160293897136,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003875370871642258,
      "loss": 5.3297,
      "mean_token_accuracy": 0.20655408054590224,
      "num_tokens": 66520970.0,
      "step": 34005
    },
    {
      "entropy": 5.74010648727417,
      "epoch": 3.3996601189583644,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038750618147712106,
      "loss": 5.2575,
      "mean_token_accuracy": 0.2157371833920479,
      "num_tokens": 66530490.0,
      "step": 34010
    },
    {
      "entropy": 5.893004989624023,
      "epoch": 3.400159944019593,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00038747527295938966,
      "loss": 5.4163,
      "mean_token_accuracy": 0.1986667260527611,
      "num_tokens": 66540718.0,
      "step": 34015
    },
    {
      "entropy": 5.9898974895477295,
      "epoch": 3.400659769080822,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003874443616118095,
      "loss": 5.5101,
      "mean_token_accuracy": 0.19256798475980758,
      "num_tokens": 66551422.0,
      "step": 34020
    },
    {
      "entropy": 5.83075065612793,
      "epoch": 3.40115959414205,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038741344743515796,
      "loss": 5.283,
      "mean_token_accuracy": 0.21037981808185577,
      "num_tokens": 66560547.0,
      "step": 34025
    },
    {
      "entropy": 5.908002185821533,
      "epoch": 3.4016594192032787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000387382530430213,
      "loss": 5.3907,
      "mean_token_accuracy": 0.20740943253040314,
      "num_tokens": 66569884.0,
      "step": 34030
    },
    {
      "entropy": 5.872682857513428,
      "epoch": 3.4021592442645074,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003873516105977526,
      "loss": 5.4853,
      "mean_token_accuracy": 0.19930442720651625,
      "num_tokens": 66579601.0,
      "step": 34035
    },
    {
      "entropy": 5.832856941223144,
      "epoch": 3.402659069325736,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038732068793855456,
      "loss": 5.3648,
      "mean_token_accuracy": 0.1999836891889572,
      "num_tokens": 66589398.0,
      "step": 34040
    },
    {
      "entropy": 5.959269285202026,
      "epoch": 3.4031588943869644,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003872897624533969,
      "loss": 5.5406,
      "mean_token_accuracy": 0.1920125499367714,
      "num_tokens": 66598907.0,
      "step": 34045
    },
    {
      "entropy": 5.926655197143555,
      "epoch": 3.403658719448193,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003872588341430577,
      "loss": 5.3523,
      "mean_token_accuracy": 0.1970810189843178,
      "num_tokens": 66608122.0,
      "step": 34050
    },
    {
      "entropy": 5.853489923477173,
      "epoch": 3.404158544509422,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003872279030083151,
      "loss": 5.324,
      "mean_token_accuracy": 0.20597028136253356,
      "num_tokens": 66617034.0,
      "step": 34055
    },
    {
      "entropy": 5.837900447845459,
      "epoch": 3.40465836957065,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038719696904994746,
      "loss": 5.414,
      "mean_token_accuracy": 0.2010474070906639,
      "num_tokens": 66626936.0,
      "step": 34060
    },
    {
      "entropy": 5.876448678970337,
      "epoch": 3.4051581946318787,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038716603226873286,
      "loss": 5.4102,
      "mean_token_accuracy": 0.19964381605386733,
      "num_tokens": 66636933.0,
      "step": 34065
    },
    {
      "entropy": 5.92913327217102,
      "epoch": 3.4056580196931074,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038713509266544973,
      "loss": 5.459,
      "mean_token_accuracy": 0.19146263897418975,
      "num_tokens": 66648394.0,
      "step": 34070
    },
    {
      "entropy": 5.9629411697387695,
      "epoch": 3.406157844754336,
      "grad_norm": 1.125,
      "learning_rate": 0.0003871041502408765,
      "loss": 5.4478,
      "mean_token_accuracy": 0.19798891246318817,
      "num_tokens": 66658407.0,
      "step": 34075
    },
    {
      "entropy": 5.822656583786011,
      "epoch": 3.4066576698155644,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038707320499579176,
      "loss": 5.3587,
      "mean_token_accuracy": 0.20014594346284867,
      "num_tokens": 66668864.0,
      "step": 34080
    },
    {
      "entropy": 5.891140365600586,
      "epoch": 3.407157494876793,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038704225693097404,
      "loss": 5.3746,
      "mean_token_accuracy": 0.203150437772274,
      "num_tokens": 66678691.0,
      "step": 34085
    },
    {
      "entropy": 5.88946647644043,
      "epoch": 3.407657319938022,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038701130604720194,
      "loss": 5.3764,
      "mean_token_accuracy": 0.20127663463354112,
      "num_tokens": 66688707.0,
      "step": 34090
    },
    {
      "entropy": 5.829930448532105,
      "epoch": 3.40815714499925,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003869803523452542,
      "loss": 5.3289,
      "mean_token_accuracy": 0.2002465695142746,
      "num_tokens": 66698166.0,
      "step": 34095
    },
    {
      "entropy": 5.818533563613892,
      "epoch": 3.4086569700604787,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003869493958259096,
      "loss": 5.3487,
      "mean_token_accuracy": 0.2022249937057495,
      "num_tokens": 66708392.0,
      "step": 34100
    },
    {
      "entropy": 5.936714696884155,
      "epoch": 3.4091567951217074,
      "grad_norm": 1.078125,
      "learning_rate": 0.000386918436489947,
      "loss": 5.371,
      "mean_token_accuracy": 0.2037203311920166,
      "num_tokens": 66717775.0,
      "step": 34105
    },
    {
      "entropy": 5.845577716827393,
      "epoch": 3.409656620182936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003868874743381453,
      "loss": 5.331,
      "mean_token_accuracy": 0.20844917595386506,
      "num_tokens": 66727036.0,
      "step": 34110
    },
    {
      "entropy": 5.970705509185791,
      "epoch": 3.4101564452441644,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003868565093712836,
      "loss": 5.6477,
      "mean_token_accuracy": 0.18734208047389983,
      "num_tokens": 66737930.0,
      "step": 34115
    },
    {
      "entropy": 5.86981167793274,
      "epoch": 3.410656270305393,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038682554159014094,
      "loss": 5.3502,
      "mean_token_accuracy": 0.21053337752819062,
      "num_tokens": 66749568.0,
      "step": 34120
    },
    {
      "entropy": 5.846520042419433,
      "epoch": 3.411156095366622,
      "grad_norm": 1.125,
      "learning_rate": 0.00038679457099549637,
      "loss": 5.3634,
      "mean_token_accuracy": 0.20228716135025024,
      "num_tokens": 66758725.0,
      "step": 34125
    },
    {
      "entropy": 5.8277411460876465,
      "epoch": 3.41165592042785,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003867635975881292,
      "loss": 5.3279,
      "mean_token_accuracy": 0.20490263700485228,
      "num_tokens": 66767774.0,
      "step": 34130
    },
    {
      "entropy": 5.804390001296997,
      "epoch": 3.4121557454890787,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003867326213688186,
      "loss": 5.3035,
      "mean_token_accuracy": 0.20819278210401534,
      "num_tokens": 66778017.0,
      "step": 34135
    },
    {
      "entropy": 5.890397644042968,
      "epoch": 3.4126555705503074,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038670164233834406,
      "loss": 5.3968,
      "mean_token_accuracy": 0.1964631199836731,
      "num_tokens": 66786736.0,
      "step": 34140
    },
    {
      "entropy": 5.963980770111084,
      "epoch": 3.413155395611536,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038667066049748487,
      "loss": 5.4619,
      "mean_token_accuracy": 0.19309945851564408,
      "num_tokens": 66797275.0,
      "step": 34145
    },
    {
      "entropy": 5.985675239562989,
      "epoch": 3.4136552206727644,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038663967584702064,
      "loss": 5.4957,
      "mean_token_accuracy": 0.19132672250270844,
      "num_tokens": 66806978.0,
      "step": 34150
    },
    {
      "entropy": 6.024144649505615,
      "epoch": 3.414155045733993,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000386608688387731,
      "loss": 5.5472,
      "mean_token_accuracy": 0.19555956423282622,
      "num_tokens": 66816865.0,
      "step": 34155
    },
    {
      "entropy": 5.9157078742980955,
      "epoch": 3.414654870795222,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038657769812039535,
      "loss": 5.359,
      "mean_token_accuracy": 0.20125223249197005,
      "num_tokens": 66826598.0,
      "step": 34160
    },
    {
      "entropy": 5.792907047271728,
      "epoch": 3.41515469585645,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003865467050457935,
      "loss": 5.3481,
      "mean_token_accuracy": 0.20684517472982406,
      "num_tokens": 66836720.0,
      "step": 34165
    },
    {
      "entropy": 5.865201139450074,
      "epoch": 3.4156545209176787,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038651570916470535,
      "loss": 5.446,
      "mean_token_accuracy": 0.1990370735526085,
      "num_tokens": 66845263.0,
      "step": 34170
    },
    {
      "entropy": 5.848092126846313,
      "epoch": 3.4161543459789074,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038648471047791064,
      "loss": 5.3472,
      "mean_token_accuracy": 0.20886092036962509,
      "num_tokens": 66854381.0,
      "step": 34175
    },
    {
      "entropy": 5.7412820816040036,
      "epoch": 3.416654171040136,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038645370898618924,
      "loss": 5.2512,
      "mean_token_accuracy": 0.2196292757987976,
      "num_tokens": 66863536.0,
      "step": 34180
    },
    {
      "entropy": 5.812742710113525,
      "epoch": 3.4171539961013644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038642270469032127,
      "loss": 5.2225,
      "mean_token_accuracy": 0.21878177672624588,
      "num_tokens": 66873570.0,
      "step": 34185
    },
    {
      "entropy": 5.931816053390503,
      "epoch": 3.417653821162593,
      "grad_norm": 1.125,
      "learning_rate": 0.0003863916975910867,
      "loss": 5.3643,
      "mean_token_accuracy": 0.19730760902166367,
      "num_tokens": 66882780.0,
      "step": 34190
    },
    {
      "entropy": 5.89862813949585,
      "epoch": 3.418153646223822,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003863606876892657,
      "loss": 5.4059,
      "mean_token_accuracy": 0.19924745857715606,
      "num_tokens": 66892704.0,
      "step": 34195
    },
    {
      "entropy": 5.875550270080566,
      "epoch": 3.41865347128505,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038632967498563843,
      "loss": 5.31,
      "mean_token_accuracy": 0.20922782272100449,
      "num_tokens": 66902773.0,
      "step": 34200
    },
    {
      "entropy": 5.808150291442871,
      "epoch": 3.4191532963462787,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038629865948098515,
      "loss": 5.3138,
      "mean_token_accuracy": 0.2077340617775917,
      "num_tokens": 66913059.0,
      "step": 34205
    },
    {
      "entropy": 5.915362930297851,
      "epoch": 3.4196531214075074,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038626764117608627,
      "loss": 5.3537,
      "mean_token_accuracy": 0.20198651105165483,
      "num_tokens": 66922220.0,
      "step": 34210
    },
    {
      "entropy": 5.957111549377442,
      "epoch": 3.420152946468736,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003862366200717222,
      "loss": 5.402,
      "mean_token_accuracy": 0.20191317796707153,
      "num_tokens": 66931419.0,
      "step": 34215
    },
    {
      "entropy": 5.860592460632324,
      "epoch": 3.4206527715299644,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003862055961686733,
      "loss": 5.4703,
      "mean_token_accuracy": 0.1915300026535988,
      "num_tokens": 66942570.0,
      "step": 34220
    },
    {
      "entropy": 5.841041469573975,
      "epoch": 3.421152596591193,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038617456946772035,
      "loss": 5.3195,
      "mean_token_accuracy": 0.2065463289618492,
      "num_tokens": 66953542.0,
      "step": 34225
    },
    {
      "entropy": 5.865471744537354,
      "epoch": 3.421652421652422,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003861435399696437,
      "loss": 5.3209,
      "mean_token_accuracy": 0.20581487268209459,
      "num_tokens": 66962821.0,
      "step": 34230
    },
    {
      "entropy": 5.889617681503296,
      "epoch": 3.42215224671365,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038611250767522427,
      "loss": 5.4918,
      "mean_token_accuracy": 0.1939401313662529,
      "num_tokens": 66972419.0,
      "step": 34235
    },
    {
      "entropy": 5.807274627685547,
      "epoch": 3.4226520717748787,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038608147258524273,
      "loss": 5.2888,
      "mean_token_accuracy": 0.20792116820812226,
      "num_tokens": 66980389.0,
      "step": 34240
    },
    {
      "entropy": 5.891101121902466,
      "epoch": 3.4231518968361074,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003860504347004799,
      "loss": 5.3904,
      "mean_token_accuracy": 0.19783766120672225,
      "num_tokens": 66991666.0,
      "step": 34245
    },
    {
      "entropy": 5.915721702575683,
      "epoch": 3.423651721897336,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038601939402171676,
      "loss": 5.3673,
      "mean_token_accuracy": 0.19489729404449463,
      "num_tokens": 67000892.0,
      "step": 34250
    },
    {
      "entropy": 5.876930141448975,
      "epoch": 3.4241515469585644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038598835054973413,
      "loss": 5.3579,
      "mean_token_accuracy": 0.20358145385980606,
      "num_tokens": 67011085.0,
      "step": 34255
    },
    {
      "entropy": 5.788974189758301,
      "epoch": 3.424651372019793,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038595730428531323,
      "loss": 5.2598,
      "mean_token_accuracy": 0.21606891602277756,
      "num_tokens": 67021182.0,
      "step": 34260
    },
    {
      "entropy": 5.922064685821534,
      "epoch": 3.425151197081022,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038592625522923513,
      "loss": 5.394,
      "mean_token_accuracy": 0.1947532743215561,
      "num_tokens": 67029749.0,
      "step": 34265
    },
    {
      "entropy": 5.82190523147583,
      "epoch": 3.42565102214225,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038589520338228094,
      "loss": 5.3381,
      "mean_token_accuracy": 0.2025637745857239,
      "num_tokens": 67039179.0,
      "step": 34270
    },
    {
      "entropy": 5.828551006317139,
      "epoch": 3.4261508472034787,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038586414874523204,
      "loss": 5.3732,
      "mean_token_accuracy": 0.20809566974639893,
      "num_tokens": 67049305.0,
      "step": 34275
    },
    {
      "entropy": 5.927449417114258,
      "epoch": 3.4266506722647074,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003858330913188696,
      "loss": 5.3695,
      "mean_token_accuracy": 0.20691993832588196,
      "num_tokens": 67058974.0,
      "step": 34280
    },
    {
      "entropy": 5.9030430793762205,
      "epoch": 3.427150497325936,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038580203110397517,
      "loss": 5.4779,
      "mean_token_accuracy": 0.19582845270633698,
      "num_tokens": 67069132.0,
      "step": 34285
    },
    {
      "entropy": 5.906346559524536,
      "epoch": 3.4276503223871644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038577096810133013,
      "loss": 5.3998,
      "mean_token_accuracy": 0.20445077270269393,
      "num_tokens": 67078696.0,
      "step": 34290
    },
    {
      "entropy": 5.913427734375,
      "epoch": 3.428150147448393,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038573990231171603,
      "loss": 5.366,
      "mean_token_accuracy": 0.2010268822312355,
      "num_tokens": 67088458.0,
      "step": 34295
    },
    {
      "entropy": 5.8823236465454105,
      "epoch": 3.428649972509622,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038570883373591447,
      "loss": 5.429,
      "mean_token_accuracy": 0.20514716506004332,
      "num_tokens": 67097664.0,
      "step": 34300
    },
    {
      "entropy": 5.919475936889649,
      "epoch": 3.42914979757085,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038567776237470716,
      "loss": 5.4977,
      "mean_token_accuracy": 0.1888515457510948,
      "num_tokens": 67109172.0,
      "step": 34305
    },
    {
      "entropy": 5.9038971900939945,
      "epoch": 3.4296496226320787,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038564668822887576,
      "loss": 5.3555,
      "mean_token_accuracy": 0.1997350350022316,
      "num_tokens": 67119034.0,
      "step": 34310
    },
    {
      "entropy": 5.828743886947632,
      "epoch": 3.4301494476933074,
      "grad_norm": 1.125,
      "learning_rate": 0.00038561561129920224,
      "loss": 5.3221,
      "mean_token_accuracy": 0.20675025433301925,
      "num_tokens": 67128880.0,
      "step": 34315
    },
    {
      "entropy": 5.8389458656311035,
      "epoch": 3.430649272754536,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003855845315864684,
      "loss": 5.3925,
      "mean_token_accuracy": 0.19920055121183394,
      "num_tokens": 67138127.0,
      "step": 34320
    },
    {
      "entropy": 5.830389833450317,
      "epoch": 3.4311490978157644,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003855534490914562,
      "loss": 5.3624,
      "mean_token_accuracy": 0.20132609605789184,
      "num_tokens": 67147731.0,
      "step": 34325
    },
    {
      "entropy": 5.8753650188446045,
      "epoch": 3.431648922876993,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003855223638149476,
      "loss": 5.3418,
      "mean_token_accuracy": 0.2098505511879921,
      "num_tokens": 67158148.0,
      "step": 34330
    },
    {
      "entropy": 5.853548431396485,
      "epoch": 3.4321487479382218,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003854912757577249,
      "loss": 5.3541,
      "mean_token_accuracy": 0.19853999763727187,
      "num_tokens": 67167280.0,
      "step": 34335
    },
    {
      "entropy": 5.7402660846710205,
      "epoch": 3.43264857299945,
      "grad_norm": 1.4375,
      "learning_rate": 0.00038546018492057,
      "loss": 5.2884,
      "mean_token_accuracy": 0.20718641877174376,
      "num_tokens": 67176937.0,
      "step": 34340
    },
    {
      "entropy": 5.905020666122437,
      "epoch": 3.4331483980606787,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038542909130426527,
      "loss": 5.4678,
      "mean_token_accuracy": 0.19513439685106276,
      "num_tokens": 67186575.0,
      "step": 34345
    },
    {
      "entropy": 5.825785303115845,
      "epoch": 3.4336482231219074,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003853979949095931,
      "loss": 5.3611,
      "mean_token_accuracy": 0.1995113894343376,
      "num_tokens": 67196036.0,
      "step": 34350
    },
    {
      "entropy": 5.971559381484985,
      "epoch": 3.434148048183136,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038536689573733575,
      "loss": 5.4204,
      "mean_token_accuracy": 0.19660134315490724,
      "num_tokens": 67206101.0,
      "step": 34355
    },
    {
      "entropy": 5.867935752868652,
      "epoch": 3.4346478732443644,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038533579378827567,
      "loss": 5.3366,
      "mean_token_accuracy": 0.19702495038509368,
      "num_tokens": 67216342.0,
      "step": 34360
    },
    {
      "entropy": 5.924527359008789,
      "epoch": 3.435147698305593,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003853046890631955,
      "loss": 5.4816,
      "mean_token_accuracy": 0.1886381044983864,
      "num_tokens": 67225932.0,
      "step": 34365
    },
    {
      "entropy": 5.894486331939698,
      "epoch": 3.4356475233668218,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038527358156287766,
      "loss": 5.4454,
      "mean_token_accuracy": 0.19330910444259644,
      "num_tokens": 67236548.0,
      "step": 34370
    },
    {
      "entropy": 5.937205123901367,
      "epoch": 3.43614734842805,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003852424712881049,
      "loss": 5.3444,
      "mean_token_accuracy": 0.20113298147916794,
      "num_tokens": 67245803.0,
      "step": 34375
    },
    {
      "entropy": 5.879409503936768,
      "epoch": 3.4366471734892787,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038521135823965994,
      "loss": 5.3765,
      "mean_token_accuracy": 0.20004098415374755,
      "num_tokens": 67254736.0,
      "step": 34380
    },
    {
      "entropy": 5.7258477210998535,
      "epoch": 3.4371469985505074,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038518024241832553,
      "loss": 5.272,
      "mean_token_accuracy": 0.22235286235809326,
      "num_tokens": 67266300.0,
      "step": 34385
    },
    {
      "entropy": 5.841140508651733,
      "epoch": 3.437646823611736,
      "grad_norm": 1.125,
      "learning_rate": 0.0003851491238248846,
      "loss": 5.3551,
      "mean_token_accuracy": 0.20074384212493895,
      "num_tokens": 67276093.0,
      "step": 34390
    },
    {
      "entropy": 5.926232957839966,
      "epoch": 3.4381466486729644,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003851180024601201,
      "loss": 5.4457,
      "mean_token_accuracy": 0.19646218568086624,
      "num_tokens": 67286433.0,
      "step": 34395
    },
    {
      "entropy": 5.9184259414672855,
      "epoch": 3.438646473734193,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000385086878324815,
      "loss": 5.3087,
      "mean_token_accuracy": 0.21274254471063614,
      "num_tokens": 67295524.0,
      "step": 34400
    },
    {
      "entropy": 5.907589244842529,
      "epoch": 3.4391462987954218,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003850557514197523,
      "loss": 5.3995,
      "mean_token_accuracy": 0.2029608443379402,
      "num_tokens": 67305948.0,
      "step": 34405
    },
    {
      "entropy": 5.924917936325073,
      "epoch": 3.43964612385665,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003850246217457152,
      "loss": 5.4759,
      "mean_token_accuracy": 0.19024756252765657,
      "num_tokens": 67315057.0,
      "step": 34410
    },
    {
      "entropy": 5.938291263580322,
      "epoch": 3.4401459489178787,
      "grad_norm": 1.203125,
      "learning_rate": 0.000384993489303487,
      "loss": 5.4886,
      "mean_token_accuracy": 0.18160293400287628,
      "num_tokens": 67324625.0,
      "step": 34415
    },
    {
      "entropy": 5.872929048538208,
      "epoch": 3.4406457739791074,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038496235409385084,
      "loss": 5.4341,
      "mean_token_accuracy": 0.19936521500349044,
      "num_tokens": 67334356.0,
      "step": 34420
    },
    {
      "entropy": 5.844847106933594,
      "epoch": 3.441145599040336,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038493121611759016,
      "loss": 5.3194,
      "mean_token_accuracy": 0.206712244451046,
      "num_tokens": 67343704.0,
      "step": 34425
    },
    {
      "entropy": 5.88604416847229,
      "epoch": 3.4416454241015644,
      "grad_norm": 1.421875,
      "learning_rate": 0.00038490007537548843,
      "loss": 5.3879,
      "mean_token_accuracy": 0.2032754123210907,
      "num_tokens": 67352861.0,
      "step": 34430
    },
    {
      "entropy": 5.86073579788208,
      "epoch": 3.442145249162793,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003848689318683291,
      "loss": 5.4074,
      "mean_token_accuracy": 0.20676328986883163,
      "num_tokens": 67361418.0,
      "step": 34435
    },
    {
      "entropy": 5.807968473434448,
      "epoch": 3.4426450742240218,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003848377855968957,
      "loss": 5.3244,
      "mean_token_accuracy": 0.2086469203233719,
      "num_tokens": 67371707.0,
      "step": 34440
    },
    {
      "entropy": 5.964219093322754,
      "epoch": 3.44314489928525,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003848066365619718,
      "loss": 5.5096,
      "mean_token_accuracy": 0.19284441471099853,
      "num_tokens": 67381004.0,
      "step": 34445
    },
    {
      "entropy": 5.988614511489868,
      "epoch": 3.4436447243464787,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003847754847643412,
      "loss": 5.517,
      "mean_token_accuracy": 0.19489117711782455,
      "num_tokens": 67390014.0,
      "step": 34450
    },
    {
      "entropy": 5.8776500701904295,
      "epoch": 3.4441445494077074,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003847443302047877,
      "loss": 5.4393,
      "mean_token_accuracy": 0.20164924412965773,
      "num_tokens": 67399963.0,
      "step": 34455
    },
    {
      "entropy": 5.838191270828247,
      "epoch": 3.444644374468936,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000384713172884095,
      "loss": 5.3308,
      "mean_token_accuracy": 0.21134380549192427,
      "num_tokens": 67408857.0,
      "step": 34460
    },
    {
      "entropy": 5.873392057418823,
      "epoch": 3.4451441995301644,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003846820128030472,
      "loss": 5.4271,
      "mean_token_accuracy": 0.20012146532535552,
      "num_tokens": 67418823.0,
      "step": 34465
    },
    {
      "entropy": 5.766800260543823,
      "epoch": 3.445644024591393,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038465084996242824,
      "loss": 5.2693,
      "mean_token_accuracy": 0.21218811720609665,
      "num_tokens": 67428446.0,
      "step": 34470
    },
    {
      "entropy": 5.849600744247437,
      "epoch": 3.4461438496526218,
      "grad_norm": 1.109375,
      "learning_rate": 0.000384619684363022,
      "loss": 5.3966,
      "mean_token_accuracy": 0.1997404143214226,
      "num_tokens": 67438731.0,
      "step": 34475
    },
    {
      "entropy": 5.81226601600647,
      "epoch": 3.44664367471385,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003845885160056127,
      "loss": 5.3057,
      "mean_token_accuracy": 0.21306370496749877,
      "num_tokens": 67449128.0,
      "step": 34480
    },
    {
      "entropy": 5.882560968399048,
      "epoch": 3.4471434997750787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038455734489098453,
      "loss": 5.3826,
      "mean_token_accuracy": 0.20873991400003433,
      "num_tokens": 67458366.0,
      "step": 34485
    },
    {
      "entropy": 5.8586009502410885,
      "epoch": 3.4476433248363074,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003845261710199218,
      "loss": 5.3358,
      "mean_token_accuracy": 0.208488392829895,
      "num_tokens": 67468031.0,
      "step": 34490
    },
    {
      "entropy": 5.911362981796264,
      "epoch": 3.4481431498975357,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038449499439320876,
      "loss": 5.5154,
      "mean_token_accuracy": 0.1928403377532959,
      "num_tokens": 67478873.0,
      "step": 34495
    },
    {
      "entropy": 5.863597631454468,
      "epoch": 3.4486429749587644,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038446381501162983,
      "loss": 5.4516,
      "mean_token_accuracy": 0.18929319083690643,
      "num_tokens": 67489943.0,
      "step": 34500
    },
    {
      "entropy": 5.888961124420166,
      "epoch": 3.449142800019993,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038443263287596966,
      "loss": 5.2668,
      "mean_token_accuracy": 0.20872745513916016,
      "num_tokens": 67500268.0,
      "step": 34505
    },
    {
      "entropy": 5.927797269821167,
      "epoch": 3.4496426250812213,
      "grad_norm": 1.125,
      "learning_rate": 0.0003844014479870124,
      "loss": 5.4386,
      "mean_token_accuracy": 0.20227955132722855,
      "num_tokens": 67510020.0,
      "step": 34510
    },
    {
      "entropy": 5.8003651142120365,
      "epoch": 3.45014245014245,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038437026034554285,
      "loss": 5.3423,
      "mean_token_accuracy": 0.21113392561674119,
      "num_tokens": 67519200.0,
      "step": 34515
    },
    {
      "entropy": 5.902327060699463,
      "epoch": 3.4506422752036787,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003843390699523458,
      "loss": 5.3684,
      "mean_token_accuracy": 0.20390342622995378,
      "num_tokens": 67528361.0,
      "step": 34520
    },
    {
      "entropy": 5.806739664077758,
      "epoch": 3.4511421002649074,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003843078768082058,
      "loss": 5.2882,
      "mean_token_accuracy": 0.21188148707151414,
      "num_tokens": 67538280.0,
      "step": 34525
    },
    {
      "entropy": 5.858570909500122,
      "epoch": 3.4516419253261357,
      "grad_norm": 1.125,
      "learning_rate": 0.00038427668091390785,
      "loss": 5.4329,
      "mean_token_accuracy": 0.19662350416183472,
      "num_tokens": 67547170.0,
      "step": 34530
    },
    {
      "entropy": 5.963037061691284,
      "epoch": 3.4521417503873644,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038424548227023657,
      "loss": 5.47,
      "mean_token_accuracy": 0.18980810344219207,
      "num_tokens": 67557425.0,
      "step": 34535
    },
    {
      "entropy": 5.839524459838867,
      "epoch": 3.452641575448593,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038421428087797725,
      "loss": 5.3273,
      "mean_token_accuracy": 0.2058877319097519,
      "num_tokens": 67567316.0,
      "step": 34540
    },
    {
      "entropy": 5.9498978614807125,
      "epoch": 3.4531414005098213,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038418307673791465,
      "loss": 5.5309,
      "mean_token_accuracy": 0.19714234173297882,
      "num_tokens": 67577359.0,
      "step": 34545
    },
    {
      "entropy": 5.887108039855957,
      "epoch": 3.45364122557105,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003841518698508339,
      "loss": 5.4216,
      "mean_token_accuracy": 0.2051377147436142,
      "num_tokens": 67587141.0,
      "step": 34550
    },
    {
      "entropy": 5.924856424331665,
      "epoch": 3.4541410506322787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003841206602175202,
      "loss": 5.39,
      "mean_token_accuracy": 0.1946903482079506,
      "num_tokens": 67596851.0,
      "step": 34555
    },
    {
      "entropy": 5.826866960525512,
      "epoch": 3.4546408756935074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003840894478387588,
      "loss": 5.3466,
      "mean_token_accuracy": 0.20468790978193283,
      "num_tokens": 67606261.0,
      "step": 34560
    },
    {
      "entropy": 5.923756694793701,
      "epoch": 3.4551407007547357,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038405823271533496,
      "loss": 5.4843,
      "mean_token_accuracy": 0.19621749967336655,
      "num_tokens": 67615524.0,
      "step": 34565
    },
    {
      "entropy": 5.908652067184448,
      "epoch": 3.4556405258159644,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038402701484803404,
      "loss": 5.3946,
      "mean_token_accuracy": 0.20125089436769486,
      "num_tokens": 67624430.0,
      "step": 34570
    },
    {
      "entropy": 5.968265676498413,
      "epoch": 3.456140350877193,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038399579423764146,
      "loss": 5.4981,
      "mean_token_accuracy": 0.19460428804159163,
      "num_tokens": 67633841.0,
      "step": 34575
    },
    {
      "entropy": 5.8580481052398685,
      "epoch": 3.4566401759384213,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003839645708849428,
      "loss": 5.3574,
      "mean_token_accuracy": 0.20671385973691941,
      "num_tokens": 67642775.0,
      "step": 34580
    },
    {
      "entropy": 5.848635005950928,
      "epoch": 3.45714000099965,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003839333447907235,
      "loss": 5.3104,
      "mean_token_accuracy": 0.21005721986293793,
      "num_tokens": 67652615.0,
      "step": 34585
    },
    {
      "entropy": 5.938879442214966,
      "epoch": 3.4576398260608787,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003839021159557693,
      "loss": 5.4711,
      "mean_token_accuracy": 0.1944447264075279,
      "num_tokens": 67662352.0,
      "step": 34590
    },
    {
      "entropy": 5.9130534648895265,
      "epoch": 3.4581396511221074,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003838708843808659,
      "loss": 5.4349,
      "mean_token_accuracy": 0.20270546674728393,
      "num_tokens": 67672488.0,
      "step": 34595
    },
    {
      "entropy": 5.816914224624634,
      "epoch": 3.4586394761833357,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038383965006679913,
      "loss": 5.2872,
      "mean_token_accuracy": 0.201017926633358,
      "num_tokens": 67682846.0,
      "step": 34600
    },
    {
      "entropy": 5.856125020980835,
      "epoch": 3.4591393012445644,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038380841301435467,
      "loss": 5.408,
      "mean_token_accuracy": 0.19837933480739595,
      "num_tokens": 67691922.0,
      "step": 34605
    },
    {
      "entropy": 5.924303340911865,
      "epoch": 3.459639126305793,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003837771732243186,
      "loss": 5.4287,
      "mean_token_accuracy": 0.1908259719610214,
      "num_tokens": 67702013.0,
      "step": 34610
    },
    {
      "entropy": 5.875902080535889,
      "epoch": 3.4601389513670213,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003837459306974768,
      "loss": 5.3271,
      "mean_token_accuracy": 0.20726002007722855,
      "num_tokens": 67711645.0,
      "step": 34615
    },
    {
      "entropy": 5.930332994461059,
      "epoch": 3.46063877642825,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003837146854346154,
      "loss": 5.4294,
      "mean_token_accuracy": 0.20024700164794923,
      "num_tokens": 67721848.0,
      "step": 34620
    },
    {
      "entropy": 5.784970045089722,
      "epoch": 3.4611386014894787,
      "grad_norm": 1.734375,
      "learning_rate": 0.00038368343743652044,
      "loss": 5.2596,
      "mean_token_accuracy": 0.21869145333766937,
      "num_tokens": 67733004.0,
      "step": 34625
    },
    {
      "entropy": 5.86284761428833,
      "epoch": 3.4616384265507074,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003836521867039782,
      "loss": 5.3428,
      "mean_token_accuracy": 0.2041326642036438,
      "num_tokens": 67742158.0,
      "step": 34630
    },
    {
      "entropy": 5.81950044631958,
      "epoch": 3.4621382516119357,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038362093323777486,
      "loss": 5.2939,
      "mean_token_accuracy": 0.20431831777095794,
      "num_tokens": 67752340.0,
      "step": 34635
    },
    {
      "entropy": 5.835430717468261,
      "epoch": 3.4626380766731644,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003835896770386968,
      "loss": 5.4261,
      "mean_token_accuracy": 0.1928269550204277,
      "num_tokens": 67761292.0,
      "step": 34640
    },
    {
      "entropy": 5.8482362747192385,
      "epoch": 3.463137901734393,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038355841810753047,
      "loss": 5.3248,
      "mean_token_accuracy": 0.20622123628854752,
      "num_tokens": 67772303.0,
      "step": 34645
    },
    {
      "entropy": 5.947666788101197,
      "epoch": 3.4636377267956213,
      "grad_norm": 1.25,
      "learning_rate": 0.0003835271564450622,
      "loss": 5.4011,
      "mean_token_accuracy": 0.19353618770837783,
      "num_tokens": 67781825.0,
      "step": 34650
    },
    {
      "entropy": 5.800951242446899,
      "epoch": 3.46413755185685,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003834958920520786,
      "loss": 5.3001,
      "mean_token_accuracy": 0.21387438625097274,
      "num_tokens": 67792443.0,
      "step": 34655
    },
    {
      "entropy": 5.675872325897217,
      "epoch": 3.4646373769180787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003834646249293664,
      "loss": 5.2338,
      "mean_token_accuracy": 0.22375573962926865,
      "num_tokens": 67802509.0,
      "step": 34660
    },
    {
      "entropy": 5.883014106750489,
      "epoch": 3.4651372019793074,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038343335507771205,
      "loss": 5.3686,
      "mean_token_accuracy": 0.20484824180603028,
      "num_tokens": 67811868.0,
      "step": 34665
    },
    {
      "entropy": 5.932747840881348,
      "epoch": 3.4656370270405357,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038340208249790236,
      "loss": 5.4531,
      "mean_token_accuracy": 0.19581473022699356,
      "num_tokens": 67822287.0,
      "step": 34670
    },
    {
      "entropy": 5.987166118621826,
      "epoch": 3.4661368521017644,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003833708071907242,
      "loss": 5.5164,
      "mean_token_accuracy": 0.1881199672818184,
      "num_tokens": 67832869.0,
      "step": 34675
    },
    {
      "entropy": 5.774255323410034,
      "epoch": 3.466636677162993,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003833395291569644,
      "loss": 5.3525,
      "mean_token_accuracy": 0.20517683625221253,
      "num_tokens": 67843245.0,
      "step": 34680
    },
    {
      "entropy": 5.844524765014649,
      "epoch": 3.4671365022242213,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038330824839740996,
      "loss": 5.406,
      "mean_token_accuracy": 0.20347477048635482,
      "num_tokens": 67854927.0,
      "step": 34685
    },
    {
      "entropy": 5.896625328063965,
      "epoch": 3.46763632728545,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038327696491284786,
      "loss": 5.305,
      "mean_token_accuracy": 0.2046024665236473,
      "num_tokens": 67864875.0,
      "step": 34690
    },
    {
      "entropy": 5.868130445480347,
      "epoch": 3.4681361523466787,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038324567870406517,
      "loss": 5.3915,
      "mean_token_accuracy": 0.20134536921977997,
      "num_tokens": 67874717.0,
      "step": 34695
    },
    {
      "entropy": 5.786448192596436,
      "epoch": 3.4686359774079074,
      "grad_norm": 1.078125,
      "learning_rate": 0.000383214389771849,
      "loss": 5.3634,
      "mean_token_accuracy": 0.19853734523057937,
      "num_tokens": 67884692.0,
      "step": 34700
    },
    {
      "entropy": 5.8782960891723635,
      "epoch": 3.4691358024691357,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003831830981169867,
      "loss": 5.386,
      "mean_token_accuracy": 0.20274029672145844,
      "num_tokens": 67894641.0,
      "step": 34705
    },
    {
      "entropy": 5.804363346099853,
      "epoch": 3.4696356275303644,
      "grad_norm": 1.125,
      "learning_rate": 0.0003831518037402655,
      "loss": 5.3231,
      "mean_token_accuracy": 0.20934001058340074,
      "num_tokens": 67904771.0,
      "step": 34710
    },
    {
      "entropy": 5.827358388900757,
      "epoch": 3.470135452591593,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003831205066424728,
      "loss": 5.2966,
      "mean_token_accuracy": 0.20546800643205643,
      "num_tokens": 67915474.0,
      "step": 34715
    },
    {
      "entropy": 5.718321466445923,
      "epoch": 3.4706352776528213,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038308920682439586,
      "loss": 5.1919,
      "mean_token_accuracy": 0.22051469534635543,
      "num_tokens": 67924884.0,
      "step": 34720
    },
    {
      "entropy": 5.871712160110474,
      "epoch": 3.47113510271405,
      "grad_norm": 1.125,
      "learning_rate": 0.0003830579042868223,
      "loss": 5.4027,
      "mean_token_accuracy": 0.1911611944437027,
      "num_tokens": 67934696.0,
      "step": 34725
    },
    {
      "entropy": 5.844212198257447,
      "epoch": 3.4716349277752787,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003830265990305397,
      "loss": 5.3474,
      "mean_token_accuracy": 0.19912776947021485,
      "num_tokens": 67943449.0,
      "step": 34730
    },
    {
      "entropy": 5.9050404071807865,
      "epoch": 3.4721347528365074,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003829952910563356,
      "loss": 5.4885,
      "mean_token_accuracy": 0.18786506056785585,
      "num_tokens": 67953716.0,
      "step": 34735
    },
    {
      "entropy": 5.870986461639404,
      "epoch": 3.4726345778977357,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038296398036499784,
      "loss": 5.3941,
      "mean_token_accuracy": 0.20375134646892548,
      "num_tokens": 67962315.0,
      "step": 34740
    },
    {
      "entropy": 5.836594104766846,
      "epoch": 3.4731344029589644,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038293266695731414,
      "loss": 5.3013,
      "mean_token_accuracy": 0.20365292131900786,
      "num_tokens": 67971116.0,
      "step": 34745
    },
    {
      "entropy": 5.863801145553589,
      "epoch": 3.473634228020193,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038290135083407235,
      "loss": 5.4094,
      "mean_token_accuracy": 0.19690825194120407,
      "num_tokens": 67980085.0,
      "step": 34750
    },
    {
      "entropy": 5.79376072883606,
      "epoch": 3.4741340530814213,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038287003199606026,
      "loss": 5.3532,
      "mean_token_accuracy": 0.20169703364372255,
      "num_tokens": 67991288.0,
      "step": 34755
    },
    {
      "entropy": 5.894032192230225,
      "epoch": 3.47463387814265,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003828387104440659,
      "loss": 5.4809,
      "mean_token_accuracy": 0.18852618634700774,
      "num_tokens": 68000723.0,
      "step": 34760
    },
    {
      "entropy": 5.925822496414185,
      "epoch": 3.4751337032038787,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003828073861788773,
      "loss": 5.3819,
      "mean_token_accuracy": 0.1969849780201912,
      "num_tokens": 68010759.0,
      "step": 34765
    },
    {
      "entropy": 5.933238410949707,
      "epoch": 3.4756335282651074,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038277605920128275,
      "loss": 5.4224,
      "mean_token_accuracy": 0.20169125944375993,
      "num_tokens": 68020752.0,
      "step": 34770
    },
    {
      "entropy": 5.869055271148682,
      "epoch": 3.4761333533263357,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003827447295120702,
      "loss": 5.4211,
      "mean_token_accuracy": 0.20704797208309172,
      "num_tokens": 68031326.0,
      "step": 34775
    },
    {
      "entropy": 5.876821279525757,
      "epoch": 3.4766331783875644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038271339711202796,
      "loss": 5.3354,
      "mean_token_accuracy": 0.20149869918823243,
      "num_tokens": 68042012.0,
      "step": 34780
    },
    {
      "entropy": 5.852310752868652,
      "epoch": 3.477133003448793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003826820620019444,
      "loss": 5.3656,
      "mean_token_accuracy": 0.20226870030164718,
      "num_tokens": 68051918.0,
      "step": 34785
    },
    {
      "entropy": 5.921930885314941,
      "epoch": 3.4776328285100213,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003826507241826079,
      "loss": 5.4738,
      "mean_token_accuracy": 0.192349411547184,
      "num_tokens": 68062509.0,
      "step": 34790
    },
    {
      "entropy": 5.876722145080566,
      "epoch": 3.47813265357125,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038261938365480677,
      "loss": 5.3964,
      "mean_token_accuracy": 0.1949095070362091,
      "num_tokens": 68071523.0,
      "step": 34795
    },
    {
      "entropy": 5.8652520179748535,
      "epoch": 3.4786324786324787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003825880404193297,
      "loss": 5.2771,
      "mean_token_accuracy": 0.21124529391527175,
      "num_tokens": 68081053.0,
      "step": 34800
    },
    {
      "entropy": 5.855381631851197,
      "epoch": 3.4791323036937074,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038255669447696523,
      "loss": 5.3892,
      "mean_token_accuracy": 0.20563681721687316,
      "num_tokens": 68089273.0,
      "step": 34805
    },
    {
      "entropy": 5.913962984085083,
      "epoch": 3.4796321287549357,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000382525345828502,
      "loss": 5.4158,
      "mean_token_accuracy": 0.20228005051612855,
      "num_tokens": 68097476.0,
      "step": 34810
    },
    {
      "entropy": 5.9368377208709715,
      "epoch": 3.4801319538161644,
      "grad_norm": 1.25,
      "learning_rate": 0.0003824939944747288,
      "loss": 5.4487,
      "mean_token_accuracy": 0.20372408479452134,
      "num_tokens": 68107360.0,
      "step": 34815
    },
    {
      "entropy": 5.900793838500976,
      "epoch": 3.480631778877393,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003824626404164343,
      "loss": 5.4005,
      "mean_token_accuracy": 0.1972043916583061,
      "num_tokens": 68118023.0,
      "step": 34820
    },
    {
      "entropy": 5.891050958633423,
      "epoch": 3.4811316039386213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038243128365440744,
      "loss": 5.341,
      "mean_token_accuracy": 0.20531893521547318,
      "num_tokens": 68127610.0,
      "step": 34825
    },
    {
      "entropy": 5.8829052448272705,
      "epoch": 3.48163142899985,
      "grad_norm": 1.265625,
      "learning_rate": 0.000382399924189437,
      "loss": 5.3939,
      "mean_token_accuracy": 0.20658969581127168,
      "num_tokens": 68137303.0,
      "step": 34830
    },
    {
      "entropy": 5.844881677627564,
      "epoch": 3.4821312540610787,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003823685620223122,
      "loss": 5.4213,
      "mean_token_accuracy": 0.19663278460502626,
      "num_tokens": 68146997.0,
      "step": 34835
    },
    {
      "entropy": 5.896483135223389,
      "epoch": 3.4826310791223074,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003823371971538221,
      "loss": 5.4381,
      "mean_token_accuracy": 0.19984713196754456,
      "num_tokens": 68156858.0,
      "step": 34840
    },
    {
      "entropy": 5.826764726638794,
      "epoch": 3.4831309041835357,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038230582958475565,
      "loss": 5.2516,
      "mean_token_accuracy": 0.20913653671741486,
      "num_tokens": 68166108.0,
      "step": 34845
    },
    {
      "entropy": 5.931205749511719,
      "epoch": 3.4836307292447644,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038227445931590215,
      "loss": 5.4243,
      "mean_token_accuracy": 0.19616835415363312,
      "num_tokens": 68174814.0,
      "step": 34850
    },
    {
      "entropy": 5.795245027542114,
      "epoch": 3.484130554305993,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003822430863480508,
      "loss": 5.3958,
      "mean_token_accuracy": 0.20304399281740187,
      "num_tokens": 68185437.0,
      "step": 34855
    },
    {
      "entropy": 5.886572170257568,
      "epoch": 3.4846303793672213,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038221171068199104,
      "loss": 5.4079,
      "mean_token_accuracy": 0.20135853737592696,
      "num_tokens": 68194811.0,
      "step": 34860
    },
    {
      "entropy": 5.924286508560181,
      "epoch": 3.48513020442845,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003821803323185123,
      "loss": 5.4037,
      "mean_token_accuracy": 0.20340768098831177,
      "num_tokens": 68204368.0,
      "step": 34865
    },
    {
      "entropy": 5.9331573963165285,
      "epoch": 3.4856300294896787,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003821489512584039,
      "loss": 5.5213,
      "mean_token_accuracy": 0.189017915725708,
      "num_tokens": 68214012.0,
      "step": 34870
    },
    {
      "entropy": 5.792575025558472,
      "epoch": 3.4861298545509074,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038211756750245547,
      "loss": 5.2792,
      "mean_token_accuracy": 0.2157947152853012,
      "num_tokens": 68224284.0,
      "step": 34875
    },
    {
      "entropy": 5.7532281398773195,
      "epoch": 3.4866296796121357,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003820861810514566,
      "loss": 5.2417,
      "mean_token_accuracy": 0.22051554322242736,
      "num_tokens": 68234576.0,
      "step": 34880
    },
    {
      "entropy": 5.901005220413208,
      "epoch": 3.4871295046733644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038205479190619703,
      "loss": 5.3994,
      "mean_token_accuracy": 0.19792788475751877,
      "num_tokens": 68243156.0,
      "step": 34885
    },
    {
      "entropy": 5.91082615852356,
      "epoch": 3.487629329734593,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038202340006746635,
      "loss": 5.3284,
      "mean_token_accuracy": 0.21252711415290831,
      "num_tokens": 68251717.0,
      "step": 34890
    },
    {
      "entropy": 5.806917715072632,
      "epoch": 3.4881291547958213,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003819920055360545,
      "loss": 5.3457,
      "mean_token_accuracy": 0.20421903282403947,
      "num_tokens": 68261565.0,
      "step": 34895
    },
    {
      "entropy": 5.855081939697266,
      "epoch": 3.48862897985705,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038196060831275134,
      "loss": 5.465,
      "mean_token_accuracy": 0.19425881505012513,
      "num_tokens": 68271965.0,
      "step": 34900
    },
    {
      "entropy": 5.902006340026856,
      "epoch": 3.4891288049182787,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003819292083983468,
      "loss": 5.3677,
      "mean_token_accuracy": 0.2027289867401123,
      "num_tokens": 68282081.0,
      "step": 34905
    },
    {
      "entropy": 5.917070484161377,
      "epoch": 3.4896286299795074,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003818978057936308,
      "loss": 5.4428,
      "mean_token_accuracy": 0.196712988615036,
      "num_tokens": 68292762.0,
      "step": 34910
    },
    {
      "entropy": 5.760815620422363,
      "epoch": 3.4901284550407357,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038186640049939365,
      "loss": 5.3402,
      "mean_token_accuracy": 0.20510243773460388,
      "num_tokens": 68302651.0,
      "step": 34915
    },
    {
      "entropy": 5.873224401473999,
      "epoch": 3.4906282801019644,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003818349925164253,
      "loss": 5.3941,
      "mean_token_accuracy": 0.19938789159059525,
      "num_tokens": 68312385.0,
      "step": 34920
    },
    {
      "entropy": 5.749425649642944,
      "epoch": 3.491128105163193,
      "grad_norm": 1.109375,
      "learning_rate": 0.000381803581845516,
      "loss": 5.2279,
      "mean_token_accuracy": 0.21722476035356522,
      "num_tokens": 68322289.0,
      "step": 34925
    },
    {
      "entropy": 5.855150890350342,
      "epoch": 3.4916279302244213,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038177216848745614,
      "loss": 5.3476,
      "mean_token_accuracy": 0.20251842141151427,
      "num_tokens": 68331606.0,
      "step": 34930
    },
    {
      "entropy": 5.78699426651001,
      "epoch": 3.49212775528565,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003817407524430359,
      "loss": 5.2399,
      "mean_token_accuracy": 0.2130677729845047,
      "num_tokens": 68341469.0,
      "step": 34935
    },
    {
      "entropy": 5.817653846740723,
      "epoch": 3.4926275803468787,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038170933371304585,
      "loss": 5.3273,
      "mean_token_accuracy": 0.20349865406751633,
      "num_tokens": 68350821.0,
      "step": 34940
    },
    {
      "entropy": 5.801052236557007,
      "epoch": 3.4931274054081074,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003816779122982765,
      "loss": 5.3694,
      "mean_token_accuracy": 0.2047739565372467,
      "num_tokens": 68360881.0,
      "step": 34945
    },
    {
      "entropy": 5.830963182449341,
      "epoch": 3.4936272304693357,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003816464881995182,
      "loss": 5.3729,
      "mean_token_accuracy": 0.20520763397216796,
      "num_tokens": 68372068.0,
      "step": 34950
    },
    {
      "entropy": 5.841994714736939,
      "epoch": 3.4941270555305644,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003816150614175617,
      "loss": 5.3458,
      "mean_token_accuracy": 0.20994785130023957,
      "num_tokens": 68382499.0,
      "step": 34955
    },
    {
      "entropy": 5.904245042800904,
      "epoch": 3.494626880591793,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003815836319531978,
      "loss": 5.3673,
      "mean_token_accuracy": 0.20379307717084885,
      "num_tokens": 68392305.0,
      "step": 34960
    },
    {
      "entropy": 5.816765451431275,
      "epoch": 3.4951267056530213,
      "grad_norm": 1.078125,
      "learning_rate": 0.000381552199807217,
      "loss": 5.2893,
      "mean_token_accuracy": 0.2084710642695427,
      "num_tokens": 68401860.0,
      "step": 34965
    },
    {
      "entropy": 5.8843159675598145,
      "epoch": 3.49562653071425,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003815207649804104,
      "loss": 5.4392,
      "mean_token_accuracy": 0.19565978050231933,
      "num_tokens": 68411189.0,
      "step": 34970
    },
    {
      "entropy": 5.793935775756836,
      "epoch": 3.4961263557754787,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003814893274735687,
      "loss": 5.2976,
      "mean_token_accuracy": 0.2166106164455414,
      "num_tokens": 68420461.0,
      "step": 34975
    },
    {
      "entropy": 5.865027332305909,
      "epoch": 3.4966261808367074,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038145788728748294,
      "loss": 5.4036,
      "mean_token_accuracy": 0.20284028500318527,
      "num_tokens": 68430835.0,
      "step": 34980
    },
    {
      "entropy": 5.804305505752564,
      "epoch": 3.4971260058979357,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003814264444229441,
      "loss": 5.2656,
      "mean_token_accuracy": 0.21974184662103652,
      "num_tokens": 68440701.0,
      "step": 34985
    },
    {
      "entropy": 5.85975923538208,
      "epoch": 3.4976258309591644,
      "grad_norm": 1.125,
      "learning_rate": 0.0003813949988807434,
      "loss": 5.4436,
      "mean_token_accuracy": 0.20140180736780167,
      "num_tokens": 68450044.0,
      "step": 34990
    },
    {
      "entropy": 5.81974606513977,
      "epoch": 3.498125656020393,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038136355066167177,
      "loss": 5.3284,
      "mean_token_accuracy": 0.20686601847410202,
      "num_tokens": 68459986.0,
      "step": 34995
    },
    {
      "entropy": 5.877404880523682,
      "epoch": 3.4986254810816213,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003813320997665207,
      "loss": 5.4447,
      "mean_token_accuracy": 0.2007819026708603,
      "num_tokens": 68469681.0,
      "step": 35000
    },
    {
      "entropy": 5.845322895050049,
      "epoch": 3.49912530614285,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038130064619608123,
      "loss": 5.3173,
      "mean_token_accuracy": 0.22121303528547287,
      "num_tokens": 68480307.0,
      "step": 35005
    },
    {
      "entropy": 5.892283344268799,
      "epoch": 3.4996251312040787,
      "grad_norm": 1.234375,
      "learning_rate": 0.000381269189951145,
      "loss": 5.4329,
      "mean_token_accuracy": 0.1978833183646202,
      "num_tokens": 68490091.0,
      "step": 35010
    },
    {
      "entropy": 5.862207126617432,
      "epoch": 3.5001249562653074,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003812377310325032,
      "loss": 5.4016,
      "mean_token_accuracy": 0.2016397535800934,
      "num_tokens": 68500898.0,
      "step": 35015
    },
    {
      "entropy": 5.885101413726806,
      "epoch": 3.5006247813265357,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003812062694409474,
      "loss": 5.4384,
      "mean_token_accuracy": 0.1978804349899292,
      "num_tokens": 68510165.0,
      "step": 35020
    },
    {
      "entropy": 5.840692138671875,
      "epoch": 3.5011246063877643,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003811748051772693,
      "loss": 5.3475,
      "mean_token_accuracy": 0.19704463481903076,
      "num_tokens": 68519766.0,
      "step": 35025
    },
    {
      "entropy": 5.85504117012024,
      "epoch": 3.5016244314489926,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038114333824226034,
      "loss": 5.3445,
      "mean_token_accuracy": 0.20281596928834916,
      "num_tokens": 68530073.0,
      "step": 35030
    },
    {
      "entropy": 5.900664806365967,
      "epoch": 3.5021242565102213,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003811118686367123,
      "loss": 5.4363,
      "mean_token_accuracy": 0.19966645389795304,
      "num_tokens": 68540069.0,
      "step": 35035
    },
    {
      "entropy": 5.866452598571778,
      "epoch": 3.50262408157145,
      "grad_norm": 1.109375,
      "learning_rate": 0.000381080396361417,
      "loss": 5.4045,
      "mean_token_accuracy": 0.1990477904677391,
      "num_tokens": 68549342.0,
      "step": 35040
    },
    {
      "entropy": 5.852326059341431,
      "epoch": 3.5031239066326787,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003810489214171662,
      "loss": 5.3333,
      "mean_token_accuracy": 0.20669275224208833,
      "num_tokens": 68558864.0,
      "step": 35045
    },
    {
      "entropy": 5.87678542137146,
      "epoch": 3.5036237316939074,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003810174438047517,
      "loss": 5.4753,
      "mean_token_accuracy": 0.19866169095039368,
      "num_tokens": 68568320.0,
      "step": 35050
    },
    {
      "entropy": 5.903407049179077,
      "epoch": 3.5041235567551356,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003809859635249658,
      "loss": 5.4261,
      "mean_token_accuracy": 0.1920968472957611,
      "num_tokens": 68577848.0,
      "step": 35055
    },
    {
      "entropy": 5.8585343837738035,
      "epoch": 3.5046233818163643,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003809544805786003,
      "loss": 5.3356,
      "mean_token_accuracy": 0.20194528698921205,
      "num_tokens": 68587554.0,
      "step": 35060
    },
    {
      "entropy": 5.870153856277466,
      "epoch": 3.5051232068775926,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003809229949664473,
      "loss": 5.4147,
      "mean_token_accuracy": 0.19737222492694856,
      "num_tokens": 68597688.0,
      "step": 35065
    },
    {
      "entropy": 5.882437992095947,
      "epoch": 3.5056230319388213,
      "grad_norm": 1.125,
      "learning_rate": 0.0003808915066892989,
      "loss": 5.3329,
      "mean_token_accuracy": 0.21349310725927353,
      "num_tokens": 68607222.0,
      "step": 35070
    },
    {
      "entropy": 5.8672935485839846,
      "epoch": 3.50612285700005,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038086001574794755,
      "loss": 5.4332,
      "mean_token_accuracy": 0.1991879403591156,
      "num_tokens": 68616022.0,
      "step": 35075
    },
    {
      "entropy": 5.822286701202392,
      "epoch": 3.5066226820612787,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003808285221431854,
      "loss": 5.3926,
      "mean_token_accuracy": 0.1996110960841179,
      "num_tokens": 68624945.0,
      "step": 35080
    },
    {
      "entropy": 5.769114923477173,
      "epoch": 3.5071225071225074,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038079702587580486,
      "loss": 5.2473,
      "mean_token_accuracy": 0.21707170605659484,
      "num_tokens": 68634704.0,
      "step": 35085
    },
    {
      "entropy": 5.909220504760742,
      "epoch": 3.5076223321837356,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038076552694659837,
      "loss": 5.3624,
      "mean_token_accuracy": 0.20345550328493117,
      "num_tokens": 68644754.0,
      "step": 35090
    },
    {
      "entropy": 5.85959095954895,
      "epoch": 3.5081221572449643,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038073402535635854,
      "loss": 5.3441,
      "mean_token_accuracy": 0.1988835319876671,
      "num_tokens": 68653750.0,
      "step": 35095
    },
    {
      "entropy": 5.908349275588989,
      "epoch": 3.5086219823061926,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038070252110587776,
      "loss": 5.44,
      "mean_token_accuracy": 0.20417943447828293,
      "num_tokens": 68664076.0,
      "step": 35100
    },
    {
      "entropy": 5.775426816940308,
      "epoch": 3.5091218073674213,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038067101419594874,
      "loss": 5.2045,
      "mean_token_accuracy": 0.22365498691797256,
      "num_tokens": 68674353.0,
      "step": 35105
    },
    {
      "entropy": 5.9235326766967775,
      "epoch": 3.50962163242865,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038063950462736425,
      "loss": 5.499,
      "mean_token_accuracy": 0.1972525343298912,
      "num_tokens": 68683571.0,
      "step": 35110
    },
    {
      "entropy": 6.010158348083496,
      "epoch": 3.5101214574898787,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003806079924009169,
      "loss": 5.5643,
      "mean_token_accuracy": 0.19165876358747483,
      "num_tokens": 68693791.0,
      "step": 35115
    },
    {
      "entropy": 5.87861967086792,
      "epoch": 3.510621282551107,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038057647751739963,
      "loss": 5.3529,
      "mean_token_accuracy": 0.20577692091464997,
      "num_tokens": 68702654.0,
      "step": 35120
    },
    {
      "entropy": 5.872904062271118,
      "epoch": 3.5111211076123356,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003805449599776055,
      "loss": 5.3963,
      "mean_token_accuracy": 0.2006140634417534,
      "num_tokens": 68713037.0,
      "step": 35125
    },
    {
      "entropy": 5.877732563018799,
      "epoch": 3.5116209326735643,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003805134397823273,
      "loss": 5.3646,
      "mean_token_accuracy": 0.20135806500911713,
      "num_tokens": 68722395.0,
      "step": 35130
    },
    {
      "entropy": 5.833263874053955,
      "epoch": 3.5121207577347926,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000380481916932358,
      "loss": 5.3086,
      "mean_token_accuracy": 0.2048608645796776,
      "num_tokens": 68732044.0,
      "step": 35135
    },
    {
      "entropy": 5.852800750732422,
      "epoch": 3.5126205827960213,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003804503914284909,
      "loss": 5.3508,
      "mean_token_accuracy": 0.2049824446439743,
      "num_tokens": 68741903.0,
      "step": 35140
    },
    {
      "entropy": 5.855718660354614,
      "epoch": 3.51312040785725,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038041886327151904,
      "loss": 5.4137,
      "mean_token_accuracy": 0.19911206662654876,
      "num_tokens": 68750743.0,
      "step": 35145
    },
    {
      "entropy": 5.827973651885986,
      "epoch": 3.5136202329184787,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038038733246223574,
      "loss": 5.3278,
      "mean_token_accuracy": 0.2080792084336281,
      "num_tokens": 68760391.0,
      "step": 35150
    },
    {
      "entropy": 5.8900899410247805,
      "epoch": 3.514120057979707,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003803557990014343,
      "loss": 5.422,
      "mean_token_accuracy": 0.19372241646051408,
      "num_tokens": 68770433.0,
      "step": 35155
    },
    {
      "entropy": 5.851507902145386,
      "epoch": 3.5146198830409356,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000380324262889908,
      "loss": 5.3261,
      "mean_token_accuracy": 0.2121310219168663,
      "num_tokens": 68779507.0,
      "step": 35160
    },
    {
      "entropy": 5.827107191085815,
      "epoch": 3.5151197081021643,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003802927241284504,
      "loss": 5.2937,
      "mean_token_accuracy": 0.2115744397044182,
      "num_tokens": 68789632.0,
      "step": 35165
    },
    {
      "entropy": 5.808831024169922,
      "epoch": 3.5156195331633926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003802611827178549,
      "loss": 5.3826,
      "mean_token_accuracy": 0.20193106085062026,
      "num_tokens": 68799049.0,
      "step": 35170
    },
    {
      "entropy": 5.885252380371094,
      "epoch": 3.5161193582246213,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038022963865891514,
      "loss": 5.3701,
      "mean_token_accuracy": 0.20453402251005173,
      "num_tokens": 68809306.0,
      "step": 35175
    },
    {
      "entropy": 5.874835395812989,
      "epoch": 3.51661918328585,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038019809195242484,
      "loss": 5.4325,
      "mean_token_accuracy": 0.2056750997900963,
      "num_tokens": 68818851.0,
      "step": 35180
    },
    {
      "entropy": 5.870796394348145,
      "epoch": 3.5171190083470787,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003801665425991775,
      "loss": 5.3452,
      "mean_token_accuracy": 0.208365735411644,
      "num_tokens": 68829132.0,
      "step": 35185
    },
    {
      "entropy": 5.885442543029785,
      "epoch": 3.517618833408307,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003801349905999671,
      "loss": 5.3748,
      "mean_token_accuracy": 0.20035126507282258,
      "num_tokens": 68838961.0,
      "step": 35190
    },
    {
      "entropy": 5.87451057434082,
      "epoch": 3.5181186584695356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003801034359555873,
      "loss": 5.3934,
      "mean_token_accuracy": 0.20405804514884948,
      "num_tokens": 68848422.0,
      "step": 35195
    },
    {
      "entropy": 5.853027296066284,
      "epoch": 3.5186184835307643,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038007187866683217,
      "loss": 5.3649,
      "mean_token_accuracy": 0.19595470130443574,
      "num_tokens": 68857657.0,
      "step": 35200
    },
    {
      "entropy": 5.866765642166138,
      "epoch": 3.5191183085919926,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003800403187344955,
      "loss": 5.4783,
      "mean_token_accuracy": 0.19391658157110214,
      "num_tokens": 68866940.0,
      "step": 35205
    },
    {
      "entropy": 5.947752189636231,
      "epoch": 3.5196181336532213,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003800087561593716,
      "loss": 5.5207,
      "mean_token_accuracy": 0.18755348771810532,
      "num_tokens": 68878600.0,
      "step": 35210
    },
    {
      "entropy": 5.932040405273438,
      "epoch": 3.52011795871445,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003799771909422543,
      "loss": 5.4083,
      "mean_token_accuracy": 0.19707875400781633,
      "num_tokens": 68889316.0,
      "step": 35215
    },
    {
      "entropy": 5.813732242584228,
      "epoch": 3.5206177837756787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037994562308393793,
      "loss": 5.2937,
      "mean_token_accuracy": 0.2057522341609001,
      "num_tokens": 68899381.0,
      "step": 35220
    },
    {
      "entropy": 5.950585842132568,
      "epoch": 3.521117608836907,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003799140525852167,
      "loss": 5.3704,
      "mean_token_accuracy": 0.20831360071897506,
      "num_tokens": 68908830.0,
      "step": 35225
    },
    {
      "entropy": 5.831218910217285,
      "epoch": 3.5216174338981356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003798824794468849,
      "loss": 5.294,
      "mean_token_accuracy": 0.21079234629869462,
      "num_tokens": 68919492.0,
      "step": 35230
    },
    {
      "entropy": 5.823147535324097,
      "epoch": 3.5221172589593643,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003798509036697369,
      "loss": 5.3964,
      "mean_token_accuracy": 0.20354301780462264,
      "num_tokens": 68930897.0,
      "step": 35235
    },
    {
      "entropy": 5.833708095550537,
      "epoch": 3.5226170840205926,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037981932525456714,
      "loss": 5.2939,
      "mean_token_accuracy": 0.2097306728363037,
      "num_tokens": 68942202.0,
      "step": 35240
    },
    {
      "entropy": 5.884766244888306,
      "epoch": 3.5231169090818213,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003797877442021701,
      "loss": 5.3817,
      "mean_token_accuracy": 0.2099976897239685,
      "num_tokens": 68950770.0,
      "step": 35245
    },
    {
      "entropy": 5.939175319671631,
      "epoch": 3.52361673414305,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003797561605133404,
      "loss": 5.4893,
      "mean_token_accuracy": 0.19598931968212127,
      "num_tokens": 68961018.0,
      "step": 35250
    },
    {
      "entropy": 5.915323686599732,
      "epoch": 3.5241165592042787,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003797245741888727,
      "loss": 5.3779,
      "mean_token_accuracy": 0.20699852854013442,
      "num_tokens": 68970832.0,
      "step": 35255
    },
    {
      "entropy": 5.860815238952637,
      "epoch": 3.524616384265507,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037969298522956156,
      "loss": 5.3855,
      "mean_token_accuracy": 0.20338721126317977,
      "num_tokens": 68980591.0,
      "step": 35260
    },
    {
      "entropy": 5.828859663009643,
      "epoch": 3.5251162093267356,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037966139363620197,
      "loss": 5.3837,
      "mean_token_accuracy": 0.20645961463451384,
      "num_tokens": 68990578.0,
      "step": 35265
    },
    {
      "entropy": 5.878554153442383,
      "epoch": 3.5256160343879643,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003796297994095886,
      "loss": 5.306,
      "mean_token_accuracy": 0.21751852631568908,
      "num_tokens": 68999671.0,
      "step": 35270
    },
    {
      "entropy": 5.762632036209107,
      "epoch": 3.5261158594491926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003795982025505164,
      "loss": 5.2665,
      "mean_token_accuracy": 0.2083053782582283,
      "num_tokens": 69009377.0,
      "step": 35275
    },
    {
      "entropy": 5.87258882522583,
      "epoch": 3.5266156845104213,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003795666030597803,
      "loss": 5.4415,
      "mean_token_accuracy": 0.2014458030462265,
      "num_tokens": 69019785.0,
      "step": 35280
    },
    {
      "entropy": 5.86114821434021,
      "epoch": 3.52711550957165,
      "grad_norm": 1.125,
      "learning_rate": 0.00037953500093817544,
      "loss": 5.2266,
      "mean_token_accuracy": 0.21567374914884568,
      "num_tokens": 69029459.0,
      "step": 35285
    },
    {
      "entropy": 5.8850993633270265,
      "epoch": 3.5276153346328787,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037950339618649675,
      "loss": 5.3682,
      "mean_token_accuracy": 0.20081277787685395,
      "num_tokens": 69038321.0,
      "step": 35290
    },
    {
      "entropy": 5.7608397006988525,
      "epoch": 3.528115159694107,
      "grad_norm": 1.125,
      "learning_rate": 0.0003794717888055397,
      "loss": 5.2943,
      "mean_token_accuracy": 0.2119860514998436,
      "num_tokens": 69047964.0,
      "step": 35295
    },
    {
      "entropy": 5.919934940338135,
      "epoch": 3.5286149847553356,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037944017879609917,
      "loss": 5.4883,
      "mean_token_accuracy": 0.18724936097860337,
      "num_tokens": 69058217.0,
      "step": 35300
    },
    {
      "entropy": 5.9896666526794435,
      "epoch": 3.5291148098165643,
      "grad_norm": 1.125,
      "learning_rate": 0.0003794085661589707,
      "loss": 5.598,
      "mean_token_accuracy": 0.18735577315092086,
      "num_tokens": 69067819.0,
      "step": 35305
    },
    {
      "entropy": 5.9221879005432125,
      "epoch": 3.5296146348777926,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037937695089494943,
      "loss": 5.3982,
      "mean_token_accuracy": 0.20162008404731752,
      "num_tokens": 69078656.0,
      "step": 35310
    },
    {
      "entropy": 5.938488340377807,
      "epoch": 3.5301144599390213,
      "grad_norm": 1.1875,
      "learning_rate": 0.000379345333004831,
      "loss": 5.4469,
      "mean_token_accuracy": 0.1984606698155403,
      "num_tokens": 69087236.0,
      "step": 35315
    },
    {
      "entropy": 5.924704599380493,
      "epoch": 3.53061428500025,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003793137124894109,
      "loss": 5.4253,
      "mean_token_accuracy": 0.19653277844190598,
      "num_tokens": 69097305.0,
      "step": 35320
    },
    {
      "entropy": 5.913994026184082,
      "epoch": 3.5311141100614787,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037928208934948467,
      "loss": 5.4884,
      "mean_token_accuracy": 0.19210747480392457,
      "num_tokens": 69107186.0,
      "step": 35325
    },
    {
      "entropy": 5.964172315597534,
      "epoch": 3.531613935122707,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037925046358584775,
      "loss": 5.5134,
      "mean_token_accuracy": 0.196640507876873,
      "num_tokens": 69116488.0,
      "step": 35330
    },
    {
      "entropy": 5.8751424789428714,
      "epoch": 3.5321137601839356,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003792188351992961,
      "loss": 5.384,
      "mean_token_accuracy": 0.19601797312498093,
      "num_tokens": 69126201.0,
      "step": 35335
    },
    {
      "entropy": 5.840430068969726,
      "epoch": 3.5326135852451643,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037918720419062536,
      "loss": 5.3929,
      "mean_token_accuracy": 0.20414265096187592,
      "num_tokens": 69135281.0,
      "step": 35340
    },
    {
      "entropy": 5.873660612106323,
      "epoch": 3.5331134103063926,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003791555705606314,
      "loss": 5.3361,
      "mean_token_accuracy": 0.20786097049713134,
      "num_tokens": 69145064.0,
      "step": 35345
    },
    {
      "entropy": 5.859607076644897,
      "epoch": 3.5336132353676213,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037912393431011005,
      "loss": 5.3371,
      "mean_token_accuracy": 0.2072366565465927,
      "num_tokens": 69154614.0,
      "step": 35350
    },
    {
      "entropy": 5.856562995910645,
      "epoch": 3.53411306042885,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003790922954398573,
      "loss": 5.4506,
      "mean_token_accuracy": 0.19717206954956054,
      "num_tokens": 69164107.0,
      "step": 35355
    },
    {
      "entropy": 5.739122200012207,
      "epoch": 3.5346128854900787,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003790606539506692,
      "loss": 5.3322,
      "mean_token_accuracy": 0.20568230450153352,
      "num_tokens": 69174038.0,
      "step": 35360
    },
    {
      "entropy": 5.885837697982788,
      "epoch": 3.535112710551307,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037902900984334176,
      "loss": 5.4082,
      "mean_token_accuracy": 0.20172822028398513,
      "num_tokens": 69182888.0,
      "step": 35365
    },
    {
      "entropy": 5.7878210067749025,
      "epoch": 3.5356125356125356,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037899736311867126,
      "loss": 5.2783,
      "mean_token_accuracy": 0.2173325836658478,
      "num_tokens": 69192466.0,
      "step": 35370
    },
    {
      "entropy": 5.929365873336792,
      "epoch": 3.5361123606737643,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037896571377745385,
      "loss": 5.4985,
      "mean_token_accuracy": 0.19016720950603486,
      "num_tokens": 69202291.0,
      "step": 35375
    },
    {
      "entropy": 5.923820447921753,
      "epoch": 3.5366121857349926,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003789340618204859,
      "loss": 5.3546,
      "mean_token_accuracy": 0.20785085558891297,
      "num_tokens": 69211982.0,
      "step": 35380
    },
    {
      "entropy": 5.8295516014099125,
      "epoch": 3.5371120107962213,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003789024072485637,
      "loss": 5.3371,
      "mean_token_accuracy": 0.20671756714582443,
      "num_tokens": 69221787.0,
      "step": 35385
    },
    {
      "entropy": 5.77665638923645,
      "epoch": 3.53761183585745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037887075006248354,
      "loss": 5.3697,
      "mean_token_accuracy": 0.20685915201902388,
      "num_tokens": 69231536.0,
      "step": 35390
    },
    {
      "entropy": 5.825467872619629,
      "epoch": 3.5381116609186787,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037883909026304206,
      "loss": 5.285,
      "mean_token_accuracy": 0.21006059050559997,
      "num_tokens": 69240320.0,
      "step": 35395
    },
    {
      "entropy": 5.796338701248169,
      "epoch": 3.538611485979907,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037880742785103584,
      "loss": 5.3071,
      "mean_token_accuracy": 0.21058594882488252,
      "num_tokens": 69249926.0,
      "step": 35400
    },
    {
      "entropy": 5.787244415283203,
      "epoch": 3.5391113110411356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037877576282726146,
      "loss": 5.3014,
      "mean_token_accuracy": 0.2102728471159935,
      "num_tokens": 69258847.0,
      "step": 35405
    },
    {
      "entropy": 5.917558479309082,
      "epoch": 3.5396111361023643,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037874409519251563,
      "loss": 5.4348,
      "mean_token_accuracy": 0.196567565202713,
      "num_tokens": 69269205.0,
      "step": 35410
    },
    {
      "entropy": 5.890255165100098,
      "epoch": 3.5401109611635926,
      "grad_norm": 1.078125,
      "learning_rate": 0.000378712424947595,
      "loss": 5.4235,
      "mean_token_accuracy": 0.19966543465852737,
      "num_tokens": 69279324.0,
      "step": 35415
    },
    {
      "entropy": 5.871117687225341,
      "epoch": 3.5406107862248213,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037868075209329644,
      "loss": 5.3434,
      "mean_token_accuracy": 0.20404942333698273,
      "num_tokens": 69289132.0,
      "step": 35420
    },
    {
      "entropy": 5.945763349533081,
      "epoch": 3.54111061128605,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037864907663041686,
      "loss": 5.3745,
      "mean_token_accuracy": 0.20882668495178222,
      "num_tokens": 69299548.0,
      "step": 35425
    },
    {
      "entropy": 5.8602893352508545,
      "epoch": 3.5416104363472787,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037861739855975316,
      "loss": 5.3512,
      "mean_token_accuracy": 0.20308931022882462,
      "num_tokens": 69310326.0,
      "step": 35430
    },
    {
      "entropy": 5.874361419677735,
      "epoch": 3.542110261408507,
      "grad_norm": 1.125,
      "learning_rate": 0.00037858571788210236,
      "loss": 5.5207,
      "mean_token_accuracy": 0.1889079213142395,
      "num_tokens": 69320731.0,
      "step": 35435
    },
    {
      "entropy": 5.894815444946289,
      "epoch": 3.5426100864697356,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003785540345982616,
      "loss": 5.4092,
      "mean_token_accuracy": 0.19947570115327834,
      "num_tokens": 69330059.0,
      "step": 35440
    },
    {
      "entropy": 5.966956901550293,
      "epoch": 3.5431099115309643,
      "grad_norm": 1.21875,
      "learning_rate": 0.000378522348709028,
      "loss": 5.4951,
      "mean_token_accuracy": 0.19287003725767135,
      "num_tokens": 69340551.0,
      "step": 35445
    },
    {
      "entropy": 5.947148084640503,
      "epoch": 3.5436097365921926,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003784906602151987,
      "loss": 5.3778,
      "mean_token_accuracy": 0.19671888500452042,
      "num_tokens": 69350307.0,
      "step": 35450
    },
    {
      "entropy": 5.802168893814087,
      "epoch": 3.5441095616534213,
      "grad_norm": 1.265625,
      "learning_rate": 0.000378458969117571,
      "loss": 5.271,
      "mean_token_accuracy": 0.20834722816944123,
      "num_tokens": 69359807.0,
      "step": 35455
    },
    {
      "entropy": 5.885007858276367,
      "epoch": 3.54460938671465,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037842727541694226,
      "loss": 5.4313,
      "mean_token_accuracy": 0.19826071113348007,
      "num_tokens": 69369328.0,
      "step": 35460
    },
    {
      "entropy": 5.845205211639405,
      "epoch": 3.5451092117758787,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003783955791141099,
      "loss": 5.3405,
      "mean_token_accuracy": 0.2127739503979683,
      "num_tokens": 69378605.0,
      "step": 35465
    },
    {
      "entropy": 5.890023803710937,
      "epoch": 3.545609036837107,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003783638802098714,
      "loss": 5.3311,
      "mean_token_accuracy": 0.20596606135368348,
      "num_tokens": 69388710.0,
      "step": 35470
    },
    {
      "entropy": 5.877843523025513,
      "epoch": 3.5461088618983356,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003783321787050243,
      "loss": 5.3808,
      "mean_token_accuracy": 0.2100265845656395,
      "num_tokens": 69397654.0,
      "step": 35475
    },
    {
      "entropy": 5.920589590072632,
      "epoch": 3.546608686959564,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003783004746003661,
      "loss": 5.4682,
      "mean_token_accuracy": 0.19808320701122284,
      "num_tokens": 69407865.0,
      "step": 35480
    },
    {
      "entropy": 5.814848756790161,
      "epoch": 3.5471085120207926,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003782687678966946,
      "loss": 5.3597,
      "mean_token_accuracy": 0.1988101825118065,
      "num_tokens": 69417469.0,
      "step": 35485
    },
    {
      "entropy": 5.880124807357788,
      "epoch": 3.5476083370820213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003782370585948074,
      "loss": 5.3031,
      "mean_token_accuracy": 0.20970036834478378,
      "num_tokens": 69426866.0,
      "step": 35490
    },
    {
      "entropy": 5.924625968933105,
      "epoch": 3.54810816214325,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003782053466955024,
      "loss": 5.4237,
      "mean_token_accuracy": 0.1911933645606041,
      "num_tokens": 69436882.0,
      "step": 35495
    },
    {
      "entropy": 5.9012082576751705,
      "epoch": 3.5486079872044787,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037817363219957745,
      "loss": 5.3837,
      "mean_token_accuracy": 0.19707193225622177,
      "num_tokens": 69445667.0,
      "step": 35500
    },
    {
      "entropy": 5.813100719451905,
      "epoch": 3.549107812265707,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037814191510783047,
      "loss": 5.3787,
      "mean_token_accuracy": 0.20296307057142257,
      "num_tokens": 69454417.0,
      "step": 35505
    },
    {
      "entropy": 5.953492069244385,
      "epoch": 3.5496076373269356,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003781101954210594,
      "loss": 5.4093,
      "mean_token_accuracy": 0.1949283242225647,
      "num_tokens": 69462982.0,
      "step": 35510
    },
    {
      "entropy": 5.945683860778809,
      "epoch": 3.550107462388164,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037807847314006246,
      "loss": 5.4915,
      "mean_token_accuracy": 0.191854327917099,
      "num_tokens": 69472887.0,
      "step": 35515
    },
    {
      "entropy": 5.877233076095581,
      "epoch": 3.5506072874493926,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003780467482656375,
      "loss": 5.347,
      "mean_token_accuracy": 0.20161988735198974,
      "num_tokens": 69482520.0,
      "step": 35520
    },
    {
      "entropy": 5.884959936141968,
      "epoch": 3.5511071125106213,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037801502079858295,
      "loss": 5.3312,
      "mean_token_accuracy": 0.20639097392559053,
      "num_tokens": 69492215.0,
      "step": 35525
    },
    {
      "entropy": 5.7864391803741455,
      "epoch": 3.55160693757185,
      "grad_norm": 1.21875,
      "learning_rate": 0.000377983290739697,
      "loss": 5.3042,
      "mean_token_accuracy": 0.21045955121517182,
      "num_tokens": 69502940.0,
      "step": 35530
    },
    {
      "entropy": 5.84486198425293,
      "epoch": 3.5521067626330787,
      "grad_norm": 1.125,
      "learning_rate": 0.0003779515580897779,
      "loss": 5.3619,
      "mean_token_accuracy": 0.20923445671796798,
      "num_tokens": 69513656.0,
      "step": 35535
    },
    {
      "entropy": 5.938121604919433,
      "epoch": 3.552606587694307,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037791982284962416,
      "loss": 5.4199,
      "mean_token_accuracy": 0.20785346180200576,
      "num_tokens": 69523824.0,
      "step": 35540
    },
    {
      "entropy": 5.843462991714477,
      "epoch": 3.5531064127555356,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037788808502003414,
      "loss": 5.3755,
      "mean_token_accuracy": 0.20605510324239731,
      "num_tokens": 69533134.0,
      "step": 35545
    },
    {
      "entropy": 5.8482757091522215,
      "epoch": 3.553606237816764,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037785634460180633,
      "loss": 5.4147,
      "mean_token_accuracy": 0.20357780456542968,
      "num_tokens": 69542370.0,
      "step": 35550
    },
    {
      "entropy": 5.907258081436157,
      "epoch": 3.5541060628779926,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037782460159573944,
      "loss": 5.3766,
      "mean_token_accuracy": 0.20935239493846894,
      "num_tokens": 69552630.0,
      "step": 35555
    },
    {
      "entropy": 5.873898077011108,
      "epoch": 3.5546058879392213,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037779285600263195,
      "loss": 5.3765,
      "mean_token_accuracy": 0.1972415864467621,
      "num_tokens": 69563629.0,
      "step": 35560
    },
    {
      "entropy": 5.879151058197022,
      "epoch": 3.55510571300045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003777611078232827,
      "loss": 5.408,
      "mean_token_accuracy": 0.20045452266931535,
      "num_tokens": 69573158.0,
      "step": 35565
    },
    {
      "entropy": 5.955469751358033,
      "epoch": 3.5556055380616787,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037772935705849046,
      "loss": 5.4262,
      "mean_token_accuracy": 0.19185985624790192,
      "num_tokens": 69583673.0,
      "step": 35570
    },
    {
      "entropy": 5.879425859451294,
      "epoch": 3.556105363122907,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003776976037090539,
      "loss": 5.3755,
      "mean_token_accuracy": 0.20083956867456437,
      "num_tokens": 69593339.0,
      "step": 35575
    },
    {
      "entropy": 5.81834774017334,
      "epoch": 3.5566051881841356,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037766584777577223,
      "loss": 5.3587,
      "mean_token_accuracy": 0.20198454707860947,
      "num_tokens": 69603709.0,
      "step": 35580
    },
    {
      "entropy": 5.917544889450073,
      "epoch": 3.557105013245364,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003776340892594441,
      "loss": 5.5176,
      "mean_token_accuracy": 0.19373028874397277,
      "num_tokens": 69614060.0,
      "step": 35585
    },
    {
      "entropy": 5.918388748168946,
      "epoch": 3.5576048383065926,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037760232816086875,
      "loss": 5.4305,
      "mean_token_accuracy": 0.2027291774749756,
      "num_tokens": 69623485.0,
      "step": 35590
    },
    {
      "entropy": 5.811666440963745,
      "epoch": 3.5581046633678213,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003775705644808452,
      "loss": 5.3204,
      "mean_token_accuracy": 0.20665358155965804,
      "num_tokens": 69633098.0,
      "step": 35595
    },
    {
      "entropy": 5.770821380615234,
      "epoch": 3.55860448842905,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037753879822017265,
      "loss": 5.2297,
      "mean_token_accuracy": 0.21747278571128845,
      "num_tokens": 69642730.0,
      "step": 35600
    },
    {
      "entropy": 5.865176343917847,
      "epoch": 3.5591043134902787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003775070293796504,
      "loss": 5.36,
      "mean_token_accuracy": 0.21034565716981887,
      "num_tokens": 69652004.0,
      "step": 35605
    },
    {
      "entropy": 5.853975677490235,
      "epoch": 3.559604138551507,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003774752579600776,
      "loss": 5.2997,
      "mean_token_accuracy": 0.20815333276987075,
      "num_tokens": 69661140.0,
      "step": 35610
    },
    {
      "entropy": 5.834923458099365,
      "epoch": 3.5601039636127356,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003774434839622537,
      "loss": 5.3301,
      "mean_token_accuracy": 0.20918819010257722,
      "num_tokens": 69671704.0,
      "step": 35615
    },
    {
      "entropy": 5.883918571472168,
      "epoch": 3.560603788673964,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037741170738697806,
      "loss": 5.3923,
      "mean_token_accuracy": 0.20431397706270218,
      "num_tokens": 69680312.0,
      "step": 35620
    },
    {
      "entropy": 5.891081285476685,
      "epoch": 3.5611036137351926,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003773799282350503,
      "loss": 5.3358,
      "mean_token_accuracy": 0.21383898258209227,
      "num_tokens": 69690515.0,
      "step": 35625
    },
    {
      "entropy": 5.829646492004395,
      "epoch": 3.5616034387964213,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003773481465072698,
      "loss": 5.3798,
      "mean_token_accuracy": 0.20436352789402007,
      "num_tokens": 69700268.0,
      "step": 35630
    },
    {
      "entropy": 5.9106546401977536,
      "epoch": 3.56210326385765,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037731636220443625,
      "loss": 5.5114,
      "mean_token_accuracy": 0.1922041341662407,
      "num_tokens": 69709349.0,
      "step": 35635
    },
    {
      "entropy": 5.906436014175415,
      "epoch": 3.5626030889188787,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037728457532734943,
      "loss": 5.4247,
      "mean_token_accuracy": 0.19789346307516098,
      "num_tokens": 69719861.0,
      "step": 35640
    },
    {
      "entropy": 5.959214305877685,
      "epoch": 3.563102913980107,
      "grad_norm": 1.25,
      "learning_rate": 0.000377252785876809,
      "loss": 5.4398,
      "mean_token_accuracy": 0.1952653631567955,
      "num_tokens": 69729290.0,
      "step": 35645
    },
    {
      "entropy": 5.914975214004516,
      "epoch": 3.5636027390413356,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037722099385361475,
      "loss": 5.4579,
      "mean_token_accuracy": 0.19624805301427842,
      "num_tokens": 69738056.0,
      "step": 35650
    },
    {
      "entropy": 5.966221666336059,
      "epoch": 3.564102564102564,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003771891992585667,
      "loss": 5.463,
      "mean_token_accuracy": 0.195696721971035,
      "num_tokens": 69747483.0,
      "step": 35655
    },
    {
      "entropy": 5.846115970611573,
      "epoch": 3.5646023891637926,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003771574020924646,
      "loss": 5.3525,
      "mean_token_accuracy": 0.2024013876914978,
      "num_tokens": 69758506.0,
      "step": 35660
    },
    {
      "entropy": 5.829296970367432,
      "epoch": 3.5651022142250213,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003771256023561086,
      "loss": 5.3336,
      "mean_token_accuracy": 0.21033655256032943,
      "num_tokens": 69768576.0,
      "step": 35665
    },
    {
      "entropy": 5.805550813674927,
      "epoch": 3.56560203928625,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003770938000502987,
      "loss": 5.3138,
      "mean_token_accuracy": 0.22005003690719604,
      "num_tokens": 69779122.0,
      "step": 35670
    },
    {
      "entropy": 5.836616468429566,
      "epoch": 3.5661018643474782,
      "grad_norm": 1.09375,
      "learning_rate": 0.000377061995175835,
      "loss": 5.311,
      "mean_token_accuracy": 0.2057464078068733,
      "num_tokens": 69789521.0,
      "step": 35675
    },
    {
      "entropy": 5.931778812408448,
      "epoch": 3.566601689408707,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003770301877335178,
      "loss": 5.5023,
      "mean_token_accuracy": 0.1944860488176346,
      "num_tokens": 69798564.0,
      "step": 35680
    },
    {
      "entropy": 5.872090148925781,
      "epoch": 3.5671015144699356,
      "grad_norm": 1.125,
      "learning_rate": 0.00037699837772414734,
      "loss": 5.3952,
      "mean_token_accuracy": 0.20276089608669282,
      "num_tokens": 69807642.0,
      "step": 35685
    },
    {
      "entropy": 5.827940130233765,
      "epoch": 3.567601339531164,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000376966565148524,
      "loss": 5.3976,
      "mean_token_accuracy": 0.20155187845230102,
      "num_tokens": 69818193.0,
      "step": 35690
    },
    {
      "entropy": 5.957048749923706,
      "epoch": 3.5681011645923926,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037693475000744807,
      "loss": 5.4881,
      "mean_token_accuracy": 0.1900451347231865,
      "num_tokens": 69827991.0,
      "step": 35695
    },
    {
      "entropy": 5.9364790439605715,
      "epoch": 3.5686009896536213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037690293230172007,
      "loss": 5.4376,
      "mean_token_accuracy": 0.19650236368179322,
      "num_tokens": 69838156.0,
      "step": 35700
    },
    {
      "entropy": 5.907529687881469,
      "epoch": 3.56910081471485,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037687111203214054,
      "loss": 5.4257,
      "mean_token_accuracy": 0.19827245473861693,
      "num_tokens": 69847711.0,
      "step": 35705
    },
    {
      "entropy": 5.873713254928589,
      "epoch": 3.569600639776078,
      "grad_norm": 1.25,
      "learning_rate": 0.00037683928919951006,
      "loss": 5.3565,
      "mean_token_accuracy": 0.20153820663690566,
      "num_tokens": 69857800.0,
      "step": 35710
    },
    {
      "entropy": 5.885822534561157,
      "epoch": 3.570100464837307,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037680746380462914,
      "loss": 5.4492,
      "mean_token_accuracy": 0.19713720828294753,
      "num_tokens": 69867436.0,
      "step": 35715
    },
    {
      "entropy": 5.906182527542114,
      "epoch": 3.5706002898985356,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003767756358482987,
      "loss": 5.4373,
      "mean_token_accuracy": 0.19589049071073533,
      "num_tokens": 69877808.0,
      "step": 35720
    },
    {
      "entropy": 5.835740995407105,
      "epoch": 3.571100114959764,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003767438053313195,
      "loss": 5.3017,
      "mean_token_accuracy": 0.21343932449817657,
      "num_tokens": 69888565.0,
      "step": 35725
    },
    {
      "entropy": 5.963356161117554,
      "epoch": 3.5715999400209926,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003767119722544923,
      "loss": 5.5659,
      "mean_token_accuracy": 0.19091629534959792,
      "num_tokens": 69898863.0,
      "step": 35730
    },
    {
      "entropy": 5.984723520278931,
      "epoch": 3.5720997650822213,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003766801366186181,
      "loss": 5.444,
      "mean_token_accuracy": 0.1933204561471939,
      "num_tokens": 69908828.0,
      "step": 35735
    },
    {
      "entropy": 5.855535840988159,
      "epoch": 3.57259959014345,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003766482984244978,
      "loss": 5.39,
      "mean_token_accuracy": 0.2029706910252571,
      "num_tokens": 69918276.0,
      "step": 35740
    },
    {
      "entropy": 5.829322481155396,
      "epoch": 3.573099415204678,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003766164576729324,
      "loss": 5.2959,
      "mean_token_accuracy": 0.2076544776558876,
      "num_tokens": 69927881.0,
      "step": 35745
    },
    {
      "entropy": 5.879998254776001,
      "epoch": 3.573599240265907,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037658461436472306,
      "loss": 5.399,
      "mean_token_accuracy": 0.21311702579259872,
      "num_tokens": 69937287.0,
      "step": 35750
    },
    {
      "entropy": 5.760423994064331,
      "epoch": 3.5740990653271356,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037655276850067103,
      "loss": 5.2544,
      "mean_token_accuracy": 0.2147312119603157,
      "num_tokens": 69946927.0,
      "step": 35755
    },
    {
      "entropy": 5.883798551559448,
      "epoch": 3.574598890388364,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003765209200815776,
      "loss": 5.3681,
      "mean_token_accuracy": 0.20398146361112596,
      "num_tokens": 69956455.0,
      "step": 35760
    },
    {
      "entropy": 5.84657998085022,
      "epoch": 3.5750987154495926,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037648906910824374,
      "loss": 5.2776,
      "mean_token_accuracy": 0.21190240383148193,
      "num_tokens": 69965918.0,
      "step": 35765
    },
    {
      "entropy": 5.871815490722656,
      "epoch": 3.5755985405108213,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037645721558147104,
      "loss": 5.3672,
      "mean_token_accuracy": 0.20337892025709153,
      "num_tokens": 69975591.0,
      "step": 35770
    },
    {
      "entropy": 5.853679656982422,
      "epoch": 3.57609836557205,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000376425359502061,
      "loss": 5.3432,
      "mean_token_accuracy": 0.21249231100082397,
      "num_tokens": 69986306.0,
      "step": 35775
    },
    {
      "entropy": 5.874852800369263,
      "epoch": 3.576598190633278,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037639350087081497,
      "loss": 5.3831,
      "mean_token_accuracy": 0.20392414927482605,
      "num_tokens": 69995762.0,
      "step": 35780
    },
    {
      "entropy": 5.9293303966522215,
      "epoch": 3.577098015694507,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037636163968853453,
      "loss": 5.3444,
      "mean_token_accuracy": 0.20530617237091064,
      "num_tokens": 70004713.0,
      "step": 35785
    },
    {
      "entropy": 5.864502477645874,
      "epoch": 3.5775978407557356,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003763297759560213,
      "loss": 5.3882,
      "mean_token_accuracy": 0.20733439177274704,
      "num_tokens": 70015074.0,
      "step": 35790
    },
    {
      "entropy": 5.907617378234863,
      "epoch": 3.578097665816964,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003762979096740771,
      "loss": 5.4568,
      "mean_token_accuracy": 0.1990286737680435,
      "num_tokens": 70026400.0,
      "step": 35795
    },
    {
      "entropy": 5.918825435638428,
      "epoch": 3.5785974908781926,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003762660408435035,
      "loss": 5.4178,
      "mean_token_accuracy": 0.19885346442461013,
      "num_tokens": 70036069.0,
      "step": 35800
    },
    {
      "entropy": 5.916077375411987,
      "epoch": 3.5790973159394213,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003762341694651023,
      "loss": 5.4618,
      "mean_token_accuracy": 0.19382222443819047,
      "num_tokens": 70046595.0,
      "step": 35805
    },
    {
      "entropy": 5.864661359786988,
      "epoch": 3.57959714100065,
      "grad_norm": 1.125,
      "learning_rate": 0.00037620229553967543,
      "loss": 5.424,
      "mean_token_accuracy": 0.20007328540086747,
      "num_tokens": 70057074.0,
      "step": 35810
    },
    {
      "entropy": 5.8091991424560545,
      "epoch": 3.580096966061878,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037617041906802496,
      "loss": 5.3535,
      "mean_token_accuracy": 0.19990360885858535,
      "num_tokens": 70067300.0,
      "step": 35815
    },
    {
      "entropy": 5.891892528533935,
      "epoch": 3.580596791123107,
      "grad_norm": 1.125,
      "learning_rate": 0.0003761385400509527,
      "loss": 5.441,
      "mean_token_accuracy": 0.19642394185066223,
      "num_tokens": 70077482.0,
      "step": 35820
    },
    {
      "entropy": 5.912742042541504,
      "epoch": 3.5810966161843356,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037610665848926086,
      "loss": 5.4007,
      "mean_token_accuracy": 0.2011036232113838,
      "num_tokens": 70087189.0,
      "step": 35825
    },
    {
      "entropy": 5.878250217437744,
      "epoch": 3.581596441245564,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003760747743837515,
      "loss": 5.3303,
      "mean_token_accuracy": 0.20799527913331986,
      "num_tokens": 70095920.0,
      "step": 35830
    },
    {
      "entropy": 5.824991273880005,
      "epoch": 3.5820962663067926,
      "grad_norm": 1.125,
      "learning_rate": 0.00037604288773522673,
      "loss": 5.3227,
      "mean_token_accuracy": 0.2051588699221611,
      "num_tokens": 70104815.0,
      "step": 35835
    },
    {
      "entropy": 5.873688220977783,
      "epoch": 3.5825960913680213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037601099854448893,
      "loss": 5.3807,
      "mean_token_accuracy": 0.20310917794704436,
      "num_tokens": 70114847.0,
      "step": 35840
    },
    {
      "entropy": 5.890455627441407,
      "epoch": 3.58309591642925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037597910681234047,
      "loss": 5.3637,
      "mean_token_accuracy": 0.2067825138568878,
      "num_tokens": 70125380.0,
      "step": 35845
    },
    {
      "entropy": 5.9219707489013675,
      "epoch": 3.583595741490478,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003759472125395837,
      "loss": 5.3787,
      "mean_token_accuracy": 0.20276925563812256,
      "num_tokens": 70134414.0,
      "step": 35850
    },
    {
      "entropy": 5.880507516860962,
      "epoch": 3.584095566551707,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003759153157270209,
      "loss": 5.4947,
      "mean_token_accuracy": 0.1921965941786766,
      "num_tokens": 70144045.0,
      "step": 35855
    },
    {
      "entropy": 5.789725017547608,
      "epoch": 3.5845953916129356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037588341637545487,
      "loss": 5.2897,
      "mean_token_accuracy": 0.21565781235694886,
      "num_tokens": 70155056.0,
      "step": 35860
    },
    {
      "entropy": 5.917118310928345,
      "epoch": 3.585095216674164,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037585151448568796,
      "loss": 5.4041,
      "mean_token_accuracy": 0.19655140191316606,
      "num_tokens": 70163898.0,
      "step": 35865
    },
    {
      "entropy": 5.8986602306365965,
      "epoch": 3.5855950417353926,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037581961005852296,
      "loss": 5.4289,
      "mean_token_accuracy": 0.19674596637487413,
      "num_tokens": 70173836.0,
      "step": 35870
    },
    {
      "entropy": 5.9390668869018555,
      "epoch": 3.5860948667966213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003757877030947624,
      "loss": 5.5597,
      "mean_token_accuracy": 0.18792446851730346,
      "num_tokens": 70184594.0,
      "step": 35875
    },
    {
      "entropy": 5.9474955081939695,
      "epoch": 3.58659469185785,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003757557935952092,
      "loss": 5.4404,
      "mean_token_accuracy": 0.19328124076128006,
      "num_tokens": 70194262.0,
      "step": 35880
    },
    {
      "entropy": 5.894625854492188,
      "epoch": 3.587094516919078,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037572388156066627,
      "loss": 5.4191,
      "mean_token_accuracy": 0.20345583111047744,
      "num_tokens": 70204247.0,
      "step": 35885
    },
    {
      "entropy": 5.911975288391114,
      "epoch": 3.587594341980307,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037569196699193633,
      "loss": 5.3727,
      "mean_token_accuracy": 0.19714045375585557,
      "num_tokens": 70214236.0,
      "step": 35890
    },
    {
      "entropy": 5.87080225944519,
      "epoch": 3.5880941670415356,
      "grad_norm": 1.125,
      "learning_rate": 0.00037566004988982235,
      "loss": 5.2838,
      "mean_token_accuracy": 0.20077928751707078,
      "num_tokens": 70223945.0,
      "step": 35895
    },
    {
      "entropy": 5.852820539474488,
      "epoch": 3.588593992102764,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003756281302551274,
      "loss": 5.3567,
      "mean_token_accuracy": 0.2091563805937767,
      "num_tokens": 70233011.0,
      "step": 35900
    },
    {
      "entropy": 5.836736583709717,
      "epoch": 3.5890938171639926,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037559620808865467,
      "loss": 5.3449,
      "mean_token_accuracy": 0.20655978173017503,
      "num_tokens": 70241970.0,
      "step": 35905
    },
    {
      "entropy": 5.84841251373291,
      "epoch": 3.5895936422252213,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037556428339120725,
      "loss": 5.3976,
      "mean_token_accuracy": 0.20391322523355485,
      "num_tokens": 70251907.0,
      "step": 35910
    },
    {
      "entropy": 5.771524333953858,
      "epoch": 3.59009346728645,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037553235616358835,
      "loss": 5.272,
      "mean_token_accuracy": 0.21626127809286116,
      "num_tokens": 70261401.0,
      "step": 35915
    },
    {
      "entropy": 5.870735120773316,
      "epoch": 3.590593292347678,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037550042640660113,
      "loss": 5.3674,
      "mean_token_accuracy": 0.199760277569294,
      "num_tokens": 70272075.0,
      "step": 35920
    },
    {
      "entropy": 5.909750747680664,
      "epoch": 3.591093117408907,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003754684941210491,
      "loss": 5.3439,
      "mean_token_accuracy": 0.19973180592060089,
      "num_tokens": 70282189.0,
      "step": 35925
    },
    {
      "entropy": 5.884809398651123,
      "epoch": 3.5915929424701356,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037543655930773564,
      "loss": 5.3986,
      "mean_token_accuracy": 0.19211284220218658,
      "num_tokens": 70292219.0,
      "step": 35930
    },
    {
      "entropy": 5.8413918018341064,
      "epoch": 3.592092767531364,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003754046219674641,
      "loss": 5.3563,
      "mean_token_accuracy": 0.20116486698389052,
      "num_tokens": 70301679.0,
      "step": 35935
    },
    {
      "entropy": 5.908225727081299,
      "epoch": 3.5925925925925926,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037537268210103826,
      "loss": 5.3582,
      "mean_token_accuracy": 0.19844117164611816,
      "num_tokens": 70310704.0,
      "step": 35940
    },
    {
      "entropy": 5.842766380310058,
      "epoch": 3.5930924176538213,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003753407397092614,
      "loss": 5.3468,
      "mean_token_accuracy": 0.19882437586784363,
      "num_tokens": 70319510.0,
      "step": 35945
    },
    {
      "entropy": 5.873372650146484,
      "epoch": 3.59359224271505,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003753087947929375,
      "loss": 5.4149,
      "mean_token_accuracy": 0.19620818495750428,
      "num_tokens": 70330307.0,
      "step": 35950
    },
    {
      "entropy": 5.905773305892945,
      "epoch": 3.594092067776278,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037527684735287007,
      "loss": 5.4557,
      "mean_token_accuracy": 0.19416215121746064,
      "num_tokens": 70340789.0,
      "step": 35955
    },
    {
      "entropy": 5.823000288009643,
      "epoch": 3.594591892837507,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003752448973898629,
      "loss": 5.2398,
      "mean_token_accuracy": 0.21918171793222427,
      "num_tokens": 70350984.0,
      "step": 35960
    },
    {
      "entropy": 5.878344011306763,
      "epoch": 3.5950917178987356,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037521294490472,
      "loss": 5.4345,
      "mean_token_accuracy": 0.1985243245959282,
      "num_tokens": 70360370.0,
      "step": 35965
    },
    {
      "entropy": 5.894321298599243,
      "epoch": 3.595591542959964,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003751809898982451,
      "loss": 5.3885,
      "mean_token_accuracy": 0.20055872797966004,
      "num_tokens": 70371138.0,
      "step": 35970
    },
    {
      "entropy": 5.893198823928833,
      "epoch": 3.5960913680211926,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003751490323712424,
      "loss": 5.4505,
      "mean_token_accuracy": 0.1932913690805435,
      "num_tokens": 70380823.0,
      "step": 35975
    },
    {
      "entropy": 5.89514856338501,
      "epoch": 3.5965911930824213,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003751170723245157,
      "loss": 5.3026,
      "mean_token_accuracy": 0.20681910067796708,
      "num_tokens": 70390979.0,
      "step": 35980
    },
    {
      "entropy": 5.847382497787476,
      "epoch": 3.59709101814365,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037508510975886925,
      "loss": 5.38,
      "mean_token_accuracy": 0.20531555116176606,
      "num_tokens": 70402738.0,
      "step": 35985
    },
    {
      "entropy": 5.895318078994751,
      "epoch": 3.597590843204878,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037505314467510726,
      "loss": 5.3739,
      "mean_token_accuracy": 0.20228261798620223,
      "num_tokens": 70411291.0,
      "step": 35990
    },
    {
      "entropy": 5.908868217468262,
      "epoch": 3.598090668266107,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003750211770740338,
      "loss": 5.4258,
      "mean_token_accuracy": 0.19816555678844452,
      "num_tokens": 70421182.0,
      "step": 35995
    },
    {
      "entropy": 5.7738879203796385,
      "epoch": 3.598590493327335,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003749892069564533,
      "loss": 5.2282,
      "mean_token_accuracy": 0.2190050721168518,
      "num_tokens": 70430603.0,
      "step": 36000
    },
    {
      "epoch": 3.598590493327335,
      "eval_entropy": 5.6587016790744356,
      "eval_loss": 5.509439945220947,
      "eval_mean_token_accuracy": 0.20547883860868635,
      "eval_num_tokens": 70430603.0,
      "eval_runtime": 23.761,
      "eval_samples_per_second": 1306.552,
      "eval_steps_per_second": 163.335,
      "step": 36000
    },
    {
      "entropy": 5.952779150009155,
      "epoch": 3.599090318388564,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003749572343231701,
      "loss": 5.4378,
      "mean_token_accuracy": 0.19602813124656676,
      "num_tokens": 70439727.0,
      "step": 36005
    },
    {
      "entropy": 5.901870059967041,
      "epoch": 3.5995901434497926,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003749252591749887,
      "loss": 5.4079,
      "mean_token_accuracy": 0.2034597545862198,
      "num_tokens": 70448316.0,
      "step": 36010
    },
    {
      "entropy": 5.846813774108886,
      "epoch": 3.6000899685110213,
      "grad_norm": 1.25,
      "learning_rate": 0.0003748932815127133,
      "loss": 5.3849,
      "mean_token_accuracy": 0.20010797381401063,
      "num_tokens": 70458384.0,
      "step": 36015
    },
    {
      "entropy": 5.8690472602844235,
      "epoch": 3.60058979357225,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037486130133714874,
      "loss": 5.3348,
      "mean_token_accuracy": 0.20365416556596755,
      "num_tokens": 70468642.0,
      "step": 36020
    },
    {
      "entropy": 5.989486742019653,
      "epoch": 3.601089618633478,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003748293186490995,
      "loss": 5.4225,
      "mean_token_accuracy": 0.2011710450053215,
      "num_tokens": 70477310.0,
      "step": 36025
    },
    {
      "entropy": 5.97604546546936,
      "epoch": 3.601589443694707,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003747973334493703,
      "loss": 5.4769,
      "mean_token_accuracy": 0.1969975098967552,
      "num_tokens": 70487288.0,
      "step": 36030
    },
    {
      "entropy": 5.9087474822998045,
      "epoch": 3.602089268755935,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003747653457387659,
      "loss": 5.4709,
      "mean_token_accuracy": 0.198569156229496,
      "num_tokens": 70497467.0,
      "step": 36035
    },
    {
      "entropy": 5.888334083557129,
      "epoch": 3.602589093817164,
      "grad_norm": 1.125,
      "learning_rate": 0.00037473335551809107,
      "loss": 5.2997,
      "mean_token_accuracy": 0.20282535403966903,
      "num_tokens": 70506146.0,
      "step": 36040
    },
    {
      "entropy": 5.900404977798462,
      "epoch": 3.6030889188783926,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037470136278815067,
      "loss": 5.3579,
      "mean_token_accuracy": 0.20827382653951645,
      "num_tokens": 70515346.0,
      "step": 36045
    },
    {
      "entropy": 5.886799478530884,
      "epoch": 3.6035887439396213,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037466936754974964,
      "loss": 5.4831,
      "mean_token_accuracy": 0.2009192258119583,
      "num_tokens": 70525894.0,
      "step": 36050
    },
    {
      "entropy": 5.863940238952637,
      "epoch": 3.60408856900085,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037463736980369297,
      "loss": 5.3208,
      "mean_token_accuracy": 0.20103347301483154,
      "num_tokens": 70535265.0,
      "step": 36055
    },
    {
      "entropy": 5.93837456703186,
      "epoch": 3.604588394062078,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003746053695507856,
      "loss": 5.4498,
      "mean_token_accuracy": 0.19680939763784408,
      "num_tokens": 70544412.0,
      "step": 36060
    },
    {
      "entropy": 5.831067085266113,
      "epoch": 3.605088219123307,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003745733667918329,
      "loss": 5.4101,
      "mean_token_accuracy": 0.19935128837823868,
      "num_tokens": 70555628.0,
      "step": 36065
    },
    {
      "entropy": 5.846049976348877,
      "epoch": 3.605588044184535,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003745413615276399,
      "loss": 5.4515,
      "mean_token_accuracy": 0.20107580870389938,
      "num_tokens": 70564748.0,
      "step": 36070
    },
    {
      "entropy": 5.746738910675049,
      "epoch": 3.606087869245764,
      "grad_norm": 1.125,
      "learning_rate": 0.00037450935375901186,
      "loss": 5.2298,
      "mean_token_accuracy": 0.22191453725099564,
      "num_tokens": 70574576.0,
      "step": 36075
    },
    {
      "entropy": 5.910451126098633,
      "epoch": 3.6065876943069926,
      "grad_norm": 1.28125,
      "learning_rate": 0.000374477343486754,
      "loss": 5.3708,
      "mean_token_accuracy": 0.20480928122997283,
      "num_tokens": 70583562.0,
      "step": 36080
    },
    {
      "entropy": 5.8664830207824705,
      "epoch": 3.6070875193682213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003744453307116719,
      "loss": 5.3908,
      "mean_token_accuracy": 0.1930817186832428,
      "num_tokens": 70594070.0,
      "step": 36085
    },
    {
      "entropy": 5.793465232849121,
      "epoch": 3.60758734442945,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037441331543457075,
      "loss": 5.3253,
      "mean_token_accuracy": 0.20558823943138121,
      "num_tokens": 70603149.0,
      "step": 36090
    },
    {
      "entropy": 5.95821361541748,
      "epoch": 3.608087169490678,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037438129765625627,
      "loss": 5.4449,
      "mean_token_accuracy": 0.1938792273402214,
      "num_tokens": 70613551.0,
      "step": 36095
    },
    {
      "entropy": 5.914823818206787,
      "epoch": 3.608586994551907,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037434927737753387,
      "loss": 5.3343,
      "mean_token_accuracy": 0.20365633368492125,
      "num_tokens": 70623606.0,
      "step": 36100
    },
    {
      "entropy": 5.8717104434967045,
      "epoch": 3.609086819613135,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037431725459920927,
      "loss": 5.4299,
      "mean_token_accuracy": 0.19162000417709352,
      "num_tokens": 70633105.0,
      "step": 36105
    },
    {
      "entropy": 5.857443618774414,
      "epoch": 3.609586644674364,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037428522932208804,
      "loss": 5.3371,
      "mean_token_accuracy": 0.20318697839975358,
      "num_tokens": 70643663.0,
      "step": 36110
    },
    {
      "entropy": 5.884540176391601,
      "epoch": 3.6100864697355926,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037425320154697595,
      "loss": 5.3288,
      "mean_token_accuracy": 0.20062772780656815,
      "num_tokens": 70653858.0,
      "step": 36115
    },
    {
      "entropy": 5.908670282363891,
      "epoch": 3.6105862947968212,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037422117127467884,
      "loss": 5.347,
      "mean_token_accuracy": 0.207684662938118,
      "num_tokens": 70663071.0,
      "step": 36120
    },
    {
      "entropy": 5.955484390258789,
      "epoch": 3.61108611985805,
      "grad_norm": 1.125,
      "learning_rate": 0.0003741891385060027,
      "loss": 5.4987,
      "mean_token_accuracy": 0.19255291819572448,
      "num_tokens": 70673678.0,
      "step": 36125
    },
    {
      "entropy": 5.907889652252197,
      "epoch": 3.611585944919278,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003741571032417533,
      "loss": 5.4255,
      "mean_token_accuracy": 0.19841596633195877,
      "num_tokens": 70683258.0,
      "step": 36130
    },
    {
      "entropy": 5.905821228027344,
      "epoch": 3.612085769980507,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003741250654827367,
      "loss": 5.4238,
      "mean_token_accuracy": 0.19848148375749589,
      "num_tokens": 70692773.0,
      "step": 36135
    },
    {
      "entropy": 5.925825214385986,
      "epoch": 3.612585595041735,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037409302522975897,
      "loss": 5.4205,
      "mean_token_accuracy": 0.20165375620126724,
      "num_tokens": 70703390.0,
      "step": 36140
    },
    {
      "entropy": 5.8962126731872555,
      "epoch": 3.613085420102964,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037406098248362625,
      "loss": 5.3684,
      "mean_token_accuracy": 0.2042121857404709,
      "num_tokens": 70712313.0,
      "step": 36145
    },
    {
      "entropy": 5.929349660873413,
      "epoch": 3.6135852451641925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037402893724514466,
      "loss": 5.4232,
      "mean_token_accuracy": 0.19461309611797334,
      "num_tokens": 70722451.0,
      "step": 36150
    },
    {
      "entropy": 5.8717714786529545,
      "epoch": 3.6140850702254212,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037399688951512044,
      "loss": 5.3877,
      "mean_token_accuracy": 0.19798318296670914,
      "num_tokens": 70731977.0,
      "step": 36155
    },
    {
      "entropy": 5.845491170883179,
      "epoch": 3.61458489528665,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003739648392943601,
      "loss": 5.2979,
      "mean_token_accuracy": 0.20796819478273393,
      "num_tokens": 70740846.0,
      "step": 36160
    },
    {
      "entropy": 5.956795310974121,
      "epoch": 3.615084720347878,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003739327865836697,
      "loss": 5.4486,
      "mean_token_accuracy": 0.1961773931980133,
      "num_tokens": 70750916.0,
      "step": 36165
    },
    {
      "entropy": 5.907831764221191,
      "epoch": 3.615584545409107,
      "grad_norm": 1.125,
      "learning_rate": 0.00037390073138385597,
      "loss": 5.4645,
      "mean_token_accuracy": 0.2006967693567276,
      "num_tokens": 70761229.0,
      "step": 36170
    },
    {
      "entropy": 5.789677286148072,
      "epoch": 3.616084370470335,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003738686736957253,
      "loss": 5.2739,
      "mean_token_accuracy": 0.206842839717865,
      "num_tokens": 70771249.0,
      "step": 36175
    },
    {
      "entropy": 5.8664045333862305,
      "epoch": 3.616584195531564,
      "grad_norm": 1.125,
      "learning_rate": 0.00037383661352008413,
      "loss": 5.3815,
      "mean_token_accuracy": 0.19736194461584092,
      "num_tokens": 70781120.0,
      "step": 36180
    },
    {
      "entropy": 5.872553205490112,
      "epoch": 3.6170840205927925,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037380455085773924,
      "loss": 5.3775,
      "mean_token_accuracy": 0.2044664740562439,
      "num_tokens": 70790463.0,
      "step": 36185
    },
    {
      "entropy": 5.954873514175415,
      "epoch": 3.6175838456540212,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037377248570949733,
      "loss": 5.4343,
      "mean_token_accuracy": 0.19137333929538727,
      "num_tokens": 70800598.0,
      "step": 36190
    },
    {
      "entropy": 5.8857869625091555,
      "epoch": 3.6180836707152495,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037374041807616506,
      "loss": 5.4037,
      "mean_token_accuracy": 0.20032188594341277,
      "num_tokens": 70809876.0,
      "step": 36195
    },
    {
      "entropy": 5.834830713272095,
      "epoch": 3.618583495776478,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037370834795854923,
      "loss": 5.3238,
      "mean_token_accuracy": 0.20986930429935455,
      "num_tokens": 70819444.0,
      "step": 36200
    },
    {
      "entropy": 5.894065952301025,
      "epoch": 3.619083320837707,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003736762753574568,
      "loss": 5.2879,
      "mean_token_accuracy": 0.2045916810631752,
      "num_tokens": 70829269.0,
      "step": 36205
    },
    {
      "entropy": 5.812096071243286,
      "epoch": 3.619583145898935,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037364420027369467,
      "loss": 5.3258,
      "mean_token_accuracy": 0.1963637113571167,
      "num_tokens": 70838869.0,
      "step": 36210
    },
    {
      "entropy": 5.805634880065918,
      "epoch": 3.620082970960164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003736121227080699,
      "loss": 5.2898,
      "mean_token_accuracy": 0.20338406413793564,
      "num_tokens": 70847789.0,
      "step": 36215
    },
    {
      "entropy": 5.833074617385864,
      "epoch": 3.6205827960213925,
      "grad_norm": 1.125,
      "learning_rate": 0.00037358004266138943,
      "loss": 5.3159,
      "mean_token_accuracy": 0.20766815096139907,
      "num_tokens": 70857928.0,
      "step": 36220
    },
    {
      "entropy": 5.883222246170044,
      "epoch": 3.6210826210826212,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037354796013446044,
      "loss": 5.334,
      "mean_token_accuracy": 0.20506392568349838,
      "num_tokens": 70866585.0,
      "step": 36225
    },
    {
      "entropy": 5.8689428806304935,
      "epoch": 3.6215824461438495,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003735158751280902,
      "loss": 5.3836,
      "mean_token_accuracy": 0.20203027427196502,
      "num_tokens": 70876760.0,
      "step": 36230
    },
    {
      "entropy": 5.915438175201416,
      "epoch": 3.622082271205078,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003734837876430859,
      "loss": 5.505,
      "mean_token_accuracy": 0.19032506942749022,
      "num_tokens": 70886405.0,
      "step": 36235
    },
    {
      "entropy": 5.859809350967407,
      "epoch": 3.622582096266307,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003734516976802548,
      "loss": 5.3383,
      "mean_token_accuracy": 0.20249305963516234,
      "num_tokens": 70896231.0,
      "step": 36240
    },
    {
      "entropy": 5.812260484695434,
      "epoch": 3.623081921327535,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003734196052404043,
      "loss": 5.3859,
      "mean_token_accuracy": 0.19992076456546784,
      "num_tokens": 70906878.0,
      "step": 36245
    },
    {
      "entropy": 5.94863715171814,
      "epoch": 3.623581746388764,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003733875103243419,
      "loss": 5.4285,
      "mean_token_accuracy": 0.20260705649852753,
      "num_tokens": 70917508.0,
      "step": 36250
    },
    {
      "entropy": 5.801815938949585,
      "epoch": 3.6240815714499925,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003733554129328751,
      "loss": 5.2896,
      "mean_token_accuracy": 0.21111200302839278,
      "num_tokens": 70928900.0,
      "step": 36255
    },
    {
      "entropy": 5.859568738937378,
      "epoch": 3.6245813965112212,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037332331306681136,
      "loss": 5.3227,
      "mean_token_accuracy": 0.19967123866081238,
      "num_tokens": 70938895.0,
      "step": 36260
    },
    {
      "entropy": 5.734337615966797,
      "epoch": 3.6250812215724495,
      "grad_norm": 1.125,
      "learning_rate": 0.00037329121072695833,
      "loss": 5.1714,
      "mean_token_accuracy": 0.21575848162174224,
      "num_tokens": 70949323.0,
      "step": 36265
    },
    {
      "entropy": 5.756920289993286,
      "epoch": 3.625581046633678,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037325910591412374,
      "loss": 5.2656,
      "mean_token_accuracy": 0.21375263184309007,
      "num_tokens": 70958036.0,
      "step": 36270
    },
    {
      "entropy": 5.787381887435913,
      "epoch": 3.626080871694907,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003732269986291154,
      "loss": 5.3096,
      "mean_token_accuracy": 0.2082212582230568,
      "num_tokens": 70968652.0,
      "step": 36275
    },
    {
      "entropy": 5.869317483901978,
      "epoch": 3.626580696756135,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037319488887274096,
      "loss": 5.4259,
      "mean_token_accuracy": 0.19708781987428664,
      "num_tokens": 70978889.0,
      "step": 36280
    },
    {
      "entropy": 5.894716072082519,
      "epoch": 3.627080521817364,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003731627766458084,
      "loss": 5.4476,
      "mean_token_accuracy": 0.20239587128162384,
      "num_tokens": 70988290.0,
      "step": 36285
    },
    {
      "entropy": 5.899331188201904,
      "epoch": 3.6275803468785925,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003731306619491257,
      "loss": 5.3936,
      "mean_token_accuracy": 0.20184194445610046,
      "num_tokens": 70997914.0,
      "step": 36290
    },
    {
      "entropy": 5.840006303787232,
      "epoch": 3.6280801719398212,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003730985447835008,
      "loss": 5.2863,
      "mean_token_accuracy": 0.212914901971817,
      "num_tokens": 71007395.0,
      "step": 36295
    },
    {
      "entropy": 5.837669563293457,
      "epoch": 3.6285799970010495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003730664251497417,
      "loss": 5.3736,
      "mean_token_accuracy": 0.20060414224863052,
      "num_tokens": 71017477.0,
      "step": 36300
    },
    {
      "entropy": 5.898233318328858,
      "epoch": 3.629079822062278,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003730343030486566,
      "loss": 5.4166,
      "mean_token_accuracy": 0.2011183723807335,
      "num_tokens": 71027634.0,
      "step": 36305
    },
    {
      "entropy": 5.9053129196167,
      "epoch": 3.629579647123507,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003730021784810536,
      "loss": 5.3894,
      "mean_token_accuracy": 0.20183207243680953,
      "num_tokens": 71037919.0,
      "step": 36310
    },
    {
      "entropy": 5.882404756546021,
      "epoch": 3.630079472184735,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003729700514477411,
      "loss": 5.3229,
      "mean_token_accuracy": 0.21141064763069153,
      "num_tokens": 71046883.0,
      "step": 36315
    },
    {
      "entropy": 5.785953664779663,
      "epoch": 3.630579297245964,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003729379219495273,
      "loss": 5.1742,
      "mean_token_accuracy": 0.22108351737260817,
      "num_tokens": 71056910.0,
      "step": 36320
    },
    {
      "entropy": 5.806769704818725,
      "epoch": 3.6310791223071925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003729057899872206,
      "loss": 5.3294,
      "mean_token_accuracy": 0.20592710822820665,
      "num_tokens": 71067305.0,
      "step": 36325
    },
    {
      "entropy": 5.862045669555664,
      "epoch": 3.6315789473684212,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003728736555616293,
      "loss": 5.2883,
      "mean_token_accuracy": 0.21113515198230742,
      "num_tokens": 71076886.0,
      "step": 36330
    },
    {
      "entropy": 5.890304374694824,
      "epoch": 3.6320787724296495,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037284151867356217,
      "loss": 5.3984,
      "mean_token_accuracy": 0.19950979501008986,
      "num_tokens": 71086063.0,
      "step": 36335
    },
    {
      "entropy": 5.914308834075928,
      "epoch": 3.632578597490878,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037280937932382754,
      "loss": 5.4563,
      "mean_token_accuracy": 0.19673918038606644,
      "num_tokens": 71095512.0,
      "step": 36340
    },
    {
      "entropy": 5.922748851776123,
      "epoch": 3.633078422552107,
      "grad_norm": 1.0625,
      "learning_rate": 0.000372777237513234,
      "loss": 5.4662,
      "mean_token_accuracy": 0.1925046846270561,
      "num_tokens": 71106941.0,
      "step": 36345
    },
    {
      "entropy": 5.862897872924805,
      "epoch": 3.633578247613335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037274509324259047,
      "loss": 5.2976,
      "mean_token_accuracy": 0.20440861731767654,
      "num_tokens": 71116886.0,
      "step": 36350
    },
    {
      "entropy": 5.827015018463134,
      "epoch": 3.634078072674564,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003727129465127055,
      "loss": 5.292,
      "mean_token_accuracy": 0.21375729739665986,
      "num_tokens": 71126245.0,
      "step": 36355
    },
    {
      "entropy": 5.894792366027832,
      "epoch": 3.6345778977357925,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003726807973243879,
      "loss": 5.4683,
      "mean_token_accuracy": 0.19898669719696044,
      "num_tokens": 71135662.0,
      "step": 36360
    },
    {
      "entropy": 5.838885736465454,
      "epoch": 3.6350777227970212,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037264864567844664,
      "loss": 5.3787,
      "mean_token_accuracy": 0.19960560500621796,
      "num_tokens": 71145523.0,
      "step": 36365
    },
    {
      "entropy": 5.948162412643432,
      "epoch": 3.6355775478582495,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003726164915756905,
      "loss": 5.4529,
      "mean_token_accuracy": 0.201816625893116,
      "num_tokens": 71155026.0,
      "step": 36370
    },
    {
      "entropy": 5.927151250839233,
      "epoch": 3.636077372919478,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037258433501692856,
      "loss": 5.4009,
      "mean_token_accuracy": 0.20197958052158355,
      "num_tokens": 71165465.0,
      "step": 36375
    },
    {
      "entropy": 5.807666349411011,
      "epoch": 3.636577197980707,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003725521760029699,
      "loss": 5.3562,
      "mean_token_accuracy": 0.20802891701459886,
      "num_tokens": 71175739.0,
      "step": 36380
    },
    {
      "entropy": 5.924394273757935,
      "epoch": 3.637077023041935,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037252001453462366,
      "loss": 5.4438,
      "mean_token_accuracy": 0.20055694729089737,
      "num_tokens": 71186318.0,
      "step": 36385
    },
    {
      "entropy": 5.971444940567016,
      "epoch": 3.637576848103164,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037248785061269883,
      "loss": 5.4541,
      "mean_token_accuracy": 0.1964578479528427,
      "num_tokens": 71195858.0,
      "step": 36390
    },
    {
      "entropy": 5.95404281616211,
      "epoch": 3.6380766731643925,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003724556842380048,
      "loss": 5.467,
      "mean_token_accuracy": 0.19879273623228072,
      "num_tokens": 71205479.0,
      "step": 36395
    },
    {
      "entropy": 5.845600986480713,
      "epoch": 3.6385764982256212,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037242351541135084,
      "loss": 5.3411,
      "mean_token_accuracy": 0.20931735932826995,
      "num_tokens": 71214745.0,
      "step": 36400
    },
    {
      "entropy": 5.908982133865356,
      "epoch": 3.6390763232868495,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037239134413354633,
      "loss": 5.4835,
      "mean_token_accuracy": 0.19421912580728531,
      "num_tokens": 71225052.0,
      "step": 36405
    },
    {
      "entropy": 5.983360862731933,
      "epoch": 3.639576148348078,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003723591704054006,
      "loss": 5.44,
      "mean_token_accuracy": 0.19419293403625487,
      "num_tokens": 71234223.0,
      "step": 36410
    },
    {
      "entropy": 5.819550943374634,
      "epoch": 3.640075973409307,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037232699422772323,
      "loss": 5.2565,
      "mean_token_accuracy": 0.21170630156993867,
      "num_tokens": 71244578.0,
      "step": 36415
    },
    {
      "entropy": 5.784280443191529,
      "epoch": 3.640575798470535,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003722948156013238,
      "loss": 5.3203,
      "mean_token_accuracy": 0.20624122768640518,
      "num_tokens": 71254859.0,
      "step": 36420
    },
    {
      "entropy": 5.8360484600067135,
      "epoch": 3.641075623531764,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037226263452701177,
      "loss": 5.3204,
      "mean_token_accuracy": 0.20698924362659454,
      "num_tokens": 71265948.0,
      "step": 36425
    },
    {
      "entropy": 5.861405658721924,
      "epoch": 3.6415754485929925,
      "grad_norm": 1.125,
      "learning_rate": 0.0003722304510055969,
      "loss": 5.3976,
      "mean_token_accuracy": 0.2017969772219658,
      "num_tokens": 71276572.0,
      "step": 36430
    },
    {
      "entropy": 5.847955226898193,
      "epoch": 3.6420752736542212,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037219826503788884,
      "loss": 5.3245,
      "mean_token_accuracy": 0.20125530362129213,
      "num_tokens": 71287486.0,
      "step": 36435
    },
    {
      "entropy": 5.895530033111572,
      "epoch": 3.6425750987154495,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003721660766246975,
      "loss": 5.3894,
      "mean_token_accuracy": 0.20082923173904418,
      "num_tokens": 71296837.0,
      "step": 36440
    },
    {
      "entropy": 5.772824478149414,
      "epoch": 3.643074923776678,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003721338857668327,
      "loss": 5.3,
      "mean_token_accuracy": 0.20739001780748367,
      "num_tokens": 71307542.0,
      "step": 36445
    },
    {
      "entropy": 5.833636474609375,
      "epoch": 3.643574748837907,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003721016924651042,
      "loss": 5.3075,
      "mean_token_accuracy": 0.21112324595451354,
      "num_tokens": 71317325.0,
      "step": 36450
    },
    {
      "entropy": 5.936425971984863,
      "epoch": 3.644074573899135,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003720694967203223,
      "loss": 5.4228,
      "mean_token_accuracy": 0.20441036075353622,
      "num_tokens": 71327180.0,
      "step": 36455
    },
    {
      "entropy": 5.923714303970337,
      "epoch": 3.644574398960364,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037203729853329675,
      "loss": 5.3616,
      "mean_token_accuracy": 0.21203464269638062,
      "num_tokens": 71337213.0,
      "step": 36460
    },
    {
      "entropy": 5.836633396148682,
      "epoch": 3.6450742240215925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037200509790483773,
      "loss": 5.3588,
      "mean_token_accuracy": 0.20520682036876678,
      "num_tokens": 71347680.0,
      "step": 36465
    },
    {
      "entropy": 5.850503158569336,
      "epoch": 3.6455740490828212,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003719728948357554,
      "loss": 5.3158,
      "mean_token_accuracy": 0.20175671726465225,
      "num_tokens": 71356525.0,
      "step": 36470
    },
    {
      "entropy": 5.851750469207763,
      "epoch": 3.6460738741440495,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037194068932686,
      "loss": 5.4393,
      "mean_token_accuracy": 0.19449986070394515,
      "num_tokens": 71365846.0,
      "step": 36475
    },
    {
      "entropy": 5.864905881881714,
      "epoch": 3.646573699205278,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037190848137896176,
      "loss": 5.309,
      "mean_token_accuracy": 0.21140700429677964,
      "num_tokens": 71374447.0,
      "step": 36480
    },
    {
      "entropy": 5.887523603439331,
      "epoch": 3.647073524266507,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037187627099287114,
      "loss": 5.3579,
      "mean_token_accuracy": 0.20423776805400848,
      "num_tokens": 71383801.0,
      "step": 36485
    },
    {
      "entropy": 5.839799690246582,
      "epoch": 3.647573349327735,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037184405816939847,
      "loss": 5.3672,
      "mean_token_accuracy": 0.20675285756587983,
      "num_tokens": 71393905.0,
      "step": 36490
    },
    {
      "entropy": 5.924411058425903,
      "epoch": 3.648073174388964,
      "grad_norm": 1.125,
      "learning_rate": 0.0003718118429093542,
      "loss": 5.3887,
      "mean_token_accuracy": 0.20616769343614577,
      "num_tokens": 71403524.0,
      "step": 36495
    },
    {
      "entropy": 5.861736059188843,
      "epoch": 3.6485729994501925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003717796252135488,
      "loss": 5.3187,
      "mean_token_accuracy": 0.20677988976240158,
      "num_tokens": 71413838.0,
      "step": 36500
    },
    {
      "entropy": 5.821837663650513,
      "epoch": 3.6490728245114212,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037174740508279296,
      "loss": 5.3516,
      "mean_token_accuracy": 0.20641265511512757,
      "num_tokens": 71423722.0,
      "step": 36505
    },
    {
      "entropy": 5.898153305053711,
      "epoch": 3.6495726495726495,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003717151825178974,
      "loss": 5.415,
      "mean_token_accuracy": 0.20855721831321716,
      "num_tokens": 71432918.0,
      "step": 36510
    },
    {
      "entropy": 5.86966462135315,
      "epoch": 3.650072474633878,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003716829575196726,
      "loss": 5.408,
      "mean_token_accuracy": 0.20406162440776826,
      "num_tokens": 71443342.0,
      "step": 36515
    },
    {
      "entropy": 5.880678415298462,
      "epoch": 3.6505722996951064,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003716507300889296,
      "loss": 5.3858,
      "mean_token_accuracy": 0.19912228882312774,
      "num_tokens": 71451908.0,
      "step": 36520
    },
    {
      "entropy": 5.885339593887329,
      "epoch": 3.651072124756335,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037161850022647905,
      "loss": 5.3331,
      "mean_token_accuracy": 0.2110485702753067,
      "num_tokens": 71461040.0,
      "step": 36525
    },
    {
      "entropy": 5.780408143997192,
      "epoch": 3.651571949817564,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003715862679331318,
      "loss": 5.1517,
      "mean_token_accuracy": 0.23330015391111375,
      "num_tokens": 71470567.0,
      "step": 36530
    },
    {
      "entropy": 5.790174913406372,
      "epoch": 3.6520717748787925,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037155403320969906,
      "loss": 5.2866,
      "mean_token_accuracy": 0.21053972393274306,
      "num_tokens": 71480063.0,
      "step": 36535
    },
    {
      "entropy": 5.809141731262207,
      "epoch": 3.6525715999400212,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003715217960569915,
      "loss": 5.4396,
      "mean_token_accuracy": 0.19865241646766663,
      "num_tokens": 71489859.0,
      "step": 36540
    },
    {
      "entropy": 5.857311773300171,
      "epoch": 3.6530714250012495,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037148955647582063,
      "loss": 5.3935,
      "mean_token_accuracy": 0.20011297464370728,
      "num_tokens": 71500759.0,
      "step": 36545
    },
    {
      "entropy": 5.926058578491211,
      "epoch": 3.653571250062478,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037145731446699715,
      "loss": 5.4122,
      "mean_token_accuracy": 0.20049784183502198,
      "num_tokens": 71510524.0,
      "step": 36550
    },
    {
      "entropy": 5.793809652328491,
      "epoch": 3.6540710751237064,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003714250700313325,
      "loss": 5.2876,
      "mean_token_accuracy": 0.20660381019115448,
      "num_tokens": 71520301.0,
      "step": 36555
    },
    {
      "entropy": 5.792587852478027,
      "epoch": 3.654570900184935,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037139282316963804,
      "loss": 5.3215,
      "mean_token_accuracy": 0.20709035843610762,
      "num_tokens": 71531800.0,
      "step": 36560
    },
    {
      "entropy": 5.821863317489624,
      "epoch": 3.655070725246164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003713605738827248,
      "loss": 5.3049,
      "mean_token_accuracy": 0.2073420464992523,
      "num_tokens": 71541725.0,
      "step": 36565
    },
    {
      "entropy": 5.8516919136047365,
      "epoch": 3.6555705503073925,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003713283221714044,
      "loss": 5.3672,
      "mean_token_accuracy": 0.19852503687143325,
      "num_tokens": 71551059.0,
      "step": 36570
    },
    {
      "entropy": 5.854003190994263,
      "epoch": 3.6560703753686212,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037129606803648814,
      "loss": 5.2811,
      "mean_token_accuracy": 0.20789351165294648,
      "num_tokens": 71560634.0,
      "step": 36575
    },
    {
      "entropy": 5.798936367034912,
      "epoch": 3.6565702004298495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037126381147878763,
      "loss": 5.4034,
      "mean_token_accuracy": 0.20937598943710328,
      "num_tokens": 71568926.0,
      "step": 36580
    },
    {
      "entropy": 5.77456111907959,
      "epoch": 3.657070025491078,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037123155249911435,
      "loss": 5.3438,
      "mean_token_accuracy": 0.21027534008026122,
      "num_tokens": 71579518.0,
      "step": 36585
    },
    {
      "entropy": 5.899652576446533,
      "epoch": 3.6575698505523064,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037119929109828,
      "loss": 5.3802,
      "mean_token_accuracy": 0.2018597200512886,
      "num_tokens": 71589071.0,
      "step": 36590
    },
    {
      "entropy": 5.871406602859497,
      "epoch": 3.658069675613535,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003711670272770962,
      "loss": 5.293,
      "mean_token_accuracy": 0.21473564803600312,
      "num_tokens": 71599346.0,
      "step": 36595
    },
    {
      "entropy": 5.958245706558228,
      "epoch": 3.658569500674764,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037113476103637483,
      "loss": 5.4743,
      "mean_token_accuracy": 0.19937346428632735,
      "num_tokens": 71609131.0,
      "step": 36600
    },
    {
      "entropy": 5.880307817459107,
      "epoch": 3.6590693257359925,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003711024923769275,
      "loss": 5.3901,
      "mean_token_accuracy": 0.20000076293945312,
      "num_tokens": 71618983.0,
      "step": 36605
    },
    {
      "entropy": 5.951070022583008,
      "epoch": 3.6595691507972212,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003710702212995662,
      "loss": 5.4797,
      "mean_token_accuracy": 0.19706238508224488,
      "num_tokens": 71628811.0,
      "step": 36610
    },
    {
      "entropy": 5.969335699081421,
      "epoch": 3.6600689758584495,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037103794780510296,
      "loss": 5.3487,
      "mean_token_accuracy": 0.20029989033937454,
      "num_tokens": 71638427.0,
      "step": 36615
    },
    {
      "entropy": 5.92476372718811,
      "epoch": 3.660568800919678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037100567189434966,
      "loss": 5.4322,
      "mean_token_accuracy": 0.20010368674993514,
      "num_tokens": 71648691.0,
      "step": 36620
    },
    {
      "entropy": 5.815634822845459,
      "epoch": 3.6610686259809064,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003709733935681183,
      "loss": 5.2916,
      "mean_token_accuracy": 0.2121470659971237,
      "num_tokens": 71657450.0,
      "step": 36625
    },
    {
      "entropy": 5.788024997711181,
      "epoch": 3.661568451042135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000370941112827221,
      "loss": 5.3254,
      "mean_token_accuracy": 0.21005641371011735,
      "num_tokens": 71666087.0,
      "step": 36630
    },
    {
      "entropy": 5.755376625061035,
      "epoch": 3.662068276103364,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037090882967247003,
      "loss": 5.2856,
      "mean_token_accuracy": 0.2082023501396179,
      "num_tokens": 71675838.0,
      "step": 36635
    },
    {
      "entropy": 5.911485767364502,
      "epoch": 3.6625681011645925,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003708765441046776,
      "loss": 5.4483,
      "mean_token_accuracy": 0.1937359094619751,
      "num_tokens": 71686210.0,
      "step": 36640
    },
    {
      "entropy": 5.8807415008544925,
      "epoch": 3.6630679262258212,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037084425612465605,
      "loss": 5.3732,
      "mean_token_accuracy": 0.20442487597465514,
      "num_tokens": 71694954.0,
      "step": 36645
    },
    {
      "entropy": 5.856083631515503,
      "epoch": 3.6635677512870495,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003708119657332176,
      "loss": 5.3479,
      "mean_token_accuracy": 0.2126806065440178,
      "num_tokens": 71704074.0,
      "step": 36650
    },
    {
      "entropy": 5.785268831253052,
      "epoch": 3.664067576348278,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037077967293117477,
      "loss": 5.2526,
      "mean_token_accuracy": 0.21128099113702775,
      "num_tokens": 71713419.0,
      "step": 36655
    },
    {
      "entropy": 5.920999193191529,
      "epoch": 3.6645674014095064,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037074737771933996,
      "loss": 5.3692,
      "mean_token_accuracy": 0.20620375871658325,
      "num_tokens": 71723779.0,
      "step": 36660
    },
    {
      "entropy": 5.852167797088623,
      "epoch": 3.665067226470735,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003707150800985259,
      "loss": 5.3584,
      "mean_token_accuracy": 0.2033916547894478,
      "num_tokens": 71733902.0,
      "step": 36665
    },
    {
      "entropy": 5.880349254608154,
      "epoch": 3.665567051531964,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037068278006954493,
      "loss": 5.3576,
      "mean_token_accuracy": 0.21047522872686386,
      "num_tokens": 71743372.0,
      "step": 36670
    },
    {
      "entropy": 5.837267446517944,
      "epoch": 3.6660668765931925,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037065047763320993,
      "loss": 5.3913,
      "mean_token_accuracy": 0.2066932961344719,
      "num_tokens": 71752768.0,
      "step": 36675
    },
    {
      "entropy": 5.883897018432617,
      "epoch": 3.6665667016544212,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003706181727903334,
      "loss": 5.378,
      "mean_token_accuracy": 0.20209911316633225,
      "num_tokens": 71762339.0,
      "step": 36680
    },
    {
      "entropy": 5.935663366317749,
      "epoch": 3.6670665267156495,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037058586554172834,
      "loss": 5.3953,
      "mean_token_accuracy": 0.20443458557128907,
      "num_tokens": 71772375.0,
      "step": 36685
    },
    {
      "entropy": 5.805738353729248,
      "epoch": 3.667566351776878,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037055355588820746,
      "loss": 5.2895,
      "mean_token_accuracy": 0.20554783046245576,
      "num_tokens": 71782002.0,
      "step": 36690
    },
    {
      "entropy": 5.888630676269531,
      "epoch": 3.6680661768381064,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003705212438305837,
      "loss": 5.3783,
      "mean_token_accuracy": 0.20302131921052932,
      "num_tokens": 71790511.0,
      "step": 36695
    },
    {
      "entropy": 5.842545700073242,
      "epoch": 3.668566001899335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003704889293696701,
      "loss": 5.3476,
      "mean_token_accuracy": 0.19916646033525467,
      "num_tokens": 71801109.0,
      "step": 36700
    },
    {
      "entropy": 5.848211097717285,
      "epoch": 3.669065826960564,
      "grad_norm": 1.125,
      "learning_rate": 0.00037045661250627955,
      "loss": 5.314,
      "mean_token_accuracy": 0.21688870787620546,
      "num_tokens": 71811072.0,
      "step": 36705
    },
    {
      "entropy": 5.83777084350586,
      "epoch": 3.6695656520217925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003704242932412252,
      "loss": 5.3929,
      "mean_token_accuracy": 0.19445801973342897,
      "num_tokens": 71821884.0,
      "step": 36710
    },
    {
      "entropy": 5.925439929962158,
      "epoch": 3.6700654770830208,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037039197157532017,
      "loss": 5.4132,
      "mean_token_accuracy": 0.20958239138126372,
      "num_tokens": 71830816.0,
      "step": 36715
    },
    {
      "entropy": 5.8964766502380375,
      "epoch": 3.6705653021442495,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003703596475093776,
      "loss": 5.3822,
      "mean_token_accuracy": 0.1986476570367813,
      "num_tokens": 71840906.0,
      "step": 36720
    },
    {
      "entropy": 5.936352157592774,
      "epoch": 3.671065127205478,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037032732104421096,
      "loss": 5.4237,
      "mean_token_accuracy": 0.20241997838020326,
      "num_tokens": 71850653.0,
      "step": 36725
    },
    {
      "entropy": 5.937456512451172,
      "epoch": 3.6715649522667064,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037029499218063346,
      "loss": 5.4225,
      "mean_token_accuracy": 0.1998118206858635,
      "num_tokens": 71859768.0,
      "step": 36730
    },
    {
      "entropy": 5.872750282287598,
      "epoch": 3.672064777327935,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037026266091945843,
      "loss": 5.3748,
      "mean_token_accuracy": 0.20110083371400833,
      "num_tokens": 71870178.0,
      "step": 36735
    },
    {
      "entropy": 5.789882612228394,
      "epoch": 3.672564602389164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003702303272614994,
      "loss": 5.2373,
      "mean_token_accuracy": 0.21506963819265365,
      "num_tokens": 71881149.0,
      "step": 36740
    },
    {
      "entropy": 5.797182607650757,
      "epoch": 3.6730644274503925,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003701979912075698,
      "loss": 5.3103,
      "mean_token_accuracy": 0.2094180628657341,
      "num_tokens": 71891120.0,
      "step": 36745
    },
    {
      "entropy": 5.882326650619507,
      "epoch": 3.6735642525116208,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003701656527584832,
      "loss": 5.4398,
      "mean_token_accuracy": 0.20535662323236464,
      "num_tokens": 71900276.0,
      "step": 36750
    },
    {
      "entropy": 5.878246927261353,
      "epoch": 3.6740640775728495,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003701333119150534,
      "loss": 5.3109,
      "mean_token_accuracy": 0.2056530937552452,
      "num_tokens": 71909234.0,
      "step": 36755
    },
    {
      "entropy": 5.85079984664917,
      "epoch": 3.674563902634078,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037010096867809384,
      "loss": 5.3239,
      "mean_token_accuracy": 0.20258254706859588,
      "num_tokens": 71918103.0,
      "step": 36760
    },
    {
      "entropy": 5.826852655410766,
      "epoch": 3.6750637276953064,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037006862304841845,
      "loss": 5.3871,
      "mean_token_accuracy": 0.20452219098806382,
      "num_tokens": 71927663.0,
      "step": 36765
    },
    {
      "entropy": 5.858156394958496,
      "epoch": 3.675563552756535,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003700362750268409,
      "loss": 5.3166,
      "mean_token_accuracy": 0.21575934290885926,
      "num_tokens": 71937940.0,
      "step": 36770
    },
    {
      "entropy": 5.823376560211182,
      "epoch": 3.676063377817764,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003700039246141752,
      "loss": 5.3133,
      "mean_token_accuracy": 0.2053188756108284,
      "num_tokens": 71947522.0,
      "step": 36775
    },
    {
      "entropy": 5.820630168914795,
      "epoch": 3.6765632028789925,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003699715718112351,
      "loss": 5.3255,
      "mean_token_accuracy": 0.2036474272608757,
      "num_tokens": 71957127.0,
      "step": 36780
    },
    {
      "entropy": 5.8204444408416744,
      "epoch": 3.6770630279402208,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036993921661883475,
      "loss": 5.3704,
      "mean_token_accuracy": 0.20483436435461044,
      "num_tokens": 71967197.0,
      "step": 36785
    },
    {
      "entropy": 5.94649748802185,
      "epoch": 3.6775628530014495,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036990685903778815,
      "loss": 5.4382,
      "mean_token_accuracy": 0.1941695511341095,
      "num_tokens": 71977115.0,
      "step": 36790
    },
    {
      "entropy": 5.924796104431152,
      "epoch": 3.678062678062678,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036987449906890937,
      "loss": 5.3371,
      "mean_token_accuracy": 0.20578284859657286,
      "num_tokens": 71986410.0,
      "step": 36795
    },
    {
      "entropy": 5.874158668518066,
      "epoch": 3.6785625031239064,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036984213671301266,
      "loss": 5.4668,
      "mean_token_accuracy": 0.20407605022192002,
      "num_tokens": 71995936.0,
      "step": 36800
    },
    {
      "entropy": 5.939153528213501,
      "epoch": 3.679062328185135,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036980977197091213,
      "loss": 5.4148,
      "mean_token_accuracy": 0.19563148468732833,
      "num_tokens": 72006732.0,
      "step": 36805
    },
    {
      "entropy": 5.969173383712769,
      "epoch": 3.679562153246364,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036977740484342214,
      "loss": 5.3664,
      "mean_token_accuracy": 0.20622149258852004,
      "num_tokens": 72016391.0,
      "step": 36810
    },
    {
      "entropy": 5.82457127571106,
      "epoch": 3.6800619783075925,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003697450353313571,
      "loss": 5.241,
      "mean_token_accuracy": 0.21127178966999055,
      "num_tokens": 72025465.0,
      "step": 36815
    },
    {
      "entropy": 5.782774782180786,
      "epoch": 3.6805618033688208,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003697126634355313,
      "loss": 5.2103,
      "mean_token_accuracy": 0.22183097451925277,
      "num_tokens": 72035786.0,
      "step": 36820
    },
    {
      "entropy": 5.882055997848511,
      "epoch": 3.6810616284300495,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036968028915675924,
      "loss": 5.446,
      "mean_token_accuracy": 0.19532289803028108,
      "num_tokens": 72046335.0,
      "step": 36825
    },
    {
      "entropy": 5.820396089553833,
      "epoch": 3.681561453491278,
      "grad_norm": 1.125,
      "learning_rate": 0.0003696479124958555,
      "loss": 5.3122,
      "mean_token_accuracy": 0.2081267222762108,
      "num_tokens": 72055569.0,
      "step": 36830
    },
    {
      "entropy": 5.8332795143127445,
      "epoch": 3.6820612785525064,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003696155334536346,
      "loss": 5.223,
      "mean_token_accuracy": 0.2113168269395828,
      "num_tokens": 72064995.0,
      "step": 36835
    },
    {
      "entropy": 5.795692443847656,
      "epoch": 3.682561103613735,
      "grad_norm": 1.125,
      "learning_rate": 0.0003695831520309113,
      "loss": 5.1987,
      "mean_token_accuracy": 0.22654290795326232,
      "num_tokens": 72075826.0,
      "step": 36840
    },
    {
      "entropy": 5.904436826705933,
      "epoch": 3.683060928674964,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036955076822850014,
      "loss": 5.3581,
      "mean_token_accuracy": 0.19899963587522507,
      "num_tokens": 72084901.0,
      "step": 36845
    },
    {
      "entropy": 5.834868240356445,
      "epoch": 3.6835607537361925,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036951838204721607,
      "loss": 5.4004,
      "mean_token_accuracy": 0.2073850378394127,
      "num_tokens": 72095032.0,
      "step": 36850
    },
    {
      "entropy": 5.919541835784912,
      "epoch": 3.6840605787974208,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036948599348787375,
      "loss": 5.4012,
      "mean_token_accuracy": 0.2069968581199646,
      "num_tokens": 72104160.0,
      "step": 36855
    },
    {
      "entropy": 5.921834325790405,
      "epoch": 3.6845604038586495,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036945360255128816,
      "loss": 5.3879,
      "mean_token_accuracy": 0.20400925874710082,
      "num_tokens": 72113421.0,
      "step": 36860
    },
    {
      "entropy": 5.825408983230591,
      "epoch": 3.685060228919878,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036942120923827426,
      "loss": 5.2861,
      "mean_token_accuracy": 0.20306050181388854,
      "num_tokens": 72123216.0,
      "step": 36865
    },
    {
      "entropy": 5.839735603332519,
      "epoch": 3.6855600539811064,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036938881354964704,
      "loss": 5.3722,
      "mean_token_accuracy": 0.20112653821706772,
      "num_tokens": 72132701.0,
      "step": 36870
    },
    {
      "entropy": 5.858726072311401,
      "epoch": 3.686059879042335,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036935641548622154,
      "loss": 5.3933,
      "mean_token_accuracy": 0.19446017146110534,
      "num_tokens": 72142526.0,
      "step": 36875
    },
    {
      "entropy": 5.887330818176269,
      "epoch": 3.686559704103564,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003693240150488129,
      "loss": 5.4308,
      "mean_token_accuracy": 0.20089459121227266,
      "num_tokens": 72151487.0,
      "step": 36880
    },
    {
      "entropy": 5.856964826583862,
      "epoch": 3.6870595291647925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003692916122382364,
      "loss": 5.3186,
      "mean_token_accuracy": 0.20978332161903382,
      "num_tokens": 72161128.0,
      "step": 36885
    },
    {
      "entropy": 5.819064474105835,
      "epoch": 3.6875593542260208,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036925920705530713,
      "loss": 5.3291,
      "mean_token_accuracy": 0.20714191198349,
      "num_tokens": 72170883.0,
      "step": 36890
    },
    {
      "entropy": 5.899317693710327,
      "epoch": 3.6880591792872495,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036922679950084043,
      "loss": 5.4215,
      "mean_token_accuracy": 0.2021089553833008,
      "num_tokens": 72180859.0,
      "step": 36895
    },
    {
      "entropy": 5.865265274047852,
      "epoch": 3.688559004348478,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003691943895756518,
      "loss": 5.4172,
      "mean_token_accuracy": 0.19125896394252778,
      "num_tokens": 72191626.0,
      "step": 36900
    },
    {
      "entropy": 5.781890487670898,
      "epoch": 3.6890588294097064,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036916197728055646,
      "loss": 5.2665,
      "mean_token_accuracy": 0.21025144606828688,
      "num_tokens": 72201847.0,
      "step": 36905
    },
    {
      "entropy": 5.937229204177856,
      "epoch": 3.689558654470935,
      "grad_norm": 1.0,
      "learning_rate": 0.00036912956261637005,
      "loss": 5.468,
      "mean_token_accuracy": 0.19469777047634124,
      "num_tokens": 72213863.0,
      "step": 36910
    },
    {
      "entropy": 5.950659275054932,
      "epoch": 3.690058479532164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003690971455839081,
      "loss": 5.3845,
      "mean_token_accuracy": 0.20296284407377244,
      "num_tokens": 72223710.0,
      "step": 36915
    },
    {
      "entropy": 5.966817903518677,
      "epoch": 3.6905583045933925,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036906472618398605,
      "loss": 5.4551,
      "mean_token_accuracy": 0.18890468180179595,
      "num_tokens": 72233349.0,
      "step": 36920
    },
    {
      "entropy": 5.817356300354004,
      "epoch": 3.6910581296546208,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036903230441741986,
      "loss": 5.2912,
      "mean_token_accuracy": 0.2103637382388115,
      "num_tokens": 72243049.0,
      "step": 36925
    },
    {
      "entropy": 5.926062774658203,
      "epoch": 3.6915579547158495,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036899988028502495,
      "loss": 5.3952,
      "mean_token_accuracy": 0.1905147299170494,
      "num_tokens": 72253560.0,
      "step": 36930
    },
    {
      "entropy": 5.880063438415528,
      "epoch": 3.692057779777078,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036896745378761734,
      "loss": 5.3951,
      "mean_token_accuracy": 0.19877127856016158,
      "num_tokens": 72262964.0,
      "step": 36935
    },
    {
      "entropy": 5.782339954376221,
      "epoch": 3.6925576048383064,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036893502492601273,
      "loss": 5.292,
      "mean_token_accuracy": 0.20810672342777253,
      "num_tokens": 72272891.0,
      "step": 36940
    },
    {
      "entropy": 5.886971378326416,
      "epoch": 3.693057429899535,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003689025937010271,
      "loss": 5.4192,
      "mean_token_accuracy": 0.20083497762680053,
      "num_tokens": 72283646.0,
      "step": 36945
    },
    {
      "entropy": 5.853969860076904,
      "epoch": 3.693557254960764,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036887016011347633,
      "loss": 5.3169,
      "mean_token_accuracy": 0.2025834932923317,
      "num_tokens": 72293220.0,
      "step": 36950
    },
    {
      "entropy": 5.817947435379028,
      "epoch": 3.6940570800219925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003688377241641765,
      "loss": 5.2734,
      "mean_token_accuracy": 0.21357757598161697,
      "num_tokens": 72301909.0,
      "step": 36955
    },
    {
      "entropy": 5.878945302963257,
      "epoch": 3.6945569050832208,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036880528585394364,
      "loss": 5.4131,
      "mean_token_accuracy": 0.1939685359597206,
      "num_tokens": 72312339.0,
      "step": 36960
    },
    {
      "entropy": 5.896247005462646,
      "epoch": 3.6950567301444495,
      "grad_norm": 1.203125,
      "learning_rate": 0.000368772845183594,
      "loss": 5.303,
      "mean_token_accuracy": 0.20574909895658494,
      "num_tokens": 72322045.0,
      "step": 36965
    },
    {
      "entropy": 5.914354753494263,
      "epoch": 3.695556555205678,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036874040215394366,
      "loss": 5.4808,
      "mean_token_accuracy": 0.19939763844013214,
      "num_tokens": 72331903.0,
      "step": 36970
    },
    {
      "entropy": 5.83381028175354,
      "epoch": 3.6960563802669064,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036870795676580885,
      "loss": 5.3264,
      "mean_token_accuracy": 0.20341651737689972,
      "num_tokens": 72341502.0,
      "step": 36975
    },
    {
      "entropy": 5.823133563995361,
      "epoch": 3.696556205328135,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003686755090200061,
      "loss": 5.2908,
      "mean_token_accuracy": 0.2073708102107048,
      "num_tokens": 72351040.0,
      "step": 36980
    },
    {
      "entropy": 5.85510482788086,
      "epoch": 3.697056030389364,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003686430589173516,
      "loss": 5.3281,
      "mean_token_accuracy": 0.2126515820622444,
      "num_tokens": 72360924.0,
      "step": 36985
    },
    {
      "entropy": 5.865300512313842,
      "epoch": 3.6975558554505925,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036861060645866184,
      "loss": 5.2873,
      "mean_token_accuracy": 0.21026142686605453,
      "num_tokens": 72369970.0,
      "step": 36990
    },
    {
      "entropy": 5.829508972167969,
      "epoch": 3.6980556805118208,
      "grad_norm": 1.125,
      "learning_rate": 0.0003685781516447533,
      "loss": 5.333,
      "mean_token_accuracy": 0.1980469599366188,
      "num_tokens": 72379689.0,
      "step": 36995
    },
    {
      "entropy": 5.851101016998291,
      "epoch": 3.6985555055730495,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003685456944764425,
      "loss": 5.4003,
      "mean_token_accuracy": 0.20047316402196885,
      "num_tokens": 72389303.0,
      "step": 37000
    },
    {
      "entropy": 5.848947525024414,
      "epoch": 3.699055330634278,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036851323495454616,
      "loss": 5.3591,
      "mean_token_accuracy": 0.2039233461022377,
      "num_tokens": 72399076.0,
      "step": 37005
    },
    {
      "entropy": 5.869585371017456,
      "epoch": 3.6995551556955064,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036848077307988084,
      "loss": 5.3887,
      "mean_token_accuracy": 0.19577763080596924,
      "num_tokens": 72408012.0,
      "step": 37010
    },
    {
      "entropy": 5.935787153244019,
      "epoch": 3.700054980756735,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036844830885326334,
      "loss": 5.4417,
      "mean_token_accuracy": 0.19533710330724716,
      "num_tokens": 72417067.0,
      "step": 37015
    },
    {
      "entropy": 5.861664295196533,
      "epoch": 3.700554805817964,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036841584227551043,
      "loss": 5.3328,
      "mean_token_accuracy": 0.20863400101661683,
      "num_tokens": 72426388.0,
      "step": 37020
    },
    {
      "entropy": 5.889753293991089,
      "epoch": 3.7010546308791925,
      "grad_norm": 1.234375,
      "learning_rate": 0.000368383373347439,
      "loss": 5.361,
      "mean_token_accuracy": 0.2078451931476593,
      "num_tokens": 72435997.0,
      "step": 37025
    },
    {
      "entropy": 5.9021909713745115,
      "epoch": 3.7015544559404208,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036835090206986594,
      "loss": 5.4768,
      "mean_token_accuracy": 0.19255456626415252,
      "num_tokens": 72446090.0,
      "step": 37030
    },
    {
      "entropy": 5.893128156661987,
      "epoch": 3.7020542810016495,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036831842844360814,
      "loss": 5.4082,
      "mean_token_accuracy": 0.20303575992584227,
      "num_tokens": 72455465.0,
      "step": 37035
    },
    {
      "entropy": 5.856558990478516,
      "epoch": 3.702554106062878,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003682859524694827,
      "loss": 5.379,
      "mean_token_accuracy": 0.2011628970503807,
      "num_tokens": 72465700.0,
      "step": 37040
    },
    {
      "entropy": 5.856666374206543,
      "epoch": 3.7030539311241064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036825347414830674,
      "loss": 5.2765,
      "mean_token_accuracy": 0.2103651061654091,
      "num_tokens": 72475719.0,
      "step": 37045
    },
    {
      "entropy": 5.943807172775268,
      "epoch": 3.703553756185335,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036822099348089737,
      "loss": 5.4827,
      "mean_token_accuracy": 0.19422122687101365,
      "num_tokens": 72484772.0,
      "step": 37050
    },
    {
      "entropy": 5.92194561958313,
      "epoch": 3.704053581246564,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036818851046807176,
      "loss": 5.3743,
      "mean_token_accuracy": 0.20416176617145537,
      "num_tokens": 72495168.0,
      "step": 37055
    },
    {
      "entropy": 5.859837675094605,
      "epoch": 3.7045534063077925,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036815602511064725,
      "loss": 5.3061,
      "mean_token_accuracy": 0.20626766085624695,
      "num_tokens": 72504400.0,
      "step": 37060
    },
    {
      "entropy": 5.874715566635132,
      "epoch": 3.7050532313690208,
      "grad_norm": 1.125,
      "learning_rate": 0.00036812353740944104,
      "loss": 5.41,
      "mean_token_accuracy": 0.19839244037866594,
      "num_tokens": 72515706.0,
      "step": 37065
    },
    {
      "entropy": 5.950783681869507,
      "epoch": 3.7055530564302495,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036809104736527064,
      "loss": 5.4728,
      "mean_token_accuracy": 0.19199186712503433,
      "num_tokens": 72527065.0,
      "step": 37070
    },
    {
      "entropy": 5.904022169113159,
      "epoch": 3.7060528814914777,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036805855497895346,
      "loss": 5.4591,
      "mean_token_accuracy": 0.19778698831796646,
      "num_tokens": 72536711.0,
      "step": 37075
    },
    {
      "entropy": 5.85899133682251,
      "epoch": 3.7065527065527064,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036802606025130696,
      "loss": 5.3599,
      "mean_token_accuracy": 0.21509512215852739,
      "num_tokens": 72546684.0,
      "step": 37080
    },
    {
      "entropy": 5.823461723327637,
      "epoch": 3.707052531613935,
      "grad_norm": 1.0,
      "learning_rate": 0.0003679935631831488,
      "loss": 5.3018,
      "mean_token_accuracy": 0.20817575752735137,
      "num_tokens": 72557107.0,
      "step": 37085
    },
    {
      "entropy": 5.833824634552002,
      "epoch": 3.707552356675164,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036796106377529644,
      "loss": 5.2724,
      "mean_token_accuracy": 0.21292609572410584,
      "num_tokens": 72567280.0,
      "step": 37090
    },
    {
      "entropy": 5.900465297698974,
      "epoch": 3.7080521817363925,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003679285620285676,
      "loss": 5.4323,
      "mean_token_accuracy": 0.20011256635189056,
      "num_tokens": 72578786.0,
      "step": 37095
    },
    {
      "entropy": 5.909240770339966,
      "epoch": 3.7085520067976208,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036789605794378013,
      "loss": 5.3673,
      "mean_token_accuracy": 0.2003122925758362,
      "num_tokens": 72588728.0,
      "step": 37100
    },
    {
      "entropy": 5.920476675033569,
      "epoch": 3.7090518318588495,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036786355152175173,
      "loss": 5.4523,
      "mean_token_accuracy": 0.1990998774766922,
      "num_tokens": 72598117.0,
      "step": 37105
    },
    {
      "entropy": 5.875774669647217,
      "epoch": 3.7095516569200777,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003678310427633003,
      "loss": 5.3262,
      "mean_token_accuracy": 0.20653311759233475,
      "num_tokens": 72607346.0,
      "step": 37110
    },
    {
      "entropy": 5.863622379302979,
      "epoch": 3.7100514819813064,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036779853166924353,
      "loss": 5.3922,
      "mean_token_accuracy": 0.20429882854223252,
      "num_tokens": 72618499.0,
      "step": 37115
    },
    {
      "entropy": 5.8647918701171875,
      "epoch": 3.710551307042535,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003677660182403998,
      "loss": 5.4004,
      "mean_token_accuracy": 0.19941361844539643,
      "num_tokens": 72629150.0,
      "step": 37120
    },
    {
      "entropy": 5.867105007171631,
      "epoch": 3.711051132103764,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036773350247758685,
      "loss": 5.3966,
      "mean_token_accuracy": 0.20128399282693862,
      "num_tokens": 72638640.0,
      "step": 37125
    },
    {
      "entropy": 5.909254550933838,
      "epoch": 3.7115509571649925,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003677009843816228,
      "loss": 5.3633,
      "mean_token_accuracy": 0.20871975570917128,
      "num_tokens": 72648211.0,
      "step": 37130
    },
    {
      "entropy": 5.895858812332153,
      "epoch": 3.7120507822262208,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003676684639533259,
      "loss": 5.3988,
      "mean_token_accuracy": 0.20383803993463517,
      "num_tokens": 72658669.0,
      "step": 37135
    },
    {
      "entropy": 5.855069828033447,
      "epoch": 3.7125506072874495,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003676359411935143,
      "loss": 5.3739,
      "mean_token_accuracy": 0.20675197392702102,
      "num_tokens": 72667297.0,
      "step": 37140
    },
    {
      "entropy": 5.958835458755493,
      "epoch": 3.7130504323486777,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003676034161030062,
      "loss": 5.4907,
      "mean_token_accuracy": 0.19049959778785705,
      "num_tokens": 72676911.0,
      "step": 37145
    },
    {
      "entropy": 5.9873302459716795,
      "epoch": 3.7135502574099064,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036757088868262,
      "loss": 5.5207,
      "mean_token_accuracy": 0.1882346138358116,
      "num_tokens": 72687373.0,
      "step": 37150
    },
    {
      "entropy": 6.0153669834136965,
      "epoch": 3.714050082471135,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036753835893317413,
      "loss": 5.5325,
      "mean_token_accuracy": 0.18875695019960403,
      "num_tokens": 72698716.0,
      "step": 37155
    },
    {
      "entropy": 5.892832660675049,
      "epoch": 3.714549907532364,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003675058268554868,
      "loss": 5.3671,
      "mean_token_accuracy": 0.208619187772274,
      "num_tokens": 72708152.0,
      "step": 37160
    },
    {
      "entropy": 5.866786336898803,
      "epoch": 3.7150497325935925,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003674732924503768,
      "loss": 5.4311,
      "mean_token_accuracy": 0.1990421250462532,
      "num_tokens": 72717541.0,
      "step": 37165
    },
    {
      "entropy": 5.850893974304199,
      "epoch": 3.7155495576548208,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003674407557186625,
      "loss": 5.3983,
      "mean_token_accuracy": 0.20592257827520372,
      "num_tokens": 72727846.0,
      "step": 37170
    },
    {
      "entropy": 5.888995122909546,
      "epoch": 3.7160493827160495,
      "grad_norm": 1.125,
      "learning_rate": 0.00036740821666116265,
      "loss": 5.4015,
      "mean_token_accuracy": 0.20228716135025024,
      "num_tokens": 72737036.0,
      "step": 37175
    },
    {
      "entropy": 5.813001537322998,
      "epoch": 3.7165492077772777,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003673756752786958,
      "loss": 5.3208,
      "mean_token_accuracy": 0.20957865118980407,
      "num_tokens": 72747674.0,
      "step": 37180
    },
    {
      "entropy": 5.799921894073487,
      "epoch": 3.7170490328385064,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036734313157208066,
      "loss": 5.3241,
      "mean_token_accuracy": 0.21555692702531815,
      "num_tokens": 72757696.0,
      "step": 37185
    },
    {
      "entropy": 5.869834947586059,
      "epoch": 3.717548857899735,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003673105855421362,
      "loss": 5.3622,
      "mean_token_accuracy": 0.20193394720554353,
      "num_tokens": 72766540.0,
      "step": 37190
    },
    {
      "entropy": 5.818830442428589,
      "epoch": 3.718048682960964,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003672780371896811,
      "loss": 5.2968,
      "mean_token_accuracy": 0.21284639835357666,
      "num_tokens": 72776158.0,
      "step": 37195
    },
    {
      "entropy": 5.935988664627075,
      "epoch": 3.7185485080221925,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036724548651553433,
      "loss": 5.4319,
      "mean_token_accuracy": 0.19163432568311692,
      "num_tokens": 72786192.0,
      "step": 37200
    },
    {
      "entropy": 5.911564683914184,
      "epoch": 3.7190483330834208,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036721293352051486,
      "loss": 5.4494,
      "mean_token_accuracy": 0.19447508007287978,
      "num_tokens": 72796892.0,
      "step": 37205
    },
    {
      "entropy": 5.752253103256225,
      "epoch": 3.7195481581446495,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003671803782054418,
      "loss": 5.1744,
      "mean_token_accuracy": 0.21694704741239548,
      "num_tokens": 72806936.0,
      "step": 37210
    },
    {
      "entropy": 5.8558472156524655,
      "epoch": 3.7200479832058777,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000367147820571134,
      "loss": 5.3244,
      "mean_token_accuracy": 0.20793746411800385,
      "num_tokens": 72816611.0,
      "step": 37215
    },
    {
      "entropy": 5.825241422653198,
      "epoch": 3.7205478082671064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036711526061841086,
      "loss": 5.3143,
      "mean_token_accuracy": 0.2020142063498497,
      "num_tokens": 72825999.0,
      "step": 37220
    },
    {
      "entropy": 5.851819467544556,
      "epoch": 3.721047633328335,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003670826983480914,
      "loss": 5.4075,
      "mean_token_accuracy": 0.2002849593758583,
      "num_tokens": 72836867.0,
      "step": 37225
    },
    {
      "entropy": 5.932523107528686,
      "epoch": 3.721547458389564,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036705013376099496,
      "loss": 5.4658,
      "mean_token_accuracy": 0.1966622292995453,
      "num_tokens": 72846678.0,
      "step": 37230
    },
    {
      "entropy": 5.854299783706665,
      "epoch": 3.7220472834507925,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003670175668579409,
      "loss": 5.2825,
      "mean_token_accuracy": 0.21026574224233627,
      "num_tokens": 72856642.0,
      "step": 37235
    },
    {
      "entropy": 5.8996875286102295,
      "epoch": 3.7225471085120208,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003669849976397484,
      "loss": 5.4417,
      "mean_token_accuracy": 0.19065888077020646,
      "num_tokens": 72866640.0,
      "step": 37240
    },
    {
      "entropy": 5.8711944103240965,
      "epoch": 3.7230469335732495,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003669524261072371,
      "loss": 5.3959,
      "mean_token_accuracy": 0.1962432846426964,
      "num_tokens": 72876723.0,
      "step": 37245
    },
    {
      "entropy": 5.871005725860596,
      "epoch": 3.7235467586344777,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003669198522612265,
      "loss": 5.3675,
      "mean_token_accuracy": 0.2042446181178093,
      "num_tokens": 72886482.0,
      "step": 37250
    },
    {
      "entropy": 5.837090063095093,
      "epoch": 3.7240465836957064,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036688727610253595,
      "loss": 5.2818,
      "mean_token_accuracy": 0.21779312789440156,
      "num_tokens": 72895548.0,
      "step": 37255
    },
    {
      "entropy": 5.822039175033569,
      "epoch": 3.724546408756935,
      "grad_norm": 1.125,
      "learning_rate": 0.0003668546976319853,
      "loss": 5.2871,
      "mean_token_accuracy": 0.20287957936525344,
      "num_tokens": 72905223.0,
      "step": 37260
    },
    {
      "entropy": 5.924599599838257,
      "epoch": 3.725046233818164,
      "grad_norm": 1.328125,
      "learning_rate": 0.000366822116850394,
      "loss": 5.4285,
      "mean_token_accuracy": 0.19971583187580108,
      "num_tokens": 72914407.0,
      "step": 37265
    },
    {
      "entropy": 5.939534759521484,
      "epoch": 3.725546058879392,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003667895337585819,
      "loss": 5.3952,
      "mean_token_accuracy": 0.1969001144170761,
      "num_tokens": 72924459.0,
      "step": 37270
    },
    {
      "entropy": 5.7882781505584715,
      "epoch": 3.7260458839406208,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036675694835736875,
      "loss": 5.3284,
      "mean_token_accuracy": 0.20431261658668518,
      "num_tokens": 72934783.0,
      "step": 37275
    },
    {
      "entropy": 5.885294818878174,
      "epoch": 3.7265457090018494,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003667243606475744,
      "loss": 5.3901,
      "mean_token_accuracy": 0.20574051141738892,
      "num_tokens": 72943615.0,
      "step": 37280
    },
    {
      "entropy": 5.855170202255249,
      "epoch": 3.7270455340630777,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00036669177063001866,
      "loss": 5.3331,
      "mean_token_accuracy": 0.20326642394065858,
      "num_tokens": 72954464.0,
      "step": 37285
    },
    {
      "entropy": 5.811943578720093,
      "epoch": 3.7275453591243064,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003666591783055216,
      "loss": 5.3054,
      "mean_token_accuracy": 0.2134775772690773,
      "num_tokens": 72964680.0,
      "step": 37290
    },
    {
      "entropy": 5.898453617095948,
      "epoch": 3.728045184185535,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003666265836749033,
      "loss": 5.3953,
      "mean_token_accuracy": 0.20383431911468505,
      "num_tokens": 72975606.0,
      "step": 37295
    },
    {
      "entropy": 5.875379467010498,
      "epoch": 3.728545009246764,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036659398673898356,
      "loss": 5.359,
      "mean_token_accuracy": 0.20149216055870056,
      "num_tokens": 72985752.0,
      "step": 37300
    },
    {
      "entropy": 5.909455347061157,
      "epoch": 3.729044834307992,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036656138749858287,
      "loss": 5.3816,
      "mean_token_accuracy": 0.20096443444490433,
      "num_tokens": 72995754.0,
      "step": 37305
    },
    {
      "entropy": 5.841073274612427,
      "epoch": 3.7295446593692207,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036652878595452105,
      "loss": 5.3403,
      "mean_token_accuracy": 0.20346746146678923,
      "num_tokens": 73004420.0,
      "step": 37310
    },
    {
      "entropy": 5.782222938537598,
      "epoch": 3.7300444844304494,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003664961821076187,
      "loss": 5.2732,
      "mean_token_accuracy": 0.21314328610897065,
      "num_tokens": 73013203.0,
      "step": 37315
    },
    {
      "entropy": 5.735241079330445,
      "epoch": 3.7305443094916777,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036646357595869586,
      "loss": 5.3147,
      "mean_token_accuracy": 0.21917928904294967,
      "num_tokens": 73022605.0,
      "step": 37320
    },
    {
      "entropy": 5.868818044662476,
      "epoch": 3.7310441345529064,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000366430967508573,
      "loss": 5.3112,
      "mean_token_accuracy": 0.20762195736169814,
      "num_tokens": 73031589.0,
      "step": 37325
    },
    {
      "entropy": 5.833606958389282,
      "epoch": 3.731543959614135,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003663983567580705,
      "loss": 5.3293,
      "mean_token_accuracy": 0.20738616734743118,
      "num_tokens": 73041694.0,
      "step": 37330
    },
    {
      "entropy": 5.90061321258545,
      "epoch": 3.732043784675364,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036636574370800887,
      "loss": 5.371,
      "mean_token_accuracy": 0.2012190505862236,
      "num_tokens": 73051192.0,
      "step": 37335
    },
    {
      "entropy": 5.793260145187378,
      "epoch": 3.732543609736592,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003663331283592086,
      "loss": 5.3112,
      "mean_token_accuracy": 0.21246671080589294,
      "num_tokens": 73061275.0,
      "step": 37340
    },
    {
      "entropy": 5.932810640335083,
      "epoch": 3.7330434347978207,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036630051071249043,
      "loss": 5.4485,
      "mean_token_accuracy": 0.1998513460159302,
      "num_tokens": 73072416.0,
      "step": 37345
    },
    {
      "entropy": 5.886567449569702,
      "epoch": 3.7335432598590494,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036626789076867484,
      "loss": 5.4067,
      "mean_token_accuracy": 0.19459582716226578,
      "num_tokens": 73082783.0,
      "step": 37350
    },
    {
      "entropy": 5.8989174365997314,
      "epoch": 3.7340430849202777,
      "grad_norm": 1.125,
      "learning_rate": 0.0003662352685285826,
      "loss": 5.3684,
      "mean_token_accuracy": 0.19546374678611755,
      "num_tokens": 73091748.0,
      "step": 37355
    },
    {
      "entropy": 5.853067302703858,
      "epoch": 3.7345429099815064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003662026439930345,
      "loss": 5.3219,
      "mean_token_accuracy": 0.20381840616464614,
      "num_tokens": 73101777.0,
      "step": 37360
    },
    {
      "entropy": 5.859487533569336,
      "epoch": 3.735042735042735,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036617001716285135,
      "loss": 5.3534,
      "mean_token_accuracy": 0.20768734514713288,
      "num_tokens": 73111602.0,
      "step": 37365
    },
    {
      "entropy": 5.802224349975586,
      "epoch": 3.735542560103964,
      "grad_norm": 1.140625,
      "learning_rate": 0.000366137388038854,
      "loss": 5.332,
      "mean_token_accuracy": 0.20297527313232422,
      "num_tokens": 73120905.0,
      "step": 37370
    },
    {
      "entropy": 5.798506736755371,
      "epoch": 3.736042385165192,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036610475662186343,
      "loss": 5.3367,
      "mean_token_accuracy": 0.20980178862810134,
      "num_tokens": 73130848.0,
      "step": 37375
    },
    {
      "entropy": 5.953382062911987,
      "epoch": 3.7365422102264207,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003660721229127006,
      "loss": 5.4367,
      "mean_token_accuracy": 0.19363584369421005,
      "num_tokens": 73139952.0,
      "step": 37380
    },
    {
      "entropy": 5.879045867919922,
      "epoch": 3.7370420352876494,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036603948691218665,
      "loss": 5.3062,
      "mean_token_accuracy": 0.20900842249393464,
      "num_tokens": 73148532.0,
      "step": 37385
    },
    {
      "entropy": 5.768174076080323,
      "epoch": 3.7375418603488777,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003660068486211426,
      "loss": 5.2425,
      "mean_token_accuracy": 0.21077940464019776,
      "num_tokens": 73158650.0,
      "step": 37390
    },
    {
      "entropy": 5.80807466506958,
      "epoch": 3.7380416854101064,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003659742080403897,
      "loss": 5.3189,
      "mean_token_accuracy": 0.20723376125097276,
      "num_tokens": 73169118.0,
      "step": 37395
    },
    {
      "entropy": 5.82758321762085,
      "epoch": 3.738541510471335,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003659415651707491,
      "loss": 5.3043,
      "mean_token_accuracy": 0.21061142832040786,
      "num_tokens": 73179307.0,
      "step": 37400
    },
    {
      "entropy": 5.9304423332214355,
      "epoch": 3.739041335532564,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003659089200130421,
      "loss": 5.4295,
      "mean_token_accuracy": 0.19400041401386262,
      "num_tokens": 73189310.0,
      "step": 37405
    },
    {
      "entropy": 5.886432218551636,
      "epoch": 3.739541160593792,
      "grad_norm": 1.125,
      "learning_rate": 0.00036587627256809,
      "loss": 5.3338,
      "mean_token_accuracy": 0.2067945957183838,
      "num_tokens": 73198650.0,
      "step": 37410
    },
    {
      "entropy": 5.805478239059449,
      "epoch": 3.7400409856550207,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003658436228367144,
      "loss": 5.3211,
      "mean_token_accuracy": 0.21543078869581223,
      "num_tokens": 73207869.0,
      "step": 37415
    },
    {
      "entropy": 5.788455867767334,
      "epoch": 3.7405408107162494,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036581097081973655,
      "loss": 5.3576,
      "mean_token_accuracy": 0.20705657005310057,
      "num_tokens": 73218102.0,
      "step": 37420
    },
    {
      "entropy": 5.881678342819214,
      "epoch": 3.7410406357774777,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003657783165179781,
      "loss": 5.4153,
      "mean_token_accuracy": 0.20101394802331923,
      "num_tokens": 73226803.0,
      "step": 37425
    },
    {
      "entropy": 5.935468101501465,
      "epoch": 3.7415404608387064,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003657456599322606,
      "loss": 5.3711,
      "mean_token_accuracy": 0.2033334940671921,
      "num_tokens": 73235826.0,
      "step": 37430
    },
    {
      "entropy": 5.831682443618774,
      "epoch": 3.742040285899935,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036571300106340553,
      "loss": 5.3273,
      "mean_token_accuracy": 0.20543310791254044,
      "num_tokens": 73245829.0,
      "step": 37435
    },
    {
      "entropy": 5.709570646286011,
      "epoch": 3.742540110961164,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036568033991223476,
      "loss": 5.1285,
      "mean_token_accuracy": 0.22454827278852463,
      "num_tokens": 73255886.0,
      "step": 37440
    },
    {
      "entropy": 5.889088344573975,
      "epoch": 3.743039936022392,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036564767647956993,
      "loss": 5.346,
      "mean_token_accuracy": 0.2079748347401619,
      "num_tokens": 73266188.0,
      "step": 37445
    },
    {
      "entropy": 5.919516277313233,
      "epoch": 3.7435397610836207,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003656150107662329,
      "loss": 5.5148,
      "mean_token_accuracy": 0.19941664040088652,
      "num_tokens": 73275993.0,
      "step": 37450
    },
    {
      "entropy": 5.90513424873352,
      "epoch": 3.7440395861448494,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036558234277304555,
      "loss": 5.3502,
      "mean_token_accuracy": 0.20625988245010377,
      "num_tokens": 73286146.0,
      "step": 37455
    },
    {
      "entropy": 5.96565351486206,
      "epoch": 3.7445394112060777,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036554967250082977,
      "loss": 5.4543,
      "mean_token_accuracy": 0.20023766309022903,
      "num_tokens": 73295390.0,
      "step": 37460
    },
    {
      "entropy": 5.924589061737061,
      "epoch": 3.7450392362673064,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036551699995040755,
      "loss": 5.3577,
      "mean_token_accuracy": 0.2057983547449112,
      "num_tokens": 73305611.0,
      "step": 37465
    },
    {
      "entropy": 5.835120344161988,
      "epoch": 3.745539061328535,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036548432512260084,
      "loss": 5.3322,
      "mean_token_accuracy": 0.19496272951364518,
      "num_tokens": 73315644.0,
      "step": 37470
    },
    {
      "entropy": 5.83770718574524,
      "epoch": 3.746038886389764,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003654516480182318,
      "loss": 5.3996,
      "mean_token_accuracy": 0.20645063370466232,
      "num_tokens": 73324584.0,
      "step": 37475
    },
    {
      "entropy": 5.8568037986755375,
      "epoch": 3.746538711450992,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0003654189686381225,
      "loss": 5.3696,
      "mean_token_accuracy": 0.2061959460377693,
      "num_tokens": 73334727.0,
      "step": 37480
    },
    {
      "entropy": 5.873660659790039,
      "epoch": 3.7470385365122207,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003653862869830953,
      "loss": 5.2772,
      "mean_token_accuracy": 0.21123117953538895,
      "num_tokens": 73344249.0,
      "step": 37485
    },
    {
      "entropy": 5.8497748374938965,
      "epoch": 3.7475383615734494,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003653536030539725,
      "loss": 5.2653,
      "mean_token_accuracy": 0.2106648713350296,
      "num_tokens": 73353799.0,
      "step": 37490
    },
    {
      "entropy": 5.848962640762329,
      "epoch": 3.7480381866346777,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036532091685157614,
      "loss": 5.3074,
      "mean_token_accuracy": 0.21201067119836808,
      "num_tokens": 73363067.0,
      "step": 37495
    },
    {
      "entropy": 5.855996322631836,
      "epoch": 3.7485380116959064,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036528822837672884,
      "loss": 5.3952,
      "mean_token_accuracy": 0.19444706290960312,
      "num_tokens": 73373003.0,
      "step": 37500
    },
    {
      "entropy": 5.80603723526001,
      "epoch": 3.749037836757135,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003652555376302528,
      "loss": 5.2952,
      "mean_token_accuracy": 0.21172418296337128,
      "num_tokens": 73383486.0,
      "step": 37505
    },
    {
      "entropy": 5.810166883468628,
      "epoch": 3.749537661818364,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036522284461297077,
      "loss": 5.3253,
      "mean_token_accuracy": 0.208937369287014,
      "num_tokens": 73393552.0,
      "step": 37510
    },
    {
      "entropy": 5.772433423995972,
      "epoch": 3.750037486879592,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036519014932570517,
      "loss": 5.2986,
      "mean_token_accuracy": 0.21246783435344696,
      "num_tokens": 73402953.0,
      "step": 37515
    },
    {
      "entropy": 5.83777813911438,
      "epoch": 3.7505373119408207,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036515745176927863,
      "loss": 5.4031,
      "mean_token_accuracy": 0.2013163298368454,
      "num_tokens": 73412811.0,
      "step": 37520
    },
    {
      "entropy": 6.019983577728271,
      "epoch": 3.7510371370020494,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003651247519445138,
      "loss": 5.5356,
      "mean_token_accuracy": 0.18827436119318008,
      "num_tokens": 73422217.0,
      "step": 37525
    },
    {
      "entropy": 5.919079303741455,
      "epoch": 3.7515369620632777,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036509204985223336,
      "loss": 5.3745,
      "mean_token_accuracy": 0.20227667689323425,
      "num_tokens": 73431176.0,
      "step": 37530
    },
    {
      "entropy": 5.809819221496582,
      "epoch": 3.7520367871245064,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003650593454932602,
      "loss": 5.2794,
      "mean_token_accuracy": 0.20798050463199616,
      "num_tokens": 73440536.0,
      "step": 37535
    },
    {
      "entropy": 5.782922315597534,
      "epoch": 3.752536612185735,
      "grad_norm": 1.15625,
      "learning_rate": 0.000365026638868417,
      "loss": 5.3721,
      "mean_token_accuracy": 0.20629789531230927,
      "num_tokens": 73449738.0,
      "step": 37540
    },
    {
      "entropy": 5.842756509780884,
      "epoch": 3.753036437246964,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036499392997852684,
      "loss": 5.3681,
      "mean_token_accuracy": 0.19826143532991408,
      "num_tokens": 73460036.0,
      "step": 37545
    },
    {
      "entropy": 5.9078741550445555,
      "epoch": 3.753536262308192,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003649612188244125,
      "loss": 5.4044,
      "mean_token_accuracy": 0.20312134474515914,
      "num_tokens": 73470331.0,
      "step": 37550
    },
    {
      "entropy": 5.829819059371948,
      "epoch": 3.7540360873694207,
      "grad_norm": 1.296875,
      "learning_rate": 0.000364928505406897,
      "loss": 5.3038,
      "mean_token_accuracy": 0.2058510348200798,
      "num_tokens": 73480530.0,
      "step": 37555
    },
    {
      "entropy": 5.813010358810425,
      "epoch": 3.7545359124306494,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003648957897268035,
      "loss": 5.4032,
      "mean_token_accuracy": 0.20223939418792725,
      "num_tokens": 73489432.0,
      "step": 37560
    },
    {
      "entropy": 5.917829513549805,
      "epoch": 3.7550357374918777,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003648630717849551,
      "loss": 5.4265,
      "mean_token_accuracy": 0.19517885744571686,
      "num_tokens": 73499176.0,
      "step": 37565
    },
    {
      "entropy": 5.943732595443725,
      "epoch": 3.7555355625531064,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036483035158217485,
      "loss": 5.4878,
      "mean_token_accuracy": 0.20098641961812974,
      "num_tokens": 73509356.0,
      "step": 37570
    },
    {
      "entropy": 5.855845594406128,
      "epoch": 3.756035387614335,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036479762911928606,
      "loss": 5.3583,
      "mean_token_accuracy": 0.21228077709674836,
      "num_tokens": 73521589.0,
      "step": 37575
    },
    {
      "entropy": 5.84791374206543,
      "epoch": 3.756535212675564,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036476490439711206,
      "loss": 5.3297,
      "mean_token_accuracy": 0.2030869424343109,
      "num_tokens": 73530159.0,
      "step": 37580
    },
    {
      "entropy": 5.688095331192017,
      "epoch": 3.757035037736792,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003647321774164762,
      "loss": 5.1622,
      "mean_token_accuracy": 0.22068938314914704,
      "num_tokens": 73541524.0,
      "step": 37585
    },
    {
      "entropy": 5.860791778564453,
      "epoch": 3.7575348627980207,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003646994481782018,
      "loss": 5.4052,
      "mean_token_accuracy": 0.19959065467119216,
      "num_tokens": 73552806.0,
      "step": 37590
    },
    {
      "entropy": 5.971860265731811,
      "epoch": 3.758034687859249,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003646667166831124,
      "loss": 5.4609,
      "mean_token_accuracy": 0.19752082228660583,
      "num_tokens": 73562619.0,
      "step": 37595
    },
    {
      "entropy": 5.877070569992066,
      "epoch": 3.7585345129204777,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003646339829320315,
      "loss": 5.361,
      "mean_token_accuracy": 0.19768978357315065,
      "num_tokens": 73573111.0,
      "step": 37600
    },
    {
      "entropy": 5.835542106628418,
      "epoch": 3.7590343379817064,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003646012469257826,
      "loss": 5.3352,
      "mean_token_accuracy": 0.21059089452028273,
      "num_tokens": 73582023.0,
      "step": 37605
    },
    {
      "entropy": 5.814394474029541,
      "epoch": 3.759534163042935,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003645685086651894,
      "loss": 5.2971,
      "mean_token_accuracy": 0.20721505433321,
      "num_tokens": 73592112.0,
      "step": 37610
    },
    {
      "entropy": 5.870262622833252,
      "epoch": 3.760033988104164,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003645357681510756,
      "loss": 5.3919,
      "mean_token_accuracy": 0.20678738355636597,
      "num_tokens": 73601462.0,
      "step": 37615
    },
    {
      "entropy": 5.782518434524536,
      "epoch": 3.760533813165392,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003645030253842649,
      "loss": 5.3339,
      "mean_token_accuracy": 0.20721350312232972,
      "num_tokens": 73611335.0,
      "step": 37620
    },
    {
      "entropy": 5.797211122512818,
      "epoch": 3.7610336382266207,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036447028036558105,
      "loss": 5.3091,
      "mean_token_accuracy": 0.2130969062447548,
      "num_tokens": 73621892.0,
      "step": 37625
    },
    {
      "entropy": 5.971917009353637,
      "epoch": 3.761533463287849,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000364437533095848,
      "loss": 5.4358,
      "mean_token_accuracy": 0.1909891441464424,
      "num_tokens": 73631914.0,
      "step": 37630
    },
    {
      "entropy": 5.827470302581787,
      "epoch": 3.7620332883490777,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036440478357588964,
      "loss": 5.3193,
      "mean_token_accuracy": 0.20929816216230393,
      "num_tokens": 73641404.0,
      "step": 37635
    },
    {
      "entropy": 5.858164072036743,
      "epoch": 3.7625331134103064,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036437203180653,
      "loss": 5.3134,
      "mean_token_accuracy": 0.21206181943416597,
      "num_tokens": 73650307.0,
      "step": 37640
    },
    {
      "entropy": 5.894265985488891,
      "epoch": 3.763032938471535,
      "grad_norm": 1.09375,
      "learning_rate": 0.000364339277788593,
      "loss": 5.4129,
      "mean_token_accuracy": 0.2006075069308281,
      "num_tokens": 73659498.0,
      "step": 37645
    },
    {
      "entropy": 5.92004714012146,
      "epoch": 3.763532763532764,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003643065215229027,
      "loss": 5.4129,
      "mean_token_accuracy": 0.20199792385101317,
      "num_tokens": 73669694.0,
      "step": 37650
    },
    {
      "entropy": 5.7602368831634525,
      "epoch": 3.764032588593992,
      "grad_norm": 1.421875,
      "learning_rate": 0.00036427376301028335,
      "loss": 5.2295,
      "mean_token_accuracy": 0.22559453248977662,
      "num_tokens": 73680080.0,
      "step": 37655
    },
    {
      "entropy": 5.874858665466308,
      "epoch": 3.7645324136552207,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003642410022515591,
      "loss": 5.4309,
      "mean_token_accuracy": 0.20008404850959777,
      "num_tokens": 73690135.0,
      "step": 37660
    },
    {
      "entropy": 5.816574192047119,
      "epoch": 3.765032238716449,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003642082392475541,
      "loss": 5.3557,
      "mean_token_accuracy": 0.21078944355249404,
      "num_tokens": 73700125.0,
      "step": 37665
    },
    {
      "entropy": 5.937102651596069,
      "epoch": 3.7655320637776777,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036417547399909285,
      "loss": 5.4275,
      "mean_token_accuracy": 0.20013104230165482,
      "num_tokens": 73710352.0,
      "step": 37670
    },
    {
      "entropy": 5.897189426422119,
      "epoch": 3.7660318888389064,
      "grad_norm": 1.25,
      "learning_rate": 0.00036414270650699963,
      "loss": 5.4543,
      "mean_token_accuracy": 0.19605837315320968,
      "num_tokens": 73719199.0,
      "step": 37675
    },
    {
      "entropy": 5.814518880844116,
      "epoch": 3.766531713900135,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036410993677209885,
      "loss": 5.2913,
      "mean_token_accuracy": 0.20979111790657043,
      "num_tokens": 73727277.0,
      "step": 37680
    },
    {
      "entropy": 5.917181158065796,
      "epoch": 3.767031538961364,
      "grad_norm": 1.046875,
      "learning_rate": 0.000364077164795215,
      "loss": 5.4332,
      "mean_token_accuracy": 0.19848672747612,
      "num_tokens": 73738074.0,
      "step": 37685
    },
    {
      "entropy": 5.913081216812134,
      "epoch": 3.767531364022592,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003640443905771726,
      "loss": 5.4452,
      "mean_token_accuracy": 0.19297428727149962,
      "num_tokens": 73747146.0,
      "step": 37690
    },
    {
      "entropy": 5.829551887512207,
      "epoch": 3.7680311890838207,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036401161411879615,
      "loss": 5.4288,
      "mean_token_accuracy": 0.2062856897711754,
      "num_tokens": 73757309.0,
      "step": 37695
    },
    {
      "entropy": 5.86508526802063,
      "epoch": 3.768531014145049,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036397883542091045,
      "loss": 5.3492,
      "mean_token_accuracy": 0.2022587016224861,
      "num_tokens": 73767062.0,
      "step": 37700
    },
    {
      "entropy": 5.818364381790161,
      "epoch": 3.7690308392062777,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036394605448434014,
      "loss": 5.2612,
      "mean_token_accuracy": 0.21749309152364732,
      "num_tokens": 73775792.0,
      "step": 37705
    },
    {
      "entropy": 5.87923789024353,
      "epoch": 3.7695306642675064,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003639132713099101,
      "loss": 5.3564,
      "mean_token_accuracy": 0.20625565499067305,
      "num_tokens": 73785770.0,
      "step": 37710
    },
    {
      "entropy": 5.859615850448608,
      "epoch": 3.770030489328735,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003638804858984449,
      "loss": 5.2767,
      "mean_token_accuracy": 0.21692000329494476,
      "num_tokens": 73795376.0,
      "step": 37715
    },
    {
      "entropy": 5.864987373352051,
      "epoch": 3.7705303143899638,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036384769825076964,
      "loss": 5.3129,
      "mean_token_accuracy": 0.20980406254529954,
      "num_tokens": 73805442.0,
      "step": 37720
    },
    {
      "entropy": 5.918309307098388,
      "epoch": 3.771030139451192,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000363814908367709,
      "loss": 5.3841,
      "mean_token_accuracy": 0.20037193298339845,
      "num_tokens": 73815425.0,
      "step": 37725
    },
    {
      "entropy": 5.893709278106689,
      "epoch": 3.7715299645124207,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036378211625008827,
      "loss": 5.3851,
      "mean_token_accuracy": 0.20069019943475724,
      "num_tokens": 73823901.0,
      "step": 37730
    },
    {
      "entropy": 5.926753234863281,
      "epoch": 3.772029789573649,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003637493218987323,
      "loss": 5.5132,
      "mean_token_accuracy": 0.20005273073911667,
      "num_tokens": 73833659.0,
      "step": 37735
    },
    {
      "entropy": 5.834343099594117,
      "epoch": 3.7725296146348777,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003637165253144662,
      "loss": 5.3112,
      "mean_token_accuracy": 0.21466686725616455,
      "num_tokens": 73844014.0,
      "step": 37740
    },
    {
      "entropy": 5.874452781677246,
      "epoch": 3.7730294396961064,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003636837264981152,
      "loss": 5.3729,
      "mean_token_accuracy": 0.20246467143297195,
      "num_tokens": 73853449.0,
      "step": 37745
    },
    {
      "entropy": 5.878138256072998,
      "epoch": 3.773529264757335,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003636509254505045,
      "loss": 5.3147,
      "mean_token_accuracy": 0.2071457639336586,
      "num_tokens": 73862322.0,
      "step": 37750
    },
    {
      "entropy": 5.794955825805664,
      "epoch": 3.7740290898185638,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003636181221724592,
      "loss": 5.3199,
      "mean_token_accuracy": 0.20649849623441696,
      "num_tokens": 73871846.0,
      "step": 37755
    },
    {
      "entropy": 5.847482442855835,
      "epoch": 3.774528914879792,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036358531666480477,
      "loss": 5.3348,
      "mean_token_accuracy": 0.21316068023443221,
      "num_tokens": 73880393.0,
      "step": 37760
    },
    {
      "entropy": 5.839756107330322,
      "epoch": 3.7750287399410207,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036355250892836663,
      "loss": 5.2975,
      "mean_token_accuracy": 0.20483437776565552,
      "num_tokens": 73889786.0,
      "step": 37765
    },
    {
      "entropy": 5.843078851699829,
      "epoch": 3.775528565002249,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036351969896397006,
      "loss": 5.3548,
      "mean_token_accuracy": 0.20636976063251494,
      "num_tokens": 73898716.0,
      "step": 37770
    },
    {
      "entropy": 5.80725507736206,
      "epoch": 3.7760283900634777,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036348688677244074,
      "loss": 5.3633,
      "mean_token_accuracy": 0.20054105222225188,
      "num_tokens": 73909477.0,
      "step": 37775
    },
    {
      "entropy": 5.863715362548828,
      "epoch": 3.7765282151247064,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036345407235460417,
      "loss": 5.406,
      "mean_token_accuracy": 0.20136978328227997,
      "num_tokens": 73919212.0,
      "step": 37780
    },
    {
      "entropy": 5.912468957901001,
      "epoch": 3.777028040185935,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036342125571128575,
      "loss": 5.3514,
      "mean_token_accuracy": 0.20105773508548735,
      "num_tokens": 73929443.0,
      "step": 37785
    },
    {
      "entropy": 5.851428699493408,
      "epoch": 3.7775278652471633,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036338843684331133,
      "loss": 5.2973,
      "mean_token_accuracy": 0.21497252583503723,
      "num_tokens": 73939431.0,
      "step": 37790
    },
    {
      "entropy": 5.817389822006225,
      "epoch": 3.778027690308392,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036335561575150663,
      "loss": 5.3066,
      "mean_token_accuracy": 0.20813185423612596,
      "num_tokens": 73948880.0,
      "step": 37795
    },
    {
      "entropy": 5.852166318893433,
      "epoch": 3.7785275153696207,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036332279243669733,
      "loss": 5.3431,
      "mean_token_accuracy": 0.2037034124135971,
      "num_tokens": 73958466.0,
      "step": 37800
    },
    {
      "entropy": 5.769900321960449,
      "epoch": 3.779027340430849,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036328996689970937,
      "loss": 5.3287,
      "mean_token_accuracy": 0.2119239553809166,
      "num_tokens": 73967936.0,
      "step": 37805
    },
    {
      "entropy": 5.897533559799195,
      "epoch": 3.7795271654920777,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003632571391413685,
      "loss": 5.4231,
      "mean_token_accuracy": 0.19824993312358857,
      "num_tokens": 73976641.0,
      "step": 37810
    },
    {
      "entropy": 5.906442022323608,
      "epoch": 3.7800269905533064,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036322430916250077,
      "loss": 5.3937,
      "mean_token_accuracy": 0.2048419162631035,
      "num_tokens": 73986360.0,
      "step": 37815
    },
    {
      "entropy": 5.934755849838257,
      "epoch": 3.780526815614535,
      "grad_norm": 1.25,
      "learning_rate": 0.00036319147696393207,
      "loss": 5.3491,
      "mean_token_accuracy": 0.20569102913141252,
      "num_tokens": 73995413.0,
      "step": 37820
    },
    {
      "entropy": 5.96516604423523,
      "epoch": 3.7810266406757633,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003631586425464885,
      "loss": 5.5087,
      "mean_token_accuracy": 0.19098259508609772,
      "num_tokens": 74005773.0,
      "step": 37825
    },
    {
      "entropy": 5.848883628845215,
      "epoch": 3.781526465736992,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036312580591099625,
      "loss": 5.3437,
      "mean_token_accuracy": 0.20204757302999496,
      "num_tokens": 74016308.0,
      "step": 37830
    },
    {
      "entropy": 5.871147012710571,
      "epoch": 3.7820262907982207,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003630929670582813,
      "loss": 5.3308,
      "mean_token_accuracy": 0.20760872215032578,
      "num_tokens": 74026895.0,
      "step": 37835
    },
    {
      "entropy": 5.7706578254699705,
      "epoch": 3.782526115859449,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036306012598917,
      "loss": 5.2963,
      "mean_token_accuracy": 0.21753647476434707,
      "num_tokens": 74036247.0,
      "step": 37840
    },
    {
      "entropy": 5.8404406070709225,
      "epoch": 3.7830259409206777,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003630272827044886,
      "loss": 5.3269,
      "mean_token_accuracy": 0.20905770510435104,
      "num_tokens": 74046061.0,
      "step": 37845
    },
    {
      "entropy": 5.8703690528869625,
      "epoch": 3.7835257659819064,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036299443720506344,
      "loss": 5.264,
      "mean_token_accuracy": 0.21321062743663788,
      "num_tokens": 74055023.0,
      "step": 37850
    },
    {
      "entropy": 5.836681127548218,
      "epoch": 3.784025591043135,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036296158949172087,
      "loss": 5.3353,
      "mean_token_accuracy": 0.2023915708065033,
      "num_tokens": 74064647.0,
      "step": 37855
    },
    {
      "entropy": 5.85214900970459,
      "epoch": 3.7845254161043633,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003629287395652873,
      "loss": 5.3774,
      "mean_token_accuracy": 0.2011621356010437,
      "num_tokens": 74074452.0,
      "step": 37860
    },
    {
      "entropy": 5.899777507781982,
      "epoch": 3.785025241165592,
      "grad_norm": 1.25,
      "learning_rate": 0.0003628958874265893,
      "loss": 5.3634,
      "mean_token_accuracy": 0.2063872292637825,
      "num_tokens": 74083606.0,
      "step": 37865
    },
    {
      "entropy": 5.907136011123657,
      "epoch": 3.7855250662268207,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003628630330764534,
      "loss": 5.4457,
      "mean_token_accuracy": 0.20014407038688659,
      "num_tokens": 74094720.0,
      "step": 37870
    },
    {
      "entropy": 5.851425409317017,
      "epoch": 3.786024891288049,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003628301765157062,
      "loss": 5.2606,
      "mean_token_accuracy": 0.20794883519411086,
      "num_tokens": 74104078.0,
      "step": 37875
    },
    {
      "entropy": 5.8718445777893065,
      "epoch": 3.7865247163492777,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036279731774517435,
      "loss": 5.3595,
      "mean_token_accuracy": 0.20290151685476304,
      "num_tokens": 74114139.0,
      "step": 37880
    },
    {
      "entropy": 5.905762720108032,
      "epoch": 3.7870245414105064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036276445676568453,
      "loss": 5.3586,
      "mean_token_accuracy": 0.20640451312065125,
      "num_tokens": 74123995.0,
      "step": 37885
    },
    {
      "entropy": 5.85124454498291,
      "epoch": 3.787524366471735,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036273159357806363,
      "loss": 5.3371,
      "mean_token_accuracy": 0.20890521854162217,
      "num_tokens": 74133571.0,
      "step": 37890
    },
    {
      "entropy": 5.799184036254883,
      "epoch": 3.7880241915329633,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036269872818313836,
      "loss": 5.3626,
      "mean_token_accuracy": 0.20842682868242263,
      "num_tokens": 74143323.0,
      "step": 37895
    },
    {
      "entropy": 5.8120239734649655,
      "epoch": 3.788524016594192,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003626658605817357,
      "loss": 5.2834,
      "mean_token_accuracy": 0.21403988301753998,
      "num_tokens": 74152791.0,
      "step": 37900
    },
    {
      "entropy": 5.858112144470215,
      "epoch": 3.7890238416554207,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036263299077468256,
      "loss": 5.2631,
      "mean_token_accuracy": 0.21559022068977357,
      "num_tokens": 74162399.0,
      "step": 37905
    },
    {
      "entropy": 5.832856559753418,
      "epoch": 3.789523666716649,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003626001187628059,
      "loss": 5.3599,
      "mean_token_accuracy": 0.20824547857046127,
      "num_tokens": 74172358.0,
      "step": 37910
    },
    {
      "entropy": 5.798845529556274,
      "epoch": 3.7900234917778777,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003625672445469327,
      "loss": 5.3181,
      "mean_token_accuracy": 0.20903094708919526,
      "num_tokens": 74182626.0,
      "step": 37915
    },
    {
      "entropy": 5.837854099273682,
      "epoch": 3.7905233168391064,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036253436812789034,
      "loss": 5.3296,
      "mean_token_accuracy": 0.2058714896440506,
      "num_tokens": 74192994.0,
      "step": 37920
    },
    {
      "entropy": 5.799303483963013,
      "epoch": 3.791023141900335,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003625014895065057,
      "loss": 5.2839,
      "mean_token_accuracy": 0.21343710869550706,
      "num_tokens": 74202947.0,
      "step": 37925
    },
    {
      "entropy": 5.874754953384399,
      "epoch": 3.7915229669615633,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003624686086836061,
      "loss": 5.4049,
      "mean_token_accuracy": 0.20398338735103608,
      "num_tokens": 74212831.0,
      "step": 37930
    },
    {
      "entropy": 5.938939905166626,
      "epoch": 3.792022792022792,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003624357256600188,
      "loss": 5.4513,
      "mean_token_accuracy": 0.20412010550498963,
      "num_tokens": 74222333.0,
      "step": 37935
    },
    {
      "entropy": 5.875602102279663,
      "epoch": 3.7925226170840207,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003624028404365712,
      "loss": 5.3347,
      "mean_token_accuracy": 0.2107146978378296,
      "num_tokens": 74233148.0,
      "step": 37940
    },
    {
      "entropy": 5.862709665298462,
      "epoch": 3.793022442145249,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003623699530140906,
      "loss": 5.326,
      "mean_token_accuracy": 0.20323613435029983,
      "num_tokens": 74242191.0,
      "step": 37945
    },
    {
      "entropy": 5.901008224487304,
      "epoch": 3.7935222672064777,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003623370633934043,
      "loss": 5.3618,
      "mean_token_accuracy": 0.2008807525038719,
      "num_tokens": 74253006.0,
      "step": 37950
    },
    {
      "entropy": 5.868520307540893,
      "epoch": 3.7940220922677064,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003623041715753402,
      "loss": 5.3299,
      "mean_token_accuracy": 0.20358492732048034,
      "num_tokens": 74263596.0,
      "step": 37955
    },
    {
      "entropy": 5.909247636795044,
      "epoch": 3.794521917328935,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036227127756072554,
      "loss": 5.5087,
      "mean_token_accuracy": 0.19589273929595946,
      "num_tokens": 74274929.0,
      "step": 37960
    },
    {
      "entropy": 5.831806898117065,
      "epoch": 3.7950217423901633,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003622383813503879,
      "loss": 5.295,
      "mean_token_accuracy": 0.20419980436563492,
      "num_tokens": 74285184.0,
      "step": 37965
    },
    {
      "entropy": 5.9088458061218265,
      "epoch": 3.795521567451392,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036220548294515516,
      "loss": 5.4232,
      "mean_token_accuracy": 0.1996854141354561,
      "num_tokens": 74295843.0,
      "step": 37970
    },
    {
      "entropy": 5.973358774185181,
      "epoch": 3.7960213925126207,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003621725823458549,
      "loss": 5.4116,
      "mean_token_accuracy": 0.19735279828310012,
      "num_tokens": 74305358.0,
      "step": 37975
    },
    {
      "entropy": 5.874614429473877,
      "epoch": 3.796521217573849,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036213967955331475,
      "loss": 5.342,
      "mean_token_accuracy": 0.19742899686098098,
      "num_tokens": 74313874.0,
      "step": 37980
    },
    {
      "entropy": 5.82993631362915,
      "epoch": 3.7970210426350777,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036210677456836285,
      "loss": 5.3508,
      "mean_token_accuracy": 0.20558635294437408,
      "num_tokens": 74323178.0,
      "step": 37985
    },
    {
      "entropy": 5.856752824783325,
      "epoch": 3.7975208676963064,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036207386739182685,
      "loss": 5.2967,
      "mean_token_accuracy": 0.20875464826822282,
      "num_tokens": 74331990.0,
      "step": 37990
    },
    {
      "entropy": 5.9417778015136715,
      "epoch": 3.798020692757535,
      "grad_norm": 1.125,
      "learning_rate": 0.0003620409580245348,
      "loss": 5.4257,
      "mean_token_accuracy": 0.19966184049844743,
      "num_tokens": 74341506.0,
      "step": 37995
    },
    {
      "entropy": 5.821560668945312,
      "epoch": 3.7985205178187633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036200804646731457,
      "loss": 5.3508,
      "mean_token_accuracy": 0.20712739676237107,
      "num_tokens": 74350829.0,
      "step": 38000
    },
    {
      "entropy": 5.911390066146851,
      "epoch": 3.799020342879992,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003619751327209943,
      "loss": 5.4192,
      "mean_token_accuracy": 0.19536685198545456,
      "num_tokens": 74360943.0,
      "step": 38005
    },
    {
      "entropy": 5.9246326923370365,
      "epoch": 3.7995201679412207,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036194221678640206,
      "loss": 5.4574,
      "mean_token_accuracy": 0.19793243557214737,
      "num_tokens": 74372539.0,
      "step": 38010
    },
    {
      "entropy": 5.935094118118286,
      "epoch": 3.800019993002449,
      "grad_norm": 1.28125,
      "learning_rate": 0.000361909298664366,
      "loss": 5.4365,
      "mean_token_accuracy": 0.19706441909074784,
      "num_tokens": 74381668.0,
      "step": 38015
    },
    {
      "entropy": 5.9307599544525145,
      "epoch": 3.8005198180636777,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003618763783557144,
      "loss": 5.4166,
      "mean_token_accuracy": 0.20051510781049728,
      "num_tokens": 74391592.0,
      "step": 38020
    },
    {
      "entropy": 5.828357172012329,
      "epoch": 3.8010196431249064,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003618434558612755,
      "loss": 5.3056,
      "mean_token_accuracy": 0.20391590297222137,
      "num_tokens": 74402281.0,
      "step": 38025
    },
    {
      "entropy": 5.755401086807251,
      "epoch": 3.801519468186135,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036181053118187757,
      "loss": 5.2017,
      "mean_token_accuracy": 0.2181548833847046,
      "num_tokens": 74412179.0,
      "step": 38030
    },
    {
      "entropy": 5.837663984298706,
      "epoch": 3.8020192932473633,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036177760431834897,
      "loss": 5.2814,
      "mean_token_accuracy": 0.21014491319656373,
      "num_tokens": 74421641.0,
      "step": 38035
    },
    {
      "entropy": 5.827477836608887,
      "epoch": 3.802519118308592,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003617446752715182,
      "loss": 5.3021,
      "mean_token_accuracy": 0.21552543044090272,
      "num_tokens": 74431906.0,
      "step": 38040
    },
    {
      "entropy": 5.978401565551758,
      "epoch": 3.8030189433698207,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036171174404221377,
      "loss": 5.4896,
      "mean_token_accuracy": 0.1956622153520584,
      "num_tokens": 74442035.0,
      "step": 38045
    },
    {
      "entropy": 5.797245931625366,
      "epoch": 3.803518768431049,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036167881063126414,
      "loss": 5.2784,
      "mean_token_accuracy": 0.21537180095911027,
      "num_tokens": 74451350.0,
      "step": 38050
    },
    {
      "entropy": 5.89180006980896,
      "epoch": 3.8040185934922777,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003616458750394979,
      "loss": 5.403,
      "mean_token_accuracy": 0.203033147752285,
      "num_tokens": 74462695.0,
      "step": 38055
    },
    {
      "entropy": 5.851149797439575,
      "epoch": 3.8045184185535064,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003616129372677438,
      "loss": 5.3467,
      "mean_token_accuracy": 0.2051297217607498,
      "num_tokens": 74473433.0,
      "step": 38060
    },
    {
      "entropy": 5.894016122817993,
      "epoch": 3.805018243614735,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003615799973168305,
      "loss": 5.4098,
      "mean_token_accuracy": 0.19398490637540816,
      "num_tokens": 74483663.0,
      "step": 38065
    },
    {
      "entropy": 5.934725618362426,
      "epoch": 3.8055180686759633,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003615470551875867,
      "loss": 5.4945,
      "mean_token_accuracy": 0.1975045308470726,
      "num_tokens": 74493831.0,
      "step": 38070
    },
    {
      "entropy": 5.981758499145508,
      "epoch": 3.806017893737192,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003615141108808412,
      "loss": 5.4575,
      "mean_token_accuracy": 0.20019208937883376,
      "num_tokens": 74503770.0,
      "step": 38075
    },
    {
      "entropy": 5.898763704299927,
      "epoch": 3.8065177187984207,
      "grad_norm": 1.171875,
      "learning_rate": 0.000361481164397423,
      "loss": 5.3105,
      "mean_token_accuracy": 0.20404798686504363,
      "num_tokens": 74512555.0,
      "step": 38080
    },
    {
      "entropy": 5.766821718215942,
      "epoch": 3.807017543859649,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036144821573816095,
      "loss": 5.2685,
      "mean_token_accuracy": 0.21208138912916183,
      "num_tokens": 74521783.0,
      "step": 38085
    },
    {
      "entropy": 5.809554767608643,
      "epoch": 3.8075173689208777,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036141526490388406,
      "loss": 5.2375,
      "mean_token_accuracy": 0.21956825405359268,
      "num_tokens": 74530667.0,
      "step": 38090
    },
    {
      "entropy": 5.873425054550171,
      "epoch": 3.8080171939821064,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036138231189542137,
      "loss": 5.3399,
      "mean_token_accuracy": 0.21377155631780625,
      "num_tokens": 74539873.0,
      "step": 38095
    },
    {
      "entropy": 5.73307991027832,
      "epoch": 3.808517019043335,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003613493567136019,
      "loss": 5.2156,
      "mean_token_accuracy": 0.21504159420728683,
      "num_tokens": 74549701.0,
      "step": 38100
    },
    {
      "entropy": 5.87944393157959,
      "epoch": 3.8090168441045633,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003613163993592548,
      "loss": 5.3779,
      "mean_token_accuracy": 0.20033796280622482,
      "num_tokens": 74558991.0,
      "step": 38105
    },
    {
      "entropy": 5.894588375091553,
      "epoch": 3.809516669165792,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003612834398332093,
      "loss": 5.387,
      "mean_token_accuracy": 0.20411255806684495,
      "num_tokens": 74568907.0,
      "step": 38110
    },
    {
      "entropy": 5.822332048416138,
      "epoch": 3.8100164942270203,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003612504781362947,
      "loss": 5.2984,
      "mean_token_accuracy": 0.21231590807437897,
      "num_tokens": 74579404.0,
      "step": 38115
    },
    {
      "entropy": 5.816324090957641,
      "epoch": 3.810516319288249,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003612175142693402,
      "loss": 5.3102,
      "mean_token_accuracy": 0.21270748674869538,
      "num_tokens": 74588672.0,
      "step": 38120
    },
    {
      "entropy": 5.816343498229981,
      "epoch": 3.8110161443494777,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003611845482331753,
      "loss": 5.3622,
      "mean_token_accuracy": 0.2033114641904831,
      "num_tokens": 74598138.0,
      "step": 38125
    },
    {
      "entropy": 5.908526468276977,
      "epoch": 3.8115159694107064,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036115158002862933,
      "loss": 5.403,
      "mean_token_accuracy": 0.20339566767215728,
      "num_tokens": 74608827.0,
      "step": 38130
    },
    {
      "entropy": 5.859445381164551,
      "epoch": 3.812015794471935,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003611186096565318,
      "loss": 5.2577,
      "mean_token_accuracy": 0.21821385025978088,
      "num_tokens": 74618148.0,
      "step": 38135
    },
    {
      "entropy": 5.841556596755981,
      "epoch": 3.8125156195331633,
      "grad_norm": 1.125,
      "learning_rate": 0.0003610856371177122,
      "loss": 5.2862,
      "mean_token_accuracy": 0.20971702188253402,
      "num_tokens": 74626694.0,
      "step": 38140
    },
    {
      "entropy": 5.857631587982178,
      "epoch": 3.813015444594392,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036105266241300005,
      "loss": 5.3186,
      "mean_token_accuracy": 0.20959365963935853,
      "num_tokens": 74635346.0,
      "step": 38145
    },
    {
      "entropy": 5.818621587753296,
      "epoch": 3.8135152696556203,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003610196855432251,
      "loss": 5.3244,
      "mean_token_accuracy": 0.20862961709499359,
      "num_tokens": 74644678.0,
      "step": 38150
    },
    {
      "entropy": 5.9142327308654785,
      "epoch": 3.814015094716849,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036098670650921706,
      "loss": 5.3301,
      "mean_token_accuracy": 0.20151551961898803,
      "num_tokens": 74654927.0,
      "step": 38155
    },
    {
      "entropy": 5.809011030197143,
      "epoch": 3.8145149197780777,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003609537253118056,
      "loss": 5.3647,
      "mean_token_accuracy": 0.20362904965877532,
      "num_tokens": 74665306.0,
      "step": 38160
    },
    {
      "entropy": 5.906500005722046,
      "epoch": 3.8150147448393064,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003609207419518204,
      "loss": 5.3773,
      "mean_token_accuracy": 0.2005243182182312,
      "num_tokens": 74674994.0,
      "step": 38165
    },
    {
      "entropy": 5.893627548217774,
      "epoch": 3.815514569900535,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003608877564300915,
      "loss": 5.332,
      "mean_token_accuracy": 0.20747236609458924,
      "num_tokens": 74685280.0,
      "step": 38170
    },
    {
      "entropy": 5.858721017837524,
      "epoch": 3.8160143949617633,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003608547687474488,
      "loss": 5.3331,
      "mean_token_accuracy": 0.20392841547727586,
      "num_tokens": 74693918.0,
      "step": 38175
    },
    {
      "entropy": 5.8686918258667,
      "epoch": 3.816514220022992,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036082177890472224,
      "loss": 5.3998,
      "mean_token_accuracy": 0.20537050515413285,
      "num_tokens": 74704435.0,
      "step": 38180
    },
    {
      "entropy": 5.885481071472168,
      "epoch": 3.8170140450842203,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036078878690274177,
      "loss": 5.3965,
      "mean_token_accuracy": 0.20034359395503998,
      "num_tokens": 74714894.0,
      "step": 38185
    },
    {
      "entropy": 5.909766435623169,
      "epoch": 3.817513870145449,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003607557927423374,
      "loss": 5.4114,
      "mean_token_accuracy": 0.20175372064113617,
      "num_tokens": 74725377.0,
      "step": 38190
    },
    {
      "entropy": 5.944500732421875,
      "epoch": 3.8180136952066777,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036072279642433946,
      "loss": 5.3513,
      "mean_token_accuracy": 0.20149630904197693,
      "num_tokens": 74734277.0,
      "step": 38195
    },
    {
      "entropy": 5.924267101287842,
      "epoch": 3.8185135202679064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036068979794957797,
      "loss": 5.4522,
      "mean_token_accuracy": 0.19727457463741302,
      "num_tokens": 74744014.0,
      "step": 38200
    },
    {
      "entropy": 5.877127933502197,
      "epoch": 3.819013345329135,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003606567973188833,
      "loss": 5.3413,
      "mean_token_accuracy": 0.20255641043186187,
      "num_tokens": 74753768.0,
      "step": 38205
    },
    {
      "entropy": 5.898213529586792,
      "epoch": 3.8195131703903633,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036062379453308556,
      "loss": 5.3906,
      "mean_token_accuracy": 0.2021253764629364,
      "num_tokens": 74763257.0,
      "step": 38210
    },
    {
      "entropy": 5.800109148025513,
      "epoch": 3.820012995451592,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036059078959301515,
      "loss": 5.3223,
      "mean_token_accuracy": 0.2082520991563797,
      "num_tokens": 74772401.0,
      "step": 38215
    },
    {
      "entropy": 5.829412508010864,
      "epoch": 3.8205128205128203,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003605577824995026,
      "loss": 5.3081,
      "mean_token_accuracy": 0.21151697188615798,
      "num_tokens": 74782966.0,
      "step": 38220
    },
    {
      "entropy": 5.891796112060547,
      "epoch": 3.821012645574049,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036052477325337824,
      "loss": 5.3894,
      "mean_token_accuracy": 0.20212597846984864,
      "num_tokens": 74792434.0,
      "step": 38225
    },
    {
      "entropy": 5.897971153259277,
      "epoch": 3.8215124706352777,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036049176185547254,
      "loss": 5.4302,
      "mean_token_accuracy": 0.20349669754505156,
      "num_tokens": 74802225.0,
      "step": 38230
    },
    {
      "entropy": 5.905653285980224,
      "epoch": 3.8220122956965064,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036045874830661613,
      "loss": 5.3277,
      "mean_token_accuracy": 0.20401953160762787,
      "num_tokens": 74811037.0,
      "step": 38235
    },
    {
      "entropy": 5.909545803070069,
      "epoch": 3.822512120757735,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003604257326076397,
      "loss": 5.3855,
      "mean_token_accuracy": 0.20057802200317382,
      "num_tokens": 74820858.0,
      "step": 38240
    },
    {
      "entropy": 5.845198297500611,
      "epoch": 3.8230119458189633,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036039271475937366,
      "loss": 5.3137,
      "mean_token_accuracy": 0.20710042268037795,
      "num_tokens": 74830091.0,
      "step": 38245
    },
    {
      "entropy": 5.828225564956665,
      "epoch": 3.823511770880192,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000360359694762649,
      "loss": 5.3766,
      "mean_token_accuracy": 0.20777156203985214,
      "num_tokens": 74840689.0,
      "step": 38250
    },
    {
      "entropy": 5.907227802276611,
      "epoch": 3.8240115959414203,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003603266726182963,
      "loss": 5.4013,
      "mean_token_accuracy": 0.2010972887277603,
      "num_tokens": 74850212.0,
      "step": 38255
    },
    {
      "entropy": 5.976600074768067,
      "epoch": 3.824511421002649,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003602936483271465,
      "loss": 5.4489,
      "mean_token_accuracy": 0.205497807264328,
      "num_tokens": 74860933.0,
      "step": 38260
    },
    {
      "entropy": 5.865457391738891,
      "epoch": 3.8250112460638777,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036026062189003046,
      "loss": 5.3263,
      "mean_token_accuracy": 0.2070230945944786,
      "num_tokens": 74871007.0,
      "step": 38265
    },
    {
      "entropy": 5.755685758590698,
      "epoch": 3.8255110711251064,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003602275933077792,
      "loss": 5.2762,
      "mean_token_accuracy": 0.21742690950632096,
      "num_tokens": 74879971.0,
      "step": 38270
    },
    {
      "entropy": 5.786479043960571,
      "epoch": 3.826010896186335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036019456258122346,
      "loss": 5.2522,
      "mean_token_accuracy": 0.21082431077957153,
      "num_tokens": 74889139.0,
      "step": 38275
    },
    {
      "entropy": 5.803353357315063,
      "epoch": 3.8265107212475633,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036016152971119457,
      "loss": 5.3114,
      "mean_token_accuracy": 0.21157339662313462,
      "num_tokens": 74898506.0,
      "step": 38280
    },
    {
      "entropy": 5.8844153881073,
      "epoch": 3.827010546308792,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003601284946985235,
      "loss": 5.3969,
      "mean_token_accuracy": 0.1991559937596321,
      "num_tokens": 74907198.0,
      "step": 38285
    },
    {
      "entropy": 5.909179544448852,
      "epoch": 3.8275103713700203,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003600954575440414,
      "loss": 5.4114,
      "mean_token_accuracy": 0.20122620314359665,
      "num_tokens": 74917099.0,
      "step": 38290
    },
    {
      "entropy": 5.864298152923584,
      "epoch": 3.828010196431249,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003600624182485794,
      "loss": 5.3686,
      "mean_token_accuracy": 0.2045236200094223,
      "num_tokens": 74926215.0,
      "step": 38295
    },
    {
      "entropy": 5.8627838611602785,
      "epoch": 3.8285100214924777,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003600293768129689,
      "loss": 5.4045,
      "mean_token_accuracy": 0.20316825211048126,
      "num_tokens": 74935872.0,
      "step": 38300
    },
    {
      "entropy": 5.8612205505371096,
      "epoch": 3.8290098465537064,
      "grad_norm": 2.03125,
      "learning_rate": 0.00035999633323804123,
      "loss": 5.28,
      "mean_token_accuracy": 0.21452339589595795,
      "num_tokens": 74946121.0,
      "step": 38305
    },
    {
      "entropy": 5.847300004959107,
      "epoch": 3.8295096716149346,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035996328752462757,
      "loss": 5.3807,
      "mean_token_accuracy": 0.20242113023996353,
      "num_tokens": 74956431.0,
      "step": 38310
    },
    {
      "entropy": 5.820830059051514,
      "epoch": 3.8300094966761633,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035993023967355954,
      "loss": 5.3126,
      "mean_token_accuracy": 0.21042001098394394,
      "num_tokens": 74966969.0,
      "step": 38315
    },
    {
      "entropy": 5.849985647201538,
      "epoch": 3.830509321737392,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035989718968566846,
      "loss": 5.3652,
      "mean_token_accuracy": 0.20612969249486923,
      "num_tokens": 74976758.0,
      "step": 38320
    },
    {
      "entropy": 5.884218740463257,
      "epoch": 3.8310091467986203,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000359864137561786,
      "loss": 5.4324,
      "mean_token_accuracy": 0.1988618031144142,
      "num_tokens": 74986257.0,
      "step": 38325
    },
    {
      "entropy": 5.8336385726928714,
      "epoch": 3.831508971859849,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003598310833027437,
      "loss": 5.3314,
      "mean_token_accuracy": 0.2021564081311226,
      "num_tokens": 74996887.0,
      "step": 38330
    },
    {
      "entropy": 5.833591890335083,
      "epoch": 3.8320087969210777,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035979802690937315,
      "loss": 5.3316,
      "mean_token_accuracy": 0.20728767216205596,
      "num_tokens": 75007188.0,
      "step": 38335
    },
    {
      "entropy": 5.878053474426269,
      "epoch": 3.8325086219823064,
      "grad_norm": 1.125,
      "learning_rate": 0.00035976496838250607,
      "loss": 5.3153,
      "mean_token_accuracy": 0.20969967395067216,
      "num_tokens": 75016402.0,
      "step": 38340
    },
    {
      "entropy": 5.848310708999634,
      "epoch": 3.8330084470435346,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035973190772297417,
      "loss": 5.3467,
      "mean_token_accuracy": 0.20584735423326492,
      "num_tokens": 75025781.0,
      "step": 38345
    },
    {
      "entropy": 5.838138103485107,
      "epoch": 3.8335082721047633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035969884493160925,
      "loss": 5.3911,
      "mean_token_accuracy": 0.20299049764871596,
      "num_tokens": 75034957.0,
      "step": 38350
    },
    {
      "entropy": 5.878546619415284,
      "epoch": 3.834008097165992,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035966578000924324,
      "loss": 5.4137,
      "mean_token_accuracy": 0.19979369044303893,
      "num_tokens": 75045310.0,
      "step": 38355
    },
    {
      "entropy": 5.853176259994507,
      "epoch": 3.8345079222272203,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003596327129567079,
      "loss": 5.3601,
      "mean_token_accuracy": 0.20145158618688583,
      "num_tokens": 75057169.0,
      "step": 38360
    },
    {
      "entropy": 5.880301666259766,
      "epoch": 3.835007747288449,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035959964377483536,
      "loss": 5.3356,
      "mean_token_accuracy": 0.20884221643209458,
      "num_tokens": 75066409.0,
      "step": 38365
    },
    {
      "entropy": 5.89373927116394,
      "epoch": 3.8355075723496777,
      "grad_norm": 1.125,
      "learning_rate": 0.0003595665724644576,
      "loss": 5.3844,
      "mean_token_accuracy": 0.20389229208230972,
      "num_tokens": 75075313.0,
      "step": 38370
    },
    {
      "entropy": 5.862022399902344,
      "epoch": 3.8360073974109063,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003595334990264065,
      "loss": 5.3555,
      "mean_token_accuracy": 0.20813409835100175,
      "num_tokens": 75083812.0,
      "step": 38375
    },
    {
      "entropy": 5.835578966140747,
      "epoch": 3.8365072224721346,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003595004234615144,
      "loss": 5.4009,
      "mean_token_accuracy": 0.2091774359345436,
      "num_tokens": 75094249.0,
      "step": 38380
    },
    {
      "entropy": 5.837056732177734,
      "epoch": 3.8370070475333633,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035946734577061345,
      "loss": 5.2962,
      "mean_token_accuracy": 0.20837867110967637,
      "num_tokens": 75105052.0,
      "step": 38385
    },
    {
      "entropy": 5.9158093452453615,
      "epoch": 3.837506872594592,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003594342659545357,
      "loss": 5.3893,
      "mean_token_accuracy": 0.20220448225736617,
      "num_tokens": 75114374.0,
      "step": 38390
    },
    {
      "entropy": 5.846032285690308,
      "epoch": 3.8380066976558203,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003594011840141136,
      "loss": 5.3317,
      "mean_token_accuracy": 0.20665618628263474,
      "num_tokens": 75124153.0,
      "step": 38395
    },
    {
      "entropy": 5.853851938247681,
      "epoch": 3.838506522717049,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035936809995017936,
      "loss": 5.4533,
      "mean_token_accuracy": 0.20424787402153016,
      "num_tokens": 75136273.0,
      "step": 38400
    },
    {
      "entropy": 5.883292627334595,
      "epoch": 3.8390063477782777,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035933501376356555,
      "loss": 5.2524,
      "mean_token_accuracy": 0.2118017256259918,
      "num_tokens": 75145567.0,
      "step": 38405
    },
    {
      "entropy": 5.908703374862671,
      "epoch": 3.8395061728395063,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003593019254551045,
      "loss": 5.4308,
      "mean_token_accuracy": 0.2068703815340996,
      "num_tokens": 75154755.0,
      "step": 38410
    },
    {
      "entropy": 5.884470367431641,
      "epoch": 3.8400059979007346,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035926883502562855,
      "loss": 5.4016,
      "mean_token_accuracy": 0.19978394508361816,
      "num_tokens": 75164109.0,
      "step": 38415
    },
    {
      "entropy": 5.8586649894714355,
      "epoch": 3.8405058229619633,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035923574247597053,
      "loss": 5.3346,
      "mean_token_accuracy": 0.21028977185487746,
      "num_tokens": 75173253.0,
      "step": 38420
    },
    {
      "entropy": 5.784244537353516,
      "epoch": 3.841005648023192,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035920264780696284,
      "loss": 5.3061,
      "mean_token_accuracy": 0.2140722543001175,
      "num_tokens": 75182553.0,
      "step": 38425
    },
    {
      "entropy": 5.872708511352539,
      "epoch": 3.8415054730844203,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003591695510194382,
      "loss": 5.3426,
      "mean_token_accuracy": 0.2017175331711769,
      "num_tokens": 75191007.0,
      "step": 38430
    },
    {
      "entropy": 5.85223355293274,
      "epoch": 3.842005298145649,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003591364521142293,
      "loss": 5.2772,
      "mean_token_accuracy": 0.20877874195575713,
      "num_tokens": 75200673.0,
      "step": 38435
    },
    {
      "entropy": 5.824548435211182,
      "epoch": 3.8425051232068776,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035910335109216895,
      "loss": 5.3165,
      "mean_token_accuracy": 0.2161339044570923,
      "num_tokens": 75209949.0,
      "step": 38440
    },
    {
      "entropy": 5.809161043167114,
      "epoch": 3.8430049482681063,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035907024795409,
      "loss": 5.2767,
      "mean_token_accuracy": 0.21271747797727586,
      "num_tokens": 75218946.0,
      "step": 38445
    },
    {
      "entropy": 5.801943111419678,
      "epoch": 3.8435047733293346,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035903714270082515,
      "loss": 5.2751,
      "mean_token_accuracy": 0.20921372324228288,
      "num_tokens": 75230773.0,
      "step": 38450
    },
    {
      "entropy": 5.803305435180664,
      "epoch": 3.8440045983905633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003590040353332074,
      "loss": 5.3919,
      "mean_token_accuracy": 0.19768657088279723,
      "num_tokens": 75241016.0,
      "step": 38455
    },
    {
      "entropy": 5.869005966186523,
      "epoch": 3.844504423451792,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003589709258520698,
      "loss": 5.4122,
      "mean_token_accuracy": 0.2029002785682678,
      "num_tokens": 75252011.0,
      "step": 38460
    },
    {
      "entropy": 5.842081880569458,
      "epoch": 3.8450042485130203,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003589378142582453,
      "loss": 5.247,
      "mean_token_accuracy": 0.2126206412911415,
      "num_tokens": 75261983.0,
      "step": 38465
    },
    {
      "entropy": 5.920661497116089,
      "epoch": 3.845504073574249,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035890470055256697,
      "loss": 5.364,
      "mean_token_accuracy": 0.20700614750385285,
      "num_tokens": 75271778.0,
      "step": 38470
    },
    {
      "entropy": 5.8544410228729244,
      "epoch": 3.8460038986354776,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035887158473586793,
      "loss": 5.3264,
      "mean_token_accuracy": 0.20475405007600783,
      "num_tokens": 75280725.0,
      "step": 38475
    },
    {
      "entropy": 5.85164155960083,
      "epoch": 3.8465037236967063,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003588384668089815,
      "loss": 5.3564,
      "mean_token_accuracy": 0.20267630219459534,
      "num_tokens": 75289770.0,
      "step": 38480
    },
    {
      "entropy": 5.825656747817993,
      "epoch": 3.8470035487579346,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035880534677274075,
      "loss": 5.3587,
      "mean_token_accuracy": 0.20524126589298247,
      "num_tokens": 75300088.0,
      "step": 38485
    },
    {
      "entropy": 5.8413636684417725,
      "epoch": 3.8475033738191633,
      "grad_norm": 1.0625,
      "learning_rate": 0.000358772224627979,
      "loss": 5.3221,
      "mean_token_accuracy": 0.20596254616975784,
      "num_tokens": 75309854.0,
      "step": 38490
    },
    {
      "entropy": 5.820237064361573,
      "epoch": 3.848003198880392,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035873910037552973,
      "loss": 5.269,
      "mean_token_accuracy": 0.2139969438314438,
      "num_tokens": 75319662.0,
      "step": 38495
    },
    {
      "entropy": 5.869282150268555,
      "epoch": 3.8485030239416202,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003587059740162262,
      "loss": 5.3589,
      "mean_token_accuracy": 0.20796797573566436,
      "num_tokens": 75329943.0,
      "step": 38500
    },
    {
      "entropy": 5.8200843334198,
      "epoch": 3.849002849002849,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003586728455509019,
      "loss": 5.337,
      "mean_token_accuracy": 0.20455944538116455,
      "num_tokens": 75339235.0,
      "step": 38505
    },
    {
      "entropy": 5.859843397140503,
      "epoch": 3.8495026740640776,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035863971498039033,
      "loss": 5.3069,
      "mean_token_accuracy": 0.2114760935306549,
      "num_tokens": 75348473.0,
      "step": 38510
    },
    {
      "entropy": 5.800118923187256,
      "epoch": 3.8500024991253063,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035860658230552506,
      "loss": 5.3194,
      "mean_token_accuracy": 0.20510145723819734,
      "num_tokens": 75357822.0,
      "step": 38515
    },
    {
      "entropy": 5.874572134017944,
      "epoch": 3.8505023241865346,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003585734475271397,
      "loss": 5.4161,
      "mean_token_accuracy": 0.19914297312498092,
      "num_tokens": 75367885.0,
      "step": 38520
    },
    {
      "entropy": 5.8511560440063475,
      "epoch": 3.8510021492477633,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003585403106460679,
      "loss": 5.3155,
      "mean_token_accuracy": 0.208677040040493,
      "num_tokens": 75377536.0,
      "step": 38525
    },
    {
      "entropy": 5.886343002319336,
      "epoch": 3.851501974308992,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003585071716631433,
      "loss": 5.3428,
      "mean_token_accuracy": 0.20681060701608658,
      "num_tokens": 75388027.0,
      "step": 38530
    },
    {
      "entropy": 5.86639084815979,
      "epoch": 3.8520017993702202,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003584740305791999,
      "loss": 5.3196,
      "mean_token_accuracy": 0.20216361433267593,
      "num_tokens": 75398385.0,
      "step": 38535
    },
    {
      "entropy": 5.913250398635864,
      "epoch": 3.852501624431449,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003584408873950713,
      "loss": 5.3576,
      "mean_token_accuracy": 0.20273967385292052,
      "num_tokens": 75407903.0,
      "step": 38540
    },
    {
      "entropy": 5.872078514099121,
      "epoch": 3.8530014494926776,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035840774211159146,
      "loss": 5.3677,
      "mean_token_accuracy": 0.2046529769897461,
      "num_tokens": 75417973.0,
      "step": 38545
    },
    {
      "entropy": 5.776837539672852,
      "epoch": 3.8535012745539063,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035837459472959425,
      "loss": 5.2655,
      "mean_token_accuracy": 0.21719880253076554,
      "num_tokens": 75426590.0,
      "step": 38550
    },
    {
      "entropy": 5.804480123519897,
      "epoch": 3.8540010996151346,
      "grad_norm": 1.25,
      "learning_rate": 0.0003583414452499137,
      "loss": 5.2861,
      "mean_token_accuracy": 0.21935774087905885,
      "num_tokens": 75435840.0,
      "step": 38555
    },
    {
      "entropy": 5.851765298843384,
      "epoch": 3.8545009246763633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035830829367338384,
      "loss": 5.3529,
      "mean_token_accuracy": 0.19748778343200685,
      "num_tokens": 75446285.0,
      "step": 38560
    },
    {
      "entropy": 5.813028240203858,
      "epoch": 3.855000749737592,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003582751400008388,
      "loss": 5.2511,
      "mean_token_accuracy": 0.21236702501773835,
      "num_tokens": 75456703.0,
      "step": 38565
    },
    {
      "entropy": 5.9032543182373045,
      "epoch": 3.8555005747988202,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035824198423311263,
      "loss": 5.4489,
      "mean_token_accuracy": 0.1979002133011818,
      "num_tokens": 75467616.0,
      "step": 38570
    },
    {
      "entropy": 5.8749754428863525,
      "epoch": 3.856000399860049,
      "grad_norm": 1.125,
      "learning_rate": 0.0003582088263710396,
      "loss": 5.368,
      "mean_token_accuracy": 0.20271042585372925,
      "num_tokens": 75477462.0,
      "step": 38575
    },
    {
      "entropy": 5.902626419067383,
      "epoch": 3.8565002249212776,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035817566641545376,
      "loss": 5.4154,
      "mean_token_accuracy": 0.19169347286224364,
      "num_tokens": 75487438.0,
      "step": 38580
    },
    {
      "entropy": 5.912590503692627,
      "epoch": 3.8570000499825063,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003581425043671897,
      "loss": 5.4194,
      "mean_token_accuracy": 0.19508148431777955,
      "num_tokens": 75497207.0,
      "step": 38585
    },
    {
      "entropy": 5.940573453903198,
      "epoch": 3.8574998750437346,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003581093402270816,
      "loss": 5.3614,
      "mean_token_accuracy": 0.20598238110542297,
      "num_tokens": 75507671.0,
      "step": 38590
    },
    {
      "entropy": 5.802586030960083,
      "epoch": 3.8579997001049633,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035807617399596383,
      "loss": 5.2867,
      "mean_token_accuracy": 0.2112416908144951,
      "num_tokens": 75517958.0,
      "step": 38595
    },
    {
      "entropy": 5.822821235656738,
      "epoch": 3.858499525166192,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003580430056746709,
      "loss": 5.3566,
      "mean_token_accuracy": 0.20790457278490065,
      "num_tokens": 75528701.0,
      "step": 38600
    },
    {
      "entropy": 5.852470016479492,
      "epoch": 3.8589993502274202,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003580098352640374,
      "loss": 5.3508,
      "mean_token_accuracy": 0.20398642718791962,
      "num_tokens": 75537913.0,
      "step": 38605
    },
    {
      "entropy": 5.956241703033447,
      "epoch": 3.859499175288649,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035797666276489773,
      "loss": 5.3885,
      "mean_token_accuracy": 0.20421511828899383,
      "num_tokens": 75548016.0,
      "step": 38610
    },
    {
      "entropy": 5.918894481658936,
      "epoch": 3.8599990003498776,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035794348817808666,
      "loss": 5.4196,
      "mean_token_accuracy": 0.2099613904953003,
      "num_tokens": 75558435.0,
      "step": 38615
    },
    {
      "entropy": 5.7980711460113525,
      "epoch": 3.8604988254111063,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035791031150443864,
      "loss": 5.3535,
      "mean_token_accuracy": 0.19945961683988572,
      "num_tokens": 75568752.0,
      "step": 38620
    },
    {
      "entropy": 5.860563850402832,
      "epoch": 3.8609986504723346,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003578771327447886,
      "loss": 5.3747,
      "mean_token_accuracy": 0.205789415538311,
      "num_tokens": 75578631.0,
      "step": 38625
    },
    {
      "entropy": 5.8409637928009035,
      "epoch": 3.8614984755335633,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003578439518999712,
      "loss": 5.272,
      "mean_token_accuracy": 0.20973225980997084,
      "num_tokens": 75588235.0,
      "step": 38630
    },
    {
      "entropy": 5.82537522315979,
      "epoch": 3.861998300594792,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035781076897082134,
      "loss": 5.3501,
      "mean_token_accuracy": 0.2076904743909836,
      "num_tokens": 75597301.0,
      "step": 38635
    },
    {
      "entropy": 5.7972412109375,
      "epoch": 3.8624981256560202,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035777758395817374,
      "loss": 5.3393,
      "mean_token_accuracy": 0.2086398869752884,
      "num_tokens": 75606914.0,
      "step": 38640
    },
    {
      "entropy": 5.663380289077759,
      "epoch": 3.862997950717249,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003577443968628635,
      "loss": 5.1229,
      "mean_token_accuracy": 0.22622082829475404,
      "num_tokens": 75616763.0,
      "step": 38645
    },
    {
      "entropy": 5.815287446975708,
      "epoch": 3.8634977757784776,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003577112076857256,
      "loss": 5.2775,
      "mean_token_accuracy": 0.21409989893436432,
      "num_tokens": 75626716.0,
      "step": 38650
    },
    {
      "entropy": 5.787504959106445,
      "epoch": 3.8639976008397063,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035767801642759485,
      "loss": 5.2741,
      "mean_token_accuracy": 0.21597410887479782,
      "num_tokens": 75636862.0,
      "step": 38655
    },
    {
      "entropy": 5.864890193939209,
      "epoch": 3.8644974259009346,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035764482308930657,
      "loss": 5.385,
      "mean_token_accuracy": 0.19810827672481537,
      "num_tokens": 75646021.0,
      "step": 38660
    },
    {
      "entropy": 5.849979114532471,
      "epoch": 3.8649972509621633,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003576116276716958,
      "loss": 5.3507,
      "mean_token_accuracy": 0.2090669721364975,
      "num_tokens": 75655404.0,
      "step": 38665
    },
    {
      "entropy": 5.829971361160278,
      "epoch": 3.8654970760233915,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003575784301755977,
      "loss": 5.3589,
      "mean_token_accuracy": 0.2103956252336502,
      "num_tokens": 75664668.0,
      "step": 38670
    },
    {
      "entropy": 5.8931962013244625,
      "epoch": 3.8659969010846202,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035754523060184753,
      "loss": 5.3156,
      "mean_token_accuracy": 0.21052024960517884,
      "num_tokens": 75674922.0,
      "step": 38675
    },
    {
      "entropy": 5.832808351516723,
      "epoch": 3.866496726145849,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035751202895128065,
      "loss": 5.2305,
      "mean_token_accuracy": 0.2157263457775116,
      "num_tokens": 75683555.0,
      "step": 38680
    },
    {
      "entropy": 5.777707862854004,
      "epoch": 3.8669965512070776,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003574788252247324,
      "loss": 5.2911,
      "mean_token_accuracy": 0.21314522325992585,
      "num_tokens": 75692621.0,
      "step": 38685
    },
    {
      "entropy": 5.85011396408081,
      "epoch": 3.8674963762683063,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035744561942303805,
      "loss": 5.3218,
      "mean_token_accuracy": 0.21016769856214523,
      "num_tokens": 75702887.0,
      "step": 38690
    },
    {
      "entropy": 5.867334461212158,
      "epoch": 3.8679962013295346,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003574124115470332,
      "loss": 5.3597,
      "mean_token_accuracy": 0.2093933790922165,
      "num_tokens": 75712390.0,
      "step": 38695
    },
    {
      "entropy": 5.8322361469268795,
      "epoch": 3.8684960263907633,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035737920159755317,
      "loss": 5.2792,
      "mean_token_accuracy": 0.2185346469283104,
      "num_tokens": 75722078.0,
      "step": 38700
    },
    {
      "entropy": 5.740361642837525,
      "epoch": 3.8689958514519915,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003573459895754337,
      "loss": 5.255,
      "mean_token_accuracy": 0.212737013399601,
      "num_tokens": 75732210.0,
      "step": 38705
    },
    {
      "entropy": 5.799743270874023,
      "epoch": 3.8694956765132202,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003573127754815103,
      "loss": 5.3407,
      "mean_token_accuracy": 0.20203393697738647,
      "num_tokens": 75741896.0,
      "step": 38710
    },
    {
      "entropy": 5.821242475509644,
      "epoch": 3.869995501574449,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035727955931661863,
      "loss": 5.3595,
      "mean_token_accuracy": 0.20460417866706848,
      "num_tokens": 75751525.0,
      "step": 38715
    },
    {
      "entropy": 5.862780666351318,
      "epoch": 3.8704953266356776,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035724634108159453,
      "loss": 5.4163,
      "mean_token_accuracy": 0.20250946432352065,
      "num_tokens": 75761176.0,
      "step": 38720
    },
    {
      "entropy": 5.84398922920227,
      "epoch": 3.8709951516969063,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003572131207772736,
      "loss": 5.2649,
      "mean_token_accuracy": 0.2152211144566536,
      "num_tokens": 75770753.0,
      "step": 38725
    },
    {
      "entropy": 5.823091077804565,
      "epoch": 3.8714949767581346,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003571798984044916,
      "loss": 5.3325,
      "mean_token_accuracy": 0.20451492220163345,
      "num_tokens": 75779587.0,
      "step": 38730
    },
    {
      "entropy": 5.850241851806641,
      "epoch": 3.8719948018193633,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035714667396408455,
      "loss": 5.3631,
      "mean_token_accuracy": 0.2112296536564827,
      "num_tokens": 75788874.0,
      "step": 38735
    },
    {
      "entropy": 5.847804212570191,
      "epoch": 3.8724946268805915,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003571134474568883,
      "loss": 5.3845,
      "mean_token_accuracy": 0.19766987413167952,
      "num_tokens": 75798120.0,
      "step": 38740
    },
    {
      "entropy": 5.848286437988281,
      "epoch": 3.8729944519418202,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000357080218883739,
      "loss": 5.3562,
      "mean_token_accuracy": 0.20437593907117843,
      "num_tokens": 75806817.0,
      "step": 38745
    },
    {
      "entropy": 5.825500917434693,
      "epoch": 3.873494277003049,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003570469882454724,
      "loss": 5.2934,
      "mean_token_accuracy": 0.20614126473665237,
      "num_tokens": 75816264.0,
      "step": 38750
    },
    {
      "entropy": 5.889812803268432,
      "epoch": 3.8739941020642776,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035701375554292477,
      "loss": 5.4228,
      "mean_token_accuracy": 0.19972701370716095,
      "num_tokens": 75826045.0,
      "step": 38755
    },
    {
      "entropy": 5.8295848846435545,
      "epoch": 3.8744939271255063,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035698052077693206,
      "loss": 5.3827,
      "mean_token_accuracy": 0.20298974663019181,
      "num_tokens": 75835983.0,
      "step": 38760
    },
    {
      "entropy": 5.870523881912232,
      "epoch": 3.8749937521867346,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003569472839483305,
      "loss": 5.3446,
      "mean_token_accuracy": 0.2039317637681961,
      "num_tokens": 75844992.0,
      "step": 38765
    },
    {
      "entropy": 5.86756272315979,
      "epoch": 3.8754935772479633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035691404505795645,
      "loss": 5.3505,
      "mean_token_accuracy": 0.2126454696059227,
      "num_tokens": 75854738.0,
      "step": 38770
    },
    {
      "entropy": 5.792322587966919,
      "epoch": 3.8759934023091915,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003568808041066461,
      "loss": 5.2613,
      "mean_token_accuracy": 0.21448912918567659,
      "num_tokens": 75864018.0,
      "step": 38775
    },
    {
      "entropy": 5.900158214569092,
      "epoch": 3.8764932273704202,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003568475610952357,
      "loss": 5.3553,
      "mean_token_accuracy": 0.20278207063674927,
      "num_tokens": 75874188.0,
      "step": 38780
    },
    {
      "entropy": 5.818155670166016,
      "epoch": 3.876993052431649,
      "grad_norm": 1.125,
      "learning_rate": 0.00035681431602456183,
      "loss": 5.3469,
      "mean_token_accuracy": 0.2031813457608223,
      "num_tokens": 75884088.0,
      "step": 38785
    },
    {
      "entropy": 5.851763153076172,
      "epoch": 3.8774928774928776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035678106889546087,
      "loss": 5.4032,
      "mean_token_accuracy": 0.1999696746468544,
      "num_tokens": 75893973.0,
      "step": 38790
    },
    {
      "entropy": 5.769983959197998,
      "epoch": 3.8779927025541063,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00035674781970876905,
      "loss": 5.2559,
      "mean_token_accuracy": 0.21642529964447021,
      "num_tokens": 75904545.0,
      "step": 38795
    },
    {
      "entropy": 5.839637374877929,
      "epoch": 3.8784925276153346,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003567145684653233,
      "loss": 5.3167,
      "mean_token_accuracy": 0.21866150945425034,
      "num_tokens": 75914622.0,
      "step": 38800
    },
    {
      "entropy": 5.753320264816284,
      "epoch": 3.8789923526765633,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035668131516596,
      "loss": 5.2526,
      "mean_token_accuracy": 0.21849234104156495,
      "num_tokens": 75925395.0,
      "step": 38805
    },
    {
      "entropy": 5.809417867660523,
      "epoch": 3.8794921777377915,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035664805981151576,
      "loss": 5.2822,
      "mean_token_accuracy": 0.21483976244926453,
      "num_tokens": 75936816.0,
      "step": 38810
    },
    {
      "entropy": 5.8167744159698485,
      "epoch": 3.8799920027990202,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003566148024028274,
      "loss": 5.2736,
      "mean_token_accuracy": 0.210096338391304,
      "num_tokens": 75946137.0,
      "step": 38815
    },
    {
      "entropy": 5.843068170547485,
      "epoch": 3.880491827860249,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035658154294073157,
      "loss": 5.3208,
      "mean_token_accuracy": 0.2145974963903427,
      "num_tokens": 75956098.0,
      "step": 38820
    },
    {
      "entropy": 5.814663028717041,
      "epoch": 3.8809916529214776,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035654828142606517,
      "loss": 5.3782,
      "mean_token_accuracy": 0.20837922543287277,
      "num_tokens": 75967324.0,
      "step": 38825
    },
    {
      "entropy": 5.832974672317505,
      "epoch": 3.8814914779827063,
      "grad_norm": 1.15625,
      "learning_rate": 0.000356515017859665,
      "loss": 5.4593,
      "mean_token_accuracy": 0.20140430629253386,
      "num_tokens": 75976147.0,
      "step": 38830
    },
    {
      "entropy": 5.9245250701904295,
      "epoch": 3.8819913030439346,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035648175224236797,
      "loss": 5.4755,
      "mean_token_accuracy": 0.19852502793073654,
      "num_tokens": 75985421.0,
      "step": 38835
    },
    {
      "entropy": 5.834830570220947,
      "epoch": 3.8824911281051633,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035644848457501096,
      "loss": 5.3201,
      "mean_token_accuracy": 0.21036941111087798,
      "num_tokens": 75994628.0,
      "step": 38840
    },
    {
      "entropy": 5.804785299301147,
      "epoch": 3.8829909531663915,
      "grad_norm": 1.0625,
      "learning_rate": 0.000356415214858431,
      "loss": 5.2999,
      "mean_token_accuracy": 0.20743838846683502,
      "num_tokens": 76004556.0,
      "step": 38845
    },
    {
      "entropy": 5.859496974945069,
      "epoch": 3.8834907782276202,
      "grad_norm": 1.125,
      "learning_rate": 0.0003563819430934652,
      "loss": 5.3479,
      "mean_token_accuracy": 0.20858951508998871,
      "num_tokens": 76014336.0,
      "step": 38850
    },
    {
      "entropy": 5.773294401168823,
      "epoch": 3.883990603288849,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003563486692809508,
      "loss": 5.3238,
      "mean_token_accuracy": 0.20991101115942,
      "num_tokens": 76023537.0,
      "step": 38855
    },
    {
      "entropy": 5.824258184432983,
      "epoch": 3.8844904283500776,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003563153934217246,
      "loss": 5.3915,
      "mean_token_accuracy": 0.19851518422365189,
      "num_tokens": 76032183.0,
      "step": 38860
    },
    {
      "entropy": 5.790609407424927,
      "epoch": 3.884990253411306,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003562821155166242,
      "loss": 5.2567,
      "mean_token_accuracy": 0.2111320123076439,
      "num_tokens": 76042287.0,
      "step": 38865
    },
    {
      "entropy": 5.834213304519653,
      "epoch": 3.8854900784725346,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035624883556648664,
      "loss": 5.4431,
      "mean_token_accuracy": 0.1991872876882553,
      "num_tokens": 76052795.0,
      "step": 38870
    },
    {
      "entropy": 5.8259130954742435,
      "epoch": 3.8859899035337633,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003562155535721492,
      "loss": 5.3737,
      "mean_token_accuracy": 0.20214440822601318,
      "num_tokens": 76063087.0,
      "step": 38875
    },
    {
      "entropy": 5.875546026229858,
      "epoch": 3.8864897285949915,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003561822695344494,
      "loss": 5.3454,
      "mean_token_accuracy": 0.2039526879787445,
      "num_tokens": 76071762.0,
      "step": 38880
    },
    {
      "entropy": 5.820380783081054,
      "epoch": 3.8869895536562202,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003561489834542246,
      "loss": 5.2886,
      "mean_token_accuracy": 0.21331300139427184,
      "num_tokens": 76080814.0,
      "step": 38885
    },
    {
      "entropy": 5.799558067321778,
      "epoch": 3.887489378717449,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035611569533231224,
      "loss": 5.3061,
      "mean_token_accuracy": 0.20029009878635406,
      "num_tokens": 76090709.0,
      "step": 38890
    },
    {
      "entropy": 5.880346584320068,
      "epoch": 3.8879892037786776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035608240516954995,
      "loss": 5.3943,
      "mean_token_accuracy": 0.20247654765844345,
      "num_tokens": 76100425.0,
      "step": 38895
    },
    {
      "entropy": 5.745716619491577,
      "epoch": 3.888489028839906,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035604911296677503,
      "loss": 5.2459,
      "mean_token_accuracy": 0.21481709629297258,
      "num_tokens": 76111579.0,
      "step": 38900
    },
    {
      "entropy": 5.87819538116455,
      "epoch": 3.8889888539011346,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003560158187248254,
      "loss": 5.3552,
      "mean_token_accuracy": 0.20724639892578126,
      "num_tokens": 76121291.0,
      "step": 38905
    },
    {
      "entropy": 5.83539137840271,
      "epoch": 3.8894886789623633,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003559825224445386,
      "loss": 5.3079,
      "mean_token_accuracy": 0.20724832117557526,
      "num_tokens": 76131154.0,
      "step": 38910
    },
    {
      "entropy": 5.890252351760864,
      "epoch": 3.8899885040235915,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003559492241267524,
      "loss": 5.4245,
      "mean_token_accuracy": 0.20264645367860795,
      "num_tokens": 76141154.0,
      "step": 38915
    },
    {
      "entropy": 5.797631072998047,
      "epoch": 3.8904883290848202,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035591592377230447,
      "loss": 5.2782,
      "mean_token_accuracy": 0.21725087463855744,
      "num_tokens": 76150335.0,
      "step": 38920
    },
    {
      "entropy": 5.867477560043335,
      "epoch": 3.890988154146049,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035588262138203276,
      "loss": 5.3816,
      "mean_token_accuracy": 0.2041119486093521,
      "num_tokens": 76159508.0,
      "step": 38925
    },
    {
      "entropy": 5.819229459762573,
      "epoch": 3.8914879792072776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035584931695677516,
      "loss": 5.3062,
      "mean_token_accuracy": 0.21705431044101714,
      "num_tokens": 76168234.0,
      "step": 38930
    },
    {
      "entropy": 5.840745067596435,
      "epoch": 3.891987804268506,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003558160104973695,
      "loss": 5.3777,
      "mean_token_accuracy": 0.20987636744976043,
      "num_tokens": 76177638.0,
      "step": 38935
    },
    {
      "entropy": 5.853730344772339,
      "epoch": 3.8924876293297346,
      "grad_norm": 1.25,
      "learning_rate": 0.0003557827020046538,
      "loss": 5.3626,
      "mean_token_accuracy": 0.20340905636548995,
      "num_tokens": 76186570.0,
      "step": 38940
    },
    {
      "entropy": 5.848280906677246,
      "epoch": 3.8929874543909633,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003557493914794661,
      "loss": 5.3785,
      "mean_token_accuracy": 0.20053339302539824,
      "num_tokens": 76196386.0,
      "step": 38945
    },
    {
      "entropy": 5.849374151229858,
      "epoch": 3.8934872794521915,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003557160789226445,
      "loss": 5.3482,
      "mean_token_accuracy": 0.21086386293172837,
      "num_tokens": 76207388.0,
      "step": 38950
    },
    {
      "entropy": 5.939233732223511,
      "epoch": 3.8939871045134202,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003556827643350271,
      "loss": 5.4498,
      "mean_token_accuracy": 0.197841639816761,
      "num_tokens": 76217490.0,
      "step": 38955
    },
    {
      "entropy": 5.8793971061706545,
      "epoch": 3.894486929574649,
      "grad_norm": 1.125,
      "learning_rate": 0.0003556494477174522,
      "loss": 5.3927,
      "mean_token_accuracy": 0.20229585468769073,
      "num_tokens": 76227688.0,
      "step": 38960
    },
    {
      "entropy": 5.834059238433838,
      "epoch": 3.8949867546358776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035561612907075794,
      "loss": 5.2858,
      "mean_token_accuracy": 0.21908741742372512,
      "num_tokens": 76237383.0,
      "step": 38965
    },
    {
      "entropy": 5.851900863647461,
      "epoch": 3.895486579697106,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035558280839578256,
      "loss": 5.269,
      "mean_token_accuracy": 0.21157976984977722,
      "num_tokens": 76246617.0,
      "step": 38970
    },
    {
      "entropy": 5.867745876312256,
      "epoch": 3.8959864047583346,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003555494856933644,
      "loss": 5.4014,
      "mean_token_accuracy": 0.20239920616149903,
      "num_tokens": 76255257.0,
      "step": 38975
    },
    {
      "entropy": 5.827837133407593,
      "epoch": 3.8964862298195633,
      "grad_norm": 1.15625,
      "learning_rate": 0.000355516160964342,
      "loss": 5.2798,
      "mean_token_accuracy": 0.2080608367919922,
      "num_tokens": 76264359.0,
      "step": 38980
    },
    {
      "entropy": 5.837845420837402,
      "epoch": 3.8969860548807915,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035548283420955377,
      "loss": 5.4386,
      "mean_token_accuracy": 0.20716082751750947,
      "num_tokens": 76274385.0,
      "step": 38985
    },
    {
      "entropy": 5.877472734451294,
      "epoch": 3.8974858799420202,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035544950542983793,
      "loss": 5.3949,
      "mean_token_accuracy": 0.20558408200740813,
      "num_tokens": 76284622.0,
      "step": 38990
    },
    {
      "entropy": 5.859451580047607,
      "epoch": 3.897985705003249,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035541617462603336,
      "loss": 5.3171,
      "mean_token_accuracy": 0.20873816460371017,
      "num_tokens": 76294551.0,
      "step": 38995
    },
    {
      "entropy": 5.84703106880188,
      "epoch": 3.8984855300644776,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003553828417989786,
      "loss": 5.3165,
      "mean_token_accuracy": 0.21042502224445342,
      "num_tokens": 76304184.0,
      "step": 39000
    },
    {
      "epoch": 3.8984855300644776,
      "eval_entropy": 5.619569937429057,
      "eval_loss": 5.470088481903076,
      "eval_mean_token_accuracy": 0.20890137820061383,
      "eval_num_tokens": 76304184.0,
      "eval_runtime": 23.6518,
      "eval_samples_per_second": 1312.588,
      "eval_steps_per_second": 164.089,
      "step": 39000
    },
    {
      "entropy": 5.848143291473389,
      "epoch": 3.898985355125706,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035534950694951213,
      "loss": 5.3115,
      "mean_token_accuracy": 0.20672922283411027,
      "num_tokens": 76313977.0,
      "step": 39005
    },
    {
      "entropy": 5.85790228843689,
      "epoch": 3.8994851801869346,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003553161700784728,
      "loss": 5.363,
      "mean_token_accuracy": 0.20417520850896836,
      "num_tokens": 76323552.0,
      "step": 39010
    },
    {
      "entropy": 5.76600866317749,
      "epoch": 3.8999850052481633,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003552828311866992,
      "loss": 5.2837,
      "mean_token_accuracy": 0.21230679899454116,
      "num_tokens": 76333431.0,
      "step": 39015
    },
    {
      "entropy": 5.802076101303101,
      "epoch": 3.9004848303093915,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003552494902750303,
      "loss": 5.2778,
      "mean_token_accuracy": 0.21029011607170106,
      "num_tokens": 76341922.0,
      "step": 39020
    },
    {
      "entropy": 5.819615840911865,
      "epoch": 3.9009846553706202,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035521614734430483,
      "loss": 5.3001,
      "mean_token_accuracy": 0.20723946988582612,
      "num_tokens": 76351693.0,
      "step": 39025
    },
    {
      "entropy": 5.835004186630249,
      "epoch": 3.901484480431849,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035518280239536174,
      "loss": 5.3641,
      "mean_token_accuracy": 0.20972472727298735,
      "num_tokens": 76361200.0,
      "step": 39030
    },
    {
      "entropy": 5.938093137741089,
      "epoch": 3.9019843054930776,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003551494554290401,
      "loss": 5.2396,
      "mean_token_accuracy": 0.21354392021894456,
      "num_tokens": 76370164.0,
      "step": 39035
    },
    {
      "entropy": 5.891850709915161,
      "epoch": 3.902484130554306,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035511610644617863,
      "loss": 5.2975,
      "mean_token_accuracy": 0.21566222608089447,
      "num_tokens": 76379310.0,
      "step": 39040
    },
    {
      "entropy": 5.7217268466949465,
      "epoch": 3.9029839556155346,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003550827554476166,
      "loss": 5.2458,
      "mean_token_accuracy": 0.21556329727172852,
      "num_tokens": 76388540.0,
      "step": 39045
    },
    {
      "entropy": 5.770640707015991,
      "epoch": 3.9034837806767633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035504940243419303,
      "loss": 5.2375,
      "mean_token_accuracy": 0.2204391360282898,
      "num_tokens": 76397606.0,
      "step": 39050
    },
    {
      "entropy": 5.814146566390991,
      "epoch": 3.9039836057379915,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003550160474067471,
      "loss": 5.2768,
      "mean_token_accuracy": 0.2036997139453888,
      "num_tokens": 76406862.0,
      "step": 39055
    },
    {
      "entropy": 5.907884502410889,
      "epoch": 3.9044834307992202,
      "grad_norm": 1.203125,
      "learning_rate": 0.000354982690366118,
      "loss": 5.4536,
      "mean_token_accuracy": 0.19937568455934523,
      "num_tokens": 76417108.0,
      "step": 39060
    },
    {
      "entropy": 5.828027296066284,
      "epoch": 3.904983255860449,
      "grad_norm": 1.140625,
      "learning_rate": 0.000354949331313145,
      "loss": 5.3023,
      "mean_token_accuracy": 0.20959648191928865,
      "num_tokens": 76425487.0,
      "step": 39065
    },
    {
      "entropy": 5.834328746795654,
      "epoch": 3.9054830809216776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035491597024866746,
      "loss": 5.2132,
      "mean_token_accuracy": 0.21641886234283447,
      "num_tokens": 76434034.0,
      "step": 39070
    },
    {
      "entropy": 5.864026021957398,
      "epoch": 3.905982905982906,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003548826071735246,
      "loss": 5.3546,
      "mean_token_accuracy": 0.20897374004125596,
      "num_tokens": 76443813.0,
      "step": 39075
    },
    {
      "entropy": 5.7570788860321045,
      "epoch": 3.9064827310441346,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003548492420885559,
      "loss": 5.2277,
      "mean_token_accuracy": 0.22058860063552857,
      "num_tokens": 76453897.0,
      "step": 39080
    },
    {
      "entropy": 5.788261222839355,
      "epoch": 3.9069825561053633,
      "grad_norm": 1.125,
      "learning_rate": 0.00035481587499460083,
      "loss": 5.3273,
      "mean_token_accuracy": 0.21282434165477754,
      "num_tokens": 76463615.0,
      "step": 39085
    },
    {
      "entropy": 5.8023213863372805,
      "epoch": 3.9074823811665915,
      "grad_norm": 1.125,
      "learning_rate": 0.0003547825058924989,
      "loss": 5.2662,
      "mean_token_accuracy": 0.2182892605662346,
      "num_tokens": 76472777.0,
      "step": 39090
    },
    {
      "entropy": 5.855275011062622,
      "epoch": 3.90798220622782,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003547491347830896,
      "loss": 5.3734,
      "mean_token_accuracy": 0.20446103066205978,
      "num_tokens": 76483476.0,
      "step": 39095
    },
    {
      "entropy": 5.893444299697876,
      "epoch": 3.908482031289049,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035471576166721267,
      "loss": 5.3869,
      "mean_token_accuracy": 0.20758388191461563,
      "num_tokens": 76493861.0,
      "step": 39100
    },
    {
      "entropy": 5.8814514636993405,
      "epoch": 3.9089818563502776,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035468238654570765,
      "loss": 5.3089,
      "mean_token_accuracy": 0.20494206100702286,
      "num_tokens": 76502745.0,
      "step": 39105
    },
    {
      "entropy": 5.7962933540344235,
      "epoch": 3.909481681411506,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003546490094194142,
      "loss": 5.2963,
      "mean_token_accuracy": 0.20820008218288422,
      "num_tokens": 76511789.0,
      "step": 39110
    },
    {
      "entropy": 5.795183086395264,
      "epoch": 3.9099815064727346,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035461563028917233,
      "loss": 5.2454,
      "mean_token_accuracy": 0.21926802396774292,
      "num_tokens": 76521910.0,
      "step": 39115
    },
    {
      "entropy": 5.9399677276611325,
      "epoch": 3.9104813315339633,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035458224915582156,
      "loss": 5.3906,
      "mean_token_accuracy": 0.20237912386655807,
      "num_tokens": 76531648.0,
      "step": 39120
    },
    {
      "entropy": 5.879639959335327,
      "epoch": 3.9109811565951915,
      "grad_norm": 1.125,
      "learning_rate": 0.00035454886602020194,
      "loss": 5.2951,
      "mean_token_accuracy": 0.20309591740369798,
      "num_tokens": 76541087.0,
      "step": 39125
    },
    {
      "entropy": 5.8631920337677,
      "epoch": 3.91148098165642,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035451548088315325,
      "loss": 5.3741,
      "mean_token_accuracy": 0.20658351927995683,
      "num_tokens": 76551354.0,
      "step": 39130
    },
    {
      "entropy": 5.889763689041137,
      "epoch": 3.911980806717649,
      "grad_norm": 1.0,
      "learning_rate": 0.0003544820937455156,
      "loss": 5.361,
      "mean_token_accuracy": 0.20659321248531343,
      "num_tokens": 76562346.0,
      "step": 39135
    },
    {
      "entropy": 5.807758617401123,
      "epoch": 3.9124806317788776,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035444870460812894,
      "loss": 5.2889,
      "mean_token_accuracy": 0.21502464413642883,
      "num_tokens": 76571068.0,
      "step": 39140
    },
    {
      "entropy": 5.761919832229614,
      "epoch": 3.912980456840106,
      "grad_norm": 1.125,
      "learning_rate": 0.00035441531347183325,
      "loss": 5.3002,
      "mean_token_accuracy": 0.22602308988571168,
      "num_tokens": 76581085.0,
      "step": 39145
    },
    {
      "entropy": 5.7386668682098385,
      "epoch": 3.9134802819013346,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003543819203374688,
      "loss": 5.1689,
      "mean_token_accuracy": 0.22588267773389817,
      "num_tokens": 76590465.0,
      "step": 39150
    },
    {
      "entropy": 5.901175022125244,
      "epoch": 3.9139801069625633,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003543485252058756,
      "loss": 5.2949,
      "mean_token_accuracy": 0.2077655702829361,
      "num_tokens": 76599980.0,
      "step": 39155
    },
    {
      "entropy": 5.832581186294556,
      "epoch": 3.9144799320237915,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003543151280778939,
      "loss": 5.3422,
      "mean_token_accuracy": 0.20474830120801926,
      "num_tokens": 76610198.0,
      "step": 39160
    },
    {
      "entropy": 5.935567903518677,
      "epoch": 3.91497975708502,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000354281728954364,
      "loss": 5.4656,
      "mean_token_accuracy": 0.19784589260816574,
      "num_tokens": 76618597.0,
      "step": 39165
    },
    {
      "entropy": 5.860767841339111,
      "epoch": 3.915479582146249,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035424832783612623,
      "loss": 5.3874,
      "mean_token_accuracy": 0.19857994318008423,
      "num_tokens": 76628608.0,
      "step": 39170
    },
    {
      "entropy": 5.8658346176147464,
      "epoch": 3.9159794072074776,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003542149247240209,
      "loss": 5.285,
      "mean_token_accuracy": 0.2088000789284706,
      "num_tokens": 76638345.0,
      "step": 39175
    },
    {
      "entropy": 5.871561861038208,
      "epoch": 3.916479232268706,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035418151961888846,
      "loss": 5.3769,
      "mean_token_accuracy": 0.20822062641382216,
      "num_tokens": 76647391.0,
      "step": 39180
    },
    {
      "entropy": 5.92044734954834,
      "epoch": 3.9169790573299346,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035414811252156936,
      "loss": 5.412,
      "mean_token_accuracy": 0.19902351796627044,
      "num_tokens": 76657095.0,
      "step": 39185
    },
    {
      "entropy": 5.827980470657349,
      "epoch": 3.917478882391163,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003541147034329041,
      "loss": 5.2533,
      "mean_token_accuracy": 0.22240989804267883,
      "num_tokens": 76666957.0,
      "step": 39190
    },
    {
      "entropy": 5.876183795928955,
      "epoch": 3.9179787074523915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035408129235373325,
      "loss": 5.3868,
      "mean_token_accuracy": 0.20475016683340072,
      "num_tokens": 76677112.0,
      "step": 39195
    },
    {
      "entropy": 5.931296539306641,
      "epoch": 3.91847853251362,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035404787928489746,
      "loss": 5.4093,
      "mean_token_accuracy": 0.20331885516643525,
      "num_tokens": 76686886.0,
      "step": 39200
    },
    {
      "entropy": 5.73927149772644,
      "epoch": 3.918978357574849,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035401446422723737,
      "loss": 5.1159,
      "mean_token_accuracy": 0.23506150096654893,
      "num_tokens": 76696271.0,
      "step": 39205
    },
    {
      "entropy": 5.856512212753296,
      "epoch": 3.9194781826360776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035398104718159366,
      "loss": 5.3799,
      "mean_token_accuracy": 0.20287849009037018,
      "num_tokens": 76706585.0,
      "step": 39210
    },
    {
      "entropy": 5.84841742515564,
      "epoch": 3.919978007697306,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003539476281488072,
      "loss": 5.306,
      "mean_token_accuracy": 0.2156092718243599,
      "num_tokens": 76716230.0,
      "step": 39215
    },
    {
      "entropy": 5.830973482131958,
      "epoch": 3.9204778327585346,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035391420712971864,
      "loss": 5.3395,
      "mean_token_accuracy": 0.20325327664613724,
      "num_tokens": 76726084.0,
      "step": 39220
    },
    {
      "entropy": 5.890242099761963,
      "epoch": 3.920977657819763,
      "grad_norm": 1.140625,
      "learning_rate": 0.000353880784125169,
      "loss": 5.4617,
      "mean_token_accuracy": 0.19918035268783568,
      "num_tokens": 76735704.0,
      "step": 39225
    },
    {
      "entropy": 5.840966749191284,
      "epoch": 3.9214774828809915,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035384735913599906,
      "loss": 5.2929,
      "mean_token_accuracy": 0.2067929610610008,
      "num_tokens": 76746391.0,
      "step": 39230
    },
    {
      "entropy": 5.774289226531982,
      "epoch": 3.92197730794222,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003538139321630499,
      "loss": 5.2005,
      "mean_token_accuracy": 0.2197247788310051,
      "num_tokens": 76755287.0,
      "step": 39235
    },
    {
      "entropy": 5.812520694732666,
      "epoch": 3.922477133003449,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003537805032071625,
      "loss": 5.3109,
      "mean_token_accuracy": 0.20387859344482423,
      "num_tokens": 76764728.0,
      "step": 39240
    },
    {
      "entropy": 5.841549110412598,
      "epoch": 3.9229769580646776,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003537470722691779,
      "loss": 5.3578,
      "mean_token_accuracy": 0.20410515367984772,
      "num_tokens": 76773756.0,
      "step": 39245
    },
    {
      "entropy": 5.863055992126465,
      "epoch": 3.923476783125906,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003537136393499372,
      "loss": 5.3591,
      "mean_token_accuracy": 0.20920788794755935,
      "num_tokens": 76783780.0,
      "step": 39250
    },
    {
      "entropy": 5.718487882614136,
      "epoch": 3.9239766081871346,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003536802044502816,
      "loss": 5.2338,
      "mean_token_accuracy": 0.21560714542865753,
      "num_tokens": 76794282.0,
      "step": 39255
    },
    {
      "entropy": 5.874641752243042,
      "epoch": 3.924476433248363,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035364676757105234,
      "loss": 5.3633,
      "mean_token_accuracy": 0.20453393459320068,
      "num_tokens": 76803871.0,
      "step": 39260
    },
    {
      "entropy": 5.893647289276123,
      "epoch": 3.9249762583095915,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003536133287130906,
      "loss": 5.3819,
      "mean_token_accuracy": 0.19844426810741425,
      "num_tokens": 76813148.0,
      "step": 39265
    },
    {
      "entropy": 5.844379854202271,
      "epoch": 3.92547608337082,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035357988787723774,
      "loss": 5.3794,
      "mean_token_accuracy": 0.19806349128484727,
      "num_tokens": 76822048.0,
      "step": 39270
    },
    {
      "entropy": 5.874217939376831,
      "epoch": 3.925975908432049,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035354644506433515,
      "loss": 5.3742,
      "mean_token_accuracy": 0.20131372809410095,
      "num_tokens": 76832366.0,
      "step": 39275
    },
    {
      "entropy": 5.880928611755371,
      "epoch": 3.9264757334932776,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003535130002752242,
      "loss": 5.3242,
      "mean_token_accuracy": 0.21041828244924546,
      "num_tokens": 76842106.0,
      "step": 39280
    },
    {
      "entropy": 5.8612407684326175,
      "epoch": 3.926975558554506,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003534795535107464,
      "loss": 5.3108,
      "mean_token_accuracy": 0.20875187665224076,
      "num_tokens": 76851885.0,
      "step": 39285
    },
    {
      "entropy": 5.917792844772339,
      "epoch": 3.9274753836157346,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035344610477174314,
      "loss": 5.4607,
      "mean_token_accuracy": 0.1983307734131813,
      "num_tokens": 76862643.0,
      "step": 39290
    },
    {
      "entropy": 5.729645919799805,
      "epoch": 3.927975208676963,
      "grad_norm": 1.5625,
      "learning_rate": 0.00035341265405905615,
      "loss": 5.2486,
      "mean_token_accuracy": 0.22006638050079347,
      "num_tokens": 76873017.0,
      "step": 39295
    },
    {
      "entropy": 5.9432567119598385,
      "epoch": 3.9284750337381915,
      "grad_norm": 1.171875,
      "learning_rate": 0.000353379201373527,
      "loss": 5.42,
      "mean_token_accuracy": 0.19887750148773192,
      "num_tokens": 76881360.0,
      "step": 39300
    },
    {
      "entropy": 5.790843343734741,
      "epoch": 3.92897485879942,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035334574671599717,
      "loss": 5.3074,
      "mean_token_accuracy": 0.20813892185688018,
      "num_tokens": 76891520.0,
      "step": 39305
    },
    {
      "entropy": 5.8117530822753904,
      "epoch": 3.929474683860649,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003533122900873086,
      "loss": 5.3084,
      "mean_token_accuracy": 0.20982145816087722,
      "num_tokens": 76901748.0,
      "step": 39310
    },
    {
      "entropy": 5.845374298095703,
      "epoch": 3.9299745089218776,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003532788314883029,
      "loss": 5.2426,
      "mean_token_accuracy": 0.2128077730536461,
      "num_tokens": 76910882.0,
      "step": 39315
    },
    {
      "entropy": 5.897785758972168,
      "epoch": 3.930474333983106,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035324537091982206,
      "loss": 5.497,
      "mean_token_accuracy": 0.1944902166724205,
      "num_tokens": 76920852.0,
      "step": 39320
    },
    {
      "entropy": 5.877722358703613,
      "epoch": 3.9309741590443346,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035321190838270773,
      "loss": 5.2989,
      "mean_token_accuracy": 0.1979812800884247,
      "num_tokens": 76929978.0,
      "step": 39325
    },
    {
      "entropy": 5.81464114189148,
      "epoch": 3.931473984105563,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000353178443877802,
      "loss": 5.3186,
      "mean_token_accuracy": 0.20866504609584807,
      "num_tokens": 76938760.0,
      "step": 39330
    },
    {
      "entropy": 5.85618314743042,
      "epoch": 3.9319738091667915,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035314497740594666,
      "loss": 5.3574,
      "mean_token_accuracy": 0.21111985594034194,
      "num_tokens": 76948692.0,
      "step": 39335
    },
    {
      "entropy": 5.848618841171264,
      "epoch": 3.93247363422802,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003531115089679837,
      "loss": 5.3581,
      "mean_token_accuracy": 0.1978414088487625,
      "num_tokens": 76957620.0,
      "step": 39340
    },
    {
      "entropy": 5.914307928085327,
      "epoch": 3.932973459289249,
      "grad_norm": 1.25,
      "learning_rate": 0.00035307803856475544,
      "loss": 5.4679,
      "mean_token_accuracy": 0.20555053055286407,
      "num_tokens": 76967942.0,
      "step": 39345
    },
    {
      "entropy": 5.901432514190674,
      "epoch": 3.9334732843504776,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003530445661971038,
      "loss": 5.3643,
      "mean_token_accuracy": 0.21047534048557281,
      "num_tokens": 76977894.0,
      "step": 39350
    },
    {
      "entropy": 5.876964139938354,
      "epoch": 3.933973109411706,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00035301109186587095,
      "loss": 5.3523,
      "mean_token_accuracy": 0.2047621190547943,
      "num_tokens": 76987955.0,
      "step": 39355
    },
    {
      "entropy": 5.745348596572876,
      "epoch": 3.9344729344729346,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035297761557189904,
      "loss": 5.2841,
      "mean_token_accuracy": 0.21156144589185716,
      "num_tokens": 76999241.0,
      "step": 39360
    },
    {
      "entropy": 5.877205562591553,
      "epoch": 3.934972759534163,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035294413731603035,
      "loss": 5.3729,
      "mean_token_accuracy": 0.20093078464269637,
      "num_tokens": 77010055.0,
      "step": 39365
    },
    {
      "entropy": 5.9761138439178465,
      "epoch": 3.9354725845953915,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035291065709910734,
      "loss": 5.41,
      "mean_token_accuracy": 0.20465686917304993,
      "num_tokens": 77020126.0,
      "step": 39370
    },
    {
      "entropy": 5.801824283599854,
      "epoch": 3.93597240965662,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003528771749219722,
      "loss": 5.1701,
      "mean_token_accuracy": 0.21817698031663896,
      "num_tokens": 77029147.0,
      "step": 39375
    },
    {
      "entropy": 5.861500263214111,
      "epoch": 3.936472234717849,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003528436907854674,
      "loss": 5.395,
      "mean_token_accuracy": 0.20157777070999144,
      "num_tokens": 77038845.0,
      "step": 39380
    },
    {
      "entropy": 5.8890016078948975,
      "epoch": 3.936972059779077,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003528102046904354,
      "loss": 5.4798,
      "mean_token_accuracy": 0.1924258217215538,
      "num_tokens": 77048607.0,
      "step": 39385
    },
    {
      "entropy": 5.903139114379883,
      "epoch": 3.937471884840306,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003527767166377186,
      "loss": 5.3723,
      "mean_token_accuracy": 0.20508427619934083,
      "num_tokens": 77057282.0,
      "step": 39390
    },
    {
      "entropy": 5.867789697647095,
      "epoch": 3.9379717099015346,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003527432266281597,
      "loss": 5.3176,
      "mean_token_accuracy": 0.20970436930656433,
      "num_tokens": 77067147.0,
      "step": 39395
    },
    {
      "entropy": 5.849299669265747,
      "epoch": 3.938471534962763,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003527097346626011,
      "loss": 5.327,
      "mean_token_accuracy": 0.213461035490036,
      "num_tokens": 77076345.0,
      "step": 39400
    },
    {
      "entropy": 5.824049234390259,
      "epoch": 3.9389713600239915,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035267624074188564,
      "loss": 5.3387,
      "mean_token_accuracy": 0.20852357149124146,
      "num_tokens": 77085414.0,
      "step": 39405
    },
    {
      "entropy": 5.824828195571899,
      "epoch": 3.93947118508522,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035264274486685593,
      "loss": 5.326,
      "mean_token_accuracy": 0.21441456377506257,
      "num_tokens": 77095086.0,
      "step": 39410
    },
    {
      "entropy": 5.7593999862670895,
      "epoch": 3.939971010146449,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003526092470383548,
      "loss": 5.203,
      "mean_token_accuracy": 0.21729733645915986,
      "num_tokens": 77104292.0,
      "step": 39415
    },
    {
      "entropy": 5.776622581481933,
      "epoch": 3.940470835207677,
      "grad_norm": 1.125,
      "learning_rate": 0.00035257574725722495,
      "loss": 5.301,
      "mean_token_accuracy": 0.203578020632267,
      "num_tokens": 77113258.0,
      "step": 39420
    },
    {
      "entropy": 5.870618677139282,
      "epoch": 3.940970660268906,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035254224552430923,
      "loss": 5.3871,
      "mean_token_accuracy": 0.21041846871376038,
      "num_tokens": 77123986.0,
      "step": 39425
    },
    {
      "entropy": 5.896091079711914,
      "epoch": 3.9414704853301346,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035250874184045063,
      "loss": 5.3192,
      "mean_token_accuracy": 0.2078562006354332,
      "num_tokens": 77132901.0,
      "step": 39430
    },
    {
      "entropy": 5.760683345794678,
      "epoch": 3.941970310391363,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035247523620649196,
      "loss": 5.3189,
      "mean_token_accuracy": 0.21325637251138688,
      "num_tokens": 77143261.0,
      "step": 39435
    },
    {
      "entropy": 5.7816699028015135,
      "epoch": 3.9424701354525915,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035244172862327635,
      "loss": 5.3158,
      "mean_token_accuracy": 0.20878227055072784,
      "num_tokens": 77152779.0,
      "step": 39440
    },
    {
      "entropy": 5.871104621887207,
      "epoch": 3.94296996051382,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035240821909164675,
      "loss": 5.3743,
      "mean_token_accuracy": 0.20727674812078475,
      "num_tokens": 77162210.0,
      "step": 39445
    },
    {
      "entropy": 5.818810939788818,
      "epoch": 3.943469785575049,
      "grad_norm": 1.125,
      "learning_rate": 0.0003523747076124462,
      "loss": 5.2337,
      "mean_token_accuracy": 0.20969536900520325,
      "num_tokens": 77171749.0,
      "step": 39450
    },
    {
      "entropy": 5.9144690990447994,
      "epoch": 3.943969610636277,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035234119418651804,
      "loss": 5.4278,
      "mean_token_accuracy": 0.19835874438285828,
      "num_tokens": 77182588.0,
      "step": 39455
    },
    {
      "entropy": 5.825549697875976,
      "epoch": 3.944469435697506,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003523076788147053,
      "loss": 5.2857,
      "mean_token_accuracy": 0.2230598583817482,
      "num_tokens": 77191655.0,
      "step": 39460
    },
    {
      "entropy": 5.893199348449707,
      "epoch": 3.9449692607587346,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003522741614978512,
      "loss": 5.3834,
      "mean_token_accuracy": 0.2095898672938347,
      "num_tokens": 77201808.0,
      "step": 39465
    },
    {
      "entropy": 5.8381153583526615,
      "epoch": 3.945469085819963,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035224064223679914,
      "loss": 5.307,
      "mean_token_accuracy": 0.20697053521871567,
      "num_tokens": 77210414.0,
      "step": 39470
    },
    {
      "entropy": 5.856251239776611,
      "epoch": 3.9459689108811915,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003522071210323924,
      "loss": 5.3548,
      "mean_token_accuracy": 0.20914637446403503,
      "num_tokens": 77219310.0,
      "step": 39475
    },
    {
      "entropy": 5.834813261032105,
      "epoch": 3.94646873594242,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003521735978854743,
      "loss": 5.248,
      "mean_token_accuracy": 0.2170690417289734,
      "num_tokens": 77229277.0,
      "step": 39480
    },
    {
      "entropy": 5.838455581665039,
      "epoch": 3.946968561003649,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003521400727968884,
      "loss": 5.3815,
      "mean_token_accuracy": 0.20582637637853624,
      "num_tokens": 77239285.0,
      "step": 39485
    },
    {
      "entropy": 5.822215461730957,
      "epoch": 3.947468386064877,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035210654576747806,
      "loss": 5.2647,
      "mean_token_accuracy": 0.20770976394414903,
      "num_tokens": 77248250.0,
      "step": 39490
    },
    {
      "entropy": 5.860601234436035,
      "epoch": 3.947968211126106,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035207301679808685,
      "loss": 5.4083,
      "mean_token_accuracy": 0.19990243166685104,
      "num_tokens": 77258353.0,
      "step": 39495
    },
    {
      "entropy": 5.859338331222534,
      "epoch": 3.9484680361873346,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003520394858895584,
      "loss": 5.3678,
      "mean_token_accuracy": 0.2061013326048851,
      "num_tokens": 77267754.0,
      "step": 39500
    },
    {
      "entropy": 5.86430253982544,
      "epoch": 3.948967861248563,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003520059530427364,
      "loss": 5.3217,
      "mean_token_accuracy": 0.21054787337779998,
      "num_tokens": 77277949.0,
      "step": 39505
    },
    {
      "entropy": 5.8464258193969725,
      "epoch": 3.9494676863097915,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003519724182584643,
      "loss": 5.3863,
      "mean_token_accuracy": 0.20724727362394332,
      "num_tokens": 77287804.0,
      "step": 39510
    },
    {
      "entropy": 5.8740428447723385,
      "epoch": 3.94996751137102,
      "grad_norm": 1.125,
      "learning_rate": 0.0003519388815375861,
      "loss": 5.3983,
      "mean_token_accuracy": 0.20015152394771576,
      "num_tokens": 77297312.0,
      "step": 39515
    },
    {
      "entropy": 5.916828393936157,
      "epoch": 3.950467336432249,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035190534288094533,
      "loss": 5.4128,
      "mean_token_accuracy": 0.20016244500875474,
      "num_tokens": 77306617.0,
      "step": 39520
    },
    {
      "entropy": 5.924597311019897,
      "epoch": 3.950967161493477,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035187180228938593,
      "loss": 5.4288,
      "mean_token_accuracy": 0.20659977793693543,
      "num_tokens": 77315786.0,
      "step": 39525
    },
    {
      "entropy": 5.907064771652221,
      "epoch": 3.951466986554706,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003518382597637518,
      "loss": 5.3293,
      "mean_token_accuracy": 0.2176923230290413,
      "num_tokens": 77325126.0,
      "step": 39530
    },
    {
      "entropy": 5.824592590332031,
      "epoch": 3.9519668116159345,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003518047153048868,
      "loss": 5.328,
      "mean_token_accuracy": 0.20450786352157593,
      "num_tokens": 77333784.0,
      "step": 39535
    },
    {
      "entropy": 5.880098485946656,
      "epoch": 3.952466636677163,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000351771168913635,
      "loss": 5.3891,
      "mean_token_accuracy": 0.19913261830806733,
      "num_tokens": 77343279.0,
      "step": 39540
    },
    {
      "entropy": 5.83465805053711,
      "epoch": 3.9529664617383915,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003517376205908403,
      "loss": 5.3275,
      "mean_token_accuracy": 0.20446455329656602,
      "num_tokens": 77353336.0,
      "step": 39545
    },
    {
      "entropy": 5.893794298171997,
      "epoch": 3.95346628679962,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003517040703373467,
      "loss": 5.4436,
      "mean_token_accuracy": 0.19595089703798294,
      "num_tokens": 77362961.0,
      "step": 39550
    },
    {
      "entropy": 5.818742656707764,
      "epoch": 3.953966111860849,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003516705181539986,
      "loss": 5.2854,
      "mean_token_accuracy": 0.21214884519577026,
      "num_tokens": 77372096.0,
      "step": 39555
    },
    {
      "entropy": 5.872999811172486,
      "epoch": 3.954465936922077,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003516369640416399,
      "loss": 5.3496,
      "mean_token_accuracy": 0.2063092917203903,
      "num_tokens": 77382477.0,
      "step": 39560
    },
    {
      "entropy": 5.81103892326355,
      "epoch": 3.954965761983306,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035160340800111496,
      "loss": 5.25,
      "mean_token_accuracy": 0.21164832264184952,
      "num_tokens": 77392042.0,
      "step": 39565
    },
    {
      "entropy": 5.835052108764648,
      "epoch": 3.9554655870445345,
      "grad_norm": 1.125,
      "learning_rate": 0.000351569850033268,
      "loss": 5.3748,
      "mean_token_accuracy": 0.2032160520553589,
      "num_tokens": 77402041.0,
      "step": 39570
    },
    {
      "entropy": 5.858169078826904,
      "epoch": 3.955965412105763,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003515362901389432,
      "loss": 5.3265,
      "mean_token_accuracy": 0.21246946305036546,
      "num_tokens": 77412253.0,
      "step": 39575
    },
    {
      "entropy": 5.858785820007324,
      "epoch": 3.9564652371669915,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003515027283189851,
      "loss": 5.3388,
      "mean_token_accuracy": 0.2020752876996994,
      "num_tokens": 77421675.0,
      "step": 39580
    },
    {
      "entropy": 5.92902102470398,
      "epoch": 3.95696506222822,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003514691645742381,
      "loss": 5.4335,
      "mean_token_accuracy": 0.2051561266183853,
      "num_tokens": 77431772.0,
      "step": 39585
    },
    {
      "entropy": 5.860202407836914,
      "epoch": 3.957464887289449,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003514355989055465,
      "loss": 5.4105,
      "mean_token_accuracy": 0.20771252363920212,
      "num_tokens": 77441162.0,
      "step": 39590
    },
    {
      "entropy": 5.844443273544312,
      "epoch": 3.957964712350677,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000351402031313755,
      "loss": 5.328,
      "mean_token_accuracy": 0.21272777915000915,
      "num_tokens": 77450666.0,
      "step": 39595
    },
    {
      "entropy": 5.871548271179199,
      "epoch": 3.958464537411906,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035136846179970797,
      "loss": 5.3552,
      "mean_token_accuracy": 0.20382148325443267,
      "num_tokens": 77460077.0,
      "step": 39600
    },
    {
      "entropy": 5.906522560119629,
      "epoch": 3.9589643624731345,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035133489036425007,
      "loss": 5.3718,
      "mean_token_accuracy": 0.20627703219652177,
      "num_tokens": 77469756.0,
      "step": 39605
    },
    {
      "entropy": 5.857908916473389,
      "epoch": 3.959464187534363,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035130131700822603,
      "loss": 5.2746,
      "mean_token_accuracy": 0.2126256451010704,
      "num_tokens": 77479235.0,
      "step": 39610
    },
    {
      "entropy": 5.804060220718384,
      "epoch": 3.9599640125955915,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003512677417324805,
      "loss": 5.32,
      "mean_token_accuracy": 0.2097933441400528,
      "num_tokens": 77488815.0,
      "step": 39615
    },
    {
      "entropy": 5.892038202285766,
      "epoch": 3.96046383765682,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003512341645378582,
      "loss": 5.4209,
      "mean_token_accuracy": 0.20062029510736465,
      "num_tokens": 77499262.0,
      "step": 39620
    },
    {
      "entropy": 5.894968175888062,
      "epoch": 3.960963662718049,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003512005854252039,
      "loss": 5.4192,
      "mean_token_accuracy": 0.20690009146928787,
      "num_tokens": 77508595.0,
      "step": 39625
    },
    {
      "entropy": 5.921145868301392,
      "epoch": 3.961463487779277,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035116700439536254,
      "loss": 5.4388,
      "mean_token_accuracy": 0.20272907018661498,
      "num_tokens": 77519579.0,
      "step": 39630
    },
    {
      "entropy": 5.943585777282715,
      "epoch": 3.961963312840506,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035113342144917894,
      "loss": 5.3991,
      "mean_token_accuracy": 0.20419021248817443,
      "num_tokens": 77529917.0,
      "step": 39635
    },
    {
      "entropy": 5.924534749984741,
      "epoch": 3.9624631379017345,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035109983658749796,
      "loss": 5.4398,
      "mean_token_accuracy": 0.19909384846687317,
      "num_tokens": 77540199.0,
      "step": 39640
    },
    {
      "entropy": 5.839155387878418,
      "epoch": 3.962962962962963,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003510662498111648,
      "loss": 5.3173,
      "mean_token_accuracy": 0.21184782683849335,
      "num_tokens": 77549300.0,
      "step": 39645
    },
    {
      "entropy": 5.783731269836426,
      "epoch": 3.9634627880241915,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035103266112102423,
      "loss": 5.3227,
      "mean_token_accuracy": 0.21652702242136002,
      "num_tokens": 77559449.0,
      "step": 39650
    },
    {
      "entropy": 5.797265720367432,
      "epoch": 3.96396261308542,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035099907051792153,
      "loss": 5.3341,
      "mean_token_accuracy": 0.20943325608968735,
      "num_tokens": 77569089.0,
      "step": 39655
    },
    {
      "entropy": 5.816582250595093,
      "epoch": 3.964462438146649,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003509654780027018,
      "loss": 5.38,
      "mean_token_accuracy": 0.20451130121946334,
      "num_tokens": 77580049.0,
      "step": 39660
    },
    {
      "entropy": 5.915019273757935,
      "epoch": 3.964962263207877,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003509318835762102,
      "loss": 5.387,
      "mean_token_accuracy": 0.20970566421747208,
      "num_tokens": 77589861.0,
      "step": 39665
    },
    {
      "entropy": 5.8441578388214115,
      "epoch": 3.965462088269106,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003508982872392919,
      "loss": 5.3223,
      "mean_token_accuracy": 0.20314589887857437,
      "num_tokens": 77599413.0,
      "step": 39670
    },
    {
      "entropy": 5.8175441265106205,
      "epoch": 3.9659619133303345,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003508646889927923,
      "loss": 5.337,
      "mean_token_accuracy": 0.20346746146678923,
      "num_tokens": 77608607.0,
      "step": 39675
    },
    {
      "entropy": 5.830837249755859,
      "epoch": 3.966461738391563,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035083108883755647,
      "loss": 5.3754,
      "mean_token_accuracy": 0.19975619167089462,
      "num_tokens": 77619446.0,
      "step": 39680
    },
    {
      "entropy": 5.860490131378174,
      "epoch": 3.9669615634527915,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035079748677443014,
      "loss": 5.3563,
      "mean_token_accuracy": 0.20524210929870607,
      "num_tokens": 77628485.0,
      "step": 39685
    },
    {
      "entropy": 5.8485767364501955,
      "epoch": 3.96746138851402,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003507638828042585,
      "loss": 5.3252,
      "mean_token_accuracy": 0.20868010967969894,
      "num_tokens": 77638029.0,
      "step": 39690
    },
    {
      "entropy": 5.903352308273315,
      "epoch": 3.967961213575249,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003507302769278871,
      "loss": 5.3234,
      "mean_token_accuracy": 0.1997020035982132,
      "num_tokens": 77646770.0,
      "step": 39695
    },
    {
      "entropy": 5.632985353469849,
      "epoch": 3.968461038636477,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003506966691461614,
      "loss": 5.1665,
      "mean_token_accuracy": 0.2408444657921791,
      "num_tokens": 77656647.0,
      "step": 39700
    },
    {
      "entropy": 5.855734062194824,
      "epoch": 3.968960863697706,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003506630594599269,
      "loss": 5.362,
      "mean_token_accuracy": 0.20842949002981187,
      "num_tokens": 77666158.0,
      "step": 39705
    },
    {
      "entropy": 5.803958988189697,
      "epoch": 3.969460688758934,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003506294478700293,
      "loss": 5.274,
      "mean_token_accuracy": 0.2072721615433693,
      "num_tokens": 77678126.0,
      "step": 39710
    },
    {
      "entropy": 5.882946968078613,
      "epoch": 3.969960513820163,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035059583437731435,
      "loss": 5.3884,
      "mean_token_accuracy": 0.20832108706235886,
      "num_tokens": 77687943.0,
      "step": 39715
    },
    {
      "entropy": 5.830236101150513,
      "epoch": 3.9704603388813915,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035056221898262756,
      "loss": 5.2894,
      "mean_token_accuracy": 0.21428388208150864,
      "num_tokens": 77697306.0,
      "step": 39720
    },
    {
      "entropy": 5.813031530380249,
      "epoch": 3.97096016394262,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035052860168681486,
      "loss": 5.3856,
      "mean_token_accuracy": 0.2097232922911644,
      "num_tokens": 77708140.0,
      "step": 39725
    },
    {
      "entropy": 5.836177158355713,
      "epoch": 3.971459989003849,
      "grad_norm": 1.078125,
      "learning_rate": 0.000350494982490722,
      "loss": 5.277,
      "mean_token_accuracy": 0.21312179565429687,
      "num_tokens": 77718427.0,
      "step": 39730
    },
    {
      "entropy": 5.807988262176513,
      "epoch": 3.971959814065077,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035046136139519474,
      "loss": 5.3196,
      "mean_token_accuracy": 0.21208993047475816,
      "num_tokens": 77727854.0,
      "step": 39735
    },
    {
      "entropy": 5.70422854423523,
      "epoch": 3.972459639126306,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000350427738401079,
      "loss": 5.1677,
      "mean_token_accuracy": 0.23039334267377853,
      "num_tokens": 77738740.0,
      "step": 39740
    },
    {
      "entropy": 5.877308988571167,
      "epoch": 3.972959464187534,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003503941135092209,
      "loss": 5.3776,
      "mean_token_accuracy": 0.20835843533277512,
      "num_tokens": 77750368.0,
      "step": 39745
    },
    {
      "entropy": 5.740317249298096,
      "epoch": 3.973459289248763,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003503604867204662,
      "loss": 5.2007,
      "mean_token_accuracy": 0.21277765780687333,
      "num_tokens": 77759005.0,
      "step": 39750
    },
    {
      "entropy": 5.91038236618042,
      "epoch": 3.9739591143099915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003503268580356611,
      "loss": 5.4333,
      "mean_token_accuracy": 0.20128938406705857,
      "num_tokens": 77769290.0,
      "step": 39755
    },
    {
      "entropy": 5.937255334854126,
      "epoch": 3.97445893937122,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035029322745565165,
      "loss": 5.378,
      "mean_token_accuracy": 0.20471943765878678,
      "num_tokens": 77779682.0,
      "step": 39760
    },
    {
      "entropy": 5.829964971542358,
      "epoch": 3.974958764432449,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003502595949812839,
      "loss": 5.2104,
      "mean_token_accuracy": 0.22162136286497117,
      "num_tokens": 77788527.0,
      "step": 39765
    },
    {
      "entropy": 5.778991937637329,
      "epoch": 3.975458589493677,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035022596061340416,
      "loss": 5.3569,
      "mean_token_accuracy": 0.21180312484502792,
      "num_tokens": 77799253.0,
      "step": 39770
    },
    {
      "entropy": 5.816784000396728,
      "epoch": 3.975958414554906,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035019232435285863,
      "loss": 5.3316,
      "mean_token_accuracy": 0.20598496794700621,
      "num_tokens": 77809320.0,
      "step": 39775
    },
    {
      "entropy": 5.919464159011841,
      "epoch": 3.976458239616134,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003501586862004935,
      "loss": 5.3842,
      "mean_token_accuracy": 0.2024470806121826,
      "num_tokens": 77819425.0,
      "step": 39780
    },
    {
      "entropy": 5.8347491264343265,
      "epoch": 3.976958064677363,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003501250461571552,
      "loss": 5.2824,
      "mean_token_accuracy": 0.21203134208917618,
      "num_tokens": 77828941.0,
      "step": 39785
    },
    {
      "entropy": 5.856714153289795,
      "epoch": 3.9774578897385915,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003500914042236901,
      "loss": 5.3122,
      "mean_token_accuracy": 0.21125588864088057,
      "num_tokens": 77838757.0,
      "step": 39790
    },
    {
      "entropy": 5.879225826263427,
      "epoch": 3.97795771479982,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003500577604009445,
      "loss": 5.4298,
      "mean_token_accuracy": 0.20012792944908142,
      "num_tokens": 77849242.0,
      "step": 39795
    },
    {
      "entropy": 5.8483967781066895,
      "epoch": 3.978457539861049,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035002411468976506,
      "loss": 5.3767,
      "mean_token_accuracy": 0.2167668893933296,
      "num_tokens": 77859123.0,
      "step": 39800
    },
    {
      "entropy": 5.786560010910034,
      "epoch": 3.978957364922277,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034999046709099815,
      "loss": 5.2674,
      "mean_token_accuracy": 0.21417910754680633,
      "num_tokens": 77868802.0,
      "step": 39805
    },
    {
      "entropy": 5.838430261611938,
      "epoch": 3.979457189983506,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003499568176054904,
      "loss": 5.4015,
      "mean_token_accuracy": 0.19767324924468993,
      "num_tokens": 77878712.0,
      "step": 39810
    },
    {
      "entropy": 5.887332010269165,
      "epoch": 3.979957015044734,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034992316623408843,
      "loss": 5.3992,
      "mean_token_accuracy": 0.20259650200605392,
      "num_tokens": 77889777.0,
      "step": 39815
    },
    {
      "entropy": 5.802850389480591,
      "epoch": 3.980456840105963,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034988951297763887,
      "loss": 5.2417,
      "mean_token_accuracy": 0.22013285756111145,
      "num_tokens": 77898846.0,
      "step": 39820
    },
    {
      "entropy": 5.783926296234131,
      "epoch": 3.9809566651671915,
      "grad_norm": 1.125,
      "learning_rate": 0.00034985585783698845,
      "loss": 5.2446,
      "mean_token_accuracy": 0.21002272218465806,
      "num_tokens": 77907829.0,
      "step": 39825
    },
    {
      "entropy": 5.843725490570068,
      "epoch": 3.98145649022842,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003498222008129838,
      "loss": 5.3526,
      "mean_token_accuracy": 0.20601067692041397,
      "num_tokens": 77917440.0,
      "step": 39830
    },
    {
      "entropy": 5.831238842010498,
      "epoch": 3.981956315289649,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034978854190647197,
      "loss": 5.376,
      "mean_token_accuracy": 0.21407220810651778,
      "num_tokens": 77926368.0,
      "step": 39835
    },
    {
      "entropy": 5.920704412460327,
      "epoch": 3.982456140350877,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034975488111829955,
      "loss": 5.396,
      "mean_token_accuracy": 0.20586888045072554,
      "num_tokens": 77935540.0,
      "step": 39840
    },
    {
      "entropy": 5.872570371627807,
      "epoch": 3.982955965412106,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034972121844931365,
      "loss": 5.3688,
      "mean_token_accuracy": 0.20902150124311447,
      "num_tokens": 77945171.0,
      "step": 39845
    },
    {
      "entropy": 5.870317840576172,
      "epoch": 3.983455790473334,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034968755390036107,
      "loss": 5.3317,
      "mean_token_accuracy": 0.1989930123090744,
      "num_tokens": 77955246.0,
      "step": 39850
    },
    {
      "entropy": 5.748978900909424,
      "epoch": 3.983955615534563,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034965388747228886,
      "loss": 5.2658,
      "mean_token_accuracy": 0.21087395548820495,
      "num_tokens": 77964217.0,
      "step": 39855
    },
    {
      "entropy": 5.883044242858887,
      "epoch": 3.9844554405957915,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034962021916594405,
      "loss": 5.3843,
      "mean_token_accuracy": 0.2056163340806961,
      "num_tokens": 77973841.0,
      "step": 39860
    },
    {
      "entropy": 5.818896675109864,
      "epoch": 3.98495526565702,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003495865489821737,
      "loss": 5.3135,
      "mean_token_accuracy": 0.20802593529224395,
      "num_tokens": 77984080.0,
      "step": 39865
    },
    {
      "entropy": 5.932604885101318,
      "epoch": 3.985455090718249,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000349552876921825,
      "loss": 5.3967,
      "mean_token_accuracy": 0.20689045488834382,
      "num_tokens": 77994882.0,
      "step": 39870
    },
    {
      "entropy": 5.855339622497558,
      "epoch": 3.985954915779477,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003495192029857452,
      "loss": 5.3051,
      "mean_token_accuracy": 0.20856972187757492,
      "num_tokens": 78004462.0,
      "step": 39875
    },
    {
      "entropy": 5.869475698471069,
      "epoch": 3.986454740840706,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003494855271747813,
      "loss": 5.4009,
      "mean_token_accuracy": 0.2002452701330185,
      "num_tokens": 78013906.0,
      "step": 39880
    },
    {
      "entropy": 5.766204929351806,
      "epoch": 3.986954565901934,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034945184948978074,
      "loss": 5.138,
      "mean_token_accuracy": 0.22054864913225175,
      "num_tokens": 78024714.0,
      "step": 39885
    },
    {
      "entropy": 5.830857658386231,
      "epoch": 3.987454390963163,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003494181699315908,
      "loss": 5.2618,
      "mean_token_accuracy": 0.20919812768697738,
      "num_tokens": 78033640.0,
      "step": 39890
    },
    {
      "entropy": 5.863530111312866,
      "epoch": 3.9879542160243915,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003493844885010589,
      "loss": 5.3016,
      "mean_token_accuracy": 0.2065328136086464,
      "num_tokens": 78044007.0,
      "step": 39895
    },
    {
      "entropy": 5.825820350646973,
      "epoch": 3.98845404108562,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034935080519903235,
      "loss": 5.3007,
      "mean_token_accuracy": 0.21295043379068374,
      "num_tokens": 78053866.0,
      "step": 39900
    },
    {
      "entropy": 5.685273170471191,
      "epoch": 3.9889538661468484,
      "grad_norm": 1.125,
      "learning_rate": 0.0003493171200263586,
      "loss": 5.1566,
      "mean_token_accuracy": 0.23283214271068572,
      "num_tokens": 78064852.0,
      "step": 39905
    },
    {
      "entropy": 5.818328332901001,
      "epoch": 3.989453691208077,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003492834329838854,
      "loss": 5.276,
      "mean_token_accuracy": 0.21329102516174317,
      "num_tokens": 78075224.0,
      "step": 39910
    },
    {
      "entropy": 5.8707129001617435,
      "epoch": 3.989953516269306,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034924974407246005,
      "loss": 5.3518,
      "mean_token_accuracy": 0.20689719766378403,
      "num_tokens": 78085090.0,
      "step": 39915
    },
    {
      "entropy": 5.829649591445923,
      "epoch": 3.990453341330534,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034921605329293025,
      "loss": 5.3657,
      "mean_token_accuracy": 0.21115216612815857,
      "num_tokens": 78094719.0,
      "step": 39920
    },
    {
      "entropy": 5.810582160949707,
      "epoch": 3.990953166391763,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034918236064614366,
      "loss": 5.3228,
      "mean_token_accuracy": 0.20634027272462846,
      "num_tokens": 78103962.0,
      "step": 39925
    },
    {
      "entropy": 5.883274793624878,
      "epoch": 3.9914529914529915,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000349148666132948,
      "loss": 5.387,
      "mean_token_accuracy": 0.19801824390888215,
      "num_tokens": 78113562.0,
      "step": 39930
    },
    {
      "entropy": 5.890722131729126,
      "epoch": 3.99195281651422,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003491149697541909,
      "loss": 5.3083,
      "mean_token_accuracy": 0.21593900620937348,
      "num_tokens": 78123042.0,
      "step": 39935
    },
    {
      "entropy": 5.859169864654541,
      "epoch": 3.9924526415754484,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034908127151072027,
      "loss": 5.3939,
      "mean_token_accuracy": 0.20349923968315126,
      "num_tokens": 78132243.0,
      "step": 39940
    },
    {
      "entropy": 5.845893716812133,
      "epoch": 3.992952466636677,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003490475714033839,
      "loss": 5.295,
      "mean_token_accuracy": 0.2173075333237648,
      "num_tokens": 78140337.0,
      "step": 39945
    },
    {
      "entropy": 5.7949076175689695,
      "epoch": 3.993452291697906,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034901386943302976,
      "loss": 5.2723,
      "mean_token_accuracy": 0.20530973076820375,
      "num_tokens": 78150357.0,
      "step": 39950
    },
    {
      "entropy": 5.859245538711548,
      "epoch": 3.993952116759134,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003489801656005057,
      "loss": 5.4059,
      "mean_token_accuracy": 0.20004096627235413,
      "num_tokens": 78159391.0,
      "step": 39955
    },
    {
      "entropy": 5.838795709609985,
      "epoch": 3.994451941820363,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034894645990665957,
      "loss": 5.3627,
      "mean_token_accuracy": 0.20815386176109313,
      "num_tokens": 78168824.0,
      "step": 39960
    },
    {
      "entropy": 5.800304794311524,
      "epoch": 3.9949517668815915,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034891275235233966,
      "loss": 5.2793,
      "mean_token_accuracy": 0.20747923851013184,
      "num_tokens": 78180223.0,
      "step": 39965
    },
    {
      "entropy": 5.9263145446777346,
      "epoch": 3.99545159194282,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034887904293839385,
      "loss": 5.381,
      "mean_token_accuracy": 0.20755790770053864,
      "num_tokens": 78191306.0,
      "step": 39970
    },
    {
      "entropy": 5.822702646255493,
      "epoch": 3.9959514170040484,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003488453316656704,
      "loss": 5.269,
      "mean_token_accuracy": 0.2152137950062752,
      "num_tokens": 78200147.0,
      "step": 39975
    },
    {
      "entropy": 5.860767793655396,
      "epoch": 3.996451242065277,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003488116185350174,
      "loss": 5.3377,
      "mean_token_accuracy": 0.20526743233203887,
      "num_tokens": 78209631.0,
      "step": 39980
    },
    {
      "entropy": 5.823101043701172,
      "epoch": 3.996951067126506,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034877790354728304,
      "loss": 5.3606,
      "mean_token_accuracy": 0.2027094691991806,
      "num_tokens": 78221030.0,
      "step": 39985
    },
    {
      "entropy": 5.780029201507569,
      "epoch": 3.997450892187734,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003487441867033156,
      "loss": 5.2687,
      "mean_token_accuracy": 0.22575363516807556,
      "num_tokens": 78229794.0,
      "step": 39990
    },
    {
      "entropy": 5.869879150390625,
      "epoch": 3.997950717248963,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003487104680039634,
      "loss": 5.3714,
      "mean_token_accuracy": 0.20358959436416627,
      "num_tokens": 78239052.0,
      "step": 39995
    },
    {
      "entropy": 5.827382230758667,
      "epoch": 3.9984505423101915,
      "grad_norm": 1.0,
      "learning_rate": 0.0003486767474500748,
      "loss": 5.3406,
      "mean_token_accuracy": 0.21542631536722184,
      "num_tokens": 78249106.0,
      "step": 40000
    },
    {
      "entropy": 5.864394474029541,
      "epoch": 3.99895036737142,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034864302504249823,
      "loss": 5.3605,
      "mean_token_accuracy": 0.20565336495637893,
      "num_tokens": 78258236.0,
      "step": 40005
    },
    {
      "entropy": 5.868437385559082,
      "epoch": 3.9994501924326484,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034860930078208207,
      "loss": 5.3917,
      "mean_token_accuracy": 0.20510825961828233,
      "num_tokens": 78267812.0,
      "step": 40010
    },
    {
      "entropy": 5.83149733543396,
      "epoch": 3.999950017493877,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034857557466967486,
      "loss": 5.3613,
      "mean_token_accuracy": 0.20586346238851547,
      "num_tokens": 78277485.0,
      "step": 40015
    },
    {
      "entropy": 5.790497832828098,
      "epoch": 4.000399860048983,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034854184670612514,
      "loss": 5.2684,
      "mean_token_accuracy": 0.21224832865926954,
      "num_tokens": 78286105.0,
      "step": 40020
    },
    {
      "entropy": 5.848835611343384,
      "epoch": 4.000899685110212,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003485081168922814,
      "loss": 5.2186,
      "mean_token_accuracy": 0.2150359719991684,
      "num_tokens": 78295080.0,
      "step": 40025
    },
    {
      "entropy": 5.757689046859741,
      "epoch": 4.00139951017144,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034847438522899246,
      "loss": 5.1779,
      "mean_token_accuracy": 0.22448572516441345,
      "num_tokens": 78304703.0,
      "step": 40030
    },
    {
      "entropy": 5.8614662170410154,
      "epoch": 4.001899335232668,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003484406517171068,
      "loss": 5.2875,
      "mean_token_accuracy": 0.2066348299384117,
      "num_tokens": 78313986.0,
      "step": 40035
    },
    {
      "entropy": 5.845341444015503,
      "epoch": 4.002399160293897,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003484069163574733,
      "loss": 5.3017,
      "mean_token_accuracy": 0.21357030272483826,
      "num_tokens": 78324453.0,
      "step": 40040
    },
    {
      "entropy": 5.811670923233033,
      "epoch": 4.002898985355126,
      "grad_norm": 1.125,
      "learning_rate": 0.00034837317915094063,
      "loss": 5.2657,
      "mean_token_accuracy": 0.2205457404255867,
      "num_tokens": 78334451.0,
      "step": 40045
    },
    {
      "entropy": 5.742589139938355,
      "epoch": 4.003398810416354,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034833944009835774,
      "loss": 5.073,
      "mean_token_accuracy": 0.2251396119594574,
      "num_tokens": 78344460.0,
      "step": 40050
    },
    {
      "entropy": 5.842817068099976,
      "epoch": 4.003898635477583,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034830569920057323,
      "loss": 5.2829,
      "mean_token_accuracy": 0.21528926491737366,
      "num_tokens": 78355110.0,
      "step": 40055
    },
    {
      "entropy": 5.8742951393127445,
      "epoch": 4.004398460538812,
      "grad_norm": 1.125,
      "learning_rate": 0.00034827195645843636,
      "loss": 5.2971,
      "mean_token_accuracy": 0.2101044699549675,
      "num_tokens": 78365394.0,
      "step": 40060
    },
    {
      "entropy": 5.848334312438965,
      "epoch": 4.00489828560004,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003482382118727959,
      "loss": 5.2428,
      "mean_token_accuracy": 0.21027636528015137,
      "num_tokens": 78375196.0,
      "step": 40065
    },
    {
      "entropy": 5.834680843353271,
      "epoch": 4.005398110661268,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003482044654445008,
      "loss": 5.2854,
      "mean_token_accuracy": 0.21082955747842788,
      "num_tokens": 78386065.0,
      "step": 40070
    },
    {
      "entropy": 5.844271326065064,
      "epoch": 4.005897935722497,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034817071717440024,
      "loss": 5.2769,
      "mean_token_accuracy": 0.21738117635250093,
      "num_tokens": 78396070.0,
      "step": 40075
    },
    {
      "entropy": 5.873344039916992,
      "epoch": 4.006397760783726,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003481369670633433,
      "loss": 5.2867,
      "mean_token_accuracy": 0.21127459406852722,
      "num_tokens": 78405680.0,
      "step": 40080
    },
    {
      "entropy": 5.825518417358398,
      "epoch": 4.006897585844954,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034810321511217894,
      "loss": 5.2903,
      "mean_token_accuracy": 0.21655621379613876,
      "num_tokens": 78414884.0,
      "step": 40085
    },
    {
      "entropy": 5.729511451721192,
      "epoch": 4.007397410906183,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003480694613217567,
      "loss": 5.1755,
      "mean_token_accuracy": 0.2197800800204277,
      "num_tokens": 78425530.0,
      "step": 40090
    },
    {
      "entropy": 5.815819120407104,
      "epoch": 4.007897235967412,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003480357056929255,
      "loss": 5.2426,
      "mean_token_accuracy": 0.21665744930505754,
      "num_tokens": 78434852.0,
      "step": 40095
    },
    {
      "entropy": 5.826985645294189,
      "epoch": 4.00839706102864,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034800194822653474,
      "loss": 5.2402,
      "mean_token_accuracy": 0.21682180613279342,
      "num_tokens": 78444900.0,
      "step": 40100
    },
    {
      "entropy": 5.851126480102539,
      "epoch": 4.008896886089868,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034796818892343374,
      "loss": 5.3006,
      "mean_token_accuracy": 0.2127360537648201,
      "num_tokens": 78455048.0,
      "step": 40105
    },
    {
      "entropy": 5.831732845306396,
      "epoch": 4.009396711151097,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034793442778447196,
      "loss": 5.2327,
      "mean_token_accuracy": 0.2161231130361557,
      "num_tokens": 78463227.0,
      "step": 40110
    },
    {
      "entropy": 5.85213532447815,
      "epoch": 4.009896536212326,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003479006648104987,
      "loss": 5.1654,
      "mean_token_accuracy": 0.22003955394029617,
      "num_tokens": 78473297.0,
      "step": 40115
    },
    {
      "entropy": 5.780253124237061,
      "epoch": 4.010396361273554,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003478669000023634,
      "loss": 5.188,
      "mean_token_accuracy": 0.2179523304104805,
      "num_tokens": 78482562.0,
      "step": 40120
    },
    {
      "entropy": 5.790191316604615,
      "epoch": 4.010896186334783,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034783313336091576,
      "loss": 5.2714,
      "mean_token_accuracy": 0.21428786665201188,
      "num_tokens": 78491816.0,
      "step": 40125
    },
    {
      "entropy": 5.794180536270142,
      "epoch": 4.011396011396012,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034779936488700514,
      "loss": 5.2035,
      "mean_token_accuracy": 0.21900035738945006,
      "num_tokens": 78501461.0,
      "step": 40130
    },
    {
      "entropy": 5.7887622833251955,
      "epoch": 4.01189583645724,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003477655945814812,
      "loss": 5.1568,
      "mean_token_accuracy": 0.22565157264471053,
      "num_tokens": 78510529.0,
      "step": 40135
    },
    {
      "entropy": 5.837430620193482,
      "epoch": 4.012395661518468,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034773182244519376,
      "loss": 5.281,
      "mean_token_accuracy": 0.21019767969846725,
      "num_tokens": 78520745.0,
      "step": 40140
    },
    {
      "entropy": 5.743340301513672,
      "epoch": 4.012895486579697,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003476980484789923,
      "loss": 5.15,
      "mean_token_accuracy": 0.22255434691905976,
      "num_tokens": 78530479.0,
      "step": 40145
    },
    {
      "entropy": 5.816694641113282,
      "epoch": 4.0133953116409256,
      "grad_norm": 1.25,
      "learning_rate": 0.0003476642726837267,
      "loss": 5.3745,
      "mean_token_accuracy": 0.20985351800918578,
      "num_tokens": 78540027.0,
      "step": 40150
    },
    {
      "entropy": 6.002013206481934,
      "epoch": 4.013895136702154,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034763049506024657,
      "loss": 5.481,
      "mean_token_accuracy": 0.19337910562753677,
      "num_tokens": 78549136.0,
      "step": 40155
    },
    {
      "entropy": 5.95108118057251,
      "epoch": 4.014394961763383,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000347596715609402,
      "loss": 5.353,
      "mean_token_accuracy": 0.20810507386922836,
      "num_tokens": 78560260.0,
      "step": 40160
    },
    {
      "entropy": 5.9076474666595455,
      "epoch": 4.014894786824612,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003475629343320427,
      "loss": 5.3481,
      "mean_token_accuracy": 0.20030403286218643,
      "num_tokens": 78570164.0,
      "step": 40165
    },
    {
      "entropy": 5.870594501495361,
      "epoch": 4.01539461188584,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034752915122901863,
      "loss": 5.2929,
      "mean_token_accuracy": 0.21271581053733826,
      "num_tokens": 78580541.0,
      "step": 40170
    },
    {
      "entropy": 5.851556968688965,
      "epoch": 4.015894436947068,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034749536630117984,
      "loss": 5.2535,
      "mean_token_accuracy": 0.22343693673610687,
      "num_tokens": 78589374.0,
      "step": 40175
    },
    {
      "entropy": 5.824409532546997,
      "epoch": 4.016394262008297,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003474615795493763,
      "loss": 5.2529,
      "mean_token_accuracy": 0.2161538064479828,
      "num_tokens": 78598698.0,
      "step": 40180
    },
    {
      "entropy": 5.864893913269043,
      "epoch": 4.0168940870695256,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034742779097445795,
      "loss": 5.2795,
      "mean_token_accuracy": 0.21919962912797927,
      "num_tokens": 78607437.0,
      "step": 40185
    },
    {
      "entropy": 5.813165473937988,
      "epoch": 4.017393912130754,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003473940005772751,
      "loss": 5.2347,
      "mean_token_accuracy": 0.21776341497898102,
      "num_tokens": 78617801.0,
      "step": 40190
    },
    {
      "entropy": 5.877412414550781,
      "epoch": 4.017893737191983,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034736020835867776,
      "loss": 5.3037,
      "mean_token_accuracy": 0.20942740887403488,
      "num_tokens": 78627048.0,
      "step": 40195
    },
    {
      "entropy": 5.834121370315552,
      "epoch": 4.018393562253212,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034732641431951626,
      "loss": 5.3183,
      "mean_token_accuracy": 0.20840972363948823,
      "num_tokens": 78637042.0,
      "step": 40200
    },
    {
      "entropy": 5.802094221115112,
      "epoch": 4.01889338731444,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003472926184606408,
      "loss": 5.2602,
      "mean_token_accuracy": 0.21201143860816957,
      "num_tokens": 78647115.0,
      "step": 40205
    },
    {
      "entropy": 5.876476097106933,
      "epoch": 4.019393212375668,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034725882078290163,
      "loss": 5.3072,
      "mean_token_accuracy": 0.20784493684768676,
      "num_tokens": 78656849.0,
      "step": 40210
    },
    {
      "entropy": 5.93442964553833,
      "epoch": 4.019893037436897,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003472250212871491,
      "loss": 5.4011,
      "mean_token_accuracy": 0.2010220155119896,
      "num_tokens": 78666685.0,
      "step": 40215
    },
    {
      "entropy": 5.822517204284668,
      "epoch": 4.0203928624981256,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034719121997423366,
      "loss": 5.2907,
      "mean_token_accuracy": 0.20995651483535765,
      "num_tokens": 78676424.0,
      "step": 40220
    },
    {
      "entropy": 5.853623580932617,
      "epoch": 4.020892687559354,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003471574168450057,
      "loss": 5.2641,
      "mean_token_accuracy": 0.2105265289545059,
      "num_tokens": 78686506.0,
      "step": 40225
    },
    {
      "entropy": 5.8723955154418945,
      "epoch": 4.021392512620583,
      "grad_norm": 0.984375,
      "learning_rate": 0.00034712361190031564,
      "loss": 5.298,
      "mean_token_accuracy": 0.21109095811843873,
      "num_tokens": 78696480.0,
      "step": 40230
    },
    {
      "entropy": 5.64198489189148,
      "epoch": 4.021892337681812,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003470898051410141,
      "loss": 5.0787,
      "mean_token_accuracy": 0.23142725676298143,
      "num_tokens": 78706081.0,
      "step": 40235
    },
    {
      "entropy": 5.829466819763184,
      "epoch": 4.02239216274304,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003470559965679516,
      "loss": 5.2826,
      "mean_token_accuracy": 0.20975716412067413,
      "num_tokens": 78715212.0,
      "step": 40240
    },
    {
      "entropy": 5.806111574172974,
      "epoch": 4.022891987804268,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034702218618197885,
      "loss": 5.1972,
      "mean_token_accuracy": 0.2283096969127655,
      "num_tokens": 78725989.0,
      "step": 40245
    },
    {
      "entropy": 5.834935569763184,
      "epoch": 4.023391812865497,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003469883739839463,
      "loss": 5.2577,
      "mean_token_accuracy": 0.21508389562368393,
      "num_tokens": 78737119.0,
      "step": 40250
    },
    {
      "entropy": 5.942488050460815,
      "epoch": 4.0238916379267256,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034695455997470486,
      "loss": 5.3311,
      "mean_token_accuracy": 0.20892323255538942,
      "num_tokens": 78745917.0,
      "step": 40255
    },
    {
      "entropy": 5.907863521575928,
      "epoch": 4.024391462987954,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034692074415510523,
      "loss": 5.3166,
      "mean_token_accuracy": 0.21076602190732957,
      "num_tokens": 78756232.0,
      "step": 40260
    },
    {
      "entropy": 5.842274284362793,
      "epoch": 4.024891288049183,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034688692652599815,
      "loss": 5.3276,
      "mean_token_accuracy": 0.21032311916351318,
      "num_tokens": 78767533.0,
      "step": 40265
    },
    {
      "entropy": 5.854567289352417,
      "epoch": 4.025391113110412,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003468531070882345,
      "loss": 5.333,
      "mean_token_accuracy": 0.2088889330625534,
      "num_tokens": 78776937.0,
      "step": 40270
    },
    {
      "entropy": 5.802263355255127,
      "epoch": 4.02589093817164,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034681928584266513,
      "loss": 5.2194,
      "mean_token_accuracy": 0.21957493126392363,
      "num_tokens": 78786547.0,
      "step": 40275
    },
    {
      "entropy": 5.808887195587158,
      "epoch": 4.026390763232868,
      "grad_norm": 1.03125,
      "learning_rate": 0.000346785462790141,
      "loss": 5.2996,
      "mean_token_accuracy": 0.21440170258283614,
      "num_tokens": 78796634.0,
      "step": 40280
    },
    {
      "entropy": 5.827723169326783,
      "epoch": 4.026890588294097,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034675163793151313,
      "loss": 5.2468,
      "mean_token_accuracy": 0.2167795330286026,
      "num_tokens": 78806260.0,
      "step": 40285
    },
    {
      "entropy": 5.833159971237182,
      "epoch": 4.0273904133553255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034671781126763263,
      "loss": 5.2906,
      "mean_token_accuracy": 0.21383937895298005,
      "num_tokens": 78817244.0,
      "step": 40290
    },
    {
      "entropy": 5.73104076385498,
      "epoch": 4.027890238416554,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034668398279935034,
      "loss": 5.0679,
      "mean_token_accuracy": 0.22932544946670533,
      "num_tokens": 78827764.0,
      "step": 40295
    },
    {
      "entropy": 5.907945919036865,
      "epoch": 4.028390063477783,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034665015252751747,
      "loss": 5.3701,
      "mean_token_accuracy": 0.2095827803015709,
      "num_tokens": 78837847.0,
      "step": 40300
    },
    {
      "entropy": 5.8373037815094,
      "epoch": 4.028889888539012,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003466163204529852,
      "loss": 5.2456,
      "mean_token_accuracy": 0.21870262622833253,
      "num_tokens": 78847144.0,
      "step": 40305
    },
    {
      "entropy": 5.799698066711426,
      "epoch": 4.02938971360024,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034658248657660476,
      "loss": 5.262,
      "mean_token_accuracy": 0.21241694837808608,
      "num_tokens": 78858250.0,
      "step": 40310
    },
    {
      "entropy": 5.887700986862183,
      "epoch": 4.029889538661468,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003465486508992274,
      "loss": 5.3023,
      "mean_token_accuracy": 0.2131428375840187,
      "num_tokens": 78867790.0,
      "step": 40315
    },
    {
      "entropy": 5.776095151901245,
      "epoch": 4.030389363722697,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003465148134217043,
      "loss": 5.1914,
      "mean_token_accuracy": 0.2223162531852722,
      "num_tokens": 78879280.0,
      "step": 40320
    },
    {
      "entropy": 5.750374221801758,
      "epoch": 4.0308891887839255,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003464809741448871,
      "loss": 5.2203,
      "mean_token_accuracy": 0.216678985953331,
      "num_tokens": 78888516.0,
      "step": 40325
    },
    {
      "entropy": 5.862652254104614,
      "epoch": 4.031389013845154,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003464471330696268,
      "loss": 5.3072,
      "mean_token_accuracy": 0.20690243393182756,
      "num_tokens": 78898420.0,
      "step": 40330
    },
    {
      "entropy": 5.893093156814575,
      "epoch": 4.031888838906383,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003464132901967751,
      "loss": 5.3471,
      "mean_token_accuracy": 0.2013581618666649,
      "num_tokens": 78907957.0,
      "step": 40335
    },
    {
      "entropy": 5.857439661026001,
      "epoch": 4.032388663967612,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003463794455271833,
      "loss": 5.2575,
      "mean_token_accuracy": 0.21735070943832396,
      "num_tokens": 78917488.0,
      "step": 40340
    },
    {
      "entropy": 5.822798013687134,
      "epoch": 4.03288848902884,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034634559906170304,
      "loss": 5.2766,
      "mean_token_accuracy": 0.20564620941877365,
      "num_tokens": 78927115.0,
      "step": 40345
    },
    {
      "entropy": 5.81534743309021,
      "epoch": 4.033388314090068,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000346311750801186,
      "loss": 5.1999,
      "mean_token_accuracy": 0.22066525965929032,
      "num_tokens": 78938170.0,
      "step": 40350
    },
    {
      "entropy": 5.9147193908691404,
      "epoch": 4.033888139151297,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003462779007464835,
      "loss": 5.3318,
      "mean_token_accuracy": 0.20444214940071107,
      "num_tokens": 78948336.0,
      "step": 40355
    },
    {
      "entropy": 5.881679677963257,
      "epoch": 4.0343879642125255,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003462440488984475,
      "loss": 5.2805,
      "mean_token_accuracy": 0.21710919141769408,
      "num_tokens": 78957736.0,
      "step": 40360
    },
    {
      "entropy": 5.85637640953064,
      "epoch": 4.034887789273754,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034621019525792946,
      "loss": 5.4119,
      "mean_token_accuracy": 0.2005866900086403,
      "num_tokens": 78967055.0,
      "step": 40365
    },
    {
      "entropy": 5.881426095962524,
      "epoch": 4.035387614334983,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034617633982578124,
      "loss": 5.2669,
      "mean_token_accuracy": 0.21214248687028886,
      "num_tokens": 78976646.0,
      "step": 40370
    },
    {
      "entropy": 5.862194776535034,
      "epoch": 4.035887439396212,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003461424826028546,
      "loss": 5.2237,
      "mean_token_accuracy": 0.20839643925428392,
      "num_tokens": 78985123.0,
      "step": 40375
    },
    {
      "entropy": 5.9079076766967775,
      "epoch": 4.03638726445744,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034610862359000144,
      "loss": 5.3724,
      "mean_token_accuracy": 0.20814478546380996,
      "num_tokens": 78994229.0,
      "step": 40380
    },
    {
      "entropy": 5.904277324676514,
      "epoch": 4.036887089518668,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003460747627880736,
      "loss": 5.3196,
      "mean_token_accuracy": 0.2097229391336441,
      "num_tokens": 79005028.0,
      "step": 40385
    },
    {
      "entropy": 5.89062466621399,
      "epoch": 4.037386914579897,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003460409001979231,
      "loss": 5.3282,
      "mean_token_accuracy": 0.20665955990552903,
      "num_tokens": 79015312.0,
      "step": 40390
    },
    {
      "entropy": 5.768894863128662,
      "epoch": 4.0378867396411255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034600703582040183,
      "loss": 5.1855,
      "mean_token_accuracy": 0.22133986055850982,
      "num_tokens": 79025216.0,
      "step": 40395
    },
    {
      "entropy": 5.859542512893677,
      "epoch": 4.038386564702354,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034597316965636173,
      "loss": 5.2757,
      "mean_token_accuracy": 0.21264679878950118,
      "num_tokens": 79034783.0,
      "step": 40400
    },
    {
      "entropy": 5.772621822357178,
      "epoch": 4.038886389763583,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034593930170665496,
      "loss": 5.1694,
      "mean_token_accuracy": 0.22567108720541001,
      "num_tokens": 79044677.0,
      "step": 40405
    },
    {
      "entropy": 5.804452419281006,
      "epoch": 4.039386214824812,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034590543197213355,
      "loss": 5.1804,
      "mean_token_accuracy": 0.2247866213321686,
      "num_tokens": 79053972.0,
      "step": 40410
    },
    {
      "entropy": 5.8429077625274655,
      "epoch": 4.0398860398860394,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034587156045364983,
      "loss": 5.2358,
      "mean_token_accuracy": 0.2194037675857544,
      "num_tokens": 79062848.0,
      "step": 40415
    },
    {
      "entropy": 5.872811412811279,
      "epoch": 4.040385864947268,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034583768715205584,
      "loss": 5.224,
      "mean_token_accuracy": 0.22160554826259612,
      "num_tokens": 79073032.0,
      "step": 40420
    },
    {
      "entropy": 5.708723878860473,
      "epoch": 4.040885690008497,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003458038120682038,
      "loss": 5.1449,
      "mean_token_accuracy": 0.22752390056848526,
      "num_tokens": 79082665.0,
      "step": 40425
    },
    {
      "entropy": 5.811044549942016,
      "epoch": 4.0413855150697255,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034576993520294623,
      "loss": 5.2929,
      "mean_token_accuracy": 0.21256163716316223,
      "num_tokens": 79093017.0,
      "step": 40430
    },
    {
      "entropy": 5.8389058113098145,
      "epoch": 4.041885340130954,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034573605655713514,
      "loss": 5.3095,
      "mean_token_accuracy": 0.20869916081428527,
      "num_tokens": 79102582.0,
      "step": 40435
    },
    {
      "entropy": 5.824842882156372,
      "epoch": 4.042385165192183,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034570217613162315,
      "loss": 5.2102,
      "mean_token_accuracy": 0.21806532442569732,
      "num_tokens": 79112130.0,
      "step": 40440
    },
    {
      "entropy": 5.830870866775513,
      "epoch": 4.042884990253412,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034566829392726256,
      "loss": 5.2337,
      "mean_token_accuracy": 0.21374700218439102,
      "num_tokens": 79121177.0,
      "step": 40445
    },
    {
      "entropy": 5.825283861160278,
      "epoch": 4.0433848153146394,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003456344099449059,
      "loss": 5.2569,
      "mean_token_accuracy": 0.22377942353487015,
      "num_tokens": 79131186.0,
      "step": 40450
    },
    {
      "entropy": 5.868358898162842,
      "epoch": 4.043884640375868,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034560052418540564,
      "loss": 5.2309,
      "mean_token_accuracy": 0.20874429047107695,
      "num_tokens": 79140376.0,
      "step": 40455
    },
    {
      "entropy": 5.906666851043701,
      "epoch": 4.044384465437097,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003455666366496145,
      "loss": 5.3334,
      "mean_token_accuracy": 0.21321525573730468,
      "num_tokens": 79150378.0,
      "step": 40460
    },
    {
      "entropy": 5.752831983566284,
      "epoch": 4.0448842904983255,
      "grad_norm": 1.25,
      "learning_rate": 0.00034553274733838476,
      "loss": 5.233,
      "mean_token_accuracy": 0.21489310264587402,
      "num_tokens": 79159235.0,
      "step": 40465
    },
    {
      "entropy": 5.900784730911255,
      "epoch": 4.045384115559554,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003454988562525693,
      "loss": 5.3995,
      "mean_token_accuracy": 0.2072784796357155,
      "num_tokens": 79170594.0,
      "step": 40470
    },
    {
      "entropy": 5.942449045181275,
      "epoch": 4.045883940620783,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034546496339302077,
      "loss": 5.3812,
      "mean_token_accuracy": 0.20474126935005188,
      "num_tokens": 79180529.0,
      "step": 40475
    },
    {
      "entropy": 5.8010626316070555,
      "epoch": 4.046383765682012,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003454310687605919,
      "loss": 5.2447,
      "mean_token_accuracy": 0.21277641505002975,
      "num_tokens": 79191349.0,
      "step": 40480
    },
    {
      "entropy": 5.863291597366333,
      "epoch": 4.0468835907432394,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034539717235613544,
      "loss": 5.3345,
      "mean_token_accuracy": 0.20912162363529205,
      "num_tokens": 79200983.0,
      "step": 40485
    },
    {
      "entropy": 5.848118734359741,
      "epoch": 4.047383415804468,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003453632741805042,
      "loss": 5.3003,
      "mean_token_accuracy": 0.21479937583208084,
      "num_tokens": 79210516.0,
      "step": 40490
    },
    {
      "entropy": 5.8319844722747805,
      "epoch": 4.047883240865697,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003453293742345512,
      "loss": 5.1906,
      "mean_token_accuracy": 0.22418806552886963,
      "num_tokens": 79220339.0,
      "step": 40495
    },
    {
      "entropy": 5.9600015640258786,
      "epoch": 4.0483830659269255,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034529547251912926,
      "loss": 5.3913,
      "mean_token_accuracy": 0.20192907452583314,
      "num_tokens": 79229775.0,
      "step": 40500
    },
    {
      "entropy": 5.8256532669067385,
      "epoch": 4.048882890988154,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034526156903509116,
      "loss": 5.2814,
      "mean_token_accuracy": 0.21203667521476746,
      "num_tokens": 79240794.0,
      "step": 40505
    },
    {
      "entropy": 5.861666536331176,
      "epoch": 4.049382716049383,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003452276637832902,
      "loss": 5.1952,
      "mean_token_accuracy": 0.22188727110624312,
      "num_tokens": 79251921.0,
      "step": 40510
    },
    {
      "entropy": 5.78933277130127,
      "epoch": 4.049882541110612,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003451937567645792,
      "loss": 5.2401,
      "mean_token_accuracy": 0.21093376278877257,
      "num_tokens": 79262219.0,
      "step": 40515
    },
    {
      "entropy": 5.846270132064819,
      "epoch": 4.050382366171839,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034515984797981136,
      "loss": 5.2626,
      "mean_token_accuracy": 0.22089436203241347,
      "num_tokens": 79273138.0,
      "step": 40520
    },
    {
      "entropy": 5.8503124713897705,
      "epoch": 4.050882191233068,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003451259374298398,
      "loss": 5.2944,
      "mean_token_accuracy": 0.20730898380279542,
      "num_tokens": 79283848.0,
      "step": 40525
    },
    {
      "entropy": 5.811955165863037,
      "epoch": 4.051382016294297,
      "grad_norm": 1.25,
      "learning_rate": 0.00034509202511551776,
      "loss": 5.1792,
      "mean_token_accuracy": 0.22028778940439225,
      "num_tokens": 79292618.0,
      "step": 40530
    },
    {
      "entropy": 5.889623308181763,
      "epoch": 4.0518818413555255,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003450581110376984,
      "loss": 5.352,
      "mean_token_accuracy": 0.20685619860887527,
      "num_tokens": 79301487.0,
      "step": 40535
    },
    {
      "entropy": 5.850316667556763,
      "epoch": 4.052381666416754,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003450241951972349,
      "loss": 5.2634,
      "mean_token_accuracy": 0.2128125622868538,
      "num_tokens": 79310987.0,
      "step": 40540
    },
    {
      "entropy": 5.789118146896362,
      "epoch": 4.052881491477983,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003449902775949807,
      "loss": 5.2454,
      "mean_token_accuracy": 0.21762560158967972,
      "num_tokens": 79320271.0,
      "step": 40545
    },
    {
      "entropy": 5.827892827987671,
      "epoch": 4.053381316539212,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034495635823178917,
      "loss": 5.2492,
      "mean_token_accuracy": 0.2120468184351921,
      "num_tokens": 79330487.0,
      "step": 40550
    },
    {
      "entropy": 5.831111860275269,
      "epoch": 4.053881141600439,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003449224371085136,
      "loss": 5.2341,
      "mean_token_accuracy": 0.22042573988437653,
      "num_tokens": 79340844.0,
      "step": 40555
    },
    {
      "entropy": 5.815172481536865,
      "epoch": 4.054380966661668,
      "grad_norm": 1.125,
      "learning_rate": 0.0003448885142260076,
      "loss": 5.2372,
      "mean_token_accuracy": 0.21532601863145828,
      "num_tokens": 79350854.0,
      "step": 40560
    },
    {
      "entropy": 5.788144826889038,
      "epoch": 4.054880791722897,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034485458958512444,
      "loss": 5.2238,
      "mean_token_accuracy": 0.2140686884522438,
      "num_tokens": 79360181.0,
      "step": 40565
    },
    {
      "entropy": 5.909129619598389,
      "epoch": 4.0553806167841255,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003448206631867179,
      "loss": 5.3378,
      "mean_token_accuracy": 0.2012469545006752,
      "num_tokens": 79369506.0,
      "step": 40570
    },
    {
      "entropy": 5.892251062393188,
      "epoch": 4.055880441845354,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034478673503164143,
      "loss": 5.3079,
      "mean_token_accuracy": 0.2167688086628914,
      "num_tokens": 79378548.0,
      "step": 40575
    },
    {
      "entropy": 5.708635044097901,
      "epoch": 4.056380266906583,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003447528051207486,
      "loss": 5.1349,
      "mean_token_accuracy": 0.22805341929197312,
      "num_tokens": 79387062.0,
      "step": 40580
    },
    {
      "entropy": 5.7520547866821286,
      "epoch": 4.056880091967812,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003447188734548932,
      "loss": 5.224,
      "mean_token_accuracy": 0.22103012055158616,
      "num_tokens": 79397380.0,
      "step": 40585
    },
    {
      "entropy": 5.89660758972168,
      "epoch": 4.057379917029039,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003446849400349289,
      "loss": 5.3795,
      "mean_token_accuracy": 0.20727864354848863,
      "num_tokens": 79408203.0,
      "step": 40590
    },
    {
      "entropy": 5.912954998016358,
      "epoch": 4.057879742090268,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003446510048617093,
      "loss": 5.298,
      "mean_token_accuracy": 0.20940227210521697,
      "num_tokens": 79418228.0,
      "step": 40595
    },
    {
      "entropy": 5.834698581695557,
      "epoch": 4.058379567151497,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003446170679360884,
      "loss": 5.2641,
      "mean_token_accuracy": 0.21607882380485535,
      "num_tokens": 79426968.0,
      "step": 40600
    },
    {
      "entropy": 5.864339780807495,
      "epoch": 4.0588793922127255,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003445831292589201,
      "loss": 5.3353,
      "mean_token_accuracy": 0.20442782938480378,
      "num_tokens": 79436672.0,
      "step": 40605
    },
    {
      "entropy": 5.798232173919677,
      "epoch": 4.059379217273954,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003445491888310581,
      "loss": 5.2326,
      "mean_token_accuracy": 0.21205345094203948,
      "num_tokens": 79446338.0,
      "step": 40610
    },
    {
      "entropy": 5.788650703430176,
      "epoch": 4.059879042335183,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003445152466533564,
      "loss": 5.2337,
      "mean_token_accuracy": 0.21960603296756745,
      "num_tokens": 79456487.0,
      "step": 40615
    },
    {
      "entropy": 5.815496015548706,
      "epoch": 4.060378867396412,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000344481302726669,
      "loss": 5.2828,
      "mean_token_accuracy": 0.2101607382297516,
      "num_tokens": 79467081.0,
      "step": 40620
    },
    {
      "entropy": 5.792648363113403,
      "epoch": 4.060878692457639,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034444735705184984,
      "loss": 5.1937,
      "mean_token_accuracy": 0.22054507285356523,
      "num_tokens": 79476815.0,
      "step": 40625
    },
    {
      "entropy": 5.923645687103272,
      "epoch": 4.061378517518868,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003444134096297531,
      "loss": 5.3199,
      "mean_token_accuracy": 0.21971730440855025,
      "num_tokens": 79487212.0,
      "step": 40630
    },
    {
      "entropy": 5.748879671096802,
      "epoch": 4.061878342580097,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003443794604612328,
      "loss": 5.1513,
      "mean_token_accuracy": 0.21411864459514618,
      "num_tokens": 79497270.0,
      "step": 40635
    },
    {
      "entropy": 5.741566514968872,
      "epoch": 4.0623781676413255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003443455095471432,
      "loss": 5.23,
      "mean_token_accuracy": 0.2087165355682373,
      "num_tokens": 79507281.0,
      "step": 40640
    },
    {
      "entropy": 5.830074405670166,
      "epoch": 4.062877992702554,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034431155688833837,
      "loss": 5.3486,
      "mean_token_accuracy": 0.2040898934006691,
      "num_tokens": 79517868.0,
      "step": 40645
    },
    {
      "entropy": 5.82875804901123,
      "epoch": 4.063377817763783,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003442776024856727,
      "loss": 5.2626,
      "mean_token_accuracy": 0.21737315654754638,
      "num_tokens": 79527932.0,
      "step": 40650
    },
    {
      "entropy": 5.721363210678101,
      "epoch": 4.063877642825012,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034424364634000023,
      "loss": 5.13,
      "mean_token_accuracy": 0.22308126240968704,
      "num_tokens": 79537993.0,
      "step": 40655
    },
    {
      "entropy": 5.916524648666382,
      "epoch": 4.064377467886239,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003442096884521754,
      "loss": 5.4265,
      "mean_token_accuracy": 0.1969923585653305,
      "num_tokens": 79548433.0,
      "step": 40660
    },
    {
      "entropy": 5.829930114746094,
      "epoch": 4.064877292947468,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034417572882305275,
      "loss": 5.2373,
      "mean_token_accuracy": 0.21434105187654495,
      "num_tokens": 79557986.0,
      "step": 40665
    },
    {
      "entropy": 5.750761842727661,
      "epoch": 4.065377118008697,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003441417674534865,
      "loss": 5.2377,
      "mean_token_accuracy": 0.21422447860240937,
      "num_tokens": 79569242.0,
      "step": 40670
    },
    {
      "entropy": 5.888313961029053,
      "epoch": 4.0658769430699255,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003441078043443313,
      "loss": 5.3289,
      "mean_token_accuracy": 0.21081772446632385,
      "num_tokens": 79579177.0,
      "step": 40675
    },
    {
      "entropy": 5.8358794212341305,
      "epoch": 4.066376768131154,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034407383949644146,
      "loss": 5.2336,
      "mean_token_accuracy": 0.2175933986902237,
      "num_tokens": 79588443.0,
      "step": 40680
    },
    {
      "entropy": 5.774157905578614,
      "epoch": 4.066876593192383,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003440398729106715,
      "loss": 5.1861,
      "mean_token_accuracy": 0.22930455952882767,
      "num_tokens": 79599130.0,
      "step": 40685
    },
    {
      "entropy": 5.901647615432739,
      "epoch": 4.067376418253612,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003440059045878761,
      "loss": 5.3021,
      "mean_token_accuracy": 0.1986316278576851,
      "num_tokens": 79608681.0,
      "step": 40690
    },
    {
      "entropy": 5.850658893585205,
      "epoch": 4.067876243314839,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034397193452891003,
      "loss": 5.2861,
      "mean_token_accuracy": 0.2025172919034958,
      "num_tokens": 79618994.0,
      "step": 40695
    },
    {
      "entropy": 5.844710111618042,
      "epoch": 4.068376068376068,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003439379627346277,
      "loss": 5.2728,
      "mean_token_accuracy": 0.21021255999803543,
      "num_tokens": 79628041.0,
      "step": 40700
    },
    {
      "entropy": 5.76519284248352,
      "epoch": 4.068875893437297,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034390398920588404,
      "loss": 5.2601,
      "mean_token_accuracy": 0.2121809735894203,
      "num_tokens": 79638109.0,
      "step": 40705
    },
    {
      "entropy": 5.870837116241455,
      "epoch": 4.0693757184985255,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003438700139435338,
      "loss": 5.3127,
      "mean_token_accuracy": 0.20933117419481279,
      "num_tokens": 79648595.0,
      "step": 40710
    },
    {
      "entropy": 5.878473472595215,
      "epoch": 4.069875543559754,
      "grad_norm": 1.25,
      "learning_rate": 0.0003438360369484317,
      "loss": 5.3486,
      "mean_token_accuracy": 0.20379365384578704,
      "num_tokens": 79658798.0,
      "step": 40715
    },
    {
      "entropy": 5.8525477886199955,
      "epoch": 4.070375368620983,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003438020582214325,
      "loss": 5.2706,
      "mean_token_accuracy": 0.21191053390502929,
      "num_tokens": 79669772.0,
      "step": 40720
    },
    {
      "entropy": 5.786713790893555,
      "epoch": 4.070875193682212,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003437680777633913,
      "loss": 5.2264,
      "mean_token_accuracy": 0.21310007870197295,
      "num_tokens": 79679408.0,
      "step": 40725
    },
    {
      "entropy": 5.921281099319458,
      "epoch": 4.071375018743439,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000343734095575163,
      "loss": 5.2878,
      "mean_token_accuracy": 0.20674212723970414,
      "num_tokens": 79689744.0,
      "step": 40730
    },
    {
      "entropy": 5.941121339797974,
      "epoch": 4.071874843804668,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003437001116576025,
      "loss": 5.3929,
      "mean_token_accuracy": 0.19878871142864227,
      "num_tokens": 79698521.0,
      "step": 40735
    },
    {
      "entropy": 5.783913803100586,
      "epoch": 4.072374668865897,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034366612601156486,
      "loss": 5.2412,
      "mean_token_accuracy": 0.22159783840179442,
      "num_tokens": 79709698.0,
      "step": 40740
    },
    {
      "entropy": 5.890796709060669,
      "epoch": 4.0728744939271255,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034363213863790527,
      "loss": 5.2817,
      "mean_token_accuracy": 0.21193003803491592,
      "num_tokens": 79719695.0,
      "step": 40745
    },
    {
      "entropy": 5.923282337188721,
      "epoch": 4.073374318988354,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034359814953747856,
      "loss": 5.3771,
      "mean_token_accuracy": 0.20374735891819,
      "num_tokens": 79730332.0,
      "step": 40750
    },
    {
      "entropy": 5.792009592056274,
      "epoch": 4.073874144049583,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034356415871114013,
      "loss": 5.2535,
      "mean_token_accuracy": 0.20986856520175934,
      "num_tokens": 79741794.0,
      "step": 40755
    },
    {
      "entropy": 5.799459838867188,
      "epoch": 4.074373969110812,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034353016615974513,
      "loss": 5.2588,
      "mean_token_accuracy": 0.22261598110198974,
      "num_tokens": 79751615.0,
      "step": 40760
    },
    {
      "entropy": 5.853361654281616,
      "epoch": 4.074873794172039,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003434961718841488,
      "loss": 5.3019,
      "mean_token_accuracy": 0.2099449008703232,
      "num_tokens": 79761146.0,
      "step": 40765
    },
    {
      "entropy": 5.958960247039795,
      "epoch": 4.075373619233268,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034346217588520636,
      "loss": 5.4171,
      "mean_token_accuracy": 0.19650344401597977,
      "num_tokens": 79771495.0,
      "step": 40770
    },
    {
      "entropy": 5.728580045700073,
      "epoch": 4.075873444294497,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003434281781637732,
      "loss": 5.1456,
      "mean_token_accuracy": 0.21961805671453477,
      "num_tokens": 79781192.0,
      "step": 40775
    },
    {
      "entropy": 5.899002933502198,
      "epoch": 4.0763732693557255,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034339417872070464,
      "loss": 5.2906,
      "mean_token_accuracy": 0.21149952262639998,
      "num_tokens": 79790534.0,
      "step": 40780
    },
    {
      "entropy": 5.768487548828125,
      "epoch": 4.076873094416954,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034336017755685623,
      "loss": 5.2899,
      "mean_token_accuracy": 0.2105783775448799,
      "num_tokens": 79800026.0,
      "step": 40785
    },
    {
      "entropy": 5.7631697177886965,
      "epoch": 4.077372919478183,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003433261746730833,
      "loss": 5.1847,
      "mean_token_accuracy": 0.21798743307590485,
      "num_tokens": 79809911.0,
      "step": 40790
    },
    {
      "entropy": 5.86497311592102,
      "epoch": 4.077872744539412,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034329217007024145,
      "loss": 5.273,
      "mean_token_accuracy": 0.20922456532716752,
      "num_tokens": 79821370.0,
      "step": 40795
    },
    {
      "entropy": 5.844345664978027,
      "epoch": 4.078372569600639,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003432581637491862,
      "loss": 5.2539,
      "mean_token_accuracy": 0.21033208966255187,
      "num_tokens": 79831131.0,
      "step": 40800
    },
    {
      "entropy": 5.807506799697876,
      "epoch": 4.078872394661868,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000343224155710773,
      "loss": 5.293,
      "mean_token_accuracy": 0.20901768505573273,
      "num_tokens": 79840187.0,
      "step": 40805
    },
    {
      "entropy": 5.8242298603057865,
      "epoch": 4.079372219723097,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034319014595585764,
      "loss": 5.3183,
      "mean_token_accuracy": 0.20665805637836457,
      "num_tokens": 79850741.0,
      "step": 40810
    },
    {
      "entropy": 5.892213439941406,
      "epoch": 4.0798720447843255,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034315613448529574,
      "loss": 5.3033,
      "mean_token_accuracy": 0.20638805031776428,
      "num_tokens": 79861324.0,
      "step": 40815
    },
    {
      "entropy": 5.817716360092163,
      "epoch": 4.080371869845554,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003431221212999431,
      "loss": 5.1891,
      "mean_token_accuracy": 0.21195404827594758,
      "num_tokens": 79871612.0,
      "step": 40820
    },
    {
      "entropy": 5.776638746261597,
      "epoch": 4.080871694906783,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034308810640065534,
      "loss": 5.2357,
      "mean_token_accuracy": 0.21630367785692214,
      "num_tokens": 79881693.0,
      "step": 40825
    },
    {
      "entropy": 5.829599094390869,
      "epoch": 4.081371519968012,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003430540897882884,
      "loss": 5.3078,
      "mean_token_accuracy": 0.2048252508044243,
      "num_tokens": 79891732.0,
      "step": 40830
    },
    {
      "entropy": 5.852621746063233,
      "epoch": 4.081871345029239,
      "grad_norm": 1.25,
      "learning_rate": 0.000343020071463698,
      "loss": 5.3422,
      "mean_token_accuracy": 0.2107517272233963,
      "num_tokens": 79899866.0,
      "step": 40835
    },
    {
      "entropy": 5.766010904312134,
      "epoch": 4.082371170090468,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034298605142774014,
      "loss": 5.2112,
      "mean_token_accuracy": 0.21634391099214553,
      "num_tokens": 79909611.0,
      "step": 40840
    },
    {
      "entropy": 5.791688060760498,
      "epoch": 4.082870995151697,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003429520296812708,
      "loss": 5.247,
      "mean_token_accuracy": 0.21464589089155198,
      "num_tokens": 79918986.0,
      "step": 40845
    },
    {
      "entropy": 5.809788942337036,
      "epoch": 4.0833708202129255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003429180062251458,
      "loss": 5.2395,
      "mean_token_accuracy": 0.21121468394994736,
      "num_tokens": 79927447.0,
      "step": 40850
    },
    {
      "entropy": 5.8483460426330565,
      "epoch": 4.083870645274154,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003428839810602212,
      "loss": 5.3213,
      "mean_token_accuracy": 0.20050057619810105,
      "num_tokens": 79935746.0,
      "step": 40855
    },
    {
      "entropy": 5.8468665599823,
      "epoch": 4.084370470335383,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003428499541873532,
      "loss": 5.2943,
      "mean_token_accuracy": 0.2137825757265091,
      "num_tokens": 79945492.0,
      "step": 40860
    },
    {
      "entropy": 5.839045667648316,
      "epoch": 4.084870295396612,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034281592560739775,
      "loss": 5.2621,
      "mean_token_accuracy": 0.216185961663723,
      "num_tokens": 79954804.0,
      "step": 40865
    },
    {
      "entropy": 5.886052036285401,
      "epoch": 4.085370120457839,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003427818953212111,
      "loss": 5.382,
      "mean_token_accuracy": 0.20402353256940842,
      "num_tokens": 79965313.0,
      "step": 40870
    },
    {
      "entropy": 5.7699363231658936,
      "epoch": 4.085869945519068,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034274786332964943,
      "loss": 5.2087,
      "mean_token_accuracy": 0.2249932035803795,
      "num_tokens": 79975070.0,
      "step": 40875
    },
    {
      "entropy": 5.875635576248169,
      "epoch": 4.086369770580297,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003427138296335688,
      "loss": 5.367,
      "mean_token_accuracy": 0.20637949854135512,
      "num_tokens": 79985140.0,
      "step": 40880
    },
    {
      "entropy": 5.8807484149932865,
      "epoch": 4.0868695956415255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003426797942338258,
      "loss": 5.2929,
      "mean_token_accuracy": 0.2121660128235817,
      "num_tokens": 79995153.0,
      "step": 40885
    },
    {
      "entropy": 5.899441385269165,
      "epoch": 4.087369420702754,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003426457571312766,
      "loss": 5.3199,
      "mean_token_accuracy": 0.20373740941286086,
      "num_tokens": 80004578.0,
      "step": 40890
    },
    {
      "entropy": 5.855451059341431,
      "epoch": 4.087869245763983,
      "grad_norm": 1.125,
      "learning_rate": 0.00034261171832677744,
      "loss": 5.2786,
      "mean_token_accuracy": 0.22008763551712035,
      "num_tokens": 80014632.0,
      "step": 40895
    },
    {
      "entropy": 5.84803581237793,
      "epoch": 4.088369070825211,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003425776778211849,
      "loss": 5.293,
      "mean_token_accuracy": 0.20610534846782685,
      "num_tokens": 80024687.0,
      "step": 40900
    },
    {
      "entropy": 5.840574312210083,
      "epoch": 4.088868895886439,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003425436356153554,
      "loss": 5.2826,
      "mean_token_accuracy": 0.20815173983573915,
      "num_tokens": 80034238.0,
      "step": 40905
    },
    {
      "entropy": 5.811600685119629,
      "epoch": 4.089368720947668,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034250959171014536,
      "loss": 5.2086,
      "mean_token_accuracy": 0.2201346278190613,
      "num_tokens": 80044733.0,
      "step": 40910
    },
    {
      "entropy": 5.802096271514893,
      "epoch": 4.089868546008897,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003424755461064114,
      "loss": 5.2648,
      "mean_token_accuracy": 0.21116699427366256,
      "num_tokens": 80054496.0,
      "step": 40915
    },
    {
      "entropy": 5.852148246765137,
      "epoch": 4.0903683710701255,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034244149880501007,
      "loss": 5.3913,
      "mean_token_accuracy": 0.2069518819451332,
      "num_tokens": 80064232.0,
      "step": 40920
    },
    {
      "entropy": 5.736364459991455,
      "epoch": 4.090868196131354,
      "grad_norm": 1.1875,
      "learning_rate": 0.000342407449806798,
      "loss": 5.1193,
      "mean_token_accuracy": 0.2252212196588516,
      "num_tokens": 80073552.0,
      "step": 40925
    },
    {
      "entropy": 5.741727066040039,
      "epoch": 4.091368021192583,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034237339911263176,
      "loss": 5.1992,
      "mean_token_accuracy": 0.21635818779468535,
      "num_tokens": 80083589.0,
      "step": 40930
    },
    {
      "entropy": 5.838922595977783,
      "epoch": 4.091867846253811,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034233934672336827,
      "loss": 5.3108,
      "mean_token_accuracy": 0.21698122173547746,
      "num_tokens": 80093494.0,
      "step": 40935
    },
    {
      "entropy": 5.9021491527557375,
      "epoch": 4.092367671315039,
      "grad_norm": 1.109375,
      "learning_rate": 0.000342305292639864,
      "loss": 5.3467,
      "mean_token_accuracy": 0.21058536171913148,
      "num_tokens": 80103438.0,
      "step": 40940
    },
    {
      "entropy": 5.864879131317139,
      "epoch": 4.092867496376268,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003422712368629759,
      "loss": 5.3734,
      "mean_token_accuracy": 0.20310524255037307,
      "num_tokens": 80114624.0,
      "step": 40945
    },
    {
      "entropy": 5.832849359512329,
      "epoch": 4.093367321437497,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034223717939356083,
      "loss": 5.2733,
      "mean_token_accuracy": 0.21088774055242537,
      "num_tokens": 80124655.0,
      "step": 40950
    },
    {
      "entropy": 5.815813255310059,
      "epoch": 4.0938671464987255,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003422031202324757,
      "loss": 5.1938,
      "mean_token_accuracy": 0.22568918019533157,
      "num_tokens": 80134055.0,
      "step": 40955
    },
    {
      "entropy": 5.8230626583099365,
      "epoch": 4.094366971559954,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003421690593805773,
      "loss": 5.3346,
      "mean_token_accuracy": 0.2115674138069153,
      "num_tokens": 80143627.0,
      "step": 40960
    },
    {
      "entropy": 5.765048170089722,
      "epoch": 4.094866796621183,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034213499683872264,
      "loss": 5.188,
      "mean_token_accuracy": 0.22460200190544127,
      "num_tokens": 80153193.0,
      "step": 40965
    },
    {
      "entropy": 5.777134370803833,
      "epoch": 4.095366621682411,
      "grad_norm": 1.125,
      "learning_rate": 0.00034210093260776865,
      "loss": 5.2247,
      "mean_token_accuracy": 0.21597013473510743,
      "num_tokens": 80163839.0,
      "step": 40970
    },
    {
      "entropy": 5.8527790069580075,
      "epoch": 4.095866446743639,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034206686668857255,
      "loss": 5.2746,
      "mean_token_accuracy": 0.20864399671554565,
      "num_tokens": 80173944.0,
      "step": 40975
    },
    {
      "entropy": 5.870839071273804,
      "epoch": 4.096366271804868,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034203279908199137,
      "loss": 5.3432,
      "mean_token_accuracy": 0.21042794734239578,
      "num_tokens": 80184595.0,
      "step": 40980
    },
    {
      "entropy": 5.748332786560058,
      "epoch": 4.096866096866097,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003419987297888822,
      "loss": 5.1745,
      "mean_token_accuracy": 0.22427140176296234,
      "num_tokens": 80195364.0,
      "step": 40985
    },
    {
      "entropy": 5.784897422790527,
      "epoch": 4.0973659219273255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003419646588101022,
      "loss": 5.1292,
      "mean_token_accuracy": 0.22276474982500077,
      "num_tokens": 80205283.0,
      "step": 40990
    },
    {
      "entropy": 5.832850742340088,
      "epoch": 4.097865746988554,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034193058614650866,
      "loss": 5.2565,
      "mean_token_accuracy": 0.20821433067321776,
      "num_tokens": 80214831.0,
      "step": 40995
    },
    {
      "entropy": 5.7409721374511715,
      "epoch": 4.098365572049783,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003418965117989587,
      "loss": 5.2388,
      "mean_token_accuracy": 0.21951255351305007,
      "num_tokens": 80224185.0,
      "step": 41000
    },
    {
      "entropy": 5.686510515213013,
      "epoch": 4.098865397111011,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034186243576830985,
      "loss": 5.1854,
      "mean_token_accuracy": 0.22993398904800416,
      "num_tokens": 80232927.0,
      "step": 41005
    },
    {
      "entropy": 5.830273151397705,
      "epoch": 4.099365222172239,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034182835805541916,
      "loss": 5.2669,
      "mean_token_accuracy": 0.21076445728540422,
      "num_tokens": 80242174.0,
      "step": 41010
    },
    {
      "entropy": 5.870353364944458,
      "epoch": 4.099865047233468,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003417942786611443,
      "loss": 5.3638,
      "mean_token_accuracy": 0.20125797092914582,
      "num_tokens": 80251344.0,
      "step": 41015
    },
    {
      "entropy": 5.814565944671631,
      "epoch": 4.100364872294697,
      "grad_norm": 1.125,
      "learning_rate": 0.0003417601975863425,
      "loss": 5.2506,
      "mean_token_accuracy": 0.2152709484100342,
      "num_tokens": 80262105.0,
      "step": 41020
    },
    {
      "entropy": 5.894606447219848,
      "epoch": 4.1008646973559255,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034172611483187143,
      "loss": 5.2915,
      "mean_token_accuracy": 0.2096600964665413,
      "num_tokens": 80271204.0,
      "step": 41025
    },
    {
      "entropy": 5.836553335189819,
      "epoch": 4.101364522417154,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003416920303985884,
      "loss": 5.2965,
      "mean_token_accuracy": 0.20707062482833863,
      "num_tokens": 80280033.0,
      "step": 41030
    },
    {
      "entropy": 5.723645973205566,
      "epoch": 4.101864347478383,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034165794428735104,
      "loss": 5.2311,
      "mean_token_accuracy": 0.2127441570162773,
      "num_tokens": 80289175.0,
      "step": 41035
    },
    {
      "entropy": 5.837332677841187,
      "epoch": 4.102364172539611,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034162385649901694,
      "loss": 5.291,
      "mean_token_accuracy": 0.20725503712892532,
      "num_tokens": 80298153.0,
      "step": 41040
    },
    {
      "entropy": 5.914792537689209,
      "epoch": 4.102863997600839,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034158976703444377,
      "loss": 5.3846,
      "mean_token_accuracy": 0.20217078179121017,
      "num_tokens": 80308076.0,
      "step": 41045
    },
    {
      "entropy": 5.774506521224976,
      "epoch": 4.103363822662068,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034155567589448925,
      "loss": 5.188,
      "mean_token_accuracy": 0.2252499058842659,
      "num_tokens": 80319546.0,
      "step": 41050
    },
    {
      "entropy": 5.851150178909302,
      "epoch": 4.103863647723297,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000341521583080011,
      "loss": 5.2489,
      "mean_token_accuracy": 0.20997188687324525,
      "num_tokens": 80328665.0,
      "step": 41055
    },
    {
      "entropy": 5.776686763763427,
      "epoch": 4.1043634727845255,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003414874885918669,
      "loss": 5.1539,
      "mean_token_accuracy": 0.21944438368082048,
      "num_tokens": 80338795.0,
      "step": 41060
    },
    {
      "entropy": 5.807731342315674,
      "epoch": 4.104863297845754,
      "grad_norm": 1.25,
      "learning_rate": 0.00034145339243091456,
      "loss": 5.2829,
      "mean_token_accuracy": 0.2096129208803177,
      "num_tokens": 80348835.0,
      "step": 41065
    },
    {
      "entropy": 5.888552284240722,
      "epoch": 4.105363122906983,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003414192945980121,
      "loss": 5.3449,
      "mean_token_accuracy": 0.212536758184433,
      "num_tokens": 80358028.0,
      "step": 41070
    },
    {
      "entropy": 5.922627544403076,
      "epoch": 4.105862947968211,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003413851950940172,
      "loss": 5.3491,
      "mean_token_accuracy": 0.2055051237344742,
      "num_tokens": 80367835.0,
      "step": 41075
    },
    {
      "entropy": 5.845208168029785,
      "epoch": 4.106362773029439,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034135109391978785,
      "loss": 5.2405,
      "mean_token_accuracy": 0.2156435117125511,
      "num_tokens": 80377673.0,
      "step": 41080
    },
    {
      "entropy": 5.795995283126831,
      "epoch": 4.106862598090668,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034131699107618215,
      "loss": 5.2649,
      "mean_token_accuracy": 0.2129230886697769,
      "num_tokens": 80387947.0,
      "step": 41085
    },
    {
      "entropy": 5.780635499954224,
      "epoch": 4.107362423151897,
      "grad_norm": 1.125,
      "learning_rate": 0.00034128288656405794,
      "loss": 5.1825,
      "mean_token_accuracy": 0.21592963337898255,
      "num_tokens": 80398204.0,
      "step": 41090
    },
    {
      "entropy": 5.8469470024108885,
      "epoch": 4.1078622482131255,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034124878038427337,
      "loss": 5.3087,
      "mean_token_accuracy": 0.2106024742126465,
      "num_tokens": 80407347.0,
      "step": 41095
    },
    {
      "entropy": 5.808268690109253,
      "epoch": 4.108362073274354,
      "grad_norm": 1.25,
      "learning_rate": 0.00034121467253768657,
      "loss": 5.2046,
      "mean_token_accuracy": 0.2152426764369011,
      "num_tokens": 80417051.0,
      "step": 41100
    },
    {
      "entropy": 5.925894689559937,
      "epoch": 4.108861898335583,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003411805630251556,
      "loss": 5.3789,
      "mean_token_accuracy": 0.20162371397018433,
      "num_tokens": 80426744.0,
      "step": 41105
    },
    {
      "entropy": 5.845630264282226,
      "epoch": 4.109361723396811,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034114645184753873,
      "loss": 5.294,
      "mean_token_accuracy": 0.2206195443868637,
      "num_tokens": 80436402.0,
      "step": 41110
    },
    {
      "entropy": 5.79308066368103,
      "epoch": 4.109861548458039,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034111233900569415,
      "loss": 5.1814,
      "mean_token_accuracy": 0.21916978061199188,
      "num_tokens": 80446323.0,
      "step": 41115
    },
    {
      "entropy": 5.87263855934143,
      "epoch": 4.110361373519268,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034107822450048014,
      "loss": 5.3592,
      "mean_token_accuracy": 0.20562680661678315,
      "num_tokens": 80456368.0,
      "step": 41120
    },
    {
      "entropy": 5.813266944885254,
      "epoch": 4.110861198580497,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000341044108332755,
      "loss": 5.2692,
      "mean_token_accuracy": 0.21514031291007996,
      "num_tokens": 80466603.0,
      "step": 41125
    },
    {
      "entropy": 5.851624393463135,
      "epoch": 4.1113610236417255,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034100999050337705,
      "loss": 5.2644,
      "mean_token_accuracy": 0.2191338449716568,
      "num_tokens": 80476926.0,
      "step": 41130
    },
    {
      "entropy": 5.812678432464599,
      "epoch": 4.111860848702954,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034097587101320474,
      "loss": 5.2431,
      "mean_token_accuracy": 0.22295210659503936,
      "num_tokens": 80486099.0,
      "step": 41135
    },
    {
      "entropy": 5.836238145828247,
      "epoch": 4.112360673764183,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034094174986309657,
      "loss": 5.3014,
      "mean_token_accuracy": 0.213235567510128,
      "num_tokens": 80496231.0,
      "step": 41140
    },
    {
      "entropy": 5.7943483829498295,
      "epoch": 4.112860498825411,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003409076270539108,
      "loss": 5.1312,
      "mean_token_accuracy": 0.22403812706470488,
      "num_tokens": 80505813.0,
      "step": 41145
    },
    {
      "entropy": 5.806009578704834,
      "epoch": 4.113360323886639,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003408735025865062,
      "loss": 5.3023,
      "mean_token_accuracy": 0.20683565884828567,
      "num_tokens": 80515138.0,
      "step": 41150
    },
    {
      "entropy": 5.856127119064331,
      "epoch": 4.113860148947868,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003408393764617412,
      "loss": 5.1982,
      "mean_token_accuracy": 0.22239865511655807,
      "num_tokens": 80526552.0,
      "step": 41155
    },
    {
      "entropy": 5.777579641342163,
      "epoch": 4.114359974009097,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034080524868047445,
      "loss": 5.2374,
      "mean_token_accuracy": 0.21987187564373017,
      "num_tokens": 80534968.0,
      "step": 41160
    },
    {
      "entropy": 5.93176040649414,
      "epoch": 4.1148597990703255,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003407711192435646,
      "loss": 5.4374,
      "mean_token_accuracy": 0.19462662041187287,
      "num_tokens": 80545050.0,
      "step": 41165
    },
    {
      "entropy": 5.781878805160522,
      "epoch": 4.115359624131554,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003407369881518702,
      "loss": 5.169,
      "mean_token_accuracy": 0.2319304198026657,
      "num_tokens": 80554283.0,
      "step": 41170
    },
    {
      "entropy": 5.852413177490234,
      "epoch": 4.115859449192783,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003407028554062502,
      "loss": 5.3628,
      "mean_token_accuracy": 0.2053602233529091,
      "num_tokens": 80564385.0,
      "step": 41175
    },
    {
      "entropy": 5.90172643661499,
      "epoch": 4.116359274254011,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003406687210075632,
      "loss": 5.3015,
      "mean_token_accuracy": 0.20761382430791855,
      "num_tokens": 80574323.0,
      "step": 41180
    },
    {
      "entropy": 5.819035577774048,
      "epoch": 4.116859099315239,
      "grad_norm": 1.125,
      "learning_rate": 0.0003406345849566681,
      "loss": 5.207,
      "mean_token_accuracy": 0.2169561579823494,
      "num_tokens": 80583446.0,
      "step": 41185
    },
    {
      "entropy": 5.817532062530518,
      "epoch": 4.117358924376468,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003406004472544237,
      "loss": 5.3249,
      "mean_token_accuracy": 0.21574763655662538,
      "num_tokens": 80593483.0,
      "step": 41190
    },
    {
      "entropy": 5.779722881317139,
      "epoch": 4.117858749437697,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003405663079016889,
      "loss": 5.2897,
      "mean_token_accuracy": 0.2064354106783867,
      "num_tokens": 80603799.0,
      "step": 41195
    },
    {
      "entropy": 5.762346315383911,
      "epoch": 4.1183585744989255,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003405321668993228,
      "loss": 5.1882,
      "mean_token_accuracy": 0.22140960693359374,
      "num_tokens": 80612972.0,
      "step": 41200
    },
    {
      "entropy": 5.868484449386597,
      "epoch": 4.118858399560154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003404980242481842,
      "loss": 5.2732,
      "mean_token_accuracy": 0.2108620896935463,
      "num_tokens": 80622683.0,
      "step": 41205
    },
    {
      "entropy": 5.797870826721192,
      "epoch": 4.119358224621383,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034046387994913216,
      "loss": 5.2,
      "mean_token_accuracy": 0.2242574691772461,
      "num_tokens": 80631745.0,
      "step": 41210
    },
    {
      "entropy": 5.80761866569519,
      "epoch": 4.119858049682611,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003404297340030257,
      "loss": 5.2011,
      "mean_token_accuracy": 0.21924546957015992,
      "num_tokens": 80640644.0,
      "step": 41215
    },
    {
      "entropy": 5.885293006896973,
      "epoch": 4.120357874743839,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034039558641072397,
      "loss": 5.3222,
      "mean_token_accuracy": 0.20607897788286209,
      "num_tokens": 80650389.0,
      "step": 41220
    },
    {
      "entropy": 5.712997484207153,
      "epoch": 4.120857699805068,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003403614371730861,
      "loss": 5.1344,
      "mean_token_accuracy": 0.21778422743082046,
      "num_tokens": 80659199.0,
      "step": 41225
    },
    {
      "entropy": 5.832421064376831,
      "epoch": 4.121357524866297,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034032728629097137,
      "loss": 5.2572,
      "mean_token_accuracy": 0.2149745225906372,
      "num_tokens": 80668767.0,
      "step": 41230
    },
    {
      "entropy": 5.850137710571289,
      "epoch": 4.1218573499275255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003402931337652389,
      "loss": 5.3539,
      "mean_token_accuracy": 0.19774605184793473,
      "num_tokens": 80680047.0,
      "step": 41235
    },
    {
      "entropy": 5.864271926879883,
      "epoch": 4.122357174988754,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000340258979596748,
      "loss": 5.3232,
      "mean_token_accuracy": 0.2040579155087471,
      "num_tokens": 80689143.0,
      "step": 41240
    },
    {
      "entropy": 5.865095615386963,
      "epoch": 4.122857000049983,
      "grad_norm": 1.125,
      "learning_rate": 0.0003402248237863579,
      "loss": 5.2087,
      "mean_token_accuracy": 0.22039630115032197,
      "num_tokens": 80698886.0,
      "step": 41245
    },
    {
      "entropy": 5.7233000755310055,
      "epoch": 4.123356825111211,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003401906663349281,
      "loss": 5.176,
      "mean_token_accuracy": 0.2276609405875206,
      "num_tokens": 80708545.0,
      "step": 41250
    },
    {
      "entropy": 5.816100358963013,
      "epoch": 4.123856650172439,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034015650724331776,
      "loss": 5.3238,
      "mean_token_accuracy": 0.20513066798448562,
      "num_tokens": 80718833.0,
      "step": 41255
    },
    {
      "entropy": 5.8287755966186525,
      "epoch": 4.124356475233668,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034012234651238654,
      "loss": 5.3129,
      "mean_token_accuracy": 0.21567070931196214,
      "num_tokens": 80728436.0,
      "step": 41260
    },
    {
      "entropy": 5.8563518047332765,
      "epoch": 4.124856300294897,
      "grad_norm": 1.125,
      "learning_rate": 0.0003400881841429939,
      "loss": 5.2847,
      "mean_token_accuracy": 0.21117523759603501,
      "num_tokens": 80738994.0,
      "step": 41265
    },
    {
      "entropy": 5.788137149810791,
      "epoch": 4.1253561253561255,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003400540201359992,
      "loss": 5.256,
      "mean_token_accuracy": 0.21135863661766052,
      "num_tokens": 80749414.0,
      "step": 41270
    },
    {
      "entropy": 5.798002767562866,
      "epoch": 4.125855950417354,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003400198544922622,
      "loss": 5.2694,
      "mean_token_accuracy": 0.21155430376529694,
      "num_tokens": 80759320.0,
      "step": 41275
    },
    {
      "entropy": 5.814213991165161,
      "epoch": 4.126355775478583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003399856872126424,
      "loss": 5.148,
      "mean_token_accuracy": 0.2202298492193222,
      "num_tokens": 80769020.0,
      "step": 41280
    },
    {
      "entropy": 5.869497919082642,
      "epoch": 4.126855600539811,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033995151829799926,
      "loss": 5.3876,
      "mean_token_accuracy": 0.20397696942090987,
      "num_tokens": 80779491.0,
      "step": 41285
    },
    {
      "entropy": 5.75635175704956,
      "epoch": 4.127355425601039,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033991734774919274,
      "loss": 5.2353,
      "mean_token_accuracy": 0.21742478013038635,
      "num_tokens": 80789941.0,
      "step": 41290
    },
    {
      "entropy": 5.799079704284668,
      "epoch": 4.127855250662268,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003398831755670825,
      "loss": 5.1337,
      "mean_token_accuracy": 0.22247946560382842,
      "num_tokens": 80798789.0,
      "step": 41295
    },
    {
      "entropy": 5.874913692474365,
      "epoch": 4.128355075723497,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003398490017525283,
      "loss": 5.3063,
      "mean_token_accuracy": 0.2047044113278389,
      "num_tokens": 80807664.0,
      "step": 41300
    },
    {
      "entropy": 5.883636713027954,
      "epoch": 4.1288549007847255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033981482630638976,
      "loss": 5.3959,
      "mean_token_accuracy": 0.20879462063312532,
      "num_tokens": 80817684.0,
      "step": 41305
    },
    {
      "entropy": 5.83868818283081,
      "epoch": 4.129354725845954,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033978064922952694,
      "loss": 5.2379,
      "mean_token_accuracy": 0.21673397868871688,
      "num_tokens": 80827424.0,
      "step": 41310
    },
    {
      "entropy": 5.766060304641724,
      "epoch": 4.129854550907183,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033974647052279963,
      "loss": 5.1597,
      "mean_token_accuracy": 0.23185315877199172,
      "num_tokens": 80836319.0,
      "step": 41315
    },
    {
      "entropy": 5.911260747909546,
      "epoch": 4.130354375968411,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003397122901870678,
      "loss": 5.366,
      "mean_token_accuracy": 0.20927503407001496,
      "num_tokens": 80847943.0,
      "step": 41320
    },
    {
      "entropy": 5.894875526428223,
      "epoch": 4.130854201029639,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003396781082231914,
      "loss": 5.3747,
      "mean_token_accuracy": 0.205142942070961,
      "num_tokens": 80859001.0,
      "step": 41325
    },
    {
      "entropy": 5.856731748580932,
      "epoch": 4.131354026090868,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033964392463203043,
      "loss": 5.2937,
      "mean_token_accuracy": 0.21275881826877593,
      "num_tokens": 80868571.0,
      "step": 41330
    },
    {
      "entropy": 5.769605588912964,
      "epoch": 4.131853851152097,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033960973941444503,
      "loss": 5.1864,
      "mean_token_accuracy": 0.21434153467416764,
      "num_tokens": 80877124.0,
      "step": 41335
    },
    {
      "entropy": 5.828870964050293,
      "epoch": 4.1323536762133255,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033957555257129514,
      "loss": 5.1945,
      "mean_token_accuracy": 0.2149093806743622,
      "num_tokens": 80886707.0,
      "step": 41340
    },
    {
      "entropy": 5.7099464416503904,
      "epoch": 4.132853501274554,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003395413641034409,
      "loss": 5.2282,
      "mean_token_accuracy": 0.22090836614370346,
      "num_tokens": 80896505.0,
      "step": 41345
    },
    {
      "entropy": 5.843861198425293,
      "epoch": 4.133353326335783,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003395071740117426,
      "loss": 5.286,
      "mean_token_accuracy": 0.20827908664941788,
      "num_tokens": 80905792.0,
      "step": 41350
    },
    {
      "entropy": 5.661135339736939,
      "epoch": 4.133853151397011,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033947298229706043,
      "loss": 5.0513,
      "mean_token_accuracy": 0.2379782646894455,
      "num_tokens": 80915091.0,
      "step": 41355
    },
    {
      "entropy": 5.796166944503784,
      "epoch": 4.134352976458239,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003394387889602546,
      "loss": 5.203,
      "mean_token_accuracy": 0.22038380801677704,
      "num_tokens": 80924831.0,
      "step": 41360
    },
    {
      "entropy": 5.822127008438111,
      "epoch": 4.134852801519468,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003394045940021854,
      "loss": 5.1944,
      "mean_token_accuracy": 0.2158053398132324,
      "num_tokens": 80934730.0,
      "step": 41365
    },
    {
      "entropy": 5.664041566848755,
      "epoch": 4.135352626580697,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033937039742371327,
      "loss": 5.0488,
      "mean_token_accuracy": 0.23203834742307664,
      "num_tokens": 80944420.0,
      "step": 41370
    },
    {
      "entropy": 5.715190172195435,
      "epoch": 4.1358524516419255,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033933619922569847,
      "loss": 5.1369,
      "mean_token_accuracy": 0.22649575024843216,
      "num_tokens": 80953494.0,
      "step": 41375
    },
    {
      "entropy": 5.8571044921875,
      "epoch": 4.136352276703154,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003393019994090014,
      "loss": 5.2288,
      "mean_token_accuracy": 0.21021503508090972,
      "num_tokens": 80963881.0,
      "step": 41380
    },
    {
      "entropy": 5.91642804145813,
      "epoch": 4.136852101764383,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033926779797448253,
      "loss": 5.3318,
      "mean_token_accuracy": 0.21093990355730058,
      "num_tokens": 80973757.0,
      "step": 41385
    },
    {
      "entropy": 5.853168487548828,
      "epoch": 4.137351926825611,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003392335949230025,
      "loss": 5.2731,
      "mean_token_accuracy": 0.22283947020769118,
      "num_tokens": 80983564.0,
      "step": 41390
    },
    {
      "entropy": 5.799913215637207,
      "epoch": 4.137851751886839,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033919939025542166,
      "loss": 5.233,
      "mean_token_accuracy": 0.21610826551914214,
      "num_tokens": 80993429.0,
      "step": 41395
    },
    {
      "entropy": 5.762307214736938,
      "epoch": 4.138351576948068,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003391651839726008,
      "loss": 5.1953,
      "mean_token_accuracy": 0.2225502237677574,
      "num_tokens": 81003461.0,
      "step": 41400
    },
    {
      "entropy": 5.784522438049317,
      "epoch": 4.138851402009297,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003391309760754004,
      "loss": 5.2539,
      "mean_token_accuracy": 0.2103325217962265,
      "num_tokens": 81012880.0,
      "step": 41405
    },
    {
      "entropy": 5.8338775634765625,
      "epoch": 4.1393512270705255,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003390967665646811,
      "loss": 5.3617,
      "mean_token_accuracy": 0.19768353104591369,
      "num_tokens": 81023248.0,
      "step": 41410
    },
    {
      "entropy": 5.823594093322754,
      "epoch": 4.139851052131754,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003390625554413036,
      "loss": 5.142,
      "mean_token_accuracy": 0.23131975829601287,
      "num_tokens": 81032591.0,
      "step": 41415
    },
    {
      "entropy": 5.898988437652588,
      "epoch": 4.140350877192983,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003390283427061288,
      "loss": 5.3268,
      "mean_token_accuracy": 0.20555420070886612,
      "num_tokens": 81043316.0,
      "step": 41420
    },
    {
      "entropy": 5.905065441131592,
      "epoch": 4.140850702254211,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003389941283600173,
      "loss": 5.3589,
      "mean_token_accuracy": 0.20614771842956542,
      "num_tokens": 81053853.0,
      "step": 41425
    },
    {
      "entropy": 5.782047080993652,
      "epoch": 4.141350527315439,
      "grad_norm": 1.125,
      "learning_rate": 0.00033895991240383,
      "loss": 5.1212,
      "mean_token_accuracy": 0.2170257404446602,
      "num_tokens": 81064294.0,
      "step": 41430
    },
    {
      "entropy": 5.717781972885132,
      "epoch": 4.141850352376668,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003389256948384278,
      "loss": 5.1877,
      "mean_token_accuracy": 0.22292715162038804,
      "num_tokens": 81074061.0,
      "step": 41435
    },
    {
      "entropy": 5.810916185379028,
      "epoch": 4.142350177437897,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003388914756646715,
      "loss": 5.2634,
      "mean_token_accuracy": 0.21010771095752717,
      "num_tokens": 81083558.0,
      "step": 41440
    },
    {
      "entropy": 5.838090515136718,
      "epoch": 4.1428500024991255,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003388572548834222,
      "loss": 5.2686,
      "mean_token_accuracy": 0.20755449682474136,
      "num_tokens": 81092672.0,
      "step": 41445
    },
    {
      "entropy": 5.833670282363892,
      "epoch": 4.143349827560354,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003388230324955407,
      "loss": 5.3519,
      "mean_token_accuracy": 0.20361728817224503,
      "num_tokens": 81102530.0,
      "step": 41450
    },
    {
      "entropy": 5.835967111587524,
      "epoch": 4.143849652621583,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033878880850188813,
      "loss": 5.3184,
      "mean_token_accuracy": 0.20765650719404222,
      "num_tokens": 81113625.0,
      "step": 41455
    },
    {
      "entropy": 5.900904369354248,
      "epoch": 4.144349477682811,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003387545829033257,
      "loss": 5.3265,
      "mean_token_accuracy": 0.2043689787387848,
      "num_tokens": 81124362.0,
      "step": 41460
    },
    {
      "entropy": 5.846835565567017,
      "epoch": 4.144849302744039,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033872035570071425,
      "loss": 5.2887,
      "mean_token_accuracy": 0.20570603162050247,
      "num_tokens": 81134244.0,
      "step": 41465
    },
    {
      "entropy": 5.856200933456421,
      "epoch": 4.145349127805268,
      "grad_norm": 1.125,
      "learning_rate": 0.00033868612689491503,
      "loss": 5.3309,
      "mean_token_accuracy": 0.21135925352573395,
      "num_tokens": 81143582.0,
      "step": 41470
    },
    {
      "entropy": 5.817958688735962,
      "epoch": 4.145848952866497,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033865189648678935,
      "loss": 5.2283,
      "mean_token_accuracy": 0.22107867598533631,
      "num_tokens": 81152398.0,
      "step": 41475
    },
    {
      "entropy": 5.8359489917755125,
      "epoch": 4.1463487779277255,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033861766447719824,
      "loss": 5.248,
      "mean_token_accuracy": 0.20844608545303345,
      "num_tokens": 81161515.0,
      "step": 41480
    },
    {
      "entropy": 5.943444442749024,
      "epoch": 4.146848602988954,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033858343086700304,
      "loss": 5.3755,
      "mean_token_accuracy": 0.20528870373964309,
      "num_tokens": 81171283.0,
      "step": 41485
    },
    {
      "entropy": 5.917056751251221,
      "epoch": 4.147348428050183,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033854919565706515,
      "loss": 5.3847,
      "mean_token_accuracy": 0.20739551782608032,
      "num_tokens": 81180419.0,
      "step": 41490
    },
    {
      "entropy": 5.779948568344116,
      "epoch": 4.147848253111411,
      "grad_norm": 1.125,
      "learning_rate": 0.00033851495884824586,
      "loss": 5.2802,
      "mean_token_accuracy": 0.20495158135890962,
      "num_tokens": 81190455.0,
      "step": 41495
    },
    {
      "entropy": 5.785981178283691,
      "epoch": 4.148348078172639,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003384807204414065,
      "loss": 5.2591,
      "mean_token_accuracy": 0.20823147743940354,
      "num_tokens": 81200078.0,
      "step": 41500
    },
    {
      "entropy": 5.833245420455933,
      "epoch": 4.148847903233868,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003384464804374086,
      "loss": 5.2507,
      "mean_token_accuracy": 0.21939258873462678,
      "num_tokens": 81209281.0,
      "step": 41505
    },
    {
      "entropy": 5.83327054977417,
      "epoch": 4.149347728295097,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003384122388371135,
      "loss": 5.2774,
      "mean_token_accuracy": 0.2162846714258194,
      "num_tokens": 81219299.0,
      "step": 41510
    },
    {
      "entropy": 5.778430986404419,
      "epoch": 4.1498475533563255,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003383779956413829,
      "loss": 5.2311,
      "mean_token_accuracy": 0.21283517628908158,
      "num_tokens": 81230007.0,
      "step": 41515
    },
    {
      "entropy": 5.802460479736328,
      "epoch": 4.150347378417554,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003383437508510781,
      "loss": 5.2559,
      "mean_token_accuracy": 0.21797244101762772,
      "num_tokens": 81238315.0,
      "step": 41520
    },
    {
      "entropy": 5.818806314468384,
      "epoch": 4.150847203478782,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003383095044670609,
      "loss": 5.2376,
      "mean_token_accuracy": 0.22243710607290268,
      "num_tokens": 81248337.0,
      "step": 41525
    },
    {
      "entropy": 5.807554626464844,
      "epoch": 4.151347028540011,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033827525649019293,
      "loss": 5.2564,
      "mean_token_accuracy": 0.22063053101301194,
      "num_tokens": 81258133.0,
      "step": 41530
    },
    {
      "entropy": 5.852791833877563,
      "epoch": 4.151846853601239,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033824100692133566,
      "loss": 5.2494,
      "mean_token_accuracy": 0.2144124314188957,
      "num_tokens": 81266261.0,
      "step": 41535
    },
    {
      "entropy": 5.844167947769165,
      "epoch": 4.152346678662468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000338206755761351,
      "loss": 5.3339,
      "mean_token_accuracy": 0.21523270457983018,
      "num_tokens": 81276075.0,
      "step": 41540
    },
    {
      "entropy": 5.840159559249878,
      "epoch": 4.152846503723697,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003381725030111006,
      "loss": 5.234,
      "mean_token_accuracy": 0.21910041123628615,
      "num_tokens": 81285629.0,
      "step": 41545
    },
    {
      "entropy": 5.792303895950317,
      "epoch": 4.1533463287849255,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003381382486714463,
      "loss": 5.2155,
      "mean_token_accuracy": 0.2179885432124138,
      "num_tokens": 81294117.0,
      "step": 41550
    },
    {
      "entropy": 5.713327693939209,
      "epoch": 4.153846153846154,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033810399274324994,
      "loss": 5.1747,
      "mean_token_accuracy": 0.22235259115695954,
      "num_tokens": 81303580.0,
      "step": 41555
    },
    {
      "entropy": 5.8145793914794925,
      "epoch": 4.154345978907382,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033806973522737326,
      "loss": 5.2993,
      "mean_token_accuracy": 0.21050120145082474,
      "num_tokens": 81312815.0,
      "step": 41560
    },
    {
      "entropy": 5.703919410705566,
      "epoch": 4.154845803968611,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003380354761246784,
      "loss": 5.1866,
      "mean_token_accuracy": 0.22330956310033798,
      "num_tokens": 81322792.0,
      "step": 41565
    },
    {
      "entropy": 5.830545949935913,
      "epoch": 4.155345629029839,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003380012154360271,
      "loss": 5.2545,
      "mean_token_accuracy": 0.2167881101369858,
      "num_tokens": 81331777.0,
      "step": 41570
    },
    {
      "entropy": 5.823186731338501,
      "epoch": 4.155845454091068,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003379669531622814,
      "loss": 5.2708,
      "mean_token_accuracy": 0.21486967653036118,
      "num_tokens": 81340450.0,
      "step": 41575
    },
    {
      "entropy": 5.827334928512573,
      "epoch": 4.156345279152297,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003379326893043034,
      "loss": 5.3095,
      "mean_token_accuracy": 0.20746292918920517,
      "num_tokens": 81350545.0,
      "step": 41580
    },
    {
      "entropy": 5.781236410140991,
      "epoch": 4.1568451042135255,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003378984238629552,
      "loss": 5.2069,
      "mean_token_accuracy": 0.21710366904735565,
      "num_tokens": 81359931.0,
      "step": 41585
    },
    {
      "entropy": 5.867203330993652,
      "epoch": 4.157344929274754,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003378641568390988,
      "loss": 5.2651,
      "mean_token_accuracy": 0.2127859741449356,
      "num_tokens": 81368801.0,
      "step": 41590
    },
    {
      "entropy": 5.798666524887085,
      "epoch": 4.157844754335982,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003378298882335963,
      "loss": 5.2287,
      "mean_token_accuracy": 0.21432369351387023,
      "num_tokens": 81378909.0,
      "step": 41595
    },
    {
      "entropy": 5.714252710342407,
      "epoch": 4.158344579397211,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033779561804731005,
      "loss": 5.1522,
      "mean_token_accuracy": 0.21626718044281007,
      "num_tokens": 81388386.0,
      "step": 41600
    },
    {
      "entropy": 5.727238655090332,
      "epoch": 4.158844404458439,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003377613462811022,
      "loss": 5.1235,
      "mean_token_accuracy": 0.22748682349920274,
      "num_tokens": 81397454.0,
      "step": 41605
    },
    {
      "entropy": 5.912668895721436,
      "epoch": 4.159344229519668,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003377270729358351,
      "loss": 5.3281,
      "mean_token_accuracy": 0.20687987059354782,
      "num_tokens": 81408048.0,
      "step": 41610
    },
    {
      "entropy": 5.810485076904297,
      "epoch": 4.159844054580897,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003376927980123709,
      "loss": 5.2467,
      "mean_token_accuracy": 0.21676390916109084,
      "num_tokens": 81416843.0,
      "step": 41615
    },
    {
      "entropy": 5.820421600341797,
      "epoch": 4.1603438796421255,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003376585215115721,
      "loss": 5.3397,
      "mean_token_accuracy": 0.20279562920331956,
      "num_tokens": 81426237.0,
      "step": 41620
    },
    {
      "entropy": 5.7798319339752195,
      "epoch": 4.160843704703354,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033762424343430106,
      "loss": 5.2411,
      "mean_token_accuracy": 0.21913889795541763,
      "num_tokens": 81435988.0,
      "step": 41625
    },
    {
      "entropy": 5.822606039047241,
      "epoch": 4.161343529764582,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033758996378142,
      "loss": 5.3174,
      "mean_token_accuracy": 0.20203717648983002,
      "num_tokens": 81446177.0,
      "step": 41630
    },
    {
      "entropy": 5.864863109588623,
      "epoch": 4.161843354825811,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033755568255379166,
      "loss": 5.2695,
      "mean_token_accuracy": 0.21808911561965943,
      "num_tokens": 81456306.0,
      "step": 41635
    },
    {
      "entropy": 5.877184772491455,
      "epoch": 4.162343179887039,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003375213997522784,
      "loss": 5.4122,
      "mean_token_accuracy": 0.200700344145298,
      "num_tokens": 81467032.0,
      "step": 41640
    },
    {
      "entropy": 5.859266996383667,
      "epoch": 4.162843004948268,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033748711537774287,
      "loss": 5.2146,
      "mean_token_accuracy": 0.21643707305192947,
      "num_tokens": 81475732.0,
      "step": 41645
    },
    {
      "entropy": 5.884094285964966,
      "epoch": 4.163342830009497,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003374528294310476,
      "loss": 5.3613,
      "mean_token_accuracy": 0.21236131191253663,
      "num_tokens": 81484982.0,
      "step": 41650
    },
    {
      "entropy": 5.728426170349121,
      "epoch": 4.1638426550707255,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033741854191305515,
      "loss": 5.1101,
      "mean_token_accuracy": 0.2263037756085396,
      "num_tokens": 81495144.0,
      "step": 41655
    },
    {
      "entropy": 5.84183931350708,
      "epoch": 4.164342480131954,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003373842528246283,
      "loss": 5.2607,
      "mean_token_accuracy": 0.22116134762763978,
      "num_tokens": 81504753.0,
      "step": 41660
    },
    {
      "entropy": 5.898527526855469,
      "epoch": 4.164842305193182,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003373499621666296,
      "loss": 5.3298,
      "mean_token_accuracy": 0.202654068171978,
      "num_tokens": 81514690.0,
      "step": 41665
    },
    {
      "entropy": 5.787746906280518,
      "epoch": 4.165342130254411,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033731566993992204,
      "loss": 5.2094,
      "mean_token_accuracy": 0.2115939661860466,
      "num_tokens": 81524307.0,
      "step": 41670
    },
    {
      "entropy": 5.752361679077149,
      "epoch": 4.165841955315639,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033728137614536815,
      "loss": 5.2269,
      "mean_token_accuracy": 0.21149950921535493,
      "num_tokens": 81534164.0,
      "step": 41675
    },
    {
      "entropy": 5.909932374954224,
      "epoch": 4.166341780376868,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003372470807838309,
      "loss": 5.3293,
      "mean_token_accuracy": 0.2001722812652588,
      "num_tokens": 81543098.0,
      "step": 41680
    },
    {
      "entropy": 5.898697376251221,
      "epoch": 4.166841605438097,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033721278385617315,
      "loss": 5.348,
      "mean_token_accuracy": 0.2015694111585617,
      "num_tokens": 81553684.0,
      "step": 41685
    },
    {
      "entropy": 5.802794885635376,
      "epoch": 4.1673414304993255,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003371784853632577,
      "loss": 5.2026,
      "mean_token_accuracy": 0.2168087840080261,
      "num_tokens": 81562632.0,
      "step": 41690
    },
    {
      "entropy": 5.692754888534546,
      "epoch": 4.167841255560554,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033714418530594767,
      "loss": 5.1398,
      "mean_token_accuracy": 0.22766151577234267,
      "num_tokens": 81571873.0,
      "step": 41695
    },
    {
      "entropy": 5.711313152313233,
      "epoch": 4.168341080621782,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003371098836851058,
      "loss": 5.1564,
      "mean_token_accuracy": 0.23070415407419204,
      "num_tokens": 81581324.0,
      "step": 41700
    },
    {
      "entropy": 5.752843809127808,
      "epoch": 4.168840905683011,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003370755805015953,
      "loss": 5.2671,
      "mean_token_accuracy": 0.2140081986784935,
      "num_tokens": 81591707.0,
      "step": 41705
    },
    {
      "entropy": 5.891847419738769,
      "epoch": 4.169340730744239,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003370412757562792,
      "loss": 5.3762,
      "mean_token_accuracy": 0.20649689584970474,
      "num_tokens": 81601824.0,
      "step": 41710
    },
    {
      "entropy": 5.7727728366851805,
      "epoch": 4.169840555805468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003370069694500205,
      "loss": 5.2018,
      "mean_token_accuracy": 0.21445361226797105,
      "num_tokens": 81611586.0,
      "step": 41715
    },
    {
      "entropy": 5.837597131729126,
      "epoch": 4.170340380866697,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003369726615836825,
      "loss": 5.252,
      "mean_token_accuracy": 0.21504818201065062,
      "num_tokens": 81620948.0,
      "step": 41720
    },
    {
      "entropy": 5.815897274017334,
      "epoch": 4.1708402059279255,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003369383521581282,
      "loss": 5.1963,
      "mean_token_accuracy": 0.22394500076770782,
      "num_tokens": 81630575.0,
      "step": 41725
    },
    {
      "entropy": 5.872657060623169,
      "epoch": 4.171340030989154,
      "grad_norm": 1.25,
      "learning_rate": 0.00033690404117422084,
      "loss": 5.2863,
      "mean_token_accuracy": 0.2073908269405365,
      "num_tokens": 81640359.0,
      "step": 41730
    },
    {
      "entropy": 5.901200151443481,
      "epoch": 4.171839856050382,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033686972863282384,
      "loss": 5.326,
      "mean_token_accuracy": 0.21113000810146332,
      "num_tokens": 81649989.0,
      "step": 41735
    },
    {
      "entropy": 5.859527730941773,
      "epoch": 4.172339681111611,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003368354145348003,
      "loss": 5.4315,
      "mean_token_accuracy": 0.1925997778773308,
      "num_tokens": 81660317.0,
      "step": 41740
    },
    {
      "entropy": 5.701200485229492,
      "epoch": 4.172839506172839,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033680109888101375,
      "loss": 5.0682,
      "mean_token_accuracy": 0.22695669829845427,
      "num_tokens": 81670665.0,
      "step": 41745
    },
    {
      "entropy": 5.812520503997803,
      "epoch": 4.173339331234068,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003367667816723274,
      "loss": 5.2799,
      "mean_token_accuracy": 0.21932430267333985,
      "num_tokens": 81680411.0,
      "step": 41750
    },
    {
      "entropy": 5.7984555721282955,
      "epoch": 4.173839156295297,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003367324629096047,
      "loss": 5.2252,
      "mean_token_accuracy": 0.21800071150064468,
      "num_tokens": 81690222.0,
      "step": 41755
    },
    {
      "entropy": 5.805366134643554,
      "epoch": 4.1743389813565255,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003366981425937091,
      "loss": 5.2839,
      "mean_token_accuracy": 0.21585042774677277,
      "num_tokens": 81700027.0,
      "step": 41760
    },
    {
      "entropy": 5.816783857345581,
      "epoch": 4.174838806417754,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033666382072550417,
      "loss": 5.2773,
      "mean_token_accuracy": 0.21499292254447938,
      "num_tokens": 81709505.0,
      "step": 41765
    },
    {
      "entropy": 5.903678512573242,
      "epoch": 4.175338631478982,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033662949730585336,
      "loss": 5.3362,
      "mean_token_accuracy": 0.2060026064515114,
      "num_tokens": 81721133.0,
      "step": 41770
    },
    {
      "entropy": 5.862806510925293,
      "epoch": 4.175838456540211,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003365951723356202,
      "loss": 5.3741,
      "mean_token_accuracy": 0.20090155750513078,
      "num_tokens": 81730647.0,
      "step": 41775
    },
    {
      "entropy": 5.845580005645752,
      "epoch": 4.176338281601439,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033656084581566844,
      "loss": 5.2662,
      "mean_token_accuracy": 0.2174952283501625,
      "num_tokens": 81740972.0,
      "step": 41780
    },
    {
      "entropy": 5.805156469345093,
      "epoch": 4.176838106662668,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033652651774686163,
      "loss": 5.2632,
      "mean_token_accuracy": 0.2155000939965248,
      "num_tokens": 81750469.0,
      "step": 41785
    },
    {
      "entropy": 5.890890741348267,
      "epoch": 4.177337931723897,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033649218813006337,
      "loss": 5.3445,
      "mean_token_accuracy": 0.21241808980703353,
      "num_tokens": 81759519.0,
      "step": 41790
    },
    {
      "entropy": 5.813356733322143,
      "epoch": 4.1778377567851255,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003364578569661376,
      "loss": 5.2589,
      "mean_token_accuracy": 0.2077874481678009,
      "num_tokens": 81769304.0,
      "step": 41795
    },
    {
      "entropy": 5.753099822998047,
      "epoch": 4.178337581846354,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033642352425594787,
      "loss": 5.172,
      "mean_token_accuracy": 0.21957820206880568,
      "num_tokens": 81779633.0,
      "step": 41800
    },
    {
      "entropy": 5.742881393432617,
      "epoch": 4.178837406907582,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033638919000035815,
      "loss": 5.1908,
      "mean_token_accuracy": 0.22613967657089235,
      "num_tokens": 81787953.0,
      "step": 41805
    },
    {
      "entropy": 5.887852144241333,
      "epoch": 4.179337231968811,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033635485420023217,
      "loss": 5.347,
      "mean_token_accuracy": 0.20758453607559205,
      "num_tokens": 81796764.0,
      "step": 41810
    },
    {
      "entropy": 5.848724460601806,
      "epoch": 4.179837057030039,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003363205168564338,
      "loss": 5.2003,
      "mean_token_accuracy": 0.22283874303102494,
      "num_tokens": 81805831.0,
      "step": 41815
    },
    {
      "entropy": 5.796861171722412,
      "epoch": 4.180336882091268,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033628617796982707,
      "loss": 5.2517,
      "mean_token_accuracy": 0.21247156113386154,
      "num_tokens": 81816213.0,
      "step": 41820
    },
    {
      "entropy": 5.813555192947388,
      "epoch": 4.180836707152497,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033625183754127584,
      "loss": 5.2883,
      "mean_token_accuracy": 0.21370406448841095,
      "num_tokens": 81826532.0,
      "step": 41825
    },
    {
      "entropy": 5.849963617324829,
      "epoch": 4.1813365322137255,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003362174955716442,
      "loss": 5.2797,
      "mean_token_accuracy": 0.22072612047195433,
      "num_tokens": 81836043.0,
      "step": 41830
    },
    {
      "entropy": 5.807043218612671,
      "epoch": 4.181836357274954,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033618315206179614,
      "loss": 5.2231,
      "mean_token_accuracy": 0.21508073657751084,
      "num_tokens": 81845644.0,
      "step": 41835
    },
    {
      "entropy": 5.770887994766236,
      "epoch": 4.182336182336182,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003361488070125956,
      "loss": 5.2333,
      "mean_token_accuracy": 0.21608883291482925,
      "num_tokens": 81854878.0,
      "step": 41840
    },
    {
      "entropy": 5.820461988449097,
      "epoch": 4.182836007397411,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003361144604249069,
      "loss": 5.2551,
      "mean_token_accuracy": 0.2136925905942917,
      "num_tokens": 81864369.0,
      "step": 41845
    },
    {
      "entropy": 5.860913228988648,
      "epoch": 4.183335832458639,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033608011229959404,
      "loss": 5.2556,
      "mean_token_accuracy": 0.20983438789844513,
      "num_tokens": 81873588.0,
      "step": 41850
    },
    {
      "entropy": 5.801246118545532,
      "epoch": 4.183835657519868,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003360457626375213,
      "loss": 5.2078,
      "mean_token_accuracy": 0.2186127319931984,
      "num_tokens": 81883264.0,
      "step": 41855
    },
    {
      "entropy": 5.832856464385986,
      "epoch": 4.184335482581097,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003360114114395529,
      "loss": 5.3667,
      "mean_token_accuracy": 0.20504181534051896,
      "num_tokens": 81892975.0,
      "step": 41860
    },
    {
      "entropy": 5.877597045898438,
      "epoch": 4.1848353076423255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000335977058706553,
      "loss": 5.402,
      "mean_token_accuracy": 0.20393298119306563,
      "num_tokens": 81903016.0,
      "step": 41865
    },
    {
      "entropy": 5.730856227874756,
      "epoch": 4.185335132703554,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033594270443938604,
      "loss": 5.1437,
      "mean_token_accuracy": 0.226765014231205,
      "num_tokens": 81913573.0,
      "step": 41870
    },
    {
      "entropy": 5.685239648818969,
      "epoch": 4.185834957764782,
      "grad_norm": 1.125,
      "learning_rate": 0.00033590834863891645,
      "loss": 5.0971,
      "mean_token_accuracy": 0.23602095991373062,
      "num_tokens": 81923655.0,
      "step": 41875
    },
    {
      "entropy": 5.888417387008667,
      "epoch": 4.186334782826011,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003358739913060083,
      "loss": 5.2454,
      "mean_token_accuracy": 0.21444191336631774,
      "num_tokens": 81933531.0,
      "step": 41880
    },
    {
      "entropy": 5.787117862701416,
      "epoch": 4.186834607887239,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003358396324415263,
      "loss": 5.1864,
      "mean_token_accuracy": 0.22228932678699492,
      "num_tokens": 81943880.0,
      "step": 41885
    },
    {
      "entropy": 5.844830131530761,
      "epoch": 4.187334432948468,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033580527204633473,
      "loss": 5.3873,
      "mean_token_accuracy": 0.1980193644762039,
      "num_tokens": 81953909.0,
      "step": 41890
    },
    {
      "entropy": 5.798803806304932,
      "epoch": 4.187834258009697,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003357709101212983,
      "loss": 5.1908,
      "mean_token_accuracy": 0.2141002297401428,
      "num_tokens": 81964183.0,
      "step": 41895
    },
    {
      "entropy": 5.813145637512207,
      "epoch": 4.1883340830709255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033573654666728127,
      "loss": 5.2132,
      "mean_token_accuracy": 0.22850026488304137,
      "num_tokens": 81974713.0,
      "step": 41900
    },
    {
      "entropy": 5.806102991104126,
      "epoch": 4.188833908132154,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003357021816851484,
      "loss": 5.2748,
      "mean_token_accuracy": 0.2190463811159134,
      "num_tokens": 81985100.0,
      "step": 41905
    },
    {
      "entropy": 5.779499912261963,
      "epoch": 4.189333733193382,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003356678151757643,
      "loss": 5.1967,
      "mean_token_accuracy": 0.21435756236314774,
      "num_tokens": 81994883.0,
      "step": 41910
    },
    {
      "entropy": 5.875057172775269,
      "epoch": 4.189833558254611,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033563344713999344,
      "loss": 5.2683,
      "mean_token_accuracy": 0.20946795046329497,
      "num_tokens": 82003755.0,
      "step": 41915
    },
    {
      "entropy": 5.882853269577026,
      "epoch": 4.190333383315839,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033559907757870074,
      "loss": 5.3581,
      "mean_token_accuracy": 0.2105857938528061,
      "num_tokens": 82013767.0,
      "step": 41920
    },
    {
      "entropy": 5.774264621734619,
      "epoch": 4.190833208377068,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033556470649275086,
      "loss": 5.2926,
      "mean_token_accuracy": 0.21793243438005447,
      "num_tokens": 82024480.0,
      "step": 41925
    },
    {
      "entropy": 5.887807083129883,
      "epoch": 4.191333033438297,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003355303338830086,
      "loss": 5.3182,
      "mean_token_accuracy": 0.21412883847951888,
      "num_tokens": 82034071.0,
      "step": 41930
    },
    {
      "entropy": 5.755815172195435,
      "epoch": 4.1918328584995255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003354959597503386,
      "loss": 5.1693,
      "mean_token_accuracy": 0.22355127781629563,
      "num_tokens": 82044406.0,
      "step": 41935
    },
    {
      "entropy": 5.839972877502442,
      "epoch": 4.192332683560753,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033546158409560587,
      "loss": 5.3442,
      "mean_token_accuracy": 0.21295441538095475,
      "num_tokens": 82054810.0,
      "step": 41940
    },
    {
      "entropy": 5.847497510910034,
      "epoch": 4.192832508621982,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003354272069196752,
      "loss": 5.2905,
      "mean_token_accuracy": 0.2073323681950569,
      "num_tokens": 82064135.0,
      "step": 41945
    },
    {
      "entropy": 5.83378963470459,
      "epoch": 4.193332333683211,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003353928282234116,
      "loss": 5.2039,
      "mean_token_accuracy": 0.22215643376111985,
      "num_tokens": 82074050.0,
      "step": 41950
    },
    {
      "entropy": 5.729189109802246,
      "epoch": 4.193832158744439,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033535844800768,
      "loss": 5.1257,
      "mean_token_accuracy": 0.23354943543672563,
      "num_tokens": 82083027.0,
      "step": 41955
    },
    {
      "entropy": 5.725520277023316,
      "epoch": 4.194331983805668,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033532406627334534,
      "loss": 5.1572,
      "mean_token_accuracy": 0.2273528277873993,
      "num_tokens": 82092527.0,
      "step": 41960
    },
    {
      "entropy": 5.7878838062286375,
      "epoch": 4.194831808866897,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003352896830212728,
      "loss": 5.1908,
      "mean_token_accuracy": 0.21681186109781264,
      "num_tokens": 82102378.0,
      "step": 41965
    },
    {
      "entropy": 5.773173856735229,
      "epoch": 4.1953316339281255,
      "grad_norm": 1.125,
      "learning_rate": 0.0003352552982523272,
      "loss": 5.2342,
      "mean_token_accuracy": 0.2130380764603615,
      "num_tokens": 82112225.0,
      "step": 41970
    },
    {
      "entropy": 5.761728239059448,
      "epoch": 4.195831458989353,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003352209119673739,
      "loss": 5.2516,
      "mean_token_accuracy": 0.2142837092280388,
      "num_tokens": 82123267.0,
      "step": 41975
    },
    {
      "entropy": 5.865330743789673,
      "epoch": 4.196331284050582,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033518652416727785,
      "loss": 5.3065,
      "mean_token_accuracy": 0.2070389986038208,
      "num_tokens": 82132711.0,
      "step": 41980
    },
    {
      "entropy": 5.82765417098999,
      "epoch": 4.196831109111811,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003351521348529044,
      "loss": 5.2456,
      "mean_token_accuracy": 0.20940017998218535,
      "num_tokens": 82142445.0,
      "step": 41985
    },
    {
      "entropy": 5.876348257064819,
      "epoch": 4.197330934173039,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033511774402511876,
      "loss": 5.3593,
      "mean_token_accuracy": 0.2082934185862541,
      "num_tokens": 82151540.0,
      "step": 41990
    },
    {
      "entropy": 5.731285333633423,
      "epoch": 4.197830759234268,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003350833516847861,
      "loss": 5.1857,
      "mean_token_accuracy": 0.2223179057240486,
      "num_tokens": 82161751.0,
      "step": 41995
    },
    {
      "entropy": 5.840321969985962,
      "epoch": 4.198330584295497,
      "grad_norm": 1.25,
      "learning_rate": 0.00033504895783277176,
      "loss": 5.2966,
      "mean_token_accuracy": 0.22019028663635254,
      "num_tokens": 82170492.0,
      "step": 42000
    },
    {
      "epoch": 4.198330584295497,
      "eval_entropy": 5.593021541488321,
      "eval_loss": 5.4478960037231445,
      "eval_mean_token_accuracy": 0.21232140653605192,
      "eval_num_tokens": 82170492.0,
      "eval_runtime": 34.5411,
      "eval_samples_per_second": 898.785,
      "eval_steps_per_second": 112.359,
      "step": 42000
    },
    {
      "entropy": 5.756327438354492,
      "epoch": 4.1988304093567255,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003350145624699412,
      "loss": 5.1829,
      "mean_token_accuracy": 0.22312007546424867,
      "num_tokens": 82179739.0,
      "step": 42005
    },
    {
      "entropy": 5.81242184638977,
      "epoch": 4.199330234417953,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033498016559715954,
      "loss": 5.3048,
      "mean_token_accuracy": 0.2061346009373665,
      "num_tokens": 82188701.0,
      "step": 42010
    },
    {
      "entropy": 5.861502408981323,
      "epoch": 4.199830059479182,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033494576721529234,
      "loss": 5.3359,
      "mean_token_accuracy": 0.20980557054281235,
      "num_tokens": 82199431.0,
      "step": 42015
    },
    {
      "entropy": 5.81865963935852,
      "epoch": 4.200329884540411,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033491136732520516,
      "loss": 5.1744,
      "mean_token_accuracy": 0.21885020434856414,
      "num_tokens": 82208860.0,
      "step": 42020
    },
    {
      "entropy": 5.871848630905151,
      "epoch": 4.200829709601639,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003348769659277634,
      "loss": 5.2817,
      "mean_token_accuracy": 0.20901186019182205,
      "num_tokens": 82218488.0,
      "step": 42025
    },
    {
      "entropy": 5.79004487991333,
      "epoch": 4.201329534662868,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033484256302383254,
      "loss": 5.2806,
      "mean_token_accuracy": 0.2150524526834488,
      "num_tokens": 82228306.0,
      "step": 42030
    },
    {
      "entropy": 5.851825809478759,
      "epoch": 4.201829359724097,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033480815861427826,
      "loss": 5.317,
      "mean_token_accuracy": 0.21231145709753035,
      "num_tokens": 82238187.0,
      "step": 42035
    },
    {
      "entropy": 5.787282943725586,
      "epoch": 4.2023291847853255,
      "grad_norm": 1.203125,
      "learning_rate": 0.000334773752699966,
      "loss": 5.1985,
      "mean_token_accuracy": 0.21712819784879683,
      "num_tokens": 82246514.0,
      "step": 42040
    },
    {
      "entropy": 5.8665283679962155,
      "epoch": 4.202829009846553,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003347393452817616,
      "loss": 5.3685,
      "mean_token_accuracy": 0.20082758218050004,
      "num_tokens": 82255988.0,
      "step": 42045
    },
    {
      "entropy": 5.825429201126099,
      "epoch": 4.203328834907782,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003347049363605306,
      "loss": 5.3252,
      "mean_token_accuracy": 0.20860601961612701,
      "num_tokens": 82265781.0,
      "step": 42050
    },
    {
      "entropy": 5.819159698486328,
      "epoch": 4.203828659969011,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003346705259371387,
      "loss": 5.2807,
      "mean_token_accuracy": 0.21772874444723128,
      "num_tokens": 82276250.0,
      "step": 42055
    },
    {
      "entropy": 5.870838642120361,
      "epoch": 4.204328485030239,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003346361140124518,
      "loss": 5.3121,
      "mean_token_accuracy": 0.21193339675664902,
      "num_tokens": 82286765.0,
      "step": 42060
    },
    {
      "entropy": 5.758634090423584,
      "epoch": 4.204828310091468,
      "grad_norm": 1.125,
      "learning_rate": 0.0003346017005873356,
      "loss": 5.213,
      "mean_token_accuracy": 0.22302042096853256,
      "num_tokens": 82295953.0,
      "step": 42065
    },
    {
      "entropy": 5.7714478969573975,
      "epoch": 4.205328135152697,
      "grad_norm": 1.25,
      "learning_rate": 0.000334567285662656,
      "loss": 5.1983,
      "mean_token_accuracy": 0.21560611128807067,
      "num_tokens": 82304673.0,
      "step": 42070
    },
    {
      "entropy": 5.729595565795899,
      "epoch": 4.2058279602139255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003345328692392787,
      "loss": 5.1819,
      "mean_token_accuracy": 0.2202826991677284,
      "num_tokens": 82313827.0,
      "step": 42075
    },
    {
      "entropy": 5.736682271957397,
      "epoch": 4.206327785275153,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033449845131806985,
      "loss": 5.2034,
      "mean_token_accuracy": 0.21817476600408553,
      "num_tokens": 82322650.0,
      "step": 42080
    },
    {
      "entropy": 5.837730884552002,
      "epoch": 4.206827610336382,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003344640318998952,
      "loss": 5.2757,
      "mean_token_accuracy": 0.21482681035995482,
      "num_tokens": 82332315.0,
      "step": 42085
    },
    {
      "entropy": 5.801802301406861,
      "epoch": 4.207327435397611,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033442961098562084,
      "loss": 5.2633,
      "mean_token_accuracy": 0.21232567578554154,
      "num_tokens": 82342146.0,
      "step": 42090
    },
    {
      "entropy": 5.84651746749878,
      "epoch": 4.207827260458839,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003343951885761127,
      "loss": 5.271,
      "mean_token_accuracy": 0.21708561182022096,
      "num_tokens": 82351638.0,
      "step": 42095
    },
    {
      "entropy": 5.715138387680054,
      "epoch": 4.208327085520068,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033436076467223695,
      "loss": 5.1413,
      "mean_token_accuracy": 0.22455849051475524,
      "num_tokens": 82361427.0,
      "step": 42100
    },
    {
      "entropy": 5.797017812728882,
      "epoch": 4.208826910581297,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003343263392748597,
      "loss": 5.2668,
      "mean_token_accuracy": 0.21988863497972488,
      "num_tokens": 82371019.0,
      "step": 42105
    },
    {
      "entropy": 5.788729858398438,
      "epoch": 4.2093267356425255,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003342919123848469,
      "loss": 5.2687,
      "mean_token_accuracy": 0.21067555993795395,
      "num_tokens": 82380539.0,
      "step": 42110
    },
    {
      "entropy": 5.752717924118042,
      "epoch": 4.209826560703753,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003342574840030649,
      "loss": 5.1854,
      "mean_token_accuracy": 0.2274412289261818,
      "num_tokens": 82390224.0,
      "step": 42115
    },
    {
      "entropy": 5.862512922286987,
      "epoch": 4.210326385764982,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003342230541303799,
      "loss": 5.2785,
      "mean_token_accuracy": 0.21699111610651017,
      "num_tokens": 82400108.0,
      "step": 42120
    },
    {
      "entropy": 5.783490371704102,
      "epoch": 4.210826210826211,
      "grad_norm": 1.171875,
      "learning_rate": 0.000334188622767658,
      "loss": 5.1872,
      "mean_token_accuracy": 0.21837456226348878,
      "num_tokens": 82410225.0,
      "step": 42125
    },
    {
      "entropy": 5.844472408294678,
      "epoch": 4.211326035887439,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003341541899157657,
      "loss": 5.2375,
      "mean_token_accuracy": 0.2160069078207016,
      "num_tokens": 82420632.0,
      "step": 42130
    },
    {
      "entropy": 5.858961486816407,
      "epoch": 4.211825860948668,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003341197555755691,
      "loss": 5.3546,
      "mean_token_accuracy": 0.2022664189338684,
      "num_tokens": 82429747.0,
      "step": 42135
    },
    {
      "entropy": 5.789462757110596,
      "epoch": 4.212325686009897,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033408531974793464,
      "loss": 5.2295,
      "mean_token_accuracy": 0.2175314173102379,
      "num_tokens": 82438833.0,
      "step": 42140
    },
    {
      "entropy": 5.81304178237915,
      "epoch": 4.2128255110711255,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003340508824337288,
      "loss": 5.2371,
      "mean_token_accuracy": 0.21046454310417176,
      "num_tokens": 82448172.0,
      "step": 42145
    },
    {
      "entropy": 5.720870542526245,
      "epoch": 4.213325336132353,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033401644363381796,
      "loss": 5.2904,
      "mean_token_accuracy": 0.21480308771133422,
      "num_tokens": 82457908.0,
      "step": 42150
    },
    {
      "entropy": 5.83367805480957,
      "epoch": 4.213825161193582,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003339820033490686,
      "loss": 5.2474,
      "mean_token_accuracy": 0.21319350749254226,
      "num_tokens": 82467889.0,
      "step": 42155
    },
    {
      "entropy": 5.850777339935303,
      "epoch": 4.214324986254811,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033394756158034724,
      "loss": 5.2849,
      "mean_token_accuracy": 0.21278860419988632,
      "num_tokens": 82478985.0,
      "step": 42160
    },
    {
      "entropy": 5.865759468078613,
      "epoch": 4.214824811316039,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003339131183285203,
      "loss": 5.3241,
      "mean_token_accuracy": 0.20090284049510956,
      "num_tokens": 82489396.0,
      "step": 42165
    },
    {
      "entropy": 5.829535484313965,
      "epoch": 4.215324636377268,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033387867359445443,
      "loss": 5.2878,
      "mean_token_accuracy": 0.21380998492240905,
      "num_tokens": 82499200.0,
      "step": 42170
    },
    {
      "entropy": 5.830836582183838,
      "epoch": 4.215824461438497,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003338442273790164,
      "loss": 5.3003,
      "mean_token_accuracy": 0.20802479088306428,
      "num_tokens": 82508843.0,
      "step": 42175
    },
    {
      "entropy": 5.766765546798706,
      "epoch": 4.2163242864997255,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033380977968307273,
      "loss": 5.2168,
      "mean_token_accuracy": 0.21532393842935563,
      "num_tokens": 82518917.0,
      "step": 42180
    },
    {
      "entropy": 5.8429296016693115,
      "epoch": 4.216824111560953,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033377533050749016,
      "loss": 5.2418,
      "mean_token_accuracy": 0.21654415279626846,
      "num_tokens": 82527788.0,
      "step": 42185
    },
    {
      "entropy": 5.793297910690308,
      "epoch": 4.217323936622182,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003337408798531353,
      "loss": 5.2204,
      "mean_token_accuracy": 0.21452336609363556,
      "num_tokens": 82537810.0,
      "step": 42190
    },
    {
      "entropy": 5.814287519454956,
      "epoch": 4.217823761683411,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033370642772087507,
      "loss": 5.3124,
      "mean_token_accuracy": 0.2182127892971039,
      "num_tokens": 82547538.0,
      "step": 42195
    },
    {
      "entropy": 5.756077909469605,
      "epoch": 4.218323586744639,
      "grad_norm": 1.125,
      "learning_rate": 0.0003336719741115762,
      "loss": 5.2245,
      "mean_token_accuracy": 0.21877883225679398,
      "num_tokens": 82557757.0,
      "step": 42200
    },
    {
      "entropy": 5.841969585418701,
      "epoch": 4.218823411805868,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003336375190261056,
      "loss": 5.2756,
      "mean_token_accuracy": 0.21698006242513657,
      "num_tokens": 82567467.0,
      "step": 42205
    },
    {
      "entropy": 5.865647888183593,
      "epoch": 4.219323236867097,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033360306246533013,
      "loss": 5.2702,
      "mean_token_accuracy": 0.21247968822717667,
      "num_tokens": 82576691.0,
      "step": 42210
    },
    {
      "entropy": 5.855646371841431,
      "epoch": 4.2198230619283255,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033356860443011665,
      "loss": 5.3241,
      "mean_token_accuracy": 0.21132765859365463,
      "num_tokens": 82587426.0,
      "step": 42215
    },
    {
      "entropy": 5.748090124130249,
      "epoch": 4.220322886989553,
      "grad_norm": 1.125,
      "learning_rate": 0.00033353414492133217,
      "loss": 5.1759,
      "mean_token_accuracy": 0.218553064763546,
      "num_tokens": 82597105.0,
      "step": 42220
    },
    {
      "entropy": 5.858190631866455,
      "epoch": 4.220822712050782,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003334996839398436,
      "loss": 5.296,
      "mean_token_accuracy": 0.21463690102100372,
      "num_tokens": 82606949.0,
      "step": 42225
    },
    {
      "entropy": 5.8156500339508055,
      "epoch": 4.221322537112011,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033346522148651817,
      "loss": 5.2047,
      "mean_token_accuracy": 0.22175967395305635,
      "num_tokens": 82616980.0,
      "step": 42230
    },
    {
      "entropy": 5.78882155418396,
      "epoch": 4.221822362173239,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033343075756222265,
      "loss": 5.2209,
      "mean_token_accuracy": 0.21547584980726242,
      "num_tokens": 82625592.0,
      "step": 42235
    },
    {
      "entropy": 5.871122932434082,
      "epoch": 4.222322187234468,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033339629216782435,
      "loss": 5.3462,
      "mean_token_accuracy": 0.20619384944438934,
      "num_tokens": 82635442.0,
      "step": 42240
    },
    {
      "entropy": 5.75014009475708,
      "epoch": 4.222822012295697,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033336182530419045,
      "loss": 5.229,
      "mean_token_accuracy": 0.21685528755187988,
      "num_tokens": 82645674.0,
      "step": 42245
    },
    {
      "entropy": 5.893288803100586,
      "epoch": 4.2233218373569255,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033332735697218804,
      "loss": 5.2227,
      "mean_token_accuracy": 0.21746248453855516,
      "num_tokens": 82655303.0,
      "step": 42250
    },
    {
      "entropy": 5.839522218704223,
      "epoch": 4.223821662418153,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003332928871726842,
      "loss": 5.2225,
      "mean_token_accuracy": 0.2149944022297859,
      "num_tokens": 82664803.0,
      "step": 42255
    },
    {
      "entropy": 5.652643251419067,
      "epoch": 4.224321487479382,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033325841590654647,
      "loss": 5.0937,
      "mean_token_accuracy": 0.22923301309347152,
      "num_tokens": 82673443.0,
      "step": 42260
    },
    {
      "entropy": 5.705167484283447,
      "epoch": 4.224821312540611,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003332239431746419,
      "loss": 5.2046,
      "mean_token_accuracy": 0.2155020132660866,
      "num_tokens": 82682102.0,
      "step": 42265
    },
    {
      "entropy": 5.8556476593017575,
      "epoch": 4.225321137601839,
      "grad_norm": 1.125,
      "learning_rate": 0.0003331894689778379,
      "loss": 5.2804,
      "mean_token_accuracy": 0.2134312629699707,
      "num_tokens": 82691993.0,
      "step": 42270
    },
    {
      "entropy": 5.730830478668213,
      "epoch": 4.225820962663068,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003331549933170019,
      "loss": 5.1356,
      "mean_token_accuracy": 0.22246070504188536,
      "num_tokens": 82701812.0,
      "step": 42275
    },
    {
      "entropy": 5.761870050430298,
      "epoch": 4.226320787724297,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003331205161930012,
      "loss": 5.231,
      "mean_token_accuracy": 0.2208924636244774,
      "num_tokens": 82712119.0,
      "step": 42280
    },
    {
      "entropy": 5.782437944412232,
      "epoch": 4.2268206127855255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003330860376067033,
      "loss": 5.1961,
      "mean_token_accuracy": 0.21786714643239974,
      "num_tokens": 82722079.0,
      "step": 42285
    },
    {
      "entropy": 5.761451578140258,
      "epoch": 4.227320437846753,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003330515575589756,
      "loss": 5.2085,
      "mean_token_accuracy": 0.21892992556095123,
      "num_tokens": 82731722.0,
      "step": 42290
    },
    {
      "entropy": 5.829269361495972,
      "epoch": 4.227820262907982,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003330170760506857,
      "loss": 5.2374,
      "mean_token_accuracy": 0.21798598617315293,
      "num_tokens": 82742787.0,
      "step": 42295
    },
    {
      "entropy": 5.87014594078064,
      "epoch": 4.228320087969211,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033298259308270096,
      "loss": 5.3451,
      "mean_token_accuracy": 0.20548882186412812,
      "num_tokens": 82751859.0,
      "step": 42300
    },
    {
      "entropy": 5.743809366226197,
      "epoch": 4.228819913030439,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033294810865588915,
      "loss": 5.2197,
      "mean_token_accuracy": 0.2235247403383255,
      "num_tokens": 82760994.0,
      "step": 42305
    },
    {
      "entropy": 5.8558642864227295,
      "epoch": 4.229319738091668,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033291362277111793,
      "loss": 5.3197,
      "mean_token_accuracy": 0.21032150834798813,
      "num_tokens": 82769964.0,
      "step": 42310
    },
    {
      "entropy": 5.775658035278321,
      "epoch": 4.229819563152897,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033287913542925484,
      "loss": 5.1754,
      "mean_token_accuracy": 0.220386278629303,
      "num_tokens": 82779136.0,
      "step": 42315
    },
    {
      "entropy": 5.748241567611695,
      "epoch": 4.2303193882141255,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003328446466311676,
      "loss": 5.1797,
      "mean_token_accuracy": 0.22946061939001083,
      "num_tokens": 82790932.0,
      "step": 42320
    },
    {
      "entropy": 5.9299787998199465,
      "epoch": 4.230819213275353,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003328101563777238,
      "loss": 5.3824,
      "mean_token_accuracy": 0.206638003885746,
      "num_tokens": 82801861.0,
      "step": 42325
    },
    {
      "entropy": 5.82640495300293,
      "epoch": 4.231319038336582,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003327756646697915,
      "loss": 5.2629,
      "mean_token_accuracy": 0.20814430713653564,
      "num_tokens": 82812145.0,
      "step": 42330
    },
    {
      "entropy": 5.686396741867066,
      "epoch": 4.231818863397811,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003327411715082382,
      "loss": 5.0974,
      "mean_token_accuracy": 0.2363678216934204,
      "num_tokens": 82822309.0,
      "step": 42335
    },
    {
      "entropy": 5.801536893844604,
      "epoch": 4.232318688459039,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000332706676893932,
      "loss": 5.3068,
      "mean_token_accuracy": 0.2149661138653755,
      "num_tokens": 82831538.0,
      "step": 42340
    },
    {
      "entropy": 5.748089933395386,
      "epoch": 4.232818513520268,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033267218082774064,
      "loss": 5.2411,
      "mean_token_accuracy": 0.21664379090070723,
      "num_tokens": 82842045.0,
      "step": 42345
    },
    {
      "entropy": 5.797688293457031,
      "epoch": 4.233318338581497,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003326376833105321,
      "loss": 5.1696,
      "mean_token_accuracy": 0.21802637726068497,
      "num_tokens": 82851725.0,
      "step": 42350
    },
    {
      "entropy": 5.8326703071594235,
      "epoch": 4.2338181636427255,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033260318434317414,
      "loss": 5.2701,
      "mean_token_accuracy": 0.21482300013303757,
      "num_tokens": 82860483.0,
      "step": 42355
    },
    {
      "entropy": 5.777642059326172,
      "epoch": 4.234317988703953,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033256868392653497,
      "loss": 5.2022,
      "mean_token_accuracy": 0.21939203888177872,
      "num_tokens": 82870580.0,
      "step": 42360
    },
    {
      "entropy": 5.799357891082764,
      "epoch": 4.234817813765182,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003325341820614825,
      "loss": 5.2519,
      "mean_token_accuracy": 0.2135571137070656,
      "num_tokens": 82880018.0,
      "step": 42365
    },
    {
      "entropy": 5.817500495910645,
      "epoch": 4.235317638826411,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003324996787488849,
      "loss": 5.2511,
      "mean_token_accuracy": 0.2046464517712593,
      "num_tokens": 82890239.0,
      "step": 42370
    },
    {
      "entropy": 5.75661997795105,
      "epoch": 4.235817463887639,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003324651739896101,
      "loss": 5.1565,
      "mean_token_accuracy": 0.22515660375356675,
      "num_tokens": 82899060.0,
      "step": 42375
    },
    {
      "entropy": 5.795968961715698,
      "epoch": 4.236317288948868,
      "grad_norm": 1.125,
      "learning_rate": 0.0003324306677845263,
      "loss": 5.2745,
      "mean_token_accuracy": 0.21353380829095842,
      "num_tokens": 82908516.0,
      "step": 42380
    },
    {
      "entropy": 5.811409854888916,
      "epoch": 4.236817114010097,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003323961601345017,
      "loss": 5.2725,
      "mean_token_accuracy": 0.21036669462919236,
      "num_tokens": 82917645.0,
      "step": 42385
    },
    {
      "entropy": 5.789304494857788,
      "epoch": 4.2373169390713255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003323616510404045,
      "loss": 5.2832,
      "mean_token_accuracy": 0.21386541426181793,
      "num_tokens": 82927946.0,
      "step": 42390
    },
    {
      "entropy": 5.92428412437439,
      "epoch": 4.237816764132553,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033232714050310285,
      "loss": 5.393,
      "mean_token_accuracy": 0.20492470413446426,
      "num_tokens": 82938828.0,
      "step": 42395
    },
    {
      "entropy": 5.866423320770264,
      "epoch": 4.238316589193782,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033229262852346514,
      "loss": 5.2786,
      "mean_token_accuracy": 0.2089170604944229,
      "num_tokens": 82947742.0,
      "step": 42400
    },
    {
      "entropy": 5.715810108184814,
      "epoch": 4.238816414255011,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033225811510235965,
      "loss": 5.2555,
      "mean_token_accuracy": 0.21650291979312897,
      "num_tokens": 82957575.0,
      "step": 42405
    },
    {
      "entropy": 5.762961387634277,
      "epoch": 4.239316239316239,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033222360024065474,
      "loss": 5.2756,
      "mean_token_accuracy": 0.21425876766443253,
      "num_tokens": 82967489.0,
      "step": 42410
    },
    {
      "entropy": 5.7405284404754635,
      "epoch": 4.239816064377468,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033218908393921874,
      "loss": 5.1625,
      "mean_token_accuracy": 0.22448007464408876,
      "num_tokens": 82977775.0,
      "step": 42415
    },
    {
      "entropy": 5.906977748870849,
      "epoch": 4.240315889438697,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003321545661989201,
      "loss": 5.3913,
      "mean_token_accuracy": 0.2043131858110428,
      "num_tokens": 82988287.0,
      "step": 42420
    },
    {
      "entropy": 5.799645566940308,
      "epoch": 4.2408157144999254,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033212004702062734,
      "loss": 5.2051,
      "mean_token_accuracy": 0.22353854775428772,
      "num_tokens": 82997631.0,
      "step": 42425
    },
    {
      "entropy": 5.802473783493042,
      "epoch": 4.241315539561153,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003320855264052088,
      "loss": 5.2463,
      "mean_token_accuracy": 0.21723154783248902,
      "num_tokens": 83007941.0,
      "step": 42430
    },
    {
      "entropy": 5.82298092842102,
      "epoch": 4.241815364622382,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033205100435353314,
      "loss": 5.316,
      "mean_token_accuracy": 0.21121736764907836,
      "num_tokens": 83017910.0,
      "step": 42435
    },
    {
      "entropy": 5.835601758956909,
      "epoch": 4.242315189683611,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003320164808664689,
      "loss": 5.3109,
      "mean_token_accuracy": 0.21436549723148346,
      "num_tokens": 83027589.0,
      "step": 42440
    },
    {
      "entropy": 5.888935279846192,
      "epoch": 4.242815014744839,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003319819559448847,
      "loss": 5.2906,
      "mean_token_accuracy": 0.2134807288646698,
      "num_tokens": 83037738.0,
      "step": 42445
    },
    {
      "entropy": 5.880699825286865,
      "epoch": 4.243314839806068,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033194742958964914,
      "loss": 5.3393,
      "mean_token_accuracy": 0.20831147134304046,
      "num_tokens": 83047355.0,
      "step": 42450
    },
    {
      "entropy": 5.7982580184936525,
      "epoch": 4.243814664867297,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033191290180163084,
      "loss": 5.2434,
      "mean_token_accuracy": 0.21526841074228287,
      "num_tokens": 83056376.0,
      "step": 42455
    },
    {
      "entropy": 5.8262639999389645,
      "epoch": 4.2443144899285254,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033187837258169864,
      "loss": 5.2832,
      "mean_token_accuracy": 0.21321196407079696,
      "num_tokens": 83065825.0,
      "step": 42460
    },
    {
      "entropy": 5.787389659881592,
      "epoch": 4.244814314989753,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003318438419307212,
      "loss": 5.1956,
      "mean_token_accuracy": 0.2165360450744629,
      "num_tokens": 83075578.0,
      "step": 42465
    },
    {
      "entropy": 5.76098222732544,
      "epoch": 4.245314140050982,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003318093098495673,
      "loss": 5.1613,
      "mean_token_accuracy": 0.22167055457830429,
      "num_tokens": 83086192.0,
      "step": 42470
    },
    {
      "entropy": 5.76558632850647,
      "epoch": 4.245813965112211,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033177477633910586,
      "loss": 5.2647,
      "mean_token_accuracy": 0.21729649752378463,
      "num_tokens": 83096502.0,
      "step": 42475
    },
    {
      "entropy": 5.937855768203735,
      "epoch": 4.246313790173439,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033174024140020556,
      "loss": 5.3912,
      "mean_token_accuracy": 0.21211264431476592,
      "num_tokens": 83107232.0,
      "step": 42480
    },
    {
      "entropy": 5.898540353775024,
      "epoch": 4.246813615234668,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033170570503373545,
      "loss": 5.3089,
      "mean_token_accuracy": 0.2174858942627907,
      "num_tokens": 83117363.0,
      "step": 42485
    },
    {
      "entropy": 5.859898042678833,
      "epoch": 4.247313440295897,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003316711672405644,
      "loss": 5.3118,
      "mean_token_accuracy": 0.21161730885505675,
      "num_tokens": 83127489.0,
      "step": 42490
    },
    {
      "entropy": 5.785232496261597,
      "epoch": 4.2478132653571254,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033163662802156136,
      "loss": 5.1678,
      "mean_token_accuracy": 0.22432128041982652,
      "num_tokens": 83136318.0,
      "step": 42495
    },
    {
      "entropy": 5.8018499374389645,
      "epoch": 4.248313090418353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033160208737759535,
      "loss": 5.2742,
      "mean_token_accuracy": 0.21386671513319017,
      "num_tokens": 83145971.0,
      "step": 42500
    },
    {
      "entropy": 5.830384826660156,
      "epoch": 4.248812915479582,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003315675453095353,
      "loss": 5.3328,
      "mean_token_accuracy": 0.20319225192070006,
      "num_tokens": 83155457.0,
      "step": 42505
    },
    {
      "entropy": 5.755175447463989,
      "epoch": 4.249312740540811,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003315330018182504,
      "loss": 5.1723,
      "mean_token_accuracy": 0.22421281635761262,
      "num_tokens": 83164309.0,
      "step": 42510
    },
    {
      "entropy": 5.771506309509277,
      "epoch": 4.249812565602039,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033149845690460974,
      "loss": 5.1963,
      "mean_token_accuracy": 0.22007442116737366,
      "num_tokens": 83174091.0,
      "step": 42515
    },
    {
      "entropy": 5.852291774749756,
      "epoch": 4.250312390663268,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033146391056948244,
      "loss": 5.3323,
      "mean_token_accuracy": 0.2082412287592888,
      "num_tokens": 83184312.0,
      "step": 42520
    },
    {
      "entropy": 5.795002698898315,
      "epoch": 4.250812215724497,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033142936281373777,
      "loss": 5.1679,
      "mean_token_accuracy": 0.23077239841222763,
      "num_tokens": 83193261.0,
      "step": 42525
    },
    {
      "entropy": 5.857014560699463,
      "epoch": 4.251312040785725,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033139481363824477,
      "loss": 5.3013,
      "mean_token_accuracy": 0.2142689049243927,
      "num_tokens": 83202810.0,
      "step": 42530
    },
    {
      "entropy": 5.814216136932373,
      "epoch": 4.251811865846953,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033136026304387276,
      "loss": 5.3001,
      "mean_token_accuracy": 0.20947002470493317,
      "num_tokens": 83212267.0,
      "step": 42535
    },
    {
      "entropy": 5.84621295928955,
      "epoch": 4.252311690908182,
      "grad_norm": 1.1875,
      "learning_rate": 0.000331325711031491,
      "loss": 5.2888,
      "mean_token_accuracy": 0.21817172467708587,
      "num_tokens": 83221709.0,
      "step": 42540
    },
    {
      "entropy": 5.859774255752564,
      "epoch": 4.252811515969411,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033129115760196885,
      "loss": 5.3007,
      "mean_token_accuracy": 0.20694604218006135,
      "num_tokens": 83231907.0,
      "step": 42545
    },
    {
      "entropy": 5.842284202575684,
      "epoch": 4.253311341030639,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033125660275617563,
      "loss": 5.2515,
      "mean_token_accuracy": 0.21893117129802703,
      "num_tokens": 83242341.0,
      "step": 42550
    },
    {
      "entropy": 5.808683252334594,
      "epoch": 4.253811166091868,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003312220464949809,
      "loss": 5.2436,
      "mean_token_accuracy": 0.2197725221514702,
      "num_tokens": 83252061.0,
      "step": 42555
    },
    {
      "entropy": 5.760774040222168,
      "epoch": 4.254310991153097,
      "grad_norm": 1.25,
      "learning_rate": 0.0003311874888192539,
      "loss": 5.1413,
      "mean_token_accuracy": 0.22495715618133544,
      "num_tokens": 83261455.0,
      "step": 42560
    },
    {
      "entropy": 5.7425884246826175,
      "epoch": 4.254810816214325,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003311529297298641,
      "loss": 5.1428,
      "mean_token_accuracy": 0.23043348640203476,
      "num_tokens": 83270833.0,
      "step": 42565
    },
    {
      "entropy": 5.703839588165283,
      "epoch": 4.255310641275553,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003311183692276811,
      "loss": 5.1281,
      "mean_token_accuracy": 0.22339571863412858,
      "num_tokens": 83280128.0,
      "step": 42570
    },
    {
      "entropy": 5.757896995544433,
      "epoch": 4.255810466336782,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033108380731357426,
      "loss": 5.2234,
      "mean_token_accuracy": 0.21312447935342788,
      "num_tokens": 83290026.0,
      "step": 42575
    },
    {
      "entropy": 5.801152563095092,
      "epoch": 4.256310291398011,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003310492439884132,
      "loss": 5.2043,
      "mean_token_accuracy": 0.214474618434906,
      "num_tokens": 83299665.0,
      "step": 42580
    },
    {
      "entropy": 5.870730876922607,
      "epoch": 4.256810116459239,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033101467925306777,
      "loss": 5.2673,
      "mean_token_accuracy": 0.216375470161438,
      "num_tokens": 83308353.0,
      "step": 42585
    },
    {
      "entropy": 5.784344673156738,
      "epoch": 4.257309941520468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003309801131084074,
      "loss": 5.2154,
      "mean_token_accuracy": 0.21240832656621933,
      "num_tokens": 83317479.0,
      "step": 42590
    },
    {
      "entropy": 5.7869688987731935,
      "epoch": 4.257809766581697,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033094554555530175,
      "loss": 5.1807,
      "mean_token_accuracy": 0.22752434015274048,
      "num_tokens": 83327493.0,
      "step": 42595
    },
    {
      "entropy": 5.814344835281372,
      "epoch": 4.258309591642925,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003309109765946206,
      "loss": 5.2287,
      "mean_token_accuracy": 0.2161886975169182,
      "num_tokens": 83336765.0,
      "step": 42600
    },
    {
      "entropy": 5.787954902648925,
      "epoch": 4.258809416704153,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033087640622723357,
      "loss": 5.2776,
      "mean_token_accuracy": 0.20812959372997283,
      "num_tokens": 83347172.0,
      "step": 42605
    },
    {
      "entropy": 5.863299751281739,
      "epoch": 4.259309241765382,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003308418344540106,
      "loss": 5.2815,
      "mean_token_accuracy": 0.21131037473678588,
      "num_tokens": 83356606.0,
      "step": 42610
    },
    {
      "entropy": 5.8534599304199215,
      "epoch": 4.259809066826611,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033080726127582147,
      "loss": 5.2675,
      "mean_token_accuracy": 0.21412750482559204,
      "num_tokens": 83367124.0,
      "step": 42615
    },
    {
      "entropy": 5.753681945800781,
      "epoch": 4.260308891887839,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003307726866935359,
      "loss": 5.172,
      "mean_token_accuracy": 0.23323168456554413,
      "num_tokens": 83377081.0,
      "step": 42620
    },
    {
      "entropy": 5.778512287139892,
      "epoch": 4.260808716949068,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003307381107080241,
      "loss": 5.1755,
      "mean_token_accuracy": 0.21442078948020935,
      "num_tokens": 83387873.0,
      "step": 42625
    },
    {
      "entropy": 5.928755521774292,
      "epoch": 4.261308542010297,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003307035333201556,
      "loss": 5.4317,
      "mean_token_accuracy": 0.20686963200569153,
      "num_tokens": 83398558.0,
      "step": 42630
    },
    {
      "entropy": 5.83690505027771,
      "epoch": 4.2618083670715246,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003306689545308006,
      "loss": 5.2498,
      "mean_token_accuracy": 0.21419091671705245,
      "num_tokens": 83408435.0,
      "step": 42635
    },
    {
      "entropy": 5.840997552871704,
      "epoch": 4.262308192132753,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033063437434082896,
      "loss": 5.3025,
      "mean_token_accuracy": 0.2060966596007347,
      "num_tokens": 83417401.0,
      "step": 42640
    },
    {
      "entropy": 5.799341154098511,
      "epoch": 4.262808017193982,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003305997927511108,
      "loss": 5.1874,
      "mean_token_accuracy": 0.2219747558236122,
      "num_tokens": 83426591.0,
      "step": 42645
    },
    {
      "entropy": 5.7316429138183596,
      "epoch": 4.263307842255211,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033056520976251617,
      "loss": 5.2413,
      "mean_token_accuracy": 0.2120353803038597,
      "num_tokens": 83435310.0,
      "step": 42650
    },
    {
      "entropy": 5.74264760017395,
      "epoch": 4.263807667316439,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033053062537591516,
      "loss": 5.2061,
      "mean_token_accuracy": 0.22216561138629914,
      "num_tokens": 83444564.0,
      "step": 42655
    },
    {
      "entropy": 5.843513250350952,
      "epoch": 4.264307492377668,
      "grad_norm": 1.125,
      "learning_rate": 0.000330496039592178,
      "loss": 5.305,
      "mean_token_accuracy": 0.2076364576816559,
      "num_tokens": 83453848.0,
      "step": 42660
    },
    {
      "entropy": 5.806061887741089,
      "epoch": 4.264807317438897,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033046145241217466,
      "loss": 5.2326,
      "mean_token_accuracy": 0.2179896727204323,
      "num_tokens": 83463380.0,
      "step": 42665
    },
    {
      "entropy": 5.801877498626709,
      "epoch": 4.2653071425001245,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003304268638367754,
      "loss": 5.2254,
      "mean_token_accuracy": 0.2190507709980011,
      "num_tokens": 83472085.0,
      "step": 42670
    },
    {
      "entropy": 5.721190166473389,
      "epoch": 4.265806967561353,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033039227386685063,
      "loss": 5.1881,
      "mean_token_accuracy": 0.22106679081916808,
      "num_tokens": 83483144.0,
      "step": 42675
    },
    {
      "entropy": 5.823439741134644,
      "epoch": 4.266306792622582,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033035768250327046,
      "loss": 5.2264,
      "mean_token_accuracy": 0.2208481103181839,
      "num_tokens": 83492726.0,
      "step": 42680
    },
    {
      "entropy": 5.841978645324707,
      "epoch": 4.266806617683811,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003303230897469052,
      "loss": 5.3,
      "mean_token_accuracy": 0.21308565735816956,
      "num_tokens": 83501935.0,
      "step": 42685
    },
    {
      "entropy": 5.809044122695923,
      "epoch": 4.267306442745039,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003302884955986253,
      "loss": 5.2056,
      "mean_token_accuracy": 0.2175329715013504,
      "num_tokens": 83512010.0,
      "step": 42690
    },
    {
      "entropy": 5.771369934082031,
      "epoch": 4.267806267806268,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003302539000593011,
      "loss": 5.1287,
      "mean_token_accuracy": 0.22438924610614777,
      "num_tokens": 83521302.0,
      "step": 42695
    },
    {
      "entropy": 5.794908714294434,
      "epoch": 4.268306092867497,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003302193031298029,
      "loss": 5.2643,
      "mean_token_accuracy": 0.21420860439538955,
      "num_tokens": 83531964.0,
      "step": 42700
    },
    {
      "entropy": 5.735476112365722,
      "epoch": 4.2688059179287245,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033018470481100125,
      "loss": 5.2485,
      "mean_token_accuracy": 0.21308553367853164,
      "num_tokens": 83541568.0,
      "step": 42705
    },
    {
      "entropy": 5.856034994125366,
      "epoch": 4.269305742989953,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033015010510376676,
      "loss": 5.3397,
      "mean_token_accuracy": 0.20298164635896682,
      "num_tokens": 83551822.0,
      "step": 42710
    },
    {
      "entropy": 5.67875919342041,
      "epoch": 4.269805568051182,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033011550400896967,
      "loss": 4.9825,
      "mean_token_accuracy": 0.2430076628923416,
      "num_tokens": 83561444.0,
      "step": 42715
    },
    {
      "entropy": 5.794744873046875,
      "epoch": 4.270305393112411,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033008090152748077,
      "loss": 5.2442,
      "mean_token_accuracy": 0.21052955687046052,
      "num_tokens": 83573176.0,
      "step": 42720
    },
    {
      "entropy": 5.802496004104614,
      "epoch": 4.270805218173639,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033004629766017046,
      "loss": 5.3046,
      "mean_token_accuracy": 0.20773365199565888,
      "num_tokens": 83585528.0,
      "step": 42725
    },
    {
      "entropy": 5.820210886001587,
      "epoch": 4.271305043234868,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033001169240790955,
      "loss": 5.3567,
      "mean_token_accuracy": 0.2063666418194771,
      "num_tokens": 83596750.0,
      "step": 42730
    },
    {
      "entropy": 5.704326105117798,
      "epoch": 4.271804868296097,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003299770857715686,
      "loss": 5.1948,
      "mean_token_accuracy": 0.22396240383386612,
      "num_tokens": 83607775.0,
      "step": 42735
    },
    {
      "entropy": 5.881990957260132,
      "epoch": 4.2723046933573245,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003299424777520183,
      "loss": 5.332,
      "mean_token_accuracy": 0.19923578798770905,
      "num_tokens": 83618416.0,
      "step": 42740
    },
    {
      "entropy": 5.845259475708008,
      "epoch": 4.272804518418553,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032990786835012943,
      "loss": 5.259,
      "mean_token_accuracy": 0.22263707220554352,
      "num_tokens": 83628163.0,
      "step": 42745
    },
    {
      "entropy": 5.810767650604248,
      "epoch": 4.273304343479782,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032987325756677266,
      "loss": 5.2426,
      "mean_token_accuracy": 0.2140217214822769,
      "num_tokens": 83637764.0,
      "step": 42750
    },
    {
      "entropy": 5.745815658569336,
      "epoch": 4.273804168541011,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003298386454028189,
      "loss": 5.1592,
      "mean_token_accuracy": 0.22856337428092957,
      "num_tokens": 83647101.0,
      "step": 42755
    },
    {
      "entropy": 5.787706184387207,
      "epoch": 4.274303993602239,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003298040318591389,
      "loss": 5.2282,
      "mean_token_accuracy": 0.22240933775901794,
      "num_tokens": 83656725.0,
      "step": 42760
    },
    {
      "entropy": 5.6800370693206785,
      "epoch": 4.274803818663468,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003297694169366036,
      "loss": 5.1264,
      "mean_token_accuracy": 0.23121309280395508,
      "num_tokens": 83667574.0,
      "step": 42765
    },
    {
      "entropy": 5.828589344024659,
      "epoch": 4.275303643724697,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003297348006360839,
      "loss": 5.2999,
      "mean_token_accuracy": 0.211051307618618,
      "num_tokens": 83677466.0,
      "step": 42770
    },
    {
      "entropy": 5.859263563156128,
      "epoch": 4.2758034687859245,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003297001829584506,
      "loss": 5.3312,
      "mean_token_accuracy": 0.20719095766544343,
      "num_tokens": 83686815.0,
      "step": 42775
    },
    {
      "entropy": 5.868814420700073,
      "epoch": 4.276303293847153,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032966556390457487,
      "loss": 5.2412,
      "mean_token_accuracy": 0.2151699736714363,
      "num_tokens": 83695596.0,
      "step": 42780
    },
    {
      "entropy": 5.830060768127441,
      "epoch": 4.276803118908382,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032963094347532757,
      "loss": 5.3294,
      "mean_token_accuracy": 0.2082052394747734,
      "num_tokens": 83706131.0,
      "step": 42785
    },
    {
      "entropy": 5.913273286819458,
      "epoch": 4.277302943969611,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032959632167157976,
      "loss": 5.2907,
      "mean_token_accuracy": 0.2140103578567505,
      "num_tokens": 83717358.0,
      "step": 42790
    },
    {
      "entropy": 5.906132078170776,
      "epoch": 4.277802769030839,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003295616984942026,
      "loss": 5.3264,
      "mean_token_accuracy": 0.20764122009277344,
      "num_tokens": 83728034.0,
      "step": 42795
    },
    {
      "entropy": 5.826951694488526,
      "epoch": 4.278302594092068,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003295270739440671,
      "loss": 5.3018,
      "mean_token_accuracy": 0.213693006336689,
      "num_tokens": 83737379.0,
      "step": 42800
    },
    {
      "entropy": 5.899770355224609,
      "epoch": 4.278802419153297,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032949244802204445,
      "loss": 5.2862,
      "mean_token_accuracy": 0.21215655654668808,
      "num_tokens": 83746923.0,
      "step": 42805
    },
    {
      "entropy": 5.853477144241333,
      "epoch": 4.2793022442145245,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032945782072900593,
      "loss": 5.1804,
      "mean_token_accuracy": 0.22074491381645203,
      "num_tokens": 83757195.0,
      "step": 42810
    },
    {
      "entropy": 5.829100513458252,
      "epoch": 4.279802069275753,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003294231920658225,
      "loss": 5.3178,
      "mean_token_accuracy": 0.20125021487474443,
      "num_tokens": 83766743.0,
      "step": 42815
    },
    {
      "entropy": 5.811333894729614,
      "epoch": 4.280301894336982,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003293885620333657,
      "loss": 5.2792,
      "mean_token_accuracy": 0.20697956383228303,
      "num_tokens": 83776766.0,
      "step": 42820
    },
    {
      "entropy": 5.819811534881592,
      "epoch": 4.280801719398211,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032935393063250655,
      "loss": 5.2662,
      "mean_token_accuracy": 0.21373796463012695,
      "num_tokens": 83786381.0,
      "step": 42825
    },
    {
      "entropy": 5.743377351760865,
      "epoch": 4.281301544459439,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003293192978641166,
      "loss": 5.1274,
      "mean_token_accuracy": 0.22801164239645005,
      "num_tokens": 83797311.0,
      "step": 42830
    },
    {
      "entropy": 5.791549634933472,
      "epoch": 4.281801369520668,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000329284663729067,
      "loss": 5.2277,
      "mean_token_accuracy": 0.21111281365156173,
      "num_tokens": 83806269.0,
      "step": 42835
    },
    {
      "entropy": 5.779216575622558,
      "epoch": 4.282301194581897,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003292500282282294,
      "loss": 5.2091,
      "mean_token_accuracy": 0.21334741860628129,
      "num_tokens": 83815834.0,
      "step": 42840
    },
    {
      "entropy": 5.845602416992188,
      "epoch": 4.2828010196431245,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032921539136247487,
      "loss": 5.3384,
      "mean_token_accuracy": 0.2061248779296875,
      "num_tokens": 83825066.0,
      "step": 42845
    },
    {
      "entropy": 5.889380645751953,
      "epoch": 4.283300844704353,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032918075313267503,
      "loss": 5.4533,
      "mean_token_accuracy": 0.201346555352211,
      "num_tokens": 83835287.0,
      "step": 42850
    },
    {
      "entropy": 5.823247671127319,
      "epoch": 4.283800669765582,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032914611353970145,
      "loss": 5.2999,
      "mean_token_accuracy": 0.20961834490299225,
      "num_tokens": 83845319.0,
      "step": 42855
    },
    {
      "entropy": 5.959419679641724,
      "epoch": 4.284300494826811,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003291114725844256,
      "loss": 5.3863,
      "mean_token_accuracy": 0.20784546583890914,
      "num_tokens": 83856298.0,
      "step": 42860
    },
    {
      "entropy": 5.79602165222168,
      "epoch": 4.284800319888039,
      "grad_norm": 1.171875,
      "learning_rate": 0.000329076830267719,
      "loss": 5.1541,
      "mean_token_accuracy": 0.23540986627340316,
      "num_tokens": 83865269.0,
      "step": 42865
    },
    {
      "entropy": 5.772212076187134,
      "epoch": 4.285300144949268,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003290421865904533,
      "loss": 5.2282,
      "mean_token_accuracy": 0.21543618589639663,
      "num_tokens": 83875030.0,
      "step": 42870
    },
    {
      "entropy": 5.778520822525024,
      "epoch": 4.285799970010497,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003290075415535001,
      "loss": 5.286,
      "mean_token_accuracy": 0.20688191056251526,
      "num_tokens": 83883730.0,
      "step": 42875
    },
    {
      "entropy": 5.767379713058472,
      "epoch": 4.2862997950717245,
      "grad_norm": 1.296875,
      "learning_rate": 0.000328972895157731,
      "loss": 5.1635,
      "mean_token_accuracy": 0.22198910862207413,
      "num_tokens": 83893066.0,
      "step": 42880
    },
    {
      "entropy": 5.7612629413604735,
      "epoch": 4.286799620132953,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032893824740401776,
      "loss": 5.2399,
      "mean_token_accuracy": 0.22385061383247376,
      "num_tokens": 83903180.0,
      "step": 42885
    },
    {
      "entropy": 5.847148656845093,
      "epoch": 4.287299445194182,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032890359829323214,
      "loss": 5.302,
      "mean_token_accuracy": 0.21146064549684523,
      "num_tokens": 83912424.0,
      "step": 42890
    },
    {
      "entropy": 5.839828681945801,
      "epoch": 4.287799270255411,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003288689478262458,
      "loss": 5.289,
      "mean_token_accuracy": 0.20853736847639084,
      "num_tokens": 83921563.0,
      "step": 42895
    },
    {
      "entropy": 5.821217012405396,
      "epoch": 4.288299095316639,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032883429600393053,
      "loss": 5.2617,
      "mean_token_accuracy": 0.21501700729131698,
      "num_tokens": 83932298.0,
      "step": 42900
    },
    {
      "entropy": 5.8118547916412355,
      "epoch": 4.288798920377868,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032879964282715843,
      "loss": 5.2145,
      "mean_token_accuracy": 0.21630696058273316,
      "num_tokens": 83942030.0,
      "step": 42905
    },
    {
      "entropy": 5.816927671432495,
      "epoch": 4.289298745439097,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000328764988296801,
      "loss": 5.1835,
      "mean_token_accuracy": 0.21929871588945388,
      "num_tokens": 83951611.0,
      "step": 42910
    },
    {
      "entropy": 5.832354307174683,
      "epoch": 4.2897985705003245,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032873033241373034,
      "loss": 5.3287,
      "mean_token_accuracy": 0.20941320359706878,
      "num_tokens": 83962314.0,
      "step": 42915
    },
    {
      "entropy": 5.789675951004028,
      "epoch": 4.290298395561553,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003286956751788184,
      "loss": 5.2466,
      "mean_token_accuracy": 0.2188074067234993,
      "num_tokens": 83971995.0,
      "step": 42920
    },
    {
      "entropy": 5.895315742492675,
      "epoch": 4.290798220622782,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032866101659293693,
      "loss": 5.3138,
      "mean_token_accuracy": 0.207717864215374,
      "num_tokens": 83982047.0,
      "step": 42925
    },
    {
      "entropy": 5.828972864151001,
      "epoch": 4.291298045684011,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032862635665695825,
      "loss": 5.2555,
      "mean_token_accuracy": 0.2064334601163864,
      "num_tokens": 83992157.0,
      "step": 42930
    },
    {
      "entropy": 5.866233968734742,
      "epoch": 4.291797870745239,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032859169537175414,
      "loss": 5.2456,
      "mean_token_accuracy": 0.2109756126999855,
      "num_tokens": 84001118.0,
      "step": 42935
    },
    {
      "entropy": 5.862402105331421,
      "epoch": 4.292297695806468,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032855703273819686,
      "loss": 5.3847,
      "mean_token_accuracy": 0.20491038858890534,
      "num_tokens": 84011561.0,
      "step": 42940
    },
    {
      "entropy": 5.8506114959716795,
      "epoch": 4.292797520867696,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032852236875715836,
      "loss": 5.2754,
      "mean_token_accuracy": 0.20971803814172746,
      "num_tokens": 84020356.0,
      "step": 42945
    },
    {
      "entropy": 5.88217945098877,
      "epoch": 4.2932973459289245,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032848770342951085,
      "loss": 5.3511,
      "mean_token_accuracy": 0.21873178631067275,
      "num_tokens": 84030390.0,
      "step": 42950
    },
    {
      "entropy": 5.796022224426269,
      "epoch": 4.293797170990153,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003284530367561264,
      "loss": 5.2091,
      "mean_token_accuracy": 0.22026447504758834,
      "num_tokens": 84039552.0,
      "step": 42955
    },
    {
      "entropy": 5.771995258331299,
      "epoch": 4.294296996051382,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003284183687378774,
      "loss": 5.1875,
      "mean_token_accuracy": 0.22126247435808183,
      "num_tokens": 84049030.0,
      "step": 42960
    },
    {
      "entropy": 5.837738037109375,
      "epoch": 4.294796821112611,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032838369937563587,
      "loss": 5.3511,
      "mean_token_accuracy": 0.20929986089468003,
      "num_tokens": 84058744.0,
      "step": 42965
    },
    {
      "entropy": 5.750711965560913,
      "epoch": 4.295296646173839,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032834902867027435,
      "loss": 5.1237,
      "mean_token_accuracy": 0.22919016778469087,
      "num_tokens": 84067921.0,
      "step": 42970
    },
    {
      "entropy": 5.869443750381469,
      "epoch": 4.295796471235068,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032831435662266495,
      "loss": 5.3241,
      "mean_token_accuracy": 0.2100876346230507,
      "num_tokens": 84077299.0,
      "step": 42975
    },
    {
      "entropy": 5.873395299911499,
      "epoch": 4.296296296296296,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003282796832336801,
      "loss": 5.3964,
      "mean_token_accuracy": 0.20731313675642013,
      "num_tokens": 84087161.0,
      "step": 42980
    },
    {
      "entropy": 5.779802846908569,
      "epoch": 4.2967961213575245,
      "grad_norm": 1.0625,
      "learning_rate": 0.000328245008504192,
      "loss": 5.2136,
      "mean_token_accuracy": 0.22354751974344253,
      "num_tokens": 84097105.0,
      "step": 42985
    },
    {
      "entropy": 5.791943597793579,
      "epoch": 4.297295946418753,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032821033243507323,
      "loss": 5.2287,
      "mean_token_accuracy": 0.2181503653526306,
      "num_tokens": 84107012.0,
      "step": 42990
    },
    {
      "entropy": 5.935891437530517,
      "epoch": 4.297795771479982,
      "grad_norm": 1.125,
      "learning_rate": 0.0003281756550271962,
      "loss": 5.3489,
      "mean_token_accuracy": 0.20446315854787828,
      "num_tokens": 84117465.0,
      "step": 42995
    },
    {
      "entropy": 5.7681221008300785,
      "epoch": 4.298295596541211,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003281409762814334,
      "loss": 5.2869,
      "mean_token_accuracy": 0.2147226020693779,
      "num_tokens": 84127593.0,
      "step": 43000
    },
    {
      "entropy": 5.853283166885376,
      "epoch": 4.298795421602439,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003281062961986573,
      "loss": 5.2516,
      "mean_token_accuracy": 0.21125716120004653,
      "num_tokens": 84136805.0,
      "step": 43005
    },
    {
      "entropy": 5.821894264221191,
      "epoch": 4.299295246663668,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003280716147797405,
      "loss": 5.2379,
      "mean_token_accuracy": 0.21480585336685182,
      "num_tokens": 84146246.0,
      "step": 43010
    },
    {
      "entropy": 5.7858476638793945,
      "epoch": 4.299795071724896,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032803693202555545,
      "loss": 5.2107,
      "mean_token_accuracy": 0.2224339723587036,
      "num_tokens": 84156353.0,
      "step": 43015
    },
    {
      "entropy": 5.718235635757447,
      "epoch": 4.3002948967861245,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032800224793697486,
      "loss": 5.175,
      "mean_token_accuracy": 0.21910586953163147,
      "num_tokens": 84166563.0,
      "step": 43020
    },
    {
      "entropy": 5.821954727172852,
      "epoch": 4.300794721847353,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032796756251487135,
      "loss": 5.2622,
      "mean_token_accuracy": 0.20649763345718383,
      "num_tokens": 84175691.0,
      "step": 43025
    },
    {
      "entropy": 5.8413074016571045,
      "epoch": 4.301294546908582,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003279328757601176,
      "loss": 5.2737,
      "mean_token_accuracy": 0.2161419376730919,
      "num_tokens": 84185525.0,
      "step": 43030
    },
    {
      "entropy": 5.832145547866821,
      "epoch": 4.301794371969811,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003278981876735864,
      "loss": 5.324,
      "mean_token_accuracy": 0.21053490936756133,
      "num_tokens": 84195165.0,
      "step": 43035
    },
    {
      "entropy": 5.860287237167358,
      "epoch": 4.302294197031039,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032786349825615033,
      "loss": 5.3329,
      "mean_token_accuracy": 0.20630977749824525,
      "num_tokens": 84205987.0,
      "step": 43040
    },
    {
      "entropy": 5.871081829071045,
      "epoch": 4.302794022092268,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003278288075086822,
      "loss": 5.2652,
      "mean_token_accuracy": 0.21159233450889586,
      "num_tokens": 84216007.0,
      "step": 43045
    },
    {
      "entropy": 5.775799751281738,
      "epoch": 4.303293847153496,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003277941154320549,
      "loss": 5.2172,
      "mean_token_accuracy": 0.22105747759342192,
      "num_tokens": 84226019.0,
      "step": 43050
    },
    {
      "entropy": 5.769150638580323,
      "epoch": 4.3037936722147245,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032775942202714126,
      "loss": 5.1994,
      "mean_token_accuracy": 0.2222549468278885,
      "num_tokens": 84236743.0,
      "step": 43055
    },
    {
      "entropy": 5.715375900268555,
      "epoch": 4.304293497275953,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032772472729481416,
      "loss": 5.1605,
      "mean_token_accuracy": 0.22840346097946168,
      "num_tokens": 84245933.0,
      "step": 43060
    },
    {
      "entropy": 5.899035930633545,
      "epoch": 4.304793322337182,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032769003123594645,
      "loss": 5.375,
      "mean_token_accuracy": 0.20599807053804398,
      "num_tokens": 84256454.0,
      "step": 43065
    },
    {
      "entropy": 5.856035900115967,
      "epoch": 4.305293147398411,
      "grad_norm": 1.125,
      "learning_rate": 0.0003276553338514111,
      "loss": 5.2645,
      "mean_token_accuracy": 0.2135786384344101,
      "num_tokens": 84266693.0,
      "step": 43070
    },
    {
      "entropy": 5.774238109588623,
      "epoch": 4.305792972459639,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003276206351420811,
      "loss": 5.2113,
      "mean_token_accuracy": 0.21704218089580535,
      "num_tokens": 84276425.0,
      "step": 43075
    },
    {
      "entropy": 5.872923517227173,
      "epoch": 4.306292797520868,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032758593510882954,
      "loss": 5.294,
      "mean_token_accuracy": 0.20967296063899993,
      "num_tokens": 84286871.0,
      "step": 43080
    },
    {
      "entropy": 5.811597871780395,
      "epoch": 4.306792622582096,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003275512337525293,
      "loss": 5.2794,
      "mean_token_accuracy": 0.20849508941173553,
      "num_tokens": 84296570.0,
      "step": 43085
    },
    {
      "entropy": 5.83207631111145,
      "epoch": 4.3072924476433245,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032751653107405363,
      "loss": 5.3042,
      "mean_token_accuracy": 0.21042310744524,
      "num_tokens": 84306066.0,
      "step": 43090
    },
    {
      "entropy": 5.849385023117065,
      "epoch": 4.307792272704553,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032748182707427546,
      "loss": 5.3068,
      "mean_token_accuracy": 0.2071021243929863,
      "num_tokens": 84315668.0,
      "step": 43095
    },
    {
      "entropy": 5.802595138549805,
      "epoch": 4.308292097765782,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032744712175406806,
      "loss": 5.2472,
      "mean_token_accuracy": 0.21829868406057357,
      "num_tokens": 84325111.0,
      "step": 43100
    },
    {
      "entropy": 5.821097326278687,
      "epoch": 4.308791922827011,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003274124151143046,
      "loss": 5.2894,
      "mean_token_accuracy": 0.21281442642211915,
      "num_tokens": 84334466.0,
      "step": 43105
    },
    {
      "entropy": 5.788319396972656,
      "epoch": 4.309291747888239,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003273777071558582,
      "loss": 5.299,
      "mean_token_accuracy": 0.21505355834960938,
      "num_tokens": 84343935.0,
      "step": 43110
    },
    {
      "entropy": 5.78162202835083,
      "epoch": 4.309791572949468,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003273429978796022,
      "loss": 5.2618,
      "mean_token_accuracy": 0.2182037740945816,
      "num_tokens": 84353654.0,
      "step": 43115
    },
    {
      "entropy": 5.790529870986939,
      "epoch": 4.310291398010696,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032730828728640983,
      "loss": 5.2145,
      "mean_token_accuracy": 0.2262528046965599,
      "num_tokens": 84363016.0,
      "step": 43120
    },
    {
      "entropy": 5.751796054840088,
      "epoch": 4.3107912230719245,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032727357537715445,
      "loss": 5.1637,
      "mean_token_accuracy": 0.21970763355493544,
      "num_tokens": 84373094.0,
      "step": 43125
    },
    {
      "entropy": 5.827004337310791,
      "epoch": 4.311291048133153,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003272388621527093,
      "loss": 5.21,
      "mean_token_accuracy": 0.2185005486011505,
      "num_tokens": 84382470.0,
      "step": 43130
    },
    {
      "entropy": 5.871889448165893,
      "epoch": 4.311790873194382,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003272041476139479,
      "loss": 5.3063,
      "mean_token_accuracy": 0.20407887399196625,
      "num_tokens": 84391705.0,
      "step": 43135
    },
    {
      "entropy": 5.80635256767273,
      "epoch": 4.312290698255611,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032716943176174353,
      "loss": 5.2315,
      "mean_token_accuracy": 0.22168980091810225,
      "num_tokens": 84401857.0,
      "step": 43140
    },
    {
      "entropy": 5.7897358417510985,
      "epoch": 4.312790523316839,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003271347145969696,
      "loss": 5.2468,
      "mean_token_accuracy": 0.2203747570514679,
      "num_tokens": 84410891.0,
      "step": 43145
    },
    {
      "entropy": 5.774423599243164,
      "epoch": 4.313290348378068,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032709999612049985,
      "loss": 5.2433,
      "mean_token_accuracy": 0.22773183584213258,
      "num_tokens": 84420373.0,
      "step": 43150
    },
    {
      "entropy": 5.720557022094726,
      "epoch": 4.313790173439296,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003270652763332075,
      "loss": 5.109,
      "mean_token_accuracy": 0.24033474624156953,
      "num_tokens": 84430494.0,
      "step": 43155
    },
    {
      "entropy": 5.783394765853882,
      "epoch": 4.3142899985005245,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003270305552359662,
      "loss": 5.1824,
      "mean_token_accuracy": 0.21799142956733703,
      "num_tokens": 84439905.0,
      "step": 43160
    },
    {
      "entropy": 5.825526571273803,
      "epoch": 4.314789823561753,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032699583282964955,
      "loss": 5.2733,
      "mean_token_accuracy": 0.21131361424922943,
      "num_tokens": 84449136.0,
      "step": 43165
    },
    {
      "entropy": 5.876477241516113,
      "epoch": 4.315289648622982,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032696110911513106,
      "loss": 5.3053,
      "mean_token_accuracy": 0.20201852172613144,
      "num_tokens": 84457884.0,
      "step": 43170
    },
    {
      "entropy": 5.857251405715942,
      "epoch": 4.315789473684211,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003269263840932845,
      "loss": 5.3091,
      "mean_token_accuracy": 0.2097030371427536,
      "num_tokens": 84468076.0,
      "step": 43175
    },
    {
      "entropy": 5.861142969131469,
      "epoch": 4.316289298745439,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032689165776498343,
      "loss": 5.2768,
      "mean_token_accuracy": 0.2170243129134178,
      "num_tokens": 84476839.0,
      "step": 43180
    },
    {
      "entropy": 5.815202283859253,
      "epoch": 4.316789123806668,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032685693013110163,
      "loss": 5.2114,
      "mean_token_accuracy": 0.2174171596765518,
      "num_tokens": 84485171.0,
      "step": 43185
    },
    {
      "entropy": 5.850450849533081,
      "epoch": 4.317288948867896,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032682220119251284,
      "loss": 5.3616,
      "mean_token_accuracy": 0.2160225987434387,
      "num_tokens": 84494049.0,
      "step": 43190
    },
    {
      "entropy": 5.820397520065308,
      "epoch": 4.3177887739291245,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032678747095009076,
      "loss": 5.3138,
      "mean_token_accuracy": 0.20830436497926713,
      "num_tokens": 84504568.0,
      "step": 43195
    },
    {
      "entropy": 5.935927629470825,
      "epoch": 4.318288598990353,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032675273940470926,
      "loss": 5.385,
      "mean_token_accuracy": 0.20304795652627944,
      "num_tokens": 84514709.0,
      "step": 43200
    },
    {
      "entropy": 5.890158605575562,
      "epoch": 4.318788424051582,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032671800655724215,
      "loss": 5.3508,
      "mean_token_accuracy": 0.2099624752998352,
      "num_tokens": 84525391.0,
      "step": 43205
    },
    {
      "entropy": 5.852384042739868,
      "epoch": 4.319288249112811,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032668327240856324,
      "loss": 5.2562,
      "mean_token_accuracy": 0.2114489659667015,
      "num_tokens": 84535924.0,
      "step": 43210
    },
    {
      "entropy": 5.810789012908936,
      "epoch": 4.319788074174039,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003266485369595465,
      "loss": 5.2256,
      "mean_token_accuracy": 0.22300231754779815,
      "num_tokens": 84545934.0,
      "step": 43215
    },
    {
      "entropy": 5.842612838745117,
      "epoch": 4.320287899235268,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032661380021106593,
      "loss": 5.3423,
      "mean_token_accuracy": 0.20959662944078444,
      "num_tokens": 84555164.0,
      "step": 43220
    },
    {
      "entropy": 5.841032075881958,
      "epoch": 4.320787724296496,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003265790621639955,
      "loss": 5.2879,
      "mean_token_accuracy": 0.2035749301314354,
      "num_tokens": 84564497.0,
      "step": 43225
    },
    {
      "entropy": 5.800917625427246,
      "epoch": 4.3212875493577245,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032654432281920904,
      "loss": 5.2681,
      "mean_token_accuracy": 0.21528624445199968,
      "num_tokens": 84573405.0,
      "step": 43230
    },
    {
      "entropy": 5.888407278060913,
      "epoch": 4.321787374418953,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032650958217758067,
      "loss": 5.3079,
      "mean_token_accuracy": 0.21131356805562973,
      "num_tokens": 84582722.0,
      "step": 43235
    },
    {
      "entropy": 5.868539571762085,
      "epoch": 4.322287199480182,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003264748402399844,
      "loss": 5.2239,
      "mean_token_accuracy": 0.21483087986707688,
      "num_tokens": 84591727.0,
      "step": 43240
    },
    {
      "entropy": 5.8414835929870605,
      "epoch": 4.322787024541411,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003264400970072945,
      "loss": 5.2721,
      "mean_token_accuracy": 0.21038057208061217,
      "num_tokens": 84601317.0,
      "step": 43245
    },
    {
      "entropy": 5.771859979629516,
      "epoch": 4.323286849602639,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032640535248038493,
      "loss": 5.2295,
      "mean_token_accuracy": 0.2200597494840622,
      "num_tokens": 84609922.0,
      "step": 43250
    },
    {
      "entropy": 5.752679681777954,
      "epoch": 4.323786674663868,
      "grad_norm": 1.125,
      "learning_rate": 0.00032637060666013004,
      "loss": 5.2348,
      "mean_token_accuracy": 0.21027793139219284,
      "num_tokens": 84620390.0,
      "step": 43255
    },
    {
      "entropy": 5.945918321609497,
      "epoch": 4.324286499725096,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003263358595474039,
      "loss": 5.4086,
      "mean_token_accuracy": 0.20418666899204255,
      "num_tokens": 84630911.0,
      "step": 43260
    },
    {
      "entropy": 5.851253175735474,
      "epoch": 4.3247863247863245,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003263011111430806,
      "loss": 5.2452,
      "mean_token_accuracy": 0.2123860239982605,
      "num_tokens": 84640070.0,
      "step": 43265
    },
    {
      "entropy": 5.906716346740723,
      "epoch": 4.325286149847553,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032626636144803456,
      "loss": 5.3227,
      "mean_token_accuracy": 0.204634328186512,
      "num_tokens": 84649234.0,
      "step": 43270
    },
    {
      "entropy": 5.825758838653565,
      "epoch": 4.325785974908782,
      "grad_norm": 1.375,
      "learning_rate": 0.0003262316104631402,
      "loss": 5.2536,
      "mean_token_accuracy": 0.22317413687705995,
      "num_tokens": 84657916.0,
      "step": 43275
    },
    {
      "entropy": 5.839883327484131,
      "epoch": 4.326285799970011,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003261968581892716,
      "loss": 5.258,
      "mean_token_accuracy": 0.21273384094238282,
      "num_tokens": 84667710.0,
      "step": 43280
    },
    {
      "entropy": 5.767826509475708,
      "epoch": 4.326785625031239,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003261621046273032,
      "loss": 5.2088,
      "mean_token_accuracy": 0.21576348692178726,
      "num_tokens": 84677453.0,
      "step": 43285
    },
    {
      "entropy": 5.829834270477295,
      "epoch": 4.327285450092468,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003261273497781095,
      "loss": 5.2176,
      "mean_token_accuracy": 0.21265143305063247,
      "num_tokens": 84687191.0,
      "step": 43290
    },
    {
      "entropy": 5.816434049606324,
      "epoch": 4.327785275153696,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003260925936425648,
      "loss": 5.2382,
      "mean_token_accuracy": 0.22092608362436295,
      "num_tokens": 84696457.0,
      "step": 43295
    },
    {
      "entropy": 5.752780485153198,
      "epoch": 4.3282851002149245,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003260578362215435,
      "loss": 5.3026,
      "mean_token_accuracy": 0.21176513582468032,
      "num_tokens": 84706520.0,
      "step": 43300
    },
    {
      "entropy": 5.845981073379517,
      "epoch": 4.328784925276153,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003260230775159203,
      "loss": 5.2933,
      "mean_token_accuracy": 0.21254049092531205,
      "num_tokens": 84716100.0,
      "step": 43305
    },
    {
      "entropy": 5.804788875579834,
      "epoch": 4.329284750337382,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032598831752656953,
      "loss": 5.2147,
      "mean_token_accuracy": 0.21893080174922944,
      "num_tokens": 84725957.0,
      "step": 43310
    },
    {
      "entropy": 5.8039428234100345,
      "epoch": 4.329784575398611,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032595355625436584,
      "loss": 5.3245,
      "mean_token_accuracy": 0.20580228120088578,
      "num_tokens": 84735821.0,
      "step": 43315
    },
    {
      "entropy": 5.8651282787323,
      "epoch": 4.330284400459839,
      "grad_norm": 1.25,
      "learning_rate": 0.00032591879370018383,
      "loss": 5.2866,
      "mean_token_accuracy": 0.21528970152139665,
      "num_tokens": 84745346.0,
      "step": 43320
    },
    {
      "entropy": 5.880578804016113,
      "epoch": 4.330784225521068,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032588402986489807,
      "loss": 5.3047,
      "mean_token_accuracy": 0.21006587594747544,
      "num_tokens": 84755765.0,
      "step": 43325
    },
    {
      "entropy": 5.774072360992432,
      "epoch": 4.331284050582296,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003258492647493831,
      "loss": 5.1926,
      "mean_token_accuracy": 0.22143829315900804,
      "num_tokens": 84765198.0,
      "step": 43330
    },
    {
      "entropy": 5.787158966064453,
      "epoch": 4.3317838756435245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003258144983545138,
      "loss": 5.2107,
      "mean_token_accuracy": 0.2154918760061264,
      "num_tokens": 84775563.0,
      "step": 43335
    },
    {
      "entropy": 5.836574411392212,
      "epoch": 4.332283700704753,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032577973068116477,
      "loss": 5.2579,
      "mean_token_accuracy": 0.21246571093797684,
      "num_tokens": 84785934.0,
      "step": 43340
    },
    {
      "entropy": 5.801532506942749,
      "epoch": 4.332783525765982,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032574496173021087,
      "loss": 5.169,
      "mean_token_accuracy": 0.2260309636592865,
      "num_tokens": 84796130.0,
      "step": 43345
    },
    {
      "entropy": 5.801450586318969,
      "epoch": 4.333283350827211,
      "grad_norm": 1.25,
      "learning_rate": 0.00032571019150252667,
      "loss": 5.2245,
      "mean_token_accuracy": 0.22088906168937683,
      "num_tokens": 84804528.0,
      "step": 43350
    },
    {
      "entropy": 5.678386211395264,
      "epoch": 4.333783175888439,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032567541999898713,
      "loss": 5.1142,
      "mean_token_accuracy": 0.22976096719503403,
      "num_tokens": 84814179.0,
      "step": 43355
    },
    {
      "entropy": 5.748899173736572,
      "epoch": 4.334283000949668,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003256406472204671,
      "loss": 5.2233,
      "mean_token_accuracy": 0.21398009210824967,
      "num_tokens": 84823867.0,
      "step": 43360
    },
    {
      "entropy": 5.788099050521851,
      "epoch": 4.334782826010896,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032560587316784133,
      "loss": 5.2041,
      "mean_token_accuracy": 0.2184078350663185,
      "num_tokens": 84834523.0,
      "step": 43365
    },
    {
      "entropy": 5.838768482208252,
      "epoch": 4.3352826510721245,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003255710978419849,
      "loss": 5.2486,
      "mean_token_accuracy": 0.21246998012065887,
      "num_tokens": 84843988.0,
      "step": 43370
    },
    {
      "entropy": 5.789622402191162,
      "epoch": 4.335782476133353,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003255363212437727,
      "loss": 5.2665,
      "mean_token_accuracy": 0.21328274458646773,
      "num_tokens": 84853465.0,
      "step": 43375
    },
    {
      "entropy": 5.7884985446929935,
      "epoch": 4.336282301194582,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032550154337407963,
      "loss": 5.2217,
      "mean_token_accuracy": 0.2192873626947403,
      "num_tokens": 84863814.0,
      "step": 43380
    },
    {
      "entropy": 5.781397962570191,
      "epoch": 4.336782126255811,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032546676423378065,
      "loss": 5.2222,
      "mean_token_accuracy": 0.2238638997077942,
      "num_tokens": 84875006.0,
      "step": 43385
    },
    {
      "entropy": 5.845161247253418,
      "epoch": 4.337281951317039,
      "grad_norm": 1.125,
      "learning_rate": 0.0003254319838237509,
      "loss": 5.3116,
      "mean_token_accuracy": 0.2100035786628723,
      "num_tokens": 84883738.0,
      "step": 43390
    },
    {
      "entropy": 5.844991445541382,
      "epoch": 4.337781776378268,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003253972021448654,
      "loss": 5.3124,
      "mean_token_accuracy": 0.2052045837044716,
      "num_tokens": 84892134.0,
      "step": 43395
    },
    {
      "entropy": 5.781297731399536,
      "epoch": 4.338281601439496,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032536241919799933,
      "loss": 5.1896,
      "mean_token_accuracy": 0.22386245727539061,
      "num_tokens": 84902232.0,
      "step": 43400
    },
    {
      "entropy": 5.923470449447632,
      "epoch": 4.3387814265007245,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032532763498402774,
      "loss": 5.3956,
      "mean_token_accuracy": 0.20056448578834535,
      "num_tokens": 84912474.0,
      "step": 43405
    },
    {
      "entropy": 5.835474252700806,
      "epoch": 4.339281251561953,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003252928495038257,
      "loss": 5.2098,
      "mean_token_accuracy": 0.21570373624563216,
      "num_tokens": 84921836.0,
      "step": 43410
    },
    {
      "entropy": 5.817287540435791,
      "epoch": 4.339781076623182,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032525806275826856,
      "loss": 5.2935,
      "mean_token_accuracy": 0.2187624990940094,
      "num_tokens": 84931727.0,
      "step": 43415
    },
    {
      "entropy": 5.795641040802002,
      "epoch": 4.340280901684411,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003252232747482316,
      "loss": 5.2453,
      "mean_token_accuracy": 0.21849471777677537,
      "num_tokens": 84941283.0,
      "step": 43420
    },
    {
      "entropy": 5.7838798522949215,
      "epoch": 4.340780726745639,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003251884854745899,
      "loss": 5.1721,
      "mean_token_accuracy": 0.22931380569934845,
      "num_tokens": 84950854.0,
      "step": 43425
    },
    {
      "entropy": 5.809481382369995,
      "epoch": 4.341280551806868,
      "grad_norm": 1.125,
      "learning_rate": 0.0003251536949382188,
      "loss": 5.2271,
      "mean_token_accuracy": 0.21873949021100997,
      "num_tokens": 84960766.0,
      "step": 43430
    },
    {
      "entropy": 5.906341981887818,
      "epoch": 4.341780376868096,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032511890313999367,
      "loss": 5.3356,
      "mean_token_accuracy": 0.20795265436172486,
      "num_tokens": 84970373.0,
      "step": 43435
    },
    {
      "entropy": 5.776531505584717,
      "epoch": 4.3422802019293245,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032508411008078976,
      "loss": 5.2051,
      "mean_token_accuracy": 0.2242858290672302,
      "num_tokens": 84980999.0,
      "step": 43440
    },
    {
      "entropy": 5.78642635345459,
      "epoch": 4.342780026990553,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003250493157614826,
      "loss": 5.2191,
      "mean_token_accuracy": 0.21905931532382966,
      "num_tokens": 84990981.0,
      "step": 43445
    },
    {
      "entropy": 5.884062671661377,
      "epoch": 4.343279852051782,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032501452018294754,
      "loss": 5.3065,
      "mean_token_accuracy": 0.20617817640304564,
      "num_tokens": 85001680.0,
      "step": 43450
    },
    {
      "entropy": 5.801793193817138,
      "epoch": 4.343779677113011,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032497972334605995,
      "loss": 5.1896,
      "mean_token_accuracy": 0.22132279723882675,
      "num_tokens": 85011364.0,
      "step": 43455
    },
    {
      "entropy": 5.775814151763916,
      "epoch": 4.344279502174239,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032494492525169544,
      "loss": 5.2017,
      "mean_token_accuracy": 0.22437286972999573,
      "num_tokens": 85021795.0,
      "step": 43460
    },
    {
      "entropy": 5.726591300964356,
      "epoch": 4.344779327235468,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003249101259007295,
      "loss": 5.1792,
      "mean_token_accuracy": 0.21874496787786485,
      "num_tokens": 85030619.0,
      "step": 43465
    },
    {
      "entropy": 5.786594104766846,
      "epoch": 4.345279152296696,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032487532529403755,
      "loss": 5.2744,
      "mean_token_accuracy": 0.21410914063453673,
      "num_tokens": 85039820.0,
      "step": 43470
    },
    {
      "entropy": 5.804339790344239,
      "epoch": 4.3457789773579245,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003248405234324953,
      "loss": 5.2707,
      "mean_token_accuracy": 0.2149697244167328,
      "num_tokens": 85051325.0,
      "step": 43475
    },
    {
      "entropy": 5.881511497497558,
      "epoch": 4.346278802419153,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032480572031697827,
      "loss": 5.2708,
      "mean_token_accuracy": 0.20892331451177598,
      "num_tokens": 85060402.0,
      "step": 43480
    },
    {
      "entropy": 5.737449741363525,
      "epoch": 4.346778627480382,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003247709159483622,
      "loss": 5.1232,
      "mean_token_accuracy": 0.23187872767448425,
      "num_tokens": 85070176.0,
      "step": 43485
    },
    {
      "entropy": 5.738439273834229,
      "epoch": 4.347278452541611,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003247361103275226,
      "loss": 5.2439,
      "mean_token_accuracy": 0.22024641633033754,
      "num_tokens": 85079283.0,
      "step": 43490
    },
    {
      "entropy": 5.736133193969726,
      "epoch": 4.347778277602839,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003247013034553353,
      "loss": 5.1681,
      "mean_token_accuracy": 0.2268220916390419,
      "num_tokens": 85088914.0,
      "step": 43495
    },
    {
      "entropy": 5.903853034973144,
      "epoch": 4.348278102664068,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000324666495332676,
      "loss": 5.3435,
      "mean_token_accuracy": 0.20355905294418336,
      "num_tokens": 85097615.0,
      "step": 43500
    },
    {
      "entropy": 5.789905548095703,
      "epoch": 4.348777927725296,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003246316859604205,
      "loss": 5.2764,
      "mean_token_accuracy": 0.2117917865514755,
      "num_tokens": 85107832.0,
      "step": 43505
    },
    {
      "entropy": 5.784666395187378,
      "epoch": 4.3492777527865245,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003245968753394445,
      "loss": 5.2224,
      "mean_token_accuracy": 0.21620065420866014,
      "num_tokens": 85116171.0,
      "step": 43510
    },
    {
      "entropy": 5.947032642364502,
      "epoch": 4.349777577847753,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003245620634706239,
      "loss": 5.4418,
      "mean_token_accuracy": 0.19430385380983353,
      "num_tokens": 85127072.0,
      "step": 43515
    },
    {
      "entropy": 5.8309361934661865,
      "epoch": 4.350277402908982,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00032452725035483454,
      "loss": 5.2091,
      "mean_token_accuracy": 0.22285878658294678,
      "num_tokens": 85136855.0,
      "step": 43520
    },
    {
      "entropy": 5.805070686340332,
      "epoch": 4.350777227970211,
      "grad_norm": 1.125,
      "learning_rate": 0.00032449243599295225,
      "loss": 5.2287,
      "mean_token_accuracy": 0.21929161101579667,
      "num_tokens": 85146012.0,
      "step": 43525
    },
    {
      "entropy": 5.824456357955933,
      "epoch": 4.351277053031439,
      "grad_norm": 1.125,
      "learning_rate": 0.0003244576203858531,
      "loss": 5.2429,
      "mean_token_accuracy": 0.21072438657283782,
      "num_tokens": 85155155.0,
      "step": 43530
    },
    {
      "entropy": 5.827112483978271,
      "epoch": 4.351776878092668,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003244228035344129,
      "loss": 5.2381,
      "mean_token_accuracy": 0.21838426738977432,
      "num_tokens": 85164072.0,
      "step": 43535
    },
    {
      "entropy": 5.793729162216186,
      "epoch": 4.352276703153896,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032438798543950765,
      "loss": 5.334,
      "mean_token_accuracy": 0.20915374606847764,
      "num_tokens": 85173848.0,
      "step": 43540
    },
    {
      "entropy": 5.91517276763916,
      "epoch": 4.3527765282151245,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003243531661020134,
      "loss": 5.3772,
      "mean_token_accuracy": 0.20674927085638045,
      "num_tokens": 85184422.0,
      "step": 43545
    },
    {
      "entropy": 5.762582540512085,
      "epoch": 4.353276353276353,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003243183455228062,
      "loss": 5.1115,
      "mean_token_accuracy": 0.2297733336687088,
      "num_tokens": 85195328.0,
      "step": 43550
    },
    {
      "entropy": 5.768802547454834,
      "epoch": 4.353776178337582,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032428352370276217,
      "loss": 5.233,
      "mean_token_accuracy": 0.2158199742436409,
      "num_tokens": 85205598.0,
      "step": 43555
    },
    {
      "entropy": 5.848062372207641,
      "epoch": 4.354276003398811,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003242487006427573,
      "loss": 5.2271,
      "mean_token_accuracy": 0.2241974726319313,
      "num_tokens": 85216639.0,
      "step": 43560
    },
    {
      "entropy": 5.788800144195557,
      "epoch": 4.354775828460039,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003242138763436678,
      "loss": 5.1875,
      "mean_token_accuracy": 0.2201929047703743,
      "num_tokens": 85225754.0,
      "step": 43565
    },
    {
      "entropy": 5.820479965209961,
      "epoch": 4.355275653521268,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032417905080636987,
      "loss": 5.2788,
      "mean_token_accuracy": 0.21093329340219497,
      "num_tokens": 85235270.0,
      "step": 43570
    },
    {
      "entropy": 5.823438453674316,
      "epoch": 4.355775478582496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003241442240317397,
      "loss": 5.2783,
      "mean_token_accuracy": 0.2111317679286003,
      "num_tokens": 85245451.0,
      "step": 43575
    },
    {
      "entropy": 5.831061983108521,
      "epoch": 4.3562753036437245,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003241093960206535,
      "loss": 5.2772,
      "mean_token_accuracy": 0.20964970588684081,
      "num_tokens": 85255231.0,
      "step": 43580
    },
    {
      "entropy": 5.829659938812256,
      "epoch": 4.356775128704953,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003240745667739874,
      "loss": 5.2793,
      "mean_token_accuracy": 0.21065742820501326,
      "num_tokens": 85265722.0,
      "step": 43585
    },
    {
      "entropy": 5.9297644138336185,
      "epoch": 4.357274953766182,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032403973629261795,
      "loss": 5.3637,
      "mean_token_accuracy": 0.20580774545669556,
      "num_tokens": 85275221.0,
      "step": 43590
    },
    {
      "entropy": 5.797859334945679,
      "epoch": 4.357774778827411,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032400490457742135,
      "loss": 5.2514,
      "mean_token_accuracy": 0.2226843774318695,
      "num_tokens": 85285449.0,
      "step": 43595
    },
    {
      "entropy": 5.841503429412842,
      "epoch": 4.358274603888639,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003239700716292739,
      "loss": 5.2823,
      "mean_token_accuracy": 0.21713411808013916,
      "num_tokens": 85295161.0,
      "step": 43600
    },
    {
      "entropy": 5.820310115814209,
      "epoch": 4.358774428949868,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032393523744905214,
      "loss": 5.2002,
      "mean_token_accuracy": 0.2126878708600998,
      "num_tokens": 85305480.0,
      "step": 43605
    },
    {
      "entropy": 5.801027774810791,
      "epoch": 4.359274254011096,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003239004020376323,
      "loss": 5.2886,
      "mean_token_accuracy": 0.21243299245834352,
      "num_tokens": 85316319.0,
      "step": 43610
    },
    {
      "entropy": 5.827766942977905,
      "epoch": 4.3597740790723245,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032386556539589097,
      "loss": 5.2698,
      "mean_token_accuracy": 0.21754447370767593,
      "num_tokens": 85326254.0,
      "step": 43615
    },
    {
      "entropy": 5.741312313079834,
      "epoch": 4.360273904133553,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003238307275247046,
      "loss": 5.1508,
      "mean_token_accuracy": 0.23180049806833267,
      "num_tokens": 85335028.0,
      "step": 43620
    },
    {
      "entropy": 5.811936616897583,
      "epoch": 4.360773729194782,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003237958884249497,
      "loss": 5.3479,
      "mean_token_accuracy": 0.2019583985209465,
      "num_tokens": 85344219.0,
      "step": 43625
    },
    {
      "entropy": 5.855065250396729,
      "epoch": 4.361273554256011,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032376104809750275,
      "loss": 5.3081,
      "mean_token_accuracy": 0.2124996453523636,
      "num_tokens": 85353472.0,
      "step": 43630
    },
    {
      "entropy": 5.768016338348389,
      "epoch": 4.361773379317239,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032372620654324034,
      "loss": 5.1971,
      "mean_token_accuracy": 0.2174692839384079,
      "num_tokens": 85363597.0,
      "step": 43635
    },
    {
      "entropy": 5.7719145774841305,
      "epoch": 4.362273204378468,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032369136376303925,
      "loss": 5.3152,
      "mean_token_accuracy": 0.20838092416524887,
      "num_tokens": 85373639.0,
      "step": 43640
    },
    {
      "entropy": 5.818607759475708,
      "epoch": 4.362773029439696,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032365651975777595,
      "loss": 5.2364,
      "mean_token_accuracy": 0.21418042182922364,
      "num_tokens": 85382928.0,
      "step": 43645
    },
    {
      "entropy": 5.893181991577149,
      "epoch": 4.3632728545009245,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003236216745283271,
      "loss": 5.3043,
      "mean_token_accuracy": 0.2094370186328888,
      "num_tokens": 85392542.0,
      "step": 43650
    },
    {
      "entropy": 5.784426832199097,
      "epoch": 4.363772679562153,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003235868280755694,
      "loss": 5.2109,
      "mean_token_accuracy": 0.2208257034420967,
      "num_tokens": 85402933.0,
      "step": 43655
    },
    {
      "entropy": 5.733833837509155,
      "epoch": 4.364272504623382,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003235519804003796,
      "loss": 5.1678,
      "mean_token_accuracy": 0.22678968757390977,
      "num_tokens": 85411742.0,
      "step": 43660
    },
    {
      "entropy": 5.764169645309448,
      "epoch": 4.364772329684611,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032351713150363453,
      "loss": 5.1572,
      "mean_token_accuracy": 0.2236732929944992,
      "num_tokens": 85421441.0,
      "step": 43665
    },
    {
      "entropy": 5.858334350585937,
      "epoch": 4.365272154745839,
      "grad_norm": 1.125,
      "learning_rate": 0.0003234822813862108,
      "loss": 5.3,
      "mean_token_accuracy": 0.2089500606060028,
      "num_tokens": 85432461.0,
      "step": 43670
    },
    {
      "entropy": 5.7981339454650875,
      "epoch": 4.365771979807068,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003234474300489856,
      "loss": 5.2526,
      "mean_token_accuracy": 0.2176418662071228,
      "num_tokens": 85442747.0,
      "step": 43675
    },
    {
      "entropy": 5.915659999847412,
      "epoch": 4.366271804868296,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032341257749283526,
      "loss": 5.3464,
      "mean_token_accuracy": 0.2058460682630539,
      "num_tokens": 85453323.0,
      "step": 43680
    },
    {
      "entropy": 5.8881267547607425,
      "epoch": 4.3667716299295245,
      "grad_norm": 1.015625,
      "learning_rate": 0.000323377723718637,
      "loss": 5.3734,
      "mean_token_accuracy": 0.19824084490537644,
      "num_tokens": 85464712.0,
      "step": 43685
    },
    {
      "entropy": 5.782730913162231,
      "epoch": 4.367271454990753,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003233428687272677,
      "loss": 5.1942,
      "mean_token_accuracy": 0.22343155294656752,
      "num_tokens": 85473631.0,
      "step": 43690
    },
    {
      "entropy": 5.765289783477783,
      "epoch": 4.367771280051982,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032330801251960424,
      "loss": 5.2227,
      "mean_token_accuracy": 0.22286861687898635,
      "num_tokens": 85484117.0,
      "step": 43695
    },
    {
      "entropy": 5.846445322036743,
      "epoch": 4.368271105113211,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032327315509652363,
      "loss": 5.2976,
      "mean_token_accuracy": 0.21014651656150818,
      "num_tokens": 85494045.0,
      "step": 43700
    },
    {
      "entropy": 5.8428253650665285,
      "epoch": 4.368770930174439,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032323829645890283,
      "loss": 5.2438,
      "mean_token_accuracy": 0.21037975698709488,
      "num_tokens": 85503408.0,
      "step": 43705
    },
    {
      "entropy": 5.793232870101929,
      "epoch": 4.369270755235667,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032320343660761897,
      "loss": 5.2392,
      "mean_token_accuracy": 0.2155853644013405,
      "num_tokens": 85513325.0,
      "step": 43710
    },
    {
      "entropy": 5.859908628463745,
      "epoch": 4.369770580296896,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000323168575543549,
      "loss": 5.3348,
      "mean_token_accuracy": 0.21020727902650832,
      "num_tokens": 85522842.0,
      "step": 43715
    },
    {
      "entropy": 5.870703554153442,
      "epoch": 4.3702704053581245,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003231337132675701,
      "loss": 5.2942,
      "mean_token_accuracy": 0.20615326315164567,
      "num_tokens": 85532366.0,
      "step": 43720
    },
    {
      "entropy": 5.846668148040772,
      "epoch": 4.370770230419353,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032309884978055925,
      "loss": 5.2534,
      "mean_token_accuracy": 0.21198282688856124,
      "num_tokens": 85541841.0,
      "step": 43725
    },
    {
      "entropy": 5.913395404815674,
      "epoch": 4.371270055480582,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003230639850833939,
      "loss": 5.3136,
      "mean_token_accuracy": 0.20438129007816314,
      "num_tokens": 85551856.0,
      "step": 43730
    },
    {
      "entropy": 5.743506717681885,
      "epoch": 4.371769880541811,
      "grad_norm": 1.25,
      "learning_rate": 0.00032302911917695093,
      "loss": 5.2073,
      "mean_token_accuracy": 0.21867403388023376,
      "num_tokens": 85562453.0,
      "step": 43735
    },
    {
      "entropy": 5.824691915512085,
      "epoch": 4.372269705603039,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032299425206210763,
      "loss": 5.1975,
      "mean_token_accuracy": 0.21580372154712676,
      "num_tokens": 85572095.0,
      "step": 43740
    },
    {
      "entropy": 5.949099540710449,
      "epoch": 4.372769530664267,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032295938373974137,
      "loss": 5.3607,
      "mean_token_accuracy": 0.1997220143675804,
      "num_tokens": 85581585.0,
      "step": 43745
    },
    {
      "entropy": 5.749648094177246,
      "epoch": 4.373269355725496,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003229245142107294,
      "loss": 5.1876,
      "mean_token_accuracy": 0.2284137099981308,
      "num_tokens": 85591492.0,
      "step": 43750
    },
    {
      "entropy": 5.842697668075561,
      "epoch": 4.3737691807867245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003228896434759489,
      "loss": 5.2371,
      "mean_token_accuracy": 0.21751405894756318,
      "num_tokens": 85601927.0,
      "step": 43755
    },
    {
      "entropy": 5.795989990234375,
      "epoch": 4.374269005847953,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003228547715362773,
      "loss": 5.2725,
      "mean_token_accuracy": 0.21240163296461106,
      "num_tokens": 85611869.0,
      "step": 43760
    },
    {
      "entropy": 5.85064377784729,
      "epoch": 4.374768830909182,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032281989839259205,
      "loss": 5.3177,
      "mean_token_accuracy": 0.2118435800075531,
      "num_tokens": 85621416.0,
      "step": 43765
    },
    {
      "entropy": 5.796152830123901,
      "epoch": 4.375268655970411,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003227850240457704,
      "loss": 5.2742,
      "mean_token_accuracy": 0.20998503118753434,
      "num_tokens": 85630048.0,
      "step": 43770
    },
    {
      "entropy": 5.784212255477906,
      "epoch": 4.375768481031639,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032275014849668984,
      "loss": 5.1785,
      "mean_token_accuracy": 0.21821454614400865,
      "num_tokens": 85640011.0,
      "step": 43775
    },
    {
      "entropy": 5.836494970321655,
      "epoch": 4.376268306092867,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003227152717462279,
      "loss": 5.332,
      "mean_token_accuracy": 0.2032942771911621,
      "num_tokens": 85649917.0,
      "step": 43780
    },
    {
      "entropy": 5.8216876029968265,
      "epoch": 4.376768131154096,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032268039379526186,
      "loss": 5.2803,
      "mean_token_accuracy": 0.21341315060853958,
      "num_tokens": 85659456.0,
      "step": 43785
    },
    {
      "entropy": 5.835726118087768,
      "epoch": 4.3772679562153245,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003226455146446695,
      "loss": 5.2037,
      "mean_token_accuracy": 0.22801631838083267,
      "num_tokens": 85669310.0,
      "step": 43790
    },
    {
      "entropy": 5.793563795089722,
      "epoch": 4.377767781276553,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032261063429532815,
      "loss": 5.1745,
      "mean_token_accuracy": 0.22812631875276565,
      "num_tokens": 85678347.0,
      "step": 43795
    },
    {
      "entropy": 5.736210870742798,
      "epoch": 4.378267606337782,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032257575274811554,
      "loss": 5.236,
      "mean_token_accuracy": 0.2221808075904846,
      "num_tokens": 85687253.0,
      "step": 43800
    },
    {
      "entropy": 5.860412502288819,
      "epoch": 4.378767431399011,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003225408700039093,
      "loss": 5.3436,
      "mean_token_accuracy": 0.21014269143342973,
      "num_tokens": 85698190.0,
      "step": 43805
    },
    {
      "entropy": 5.829698085784912,
      "epoch": 4.379267256460239,
      "grad_norm": 1.25,
      "learning_rate": 0.0003225059860635869,
      "loss": 5.3249,
      "mean_token_accuracy": 0.2122911348938942,
      "num_tokens": 85708214.0,
      "step": 43810
    },
    {
      "entropy": 5.832491397857666,
      "epoch": 4.379767081521467,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032247110092802625,
      "loss": 5.3467,
      "mean_token_accuracy": 0.2094010666012764,
      "num_tokens": 85718033.0,
      "step": 43815
    },
    {
      "entropy": 5.836159896850586,
      "epoch": 4.380266906582696,
      "grad_norm": 1.125,
      "learning_rate": 0.0003224362145981048,
      "loss": 5.2897,
      "mean_token_accuracy": 0.21123270243406295,
      "num_tokens": 85728399.0,
      "step": 43820
    },
    {
      "entropy": 5.796245098114014,
      "epoch": 4.3807667316439245,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032240132707470043,
      "loss": 5.1413,
      "mean_token_accuracy": 0.22916973680257796,
      "num_tokens": 85737205.0,
      "step": 43825
    },
    {
      "entropy": 5.847028827667236,
      "epoch": 4.381266556705153,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032236643835869085,
      "loss": 5.2465,
      "mean_token_accuracy": 0.2141183391213417,
      "num_tokens": 85746790.0,
      "step": 43830
    },
    {
      "entropy": 5.766628217697144,
      "epoch": 4.381766381766382,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032233154845095384,
      "loss": 5.1822,
      "mean_token_accuracy": 0.21196554154157637,
      "num_tokens": 85755757.0,
      "step": 43835
    },
    {
      "entropy": 5.7756353378295895,
      "epoch": 4.382266206827611,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003222966573523673,
      "loss": 5.2603,
      "mean_token_accuracy": 0.21163840889930724,
      "num_tokens": 85765751.0,
      "step": 43840
    },
    {
      "entropy": 5.793716049194336,
      "epoch": 4.382766031888839,
      "grad_norm": 1.25,
      "learning_rate": 0.000322261765063809,
      "loss": 5.2753,
      "mean_token_accuracy": 0.22014611810445786,
      "num_tokens": 85774561.0,
      "step": 43845
    },
    {
      "entropy": 5.782850980758667,
      "epoch": 4.383265856950067,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003222268715861569,
      "loss": 5.2116,
      "mean_token_accuracy": 0.21234560906887054,
      "num_tokens": 85784105.0,
      "step": 43850
    },
    {
      "entropy": 5.778852891921997,
      "epoch": 4.383765682011296,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003221919769202889,
      "loss": 5.1827,
      "mean_token_accuracy": 0.2281241610646248,
      "num_tokens": 85793399.0,
      "step": 43855
    },
    {
      "entropy": 5.828179740905762,
      "epoch": 4.3842655070725245,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032215708106708284,
      "loss": 5.2352,
      "mean_token_accuracy": 0.20957648307085036,
      "num_tokens": 85802497.0,
      "step": 43860
    },
    {
      "entropy": 5.766715049743652,
      "epoch": 4.384765332133753,
      "grad_norm": 1.125,
      "learning_rate": 0.0003221221840274168,
      "loss": 5.1693,
      "mean_token_accuracy": 0.22124226540327072,
      "num_tokens": 85811630.0,
      "step": 43865
    },
    {
      "entropy": 5.780120182037353,
      "epoch": 4.385265157194982,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032208728580216864,
      "loss": 5.2689,
      "mean_token_accuracy": 0.21420808434486388,
      "num_tokens": 85820859.0,
      "step": 43870
    },
    {
      "entropy": 5.688962602615357,
      "epoch": 4.385764982256211,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032205238639221653,
      "loss": 5.0461,
      "mean_token_accuracy": 0.23112708032131196,
      "num_tokens": 85829941.0,
      "step": 43875
    },
    {
      "entropy": 5.754291582107544,
      "epoch": 4.386264807317439,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003220174857984385,
      "loss": 5.1485,
      "mean_token_accuracy": 0.2196160599589348,
      "num_tokens": 85839441.0,
      "step": 43880
    },
    {
      "entropy": 5.752842140197754,
      "epoch": 4.386764632378667,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032198258402171267,
      "loss": 5.2443,
      "mean_token_accuracy": 0.21807799637317657,
      "num_tokens": 85849315.0,
      "step": 43885
    },
    {
      "entropy": 5.75939450263977,
      "epoch": 4.387264457439896,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032194768106291705,
      "loss": 5.2239,
      "mean_token_accuracy": 0.22061410397291184,
      "num_tokens": 85859299.0,
      "step": 43890
    },
    {
      "entropy": 5.84769639968872,
      "epoch": 4.3877642825011245,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032191277692292984,
      "loss": 5.2908,
      "mean_token_accuracy": 0.21289280652999878,
      "num_tokens": 85869104.0,
      "step": 43895
    },
    {
      "entropy": 5.888578462600708,
      "epoch": 4.388264107562353,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032187787160262924,
      "loss": 5.3187,
      "mean_token_accuracy": 0.21012745052576065,
      "num_tokens": 85879556.0,
      "step": 43900
    },
    {
      "entropy": 5.812680053710937,
      "epoch": 4.388763932623582,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003218429651028935,
      "loss": 5.2315,
      "mean_token_accuracy": 0.2151745229959488,
      "num_tokens": 85888815.0,
      "step": 43905
    },
    {
      "entropy": 5.771393346786499,
      "epoch": 4.389263757684811,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032180805742460086,
      "loss": 5.1992,
      "mean_token_accuracy": 0.21547148525714874,
      "num_tokens": 85897238.0,
      "step": 43910
    },
    {
      "entropy": 5.78593864440918,
      "epoch": 4.389763582746039,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032177314856862943,
      "loss": 5.2683,
      "mean_token_accuracy": 0.21417679637670517,
      "num_tokens": 85907317.0,
      "step": 43915
    },
    {
      "entropy": 5.739805841445923,
      "epoch": 4.390263407807267,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003217382385358577,
      "loss": 5.1454,
      "mean_token_accuracy": 0.22865383476018905,
      "num_tokens": 85916857.0,
      "step": 43920
    },
    {
      "entropy": 5.926377201080323,
      "epoch": 4.390763232868496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003217033273271639,
      "loss": 5.3817,
      "mean_token_accuracy": 0.20495086759328843,
      "num_tokens": 85927128.0,
      "step": 43925
    },
    {
      "entropy": 5.815637016296387,
      "epoch": 4.3912630579297245,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003216684149434263,
      "loss": 5.1846,
      "mean_token_accuracy": 0.22158605456352234,
      "num_tokens": 85937488.0,
      "step": 43930
    },
    {
      "entropy": 5.84282169342041,
      "epoch": 4.391762882990953,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003216335013855235,
      "loss": 5.2777,
      "mean_token_accuracy": 0.2130055770277977,
      "num_tokens": 85946863.0,
      "step": 43935
    },
    {
      "entropy": 5.812851619720459,
      "epoch": 4.392262708052182,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032159858665433376,
      "loss": 5.2758,
      "mean_token_accuracy": 0.21275351494550704,
      "num_tokens": 85956631.0,
      "step": 43940
    },
    {
      "entropy": 5.763924026489258,
      "epoch": 4.392762533113411,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032156367075073563,
      "loss": 5.1946,
      "mean_token_accuracy": 0.21834511458873748,
      "num_tokens": 85966310.0,
      "step": 43945
    },
    {
      "entropy": 5.839864063262939,
      "epoch": 4.393262358174639,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003215287536756074,
      "loss": 5.3317,
      "mean_token_accuracy": 0.21384965628385544,
      "num_tokens": 85976767.0,
      "step": 43950
    },
    {
      "entropy": 5.791072273254395,
      "epoch": 4.393762183235867,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032149383542982786,
      "loss": 5.2026,
      "mean_token_accuracy": 0.22789547741413116,
      "num_tokens": 85987193.0,
      "step": 43955
    },
    {
      "entropy": 5.779657363891602,
      "epoch": 4.394262008297096,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003214589160142753,
      "loss": 5.2009,
      "mean_token_accuracy": 0.2244940459728241,
      "num_tokens": 85996603.0,
      "step": 43960
    },
    {
      "entropy": 5.815276384353638,
      "epoch": 4.3947618333583245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032142399542982834,
      "loss": 5.2718,
      "mean_token_accuracy": 0.21546627581119537,
      "num_tokens": 86007036.0,
      "step": 43965
    },
    {
      "entropy": 5.77119779586792,
      "epoch": 4.395261658419553,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032138907367736564,
      "loss": 5.2226,
      "mean_token_accuracy": 0.218208846449852,
      "num_tokens": 86016288.0,
      "step": 43970
    },
    {
      "entropy": 5.902249145507812,
      "epoch": 4.395761483480782,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003213541507577658,
      "loss": 5.3638,
      "mean_token_accuracy": 0.20069007575511932,
      "num_tokens": 86026962.0,
      "step": 43975
    },
    {
      "entropy": 5.910299348831177,
      "epoch": 4.396261308542011,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003213192266719074,
      "loss": 5.2777,
      "mean_token_accuracy": 0.2117451936006546,
      "num_tokens": 86036885.0,
      "step": 43980
    },
    {
      "entropy": 5.927409744262695,
      "epoch": 4.396761133603239,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032128430142066926,
      "loss": 5.3629,
      "mean_token_accuracy": 0.1987860307097435,
      "num_tokens": 86046117.0,
      "step": 43985
    },
    {
      "entropy": 5.799274349212647,
      "epoch": 4.397260958664467,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032124937500492995,
      "loss": 5.1999,
      "mean_token_accuracy": 0.2166141763329506,
      "num_tokens": 86056188.0,
      "step": 43990
    },
    {
      "entropy": 5.769575452804565,
      "epoch": 4.397760783725696,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003212144474255682,
      "loss": 5.31,
      "mean_token_accuracy": 0.20538664162158965,
      "num_tokens": 86066904.0,
      "step": 43995
    },
    {
      "entropy": 5.803809595108032,
      "epoch": 4.3982606087869245,
      "grad_norm": 1.125,
      "learning_rate": 0.000321179518683463,
      "loss": 5.2408,
      "mean_token_accuracy": 0.21481959521770477,
      "num_tokens": 86076223.0,
      "step": 44000
    },
    {
      "entropy": 5.899402332305908,
      "epoch": 4.398760433848153,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032114458877949287,
      "loss": 5.3106,
      "mean_token_accuracy": 0.2123078763484955,
      "num_tokens": 86087094.0,
      "step": 44005
    },
    {
      "entropy": 5.806591129302978,
      "epoch": 4.399260258909382,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003211096577145368,
      "loss": 5.2716,
      "mean_token_accuracy": 0.21325514912605287,
      "num_tokens": 86097405.0,
      "step": 44010
    },
    {
      "entropy": 5.796134042739868,
      "epoch": 4.399760083970611,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003210747254894736,
      "loss": 5.2597,
      "mean_token_accuracy": 0.22055429220199585,
      "num_tokens": 86106239.0,
      "step": 44015
    },
    {
      "entropy": 5.772617387771606,
      "epoch": 4.400259909031838,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003210397921051822,
      "loss": 5.2278,
      "mean_token_accuracy": 0.21670147329568862,
      "num_tokens": 86116468.0,
      "step": 44020
    },
    {
      "entropy": 5.8222557544708256,
      "epoch": 4.400759734093067,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003210048575625414,
      "loss": 5.1981,
      "mean_token_accuracy": 0.2224741369485855,
      "num_tokens": 86125636.0,
      "step": 44025
    },
    {
      "entropy": 5.856077718734741,
      "epoch": 4.401259559154296,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032096992186243025,
      "loss": 5.2751,
      "mean_token_accuracy": 0.21234101057052612,
      "num_tokens": 86135263.0,
      "step": 44030
    },
    {
      "entropy": 5.7911975383758545,
      "epoch": 4.4017593842155245,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003209349850057277,
      "loss": 5.2849,
      "mean_token_accuracy": 0.20767782330513002,
      "num_tokens": 86146081.0,
      "step": 44035
    },
    {
      "entropy": 5.879255580902099,
      "epoch": 4.402259209276753,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032090004699331277,
      "loss": 5.3271,
      "mean_token_accuracy": 0.20706759989261628,
      "num_tokens": 86156200.0,
      "step": 44040
    },
    {
      "entropy": 5.7926328659057615,
      "epoch": 4.402759034337982,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032086510782606447,
      "loss": 5.1775,
      "mean_token_accuracy": 0.22339701652526855,
      "num_tokens": 86166744.0,
      "step": 44045
    },
    {
      "entropy": 5.815834617614746,
      "epoch": 4.403258859399211,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003208301675048619,
      "loss": 5.3164,
      "mean_token_accuracy": 0.2141580328345299,
      "num_tokens": 86175577.0,
      "step": 44050
    },
    {
      "entropy": 5.810608148574829,
      "epoch": 4.403758684460438,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032079522603058403,
      "loss": 5.2653,
      "mean_token_accuracy": 0.21325102895498277,
      "num_tokens": 86186169.0,
      "step": 44055
    },
    {
      "entropy": 5.876827955245972,
      "epoch": 4.404258509521667,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003207602834041101,
      "loss": 5.3703,
      "mean_token_accuracy": 0.21326131522655487,
      "num_tokens": 86195972.0,
      "step": 44060
    },
    {
      "entropy": 5.8989830493927,
      "epoch": 4.404758334582896,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032072533962631915,
      "loss": 5.3171,
      "mean_token_accuracy": 0.2080575406551361,
      "num_tokens": 86205815.0,
      "step": 44065
    },
    {
      "entropy": 5.8516199588775635,
      "epoch": 4.4052581596441245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032069039469809045,
      "loss": 5.2876,
      "mean_token_accuracy": 0.21052275598049164,
      "num_tokens": 86215304.0,
      "step": 44070
    },
    {
      "entropy": 5.810554027557373,
      "epoch": 4.405757984705353,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003206554486203032,
      "loss": 5.1925,
      "mean_token_accuracy": 0.21691055297851564,
      "num_tokens": 86224702.0,
      "step": 44075
    },
    {
      "entropy": 5.762700080871582,
      "epoch": 4.406257809766582,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003206205013938366,
      "loss": 5.2393,
      "mean_token_accuracy": 0.22284943759441375,
      "num_tokens": 86235445.0,
      "step": 44080
    },
    {
      "entropy": 5.858736944198609,
      "epoch": 4.406757634827811,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032058555301957,
      "loss": 5.2724,
      "mean_token_accuracy": 0.2048563301563263,
      "num_tokens": 86244803.0,
      "step": 44085
    },
    {
      "entropy": 5.812299680709839,
      "epoch": 4.407257459889038,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003205506034983825,
      "loss": 5.3038,
      "mean_token_accuracy": 0.2090870976448059,
      "num_tokens": 86255065.0,
      "step": 44090
    },
    {
      "entropy": 5.71130223274231,
      "epoch": 4.407757284950267,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032051565283115356,
      "loss": 5.1313,
      "mean_token_accuracy": 0.2276747331023216,
      "num_tokens": 86265175.0,
      "step": 44095
    },
    {
      "entropy": 5.896908092498779,
      "epoch": 4.408257110011496,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003204807010187625,
      "loss": 5.3106,
      "mean_token_accuracy": 0.207182440161705,
      "num_tokens": 86275275.0,
      "step": 44100
    },
    {
      "entropy": 5.833761215209961,
      "epoch": 4.4087569350727245,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032044574806208877,
      "loss": 5.2779,
      "mean_token_accuracy": 0.21686905324459077,
      "num_tokens": 86285239.0,
      "step": 44105
    },
    {
      "entropy": 5.836661243438721,
      "epoch": 4.409256760133953,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032041079396201167,
      "loss": 5.262,
      "mean_token_accuracy": 0.20979850739240646,
      "num_tokens": 86295736.0,
      "step": 44110
    },
    {
      "entropy": 5.786052465438843,
      "epoch": 4.409756585195182,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032037583871941073,
      "loss": 5.1785,
      "mean_token_accuracy": 0.22114024609327315,
      "num_tokens": 86305488.0,
      "step": 44115
    },
    {
      "entropy": 5.936246490478515,
      "epoch": 4.410256410256411,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003203408823351653,
      "loss": 5.4837,
      "mean_token_accuracy": 0.19430067986249924,
      "num_tokens": 86315864.0,
      "step": 44120
    },
    {
      "entropy": 5.85382719039917,
      "epoch": 4.410756235317638,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032030592481015495,
      "loss": 5.2927,
      "mean_token_accuracy": 0.20782275944948198,
      "num_tokens": 86327092.0,
      "step": 44125
    },
    {
      "entropy": 5.855951642990112,
      "epoch": 4.411256060378867,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032027096614525917,
      "loss": 5.3239,
      "mean_token_accuracy": 0.2118848592042923,
      "num_tokens": 86337476.0,
      "step": 44130
    },
    {
      "entropy": 5.782603597640991,
      "epoch": 4.411755885440096,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032023600634135756,
      "loss": 5.2401,
      "mean_token_accuracy": 0.22527172714471816,
      "num_tokens": 86348380.0,
      "step": 44135
    },
    {
      "entropy": 5.896282911300659,
      "epoch": 4.4122557105013245,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003202010453993297,
      "loss": 5.3502,
      "mean_token_accuracy": 0.20775092244148255,
      "num_tokens": 86357830.0,
      "step": 44140
    },
    {
      "entropy": 5.819809007644653,
      "epoch": 4.412755535562553,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032016608332005516,
      "loss": 5.2395,
      "mean_token_accuracy": 0.21703240871429444,
      "num_tokens": 86366605.0,
      "step": 44145
    },
    {
      "entropy": 5.731888389587402,
      "epoch": 4.413255360623782,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003201311201044136,
      "loss": 5.1787,
      "mean_token_accuracy": 0.21683312803506852,
      "num_tokens": 86375317.0,
      "step": 44150
    },
    {
      "entropy": 5.846306991577149,
      "epoch": 4.413755185685011,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032009615575328453,
      "loss": 5.288,
      "mean_token_accuracy": 0.21321294605731964,
      "num_tokens": 86385270.0,
      "step": 44155
    },
    {
      "entropy": 5.922647190093994,
      "epoch": 4.414255010746238,
      "grad_norm": 1.234375,
      "learning_rate": 0.000320061190267548,
      "loss": 5.3646,
      "mean_token_accuracy": 0.20356098115444182,
      "num_tokens": 86394926.0,
      "step": 44160
    },
    {
      "entropy": 5.826720333099365,
      "epoch": 4.414754835807467,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032002622364808334,
      "loss": 5.2396,
      "mean_token_accuracy": 0.21636108309030533,
      "num_tokens": 86404184.0,
      "step": 44165
    },
    {
      "entropy": 5.731060028076172,
      "epoch": 4.415254660868696,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031999125589577055,
      "loss": 5.1265,
      "mean_token_accuracy": 0.22667875438928603,
      "num_tokens": 86414056.0,
      "step": 44170
    },
    {
      "entropy": 5.854810428619385,
      "epoch": 4.4157544859299245,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003199562870114894,
      "loss": 5.3072,
      "mean_token_accuracy": 0.20887595415115356,
      "num_tokens": 86423429.0,
      "step": 44175
    },
    {
      "entropy": 5.732721853256225,
      "epoch": 4.416254310991153,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031992131699611954,
      "loss": 5.2559,
      "mean_token_accuracy": 0.21676035225391388,
      "num_tokens": 86433867.0,
      "step": 44180
    },
    {
      "entropy": 5.792721605300903,
      "epoch": 4.416754136052382,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003198863458505409,
      "loss": 5.2877,
      "mean_token_accuracy": 0.2139852151274681,
      "num_tokens": 86443688.0,
      "step": 44185
    },
    {
      "entropy": 5.875492811203003,
      "epoch": 4.417253961113611,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003198513735756334,
      "loss": 5.2199,
      "mean_token_accuracy": 0.2220730572938919,
      "num_tokens": 86453337.0,
      "step": 44190
    },
    {
      "entropy": 5.868581390380859,
      "epoch": 4.417753786174838,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031981640017227694,
      "loss": 5.3731,
      "mean_token_accuracy": 0.20737848430871964,
      "num_tokens": 86461990.0,
      "step": 44195
    },
    {
      "entropy": 5.755158853530884,
      "epoch": 4.418253611236067,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031978142564135126,
      "loss": 5.1809,
      "mean_token_accuracy": 0.22317709773778915,
      "num_tokens": 86471767.0,
      "step": 44200
    },
    {
      "entropy": 5.746924495697021,
      "epoch": 4.418753436297296,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031974644998373645,
      "loss": 5.2499,
      "mean_token_accuracy": 0.21437172293663026,
      "num_tokens": 86481510.0,
      "step": 44205
    },
    {
      "entropy": 5.805189752578736,
      "epoch": 4.4192532613585245,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031971147320031257,
      "loss": 5.2379,
      "mean_token_accuracy": 0.21496354788541794,
      "num_tokens": 86491323.0,
      "step": 44210
    },
    {
      "entropy": 5.8794340133667,
      "epoch": 4.419753086419753,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031967649529195943,
      "loss": 5.2227,
      "mean_token_accuracy": 0.21540471613407136,
      "num_tokens": 86500632.0,
      "step": 44215
    },
    {
      "entropy": 5.855917835235596,
      "epoch": 4.420252911480982,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003196415162595572,
      "loss": 5.2788,
      "mean_token_accuracy": 0.2134341686964035,
      "num_tokens": 86509809.0,
      "step": 44220
    },
    {
      "entropy": 5.776938152313233,
      "epoch": 4.420752736542211,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003196065361039859,
      "loss": 5.2924,
      "mean_token_accuracy": 0.21261335164308548,
      "num_tokens": 86519055.0,
      "step": 44225
    },
    {
      "entropy": 5.703144359588623,
      "epoch": 4.421252561603438,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031957155482612563,
      "loss": 5.1937,
      "mean_token_accuracy": 0.22099410593509675,
      "num_tokens": 86528618.0,
      "step": 44230
    },
    {
      "entropy": 5.77115273475647,
      "epoch": 4.421752386664667,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031953657242685655,
      "loss": 5.1899,
      "mean_token_accuracy": 0.2249274358153343,
      "num_tokens": 86537821.0,
      "step": 44235
    },
    {
      "entropy": 5.763142538070679,
      "epoch": 4.422252211725896,
      "grad_norm": 1.125,
      "learning_rate": 0.0003195015889070588,
      "loss": 5.2262,
      "mean_token_accuracy": 0.21800911724567412,
      "num_tokens": 86548608.0,
      "step": 44240
    },
    {
      "entropy": 5.780586910247803,
      "epoch": 4.4227520367871245,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031946660426761237,
      "loss": 5.1891,
      "mean_token_accuracy": 0.21807741224765778,
      "num_tokens": 86557401.0,
      "step": 44245
    },
    {
      "entropy": 5.799337577819824,
      "epoch": 4.423251861848353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031943161850939774,
      "loss": 5.35,
      "mean_token_accuracy": 0.20118409395217896,
      "num_tokens": 86568132.0,
      "step": 44250
    },
    {
      "entropy": 5.753026962280273,
      "epoch": 4.423751686909582,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031939663163329504,
      "loss": 5.1909,
      "mean_token_accuracy": 0.2277993842959404,
      "num_tokens": 86578364.0,
      "step": 44255
    },
    {
      "entropy": 5.808361387252807,
      "epoch": 4.424251511970811,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003193616436401844,
      "loss": 5.2826,
      "mean_token_accuracy": 0.21157752424478532,
      "num_tokens": 86588081.0,
      "step": 44260
    },
    {
      "entropy": 5.858316659927368,
      "epoch": 4.424751337032038,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003193266545309464,
      "loss": 5.3602,
      "mean_token_accuracy": 0.2031092882156372,
      "num_tokens": 86597616.0,
      "step": 44265
    },
    {
      "entropy": 5.791487455368042,
      "epoch": 4.425251162093267,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031929166430646105,
      "loss": 5.2385,
      "mean_token_accuracy": 0.21816300302743913,
      "num_tokens": 86607769.0,
      "step": 44270
    },
    {
      "entropy": 5.825230455398559,
      "epoch": 4.425750987154496,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003192566729676088,
      "loss": 5.232,
      "mean_token_accuracy": 0.20925703793764114,
      "num_tokens": 86617595.0,
      "step": 44275
    },
    {
      "entropy": 5.805787467956543,
      "epoch": 4.4262508122157245,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031922168051527,
      "loss": 5.1881,
      "mean_token_accuracy": 0.2180739685893059,
      "num_tokens": 86627569.0,
      "step": 44280
    },
    {
      "entropy": 5.742336273193359,
      "epoch": 4.426750637276953,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003191866869503252,
      "loss": 5.1774,
      "mean_token_accuracy": 0.2160977303981781,
      "num_tokens": 86636833.0,
      "step": 44285
    },
    {
      "entropy": 5.7360944747924805,
      "epoch": 4.427250462338182,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003191516922736547,
      "loss": 5.2399,
      "mean_token_accuracy": 0.22183759659528732,
      "num_tokens": 86648037.0,
      "step": 44290
    },
    {
      "entropy": 5.919267320632935,
      "epoch": 4.427750287399411,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000319116696486139,
      "loss": 5.3412,
      "mean_token_accuracy": 0.19793926030397416,
      "num_tokens": 86656582.0,
      "step": 44295
    },
    {
      "entropy": 5.746321105957032,
      "epoch": 4.428250112460638,
      "grad_norm": 1.125,
      "learning_rate": 0.0003190816995886586,
      "loss": 5.1999,
      "mean_token_accuracy": 0.21954394727945328,
      "num_tokens": 86666392.0,
      "step": 44300
    },
    {
      "entropy": 5.800011682510376,
      "epoch": 4.428749937521867,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031904670158209404,
      "loss": 5.2687,
      "mean_token_accuracy": 0.21289449483156203,
      "num_tokens": 86676845.0,
      "step": 44305
    },
    {
      "entropy": 5.861137723922729,
      "epoch": 4.429249762583096,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031901170246732564,
      "loss": 5.3188,
      "mean_token_accuracy": 0.2060419961810112,
      "num_tokens": 86685788.0,
      "step": 44310
    },
    {
      "entropy": 5.78772382736206,
      "epoch": 4.4297495876443245,
      "grad_norm": 1.25,
      "learning_rate": 0.00031897670224523413,
      "loss": 5.2145,
      "mean_token_accuracy": 0.22424056082963945,
      "num_tokens": 86696181.0,
      "step": 44315
    },
    {
      "entropy": 5.8546360492706295,
      "epoch": 4.430249412705553,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031894170091670016,
      "loss": 5.2331,
      "mean_token_accuracy": 0.21807568669319152,
      "num_tokens": 86705915.0,
      "step": 44320
    },
    {
      "entropy": 5.8460651397705075,
      "epoch": 4.430749237766782,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031890669848260437,
      "loss": 5.3389,
      "mean_token_accuracy": 0.20879140347242356,
      "num_tokens": 86715102.0,
      "step": 44325
    },
    {
      "entropy": 5.771598434448242,
      "epoch": 4.431249062828011,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003188716949438273,
      "loss": 5.2169,
      "mean_token_accuracy": 0.2173983335494995,
      "num_tokens": 86724254.0,
      "step": 44330
    },
    {
      "entropy": 5.8144104957580565,
      "epoch": 4.431748887889238,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031883669030124967,
      "loss": 5.2563,
      "mean_token_accuracy": 0.211177459359169,
      "num_tokens": 86734297.0,
      "step": 44335
    },
    {
      "entropy": 5.853423547744751,
      "epoch": 4.432248712950467,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031880168455575216,
      "loss": 5.2924,
      "mean_token_accuracy": 0.21333152800798416,
      "num_tokens": 86744723.0,
      "step": 44340
    },
    {
      "entropy": 5.907296752929687,
      "epoch": 4.432748538011696,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031876667770821556,
      "loss": 5.2776,
      "mean_token_accuracy": 0.21276756525039672,
      "num_tokens": 86753749.0,
      "step": 44345
    },
    {
      "entropy": 5.741182518005371,
      "epoch": 4.4332483630729245,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003187316697595206,
      "loss": 5.2175,
      "mean_token_accuracy": 0.21644788086414338,
      "num_tokens": 86763731.0,
      "step": 44350
    },
    {
      "entropy": 5.8208959102630615,
      "epoch": 4.433748188134153,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003186966607105481,
      "loss": 5.3826,
      "mean_token_accuracy": 0.20288134962320328,
      "num_tokens": 86773574.0,
      "step": 44355
    },
    {
      "entropy": 5.722453975677491,
      "epoch": 4.434248013195382,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031866165056217887,
      "loss": 5.0366,
      "mean_token_accuracy": 0.23074718862771987,
      "num_tokens": 86784122.0,
      "step": 44360
    },
    {
      "entropy": 5.82885627746582,
      "epoch": 4.434747838256611,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003186266393152937,
      "loss": 5.2043,
      "mean_token_accuracy": 0.21920156180858613,
      "num_tokens": 86794301.0,
      "step": 44365
    },
    {
      "entropy": 5.767764282226563,
      "epoch": 4.435247663317838,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003185916269707736,
      "loss": 5.2273,
      "mean_token_accuracy": 0.22066255211830138,
      "num_tokens": 86803150.0,
      "step": 44370
    },
    {
      "entropy": 5.921256351470947,
      "epoch": 4.435747488379067,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031855661352949934,
      "loss": 5.3739,
      "mean_token_accuracy": 0.19959839284420014,
      "num_tokens": 86812984.0,
      "step": 44375
    },
    {
      "entropy": 5.7484804630279545,
      "epoch": 4.436247313440296,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003185215989923519,
      "loss": 5.1308,
      "mean_token_accuracy": 0.23350217193365097,
      "num_tokens": 86824544.0,
      "step": 44380
    },
    {
      "entropy": 5.803786659240723,
      "epoch": 4.4367471385015245,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003184865833602121,
      "loss": 5.2132,
      "mean_token_accuracy": 0.21837296485900878,
      "num_tokens": 86833773.0,
      "step": 44385
    },
    {
      "entropy": 5.774668741226196,
      "epoch": 4.437246963562753,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031845156663396125,
      "loss": 5.2365,
      "mean_token_accuracy": 0.2187248483300209,
      "num_tokens": 86843008.0,
      "step": 44390
    },
    {
      "entropy": 5.82999963760376,
      "epoch": 4.437746788623982,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031841654881448005,
      "loss": 5.2763,
      "mean_token_accuracy": 0.21277386099100112,
      "num_tokens": 86852837.0,
      "step": 44395
    },
    {
      "entropy": 5.851885890960693,
      "epoch": 4.438246613685211,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031838152990264966,
      "loss": 5.3483,
      "mean_token_accuracy": 0.20997357070446016,
      "num_tokens": 86863278.0,
      "step": 44400
    },
    {
      "entropy": 5.794516849517822,
      "epoch": 4.438746438746438,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003183465098993512,
      "loss": 5.273,
      "mean_token_accuracy": 0.2116060435771942,
      "num_tokens": 86873410.0,
      "step": 44405
    },
    {
      "entropy": 5.944666814804077,
      "epoch": 4.439246263807667,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031831148880546566,
      "loss": 5.4019,
      "mean_token_accuracy": 0.20172208398580552,
      "num_tokens": 86883395.0,
      "step": 44410
    },
    {
      "entropy": 5.879328489303589,
      "epoch": 4.439746088868896,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003182764666218742,
      "loss": 5.2809,
      "mean_token_accuracy": 0.21373596638441086,
      "num_tokens": 86892788.0,
      "step": 44415
    },
    {
      "entropy": 5.848212623596192,
      "epoch": 4.4402459139301245,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000318241443349458,
      "loss": 5.2523,
      "mean_token_accuracy": 0.21369733810424804,
      "num_tokens": 86902003.0,
      "step": 44420
    },
    {
      "entropy": 5.740780448913574,
      "epoch": 4.440745738991353,
      "grad_norm": 1.125,
      "learning_rate": 0.0003182064189890982,
      "loss": 5.2754,
      "mean_token_accuracy": 0.21347632855176926,
      "num_tokens": 86912036.0,
      "step": 44425
    },
    {
      "entropy": 5.861412239074707,
      "epoch": 4.441245564052582,
      "grad_norm": 1.28125,
      "learning_rate": 0.000318171393541676,
      "loss": 5.2784,
      "mean_token_accuracy": 0.21563603729009628,
      "num_tokens": 86921277.0,
      "step": 44430
    },
    {
      "entropy": 5.815994739532471,
      "epoch": 4.441745389113811,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031813636700807265,
      "loss": 5.2384,
      "mean_token_accuracy": 0.21312022358179092,
      "num_tokens": 86931127.0,
      "step": 44435
    },
    {
      "entropy": 5.784157419204712,
      "epoch": 4.442245214175038,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003181013393891695,
      "loss": 5.1773,
      "mean_token_accuracy": 0.22638918757438659,
      "num_tokens": 86940815.0,
      "step": 44440
    },
    {
      "entropy": 5.729798936843872,
      "epoch": 4.442745039236267,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031806631068584764,
      "loss": 5.2017,
      "mean_token_accuracy": 0.21937546133995056,
      "num_tokens": 86950267.0,
      "step": 44445
    },
    {
      "entropy": 5.812269067764282,
      "epoch": 4.443244864297496,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031803128089898845,
      "loss": 5.286,
      "mean_token_accuracy": 0.21178678423166275,
      "num_tokens": 86960540.0,
      "step": 44450
    },
    {
      "entropy": 5.859299039840698,
      "epoch": 4.4437446893587245,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031799625002947333,
      "loss": 5.2835,
      "mean_token_accuracy": 0.20969225466251373,
      "num_tokens": 86969981.0,
      "step": 44455
    },
    {
      "entropy": 5.806242418289185,
      "epoch": 4.444244514419953,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031796121807818367,
      "loss": 5.2844,
      "mean_token_accuracy": 0.20993402749300002,
      "num_tokens": 86979799.0,
      "step": 44460
    },
    {
      "entropy": 5.799554300308228,
      "epoch": 4.444744339481182,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003179261850460008,
      "loss": 5.2156,
      "mean_token_accuracy": 0.22052816152572632,
      "num_tokens": 86988930.0,
      "step": 44465
    },
    {
      "entropy": 5.809148216247559,
      "epoch": 4.445244164542411,
      "grad_norm": 1.125,
      "learning_rate": 0.00031789115093380614,
      "loss": 5.3182,
      "mean_token_accuracy": 0.21058106869459153,
      "num_tokens": 86998613.0,
      "step": 44470
    },
    {
      "entropy": 5.863146734237671,
      "epoch": 4.445743989603638,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031785611574248123,
      "loss": 5.288,
      "mean_token_accuracy": 0.21352393478155135,
      "num_tokens": 87007488.0,
      "step": 44475
    },
    {
      "entropy": 5.86735200881958,
      "epoch": 4.446243814664867,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031782107947290745,
      "loss": 5.2835,
      "mean_token_accuracy": 0.2098905399441719,
      "num_tokens": 87016831.0,
      "step": 44480
    },
    {
      "entropy": 5.741169071197509,
      "epoch": 4.446743639726096,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003177860421259663,
      "loss": 5.1445,
      "mean_token_accuracy": 0.22222742140293122,
      "num_tokens": 87025989.0,
      "step": 44485
    },
    {
      "entropy": 5.7891610622406,
      "epoch": 4.4472434647873245,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003177510037025394,
      "loss": 5.1567,
      "mean_token_accuracy": 0.22406167089939116,
      "num_tokens": 87036287.0,
      "step": 44490
    },
    {
      "entropy": 5.828827953338623,
      "epoch": 4.447743289848553,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031771596420350815,
      "loss": 5.307,
      "mean_token_accuracy": 0.2082522839307785,
      "num_tokens": 87045546.0,
      "step": 44495
    },
    {
      "entropy": 5.745552587509155,
      "epoch": 4.448243114909782,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003176809236297542,
      "loss": 5.1887,
      "mean_token_accuracy": 0.21843431144952774,
      "num_tokens": 87056717.0,
      "step": 44500
    },
    {
      "entropy": 5.877780723571777,
      "epoch": 4.448742939971011,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003176458819821593,
      "loss": 5.3269,
      "mean_token_accuracy": 0.20250075161457062,
      "num_tokens": 87066742.0,
      "step": 44505
    },
    {
      "entropy": 5.759248828887939,
      "epoch": 4.449242765032238,
      "grad_norm": 1.140625,
      "learning_rate": 0.000317610839261605,
      "loss": 5.1444,
      "mean_token_accuracy": 0.22317058593034744,
      "num_tokens": 87076230.0,
      "step": 44510
    },
    {
      "entropy": 5.830898714065552,
      "epoch": 4.449742590093467,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031757579546897287,
      "loss": 5.2769,
      "mean_token_accuracy": 0.21184243112802506,
      "num_tokens": 87085649.0,
      "step": 44515
    },
    {
      "entropy": 5.8383469581604,
      "epoch": 4.450242415154696,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031754075060514477,
      "loss": 5.2806,
      "mean_token_accuracy": 0.21963519901037215,
      "num_tokens": 87094871.0,
      "step": 44520
    },
    {
      "entropy": 5.854557991027832,
      "epoch": 4.4507422402159245,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031750570467100226,
      "loss": 5.2172,
      "mean_token_accuracy": 0.2220625475049019,
      "num_tokens": 87104647.0,
      "step": 44525
    },
    {
      "entropy": 5.776945352554321,
      "epoch": 4.451242065277153,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003174706576674271,
      "loss": 5.1733,
      "mean_token_accuracy": 0.22317405939102172,
      "num_tokens": 87116179.0,
      "step": 44530
    },
    {
      "entropy": 5.65566987991333,
      "epoch": 4.451741890338382,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003174356095953012,
      "loss": 5.0812,
      "mean_token_accuracy": 0.23287303745746613,
      "num_tokens": 87125876.0,
      "step": 44535
    },
    {
      "entropy": 5.787786293029785,
      "epoch": 4.4522417153996106,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003174005604555063,
      "loss": 5.2772,
      "mean_token_accuracy": 0.21058003902435302,
      "num_tokens": 87136228.0,
      "step": 44540
    },
    {
      "entropy": 5.836726713180542,
      "epoch": 4.452741540460838,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031736551024892423,
      "loss": 5.3052,
      "mean_token_accuracy": 0.2109356179833412,
      "num_tokens": 87147612.0,
      "step": 44545
    },
    {
      "entropy": 5.925019979476929,
      "epoch": 4.453241365522067,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003173304589764368,
      "loss": 5.3199,
      "mean_token_accuracy": 0.20478460639715196,
      "num_tokens": 87159049.0,
      "step": 44550
    },
    {
      "entropy": 5.837993478775024,
      "epoch": 4.453741190583296,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031729540663892586,
      "loss": 5.2198,
      "mean_token_accuracy": 0.21720300018787383,
      "num_tokens": 87168304.0,
      "step": 44555
    },
    {
      "entropy": 5.786194324493408,
      "epoch": 4.4542410156445245,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031726035323727334,
      "loss": 5.2769,
      "mean_token_accuracy": 0.22001359909772872,
      "num_tokens": 87177048.0,
      "step": 44560
    },
    {
      "entropy": 5.841835069656372,
      "epoch": 4.454740840705753,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031722529877236126,
      "loss": 5.284,
      "mean_token_accuracy": 0.20666441917419434,
      "num_tokens": 87188033.0,
      "step": 44565
    },
    {
      "entropy": 5.789891624450684,
      "epoch": 4.455240665766982,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003171902432450716,
      "loss": 5.2161,
      "mean_token_accuracy": 0.2266052857041359,
      "num_tokens": 87198996.0,
      "step": 44570
    },
    {
      "entropy": 5.759930419921875,
      "epoch": 4.4557404908282106,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031715518665628625,
      "loss": 5.1926,
      "mean_token_accuracy": 0.22599462568759918,
      "num_tokens": 87208304.0,
      "step": 44575
    },
    {
      "entropy": 5.7808033466339115,
      "epoch": 4.456240315889438,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031712012900688726,
      "loss": 5.2186,
      "mean_token_accuracy": 0.22102925032377244,
      "num_tokens": 87217271.0,
      "step": 44580
    },
    {
      "entropy": 5.783459234237671,
      "epoch": 4.456740140950667,
      "grad_norm": 1.25,
      "learning_rate": 0.0003170850702977566,
      "loss": 5.2104,
      "mean_token_accuracy": 0.22140742242336273,
      "num_tokens": 87225567.0,
      "step": 44585
    },
    {
      "entropy": 5.785086393356323,
      "epoch": 4.457239966011896,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031705001052977647,
      "loss": 5.1622,
      "mean_token_accuracy": 0.22744390815496446,
      "num_tokens": 87235415.0,
      "step": 44590
    },
    {
      "entropy": 5.853301048278809,
      "epoch": 4.4577397910731245,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003170149497038288,
      "loss": 5.3282,
      "mean_token_accuracy": 0.21513697803020476,
      "num_tokens": 87246026.0,
      "step": 44595
    },
    {
      "entropy": 5.859765481948853,
      "epoch": 4.458239616134353,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031697988782079583,
      "loss": 5.2908,
      "mean_token_accuracy": 0.2107007011771202,
      "num_tokens": 87255483.0,
      "step": 44600
    },
    {
      "entropy": 5.858791923522949,
      "epoch": 4.458739441195582,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031694482488155975,
      "loss": 5.2607,
      "mean_token_accuracy": 0.21961895674467086,
      "num_tokens": 87265678.0,
      "step": 44605
    },
    {
      "entropy": 5.83416976928711,
      "epoch": 4.4592392662568106,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003169097608870026,
      "loss": 5.2562,
      "mean_token_accuracy": 0.21127282083034515,
      "num_tokens": 87275847.0,
      "step": 44610
    },
    {
      "entropy": 5.7592260360717775,
      "epoch": 4.459739091318038,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003168746958380067,
      "loss": 5.2108,
      "mean_token_accuracy": 0.217323400080204,
      "num_tokens": 87285426.0,
      "step": 44615
    },
    {
      "entropy": 5.860729503631592,
      "epoch": 4.460238916379267,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031683962973545423,
      "loss": 5.2468,
      "mean_token_accuracy": 0.21732814610004425,
      "num_tokens": 87294976.0,
      "step": 44620
    },
    {
      "entropy": 5.893292045593261,
      "epoch": 4.460738741440496,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003168045625802273,
      "loss": 5.3474,
      "mean_token_accuracy": 0.20129259675741196,
      "num_tokens": 87304449.0,
      "step": 44625
    },
    {
      "entropy": 5.825392770767212,
      "epoch": 4.4612385665017245,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003167694943732085,
      "loss": 5.2433,
      "mean_token_accuracy": 0.21294234842061996,
      "num_tokens": 87314276.0,
      "step": 44630
    },
    {
      "entropy": 5.824283266067505,
      "epoch": 4.461738391562953,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031673442511527994,
      "loss": 5.2809,
      "mean_token_accuracy": 0.2126771628856659,
      "num_tokens": 87323487.0,
      "step": 44635
    },
    {
      "entropy": 5.864872026443481,
      "epoch": 4.462238216624182,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031669935480732397,
      "loss": 5.2916,
      "mean_token_accuracy": 0.21095860600471497,
      "num_tokens": 87333275.0,
      "step": 44640
    },
    {
      "entropy": 5.797276020050049,
      "epoch": 4.4627380416854106,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031666428345022296,
      "loss": 5.2471,
      "mean_token_accuracy": 0.21991163343191147,
      "num_tokens": 87342202.0,
      "step": 44645
    },
    {
      "entropy": 5.775160217285157,
      "epoch": 4.463237866746638,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003166292110448593,
      "loss": 5.1934,
      "mean_token_accuracy": 0.2185912936925888,
      "num_tokens": 87352014.0,
      "step": 44650
    },
    {
      "entropy": 5.823655128479004,
      "epoch": 4.463737691807867,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003165941375921154,
      "loss": 5.2768,
      "mean_token_accuracy": 0.21020329296588897,
      "num_tokens": 87363204.0,
      "step": 44655
    },
    {
      "entropy": 5.837252426147461,
      "epoch": 4.464237516869096,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003165590630928737,
      "loss": 5.2966,
      "mean_token_accuracy": 0.21063332110643387,
      "num_tokens": 87373023.0,
      "step": 44660
    },
    {
      "entropy": 5.931497716903687,
      "epoch": 4.4647373419303245,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031652398754801665,
      "loss": 5.3659,
      "mean_token_accuracy": 0.20132860243320466,
      "num_tokens": 87383547.0,
      "step": 44665
    },
    {
      "entropy": 5.817470455169678,
      "epoch": 4.465237166991553,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003164889109584268,
      "loss": 5.1578,
      "mean_token_accuracy": 0.2290479749441147,
      "num_tokens": 87392906.0,
      "step": 44670
    },
    {
      "entropy": 5.839428901672363,
      "epoch": 4.465736992052782,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003164538333249866,
      "loss": 5.2598,
      "mean_token_accuracy": 0.20842488706111909,
      "num_tokens": 87401825.0,
      "step": 44675
    },
    {
      "entropy": 5.835538101196289,
      "epoch": 4.4662368171140105,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003164187546485786,
      "loss": 5.3177,
      "mean_token_accuracy": 0.20725997239351274,
      "num_tokens": 87411161.0,
      "step": 44680
    },
    {
      "entropy": 5.841520023345947,
      "epoch": 4.466736642175238,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031638367493008545,
      "loss": 5.3112,
      "mean_token_accuracy": 0.2155770018696785,
      "num_tokens": 87420543.0,
      "step": 44685
    },
    {
      "entropy": 5.843414163589477,
      "epoch": 4.467236467236467,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031634859417038963,
      "loss": 5.2393,
      "mean_token_accuracy": 0.21877165287733077,
      "num_tokens": 87431573.0,
      "step": 44690
    },
    {
      "entropy": 5.9356217861175535,
      "epoch": 4.467736292297696,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031631351237037386,
      "loss": 5.4083,
      "mean_token_accuracy": 0.20545955747365952,
      "num_tokens": 87442157.0,
      "step": 44695
    },
    {
      "entropy": 5.842235803604126,
      "epoch": 4.4682361173589245,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031627842953092073,
      "loss": 5.2612,
      "mean_token_accuracy": 0.2155841276049614,
      "num_tokens": 87451477.0,
      "step": 44700
    },
    {
      "entropy": 5.721321964263916,
      "epoch": 4.468735942420153,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031624334565291294,
      "loss": 5.1086,
      "mean_token_accuracy": 0.22614314705133437,
      "num_tokens": 87460784.0,
      "step": 44705
    },
    {
      "entropy": 5.8856360912323,
      "epoch": 4.469235767481382,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003162082607372332,
      "loss": 5.2783,
      "mean_token_accuracy": 0.21102318912744522,
      "num_tokens": 87470045.0,
      "step": 44710
    },
    {
      "entropy": 5.84807014465332,
      "epoch": 4.4697355925426105,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003161731747847642,
      "loss": 5.2208,
      "mean_token_accuracy": 0.22263634353876113,
      "num_tokens": 87479322.0,
      "step": 44715
    },
    {
      "entropy": 5.787328433990479,
      "epoch": 4.470235417603838,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031613808779638874,
      "loss": 5.1665,
      "mean_token_accuracy": 0.22179065197706221,
      "num_tokens": 87488674.0,
      "step": 44720
    },
    {
      "entropy": 5.767345857620239,
      "epoch": 4.470735242665067,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003161029997729895,
      "loss": 5.2762,
      "mean_token_accuracy": 0.21969435811042787,
      "num_tokens": 87498172.0,
      "step": 44725
    },
    {
      "entropy": 5.8837165355682375,
      "epoch": 4.471235067726296,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031606791071544946,
      "loss": 5.2972,
      "mean_token_accuracy": 0.21021280884742738,
      "num_tokens": 87508175.0,
      "step": 44730
    },
    {
      "entropy": 5.907639265060425,
      "epoch": 4.4717348927875245,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003160328206246514,
      "loss": 5.3181,
      "mean_token_accuracy": 0.20743184685707092,
      "num_tokens": 87518282.0,
      "step": 44735
    },
    {
      "entropy": 5.767623472213745,
      "epoch": 4.472234717848753,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000315997729501478,
      "loss": 5.1772,
      "mean_token_accuracy": 0.22801485359668733,
      "num_tokens": 87527824.0,
      "step": 44740
    },
    {
      "entropy": 5.825658130645752,
      "epoch": 4.472734542909982,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003159626373468123,
      "loss": 5.3004,
      "mean_token_accuracy": 0.211063215136528,
      "num_tokens": 87537035.0,
      "step": 44745
    },
    {
      "entropy": 5.867188119888306,
      "epoch": 4.4732343679712105,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031592754416153725,
      "loss": 5.3297,
      "mean_token_accuracy": 0.20812448859214783,
      "num_tokens": 87545931.0,
      "step": 44750
    },
    {
      "entropy": 5.820956611633301,
      "epoch": 4.473734193032438,
      "grad_norm": 1.125,
      "learning_rate": 0.00031589244994653567,
      "loss": 5.2195,
      "mean_token_accuracy": 0.21958927661180497,
      "num_tokens": 87555175.0,
      "step": 44755
    },
    {
      "entropy": 5.826110410690307,
      "epoch": 4.474234018093667,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031585735470269057,
      "loss": 5.2455,
      "mean_token_accuracy": 0.21663086861371994,
      "num_tokens": 87565650.0,
      "step": 44760
    },
    {
      "entropy": 5.7483762264251705,
      "epoch": 4.474733843154896,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000315822258430885,
      "loss": 5.2065,
      "mean_token_accuracy": 0.2175835520029068,
      "num_tokens": 87574617.0,
      "step": 44765
    },
    {
      "entropy": 5.7500159740448,
      "epoch": 4.4752336682161244,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003157871611320018,
      "loss": 5.2218,
      "mean_token_accuracy": 0.21681938618421553,
      "num_tokens": 87583735.0,
      "step": 44770
    },
    {
      "entropy": 5.844129276275635,
      "epoch": 4.475733493277353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003157520628069242,
      "loss": 5.2949,
      "mean_token_accuracy": 0.2125509798526764,
      "num_tokens": 87594415.0,
      "step": 44775
    },
    {
      "entropy": 5.862101125717163,
      "epoch": 4.476233318338582,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031571696345653516,
      "loss": 5.2489,
      "mean_token_accuracy": 0.22130066603422166,
      "num_tokens": 87603380.0,
      "step": 44780
    },
    {
      "entropy": 5.675027227401733,
      "epoch": 4.47673314339981,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031568186308171775,
      "loss": 5.0652,
      "mean_token_accuracy": 0.23924848735332488,
      "num_tokens": 87613697.0,
      "step": 44785
    },
    {
      "entropy": 5.818321609497071,
      "epoch": 4.477232968461038,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031564676168335527,
      "loss": 5.2564,
      "mean_token_accuracy": 0.21521953195333482,
      "num_tokens": 87623160.0,
      "step": 44790
    },
    {
      "entropy": 5.840770435333252,
      "epoch": 4.477732793522267,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031561165926233064,
      "loss": 5.2554,
      "mean_token_accuracy": 0.21830411106348038,
      "num_tokens": 87632649.0,
      "step": 44795
    },
    {
      "entropy": 5.814417457580566,
      "epoch": 4.478232618583496,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031557655581952715,
      "loss": 5.3109,
      "mean_token_accuracy": 0.2140636458992958,
      "num_tokens": 87643442.0,
      "step": 44800
    },
    {
      "entropy": 5.981095647811889,
      "epoch": 4.4787324436447244,
      "grad_norm": 1.25,
      "learning_rate": 0.00031554145135582784,
      "loss": 5.411,
      "mean_token_accuracy": 0.1998259410262108,
      "num_tokens": 87653939.0,
      "step": 44805
    },
    {
      "entropy": 5.909384822845459,
      "epoch": 4.479232268705953,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031550634587211617,
      "loss": 5.27,
      "mean_token_accuracy": 0.21028752774000167,
      "num_tokens": 87664082.0,
      "step": 44810
    },
    {
      "entropy": 5.862366485595703,
      "epoch": 4.479732093767182,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031547123936927516,
      "loss": 5.2354,
      "mean_token_accuracy": 0.2126054733991623,
      "num_tokens": 87672726.0,
      "step": 44815
    },
    {
      "entropy": 5.744838666915894,
      "epoch": 4.48023191882841,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031543613184818834,
      "loss": 5.2115,
      "mean_token_accuracy": 0.21879784017801285,
      "num_tokens": 87684124.0,
      "step": 44820
    },
    {
      "entropy": 5.888264417648315,
      "epoch": 4.480731743889638,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003154010233097387,
      "loss": 5.3153,
      "mean_token_accuracy": 0.21157014816999437,
      "num_tokens": 87693852.0,
      "step": 44825
    },
    {
      "entropy": 5.885940504074097,
      "epoch": 4.481231568950867,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031536591375480984,
      "loss": 5.2512,
      "mean_token_accuracy": 0.22020809650421141,
      "num_tokens": 87702825.0,
      "step": 44830
    },
    {
      "entropy": 5.821665334701538,
      "epoch": 4.481731394012096,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003153308031842849,
      "loss": 5.1958,
      "mean_token_accuracy": 0.22347555458545684,
      "num_tokens": 87712420.0,
      "step": 44835
    },
    {
      "entropy": 5.780101585388183,
      "epoch": 4.4822312190733244,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031529569159904723,
      "loss": 5.2828,
      "mean_token_accuracy": 0.21425458043813705,
      "num_tokens": 87722675.0,
      "step": 44840
    },
    {
      "entropy": 5.767142343521118,
      "epoch": 4.482731044134553,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031526057899998043,
      "loss": 5.1592,
      "mean_token_accuracy": 0.2186513915657997,
      "num_tokens": 87732453.0,
      "step": 44845
    },
    {
      "entropy": 5.895485734939575,
      "epoch": 4.483230869195782,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003152254653879679,
      "loss": 5.3043,
      "mean_token_accuracy": 0.21131318509578706,
      "num_tokens": 87741772.0,
      "step": 44850
    },
    {
      "entropy": 5.90304388999939,
      "epoch": 4.48373069425701,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031519035076389284,
      "loss": 5.3966,
      "mean_token_accuracy": 0.20909662097692489,
      "num_tokens": 87750657.0,
      "step": 44855
    },
    {
      "entropy": 5.849650192260742,
      "epoch": 4.484230519318238,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000315155235128639,
      "loss": 5.228,
      "mean_token_accuracy": 0.2191192090511322,
      "num_tokens": 87760342.0,
      "step": 44860
    },
    {
      "entropy": 5.880076456069946,
      "epoch": 4.484730344379467,
      "grad_norm": 1.25,
      "learning_rate": 0.0003151201184830899,
      "loss": 5.3126,
      "mean_token_accuracy": 0.20714633613824845,
      "num_tokens": 87769950.0,
      "step": 44865
    },
    {
      "entropy": 5.8494609832763675,
      "epoch": 4.485230169440696,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031508500082812874,
      "loss": 5.2862,
      "mean_token_accuracy": 0.2189214289188385,
      "num_tokens": 87779210.0,
      "step": 44870
    },
    {
      "entropy": 5.8883696556091305,
      "epoch": 4.4857299945019244,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031504988216463927,
      "loss": 5.2574,
      "mean_token_accuracy": 0.21675199419260024,
      "num_tokens": 87789096.0,
      "step": 44875
    },
    {
      "entropy": 5.763055515289307,
      "epoch": 4.486229819563153,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003150147624935052,
      "loss": 5.1914,
      "mean_token_accuracy": 0.22153624296188354,
      "num_tokens": 87798367.0,
      "step": 44880
    },
    {
      "entropy": 5.807170391082764,
      "epoch": 4.486729644624382,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003149796418156099,
      "loss": 5.2825,
      "mean_token_accuracy": 0.2169631689786911,
      "num_tokens": 87809028.0,
      "step": 44885
    },
    {
      "entropy": 5.811135005950928,
      "epoch": 4.48722946968561,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003149445201318372,
      "loss": 5.2207,
      "mean_token_accuracy": 0.21532949656248093,
      "num_tokens": 87819474.0,
      "step": 44890
    },
    {
      "entropy": 5.858116912841797,
      "epoch": 4.487729294746838,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031490939744307063,
      "loss": 5.2703,
      "mean_token_accuracy": 0.20796445906162261,
      "num_tokens": 87828912.0,
      "step": 44895
    },
    {
      "entropy": 5.781684923171997,
      "epoch": 4.488229119808067,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003148742737501938,
      "loss": 5.184,
      "mean_token_accuracy": 0.2301480382680893,
      "num_tokens": 87839377.0,
      "step": 44900
    },
    {
      "entropy": 5.752246570587158,
      "epoch": 4.488728944869296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031483914905409055,
      "loss": 5.1319,
      "mean_token_accuracy": 0.2303673043847084,
      "num_tokens": 87849292.0,
      "step": 44905
    },
    {
      "entropy": 5.602516841888428,
      "epoch": 4.489228769930524,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003148040233556446,
      "loss": 5.0644,
      "mean_token_accuracy": 0.24387278258800507,
      "num_tokens": 87860148.0,
      "step": 44910
    },
    {
      "entropy": 5.791743326187134,
      "epoch": 4.489728594991753,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003147688966557396,
      "loss": 5.2806,
      "mean_token_accuracy": 0.2171940788626671,
      "num_tokens": 87869527.0,
      "step": 44915
    },
    {
      "entropy": 5.8618738651275635,
      "epoch": 4.490228420052982,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003147337689552594,
      "loss": 5.2792,
      "mean_token_accuracy": 0.21468922346830369,
      "num_tokens": 87879060.0,
      "step": 44920
    },
    {
      "entropy": 5.854047346115112,
      "epoch": 4.49072824511421,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003146986402550878,
      "loss": 5.2322,
      "mean_token_accuracy": 0.2144627243280411,
      "num_tokens": 87887546.0,
      "step": 44925
    },
    {
      "entropy": 5.8625650882720945,
      "epoch": 4.491228070175438,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031466351055610875,
      "loss": 5.3094,
      "mean_token_accuracy": 0.21681994795799256,
      "num_tokens": 87898385.0,
      "step": 44930
    },
    {
      "entropy": 5.859922599792481,
      "epoch": 4.491727895236667,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031462837985920584,
      "loss": 5.3053,
      "mean_token_accuracy": 0.20973211228847505,
      "num_tokens": 87907705.0,
      "step": 44935
    },
    {
      "entropy": 5.7709949016571045,
      "epoch": 4.492227720297896,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031459324816526313,
      "loss": 5.2065,
      "mean_token_accuracy": 0.22140643000602722,
      "num_tokens": 87917159.0,
      "step": 44940
    },
    {
      "entropy": 5.77619309425354,
      "epoch": 4.492727545359124,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031455811547516447,
      "loss": 5.1857,
      "mean_token_accuracy": 0.22811839431524278,
      "num_tokens": 87927210.0,
      "step": 44945
    },
    {
      "entropy": 5.94352650642395,
      "epoch": 4.493227370420353,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031452298178979385,
      "loss": 5.4378,
      "mean_token_accuracy": 0.19341589510440826,
      "num_tokens": 87937045.0,
      "step": 44950
    },
    {
      "entropy": 5.872874689102173,
      "epoch": 4.493727195481582,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003144878471100351,
      "loss": 5.2823,
      "mean_token_accuracy": 0.2148751810193062,
      "num_tokens": 87947051.0,
      "step": 44955
    },
    {
      "entropy": 5.694797801971435,
      "epoch": 4.49422702054281,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031445271143677235,
      "loss": 5.0478,
      "mean_token_accuracy": 0.23554387390613557,
      "num_tokens": 87956847.0,
      "step": 44960
    },
    {
      "entropy": 5.75282917022705,
      "epoch": 4.494726845604038,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031441757477088957,
      "loss": 5.2458,
      "mean_token_accuracy": 0.21538978219032287,
      "num_tokens": 87966975.0,
      "step": 44965
    },
    {
      "entropy": 5.799750471115113,
      "epoch": 4.495226670665267,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031438243711327064,
      "loss": 5.2462,
      "mean_token_accuracy": 0.2169017270207405,
      "num_tokens": 87977960.0,
      "step": 44970
    },
    {
      "entropy": 5.855585336685181,
      "epoch": 4.495726495726496,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003143472984647998,
      "loss": 5.3262,
      "mean_token_accuracy": 0.21490954607725143,
      "num_tokens": 87988579.0,
      "step": 44975
    },
    {
      "entropy": 5.7718860626220705,
      "epoch": 4.496226320787724,
      "grad_norm": 1.203125,
      "learning_rate": 0.000314312158826361,
      "loss": 5.1861,
      "mean_token_accuracy": 0.2214513584971428,
      "num_tokens": 87997043.0,
      "step": 44980
    },
    {
      "entropy": 5.782233953475952,
      "epoch": 4.496726145848953,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031427701819883845,
      "loss": 5.175,
      "mean_token_accuracy": 0.21994477659463882,
      "num_tokens": 88007140.0,
      "step": 44985
    },
    {
      "entropy": 5.805068635940552,
      "epoch": 4.497225970910182,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003142418765831162,
      "loss": 5.2583,
      "mean_token_accuracy": 0.2169656604528427,
      "num_tokens": 88016273.0,
      "step": 44990
    },
    {
      "entropy": 5.774663209915161,
      "epoch": 4.49772579597141,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031420673398007846,
      "loss": 5.1651,
      "mean_token_accuracy": 0.22439159899950029,
      "num_tokens": 88025455.0,
      "step": 44995
    },
    {
      "entropy": 5.810020160675049,
      "epoch": 4.498225621032638,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031417159039060936,
      "loss": 5.3609,
      "mean_token_accuracy": 0.21191780865192414,
      "num_tokens": 88034107.0,
      "step": 45000
    },
    {
      "epoch": 4.498225621032638,
      "eval_entropy": 5.614085921857874,
      "eval_loss": 5.4199910163879395,
      "eval_mean_token_accuracy": 0.2158659838813846,
      "eval_num_tokens": 88034107.0,
      "eval_runtime": 30.2053,
      "eval_samples_per_second": 1027.801,
      "eval_steps_per_second": 128.488,
      "step": 45000
    },
    {
      "entropy": 5.739853286743164,
      "epoch": 4.498725446093867,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003141364458155931,
      "loss": 5.0622,
      "mean_token_accuracy": 0.2339621439576149,
      "num_tokens": 88044155.0,
      "step": 45005
    },
    {
      "entropy": 5.764558553695679,
      "epoch": 4.499225271155096,
      "grad_norm": 1.203125,
      "learning_rate": 0.000314101300255914,
      "loss": 5.1166,
      "mean_token_accuracy": 0.22678057253360748,
      "num_tokens": 88053670.0,
      "step": 45010
    },
    {
      "entropy": 5.90524845123291,
      "epoch": 4.499725096216324,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003140661537124562,
      "loss": 5.3384,
      "mean_token_accuracy": 0.20380779653787612,
      "num_tokens": 88062800.0,
      "step": 45015
    },
    {
      "entropy": 5.837892484664917,
      "epoch": 4.500224921277553,
      "grad_norm": 1.078125,
      "learning_rate": 0.000314031006186104,
      "loss": 5.2328,
      "mean_token_accuracy": 0.21917279213666915,
      "num_tokens": 88072631.0,
      "step": 45020
    },
    {
      "entropy": 5.800565958023071,
      "epoch": 4.500724746338781,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003139958576777418,
      "loss": 5.2538,
      "mean_token_accuracy": 0.20958276838064194,
      "num_tokens": 88081439.0,
      "step": 45025
    },
    {
      "entropy": 5.813790035247803,
      "epoch": 4.50122457140001,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003139607081882538,
      "loss": 5.3206,
      "mean_token_accuracy": 0.2031514510512352,
      "num_tokens": 88090093.0,
      "step": 45030
    },
    {
      "entropy": 5.82219762802124,
      "epoch": 4.501724396461238,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031392555771852447,
      "loss": 5.2622,
      "mean_token_accuracy": 0.22019321322441102,
      "num_tokens": 88100001.0,
      "step": 45035
    },
    {
      "entropy": 5.855831956863403,
      "epoch": 4.502224221522467,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003138904062694381,
      "loss": 5.2805,
      "mean_token_accuracy": 0.21662641167640687,
      "num_tokens": 88109410.0,
      "step": 45040
    },
    {
      "entropy": 5.817545652389526,
      "epoch": 4.502724046583696,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003138552538418791,
      "loss": 5.2595,
      "mean_token_accuracy": 0.21700579673051834,
      "num_tokens": 88119718.0,
      "step": 45045
    },
    {
      "entropy": 5.749893617630005,
      "epoch": 4.503223871644924,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031382010043673193,
      "loss": 5.225,
      "mean_token_accuracy": 0.21593547910451888,
      "num_tokens": 88128663.0,
      "step": 45050
    },
    {
      "entropy": 5.864526224136353,
      "epoch": 4.503723696706153,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003137849460548809,
      "loss": 5.3067,
      "mean_token_accuracy": 0.20897428393363954,
      "num_tokens": 88137469.0,
      "step": 45055
    },
    {
      "entropy": 5.920152711868286,
      "epoch": 4.504223521767381,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031374979069721077,
      "loss": 5.3807,
      "mean_token_accuracy": 0.20814260393381118,
      "num_tokens": 88146475.0,
      "step": 45060
    },
    {
      "entropy": 5.829674482345581,
      "epoch": 4.50472334682861,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031371463436460586,
      "loss": 5.1972,
      "mean_token_accuracy": 0.21888423413038255,
      "num_tokens": 88156211.0,
      "step": 45065
    },
    {
      "entropy": 5.87853364944458,
      "epoch": 4.505223171889838,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031367947705795063,
      "loss": 5.329,
      "mean_token_accuracy": 0.2035890579223633,
      "num_tokens": 88165472.0,
      "step": 45070
    },
    {
      "entropy": 5.700887870788574,
      "epoch": 4.505722996951067,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003136443187781298,
      "loss": 5.1678,
      "mean_token_accuracy": 0.2315901681780815,
      "num_tokens": 88176531.0,
      "step": 45075
    },
    {
      "entropy": 5.785906362533569,
      "epoch": 4.506222822012296,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031360915952602776,
      "loss": 5.1993,
      "mean_token_accuracy": 0.21559841334819793,
      "num_tokens": 88185816.0,
      "step": 45080
    },
    {
      "entropy": 5.7914642810821535,
      "epoch": 4.506722647073524,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003135739993025292,
      "loss": 5.1698,
      "mean_token_accuracy": 0.23041631877422333,
      "num_tokens": 88195409.0,
      "step": 45085
    },
    {
      "entropy": 5.8486110210418705,
      "epoch": 4.507222472134753,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003135388381085188,
      "loss": 5.3129,
      "mean_token_accuracy": 0.2146872192621231,
      "num_tokens": 88205056.0,
      "step": 45090
    },
    {
      "entropy": 5.7791201114654545,
      "epoch": 4.507722297195981,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031350367594488115,
      "loss": 5.2499,
      "mean_token_accuracy": 0.2147106260061264,
      "num_tokens": 88214248.0,
      "step": 45095
    },
    {
      "entropy": 5.745128726959228,
      "epoch": 4.50822212225721,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031346851281250087,
      "loss": 5.2307,
      "mean_token_accuracy": 0.21538320034742356,
      "num_tokens": 88224073.0,
      "step": 45100
    },
    {
      "entropy": 5.744175100326538,
      "epoch": 4.508721947318438,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003134333487122628,
      "loss": 5.2248,
      "mean_token_accuracy": 0.22339775115251542,
      "num_tokens": 88235663.0,
      "step": 45105
    },
    {
      "entropy": 5.8659717559814455,
      "epoch": 4.509221772379667,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031339818364505153,
      "loss": 5.319,
      "mean_token_accuracy": 0.20719255059957503,
      "num_tokens": 88246238.0,
      "step": 45110
    },
    {
      "entropy": 5.904253530502319,
      "epoch": 4.509721597440896,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003133630176117518,
      "loss": 5.2471,
      "mean_token_accuracy": 0.21951813995838165,
      "num_tokens": 88255114.0,
      "step": 45115
    },
    {
      "entropy": 5.849960470199585,
      "epoch": 4.510221422502124,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003133278506132484,
      "loss": 5.2819,
      "mean_token_accuracy": 0.21000064313411712,
      "num_tokens": 88265704.0,
      "step": 45120
    },
    {
      "entropy": 5.937980985641479,
      "epoch": 4.510721247563353,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003132926826504262,
      "loss": 5.3772,
      "mean_token_accuracy": 0.21182582527399063,
      "num_tokens": 88276080.0,
      "step": 45125
    },
    {
      "entropy": 5.841256904602051,
      "epoch": 4.511221072624581,
      "grad_norm": 1.125,
      "learning_rate": 0.00031325751372416987,
      "loss": 5.2985,
      "mean_token_accuracy": 0.2133651092648506,
      "num_tokens": 88286019.0,
      "step": 45130
    },
    {
      "entropy": 5.789185333251953,
      "epoch": 4.51172089768581,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003132223438353644,
      "loss": 5.2747,
      "mean_token_accuracy": 0.21734680235385895,
      "num_tokens": 88295768.0,
      "step": 45135
    },
    {
      "entropy": 5.700953674316406,
      "epoch": 4.512220722747038,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003131871729848947,
      "loss": 5.0788,
      "mean_token_accuracy": 0.2296571746468544,
      "num_tokens": 88305377.0,
      "step": 45140
    },
    {
      "entropy": 5.872838020324707,
      "epoch": 4.512720547808267,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031315200117364546,
      "loss": 5.3002,
      "mean_token_accuracy": 0.20740179866552352,
      "num_tokens": 88314927.0,
      "step": 45145
    },
    {
      "entropy": 5.811081790924073,
      "epoch": 4.513220372869496,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003131168284025017,
      "loss": 5.2239,
      "mean_token_accuracy": 0.2162394106388092,
      "num_tokens": 88324480.0,
      "step": 45150
    },
    {
      "entropy": 5.82098650932312,
      "epoch": 4.513720197930724,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031308165467234834,
      "loss": 5.2503,
      "mean_token_accuracy": 0.2198092132806778,
      "num_tokens": 88333072.0,
      "step": 45155
    },
    {
      "entropy": 5.721682596206665,
      "epoch": 4.514220022991953,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031304647998407027,
      "loss": 5.2048,
      "mean_token_accuracy": 0.22528823763132094,
      "num_tokens": 88342993.0,
      "step": 45160
    },
    {
      "entropy": 5.77793869972229,
      "epoch": 4.514719848053181,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031301130433855265,
      "loss": 5.2856,
      "mean_token_accuracy": 0.21260148137807847,
      "num_tokens": 88352615.0,
      "step": 45165
    },
    {
      "entropy": 5.82330174446106,
      "epoch": 4.51521967311441,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003129761277366803,
      "loss": 5.1671,
      "mean_token_accuracy": 0.21475086808204652,
      "num_tokens": 88362496.0,
      "step": 45170
    },
    {
      "entropy": 5.900026607513428,
      "epoch": 4.515719498175638,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031294095017933847,
      "loss": 5.3755,
      "mean_token_accuracy": 0.20678658038377762,
      "num_tokens": 88373776.0,
      "step": 45175
    },
    {
      "entropy": 5.720180606842041,
      "epoch": 4.516219323236867,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031290577166741203,
      "loss": 5.219,
      "mean_token_accuracy": 0.2204141288995743,
      "num_tokens": 88384169.0,
      "step": 45180
    },
    {
      "entropy": 5.789909362792969,
      "epoch": 4.516719148298096,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003128705922017862,
      "loss": 5.2248,
      "mean_token_accuracy": 0.21566945910453797,
      "num_tokens": 88393736.0,
      "step": 45185
    },
    {
      "entropy": 5.767181253433227,
      "epoch": 4.517218973359324,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031283541178334594,
      "loss": 5.1717,
      "mean_token_accuracy": 0.2226429581642151,
      "num_tokens": 88402937.0,
      "step": 45190
    },
    {
      "entropy": 5.801085710525513,
      "epoch": 4.517718798420553,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003128002304129764,
      "loss": 5.2236,
      "mean_token_accuracy": 0.2162722408771515,
      "num_tokens": 88412065.0,
      "step": 45195
    },
    {
      "entropy": 5.785986661911011,
      "epoch": 4.518218623481781,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031276504809156285,
      "loss": 5.2208,
      "mean_token_accuracy": 0.21799101829528808,
      "num_tokens": 88421860.0,
      "step": 45200
    },
    {
      "entropy": 5.815335512161255,
      "epoch": 4.51871844854301,
      "grad_norm": 1.125,
      "learning_rate": 0.00031272986481999045,
      "loss": 5.2642,
      "mean_token_accuracy": 0.20934195816516876,
      "num_tokens": 88431467.0,
      "step": 45205
    },
    {
      "entropy": 5.822039985656739,
      "epoch": 4.519218273604238,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031269468059914434,
      "loss": 5.2117,
      "mean_token_accuracy": 0.22021090537309645,
      "num_tokens": 88440638.0,
      "step": 45210
    },
    {
      "entropy": 5.8422629833221436,
      "epoch": 4.519718098665467,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031265949542990973,
      "loss": 5.381,
      "mean_token_accuracy": 0.20458482950925827,
      "num_tokens": 88449822.0,
      "step": 45215
    },
    {
      "entropy": 5.837202453613282,
      "epoch": 4.520217923726696,
      "grad_norm": 1.15625,
      "learning_rate": 0.000312624309313172,
      "loss": 5.3854,
      "mean_token_accuracy": 0.20330490469932555,
      "num_tokens": 88460085.0,
      "step": 45220
    },
    {
      "entropy": 5.8275469779968265,
      "epoch": 4.520717748787924,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031258912224981617,
      "loss": 5.2385,
      "mean_token_accuracy": 0.21723705977201463,
      "num_tokens": 88469246.0,
      "step": 45225
    },
    {
      "entropy": 5.816809129714966,
      "epoch": 4.521217573849153,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031255393424072783,
      "loss": 5.2608,
      "mean_token_accuracy": 0.21389424949884414,
      "num_tokens": 88478999.0,
      "step": 45230
    },
    {
      "entropy": 5.747324991226196,
      "epoch": 4.521717398910381,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003125187452867921,
      "loss": 5.0796,
      "mean_token_accuracy": 0.2235329493880272,
      "num_tokens": 88487460.0,
      "step": 45235
    },
    {
      "entropy": 5.7262458324432375,
      "epoch": 4.52221722397161,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003124835553888944,
      "loss": 5.1593,
      "mean_token_accuracy": 0.22598715126514435,
      "num_tokens": 88497143.0,
      "step": 45240
    },
    {
      "entropy": 5.766931200027466,
      "epoch": 4.522717049032838,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003124483645479201,
      "loss": 5.2106,
      "mean_token_accuracy": 0.2194676712155342,
      "num_tokens": 88506771.0,
      "step": 45245
    },
    {
      "entropy": 5.802984428405762,
      "epoch": 4.523216874094067,
      "grad_norm": 1.125,
      "learning_rate": 0.0003124131727647547,
      "loss": 5.3243,
      "mean_token_accuracy": 0.2092401683330536,
      "num_tokens": 88516619.0,
      "step": 45250
    },
    {
      "entropy": 5.7273180961608885,
      "epoch": 4.523716699155296,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031237798004028327,
      "loss": 5.0701,
      "mean_token_accuracy": 0.23104711771011352,
      "num_tokens": 88526174.0,
      "step": 45255
    },
    {
      "entropy": 5.7882341861724855,
      "epoch": 4.524216524216524,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031234278637539163,
      "loss": 5.2306,
      "mean_token_accuracy": 0.21336816549301146,
      "num_tokens": 88536184.0,
      "step": 45260
    },
    {
      "entropy": 5.809139966964722,
      "epoch": 4.524716349277753,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000312307591770965,
      "loss": 5.2205,
      "mean_token_accuracy": 0.21622334420681,
      "num_tokens": 88545406.0,
      "step": 45265
    },
    {
      "entropy": 5.793089962005615,
      "epoch": 4.525216174338981,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031227239622788904,
      "loss": 5.186,
      "mean_token_accuracy": 0.22302854657173157,
      "num_tokens": 88555595.0,
      "step": 45270
    },
    {
      "entropy": 5.765991640090943,
      "epoch": 4.52571599940021,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003122371997470492,
      "loss": 5.2129,
      "mean_token_accuracy": 0.21756531894207,
      "num_tokens": 88564845.0,
      "step": 45275
    },
    {
      "entropy": 5.76388955116272,
      "epoch": 4.526215824461438,
      "grad_norm": 1.15625,
      "learning_rate": 0.000312202002329331,
      "loss": 5.1717,
      "mean_token_accuracy": 0.23163320124149323,
      "num_tokens": 88574653.0,
      "step": 45280
    },
    {
      "entropy": 5.82817587852478,
      "epoch": 4.526715649522667,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003121668039756199,
      "loss": 5.2202,
      "mean_token_accuracy": 0.2214185506105423,
      "num_tokens": 88585780.0,
      "step": 45285
    },
    {
      "entropy": 5.819263935089111,
      "epoch": 4.527215474583896,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031213160468680166,
      "loss": 5.2237,
      "mean_token_accuracy": 0.21913135349750518,
      "num_tokens": 88596266.0,
      "step": 45290
    },
    {
      "entropy": 5.816896581649781,
      "epoch": 4.527715299645124,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031209640446376173,
      "loss": 5.3238,
      "mean_token_accuracy": 0.21166047155857087,
      "num_tokens": 88606991.0,
      "step": 45295
    },
    {
      "entropy": 5.701585292816162,
      "epoch": 4.528215124706353,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031206120330738593,
      "loss": 5.1364,
      "mean_token_accuracy": 0.23013996630907058,
      "num_tokens": 88617150.0,
      "step": 45300
    },
    {
      "entropy": 5.659202623367309,
      "epoch": 4.528714949767581,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003120260012185597,
      "loss": 5.0444,
      "mean_token_accuracy": 0.240819288790226,
      "num_tokens": 88626379.0,
      "step": 45305
    },
    {
      "entropy": 5.766166639328003,
      "epoch": 4.52921477482881,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003119907981981689,
      "loss": 5.2176,
      "mean_token_accuracy": 0.21408906430006028,
      "num_tokens": 88636617.0,
      "step": 45310
    },
    {
      "entropy": 5.795148277282715,
      "epoch": 4.529714599890038,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031195559424709895,
      "loss": 5.2962,
      "mean_token_accuracy": 0.20696952641010286,
      "num_tokens": 88646632.0,
      "step": 45315
    },
    {
      "entropy": 5.888282299041748,
      "epoch": 4.530214424951267,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031192038936623585,
      "loss": 5.317,
      "mean_token_accuracy": 0.20862434804439545,
      "num_tokens": 88657263.0,
      "step": 45320
    },
    {
      "entropy": 5.820746183395386,
      "epoch": 4.530714250012496,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031188518355646536,
      "loss": 5.2663,
      "mean_token_accuracy": 0.2085834860801697,
      "num_tokens": 88666978.0,
      "step": 45325
    },
    {
      "entropy": 5.846737813949585,
      "epoch": 4.531214075073724,
      "grad_norm": 1.1875,
      "learning_rate": 0.000311849976818673,
      "loss": 5.3294,
      "mean_token_accuracy": 0.21074243485927582,
      "num_tokens": 88677282.0,
      "step": 45330
    },
    {
      "entropy": 5.7240447998046875,
      "epoch": 4.531713900134953,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003118147691537448,
      "loss": 5.0877,
      "mean_token_accuracy": 0.23643593192100526,
      "num_tokens": 88686329.0,
      "step": 45335
    },
    {
      "entropy": 5.823894453048706,
      "epoch": 4.532213725196181,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003117795605625664,
      "loss": 5.2774,
      "mean_token_accuracy": 0.2129955470561981,
      "num_tokens": 88697561.0,
      "step": 45340
    },
    {
      "entropy": 5.857670450210572,
      "epoch": 4.53271355025741,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031174435104602377,
      "loss": 5.2595,
      "mean_token_accuracy": 0.22142484188079833,
      "num_tokens": 88707970.0,
      "step": 45345
    },
    {
      "entropy": 5.834522867202759,
      "epoch": 4.533213375318638,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003117091406050027,
      "loss": 5.2667,
      "mean_token_accuracy": 0.21483464241027833,
      "num_tokens": 88718178.0,
      "step": 45350
    },
    {
      "entropy": 5.824514102935791,
      "epoch": 4.533713200379867,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003116739292403891,
      "loss": 5.1954,
      "mean_token_accuracy": 0.2253935769200325,
      "num_tokens": 88728861.0,
      "step": 45355
    },
    {
      "entropy": 5.83847074508667,
      "epoch": 4.534213025441096,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031163871695306895,
      "loss": 5.3427,
      "mean_token_accuracy": 0.20644401162862777,
      "num_tokens": 88739834.0,
      "step": 45360
    },
    {
      "entropy": 5.837659931182861,
      "epoch": 4.534712850502324,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000311603503743928,
      "loss": 5.3397,
      "mean_token_accuracy": 0.2138324111700058,
      "num_tokens": 88749296.0,
      "step": 45365
    },
    {
      "entropy": 5.776829051971435,
      "epoch": 4.535212675563553,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031156828961385235,
      "loss": 5.1816,
      "mean_token_accuracy": 0.2300407275557518,
      "num_tokens": 88758912.0,
      "step": 45370
    },
    {
      "entropy": 5.836223936080932,
      "epoch": 4.535712500624781,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000311533074563728,
      "loss": 5.2503,
      "mean_token_accuracy": 0.21759783625602722,
      "num_tokens": 88768610.0,
      "step": 45375
    },
    {
      "entropy": 5.888637399673462,
      "epoch": 4.53621232568601,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031149785859444087,
      "loss": 5.3227,
      "mean_token_accuracy": 0.20922306329011917,
      "num_tokens": 88778246.0,
      "step": 45380
    },
    {
      "entropy": 5.776965808868408,
      "epoch": 4.536712150747238,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031146264170687696,
      "loss": 5.2273,
      "mean_token_accuracy": 0.22204288244247436,
      "num_tokens": 88788732.0,
      "step": 45385
    },
    {
      "entropy": 5.8261909008026125,
      "epoch": 4.537211975808467,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031142742390192243,
      "loss": 5.2819,
      "mean_token_accuracy": 0.2135410413146019,
      "num_tokens": 88797962.0,
      "step": 45390
    },
    {
      "entropy": 5.772514963150025,
      "epoch": 4.537711800869696,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031139220518046324,
      "loss": 5.2075,
      "mean_token_accuracy": 0.21833599656820296,
      "num_tokens": 88809062.0,
      "step": 45395
    },
    {
      "entropy": 5.792853546142578,
      "epoch": 4.538211625930924,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031135698554338555,
      "loss": 5.2479,
      "mean_token_accuracy": 0.22025878727436066,
      "num_tokens": 88818378.0,
      "step": 45400
    },
    {
      "entropy": 5.835050487518311,
      "epoch": 4.538711450992153,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031132176499157544,
      "loss": 5.3236,
      "mean_token_accuracy": 0.21046533435583115,
      "num_tokens": 88828008.0,
      "step": 45405
    },
    {
      "entropy": 5.800335025787353,
      "epoch": 4.539211276053381,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031128654352591907,
      "loss": 5.2164,
      "mean_token_accuracy": 0.21447995007038118,
      "num_tokens": 88837793.0,
      "step": 45410
    },
    {
      "entropy": 5.797150611877441,
      "epoch": 4.53971110111461,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031125132114730267,
      "loss": 5.2046,
      "mean_token_accuracy": 0.21791865676641464,
      "num_tokens": 88848246.0,
      "step": 45415
    },
    {
      "entropy": 5.8242181777954105,
      "epoch": 4.540210926175838,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031121609785661225,
      "loss": 5.3011,
      "mean_token_accuracy": 0.21758143603801727,
      "num_tokens": 88857935.0,
      "step": 45420
    },
    {
      "entropy": 5.8327678680419925,
      "epoch": 4.540710751237067,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031118087365473426,
      "loss": 5.2724,
      "mean_token_accuracy": 0.21543115377426147,
      "num_tokens": 88867387.0,
      "step": 45425
    },
    {
      "entropy": 5.848909616470337,
      "epoch": 4.541210576298296,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031114564854255465,
      "loss": 5.2961,
      "mean_token_accuracy": 0.21215808391571045,
      "num_tokens": 88877327.0,
      "step": 45430
    },
    {
      "entropy": 5.802414989471435,
      "epoch": 4.541710401359524,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031111042252095997,
      "loss": 5.2734,
      "mean_token_accuracy": 0.21834907084703445,
      "num_tokens": 88886995.0,
      "step": 45435
    },
    {
      "entropy": 5.826169729232788,
      "epoch": 4.542210226420753,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031107519559083624,
      "loss": 5.2686,
      "mean_token_accuracy": 0.22005391567945481,
      "num_tokens": 88896698.0,
      "step": 45440
    },
    {
      "entropy": 5.796889305114746,
      "epoch": 4.542710051481981,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003110399677530699,
      "loss": 5.2253,
      "mean_token_accuracy": 0.2211455762386322,
      "num_tokens": 88906742.0,
      "step": 45445
    },
    {
      "entropy": 5.832339668273926,
      "epoch": 4.54320987654321,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031100473900854726,
      "loss": 5.2646,
      "mean_token_accuracy": 0.21117539405822755,
      "num_tokens": 88916755.0,
      "step": 45450
    },
    {
      "entropy": 5.8178201675415036,
      "epoch": 4.543709701604438,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031096950935815464,
      "loss": 5.2652,
      "mean_token_accuracy": 0.21789567321538925,
      "num_tokens": 88925786.0,
      "step": 45455
    },
    {
      "entropy": 5.804132604598999,
      "epoch": 4.544209526665667,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003109342788027784,
      "loss": 5.2564,
      "mean_token_accuracy": 0.21756273359060288,
      "num_tokens": 88936863.0,
      "step": 45460
    },
    {
      "entropy": 5.82542929649353,
      "epoch": 4.544709351726896,
      "grad_norm": 1.1875,
      "learning_rate": 0.000310899047343305,
      "loss": 5.2223,
      "mean_token_accuracy": 0.21733598411083221,
      "num_tokens": 88945927.0,
      "step": 45465
    },
    {
      "entropy": 5.730693531036377,
      "epoch": 4.545209176788124,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003108638149806208,
      "loss": 5.2037,
      "mean_token_accuracy": 0.22340935170650483,
      "num_tokens": 88955712.0,
      "step": 45470
    },
    {
      "entropy": 5.775151872634888,
      "epoch": 4.545709001849353,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003108285817156122,
      "loss": 5.1635,
      "mean_token_accuracy": 0.225080469250679,
      "num_tokens": 88964052.0,
      "step": 45475
    },
    {
      "entropy": 5.752158498764038,
      "epoch": 4.546208826910581,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031079334754916573,
      "loss": 5.1485,
      "mean_token_accuracy": 0.22151713520288469,
      "num_tokens": 88973564.0,
      "step": 45480
    },
    {
      "entropy": 5.671452236175537,
      "epoch": 4.54670865197181,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0003107581124821679,
      "loss": 5.1299,
      "mean_token_accuracy": 0.24208773970603942,
      "num_tokens": 88983511.0,
      "step": 45485
    },
    {
      "entropy": 5.801620197296143,
      "epoch": 4.547208477033038,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031072287651550507,
      "loss": 5.1883,
      "mean_token_accuracy": 0.21561166495084763,
      "num_tokens": 88993182.0,
      "step": 45490
    },
    {
      "entropy": 5.806864595413208,
      "epoch": 4.547708302094267,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003106876396500639,
      "loss": 5.2298,
      "mean_token_accuracy": 0.22043537348508835,
      "num_tokens": 89002432.0,
      "step": 45495
    },
    {
      "entropy": 5.793449401855469,
      "epoch": 4.548208127155496,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003106524018867309,
      "loss": 5.2378,
      "mean_token_accuracy": 0.2233164280653,
      "num_tokens": 89011405.0,
      "step": 45500
    },
    {
      "entropy": 5.920573711395264,
      "epoch": 4.548707952216724,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031061716322639265,
      "loss": 5.3474,
      "mean_token_accuracy": 0.2076001212000847,
      "num_tokens": 89023327.0,
      "step": 45505
    },
    {
      "entropy": 5.88239631652832,
      "epoch": 4.549207777277953,
      "grad_norm": 1.25,
      "learning_rate": 0.00031058192366993566,
      "loss": 5.3409,
      "mean_token_accuracy": 0.2054532065987587,
      "num_tokens": 89033382.0,
      "step": 45510
    },
    {
      "entropy": 5.818471050262451,
      "epoch": 4.549707602339181,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031054668321824677,
      "loss": 5.2403,
      "mean_token_accuracy": 0.22171510457992555,
      "num_tokens": 89044478.0,
      "step": 45515
    },
    {
      "entropy": 5.836854457855225,
      "epoch": 4.55020742740041,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003105114418722124,
      "loss": 5.3153,
      "mean_token_accuracy": 0.21041550785303115,
      "num_tokens": 89053361.0,
      "step": 45520
    },
    {
      "entropy": 5.801882028579712,
      "epoch": 4.550707252461638,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003104761996327193,
      "loss": 5.1816,
      "mean_token_accuracy": 0.22355007529258727,
      "num_tokens": 89062710.0,
      "step": 45525
    },
    {
      "entropy": 5.857425546646118,
      "epoch": 4.551207077522867,
      "grad_norm": 1.375,
      "learning_rate": 0.0003104409565006541,
      "loss": 5.2938,
      "mean_token_accuracy": 0.20636566877365112,
      "num_tokens": 89071897.0,
      "step": 45530
    },
    {
      "entropy": 5.818840885162354,
      "epoch": 4.551706902584096,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003104057124769036,
      "loss": 5.2338,
      "mean_token_accuracy": 0.21350340098142623,
      "num_tokens": 89081536.0,
      "step": 45535
    },
    {
      "entropy": 5.875035524368286,
      "epoch": 4.552206727645324,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003103704675623544,
      "loss": 5.3484,
      "mean_token_accuracy": 0.2103116914629936,
      "num_tokens": 89092082.0,
      "step": 45540
    },
    {
      "entropy": 5.903117704391479,
      "epoch": 4.552706552706553,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003103352217578935,
      "loss": 5.3379,
      "mean_token_accuracy": 0.20403692573308946,
      "num_tokens": 89101665.0,
      "step": 45545
    },
    {
      "entropy": 5.835641622543335,
      "epoch": 4.553206377767781,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031029997506440735,
      "loss": 5.2687,
      "mean_token_accuracy": 0.2123265564441681,
      "num_tokens": 89110555.0,
      "step": 45550
    },
    {
      "entropy": 5.751908636093139,
      "epoch": 4.55370620282901,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003102647274827831,
      "loss": 5.2497,
      "mean_token_accuracy": 0.2157052204012871,
      "num_tokens": 89119850.0,
      "step": 45555
    },
    {
      "entropy": 5.810051298141479,
      "epoch": 4.554206027890238,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031022947901390724,
      "loss": 5.24,
      "mean_token_accuracy": 0.21748343408107756,
      "num_tokens": 89129786.0,
      "step": 45560
    },
    {
      "entropy": 5.8709272861480715,
      "epoch": 4.554705852951467,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031019422965866683,
      "loss": 5.2229,
      "mean_token_accuracy": 0.21939784288406372,
      "num_tokens": 89139187.0,
      "step": 45565
    },
    {
      "entropy": 5.770851278305054,
      "epoch": 4.555205678012696,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003101589794179486,
      "loss": 5.1344,
      "mean_token_accuracy": 0.22416032403707503,
      "num_tokens": 89148832.0,
      "step": 45570
    },
    {
      "entropy": 5.843187856674194,
      "epoch": 4.555705503073924,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003101237282926394,
      "loss": 5.3426,
      "mean_token_accuracy": 0.21094538122415543,
      "num_tokens": 89157762.0,
      "step": 45575
    },
    {
      "entropy": 5.832079648971558,
      "epoch": 4.556205328135153,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003100884762836264,
      "loss": 5.2599,
      "mean_token_accuracy": 0.21539702862501145,
      "num_tokens": 89167430.0,
      "step": 45580
    },
    {
      "entropy": 5.83742127418518,
      "epoch": 4.556705153196381,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003100532233917963,
      "loss": 5.2822,
      "mean_token_accuracy": 0.2241617664694786,
      "num_tokens": 89177847.0,
      "step": 45585
    },
    {
      "entropy": 5.810140705108642,
      "epoch": 4.55720497825761,
      "grad_norm": 1.125,
      "learning_rate": 0.00031001796961803614,
      "loss": 5.2469,
      "mean_token_accuracy": 0.21992963701486587,
      "num_tokens": 89187846.0,
      "step": 45590
    },
    {
      "entropy": 5.874144268035889,
      "epoch": 4.557704803318838,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030998271496323286,
      "loss": 5.3116,
      "mean_token_accuracy": 0.21046017706394196,
      "num_tokens": 89196974.0,
      "step": 45595
    },
    {
      "entropy": 5.845656967163086,
      "epoch": 4.558204628380067,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030994745942827345,
      "loss": 5.2673,
      "mean_token_accuracy": 0.21232899129390717,
      "num_tokens": 89207224.0,
      "step": 45600
    },
    {
      "entropy": 5.7835687637329105,
      "epoch": 4.558704453441296,
      "grad_norm": 1.296875,
      "learning_rate": 0.000309912203014045,
      "loss": 5.2252,
      "mean_token_accuracy": 0.21880197077989577,
      "num_tokens": 89216656.0,
      "step": 45605
    },
    {
      "entropy": 5.809775686264038,
      "epoch": 4.559204278502524,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003098769457214345,
      "loss": 5.2624,
      "mean_token_accuracy": 0.2122936874628067,
      "num_tokens": 89227036.0,
      "step": 45610
    },
    {
      "entropy": 5.84205265045166,
      "epoch": 4.559704103563753,
      "grad_norm": 1.09375,
      "learning_rate": 0.000309841687551329,
      "loss": 5.2508,
      "mean_token_accuracy": 0.2121261715888977,
      "num_tokens": 89237239.0,
      "step": 45615
    },
    {
      "entropy": 5.873048496246338,
      "epoch": 4.560203928624981,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003098064285046156,
      "loss": 5.3215,
      "mean_token_accuracy": 0.2078541949391365,
      "num_tokens": 89247267.0,
      "step": 45620
    },
    {
      "entropy": 5.703115797042846,
      "epoch": 4.56070375368621,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003097711685821815,
      "loss": 5.1742,
      "mean_token_accuracy": 0.22506116181612015,
      "num_tokens": 89257688.0,
      "step": 45625
    },
    {
      "entropy": 5.745402908325195,
      "epoch": 4.561203578747438,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030973590778491354,
      "loss": 5.1725,
      "mean_token_accuracy": 0.21961015313863755,
      "num_tokens": 89267573.0,
      "step": 45630
    },
    {
      "entropy": 5.807851696014405,
      "epoch": 4.561703403808667,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030970064611369927,
      "loss": 5.2193,
      "mean_token_accuracy": 0.2167785257101059,
      "num_tokens": 89276537.0,
      "step": 45635
    },
    {
      "entropy": 5.815820932388306,
      "epoch": 4.562203228869896,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003096653835694256,
      "loss": 5.2373,
      "mean_token_accuracy": 0.2226124405860901,
      "num_tokens": 89286871.0,
      "step": 45640
    },
    {
      "entropy": 5.8550371646881105,
      "epoch": 4.562703053931124,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003096301201529798,
      "loss": 5.3153,
      "mean_token_accuracy": 0.20964418053627015,
      "num_tokens": 89297553.0,
      "step": 45645
    },
    {
      "entropy": 5.85530424118042,
      "epoch": 4.563202878992353,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003095948558652492,
      "loss": 5.3527,
      "mean_token_accuracy": 0.21121478229761123,
      "num_tokens": 89307669.0,
      "step": 45650
    },
    {
      "entropy": 5.842284345626831,
      "epoch": 4.563702704053581,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030955959070712075,
      "loss": 5.2102,
      "mean_token_accuracy": 0.21978096961975097,
      "num_tokens": 89318837.0,
      "step": 45655
    },
    {
      "entropy": 5.869230794906616,
      "epoch": 4.56420252911481,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030952432467948196,
      "loss": 5.2697,
      "mean_token_accuracy": 0.20738609433174132,
      "num_tokens": 89328227.0,
      "step": 45660
    },
    {
      "entropy": 5.865582132339478,
      "epoch": 4.564702354176038,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003094890577832201,
      "loss": 5.2758,
      "mean_token_accuracy": 0.21912129521369933,
      "num_tokens": 89338757.0,
      "step": 45665
    },
    {
      "entropy": 5.726826143264771,
      "epoch": 4.565202179237267,
      "grad_norm": 1.125,
      "learning_rate": 0.00030945379001922236,
      "loss": 5.2285,
      "mean_token_accuracy": 0.22316578179597854,
      "num_tokens": 89348705.0,
      "step": 45670
    },
    {
      "entropy": 5.795433521270752,
      "epoch": 4.565702004298496,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003094185213883761,
      "loss": 5.3084,
      "mean_token_accuracy": 0.21138611286878586,
      "num_tokens": 89358645.0,
      "step": 45675
    },
    {
      "entropy": 5.771960210800171,
      "epoch": 4.566201829359724,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003093832518915688,
      "loss": 5.1028,
      "mean_token_accuracy": 0.22466339468955993,
      "num_tokens": 89367975.0,
      "step": 45680
    },
    {
      "entropy": 5.944163656234741,
      "epoch": 4.566701654420953,
      "grad_norm": 1.0,
      "learning_rate": 0.0003093479815296877,
      "loss": 5.3231,
      "mean_token_accuracy": 0.21611522734165192,
      "num_tokens": 89378499.0,
      "step": 45685
    },
    {
      "entropy": 5.811908292770386,
      "epoch": 4.567201479482181,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030931271030362013,
      "loss": 5.2632,
      "mean_token_accuracy": 0.2214302107691765,
      "num_tokens": 89388279.0,
      "step": 45690
    },
    {
      "entropy": 5.738385391235352,
      "epoch": 4.56770130454341,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003092774382142537,
      "loss": 5.2467,
      "mean_token_accuracy": 0.22985470294952393,
      "num_tokens": 89399057.0,
      "step": 45695
    },
    {
      "entropy": 5.861955785751343,
      "epoch": 4.568201129604638,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030924216526247573,
      "loss": 5.2208,
      "mean_token_accuracy": 0.22701186388731004,
      "num_tokens": 89408410.0,
      "step": 45700
    },
    {
      "entropy": 5.877302503585815,
      "epoch": 4.568700954665867,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003092068914491736,
      "loss": 5.3175,
      "mean_token_accuracy": 0.21306084990501403,
      "num_tokens": 89417621.0,
      "step": 45705
    },
    {
      "entropy": 5.880350828170776,
      "epoch": 4.569200779727096,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030917161677523496,
      "loss": 5.2846,
      "mean_token_accuracy": 0.2077166959643364,
      "num_tokens": 89427992.0,
      "step": 45710
    },
    {
      "entropy": 5.871936225891114,
      "epoch": 4.569700604788324,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003091363412415473,
      "loss": 5.3564,
      "mean_token_accuracy": 0.21018664836883544,
      "num_tokens": 89437507.0,
      "step": 45715
    },
    {
      "entropy": 5.809359502792359,
      "epoch": 4.570200429849553,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030910106484899796,
      "loss": 5.268,
      "mean_token_accuracy": 0.2109096899628639,
      "num_tokens": 89446957.0,
      "step": 45720
    },
    {
      "entropy": 5.734596920013428,
      "epoch": 4.570700254910781,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003090657875984747,
      "loss": 5.1066,
      "mean_token_accuracy": 0.22971229553222655,
      "num_tokens": 89455743.0,
      "step": 45725
    },
    {
      "entropy": 5.826987552642822,
      "epoch": 4.57120007997201,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030903050949086493,
      "loss": 5.2189,
      "mean_token_accuracy": 0.21625320464372635,
      "num_tokens": 89464868.0,
      "step": 45730
    },
    {
      "entropy": 5.796446037292481,
      "epoch": 4.571699905033238,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003089952305270563,
      "loss": 5.1888,
      "mean_token_accuracy": 0.21717164367437364,
      "num_tokens": 89474406.0,
      "step": 45735
    },
    {
      "entropy": 5.807218170166015,
      "epoch": 4.572199730094467,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003089599507079364,
      "loss": 5.2472,
      "mean_token_accuracy": 0.21919426023960115,
      "num_tokens": 89483594.0,
      "step": 45740
    },
    {
      "entropy": 5.647482776641846,
      "epoch": 4.572699555155696,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030892467003439296,
      "loss": 5.0976,
      "mean_token_accuracy": 0.23484918624162673,
      "num_tokens": 89493501.0,
      "step": 45745
    },
    {
      "entropy": 5.792604255676269,
      "epoch": 4.573199380216924,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030888938850731347,
      "loss": 5.228,
      "mean_token_accuracy": 0.20973169356584548,
      "num_tokens": 89504040.0,
      "step": 45750
    },
    {
      "entropy": 5.81112961769104,
      "epoch": 4.573699205278153,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030885410612758573,
      "loss": 5.246,
      "mean_token_accuracy": 0.21416978687047958,
      "num_tokens": 89513857.0,
      "step": 45755
    },
    {
      "entropy": 5.809723663330078,
      "epoch": 4.574199030339381,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030881882289609733,
      "loss": 5.249,
      "mean_token_accuracy": 0.22338341772556305,
      "num_tokens": 89522768.0,
      "step": 45760
    },
    {
      "entropy": 5.88418140411377,
      "epoch": 4.57469885540061,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003087835388137361,
      "loss": 5.2139,
      "mean_token_accuracy": 0.2188856765627861,
      "num_tokens": 89533327.0,
      "step": 45765
    },
    {
      "entropy": 5.817886018753052,
      "epoch": 4.575198680461838,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003087482538813897,
      "loss": 5.2315,
      "mean_token_accuracy": 0.21812163591384887,
      "num_tokens": 89543145.0,
      "step": 45770
    },
    {
      "entropy": 5.896750593185425,
      "epoch": 4.575698505523067,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030871296809994596,
      "loss": 5.3117,
      "mean_token_accuracy": 0.20569125264883042,
      "num_tokens": 89553007.0,
      "step": 45775
    },
    {
      "entropy": 5.723104238510132,
      "epoch": 4.576198330584296,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003086776814702926,
      "loss": 5.1894,
      "mean_token_accuracy": 0.22625095546245574,
      "num_tokens": 89563196.0,
      "step": 45780
    },
    {
      "entropy": 5.860363245010376,
      "epoch": 4.576698155645524,
      "grad_norm": 1.25,
      "learning_rate": 0.00030864239399331744,
      "loss": 5.3014,
      "mean_token_accuracy": 0.2059067592024803,
      "num_tokens": 89573265.0,
      "step": 45785
    },
    {
      "entropy": 5.79345440864563,
      "epoch": 4.577197980706753,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003086071056699083,
      "loss": 5.2194,
      "mean_token_accuracy": 0.2195821687579155,
      "num_tokens": 89583196.0,
      "step": 45790
    },
    {
      "entropy": 5.892476320266724,
      "epoch": 4.577697805767981,
      "grad_norm": 1.25,
      "learning_rate": 0.000308571816500953,
      "loss": 5.3393,
      "mean_token_accuracy": 0.20929695218801497,
      "num_tokens": 89592349.0,
      "step": 45795
    },
    {
      "entropy": 5.735981559753418,
      "epoch": 4.57819763082921,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003085365264873395,
      "loss": 5.1512,
      "mean_token_accuracy": 0.22897801101207732,
      "num_tokens": 89602239.0,
      "step": 45800
    },
    {
      "entropy": 5.827107858657837,
      "epoch": 4.578697455890438,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003085012356299556,
      "loss": 5.2298,
      "mean_token_accuracy": 0.216062992811203,
      "num_tokens": 89611150.0,
      "step": 45805
    },
    {
      "entropy": 5.866182088851929,
      "epoch": 4.579197280951667,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030846594392968925,
      "loss": 5.2846,
      "mean_token_accuracy": 0.216497066617012,
      "num_tokens": 89620360.0,
      "step": 45810
    },
    {
      "entropy": 5.785254573822021,
      "epoch": 4.579697106012896,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030843065138742826,
      "loss": 5.2983,
      "mean_token_accuracy": 0.2131400778889656,
      "num_tokens": 89629860.0,
      "step": 45815
    },
    {
      "entropy": 5.787085628509521,
      "epoch": 4.580196931074124,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030839535800406085,
      "loss": 5.2854,
      "mean_token_accuracy": 0.21335234940052034,
      "num_tokens": 89639465.0,
      "step": 45820
    },
    {
      "entropy": 5.821726846694946,
      "epoch": 4.580696756135353,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030836006378047464,
      "loss": 5.2829,
      "mean_token_accuracy": 0.21954575181007385,
      "num_tokens": 89648398.0,
      "step": 45825
    },
    {
      "entropy": 5.947697687149048,
      "epoch": 4.581196581196581,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003083247687175579,
      "loss": 5.4071,
      "mean_token_accuracy": 0.20578573793172836,
      "num_tokens": 89658587.0,
      "step": 45830
    },
    {
      "entropy": 5.846202278137207,
      "epoch": 4.58169640625781,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030828947281619863,
      "loss": 5.2516,
      "mean_token_accuracy": 0.2170817345380783,
      "num_tokens": 89667841.0,
      "step": 45835
    },
    {
      "entropy": 5.852562952041626,
      "epoch": 4.582196231319038,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003082541760772847,
      "loss": 5.2813,
      "mean_token_accuracy": 0.21450962126255035,
      "num_tokens": 89677019.0,
      "step": 45840
    },
    {
      "entropy": 5.742733907699585,
      "epoch": 4.582696056380267,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030821887850170424,
      "loss": 5.1425,
      "mean_token_accuracy": 0.22904059439897537,
      "num_tokens": 89687096.0,
      "step": 45845
    },
    {
      "entropy": 5.839539384841919,
      "epoch": 4.583195881441496,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030818358009034535,
      "loss": 5.2701,
      "mean_token_accuracy": 0.2144462510943413,
      "num_tokens": 89697533.0,
      "step": 45850
    },
    {
      "entropy": 5.902737283706665,
      "epoch": 4.583695706502724,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030814828084409614,
      "loss": 5.3939,
      "mean_token_accuracy": 0.2046980768442154,
      "num_tokens": 89706683.0,
      "step": 45855
    },
    {
      "entropy": 5.801454544067383,
      "epoch": 4.584195531563953,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030811298076384474,
      "loss": 5.2461,
      "mean_token_accuracy": 0.21528688967227935,
      "num_tokens": 89716743.0,
      "step": 45860
    },
    {
      "entropy": 5.91715726852417,
      "epoch": 4.584695356625181,
      "grad_norm": 1.25,
      "learning_rate": 0.00030807767985047927,
      "loss": 5.3386,
      "mean_token_accuracy": 0.2091815546154976,
      "num_tokens": 89727260.0,
      "step": 45865
    },
    {
      "entropy": 5.717611646652221,
      "epoch": 4.58519518168641,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003080423781048879,
      "loss": 5.1307,
      "mean_token_accuracy": 0.22943572252988814,
      "num_tokens": 89737248.0,
      "step": 45870
    },
    {
      "entropy": 5.808887672424317,
      "epoch": 4.585695006747638,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030800707552795873,
      "loss": 5.2657,
      "mean_token_accuracy": 0.21845489591360093,
      "num_tokens": 89747526.0,
      "step": 45875
    },
    {
      "entropy": 5.875220251083374,
      "epoch": 4.586194831808867,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030797177212058,
      "loss": 5.3111,
      "mean_token_accuracy": 0.20643401145935059,
      "num_tokens": 89758672.0,
      "step": 45880
    },
    {
      "entropy": 5.749479198455811,
      "epoch": 4.586694656870096,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003079364678836401,
      "loss": 5.1231,
      "mean_token_accuracy": 0.229768505692482,
      "num_tokens": 89768518.0,
      "step": 45885
    },
    {
      "entropy": 5.836475896835327,
      "epoch": 4.587194481931324,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030790116281802704,
      "loss": 5.3533,
      "mean_token_accuracy": 0.20975780636072158,
      "num_tokens": 89777629.0,
      "step": 45890
    },
    {
      "entropy": 5.727393388748169,
      "epoch": 4.587694306992553,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003078658569246293,
      "loss": 5.1532,
      "mean_token_accuracy": 0.224287810921669,
      "num_tokens": 89788211.0,
      "step": 45895
    },
    {
      "entropy": 5.88448133468628,
      "epoch": 4.588194132053781,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003078305502043349,
      "loss": 5.2749,
      "mean_token_accuracy": 0.21103160381317138,
      "num_tokens": 89798596.0,
      "step": 45900
    },
    {
      "entropy": 5.788299417495727,
      "epoch": 4.58869395711501,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003077952426580324,
      "loss": 5.245,
      "mean_token_accuracy": 0.21675177961587905,
      "num_tokens": 89806970.0,
      "step": 45905
    },
    {
      "entropy": 5.791986179351807,
      "epoch": 4.589193782176238,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003077599342866101,
      "loss": 5.2044,
      "mean_token_accuracy": 0.22105135917663574,
      "num_tokens": 89816337.0,
      "step": 45910
    },
    {
      "entropy": 5.858228540420532,
      "epoch": 4.589693607237467,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030772462509095615,
      "loss": 5.3193,
      "mean_token_accuracy": 0.21008588075637818,
      "num_tokens": 89826272.0,
      "step": 45915
    },
    {
      "entropy": 5.775860357284546,
      "epoch": 4.590193432298696,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030768931507195907,
      "loss": 5.1907,
      "mean_token_accuracy": 0.21989800781011581,
      "num_tokens": 89836086.0,
      "step": 45920
    },
    {
      "entropy": 5.7909246444702145,
      "epoch": 4.590693257359924,
      "grad_norm": 1.125,
      "learning_rate": 0.00030765400423050725,
      "loss": 5.2,
      "mean_token_accuracy": 0.22402335554361344,
      "num_tokens": 89846795.0,
      "step": 45925
    },
    {
      "entropy": 5.836011981964111,
      "epoch": 4.591193082421153,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030761869256748915,
      "loss": 5.2307,
      "mean_token_accuracy": 0.216341795027256,
      "num_tokens": 89856482.0,
      "step": 45930
    },
    {
      "entropy": 5.762167310714721,
      "epoch": 4.591692907482381,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003075833800837931,
      "loss": 5.3003,
      "mean_token_accuracy": 0.21789863109588622,
      "num_tokens": 89867268.0,
      "step": 45935
    },
    {
      "entropy": 5.824745845794678,
      "epoch": 4.59219273254361,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003075480667803076,
      "loss": 5.2127,
      "mean_token_accuracy": 0.22199615836143494,
      "num_tokens": 89877742.0,
      "step": 45940
    },
    {
      "entropy": 5.849345016479492,
      "epoch": 4.592692557604838,
      "grad_norm": 1.125,
      "learning_rate": 0.0003075127526579211,
      "loss": 5.2741,
      "mean_token_accuracy": 0.212773634493351,
      "num_tokens": 89887505.0,
      "step": 45945
    },
    {
      "entropy": 5.66997652053833,
      "epoch": 4.593192382666067,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003074774377175221,
      "loss": 5.1032,
      "mean_token_accuracy": 0.22801918983459474,
      "num_tokens": 89898527.0,
      "step": 45950
    },
    {
      "entropy": 5.863683891296387,
      "epoch": 4.593692207727296,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003074421219599992,
      "loss": 5.2422,
      "mean_token_accuracy": 0.21784938424825667,
      "num_tokens": 89907668.0,
      "step": 45955
    },
    {
      "entropy": 5.832021474838257,
      "epoch": 4.594192032788524,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030740680538624075,
      "loss": 5.2775,
      "mean_token_accuracy": 0.2137498900294304,
      "num_tokens": 89916571.0,
      "step": 45960
    },
    {
      "entropy": 5.776120138168335,
      "epoch": 4.594691857849752,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003073714879971355,
      "loss": 5.2453,
      "mean_token_accuracy": 0.22049985080957413,
      "num_tokens": 89926602.0,
      "step": 45965
    },
    {
      "entropy": 5.789232397079468,
      "epoch": 4.595191682910981,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030733616979357185,
      "loss": 5.2771,
      "mean_token_accuracy": 0.21364939957857132,
      "num_tokens": 89937190.0,
      "step": 45970
    },
    {
      "entropy": 5.898197841644287,
      "epoch": 4.59569150797221,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003073008507764386,
      "loss": 5.3391,
      "mean_token_accuracy": 0.20651781409978867,
      "num_tokens": 89946156.0,
      "step": 45975
    },
    {
      "entropy": 5.776095724105835,
      "epoch": 4.596191333033438,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030726553094662416,
      "loss": 5.1532,
      "mean_token_accuracy": 0.23340709507465363,
      "num_tokens": 89955160.0,
      "step": 45980
    },
    {
      "entropy": 5.883873462677002,
      "epoch": 4.596691158094667,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003072302103050173,
      "loss": 5.332,
      "mean_token_accuracy": 0.20988837033510208,
      "num_tokens": 89964741.0,
      "step": 45985
    },
    {
      "entropy": 5.837540912628174,
      "epoch": 4.597190983155896,
      "grad_norm": 1.25,
      "learning_rate": 0.0003071948888525067,
      "loss": 5.2466,
      "mean_token_accuracy": 0.21988755762577056,
      "num_tokens": 89975161.0,
      "step": 45990
    },
    {
      "entropy": 5.7771249294281,
      "epoch": 4.597690808217124,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003071595665899809,
      "loss": 5.2097,
      "mean_token_accuracy": 0.21967030167579651,
      "num_tokens": 89984669.0,
      "step": 45995
    },
    {
      "entropy": 5.810055255889893,
      "epoch": 4.598190633278352,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030712424351832875,
      "loss": 5.2551,
      "mean_token_accuracy": 0.21216138005256652,
      "num_tokens": 89994844.0,
      "step": 46000
    },
    {
      "entropy": 5.802334833145141,
      "epoch": 4.598690458339581,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003070889196384388,
      "loss": 5.2057,
      "mean_token_accuracy": 0.22509087920188903,
      "num_tokens": 90004508.0,
      "step": 46005
    },
    {
      "entropy": 5.8267786502838135,
      "epoch": 4.59919028340081,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030705359495119996,
      "loss": 5.2557,
      "mean_token_accuracy": 0.2169906184077263,
      "num_tokens": 90014353.0,
      "step": 46010
    },
    {
      "entropy": 5.78232102394104,
      "epoch": 4.599690108462038,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003070182694575009,
      "loss": 5.1662,
      "mean_token_accuracy": 0.22886427491903305,
      "num_tokens": 90023954.0,
      "step": 46015
    },
    {
      "entropy": 5.856798219680786,
      "epoch": 4.600189933523267,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003069829431582304,
      "loss": 5.2936,
      "mean_token_accuracy": 0.2135015532374382,
      "num_tokens": 90033274.0,
      "step": 46020
    },
    {
      "entropy": 5.855597066879272,
      "epoch": 4.600689758584496,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003069476160542773,
      "loss": 5.316,
      "mean_token_accuracy": 0.21095429062843324,
      "num_tokens": 90043143.0,
      "step": 46025
    },
    {
      "entropy": 5.731807231903076,
      "epoch": 4.6011895836457235,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003069122881465305,
      "loss": 5.2168,
      "mean_token_accuracy": 0.2243155688047409,
      "num_tokens": 90052322.0,
      "step": 46030
    },
    {
      "entropy": 5.833796119689941,
      "epoch": 4.601689408706952,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030687695943587854,
      "loss": 5.2877,
      "mean_token_accuracy": 0.20917681902647017,
      "num_tokens": 90063197.0,
      "step": 46035
    },
    {
      "entropy": 5.800261735916138,
      "epoch": 4.602189233768181,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003068416299232107,
      "loss": 5.1833,
      "mean_token_accuracy": 0.22728277891874313,
      "num_tokens": 90072841.0,
      "step": 46040
    },
    {
      "entropy": 5.858745002746582,
      "epoch": 4.60268905882941,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003068062996094155,
      "loss": 5.2835,
      "mean_token_accuracy": 0.21167115718126298,
      "num_tokens": 90083374.0,
      "step": 46045
    },
    {
      "entropy": 5.805426120758057,
      "epoch": 4.603188883890638,
      "grad_norm": 1.015625,
      "learning_rate": 0.000306770968495382,
      "loss": 5.3178,
      "mean_token_accuracy": 0.21169829815626146,
      "num_tokens": 90094780.0,
      "step": 46050
    },
    {
      "entropy": 5.853502368927002,
      "epoch": 4.603688708951867,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003067356365819992,
      "loss": 5.2184,
      "mean_token_accuracy": 0.21690236926078796,
      "num_tokens": 90104511.0,
      "step": 46055
    },
    {
      "entropy": 5.87246298789978,
      "epoch": 4.604188534013096,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030670030387015585,
      "loss": 5.2501,
      "mean_token_accuracy": 0.2104841575026512,
      "num_tokens": 90113724.0,
      "step": 46060
    },
    {
      "entropy": 5.763779258728027,
      "epoch": 4.6046883590743235,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030666497036074104,
      "loss": 5.213,
      "mean_token_accuracy": 0.21641767024993896,
      "num_tokens": 90123892.0,
      "step": 46065
    },
    {
      "entropy": 5.767425537109375,
      "epoch": 4.605188184135552,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003066296360546437,
      "loss": 5.2324,
      "mean_token_accuracy": 0.2177818387746811,
      "num_tokens": 90134248.0,
      "step": 46070
    },
    {
      "entropy": 5.853458261489868,
      "epoch": 4.605688009196781,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030659430095275283,
      "loss": 5.3199,
      "mean_token_accuracy": 0.20589226335287095,
      "num_tokens": 90144304.0,
      "step": 46075
    },
    {
      "entropy": 5.823599529266358,
      "epoch": 4.60618783425801,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003065589650559575,
      "loss": 5.2185,
      "mean_token_accuracy": 0.21744061410427093,
      "num_tokens": 90154625.0,
      "step": 46080
    },
    {
      "entropy": 5.838016748428345,
      "epoch": 4.606687659319238,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030652362836514687,
      "loss": 5.2262,
      "mean_token_accuracy": 0.21539802849292755,
      "num_tokens": 90164027.0,
      "step": 46085
    },
    {
      "entropy": 5.742555284500122,
      "epoch": 4.607187484380467,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030648829088120974,
      "loss": 5.186,
      "mean_token_accuracy": 0.2218975305557251,
      "num_tokens": 90173803.0,
      "step": 46090
    },
    {
      "entropy": 5.864400243759155,
      "epoch": 4.607687309441696,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003064529526050353,
      "loss": 5.2457,
      "mean_token_accuracy": 0.21283608675003052,
      "num_tokens": 90182495.0,
      "step": 46095
    },
    {
      "entropy": 5.832816696166992,
      "epoch": 4.6081871345029235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003064176135375127,
      "loss": 5.3095,
      "mean_token_accuracy": 0.20747362524271012,
      "num_tokens": 90192925.0,
      "step": 46100
    },
    {
      "entropy": 5.828697633743286,
      "epoch": 4.608686959564152,
      "grad_norm": 1.4375,
      "learning_rate": 0.000306382273679531,
      "loss": 5.3673,
      "mean_token_accuracy": 0.20702870935201645,
      "num_tokens": 90203578.0,
      "step": 46105
    },
    {
      "entropy": 5.668605995178223,
      "epoch": 4.609186784625381,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030634693303197935,
      "loss": 5.0536,
      "mean_token_accuracy": 0.23591149598360062,
      "num_tokens": 90213733.0,
      "step": 46110
    },
    {
      "entropy": 5.6860432624816895,
      "epoch": 4.60968660968661,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030631159159574703,
      "loss": 5.1126,
      "mean_token_accuracy": 0.2328905090689659,
      "num_tokens": 90223489.0,
      "step": 46115
    },
    {
      "entropy": 5.767615127563476,
      "epoch": 4.610186434747838,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030627624937172305,
      "loss": 5.2226,
      "mean_token_accuracy": 0.2286679819226265,
      "num_tokens": 90233713.0,
      "step": 46120
    },
    {
      "entropy": 5.83164119720459,
      "epoch": 4.610686259809067,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003062409063607967,
      "loss": 5.2175,
      "mean_token_accuracy": 0.21664734482765197,
      "num_tokens": 90243747.0,
      "step": 46125
    },
    {
      "entropy": 5.681087684631348,
      "epoch": 4.611186084870296,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003062055625638572,
      "loss": 5.1324,
      "mean_token_accuracy": 0.23098485320806503,
      "num_tokens": 90254169.0,
      "step": 46130
    },
    {
      "entropy": 5.700582981109619,
      "epoch": 4.6116859099315235,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003061702179817937,
      "loss": 5.2012,
      "mean_token_accuracy": 0.22524535953998565,
      "num_tokens": 90263596.0,
      "step": 46135
    },
    {
      "entropy": 5.77263445854187,
      "epoch": 4.612185734992752,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003061348726154956,
      "loss": 5.2103,
      "mean_token_accuracy": 0.2316751077771187,
      "num_tokens": 90272928.0,
      "step": 46140
    },
    {
      "entropy": 5.822495317459106,
      "epoch": 4.612685560053981,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003060995264658521,
      "loss": 5.3142,
      "mean_token_accuracy": 0.21129002124071122,
      "num_tokens": 90282921.0,
      "step": 46145
    },
    {
      "entropy": 5.830654859542847,
      "epoch": 4.61318538511521,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003060641795337526,
      "loss": 5.2099,
      "mean_token_accuracy": 0.22220200598239898,
      "num_tokens": 90292108.0,
      "step": 46150
    },
    {
      "entropy": 5.815455484390259,
      "epoch": 4.613685210176438,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030602883182008625,
      "loss": 5.3116,
      "mean_token_accuracy": 0.20899001359939576,
      "num_tokens": 90301919.0,
      "step": 46155
    },
    {
      "entropy": 5.841420364379883,
      "epoch": 4.614185035237667,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003059934833257425,
      "loss": 5.2947,
      "mean_token_accuracy": 0.20909155756235123,
      "num_tokens": 90312661.0,
      "step": 46160
    },
    {
      "entropy": 5.805564451217651,
      "epoch": 4.614684860298896,
      "grad_norm": 1.25,
      "learning_rate": 0.00030595813405161063,
      "loss": 5.1951,
      "mean_token_accuracy": 0.22073791921138763,
      "num_tokens": 90322621.0,
      "step": 46165
    },
    {
      "entropy": 5.86983003616333,
      "epoch": 4.6151846853601235,
      "grad_norm": 1.125,
      "learning_rate": 0.0003059227839985801,
      "loss": 5.2823,
      "mean_token_accuracy": 0.2078564077615738,
      "num_tokens": 90332431.0,
      "step": 46170
    },
    {
      "entropy": 5.820374774932861,
      "epoch": 4.615684510421352,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003058874331675403,
      "loss": 5.2282,
      "mean_token_accuracy": 0.22433091849088668,
      "num_tokens": 90341875.0,
      "step": 46175
    },
    {
      "entropy": 5.778879117965698,
      "epoch": 4.616184335482581,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003058520815593807,
      "loss": 5.2825,
      "mean_token_accuracy": 0.20697410702705382,
      "num_tokens": 90350986.0,
      "step": 46180
    },
    {
      "entropy": 5.83581805229187,
      "epoch": 4.61668416054381,
      "grad_norm": 1.25,
      "learning_rate": 0.00030581672917499056,
      "loss": 5.3405,
      "mean_token_accuracy": 0.21204326152801514,
      "num_tokens": 90360015.0,
      "step": 46185
    },
    {
      "entropy": 5.895945644378662,
      "epoch": 4.617183985605038,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030578137601525954,
      "loss": 5.3066,
      "mean_token_accuracy": 0.20661678463220595,
      "num_tokens": 90369656.0,
      "step": 46190
    },
    {
      "entropy": 5.831919765472412,
      "epoch": 4.617683810666267,
      "grad_norm": 1.25,
      "learning_rate": 0.0003057460220810769,
      "loss": 5.2732,
      "mean_token_accuracy": 0.21342894732952117,
      "num_tokens": 90379711.0,
      "step": 46195
    },
    {
      "entropy": 5.773853015899658,
      "epoch": 4.618183635727496,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030571066737333227,
      "loss": 5.1533,
      "mean_token_accuracy": 0.22490838915109634,
      "num_tokens": 90389888.0,
      "step": 46200
    },
    {
      "entropy": 5.725597143173218,
      "epoch": 4.6186834607887235,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003056753118929152,
      "loss": 5.1439,
      "mean_token_accuracy": 0.22283715158700942,
      "num_tokens": 90400166.0,
      "step": 46205
    },
    {
      "entropy": 5.717858934402466,
      "epoch": 4.619183285849952,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003056399556407152,
      "loss": 5.0871,
      "mean_token_accuracy": 0.22768468111753465,
      "num_tokens": 90409661.0,
      "step": 46210
    },
    {
      "entropy": 5.748881721496582,
      "epoch": 4.619683110911181,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030560459861762174,
      "loss": 5.2759,
      "mean_token_accuracy": 0.21184435933828355,
      "num_tokens": 90420164.0,
      "step": 46215
    },
    {
      "entropy": 5.762475776672363,
      "epoch": 4.62018293597241,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030556924082452455,
      "loss": 5.184,
      "mean_token_accuracy": 0.22508643716573715,
      "num_tokens": 90430903.0,
      "step": 46220
    },
    {
      "entropy": 5.837968492507935,
      "epoch": 4.620682761033638,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030553388226231304,
      "loss": 5.1805,
      "mean_token_accuracy": 0.2194441720843315,
      "num_tokens": 90441270.0,
      "step": 46225
    },
    {
      "entropy": 5.809137916564941,
      "epoch": 4.621182586094867,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030549852293187686,
      "loss": 5.1673,
      "mean_token_accuracy": 0.22461032569408418,
      "num_tokens": 90450171.0,
      "step": 46230
    },
    {
      "entropy": 5.808086633682251,
      "epoch": 4.621682411156096,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003054631628341058,
      "loss": 5.2761,
      "mean_token_accuracy": 0.21136991083621978,
      "num_tokens": 90460103.0,
      "step": 46235
    },
    {
      "entropy": 5.859152507781983,
      "epoch": 4.6221822362173235,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030542780196988936,
      "loss": 5.3217,
      "mean_token_accuracy": 0.20938776433467865,
      "num_tokens": 90469414.0,
      "step": 46240
    },
    {
      "entropy": 5.797773647308349,
      "epoch": 4.622682061278552,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030539244034011727,
      "loss": 5.2481,
      "mean_token_accuracy": 0.21927333176136016,
      "num_tokens": 90478932.0,
      "step": 46245
    },
    {
      "entropy": 5.8189996719360355,
      "epoch": 4.623181886339781,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003053570779456792,
      "loss": 5.2658,
      "mean_token_accuracy": 0.2150947332382202,
      "num_tokens": 90488910.0,
      "step": 46250
    },
    {
      "entropy": 5.877464675903321,
      "epoch": 4.62368171140101,
      "grad_norm": 1.125,
      "learning_rate": 0.00030532171478746484,
      "loss": 5.2974,
      "mean_token_accuracy": 0.2129878282546997,
      "num_tokens": 90498277.0,
      "step": 46255
    },
    {
      "entropy": 5.771207571029663,
      "epoch": 4.624181536462238,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003052863508663639,
      "loss": 5.1227,
      "mean_token_accuracy": 0.22835996150970458,
      "num_tokens": 90507687.0,
      "step": 46260
    },
    {
      "entropy": 5.872611618041992,
      "epoch": 4.624681361523467,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030525098618326625,
      "loss": 5.3258,
      "mean_token_accuracy": 0.20617612302303315,
      "num_tokens": 90516760.0,
      "step": 46265
    },
    {
      "entropy": 5.8127354145050045,
      "epoch": 4.625181186584696,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030521562073906156,
      "loss": 5.2395,
      "mean_token_accuracy": 0.21659239381551743,
      "num_tokens": 90524881.0,
      "step": 46270
    },
    {
      "entropy": 5.8273858547210695,
      "epoch": 4.6256810116459235,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003051802545346396,
      "loss": 5.2433,
      "mean_token_accuracy": 0.21256582885980607,
      "num_tokens": 90535109.0,
      "step": 46275
    },
    {
      "entropy": 5.873070526123047,
      "epoch": 4.626180836707152,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030514488757089013,
      "loss": 5.3643,
      "mean_token_accuracy": 0.21807502955198288,
      "num_tokens": 90546406.0,
      "step": 46280
    },
    {
      "entropy": 5.841071557998657,
      "epoch": 4.626680661768381,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003051095198487032,
      "loss": 5.2855,
      "mean_token_accuracy": 0.21474418938159942,
      "num_tokens": 90557898.0,
      "step": 46285
    },
    {
      "entropy": 5.757690286636352,
      "epoch": 4.62718048682961,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030507415136896837,
      "loss": 5.1277,
      "mean_token_accuracy": 0.22587536871433259,
      "num_tokens": 90566807.0,
      "step": 46290
    },
    {
      "entropy": 5.835461044311524,
      "epoch": 4.627680311890838,
      "grad_norm": 1.125,
      "learning_rate": 0.0003050387821325757,
      "loss": 5.3148,
      "mean_token_accuracy": 0.20864694863557814,
      "num_tokens": 90577027.0,
      "step": 46295
    },
    {
      "entropy": 5.769676399230957,
      "epoch": 4.628180136952067,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000305003412140415,
      "loss": 5.1824,
      "mean_token_accuracy": 0.2196197122335434,
      "num_tokens": 90586304.0,
      "step": 46300
    },
    {
      "entropy": 5.839774656295776,
      "epoch": 4.628679962013296,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030496804139337617,
      "loss": 5.3269,
      "mean_token_accuracy": 0.20657926946878433,
      "num_tokens": 90596673.0,
      "step": 46305
    },
    {
      "entropy": 5.781647157669068,
      "epoch": 4.6291797870745235,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030493266989234906,
      "loss": 5.2084,
      "mean_token_accuracy": 0.2215774714946747,
      "num_tokens": 90607770.0,
      "step": 46310
    },
    {
      "entropy": 5.779181623458863,
      "epoch": 4.629679612135752,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030489729763822374,
      "loss": 5.1905,
      "mean_token_accuracy": 0.221956205368042,
      "num_tokens": 90617734.0,
      "step": 46315
    },
    {
      "entropy": 5.622830581665039,
      "epoch": 4.630179437196981,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030486192463189017,
      "loss": 5.0903,
      "mean_token_accuracy": 0.2387862354516983,
      "num_tokens": 90627755.0,
      "step": 46320
    },
    {
      "entropy": 5.8374990940094,
      "epoch": 4.63067926225821,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003048265508742381,
      "loss": 5.258,
      "mean_token_accuracy": 0.21572302132844925,
      "num_tokens": 90637860.0,
      "step": 46325
    },
    {
      "entropy": 5.824564838409424,
      "epoch": 4.631179087319438,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030479117636615785,
      "loss": 5.2061,
      "mean_token_accuracy": 0.21727451235055922,
      "num_tokens": 90647829.0,
      "step": 46330
    },
    {
      "entropy": 5.861134481430054,
      "epoch": 4.631678912380667,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030475580110853913,
      "loss": 5.2828,
      "mean_token_accuracy": 0.21449823081493377,
      "num_tokens": 90656711.0,
      "step": 46335
    },
    {
      "entropy": 5.759682035446167,
      "epoch": 4.632178737441896,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003047204251022722,
      "loss": 5.1636,
      "mean_token_accuracy": 0.22326277643442155,
      "num_tokens": 90666995.0,
      "step": 46340
    },
    {
      "entropy": 5.792923212051392,
      "epoch": 4.6326785625031235,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000304685048348247,
      "loss": 5.2929,
      "mean_token_accuracy": 0.21550897806882857,
      "num_tokens": 90677689.0,
      "step": 46345
    },
    {
      "entropy": 5.819082117080688,
      "epoch": 4.633178387564352,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030464967084735357,
      "loss": 5.2422,
      "mean_token_accuracy": 0.22530325055122374,
      "num_tokens": 90688176.0,
      "step": 46350
    },
    {
      "entropy": 5.869632530212402,
      "epoch": 4.633678212625581,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030461429260048213,
      "loss": 5.2512,
      "mean_token_accuracy": 0.2140304207801819,
      "num_tokens": 90697560.0,
      "step": 46355
    },
    {
      "entropy": 5.886376667022705,
      "epoch": 4.63417803768681,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030457891360852267,
      "loss": 5.3491,
      "mean_token_accuracy": 0.2081536367535591,
      "num_tokens": 90707146.0,
      "step": 46360
    },
    {
      "entropy": 5.831919527053833,
      "epoch": 4.634677862748038,
      "grad_norm": 1.25,
      "learning_rate": 0.0003045435338723653,
      "loss": 5.2755,
      "mean_token_accuracy": 0.20458683371543884,
      "num_tokens": 90717429.0,
      "step": 46365
    },
    {
      "entropy": 5.807948589324951,
      "epoch": 4.635177687809267,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030450815339290026,
      "loss": 5.2282,
      "mean_token_accuracy": 0.22524716705083847,
      "num_tokens": 90726439.0,
      "step": 46370
    },
    {
      "entropy": 5.813843202590943,
      "epoch": 4.635677512870496,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030447277217101774,
      "loss": 5.269,
      "mean_token_accuracy": 0.21054776906967163,
      "num_tokens": 90735770.0,
      "step": 46375
    },
    {
      "entropy": 5.804624366760254,
      "epoch": 4.6361773379317235,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003044373902076079,
      "loss": 5.2097,
      "mean_token_accuracy": 0.2168782591819763,
      "num_tokens": 90744288.0,
      "step": 46380
    },
    {
      "entropy": 5.792742681503296,
      "epoch": 4.636677162992952,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030440200750356066,
      "loss": 5.1685,
      "mean_token_accuracy": 0.2189010888338089,
      "num_tokens": 90753179.0,
      "step": 46385
    },
    {
      "entropy": 5.810815906524658,
      "epoch": 4.637176988054181,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003043666240597667,
      "loss": 5.3039,
      "mean_token_accuracy": 0.21675096452236176,
      "num_tokens": 90763732.0,
      "step": 46390
    },
    {
      "entropy": 5.812212705612183,
      "epoch": 4.63767681311541,
      "grad_norm": 1.25,
      "learning_rate": 0.0003043312398771159,
      "loss": 5.2225,
      "mean_token_accuracy": 0.21691542267799377,
      "num_tokens": 90772654.0,
      "step": 46395
    },
    {
      "entropy": 5.814590501785278,
      "epoch": 4.638176638176638,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030429585495649877,
      "loss": 5.3019,
      "mean_token_accuracy": 0.21090812981128693,
      "num_tokens": 90782231.0,
      "step": 46400
    },
    {
      "entropy": 5.875035285949707,
      "epoch": 4.638676463237867,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003042604692988054,
      "loss": 5.2726,
      "mean_token_accuracy": 0.21635109931230545,
      "num_tokens": 90792475.0,
      "step": 46405
    },
    {
      "entropy": 5.871121740341186,
      "epoch": 4.639176288299096,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003042250829049261,
      "loss": 5.2047,
      "mean_token_accuracy": 0.22186841070652008,
      "num_tokens": 90801533.0,
      "step": 46410
    },
    {
      "entropy": 5.834745931625366,
      "epoch": 4.6396761133603235,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003041896957757513,
      "loss": 5.3101,
      "mean_token_accuracy": 0.21438902318477632,
      "num_tokens": 90810626.0,
      "step": 46415
    },
    {
      "entropy": 5.816391849517823,
      "epoch": 4.640175938421552,
      "grad_norm": 1.25,
      "learning_rate": 0.0003041543079121713,
      "loss": 5.2319,
      "mean_token_accuracy": 0.21703548431396485,
      "num_tokens": 90819518.0,
      "step": 46420
    },
    {
      "entropy": 5.916394281387329,
      "epoch": 4.640675763482781,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030411891931507635,
      "loss": 5.4101,
      "mean_token_accuracy": 0.20409306287765502,
      "num_tokens": 90828854.0,
      "step": 46425
    },
    {
      "entropy": 5.829104948043823,
      "epoch": 4.64117558854401,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030408352998535695,
      "loss": 5.208,
      "mean_token_accuracy": 0.22058454900979996,
      "num_tokens": 90838236.0,
      "step": 46430
    },
    {
      "entropy": 5.828600454330444,
      "epoch": 4.641675413605238,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030404813992390337,
      "loss": 5.255,
      "mean_token_accuracy": 0.2170874744653702,
      "num_tokens": 90848332.0,
      "step": 46435
    },
    {
      "entropy": 5.756506013870239,
      "epoch": 4.642175238666467,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003040127491316061,
      "loss": 5.1829,
      "mean_token_accuracy": 0.22162582576274872,
      "num_tokens": 90857531.0,
      "step": 46440
    },
    {
      "entropy": 5.820285224914551,
      "epoch": 4.642675063727696,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030397735760935547,
      "loss": 5.2313,
      "mean_token_accuracy": 0.21076217144727707,
      "num_tokens": 90868164.0,
      "step": 46445
    },
    {
      "entropy": 5.796012020111084,
      "epoch": 4.6431748887889235,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030394196535804205,
      "loss": 5.2695,
      "mean_token_accuracy": 0.2227075830101967,
      "num_tokens": 90878300.0,
      "step": 46450
    },
    {
      "entropy": 5.85450849533081,
      "epoch": 4.643674713850152,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030390657237855606,
      "loss": 5.3454,
      "mean_token_accuracy": 0.21380623281002045,
      "num_tokens": 90889228.0,
      "step": 46455
    },
    {
      "entropy": 5.73270812034607,
      "epoch": 4.644174538911381,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003038711786717883,
      "loss": 5.1298,
      "mean_token_accuracy": 0.23428308069705964,
      "num_tokens": 90899077.0,
      "step": 46460
    },
    {
      "entropy": 5.772976684570312,
      "epoch": 4.64467436397261,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030383578423862914,
      "loss": 5.1965,
      "mean_token_accuracy": 0.22129617184400557,
      "num_tokens": 90907914.0,
      "step": 46465
    },
    {
      "entropy": 5.7946006774902346,
      "epoch": 4.645174189033838,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030380038907996905,
      "loss": 5.2237,
      "mean_token_accuracy": 0.215922711789608,
      "num_tokens": 90918191.0,
      "step": 46470
    },
    {
      "entropy": 5.6226459503173825,
      "epoch": 4.645674014095067,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003037649931966985,
      "loss": 5.046,
      "mean_token_accuracy": 0.24274612367153167,
      "num_tokens": 90929829.0,
      "step": 46475
    },
    {
      "entropy": 5.7085975170135494,
      "epoch": 4.646173839156296,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003037295965897082,
      "loss": 5.1651,
      "mean_token_accuracy": 0.22915525436401368,
      "num_tokens": 90939849.0,
      "step": 46480
    },
    {
      "entropy": 5.825590896606445,
      "epoch": 4.6466736642175235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003036941992598885,
      "loss": 5.2392,
      "mean_token_accuracy": 0.2226724937558174,
      "num_tokens": 90949166.0,
      "step": 46485
    },
    {
      "entropy": 5.812509202957154,
      "epoch": 4.647173489278752,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003036588012081303,
      "loss": 5.1803,
      "mean_token_accuracy": 0.22394855469465255,
      "num_tokens": 90959756.0,
      "step": 46490
    },
    {
      "entropy": 5.771692943572998,
      "epoch": 4.647673314339981,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030362340243532393,
      "loss": 5.2091,
      "mean_token_accuracy": 0.22419802099466324,
      "num_tokens": 90969964.0,
      "step": 46495
    },
    {
      "entropy": 5.829090738296509,
      "epoch": 4.64817313940121,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003035880029423602,
      "loss": 5.2491,
      "mean_token_accuracy": 0.21212011873722075,
      "num_tokens": 90980572.0,
      "step": 46500
    },
    {
      "entropy": 5.841134119033813,
      "epoch": 4.648672964462438,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003035526027301297,
      "loss": 5.2672,
      "mean_token_accuracy": 0.2130598768591881,
      "num_tokens": 90991618.0,
      "step": 46505
    },
    {
      "entropy": 5.787009239196777,
      "epoch": 4.649172789523667,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030351720179952295,
      "loss": 5.1525,
      "mean_token_accuracy": 0.2261809378862381,
      "num_tokens": 91001703.0,
      "step": 46510
    },
    {
      "entropy": 5.856985139846802,
      "epoch": 4.649672614584896,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030348180015143077,
      "loss": 5.3817,
      "mean_token_accuracy": 0.20581942200660705,
      "num_tokens": 91011994.0,
      "step": 46515
    },
    {
      "entropy": 5.823092460632324,
      "epoch": 4.6501724396461235,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003034463977867439,
      "loss": 5.2714,
      "mean_token_accuracy": 0.2190469980239868,
      "num_tokens": 91021756.0,
      "step": 46520
    },
    {
      "entropy": 5.826218891143799,
      "epoch": 4.650672264707352,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030341099470635293,
      "loss": 5.2683,
      "mean_token_accuracy": 0.2163106009364128,
      "num_tokens": 91031865.0,
      "step": 46525
    },
    {
      "entropy": 5.784422063827515,
      "epoch": 4.651172089768581,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030337559091114856,
      "loss": 5.2305,
      "mean_token_accuracy": 0.2174073874950409,
      "num_tokens": 91041644.0,
      "step": 46530
    },
    {
      "entropy": 5.79227933883667,
      "epoch": 4.65167191482981,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030334018640202184,
      "loss": 5.2023,
      "mean_token_accuracy": 0.22180493772029877,
      "num_tokens": 91052109.0,
      "step": 46535
    },
    {
      "entropy": 5.869599771499634,
      "epoch": 4.652171739891038,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030330478117986316,
      "loss": 5.3087,
      "mean_token_accuracy": 0.21037432849407195,
      "num_tokens": 91061681.0,
      "step": 46540
    },
    {
      "entropy": 5.856968879699707,
      "epoch": 4.652671564952267,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030326937524556346,
      "loss": 5.289,
      "mean_token_accuracy": 0.21440234631299973,
      "num_tokens": 91072618.0,
      "step": 46545
    },
    {
      "entropy": 5.7980344772338865,
      "epoch": 4.653171390013496,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030323396860001365,
      "loss": 5.2165,
      "mean_token_accuracy": 0.21631668955087663,
      "num_tokens": 91081212.0,
      "step": 46550
    },
    {
      "entropy": 5.817701387405395,
      "epoch": 4.6536712150747235,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003031985612441044,
      "loss": 5.25,
      "mean_token_accuracy": 0.21040839850902557,
      "num_tokens": 91091487.0,
      "step": 46555
    },
    {
      "entropy": 5.860548686981201,
      "epoch": 4.654171040135952,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030316315317872655,
      "loss": 5.2678,
      "mean_token_accuracy": 0.21292934715747833,
      "num_tokens": 91100547.0,
      "step": 46560
    },
    {
      "entropy": 5.830198431015015,
      "epoch": 4.654670865197181,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003031277444047711,
      "loss": 5.2181,
      "mean_token_accuracy": 0.2145014613866806,
      "num_tokens": 91110161.0,
      "step": 46565
    },
    {
      "entropy": 5.8385758876800535,
      "epoch": 4.65517069025841,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003030923349231288,
      "loss": 5.2654,
      "mean_token_accuracy": 0.2159135326743126,
      "num_tokens": 91119155.0,
      "step": 46570
    },
    {
      "entropy": 5.769393444061279,
      "epoch": 4.655670515319638,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003030569247346906,
      "loss": 5.2415,
      "mean_token_accuracy": 0.21949480324983597,
      "num_tokens": 91128402.0,
      "step": 46575
    },
    {
      "entropy": 5.797098112106323,
      "epoch": 4.656170340380867,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003030215138403474,
      "loss": 5.2509,
      "mean_token_accuracy": 0.21543549448251725,
      "num_tokens": 91137857.0,
      "step": 46580
    },
    {
      "entropy": 5.795443296432495,
      "epoch": 4.656670165442096,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030298610224099,
      "loss": 5.1866,
      "mean_token_accuracy": 0.22608198523521422,
      "num_tokens": 91147053.0,
      "step": 46585
    },
    {
      "entropy": 5.849261283874512,
      "epoch": 4.6571699905033235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030295068993750965,
      "loss": 5.3073,
      "mean_token_accuracy": 0.20876002460718154,
      "num_tokens": 91155929.0,
      "step": 46590
    },
    {
      "entropy": 5.855413246154785,
      "epoch": 4.657669815564552,
      "grad_norm": 1.015625,
      "learning_rate": 0.00030291527693079704,
      "loss": 5.3154,
      "mean_token_accuracy": 0.21253955364227295,
      "num_tokens": 91167774.0,
      "step": 46595
    },
    {
      "entropy": 5.845234346389771,
      "epoch": 4.658169640625781,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030287986322174326,
      "loss": 5.2342,
      "mean_token_accuracy": 0.2159671053290367,
      "num_tokens": 91177685.0,
      "step": 46600
    },
    {
      "entropy": 5.878548908233642,
      "epoch": 4.65866946568701,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030284444881123934,
      "loss": 5.2988,
      "mean_token_accuracy": 0.21327624171972276,
      "num_tokens": 91188021.0,
      "step": 46605
    },
    {
      "entropy": 5.768446350097657,
      "epoch": 4.659169290748238,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003028090337001762,
      "loss": 5.1512,
      "mean_token_accuracy": 0.22010829895734788,
      "num_tokens": 91197681.0,
      "step": 46610
    },
    {
      "entropy": 5.772907781600952,
      "epoch": 4.659669115809467,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030277361788944495,
      "loss": 5.206,
      "mean_token_accuracy": 0.21327273100614547,
      "num_tokens": 91208205.0,
      "step": 46615
    },
    {
      "entropy": 5.874415731430053,
      "epoch": 4.660168940870696,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003027382013799366,
      "loss": 5.284,
      "mean_token_accuracy": 0.2114109143614769,
      "num_tokens": 91218582.0,
      "step": 46620
    },
    {
      "entropy": 5.9087381839752195,
      "epoch": 4.6606687659319235,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030270278417254224,
      "loss": 5.3987,
      "mean_token_accuracy": 0.21074037551879882,
      "num_tokens": 91228650.0,
      "step": 46625
    },
    {
      "entropy": 5.836193513870239,
      "epoch": 4.661168590993152,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000302667366268153,
      "loss": 5.301,
      "mean_token_accuracy": 0.21231211870908737,
      "num_tokens": 91238836.0,
      "step": 46630
    },
    {
      "entropy": 5.848719882965088,
      "epoch": 4.661668416054381,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003026319476676599,
      "loss": 5.2321,
      "mean_token_accuracy": 0.2195258170366287,
      "num_tokens": 91249890.0,
      "step": 46635
    },
    {
      "entropy": 5.7632488250732425,
      "epoch": 4.66216824111561,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030259652837195413,
      "loss": 5.2121,
      "mean_token_accuracy": 0.2230358824133873,
      "num_tokens": 91259052.0,
      "step": 46640
    },
    {
      "entropy": 5.866198205947876,
      "epoch": 4.662668066176838,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030256110838192677,
      "loss": 5.3025,
      "mean_token_accuracy": 0.21282820850610734,
      "num_tokens": 91268899.0,
      "step": 46645
    },
    {
      "entropy": 5.883970022201538,
      "epoch": 4.663167891238067,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003025256876984691,
      "loss": 5.2909,
      "mean_token_accuracy": 0.2088240921497345,
      "num_tokens": 91277769.0,
      "step": 46650
    },
    {
      "entropy": 5.85172176361084,
      "epoch": 4.663667716299296,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030249026632247214,
      "loss": 5.3127,
      "mean_token_accuracy": 0.21484708189964294,
      "num_tokens": 91287443.0,
      "step": 46655
    },
    {
      "entropy": 5.900146007537842,
      "epoch": 4.6641675413605235,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030245484425482714,
      "loss": 5.2955,
      "mean_token_accuracy": 0.21289748549461365,
      "num_tokens": 91296783.0,
      "step": 46660
    },
    {
      "entropy": 5.777556085586548,
      "epoch": 4.664667366421752,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030241942149642533,
      "loss": 5.1157,
      "mean_token_accuracy": 0.22973484247922898,
      "num_tokens": 91306564.0,
      "step": 46665
    },
    {
      "entropy": 5.833661937713623,
      "epoch": 4.665167191482981,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000302383998048158,
      "loss": 5.2399,
      "mean_token_accuracy": 0.221716046333313,
      "num_tokens": 91316012.0,
      "step": 46670
    },
    {
      "entropy": 5.681146717071533,
      "epoch": 4.66566701654421,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030234857391091627,
      "loss": 5.104,
      "mean_token_accuracy": 0.2343247801065445,
      "num_tokens": 91325755.0,
      "step": 46675
    },
    {
      "entropy": 5.791825628280639,
      "epoch": 4.666166841605438,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030231314908559144,
      "loss": 5.2798,
      "mean_token_accuracy": 0.2199803724884987,
      "num_tokens": 91334740.0,
      "step": 46680
    },
    {
      "entropy": 5.785341596603393,
      "epoch": 4.666666666666667,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003022777235730749,
      "loss": 5.2279,
      "mean_token_accuracy": 0.21842733323574065,
      "num_tokens": 91344382.0,
      "step": 46685
    },
    {
      "entropy": 5.942310762405396,
      "epoch": 4.667166491727896,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030224229737425774,
      "loss": 5.3912,
      "mean_token_accuracy": 0.20050758123397827,
      "num_tokens": 91354665.0,
      "step": 46690
    },
    {
      "entropy": 5.868501949310303,
      "epoch": 4.6676663167891235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003022068704900314,
      "loss": 5.2415,
      "mean_token_accuracy": 0.21640914529561997,
      "num_tokens": 91364265.0,
      "step": 46695
    },
    {
      "entropy": 5.873986101150512,
      "epoch": 4.668166141850352,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003021714429212872,
      "loss": 5.3248,
      "mean_token_accuracy": 0.21136102974414825,
      "num_tokens": 91373602.0,
      "step": 46700
    },
    {
      "entropy": 5.75909013748169,
      "epoch": 4.668665966911581,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030213601466891663,
      "loss": 5.1239,
      "mean_token_accuracy": 0.22198963761329651,
      "num_tokens": 91383676.0,
      "step": 46705
    },
    {
      "entropy": 5.690815591812134,
      "epoch": 4.66916579197281,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003021005857338107,
      "loss": 5.1374,
      "mean_token_accuracy": 0.22277835756540298,
      "num_tokens": 91393999.0,
      "step": 46710
    },
    {
      "entropy": 5.821980619430542,
      "epoch": 4.669665617034038,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003020651561168612,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21315957605838776,
      "num_tokens": 91403628.0,
      "step": 46715
    },
    {
      "entropy": 5.795641374588013,
      "epoch": 4.670165442095267,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030202972581895923,
      "loss": 5.2447,
      "mean_token_accuracy": 0.21893989741802217,
      "num_tokens": 91414009.0,
      "step": 46720
    },
    {
      "entropy": 5.8355237483978275,
      "epoch": 4.670665267156496,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003019942948409963,
      "loss": 5.222,
      "mean_token_accuracy": 0.22087506502866744,
      "num_tokens": 91423079.0,
      "step": 46725
    },
    {
      "entropy": 5.886212587356567,
      "epoch": 4.6711650922177235,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030195886318386397,
      "loss": 5.32,
      "mean_token_accuracy": 0.2061651274561882,
      "num_tokens": 91432548.0,
      "step": 46730
    },
    {
      "entropy": 5.786844873428345,
      "epoch": 4.671664917278952,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030192343084845343,
      "loss": 5.2233,
      "mean_token_accuracy": 0.2132358655333519,
      "num_tokens": 91442449.0,
      "step": 46735
    },
    {
      "entropy": 5.766799545288086,
      "epoch": 4.672164742340181,
      "grad_norm": 1.125,
      "learning_rate": 0.0003018879978356564,
      "loss": 5.1383,
      "mean_token_accuracy": 0.22198978811502457,
      "num_tokens": 91452090.0,
      "step": 46740
    },
    {
      "entropy": 5.7644147872924805,
      "epoch": 4.67266456740141,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030185256414636425,
      "loss": 5.1937,
      "mean_token_accuracy": 0.22361682802438737,
      "num_tokens": 91461750.0,
      "step": 46745
    },
    {
      "entropy": 5.787861347198486,
      "epoch": 4.673164392462638,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030181712978146846,
      "loss": 5.3336,
      "mean_token_accuracy": 0.2041833892464638,
      "num_tokens": 91471729.0,
      "step": 46750
    },
    {
      "entropy": 5.8308793067932125,
      "epoch": 4.673664217523867,
      "grad_norm": 1.125,
      "learning_rate": 0.00030178169474186055,
      "loss": 5.2879,
      "mean_token_accuracy": 0.20709017664194107,
      "num_tokens": 91481362.0,
      "step": 46755
    },
    {
      "entropy": 5.829749250411988,
      "epoch": 4.674164042585096,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030174625902843215,
      "loss": 5.1729,
      "mean_token_accuracy": 0.22471379935741426,
      "num_tokens": 91491136.0,
      "step": 46760
    },
    {
      "entropy": 5.720038032531738,
      "epoch": 4.6746638676463235,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003017108226420748,
      "loss": 5.1626,
      "mean_token_accuracy": 0.22559285610914231,
      "num_tokens": 91501474.0,
      "step": 46765
    },
    {
      "entropy": 5.852313995361328,
      "epoch": 4.675163692707552,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003016753855836799,
      "loss": 5.2431,
      "mean_token_accuracy": 0.21493891328573228,
      "num_tokens": 91511555.0,
      "step": 46770
    },
    {
      "entropy": 5.841338348388672,
      "epoch": 4.675663517768781,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003016399478541393,
      "loss": 5.2778,
      "mean_token_accuracy": 0.21032430082559586,
      "num_tokens": 91521317.0,
      "step": 46775
    },
    {
      "entropy": 5.777031278610229,
      "epoch": 4.67616334283001,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030160450945434435,
      "loss": 5.1916,
      "mean_token_accuracy": 0.22123087644577027,
      "num_tokens": 91531448.0,
      "step": 46780
    },
    {
      "entropy": 5.805846118927002,
      "epoch": 4.676663167891238,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003015690703851869,
      "loss": 5.2185,
      "mean_token_accuracy": 0.2225275531411171,
      "num_tokens": 91542089.0,
      "step": 46785
    },
    {
      "entropy": 5.780720949172974,
      "epoch": 4.677162992952467,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003015336306475584,
      "loss": 5.1913,
      "mean_token_accuracy": 0.22244220077991486,
      "num_tokens": 91551283.0,
      "step": 46790
    },
    {
      "entropy": 5.8978136539459225,
      "epoch": 4.677662818013696,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003014981902423506,
      "loss": 5.3201,
      "mean_token_accuracy": 0.21062080413103104,
      "num_tokens": 91560460.0,
      "step": 46795
    },
    {
      "entropy": 5.877269268035889,
      "epoch": 4.6781626430749235,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003014627491704551,
      "loss": 5.3442,
      "mean_token_accuracy": 0.21448458433151246,
      "num_tokens": 91570768.0,
      "step": 46800
    },
    {
      "entropy": 5.751779842376709,
      "epoch": 4.678662468136152,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003014273074327637,
      "loss": 5.1515,
      "mean_token_accuracy": 0.22279605716466905,
      "num_tokens": 91579394.0,
      "step": 46805
    },
    {
      "entropy": 5.746805953979492,
      "epoch": 4.679162293197381,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003013918650301681,
      "loss": 5.1041,
      "mean_token_accuracy": 0.2282474622130394,
      "num_tokens": 91588377.0,
      "step": 46810
    },
    {
      "entropy": 5.753546190261841,
      "epoch": 4.67966211825861,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030135642196355987,
      "loss": 5.168,
      "mean_token_accuracy": 0.21816461384296418,
      "num_tokens": 91596741.0,
      "step": 46815
    },
    {
      "entropy": 5.751318836212159,
      "epoch": 4.680161943319838,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003013209782338309,
      "loss": 5.2809,
      "mean_token_accuracy": 0.2185034468770027,
      "num_tokens": 91605695.0,
      "step": 46820
    },
    {
      "entropy": 5.848953485488892,
      "epoch": 4.680661768381067,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003012855338418729,
      "loss": 5.2329,
      "mean_token_accuracy": 0.2231995239853859,
      "num_tokens": 91615810.0,
      "step": 46825
    },
    {
      "entropy": 5.908376312255859,
      "epoch": 4.681161593442296,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003012500887885775,
      "loss": 5.2301,
      "mean_token_accuracy": 0.22157563418149948,
      "num_tokens": 91625697.0,
      "step": 46830
    },
    {
      "entropy": 5.821082830429077,
      "epoch": 4.6816614185035235,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030121464307483677,
      "loss": 5.2687,
      "mean_token_accuracy": 0.21329160779714584,
      "num_tokens": 91635825.0,
      "step": 46835
    },
    {
      "entropy": 5.798488187789917,
      "epoch": 4.682161243564752,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003011791967015423,
      "loss": 5.2682,
      "mean_token_accuracy": 0.22480989694595338,
      "num_tokens": 91646096.0,
      "step": 46840
    },
    {
      "entropy": 5.978400325775146,
      "epoch": 4.682661068625981,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030114374966958604,
      "loss": 5.3893,
      "mean_token_accuracy": 0.20307547301054002,
      "num_tokens": 91655744.0,
      "step": 46845
    },
    {
      "entropy": 5.861075592041016,
      "epoch": 4.68316089368721,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003011083019798597,
      "loss": 5.1919,
      "mean_token_accuracy": 0.22401715964078903,
      "num_tokens": 91664365.0,
      "step": 46850
    },
    {
      "entropy": 5.782629489898682,
      "epoch": 4.683660718748438,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003010728536332552,
      "loss": 5.221,
      "mean_token_accuracy": 0.2182271122932434,
      "num_tokens": 91674525.0,
      "step": 46855
    },
    {
      "entropy": 5.72868332862854,
      "epoch": 4.684160543809667,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003010374046306644,
      "loss": 5.1833,
      "mean_token_accuracy": 0.22534913569688797,
      "num_tokens": 91683935.0,
      "step": 46860
    },
    {
      "entropy": 5.707022857666016,
      "epoch": 4.684660368870896,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003010019549729793,
      "loss": 5.1811,
      "mean_token_accuracy": 0.22220584154129028,
      "num_tokens": 91694239.0,
      "step": 46865
    },
    {
      "entropy": 5.805554437637329,
      "epoch": 4.6851601939321235,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003009665046610916,
      "loss": 5.1557,
      "mean_token_accuracy": 0.22759063839912413,
      "num_tokens": 91703624.0,
      "step": 46870
    },
    {
      "entropy": 5.672909307479858,
      "epoch": 4.685660018993352,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003009310536958934,
      "loss": 5.1163,
      "mean_token_accuracy": 0.23785228580236434,
      "num_tokens": 91712744.0,
      "step": 46875
    },
    {
      "entropy": 5.689020776748658,
      "epoch": 4.686159844054581,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030089560207827653,
      "loss": 5.1229,
      "mean_token_accuracy": 0.23316417783498763,
      "num_tokens": 91722171.0,
      "step": 46880
    },
    {
      "entropy": 5.833604860305786,
      "epoch": 4.68665966911581,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030086014980913294,
      "loss": 5.2268,
      "mean_token_accuracy": 0.21609679162502288,
      "num_tokens": 91731318.0,
      "step": 46885
    },
    {
      "entropy": 5.916919279098511,
      "epoch": 4.687159494177038,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030082469688935465,
      "loss": 5.38,
      "mean_token_accuracy": 0.20322090834379197,
      "num_tokens": 91740220.0,
      "step": 46890
    },
    {
      "entropy": 5.8227746963500975,
      "epoch": 4.687659319238267,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003007892433198336,
      "loss": 5.1908,
      "mean_token_accuracy": 0.2256064385175705,
      "num_tokens": 91749626.0,
      "step": 46895
    },
    {
      "entropy": 5.825919818878174,
      "epoch": 4.688159144299496,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000300753789101462,
      "loss": 5.2376,
      "mean_token_accuracy": 0.2142983630299568,
      "num_tokens": 91758212.0,
      "step": 46900
    },
    {
      "entropy": 5.81083116531372,
      "epoch": 4.6886589693607235,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003007183342351315,
      "loss": 5.2668,
      "mean_token_accuracy": 0.21220327019691468,
      "num_tokens": 91768416.0,
      "step": 46905
    },
    {
      "entropy": 5.788816022872925,
      "epoch": 4.689158794421952,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003006828787217344,
      "loss": 5.236,
      "mean_token_accuracy": 0.21919914782047273,
      "num_tokens": 91777559.0,
      "step": 46910
    },
    {
      "entropy": 5.804159784317017,
      "epoch": 4.689658619483181,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030064742256216264,
      "loss": 5.2296,
      "mean_token_accuracy": 0.21839151680469512,
      "num_tokens": 91787478.0,
      "step": 46915
    },
    {
      "entropy": 5.8272504806518555,
      "epoch": 4.6901584445444096,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003006119657573084,
      "loss": 5.2044,
      "mean_token_accuracy": 0.21802067309617995,
      "num_tokens": 91796437.0,
      "step": 46920
    },
    {
      "entropy": 5.837101411819458,
      "epoch": 4.690658269605638,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030057650830806365,
      "loss": 5.2514,
      "mean_token_accuracy": 0.22200120538473128,
      "num_tokens": 91806270.0,
      "step": 46925
    },
    {
      "entropy": 5.853644704818725,
      "epoch": 4.691158094666867,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030054105021532054,
      "loss": 5.2736,
      "mean_token_accuracy": 0.20885480493307113,
      "num_tokens": 91815660.0,
      "step": 46930
    },
    {
      "entropy": 5.748585987091064,
      "epoch": 4.691657919728096,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030050559147997113,
      "loss": 5.2207,
      "mean_token_accuracy": 0.219575035572052,
      "num_tokens": 91825405.0,
      "step": 46935
    },
    {
      "entropy": 5.823184061050415,
      "epoch": 4.6921577447893235,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003004701321029076,
      "loss": 5.2666,
      "mean_token_accuracy": 0.21886527091264724,
      "num_tokens": 91836028.0,
      "step": 46940
    },
    {
      "entropy": 5.789985466003418,
      "epoch": 4.692657569850552,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030043467208502216,
      "loss": 5.2338,
      "mean_token_accuracy": 0.21757513284683228,
      "num_tokens": 91845607.0,
      "step": 46945
    },
    {
      "entropy": 5.819101476669312,
      "epoch": 4.693157394911781,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000300399211427207,
      "loss": 5.2828,
      "mean_token_accuracy": 0.206280180811882,
      "num_tokens": 91855174.0,
      "step": 46950
    },
    {
      "entropy": 5.8946319103240965,
      "epoch": 4.6936572199730096,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030036375013035404,
      "loss": 5.2779,
      "mean_token_accuracy": 0.2219721183180809,
      "num_tokens": 91864588.0,
      "step": 46955
    },
    {
      "entropy": 5.788314914703369,
      "epoch": 4.694157045034238,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030032828819535575,
      "loss": 5.1583,
      "mean_token_accuracy": 0.22359221279621125,
      "num_tokens": 91873895.0,
      "step": 46960
    },
    {
      "entropy": 5.711716079711914,
      "epoch": 4.694656870095467,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030029282562310424,
      "loss": 5.2167,
      "mean_token_accuracy": 0.21880985647439957,
      "num_tokens": 91884615.0,
      "step": 46965
    },
    {
      "entropy": 5.785926437377929,
      "epoch": 4.695156695156696,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003002573624144917,
      "loss": 5.3053,
      "mean_token_accuracy": 0.21907673627138138,
      "num_tokens": 91894198.0,
      "step": 46970
    },
    {
      "entropy": 5.914644765853882,
      "epoch": 4.6956565202179235,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030022189857041045,
      "loss": 5.3228,
      "mean_token_accuracy": 0.20370325297117234,
      "num_tokens": 91903878.0,
      "step": 46975
    },
    {
      "entropy": 5.796762704849243,
      "epoch": 4.696156345279152,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003001864340917527,
      "loss": 5.1494,
      "mean_token_accuracy": 0.22593581676483154,
      "num_tokens": 91914097.0,
      "step": 46980
    },
    {
      "entropy": 5.839614486694336,
      "epoch": 4.696656170340381,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003001509689794108,
      "loss": 5.248,
      "mean_token_accuracy": 0.21485516875982286,
      "num_tokens": 91923166.0,
      "step": 46985
    },
    {
      "entropy": 5.739768552780151,
      "epoch": 4.6971559954016096,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030011550323427694,
      "loss": 5.2225,
      "mean_token_accuracy": 0.22516253888607024,
      "num_tokens": 91933887.0,
      "step": 46990
    },
    {
      "entropy": 5.893611049652099,
      "epoch": 4.697655820462838,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030008003685724353,
      "loss": 5.3024,
      "mean_token_accuracy": 0.21673304736614227,
      "num_tokens": 91944072.0,
      "step": 46995
    },
    {
      "entropy": 5.80093731880188,
      "epoch": 4.698155645524067,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030004456984920283,
      "loss": 5.219,
      "mean_token_accuracy": 0.2155776545405388,
      "num_tokens": 91953981.0,
      "step": 47000
    },
    {
      "entropy": 5.805992174148559,
      "epoch": 4.698655470585296,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003000091022110472,
      "loss": 5.1881,
      "mean_token_accuracy": 0.2210484817624092,
      "num_tokens": 91963551.0,
      "step": 47005
    },
    {
      "entropy": 5.829124164581299,
      "epoch": 4.6991552956465235,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002999736339436689,
      "loss": 5.2786,
      "mean_token_accuracy": 0.2142673283815384,
      "num_tokens": 91972237.0,
      "step": 47010
    },
    {
      "entropy": 5.823067951202392,
      "epoch": 4.699655120707752,
      "grad_norm": 1.125,
      "learning_rate": 0.00029993816504796057,
      "loss": 5.2597,
      "mean_token_accuracy": 0.2092686414718628,
      "num_tokens": 91982574.0,
      "step": 47015
    },
    {
      "entropy": 5.887418651580811,
      "epoch": 4.700154945768981,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002999026955248143,
      "loss": 5.2724,
      "mean_token_accuracy": 0.2120761454105377,
      "num_tokens": 91992328.0,
      "step": 47020
    },
    {
      "entropy": 5.797320985794068,
      "epoch": 4.7006547708302096,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002998672253751227,
      "loss": 5.2422,
      "mean_token_accuracy": 0.21376658380031585,
      "num_tokens": 92001094.0,
      "step": 47025
    },
    {
      "entropy": 5.812450838088989,
      "epoch": 4.701154595891438,
      "grad_norm": 1.25,
      "learning_rate": 0.00029983175459977803,
      "loss": 5.2779,
      "mean_token_accuracy": 0.2148577705025673,
      "num_tokens": 92010212.0,
      "step": 47030
    },
    {
      "entropy": 5.759663391113281,
      "epoch": 4.701654420952667,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002997962831996728,
      "loss": 5.1845,
      "mean_token_accuracy": 0.22205598801374435,
      "num_tokens": 92020148.0,
      "step": 47035
    },
    {
      "entropy": 5.853768539428711,
      "epoch": 4.702154246013895,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029976081117569957,
      "loss": 5.2923,
      "mean_token_accuracy": 0.21376391798257827,
      "num_tokens": 92029056.0,
      "step": 47040
    },
    {
      "entropy": 5.757015752792358,
      "epoch": 4.7026540710751235,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029972533852875064,
      "loss": 5.2195,
      "mean_token_accuracy": 0.21772062182426452,
      "num_tokens": 92038978.0,
      "step": 47045
    },
    {
      "entropy": 5.785035610198975,
      "epoch": 4.703153896136352,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029968986525971854,
      "loss": 5.1646,
      "mean_token_accuracy": 0.22502707540988923,
      "num_tokens": 92047962.0,
      "step": 47050
    },
    {
      "entropy": 5.862910413742066,
      "epoch": 4.703653721197581,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002996543913694958,
      "loss": 5.3116,
      "mean_token_accuracy": 0.21019013971090317,
      "num_tokens": 92058226.0,
      "step": 47055
    },
    {
      "entropy": 5.8089296340942385,
      "epoch": 4.7041535462588095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029961891685897484,
      "loss": 5.2856,
      "mean_token_accuracy": 0.2130971372127533,
      "num_tokens": 92068942.0,
      "step": 47060
    },
    {
      "entropy": 5.798260068893432,
      "epoch": 4.704653371320038,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002995834417290484,
      "loss": 5.2329,
      "mean_token_accuracy": 0.22057520002126693,
      "num_tokens": 92078777.0,
      "step": 47065
    },
    {
      "entropy": 5.8011627197265625,
      "epoch": 4.705153196381267,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002995479659806088,
      "loss": 5.3131,
      "mean_token_accuracy": 0.2155759036540985,
      "num_tokens": 92087750.0,
      "step": 47070
    },
    {
      "entropy": 5.7782995223999025,
      "epoch": 4.705653021442495,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029951248961454876,
      "loss": 5.2063,
      "mean_token_accuracy": 0.2134034276008606,
      "num_tokens": 92097281.0,
      "step": 47075
    },
    {
      "entropy": 5.833664226531982,
      "epoch": 4.7061528465037235,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029947701263176074,
      "loss": 5.3167,
      "mean_token_accuracy": 0.20996909588575363,
      "num_tokens": 92107419.0,
      "step": 47080
    },
    {
      "entropy": 5.775703239440918,
      "epoch": 4.706652671564952,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002994415350331373,
      "loss": 5.1564,
      "mean_token_accuracy": 0.2200787752866745,
      "num_tokens": 92117662.0,
      "step": 47085
    },
    {
      "entropy": 5.874575853347778,
      "epoch": 4.707152496626181,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002994060568195713,
      "loss": 5.2683,
      "mean_token_accuracy": 0.21260226666927337,
      "num_tokens": 92128058.0,
      "step": 47090
    },
    {
      "entropy": 5.739095306396484,
      "epoch": 4.7076523216874095,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002993705779919551,
      "loss": 5.1475,
      "mean_token_accuracy": 0.22976538091897963,
      "num_tokens": 92138620.0,
      "step": 47095
    },
    {
      "entropy": 5.784007406234741,
      "epoch": 4.708152146748638,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002993350985511814,
      "loss": 5.2917,
      "mean_token_accuracy": 0.21469557881355286,
      "num_tokens": 92148976.0,
      "step": 47100
    },
    {
      "entropy": 5.787513589859008,
      "epoch": 4.708651971809866,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002992996184981429,
      "loss": 5.2776,
      "mean_token_accuracy": 0.21068612337112427,
      "num_tokens": 92159148.0,
      "step": 47105
    },
    {
      "entropy": 5.781657743453979,
      "epoch": 4.709151796871095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029926413783373237,
      "loss": 5.2145,
      "mean_token_accuracy": 0.22362405806779861,
      "num_tokens": 92168679.0,
      "step": 47110
    },
    {
      "entropy": 5.839964580535889,
      "epoch": 4.7096516219323235,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002992286565588422,
      "loss": 5.2107,
      "mean_token_accuracy": 0.22050308883190156,
      "num_tokens": 92177713.0,
      "step": 47115
    },
    {
      "entropy": 5.79231276512146,
      "epoch": 4.710151446993552,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002991931746743653,
      "loss": 5.1948,
      "mean_token_accuracy": 0.22142457365989685,
      "num_tokens": 92187516.0,
      "step": 47120
    },
    {
      "entropy": 5.800592613220215,
      "epoch": 4.710651272054781,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002991576921811945,
      "loss": 5.2675,
      "mean_token_accuracy": 0.21688567996025085,
      "num_tokens": 92197170.0,
      "step": 47125
    },
    {
      "entropy": 5.846626853942871,
      "epoch": 4.7111510971160095,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029912220908022224,
      "loss": 5.3426,
      "mean_token_accuracy": 0.2116747424006462,
      "num_tokens": 92206913.0,
      "step": 47130
    },
    {
      "entropy": 5.889289665222168,
      "epoch": 4.711650922177238,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029908672537234146,
      "loss": 5.2836,
      "mean_token_accuracy": 0.20967742502689363,
      "num_tokens": 92216443.0,
      "step": 47135
    },
    {
      "entropy": 5.871772050857544,
      "epoch": 4.712150747238466,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029905124105844485,
      "loss": 5.2979,
      "mean_token_accuracy": 0.2121924117207527,
      "num_tokens": 92226570.0,
      "step": 47140
    },
    {
      "entropy": 5.727559614181518,
      "epoch": 4.712650572299695,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002990157561394252,
      "loss": 5.162,
      "mean_token_accuracy": 0.23068303763866424,
      "num_tokens": 92236218.0,
      "step": 47145
    },
    {
      "entropy": 5.8150218486785885,
      "epoch": 4.7131503973609234,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029898027061617526,
      "loss": 5.3041,
      "mean_token_accuracy": 0.208733494579792,
      "num_tokens": 92246329.0,
      "step": 47150
    },
    {
      "entropy": 5.819461297988892,
      "epoch": 4.713650222422152,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002989447844895879,
      "loss": 5.2589,
      "mean_token_accuracy": 0.2220943510532379,
      "num_tokens": 92256048.0,
      "step": 47155
    },
    {
      "entropy": 5.818028736114502,
      "epoch": 4.714150047483381,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000298909297760556,
      "loss": 5.1757,
      "mean_token_accuracy": 0.23067831099033356,
      "num_tokens": 92265557.0,
      "step": 47160
    },
    {
      "entropy": 5.891064167022705,
      "epoch": 4.7146498725446095,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029887381042997227,
      "loss": 5.3416,
      "mean_token_accuracy": 0.2052633911371231,
      "num_tokens": 92275852.0,
      "step": 47165
    },
    {
      "entropy": 5.756803226470947,
      "epoch": 4.715149697605838,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002988383224987296,
      "loss": 5.1726,
      "mean_token_accuracy": 0.2278142675757408,
      "num_tokens": 92285862.0,
      "step": 47170
    },
    {
      "entropy": 5.886232233047485,
      "epoch": 4.715649522667066,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029880283396772087,
      "loss": 5.2911,
      "mean_token_accuracy": 0.20709967911243438,
      "num_tokens": 92295341.0,
      "step": 47175
    },
    {
      "entropy": 5.653800821304321,
      "epoch": 4.716149347728295,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029876734483783897,
      "loss": 5.0934,
      "mean_token_accuracy": 0.23500171303749084,
      "num_tokens": 92304917.0,
      "step": 47180
    },
    {
      "entropy": 5.839961338043213,
      "epoch": 4.7166491727895234,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029873185510997686,
      "loss": 5.1778,
      "mean_token_accuracy": 0.22257585525512696,
      "num_tokens": 92314780.0,
      "step": 47185
    },
    {
      "entropy": 5.830743885040283,
      "epoch": 4.717148997850752,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029869636478502726,
      "loss": 5.2652,
      "mean_token_accuracy": 0.21399615406990052,
      "num_tokens": 92324429.0,
      "step": 47190
    },
    {
      "entropy": 5.81647891998291,
      "epoch": 4.717648822911981,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029866087386388334,
      "loss": 5.3061,
      "mean_token_accuracy": 0.20983860790729522,
      "num_tokens": 92333973.0,
      "step": 47195
    },
    {
      "entropy": 5.884367752075195,
      "epoch": 4.7181486479732095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029862538234743784,
      "loss": 5.2681,
      "mean_token_accuracy": 0.21610599309206008,
      "num_tokens": 92344075.0,
      "step": 47200
    },
    {
      "entropy": 5.783160543441772,
      "epoch": 4.718648473034438,
      "grad_norm": 1.25,
      "learning_rate": 0.00029858989023658376,
      "loss": 5.2735,
      "mean_token_accuracy": 0.21379240304231645,
      "num_tokens": 92354056.0,
      "step": 47205
    },
    {
      "entropy": 5.81106276512146,
      "epoch": 4.719148298095666,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002985543975322142,
      "loss": 5.3253,
      "mean_token_accuracy": 0.2085249051451683,
      "num_tokens": 92363460.0,
      "step": 47210
    },
    {
      "entropy": 5.816472148895263,
      "epoch": 4.719648123156895,
      "grad_norm": 1.171875,
      "learning_rate": 0.000298518904235222,
      "loss": 5.2524,
      "mean_token_accuracy": 0.21917722523212432,
      "num_tokens": 92373294.0,
      "step": 47215
    },
    {
      "entropy": 5.840943050384522,
      "epoch": 4.7201479482181234,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029848341034650026,
      "loss": 5.2844,
      "mean_token_accuracy": 0.21810318380594254,
      "num_tokens": 92384385.0,
      "step": 47220
    },
    {
      "entropy": 5.858893394470215,
      "epoch": 4.720647773279352,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000298447915866942,
      "loss": 5.274,
      "mean_token_accuracy": 0.214875628054142,
      "num_tokens": 92394326.0,
      "step": 47225
    },
    {
      "entropy": 5.778936767578125,
      "epoch": 4.721147598340581,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002984124207974402,
      "loss": 5.2303,
      "mean_token_accuracy": 0.2244235470890999,
      "num_tokens": 92403844.0,
      "step": 47230
    },
    {
      "entropy": 5.750623941421509,
      "epoch": 4.7216474234018095,
      "grad_norm": 1.25,
      "learning_rate": 0.00029837692513888785,
      "loss": 5.1931,
      "mean_token_accuracy": 0.2251914843916893,
      "num_tokens": 92413204.0,
      "step": 47235
    },
    {
      "entropy": 5.879278182983398,
      "epoch": 4.722147248463038,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002983414288921781,
      "loss": 5.2794,
      "mean_token_accuracy": 0.2135940596461296,
      "num_tokens": 92421989.0,
      "step": 47240
    },
    {
      "entropy": 5.854074764251709,
      "epoch": 4.722647073524266,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029830593205820407,
      "loss": 5.2294,
      "mean_token_accuracy": 0.22684367299079894,
      "num_tokens": 92433231.0,
      "step": 47245
    },
    {
      "entropy": 5.806501483917236,
      "epoch": 4.723146898585495,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002982704346378587,
      "loss": 5.2035,
      "mean_token_accuracy": 0.22453518956899643,
      "num_tokens": 92443063.0,
      "step": 47250
    },
    {
      "entropy": 5.860188817977905,
      "epoch": 4.7236467236467234,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029823493663203515,
      "loss": 5.3485,
      "mean_token_accuracy": 0.2152205914258957,
      "num_tokens": 92453410.0,
      "step": 47255
    },
    {
      "entropy": 5.811329889297485,
      "epoch": 4.724146548707952,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002981994380416267,
      "loss": 5.18,
      "mean_token_accuracy": 0.22200072556734085,
      "num_tokens": 92461523.0,
      "step": 47260
    },
    {
      "entropy": 5.857751941680908,
      "epoch": 4.724646373769181,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002981639388675263,
      "loss": 5.3086,
      "mean_token_accuracy": 0.21207997798919678,
      "num_tokens": 92469959.0,
      "step": 47265
    },
    {
      "entropy": 5.8703789710998535,
      "epoch": 4.7251461988304095,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002981284391106271,
      "loss": 5.3662,
      "mean_token_accuracy": 0.21107104569673538,
      "num_tokens": 92480271.0,
      "step": 47270
    },
    {
      "entropy": 5.814873933792114,
      "epoch": 4.725646023891638,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002980929387718224,
      "loss": 5.2545,
      "mean_token_accuracy": 0.22243065387010574,
      "num_tokens": 92490440.0,
      "step": 47275
    },
    {
      "entropy": 5.844959115982055,
      "epoch": 4.726145848952866,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002980574378520053,
      "loss": 5.2279,
      "mean_token_accuracy": 0.22479980885982515,
      "num_tokens": 92501935.0,
      "step": 47280
    },
    {
      "entropy": 5.872433137893677,
      "epoch": 4.726645674014095,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029802193635206895,
      "loss": 5.2269,
      "mean_token_accuracy": 0.2198603093624115,
      "num_tokens": 92511288.0,
      "step": 47285
    },
    {
      "entropy": 5.755534887313843,
      "epoch": 4.727145499075323,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002979864342729067,
      "loss": 5.1705,
      "mean_token_accuracy": 0.21839484721422195,
      "num_tokens": 92520621.0,
      "step": 47290
    },
    {
      "entropy": 5.821049308776855,
      "epoch": 4.727645324136552,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029795093161541156,
      "loss": 5.2825,
      "mean_token_accuracy": 0.2184424638748169,
      "num_tokens": 92530294.0,
      "step": 47295
    },
    {
      "entropy": 5.878725242614746,
      "epoch": 4.728145149197781,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000297915428380477,
      "loss": 5.2907,
      "mean_token_accuracy": 0.21730652451515198,
      "num_tokens": 92540858.0,
      "step": 47300
    },
    {
      "entropy": 5.877109575271606,
      "epoch": 4.7286449742590095,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029787992456899604,
      "loss": 5.2785,
      "mean_token_accuracy": 0.21660797595977782,
      "num_tokens": 92551118.0,
      "step": 47305
    },
    {
      "entropy": 5.789344930648804,
      "epoch": 4.729144799320238,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002978444201818621,
      "loss": 5.2537,
      "mean_token_accuracy": 0.21616043895483017,
      "num_tokens": 92560479.0,
      "step": 47310
    },
    {
      "entropy": 5.802843332290649,
      "epoch": 4.729644624381466,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029780891521996843,
      "loss": 5.2332,
      "mean_token_accuracy": 0.2195326566696167,
      "num_tokens": 92569489.0,
      "step": 47315
    },
    {
      "entropy": 5.8620078563690186,
      "epoch": 4.730144449442695,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002977734096842084,
      "loss": 5.306,
      "mean_token_accuracy": 0.2073813185095787,
      "num_tokens": 92580270.0,
      "step": 47320
    },
    {
      "entropy": 5.737445878982544,
      "epoch": 4.730644274503923,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029773790357547514,
      "loss": 5.1467,
      "mean_token_accuracy": 0.22636569440364837,
      "num_tokens": 92588967.0,
      "step": 47325
    },
    {
      "entropy": 5.764883804321289,
      "epoch": 4.731144099565152,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029770239689466216,
      "loss": 5.1785,
      "mean_token_accuracy": 0.21942876428365707,
      "num_tokens": 92598163.0,
      "step": 47330
    },
    {
      "entropy": 5.885549068450928,
      "epoch": 4.731643924626381,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029766688964266263,
      "loss": 5.2447,
      "mean_token_accuracy": 0.22030724734067916,
      "num_tokens": 92607737.0,
      "step": 47335
    },
    {
      "entropy": 5.857285976409912,
      "epoch": 4.7321437496876095,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002976313818203701,
      "loss": 5.3444,
      "mean_token_accuracy": 0.20687295198440553,
      "num_tokens": 92618517.0,
      "step": 47340
    },
    {
      "entropy": 5.816691923141479,
      "epoch": 4.732643574748838,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002975958734286777,
      "loss": 5.3087,
      "mean_token_accuracy": 0.2076929211616516,
      "num_tokens": 92628179.0,
      "step": 47345
    },
    {
      "entropy": 5.828852033615112,
      "epoch": 4.733143399810066,
      "grad_norm": 1.171875,
      "learning_rate": 0.000297560364468479,
      "loss": 5.2822,
      "mean_token_accuracy": 0.21362409889698028,
      "num_tokens": 92638600.0,
      "step": 47350
    },
    {
      "entropy": 5.810273742675781,
      "epoch": 4.733643224871295,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002975248549406673,
      "loss": 5.2286,
      "mean_token_accuracy": 0.21706045866012574,
      "num_tokens": 92649067.0,
      "step": 47355
    },
    {
      "entropy": 5.788121843338013,
      "epoch": 4.734143049932523,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002974893448461361,
      "loss": 5.2091,
      "mean_token_accuracy": 0.22250732481479646,
      "num_tokens": 92658895.0,
      "step": 47360
    },
    {
      "entropy": 5.730167245864868,
      "epoch": 4.734642874993752,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029745383418577876,
      "loss": 5.1977,
      "mean_token_accuracy": 0.2202197402715683,
      "num_tokens": 92669658.0,
      "step": 47365
    },
    {
      "entropy": 5.9021069526672365,
      "epoch": 4.735142700054981,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002974183229604888,
      "loss": 5.3704,
      "mean_token_accuracy": 0.20468994081020356,
      "num_tokens": 92680439.0,
      "step": 47370
    },
    {
      "entropy": 5.8430823802948,
      "epoch": 4.7356425251162095,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029738281117115944,
      "loss": 5.2142,
      "mean_token_accuracy": 0.22354355454444885,
      "num_tokens": 92690480.0,
      "step": 47375
    },
    {
      "entropy": 5.784356689453125,
      "epoch": 4.736142350177438,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002973472988186844,
      "loss": 5.1613,
      "mean_token_accuracy": 0.2241012006998062,
      "num_tokens": 92700115.0,
      "step": 47380
    },
    {
      "entropy": 5.819198656082153,
      "epoch": 4.736642175238666,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002973117859039571,
      "loss": 5.2706,
      "mean_token_accuracy": 0.21912865936756135,
      "num_tokens": 92708887.0,
      "step": 47385
    },
    {
      "entropy": 5.8181085109710695,
      "epoch": 4.737142000299895,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029727627242787097,
      "loss": 5.2482,
      "mean_token_accuracy": 0.21685048639774324,
      "num_tokens": 92718337.0,
      "step": 47390
    },
    {
      "entropy": 5.837231731414795,
      "epoch": 4.737641825361123,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029724075839131957,
      "loss": 5.2579,
      "mean_token_accuracy": 0.22028463929891587,
      "num_tokens": 92727520.0,
      "step": 47395
    },
    {
      "entropy": 5.786226749420166,
      "epoch": 4.738141650422352,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002972052437951963,
      "loss": 5.2572,
      "mean_token_accuracy": 0.21632723808288573,
      "num_tokens": 92736999.0,
      "step": 47400
    },
    {
      "entropy": 5.725975513458252,
      "epoch": 4.738641475483581,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029716972864039505,
      "loss": 5.2284,
      "mean_token_accuracy": 0.21894727349281312,
      "num_tokens": 92746606.0,
      "step": 47405
    },
    {
      "entropy": 5.764142799377441,
      "epoch": 4.7391413005448095,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029713421292780896,
      "loss": 5.1842,
      "mean_token_accuracy": 0.22096841037273407,
      "num_tokens": 92757386.0,
      "step": 47410
    },
    {
      "entropy": 5.86564998626709,
      "epoch": 4.739641125606038,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029709869665833186,
      "loss": 5.2572,
      "mean_token_accuracy": 0.22313945591449738,
      "num_tokens": 92766868.0,
      "step": 47415
    },
    {
      "entropy": 5.857166528701782,
      "epoch": 4.740140950667266,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029706317983285713,
      "loss": 5.3008,
      "mean_token_accuracy": 0.21550676077604294,
      "num_tokens": 92776496.0,
      "step": 47420
    },
    {
      "entropy": 5.747288417816162,
      "epoch": 4.740640775728495,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002970276624522785,
      "loss": 5.2174,
      "mean_token_accuracy": 0.22246475964784623,
      "num_tokens": 92786185.0,
      "step": 47425
    },
    {
      "entropy": 5.729208517074585,
      "epoch": 4.741140600789723,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029699214451748955,
      "loss": 5.0852,
      "mean_token_accuracy": 0.22565306574106217,
      "num_tokens": 92796324.0,
      "step": 47430
    },
    {
      "entropy": 5.836783790588379,
      "epoch": 4.741640425850952,
      "grad_norm": 1.125,
      "learning_rate": 0.00029695662602938395,
      "loss": 5.2925,
      "mean_token_accuracy": 0.21486320346593857,
      "num_tokens": 92807122.0,
      "step": 47435
    },
    {
      "entropy": 5.843992805480957,
      "epoch": 4.742140250912181,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029692110698885534,
      "loss": 5.2763,
      "mean_token_accuracy": 0.2131924569606781,
      "num_tokens": 92818413.0,
      "step": 47440
    },
    {
      "entropy": 5.813348770141602,
      "epoch": 4.7426400759734095,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002968855873967972,
      "loss": 5.2606,
      "mean_token_accuracy": 0.20872217863798143,
      "num_tokens": 92827878.0,
      "step": 47445
    },
    {
      "entropy": 5.821229124069214,
      "epoch": 4.743139901034638,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002968500672541034,
      "loss": 5.2822,
      "mean_token_accuracy": 0.21848077476024627,
      "num_tokens": 92838687.0,
      "step": 47450
    },
    {
      "entropy": 5.8545482635498045,
      "epoch": 4.743639726095866,
      "grad_norm": 1.25,
      "learning_rate": 0.00029681454656166746,
      "loss": 5.2384,
      "mean_token_accuracy": 0.22030838280916215,
      "num_tokens": 92847443.0,
      "step": 47455
    },
    {
      "entropy": 5.835563230514526,
      "epoch": 4.744139551157095,
      "grad_norm": 1.125,
      "learning_rate": 0.00029677902532038316,
      "loss": 5.2826,
      "mean_token_accuracy": 0.2177327200770378,
      "num_tokens": 92857580.0,
      "step": 47460
    },
    {
      "entropy": 5.648082637786866,
      "epoch": 4.744639376218323,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002967435035311442,
      "loss": 5.01,
      "mean_token_accuracy": 0.23113665878772735,
      "num_tokens": 92868022.0,
      "step": 47465
    },
    {
      "entropy": 5.8675452709198,
      "epoch": 4.745139201279552,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002967079811948443,
      "loss": 5.3294,
      "mean_token_accuracy": 0.21357798874378203,
      "num_tokens": 92877983.0,
      "step": 47470
    },
    {
      "entropy": 5.872823905944824,
      "epoch": 4.745639026340781,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002966724583123772,
      "loss": 5.2364,
      "mean_token_accuracy": 0.22408159971237182,
      "num_tokens": 92886763.0,
      "step": 47475
    },
    {
      "entropy": 5.808623790740967,
      "epoch": 4.7461388514020095,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002966369348846366,
      "loss": 5.2314,
      "mean_token_accuracy": 0.22174127846956254,
      "num_tokens": 92895899.0,
      "step": 47480
    },
    {
      "entropy": 5.752199935913086,
      "epoch": 4.746638676463238,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029660141091251627,
      "loss": 5.2202,
      "mean_token_accuracy": 0.21821910589933396,
      "num_tokens": 92905958.0,
      "step": 47485
    },
    {
      "entropy": 5.841016244888306,
      "epoch": 4.747138501524466,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002965658863969099,
      "loss": 5.344,
      "mean_token_accuracy": 0.20844752192497254,
      "num_tokens": 92917076.0,
      "step": 47490
    },
    {
      "entropy": 5.797690486907959,
      "epoch": 4.747638326585695,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029653036133871154,
      "loss": 5.2261,
      "mean_token_accuracy": 0.22058231383562088,
      "num_tokens": 92925933.0,
      "step": 47495
    },
    {
      "entropy": 5.760680723190307,
      "epoch": 4.748138151646923,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002964948357388148,
      "loss": 5.1673,
      "mean_token_accuracy": 0.22857381105422975,
      "num_tokens": 92935285.0,
      "step": 47500
    },
    {
      "entropy": 5.80223708152771,
      "epoch": 4.748637976708152,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002964593095981135,
      "loss": 5.216,
      "mean_token_accuracy": 0.2181653246283531,
      "num_tokens": 92945622.0,
      "step": 47505
    },
    {
      "entropy": 5.865049028396607,
      "epoch": 4.749137801769381,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002964237829175015,
      "loss": 5.2383,
      "mean_token_accuracy": 0.2202371433377266,
      "num_tokens": 92955704.0,
      "step": 47510
    },
    {
      "entropy": 5.828684949874878,
      "epoch": 4.7496376268306095,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002963882556978726,
      "loss": 5.2229,
      "mean_token_accuracy": 0.22236986458301544,
      "num_tokens": 92964230.0,
      "step": 47515
    },
    {
      "entropy": 5.845533657073974,
      "epoch": 4.750137451891838,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029635272794012067,
      "loss": 5.2463,
      "mean_token_accuracy": 0.2221672996878624,
      "num_tokens": 92973588.0,
      "step": 47520
    },
    {
      "entropy": 5.786330509185791,
      "epoch": 4.750637276953066,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002963171996451397,
      "loss": 5.2118,
      "mean_token_accuracy": 0.22058846205472946,
      "num_tokens": 92983072.0,
      "step": 47525
    },
    {
      "entropy": 5.808532953262329,
      "epoch": 4.751137102014295,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002962816708138234,
      "loss": 5.3225,
      "mean_token_accuracy": 0.20716114044189454,
      "num_tokens": 92994381.0,
      "step": 47530
    },
    {
      "entropy": 5.879029130935669,
      "epoch": 4.751636927075523,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002962461414470658,
      "loss": 5.2941,
      "mean_token_accuracy": 0.2148963287472725,
      "num_tokens": 93004773.0,
      "step": 47535
    },
    {
      "entropy": 5.913109016418457,
      "epoch": 4.752136752136752,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002962106115457607,
      "loss": 5.3365,
      "mean_token_accuracy": 0.21121536791324616,
      "num_tokens": 93015100.0,
      "step": 47540
    },
    {
      "entropy": 5.79461932182312,
      "epoch": 4.752636577197981,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000296175081110802,
      "loss": 5.2002,
      "mean_token_accuracy": 0.22261871695518493,
      "num_tokens": 93024974.0,
      "step": 47545
    },
    {
      "entropy": 5.766325521469116,
      "epoch": 4.7531364022592095,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029613955014308387,
      "loss": 5.2071,
      "mean_token_accuracy": 0.21959704607725145,
      "num_tokens": 93034270.0,
      "step": 47550
    },
    {
      "entropy": 5.780379056930542,
      "epoch": 4.753636227320438,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029610401864349993,
      "loss": 5.2966,
      "mean_token_accuracy": 0.21671161353588103,
      "num_tokens": 93043884.0,
      "step": 47555
    },
    {
      "entropy": 5.789208698272705,
      "epoch": 4.754136052381666,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029606848661294446,
      "loss": 5.2237,
      "mean_token_accuracy": 0.22326773703098296,
      "num_tokens": 93052451.0,
      "step": 47560
    },
    {
      "entropy": 5.780182266235352,
      "epoch": 4.754635877442895,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029603295405231117,
      "loss": 5.1582,
      "mean_token_accuracy": 0.22757903188467027,
      "num_tokens": 93063597.0,
      "step": 47565
    },
    {
      "entropy": 5.924140071868896,
      "epoch": 4.755135702504123,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002959974209624943,
      "loss": 5.3584,
      "mean_token_accuracy": 0.20430873930454255,
      "num_tokens": 93072914.0,
      "step": 47570
    },
    {
      "entropy": 5.816623163223267,
      "epoch": 4.755635527565352,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029596188734438755,
      "loss": 5.1737,
      "mean_token_accuracy": 0.22140656411647797,
      "num_tokens": 93081800.0,
      "step": 47575
    },
    {
      "entropy": 5.884450197219849,
      "epoch": 4.756135352626581,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002959263531988853,
      "loss": 5.2956,
      "mean_token_accuracy": 0.20944836586713791,
      "num_tokens": 93091020.0,
      "step": 47580
    },
    {
      "entropy": 5.835002040863037,
      "epoch": 4.7566351776878095,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002958908185268812,
      "loss": 5.2635,
      "mean_token_accuracy": 0.21571237593889236,
      "num_tokens": 93100529.0,
      "step": 47585
    },
    {
      "entropy": 5.898374271392822,
      "epoch": 4.757135002749038,
      "grad_norm": 1.25,
      "learning_rate": 0.0002958552833292696,
      "loss": 5.305,
      "mean_token_accuracy": 0.20544545501470565,
      "num_tokens": 93110541.0,
      "step": 47590
    },
    {
      "entropy": 5.875619983673095,
      "epoch": 4.757634827810266,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002958197476069444,
      "loss": 5.3424,
      "mean_token_accuracy": 0.20731985121965407,
      "num_tokens": 93120664.0,
      "step": 47595
    },
    {
      "entropy": 5.838412570953369,
      "epoch": 4.758134652871495,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002957842113607997,
      "loss": 5.2431,
      "mean_token_accuracy": 0.21201371997594834,
      "num_tokens": 93129967.0,
      "step": 47600
    },
    {
      "entropy": 5.817788219451904,
      "epoch": 4.758634477932723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002957486745917296,
      "loss": 5.2041,
      "mean_token_accuracy": 0.21543725579977036,
      "num_tokens": 93139537.0,
      "step": 47605
    },
    {
      "entropy": 5.92238416671753,
      "epoch": 4.759134302993952,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002957131373006282,
      "loss": 5.3453,
      "mean_token_accuracy": 0.20584456026554107,
      "num_tokens": 93150300.0,
      "step": 47610
    },
    {
      "entropy": 5.78009729385376,
      "epoch": 4.759634128055181,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029567759948838964,
      "loss": 5.1913,
      "mean_token_accuracy": 0.22308164834976196,
      "num_tokens": 93159280.0,
      "step": 47615
    },
    {
      "entropy": 5.804564476013184,
      "epoch": 4.7601339531164095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002956420611559079,
      "loss": 5.2572,
      "mean_token_accuracy": 0.2179785668849945,
      "num_tokens": 93169062.0,
      "step": 47620
    },
    {
      "entropy": 5.800934886932373,
      "epoch": 4.760633778177638,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029560652230407723,
      "loss": 5.1969,
      "mean_token_accuracy": 0.21683420538902282,
      "num_tokens": 93177921.0,
      "step": 47625
    },
    {
      "entropy": 5.858273839950561,
      "epoch": 4.761133603238866,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029557098293379186,
      "loss": 5.3961,
      "mean_token_accuracy": 0.2119354262948036,
      "num_tokens": 93188249.0,
      "step": 47630
    },
    {
      "entropy": 5.8370680809021,
      "epoch": 4.761633428300095,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002955354430459457,
      "loss": 5.2662,
      "mean_token_accuracy": 0.21656986624002456,
      "num_tokens": 93198168.0,
      "step": 47635
    },
    {
      "entropy": 5.868140172958374,
      "epoch": 4.762133253361323,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029549990264143323,
      "loss": 5.2958,
      "mean_token_accuracy": 0.21756925135850907,
      "num_tokens": 93208310.0,
      "step": 47640
    },
    {
      "entropy": 5.7876996994018555,
      "epoch": 4.762633078422552,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029546436172114837,
      "loss": 5.1371,
      "mean_token_accuracy": 0.22079551368951797,
      "num_tokens": 93218317.0,
      "step": 47645
    },
    {
      "entropy": 5.769620656967163,
      "epoch": 4.763132903483781,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002954288202859855,
      "loss": 5.2501,
      "mean_token_accuracy": 0.21908589601516723,
      "num_tokens": 93227377.0,
      "step": 47650
    },
    {
      "entropy": 5.661456108093262,
      "epoch": 4.7636327285450095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002953932783368388,
      "loss": 5.0969,
      "mean_token_accuracy": 0.24051095694303512,
      "num_tokens": 93236386.0,
      "step": 47655
    },
    {
      "entropy": 5.833033704757691,
      "epoch": 4.764132553606238,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029535773587460235,
      "loss": 5.2783,
      "mean_token_accuracy": 0.21329327076673507,
      "num_tokens": 93246461.0,
      "step": 47660
    },
    {
      "entropy": 5.782737684249878,
      "epoch": 4.764632378667466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002953221929001706,
      "loss": 5.2526,
      "mean_token_accuracy": 0.21731134355068207,
      "num_tokens": 93256303.0,
      "step": 47665
    },
    {
      "entropy": 5.781830596923828,
      "epoch": 4.765132203728695,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029528664941443766,
      "loss": 5.1783,
      "mean_token_accuracy": 0.2270931273698807,
      "num_tokens": 93267286.0,
      "step": 47670
    },
    {
      "entropy": 5.878135442733765,
      "epoch": 4.765632028789923,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002952511054182978,
      "loss": 5.292,
      "mean_token_accuracy": 0.21071041524410247,
      "num_tokens": 93275807.0,
      "step": 47675
    },
    {
      "entropy": 5.796392154693604,
      "epoch": 4.766131853851152,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029521556091264547,
      "loss": 5.2331,
      "mean_token_accuracy": 0.2172156110405922,
      "num_tokens": 93284956.0,
      "step": 47680
    },
    {
      "entropy": 5.831582546234131,
      "epoch": 4.766631678912381,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029518001589837466,
      "loss": 5.2047,
      "mean_token_accuracy": 0.21834747046232222,
      "num_tokens": 93294028.0,
      "step": 47685
    },
    {
      "entropy": 5.868341159820557,
      "epoch": 4.7671315039736095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029514447037638,
      "loss": 5.2634,
      "mean_token_accuracy": 0.21086618155241013,
      "num_tokens": 93304094.0,
      "step": 47690
    },
    {
      "entropy": 5.7252942562103275,
      "epoch": 4.767631329034838,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029510892434755556,
      "loss": 5.0687,
      "mean_token_accuracy": 0.2257650837302208,
      "num_tokens": 93312901.0,
      "step": 47695
    },
    {
      "entropy": 5.772397708892822,
      "epoch": 4.768131154096066,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002950733778127958,
      "loss": 5.2787,
      "mean_token_accuracy": 0.21102288216352463,
      "num_tokens": 93322277.0,
      "step": 47700
    },
    {
      "entropy": 5.865954256057739,
      "epoch": 4.768630979157295,
      "grad_norm": 1.15625,
      "learning_rate": 0.000295037830772995,
      "loss": 5.2379,
      "mean_token_accuracy": 0.21021449863910674,
      "num_tokens": 93332550.0,
      "step": 47705
    },
    {
      "entropy": 5.947641134262085,
      "epoch": 4.769130804218523,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002950022832290475,
      "loss": 5.4117,
      "mean_token_accuracy": 0.19860086739063262,
      "num_tokens": 93342667.0,
      "step": 47710
    },
    {
      "entropy": 5.868178462982177,
      "epoch": 4.769630629279752,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029496673518184777,
      "loss": 5.3116,
      "mean_token_accuracy": 0.21048267036676407,
      "num_tokens": 93352198.0,
      "step": 47715
    },
    {
      "entropy": 5.81578950881958,
      "epoch": 4.770130454340981,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002949311866322901,
      "loss": 5.2655,
      "mean_token_accuracy": 0.21471429765224456,
      "num_tokens": 93362751.0,
      "step": 47720
    },
    {
      "entropy": 5.783129024505615,
      "epoch": 4.7706302794022095,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029489563758126886,
      "loss": 5.2068,
      "mean_token_accuracy": 0.219323094189167,
      "num_tokens": 93372616.0,
      "step": 47725
    },
    {
      "entropy": 5.642757749557495,
      "epoch": 4.771130104463438,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002948600880296785,
      "loss": 5.067,
      "mean_token_accuracy": 0.23965440541505814,
      "num_tokens": 93382593.0,
      "step": 47730
    },
    {
      "entropy": 5.823071146011353,
      "epoch": 4.771629929524666,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002948245379784135,
      "loss": 5.2114,
      "mean_token_accuracy": 0.22242758721113204,
      "num_tokens": 93390770.0,
      "step": 47735
    },
    {
      "entropy": 5.761183595657348,
      "epoch": 4.772129754585895,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029478898742836817,
      "loss": 5.1389,
      "mean_token_accuracy": 0.2244297131896019,
      "num_tokens": 93400166.0,
      "step": 47740
    },
    {
      "entropy": 5.782255744934082,
      "epoch": 4.772629579647123,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029475343638043697,
      "loss": 5.1919,
      "mean_token_accuracy": 0.22189483791589737,
      "num_tokens": 93409856.0,
      "step": 47745
    },
    {
      "entropy": 5.7054431438446045,
      "epoch": 4.773129404708352,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029471788483551443,
      "loss": 5.1091,
      "mean_token_accuracy": 0.22066725492477418,
      "num_tokens": 93419326.0,
      "step": 47750
    },
    {
      "entropy": 5.812247228622437,
      "epoch": 4.773629229769581,
      "grad_norm": 1.125,
      "learning_rate": 0.0002946823327944951,
      "loss": 5.2646,
      "mean_token_accuracy": 0.20852500647306443,
      "num_tokens": 93429083.0,
      "step": 47755
    },
    {
      "entropy": 5.778392362594604,
      "epoch": 4.7741290548308095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002946467802582732,
      "loss": 5.1412,
      "mean_token_accuracy": 0.2209995359182358,
      "num_tokens": 93438582.0,
      "step": 47760
    },
    {
      "entropy": 5.864433145523071,
      "epoch": 4.774628879892038,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002946112272277434,
      "loss": 5.2892,
      "mean_token_accuracy": 0.21051009893417358,
      "num_tokens": 93447791.0,
      "step": 47765
    },
    {
      "entropy": 5.775034761428833,
      "epoch": 4.775128704953266,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002945756737038002,
      "loss": 5.1866,
      "mean_token_accuracy": 0.22528911978006363,
      "num_tokens": 93458506.0,
      "step": 47770
    },
    {
      "entropy": 5.798443412780761,
      "epoch": 4.775628530014495,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029454011968733804,
      "loss": 5.1804,
      "mean_token_accuracy": 0.21947077810764312,
      "num_tokens": 93468155.0,
      "step": 47775
    },
    {
      "entropy": 5.805390882492065,
      "epoch": 4.776128355075723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002945045651792515,
      "loss": 5.2554,
      "mean_token_accuracy": 0.21523274928331376,
      "num_tokens": 93476839.0,
      "step": 47780
    },
    {
      "entropy": 5.756129598617553,
      "epoch": 4.776628180136952,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029446901018043516,
      "loss": 5.2149,
      "mean_token_accuracy": 0.22367593348026277,
      "num_tokens": 93486590.0,
      "step": 47785
    },
    {
      "entropy": 5.7276694774627686,
      "epoch": 4.777128005198181,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029443345469178355,
      "loss": 5.1143,
      "mean_token_accuracy": 0.23436820209026338,
      "num_tokens": 93495982.0,
      "step": 47790
    },
    {
      "entropy": 5.785484361648559,
      "epoch": 4.7776278302594095,
      "grad_norm": 1.25,
      "learning_rate": 0.00029439789871419113,
      "loss": 5.1929,
      "mean_token_accuracy": 0.23836110383272172,
      "num_tokens": 93505122.0,
      "step": 47795
    },
    {
      "entropy": 5.72336597442627,
      "epoch": 4.778127655320638,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002943623422485527,
      "loss": 5.152,
      "mean_token_accuracy": 0.23195490390062332,
      "num_tokens": 93515423.0,
      "step": 47800
    },
    {
      "entropy": 5.855371475219727,
      "epoch": 4.778627480381866,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002943267852957626,
      "loss": 5.2843,
      "mean_token_accuracy": 0.21515370458364486,
      "num_tokens": 93526007.0,
      "step": 47805
    },
    {
      "entropy": 5.754363584518432,
      "epoch": 4.779127305443095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029429122785671556,
      "loss": 5.2114,
      "mean_token_accuracy": 0.21292911171913148,
      "num_tokens": 93535727.0,
      "step": 47810
    },
    {
      "entropy": 5.811377668380738,
      "epoch": 4.779627130504323,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029425566993230624,
      "loss": 5.2449,
      "mean_token_accuracy": 0.21486254185438156,
      "num_tokens": 93546347.0,
      "step": 47815
    },
    {
      "entropy": 5.758086013793945,
      "epoch": 4.780126955565552,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002942201115234292,
      "loss": 5.1615,
      "mean_token_accuracy": 0.21544632762670518,
      "num_tokens": 93556225.0,
      "step": 47820
    },
    {
      "entropy": 5.841309881210327,
      "epoch": 4.780626780626781,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029418455263097913,
      "loss": 5.2979,
      "mean_token_accuracy": 0.21063094437122346,
      "num_tokens": 93565713.0,
      "step": 47825
    },
    {
      "entropy": 5.81339659690857,
      "epoch": 4.7811266056880095,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029414899325585066,
      "loss": 5.2323,
      "mean_token_accuracy": 0.22441661208868027,
      "num_tokens": 93576841.0,
      "step": 47830
    },
    {
      "entropy": 5.817865610122681,
      "epoch": 4.781626430749238,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029411343339893835,
      "loss": 5.2641,
      "mean_token_accuracy": 0.21352846920490265,
      "num_tokens": 93587351.0,
      "step": 47835
    },
    {
      "entropy": 5.792579984664917,
      "epoch": 4.782126255810466,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029407787306113707,
      "loss": 5.2551,
      "mean_token_accuracy": 0.22686635851860046,
      "num_tokens": 93596796.0,
      "step": 47840
    },
    {
      "entropy": 5.765171194076538,
      "epoch": 4.782626080871695,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029404231224334137,
      "loss": 5.1598,
      "mean_token_accuracy": 0.22522784769535065,
      "num_tokens": 93606445.0,
      "step": 47845
    },
    {
      "entropy": 5.689988708496093,
      "epoch": 4.783125905932923,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029400675094644606,
      "loss": 5.1595,
      "mean_token_accuracy": 0.22887588590383529,
      "num_tokens": 93617233.0,
      "step": 47850
    },
    {
      "entropy": 5.819366312026977,
      "epoch": 4.783625730994152,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029397118917134576,
      "loss": 5.1974,
      "mean_token_accuracy": 0.2224247395992279,
      "num_tokens": 93626638.0,
      "step": 47855
    },
    {
      "entropy": 5.825408887863159,
      "epoch": 4.784125556055381,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002939356269189352,
      "loss": 5.2184,
      "mean_token_accuracy": 0.21688461303710938,
      "num_tokens": 93636869.0,
      "step": 47860
    },
    {
      "entropy": 5.85829439163208,
      "epoch": 4.7846253811166095,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029390006419010916,
      "loss": 5.3173,
      "mean_token_accuracy": 0.2136480212211609,
      "num_tokens": 93647009.0,
      "step": 47865
    },
    {
      "entropy": 5.783406019210815,
      "epoch": 4.785125206177838,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002938645009857623,
      "loss": 5.2047,
      "mean_token_accuracy": 0.22050586789846421,
      "num_tokens": 93657940.0,
      "step": 47870
    },
    {
      "entropy": 5.840200471878052,
      "epoch": 4.785625031239066,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002938289373067895,
      "loss": 5.2487,
      "mean_token_accuracy": 0.20951860249042512,
      "num_tokens": 93666962.0,
      "step": 47875
    },
    {
      "entropy": 5.838033723831177,
      "epoch": 4.786124856300295,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002937933731540854,
      "loss": 5.2975,
      "mean_token_accuracy": 0.20324519723653794,
      "num_tokens": 93677642.0,
      "step": 47880
    },
    {
      "entropy": 5.780894804000854,
      "epoch": 4.786624681361523,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000293757808528545,
      "loss": 5.1336,
      "mean_token_accuracy": 0.23076663613319398,
      "num_tokens": 93687873.0,
      "step": 47885
    },
    {
      "entropy": 5.8220209121704105,
      "epoch": 4.787124506422752,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002937222434310629,
      "loss": 5.2955,
      "mean_token_accuracy": 0.21357200741767884,
      "num_tokens": 93698363.0,
      "step": 47890
    },
    {
      "entropy": 5.917741489410401,
      "epoch": 4.787624331483981,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029368667786253404,
      "loss": 5.349,
      "mean_token_accuracy": 0.21278341263532638,
      "num_tokens": 93707562.0,
      "step": 47895
    },
    {
      "entropy": 5.728090238571167,
      "epoch": 4.7881241565452095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029365111182385305,
      "loss": 5.0955,
      "mean_token_accuracy": 0.23560260683298112,
      "num_tokens": 93716664.0,
      "step": 47900
    },
    {
      "entropy": 5.712009620666504,
      "epoch": 4.788623981606438,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000293615545315915,
      "loss": 5.1437,
      "mean_token_accuracy": 0.23016658574342727,
      "num_tokens": 93726710.0,
      "step": 47905
    },
    {
      "entropy": 5.752905702590942,
      "epoch": 4.789123806667666,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029357997833961454,
      "loss": 5.1377,
      "mean_token_accuracy": 0.2304869621992111,
      "num_tokens": 93736949.0,
      "step": 47910
    },
    {
      "entropy": 5.81825361251831,
      "epoch": 4.789623631728895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029354441089584667,
      "loss": 5.2739,
      "mean_token_accuracy": 0.2119083508849144,
      "num_tokens": 93746510.0,
      "step": 47915
    },
    {
      "entropy": 5.746563577651978,
      "epoch": 4.790123456790123,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029350884298550616,
      "loss": 5.217,
      "mean_token_accuracy": 0.2246015563607216,
      "num_tokens": 93755791.0,
      "step": 47920
    },
    {
      "entropy": 5.782573652267456,
      "epoch": 4.790623281851352,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029347327460948786,
      "loss": 5.2155,
      "mean_token_accuracy": 0.22163795232772826,
      "num_tokens": 93765868.0,
      "step": 47925
    },
    {
      "entropy": 5.737606143951416,
      "epoch": 4.791123106912581,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029343770576868684,
      "loss": 5.1492,
      "mean_token_accuracy": 0.2348059445619583,
      "num_tokens": 93777539.0,
      "step": 47930
    },
    {
      "entropy": 5.789166879653931,
      "epoch": 4.7916229319738095,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002934021364639978,
      "loss": 5.2617,
      "mean_token_accuracy": 0.2127280592918396,
      "num_tokens": 93789214.0,
      "step": 47935
    },
    {
      "entropy": 5.8353596210479735,
      "epoch": 4.792122757035038,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002933665666963158,
      "loss": 5.2003,
      "mean_token_accuracy": 0.21526670455932617,
      "num_tokens": 93798780.0,
      "step": 47940
    },
    {
      "entropy": 5.834532880783081,
      "epoch": 4.792622582096266,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029333099646653565,
      "loss": 5.3222,
      "mean_token_accuracy": 0.2101569339632988,
      "num_tokens": 93809205.0,
      "step": 47945
    },
    {
      "entropy": 5.823121643066406,
      "epoch": 4.793122407157495,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002932954257755524,
      "loss": 5.1893,
      "mean_token_accuracy": 0.226746267080307,
      "num_tokens": 93818164.0,
      "step": 47950
    },
    {
      "entropy": 5.868682765960694,
      "epoch": 4.793622232218723,
      "grad_norm": 1.203125,
      "learning_rate": 0.000293259854624261,
      "loss": 5.2809,
      "mean_token_accuracy": 0.2092116191983223,
      "num_tokens": 93827582.0,
      "step": 47955
    },
    {
      "entropy": 5.836835193634033,
      "epoch": 4.794122057279952,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029322428301355624,
      "loss": 5.1991,
      "mean_token_accuracy": 0.22493767142295837,
      "num_tokens": 93837594.0,
      "step": 47960
    },
    {
      "entropy": 5.7689821243286135,
      "epoch": 4.794621882341181,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029318871094433324,
      "loss": 5.13,
      "mean_token_accuracy": 0.22999872714281083,
      "num_tokens": 93846563.0,
      "step": 47965
    },
    {
      "entropy": 5.81068229675293,
      "epoch": 4.7951217074024095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029315313841748704,
      "loss": 5.2457,
      "mean_token_accuracy": 0.21385176479816437,
      "num_tokens": 93857052.0,
      "step": 47970
    },
    {
      "entropy": 5.77226939201355,
      "epoch": 4.795621532463638,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029311756543391244,
      "loss": 5.2504,
      "mean_token_accuracy": 0.21396289467811586,
      "num_tokens": 93867211.0,
      "step": 47975
    },
    {
      "entropy": 5.810956239700317,
      "epoch": 4.796121357524866,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002930819919945046,
      "loss": 5.2469,
      "mean_token_accuracy": 0.21074824631214142,
      "num_tokens": 93876990.0,
      "step": 47980
    },
    {
      "entropy": 5.887476921081543,
      "epoch": 4.796621182586095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002930464181001585,
      "loss": 5.3605,
      "mean_token_accuracy": 0.20811742544174194,
      "num_tokens": 93886388.0,
      "step": 47985
    },
    {
      "entropy": 5.8300048351287845,
      "epoch": 4.797121007647323,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002930108437517691,
      "loss": 5.2964,
      "mean_token_accuracy": 0.20729176700115204,
      "num_tokens": 93895522.0,
      "step": 47990
    },
    {
      "entropy": 5.87089433670044,
      "epoch": 4.797620832708552,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029297526895023156,
      "loss": 5.2861,
      "mean_token_accuracy": 0.210190948843956,
      "num_tokens": 93905181.0,
      "step": 47995
    },
    {
      "entropy": 5.894828605651855,
      "epoch": 4.798120657769781,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002929396936964409,
      "loss": 5.3471,
      "mean_token_accuracy": 0.20841705054044724,
      "num_tokens": 93915201.0,
      "step": 48000
    },
    {
      "epoch": 4.798120657769781,
      "eval_entropy": 5.603181414922406,
      "eval_loss": 5.3974289894104,
      "eval_mean_token_accuracy": 0.2183662213997508,
      "eval_num_tokens": 93915201.0,
      "eval_runtime": 23.8172,
      "eval_samples_per_second": 1303.467,
      "eval_steps_per_second": 162.949,
      "step": 48000
    },
    {
      "entropy": 5.8535724639892575,
      "epoch": 4.7986204828310095,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002929041179912921,
      "loss": 5.2003,
      "mean_token_accuracy": 0.21640380769968032,
      "num_tokens": 93924643.0,
      "step": 48005
    },
    {
      "entropy": 5.8378993511199955,
      "epoch": 4.799120307892238,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029286854183568033,
      "loss": 5.2726,
      "mean_token_accuracy": 0.21453656405210494,
      "num_tokens": 93934459.0,
      "step": 48010
    },
    {
      "entropy": 5.763594007492065,
      "epoch": 4.799620132953466,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002928329652305007,
      "loss": 5.3164,
      "mean_token_accuracy": 0.20927723944187165,
      "num_tokens": 93945090.0,
      "step": 48015
    },
    {
      "entropy": 5.741130447387695,
      "epoch": 4.800119958014695,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002927973881766482,
      "loss": 5.1604,
      "mean_token_accuracy": 0.2218460351228714,
      "num_tokens": 93954121.0,
      "step": 48020
    },
    {
      "entropy": 5.873915767669677,
      "epoch": 4.800619783075923,
      "grad_norm": 1.1875,
      "learning_rate": 0.000292761810675018,
      "loss": 5.2994,
      "mean_token_accuracy": 0.21371822506189347,
      "num_tokens": 93964801.0,
      "step": 48025
    },
    {
      "entropy": 5.870990991592407,
      "epoch": 4.801119608137152,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002927262327265052,
      "loss": 5.2409,
      "mean_token_accuracy": 0.21281859129667283,
      "num_tokens": 93974809.0,
      "step": 48030
    },
    {
      "entropy": 5.911225748062134,
      "epoch": 4.801619433198381,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029269065433200494,
      "loss": 5.3074,
      "mean_token_accuracy": 0.20694425851106643,
      "num_tokens": 93983671.0,
      "step": 48035
    },
    {
      "entropy": 5.82842755317688,
      "epoch": 4.8021192582596095,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002926550754924124,
      "loss": 5.2542,
      "mean_token_accuracy": 0.21985478848218917,
      "num_tokens": 93993233.0,
      "step": 48040
    },
    {
      "entropy": 5.765724086761475,
      "epoch": 4.802619083320838,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002926194962086227,
      "loss": 5.2112,
      "mean_token_accuracy": 0.21763395220041276,
      "num_tokens": 94002763.0,
      "step": 48045
    },
    {
      "entropy": 5.808754587173462,
      "epoch": 4.803118908382066,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000292583916481531,
      "loss": 5.2373,
      "mean_token_accuracy": 0.22701563835144042,
      "num_tokens": 94012254.0,
      "step": 48050
    },
    {
      "entropy": 5.759091329574585,
      "epoch": 4.803618733443295,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002925483363120325,
      "loss": 5.1159,
      "mean_token_accuracy": 0.23511704951524734,
      "num_tokens": 94020656.0,
      "step": 48055
    },
    {
      "entropy": 5.770436811447143,
      "epoch": 4.804118558504523,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002925127557010223,
      "loss": 5.1586,
      "mean_token_accuracy": 0.22835119366645812,
      "num_tokens": 94030626.0,
      "step": 48060
    },
    {
      "entropy": 5.673846864700318,
      "epoch": 4.804618383565752,
      "grad_norm": 1.25,
      "learning_rate": 0.0002924771746493958,
      "loss": 5.1381,
      "mean_token_accuracy": 0.2229207932949066,
      "num_tokens": 94040547.0,
      "step": 48065
    },
    {
      "entropy": 5.872878170013427,
      "epoch": 4.805118208626981,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029244159315804796,
      "loss": 5.3286,
      "mean_token_accuracy": 0.20745191127061843,
      "num_tokens": 94050376.0,
      "step": 48070
    },
    {
      "entropy": 5.799742126464844,
      "epoch": 4.8056180336882095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002924060112278742,
      "loss": 5.2493,
      "mean_token_accuracy": 0.22375667244195938,
      "num_tokens": 94060620.0,
      "step": 48075
    },
    {
      "entropy": 5.827837181091309,
      "epoch": 4.806117858749437,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029237042885976966,
      "loss": 5.2067,
      "mean_token_accuracy": 0.2185348764061928,
      "num_tokens": 94069769.0,
      "step": 48080
    },
    {
      "entropy": 5.767286396026611,
      "epoch": 4.806617683810666,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029233484605462956,
      "loss": 5.2021,
      "mean_token_accuracy": 0.22302276343107225,
      "num_tokens": 94079307.0,
      "step": 48085
    },
    {
      "entropy": 5.817012023925781,
      "epoch": 4.807117508871895,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029229926281334925,
      "loss": 5.2995,
      "mean_token_accuracy": 0.21089343577623368,
      "num_tokens": 94088811.0,
      "step": 48090
    },
    {
      "entropy": 5.71518292427063,
      "epoch": 4.807617333933123,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002922636791368239,
      "loss": 5.1576,
      "mean_token_accuracy": 0.23097544461488723,
      "num_tokens": 94099498.0,
      "step": 48095
    },
    {
      "entropy": 5.855770540237427,
      "epoch": 4.808117158994352,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029222809502594887,
      "loss": 5.275,
      "mean_token_accuracy": 0.2083741992712021,
      "num_tokens": 94108444.0,
      "step": 48100
    },
    {
      "entropy": 5.88870096206665,
      "epoch": 4.808616984055581,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029219251048161936,
      "loss": 5.2551,
      "mean_token_accuracy": 0.21266642808914185,
      "num_tokens": 94118288.0,
      "step": 48105
    },
    {
      "entropy": 5.770280027389527,
      "epoch": 4.8091168091168095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029215692550473075,
      "loss": 5.1837,
      "mean_token_accuracy": 0.2176469773054123,
      "num_tokens": 94127821.0,
      "step": 48110
    },
    {
      "entropy": 5.821003484725952,
      "epoch": 4.809616634178037,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002921213400961783,
      "loss": 5.2558,
      "mean_token_accuracy": 0.21407032757997513,
      "num_tokens": 94137445.0,
      "step": 48115
    },
    {
      "entropy": 5.714313650131226,
      "epoch": 4.810116459239266,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029208575425685725,
      "loss": 5.1091,
      "mean_token_accuracy": 0.2403782993555069,
      "num_tokens": 94147802.0,
      "step": 48120
    },
    {
      "entropy": 5.620806789398193,
      "epoch": 4.810616284300495,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002920501679876631,
      "loss": 5.0778,
      "mean_token_accuracy": 0.2391015872359276,
      "num_tokens": 94157429.0,
      "step": 48125
    },
    {
      "entropy": 5.730509281158447,
      "epoch": 4.811116109361723,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002920145812894911,
      "loss": 5.1302,
      "mean_token_accuracy": 0.2224692478775978,
      "num_tokens": 94167176.0,
      "step": 48130
    },
    {
      "entropy": 5.780545806884765,
      "epoch": 4.811615934422952,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002919789941632366,
      "loss": 5.2119,
      "mean_token_accuracy": 0.22129747420549392,
      "num_tokens": 94175988.0,
      "step": 48135
    },
    {
      "entropy": 5.820768880844116,
      "epoch": 4.812115759484181,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029194340660979506,
      "loss": 5.2414,
      "mean_token_accuracy": 0.21386076509952545,
      "num_tokens": 94185969.0,
      "step": 48140
    },
    {
      "entropy": 5.832958602905274,
      "epoch": 4.812615584545409,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029190781863006174,
      "loss": 5.3045,
      "mean_token_accuracy": 0.21618153005838395,
      "num_tokens": 94196613.0,
      "step": 48145
    },
    {
      "entropy": 5.786663627624511,
      "epoch": 4.813115409606637,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000291872230224932,
      "loss": 5.2138,
      "mean_token_accuracy": 0.22202938199043273,
      "num_tokens": 94206217.0,
      "step": 48150
    },
    {
      "entropy": 5.771391487121582,
      "epoch": 4.813615234667866,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029183664139530125,
      "loss": 5.2332,
      "mean_token_accuracy": 0.2191184401512146,
      "num_tokens": 94216018.0,
      "step": 48155
    },
    {
      "entropy": 5.733765649795532,
      "epoch": 4.814115059729095,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000291801052142065,
      "loss": 5.2481,
      "mean_token_accuracy": 0.22111430317163466,
      "num_tokens": 94226409.0,
      "step": 48160
    },
    {
      "entropy": 5.826663303375244,
      "epoch": 4.814614884790323,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029176546246611855,
      "loss": 5.1937,
      "mean_token_accuracy": 0.22074921578168868,
      "num_tokens": 94235397.0,
      "step": 48165
    },
    {
      "entropy": 5.796193933486938,
      "epoch": 4.815114709851552,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002917298723683575,
      "loss": 5.2651,
      "mean_token_accuracy": 0.21418033093214034,
      "num_tokens": 94245844.0,
      "step": 48170
    },
    {
      "entropy": 5.837117099761963,
      "epoch": 4.815614534912781,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029169428184967706,
      "loss": 5.2529,
      "mean_token_accuracy": 0.22328338474035264,
      "num_tokens": 94256372.0,
      "step": 48175
    },
    {
      "entropy": 5.723513174057007,
      "epoch": 4.816114359974009,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029165869091097283,
      "loss": 5.1885,
      "mean_token_accuracy": 0.2267742708325386,
      "num_tokens": 94267667.0,
      "step": 48180
    },
    {
      "entropy": 5.8045590877532955,
      "epoch": 4.816614185035237,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002916230995531402,
      "loss": 5.2143,
      "mean_token_accuracy": 0.22105358839035033,
      "num_tokens": 94276551.0,
      "step": 48185
    },
    {
      "entropy": 5.770199203491211,
      "epoch": 4.817114010096466,
      "grad_norm": 1.25,
      "learning_rate": 0.00029158750777707467,
      "loss": 5.2601,
      "mean_token_accuracy": 0.22423477619886398,
      "num_tokens": 94286641.0,
      "step": 48190
    },
    {
      "entropy": 5.876237201690674,
      "epoch": 4.817613835157695,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002915519155836717,
      "loss": 5.3152,
      "mean_token_accuracy": 0.20357059985399245,
      "num_tokens": 94295874.0,
      "step": 48195
    },
    {
      "entropy": 5.879189157485962,
      "epoch": 4.818113660218923,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002915163229738268,
      "loss": 5.2725,
      "mean_token_accuracy": 0.21711881458759308,
      "num_tokens": 94306199.0,
      "step": 48200
    },
    {
      "entropy": 5.676439619064331,
      "epoch": 4.818613485280152,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002914807299484355,
      "loss": 5.0524,
      "mean_token_accuracy": 0.2510624065995216,
      "num_tokens": 94316329.0,
      "step": 48205
    },
    {
      "entropy": 5.787962913513184,
      "epoch": 4.819113310341381,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002914451365083932,
      "loss": 5.2195,
      "mean_token_accuracy": 0.22366165816783906,
      "num_tokens": 94325844.0,
      "step": 48210
    },
    {
      "entropy": 5.692712831497192,
      "epoch": 4.819613135402609,
      "grad_norm": 1.125,
      "learning_rate": 0.00029140954265459554,
      "loss": 5.1602,
      "mean_token_accuracy": 0.22351760268211365,
      "num_tokens": 94336398.0,
      "step": 48215
    },
    {
      "entropy": 5.717973709106445,
      "epoch": 4.820112960463837,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000291373948387938,
      "loss": 5.2348,
      "mean_token_accuracy": 0.2229483023285866,
      "num_tokens": 94347187.0,
      "step": 48220
    },
    {
      "entropy": 5.924737405776978,
      "epoch": 4.820612785525066,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002913383537093161,
      "loss": 5.3889,
      "mean_token_accuracy": 0.20333433002233506,
      "num_tokens": 94357594.0,
      "step": 48225
    },
    {
      "entropy": 5.856863021850586,
      "epoch": 4.821112610586295,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002913027586196255,
      "loss": 5.2191,
      "mean_token_accuracy": 0.21558418422937392,
      "num_tokens": 94368172.0,
      "step": 48230
    },
    {
      "entropy": 5.736045026779175,
      "epoch": 4.821612435647523,
      "grad_norm": 1.125,
      "learning_rate": 0.0002912671631197616,
      "loss": 5.2362,
      "mean_token_accuracy": 0.21838701963424684,
      "num_tokens": 94377532.0,
      "step": 48235
    },
    {
      "entropy": 5.871310472488403,
      "epoch": 4.822112260708752,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002912315672106202,
      "loss": 5.3637,
      "mean_token_accuracy": 0.21063364595174788,
      "num_tokens": 94388028.0,
      "step": 48240
    },
    {
      "entropy": 5.774447441101074,
      "epoch": 4.822612085769981,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002911959708930966,
      "loss": 5.1856,
      "mean_token_accuracy": 0.2258029028773308,
      "num_tokens": 94397777.0,
      "step": 48245
    },
    {
      "entropy": 5.8197722911834715,
      "epoch": 4.823111910831209,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002911603741680866,
      "loss": 5.289,
      "mean_token_accuracy": 0.21036362797021865,
      "num_tokens": 94408371.0,
      "step": 48250
    },
    {
      "entropy": 5.732560682296753,
      "epoch": 4.823611735892437,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029112477703648575,
      "loss": 5.217,
      "mean_token_accuracy": 0.22736431658267975,
      "num_tokens": 94417652.0,
      "step": 48255
    },
    {
      "entropy": 5.848376655578614,
      "epoch": 4.824111560953666,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002910891794991897,
      "loss": 5.2909,
      "mean_token_accuracy": 0.21030962318181992,
      "num_tokens": 94426519.0,
      "step": 48260
    },
    {
      "entropy": 5.8629225730896,
      "epoch": 4.824611386014895,
      "grad_norm": 1.265625,
      "learning_rate": 0.000291053581557094,
      "loss": 5.2763,
      "mean_token_accuracy": 0.21356878727674483,
      "num_tokens": 94436175.0,
      "step": 48265
    },
    {
      "entropy": 5.836793661117554,
      "epoch": 4.825111211076123,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029101798321109427,
      "loss": 5.2365,
      "mean_token_accuracy": 0.21589066088199615,
      "num_tokens": 94445663.0,
      "step": 48270
    },
    {
      "entropy": 5.665655899047851,
      "epoch": 4.825611036137352,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029098238446208626,
      "loss": 5.086,
      "mean_token_accuracy": 0.2318326562643051,
      "num_tokens": 94455575.0,
      "step": 48275
    },
    {
      "entropy": 5.801303720474243,
      "epoch": 4.826110861198581,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002909467853109656,
      "loss": 5.2159,
      "mean_token_accuracy": 0.22106779217720032,
      "num_tokens": 94465088.0,
      "step": 48280
    },
    {
      "entropy": 5.863470506668091,
      "epoch": 4.826610686259809,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002909111857586278,
      "loss": 5.3025,
      "mean_token_accuracy": 0.2110141396522522,
      "num_tokens": 94473989.0,
      "step": 48285
    },
    {
      "entropy": 5.876831912994385,
      "epoch": 4.827110511321037,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002908755858059688,
      "loss": 5.2858,
      "mean_token_accuracy": 0.2191615954041481,
      "num_tokens": 94483999.0,
      "step": 48290
    },
    {
      "entropy": 5.884886026382446,
      "epoch": 4.827610336382266,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029083998545388406,
      "loss": 5.3309,
      "mean_token_accuracy": 0.2119506224989891,
      "num_tokens": 94494701.0,
      "step": 48295
    },
    {
      "entropy": 5.818874883651733,
      "epoch": 4.828110161443495,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002908043847032694,
      "loss": 5.1839,
      "mean_token_accuracy": 0.22136540859937667,
      "num_tokens": 94504067.0,
      "step": 48300
    },
    {
      "entropy": 5.8010107517242435,
      "epoch": 4.828609986504723,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029076878355502046,
      "loss": 5.246,
      "mean_token_accuracy": 0.2145160898566246,
      "num_tokens": 94513409.0,
      "step": 48305
    },
    {
      "entropy": 5.847391605377197,
      "epoch": 4.829109811565952,
      "grad_norm": 1.3125,
      "learning_rate": 0.000290733182010033,
      "loss": 5.2392,
      "mean_token_accuracy": 0.21593121141195298,
      "num_tokens": 94522408.0,
      "step": 48310
    },
    {
      "entropy": 5.778770542144775,
      "epoch": 4.829609636627181,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002906975800692027,
      "loss": 5.1354,
      "mean_token_accuracy": 0.22908181995153426,
      "num_tokens": 94532052.0,
      "step": 48315
    },
    {
      "entropy": 5.852521991729736,
      "epoch": 4.830109461688409,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029066197773342536,
      "loss": 5.3231,
      "mean_token_accuracy": 0.20967686921358109,
      "num_tokens": 94541400.0,
      "step": 48320
    },
    {
      "entropy": 5.772142219543457,
      "epoch": 4.830609286749637,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002906263750035967,
      "loss": 5.2062,
      "mean_token_accuracy": 0.21802634000778198,
      "num_tokens": 94551369.0,
      "step": 48325
    },
    {
      "entropy": 5.726253318786621,
      "epoch": 4.831109111810866,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029059077188061245,
      "loss": 5.2002,
      "mean_token_accuracy": 0.22370177507400513,
      "num_tokens": 94562387.0,
      "step": 48330
    },
    {
      "entropy": 5.791087007522583,
      "epoch": 4.831608936872095,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002905551683653683,
      "loss": 5.1891,
      "mean_token_accuracy": 0.21920527815818786,
      "num_tokens": 94572283.0,
      "step": 48335
    },
    {
      "entropy": 5.733978652954102,
      "epoch": 4.832108761933323,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002905195644587603,
      "loss": 5.0469,
      "mean_token_accuracy": 0.24421666115522384,
      "num_tokens": 94582348.0,
      "step": 48340
    },
    {
      "entropy": 5.827839040756226,
      "epoch": 4.832608586994552,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029048396016168396,
      "loss": 5.2499,
      "mean_token_accuracy": 0.21687576025724412,
      "num_tokens": 94592273.0,
      "step": 48345
    },
    {
      "entropy": 5.7466224193572994,
      "epoch": 4.833108412055781,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029044835547503516,
      "loss": 5.1256,
      "mean_token_accuracy": 0.22376050800085068,
      "num_tokens": 94601914.0,
      "step": 48350
    },
    {
      "entropy": 5.750549364089966,
      "epoch": 4.833608237117009,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029041275039970964,
      "loss": 5.2235,
      "mean_token_accuracy": 0.21653195172548295,
      "num_tokens": 94611965.0,
      "step": 48355
    },
    {
      "entropy": 5.641594696044922,
      "epoch": 4.834108062178237,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002903771449366034,
      "loss": 5.1081,
      "mean_token_accuracy": 0.2377405896782875,
      "num_tokens": 94623372.0,
      "step": 48360
    },
    {
      "entropy": 5.861559343338013,
      "epoch": 4.834607887239466,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002903415390866121,
      "loss": 5.2264,
      "mean_token_accuracy": 0.2145664170384407,
      "num_tokens": 94632354.0,
      "step": 48365
    },
    {
      "entropy": 5.832646179199219,
      "epoch": 4.835107712300695,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002903059328506317,
      "loss": 5.1997,
      "mean_token_accuracy": 0.22251516431570054,
      "num_tokens": 94642502.0,
      "step": 48370
    },
    {
      "entropy": 5.7655078887939455,
      "epoch": 4.835607537361923,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029027032622955777,
      "loss": 5.1494,
      "mean_token_accuracy": 0.22200191915035247,
      "num_tokens": 94651496.0,
      "step": 48375
    },
    {
      "entropy": 5.789140844345093,
      "epoch": 4.836107362423152,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029023471922428657,
      "loss": 5.2825,
      "mean_token_accuracy": 0.2154518559575081,
      "num_tokens": 94660472.0,
      "step": 48380
    },
    {
      "entropy": 5.801126766204834,
      "epoch": 4.836607187484381,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002901991118357137,
      "loss": 5.2611,
      "mean_token_accuracy": 0.22055306285619736,
      "num_tokens": 94669655.0,
      "step": 48385
    },
    {
      "entropy": 5.837733507156372,
      "epoch": 4.837107012545609,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029016350406473507,
      "loss": 5.3039,
      "mean_token_accuracy": 0.21358225047588347,
      "num_tokens": 94679708.0,
      "step": 48390
    },
    {
      "entropy": 5.7975095272064205,
      "epoch": 4.837606837606837,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029012789591224657,
      "loss": 5.2084,
      "mean_token_accuracy": 0.22656022161245346,
      "num_tokens": 94689574.0,
      "step": 48395
    },
    {
      "entropy": 5.73718786239624,
      "epoch": 4.838106662668066,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002900922873791441,
      "loss": 5.1225,
      "mean_token_accuracy": 0.2340715229511261,
      "num_tokens": 94699720.0,
      "step": 48400
    },
    {
      "entropy": 5.792519044876099,
      "epoch": 4.838606487729295,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002900566784663235,
      "loss": 5.1341,
      "mean_token_accuracy": 0.22717469036579133,
      "num_tokens": 94709272.0,
      "step": 48405
    },
    {
      "entropy": 5.880201959609986,
      "epoch": 4.839106312790523,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029002106917468086,
      "loss": 5.2983,
      "mean_token_accuracy": 0.20872645527124406,
      "num_tokens": 94719128.0,
      "step": 48410
    },
    {
      "entropy": 5.812291622161865,
      "epoch": 4.839606137851752,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028998545950511195,
      "loss": 5.1792,
      "mean_token_accuracy": 0.22745523005723953,
      "num_tokens": 94728384.0,
      "step": 48415
    },
    {
      "entropy": 5.789049768447876,
      "epoch": 4.840105962912981,
      "grad_norm": 1.125,
      "learning_rate": 0.00028994984945851277,
      "loss": 5.2373,
      "mean_token_accuracy": 0.2277359813451767,
      "num_tokens": 94737657.0,
      "step": 48420
    },
    {
      "entropy": 5.734743404388428,
      "epoch": 4.840605787974209,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002899142390357792,
      "loss": 5.1388,
      "mean_token_accuracy": 0.22606067806482316,
      "num_tokens": 94747436.0,
      "step": 48425
    },
    {
      "entropy": 5.7436377048492435,
      "epoch": 4.841105613035437,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028987862823780713,
      "loss": 5.1381,
      "mean_token_accuracy": 0.22144332975149156,
      "num_tokens": 94757334.0,
      "step": 48430
    },
    {
      "entropy": 5.869490623474121,
      "epoch": 4.841605438096666,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028984301706549265,
      "loss": 5.3633,
      "mean_token_accuracy": 0.20576487481594086,
      "num_tokens": 94768481.0,
      "step": 48435
    },
    {
      "entropy": 5.739830017089844,
      "epoch": 4.842105263157895,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028980740551973167,
      "loss": 5.1447,
      "mean_token_accuracy": 0.2232888713479042,
      "num_tokens": 94779095.0,
      "step": 48440
    },
    {
      "entropy": 5.838726902008057,
      "epoch": 4.842605088219123,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002897717936014202,
      "loss": 5.3064,
      "mean_token_accuracy": 0.21015710681676864,
      "num_tokens": 94788838.0,
      "step": 48445
    },
    {
      "entropy": 5.86359601020813,
      "epoch": 4.843104913280352,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002897361813114543,
      "loss": 5.2697,
      "mean_token_accuracy": 0.2182282567024231,
      "num_tokens": 94797971.0,
      "step": 48450
    },
    {
      "entropy": 5.771420860290528,
      "epoch": 4.843604738341581,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028970056865072974,
      "loss": 5.2111,
      "mean_token_accuracy": 0.21907562911510467,
      "num_tokens": 94808197.0,
      "step": 48455
    },
    {
      "entropy": 5.758467960357666,
      "epoch": 4.844104563402809,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028966495562014267,
      "loss": 5.1349,
      "mean_token_accuracy": 0.2272206500172615,
      "num_tokens": 94817887.0,
      "step": 48460
    },
    {
      "entropy": 5.787207984924317,
      "epoch": 4.844604388464037,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002896293422205891,
      "loss": 5.1699,
      "mean_token_accuracy": 0.21905305683612825,
      "num_tokens": 94827472.0,
      "step": 48465
    },
    {
      "entropy": 5.707177925109863,
      "epoch": 4.845104213525266,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028959372845296504,
      "loss": 5.1332,
      "mean_token_accuracy": 0.2331377625465393,
      "num_tokens": 94837313.0,
      "step": 48470
    },
    {
      "entropy": 5.8219469547271725,
      "epoch": 4.845604038586495,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002895581143181665,
      "loss": 5.3656,
      "mean_token_accuracy": 0.2159707710146904,
      "num_tokens": 94848332.0,
      "step": 48475
    },
    {
      "entropy": 5.847135734558106,
      "epoch": 4.846103863647723,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002895224998170897,
      "loss": 5.227,
      "mean_token_accuracy": 0.22658550441265107,
      "num_tokens": 94857953.0,
      "step": 48480
    },
    {
      "entropy": 5.898191261291504,
      "epoch": 4.846603688708952,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002894868849506304,
      "loss": 5.325,
      "mean_token_accuracy": 0.2099463924765587,
      "num_tokens": 94866692.0,
      "step": 48485
    },
    {
      "entropy": 5.814535665512085,
      "epoch": 4.847103513770181,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002894512697196848,
      "loss": 5.2298,
      "mean_token_accuracy": 0.2245158299803734,
      "num_tokens": 94876203.0,
      "step": 48490
    },
    {
      "entropy": 5.848943758010864,
      "epoch": 4.847603338831409,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028941565412514907,
      "loss": 5.3097,
      "mean_token_accuracy": 0.21944934278726577,
      "num_tokens": 94885060.0,
      "step": 48495
    },
    {
      "entropy": 5.900507831573487,
      "epoch": 4.848103163892637,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028938003816791907,
      "loss": 5.3521,
      "mean_token_accuracy": 0.20915553569793702,
      "num_tokens": 94895268.0,
      "step": 48500
    },
    {
      "entropy": 5.864008092880249,
      "epoch": 4.848602988953866,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002893444218488911,
      "loss": 5.2781,
      "mean_token_accuracy": 0.21560501009225846,
      "num_tokens": 94904659.0,
      "step": 48505
    },
    {
      "entropy": 5.736719083786011,
      "epoch": 4.849102814015095,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028930880516896113,
      "loss": 5.1912,
      "mean_token_accuracy": 0.2211427092552185,
      "num_tokens": 94912709.0,
      "step": 48510
    },
    {
      "entropy": 5.683274126052856,
      "epoch": 4.849602639076323,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002892731881290253,
      "loss": 5.1404,
      "mean_token_accuracy": 0.23433176428079605,
      "num_tokens": 94922403.0,
      "step": 48515
    },
    {
      "entropy": 5.796094465255737,
      "epoch": 4.850102464137552,
      "grad_norm": 1.25,
      "learning_rate": 0.0002892375707299798,
      "loss": 5.2661,
      "mean_token_accuracy": 0.22111051827669143,
      "num_tokens": 94931254.0,
      "step": 48520
    },
    {
      "entropy": 5.897945356369019,
      "epoch": 4.850602289198781,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002892019529727206,
      "loss": 5.3835,
      "mean_token_accuracy": 0.19795309603214264,
      "num_tokens": 94942297.0,
      "step": 48525
    },
    {
      "entropy": 5.816406297683716,
      "epoch": 4.851102114260009,
      "grad_norm": 1.25,
      "learning_rate": 0.0002891663348581439,
      "loss": 5.2009,
      "mean_token_accuracy": 0.2144999086856842,
      "num_tokens": 94951857.0,
      "step": 48530
    },
    {
      "entropy": 5.914576578140259,
      "epoch": 4.851601939321237,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002891307163871459,
      "loss": 5.3423,
      "mean_token_accuracy": 0.20705051422119142,
      "num_tokens": 94962764.0,
      "step": 48535
    },
    {
      "entropy": 5.897637605667114,
      "epoch": 4.852101764382466,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028909509756062273,
      "loss": 5.3369,
      "mean_token_accuracy": 0.2110355481505394,
      "num_tokens": 94973854.0,
      "step": 48540
    },
    {
      "entropy": 5.776208686828613,
      "epoch": 4.852601589443695,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028905947837947057,
      "loss": 5.1437,
      "mean_token_accuracy": 0.22290263921022416,
      "num_tokens": 94982858.0,
      "step": 48545
    },
    {
      "entropy": 5.735943794250488,
      "epoch": 4.853101414504923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002890238588445854,
      "loss": 5.2231,
      "mean_token_accuracy": 0.22925333082675933,
      "num_tokens": 94993606.0,
      "step": 48550
    },
    {
      "entropy": 5.83131628036499,
      "epoch": 4.853601239566152,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002889882389568638,
      "loss": 5.3024,
      "mean_token_accuracy": 0.2106259882450104,
      "num_tokens": 95004398.0,
      "step": 48555
    },
    {
      "entropy": 5.826123428344727,
      "epoch": 4.854101064627381,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002889526187172015,
      "loss": 5.1741,
      "mean_token_accuracy": 0.2227873384952545,
      "num_tokens": 95013468.0,
      "step": 48560
    },
    {
      "entropy": 5.828402519226074,
      "epoch": 4.854600889688609,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028891699812649493,
      "loss": 5.2815,
      "mean_token_accuracy": 0.21534111499786376,
      "num_tokens": 95023801.0,
      "step": 48565
    },
    {
      "entropy": 5.874734497070312,
      "epoch": 4.855100714749837,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002888813771856403,
      "loss": 5.2358,
      "mean_token_accuracy": 0.2139451414346695,
      "num_tokens": 95033769.0,
      "step": 48570
    },
    {
      "entropy": 5.804473352432251,
      "epoch": 4.855600539811066,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002888457558955338,
      "loss": 5.2076,
      "mean_token_accuracy": 0.22317716777324675,
      "num_tokens": 95043822.0,
      "step": 48575
    },
    {
      "entropy": 5.765447902679443,
      "epoch": 4.856100364872295,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028881013425707163,
      "loss": 5.1975,
      "mean_token_accuracy": 0.22244254052639006,
      "num_tokens": 95053439.0,
      "step": 48580
    },
    {
      "entropy": 5.739242887496948,
      "epoch": 4.856600189933523,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028877451227115003,
      "loss": 5.1247,
      "mean_token_accuracy": 0.22775428146123886,
      "num_tokens": 95063468.0,
      "step": 48585
    },
    {
      "entropy": 5.93995885848999,
      "epoch": 4.857100014994752,
      "grad_norm": 1.25,
      "learning_rate": 0.00028873888993866527,
      "loss": 5.3838,
      "mean_token_accuracy": 0.20269775390625,
      "num_tokens": 95073177.0,
      "step": 48590
    },
    {
      "entropy": 5.744062423706055,
      "epoch": 4.857599840055981,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002887032672605136,
      "loss": 5.2073,
      "mean_token_accuracy": 0.22419218122959136,
      "num_tokens": 95083851.0,
      "step": 48595
    },
    {
      "entropy": 5.754631900787354,
      "epoch": 4.858099665117209,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028866764423759116,
      "loss": 5.2726,
      "mean_token_accuracy": 0.22158454358577728,
      "num_tokens": 95094156.0,
      "step": 48600
    },
    {
      "entropy": 5.836094713211059,
      "epoch": 4.858599490178437,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002886320208707944,
      "loss": 5.1852,
      "mean_token_accuracy": 0.2231227993965149,
      "num_tokens": 95103533.0,
      "step": 48605
    },
    {
      "entropy": 5.830214500427246,
      "epoch": 4.859099315239666,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002885963971610194,
      "loss": 5.3405,
      "mean_token_accuracy": 0.20941996127367019,
      "num_tokens": 95111967.0,
      "step": 48610
    },
    {
      "entropy": 5.817248106002808,
      "epoch": 4.859599140300895,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028856077310916267,
      "loss": 5.2464,
      "mean_token_accuracy": 0.21594811230897903,
      "num_tokens": 95122054.0,
      "step": 48615
    },
    {
      "entropy": 5.8086510181427,
      "epoch": 4.860098965362123,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002885251487161203,
      "loss": 5.234,
      "mean_token_accuracy": 0.21611903309822084,
      "num_tokens": 95132185.0,
      "step": 48620
    },
    {
      "entropy": 5.803645086288452,
      "epoch": 4.860598790423352,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028848952398278865,
      "loss": 5.2265,
      "mean_token_accuracy": 0.2200348511338234,
      "num_tokens": 95142216.0,
      "step": 48625
    },
    {
      "entropy": 5.744081974029541,
      "epoch": 4.861098615484581,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028845389891006415,
      "loss": 5.1218,
      "mean_token_accuracy": 0.2279597580432892,
      "num_tokens": 95151385.0,
      "step": 48630
    },
    {
      "entropy": 5.840126276016235,
      "epoch": 4.861598440545809,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002884182734988429,
      "loss": 5.31,
      "mean_token_accuracy": 0.2115102455019951,
      "num_tokens": 95160241.0,
      "step": 48635
    },
    {
      "entropy": 5.807007694244385,
      "epoch": 4.862098265607037,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028838264775002143,
      "loss": 5.2266,
      "mean_token_accuracy": 0.2174956753849983,
      "num_tokens": 95169075.0,
      "step": 48640
    },
    {
      "entropy": 5.779291343688965,
      "epoch": 4.862598090668266,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028834702166449593,
      "loss": 5.1501,
      "mean_token_accuracy": 0.22250983715057374,
      "num_tokens": 95178025.0,
      "step": 48645
    },
    {
      "entropy": 5.7940582752227785,
      "epoch": 4.863097915729495,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002883113952431629,
      "loss": 5.2129,
      "mean_token_accuracy": 0.22144452184438707,
      "num_tokens": 95187105.0,
      "step": 48650
    },
    {
      "entropy": 5.7771800518035885,
      "epoch": 4.863597740790723,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028827576848691836,
      "loss": 5.1886,
      "mean_token_accuracy": 0.2164718344807625,
      "num_tokens": 95197115.0,
      "step": 48655
    },
    {
      "entropy": 5.846317386627197,
      "epoch": 4.864097565851952,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028824014139665913,
      "loss": 5.2525,
      "mean_token_accuracy": 0.2168487101793289,
      "num_tokens": 95205871.0,
      "step": 48660
    },
    {
      "entropy": 5.601408576965332,
      "epoch": 4.864597390913181,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002882045139732813,
      "loss": 5.0702,
      "mean_token_accuracy": 0.2394086703658104,
      "num_tokens": 95216119.0,
      "step": 48665
    },
    {
      "entropy": 5.775272560119629,
      "epoch": 4.865097215974409,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002881688862176812,
      "loss": 5.2491,
      "mean_token_accuracy": 0.21664804667234422,
      "num_tokens": 95225849.0,
      "step": 48670
    },
    {
      "entropy": 5.78877329826355,
      "epoch": 4.865597041035637,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002881332581307554,
      "loss": 5.2616,
      "mean_token_accuracy": 0.21728747189044953,
      "num_tokens": 95234527.0,
      "step": 48675
    },
    {
      "entropy": 5.663978242874146,
      "epoch": 4.866096866096866,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028809762971340015,
      "loss": 5.0688,
      "mean_token_accuracy": 0.24295303970575333,
      "num_tokens": 95244491.0,
      "step": 48680
    },
    {
      "entropy": 5.886888933181763,
      "epoch": 4.866596691158095,
      "grad_norm": 1.203125,
      "learning_rate": 0.000288062000966512,
      "loss": 5.2931,
      "mean_token_accuracy": 0.2141903355717659,
      "num_tokens": 95253452.0,
      "step": 48685
    },
    {
      "entropy": 5.71809229850769,
      "epoch": 4.867096516219323,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002880263718909871,
      "loss": 5.14,
      "mean_token_accuracy": 0.2361612156033516,
      "num_tokens": 95262514.0,
      "step": 48690
    },
    {
      "entropy": 5.696096706390381,
      "epoch": 4.867596341280552,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028799074248772214,
      "loss": 5.1166,
      "mean_token_accuracy": 0.22552868723869324,
      "num_tokens": 95271710.0,
      "step": 48695
    },
    {
      "entropy": 5.788973474502564,
      "epoch": 4.868096166341781,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028795511275761347,
      "loss": 5.2535,
      "mean_token_accuracy": 0.21309757828712464,
      "num_tokens": 95281527.0,
      "step": 48700
    },
    {
      "entropy": 5.740333127975464,
      "epoch": 4.868595991403009,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028791948270155747,
      "loss": 5.1952,
      "mean_token_accuracy": 0.21756596714258195,
      "num_tokens": 95290493.0,
      "step": 48705
    },
    {
      "entropy": 5.772379207611084,
      "epoch": 4.869095816464237,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028788385232045066,
      "loss": 5.1643,
      "mean_token_accuracy": 0.22880397588014603,
      "num_tokens": 95300177.0,
      "step": 48710
    },
    {
      "entropy": 5.773220586776733,
      "epoch": 4.869595641525466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002878482216151894,
      "loss": 5.1887,
      "mean_token_accuracy": 0.22276372760534285,
      "num_tokens": 95310117.0,
      "step": 48715
    },
    {
      "entropy": 5.906521320343018,
      "epoch": 4.870095466586695,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002878125905866701,
      "loss": 5.3451,
      "mean_token_accuracy": 0.2165274903178215,
      "num_tokens": 95320164.0,
      "step": 48720
    },
    {
      "entropy": 5.6391894817352295,
      "epoch": 4.870595291647923,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028777695923578944,
      "loss": 5.0163,
      "mean_token_accuracy": 0.23451687693595885,
      "num_tokens": 95329594.0,
      "step": 48725
    },
    {
      "entropy": 5.672423219680786,
      "epoch": 4.871095116709152,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002877413275634438,
      "loss": 5.1438,
      "mean_token_accuracy": 0.236103855073452,
      "num_tokens": 95339520.0,
      "step": 48730
    },
    {
      "entropy": 5.816253042221069,
      "epoch": 4.871594941770381,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002877056955705296,
      "loss": 5.2791,
      "mean_token_accuracy": 0.22243204116821289,
      "num_tokens": 95348349.0,
      "step": 48735
    },
    {
      "entropy": 5.783435821533203,
      "epoch": 4.872094766831609,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002876700632579433,
      "loss": 5.2026,
      "mean_token_accuracy": 0.22169436514377594,
      "num_tokens": 95358016.0,
      "step": 48740
    },
    {
      "entropy": 5.790235424041748,
      "epoch": 4.872594591892837,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002876344306265815,
      "loss": 5.2264,
      "mean_token_accuracy": 0.21904952079057693,
      "num_tokens": 95366942.0,
      "step": 48745
    },
    {
      "entropy": 5.8974127769470215,
      "epoch": 4.873094416954066,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028759879767734065,
      "loss": 5.284,
      "mean_token_accuracy": 0.2084643319249153,
      "num_tokens": 95375968.0,
      "step": 48750
    },
    {
      "entropy": 5.793096399307251,
      "epoch": 4.873594242015295,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002875631644111172,
      "loss": 5.1777,
      "mean_token_accuracy": 0.21885833144187927,
      "num_tokens": 95384648.0,
      "step": 48755
    },
    {
      "entropy": 5.798482227325439,
      "epoch": 4.874094067076523,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002875275308288079,
      "loss": 5.2887,
      "mean_token_accuracy": 0.2160811871290207,
      "num_tokens": 95395327.0,
      "step": 48760
    },
    {
      "entropy": 5.845179033279419,
      "epoch": 4.874593892137752,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028749189693130924,
      "loss": 5.2409,
      "mean_token_accuracy": 0.22304711639881133,
      "num_tokens": 95405501.0,
      "step": 48765
    },
    {
      "entropy": 5.793375015258789,
      "epoch": 4.875093717198981,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002874562627195176,
      "loss": 5.1055,
      "mean_token_accuracy": 0.23268313258886336,
      "num_tokens": 95414552.0,
      "step": 48770
    },
    {
      "entropy": 5.708884000778198,
      "epoch": 4.875593542260209,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028742062819432955,
      "loss": 5.1248,
      "mean_token_accuracy": 0.2260858327150345,
      "num_tokens": 95424732.0,
      "step": 48775
    },
    {
      "entropy": 5.793675088882447,
      "epoch": 4.876093367321437,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002873849933566417,
      "loss": 5.2255,
      "mean_token_accuracy": 0.21627459973096846,
      "num_tokens": 95434733.0,
      "step": 48780
    },
    {
      "entropy": 5.829454469680786,
      "epoch": 4.876593192382666,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002873493582073506,
      "loss": 5.2548,
      "mean_token_accuracy": 0.2126930609345436,
      "num_tokens": 95444557.0,
      "step": 48785
    },
    {
      "entropy": 5.973088979721069,
      "epoch": 4.877093017443895,
      "grad_norm": 1.125,
      "learning_rate": 0.0002873137227473527,
      "loss": 5.3756,
      "mean_token_accuracy": 0.21109994798898696,
      "num_tokens": 95453867.0,
      "step": 48790
    },
    {
      "entropy": 5.87469973564148,
      "epoch": 4.877592842505123,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002872780869775449,
      "loss": 5.2617,
      "mean_token_accuracy": 0.21541307866573334,
      "num_tokens": 95463129.0,
      "step": 48795
    },
    {
      "entropy": 5.784485912322998,
      "epoch": 4.878092667566352,
      "grad_norm": 1.25,
      "learning_rate": 0.0002872424508988236,
      "loss": 5.3087,
      "mean_token_accuracy": 0.21782918870449067,
      "num_tokens": 95473448.0,
      "step": 48800
    },
    {
      "entropy": 5.814182710647583,
      "epoch": 4.878592492627581,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002872068145120853,
      "loss": 5.2763,
      "mean_token_accuracy": 0.21795214116573333,
      "num_tokens": 95483648.0,
      "step": 48805
    },
    {
      "entropy": 5.826685190200806,
      "epoch": 4.879092317688809,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002871711778182267,
      "loss": 5.2722,
      "mean_token_accuracy": 0.2202293962240219,
      "num_tokens": 95493065.0,
      "step": 48810
    },
    {
      "entropy": 5.837626314163208,
      "epoch": 4.879592142750037,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002871355408181444,
      "loss": 5.2896,
      "mean_token_accuracy": 0.21809643507003784,
      "num_tokens": 95503399.0,
      "step": 48815
    },
    {
      "entropy": 5.8276091575622555,
      "epoch": 4.880091967811266,
      "grad_norm": 1.125,
      "learning_rate": 0.000287099903512735,
      "loss": 5.2527,
      "mean_token_accuracy": 0.2154105857014656,
      "num_tokens": 95513176.0,
      "step": 48820
    },
    {
      "entropy": 5.717273473739624,
      "epoch": 4.880591792872495,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028706426590289515,
      "loss": 5.1615,
      "mean_token_accuracy": 0.2311352849006653,
      "num_tokens": 95522687.0,
      "step": 48825
    },
    {
      "entropy": 5.776536750793457,
      "epoch": 4.881091617933723,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028702862798952143,
      "loss": 5.2526,
      "mean_token_accuracy": 0.21227207630872727,
      "num_tokens": 95532674.0,
      "step": 48830
    },
    {
      "entropy": 5.900971603393555,
      "epoch": 4.881591442994952,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002869929897735106,
      "loss": 5.3423,
      "mean_token_accuracy": 0.20236956477165222,
      "num_tokens": 95543460.0,
      "step": 48835
    },
    {
      "entropy": 5.71451587677002,
      "epoch": 4.882091268056181,
      "grad_norm": 1.125,
      "learning_rate": 0.00028695735125575924,
      "loss": 5.1256,
      "mean_token_accuracy": 0.22666216492652894,
      "num_tokens": 95554164.0,
      "step": 48840
    },
    {
      "entropy": 5.888636684417724,
      "epoch": 4.882591093117409,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028692171243716394,
      "loss": 5.3325,
      "mean_token_accuracy": 0.21909181028604507,
      "num_tokens": 95563415.0,
      "step": 48845
    },
    {
      "entropy": 5.814065742492676,
      "epoch": 4.883090918178637,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028688607331862144,
      "loss": 5.2463,
      "mean_token_accuracy": 0.21702223569154738,
      "num_tokens": 95573010.0,
      "step": 48850
    },
    {
      "entropy": 5.7581991195678714,
      "epoch": 4.883590743239866,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028685043390102845,
      "loss": 5.1434,
      "mean_token_accuracy": 0.22585611194372177,
      "num_tokens": 95583547.0,
      "step": 48855
    },
    {
      "entropy": 5.79010157585144,
      "epoch": 4.884090568301095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002868147941852815,
      "loss": 5.2023,
      "mean_token_accuracy": 0.22135880440473557,
      "num_tokens": 95593476.0,
      "step": 48860
    },
    {
      "entropy": 5.753040838241577,
      "epoch": 4.884590393362323,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002867791541722774,
      "loss": 5.1971,
      "mean_token_accuracy": 0.22090595960617065,
      "num_tokens": 95602970.0,
      "step": 48865
    },
    {
      "entropy": 5.808763837814331,
      "epoch": 4.885090218423552,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002867435138629129,
      "loss": 5.2099,
      "mean_token_accuracy": 0.21946002244949342,
      "num_tokens": 95612893.0,
      "step": 48870
    },
    {
      "entropy": 5.790735483169556,
      "epoch": 4.885590043484781,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002867078732580845,
      "loss": 5.2171,
      "mean_token_accuracy": 0.22043040543794631,
      "num_tokens": 95622940.0,
      "step": 48875
    },
    {
      "entropy": 5.798035430908203,
      "epoch": 4.886089868546009,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028667223235868906,
      "loss": 5.1959,
      "mean_token_accuracy": 0.22066187709569932,
      "num_tokens": 95632032.0,
      "step": 48880
    },
    {
      "entropy": 5.824421787261963,
      "epoch": 4.886589693607237,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028663659116562323,
      "loss": 5.2557,
      "mean_token_accuracy": 0.21703149676322936,
      "num_tokens": 95643632.0,
      "step": 48885
    },
    {
      "entropy": 5.760684442520142,
      "epoch": 4.887089518668466,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002866009496797838,
      "loss": 5.1845,
      "mean_token_accuracy": 0.2176496058702469,
      "num_tokens": 95653705.0,
      "step": 48890
    },
    {
      "entropy": 5.798341512680054,
      "epoch": 4.887589343729695,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002865653079020674,
      "loss": 5.2336,
      "mean_token_accuracy": 0.21880154460668563,
      "num_tokens": 95663959.0,
      "step": 48895
    },
    {
      "entropy": 5.789427995681763,
      "epoch": 4.888089168790923,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002865296658333709,
      "loss": 5.2154,
      "mean_token_accuracy": 0.22081068605184556,
      "num_tokens": 95674008.0,
      "step": 48900
    },
    {
      "entropy": 5.856543970108032,
      "epoch": 4.888588993852152,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028649402347459096,
      "loss": 5.2916,
      "mean_token_accuracy": 0.2134307950735092,
      "num_tokens": 95684398.0,
      "step": 48905
    },
    {
      "entropy": 5.800591611862183,
      "epoch": 4.889088818913381,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002864583808266244,
      "loss": 5.2148,
      "mean_token_accuracy": 0.22013709992170333,
      "num_tokens": 95693211.0,
      "step": 48910
    },
    {
      "entropy": 5.835326051712036,
      "epoch": 4.889588643974609,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028642273789036783,
      "loss": 5.277,
      "mean_token_accuracy": 0.2177064076066017,
      "num_tokens": 95702820.0,
      "step": 48915
    },
    {
      "entropy": 5.858919382095337,
      "epoch": 4.890088469035837,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028638709466671813,
      "loss": 5.2394,
      "mean_token_accuracy": 0.21524790227413176,
      "num_tokens": 95712453.0,
      "step": 48920
    },
    {
      "entropy": 5.77006049156189,
      "epoch": 4.890588294097066,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028635145115657207,
      "loss": 5.1737,
      "mean_token_accuracy": 0.22253418564796448,
      "num_tokens": 95722098.0,
      "step": 48925
    },
    {
      "entropy": 5.849715375900269,
      "epoch": 4.891088119158295,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028631580736082645,
      "loss": 5.2909,
      "mean_token_accuracy": 0.21729015856981276,
      "num_tokens": 95732483.0,
      "step": 48930
    },
    {
      "entropy": 5.781710958480835,
      "epoch": 4.891587944219523,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000286280163280378,
      "loss": 5.2571,
      "mean_token_accuracy": 0.21408986747264863,
      "num_tokens": 95743268.0,
      "step": 48935
    },
    {
      "entropy": 5.831787633895874,
      "epoch": 4.892087769280752,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028624451891612356,
      "loss": 5.3133,
      "mean_token_accuracy": 0.20501006543636321,
      "num_tokens": 95753199.0,
      "step": 48940
    },
    {
      "entropy": 5.843409776687622,
      "epoch": 4.892587594341981,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028620887426895994,
      "loss": 5.3139,
      "mean_token_accuracy": 0.20828796327114105,
      "num_tokens": 95762490.0,
      "step": 48945
    },
    {
      "entropy": 5.831299304962158,
      "epoch": 4.893087419403209,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002861732293397839,
      "loss": 5.295,
      "mean_token_accuracy": 0.21384789943695068,
      "num_tokens": 95772572.0,
      "step": 48950
    },
    {
      "entropy": 5.794043445587159,
      "epoch": 4.893587244464437,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002861375841294923,
      "loss": 5.1956,
      "mean_token_accuracy": 0.21823487132787706,
      "num_tokens": 95781667.0,
      "step": 48955
    },
    {
      "entropy": 5.817412328720093,
      "epoch": 4.894087069525666,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002861019386389819,
      "loss": 5.2744,
      "mean_token_accuracy": 0.21347686797380447,
      "num_tokens": 95792056.0,
      "step": 48960
    },
    {
      "entropy": 5.832958316802978,
      "epoch": 4.894586894586895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002860662928691496,
      "loss": 5.274,
      "mean_token_accuracy": 0.21321569085121156,
      "num_tokens": 95801226.0,
      "step": 48965
    },
    {
      "entropy": 5.8150187015533445,
      "epoch": 4.895086719648123,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028603064682089225,
      "loss": 5.2092,
      "mean_token_accuracy": 0.22176895290613174,
      "num_tokens": 95810364.0,
      "step": 48970
    },
    {
      "entropy": 5.894062089920044,
      "epoch": 4.895586544709352,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028599500049510666,
      "loss": 5.3288,
      "mean_token_accuracy": 0.2113850012421608,
      "num_tokens": 95818436.0,
      "step": 48975
    },
    {
      "entropy": 5.787242889404297,
      "epoch": 4.896086369770581,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002859593538926897,
      "loss": 5.1421,
      "mean_token_accuracy": 0.2279365912079811,
      "num_tokens": 95827586.0,
      "step": 48980
    },
    {
      "entropy": 5.807066869735718,
      "epoch": 4.896586194831809,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002859237070145381,
      "loss": 5.2419,
      "mean_token_accuracy": 0.21798657029867172,
      "num_tokens": 95838021.0,
      "step": 48985
    },
    {
      "entropy": 5.73924036026001,
      "epoch": 4.897086019893037,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028588805986154897,
      "loss": 5.1574,
      "mean_token_accuracy": 0.23219552338123323,
      "num_tokens": 95847590.0,
      "step": 48990
    },
    {
      "entropy": 5.8096027851104735,
      "epoch": 4.897585844954266,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000285852412434619,
      "loss": 5.309,
      "mean_token_accuracy": 0.20833013206720352,
      "num_tokens": 95857513.0,
      "step": 48995
    },
    {
      "entropy": 5.7248300075531,
      "epoch": 4.898085670015495,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028581676473464517,
      "loss": 5.1587,
      "mean_token_accuracy": 0.22920697927474976,
      "num_tokens": 95868373.0,
      "step": 49000
    },
    {
      "entropy": 5.8284485816955565,
      "epoch": 4.898585495076723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028578111676252427,
      "loss": 5.1864,
      "mean_token_accuracy": 0.21440774351358413,
      "num_tokens": 95878138.0,
      "step": 49005
    },
    {
      "entropy": 5.836620092391968,
      "epoch": 4.899085320137952,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002857454685191533,
      "loss": 5.2406,
      "mean_token_accuracy": 0.22037727236747742,
      "num_tokens": 95888099.0,
      "step": 49010
    },
    {
      "entropy": 5.710585880279541,
      "epoch": 4.899585145199181,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002857098200054291,
      "loss": 5.1359,
      "mean_token_accuracy": 0.22350738644599916,
      "num_tokens": 95897859.0,
      "step": 49015
    },
    {
      "entropy": 5.761330556869507,
      "epoch": 4.900084970260409,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028567417122224854,
      "loss": 5.2254,
      "mean_token_accuracy": 0.23668622821569443,
      "num_tokens": 95908831.0,
      "step": 49020
    },
    {
      "entropy": 5.869406747817993,
      "epoch": 4.900584795321637,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002856385221705086,
      "loss": 5.3226,
      "mean_token_accuracy": 0.21134999245405198,
      "num_tokens": 95919445.0,
      "step": 49025
    },
    {
      "entropy": 5.736685800552368,
      "epoch": 4.901084620382866,
      "grad_norm": 1.125,
      "learning_rate": 0.00028560287285110615,
      "loss": 5.1129,
      "mean_token_accuracy": 0.2309000685811043,
      "num_tokens": 95929269.0,
      "step": 49030
    },
    {
      "entropy": 5.745718002319336,
      "epoch": 4.901584445444095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002855672232649382,
      "loss": 5.1912,
      "mean_token_accuracy": 0.21666420251131058,
      "num_tokens": 95938949.0,
      "step": 49035
    },
    {
      "entropy": 5.567536401748657,
      "epoch": 4.902084270505323,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002855315734129016,
      "loss": 4.9371,
      "mean_token_accuracy": 0.24689733535051345,
      "num_tokens": 95949764.0,
      "step": 49040
    },
    {
      "entropy": 5.723054981231689,
      "epoch": 4.902584095566552,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028549592329589333,
      "loss": 5.1545,
      "mean_token_accuracy": 0.22530522644519807,
      "num_tokens": 95959269.0,
      "step": 49045
    },
    {
      "entropy": 5.816400480270386,
      "epoch": 4.903083920627781,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028546027291481035,
      "loss": 5.3588,
      "mean_token_accuracy": 0.20933820009231568,
      "num_tokens": 95969744.0,
      "step": 49050
    },
    {
      "entropy": 5.786762523651123,
      "epoch": 4.903583745689009,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002854246222705496,
      "loss": 5.1195,
      "mean_token_accuracy": 0.23428947031497954,
      "num_tokens": 95979132.0,
      "step": 49055
    },
    {
      "entropy": 5.76359486579895,
      "epoch": 4.904083570750237,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028538897136400797,
      "loss": 5.1328,
      "mean_token_accuracy": 0.22726152688264847,
      "num_tokens": 95988225.0,
      "step": 49060
    },
    {
      "entropy": 5.771934127807617,
      "epoch": 4.904583395811466,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028535332019608256,
      "loss": 5.1896,
      "mean_token_accuracy": 0.22550583332777024,
      "num_tokens": 95998892.0,
      "step": 49065
    },
    {
      "entropy": 5.811882638931275,
      "epoch": 4.905083220872695,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028531766876767025,
      "loss": 5.204,
      "mean_token_accuracy": 0.21081583499908446,
      "num_tokens": 96008823.0,
      "step": 49070
    },
    {
      "entropy": 5.797217226028442,
      "epoch": 4.905583045933923,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028528201707966807,
      "loss": 5.2075,
      "mean_token_accuracy": 0.22017315179109573,
      "num_tokens": 96019748.0,
      "step": 49075
    },
    {
      "entropy": 5.872135162353516,
      "epoch": 4.906082870995152,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028524636513297297,
      "loss": 5.393,
      "mean_token_accuracy": 0.19826363325119017,
      "num_tokens": 96029074.0,
      "step": 49080
    },
    {
      "entropy": 5.862930011749268,
      "epoch": 4.906582696056381,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028521071292848197,
      "loss": 5.2413,
      "mean_token_accuracy": 0.21903014332056045,
      "num_tokens": 96038394.0,
      "step": 49085
    },
    {
      "entropy": 5.781032419204712,
      "epoch": 4.907082521117609,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002851750604670921,
      "loss": 5.1922,
      "mean_token_accuracy": 0.21989633291959762,
      "num_tokens": 96048311.0,
      "step": 49090
    },
    {
      "entropy": 5.862139463424683,
      "epoch": 4.907582346178837,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002851394077497003,
      "loss": 5.3348,
      "mean_token_accuracy": 0.21006498336791993,
      "num_tokens": 96057578.0,
      "step": 49095
    },
    {
      "entropy": 5.801079797744751,
      "epoch": 4.908082171240066,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002851037547772035,
      "loss": 5.1338,
      "mean_token_accuracy": 0.22769396305084227,
      "num_tokens": 96065369.0,
      "step": 49100
    },
    {
      "entropy": 5.79930157661438,
      "epoch": 4.908581996301295,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028506810155049894,
      "loss": 5.2659,
      "mean_token_accuracy": 0.22033033668994903,
      "num_tokens": 96074546.0,
      "step": 49105
    },
    {
      "entropy": 5.778039121627808,
      "epoch": 4.909081821362523,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002850324480704835,
      "loss": 5.2428,
      "mean_token_accuracy": 0.21929406374692917,
      "num_tokens": 96083282.0,
      "step": 49110
    },
    {
      "entropy": 5.924678754806519,
      "epoch": 4.909581646423752,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002849967943380543,
      "loss": 5.3166,
      "mean_token_accuracy": 0.20968541502952576,
      "num_tokens": 96093700.0,
      "step": 49115
    },
    {
      "entropy": 5.888912725448608,
      "epoch": 4.910081471484981,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002849611403541083,
      "loss": 5.2489,
      "mean_token_accuracy": 0.21654735654592513,
      "num_tokens": 96102895.0,
      "step": 49120
    },
    {
      "entropy": 5.820675230026245,
      "epoch": 4.910581296546209,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028492548611954257,
      "loss": 5.2628,
      "mean_token_accuracy": 0.2230367913842201,
      "num_tokens": 96111341.0,
      "step": 49125
    },
    {
      "entropy": 5.8501701831817625,
      "epoch": 4.911081121607437,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002848898316352541,
      "loss": 5.2716,
      "mean_token_accuracy": 0.2202148735523224,
      "num_tokens": 96120619.0,
      "step": 49130
    },
    {
      "entropy": 5.802032709121704,
      "epoch": 4.911580946668666,
      "grad_norm": 1.375,
      "learning_rate": 0.00028485417690214004,
      "loss": 5.1763,
      "mean_token_accuracy": 0.22034551799297333,
      "num_tokens": 96129846.0,
      "step": 49135
    },
    {
      "entropy": 5.751726961135864,
      "epoch": 4.912080771729895,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002848185219210975,
      "loss": 5.2622,
      "mean_token_accuracy": 0.22226408421993255,
      "num_tokens": 96140063.0,
      "step": 49140
    },
    {
      "entropy": 5.796066999435425,
      "epoch": 4.912580596791123,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028478286669302336,
      "loss": 5.1615,
      "mean_token_accuracy": 0.22363665103912353,
      "num_tokens": 96148548.0,
      "step": 49145
    },
    {
      "entropy": 5.76943187713623,
      "epoch": 4.913080421852352,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028474721121881497,
      "loss": 5.1783,
      "mean_token_accuracy": 0.21840940564870834,
      "num_tokens": 96158233.0,
      "step": 49150
    },
    {
      "entropy": 5.786449766159057,
      "epoch": 4.91358024691358,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002847115554993691,
      "loss": 5.1745,
      "mean_token_accuracy": 0.2179355040192604,
      "num_tokens": 96168131.0,
      "step": 49155
    },
    {
      "entropy": 5.776802921295166,
      "epoch": 4.914080071974809,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002846758995355831,
      "loss": 5.1907,
      "mean_token_accuracy": 0.22274086028337478,
      "num_tokens": 96177234.0,
      "step": 49160
    },
    {
      "entropy": 5.808670330047607,
      "epoch": 4.914579897036037,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028464024332835384,
      "loss": 5.2582,
      "mean_token_accuracy": 0.2148997113108635,
      "num_tokens": 96187384.0,
      "step": 49165
    },
    {
      "entropy": 5.681070709228516,
      "epoch": 4.915079722097266,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028460458687857865,
      "loss": 5.0706,
      "mean_token_accuracy": 0.2322628304362297,
      "num_tokens": 96196868.0,
      "step": 49170
    },
    {
      "entropy": 5.775576639175415,
      "epoch": 4.915579547158495,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028456893018715453,
      "loss": 5.1687,
      "mean_token_accuracy": 0.22329991012811662,
      "num_tokens": 96206141.0,
      "step": 49175
    },
    {
      "entropy": 5.849920415878296,
      "epoch": 4.916079372219723,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002845332732549785,
      "loss": 5.2755,
      "mean_token_accuracy": 0.2134491428732872,
      "num_tokens": 96216331.0,
      "step": 49180
    },
    {
      "entropy": 5.884139680862427,
      "epoch": 4.916579197280952,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002844976160829478,
      "loss": 5.3024,
      "mean_token_accuracy": 0.20965173840522766,
      "num_tokens": 96226731.0,
      "step": 49185
    },
    {
      "entropy": 5.880835962295532,
      "epoch": 4.91707902234218,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028446195867195954,
      "loss": 5.292,
      "mean_token_accuracy": 0.2108748286962509,
      "num_tokens": 96236797.0,
      "step": 49190
    },
    {
      "entropy": 5.783713912963867,
      "epoch": 4.917578847403409,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002844263010229108,
      "loss": 5.21,
      "mean_token_accuracy": 0.22071605026721955,
      "num_tokens": 96247356.0,
      "step": 49195
    },
    {
      "entropy": 5.8797753810882565,
      "epoch": 4.918078672464637,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002843906431366988,
      "loss": 5.4118,
      "mean_token_accuracy": 0.20563421696424483,
      "num_tokens": 96257397.0,
      "step": 49200
    },
    {
      "entropy": 5.6372700214385985,
      "epoch": 4.918578497525866,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028435498501422064,
      "loss": 5.0257,
      "mean_token_accuracy": 0.23190736919641494,
      "num_tokens": 96268170.0,
      "step": 49205
    },
    {
      "entropy": 5.7423382759094235,
      "epoch": 4.919078322587095,
      "grad_norm": 1.125,
      "learning_rate": 0.0002843193266563734,
      "loss": 5.209,
      "mean_token_accuracy": 0.21944830268621446,
      "num_tokens": 96278172.0,
      "step": 49210
    },
    {
      "entropy": 5.876405477523804,
      "epoch": 4.919578147648323,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028428366806405436,
      "loss": 5.3904,
      "mean_token_accuracy": 0.20737043470144273,
      "num_tokens": 96287989.0,
      "step": 49215
    },
    {
      "entropy": 5.830421733856201,
      "epoch": 4.920077972709551,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028424800923816066,
      "loss": 5.2609,
      "mean_token_accuracy": 0.22240186333656312,
      "num_tokens": 96298934.0,
      "step": 49220
    },
    {
      "entropy": 5.7204179763793945,
      "epoch": 4.92057779777078,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028421235017958937,
      "loss": 5.1061,
      "mean_token_accuracy": 0.2367246478796005,
      "num_tokens": 96308130.0,
      "step": 49225
    },
    {
      "entropy": 5.835206365585327,
      "epoch": 4.921077622832009,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002841766908892377,
      "loss": 5.3083,
      "mean_token_accuracy": 0.2050669550895691,
      "num_tokens": 96317899.0,
      "step": 49230
    },
    {
      "entropy": 5.7853888988494875,
      "epoch": 4.921577447893237,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028414103136800283,
      "loss": 5.1733,
      "mean_token_accuracy": 0.21821951419115065,
      "num_tokens": 96326814.0,
      "step": 49235
    },
    {
      "entropy": 5.780437326431274,
      "epoch": 4.922077272954466,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028410537161678204,
      "loss": 5.1721,
      "mean_token_accuracy": 0.22976600080728532,
      "num_tokens": 96337825.0,
      "step": 49240
    },
    {
      "entropy": 5.816331052780152,
      "epoch": 4.922577098015695,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002840697116364724,
      "loss": 5.3018,
      "mean_token_accuracy": 0.22120245397090912,
      "num_tokens": 96347525.0,
      "step": 49245
    },
    {
      "entropy": 5.846827793121338,
      "epoch": 4.923076923076923,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002840340514279711,
      "loss": 5.3046,
      "mean_token_accuracy": 0.2131512477993965,
      "num_tokens": 96357633.0,
      "step": 49250
    },
    {
      "entropy": 5.830612468719482,
      "epoch": 4.923576748138151,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002839983909921755,
      "loss": 5.2627,
      "mean_token_accuracy": 0.20921504944562913,
      "num_tokens": 96366750.0,
      "step": 49255
    },
    {
      "entropy": 5.829584217071533,
      "epoch": 4.92407657319938,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002839627303299826,
      "loss": 5.2333,
      "mean_token_accuracy": 0.22373709380626677,
      "num_tokens": 96376057.0,
      "step": 49260
    },
    {
      "entropy": 5.70792932510376,
      "epoch": 4.924576398260609,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028392706944228976,
      "loss": 5.158,
      "mean_token_accuracy": 0.22778059989213945,
      "num_tokens": 96385917.0,
      "step": 49265
    },
    {
      "entropy": 5.818166208267212,
      "epoch": 4.925076223321837,
      "grad_norm": 1.046875,
      "learning_rate": 0.00028389140832999415,
      "loss": 5.2912,
      "mean_token_accuracy": 0.21185883283615112,
      "num_tokens": 96395370.0,
      "step": 49270
    },
    {
      "entropy": 5.788072872161865,
      "epoch": 4.925576048383066,
      "grad_norm": 1.234375,
      "learning_rate": 0.000283855746993993,
      "loss": 5.1537,
      "mean_token_accuracy": 0.22308834940195083,
      "num_tokens": 96404049.0,
      "step": 49275
    },
    {
      "entropy": 5.762806510925293,
      "epoch": 4.926075873444295,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002838200854351835,
      "loss": 5.2306,
      "mean_token_accuracy": 0.22412841618061066,
      "num_tokens": 96413166.0,
      "step": 49280
    },
    {
      "entropy": 5.746399927139282,
      "epoch": 4.926575698505523,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028378442365446294,
      "loss": 5.156,
      "mean_token_accuracy": 0.23635386824607849,
      "num_tokens": 96423509.0,
      "step": 49285
    },
    {
      "entropy": 5.835346603393555,
      "epoch": 4.927075523566751,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002837487616527286,
      "loss": 5.3032,
      "mean_token_accuracy": 0.21831198930740356,
      "num_tokens": 96434191.0,
      "step": 49290
    },
    {
      "entropy": 5.8964849472045895,
      "epoch": 4.92757534862798,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002837130994308776,
      "loss": 5.2467,
      "mean_token_accuracy": 0.21666827648878098,
      "num_tokens": 96443187.0,
      "step": 49295
    },
    {
      "entropy": 5.816350650787354,
      "epoch": 4.9280751736892086,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028367743698980727,
      "loss": 5.2149,
      "mean_token_accuracy": 0.21551958322525025,
      "num_tokens": 96452313.0,
      "step": 49300
    },
    {
      "entropy": 5.818268489837647,
      "epoch": 4.928574998750437,
      "grad_norm": 1.25,
      "learning_rate": 0.0002836417743304149,
      "loss": 5.2784,
      "mean_token_accuracy": 0.2102631151676178,
      "num_tokens": 96462185.0,
      "step": 49305
    },
    {
      "entropy": 5.7858985424041744,
      "epoch": 4.929074823811666,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002836061114535976,
      "loss": 5.2589,
      "mean_token_accuracy": 0.217421755194664,
      "num_tokens": 96472185.0,
      "step": 49310
    },
    {
      "entropy": 5.766721296310425,
      "epoch": 4.929574648872895,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002835704483602528,
      "loss": 5.1869,
      "mean_token_accuracy": 0.2245575338602066,
      "num_tokens": 96483876.0,
      "step": 49315
    },
    {
      "entropy": 5.776400136947632,
      "epoch": 4.930074473934123,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002835347850512778,
      "loss": 5.2338,
      "mean_token_accuracy": 0.23028988987207413,
      "num_tokens": 96494607.0,
      "step": 49320
    },
    {
      "entropy": 5.778414869308472,
      "epoch": 4.930574298995351,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028349912152756974,
      "loss": 5.1611,
      "mean_token_accuracy": 0.22163419276475907,
      "num_tokens": 96503857.0,
      "step": 49325
    },
    {
      "entropy": 5.783723163604736,
      "epoch": 4.93107412405658,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002834634577900259,
      "loss": 5.2753,
      "mean_token_accuracy": 0.21936400532722472,
      "num_tokens": 96514285.0,
      "step": 49330
    },
    {
      "entropy": 5.79677906036377,
      "epoch": 4.9315739491178086,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028342779383954374,
      "loss": 5.321,
      "mean_token_accuracy": 0.21622001081705094,
      "num_tokens": 96524858.0,
      "step": 49335
    },
    {
      "entropy": 5.829264116287232,
      "epoch": 4.932073774179037,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028339212967702046,
      "loss": 5.1992,
      "mean_token_accuracy": 0.2181227445602417,
      "num_tokens": 96533503.0,
      "step": 49340
    },
    {
      "entropy": 5.882691526412964,
      "epoch": 4.932573599240266,
      "grad_norm": 1.25,
      "learning_rate": 0.00028335646530335337,
      "loss": 5.21,
      "mean_token_accuracy": 0.21589916050434113,
      "num_tokens": 96543751.0,
      "step": 49345
    },
    {
      "entropy": 5.7171392917633055,
      "epoch": 4.933073424301495,
      "grad_norm": 1.375,
      "learning_rate": 0.00028332080071943964,
      "loss": 5.1918,
      "mean_token_accuracy": 0.22545184046030045,
      "num_tokens": 96553314.0,
      "step": 49350
    },
    {
      "entropy": 5.796466922760009,
      "epoch": 4.933573249362723,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002832851359261768,
      "loss": 5.3019,
      "mean_token_accuracy": 0.21644217222929002,
      "num_tokens": 96563005.0,
      "step": 49355
    },
    {
      "entropy": 5.876365709304809,
      "epoch": 4.934073074423951,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028324947092446214,
      "loss": 5.2902,
      "mean_token_accuracy": 0.2098639726638794,
      "num_tokens": 96572960.0,
      "step": 49360
    },
    {
      "entropy": 5.80910096168518,
      "epoch": 4.93457289948518,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028321380571519285,
      "loss": 5.1971,
      "mean_token_accuracy": 0.2200228065252304,
      "num_tokens": 96583526.0,
      "step": 49365
    },
    {
      "entropy": 5.762747716903687,
      "epoch": 4.9350727245464086,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002831781402992663,
      "loss": 5.1522,
      "mean_token_accuracy": 0.23095661401748657,
      "num_tokens": 96592200.0,
      "step": 49370
    },
    {
      "entropy": 5.837825441360474,
      "epoch": 4.935572549607637,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028314247467757986,
      "loss": 5.3073,
      "mean_token_accuracy": 0.21677956432104112,
      "num_tokens": 96601293.0,
      "step": 49375
    },
    {
      "entropy": 5.902195930480957,
      "epoch": 4.936072374668866,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002831068088510308,
      "loss": 5.275,
      "mean_token_accuracy": 0.21969014555215835,
      "num_tokens": 96611094.0,
      "step": 49380
    },
    {
      "entropy": 5.8353218078613285,
      "epoch": 4.936572199730095,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002830711428205165,
      "loss": 5.2335,
      "mean_token_accuracy": 0.20862644612789155,
      "num_tokens": 96620616.0,
      "step": 49385
    },
    {
      "entropy": 5.788969087600708,
      "epoch": 4.937072024791323,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028303547658693447,
      "loss": 5.1491,
      "mean_token_accuracy": 0.23049009889364241,
      "num_tokens": 96629587.0,
      "step": 49390
    },
    {
      "entropy": 5.781395483016968,
      "epoch": 4.937571849852551,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002829998101511818,
      "loss": 5.2281,
      "mean_token_accuracy": 0.2189992129802704,
      "num_tokens": 96639294.0,
      "step": 49395
    },
    {
      "entropy": 5.7622106075286865,
      "epoch": 4.93807167491378,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000282964143514156,
      "loss": 5.2946,
      "mean_token_accuracy": 0.2088506668806076,
      "num_tokens": 96648831.0,
      "step": 49400
    },
    {
      "entropy": 5.817257118225098,
      "epoch": 4.9385714999750085,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028292847667675444,
      "loss": 5.2284,
      "mean_token_accuracy": 0.21606225222349168,
      "num_tokens": 96658279.0,
      "step": 49405
    },
    {
      "entropy": 5.843402576446533,
      "epoch": 4.939071325036237,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002828928096398743,
      "loss": 5.2835,
      "mean_token_accuracy": 0.21634072214365005,
      "num_tokens": 96667063.0,
      "step": 49410
    },
    {
      "entropy": 5.709258890151977,
      "epoch": 4.939571150097466,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028285714240441326,
      "loss": 5.1597,
      "mean_token_accuracy": 0.23002966046333312,
      "num_tokens": 96676315.0,
      "step": 49415
    },
    {
      "entropy": 5.8023255348205565,
      "epoch": 4.940070975158695,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002828214749712685,
      "loss": 5.2436,
      "mean_token_accuracy": 0.2142302230000496,
      "num_tokens": 96686604.0,
      "step": 49420
    },
    {
      "entropy": 5.88891863822937,
      "epoch": 4.940570800219923,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002827858073413374,
      "loss": 5.2787,
      "mean_token_accuracy": 0.20908612459897996,
      "num_tokens": 96696123.0,
      "step": 49425
    },
    {
      "entropy": 5.902321434020996,
      "epoch": 4.941070625281151,
      "grad_norm": 1.25,
      "learning_rate": 0.0002827501395155174,
      "loss": 5.4049,
      "mean_token_accuracy": 0.2077740654349327,
      "num_tokens": 96706943.0,
      "step": 49430
    },
    {
      "entropy": 5.791131162643433,
      "epoch": 4.94157045034238,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002827144714947059,
      "loss": 5.1586,
      "mean_token_accuracy": 0.22532527446746825,
      "num_tokens": 96716351.0,
      "step": 49435
    },
    {
      "entropy": 5.931128311157226,
      "epoch": 4.9420702754036085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002826788032798003,
      "loss": 5.3191,
      "mean_token_accuracy": 0.2105119228363037,
      "num_tokens": 96725825.0,
      "step": 49440
    },
    {
      "entropy": 5.8344930648803714,
      "epoch": 4.942570100464837,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028264313487169797,
      "loss": 5.2641,
      "mean_token_accuracy": 0.21384575963020325,
      "num_tokens": 96734628.0,
      "step": 49445
    },
    {
      "entropy": 5.810029125213623,
      "epoch": 4.943069925526066,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028260746627129624,
      "loss": 5.3012,
      "mean_token_accuracy": 0.21647679209709167,
      "num_tokens": 96744836.0,
      "step": 49450
    },
    {
      "entropy": 5.933159923553466,
      "epoch": 4.943569750587295,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002825717974794927,
      "loss": 5.4136,
      "mean_token_accuracy": 0.20089834779500962,
      "num_tokens": 96754369.0,
      "step": 49455
    },
    {
      "entropy": 5.8682598114013675,
      "epoch": 4.944069575648523,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028253612849718474,
      "loss": 5.2931,
      "mean_token_accuracy": 0.21423530280590058,
      "num_tokens": 96764081.0,
      "step": 49460
    },
    {
      "entropy": 5.792017316818237,
      "epoch": 4.944569400709751,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002825004593252697,
      "loss": 5.1185,
      "mean_token_accuracy": 0.23331914991140365,
      "num_tokens": 96773835.0,
      "step": 49465
    },
    {
      "entropy": 5.825844383239746,
      "epoch": 4.94506922577098,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028246478996464503,
      "loss": 5.2323,
      "mean_token_accuracy": 0.2248783752322197,
      "num_tokens": 96783130.0,
      "step": 49470
    },
    {
      "entropy": 5.840491247177124,
      "epoch": 4.9455690508322085,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028242912041620804,
      "loss": 5.2825,
      "mean_token_accuracy": 0.2062183290719986,
      "num_tokens": 96793255.0,
      "step": 49475
    },
    {
      "entropy": 5.875999832153321,
      "epoch": 4.946068875893437,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002823934506808565,
      "loss": 5.2468,
      "mean_token_accuracy": 0.2129814714193344,
      "num_tokens": 96802136.0,
      "step": 49480
    },
    {
      "entropy": 5.762973022460938,
      "epoch": 4.946568700954666,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028235778075948756,
      "loss": 5.1733,
      "mean_token_accuracy": 0.22389235943555832,
      "num_tokens": 96811712.0,
      "step": 49485
    },
    {
      "entropy": 5.799658250808716,
      "epoch": 4.947068526015895,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028232211065299864,
      "loss": 5.2896,
      "mean_token_accuracy": 0.21529206931591033,
      "num_tokens": 96821034.0,
      "step": 49490
    },
    {
      "entropy": 5.749050903320312,
      "epoch": 4.947568351077123,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028228644036228746,
      "loss": 5.1691,
      "mean_token_accuracy": 0.2306289479136467,
      "num_tokens": 96831110.0,
      "step": 49495
    },
    {
      "entropy": 5.773786735534668,
      "epoch": 4.948068176138351,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028225076988825124,
      "loss": 5.1489,
      "mean_token_accuracy": 0.22925963252782822,
      "num_tokens": 96840376.0,
      "step": 49500
    },
    {
      "entropy": 5.748333358764649,
      "epoch": 4.94856800119958,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028221509923178744,
      "loss": 5.105,
      "mean_token_accuracy": 0.2392527312040329,
      "num_tokens": 96850552.0,
      "step": 49505
    },
    {
      "entropy": 5.792959070205688,
      "epoch": 4.9490678262608085,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002821794283937937,
      "loss": 5.241,
      "mean_token_accuracy": 0.21689738631248473,
      "num_tokens": 96860228.0,
      "step": 49510
    },
    {
      "entropy": 5.804396152496338,
      "epoch": 4.949567651322037,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002821437573751674,
      "loss": 5.1447,
      "mean_token_accuracy": 0.22155736088752748,
      "num_tokens": 96869422.0,
      "step": 49515
    },
    {
      "entropy": 5.7816568374633786,
      "epoch": 4.950067476383266,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028210808617680597,
      "loss": 5.1411,
      "mean_token_accuracy": 0.2209539070725441,
      "num_tokens": 96879112.0,
      "step": 49520
    },
    {
      "entropy": 5.8137232780456545,
      "epoch": 4.950567301444495,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002820724147996069,
      "loss": 5.2522,
      "mean_token_accuracy": 0.21238211244344712,
      "num_tokens": 96887668.0,
      "step": 49525
    },
    {
      "entropy": 5.802844381332397,
      "epoch": 4.951067126505723,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028203674324446776,
      "loss": 5.1902,
      "mean_token_accuracy": 0.2167900800704956,
      "num_tokens": 96897948.0,
      "step": 49530
    },
    {
      "entropy": 5.783225965499878,
      "epoch": 4.951566951566951,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028200107151228593,
      "loss": 5.2311,
      "mean_token_accuracy": 0.21774446219205856,
      "num_tokens": 96907835.0,
      "step": 49535
    },
    {
      "entropy": 5.787386083602906,
      "epoch": 4.95206677662818,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002819653996039589,
      "loss": 5.1698,
      "mean_token_accuracy": 0.22673189789056777,
      "num_tokens": 96917326.0,
      "step": 49540
    },
    {
      "entropy": 5.800004339218139,
      "epoch": 4.9525666016894085,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002819297275203842,
      "loss": 5.2199,
      "mean_token_accuracy": 0.2156592294573784,
      "num_tokens": 96926337.0,
      "step": 49545
    },
    {
      "entropy": 5.838723754882812,
      "epoch": 4.953066426750637,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002818940552624594,
      "loss": 5.3838,
      "mean_token_accuracy": 0.21190925240516661,
      "num_tokens": 96936354.0,
      "step": 49550
    },
    {
      "entropy": 5.846911191940308,
      "epoch": 4.953566251811866,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028185838283108195,
      "loss": 5.2317,
      "mean_token_accuracy": 0.2155361294746399,
      "num_tokens": 96946317.0,
      "step": 49555
    },
    {
      "entropy": 5.800507307052612,
      "epoch": 4.954066076873095,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002818227102271493,
      "loss": 5.2024,
      "mean_token_accuracy": 0.21927828937768937,
      "num_tokens": 96955305.0,
      "step": 49560
    },
    {
      "entropy": 5.869993019104004,
      "epoch": 4.954565901934323,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002817870374515591,
      "loss": 5.3368,
      "mean_token_accuracy": 0.2104467749595642,
      "num_tokens": 96965171.0,
      "step": 49565
    },
    {
      "entropy": 5.876949119567871,
      "epoch": 4.955065726995551,
      "grad_norm": 1.25,
      "learning_rate": 0.00028175136450520874,
      "loss": 5.2593,
      "mean_token_accuracy": 0.21819100081920623,
      "num_tokens": 96974795.0,
      "step": 49570
    },
    {
      "entropy": 5.764605712890625,
      "epoch": 4.95556555205678,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028171569138899573,
      "loss": 5.199,
      "mean_token_accuracy": 0.22068546563386918,
      "num_tokens": 96984847.0,
      "step": 49575
    },
    {
      "entropy": 5.849723243713379,
      "epoch": 4.9560653771180085,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028168001810381773,
      "loss": 5.3043,
      "mean_token_accuracy": 0.21056387275457383,
      "num_tokens": 96994457.0,
      "step": 49580
    },
    {
      "entropy": 5.872985219955444,
      "epoch": 4.956565202179237,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028164434465057214,
      "loss": 5.2973,
      "mean_token_accuracy": 0.21183741837739944,
      "num_tokens": 97003994.0,
      "step": 49585
    },
    {
      "entropy": 5.772653532028198,
      "epoch": 4.957065027240466,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002816086710301566,
      "loss": 5.1973,
      "mean_token_accuracy": 0.21931889802217483,
      "num_tokens": 97014338.0,
      "step": 49590
    },
    {
      "entropy": 5.856957244873047,
      "epoch": 4.957564852301695,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002815729972434686,
      "loss": 5.2965,
      "mean_token_accuracy": 0.21608444601297377,
      "num_tokens": 97024158.0,
      "step": 49595
    },
    {
      "entropy": 5.7457661628723145,
      "epoch": 4.958064677362923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028153732329140563,
      "loss": 5.2297,
      "mean_token_accuracy": 0.22045696675777435,
      "num_tokens": 97033303.0,
      "step": 49600
    },
    {
      "entropy": 5.881831312179566,
      "epoch": 4.958564502424151,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002815016491748653,
      "loss": 5.288,
      "mean_token_accuracy": 0.2070052966475487,
      "num_tokens": 97042375.0,
      "step": 49605
    },
    {
      "entropy": 5.851139163970947,
      "epoch": 4.95906432748538,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002814659748947452,
      "loss": 5.2407,
      "mean_token_accuracy": 0.21969990134239198,
      "num_tokens": 97051805.0,
      "step": 49610
    },
    {
      "entropy": 5.660289669036866,
      "epoch": 4.9595641525466085,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002814303004519427,
      "loss": 5.067,
      "mean_token_accuracy": 0.23649229407310485,
      "num_tokens": 97061335.0,
      "step": 49615
    },
    {
      "entropy": 5.783500909805298,
      "epoch": 4.960063977607837,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002813946258473556,
      "loss": 5.2231,
      "mean_token_accuracy": 0.2146571919322014,
      "num_tokens": 97070334.0,
      "step": 49620
    },
    {
      "entropy": 5.7664222240448,
      "epoch": 4.960563802669066,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028135895108188144,
      "loss": 5.2536,
      "mean_token_accuracy": 0.2176750421524048,
      "num_tokens": 97079527.0,
      "step": 49625
    },
    {
      "entropy": 5.84877872467041,
      "epoch": 4.961063627730295,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028132327615641755,
      "loss": 5.2508,
      "mean_token_accuracy": 0.2153997913002968,
      "num_tokens": 97089156.0,
      "step": 49630
    },
    {
      "entropy": 5.845317125320435,
      "epoch": 4.961563452791523,
      "grad_norm": 1.25,
      "learning_rate": 0.00028128760107186174,
      "loss": 5.325,
      "mean_token_accuracy": 0.21626098155975343,
      "num_tokens": 97098775.0,
      "step": 49635
    },
    {
      "entropy": 5.838347387313843,
      "epoch": 4.962063277852751,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028125192582911145,
      "loss": 5.2272,
      "mean_token_accuracy": 0.21823787540197373,
      "num_tokens": 97108305.0,
      "step": 49640
    },
    {
      "entropy": 5.847036647796631,
      "epoch": 4.96256310291398,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002812162504290644,
      "loss": 5.2544,
      "mean_token_accuracy": 0.21401134133338928,
      "num_tokens": 97118910.0,
      "step": 49645
    },
    {
      "entropy": 5.772724580764771,
      "epoch": 4.9630629279752085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028118057487261806,
      "loss": 5.2055,
      "mean_token_accuracy": 0.22109249383211135,
      "num_tokens": 97129078.0,
      "step": 49650
    },
    {
      "entropy": 5.763043165206909,
      "epoch": 4.963562753036437,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028114489916067006,
      "loss": 5.1473,
      "mean_token_accuracy": 0.22697223871946334,
      "num_tokens": 97138090.0,
      "step": 49655
    },
    {
      "entropy": 5.885312271118164,
      "epoch": 4.964062578097666,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028110922329411793,
      "loss": 5.3641,
      "mean_token_accuracy": 0.21150592416524888,
      "num_tokens": 97148774.0,
      "step": 49660
    },
    {
      "entropy": 5.8260327816009525,
      "epoch": 4.964562403158895,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002810735472738593,
      "loss": 5.2197,
      "mean_token_accuracy": 0.21813114434480668,
      "num_tokens": 97159521.0,
      "step": 49665
    },
    {
      "entropy": 5.801687145233155,
      "epoch": 4.965062228220123,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002810378711007918,
      "loss": 5.2022,
      "mean_token_accuracy": 0.22073320001363755,
      "num_tokens": 97168825.0,
      "step": 49670
    },
    {
      "entropy": 5.752173376083374,
      "epoch": 4.965562053281351,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002810021947758131,
      "loss": 5.1683,
      "mean_token_accuracy": 0.2188665971159935,
      "num_tokens": 97178304.0,
      "step": 49675
    },
    {
      "entropy": 5.754572582244873,
      "epoch": 4.96606187834258,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002809665182998206,
      "loss": 5.1883,
      "mean_token_accuracy": 0.2178625538945198,
      "num_tokens": 97188846.0,
      "step": 49680
    },
    {
      "entropy": 5.804746961593628,
      "epoch": 4.9665617034038085,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028093084167371215,
      "loss": 5.2554,
      "mean_token_accuracy": 0.21718631833791732,
      "num_tokens": 97198086.0,
      "step": 49685
    },
    {
      "entropy": 5.7970240116119385,
      "epoch": 4.967061528465037,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002808951648983852,
      "loss": 5.2902,
      "mean_token_accuracy": 0.21884737610816957,
      "num_tokens": 97208586.0,
      "step": 49690
    },
    {
      "entropy": 5.801386165618896,
      "epoch": 4.967561353526266,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002808594879747374,
      "loss": 5.1997,
      "mean_token_accuracy": 0.21758804321289063,
      "num_tokens": 97217799.0,
      "step": 49695
    },
    {
      "entropy": 5.909014368057251,
      "epoch": 4.968061178587495,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002808238109036663,
      "loss": 5.3077,
      "mean_token_accuracy": 0.21519827097654343,
      "num_tokens": 97226960.0,
      "step": 49700
    },
    {
      "entropy": 5.646469020843506,
      "epoch": 4.968561003648723,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002807881336860698,
      "loss": 5.0092,
      "mean_token_accuracy": 0.2403663694858551,
      "num_tokens": 97237075.0,
      "step": 49705
    },
    {
      "entropy": 5.777372694015503,
      "epoch": 4.969060828709951,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028075245632284523,
      "loss": 5.1905,
      "mean_token_accuracy": 0.23193349093198776,
      "num_tokens": 97247488.0,
      "step": 49710
    },
    {
      "entropy": 5.757132959365845,
      "epoch": 4.96956065377118,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002807167788148904,
      "loss": 5.1453,
      "mean_token_accuracy": 0.22517655193805694,
      "num_tokens": 97257356.0,
      "step": 49715
    },
    {
      "entropy": 5.783955335617065,
      "epoch": 4.9700604788324085,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028068110116310287,
      "loss": 5.1811,
      "mean_token_accuracy": 0.22873401492834092,
      "num_tokens": 97266729.0,
      "step": 49720
    },
    {
      "entropy": 5.791501522064209,
      "epoch": 4.970560303893637,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002806454233683803,
      "loss": 5.1788,
      "mean_token_accuracy": 0.22235502153635026,
      "num_tokens": 97275761.0,
      "step": 49725
    },
    {
      "entropy": 5.890665197372437,
      "epoch": 4.971060128954866,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002806097454316202,
      "loss": 5.262,
      "mean_token_accuracy": 0.21961592733860016,
      "num_tokens": 97285581.0,
      "step": 49730
    },
    {
      "entropy": 5.825987100601196,
      "epoch": 4.971559954016095,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002805740673537205,
      "loss": 5.2041,
      "mean_token_accuracy": 0.2196364849805832,
      "num_tokens": 97295344.0,
      "step": 49735
    },
    {
      "entropy": 5.801621580123902,
      "epoch": 4.972059779077323,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002805383891355787,
      "loss": 5.2209,
      "mean_token_accuracy": 0.21684653609991072,
      "num_tokens": 97305193.0,
      "step": 49740
    },
    {
      "entropy": 5.763424396514893,
      "epoch": 4.972559604138551,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028050271077809244,
      "loss": 5.2566,
      "mean_token_accuracy": 0.2190885990858078,
      "num_tokens": 97315159.0,
      "step": 49745
    },
    {
      "entropy": 5.843927907943725,
      "epoch": 4.97305942919978,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002804670322821594,
      "loss": 5.2722,
      "mean_token_accuracy": 0.21417490690946578,
      "num_tokens": 97324783.0,
      "step": 49750
    },
    {
      "entropy": 5.784912776947022,
      "epoch": 4.9735592542610085,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028043135364867717,
      "loss": 5.205,
      "mean_token_accuracy": 0.2193819910287857,
      "num_tokens": 97333484.0,
      "step": 49755
    },
    {
      "entropy": 5.796416616439819,
      "epoch": 4.974059079322237,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00028039567487854347,
      "loss": 5.1516,
      "mean_token_accuracy": 0.22273956388235092,
      "num_tokens": 97344061.0,
      "step": 49760
    },
    {
      "entropy": 5.793261289596558,
      "epoch": 4.974558904383466,
      "grad_norm": 1.296875,
      "learning_rate": 0.000280359995972656,
      "loss": 5.2277,
      "mean_token_accuracy": 0.2196686699986458,
      "num_tokens": 97353586.0,
      "step": 49765
    },
    {
      "entropy": 5.772048664093018,
      "epoch": 4.975058729444695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028032431693191244,
      "loss": 5.2303,
      "mean_token_accuracy": 0.22576259076595306,
      "num_tokens": 97364014.0,
      "step": 49770
    },
    {
      "entropy": 5.764096212387085,
      "epoch": 4.975558554505923,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002802886377572105,
      "loss": 5.1539,
      "mean_token_accuracy": 0.2232454627752304,
      "num_tokens": 97373530.0,
      "step": 49775
    },
    {
      "entropy": 5.7739458084106445,
      "epoch": 4.976058379567151,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002802529584494477,
      "loss": 5.2001,
      "mean_token_accuracy": 0.2221078023314476,
      "num_tokens": 97382891.0,
      "step": 49780
    },
    {
      "entropy": 5.81282901763916,
      "epoch": 4.97655820462838,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002802172790095219,
      "loss": 5.2483,
      "mean_token_accuracy": 0.21878458559513092,
      "num_tokens": 97392202.0,
      "step": 49785
    },
    {
      "entropy": 5.874989652633667,
      "epoch": 4.9770580296896085,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028018159943833047,
      "loss": 5.2893,
      "mean_token_accuracy": 0.20616048127412795,
      "num_tokens": 97402068.0,
      "step": 49790
    },
    {
      "entropy": 5.874148511886597,
      "epoch": 4.977557854750837,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002801459197367716,
      "loss": 5.3155,
      "mean_token_accuracy": 0.20875300019979476,
      "num_tokens": 97411873.0,
      "step": 49795
    },
    {
      "entropy": 5.840721225738525,
      "epoch": 4.978057679812066,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002801102399057426,
      "loss": 5.274,
      "mean_token_accuracy": 0.21610419005155562,
      "num_tokens": 97422393.0,
      "step": 49800
    },
    {
      "entropy": 5.882505416870117,
      "epoch": 4.978557504873295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028007455994614127,
      "loss": 5.2707,
      "mean_token_accuracy": 0.21225162148475646,
      "num_tokens": 97431140.0,
      "step": 49805
    },
    {
      "entropy": 5.848910284042359,
      "epoch": 4.979057329934523,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002800388798588654,
      "loss": 5.2412,
      "mean_token_accuracy": 0.21660330146551132,
      "num_tokens": 97440271.0,
      "step": 49810
    },
    {
      "entropy": 5.805694818496704,
      "epoch": 4.979557154995751,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028000319964481253,
      "loss": 5.2394,
      "mean_token_accuracy": 0.22297109365463258,
      "num_tokens": 97451516.0,
      "step": 49815
    },
    {
      "entropy": 5.76812162399292,
      "epoch": 4.98005698005698,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027996751930488045,
      "loss": 5.1974,
      "mean_token_accuracy": 0.22318131923675538,
      "num_tokens": 97460663.0,
      "step": 49820
    },
    {
      "entropy": 5.922189712524414,
      "epoch": 4.9805568051182085,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027993183883996686,
      "loss": 5.4495,
      "mean_token_accuracy": 0.202401864528656,
      "num_tokens": 97471476.0,
      "step": 49825
    },
    {
      "entropy": 5.797245264053345,
      "epoch": 4.981056630179437,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027989615825096953,
      "loss": 5.2541,
      "mean_token_accuracy": 0.22420429885387422,
      "num_tokens": 97480927.0,
      "step": 49830
    },
    {
      "entropy": 5.722223854064941,
      "epoch": 4.981556455240666,
      "grad_norm": 1.125,
      "learning_rate": 0.00027986047753878607,
      "loss": 5.073,
      "mean_token_accuracy": 0.2375298798084259,
      "num_tokens": 97489972.0,
      "step": 49835
    },
    {
      "entropy": 5.869152498245239,
      "epoch": 4.982056280301895,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027982479670431425,
      "loss": 5.3001,
      "mean_token_accuracy": 0.21225457042455673,
      "num_tokens": 97499917.0,
      "step": 49840
    },
    {
      "entropy": 5.828995132446289,
      "epoch": 4.982556105363123,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002797891157484519,
      "loss": 5.2232,
      "mean_token_accuracy": 0.2202902153134346,
      "num_tokens": 97509915.0,
      "step": 49845
    },
    {
      "entropy": 5.815642356872559,
      "epoch": 4.983055930424351,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002797534346720965,
      "loss": 5.2005,
      "mean_token_accuracy": 0.22636737078428268,
      "num_tokens": 97519887.0,
      "step": 49850
    },
    {
      "entropy": 5.74957594871521,
      "epoch": 4.98355575548558,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027971775347614586,
      "loss": 5.1204,
      "mean_token_accuracy": 0.22354640066623688,
      "num_tokens": 97529527.0,
      "step": 49855
    },
    {
      "entropy": 5.762928628921509,
      "epoch": 4.9840555805468085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002796820721614978,
      "loss": 5.2359,
      "mean_token_accuracy": 0.2159373328089714,
      "num_tokens": 97538322.0,
      "step": 49860
    },
    {
      "entropy": 5.732376766204834,
      "epoch": 4.984555405608037,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027964639072905007,
      "loss": 5.1274,
      "mean_token_accuracy": 0.2222355931997299,
      "num_tokens": 97547084.0,
      "step": 49865
    },
    {
      "entropy": 5.876184701919556,
      "epoch": 4.985055230669266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027961070917970033,
      "loss": 5.3177,
      "mean_token_accuracy": 0.21249826848506928,
      "num_tokens": 97555902.0,
      "step": 49870
    },
    {
      "entropy": 5.738709878921509,
      "epoch": 4.985555055730495,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027957502751434633,
      "loss": 5.2238,
      "mean_token_accuracy": 0.2217195987701416,
      "num_tokens": 97567090.0,
      "step": 49875
    },
    {
      "entropy": 5.813723802566528,
      "epoch": 4.986054880791723,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002795393457338858,
      "loss": 5.2212,
      "mean_token_accuracy": 0.2086929649114609,
      "num_tokens": 97576132.0,
      "step": 49880
    },
    {
      "entropy": 5.85732626914978,
      "epoch": 4.986554705852951,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002795036638392165,
      "loss": 5.1911,
      "mean_token_accuracy": 0.2252056360244751,
      "num_tokens": 97586011.0,
      "step": 49885
    },
    {
      "entropy": 5.791554546356201,
      "epoch": 4.98705453091418,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002794679818312362,
      "loss": 5.2295,
      "mean_token_accuracy": 0.22292397022247315,
      "num_tokens": 97596587.0,
      "step": 49890
    },
    {
      "entropy": 5.885250091552734,
      "epoch": 4.9875543559754085,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027943229971084255,
      "loss": 5.3185,
      "mean_token_accuracy": 0.2131537526845932,
      "num_tokens": 97606023.0,
      "step": 49895
    },
    {
      "entropy": 5.810583829879761,
      "epoch": 4.988054181036637,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027939661747893344,
      "loss": 5.18,
      "mean_token_accuracy": 0.22251253128051757,
      "num_tokens": 97614817.0,
      "step": 49900
    },
    {
      "entropy": 5.82873044013977,
      "epoch": 4.988554006097866,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027936093513640655,
      "loss": 5.2718,
      "mean_token_accuracy": 0.21579505056142806,
      "num_tokens": 97624129.0,
      "step": 49905
    },
    {
      "entropy": 5.794785070419311,
      "epoch": 4.989053831159095,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002793252526841597,
      "loss": 5.2669,
      "mean_token_accuracy": 0.2153763696551323,
      "num_tokens": 97633874.0,
      "step": 49910
    },
    {
      "entropy": 5.818808698654175,
      "epoch": 4.989553656220323,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002792895701230906,
      "loss": 5.1847,
      "mean_token_accuracy": 0.2211230367422104,
      "num_tokens": 97643580.0,
      "step": 49915
    },
    {
      "entropy": 5.791567611694336,
      "epoch": 4.990053481281551,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027925388745409695,
      "loss": 5.1721,
      "mean_token_accuracy": 0.22519259601831437,
      "num_tokens": 97653402.0,
      "step": 49920
    },
    {
      "entropy": 5.84647536277771,
      "epoch": 4.99055330634278,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002792182046780767,
      "loss": 5.3653,
      "mean_token_accuracy": 0.2099542036652565,
      "num_tokens": 97663070.0,
      "step": 49925
    },
    {
      "entropy": 5.763317728042603,
      "epoch": 4.9910531314040085,
      "grad_norm": 1.125,
      "learning_rate": 0.0002791825217959274,
      "loss": 5.1189,
      "mean_token_accuracy": 0.2262927085161209,
      "num_tokens": 97673111.0,
      "step": 49930
    },
    {
      "entropy": 5.751434803009033,
      "epoch": 4.991552956465237,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002791468388085469,
      "loss": 5.1925,
      "mean_token_accuracy": 0.22298935651779175,
      "num_tokens": 97682442.0,
      "step": 49935
    },
    {
      "entropy": 5.739949703216553,
      "epoch": 4.992052781526466,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002791111557168331,
      "loss": 5.1182,
      "mean_token_accuracy": 0.2291396826505661,
      "num_tokens": 97692083.0,
      "step": 49940
    },
    {
      "entropy": 5.77383189201355,
      "epoch": 4.992552606587695,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002790754725216836,
      "loss": 5.2465,
      "mean_token_accuracy": 0.21497172862291336,
      "num_tokens": 97700626.0,
      "step": 49945
    },
    {
      "entropy": 5.861896514892578,
      "epoch": 4.993052431648923,
      "grad_norm": 1.03125,
      "learning_rate": 0.00027903978922399635,
      "loss": 5.2691,
      "mean_token_accuracy": 0.21284680515527726,
      "num_tokens": 97710353.0,
      "step": 49950
    },
    {
      "entropy": 5.865693044662476,
      "epoch": 4.993552256710151,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027900410582466903,
      "loss": 5.3064,
      "mean_token_accuracy": 0.2096031501889229,
      "num_tokens": 97719807.0,
      "step": 49955
    },
    {
      "entropy": 5.791491985321045,
      "epoch": 4.99405208177138,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002789684223245993,
      "loss": 5.1957,
      "mean_token_accuracy": 0.2220311924815178,
      "num_tokens": 97730780.0,
      "step": 49960
    },
    {
      "entropy": 5.800440549850464,
      "epoch": 4.9945519068326085,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027893273872468523,
      "loss": 5.2287,
      "mean_token_accuracy": 0.21091035455465318,
      "num_tokens": 97741369.0,
      "step": 49965
    },
    {
      "entropy": 5.879392957687378,
      "epoch": 4.995051731893837,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002788970550258244,
      "loss": 5.3365,
      "mean_token_accuracy": 0.21447816193103791,
      "num_tokens": 97751734.0,
      "step": 49970
    },
    {
      "entropy": 5.773026418685913,
      "epoch": 4.995551556955066,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002788613712289147,
      "loss": 5.1918,
      "mean_token_accuracy": 0.2264362931251526,
      "num_tokens": 97760985.0,
      "step": 49975
    },
    {
      "entropy": 5.814297437667847,
      "epoch": 4.996051382016295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002788256873348539,
      "loss": 5.2202,
      "mean_token_accuracy": 0.22046064585447311,
      "num_tokens": 97771590.0,
      "step": 49980
    },
    {
      "entropy": 5.799531269073486,
      "epoch": 4.996551207077523,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002787900033445398,
      "loss": 5.2397,
      "mean_token_accuracy": 0.21671798676252366,
      "num_tokens": 97780477.0,
      "step": 49985
    },
    {
      "entropy": 5.797877931594849,
      "epoch": 4.997051032138751,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002787543192588702,
      "loss": 5.1977,
      "mean_token_accuracy": 0.21981725692749024,
      "num_tokens": 97790006.0,
      "step": 49990
    },
    {
      "entropy": 5.827411222457886,
      "epoch": 4.99755085719998,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027871863507874286,
      "loss": 5.2226,
      "mean_token_accuracy": 0.21992795616388322,
      "num_tokens": 97799239.0,
      "step": 49995
    },
    {
      "entropy": 5.830237245559692,
      "epoch": 4.9980506822612085,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027868295080505555,
      "loss": 5.2298,
      "mean_token_accuracy": 0.22654162496328353,
      "num_tokens": 97808181.0,
      "step": 50000
    },
    {
      "entropy": 5.724841070175171,
      "epoch": 4.998550507322437,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027864726643870625,
      "loss": 5.1438,
      "mean_token_accuracy": 0.22566040754318237,
      "num_tokens": 97818333.0,
      "step": 50005
    },
    {
      "entropy": 5.720878553390503,
      "epoch": 4.999050332383666,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027861158198059263,
      "loss": 5.2285,
      "mean_token_accuracy": 0.2145905077457428,
      "num_tokens": 97829309.0,
      "step": 50010
    },
    {
      "entropy": 5.727563762664795,
      "epoch": 4.999550157444895,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002785758974316125,
      "loss": 5.1311,
      "mean_token_accuracy": 0.21973737925291062,
      "num_tokens": 97839137.0,
      "step": 50015
    },
    {
      "entropy": 5.808529324001736,
      "epoch": 5.0,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002785402127926638,
      "loss": 5.2447,
      "mean_token_accuracy": 0.21930259300602806,
      "num_tokens": 97848300.0,
      "step": 50020
    },
    {
      "entropy": 5.741339826583863,
      "epoch": 5.000499825061229,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002785045280646442,
      "loss": 5.0681,
      "mean_token_accuracy": 0.23803022205829621,
      "num_tokens": 97857050.0,
      "step": 50025
    },
    {
      "entropy": 5.7899247169494625,
      "epoch": 5.000999650122457,
      "grad_norm": 1.375,
      "learning_rate": 0.00027846884324845157,
      "loss": 5.1738,
      "mean_token_accuracy": 0.22954004108905793,
      "num_tokens": 97865142.0,
      "step": 50030
    },
    {
      "entropy": 5.7323685646057125,
      "epoch": 5.001499475183686,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027843315834498374,
      "loss": 5.0449,
      "mean_token_accuracy": 0.2375782996416092,
      "num_tokens": 97875189.0,
      "step": 50035
    },
    {
      "entropy": 5.82108588218689,
      "epoch": 5.001999300244914,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027839747335513856,
      "loss": 5.1561,
      "mean_token_accuracy": 0.2255665123462677,
      "num_tokens": 97885112.0,
      "step": 50040
    },
    {
      "entropy": 5.774353313446045,
      "epoch": 5.002499125306143,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027836178827981374,
      "loss": 5.1235,
      "mean_token_accuracy": 0.23034408688545227,
      "num_tokens": 97895213.0,
      "step": 50045
    },
    {
      "entropy": 5.832403326034546,
      "epoch": 5.002998950367371,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027832610311990727,
      "loss": 5.2143,
      "mean_token_accuracy": 0.22090688049793245,
      "num_tokens": 97905484.0,
      "step": 50050
    },
    {
      "entropy": 5.670369577407837,
      "epoch": 5.0034987754286,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002782904178763169,
      "loss": 5.0166,
      "mean_token_accuracy": 0.24508069604635238,
      "num_tokens": 97914867.0,
      "step": 50055
    },
    {
      "entropy": 5.839645481109619,
      "epoch": 5.003998600489829,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027825473254994044,
      "loss": 5.1925,
      "mean_token_accuracy": 0.22485744059085847,
      "num_tokens": 97923883.0,
      "step": 50060
    },
    {
      "entropy": 5.768805074691772,
      "epoch": 5.004498425551057,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002782190471416758,
      "loss": 5.2072,
      "mean_token_accuracy": 0.22247035950422286,
      "num_tokens": 97934639.0,
      "step": 50065
    },
    {
      "entropy": 5.70389575958252,
      "epoch": 5.004998250612286,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002781833616524207,
      "loss": 5.0347,
      "mean_token_accuracy": 0.23748814910650254,
      "num_tokens": 97944982.0,
      "step": 50070
    },
    {
      "entropy": 5.798644018173218,
      "epoch": 5.005498075673514,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027814767608307295,
      "loss": 5.1143,
      "mean_token_accuracy": 0.22757604271173476,
      "num_tokens": 97955135.0,
      "step": 50075
    },
    {
      "entropy": 5.730957508087158,
      "epoch": 5.005997900734743,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027811199043453066,
      "loss": 5.1376,
      "mean_token_accuracy": 0.2266934558749199,
      "num_tokens": 97964365.0,
      "step": 50080
    },
    {
      "entropy": 5.904668188095092,
      "epoch": 5.006497725795971,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002780763047076914,
      "loss": 5.2786,
      "mean_token_accuracy": 0.21790696829557418,
      "num_tokens": 97974535.0,
      "step": 50085
    },
    {
      "entropy": 5.746056079864502,
      "epoch": 5.0069975508572,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002780406189034532,
      "loss": 5.0724,
      "mean_token_accuracy": 0.22885963171720505,
      "num_tokens": 97983552.0,
      "step": 50090
    },
    {
      "entropy": 5.722314167022705,
      "epoch": 5.007497375918429,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027800493302271366,
      "loss": 4.9959,
      "mean_token_accuracy": 0.24254221767187117,
      "num_tokens": 97993205.0,
      "step": 50095
    },
    {
      "entropy": 5.820800161361694,
      "epoch": 5.007997200979657,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027796924706637087,
      "loss": 5.1728,
      "mean_token_accuracy": 0.22214791625738145,
      "num_tokens": 98002726.0,
      "step": 50100
    },
    {
      "entropy": 5.79942889213562,
      "epoch": 5.008497026040886,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027793356103532254,
      "loss": 5.1546,
      "mean_token_accuracy": 0.23556865602731705,
      "num_tokens": 98013387.0,
      "step": 50105
    },
    {
      "entropy": 5.788008737564087,
      "epoch": 5.008996851102114,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002778978749304666,
      "loss": 5.1133,
      "mean_token_accuracy": 0.22722418159246444,
      "num_tokens": 98022736.0,
      "step": 50110
    },
    {
      "entropy": 5.631098747253418,
      "epoch": 5.009496676163343,
      "grad_norm": 1.125,
      "learning_rate": 0.00027786218875270084,
      "loss": 4.9937,
      "mean_token_accuracy": 0.24355061799287797,
      "num_tokens": 98032928.0,
      "step": 50115
    },
    {
      "entropy": 5.827776002883911,
      "epoch": 5.009996501224571,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027782650250292323,
      "loss": 5.2012,
      "mean_token_accuracy": 0.22098876386880875,
      "num_tokens": 98042631.0,
      "step": 50120
    },
    {
      "entropy": 5.66122350692749,
      "epoch": 5.0104963262858,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002777908161820315,
      "loss": 5.1517,
      "mean_token_accuracy": 0.22542708069086076,
      "num_tokens": 98053147.0,
      "step": 50125
    },
    {
      "entropy": 5.719797086715698,
      "epoch": 5.010996151347029,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002777551297909236,
      "loss": 5.0992,
      "mean_token_accuracy": 0.22989799976348876,
      "num_tokens": 98063698.0,
      "step": 50130
    },
    {
      "entropy": 5.904319190979004,
      "epoch": 5.011495976408257,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002777194433304972,
      "loss": 5.2144,
      "mean_token_accuracy": 0.2253866046667099,
      "num_tokens": 98072420.0,
      "step": 50135
    },
    {
      "entropy": 5.801671266555786,
      "epoch": 5.011995801469486,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002776837568016504,
      "loss": 5.1789,
      "mean_token_accuracy": 0.2220448836684227,
      "num_tokens": 98082174.0,
      "step": 50140
    },
    {
      "entropy": 5.7076630115509035,
      "epoch": 5.012495626530714,
      "grad_norm": 1.203125,
      "learning_rate": 0.000277648070205281,
      "loss": 5.0182,
      "mean_token_accuracy": 0.24006677120923997,
      "num_tokens": 98091251.0,
      "step": 50145
    },
    {
      "entropy": 5.783073139190674,
      "epoch": 5.012995451591943,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002776123835422868,
      "loss": 5.1464,
      "mean_token_accuracy": 0.22449579685926438,
      "num_tokens": 98101482.0,
      "step": 50150
    },
    {
      "entropy": 5.851728630065918,
      "epoch": 5.013495276653171,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002775766968135657,
      "loss": 5.1627,
      "mean_token_accuracy": 0.21921915113925933,
      "num_tokens": 98109803.0,
      "step": 50155
    },
    {
      "entropy": 5.727922534942627,
      "epoch": 5.0139951017144,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002775410100200156,
      "loss": 5.1135,
      "mean_token_accuracy": 0.23640238642692565,
      "num_tokens": 98119231.0,
      "step": 50160
    },
    {
      "entropy": 5.742554759979248,
      "epoch": 5.014494926775629,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027750532316253427,
      "loss": 5.1418,
      "mean_token_accuracy": 0.2338918924331665,
      "num_tokens": 98129564.0,
      "step": 50165
    },
    {
      "entropy": 5.731126022338867,
      "epoch": 5.014994751836857,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027746963624201965,
      "loss": 5.1268,
      "mean_token_accuracy": 0.23171427845954895,
      "num_tokens": 98138206.0,
      "step": 50170
    },
    {
      "entropy": 5.827392101287842,
      "epoch": 5.015494576898086,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002774339492593697,
      "loss": 5.2612,
      "mean_token_accuracy": 0.22007388025522232,
      "num_tokens": 98148107.0,
      "step": 50175
    },
    {
      "entropy": 5.888911962509155,
      "epoch": 5.015994401959314,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002773982622154821,
      "loss": 5.1728,
      "mean_token_accuracy": 0.22615784853696824,
      "num_tokens": 98158293.0,
      "step": 50180
    },
    {
      "entropy": 5.76409158706665,
      "epoch": 5.016494227020543,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027736257511125497,
      "loss": 5.0448,
      "mean_token_accuracy": 0.23999533504247667,
      "num_tokens": 98168001.0,
      "step": 50185
    },
    {
      "entropy": 5.570001220703125,
      "epoch": 5.016994052081771,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027732688794758606,
      "loss": 4.9229,
      "mean_token_accuracy": 0.25011285543441775,
      "num_tokens": 98178050.0,
      "step": 50190
    },
    {
      "entropy": 5.794434642791748,
      "epoch": 5.017493877143,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002772912007253732,
      "loss": 5.1509,
      "mean_token_accuracy": 0.22767349034547807,
      "num_tokens": 98187259.0,
      "step": 50195
    },
    {
      "entropy": 5.820520973205566,
      "epoch": 5.017993702204229,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002772555134455143,
      "loss": 5.2454,
      "mean_token_accuracy": 0.21433551013469695,
      "num_tokens": 98198020.0,
      "step": 50200
    },
    {
      "entropy": 5.824642848968506,
      "epoch": 5.018493527265457,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027721982610890724,
      "loss": 5.1893,
      "mean_token_accuracy": 0.21959538161754608,
      "num_tokens": 98208657.0,
      "step": 50205
    },
    {
      "entropy": 5.877833986282349,
      "epoch": 5.018993352326686,
      "grad_norm": 1.25,
      "learning_rate": 0.00027718413871645,
      "loss": 5.1377,
      "mean_token_accuracy": 0.2242721438407898,
      "num_tokens": 98218391.0,
      "step": 50210
    },
    {
      "entropy": 5.773988342285156,
      "epoch": 5.019493177387914,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002771484512690403,
      "loss": 5.1484,
      "mean_token_accuracy": 0.23165539652109146,
      "num_tokens": 98228201.0,
      "step": 50215
    },
    {
      "entropy": 5.753559923171997,
      "epoch": 5.019993002449143,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002771127637675762,
      "loss": 5.1058,
      "mean_token_accuracy": 0.2276701584458351,
      "num_tokens": 98237380.0,
      "step": 50220
    },
    {
      "entropy": 5.78422064781189,
      "epoch": 5.020492827510371,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027707707621295547,
      "loss": 5.1126,
      "mean_token_accuracy": 0.22753612250089644,
      "num_tokens": 98247084.0,
      "step": 50225
    },
    {
      "entropy": 5.749614524841308,
      "epoch": 5.0209926525716,
      "grad_norm": 1.25,
      "learning_rate": 0.0002770413886060761,
      "loss": 5.1156,
      "mean_token_accuracy": 0.22990636378526688,
      "num_tokens": 98256697.0,
      "step": 50230
    },
    {
      "entropy": 5.847087383270264,
      "epoch": 5.021492477632829,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027700570094783583,
      "loss": 5.2217,
      "mean_token_accuracy": 0.22400470823049545,
      "num_tokens": 98266974.0,
      "step": 50235
    },
    {
      "entropy": 5.845553255081176,
      "epoch": 5.021992302694057,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002769700132391327,
      "loss": 5.2335,
      "mean_token_accuracy": 0.2139231413602829,
      "num_tokens": 98276343.0,
      "step": 50240
    },
    {
      "entropy": 5.807860088348389,
      "epoch": 5.022492127755286,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002769343254808645,
      "loss": 5.0895,
      "mean_token_accuracy": 0.2336229383945465,
      "num_tokens": 98285221.0,
      "step": 50245
    },
    {
      "entropy": 5.708049583435058,
      "epoch": 5.022991952816514,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027689863767392915,
      "loss": 5.0882,
      "mean_token_accuracy": 0.23714621663093566,
      "num_tokens": 98296288.0,
      "step": 50250
    },
    {
      "entropy": 5.834920835494995,
      "epoch": 5.023491777877743,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002768629498192247,
      "loss": 5.1774,
      "mean_token_accuracy": 0.2255142167210579,
      "num_tokens": 98306495.0,
      "step": 50255
    },
    {
      "entropy": 5.819962024688721,
      "epoch": 5.023991602938971,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002768272619176488,
      "loss": 5.2291,
      "mean_token_accuracy": 0.21863873451948165,
      "num_tokens": 98316825.0,
      "step": 50260
    },
    {
      "entropy": 5.746790266036987,
      "epoch": 5.0244914280002,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027679157397009945,
      "loss": 5.0352,
      "mean_token_accuracy": 0.24228092432022094,
      "num_tokens": 98326523.0,
      "step": 50265
    },
    {
      "entropy": 5.830479288101197,
      "epoch": 5.024991253061429,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027675588597747455,
      "loss": 5.1922,
      "mean_token_accuracy": 0.216902257502079,
      "num_tokens": 98335000.0,
      "step": 50270
    },
    {
      "entropy": 5.78275990486145,
      "epoch": 5.025491078122657,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027672019794067204,
      "loss": 5.18,
      "mean_token_accuracy": 0.2241313248872757,
      "num_tokens": 98344694.0,
      "step": 50275
    },
    {
      "entropy": 5.815738868713379,
      "epoch": 5.025990903183885,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002766845098605898,
      "loss": 5.1901,
      "mean_token_accuracy": 0.22182462364435196,
      "num_tokens": 98354075.0,
      "step": 50280
    },
    {
      "entropy": 5.788162517547607,
      "epoch": 5.026490728245114,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002766488217381258,
      "loss": 5.128,
      "mean_token_accuracy": 0.22476030141115189,
      "num_tokens": 98364006.0,
      "step": 50285
    },
    {
      "entropy": 5.8457146167755125,
      "epoch": 5.026990553306343,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002766131335741778,
      "loss": 5.2229,
      "mean_token_accuracy": 0.21755639910697938,
      "num_tokens": 98373917.0,
      "step": 50290
    },
    {
      "entropy": 5.796451139450073,
      "epoch": 5.027490378367571,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027657744536964374,
      "loss": 5.1637,
      "mean_token_accuracy": 0.22915271520614625,
      "num_tokens": 98384773.0,
      "step": 50295
    },
    {
      "entropy": 5.776569938659668,
      "epoch": 5.0279902034288,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027654175712542154,
      "loss": 5.1362,
      "mean_token_accuracy": 0.224607415497303,
      "num_tokens": 98393529.0,
      "step": 50300
    },
    {
      "entropy": 5.81063814163208,
      "epoch": 5.028490028490029,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002765060688424092,
      "loss": 5.2233,
      "mean_token_accuracy": 0.21966415792703628,
      "num_tokens": 98403407.0,
      "step": 50305
    },
    {
      "entropy": 5.763628768920898,
      "epoch": 5.028989853551257,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002764703805215045,
      "loss": 5.1604,
      "mean_token_accuracy": 0.2200721338391304,
      "num_tokens": 98413068.0,
      "step": 50310
    },
    {
      "entropy": 5.811252307891846,
      "epoch": 5.029489678612485,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027643469216360545,
      "loss": 5.1581,
      "mean_token_accuracy": 0.22799161225557327,
      "num_tokens": 98421630.0,
      "step": 50315
    },
    {
      "entropy": 5.808308076858521,
      "epoch": 5.029989503673714,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027639900376960993,
      "loss": 5.1638,
      "mean_token_accuracy": 0.2148326575756073,
      "num_tokens": 98431612.0,
      "step": 50320
    },
    {
      "entropy": 5.79146728515625,
      "epoch": 5.030489328734943,
      "grad_norm": 1.25,
      "learning_rate": 0.00027636331534041574,
      "loss": 5.1178,
      "mean_token_accuracy": 0.23439791053533554,
      "num_tokens": 98440608.0,
      "step": 50325
    },
    {
      "entropy": 5.8307945728302,
      "epoch": 5.030989153796171,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002763276268769209,
      "loss": 5.2476,
      "mean_token_accuracy": 0.21889277100563048,
      "num_tokens": 98449596.0,
      "step": 50330
    },
    {
      "entropy": 5.746927070617676,
      "epoch": 5.0314889788574,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002762919383800235,
      "loss": 5.1836,
      "mean_token_accuracy": 0.22046967148780822,
      "num_tokens": 98459382.0,
      "step": 50335
    },
    {
      "entropy": 5.7969340801239015,
      "epoch": 5.031988803918629,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002762562498506211,
      "loss": 5.1528,
      "mean_token_accuracy": 0.22784011363983153,
      "num_tokens": 98468871.0,
      "step": 50340
    },
    {
      "entropy": 5.8054218769073485,
      "epoch": 5.032488628979857,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027622056128961183,
      "loss": 5.1322,
      "mean_token_accuracy": 0.2285596713423729,
      "num_tokens": 98478029.0,
      "step": 50345
    },
    {
      "entropy": 5.800421953201294,
      "epoch": 5.032988454041085,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002761848726978935,
      "loss": 5.181,
      "mean_token_accuracy": 0.22803892344236373,
      "num_tokens": 98488339.0,
      "step": 50350
    },
    {
      "entropy": 5.6752612590789795,
      "epoch": 5.033488279102314,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027614918407636407,
      "loss": 5.004,
      "mean_token_accuracy": 0.23603529930114747,
      "num_tokens": 98498806.0,
      "step": 50355
    },
    {
      "entropy": 5.852773427963257,
      "epoch": 5.033988104163543,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027611349542592146,
      "loss": 5.2119,
      "mean_token_accuracy": 0.21096826195716858,
      "num_tokens": 98508159.0,
      "step": 50360
    },
    {
      "entropy": 5.749150228500366,
      "epoch": 5.034487929224771,
      "grad_norm": 1.25,
      "learning_rate": 0.00027607780674746365,
      "loss": 5.1484,
      "mean_token_accuracy": 0.22913335561752318,
      "num_tokens": 98518628.0,
      "step": 50365
    },
    {
      "entropy": 5.778139877319336,
      "epoch": 5.034987754286,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002760421180418885,
      "loss": 5.1298,
      "mean_token_accuracy": 0.2211672842502594,
      "num_tokens": 98529358.0,
      "step": 50370
    },
    {
      "entropy": 5.74088044166565,
      "epoch": 5.035487579347229,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002760064293100939,
      "loss": 5.0525,
      "mean_token_accuracy": 0.24133833050727843,
      "num_tokens": 98538914.0,
      "step": 50375
    },
    {
      "entropy": 5.813129854202271,
      "epoch": 5.035987404408457,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027597074055297783,
      "loss": 5.2172,
      "mean_token_accuracy": 0.2212151750922203,
      "num_tokens": 98549795.0,
      "step": 50380
    },
    {
      "entropy": 5.7494707107543945,
      "epoch": 5.036487229469685,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027593505177143815,
      "loss": 5.1132,
      "mean_token_accuracy": 0.23155697137117387,
      "num_tokens": 98559643.0,
      "step": 50385
    },
    {
      "entropy": 5.786699533462524,
      "epoch": 5.036987054530914,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002758993629663727,
      "loss": 5.1213,
      "mean_token_accuracy": 0.22079743593931198,
      "num_tokens": 98568049.0,
      "step": 50390
    },
    {
      "entropy": 5.748332118988037,
      "epoch": 5.037486879592143,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002758636741386796,
      "loss": 5.1585,
      "mean_token_accuracy": 0.22610264718532563,
      "num_tokens": 98577210.0,
      "step": 50395
    },
    {
      "entropy": 5.792299747467041,
      "epoch": 5.037986704653371,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002758279852892568,
      "loss": 5.2183,
      "mean_token_accuracy": 0.21723132729530334,
      "num_tokens": 98585935.0,
      "step": 50400
    },
    {
      "entropy": 5.732297134399414,
      "epoch": 5.0384865297146,
      "grad_norm": 1.203125,
      "learning_rate": 0.000275792296419002,
      "loss": 5.1313,
      "mean_token_accuracy": 0.22857339978218078,
      "num_tokens": 98597043.0,
      "step": 50405
    },
    {
      "entropy": 5.805272769927979,
      "epoch": 5.038986354775829,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002757566075288132,
      "loss": 5.0968,
      "mean_token_accuracy": 0.23347013294696808,
      "num_tokens": 98606509.0,
      "step": 50410
    },
    {
      "entropy": 5.80228362083435,
      "epoch": 5.039486179837057,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027572091861958835,
      "loss": 5.1197,
      "mean_token_accuracy": 0.22538021951913834,
      "num_tokens": 98616071.0,
      "step": 50415
    },
    {
      "entropy": 5.816283178329468,
      "epoch": 5.039986004898285,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002756852296922254,
      "loss": 5.2308,
      "mean_token_accuracy": 0.21366625428199768,
      "num_tokens": 98625052.0,
      "step": 50420
    },
    {
      "entropy": 5.760176038742065,
      "epoch": 5.040485829959514,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002756495407476223,
      "loss": 5.0165,
      "mean_token_accuracy": 0.238512621819973,
      "num_tokens": 98635473.0,
      "step": 50425
    },
    {
      "entropy": 5.698930978775024,
      "epoch": 5.040985655020743,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002756138517866768,
      "loss": 5.0599,
      "mean_token_accuracy": 0.23187771439552307,
      "num_tokens": 98645554.0,
      "step": 50430
    },
    {
      "entropy": 5.806757593154908,
      "epoch": 5.041485480081971,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027557816281028707,
      "loss": 5.2423,
      "mean_token_accuracy": 0.21369960755109788,
      "num_tokens": 98654525.0,
      "step": 50435
    },
    {
      "entropy": 5.7478996276855465,
      "epoch": 5.0419853051432,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002755424738193509,
      "loss": 5.097,
      "mean_token_accuracy": 0.22706826031208038,
      "num_tokens": 98663894.0,
      "step": 50440
    },
    {
      "entropy": 5.7640830993652346,
      "epoch": 5.042485130204429,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027550678481476614,
      "loss": 5.1675,
      "mean_token_accuracy": 0.22551886290311812,
      "num_tokens": 98672638.0,
      "step": 50445
    },
    {
      "entropy": 5.76674222946167,
      "epoch": 5.042984955265657,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027547109579743087,
      "loss": 5.0923,
      "mean_token_accuracy": 0.22745743095874787,
      "num_tokens": 98681535.0,
      "step": 50450
    },
    {
      "entropy": 5.7774975299835205,
      "epoch": 5.043484780326885,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002754354067682429,
      "loss": 5.189,
      "mean_token_accuracy": 0.22289083600044252,
      "num_tokens": 98692371.0,
      "step": 50455
    },
    {
      "entropy": 5.857534170150757,
      "epoch": 5.043984605388114,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002753997177281002,
      "loss": 5.268,
      "mean_token_accuracy": 0.21391710489988328,
      "num_tokens": 98702745.0,
      "step": 50460
    },
    {
      "entropy": 5.852266693115235,
      "epoch": 5.044484430449343,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027536402867790077,
      "loss": 5.2138,
      "mean_token_accuracy": 0.21800573021173478,
      "num_tokens": 98712173.0,
      "step": 50465
    },
    {
      "entropy": 5.759313678741455,
      "epoch": 5.044984255510571,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027532833961854247,
      "loss": 5.0815,
      "mean_token_accuracy": 0.23469312787055968,
      "num_tokens": 98721818.0,
      "step": 50470
    },
    {
      "entropy": 5.747949409484863,
      "epoch": 5.0454840805718,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027529265055092325,
      "loss": 5.0972,
      "mean_token_accuracy": 0.23287010043859482,
      "num_tokens": 98731378.0,
      "step": 50475
    },
    {
      "entropy": 5.833069515228272,
      "epoch": 5.045983905633029,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027525696147594095,
      "loss": 5.2917,
      "mean_token_accuracy": 0.21767496615648269,
      "num_tokens": 98742121.0,
      "step": 50480
    },
    {
      "entropy": 5.857402229309082,
      "epoch": 5.046483730694257,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002752212723944935,
      "loss": 5.1673,
      "mean_token_accuracy": 0.22061726599931716,
      "num_tokens": 98753116.0,
      "step": 50485
    },
    {
      "entropy": 5.77315320968628,
      "epoch": 5.046983555755485,
      "grad_norm": 1.171875,
      "learning_rate": 0.000275185583307479,
      "loss": 5.0662,
      "mean_token_accuracy": 0.2229374185204506,
      "num_tokens": 98762986.0,
      "step": 50490
    },
    {
      "entropy": 5.754674959182739,
      "epoch": 5.047483380816714,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027514989421579525,
      "loss": 5.1447,
      "mean_token_accuracy": 0.22498637437820435,
      "num_tokens": 98772350.0,
      "step": 50495
    },
    {
      "entropy": 5.853113508224487,
      "epoch": 5.047983205877943,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002751142051203402,
      "loss": 5.2587,
      "mean_token_accuracy": 0.21358796805143357,
      "num_tokens": 98782740.0,
      "step": 50500
    },
    {
      "entropy": 5.849034261703491,
      "epoch": 5.048483030939171,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002750785160220118,
      "loss": 5.1892,
      "mean_token_accuracy": 0.22722983807325364,
      "num_tokens": 98792359.0,
      "step": 50505
    },
    {
      "entropy": 5.814743185043335,
      "epoch": 5.0489828560004,
      "grad_norm": 1.234375,
      "learning_rate": 0.000275042826921708,
      "loss": 5.1558,
      "mean_token_accuracy": 0.23158910125494003,
      "num_tokens": 98802104.0,
      "step": 50510
    },
    {
      "entropy": 5.720367908477783,
      "epoch": 5.049482681061629,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027500713782032657,
      "loss": 5.0307,
      "mean_token_accuracy": 0.23465381115674971,
      "num_tokens": 98811399.0,
      "step": 50515
    },
    {
      "entropy": 5.763923454284668,
      "epoch": 5.049982506122857,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027497144871876553,
      "loss": 5.1437,
      "mean_token_accuracy": 0.22524382472038268,
      "num_tokens": 98821667.0,
      "step": 50520
    },
    {
      "entropy": 5.723640394210816,
      "epoch": 5.050482331184085,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027493575961792306,
      "loss": 5.0625,
      "mean_token_accuracy": 0.22989982217550278,
      "num_tokens": 98831940.0,
      "step": 50525
    },
    {
      "entropy": 5.8411486625671385,
      "epoch": 5.050982156245314,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002749000705186967,
      "loss": 5.2488,
      "mean_token_accuracy": 0.21915586441755294,
      "num_tokens": 98841160.0,
      "step": 50530
    },
    {
      "entropy": 5.809846830368042,
      "epoch": 5.051481981306543,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027486438142198454,
      "loss": 5.2079,
      "mean_token_accuracy": 0.2188231199979782,
      "num_tokens": 98850603.0,
      "step": 50535
    },
    {
      "entropy": 5.782731723785401,
      "epoch": 5.051981806367771,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002748286923286845,
      "loss": 5.2045,
      "mean_token_accuracy": 0.21624965369701385,
      "num_tokens": 98860334.0,
      "step": 50540
    },
    {
      "entropy": 5.840532732009888,
      "epoch": 5.052481631429,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027479300323969455,
      "loss": 5.1622,
      "mean_token_accuracy": 0.21721533834934234,
      "num_tokens": 98869756.0,
      "step": 50545
    },
    {
      "entropy": 5.8559596061706545,
      "epoch": 5.052981456490229,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027475731415591253,
      "loss": 5.2249,
      "mean_token_accuracy": 0.21382110863924025,
      "num_tokens": 98879024.0,
      "step": 50550
    },
    {
      "entropy": 5.8320292949676515,
      "epoch": 5.053481281551457,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027472162507823653,
      "loss": 5.2116,
      "mean_token_accuracy": 0.21889074593782426,
      "num_tokens": 98888657.0,
      "step": 50555
    },
    {
      "entropy": 5.779442691802979,
      "epoch": 5.053981106612685,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027468593600756425,
      "loss": 5.1784,
      "mean_token_accuracy": 0.2184527486562729,
      "num_tokens": 98897697.0,
      "step": 50560
    },
    {
      "entropy": 5.961690187454224,
      "epoch": 5.054480931673914,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027465024694479386,
      "loss": 5.3711,
      "mean_token_accuracy": 0.2068707302212715,
      "num_tokens": 98907605.0,
      "step": 50565
    },
    {
      "entropy": 5.8039665699005125,
      "epoch": 5.054980756735143,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027461455789082307,
      "loss": 5.1603,
      "mean_token_accuracy": 0.2261606350541115,
      "num_tokens": 98916568.0,
      "step": 50570
    },
    {
      "entropy": 5.810457992553711,
      "epoch": 5.055480581796371,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002745788688465499,
      "loss": 5.2075,
      "mean_token_accuracy": 0.21879328787326813,
      "num_tokens": 98925907.0,
      "step": 50575
    },
    {
      "entropy": 5.756629991531372,
      "epoch": 5.0559804068576,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027454317981287236,
      "loss": 5.119,
      "mean_token_accuracy": 0.2289721965789795,
      "num_tokens": 98936069.0,
      "step": 50580
    },
    {
      "entropy": 5.770047569274903,
      "epoch": 5.056480231918829,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002745074907906882,
      "loss": 5.1218,
      "mean_token_accuracy": 0.2279893636703491,
      "num_tokens": 98945072.0,
      "step": 50585
    },
    {
      "entropy": 5.820612859725952,
      "epoch": 5.056980056980057,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002744718017808955,
      "loss": 5.2274,
      "mean_token_accuracy": 0.22785576581954955,
      "num_tokens": 98955211.0,
      "step": 50590
    },
    {
      "entropy": 5.814574575424194,
      "epoch": 5.057479882041285,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027443611278439226,
      "loss": 5.1828,
      "mean_token_accuracy": 0.2204620435833931,
      "num_tokens": 98965429.0,
      "step": 50595
    },
    {
      "entropy": 5.863741779327393,
      "epoch": 5.057979707102514,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002744004238020761,
      "loss": 5.193,
      "mean_token_accuracy": 0.2192537948489189,
      "num_tokens": 98975229.0,
      "step": 50600
    },
    {
      "entropy": 5.81267032623291,
      "epoch": 5.058479532163743,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027436473483484513,
      "loss": 5.2351,
      "mean_token_accuracy": 0.2203196331858635,
      "num_tokens": 98985351.0,
      "step": 50605
    },
    {
      "entropy": 5.7166375637054445,
      "epoch": 5.058979357224971,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027432904588359735,
      "loss": 5.0354,
      "mean_token_accuracy": 0.23161893337965012,
      "num_tokens": 98995350.0,
      "step": 50610
    },
    {
      "entropy": 5.7287345886230465,
      "epoch": 5.0594791822862,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002742933569492306,
      "loss": 4.9734,
      "mean_token_accuracy": 0.23786280751228334,
      "num_tokens": 99004415.0,
      "step": 50615
    },
    {
      "entropy": 5.812160396575928,
      "epoch": 5.059979007347429,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027425766803264274,
      "loss": 5.1969,
      "mean_token_accuracy": 0.22131207585334778,
      "num_tokens": 99015059.0,
      "step": 50620
    },
    {
      "entropy": 5.80750880241394,
      "epoch": 5.060478832408657,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002742219791347318,
      "loss": 5.148,
      "mean_token_accuracy": 0.22859860360622405,
      "num_tokens": 99024903.0,
      "step": 50625
    },
    {
      "entropy": 5.546170854568482,
      "epoch": 5.060978657469885,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002741862902563957,
      "loss": 4.8582,
      "mean_token_accuracy": 0.26497390419244765,
      "num_tokens": 99034831.0,
      "step": 50630
    },
    {
      "entropy": 5.724675035476684,
      "epoch": 5.061478482531114,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002741506013985323,
      "loss": 5.0838,
      "mean_token_accuracy": 0.23495964407920839,
      "num_tokens": 99045011.0,
      "step": 50635
    },
    {
      "entropy": 5.775168704986572,
      "epoch": 5.061978307592343,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027411491256203944,
      "loss": 5.1565,
      "mean_token_accuracy": 0.2298825889825821,
      "num_tokens": 99055193.0,
      "step": 50640
    },
    {
      "entropy": 5.766234827041626,
      "epoch": 5.062478132653571,
      "grad_norm": 1.25,
      "learning_rate": 0.00027407922374781524,
      "loss": 5.1072,
      "mean_token_accuracy": 0.2291930288076401,
      "num_tokens": 99063781.0,
      "step": 50645
    },
    {
      "entropy": 5.753731441497803,
      "epoch": 5.0629779577148,
      "grad_norm": 1.25,
      "learning_rate": 0.00027404353495675757,
      "loss": 5.1738,
      "mean_token_accuracy": 0.22437804490327834,
      "num_tokens": 99074147.0,
      "step": 50650
    },
    {
      "entropy": 5.758320331573486,
      "epoch": 5.063477782776029,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027400784618976425,
      "loss": 5.0979,
      "mean_token_accuracy": 0.23586748987436296,
      "num_tokens": 99082946.0,
      "step": 50655
    },
    {
      "entropy": 5.68792495727539,
      "epoch": 5.063977607837257,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027397215744773333,
      "loss": 5.0416,
      "mean_token_accuracy": 0.24097394645214082,
      "num_tokens": 99093222.0,
      "step": 50660
    },
    {
      "entropy": 5.716043615341187,
      "epoch": 5.064477432898485,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002739364687315627,
      "loss": 5.1145,
      "mean_token_accuracy": 0.23127248883247375,
      "num_tokens": 99102486.0,
      "step": 50665
    },
    {
      "entropy": 5.679377222061158,
      "epoch": 5.064977257959714,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027390078004215017,
      "loss": 5.0365,
      "mean_token_accuracy": 0.2448124334216118,
      "num_tokens": 99112133.0,
      "step": 50670
    },
    {
      "entropy": 5.755318212509155,
      "epoch": 5.065477083020943,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027386509138039374,
      "loss": 5.1312,
      "mean_token_accuracy": 0.22370551377534867,
      "num_tokens": 99121128.0,
      "step": 50675
    },
    {
      "entropy": 5.788215446472168,
      "epoch": 5.065976908082171,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002738294027471913,
      "loss": 5.2322,
      "mean_token_accuracy": 0.22436809539794922,
      "num_tokens": 99132424.0,
      "step": 50680
    },
    {
      "entropy": 5.770577049255371,
      "epoch": 5.0664767331434,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002737937141434408,
      "loss": 5.0481,
      "mean_token_accuracy": 0.2298956871032715,
      "num_tokens": 99141583.0,
      "step": 50685
    },
    {
      "entropy": 5.774998188018799,
      "epoch": 5.066976558204629,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002737580255700402,
      "loss": 5.145,
      "mean_token_accuracy": 0.23188503086566925,
      "num_tokens": 99151001.0,
      "step": 50690
    },
    {
      "entropy": 5.745249938964844,
      "epoch": 5.067476383265857,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027372233702788727,
      "loss": 5.1413,
      "mean_token_accuracy": 0.22934041023254395,
      "num_tokens": 99159801.0,
      "step": 50695
    },
    {
      "entropy": 5.8654357433319095,
      "epoch": 5.067976208327085,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027368664851788006,
      "loss": 5.1937,
      "mean_token_accuracy": 0.21568788439035416,
      "num_tokens": 99169552.0,
      "step": 50700
    },
    {
      "entropy": 5.785890007019043,
      "epoch": 5.068476033388314,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002736509600409164,
      "loss": 5.2156,
      "mean_token_accuracy": 0.22445307821035385,
      "num_tokens": 99178500.0,
      "step": 50705
    },
    {
      "entropy": 5.808284425735474,
      "epoch": 5.068975858449543,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027361527159789424,
      "loss": 5.1503,
      "mean_token_accuracy": 0.21402340084314347,
      "num_tokens": 99188398.0,
      "step": 50710
    },
    {
      "entropy": 5.7395772457122805,
      "epoch": 5.069475683510771,
      "grad_norm": 1.25,
      "learning_rate": 0.0002735795831897115,
      "loss": 5.0421,
      "mean_token_accuracy": 0.2367873251438141,
      "num_tokens": 99198492.0,
      "step": 50715
    },
    {
      "entropy": 5.732904291152954,
      "epoch": 5.069975508572,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002735438948172661,
      "loss": 5.1489,
      "mean_token_accuracy": 0.23321365416049958,
      "num_tokens": 99207859.0,
      "step": 50720
    },
    {
      "entropy": 5.7633240699768065,
      "epoch": 5.070475333633229,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002735082064814559,
      "loss": 5.1199,
      "mean_token_accuracy": 0.22641810178756713,
      "num_tokens": 99217132.0,
      "step": 50725
    },
    {
      "entropy": 5.791251802444458,
      "epoch": 5.070975158694457,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027347251818317885,
      "loss": 5.178,
      "mean_token_accuracy": 0.22023964375257493,
      "num_tokens": 99226980.0,
      "step": 50730
    },
    {
      "entropy": 5.863850688934326,
      "epoch": 5.071474983755685,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027343682992333297,
      "loss": 5.2892,
      "mean_token_accuracy": 0.21506565809249878,
      "num_tokens": 99237176.0,
      "step": 50735
    },
    {
      "entropy": 5.843474578857422,
      "epoch": 5.071974808816914,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002734011417028159,
      "loss": 5.137,
      "mean_token_accuracy": 0.2244459480047226,
      "num_tokens": 99246025.0,
      "step": 50740
    },
    {
      "entropy": 5.698772192001343,
      "epoch": 5.072474633878143,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002733654535225257,
      "loss": 5.0654,
      "mean_token_accuracy": 0.23047292828559876,
      "num_tokens": 99255110.0,
      "step": 50745
    },
    {
      "entropy": 5.787006759643555,
      "epoch": 5.072974458939371,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027332976538336037,
      "loss": 5.1715,
      "mean_token_accuracy": 0.22381090819835664,
      "num_tokens": 99265158.0,
      "step": 50750
    },
    {
      "entropy": 5.787304353713989,
      "epoch": 5.0734742840006,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002732940772862176,
      "loss": 5.1586,
      "mean_token_accuracy": 0.22641617506742479,
      "num_tokens": 99275534.0,
      "step": 50755
    },
    {
      "entropy": 5.8226312637329105,
      "epoch": 5.073974109061829,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027325838923199545,
      "loss": 5.1908,
      "mean_token_accuracy": 0.2233033612370491,
      "num_tokens": 99285907.0,
      "step": 50760
    },
    {
      "entropy": 5.813615798950195,
      "epoch": 5.074473934123057,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027322270122159175,
      "loss": 5.2005,
      "mean_token_accuracy": 0.21720333099365235,
      "num_tokens": 99295425.0,
      "step": 50765
    },
    {
      "entropy": 5.903329372406006,
      "epoch": 5.074973759184285,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027318701325590444,
      "loss": 5.2256,
      "mean_token_accuracy": 0.21464893519878386,
      "num_tokens": 99304864.0,
      "step": 50770
    },
    {
      "entropy": 5.778205537796021,
      "epoch": 5.075473584245514,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002731513253358315,
      "loss": 5.2151,
      "mean_token_accuracy": 0.21887294501066207,
      "num_tokens": 99314479.0,
      "step": 50775
    },
    {
      "entropy": 5.696797704696655,
      "epoch": 5.075973409306743,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002731156374622706,
      "loss": 5.053,
      "mean_token_accuracy": 0.23773396164178848,
      "num_tokens": 99323921.0,
      "step": 50780
    },
    {
      "entropy": 5.779649829864502,
      "epoch": 5.076473234367971,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027307994963611987,
      "loss": 5.1309,
      "mean_token_accuracy": 0.23525257110595704,
      "num_tokens": 99334897.0,
      "step": 50785
    },
    {
      "entropy": 5.779289627075196,
      "epoch": 5.0769730594292,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002730442618582771,
      "loss": 5.15,
      "mean_token_accuracy": 0.2350068524479866,
      "num_tokens": 99345265.0,
      "step": 50790
    },
    {
      "entropy": 5.693783187866211,
      "epoch": 5.077472884490429,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002730085741296402,
      "loss": 5.0794,
      "mean_token_accuracy": 0.23836585730314255,
      "num_tokens": 99354886.0,
      "step": 50795
    },
    {
      "entropy": 5.740339231491089,
      "epoch": 5.077972709551657,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00027297288645110703,
      "loss": 5.1396,
      "mean_token_accuracy": 0.232313571870327,
      "num_tokens": 99365086.0,
      "step": 50800
    },
    {
      "entropy": 5.857637214660644,
      "epoch": 5.078472534612885,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002729371988235756,
      "loss": 5.224,
      "mean_token_accuracy": 0.221421916782856,
      "num_tokens": 99374827.0,
      "step": 50805
    },
    {
      "entropy": 5.828780841827393,
      "epoch": 5.078972359674114,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027290151124794365,
      "loss": 5.1738,
      "mean_token_accuracy": 0.22597666531801225,
      "num_tokens": 99385245.0,
      "step": 50810
    },
    {
      "entropy": 5.761242628097534,
      "epoch": 5.079472184735343,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027286582372510913,
      "loss": 5.1345,
      "mean_token_accuracy": 0.23245518058538436,
      "num_tokens": 99394570.0,
      "step": 50815
    },
    {
      "entropy": 5.803978633880615,
      "epoch": 5.079972009796571,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027283013625597,
      "loss": 5.2401,
      "mean_token_accuracy": 0.21618133187294006,
      "num_tokens": 99403764.0,
      "step": 50820
    },
    {
      "entropy": 5.769129991531372,
      "epoch": 5.0804718348578,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027279444884142397,
      "loss": 5.099,
      "mean_token_accuracy": 0.23627379834651946,
      "num_tokens": 99413797.0,
      "step": 50825
    },
    {
      "entropy": 5.767573881149292,
      "epoch": 5.080971659919029,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027275876148236903,
      "loss": 5.084,
      "mean_token_accuracy": 0.24032808244228362,
      "num_tokens": 99422371.0,
      "step": 50830
    },
    {
      "entropy": 5.746330404281617,
      "epoch": 5.081471484980257,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027272307417970323,
      "loss": 5.0751,
      "mean_token_accuracy": 0.2310320720076561,
      "num_tokens": 99431475.0,
      "step": 50835
    },
    {
      "entropy": 5.804688262939453,
      "epoch": 5.081971310041485,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002726873869343242,
      "loss": 5.1695,
      "mean_token_accuracy": 0.2263142868876457,
      "num_tokens": 99441501.0,
      "step": 50840
    },
    {
      "entropy": 5.796480798721314,
      "epoch": 5.082471135102714,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027265169974712987,
      "loss": 5.0824,
      "mean_token_accuracy": 0.2346719205379486,
      "num_tokens": 99450602.0,
      "step": 50845
    },
    {
      "entropy": 5.791384363174439,
      "epoch": 5.082970960163943,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027261601261901824,
      "loss": 5.2181,
      "mean_token_accuracy": 0.22232684195041658,
      "num_tokens": 99460808.0,
      "step": 50850
    },
    {
      "entropy": 5.782971525192261,
      "epoch": 5.083470785225171,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002725803255508871,
      "loss": 5.1746,
      "mean_token_accuracy": 0.22563944905996322,
      "num_tokens": 99471234.0,
      "step": 50855
    },
    {
      "entropy": 5.7596352100372314,
      "epoch": 5.0839706102864,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027254463854363436,
      "loss": 5.1405,
      "mean_token_accuracy": 0.2306041195988655,
      "num_tokens": 99480673.0,
      "step": 50860
    },
    {
      "entropy": 5.7284129619598385,
      "epoch": 5.084470435347629,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002725089515981579,
      "loss": 5.1611,
      "mean_token_accuracy": 0.23131401538848878,
      "num_tokens": 99490225.0,
      "step": 50865
    },
    {
      "entropy": 5.775336694717407,
      "epoch": 5.0849702604088565,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002724732647153555,
      "loss": 5.1506,
      "mean_token_accuracy": 0.23220458030700683,
      "num_tokens": 99500285.0,
      "step": 50870
    },
    {
      "entropy": 5.932348537445068,
      "epoch": 5.085470085470085,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027243757789612526,
      "loss": 5.2677,
      "mean_token_accuracy": 0.2225899338722229,
      "num_tokens": 99509638.0,
      "step": 50875
    },
    {
      "entropy": 5.827115726470947,
      "epoch": 5.085969910531314,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002724018911413648,
      "loss": 5.1538,
      "mean_token_accuracy": 0.23057102411985397,
      "num_tokens": 99519182.0,
      "step": 50880
    },
    {
      "entropy": 5.735175609588623,
      "epoch": 5.086469735592543,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027236620445197213,
      "loss": 5.0926,
      "mean_token_accuracy": 0.22924406826496124,
      "num_tokens": 99528783.0,
      "step": 50885
    },
    {
      "entropy": 5.788478088378906,
      "epoch": 5.086969560653771,
      "grad_norm": 1.296875,
      "learning_rate": 0.000272330517828845,
      "loss": 5.1507,
      "mean_token_accuracy": 0.23197167366743088,
      "num_tokens": 99537885.0,
      "step": 50890
    },
    {
      "entropy": 5.734203195571899,
      "epoch": 5.087469385715,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027229483127288133,
      "loss": 5.0737,
      "mean_token_accuracy": 0.2374486058950424,
      "num_tokens": 99546841.0,
      "step": 50895
    },
    {
      "entropy": 5.826677322387695,
      "epoch": 5.087969210776229,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002722591447849791,
      "loss": 5.2461,
      "mean_token_accuracy": 0.22163718193769455,
      "num_tokens": 99556841.0,
      "step": 50900
    },
    {
      "entropy": 5.822218990325927,
      "epoch": 5.0884690358374565,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002722234583660361,
      "loss": 5.1442,
      "mean_token_accuracy": 0.22286846339702607,
      "num_tokens": 99566681.0,
      "step": 50905
    },
    {
      "entropy": 5.735228967666626,
      "epoch": 5.088968860898685,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002721877720169501,
      "loss": 5.1259,
      "mean_token_accuracy": 0.2347855031490326,
      "num_tokens": 99576911.0,
      "step": 50910
    },
    {
      "entropy": 5.773270320892334,
      "epoch": 5.089468685959914,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002721520857386191,
      "loss": 5.1887,
      "mean_token_accuracy": 0.2166134759783745,
      "num_tokens": 99586502.0,
      "step": 50915
    },
    {
      "entropy": 5.805353212356567,
      "epoch": 5.089968511021143,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002721163995319408,
      "loss": 5.1326,
      "mean_token_accuracy": 0.2244008868932724,
      "num_tokens": 99596220.0,
      "step": 50920
    },
    {
      "entropy": 5.708923959732056,
      "epoch": 5.090468336082371,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002720807133978133,
      "loss": 4.9826,
      "mean_token_accuracy": 0.24430614411830903,
      "num_tokens": 99604720.0,
      "step": 50925
    },
    {
      "entropy": 5.826837635040283,
      "epoch": 5.0909681611436,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002720450273371341,
      "loss": 5.2122,
      "mean_token_accuracy": 0.22262488752603532,
      "num_tokens": 99614203.0,
      "step": 50930
    },
    {
      "entropy": 5.576394844055176,
      "epoch": 5.091467986204829,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027200934135080133,
      "loss": 4.9258,
      "mean_token_accuracy": 0.25305506139993666,
      "num_tokens": 99624786.0,
      "step": 50935
    },
    {
      "entropy": 5.795478439331054,
      "epoch": 5.0919678112660565,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002719736554397128,
      "loss": 5.1906,
      "mean_token_accuracy": 0.23275974839925767,
      "num_tokens": 99634141.0,
      "step": 50940
    },
    {
      "entropy": 5.773385238647461,
      "epoch": 5.092467636327285,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002719379696047664,
      "loss": 5.1083,
      "mean_token_accuracy": 0.23089908212423324,
      "num_tokens": 99644327.0,
      "step": 50945
    },
    {
      "entropy": 5.8249897956848145,
      "epoch": 5.092967461388514,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002719022838468598,
      "loss": 5.2491,
      "mean_token_accuracy": 0.21595589369535445,
      "num_tokens": 99652877.0,
      "step": 50950
    },
    {
      "entropy": 5.826173305511475,
      "epoch": 5.093467286449743,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002718665981668909,
      "loss": 5.188,
      "mean_token_accuracy": 0.22076037675142288,
      "num_tokens": 99662944.0,
      "step": 50955
    },
    {
      "entropy": 5.775108337402344,
      "epoch": 5.093967111510971,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027183091256575764,
      "loss": 5.1249,
      "mean_token_accuracy": 0.2306169182062149,
      "num_tokens": 99673587.0,
      "step": 50960
    },
    {
      "entropy": 5.786681509017944,
      "epoch": 5.0944669365722,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002717952270443578,
      "loss": 5.2683,
      "mean_token_accuracy": 0.21682628244161606,
      "num_tokens": 99682794.0,
      "step": 50965
    },
    {
      "entropy": 5.738982009887695,
      "epoch": 5.094966761633429,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002717595416035892,
      "loss": 5.0948,
      "mean_token_accuracy": 0.2307289198040962,
      "num_tokens": 99691750.0,
      "step": 50970
    },
    {
      "entropy": 5.7268260478973385,
      "epoch": 5.0954665866946565,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002717238562443497,
      "loss": 5.1689,
      "mean_token_accuracy": 0.22483589202165605,
      "num_tokens": 99701732.0,
      "step": 50975
    },
    {
      "entropy": 5.8707029819488525,
      "epoch": 5.095966411755885,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002716881709675372,
      "loss": 5.1993,
      "mean_token_accuracy": 0.21963951140642166,
      "num_tokens": 99712288.0,
      "step": 50980
    },
    {
      "entropy": 5.782635164260864,
      "epoch": 5.096466236817114,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002716524857740494,
      "loss": 5.1463,
      "mean_token_accuracy": 0.23378449380397798,
      "num_tokens": 99722005.0,
      "step": 50985
    },
    {
      "entropy": 5.7885209083557125,
      "epoch": 5.096966061878343,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002716168006647841,
      "loss": 5.1188,
      "mean_token_accuracy": 0.22980444580316545,
      "num_tokens": 99731828.0,
      "step": 50990
    },
    {
      "entropy": 5.876633644104004,
      "epoch": 5.097465886939571,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002715811156406394,
      "loss": 5.2933,
      "mean_token_accuracy": 0.21742285639047623,
      "num_tokens": 99740904.0,
      "step": 50995
    },
    {
      "entropy": 5.7065380096435545,
      "epoch": 5.0979657120008,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027154543070251287,
      "loss": 5.0626,
      "mean_token_accuracy": 0.23612882643938066,
      "num_tokens": 99751002.0,
      "step": 51000
    },
    {
      "epoch": 5.0979657120008,
      "eval_entropy": 5.51761546708726,
      "eval_loss": 5.37973690032959,
      "eval_mean_token_accuracy": 0.22080750448767963,
      "eval_num_tokens": 99751002.0,
      "eval_runtime": 23.8998,
      "eval_samples_per_second": 1298.963,
      "eval_steps_per_second": 162.386,
      "step": 51000
    },
    {
      "entropy": 5.753362131118775,
      "epoch": 5.098465537062029,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002715097458513024,
      "loss": 5.1738,
      "mean_token_accuracy": 0.21809256225824356,
      "num_tokens": 99759971.0,
      "step": 51005
    },
    {
      "entropy": 5.811737871170044,
      "epoch": 5.0989653621232565,
      "grad_norm": 1.25,
      "learning_rate": 0.0002714740610879059,
      "loss": 5.1572,
      "mean_token_accuracy": 0.22357241809368134,
      "num_tokens": 99769307.0,
      "step": 51010
    },
    {
      "entropy": 5.726041507720947,
      "epoch": 5.099465187184485,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002714383764132211,
      "loss": 5.0688,
      "mean_token_accuracy": 0.23515217751264572,
      "num_tokens": 99778805.0,
      "step": 51015
    },
    {
      "entropy": 5.840269756317139,
      "epoch": 5.099965012245714,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002714026918281458,
      "loss": 5.1645,
      "mean_token_accuracy": 0.224900484085083,
      "num_tokens": 99789099.0,
      "step": 51020
    },
    {
      "entropy": 5.789171504974365,
      "epoch": 5.100464837306943,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002713670073335778,
      "loss": 5.1848,
      "mean_token_accuracy": 0.21486856490373613,
      "num_tokens": 99798472.0,
      "step": 51025
    },
    {
      "entropy": 5.637106609344483,
      "epoch": 5.100964662368171,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002713313229304151,
      "loss": 4.9799,
      "mean_token_accuracy": 0.24339328408241273,
      "num_tokens": 99808090.0,
      "step": 51030
    },
    {
      "entropy": 5.7172276973724365,
      "epoch": 5.1014644874294,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027129563861955525,
      "loss": 5.1351,
      "mean_token_accuracy": 0.22727967351675032,
      "num_tokens": 99816971.0,
      "step": 51035
    },
    {
      "entropy": 5.793007659912109,
      "epoch": 5.101964312490629,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002712599544018962,
      "loss": 5.2012,
      "mean_token_accuracy": 0.22004556208848952,
      "num_tokens": 99826666.0,
      "step": 51040
    },
    {
      "entropy": 5.774748420715332,
      "epoch": 5.1024641375518565,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027122427027833577,
      "loss": 5.104,
      "mean_token_accuracy": 0.22222414314746858,
      "num_tokens": 99837079.0,
      "step": 51045
    },
    {
      "entropy": 5.8864655017852785,
      "epoch": 5.102963962613085,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027118858624977166,
      "loss": 5.2949,
      "mean_token_accuracy": 0.21427605003118516,
      "num_tokens": 99847535.0,
      "step": 51050
    },
    {
      "entropy": 5.775228595733642,
      "epoch": 5.103463787674314,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027115290231710177,
      "loss": 5.1493,
      "mean_token_accuracy": 0.22289182841777802,
      "num_tokens": 99857284.0,
      "step": 51055
    },
    {
      "entropy": 5.773278617858887,
      "epoch": 5.1039636127355426,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002711172184812239,
      "loss": 5.1401,
      "mean_token_accuracy": 0.22720126211643218,
      "num_tokens": 99867458.0,
      "step": 51060
    },
    {
      "entropy": 5.816163539886475,
      "epoch": 5.104463437796771,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002710815347430358,
      "loss": 5.1163,
      "mean_token_accuracy": 0.22574896663427352,
      "num_tokens": 99878589.0,
      "step": 51065
    },
    {
      "entropy": 5.866818141937256,
      "epoch": 5.104963262858,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027104585110343526,
      "loss": 5.2199,
      "mean_token_accuracy": 0.21484897881746293,
      "num_tokens": 99887936.0,
      "step": 51070
    },
    {
      "entropy": 5.813540410995484,
      "epoch": 5.105463087919229,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002710101675633201,
      "loss": 5.2448,
      "mean_token_accuracy": 0.218938909471035,
      "num_tokens": 99898293.0,
      "step": 51075
    },
    {
      "entropy": 5.730666065216065,
      "epoch": 5.1059629129804565,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002709744841235881,
      "loss": 5.1043,
      "mean_token_accuracy": 0.2265544757246971,
      "num_tokens": 99907944.0,
      "step": 51080
    },
    {
      "entropy": 5.6970734119415285,
      "epoch": 5.106462738041685,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002709388007851371,
      "loss": 5.0835,
      "mean_token_accuracy": 0.23163426220417022,
      "num_tokens": 99916642.0,
      "step": 51085
    },
    {
      "entropy": 5.797263717651367,
      "epoch": 5.106962563102914,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002709031175488648,
      "loss": 5.1245,
      "mean_token_accuracy": 0.22923059314489364,
      "num_tokens": 99926221.0,
      "step": 51090
    },
    {
      "entropy": 5.77141432762146,
      "epoch": 5.1074623881641426,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027086743441566896,
      "loss": 5.1703,
      "mean_token_accuracy": 0.2285247266292572,
      "num_tokens": 99936971.0,
      "step": 51095
    },
    {
      "entropy": 5.8072772979736325,
      "epoch": 5.107962213225371,
      "grad_norm": 1.25,
      "learning_rate": 0.00027083175138644744,
      "loss": 5.2231,
      "mean_token_accuracy": 0.22266023755073547,
      "num_tokens": 99947714.0,
      "step": 51100
    },
    {
      "entropy": 5.809925365447998,
      "epoch": 5.1084620382866,
      "grad_norm": 1.21875,
      "learning_rate": 0.000270796068462098,
      "loss": 5.172,
      "mean_token_accuracy": 0.22538554221391677,
      "num_tokens": 99956496.0,
      "step": 51105
    },
    {
      "entropy": 5.763925504684448,
      "epoch": 5.108961863347829,
      "grad_norm": 1.125,
      "learning_rate": 0.00027076038564351836,
      "loss": 5.2388,
      "mean_token_accuracy": 0.2149604946374893,
      "num_tokens": 99967320.0,
      "step": 51110
    },
    {
      "entropy": 5.830043172836303,
      "epoch": 5.1094616884090565,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002707247029316062,
      "loss": 5.1555,
      "mean_token_accuracy": 0.21956864595413209,
      "num_tokens": 99977325.0,
      "step": 51115
    },
    {
      "entropy": 5.788533163070679,
      "epoch": 5.109961513470285,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027068902032725963,
      "loss": 5.1468,
      "mean_token_accuracy": 0.2333104819059372,
      "num_tokens": 99986349.0,
      "step": 51120
    },
    {
      "entropy": 5.787645149230957,
      "epoch": 5.110461338531514,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002706533378313761,
      "loss": 5.1274,
      "mean_token_accuracy": 0.22464438080787658,
      "num_tokens": 99996625.0,
      "step": 51125
    },
    {
      "entropy": 5.7719231128692625,
      "epoch": 5.1109611635927426,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027061765544485346,
      "loss": 5.0553,
      "mean_token_accuracy": 0.23849812597036363,
      "num_tokens": 100006048.0,
      "step": 51130
    },
    {
      "entropy": 5.65424656867981,
      "epoch": 5.111460988653971,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002705819731685895,
      "loss": 5.0129,
      "mean_token_accuracy": 0.23541614711284636,
      "num_tokens": 100015611.0,
      "step": 51135
    },
    {
      "entropy": 5.728546476364135,
      "epoch": 5.1119608137152,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000270546291003482,
      "loss": 5.2011,
      "mean_token_accuracy": 0.22120410501956939,
      "num_tokens": 100025210.0,
      "step": 51140
    },
    {
      "entropy": 5.838113069534302,
      "epoch": 5.112460638776429,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002705106089504286,
      "loss": 5.1927,
      "mean_token_accuracy": 0.22089126855134963,
      "num_tokens": 100034435.0,
      "step": 51145
    },
    {
      "entropy": 5.920604944229126,
      "epoch": 5.1129604638376565,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027047492701032716,
      "loss": 5.2339,
      "mean_token_accuracy": 0.22225204557180406,
      "num_tokens": 100045195.0,
      "step": 51150
    },
    {
      "entropy": 5.785375690460205,
      "epoch": 5.113460288898885,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002704392451840755,
      "loss": 5.0939,
      "mean_token_accuracy": 0.227019564807415,
      "num_tokens": 100055058.0,
      "step": 51155
    },
    {
      "entropy": 5.7835853099823,
      "epoch": 5.113960113960114,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002704035634725711,
      "loss": 5.1011,
      "mean_token_accuracy": 0.22978938966989518,
      "num_tokens": 100064418.0,
      "step": 51160
    },
    {
      "entropy": 5.8079570770263675,
      "epoch": 5.1144599390213425,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027036788187671194,
      "loss": 5.1845,
      "mean_token_accuracy": 0.22688582688570022,
      "num_tokens": 100073502.0,
      "step": 51165
    },
    {
      "entropy": 5.72230863571167,
      "epoch": 5.114959764082571,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027033220039739564,
      "loss": 5.0296,
      "mean_token_accuracy": 0.2333139568567276,
      "num_tokens": 100083278.0,
      "step": 51170
    },
    {
      "entropy": 5.76015191078186,
      "epoch": 5.1154595891438,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002702965190355199,
      "loss": 5.1283,
      "mean_token_accuracy": 0.2328397199511528,
      "num_tokens": 100092804.0,
      "step": 51175
    },
    {
      "entropy": 5.7960914134979244,
      "epoch": 5.115959414205029,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002702608377919827,
      "loss": 5.1721,
      "mean_token_accuracy": 0.22256057262420653,
      "num_tokens": 100102248.0,
      "step": 51180
    },
    {
      "entropy": 5.874544858932495,
      "epoch": 5.1164592392662565,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002702251566676815,
      "loss": 5.2562,
      "mean_token_accuracy": 0.21731923371553422,
      "num_tokens": 100111530.0,
      "step": 51185
    },
    {
      "entropy": 5.790547657012939,
      "epoch": 5.116959064327485,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027018947566351416,
      "loss": 5.1546,
      "mean_token_accuracy": 0.21914204955101013,
      "num_tokens": 100121885.0,
      "step": 51190
    },
    {
      "entropy": 5.787432765960693,
      "epoch": 5.117458889388714,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002701537947803783,
      "loss": 5.1878,
      "mean_token_accuracy": 0.21840695589780806,
      "num_tokens": 100132047.0,
      "step": 51195
    },
    {
      "entropy": 5.73820276260376,
      "epoch": 5.1179587144499425,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027011811401917184,
      "loss": 5.1329,
      "mean_token_accuracy": 0.23056559413671493,
      "num_tokens": 100141240.0,
      "step": 51200
    },
    {
      "entropy": 5.8642497062683105,
      "epoch": 5.118458539511171,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002700824333807923,
      "loss": 5.2687,
      "mean_token_accuracy": 0.21720639765262603,
      "num_tokens": 100152924.0,
      "step": 51205
    },
    {
      "entropy": 5.9013269424438475,
      "epoch": 5.1189583645724,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002700467528661375,
      "loss": 5.1841,
      "mean_token_accuracy": 0.2214805394411087,
      "num_tokens": 100163245.0,
      "step": 51210
    },
    {
      "entropy": 5.830171251296997,
      "epoch": 5.119458189633629,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027001107247610505,
      "loss": 5.1878,
      "mean_token_accuracy": 0.22562671452760696,
      "num_tokens": 100172668.0,
      "step": 51215
    },
    {
      "entropy": 5.769739580154419,
      "epoch": 5.1199580146948565,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002699753922115928,
      "loss": 5.1746,
      "mean_token_accuracy": 0.22371030747890472,
      "num_tokens": 100180823.0,
      "step": 51220
    },
    {
      "entropy": 5.78728232383728,
      "epoch": 5.120457839756085,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002699397120734985,
      "loss": 5.1583,
      "mean_token_accuracy": 0.2280287727713585,
      "num_tokens": 100189963.0,
      "step": 51225
    },
    {
      "entropy": 5.838852453231811,
      "epoch": 5.120957664817314,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002699040320627196,
      "loss": 5.169,
      "mean_token_accuracy": 0.22640891820192338,
      "num_tokens": 100199337.0,
      "step": 51230
    },
    {
      "entropy": 5.8185789585113525,
      "epoch": 5.1214574898785425,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026986835218015395,
      "loss": 5.1431,
      "mean_token_accuracy": 0.2210204631090164,
      "num_tokens": 100208389.0,
      "step": 51235
    },
    {
      "entropy": 5.74406213760376,
      "epoch": 5.121957314939771,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026983267242669924,
      "loss": 5.1,
      "mean_token_accuracy": 0.22675098925828935,
      "num_tokens": 100217762.0,
      "step": 51240
    },
    {
      "entropy": 5.754914712905884,
      "epoch": 5.122457140001,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026979699280325317,
      "loss": 5.1176,
      "mean_token_accuracy": 0.225554022192955,
      "num_tokens": 100226818.0,
      "step": 51245
    },
    {
      "entropy": 5.714050769805908,
      "epoch": 5.122956965062229,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026976131331071345,
      "loss": 5.0707,
      "mean_token_accuracy": 0.23121072500944137,
      "num_tokens": 100235522.0,
      "step": 51250
    },
    {
      "entropy": 5.82019944190979,
      "epoch": 5.1234567901234565,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026972563394997776,
      "loss": 5.1988,
      "mean_token_accuracy": 0.22299302071332933,
      "num_tokens": 100245317.0,
      "step": 51255
    },
    {
      "entropy": 5.729621696472168,
      "epoch": 5.123956615184685,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002696899547219437,
      "loss": 5.1304,
      "mean_token_accuracy": 0.2296872153878212,
      "num_tokens": 100254745.0,
      "step": 51260
    },
    {
      "entropy": 5.786914682388305,
      "epoch": 5.124456440245914,
      "grad_norm": 1.15625,
      "learning_rate": 0.000269654275627509,
      "loss": 5.1992,
      "mean_token_accuracy": 0.2219938114285469,
      "num_tokens": 100263607.0,
      "step": 51265
    },
    {
      "entropy": 5.742400789260865,
      "epoch": 5.1249562653071425,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026961859666757137,
      "loss": 5.1504,
      "mean_token_accuracy": 0.23041100651025773,
      "num_tokens": 100274253.0,
      "step": 51270
    },
    {
      "entropy": 5.805403804779052,
      "epoch": 5.125456090368371,
      "grad_norm": 1.25,
      "learning_rate": 0.00026958291784302845,
      "loss": 5.1214,
      "mean_token_accuracy": 0.23206258863210677,
      "num_tokens": 100284838.0,
      "step": 51275
    },
    {
      "entropy": 5.793075037002564,
      "epoch": 5.1259559154296,
      "grad_norm": 1.296875,
      "learning_rate": 0.000269547239154778,
      "loss": 5.1205,
      "mean_token_accuracy": 0.222331266105175,
      "num_tokens": 100295064.0,
      "step": 51280
    },
    {
      "entropy": 5.809178829193115,
      "epoch": 5.126455740490829,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026951156060371753,
      "loss": 5.1555,
      "mean_token_accuracy": 0.22975969612598418,
      "num_tokens": 100304224.0,
      "step": 51285
    },
    {
      "entropy": 5.69410400390625,
      "epoch": 5.1269555655520564,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002694758821907448,
      "loss": 5.0933,
      "mean_token_accuracy": 0.23569890409708022,
      "num_tokens": 100313429.0,
      "step": 51290
    },
    {
      "entropy": 5.764275217056275,
      "epoch": 5.127455390613285,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026944020391675747,
      "loss": 5.2258,
      "mean_token_accuracy": 0.22239215970039367,
      "num_tokens": 100323263.0,
      "step": 51295
    },
    {
      "entropy": 5.772494268417359,
      "epoch": 5.127955215674514,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026940452578265325,
      "loss": 5.1319,
      "mean_token_accuracy": 0.23378317803144455,
      "num_tokens": 100333259.0,
      "step": 51300
    },
    {
      "entropy": 5.776280403137207,
      "epoch": 5.1284550407357425,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002693688477893296,
      "loss": 5.1003,
      "mean_token_accuracy": 0.2219247490167618,
      "num_tokens": 100342473.0,
      "step": 51305
    },
    {
      "entropy": 5.768207740783692,
      "epoch": 5.128954865796971,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026933316993768433,
      "loss": 5.1332,
      "mean_token_accuracy": 0.22315005213022232,
      "num_tokens": 100352336.0,
      "step": 51310
    },
    {
      "entropy": 5.812569093704224,
      "epoch": 5.1294546908582,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026929749222861507,
      "loss": 5.1873,
      "mean_token_accuracy": 0.2233409658074379,
      "num_tokens": 100363098.0,
      "step": 51315
    },
    {
      "entropy": 5.861482810974121,
      "epoch": 5.129954515919428,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002692618146630194,
      "loss": 5.2092,
      "mean_token_accuracy": 0.22255096733570098,
      "num_tokens": 100372140.0,
      "step": 51320
    },
    {
      "entropy": 5.782766532897949,
      "epoch": 5.1304543409806564,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026922613724179504,
      "loss": 5.1547,
      "mean_token_accuracy": 0.22895591109991073,
      "num_tokens": 100381063.0,
      "step": 51325
    },
    {
      "entropy": 5.761558389663696,
      "epoch": 5.130954166041885,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026919045996583955,
      "loss": 5.1227,
      "mean_token_accuracy": 0.22833715826272966,
      "num_tokens": 100392114.0,
      "step": 51330
    },
    {
      "entropy": 5.710898399353027,
      "epoch": 5.131453991103114,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026915478283605056,
      "loss": 5.0857,
      "mean_token_accuracy": 0.23428264260292053,
      "num_tokens": 100402817.0,
      "step": 51335
    },
    {
      "entropy": 5.889938640594482,
      "epoch": 5.1319538161643425,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002691191058533258,
      "loss": 5.1919,
      "mean_token_accuracy": 0.22166251093149186,
      "num_tokens": 100412082.0,
      "step": 51340
    },
    {
      "entropy": 5.841456747055053,
      "epoch": 5.132453641225571,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026908342901856274,
      "loss": 5.1982,
      "mean_token_accuracy": 0.21792595088481903,
      "num_tokens": 100421365.0,
      "step": 51345
    },
    {
      "entropy": 5.78624324798584,
      "epoch": 5.1329534662868,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026904775233265913,
      "loss": 5.1377,
      "mean_token_accuracy": 0.22693527042865752,
      "num_tokens": 100431248.0,
      "step": 51350
    },
    {
      "entropy": 5.7866637229919435,
      "epoch": 5.133453291348028,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026901207579651257,
      "loss": 5.2588,
      "mean_token_accuracy": 0.21861800998449327,
      "num_tokens": 100441736.0,
      "step": 51355
    },
    {
      "entropy": 5.816161823272705,
      "epoch": 5.1339531164092564,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002689763994110207,
      "loss": 5.1822,
      "mean_token_accuracy": 0.21777766346931457,
      "num_tokens": 100450842.0,
      "step": 51360
    },
    {
      "entropy": 5.789847040176392,
      "epoch": 5.134452941470485,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026894072317708095,
      "loss": 5.1905,
      "mean_token_accuracy": 0.22055551558732986,
      "num_tokens": 100460885.0,
      "step": 51365
    },
    {
      "entropy": 5.733582305908203,
      "epoch": 5.134952766531714,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002689050470955911,
      "loss": 5.1025,
      "mean_token_accuracy": 0.23662927150726318,
      "num_tokens": 100470592.0,
      "step": 51370
    },
    {
      "entropy": 5.8524942874908445,
      "epoch": 5.1354525915929425,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002688693711674487,
      "loss": 5.2869,
      "mean_token_accuracy": 0.2119527667760849,
      "num_tokens": 100480258.0,
      "step": 51375
    },
    {
      "entropy": 5.774976301193237,
      "epoch": 5.135952416654171,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026883369539355134,
      "loss": 5.1924,
      "mean_token_accuracy": 0.22562195956707,
      "num_tokens": 100490049.0,
      "step": 51380
    },
    {
      "entropy": 5.772374010086059,
      "epoch": 5.1364522417154,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026879801977479654,
      "loss": 5.1027,
      "mean_token_accuracy": 0.23008862435817717,
      "num_tokens": 100499386.0,
      "step": 51385
    },
    {
      "entropy": 5.845327520370484,
      "epoch": 5.136952066776628,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002687623443120821,
      "loss": 5.2197,
      "mean_token_accuracy": 0.2161787673830986,
      "num_tokens": 100509756.0,
      "step": 51390
    },
    {
      "entropy": 5.692055630683899,
      "epoch": 5.137451891837856,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002687266690063053,
      "loss": 5.0252,
      "mean_token_accuracy": 0.22987312823534012,
      "num_tokens": 100519188.0,
      "step": 51395
    },
    {
      "entropy": 5.738663578033448,
      "epoch": 5.137951716899085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000268690993858364,
      "loss": 5.0455,
      "mean_token_accuracy": 0.2365707442164421,
      "num_tokens": 100528444.0,
      "step": 51400
    },
    {
      "entropy": 5.721014833450317,
      "epoch": 5.138451541960314,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026865531886915566,
      "loss": 5.1402,
      "mean_token_accuracy": 0.22588685899972916,
      "num_tokens": 100538552.0,
      "step": 51405
    },
    {
      "entropy": 5.727406930923462,
      "epoch": 5.1389513670215425,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002686196440395779,
      "loss": 5.1691,
      "mean_token_accuracy": 0.2317688062787056,
      "num_tokens": 100548909.0,
      "step": 51410
    },
    {
      "entropy": 5.877586889266968,
      "epoch": 5.139451192082771,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002685839693705282,
      "loss": 5.1869,
      "mean_token_accuracy": 0.219836962223053,
      "num_tokens": 100557963.0,
      "step": 51415
    },
    {
      "entropy": 5.895894861221313,
      "epoch": 5.139951017144,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026854829486290424,
      "loss": 5.2255,
      "mean_token_accuracy": 0.22254960536956786,
      "num_tokens": 100567253.0,
      "step": 51420
    },
    {
      "entropy": 5.817276859283448,
      "epoch": 5.140450842205228,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002685126205176035,
      "loss": 5.2519,
      "mean_token_accuracy": 0.2141656756401062,
      "num_tokens": 100577771.0,
      "step": 51425
    },
    {
      "entropy": 5.814854145050049,
      "epoch": 5.140950667266456,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002684769463355236,
      "loss": 5.2015,
      "mean_token_accuracy": 0.2272321492433548,
      "num_tokens": 100588236.0,
      "step": 51430
    },
    {
      "entropy": 5.736982297897339,
      "epoch": 5.141450492327685,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026844127231756196,
      "loss": 5.0603,
      "mean_token_accuracy": 0.2290283203125,
      "num_tokens": 100597709.0,
      "step": 51435
    },
    {
      "entropy": 5.5930859565734865,
      "epoch": 5.141950317388914,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002684055984646163,
      "loss": 5.0204,
      "mean_token_accuracy": 0.24540239423513413,
      "num_tokens": 100607908.0,
      "step": 51440
    },
    {
      "entropy": 5.726786375045776,
      "epoch": 5.1424501424501425,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000268369924777584,
      "loss": 5.1435,
      "mean_token_accuracy": 0.22585287988185881,
      "num_tokens": 100618794.0,
      "step": 51445
    },
    {
      "entropy": 5.8199139595031735,
      "epoch": 5.142949967511371,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002683342512573628,
      "loss": 5.1631,
      "mean_token_accuracy": 0.22094788104295732,
      "num_tokens": 100627577.0,
      "step": 51450
    },
    {
      "entropy": 5.754178667068482,
      "epoch": 5.1434497925726,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026829857790485,
      "loss": 5.1108,
      "mean_token_accuracy": 0.22880731523036957,
      "num_tokens": 100637443.0,
      "step": 51455
    },
    {
      "entropy": 5.679537105560303,
      "epoch": 5.143949617633828,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026826290472094325,
      "loss": 4.9893,
      "mean_token_accuracy": 0.24743707329034806,
      "num_tokens": 100646329.0,
      "step": 51460
    },
    {
      "entropy": 5.735430908203125,
      "epoch": 5.144449442695056,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026822723170654014,
      "loss": 5.1597,
      "mean_token_accuracy": 0.22166595309972764,
      "num_tokens": 100655835.0,
      "step": 51465
    },
    {
      "entropy": 5.721211242675781,
      "epoch": 5.144949267756285,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026819155886253815,
      "loss": 5.1327,
      "mean_token_accuracy": 0.2312879353761673,
      "num_tokens": 100665107.0,
      "step": 51470
    },
    {
      "entropy": 5.749836540222168,
      "epoch": 5.145449092817514,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002681558861898348,
      "loss": 5.1038,
      "mean_token_accuracy": 0.22629644125699996,
      "num_tokens": 100673727.0,
      "step": 51475
    },
    {
      "entropy": 5.839484119415284,
      "epoch": 5.1459489178787425,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002681202136893275,
      "loss": 5.2481,
      "mean_token_accuracy": 0.21470581144094467,
      "num_tokens": 100683308.0,
      "step": 51480
    },
    {
      "entropy": 5.792268562316894,
      "epoch": 5.146448742939971,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026808454136191393,
      "loss": 5.1347,
      "mean_token_accuracy": 0.2289538562297821,
      "num_tokens": 100692941.0,
      "step": 51485
    },
    {
      "entropy": 5.786446237564087,
      "epoch": 5.1469485680012,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026804886920849154,
      "loss": 5.1009,
      "mean_token_accuracy": 0.23388078063726425,
      "num_tokens": 100702072.0,
      "step": 51490
    },
    {
      "entropy": 5.876263809204102,
      "epoch": 5.147448393062428,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026801319722995784,
      "loss": 5.1927,
      "mean_token_accuracy": 0.2223934769630432,
      "num_tokens": 100711948.0,
      "step": 51495
    },
    {
      "entropy": 5.824765348434449,
      "epoch": 5.147948218123656,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026797752542721023,
      "loss": 5.2295,
      "mean_token_accuracy": 0.22447639256715773,
      "num_tokens": 100721415.0,
      "step": 51500
    },
    {
      "entropy": 5.786394786834717,
      "epoch": 5.148448043184885,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026794185380114627,
      "loss": 5.1914,
      "mean_token_accuracy": 0.21620631366968154,
      "num_tokens": 100731577.0,
      "step": 51505
    },
    {
      "entropy": 5.768067789077759,
      "epoch": 5.148947868246114,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002679061823526635,
      "loss": 5.1195,
      "mean_token_accuracy": 0.22375988513231276,
      "num_tokens": 100741323.0,
      "step": 51510
    },
    {
      "entropy": 5.907409572601319,
      "epoch": 5.1494476933073425,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002678705110826593,
      "loss": 5.3763,
      "mean_token_accuracy": 0.20583563446998596,
      "num_tokens": 100751873.0,
      "step": 51515
    },
    {
      "entropy": 5.71413631439209,
      "epoch": 5.149947518368571,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002678348399920312,
      "loss": 5.0572,
      "mean_token_accuracy": 0.23989670276641845,
      "num_tokens": 100761687.0,
      "step": 51520
    },
    {
      "entropy": 5.845649528503418,
      "epoch": 5.1504473434298,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002677991690816767,
      "loss": 5.1888,
      "mean_token_accuracy": 0.22007959336042404,
      "num_tokens": 100770623.0,
      "step": 51525
    },
    {
      "entropy": 5.792565822601318,
      "epoch": 5.150947168491028,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002677634983524933,
      "loss": 5.1947,
      "mean_token_accuracy": 0.21883547008037568,
      "num_tokens": 100779786.0,
      "step": 51530
    },
    {
      "entropy": 5.802640295028686,
      "epoch": 5.151446993552256,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026772782780537833,
      "loss": 5.2408,
      "mean_token_accuracy": 0.22466055303812027,
      "num_tokens": 100789192.0,
      "step": 51535
    },
    {
      "entropy": 5.874287176132202,
      "epoch": 5.151946818613485,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002676921574412294,
      "loss": 5.2906,
      "mean_token_accuracy": 0.21584732681512833,
      "num_tokens": 100799322.0,
      "step": 51540
    },
    {
      "entropy": 5.784068822860718,
      "epoch": 5.152446643674714,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026765648726094396,
      "loss": 5.0478,
      "mean_token_accuracy": 0.23048424869775772,
      "num_tokens": 100807581.0,
      "step": 51545
    },
    {
      "entropy": 5.794895315170288,
      "epoch": 5.1529464687359425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026762081726541934,
      "loss": 5.1897,
      "mean_token_accuracy": 0.21987762153148652,
      "num_tokens": 100818052.0,
      "step": 51550
    },
    {
      "entropy": 5.773079967498779,
      "epoch": 5.153446293797171,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026758514745555306,
      "loss": 5.21,
      "mean_token_accuracy": 0.2188539609313011,
      "num_tokens": 100827826.0,
      "step": 51555
    },
    {
      "entropy": 5.716075992584228,
      "epoch": 5.1539461188584,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026754947783224253,
      "loss": 5.0569,
      "mean_token_accuracy": 0.23217764794826506,
      "num_tokens": 100837518.0,
      "step": 51560
    },
    {
      "entropy": 5.8281392574310305,
      "epoch": 5.154445943919628,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002675138083963853,
      "loss": 5.1801,
      "mean_token_accuracy": 0.22473290264606477,
      "num_tokens": 100846677.0,
      "step": 51565
    },
    {
      "entropy": 5.843825578689575,
      "epoch": 5.154945768980856,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002674781391488788,
      "loss": 5.2354,
      "mean_token_accuracy": 0.21680028289556502,
      "num_tokens": 100856078.0,
      "step": 51570
    },
    {
      "entropy": 5.8174539566040036,
      "epoch": 5.155445594042085,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002674424700906203,
      "loss": 5.168,
      "mean_token_accuracy": 0.22527939975261688,
      "num_tokens": 100866211.0,
      "step": 51575
    },
    {
      "entropy": 5.861810970306396,
      "epoch": 5.155945419103314,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002674068012225072,
      "loss": 5.2708,
      "mean_token_accuracy": 0.22099674940109254,
      "num_tokens": 100875629.0,
      "step": 51580
    },
    {
      "entropy": 5.726769018173218,
      "epoch": 5.1564452441645425,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002673711325454371,
      "loss": 5.1426,
      "mean_token_accuracy": 0.2294452369213104,
      "num_tokens": 100884352.0,
      "step": 51585
    },
    {
      "entropy": 5.802477312088013,
      "epoch": 5.156945069225771,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002673354640603073,
      "loss": 5.1658,
      "mean_token_accuracy": 0.22861281782388687,
      "num_tokens": 100893340.0,
      "step": 51590
    },
    {
      "entropy": 5.728754329681396,
      "epoch": 5.157444894287,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026729979576801537,
      "loss": 5.1119,
      "mean_token_accuracy": 0.23398513048887254,
      "num_tokens": 100902870.0,
      "step": 51595
    },
    {
      "entropy": 5.7668318271636965,
      "epoch": 5.157944719348228,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002672641276694585,
      "loss": 5.2408,
      "mean_token_accuracy": 0.22125123888254167,
      "num_tokens": 100911982.0,
      "step": 51600
    },
    {
      "entropy": 5.791902542114258,
      "epoch": 5.158444544409456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026722845976553423,
      "loss": 5.1913,
      "mean_token_accuracy": 0.23379833847284318,
      "num_tokens": 100922475.0,
      "step": 51605
    },
    {
      "entropy": 5.849876642227173,
      "epoch": 5.158944369470685,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002671927920571399,
      "loss": 5.1466,
      "mean_token_accuracy": 0.22690588980913162,
      "num_tokens": 100932096.0,
      "step": 51610
    },
    {
      "entropy": 5.86428656578064,
      "epoch": 5.159444194531914,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002671571245451729,
      "loss": 5.206,
      "mean_token_accuracy": 0.2169743835926056,
      "num_tokens": 100941824.0,
      "step": 51615
    },
    {
      "entropy": 5.7832419872283936,
      "epoch": 5.1599440195931425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026712145723053066,
      "loss": 5.1403,
      "mean_token_accuracy": 0.23007318526506423,
      "num_tokens": 100950984.0,
      "step": 51620
    },
    {
      "entropy": 5.874851179122925,
      "epoch": 5.160443844654371,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002670857901141105,
      "loss": 5.2678,
      "mean_token_accuracy": 0.21696273535490035,
      "num_tokens": 100961326.0,
      "step": 51625
    },
    {
      "entropy": 5.779034614562988,
      "epoch": 5.1609436697156,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002670501231968098,
      "loss": 5.0883,
      "mean_token_accuracy": 0.2309202179312706,
      "num_tokens": 100970429.0,
      "step": 51630
    },
    {
      "entropy": 5.885172557830811,
      "epoch": 5.161443494776828,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000267014456479526,
      "loss": 5.3203,
      "mean_token_accuracy": 0.21144025325775145,
      "num_tokens": 100979745.0,
      "step": 51635
    },
    {
      "entropy": 5.819743299484253,
      "epoch": 5.161943319838056,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026697878996315654,
      "loss": 5.1458,
      "mean_token_accuracy": 0.22758897989988328,
      "num_tokens": 100988241.0,
      "step": 51640
    },
    {
      "entropy": 5.812612247467041,
      "epoch": 5.162443144899285,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002669431236485985,
      "loss": 5.1865,
      "mean_token_accuracy": 0.22468767762184144,
      "num_tokens": 100997408.0,
      "step": 51645
    },
    {
      "entropy": 5.702940130233765,
      "epoch": 5.162942969960514,
      "grad_norm": 1.375,
      "learning_rate": 0.0002669074575367494,
      "loss": 5.0665,
      "mean_token_accuracy": 0.2326718881726265,
      "num_tokens": 101006070.0,
      "step": 51650
    },
    {
      "entropy": 5.759203815460205,
      "epoch": 5.1634427950217425,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026687179162850663,
      "loss": 5.1087,
      "mean_token_accuracy": 0.22182652801275254,
      "num_tokens": 101015709.0,
      "step": 51655
    },
    {
      "entropy": 5.783723402023315,
      "epoch": 5.163942620082971,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026683612592476747,
      "loss": 5.1437,
      "mean_token_accuracy": 0.22118050754070281,
      "num_tokens": 101025817.0,
      "step": 51660
    },
    {
      "entropy": 5.806941509246826,
      "epoch": 5.1644424451442,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026680046042642933,
      "loss": 5.2414,
      "mean_token_accuracy": 0.22494736611843108,
      "num_tokens": 101036217.0,
      "step": 51665
    },
    {
      "entropy": 5.7677717208862305,
      "epoch": 5.164942270205428,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002667647951343894,
      "loss": 5.1403,
      "mean_token_accuracy": 0.2243684709072113,
      "num_tokens": 101046193.0,
      "step": 51670
    },
    {
      "entropy": 5.906327581405639,
      "epoch": 5.165442095266656,
      "grad_norm": 1.25,
      "learning_rate": 0.00026672913004954513,
      "loss": 5.2506,
      "mean_token_accuracy": 0.2169728770852089,
      "num_tokens": 101055476.0,
      "step": 51675
    },
    {
      "entropy": 5.850387716293335,
      "epoch": 5.165941920327885,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002666934651727938,
      "loss": 5.1556,
      "mean_token_accuracy": 0.22527254074811937,
      "num_tokens": 101065678.0,
      "step": 51680
    },
    {
      "entropy": 5.858238935470581,
      "epoch": 5.166441745389114,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002666578005050328,
      "loss": 5.2186,
      "mean_token_accuracy": 0.2099147379398346,
      "num_tokens": 101075599.0,
      "step": 51685
    },
    {
      "entropy": 5.769926691055298,
      "epoch": 5.1669415704503425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026662213604715943,
      "loss": 5.13,
      "mean_token_accuracy": 0.2208623006939888,
      "num_tokens": 101085754.0,
      "step": 51690
    },
    {
      "entropy": 5.759210109710693,
      "epoch": 5.167441395511571,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002665864718000709,
      "loss": 5.1636,
      "mean_token_accuracy": 0.22313259541988373,
      "num_tokens": 101096301.0,
      "step": 51695
    },
    {
      "entropy": 5.73471508026123,
      "epoch": 5.1679412205728,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002665508077646646,
      "loss": 5.1684,
      "mean_token_accuracy": 0.22152531147003174,
      "num_tokens": 101107053.0,
      "step": 51700
    },
    {
      "entropy": 5.761846017837525,
      "epoch": 5.168441045634028,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026651514394183777,
      "loss": 5.0468,
      "mean_token_accuracy": 0.2375819504261017,
      "num_tokens": 101118063.0,
      "step": 51705
    },
    {
      "entropy": 5.7198248386383055,
      "epoch": 5.168940870695256,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026647948033248784,
      "loss": 5.1107,
      "mean_token_accuracy": 0.2320726603269577,
      "num_tokens": 101127268.0,
      "step": 51710
    },
    {
      "entropy": 5.768090486526489,
      "epoch": 5.169440695756485,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002664438169375119,
      "loss": 5.1485,
      "mean_token_accuracy": 0.23282405138015747,
      "num_tokens": 101136258.0,
      "step": 51715
    },
    {
      "entropy": 5.78148250579834,
      "epoch": 5.169940520817714,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026640815375780736,
      "loss": 5.1874,
      "mean_token_accuracy": 0.21908533275127412,
      "num_tokens": 101146192.0,
      "step": 51720
    },
    {
      "entropy": 5.834186649322509,
      "epoch": 5.1704403458789425,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002663724907942714,
      "loss": 5.1322,
      "mean_token_accuracy": 0.22400073260068892,
      "num_tokens": 101155221.0,
      "step": 51725
    },
    {
      "entropy": 5.777172327041626,
      "epoch": 5.170940170940171,
      "grad_norm": 1.25,
      "learning_rate": 0.0002663368280478014,
      "loss": 5.2239,
      "mean_token_accuracy": 0.22163686603307725,
      "num_tokens": 101165524.0,
      "step": 51730
    },
    {
      "entropy": 5.7638906955719,
      "epoch": 5.1714399960014,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026630116551929454,
      "loss": 5.1555,
      "mean_token_accuracy": 0.232869653403759,
      "num_tokens": 101174498.0,
      "step": 51735
    },
    {
      "entropy": 5.792409515380859,
      "epoch": 5.171939821062628,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026626550320964817,
      "loss": 5.1288,
      "mean_token_accuracy": 0.2268853411078453,
      "num_tokens": 101183997.0,
      "step": 51740
    },
    {
      "entropy": 5.792769956588745,
      "epoch": 5.172439646123856,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002662298411197594,
      "loss": 5.0924,
      "mean_token_accuracy": 0.22877470552921295,
      "num_tokens": 101193198.0,
      "step": 51745
    },
    {
      "entropy": 5.838634586334228,
      "epoch": 5.172939471185085,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026619417925052564,
      "loss": 5.2347,
      "mean_token_accuracy": 0.2176954835653305,
      "num_tokens": 101203401.0,
      "step": 51750
    },
    {
      "entropy": 5.696364784240723,
      "epoch": 5.173439296246314,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002661585176028441,
      "loss": 5.0702,
      "mean_token_accuracy": 0.22926406413316727,
      "num_tokens": 101212007.0,
      "step": 51755
    },
    {
      "entropy": 5.829201602935791,
      "epoch": 5.1739391213075425,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002661228561776119,
      "loss": 5.2014,
      "mean_token_accuracy": 0.21997696310281753,
      "num_tokens": 101221770.0,
      "step": 51760
    },
    {
      "entropy": 5.846955108642578,
      "epoch": 5.174438946368771,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002660871949757264,
      "loss": 5.2312,
      "mean_token_accuracy": 0.21308554261922835,
      "num_tokens": 101231172.0,
      "step": 51765
    },
    {
      "entropy": 5.806149005889893,
      "epoch": 5.17493877143,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002660515339980848,
      "loss": 5.1873,
      "mean_token_accuracy": 0.22139165103435515,
      "num_tokens": 101241125.0,
      "step": 51770
    },
    {
      "entropy": 5.844179773330689,
      "epoch": 5.175438596491228,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002660158732455843,
      "loss": 5.2525,
      "mean_token_accuracy": 0.22128809988498688,
      "num_tokens": 101251857.0,
      "step": 51775
    },
    {
      "entropy": 5.777693128585815,
      "epoch": 5.175938421552456,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026598021271912207,
      "loss": 5.1531,
      "mean_token_accuracy": 0.22339731752872466,
      "num_tokens": 101261039.0,
      "step": 51780
    },
    {
      "entropy": 5.8236644744873045,
      "epoch": 5.176438246613685,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002659445524195955,
      "loss": 5.1828,
      "mean_token_accuracy": 0.21447802931070328,
      "num_tokens": 101269967.0,
      "step": 51785
    },
    {
      "entropy": 5.792267084121704,
      "epoch": 5.176938071674914,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002659088923479015,
      "loss": 5.0927,
      "mean_token_accuracy": 0.23299963176250457,
      "num_tokens": 101279575.0,
      "step": 51790
    },
    {
      "entropy": 5.915919351577759,
      "epoch": 5.1774378967361425,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002658732325049375,
      "loss": 5.3431,
      "mean_token_accuracy": 0.20917152911424636,
      "num_tokens": 101289765.0,
      "step": 51795
    },
    {
      "entropy": 5.752347087860107,
      "epoch": 5.177937721797371,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002658375728916006,
      "loss": 5.1344,
      "mean_token_accuracy": 0.22468120753765106,
      "num_tokens": 101300078.0,
      "step": 51800
    },
    {
      "entropy": 5.937742137908936,
      "epoch": 5.1784375468586,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026580191350878795,
      "loss": 5.3579,
      "mean_token_accuracy": 0.20697163343429564,
      "num_tokens": 101311078.0,
      "step": 51805
    },
    {
      "entropy": 5.88294620513916,
      "epoch": 5.178937371919828,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026576625435739687,
      "loss": 5.1902,
      "mean_token_accuracy": 0.22588142156600952,
      "num_tokens": 101320990.0,
      "step": 51810
    },
    {
      "entropy": 5.769101333618164,
      "epoch": 5.179437196981056,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026573059543832445,
      "loss": 5.1712,
      "mean_token_accuracy": 0.22055114358663558,
      "num_tokens": 101331210.0,
      "step": 51815
    },
    {
      "entropy": 5.787077569961548,
      "epoch": 5.179937022042285,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002656949367524679,
      "loss": 5.2273,
      "mean_token_accuracy": 0.2219620391726494,
      "num_tokens": 101341370.0,
      "step": 51820
    },
    {
      "entropy": 5.725761842727661,
      "epoch": 5.180436847103514,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002656592783007242,
      "loss": 5.098,
      "mean_token_accuracy": 0.22960303276777266,
      "num_tokens": 101352014.0,
      "step": 51825
    },
    {
      "entropy": 5.771445417404175,
      "epoch": 5.1809366721647425,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026562362008399076,
      "loss": 5.1751,
      "mean_token_accuracy": 0.23073949664831161,
      "num_tokens": 101362008.0,
      "step": 51830
    },
    {
      "entropy": 5.788714742660522,
      "epoch": 5.181436497225971,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002655879621031647,
      "loss": 5.1522,
      "mean_token_accuracy": 0.22092693001031877,
      "num_tokens": 101372178.0,
      "step": 51835
    },
    {
      "entropy": 5.823539924621582,
      "epoch": 5.1819363222872,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000265552304359143,
      "loss": 5.2103,
      "mean_token_accuracy": 0.2206999972462654,
      "num_tokens": 101382638.0,
      "step": 51840
    },
    {
      "entropy": 5.831222295761108,
      "epoch": 5.182436147348428,
      "grad_norm": 1.25,
      "learning_rate": 0.0002655166468528228,
      "loss": 5.1937,
      "mean_token_accuracy": 0.21825207769870758,
      "num_tokens": 101392907.0,
      "step": 51845
    },
    {
      "entropy": 5.776800298690796,
      "epoch": 5.182935972409656,
      "grad_norm": 1.125,
      "learning_rate": 0.00026548098958510154,
      "loss": 5.1973,
      "mean_token_accuracy": 0.21474234163761138,
      "num_tokens": 101402809.0,
      "step": 51850
    },
    {
      "entropy": 5.817693281173706,
      "epoch": 5.183435797470885,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026544533255687603,
      "loss": 5.1277,
      "mean_token_accuracy": 0.228373920917511,
      "num_tokens": 101412648.0,
      "step": 51855
    },
    {
      "entropy": 5.783969306945801,
      "epoch": 5.183935622532114,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026540967576904346,
      "loss": 5.1344,
      "mean_token_accuracy": 0.22330974638462067,
      "num_tokens": 101422567.0,
      "step": 51860
    },
    {
      "entropy": 5.611084985733032,
      "epoch": 5.1844354475933425,
      "grad_norm": 1.296875,
      "learning_rate": 0.000265374019222501,
      "loss": 4.8703,
      "mean_token_accuracy": 0.2601349249482155,
      "num_tokens": 101432584.0,
      "step": 51865
    },
    {
      "entropy": 5.775690698623658,
      "epoch": 5.184935272654571,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026533836291814573,
      "loss": 5.1571,
      "mean_token_accuracy": 0.22960441559553146,
      "num_tokens": 101442780.0,
      "step": 51870
    },
    {
      "entropy": 5.790964412689209,
      "epoch": 5.1854350977158,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026530270685687477,
      "loss": 5.0856,
      "mean_token_accuracy": 0.2330562502145767,
      "num_tokens": 101451439.0,
      "step": 51875
    },
    {
      "entropy": 5.813404703140259,
      "epoch": 5.185934922777028,
      "grad_norm": 1.25,
      "learning_rate": 0.00026526705103958523,
      "loss": 5.1579,
      "mean_token_accuracy": 0.22388888001441956,
      "num_tokens": 101461056.0,
      "step": 51880
    },
    {
      "entropy": 5.801596641540527,
      "epoch": 5.186434747838256,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026523139546717414,
      "loss": 5.23,
      "mean_token_accuracy": 0.2191889226436615,
      "num_tokens": 101471691.0,
      "step": 51885
    },
    {
      "entropy": 5.757204151153564,
      "epoch": 5.186934572899485,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026519574014053867,
      "loss": 5.1605,
      "mean_token_accuracy": 0.22614913880825044,
      "num_tokens": 101481961.0,
      "step": 51890
    },
    {
      "entropy": 5.6732683181762695,
      "epoch": 5.187434397960714,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026516008506057584,
      "loss": 5.0408,
      "mean_token_accuracy": 0.22695264518260955,
      "num_tokens": 101491790.0,
      "step": 51895
    },
    {
      "entropy": 5.819476079940796,
      "epoch": 5.1879342230219425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002651244302281827,
      "loss": 5.292,
      "mean_token_accuracy": 0.21568183302879335,
      "num_tokens": 101500401.0,
      "step": 51900
    },
    {
      "entropy": 5.777219915390015,
      "epoch": 5.188434048083171,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026508877564425636,
      "loss": 5.155,
      "mean_token_accuracy": 0.21905574351549148,
      "num_tokens": 101510635.0,
      "step": 51905
    },
    {
      "entropy": 5.80887360572815,
      "epoch": 5.1889338731444,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002650531213096939,
      "loss": 5.1474,
      "mean_token_accuracy": 0.22298979312181472,
      "num_tokens": 101519592.0,
      "step": 51910
    },
    {
      "entropy": 5.82970027923584,
      "epoch": 5.189433698205628,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026501746722539237,
      "loss": 5.2622,
      "mean_token_accuracy": 0.2159685805439949,
      "num_tokens": 101530567.0,
      "step": 51915
    },
    {
      "entropy": 5.693635559082031,
      "epoch": 5.189933523266856,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026498181339224886,
      "loss": 5.0876,
      "mean_token_accuracy": 0.23954839557409285,
      "num_tokens": 101540884.0,
      "step": 51920
    },
    {
      "entropy": 5.728108835220337,
      "epoch": 5.190433348328085,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002649461598111602,
      "loss": 5.099,
      "mean_token_accuracy": 0.24123583137989044,
      "num_tokens": 101551267.0,
      "step": 51925
    },
    {
      "entropy": 5.850828552246094,
      "epoch": 5.190933173389314,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002649105064830237,
      "loss": 5.2262,
      "mean_token_accuracy": 0.22788822054862976,
      "num_tokens": 101560996.0,
      "step": 51930
    },
    {
      "entropy": 5.7283287525177,
      "epoch": 5.1914329984505425,
      "grad_norm": 1.375,
      "learning_rate": 0.0002648748534087361,
      "loss": 5.1057,
      "mean_token_accuracy": 0.23832689523696898,
      "num_tokens": 101570312.0,
      "step": 51935
    },
    {
      "entropy": 5.778324270248413,
      "epoch": 5.191932823511771,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002648392005891947,
      "loss": 5.1403,
      "mean_token_accuracy": 0.22444833666086197,
      "num_tokens": 101579957.0,
      "step": 51940
    },
    {
      "entropy": 5.720019912719726,
      "epoch": 5.192432648572999,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026480354802529645,
      "loss": 5.0796,
      "mean_token_accuracy": 0.23937107771635055,
      "num_tokens": 101590985.0,
      "step": 51945
    },
    {
      "entropy": 5.765087223052978,
      "epoch": 5.192932473634228,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002647678957179382,
      "loss": 5.1074,
      "mean_token_accuracy": 0.22728779315948486,
      "num_tokens": 101599607.0,
      "step": 51950
    },
    {
      "entropy": 5.773763608932495,
      "epoch": 5.193432298695456,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026473224366801713,
      "loss": 5.1897,
      "mean_token_accuracy": 0.21823360323905944,
      "num_tokens": 101608858.0,
      "step": 51955
    },
    {
      "entropy": 5.805203676223755,
      "epoch": 5.193932123756685,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002646965918764301,
      "loss": 5.1641,
      "mean_token_accuracy": 0.21848368048667907,
      "num_tokens": 101620052.0,
      "step": 51960
    },
    {
      "entropy": 5.75172700881958,
      "epoch": 5.194431948817914,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026466094034407415,
      "loss": 5.166,
      "mean_token_accuracy": 0.22093774378299713,
      "num_tokens": 101629887.0,
      "step": 51965
    },
    {
      "entropy": 5.847396469116211,
      "epoch": 5.1949317738791425,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026462528907184637,
      "loss": 5.1993,
      "mean_token_accuracy": 0.21519881933927537,
      "num_tokens": 101639355.0,
      "step": 51970
    },
    {
      "entropy": 5.919387340545654,
      "epoch": 5.195431598940371,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002645896380606436,
      "loss": 5.3066,
      "mean_token_accuracy": 0.20975254476070404,
      "num_tokens": 101648949.0,
      "step": 51975
    },
    {
      "entropy": 5.801178216934204,
      "epoch": 5.195931424001599,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002645539873113629,
      "loss": 5.1279,
      "mean_token_accuracy": 0.22712661772966386,
      "num_tokens": 101658002.0,
      "step": 51980
    },
    {
      "entropy": 5.800029945373535,
      "epoch": 5.196431249062828,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026451833682490113,
      "loss": 5.2129,
      "mean_token_accuracy": 0.22712988704442977,
      "num_tokens": 101668293.0,
      "step": 51985
    },
    {
      "entropy": 5.775330352783203,
      "epoch": 5.196931074124056,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002644826866021554,
      "loss": 5.0999,
      "mean_token_accuracy": 0.2305322453379631,
      "num_tokens": 101677789.0,
      "step": 51990
    },
    {
      "entropy": 5.7049096584320065,
      "epoch": 5.197430899185285,
      "grad_norm": 1.25,
      "learning_rate": 0.00026444703664402246,
      "loss": 5.1204,
      "mean_token_accuracy": 0.2358259528875351,
      "num_tokens": 101686807.0,
      "step": 51995
    },
    {
      "entropy": 5.763421678543091,
      "epoch": 5.197930724246514,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026441138695139946,
      "loss": 5.148,
      "mean_token_accuracy": 0.22794456034898758,
      "num_tokens": 101697352.0,
      "step": 52000
    },
    {
      "entropy": 5.787208175659179,
      "epoch": 5.1984305493077425,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002643757375251832,
      "loss": 5.1765,
      "mean_token_accuracy": 0.22179801911115646,
      "num_tokens": 101706099.0,
      "step": 52005
    },
    {
      "entropy": 5.791226100921631,
      "epoch": 5.198930374368971,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002643400883662706,
      "loss": 5.1465,
      "mean_token_accuracy": 0.22421692311763763,
      "num_tokens": 101715782.0,
      "step": 52010
    },
    {
      "entropy": 5.847907304763794,
      "epoch": 5.199430199430199,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002643044394755587,
      "loss": 5.2605,
      "mean_token_accuracy": 0.21807310581207276,
      "num_tokens": 101726107.0,
      "step": 52015
    },
    {
      "entropy": 5.726264142990113,
      "epoch": 5.199930024491428,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002642687908539443,
      "loss": 5.0915,
      "mean_token_accuracy": 0.23364797234535217,
      "num_tokens": 101736678.0,
      "step": 52020
    },
    {
      "entropy": 5.781699991226196,
      "epoch": 5.200429849552656,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002642331425023244,
      "loss": 5.1843,
      "mean_token_accuracy": 0.2214186817407608,
      "num_tokens": 101746347.0,
      "step": 52025
    },
    {
      "entropy": 5.804554653167725,
      "epoch": 5.200929674613885,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002641974944215958,
      "loss": 5.2236,
      "mean_token_accuracy": 0.21875008642673494,
      "num_tokens": 101755906.0,
      "step": 52030
    },
    {
      "entropy": 5.8315574645996096,
      "epoch": 5.201429499675114,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026416184661265553,
      "loss": 5.239,
      "mean_token_accuracy": 0.2136923059821129,
      "num_tokens": 101766245.0,
      "step": 52035
    },
    {
      "entropy": 5.774095439910889,
      "epoch": 5.2019293247363425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002641261990764004,
      "loss": 5.1517,
      "mean_token_accuracy": 0.23150647729635238,
      "num_tokens": 101776572.0,
      "step": 52040
    },
    {
      "entropy": 5.851140022277832,
      "epoch": 5.202429149797571,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002640905518137273,
      "loss": 5.2201,
      "mean_token_accuracy": 0.22327466458082199,
      "num_tokens": 101785171.0,
      "step": 52045
    },
    {
      "entropy": 5.822575616836548,
      "epoch": 5.202928974858799,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002640549048255331,
      "loss": 5.1891,
      "mean_token_accuracy": 0.2242568850517273,
      "num_tokens": 101796472.0,
      "step": 52050
    },
    {
      "entropy": 5.790705919265747,
      "epoch": 5.203428799920028,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002640192581127147,
      "loss": 5.1541,
      "mean_token_accuracy": 0.23176359385252,
      "num_tokens": 101806125.0,
      "step": 52055
    },
    {
      "entropy": 5.779086446762085,
      "epoch": 5.203928624981256,
      "grad_norm": 1.25,
      "learning_rate": 0.0002639836116761689,
      "loss": 5.1618,
      "mean_token_accuracy": 0.22468895316123963,
      "num_tokens": 101816024.0,
      "step": 52060
    },
    {
      "entropy": 5.804253244400025,
      "epoch": 5.204428450042485,
      "grad_norm": 1.25,
      "learning_rate": 0.0002639479655167926,
      "loss": 5.2269,
      "mean_token_accuracy": 0.22524333894252777,
      "num_tokens": 101825922.0,
      "step": 52065
    },
    {
      "entropy": 5.692032670974731,
      "epoch": 5.204928275103714,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002639123196354827,
      "loss": 5.1245,
      "mean_token_accuracy": 0.22933796793222427,
      "num_tokens": 101836702.0,
      "step": 52070
    },
    {
      "entropy": 5.8219820022583,
      "epoch": 5.2054281001649425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002638766740331359,
      "loss": 5.1981,
      "mean_token_accuracy": 0.21654190570116044,
      "num_tokens": 101847015.0,
      "step": 52075
    },
    {
      "entropy": 5.789684534072876,
      "epoch": 5.205927925226171,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002638410287106492,
      "loss": 5.1588,
      "mean_token_accuracy": 0.2243436336517334,
      "num_tokens": 101856938.0,
      "step": 52080
    },
    {
      "entropy": 5.835767841339111,
      "epoch": 5.206427750287399,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002638053836689193,
      "loss": 5.2094,
      "mean_token_accuracy": 0.2190855085849762,
      "num_tokens": 101866794.0,
      "step": 52085
    },
    {
      "entropy": 5.739429569244384,
      "epoch": 5.206927575348628,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026376973890884304,
      "loss": 5.0851,
      "mean_token_accuracy": 0.23543894439935684,
      "num_tokens": 101876465.0,
      "step": 52090
    },
    {
      "entropy": 5.809225940704346,
      "epoch": 5.207427400409856,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002637340944313173,
      "loss": 5.1718,
      "mean_token_accuracy": 0.22527812868356706,
      "num_tokens": 101886004.0,
      "step": 52095
    },
    {
      "entropy": 5.8772218227386475,
      "epoch": 5.207927225471085,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002636984502372388,
      "loss": 5.3305,
      "mean_token_accuracy": 0.2162906840443611,
      "num_tokens": 101896473.0,
      "step": 52100
    },
    {
      "entropy": 5.786329555511474,
      "epoch": 5.208427050532314,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026366280632750445,
      "loss": 5.1572,
      "mean_token_accuracy": 0.22572141289710998,
      "num_tokens": 101905534.0,
      "step": 52105
    },
    {
      "entropy": 5.801580286026001,
      "epoch": 5.2089268755935425,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026362716270301095,
      "loss": 5.1622,
      "mean_token_accuracy": 0.22858504503965377,
      "num_tokens": 101916436.0,
      "step": 52110
    },
    {
      "entropy": 5.764150810241699,
      "epoch": 5.209426700654771,
      "grad_norm": 1.25,
      "learning_rate": 0.00026359151936465516,
      "loss": 5.1414,
      "mean_token_accuracy": 0.22722635567188262,
      "num_tokens": 101926145.0,
      "step": 52115
    },
    {
      "entropy": 5.780738449096679,
      "epoch": 5.209926525715999,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002635558763133337,
      "loss": 5.2032,
      "mean_token_accuracy": 0.2215114489197731,
      "num_tokens": 101935197.0,
      "step": 52120
    },
    {
      "entropy": 5.7288109302520756,
      "epoch": 5.210426350777228,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026352023354994354,
      "loss": 5.1489,
      "mean_token_accuracy": 0.22858712077140808,
      "num_tokens": 101944925.0,
      "step": 52125
    },
    {
      "entropy": 5.893429613113403,
      "epoch": 5.210926175838456,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002634845910753814,
      "loss": 5.2273,
      "mean_token_accuracy": 0.22057601511478425,
      "num_tokens": 101954075.0,
      "step": 52130
    },
    {
      "entropy": 5.8840375423431395,
      "epoch": 5.211426000899685,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026344894889054383,
      "loss": 5.2223,
      "mean_token_accuracy": 0.22879345417022706,
      "num_tokens": 101962764.0,
      "step": 52135
    },
    {
      "entropy": 5.83792929649353,
      "epoch": 5.211925825960914,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002634133069963278,
      "loss": 5.2295,
      "mean_token_accuracy": 0.21886559277772905,
      "num_tokens": 101973434.0,
      "step": 52140
    },
    {
      "entropy": 5.838623094558716,
      "epoch": 5.2124256510221425,
      "grad_norm": 1.25,
      "learning_rate": 0.00026337766539363003,
      "loss": 5.2607,
      "mean_token_accuracy": 0.2160674050450325,
      "num_tokens": 101983105.0,
      "step": 52145
    },
    {
      "entropy": 5.7960508346557615,
      "epoch": 5.212925476083371,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026334202408334715,
      "loss": 5.2008,
      "mean_token_accuracy": 0.22191466391086578,
      "num_tokens": 101993769.0,
      "step": 52150
    },
    {
      "entropy": 5.783133792877197,
      "epoch": 5.213425301144599,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026330638306637586,
      "loss": 5.129,
      "mean_token_accuracy": 0.22714078575372695,
      "num_tokens": 102002981.0,
      "step": 52155
    },
    {
      "entropy": 5.732349061965943,
      "epoch": 5.213925126205828,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002632707423436131,
      "loss": 5.0462,
      "mean_token_accuracy": 0.23507696390151978,
      "num_tokens": 102012264.0,
      "step": 52160
    },
    {
      "entropy": 5.7792401790618895,
      "epoch": 5.214424951267056,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002632351019159554,
      "loss": 5.1773,
      "mean_token_accuracy": 0.22856451869010924,
      "num_tokens": 102022583.0,
      "step": 52165
    },
    {
      "entropy": 5.799162149429321,
      "epoch": 5.214924776328285,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026319946178429953,
      "loss": 5.2366,
      "mean_token_accuracy": 0.21522675603628158,
      "num_tokens": 102031902.0,
      "step": 52170
    },
    {
      "entropy": 5.813452863693238,
      "epoch": 5.215424601389514,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026316382194954214,
      "loss": 5.1275,
      "mean_token_accuracy": 0.22255093306303025,
      "num_tokens": 102041187.0,
      "step": 52175
    },
    {
      "entropy": 5.735652351379395,
      "epoch": 5.2159244264507425,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026312818241257995,
      "loss": 5.1063,
      "mean_token_accuracy": 0.22672808915376663,
      "num_tokens": 102049860.0,
      "step": 52180
    },
    {
      "entropy": 5.819703054428101,
      "epoch": 5.216424251511971,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026309254317430966,
      "loss": 5.1858,
      "mean_token_accuracy": 0.22723279744386674,
      "num_tokens": 102059450.0,
      "step": 52185
    },
    {
      "entropy": 5.777609252929688,
      "epoch": 5.216924076573199,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026305690423562785,
      "loss": 5.1476,
      "mean_token_accuracy": 0.2321559265255928,
      "num_tokens": 102068857.0,
      "step": 52190
    },
    {
      "entropy": 5.811407279968262,
      "epoch": 5.217423901634428,
      "grad_norm": 1.25,
      "learning_rate": 0.00026302126559743126,
      "loss": 5.2504,
      "mean_token_accuracy": 0.21639348566532135,
      "num_tokens": 102078790.0,
      "step": 52195
    },
    {
      "entropy": 5.712381410598755,
      "epoch": 5.217923726695656,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026298562726061674,
      "loss": 5.0853,
      "mean_token_accuracy": 0.23175066858530044,
      "num_tokens": 102090090.0,
      "step": 52200
    },
    {
      "entropy": 5.796857595443726,
      "epoch": 5.218423551756885,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002629499892260806,
      "loss": 5.2265,
      "mean_token_accuracy": 0.22207064777612687,
      "num_tokens": 102100163.0,
      "step": 52205
    },
    {
      "entropy": 5.778532791137695,
      "epoch": 5.218923376818114,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002629143514947196,
      "loss": 5.1797,
      "mean_token_accuracy": 0.2268250420689583,
      "num_tokens": 102110183.0,
      "step": 52210
    },
    {
      "entropy": 5.759098100662231,
      "epoch": 5.2194232018793425,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026287871406743043,
      "loss": 5.1924,
      "mean_token_accuracy": 0.22153674066066742,
      "num_tokens": 102120809.0,
      "step": 52215
    },
    {
      "entropy": 5.777885770797729,
      "epoch": 5.219923026940571,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002628430769451097,
      "loss": 5.1258,
      "mean_token_accuracy": 0.2357493206858635,
      "num_tokens": 102130661.0,
      "step": 52220
    },
    {
      "entropy": 5.818612432479858,
      "epoch": 5.220422852001799,
      "grad_norm": 1.25,
      "learning_rate": 0.000262807440128654,
      "loss": 5.2265,
      "mean_token_accuracy": 0.22200414687395095,
      "num_tokens": 102139866.0,
      "step": 52225
    },
    {
      "entropy": 5.715329599380493,
      "epoch": 5.220922677063028,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026277180361896,
      "loss": 5.1254,
      "mean_token_accuracy": 0.23293357342481613,
      "num_tokens": 102150290.0,
      "step": 52230
    },
    {
      "entropy": 5.757210779190063,
      "epoch": 5.221422502124256,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002627361674169243,
      "loss": 5.1309,
      "mean_token_accuracy": 0.2249365359544754,
      "num_tokens": 102159729.0,
      "step": 52235
    },
    {
      "entropy": 5.726903676986694,
      "epoch": 5.221922327185485,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002627005315234434,
      "loss": 5.08,
      "mean_token_accuracy": 0.23511593788862228,
      "num_tokens": 102169274.0,
      "step": 52240
    },
    {
      "entropy": 5.848077201843262,
      "epoch": 5.222422152246714,
      "grad_norm": 1.40625,
      "learning_rate": 0.000262664895939414,
      "loss": 5.1494,
      "mean_token_accuracy": 0.22577578127384185,
      "num_tokens": 102177893.0,
      "step": 52245
    },
    {
      "entropy": 5.800720119476319,
      "epoch": 5.2229219773079425,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002626292606657326,
      "loss": 5.1189,
      "mean_token_accuracy": 0.2244580551981926,
      "num_tokens": 102187624.0,
      "step": 52250
    },
    {
      "entropy": 5.535694718360901,
      "epoch": 5.223421802369171,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026259362570329585,
      "loss": 5.0093,
      "mean_token_accuracy": 0.2563331171870232,
      "num_tokens": 102199979.0,
      "step": 52255
    },
    {
      "entropy": 5.787581157684326,
      "epoch": 5.223921627430399,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002625579910530003,
      "loss": 5.0877,
      "mean_token_accuracy": 0.2317567840218544,
      "num_tokens": 102209098.0,
      "step": 52260
    },
    {
      "entropy": 5.752367162704468,
      "epoch": 5.224421452491628,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026252235671574256,
      "loss": 5.2097,
      "mean_token_accuracy": 0.22751939594745635,
      "num_tokens": 102218549.0,
      "step": 52265
    },
    {
      "entropy": 5.752228593826294,
      "epoch": 5.224921277552856,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026248672269241907,
      "loss": 5.0634,
      "mean_token_accuracy": 0.23058840483427048,
      "num_tokens": 102228565.0,
      "step": 52270
    },
    {
      "entropy": 5.881562042236328,
      "epoch": 5.225421102614085,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026245108898392637,
      "loss": 5.2377,
      "mean_token_accuracy": 0.21976538449525834,
      "num_tokens": 102238303.0,
      "step": 52275
    },
    {
      "entropy": 5.800264358520508,
      "epoch": 5.225920927675314,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002624154555911611,
      "loss": 5.1486,
      "mean_token_accuracy": 0.22861243635416031,
      "num_tokens": 102247167.0,
      "step": 52280
    },
    {
      "entropy": 5.692451000213623,
      "epoch": 5.2264207527365425,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026237982251501967,
      "loss": 5.1053,
      "mean_token_accuracy": 0.22617084085941314,
      "num_tokens": 102256732.0,
      "step": 52285
    },
    {
      "entropy": 5.754989576339722,
      "epoch": 5.226920577797771,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026234418975639866,
      "loss": 5.1513,
      "mean_token_accuracy": 0.2328181803226471,
      "num_tokens": 102265739.0,
      "step": 52290
    },
    {
      "entropy": 5.796884059906006,
      "epoch": 5.227420402858999,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026230855731619466,
      "loss": 5.1334,
      "mean_token_accuracy": 0.22956014126539231,
      "num_tokens": 102275613.0,
      "step": 52295
    },
    {
      "entropy": 5.750951290130615,
      "epoch": 5.227920227920228,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026227292519530397,
      "loss": 5.1127,
      "mean_token_accuracy": 0.2286008521914482,
      "num_tokens": 102285015.0,
      "step": 52300
    },
    {
      "entropy": 5.673743200302124,
      "epoch": 5.228420052981456,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026223729339462334,
      "loss": 5.0716,
      "mean_token_accuracy": 0.24062468856573105,
      "num_tokens": 102294643.0,
      "step": 52305
    },
    {
      "entropy": 5.763764381408691,
      "epoch": 5.228919878042685,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026220166191504897,
      "loss": 5.2156,
      "mean_token_accuracy": 0.22372389882802962,
      "num_tokens": 102304132.0,
      "step": 52310
    },
    {
      "entropy": 5.813705253601074,
      "epoch": 5.229419703103914,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002621660307574776,
      "loss": 5.1125,
      "mean_token_accuracy": 0.22766916304826737,
      "num_tokens": 102313764.0,
      "step": 52315
    },
    {
      "entropy": 5.717610883712768,
      "epoch": 5.2299195281651425,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026213039992280555,
      "loss": 5.0967,
      "mean_token_accuracy": 0.23108610659837722,
      "num_tokens": 102323286.0,
      "step": 52320
    },
    {
      "entropy": 5.743986988067627,
      "epoch": 5.230419353226371,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002620947694119294,
      "loss": 5.1204,
      "mean_token_accuracy": 0.23358843475580215,
      "num_tokens": 102333678.0,
      "step": 52325
    },
    {
      "entropy": 5.798312139511109,
      "epoch": 5.230919178287599,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026205913922574544,
      "loss": 5.1978,
      "mean_token_accuracy": 0.22249855548143388,
      "num_tokens": 102344011.0,
      "step": 52330
    },
    {
      "entropy": 5.7798743724823,
      "epoch": 5.231419003348828,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002620235093651502,
      "loss": 5.1493,
      "mean_token_accuracy": 0.2252457171678543,
      "num_tokens": 102353137.0,
      "step": 52335
    },
    {
      "entropy": 5.642084741592408,
      "epoch": 5.231918828410056,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026198787983104026,
      "loss": 5.0132,
      "mean_token_accuracy": 0.2506032183766365,
      "num_tokens": 102364361.0,
      "step": 52340
    },
    {
      "entropy": 5.732252073287964,
      "epoch": 5.232418653471285,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002619522506243118,
      "loss": 5.1196,
      "mean_token_accuracy": 0.22237569987773895,
      "num_tokens": 102374001.0,
      "step": 52345
    },
    {
      "entropy": 5.872843837738037,
      "epoch": 5.232918478532514,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026191662174586145,
      "loss": 5.2971,
      "mean_token_accuracy": 0.2177630767226219,
      "num_tokens": 102383681.0,
      "step": 52350
    },
    {
      "entropy": 5.750691747665405,
      "epoch": 5.2334183035937425,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026188099319658545,
      "loss": 5.1254,
      "mean_token_accuracy": 0.22358918339014053,
      "num_tokens": 102392840.0,
      "step": 52355
    },
    {
      "entropy": 5.793955516815186,
      "epoch": 5.23391812865497,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002618453649773803,
      "loss": 5.1563,
      "mean_token_accuracy": 0.22599893063306808,
      "num_tokens": 102402894.0,
      "step": 52360
    },
    {
      "entropy": 5.724733352661133,
      "epoch": 5.234417953716199,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002618097370891423,
      "loss": 5.0032,
      "mean_token_accuracy": 0.2430989161133766,
      "num_tokens": 102412799.0,
      "step": 52365
    },
    {
      "entropy": 5.850288581848145,
      "epoch": 5.234917778777428,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002617741095327681,
      "loss": 5.2001,
      "mean_token_accuracy": 0.2255416840314865,
      "num_tokens": 102423800.0,
      "step": 52370
    },
    {
      "entropy": 5.706450700759888,
      "epoch": 5.235417603838656,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002617384823091537,
      "loss": 5.0681,
      "mean_token_accuracy": 0.2382604017853737,
      "num_tokens": 102433432.0,
      "step": 52375
    },
    {
      "entropy": 5.812264251708984,
      "epoch": 5.235917428899885,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002617028554191958,
      "loss": 5.2329,
      "mean_token_accuracy": 0.21835968494415284,
      "num_tokens": 102442252.0,
      "step": 52380
    },
    {
      "entropy": 5.856234312057495,
      "epoch": 5.236417253961114,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026166722886379064,
      "loss": 5.2132,
      "mean_token_accuracy": 0.21731618344783782,
      "num_tokens": 102451721.0,
      "step": 52385
    },
    {
      "entropy": 5.711640071868897,
      "epoch": 5.2369170790223425,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002616316026438346,
      "loss": 5.0738,
      "mean_token_accuracy": 0.2371349811553955,
      "num_tokens": 102463091.0,
      "step": 52390
    },
    {
      "entropy": 5.743600273132325,
      "epoch": 5.23741690408357,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002615959767602239,
      "loss": 5.136,
      "mean_token_accuracy": 0.2278431698679924,
      "num_tokens": 102473448.0,
      "step": 52395
    },
    {
      "entropy": 5.776070022583008,
      "epoch": 5.237916729144799,
      "grad_norm": 1.078125,
      "learning_rate": 0.000261560351213855,
      "loss": 5.1544,
      "mean_token_accuracy": 0.22190861850976945,
      "num_tokens": 102483312.0,
      "step": 52400
    },
    {
      "entropy": 5.803842639923095,
      "epoch": 5.238416554206028,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026152472600562423,
      "loss": 5.1713,
      "mean_token_accuracy": 0.21991289258003235,
      "num_tokens": 102492898.0,
      "step": 52405
    },
    {
      "entropy": 5.814144468307495,
      "epoch": 5.238916379267256,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026148910113642796,
      "loss": 5.1491,
      "mean_token_accuracy": 0.23001058548688888,
      "num_tokens": 102502963.0,
      "step": 52410
    },
    {
      "entropy": 5.7135645866394045,
      "epoch": 5.239416204328485,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002614534766071623,
      "loss": 5.0561,
      "mean_token_accuracy": 0.2372412085533142,
      "num_tokens": 102513236.0,
      "step": 52415
    },
    {
      "entropy": 5.80870246887207,
      "epoch": 5.239916029389714,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002614178524187237,
      "loss": 5.1345,
      "mean_token_accuracy": 0.2302390530705452,
      "num_tokens": 102523755.0,
      "step": 52420
    },
    {
      "entropy": 5.7478560447692875,
      "epoch": 5.2404158544509425,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002613822285720085,
      "loss": 5.159,
      "mean_token_accuracy": 0.23001234382390975,
      "num_tokens": 102533827.0,
      "step": 52425
    },
    {
      "entropy": 5.7386456489562985,
      "epoch": 5.24091567951217,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002613466050679129,
      "loss": 5.1089,
      "mean_token_accuracy": 0.22769679576158525,
      "num_tokens": 102543969.0,
      "step": 52430
    },
    {
      "entropy": 5.7229164123535154,
      "epoch": 5.241415504573399,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026131098190733317,
      "loss": 5.1161,
      "mean_token_accuracy": 0.23230630159378052,
      "num_tokens": 102553690.0,
      "step": 52435
    },
    {
      "entropy": 5.793640041351319,
      "epoch": 5.241915329634628,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002612753590911656,
      "loss": 5.1603,
      "mean_token_accuracy": 0.22419596165418626,
      "num_tokens": 102563011.0,
      "step": 52440
    },
    {
      "entropy": 5.751243782043457,
      "epoch": 5.242415154695856,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026123973662030655,
      "loss": 5.0908,
      "mean_token_accuracy": 0.2305256262421608,
      "num_tokens": 102573906.0,
      "step": 52445
    },
    {
      "entropy": 5.774703550338745,
      "epoch": 5.242914979757085,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026120411449565207,
      "loss": 5.16,
      "mean_token_accuracy": 0.224991674721241,
      "num_tokens": 102583997.0,
      "step": 52450
    },
    {
      "entropy": 5.727332401275635,
      "epoch": 5.243414804818314,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002611684927180986,
      "loss": 5.0906,
      "mean_token_accuracy": 0.23552947342395783,
      "num_tokens": 102594521.0,
      "step": 52455
    },
    {
      "entropy": 5.877046394348144,
      "epoch": 5.2439146298795425,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002611328712885423,
      "loss": 5.2256,
      "mean_token_accuracy": 0.2219441145658493,
      "num_tokens": 102604040.0,
      "step": 52460
    },
    {
      "entropy": 5.760520648956299,
      "epoch": 5.24441445494077,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002610972502078794,
      "loss": 5.1151,
      "mean_token_accuracy": 0.2347361460328102,
      "num_tokens": 102614711.0,
      "step": 52465
    },
    {
      "entropy": 5.702761316299439,
      "epoch": 5.244914280001999,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026106162947700604,
      "loss": 5.0786,
      "mean_token_accuracy": 0.23825484812259673,
      "num_tokens": 102624956.0,
      "step": 52470
    },
    {
      "entropy": 5.717621326446533,
      "epoch": 5.245414105063228,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002610260090968185,
      "loss": 5.0668,
      "mean_token_accuracy": 0.2358429253101349,
      "num_tokens": 102635634.0,
      "step": 52475
    },
    {
      "entropy": 5.787335395812988,
      "epoch": 5.245913930124456,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002609903890682131,
      "loss": 5.1219,
      "mean_token_accuracy": 0.22104052901268006,
      "num_tokens": 102645490.0,
      "step": 52480
    },
    {
      "entropy": 5.774110317230225,
      "epoch": 5.246413755185685,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00026095476939208583,
      "loss": 5.1862,
      "mean_token_accuracy": 0.22594767957925796,
      "num_tokens": 102655384.0,
      "step": 52485
    },
    {
      "entropy": 5.844509887695312,
      "epoch": 5.246913580246914,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002609191500693329,
      "loss": 5.1814,
      "mean_token_accuracy": 0.22513842433691025,
      "num_tokens": 102665900.0,
      "step": 52490
    },
    {
      "entropy": 5.775953245162964,
      "epoch": 5.2474134053081425,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002608835311008506,
      "loss": 5.1342,
      "mean_token_accuracy": 0.2245892658829689,
      "num_tokens": 102675867.0,
      "step": 52495
    },
    {
      "entropy": 5.7796731948852536,
      "epoch": 5.24791323036937,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000260847912487535,
      "loss": 5.1657,
      "mean_token_accuracy": 0.22061028629541396,
      "num_tokens": 102685426.0,
      "step": 52500
    },
    {
      "entropy": 5.776322507858277,
      "epoch": 5.248413055430599,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002608122942302822,
      "loss": 5.1468,
      "mean_token_accuracy": 0.23355961740016937,
      "num_tokens": 102695545.0,
      "step": 52505
    },
    {
      "entropy": 5.795589351654053,
      "epoch": 5.248912880491828,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026077667632998854,
      "loss": 5.2073,
      "mean_token_accuracy": 0.21950169354677201,
      "num_tokens": 102706161.0,
      "step": 52510
    },
    {
      "entropy": 5.699768352508545,
      "epoch": 5.249412705553056,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026074105878754996,
      "loss": 5.0536,
      "mean_token_accuracy": 0.23217568099498748,
      "num_tokens": 102715954.0,
      "step": 52515
    },
    {
      "entropy": 5.872931814193725,
      "epoch": 5.249912530614285,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002607054416038627,
      "loss": 5.2508,
      "mean_token_accuracy": 0.2221939593553543,
      "num_tokens": 102726530.0,
      "step": 52520
    },
    {
      "entropy": 5.787307071685791,
      "epoch": 5.250412355675514,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002606698247798229,
      "loss": 5.1146,
      "mean_token_accuracy": 0.22396588176488877,
      "num_tokens": 102735872.0,
      "step": 52525
    },
    {
      "entropy": 5.734575605392456,
      "epoch": 5.2509121807367425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002606342083163266,
      "loss": 5.1375,
      "mean_token_accuracy": 0.23260378390550612,
      "num_tokens": 102745987.0,
      "step": 52530
    },
    {
      "entropy": 5.746028804779053,
      "epoch": 5.25141200579797,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026059859221426977,
      "loss": 5.0739,
      "mean_token_accuracy": 0.235298228263855,
      "num_tokens": 102755762.0,
      "step": 52535
    },
    {
      "entropy": 5.682362508773804,
      "epoch": 5.251911830859199,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026056297647454883,
      "loss": 5.03,
      "mean_token_accuracy": 0.23755033761262895,
      "num_tokens": 102764887.0,
      "step": 52540
    },
    {
      "entropy": 5.633777761459351,
      "epoch": 5.252411655920428,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002605273610980597,
      "loss": 4.9571,
      "mean_token_accuracy": 0.25128167420625686,
      "num_tokens": 102774871.0,
      "step": 52545
    },
    {
      "entropy": 5.847788619995117,
      "epoch": 5.252911480981656,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002604917460856985,
      "loss": 5.2172,
      "mean_token_accuracy": 0.22375622391700745,
      "num_tokens": 102785351.0,
      "step": 52550
    },
    {
      "entropy": 5.778202199935913,
      "epoch": 5.253411306042885,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026045613143836106,
      "loss": 5.1588,
      "mean_token_accuracy": 0.22365263402462005,
      "num_tokens": 102795490.0,
      "step": 52555
    },
    {
      "entropy": 5.728235864639283,
      "epoch": 5.253911131104114,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026042051715694366,
      "loss": 5.1054,
      "mean_token_accuracy": 0.23502610623836517,
      "num_tokens": 102805022.0,
      "step": 52560
    },
    {
      "entropy": 5.798934555053711,
      "epoch": 5.2544109561653425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026038490324234225,
      "loss": 5.2425,
      "mean_token_accuracy": 0.21985455155372619,
      "num_tokens": 102815389.0,
      "step": 52565
    },
    {
      "entropy": 5.700432538986206,
      "epoch": 5.25491078122657,
      "grad_norm": 1.265625,
      "learning_rate": 0.000260349289695453,
      "loss": 4.974,
      "mean_token_accuracy": 0.24148349910974504,
      "num_tokens": 102826271.0,
      "step": 52570
    },
    {
      "entropy": 5.833459043502808,
      "epoch": 5.255410606287799,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026031367651717184,
      "loss": 5.1856,
      "mean_token_accuracy": 0.22638312131166458,
      "num_tokens": 102835406.0,
      "step": 52575
    },
    {
      "entropy": 5.805955457687378,
      "epoch": 5.255910431349028,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002602780637083948,
      "loss": 5.2509,
      "mean_token_accuracy": 0.21732568144798278,
      "num_tokens": 102844530.0,
      "step": 52580
    },
    {
      "entropy": 5.816052341461182,
      "epoch": 5.256410256410256,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002602424512700178,
      "loss": 5.2511,
      "mean_token_accuracy": 0.2172977387905121,
      "num_tokens": 102853518.0,
      "step": 52585
    },
    {
      "entropy": 5.651276016235352,
      "epoch": 5.256910081471485,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002602068392029371,
      "loss": 5.0415,
      "mean_token_accuracy": 0.23816165775060655,
      "num_tokens": 102864473.0,
      "step": 52590
    },
    {
      "entropy": 5.802323532104492,
      "epoch": 5.257409906532714,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026017122750804833,
      "loss": 5.0983,
      "mean_token_accuracy": 0.2281664490699768,
      "num_tokens": 102874376.0,
      "step": 52595
    },
    {
      "entropy": 5.843246603012085,
      "epoch": 5.2579097315939425,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026013561618624774,
      "loss": 5.2196,
      "mean_token_accuracy": 0.2221165955066681,
      "num_tokens": 102884283.0,
      "step": 52600
    },
    {
      "entropy": 5.8119700908660885,
      "epoch": 5.25840955665517,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026010000523843125,
      "loss": 5.2208,
      "mean_token_accuracy": 0.20785485953092575,
      "num_tokens": 102893625.0,
      "step": 52605
    },
    {
      "entropy": 5.823038578033447,
      "epoch": 5.258909381716399,
      "grad_norm": 1.125,
      "learning_rate": 0.00026006439466549474,
      "loss": 5.1891,
      "mean_token_accuracy": 0.21844321489334106,
      "num_tokens": 102905016.0,
      "step": 52610
    },
    {
      "entropy": 5.700334548950195,
      "epoch": 5.259409206777628,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002600287844683343,
      "loss": 5.049,
      "mean_token_accuracy": 0.23389241546392442,
      "num_tokens": 102914914.0,
      "step": 52615
    },
    {
      "entropy": 5.790295934677124,
      "epoch": 5.259909031838856,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002599931746478458,
      "loss": 5.1963,
      "mean_token_accuracy": 0.22316782772541047,
      "num_tokens": 102924404.0,
      "step": 52620
    },
    {
      "entropy": 5.855201864242554,
      "epoch": 5.260408856900085,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002599575652049251,
      "loss": 5.2605,
      "mean_token_accuracy": 0.22226197570562362,
      "num_tokens": 102934440.0,
      "step": 52625
    },
    {
      "entropy": 5.841542148590088,
      "epoch": 5.260908681961314,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002599219561404682,
      "loss": 5.1613,
      "mean_token_accuracy": 0.22744958847761154,
      "num_tokens": 102944362.0,
      "step": 52630
    },
    {
      "entropy": 5.759908676147461,
      "epoch": 5.2614085070225425,
      "grad_norm": 1.296875,
      "learning_rate": 0.000259886347455371,
      "loss": 5.1604,
      "mean_token_accuracy": 0.22048152834177018,
      "num_tokens": 102953740.0,
      "step": 52635
    },
    {
      "entropy": 5.763352966308593,
      "epoch": 5.26190833208377,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002598507391505294,
      "loss": 5.1804,
      "mean_token_accuracy": 0.2273368313908577,
      "num_tokens": 102962944.0,
      "step": 52640
    },
    {
      "entropy": 5.810108995437622,
      "epoch": 5.262408157144999,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002598151312268393,
      "loss": 5.1459,
      "mean_token_accuracy": 0.22969448268413545,
      "num_tokens": 102972624.0,
      "step": 52645
    },
    {
      "entropy": 5.739401483535767,
      "epoch": 5.262907982206228,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025977952368519663,
      "loss": 5.1633,
      "mean_token_accuracy": 0.22446130067110062,
      "num_tokens": 102983527.0,
      "step": 52650
    },
    {
      "entropy": 5.801274347305298,
      "epoch": 5.263407807267456,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002597439165264972,
      "loss": 5.1516,
      "mean_token_accuracy": 0.21986725479364394,
      "num_tokens": 102994541.0,
      "step": 52655
    },
    {
      "entropy": 5.8007019519805905,
      "epoch": 5.263907632328685,
      "grad_norm": 1.25,
      "learning_rate": 0.0002597083097516368,
      "loss": 5.1653,
      "mean_token_accuracy": 0.21989327818155288,
      "num_tokens": 103005690.0,
      "step": 52660
    },
    {
      "entropy": 5.895710372924805,
      "epoch": 5.264407457389914,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002596727033615116,
      "loss": 5.2569,
      "mean_token_accuracy": 0.21418935507535936,
      "num_tokens": 103014028.0,
      "step": 52665
    },
    {
      "entropy": 5.793761157989502,
      "epoch": 5.2649072824511425,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025963709735701715,
      "loss": 5.1145,
      "mean_token_accuracy": 0.2337503507733345,
      "num_tokens": 103024694.0,
      "step": 52670
    },
    {
      "entropy": 5.769284200668335,
      "epoch": 5.26540710751237,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025960149173904937,
      "loss": 5.1782,
      "mean_token_accuracy": 0.22717241197824478,
      "num_tokens": 103035264.0,
      "step": 52675
    },
    {
      "entropy": 5.7588210105896,
      "epoch": 5.265906932573599,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025956588650850404,
      "loss": 5.1739,
      "mean_token_accuracy": 0.23203705251216888,
      "num_tokens": 103044961.0,
      "step": 52680
    },
    {
      "entropy": 5.811165142059326,
      "epoch": 5.266406757634828,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002595302816662771,
      "loss": 5.2362,
      "mean_token_accuracy": 0.21361526995897293,
      "num_tokens": 103056019.0,
      "step": 52685
    },
    {
      "entropy": 5.895565700531006,
      "epoch": 5.266906582696056,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002594946772132643,
      "loss": 5.2412,
      "mean_token_accuracy": 0.2137683480978012,
      "num_tokens": 103066917.0,
      "step": 52690
    },
    {
      "entropy": 5.899509906768799,
      "epoch": 5.267406407757285,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025945907315036144,
      "loss": 5.3508,
      "mean_token_accuracy": 0.20666868686676027,
      "num_tokens": 103076900.0,
      "step": 52695
    },
    {
      "entropy": 5.83478856086731,
      "epoch": 5.267906232818514,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025942346947846424,
      "loss": 5.1919,
      "mean_token_accuracy": 0.2176658794283867,
      "num_tokens": 103086901.0,
      "step": 52700
    },
    {
      "entropy": 5.746528816223145,
      "epoch": 5.2684060578797425,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002593878661984686,
      "loss": 5.1311,
      "mean_token_accuracy": 0.22685520946979523,
      "num_tokens": 103097134.0,
      "step": 52705
    },
    {
      "entropy": 5.769997596740723,
      "epoch": 5.26890588294097,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002593522633112702,
      "loss": 5.1398,
      "mean_token_accuracy": 0.22369784861803055,
      "num_tokens": 103106300.0,
      "step": 52710
    },
    {
      "entropy": 5.788970422744751,
      "epoch": 5.269405708002199,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002593166608177648,
      "loss": 5.1538,
      "mean_token_accuracy": 0.23122130930423737,
      "num_tokens": 103115749.0,
      "step": 52715
    },
    {
      "entropy": 5.872012805938721,
      "epoch": 5.269905533063428,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002592810587188482,
      "loss": 5.2463,
      "mean_token_accuracy": 0.2162238121032715,
      "num_tokens": 103126837.0,
      "step": 52720
    },
    {
      "entropy": 5.9280445098876955,
      "epoch": 5.270405358124656,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002592454570154161,
      "loss": 5.235,
      "mean_token_accuracy": 0.21386554837226868,
      "num_tokens": 103135800.0,
      "step": 52725
    },
    {
      "entropy": 5.805944395065308,
      "epoch": 5.270905183185885,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025920985570836424,
      "loss": 5.1619,
      "mean_token_accuracy": 0.2298886403441429,
      "num_tokens": 103144727.0,
      "step": 52730
    },
    {
      "entropy": 5.77387375831604,
      "epoch": 5.271405008247114,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002591742547985884,
      "loss": 5.2103,
      "mean_token_accuracy": 0.21284937858581543,
      "num_tokens": 103154136.0,
      "step": 52735
    },
    {
      "entropy": 5.716011095046997,
      "epoch": 5.2719048333083425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025913865428698425,
      "loss": 5.1693,
      "mean_token_accuracy": 0.22215899378061293,
      "num_tokens": 103164705.0,
      "step": 52740
    },
    {
      "entropy": 5.815771627426147,
      "epoch": 5.27240465836957,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002591030541744475,
      "loss": 5.1555,
      "mean_token_accuracy": 0.22255031317472457,
      "num_tokens": 103173493.0,
      "step": 52745
    },
    {
      "entropy": 5.794795179367066,
      "epoch": 5.272904483430799,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002590674544618737,
      "loss": 5.1582,
      "mean_token_accuracy": 0.22972292602062225,
      "num_tokens": 103183054.0,
      "step": 52750
    },
    {
      "entropy": 5.791912698745728,
      "epoch": 5.273404308492028,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025903185515015875,
      "loss": 5.14,
      "mean_token_accuracy": 0.23073821514844894,
      "num_tokens": 103193885.0,
      "step": 52755
    },
    {
      "entropy": 5.631897068023681,
      "epoch": 5.273904133553256,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025899625624019817,
      "loss": 5.0457,
      "mean_token_accuracy": 0.23798799216747285,
      "num_tokens": 103202914.0,
      "step": 52760
    },
    {
      "entropy": 5.741938161849975,
      "epoch": 5.274403958614485,
      "grad_norm": 1.25,
      "learning_rate": 0.0002589606577328876,
      "loss": 5.0915,
      "mean_token_accuracy": 0.23107265681028366,
      "num_tokens": 103212331.0,
      "step": 52765
    },
    {
      "entropy": 5.764460515975952,
      "epoch": 5.274903783675714,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002589250596291229,
      "loss": 5.1362,
      "mean_token_accuracy": 0.23055716454982758,
      "num_tokens": 103222167.0,
      "step": 52770
    },
    {
      "entropy": 5.814035224914551,
      "epoch": 5.2754036087369425,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025888946192979947,
      "loss": 5.1601,
      "mean_token_accuracy": 0.22445416301488877,
      "num_tokens": 103232704.0,
      "step": 52775
    },
    {
      "entropy": 5.757168436050415,
      "epoch": 5.27590343379817,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000258853864635813,
      "loss": 5.1287,
      "mean_token_accuracy": 0.22675791382789612,
      "num_tokens": 103243008.0,
      "step": 52780
    },
    {
      "entropy": 5.854746770858765,
      "epoch": 5.276403258859399,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00025881826774805914,
      "loss": 5.2941,
      "mean_token_accuracy": 0.20951172709465027,
      "num_tokens": 103253193.0,
      "step": 52785
    },
    {
      "entropy": 5.683731317520142,
      "epoch": 5.276903083920628,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025878267126743356,
      "loss": 5.0519,
      "mean_token_accuracy": 0.24116577804088593,
      "num_tokens": 103263100.0,
      "step": 52790
    },
    {
      "entropy": 5.76290545463562,
      "epoch": 5.277402908981856,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025874707519483184,
      "loss": 5.1295,
      "mean_token_accuracy": 0.22681333273649215,
      "num_tokens": 103273389.0,
      "step": 52795
    },
    {
      "entropy": 5.81973876953125,
      "epoch": 5.277902734043085,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025871147953114945,
      "loss": 5.2345,
      "mean_token_accuracy": 0.21748311072587967,
      "num_tokens": 103283222.0,
      "step": 52800
    },
    {
      "entropy": 5.82376127243042,
      "epoch": 5.278402559104314,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002586758842772821,
      "loss": 5.1815,
      "mean_token_accuracy": 0.22122040390968323,
      "num_tokens": 103292312.0,
      "step": 52805
    },
    {
      "entropy": 5.880968284606934,
      "epoch": 5.2789023841655425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002586402894341252,
      "loss": 5.2749,
      "mean_token_accuracy": 0.21692135632038118,
      "num_tokens": 103302064.0,
      "step": 52810
    },
    {
      "entropy": 5.817407894134521,
      "epoch": 5.27940220922677,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002586046950025745,
      "loss": 5.1465,
      "mean_token_accuracy": 0.22473097890615462,
      "num_tokens": 103310608.0,
      "step": 52815
    },
    {
      "entropy": 5.750941896438599,
      "epoch": 5.279902034287999,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025856910098352536,
      "loss": 5.1072,
      "mean_token_accuracy": 0.2327498108148575,
      "num_tokens": 103321585.0,
      "step": 52820
    },
    {
      "entropy": 5.748891067504883,
      "epoch": 5.280401859349228,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002585335073778736,
      "loss": 5.1782,
      "mean_token_accuracy": 0.22186422646045684,
      "num_tokens": 103331669.0,
      "step": 52825
    },
    {
      "entropy": 5.789378309249878,
      "epoch": 5.280901684410456,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002584979141865144,
      "loss": 5.208,
      "mean_token_accuracy": 0.214313043653965,
      "num_tokens": 103341658.0,
      "step": 52830
    },
    {
      "entropy": 5.801794767379761,
      "epoch": 5.281401509471685,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002584623214103435,
      "loss": 5.1199,
      "mean_token_accuracy": 0.22648383677005768,
      "num_tokens": 103350686.0,
      "step": 52835
    },
    {
      "entropy": 5.801869440078735,
      "epoch": 5.281901334532914,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002584267290502562,
      "loss": 5.1927,
      "mean_token_accuracy": 0.22386714816093445,
      "num_tokens": 103361307.0,
      "step": 52840
    },
    {
      "entropy": 5.7362189292907715,
      "epoch": 5.2824011595941425,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002583911371071483,
      "loss": 5.1274,
      "mean_token_accuracy": 0.23214596956968309,
      "num_tokens": 103370440.0,
      "step": 52845
    },
    {
      "entropy": 5.783230829238891,
      "epoch": 5.28290098465537,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025835554558191503,
      "loss": 5.2044,
      "mean_token_accuracy": 0.22821252346038817,
      "num_tokens": 103380021.0,
      "step": 52850
    },
    {
      "entropy": 5.759934377670288,
      "epoch": 5.283400809716599,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000258319954475452,
      "loss": 5.131,
      "mean_token_accuracy": 0.22458379715681076,
      "num_tokens": 103389686.0,
      "step": 52855
    },
    {
      "entropy": 5.7548433303833,
      "epoch": 5.283900634777828,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002582843637886546,
      "loss": 5.0973,
      "mean_token_accuracy": 0.22122343331575395,
      "num_tokens": 103399194.0,
      "step": 52860
    },
    {
      "entropy": 5.805538034439087,
      "epoch": 5.284400459839056,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002582487735224183,
      "loss": 5.1438,
      "mean_token_accuracy": 0.2222096472978592,
      "num_tokens": 103408290.0,
      "step": 52865
    },
    {
      "entropy": 5.683002948760986,
      "epoch": 5.284900284900285,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025821318367763856,
      "loss": 5.1189,
      "mean_token_accuracy": 0.22748589217662812,
      "num_tokens": 103418355.0,
      "step": 52870
    },
    {
      "entropy": 5.771655654907226,
      "epoch": 5.285400109961514,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025817759425521075,
      "loss": 5.0902,
      "mean_token_accuracy": 0.2317323476076126,
      "num_tokens": 103428416.0,
      "step": 52875
    },
    {
      "entropy": 5.7768007755279545,
      "epoch": 5.2858999350227425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002581420052560303,
      "loss": 5.153,
      "mean_token_accuracy": 0.21840884238481523,
      "num_tokens": 103438355.0,
      "step": 52880
    },
    {
      "entropy": 5.690449237823486,
      "epoch": 5.28639976008397,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002581064166809928,
      "loss": 5.1172,
      "mean_token_accuracy": 0.2386813670396805,
      "num_tokens": 103448612.0,
      "step": 52885
    },
    {
      "entropy": 5.779743576049805,
      "epoch": 5.286899585145199,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025807082853099343,
      "loss": 5.1173,
      "mean_token_accuracy": 0.2275390937924385,
      "num_tokens": 103457227.0,
      "step": 52890
    },
    {
      "entropy": 5.910595512390136,
      "epoch": 5.287399410206428,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025803524080692783,
      "loss": 5.2905,
      "mean_token_accuracy": 0.2162891924381256,
      "num_tokens": 103466594.0,
      "step": 52895
    },
    {
      "entropy": 5.8694699764251705,
      "epoch": 5.287899235267656,
      "grad_norm": 1.265625,
      "learning_rate": 0.000257999653509691,
      "loss": 5.2351,
      "mean_token_accuracy": 0.2229229673743248,
      "num_tokens": 103476704.0,
      "step": 52900
    },
    {
      "entropy": 5.834773159027099,
      "epoch": 5.288399060328885,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002579640666401785,
      "loss": 5.161,
      "mean_token_accuracy": 0.22727852314710617,
      "num_tokens": 103486993.0,
      "step": 52905
    },
    {
      "entropy": 5.7802770137786865,
      "epoch": 5.288898885390114,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025792848019928576,
      "loss": 5.17,
      "mean_token_accuracy": 0.22549373507499695,
      "num_tokens": 103497824.0,
      "step": 52910
    },
    {
      "entropy": 5.802233123779297,
      "epoch": 5.2893987104513425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025789289418790814,
      "loss": 5.1562,
      "mean_token_accuracy": 0.23179354667663574,
      "num_tokens": 103506906.0,
      "step": 52915
    },
    {
      "entropy": 5.6989648818969725,
      "epoch": 5.28989853551257,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002578573086069408,
      "loss": 5.1224,
      "mean_token_accuracy": 0.23284824639558793,
      "num_tokens": 103516723.0,
      "step": 52920
    },
    {
      "entropy": 5.861837530136109,
      "epoch": 5.290398360573799,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002578217234572792,
      "loss": 5.212,
      "mean_token_accuracy": 0.22028430700302123,
      "num_tokens": 103527173.0,
      "step": 52925
    },
    {
      "entropy": 5.821815013885498,
      "epoch": 5.290898185635028,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002577861387398186,
      "loss": 5.151,
      "mean_token_accuracy": 0.22868277579545976,
      "num_tokens": 103537209.0,
      "step": 52930
    },
    {
      "entropy": 5.837669944763183,
      "epoch": 5.291398010696256,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002577505544554543,
      "loss": 5.1939,
      "mean_token_accuracy": 0.21717049330472946,
      "num_tokens": 103547294.0,
      "step": 52935
    },
    {
      "entropy": 5.799870920181275,
      "epoch": 5.291897835757485,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002577149706050817,
      "loss": 5.1248,
      "mean_token_accuracy": 0.231715789437294,
      "num_tokens": 103556272.0,
      "step": 52940
    },
    {
      "entropy": 5.778443098068237,
      "epoch": 5.292397660818714,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002576793871895959,
      "loss": 5.1537,
      "mean_token_accuracy": 0.22637194842100145,
      "num_tokens": 103566057.0,
      "step": 52945
    },
    {
      "entropy": 5.790244436264038,
      "epoch": 5.2928974858799425,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025764380420989226,
      "loss": 5.2176,
      "mean_token_accuracy": 0.21628812700510025,
      "num_tokens": 103576382.0,
      "step": 52950
    },
    {
      "entropy": 5.861952018737793,
      "epoch": 5.29339731094117,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002576082216668661,
      "loss": 5.3253,
      "mean_token_accuracy": 0.20849028676748277,
      "num_tokens": 103587374.0,
      "step": 52955
    },
    {
      "entropy": 5.80858473777771,
      "epoch": 5.293897136002399,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002575726395614126,
      "loss": 5.1737,
      "mean_token_accuracy": 0.22710408866405488,
      "num_tokens": 103596943.0,
      "step": 52960
    },
    {
      "entropy": 5.78536901473999,
      "epoch": 5.294396961063628,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000257537057894427,
      "loss": 5.195,
      "mean_token_accuracy": 0.2238062098622322,
      "num_tokens": 103606731.0,
      "step": 52965
    },
    {
      "entropy": 5.751351928710937,
      "epoch": 5.294896786124856,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025750147666680454,
      "loss": 5.1742,
      "mean_token_accuracy": 0.22203251719474792,
      "num_tokens": 103616260.0,
      "step": 52970
    },
    {
      "entropy": 5.827448749542237,
      "epoch": 5.295396611186085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002574658958794404,
      "loss": 5.2377,
      "mean_token_accuracy": 0.21834697872400283,
      "num_tokens": 103626053.0,
      "step": 52975
    },
    {
      "entropy": 5.816322326660156,
      "epoch": 5.295896436247314,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002574303155332298,
      "loss": 5.1975,
      "mean_token_accuracy": 0.22755162864923478,
      "num_tokens": 103635511.0,
      "step": 52980
    },
    {
      "entropy": 5.739516353607177,
      "epoch": 5.2963962613085425,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025739473562906795,
      "loss": 5.1176,
      "mean_token_accuracy": 0.23441435396671295,
      "num_tokens": 103645574.0,
      "step": 52985
    },
    {
      "entropy": 5.809647560119629,
      "epoch": 5.29689608636977,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025735915616785005,
      "loss": 5.1813,
      "mean_token_accuracy": 0.23021538704633712,
      "num_tokens": 103655455.0,
      "step": 52990
    },
    {
      "entropy": 5.656716632843017,
      "epoch": 5.297395911430999,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025732357715047124,
      "loss": 4.921,
      "mean_token_accuracy": 0.2335398390889168,
      "num_tokens": 103664628.0,
      "step": 52995
    },
    {
      "entropy": 5.668472003936768,
      "epoch": 5.297895736492228,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025728799857782663,
      "loss": 5.0153,
      "mean_token_accuracy": 0.23379296213388442,
      "num_tokens": 103673794.0,
      "step": 53000
    },
    {
      "entropy": 5.776722097396851,
      "epoch": 5.298395561553456,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002572524204508114,
      "loss": 5.24,
      "mean_token_accuracy": 0.2255443587899208,
      "num_tokens": 103682512.0,
      "step": 53005
    },
    {
      "entropy": 5.847726583480835,
      "epoch": 5.298895386614685,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025721684277032076,
      "loss": 5.2172,
      "mean_token_accuracy": 0.2215162545442581,
      "num_tokens": 103692485.0,
      "step": 53010
    },
    {
      "entropy": 5.826380109786987,
      "epoch": 5.299395211675914,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025718126553724973,
      "loss": 5.148,
      "mean_token_accuracy": 0.22786375284194946,
      "num_tokens": 103703363.0,
      "step": 53015
    },
    {
      "entropy": 5.7522838592529295,
      "epoch": 5.2998950367371425,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025714568875249353,
      "loss": 5.0549,
      "mean_token_accuracy": 0.23248248398303986,
      "num_tokens": 103713057.0,
      "step": 53020
    },
    {
      "entropy": 5.842968225479126,
      "epoch": 5.30039486179837,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002571101124169472,
      "loss": 5.2552,
      "mean_token_accuracy": 0.22064816504716872,
      "num_tokens": 103722785.0,
      "step": 53025
    },
    {
      "entropy": 5.829400491714478,
      "epoch": 5.300894686859599,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002570745365315058,
      "loss": 5.1967,
      "mean_token_accuracy": 0.21795299053192138,
      "num_tokens": 103732932.0,
      "step": 53030
    },
    {
      "entropy": 5.705984306335449,
      "epoch": 5.301394511920828,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002570389610970645,
      "loss": 5.0748,
      "mean_token_accuracy": 0.23171852827072142,
      "num_tokens": 103742539.0,
      "step": 53035
    },
    {
      "entropy": 5.8479743003845215,
      "epoch": 5.301894336982056,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025700338611451835,
      "loss": 5.2641,
      "mean_token_accuracy": 0.21778205633163453,
      "num_tokens": 103752739.0,
      "step": 53040
    },
    {
      "entropy": 5.758499336242676,
      "epoch": 5.302394162043285,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025696781158476227,
      "loss": 5.115,
      "mean_token_accuracy": 0.22483340203762053,
      "num_tokens": 103762710.0,
      "step": 53045
    },
    {
      "entropy": 5.852488803863525,
      "epoch": 5.302893987104514,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002569322375086915,
      "loss": 5.2439,
      "mean_token_accuracy": 0.2158198893070221,
      "num_tokens": 103772465.0,
      "step": 53050
    },
    {
      "entropy": 5.816619157791138,
      "epoch": 5.3033938121657425,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025689666388720097,
      "loss": 5.1451,
      "mean_token_accuracy": 0.2274508684873581,
      "num_tokens": 103781592.0,
      "step": 53055
    },
    {
      "entropy": 5.828047084808349,
      "epoch": 5.30389363722697,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025686109072118566,
      "loss": 5.3663,
      "mean_token_accuracy": 0.2102663606405258,
      "num_tokens": 103790547.0,
      "step": 53060
    },
    {
      "entropy": 5.848613739013672,
      "epoch": 5.304393462288199,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025682551801154064,
      "loss": 5.1927,
      "mean_token_accuracy": 0.22199519127607345,
      "num_tokens": 103799405.0,
      "step": 53065
    },
    {
      "entropy": 5.765083360671997,
      "epoch": 5.304893287349428,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002567899457591609,
      "loss": 5.041,
      "mean_token_accuracy": 0.24457217156887054,
      "num_tokens": 103807993.0,
      "step": 53070
    },
    {
      "entropy": 5.752125215530396,
      "epoch": 5.305393112410656,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002567543739649415,
      "loss": 5.1956,
      "mean_token_accuracy": 0.2234192579984665,
      "num_tokens": 103817703.0,
      "step": 53075
    },
    {
      "entropy": 5.7467255115509035,
      "epoch": 5.305892937471885,
      "grad_norm": 1.25,
      "learning_rate": 0.00025671880262977733,
      "loss": 5.1768,
      "mean_token_accuracy": 0.22776924222707748,
      "num_tokens": 103827314.0,
      "step": 53080
    },
    {
      "entropy": 5.703102684020996,
      "epoch": 5.306392762533114,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025668323175456336,
      "loss": 5.046,
      "mean_token_accuracy": 0.23482533246278764,
      "num_tokens": 103836543.0,
      "step": 53085
    },
    {
      "entropy": 5.769309854507446,
      "epoch": 5.306892587594342,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002566476613401946,
      "loss": 5.2261,
      "mean_token_accuracy": 0.2217671737074852,
      "num_tokens": 103846851.0,
      "step": 53090
    },
    {
      "entropy": 5.689254474639893,
      "epoch": 5.30739241265557,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025661209138756595,
      "loss": 4.9584,
      "mean_token_accuracy": 0.25107315331697466,
      "num_tokens": 103855735.0,
      "step": 53095
    },
    {
      "entropy": 5.7689430713653564,
      "epoch": 5.307892237716799,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025657652189757236,
      "loss": 5.2025,
      "mean_token_accuracy": 0.22087654769420623,
      "num_tokens": 103865154.0,
      "step": 53100
    },
    {
      "entropy": 5.777610635757446,
      "epoch": 5.308392062778028,
      "grad_norm": 1.25,
      "learning_rate": 0.00025654095287110877,
      "loss": 5.1376,
      "mean_token_accuracy": 0.23235800862312317,
      "num_tokens": 103874945.0,
      "step": 53105
    },
    {
      "entropy": 5.751032209396362,
      "epoch": 5.308891887839256,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025650538430907,
      "loss": 5.1517,
      "mean_token_accuracy": 0.2176075205206871,
      "num_tokens": 103884436.0,
      "step": 53110
    },
    {
      "entropy": 5.797897291183472,
      "epoch": 5.309391712900485,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025646981621235107,
      "loss": 5.1308,
      "mean_token_accuracy": 0.22303727865219117,
      "num_tokens": 103893757.0,
      "step": 53115
    },
    {
      "entropy": 5.866995429992675,
      "epoch": 5.309891537961714,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002564342485818467,
      "loss": 5.2842,
      "mean_token_accuracy": 0.21878100484609603,
      "num_tokens": 103903110.0,
      "step": 53120
    },
    {
      "entropy": 5.837282514572143,
      "epoch": 5.310391363022942,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025639868141845185,
      "loss": 5.2436,
      "mean_token_accuracy": 0.22224178463220595,
      "num_tokens": 103912562.0,
      "step": 53125
    },
    {
      "entropy": 5.7588536739349365,
      "epoch": 5.31089118808417,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025636311472306147,
      "loss": 5.1587,
      "mean_token_accuracy": 0.2271958887577057,
      "num_tokens": 103922773.0,
      "step": 53130
    },
    {
      "entropy": 5.800306606292724,
      "epoch": 5.311391013145399,
      "grad_norm": 1.25,
      "learning_rate": 0.00025632754849657025,
      "loss": 5.1587,
      "mean_token_accuracy": 0.22384865432977677,
      "num_tokens": 103932035.0,
      "step": 53135
    },
    {
      "entropy": 5.787423515319825,
      "epoch": 5.311890838206628,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002562919827398732,
      "loss": 5.1193,
      "mean_token_accuracy": 0.22901103496551514,
      "num_tokens": 103942036.0,
      "step": 53140
    },
    {
      "entropy": 5.760022401809692,
      "epoch": 5.312390663267856,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000256256417453865,
      "loss": 5.14,
      "mean_token_accuracy": 0.22818139046430588,
      "num_tokens": 103951968.0,
      "step": 53145
    },
    {
      "entropy": 5.785780811309815,
      "epoch": 5.312890488329085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002562208526394405,
      "loss": 5.2794,
      "mean_token_accuracy": 0.2186596214771271,
      "num_tokens": 103962567.0,
      "step": 53150
    },
    {
      "entropy": 5.749938726425171,
      "epoch": 5.313390313390314,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002561852882974945,
      "loss": 5.0593,
      "mean_token_accuracy": 0.23668616116046906,
      "num_tokens": 103972097.0,
      "step": 53155
    },
    {
      "entropy": 5.888806295394898,
      "epoch": 5.313890138451542,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025614972442892186,
      "loss": 5.2758,
      "mean_token_accuracy": 0.21215032786130905,
      "num_tokens": 103982255.0,
      "step": 53160
    },
    {
      "entropy": 5.880667018890381,
      "epoch": 5.31438996351277,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002561141610346172,
      "loss": 5.3106,
      "mean_token_accuracy": 0.2161382794380188,
      "num_tokens": 103992432.0,
      "step": 53165
    },
    {
      "entropy": 5.801206350326538,
      "epoch": 5.314889788573999,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002560785981154755,
      "loss": 5.2036,
      "mean_token_accuracy": 0.22711200565099715,
      "num_tokens": 104002880.0,
      "step": 53170
    },
    {
      "entropy": 5.806043386459351,
      "epoch": 5.315389613635228,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025604303567239143,
      "loss": 5.1427,
      "mean_token_accuracy": 0.22094618380069733,
      "num_tokens": 104013243.0,
      "step": 53175
    },
    {
      "entropy": 5.770296573638916,
      "epoch": 5.315889438696456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025600747370625965,
      "loss": 5.14,
      "mean_token_accuracy": 0.2294919416308403,
      "num_tokens": 104024465.0,
      "step": 53180
    },
    {
      "entropy": 5.885375118255615,
      "epoch": 5.316389263757685,
      "grad_norm": 1.234375,
      "learning_rate": 0.000255971912217975,
      "loss": 5.2942,
      "mean_token_accuracy": 0.21336025148630142,
      "num_tokens": 104034248.0,
      "step": 53185
    },
    {
      "entropy": 5.771190929412842,
      "epoch": 5.316889088818914,
      "grad_norm": 1.25,
      "learning_rate": 0.00025593635120843203,
      "loss": 5.126,
      "mean_token_accuracy": 0.23295143097639084,
      "num_tokens": 104044806.0,
      "step": 53190
    },
    {
      "entropy": 5.8207399368286135,
      "epoch": 5.317388913880142,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025590079067852557,
      "loss": 5.2071,
      "mean_token_accuracy": 0.21680368930101396,
      "num_tokens": 104053799.0,
      "step": 53195
    },
    {
      "entropy": 5.834380340576172,
      "epoch": 5.31788873894137,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025586523062915036,
      "loss": 5.2385,
      "mean_token_accuracy": 0.2175998494029045,
      "num_tokens": 104063457.0,
      "step": 53200
    },
    {
      "entropy": 5.778292465209961,
      "epoch": 5.318388564002599,
      "grad_norm": 1.125,
      "learning_rate": 0.00025582967106120097,
      "loss": 5.2176,
      "mean_token_accuracy": 0.22007691413164138,
      "num_tokens": 104073070.0,
      "step": 53205
    },
    {
      "entropy": 5.754074048995972,
      "epoch": 5.318888389063828,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002557941119755722,
      "loss": 5.0743,
      "mean_token_accuracy": 0.2321416899561882,
      "num_tokens": 104082890.0,
      "step": 53210
    },
    {
      "entropy": 5.707961893081665,
      "epoch": 5.319388214125056,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002557585533731586,
      "loss": 5.0751,
      "mean_token_accuracy": 0.2401389881968498,
      "num_tokens": 104092629.0,
      "step": 53215
    },
    {
      "entropy": 5.706487369537354,
      "epoch": 5.319888039186285,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002557229952548548,
      "loss": 5.0523,
      "mean_token_accuracy": 0.23887917548418044,
      "num_tokens": 104102475.0,
      "step": 53220
    },
    {
      "entropy": 5.767553377151489,
      "epoch": 5.320387864247514,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002556874376215555,
      "loss": 5.1818,
      "mean_token_accuracy": 0.23238900899887086,
      "num_tokens": 104112027.0,
      "step": 53225
    },
    {
      "entropy": 5.778106164932251,
      "epoch": 5.320887689308742,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025565188047415525,
      "loss": 5.1623,
      "mean_token_accuracy": 0.23083152621984482,
      "num_tokens": 104121992.0,
      "step": 53230
    },
    {
      "entropy": 5.78792781829834,
      "epoch": 5.32138751436997,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002556163238135488,
      "loss": 5.1685,
      "mean_token_accuracy": 0.2279927760362625,
      "num_tokens": 104132267.0,
      "step": 53235
    },
    {
      "entropy": 5.783376455307007,
      "epoch": 5.321887339431199,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025558076764063063,
      "loss": 5.1443,
      "mean_token_accuracy": 0.22516191452741624,
      "num_tokens": 104141467.0,
      "step": 53240
    },
    {
      "entropy": 5.793749475479126,
      "epoch": 5.322387164492428,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002555452119562955,
      "loss": 5.1367,
      "mean_token_accuracy": 0.22414681315422058,
      "num_tokens": 104151596.0,
      "step": 53245
    },
    {
      "entropy": 5.794854640960693,
      "epoch": 5.322886989553656,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025550965676143756,
      "loss": 5.1344,
      "mean_token_accuracy": 0.22982621937990189,
      "num_tokens": 104160956.0,
      "step": 53250
    },
    {
      "entropy": 5.740538692474365,
      "epoch": 5.323386814614885,
      "grad_norm": 1.25,
      "learning_rate": 0.00025547410205695176,
      "loss": 5.1716,
      "mean_token_accuracy": 0.22110863626003266,
      "num_tokens": 104170236.0,
      "step": 53255
    },
    {
      "entropy": 5.818103790283203,
      "epoch": 5.323886639676114,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025543854784373254,
      "loss": 5.2107,
      "mean_token_accuracy": 0.2227917030453682,
      "num_tokens": 104179398.0,
      "step": 53260
    },
    {
      "entropy": 5.817873811721801,
      "epoch": 5.324386464737342,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025540299412267443,
      "loss": 5.216,
      "mean_token_accuracy": 0.21903896927833558,
      "num_tokens": 104190337.0,
      "step": 53265
    },
    {
      "entropy": 5.890847826004029,
      "epoch": 5.32488628979857,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002553674408946719,
      "loss": 5.2716,
      "mean_token_accuracy": 0.2094300076365471,
      "num_tokens": 104199809.0,
      "step": 53270
    },
    {
      "entropy": 5.75649585723877,
      "epoch": 5.325386114859799,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002553318881606195,
      "loss": 5.0705,
      "mean_token_accuracy": 0.2380363568663597,
      "num_tokens": 104210511.0,
      "step": 53275
    },
    {
      "entropy": 5.743492269515992,
      "epoch": 5.325885939921028,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025529633592141185,
      "loss": 5.1158,
      "mean_token_accuracy": 0.23143586665391921,
      "num_tokens": 104220466.0,
      "step": 53280
    },
    {
      "entropy": 5.745902442932129,
      "epoch": 5.326385764982256,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025526078417794317,
      "loss": 5.0535,
      "mean_token_accuracy": 0.24079471528530122,
      "num_tokens": 104230426.0,
      "step": 53285
    },
    {
      "entropy": 5.858150911331177,
      "epoch": 5.326885590043485,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002552252329311081,
      "loss": 5.2696,
      "mean_token_accuracy": 0.21698497235774994,
      "num_tokens": 104240669.0,
      "step": 53290
    },
    {
      "entropy": 5.832006168365479,
      "epoch": 5.327385415104714,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025518968218180113,
      "loss": 5.1323,
      "mean_token_accuracy": 0.22688308209180832,
      "num_tokens": 104250756.0,
      "step": 53295
    },
    {
      "entropy": 5.816184616088867,
      "epoch": 5.327885240165942,
      "grad_norm": 1.125,
      "learning_rate": 0.0002551541319309166,
      "loss": 5.2164,
      "mean_token_accuracy": 0.21340789049863815,
      "num_tokens": 104261766.0,
      "step": 53300
    },
    {
      "entropy": 5.778514623641968,
      "epoch": 5.32838506522717,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025511858217934914,
      "loss": 5.1869,
      "mean_token_accuracy": 0.22679607570171356,
      "num_tokens": 104272261.0,
      "step": 53305
    },
    {
      "entropy": 5.704246234893799,
      "epoch": 5.328884890288399,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002550830329279929,
      "loss": 5.0768,
      "mean_token_accuracy": 0.23176470696926116,
      "num_tokens": 104281949.0,
      "step": 53310
    },
    {
      "entropy": 5.793335437774658,
      "epoch": 5.329384715349628,
      "grad_norm": 1.125,
      "learning_rate": 0.00025504748417774255,
      "loss": 5.1593,
      "mean_token_accuracy": 0.2197689637541771,
      "num_tokens": 104291053.0,
      "step": 53315
    },
    {
      "entropy": 5.789496612548828,
      "epoch": 5.329884540410856,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002550119359294923,
      "loss": 5.1952,
      "mean_token_accuracy": 0.22096786946058272,
      "num_tokens": 104301524.0,
      "step": 53320
    },
    {
      "entropy": 5.784981441497803,
      "epoch": 5.330384365472085,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025497638818413655,
      "loss": 5.1525,
      "mean_token_accuracy": 0.2257826566696167,
      "num_tokens": 104310494.0,
      "step": 53325
    },
    {
      "entropy": 5.801744842529297,
      "epoch": 5.330884190533314,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025494084094256975,
      "loss": 5.1492,
      "mean_token_accuracy": 0.2227514773607254,
      "num_tokens": 104320852.0,
      "step": 53330
    },
    {
      "entropy": 5.889050531387329,
      "epoch": 5.331384015594542,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002549052942056863,
      "loss": 5.2618,
      "mean_token_accuracy": 0.21328273117542268,
      "num_tokens": 104329525.0,
      "step": 53335
    },
    {
      "entropy": 5.722319269180298,
      "epoch": 5.33188384065577,
      "grad_norm": 1.125,
      "learning_rate": 0.0002548697479743804,
      "loss": 5.1012,
      "mean_token_accuracy": 0.23791380375623702,
      "num_tokens": 104339117.0,
      "step": 53340
    },
    {
      "entropy": 5.757438373565674,
      "epoch": 5.332383665716999,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025483420224954644,
      "loss": 5.1136,
      "mean_token_accuracy": 0.2264233186841011,
      "num_tokens": 104348986.0,
      "step": 53345
    },
    {
      "entropy": 5.793820381164551,
      "epoch": 5.332883490778228,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002547986570320788,
      "loss": 5.1718,
      "mean_token_accuracy": 0.22970013171434403,
      "num_tokens": 104358546.0,
      "step": 53350
    },
    {
      "entropy": 5.885388946533203,
      "epoch": 5.333383315839456,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025476311232287173,
      "loss": 5.225,
      "mean_token_accuracy": 0.21710736453533172,
      "num_tokens": 104368919.0,
      "step": 53355
    },
    {
      "entropy": 5.804857015609741,
      "epoch": 5.333883140900685,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002547275681228196,
      "loss": 5.1709,
      "mean_token_accuracy": 0.2241965189576149,
      "num_tokens": 104378421.0,
      "step": 53360
    },
    {
      "entropy": 5.765453720092774,
      "epoch": 5.334382965961913,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002546920244328166,
      "loss": 5.1557,
      "mean_token_accuracy": 0.22512932270765304,
      "num_tokens": 104388417.0,
      "step": 53365
    },
    {
      "entropy": 5.807609605789184,
      "epoch": 5.334882791023142,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025465648125375703,
      "loss": 5.1557,
      "mean_token_accuracy": 0.2329445943236351,
      "num_tokens": 104398324.0,
      "step": 53370
    },
    {
      "entropy": 5.7905560493469235,
      "epoch": 5.33538261608437,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025462093858653515,
      "loss": 5.188,
      "mean_token_accuracy": 0.23206033408641816,
      "num_tokens": 104407684.0,
      "step": 53375
    },
    {
      "entropy": 5.781275701522827,
      "epoch": 5.335882441145599,
      "grad_norm": 1.125,
      "learning_rate": 0.0002545853964320453,
      "loss": 5.1315,
      "mean_token_accuracy": 0.22918639779090882,
      "num_tokens": 104418241.0,
      "step": 53380
    },
    {
      "entropy": 5.778073263168335,
      "epoch": 5.336382266206828,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002545498547911816,
      "loss": 5.103,
      "mean_token_accuracy": 0.22977495789527894,
      "num_tokens": 104428358.0,
      "step": 53385
    },
    {
      "entropy": 5.87214412689209,
      "epoch": 5.336882091268056,
      "grad_norm": 1.125,
      "learning_rate": 0.0002545143136648382,
      "loss": 5.2936,
      "mean_token_accuracy": 0.20659835636615753,
      "num_tokens": 104438520.0,
      "step": 53390
    },
    {
      "entropy": 5.799662256240845,
      "epoch": 5.337381916329285,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025447877305390945,
      "loss": 5.155,
      "mean_token_accuracy": 0.2299586668610573,
      "num_tokens": 104447396.0,
      "step": 53395
    },
    {
      "entropy": 5.88908576965332,
      "epoch": 5.337881741390513,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025444323295928953,
      "loss": 5.1995,
      "mean_token_accuracy": 0.2308349773287773,
      "num_tokens": 104457310.0,
      "step": 53400
    },
    {
      "entropy": 5.816137456893921,
      "epoch": 5.3383815664517416,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002544076933818725,
      "loss": 5.1817,
      "mean_token_accuracy": 0.23188392966985702,
      "num_tokens": 104466991.0,
      "step": 53405
    },
    {
      "entropy": 5.737444734573364,
      "epoch": 5.33888139151297,
      "grad_norm": 1.375,
      "learning_rate": 0.0002543721543225526,
      "loss": 5.0715,
      "mean_token_accuracy": 0.22908291667699815,
      "num_tokens": 104476669.0,
      "step": 53410
    },
    {
      "entropy": 5.809185361862182,
      "epoch": 5.339381216574199,
      "grad_norm": 1.265625,
      "learning_rate": 0.000254336615782224,
      "loss": 5.1834,
      "mean_token_accuracy": 0.22475217133760453,
      "num_tokens": 104486265.0,
      "step": 53415
    },
    {
      "entropy": 5.812472867965698,
      "epoch": 5.339881041635428,
      "grad_norm": 1.09375,
      "learning_rate": 0.00025430107776178086,
      "loss": 5.2428,
      "mean_token_accuracy": 0.22151806354522705,
      "num_tokens": 104497838.0,
      "step": 53420
    },
    {
      "entropy": 5.6956480026245115,
      "epoch": 5.340380866696656,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002542655402621173,
      "loss": 5.0127,
      "mean_token_accuracy": 0.23285801857709884,
      "num_tokens": 104507210.0,
      "step": 53425
    },
    {
      "entropy": 5.713643598556518,
      "epoch": 5.340880691757885,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002542300032841274,
      "loss": 5.0789,
      "mean_token_accuracy": 0.24199433028697967,
      "num_tokens": 104516690.0,
      "step": 53430
    },
    {
      "entropy": 5.800882196426391,
      "epoch": 5.341380516819113,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002541944668287052,
      "loss": 5.1878,
      "mean_token_accuracy": 0.22380493730306625,
      "num_tokens": 104526668.0,
      "step": 53435
    },
    {
      "entropy": 5.791344070434571,
      "epoch": 5.3418803418803416,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000254158930896745,
      "loss": 5.1465,
      "mean_token_accuracy": 0.22345794141292571,
      "num_tokens": 104536268.0,
      "step": 53440
    },
    {
      "entropy": 5.738384485244751,
      "epoch": 5.34238016694157,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002541233954891406,
      "loss": 5.0181,
      "mean_token_accuracy": 0.24156279265880584,
      "num_tokens": 104545656.0,
      "step": 53445
    },
    {
      "entropy": 5.75295467376709,
      "epoch": 5.342879992002799,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025408786060678626,
      "loss": 5.1203,
      "mean_token_accuracy": 0.22787025719881057,
      "num_tokens": 104555983.0,
      "step": 53450
    },
    {
      "entropy": 5.756746912002564,
      "epoch": 5.343379817064028,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000254052326250576,
      "loss": 5.1372,
      "mean_token_accuracy": 0.22291099280118942,
      "num_tokens": 104564693.0,
      "step": 53455
    },
    {
      "entropy": 5.792466402053833,
      "epoch": 5.343879642125256,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025401679242140386,
      "loss": 5.1725,
      "mean_token_accuracy": 0.22516136467456818,
      "num_tokens": 104574661.0,
      "step": 53460
    },
    {
      "entropy": 5.805507850646973,
      "epoch": 5.344379467186485,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025398125912016366,
      "loss": 5.2238,
      "mean_token_accuracy": 0.2188225880265236,
      "num_tokens": 104585186.0,
      "step": 53465
    },
    {
      "entropy": 5.830627012252807,
      "epoch": 5.344879292247713,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002539457263477497,
      "loss": 5.1758,
      "mean_token_accuracy": 0.22523312270641327,
      "num_tokens": 104593895.0,
      "step": 53470
    },
    {
      "entropy": 5.7853141784667965,
      "epoch": 5.3453791173089416,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025391019410505575,
      "loss": 5.2114,
      "mean_token_accuracy": 0.21693249344825744,
      "num_tokens": 104603762.0,
      "step": 53475
    },
    {
      "entropy": 5.657917165756226,
      "epoch": 5.34587894237017,
      "grad_norm": 1.28125,
      "learning_rate": 0.000253874662392976,
      "loss": 5.0104,
      "mean_token_accuracy": 0.24310302138328552,
      "num_tokens": 104613995.0,
      "step": 53480
    },
    {
      "entropy": 5.646525049209595,
      "epoch": 5.346378767431399,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025383913121240423,
      "loss": 4.9944,
      "mean_token_accuracy": 0.2410804033279419,
      "num_tokens": 104623657.0,
      "step": 53485
    },
    {
      "entropy": 5.903172636032105,
      "epoch": 5.346878592492628,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025380360056423445,
      "loss": 5.2794,
      "mean_token_accuracy": 0.2213760420680046,
      "num_tokens": 104633326.0,
      "step": 53490
    },
    {
      "entropy": 5.672009992599487,
      "epoch": 5.347378417553856,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002537680704493607,
      "loss": 5.0258,
      "mean_token_accuracy": 0.23955013155937194,
      "num_tokens": 104644305.0,
      "step": 53495
    },
    {
      "entropy": 5.759589624404907,
      "epoch": 5.347878242615085,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025373254086867683,
      "loss": 5.1125,
      "mean_token_accuracy": 0.23008694350719452,
      "num_tokens": 104652412.0,
      "step": 53500
    },
    {
      "entropy": 5.710273694992066,
      "epoch": 5.348378067676313,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002536970118230767,
      "loss": 5.1006,
      "mean_token_accuracy": 0.23549072444438934,
      "num_tokens": 104661827.0,
      "step": 53505
    },
    {
      "entropy": 5.783714056015015,
      "epoch": 5.3488778927375416,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002536614833134543,
      "loss": 5.1624,
      "mean_token_accuracy": 0.22261994928121567,
      "num_tokens": 104671530.0,
      "step": 53510
    },
    {
      "entropy": 5.880230283737182,
      "epoch": 5.34937771779877,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002536259553407034,
      "loss": 5.2401,
      "mean_token_accuracy": 0.21990704238414766,
      "num_tokens": 104680656.0,
      "step": 53515
    },
    {
      "entropy": 5.788092231750488,
      "epoch": 5.349877542859999,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002535904279057181,
      "loss": 5.1899,
      "mean_token_accuracy": 0.22199291586875916,
      "num_tokens": 104690406.0,
      "step": 53520
    },
    {
      "entropy": 5.653230476379394,
      "epoch": 5.350377367921228,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002535549010093922,
      "loss": 4.983,
      "mean_token_accuracy": 0.2410133436322212,
      "num_tokens": 104700313.0,
      "step": 53525
    },
    {
      "entropy": 5.753574705123901,
      "epoch": 5.350877192982456,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002535193746526194,
      "loss": 5.1776,
      "mean_token_accuracy": 0.22352349758148193,
      "num_tokens": 104709787.0,
      "step": 53530
    },
    {
      "entropy": 5.595990109443664,
      "epoch": 5.351377018043685,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025348384883629345,
      "loss": 4.9268,
      "mean_token_accuracy": 0.25061275213956835,
      "num_tokens": 104719590.0,
      "step": 53535
    },
    {
      "entropy": 5.790813446044922,
      "epoch": 5.351876843104913,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002534483235613085,
      "loss": 5.1361,
      "mean_token_accuracy": 0.23591817170381546,
      "num_tokens": 104729433.0,
      "step": 53540
    },
    {
      "entropy": 5.804903364181518,
      "epoch": 5.3523766681661415,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025341279882855805,
      "loss": 5.1532,
      "mean_token_accuracy": 0.2229967772960663,
      "num_tokens": 104738695.0,
      "step": 53545
    },
    {
      "entropy": 5.800535488128662,
      "epoch": 5.35287649322737,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002533772746389361,
      "loss": 5.2092,
      "mean_token_accuracy": 0.23088542968034745,
      "num_tokens": 104748365.0,
      "step": 53550
    },
    {
      "entropy": 5.8420209884643555,
      "epoch": 5.353376318288599,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002533417509933364,
      "loss": 5.2717,
      "mean_token_accuracy": 0.21882434487342833,
      "num_tokens": 104757620.0,
      "step": 53555
    },
    {
      "entropy": 5.760773468017578,
      "epoch": 5.353876143349828,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025330622789265263,
      "loss": 5.1445,
      "mean_token_accuracy": 0.22843940556049347,
      "num_tokens": 104766208.0,
      "step": 53560
    },
    {
      "entropy": 5.7658415794372555,
      "epoch": 5.354375968411056,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002532707053377786,
      "loss": 5.0828,
      "mean_token_accuracy": 0.2248542606830597,
      "num_tokens": 104775577.0,
      "step": 53565
    },
    {
      "entropy": 5.7597838878631595,
      "epoch": 5.354875793472285,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025323518332960793,
      "loss": 5.1498,
      "mean_token_accuracy": 0.2209981858730316,
      "num_tokens": 104785318.0,
      "step": 53570
    },
    {
      "entropy": 5.810941886901856,
      "epoch": 5.355375618533513,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002531996618690345,
      "loss": 5.1722,
      "mean_token_accuracy": 0.22415666580200194,
      "num_tokens": 104795173.0,
      "step": 53575
    },
    {
      "entropy": 5.797458791732788,
      "epoch": 5.3558754435947415,
      "grad_norm": 1.25,
      "learning_rate": 0.000253164140956952,
      "loss": 5.1517,
      "mean_token_accuracy": 0.22812752723693847,
      "num_tokens": 104805921.0,
      "step": 53580
    },
    {
      "entropy": 5.860481119155883,
      "epoch": 5.35637526865597,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025312862059425413,
      "loss": 5.2719,
      "mean_token_accuracy": 0.2172760158777237,
      "num_tokens": 104816352.0,
      "step": 53585
    },
    {
      "entropy": 5.888976192474365,
      "epoch": 5.356875093717199,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025309310078183445,
      "loss": 5.2395,
      "mean_token_accuracy": 0.21689113825559617,
      "num_tokens": 104826223.0,
      "step": 53590
    },
    {
      "entropy": 5.790122652053833,
      "epoch": 5.357374918778428,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025305758152058684,
      "loss": 5.103,
      "mean_token_accuracy": 0.2307119145989418,
      "num_tokens": 104835567.0,
      "step": 53595
    },
    {
      "entropy": 5.7683281898498535,
      "epoch": 5.357874743839656,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002530220628114048,
      "loss": 5.0859,
      "mean_token_accuracy": 0.22444422394037247,
      "num_tokens": 104845509.0,
      "step": 53600
    },
    {
      "entropy": 5.721364545822143,
      "epoch": 5.358374568900885,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002529865446551819,
      "loss": 5.0373,
      "mean_token_accuracy": 0.24190520346164704,
      "num_tokens": 104854118.0,
      "step": 53605
    },
    {
      "entropy": 5.797693061828613,
      "epoch": 5.358874393962113,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000252951027052812,
      "loss": 5.1911,
      "mean_token_accuracy": 0.22537602186203004,
      "num_tokens": 104864130.0,
      "step": 53610
    },
    {
      "entropy": 5.8052538394927975,
      "epoch": 5.3593742190233415,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025291551000518846,
      "loss": 5.1393,
      "mean_token_accuracy": 0.22999977320432663,
      "num_tokens": 104873450.0,
      "step": 53615
    },
    {
      "entropy": 5.752331733703613,
      "epoch": 5.35987404408457,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025287999351320504,
      "loss": 5.1135,
      "mean_token_accuracy": 0.22718360871076584,
      "num_tokens": 104881930.0,
      "step": 53620
    },
    {
      "entropy": 5.7853423118591305,
      "epoch": 5.360373869145799,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00025284447757775534,
      "loss": 5.0897,
      "mean_token_accuracy": 0.22308663725852967,
      "num_tokens": 104892160.0,
      "step": 53625
    },
    {
      "entropy": 5.808513402938843,
      "epoch": 5.360873694207028,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002528089621997329,
      "loss": 5.2047,
      "mean_token_accuracy": 0.22732383608818055,
      "num_tokens": 104902333.0,
      "step": 53630
    },
    {
      "entropy": 5.748247146606445,
      "epoch": 5.361373519268256,
      "grad_norm": 1.375,
      "learning_rate": 0.00025277344738003116,
      "loss": 5.0688,
      "mean_token_accuracy": 0.23588966578245163,
      "num_tokens": 104912698.0,
      "step": 53635
    },
    {
      "entropy": 5.794940090179443,
      "epoch": 5.361873344329485,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002527379331195439,
      "loss": 5.1166,
      "mean_token_accuracy": 0.2364719480276108,
      "num_tokens": 104922271.0,
      "step": 53640
    },
    {
      "entropy": 5.7635729789733885,
      "epoch": 5.362373169390713,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002527024194191644,
      "loss": 5.2465,
      "mean_token_accuracy": 0.22021040916442872,
      "num_tokens": 104932347.0,
      "step": 53645
    },
    {
      "entropy": 5.868723440170288,
      "epoch": 5.3628729944519415,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025266690627978627,
      "loss": 5.2096,
      "mean_token_accuracy": 0.22389870136976242,
      "num_tokens": 104941707.0,
      "step": 53650
    },
    {
      "entropy": 5.781741142272949,
      "epoch": 5.36337281951317,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002526313937023031,
      "loss": 5.1092,
      "mean_token_accuracy": 0.22465988248586655,
      "num_tokens": 104951135.0,
      "step": 53655
    },
    {
      "entropy": 5.75066065788269,
      "epoch": 5.363872644574399,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025259588168760833,
      "loss": 5.103,
      "mean_token_accuracy": 0.23893254548311232,
      "num_tokens": 104961263.0,
      "step": 53660
    },
    {
      "entropy": 5.848050975799561,
      "epoch": 5.364372469635628,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002525603702365954,
      "loss": 5.2551,
      "mean_token_accuracy": 0.2136605367064476,
      "num_tokens": 104971501.0,
      "step": 53665
    },
    {
      "entropy": 5.721495676040649,
      "epoch": 5.364872294696856,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002525248593501578,
      "loss": 5.15,
      "mean_token_accuracy": 0.22198071628808974,
      "num_tokens": 104981377.0,
      "step": 53670
    },
    {
      "entropy": 5.689824056625366,
      "epoch": 5.365372119758085,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002524893490291889,
      "loss": 5.1249,
      "mean_token_accuracy": 0.2321523830294609,
      "num_tokens": 104991334.0,
      "step": 53675
    },
    {
      "entropy": 5.769001150131226,
      "epoch": 5.365871944819313,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002524538392745822,
      "loss": 5.0929,
      "mean_token_accuracy": 0.2356332302093506,
      "num_tokens": 105000495.0,
      "step": 53680
    },
    {
      "entropy": 5.726848220825195,
      "epoch": 5.3663717698805415,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002524183300872311,
      "loss": 5.087,
      "mean_token_accuracy": 0.2365215614438057,
      "num_tokens": 105010123.0,
      "step": 53685
    },
    {
      "entropy": 5.721283054351806,
      "epoch": 5.36687159494177,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025238282146802907,
      "loss": 5.0657,
      "mean_token_accuracy": 0.23298803418874742,
      "num_tokens": 105019505.0,
      "step": 53690
    },
    {
      "entropy": 5.864450407028198,
      "epoch": 5.367371420002999,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025234731341786934,
      "loss": 5.2752,
      "mean_token_accuracy": 0.21305576115846633,
      "num_tokens": 105030706.0,
      "step": 53695
    },
    {
      "entropy": 5.863615036010742,
      "epoch": 5.367871245064228,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025231180593764536,
      "loss": 5.1535,
      "mean_token_accuracy": 0.2228919804096222,
      "num_tokens": 105039146.0,
      "step": 53700
    },
    {
      "entropy": 5.817164468765259,
      "epoch": 5.368371070125456,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002522762990282506,
      "loss": 5.1838,
      "mean_token_accuracy": 0.22806286215782165,
      "num_tokens": 105049028.0,
      "step": 53705
    },
    {
      "entropy": 5.768606233596802,
      "epoch": 5.368870895186685,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002522407926905782,
      "loss": 5.1204,
      "mean_token_accuracy": 0.2367040768265724,
      "num_tokens": 105057984.0,
      "step": 53710
    },
    {
      "entropy": 5.776456737518311,
      "epoch": 5.369370720247913,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002522052869255217,
      "loss": 5.1522,
      "mean_token_accuracy": 0.22899897247552872,
      "num_tokens": 105067089.0,
      "step": 53715
    },
    {
      "entropy": 5.785245704650879,
      "epoch": 5.3698705453091415,
      "grad_norm": 1.25,
      "learning_rate": 0.00025216978173397427,
      "loss": 5.1508,
      "mean_token_accuracy": 0.22400080412626266,
      "num_tokens": 105076092.0,
      "step": 53720
    },
    {
      "entropy": 5.75287561416626,
      "epoch": 5.37037037037037,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025213427711682926,
      "loss": 5.0976,
      "mean_token_accuracy": 0.22979717552661896,
      "num_tokens": 105085146.0,
      "step": 53725
    },
    {
      "entropy": 5.756229400634766,
      "epoch": 5.370870195431599,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002520987730749799,
      "loss": 5.1567,
      "mean_token_accuracy": 0.22272273004055024,
      "num_tokens": 105095205.0,
      "step": 53730
    },
    {
      "entropy": 5.77058777809143,
      "epoch": 5.371370020492828,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025206326960931964,
      "loss": 5.1042,
      "mean_token_accuracy": 0.2400142014026642,
      "num_tokens": 105104648.0,
      "step": 53735
    },
    {
      "entropy": 5.728450679779053,
      "epoch": 5.371869845554056,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002520277667207416,
      "loss": 5.1002,
      "mean_token_accuracy": 0.231487475335598,
      "num_tokens": 105114530.0,
      "step": 53740
    },
    {
      "entropy": 5.830158424377442,
      "epoch": 5.372369670615285,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025199226441013895,
      "loss": 5.2254,
      "mean_token_accuracy": 0.21819366365671158,
      "num_tokens": 105124716.0,
      "step": 53745
    },
    {
      "entropy": 5.772002458572388,
      "epoch": 5.372869495676513,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025195676267840503,
      "loss": 5.1385,
      "mean_token_accuracy": 0.23142073005437852,
      "num_tokens": 105134162.0,
      "step": 53750
    },
    {
      "entropy": 5.767509937286377,
      "epoch": 5.3733693207377415,
      "grad_norm": 1.796875,
      "learning_rate": 0.00025192126152643305,
      "loss": 5.108,
      "mean_token_accuracy": 0.23534816056489943,
      "num_tokens": 105143583.0,
      "step": 53755
    },
    {
      "entropy": 5.816254186630249,
      "epoch": 5.37386914579897,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025188576095511617,
      "loss": 5.1937,
      "mean_token_accuracy": 0.21901217252016067,
      "num_tokens": 105152380.0,
      "step": 53760
    },
    {
      "entropy": 5.848938798904419,
      "epoch": 5.374368970860199,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025185026096534755,
      "loss": 5.1918,
      "mean_token_accuracy": 0.23025655895471572,
      "num_tokens": 105162379.0,
      "step": 53765
    },
    {
      "entropy": 5.775945711135864,
      "epoch": 5.374868795921428,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025181476155802053,
      "loss": 5.1278,
      "mean_token_accuracy": 0.23551561683416367,
      "num_tokens": 105171699.0,
      "step": 53770
    },
    {
      "entropy": 5.720866870880127,
      "epoch": 5.375368620982656,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025177926273402805,
      "loss": 5.0881,
      "mean_token_accuracy": 0.2386612683534622,
      "num_tokens": 105182031.0,
      "step": 53775
    },
    {
      "entropy": 5.880202960968018,
      "epoch": 5.375868446043885,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002517437644942634,
      "loss": 5.2472,
      "mean_token_accuracy": 0.210963873565197,
      "num_tokens": 105191079.0,
      "step": 53780
    },
    {
      "entropy": 5.76935043334961,
      "epoch": 5.376368271105113,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025170826683961964,
      "loss": 5.1584,
      "mean_token_accuracy": 0.225346440076828,
      "num_tokens": 105201651.0,
      "step": 53785
    },
    {
      "entropy": 5.787536525726319,
      "epoch": 5.3768680961663415,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025167276977098983,
      "loss": 5.1831,
      "mean_token_accuracy": 0.2312415823340416,
      "num_tokens": 105211220.0,
      "step": 53790
    },
    {
      "entropy": 5.730712413787842,
      "epoch": 5.37736792122757,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025163727328926717,
      "loss": 5.1704,
      "mean_token_accuracy": 0.21876238584518432,
      "num_tokens": 105221104.0,
      "step": 53795
    },
    {
      "entropy": 5.74732723236084,
      "epoch": 5.377867746288799,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002516017773953447,
      "loss": 5.1305,
      "mean_token_accuracy": 0.23568220287561417,
      "num_tokens": 105230449.0,
      "step": 53800
    },
    {
      "entropy": 5.763514757156372,
      "epoch": 5.378367571350028,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025156628209011564,
      "loss": 5.1387,
      "mean_token_accuracy": 0.22832028120756148,
      "num_tokens": 105240526.0,
      "step": 53805
    },
    {
      "entropy": 5.744978523254394,
      "epoch": 5.378867396411256,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002515307873744727,
      "loss": 5.125,
      "mean_token_accuracy": 0.24328053146600723,
      "num_tokens": 105250704.0,
      "step": 53810
    },
    {
      "entropy": 5.816682147979736,
      "epoch": 5.379367221472485,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002514952932493092,
      "loss": 5.1801,
      "mean_token_accuracy": 0.2220791220664978,
      "num_tokens": 105259565.0,
      "step": 53815
    },
    {
      "entropy": 5.776864147186279,
      "epoch": 5.379867046533713,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025145979971551803,
      "loss": 5.1476,
      "mean_token_accuracy": 0.22619072645902633,
      "num_tokens": 105268549.0,
      "step": 53820
    },
    {
      "entropy": 5.783485794067383,
      "epoch": 5.3803668715949415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002514243067739922,
      "loss": 5.1118,
      "mean_token_accuracy": 0.24102537631988524,
      "num_tokens": 105279337.0,
      "step": 53825
    },
    {
      "entropy": 5.734148263931274,
      "epoch": 5.38086669665617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002513888144256248,
      "loss": 5.1213,
      "mean_token_accuracy": 0.23346594274044036,
      "num_tokens": 105289507.0,
      "step": 53830
    },
    {
      "entropy": 5.625352764129639,
      "epoch": 5.381366521717399,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002513533226713088,
      "loss": 5.0195,
      "mean_token_accuracy": 0.23708058297634124,
      "num_tokens": 105298425.0,
      "step": 53835
    },
    {
      "entropy": 5.739516735076904,
      "epoch": 5.381866346778628,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025131783151193703,
      "loss": 5.104,
      "mean_token_accuracy": 0.22599755972623825,
      "num_tokens": 105307963.0,
      "step": 53840
    },
    {
      "entropy": 5.732106733322143,
      "epoch": 5.382366171839856,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025128234094840256,
      "loss": 5.0917,
      "mean_token_accuracy": 0.2341120183467865,
      "num_tokens": 105317463.0,
      "step": 53845
    },
    {
      "entropy": 5.880716037750244,
      "epoch": 5.382865996901085,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002512468509815983,
      "loss": 5.3376,
      "mean_token_accuracy": 0.21198750734329225,
      "num_tokens": 105328882.0,
      "step": 53850
    },
    {
      "entropy": 5.765270662307739,
      "epoch": 5.383365821962313,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025121136161241714,
      "loss": 5.0839,
      "mean_token_accuracy": 0.23686848878860473,
      "num_tokens": 105337487.0,
      "step": 53855
    },
    {
      "entropy": 5.824220514297485,
      "epoch": 5.3838656470235415,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025117587284175197,
      "loss": 5.1753,
      "mean_token_accuracy": 0.21759494096040727,
      "num_tokens": 105347395.0,
      "step": 53860
    },
    {
      "entropy": 5.7792308807373045,
      "epoch": 5.38436547208477,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002511403846704957,
      "loss": 5.1913,
      "mean_token_accuracy": 0.22919658869504927,
      "num_tokens": 105356311.0,
      "step": 53865
    },
    {
      "entropy": 5.810400009155273,
      "epoch": 5.384865297145999,
      "grad_norm": 1.25,
      "learning_rate": 0.00025110489709954134,
      "loss": 5.2555,
      "mean_token_accuracy": 0.21832794696092606,
      "num_tokens": 105366313.0,
      "step": 53870
    },
    {
      "entropy": 5.808086347579956,
      "epoch": 5.385365122207228,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025106941012978157,
      "loss": 5.1719,
      "mean_token_accuracy": 0.22365859746932984,
      "num_tokens": 105376514.0,
      "step": 53875
    },
    {
      "entropy": 5.692718076705932,
      "epoch": 5.385864947268456,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002510339237621092,
      "loss": 5.0195,
      "mean_token_accuracy": 0.23757107704877853,
      "num_tokens": 105385567.0,
      "step": 53880
    },
    {
      "entropy": 5.702941751480102,
      "epoch": 5.386364772329685,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025099843799741725,
      "loss": 5.0525,
      "mean_token_accuracy": 0.2358887642621994,
      "num_tokens": 105394972.0,
      "step": 53885
    },
    {
      "entropy": 5.701433801651001,
      "epoch": 5.386864597390913,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002509629528365984,
      "loss": 5.0337,
      "mean_token_accuracy": 0.23381958156824112,
      "num_tokens": 105403739.0,
      "step": 53890
    },
    {
      "entropy": 5.8244908332824705,
      "epoch": 5.3873644224521415,
      "grad_norm": 1.1875,
      "learning_rate": 0.00025092746828054544,
      "loss": 5.2301,
      "mean_token_accuracy": 0.22399598807096482,
      "num_tokens": 105414417.0,
      "step": 53895
    },
    {
      "entropy": 5.836722898483276,
      "epoch": 5.38786424751337,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025089198433015113,
      "loss": 5.1622,
      "mean_token_accuracy": 0.23025309592485427,
      "num_tokens": 105424497.0,
      "step": 53900
    },
    {
      "entropy": 5.885727024078369,
      "epoch": 5.388364072574599,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025085650098630836,
      "loss": 5.2455,
      "mean_token_accuracy": 0.2176447257399559,
      "num_tokens": 105436728.0,
      "step": 53905
    },
    {
      "entropy": 5.796839237213135,
      "epoch": 5.388863897635828,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025082101824990975,
      "loss": 5.2281,
      "mean_token_accuracy": 0.218881094455719,
      "num_tokens": 105446150.0,
      "step": 53910
    },
    {
      "entropy": 5.8693396091461185,
      "epoch": 5.389363722697056,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002507855361218482,
      "loss": 5.3099,
      "mean_token_accuracy": 0.21839727461338043,
      "num_tokens": 105457146.0,
      "step": 53915
    },
    {
      "entropy": 5.767719364166259,
      "epoch": 5.389863547758285,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002507500546030162,
      "loss": 5.1923,
      "mean_token_accuracy": 0.2298763006925583,
      "num_tokens": 105467742.0,
      "step": 53920
    },
    {
      "entropy": 5.820969820022583,
      "epoch": 5.390363372819513,
      "grad_norm": 1.125,
      "learning_rate": 0.00025071457369430666,
      "loss": 5.0987,
      "mean_token_accuracy": 0.2242877423763275,
      "num_tokens": 105477741.0,
      "step": 53925
    },
    {
      "entropy": 5.7751997947692875,
      "epoch": 5.3908631978807415,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025067909339661224,
      "loss": 5.1595,
      "mean_token_accuracy": 0.22415812760591508,
      "num_tokens": 105487864.0,
      "step": 53930
    },
    {
      "entropy": 5.768358325958252,
      "epoch": 5.39136302294197,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002506436137108255,
      "loss": 5.1379,
      "mean_token_accuracy": 0.23000444918870927,
      "num_tokens": 105497772.0,
      "step": 53935
    },
    {
      "entropy": 5.802756023406983,
      "epoch": 5.391862848003199,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002506081346378392,
      "loss": 5.1509,
      "mean_token_accuracy": 0.22251757979393005,
      "num_tokens": 105509054.0,
      "step": 53940
    },
    {
      "entropy": 5.733621120452881,
      "epoch": 5.392362673064428,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025057265617854603,
      "loss": 5.1057,
      "mean_token_accuracy": 0.2255749598145485,
      "num_tokens": 105517994.0,
      "step": 53945
    },
    {
      "entropy": 5.79025444984436,
      "epoch": 5.392862498125656,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002505371783338385,
      "loss": 5.2205,
      "mean_token_accuracy": 0.2119619995355606,
      "num_tokens": 105526750.0,
      "step": 53950
    },
    {
      "entropy": 5.8012511253356935,
      "epoch": 5.393362323186885,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002505017011046092,
      "loss": 5.2055,
      "mean_token_accuracy": 0.22495187371969222,
      "num_tokens": 105537876.0,
      "step": 53955
    },
    {
      "entropy": 5.786267185211182,
      "epoch": 5.393862148248113,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025046622449175084,
      "loss": 5.1067,
      "mean_token_accuracy": 0.22572419792413712,
      "num_tokens": 105547292.0,
      "step": 53960
    },
    {
      "entropy": 5.88256049156189,
      "epoch": 5.3943619733093415,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025043074849615595,
      "loss": 5.288,
      "mean_token_accuracy": 0.215806445479393,
      "num_tokens": 105556598.0,
      "step": 53965
    },
    {
      "entropy": 5.702832794189453,
      "epoch": 5.39486179837057,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025039527311871714,
      "loss": 5.1149,
      "mean_token_accuracy": 0.23205333352088928,
      "num_tokens": 105566185.0,
      "step": 53970
    },
    {
      "entropy": 5.751948070526123,
      "epoch": 5.395361623431799,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025035979836032687,
      "loss": 5.1452,
      "mean_token_accuracy": 0.22750134319067,
      "num_tokens": 105576201.0,
      "step": 53975
    },
    {
      "entropy": 5.8132730484008786,
      "epoch": 5.395861448493028,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002503243242218777,
      "loss": 5.1729,
      "mean_token_accuracy": 0.22956819236278533,
      "num_tokens": 105586958.0,
      "step": 53980
    },
    {
      "entropy": 5.895761108398437,
      "epoch": 5.396361273554256,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002502888507042623,
      "loss": 5.3028,
      "mean_token_accuracy": 0.2095980703830719,
      "num_tokens": 105596287.0,
      "step": 53985
    },
    {
      "entropy": 5.801318073272705,
      "epoch": 5.396861098615485,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000250253377808373,
      "loss": 5.1583,
      "mean_token_accuracy": 0.22761653363704681,
      "num_tokens": 105605907.0,
      "step": 53990
    },
    {
      "entropy": 5.766356611251831,
      "epoch": 5.397360923676713,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002502179055351023,
      "loss": 5.1279,
      "mean_token_accuracy": 0.22617551684379578,
      "num_tokens": 105615347.0,
      "step": 53995
    },
    {
      "entropy": 5.757268810272217,
      "epoch": 5.3978607487379415,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002501824338853428,
      "loss": 5.1299,
      "mean_token_accuracy": 0.23365454971790314,
      "num_tokens": 105624921.0,
      "step": 54000
    },
    {
      "epoch": 5.3978607487379415,
      "eval_entropy": 5.545748165359635,
      "eval_loss": 5.363903045654297,
      "eval_mean_token_accuracy": 0.22265285456059244,
      "eval_num_tokens": 105624921.0,
      "eval_runtime": 23.7935,
      "eval_samples_per_second": 1304.771,
      "eval_steps_per_second": 163.112,
      "step": 54000
    },
    {
      "entropy": 5.8232097148895265,
      "epoch": 5.39836057379917,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025014696285998686,
      "loss": 5.2273,
      "mean_token_accuracy": 0.2130809471011162,
      "num_tokens": 105633936.0,
      "step": 54005
    },
    {
      "entropy": 5.829519939422608,
      "epoch": 5.398860398860399,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002501114924599269,
      "loss": 5.1827,
      "mean_token_accuracy": 0.22150868624448777,
      "num_tokens": 105643436.0,
      "step": 54010
    },
    {
      "entropy": 5.7891288757324215,
      "epoch": 5.399360223921628,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025007602268605546,
      "loss": 5.1629,
      "mean_token_accuracy": 0.23468869030475617,
      "num_tokens": 105654502.0,
      "step": 54015
    },
    {
      "entropy": 5.742329740524292,
      "epoch": 5.399860048982856,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025004055353926486,
      "loss": 5.115,
      "mean_token_accuracy": 0.23259345889091493,
      "num_tokens": 105662901.0,
      "step": 54020
    },
    {
      "entropy": 5.850857448577881,
      "epoch": 5.400359874044085,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002500050850204474,
      "loss": 5.2251,
      "mean_token_accuracy": 0.218539497256279,
      "num_tokens": 105671893.0,
      "step": 54025
    },
    {
      "entropy": 5.847493600845337,
      "epoch": 5.400859699105313,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024996961713049564,
      "loss": 5.2316,
      "mean_token_accuracy": 0.2194249376654625,
      "num_tokens": 105681513.0,
      "step": 54030
    },
    {
      "entropy": 5.853308248519897,
      "epoch": 5.4013595241665415,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024993414987030193,
      "loss": 5.2234,
      "mean_token_accuracy": 0.22270411998033524,
      "num_tokens": 105691340.0,
      "step": 54035
    },
    {
      "entropy": 5.7870124816894535,
      "epoch": 5.40185934922777,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024989868324075843,
      "loss": 5.1033,
      "mean_token_accuracy": 0.23158072531223298,
      "num_tokens": 105700485.0,
      "step": 54040
    },
    {
      "entropy": 5.704996490478516,
      "epoch": 5.402359174288999,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002498632172427577,
      "loss": 5.0447,
      "mean_token_accuracy": 0.23287152349948884,
      "num_tokens": 105709800.0,
      "step": 54045
    },
    {
      "entropy": 5.769640970230102,
      "epoch": 5.402858999350228,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002498277518771919,
      "loss": 5.2115,
      "mean_token_accuracy": 0.22661124169826508,
      "num_tokens": 105719766.0,
      "step": 54050
    },
    {
      "entropy": 5.702660369873047,
      "epoch": 5.403358824411456,
      "grad_norm": 1.25,
      "learning_rate": 0.00024979228714495347,
      "loss": 5.0312,
      "mean_token_accuracy": 0.236152184009552,
      "num_tokens": 105730190.0,
      "step": 54055
    },
    {
      "entropy": 5.783793258666992,
      "epoch": 5.403858649472685,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024975682304693455,
      "loss": 5.1305,
      "mean_token_accuracy": 0.22681175619363786,
      "num_tokens": 105740528.0,
      "step": 54060
    },
    {
      "entropy": 5.784833192825317,
      "epoch": 5.404358474533913,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002497213595840275,
      "loss": 5.1475,
      "mean_token_accuracy": 0.22432292401790618,
      "num_tokens": 105750160.0,
      "step": 54065
    },
    {
      "entropy": 5.825800561904908,
      "epoch": 5.4048582995951415,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002496858967571245,
      "loss": 5.2073,
      "mean_token_accuracy": 0.22820006161928177,
      "num_tokens": 105760342.0,
      "step": 54070
    },
    {
      "entropy": 5.795286417007446,
      "epoch": 5.40535812465637,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024965043456711783,
      "loss": 5.217,
      "mean_token_accuracy": 0.21574353128671647,
      "num_tokens": 105769709.0,
      "step": 54075
    },
    {
      "entropy": 5.7696422100067135,
      "epoch": 5.405857949717599,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024961497301489977,
      "loss": 5.179,
      "mean_token_accuracy": 0.22034643739461898,
      "num_tokens": 105779834.0,
      "step": 54080
    },
    {
      "entropy": 5.766673517227173,
      "epoch": 5.406357774778828,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002495795121013624,
      "loss": 5.1328,
      "mean_token_accuracy": 0.23357392698526383,
      "num_tokens": 105789651.0,
      "step": 54085
    },
    {
      "entropy": 5.84404878616333,
      "epoch": 5.406857599840056,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024954405182739793,
      "loss": 5.1953,
      "mean_token_accuracy": 0.22055042684078216,
      "num_tokens": 105799269.0,
      "step": 54090
    },
    {
      "entropy": 5.787008476257324,
      "epoch": 5.407357424901285,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002495085921938986,
      "loss": 5.1278,
      "mean_token_accuracy": 0.22506877779960632,
      "num_tokens": 105807786.0,
      "step": 54095
    },
    {
      "entropy": 5.716583681106568,
      "epoch": 5.407857249962513,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024947313320175654,
      "loss": 5.033,
      "mean_token_accuracy": 0.23826221525669097,
      "num_tokens": 105817875.0,
      "step": 54100
    },
    {
      "entropy": 5.725995349884033,
      "epoch": 5.4083570750237415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002494376748518638,
      "loss": 5.0175,
      "mean_token_accuracy": 0.2344806209206581,
      "num_tokens": 105827519.0,
      "step": 54105
    },
    {
      "entropy": 5.738178873062134,
      "epoch": 5.40885690008497,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002494022171451127,
      "loss": 5.0729,
      "mean_token_accuracy": 0.23747360706329346,
      "num_tokens": 105837058.0,
      "step": 54110
    },
    {
      "entropy": 5.703749418258667,
      "epoch": 5.409356725146199,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024936676008239513,
      "loss": 5.1628,
      "mean_token_accuracy": 0.2236175552010536,
      "num_tokens": 105846797.0,
      "step": 54115
    },
    {
      "entropy": 5.844391870498657,
      "epoch": 5.409856550207428,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002493313036646033,
      "loss": 5.2759,
      "mean_token_accuracy": 0.21941557228565217,
      "num_tokens": 105855651.0,
      "step": 54120
    },
    {
      "entropy": 5.821986293792724,
      "epoch": 5.410356375268656,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024929584789262934,
      "loss": 5.1789,
      "mean_token_accuracy": 0.22573791444301605,
      "num_tokens": 105865554.0,
      "step": 54125
    },
    {
      "entropy": 5.8865210056304935,
      "epoch": 5.410856200329884,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024926039276736524,
      "loss": 5.2089,
      "mean_token_accuracy": 0.21908122450113296,
      "num_tokens": 105875207.0,
      "step": 54130
    },
    {
      "entropy": 5.741619491577149,
      "epoch": 5.411356025391113,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002492249382897029,
      "loss": 5.1386,
      "mean_token_accuracy": 0.2285382255911827,
      "num_tokens": 105885368.0,
      "step": 54135
    },
    {
      "entropy": 5.8277916431427,
      "epoch": 5.4118558504523415,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002491894844605346,
      "loss": 5.2399,
      "mean_token_accuracy": 0.2163827136158943,
      "num_tokens": 105895340.0,
      "step": 54140
    },
    {
      "entropy": 5.796387434005737,
      "epoch": 5.41235567551357,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024915403128075215,
      "loss": 5.1688,
      "mean_token_accuracy": 0.21877461224794387,
      "num_tokens": 105904127.0,
      "step": 54145
    },
    {
      "entropy": 5.803868818283081,
      "epoch": 5.412855500574799,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024911857875124777,
      "loss": 5.1784,
      "mean_token_accuracy": 0.2256171226501465,
      "num_tokens": 105913614.0,
      "step": 54150
    },
    {
      "entropy": 5.69631142616272,
      "epoch": 5.413355325636028,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002490831268729132,
      "loss": 5.042,
      "mean_token_accuracy": 0.2390824094414711,
      "num_tokens": 105924031.0,
      "step": 54155
    },
    {
      "entropy": 5.7448944568634035,
      "epoch": 5.413855150697256,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024904767564664043,
      "loss": 5.1357,
      "mean_token_accuracy": 0.22236722111701965,
      "num_tokens": 105934138.0,
      "step": 54160
    },
    {
      "entropy": 5.83246488571167,
      "epoch": 5.414354975758484,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002490122250733215,
      "loss": 5.196,
      "mean_token_accuracy": 0.2248922348022461,
      "num_tokens": 105943988.0,
      "step": 54165
    },
    {
      "entropy": 5.693667697906494,
      "epoch": 5.414854800819713,
      "grad_norm": 1.25,
      "learning_rate": 0.00024897677515384826,
      "loss": 4.9861,
      "mean_token_accuracy": 0.24595885574817658,
      "num_tokens": 105954359.0,
      "step": 54170
    },
    {
      "entropy": 5.797590684890747,
      "epoch": 5.4153546258809415,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002489413258891127,
      "loss": 5.1871,
      "mean_token_accuracy": 0.2220962554216385,
      "num_tokens": 105962961.0,
      "step": 54175
    },
    {
      "entropy": 5.742543458938599,
      "epoch": 5.41585445094217,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024890587728000667,
      "loss": 5.0468,
      "mean_token_accuracy": 0.23167920708656312,
      "num_tokens": 105971603.0,
      "step": 54180
    },
    {
      "entropy": 5.781041097640991,
      "epoch": 5.416354276003399,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002488704293274221,
      "loss": 5.1195,
      "mean_token_accuracy": 0.2298632100224495,
      "num_tokens": 105980968.0,
      "step": 54185
    },
    {
      "entropy": 5.754049158096313,
      "epoch": 5.416854101064628,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002488349820322508,
      "loss": 5.198,
      "mean_token_accuracy": 0.21848685890436173,
      "num_tokens": 105991217.0,
      "step": 54190
    },
    {
      "entropy": 5.641528558731079,
      "epoch": 5.417353926125856,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024879953539538456,
      "loss": 4.9089,
      "mean_token_accuracy": 0.253588804602623,
      "num_tokens": 106001127.0,
      "step": 54195
    },
    {
      "entropy": 5.704527854919434,
      "epoch": 5.417853751187084,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024876408941771526,
      "loss": 5.0529,
      "mean_token_accuracy": 0.2340064138174057,
      "num_tokens": 106010475.0,
      "step": 54200
    },
    {
      "entropy": 5.722080707550049,
      "epoch": 5.418353576248313,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024872864410013475,
      "loss": 5.0651,
      "mean_token_accuracy": 0.23009866923093797,
      "num_tokens": 106019368.0,
      "step": 54205
    },
    {
      "entropy": 5.856852912902832,
      "epoch": 5.4188534013095415,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002486931994435348,
      "loss": 5.279,
      "mean_token_accuracy": 0.21677054166793824,
      "num_tokens": 106030647.0,
      "step": 54210
    },
    {
      "entropy": 5.7268272876739506,
      "epoch": 5.41935322637077,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002486577554488072,
      "loss": 5.1174,
      "mean_token_accuracy": 0.22465519458055497,
      "num_tokens": 106040582.0,
      "step": 54215
    },
    {
      "entropy": 5.848059606552124,
      "epoch": 5.419853051431999,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002486223121168437,
      "loss": 5.2614,
      "mean_token_accuracy": 0.2162187233567238,
      "num_tokens": 106050632.0,
      "step": 54220
    },
    {
      "entropy": 5.775995302200317,
      "epoch": 5.420352876493228,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024858686944853603,
      "loss": 5.1482,
      "mean_token_accuracy": 0.231770621240139,
      "num_tokens": 106060671.0,
      "step": 54225
    },
    {
      "entropy": 5.748905897140503,
      "epoch": 5.420852701554456,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002485514274447759,
      "loss": 5.1163,
      "mean_token_accuracy": 0.23077060729265214,
      "num_tokens": 106070079.0,
      "step": 54230
    },
    {
      "entropy": 5.804970455169678,
      "epoch": 5.421352526615684,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024851598610645505,
      "loss": 5.2131,
      "mean_token_accuracy": 0.2243623122572899,
      "num_tokens": 106081493.0,
      "step": 54235
    },
    {
      "entropy": 5.801847743988037,
      "epoch": 5.421852351676913,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002484805454344652,
      "loss": 5.1487,
      "mean_token_accuracy": 0.22492467015981674,
      "num_tokens": 106091536.0,
      "step": 54240
    },
    {
      "entropy": 5.837334156036377,
      "epoch": 5.4223521767381415,
      "grad_norm": 1.140625,
      "learning_rate": 0.000248445105429698,
      "loss": 5.1832,
      "mean_token_accuracy": 0.22776661068201065,
      "num_tokens": 106101541.0,
      "step": 54245
    },
    {
      "entropy": 5.756748008728027,
      "epoch": 5.42285200179937,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002484096660930451,
      "loss": 5.1249,
      "mean_token_accuracy": 0.2336085855960846,
      "num_tokens": 106110890.0,
      "step": 54250
    },
    {
      "entropy": 5.6513549327850345,
      "epoch": 5.423351826860599,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024837422742539817,
      "loss": 5.0193,
      "mean_token_accuracy": 0.241683666408062,
      "num_tokens": 106121913.0,
      "step": 54255
    },
    {
      "entropy": 5.826915216445923,
      "epoch": 5.423851651921828,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002483387894276488,
      "loss": 5.1649,
      "mean_token_accuracy": 0.228518944978714,
      "num_tokens": 106132013.0,
      "step": 54260
    },
    {
      "entropy": 5.736591100692749,
      "epoch": 5.424351476983056,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002483033521006886,
      "loss": 5.1272,
      "mean_token_accuracy": 0.22930319458246232,
      "num_tokens": 106141395.0,
      "step": 54265
    },
    {
      "entropy": 5.750751590728759,
      "epoch": 5.424851302044284,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002482679154454093,
      "loss": 5.1658,
      "mean_token_accuracy": 0.2189067170023918,
      "num_tokens": 106150888.0,
      "step": 54270
    },
    {
      "entropy": 5.745218658447266,
      "epoch": 5.425351127105513,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024823247946270227,
      "loss": 5.1751,
      "mean_token_accuracy": 0.2209399327635765,
      "num_tokens": 106159527.0,
      "step": 54275
    },
    {
      "entropy": 5.7918312549591064,
      "epoch": 5.4258509521667415,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002481970441534592,
      "loss": 5.1548,
      "mean_token_accuracy": 0.22054877430200576,
      "num_tokens": 106168361.0,
      "step": 54280
    },
    {
      "entropy": 5.752208948135376,
      "epoch": 5.42635077722797,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024816160951857157,
      "loss": 5.1246,
      "mean_token_accuracy": 0.22104505598545074,
      "num_tokens": 106177942.0,
      "step": 54285
    },
    {
      "entropy": 5.865782976150513,
      "epoch": 5.426850602289199,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002481261755589311,
      "loss": 5.2801,
      "mean_token_accuracy": 0.21524893194437028,
      "num_tokens": 106187704.0,
      "step": 54290
    },
    {
      "entropy": 5.929987382888794,
      "epoch": 5.427350427350428,
      "grad_norm": 1.15625,
      "learning_rate": 0.000248090742275429,
      "loss": 5.2903,
      "mean_token_accuracy": 0.21514782905578614,
      "num_tokens": 106197693.0,
      "step": 54295
    },
    {
      "entropy": 5.8316271781921385,
      "epoch": 5.427850252411656,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002480553096689569,
      "loss": 5.161,
      "mean_token_accuracy": 0.22095453143119811,
      "num_tokens": 106206897.0,
      "step": 54300
    },
    {
      "entropy": 5.745049142837525,
      "epoch": 5.428350077472884,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002480198777404063,
      "loss": 5.1723,
      "mean_token_accuracy": 0.22774586975574493,
      "num_tokens": 106216582.0,
      "step": 54305
    },
    {
      "entropy": 5.7842698097229,
      "epoch": 5.428849902534113,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024798444649066863,
      "loss": 5.2036,
      "mean_token_accuracy": 0.22004162818193435,
      "num_tokens": 106226462.0,
      "step": 54310
    },
    {
      "entropy": 5.727097892761231,
      "epoch": 5.4293497275953415,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002479490159206353,
      "loss": 5.084,
      "mean_token_accuracy": 0.23339140713214873,
      "num_tokens": 106236491.0,
      "step": 54315
    },
    {
      "entropy": 5.7828281879425045,
      "epoch": 5.42984955265657,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002479135860311978,
      "loss": 5.1536,
      "mean_token_accuracy": 0.23191023468971253,
      "num_tokens": 106245985.0,
      "step": 54320
    },
    {
      "entropy": 5.75238265991211,
      "epoch": 5.430349377717799,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002478781568232475,
      "loss": 5.1347,
      "mean_token_accuracy": 0.23144128322601318,
      "num_tokens": 106254843.0,
      "step": 54325
    },
    {
      "entropy": 5.7793666362762455,
      "epoch": 5.430849202779028,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024784272829767576,
      "loss": 5.1821,
      "mean_token_accuracy": 0.22468429207801818,
      "num_tokens": 106264880.0,
      "step": 54330
    },
    {
      "entropy": 5.774009656906128,
      "epoch": 5.431349027840256,
      "grad_norm": 1.234375,
      "learning_rate": 0.000247807300455374,
      "loss": 5.1511,
      "mean_token_accuracy": 0.22881596982479097,
      "num_tokens": 106274081.0,
      "step": 54335
    },
    {
      "entropy": 5.725677585601806,
      "epoch": 5.431848852901484,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002477718732972336,
      "loss": 5.0719,
      "mean_token_accuracy": 0.23674449622631072,
      "num_tokens": 106283653.0,
      "step": 54340
    },
    {
      "entropy": 5.831588506698608,
      "epoch": 5.432348677962713,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024773644682414585,
      "loss": 5.2643,
      "mean_token_accuracy": 0.21589448004961015,
      "num_tokens": 106292582.0,
      "step": 54345
    },
    {
      "entropy": 5.831453514099121,
      "epoch": 5.4328485030239415,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002477010210370021,
      "loss": 5.2224,
      "mean_token_accuracy": 0.22141729891300202,
      "num_tokens": 106303045.0,
      "step": 54350
    },
    {
      "entropy": 5.771727275848389,
      "epoch": 5.43334832808517,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024766559593669365,
      "loss": 5.1071,
      "mean_token_accuracy": 0.229828342795372,
      "num_tokens": 106312140.0,
      "step": 54355
    },
    {
      "entropy": 5.778910970687866,
      "epoch": 5.433848153146399,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002476301715241118,
      "loss": 5.1699,
      "mean_token_accuracy": 0.22428708225488664,
      "num_tokens": 106322249.0,
      "step": 54360
    },
    {
      "entropy": 5.806325244903564,
      "epoch": 5.434347978207628,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024759474780014767,
      "loss": 5.2039,
      "mean_token_accuracy": 0.21956128478050232,
      "num_tokens": 106332057.0,
      "step": 54365
    },
    {
      "entropy": 5.73859658241272,
      "epoch": 5.434847803268856,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024755932476569275,
      "loss": 5.0906,
      "mean_token_accuracy": 0.2373790130019188,
      "num_tokens": 106342750.0,
      "step": 54370
    },
    {
      "entropy": 5.804271507263183,
      "epoch": 5.435347628330084,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002475239024216381,
      "loss": 5.2472,
      "mean_token_accuracy": 0.21691899597644806,
      "num_tokens": 106353439.0,
      "step": 54375
    },
    {
      "entropy": 5.770735502243042,
      "epoch": 5.435847453391313,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000247488480768875,
      "loss": 5.2116,
      "mean_token_accuracy": 0.2202276334166527,
      "num_tokens": 106363863.0,
      "step": 54380
    },
    {
      "entropy": 5.6706154346466064,
      "epoch": 5.4363472784525415,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002474530598082946,
      "loss": 5.0095,
      "mean_token_accuracy": 0.23626244068145752,
      "num_tokens": 106373031.0,
      "step": 54385
    },
    {
      "entropy": 5.78821210861206,
      "epoch": 5.43684710351377,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024741763954078816,
      "loss": 5.2068,
      "mean_token_accuracy": 0.21990958899259566,
      "num_tokens": 106382008.0,
      "step": 54390
    },
    {
      "entropy": 5.724336767196656,
      "epoch": 5.437346928574999,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002473822199672468,
      "loss": 5.0894,
      "mean_token_accuracy": 0.23254686444997788,
      "num_tokens": 106391758.0,
      "step": 54395
    },
    {
      "entropy": 5.90145788192749,
      "epoch": 5.437846753636228,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002473468010885617,
      "loss": 5.311,
      "mean_token_accuracy": 0.21804715245962142,
      "num_tokens": 106402274.0,
      "step": 54400
    },
    {
      "entropy": 5.808587265014649,
      "epoch": 5.438346578697456,
      "grad_norm": 1.296875,
      "learning_rate": 0.000247311382905624,
      "loss": 5.1259,
      "mean_token_accuracy": 0.2325017884373665,
      "num_tokens": 106411879.0,
      "step": 54405
    },
    {
      "entropy": 5.712805986404419,
      "epoch": 5.438846403758684,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024727596541932476,
      "loss": 5.1098,
      "mean_token_accuracy": 0.24252612590789796,
      "num_tokens": 106422529.0,
      "step": 54410
    },
    {
      "entropy": 5.735489416122436,
      "epoch": 5.439346228819913,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002472405486305551,
      "loss": 5.1006,
      "mean_token_accuracy": 0.2420354202389717,
      "num_tokens": 106433364.0,
      "step": 54415
    },
    {
      "entropy": 5.8182135105133055,
      "epoch": 5.4398460538811415,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024720513254020604,
      "loss": 5.1432,
      "mean_token_accuracy": 0.2317993387579918,
      "num_tokens": 106442631.0,
      "step": 54420
    },
    {
      "entropy": 5.723421716690064,
      "epoch": 5.44034587894237,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002471697171491687,
      "loss": 4.9945,
      "mean_token_accuracy": 0.2373335212469101,
      "num_tokens": 106451013.0,
      "step": 54425
    },
    {
      "entropy": 5.795549392700195,
      "epoch": 5.440845704003599,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024713430245833424,
      "loss": 5.1545,
      "mean_token_accuracy": 0.2202114537358284,
      "num_tokens": 106460660.0,
      "step": 54430
    },
    {
      "entropy": 5.771597957611084,
      "epoch": 5.441345529064828,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024709888846859347,
      "loss": 5.1853,
      "mean_token_accuracy": 0.2257935255765915,
      "num_tokens": 106471973.0,
      "step": 54435
    },
    {
      "entropy": 5.815413570404052,
      "epoch": 5.441845354126055,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002470634751808375,
      "loss": 5.2507,
      "mean_token_accuracy": 0.21761596351861953,
      "num_tokens": 106482093.0,
      "step": 54440
    },
    {
      "entropy": 5.773063993453979,
      "epoch": 5.442345179187284,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002470280625959573,
      "loss": 5.1342,
      "mean_token_accuracy": 0.2331322804093361,
      "num_tokens": 106491954.0,
      "step": 54445
    },
    {
      "entropy": 5.787621116638183,
      "epoch": 5.442845004248513,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002469926507148438,
      "loss": 5.1624,
      "mean_token_accuracy": 0.2227422058582306,
      "num_tokens": 106501464.0,
      "step": 54450
    },
    {
      "entropy": 5.792415380477905,
      "epoch": 5.4433448293097415,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000246957239538388,
      "loss": 5.2192,
      "mean_token_accuracy": 0.22220133244991302,
      "num_tokens": 106510963.0,
      "step": 54455
    },
    {
      "entropy": 5.794414901733399,
      "epoch": 5.44384465437097,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002469218290674809,
      "loss": 5.1593,
      "mean_token_accuracy": 0.22768452167510986,
      "num_tokens": 106520341.0,
      "step": 54460
    },
    {
      "entropy": 5.790568017959595,
      "epoch": 5.444344479432199,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002468864193030133,
      "loss": 5.1291,
      "mean_token_accuracy": 0.23453871607780458,
      "num_tokens": 106529110.0,
      "step": 54465
    },
    {
      "entropy": 5.818732404708863,
      "epoch": 5.444844304493428,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002468510102458762,
      "loss": 5.1898,
      "mean_token_accuracy": 0.21944183707237244,
      "num_tokens": 106538552.0,
      "step": 54470
    },
    {
      "entropy": 5.782999420166016,
      "epoch": 5.445344129554655,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024681560189696044,
      "loss": 5.1994,
      "mean_token_accuracy": 0.2206276088953018,
      "num_tokens": 106548204.0,
      "step": 54475
    },
    {
      "entropy": 5.829898166656494,
      "epoch": 5.445843954615884,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024678019425715685,
      "loss": 5.2564,
      "mean_token_accuracy": 0.21910861134529114,
      "num_tokens": 106558111.0,
      "step": 54480
    },
    {
      "entropy": 5.7195885181427,
      "epoch": 5.446343779677113,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002467447873273564,
      "loss": 5.1024,
      "mean_token_accuracy": 0.2348496824502945,
      "num_tokens": 106566932.0,
      "step": 54485
    },
    {
      "entropy": 5.827306079864502,
      "epoch": 5.4468436047383415,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024670938110844966,
      "loss": 5.2489,
      "mean_token_accuracy": 0.21975573152303696,
      "num_tokens": 106576505.0,
      "step": 54490
    },
    {
      "entropy": 5.80193567276001,
      "epoch": 5.44734342979957,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024667397560132774,
      "loss": 5.1726,
      "mean_token_accuracy": 0.22386716902256013,
      "num_tokens": 106585945.0,
      "step": 54495
    },
    {
      "entropy": 5.747617578506469,
      "epoch": 5.447843254860799,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024663857080688134,
      "loss": 5.105,
      "mean_token_accuracy": 0.2265584260225296,
      "num_tokens": 106595775.0,
      "step": 54500
    },
    {
      "entropy": 5.697622442245484,
      "epoch": 5.448343079922028,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024660316672600114,
      "loss": 5.0704,
      "mean_token_accuracy": 0.2428264394402504,
      "num_tokens": 106604784.0,
      "step": 54505
    },
    {
      "entropy": 5.788017225265503,
      "epoch": 5.448842904983255,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002465677633595779,
      "loss": 5.1251,
      "mean_token_accuracy": 0.23046947717666627,
      "num_tokens": 106614214.0,
      "step": 54510
    },
    {
      "entropy": 5.77395920753479,
      "epoch": 5.449342730044484,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002465323607085025,
      "loss": 5.1793,
      "mean_token_accuracy": 0.22521984875202178,
      "num_tokens": 106623678.0,
      "step": 54515
    },
    {
      "entropy": 5.747599363327026,
      "epoch": 5.449842555105713,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024649695877366557,
      "loss": 5.1009,
      "mean_token_accuracy": 0.22374343276023864,
      "num_tokens": 106633911.0,
      "step": 54520
    },
    {
      "entropy": 5.747092437744141,
      "epoch": 5.4503423801669415,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002464615575559578,
      "loss": 5.1261,
      "mean_token_accuracy": 0.22607556432485582,
      "num_tokens": 106641897.0,
      "step": 54525
    },
    {
      "entropy": 5.863976669311524,
      "epoch": 5.45084220522817,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002464261570562699,
      "loss": 5.202,
      "mean_token_accuracy": 0.22118854969739915,
      "num_tokens": 106651770.0,
      "step": 54530
    },
    {
      "entropy": 5.778400230407715,
      "epoch": 5.451342030289399,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024639075727549256,
      "loss": 5.157,
      "mean_token_accuracy": 0.2209418684244156,
      "num_tokens": 106661993.0,
      "step": 54535
    },
    {
      "entropy": 5.701962423324585,
      "epoch": 5.451841855350628,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002463553582145164,
      "loss": 5.1161,
      "mean_token_accuracy": 0.23205879032611848,
      "num_tokens": 106672565.0,
      "step": 54540
    },
    {
      "entropy": 5.720036411285401,
      "epoch": 5.452341680411855,
      "grad_norm": 1.296875,
      "learning_rate": 0.000246319959874232,
      "loss": 4.9983,
      "mean_token_accuracy": 0.23656755238771437,
      "num_tokens": 106681698.0,
      "step": 54545
    },
    {
      "entropy": 5.835110092163086,
      "epoch": 5.452841505473084,
      "grad_norm": 1.09375,
      "learning_rate": 0.00024628456225553005,
      "loss": 5.2674,
      "mean_token_accuracy": 0.21309762299060822,
      "num_tokens": 106691488.0,
      "step": 54550
    },
    {
      "entropy": 5.764057683944702,
      "epoch": 5.453341330534313,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024624916535930116,
      "loss": 5.1068,
      "mean_token_accuracy": 0.22896192818880082,
      "num_tokens": 106700418.0,
      "step": 54555
    },
    {
      "entropy": 5.70787501335144,
      "epoch": 5.4538411555955415,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024621376918643585,
      "loss": 5.1176,
      "mean_token_accuracy": 0.2278638154268265,
      "num_tokens": 106709067.0,
      "step": 54560
    },
    {
      "entropy": 5.746152782440186,
      "epoch": 5.45434098065677,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002461783737378247,
      "loss": 5.0922,
      "mean_token_accuracy": 0.22908540666103364,
      "num_tokens": 106719047.0,
      "step": 54565
    },
    {
      "entropy": 5.822819805145263,
      "epoch": 5.454840805717999,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002461429790143583,
      "loss": 5.1926,
      "mean_token_accuracy": 0.22699822038412093,
      "num_tokens": 106729780.0,
      "step": 54570
    },
    {
      "entropy": 5.740262460708618,
      "epoch": 5.455340630779228,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002461075850169271,
      "loss": 5.0476,
      "mean_token_accuracy": 0.23768712729215621,
      "num_tokens": 106738591.0,
      "step": 54575
    },
    {
      "entropy": 5.720627641677856,
      "epoch": 5.455840455840455,
      "grad_norm": 1.25,
      "learning_rate": 0.0002460721917464216,
      "loss": 5.1149,
      "mean_token_accuracy": 0.22819972932338714,
      "num_tokens": 106747679.0,
      "step": 54580
    },
    {
      "entropy": 5.783532428741455,
      "epoch": 5.456340280901684,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024603679920373237,
      "loss": 5.1357,
      "mean_token_accuracy": 0.22854113727807998,
      "num_tokens": 106756625.0,
      "step": 54585
    },
    {
      "entropy": 5.777715826034546,
      "epoch": 5.456840105962913,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024600140738974973,
      "loss": 5.1586,
      "mean_token_accuracy": 0.2264877066016197,
      "num_tokens": 106766126.0,
      "step": 54590
    },
    {
      "entropy": 5.873045206069946,
      "epoch": 5.4573399310241415,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024596601630536425,
      "loss": 5.3469,
      "mean_token_accuracy": 0.20882431268692017,
      "num_tokens": 106776252.0,
      "step": 54595
    },
    {
      "entropy": 5.8120424270629885,
      "epoch": 5.45783975608537,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024593062595146633,
      "loss": 5.1663,
      "mean_token_accuracy": 0.21491728276014327,
      "num_tokens": 106785171.0,
      "step": 54600
    },
    {
      "entropy": 5.74042010307312,
      "epoch": 5.458339581146599,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002458952363289464,
      "loss": 5.0225,
      "mean_token_accuracy": 0.23588740527629853,
      "num_tokens": 106794738.0,
      "step": 54605
    },
    {
      "entropy": 5.797840595245361,
      "epoch": 5.458839406207828,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024585984743869476,
      "loss": 5.1715,
      "mean_token_accuracy": 0.22165420949459075,
      "num_tokens": 106803976.0,
      "step": 54610
    },
    {
      "entropy": 5.759942674636841,
      "epoch": 5.459339231269055,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002458244592816019,
      "loss": 5.1001,
      "mean_token_accuracy": 0.23631076663732528,
      "num_tokens": 106813091.0,
      "step": 54615
    },
    {
      "entropy": 5.879911851882935,
      "epoch": 5.459839056330284,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002457890718585582,
      "loss": 5.31,
      "mean_token_accuracy": 0.20965418219566345,
      "num_tokens": 106822834.0,
      "step": 54620
    },
    {
      "entropy": 5.75465350151062,
      "epoch": 5.460338881391513,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002457536851704539,
      "loss": 5.1317,
      "mean_token_accuracy": 0.23047349750995635,
      "num_tokens": 106832960.0,
      "step": 54625
    },
    {
      "entropy": 5.8340569019317625,
      "epoch": 5.4608387064527415,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024571829921817934,
      "loss": 5.2191,
      "mean_token_accuracy": 0.22042547911405563,
      "num_tokens": 106842060.0,
      "step": 54630
    },
    {
      "entropy": 5.71514892578125,
      "epoch": 5.46133853151397,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024568291400262483,
      "loss": 5.0918,
      "mean_token_accuracy": 0.2313172146677971,
      "num_tokens": 106852265.0,
      "step": 54635
    },
    {
      "entropy": 5.685608196258545,
      "epoch": 5.461838356575199,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002456475295246807,
      "loss": 5.0839,
      "mean_token_accuracy": 0.2323439225554466,
      "num_tokens": 106863577.0,
      "step": 54640
    },
    {
      "entropy": 5.849469232559204,
      "epoch": 5.462338181636428,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024561214578523716,
      "loss": 5.1772,
      "mean_token_accuracy": 0.2206227019429207,
      "num_tokens": 106872187.0,
      "step": 54645
    },
    {
      "entropy": 5.889171695709228,
      "epoch": 5.462838006697655,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024557676278518447,
      "loss": 5.2272,
      "mean_token_accuracy": 0.21564674079418183,
      "num_tokens": 106882489.0,
      "step": 54650
    },
    {
      "entropy": 5.776342344284058,
      "epoch": 5.463337831758884,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002455413805254129,
      "loss": 5.185,
      "mean_token_accuracy": 0.22395627796649933,
      "num_tokens": 106891463.0,
      "step": 54655
    },
    {
      "entropy": 5.693857049942016,
      "epoch": 5.463837656820113,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002455059990068125,
      "loss": 5.0953,
      "mean_token_accuracy": 0.23029527366161345,
      "num_tokens": 106901239.0,
      "step": 54660
    },
    {
      "entropy": 5.7401000499725345,
      "epoch": 5.4643374818813415,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002454706182302737,
      "loss": 5.173,
      "mean_token_accuracy": 0.2301143988966942,
      "num_tokens": 106910094.0,
      "step": 54665
    },
    {
      "entropy": 5.712040853500366,
      "epoch": 5.46483730694257,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024543523819668645,
      "loss": 4.9814,
      "mean_token_accuracy": 0.2337876632809639,
      "num_tokens": 106919715.0,
      "step": 54670
    },
    {
      "entropy": 5.803408098220825,
      "epoch": 5.465337132003799,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000245399858906941,
      "loss": 5.2122,
      "mean_token_accuracy": 0.22074911147356033,
      "num_tokens": 106928901.0,
      "step": 54675
    },
    {
      "entropy": 5.774859380722046,
      "epoch": 5.465836957065028,
      "grad_norm": 1.125,
      "learning_rate": 0.0002453644803619275,
      "loss": 5.1427,
      "mean_token_accuracy": 0.23337201178073883,
      "num_tokens": 106939157.0,
      "step": 54680
    },
    {
      "entropy": 5.81385498046875,
      "epoch": 5.466336782126255,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024532910256253607,
      "loss": 5.1491,
      "mean_token_accuracy": 0.22635643184185028,
      "num_tokens": 106949039.0,
      "step": 54685
    },
    {
      "entropy": 5.767543172836303,
      "epoch": 5.466836607187484,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002452937255096568,
      "loss": 5.1168,
      "mean_token_accuracy": 0.22873904556035995,
      "num_tokens": 106958824.0,
      "step": 54690
    },
    {
      "entropy": 5.802433776855469,
      "epoch": 5.467336432248713,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002452583492041797,
      "loss": 5.2644,
      "mean_token_accuracy": 0.21510399878025055,
      "num_tokens": 106967898.0,
      "step": 54695
    },
    {
      "entropy": 5.761160469055175,
      "epoch": 5.4678362573099415,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002452229736469949,
      "loss": 5.1135,
      "mean_token_accuracy": 0.23200565427541733,
      "num_tokens": 106977475.0,
      "step": 54700
    },
    {
      "entropy": 5.852419424057007,
      "epoch": 5.46833608237117,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024518759883899245,
      "loss": 5.3265,
      "mean_token_accuracy": 0.213639472424984,
      "num_tokens": 106987171.0,
      "step": 54705
    },
    {
      "entropy": 5.766679286956787,
      "epoch": 5.468835907432399,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024515222478106236,
      "loss": 5.129,
      "mean_token_accuracy": 0.22859131991863252,
      "num_tokens": 106997684.0,
      "step": 54710
    },
    {
      "entropy": 5.803356027603149,
      "epoch": 5.469335732493628,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002451168514740946,
      "loss": 5.2532,
      "mean_token_accuracy": 0.22383484542369841,
      "num_tokens": 107008070.0,
      "step": 54715
    },
    {
      "entropy": 5.7454505443573,
      "epoch": 5.469835557554855,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002450814789189792,
      "loss": 5.1126,
      "mean_token_accuracy": 0.23166966587305068,
      "num_tokens": 107018098.0,
      "step": 54720
    },
    {
      "entropy": 5.807083511352539,
      "epoch": 5.470335382616084,
      "grad_norm": 1.25,
      "learning_rate": 0.000245046107116606,
      "loss": 5.2015,
      "mean_token_accuracy": 0.22673196643590926,
      "num_tokens": 107026320.0,
      "step": 54725
    },
    {
      "entropy": 5.870071029663086,
      "epoch": 5.470835207677313,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002450107360678651,
      "loss": 5.2557,
      "mean_token_accuracy": 0.21071226000785828,
      "num_tokens": 107035639.0,
      "step": 54730
    },
    {
      "entropy": 5.799672317504883,
      "epoch": 5.4713350327385415,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024497536577364627,
      "loss": 5.2327,
      "mean_token_accuracy": 0.21835507303476334,
      "num_tokens": 107045361.0,
      "step": 54735
    },
    {
      "entropy": 5.927658653259277,
      "epoch": 5.47183485779977,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002449399962348396,
      "loss": 5.3321,
      "mean_token_accuracy": 0.21037240773439408,
      "num_tokens": 107055334.0,
      "step": 54740
    },
    {
      "entropy": 5.717775869369507,
      "epoch": 5.472334682860999,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002449046274523348,
      "loss": 5.1011,
      "mean_token_accuracy": 0.22647431939840318,
      "num_tokens": 107065223.0,
      "step": 54745
    },
    {
      "entropy": 5.816263341903687,
      "epoch": 5.472834507922228,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024486925942702195,
      "loss": 5.2147,
      "mean_token_accuracy": 0.2246063992381096,
      "num_tokens": 107073417.0,
      "step": 54750
    },
    {
      "entropy": 5.827822637557984,
      "epoch": 5.473334332983455,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002448338921597907,
      "loss": 5.2403,
      "mean_token_accuracy": 0.21226340681314468,
      "num_tokens": 107082637.0,
      "step": 54755
    },
    {
      "entropy": 5.763302040100098,
      "epoch": 5.473834158044684,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024479852565153105,
      "loss": 5.0873,
      "mean_token_accuracy": 0.23340083956718444,
      "num_tokens": 107092221.0,
      "step": 54760
    },
    {
      "entropy": 5.740342617034912,
      "epoch": 5.474333983105913,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002447631599031326,
      "loss": 5.1088,
      "mean_token_accuracy": 0.2319633334875107,
      "num_tokens": 107102475.0,
      "step": 54765
    },
    {
      "entropy": 5.736119127273559,
      "epoch": 5.4748338081671415,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024472779491548535,
      "loss": 5.1694,
      "mean_token_accuracy": 0.23195431530475616,
      "num_tokens": 107111681.0,
      "step": 54770
    },
    {
      "entropy": 5.78276219367981,
      "epoch": 5.47533363322837,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024469243068947895,
      "loss": 5.18,
      "mean_token_accuracy": 0.22137493938207625,
      "num_tokens": 107120688.0,
      "step": 54775
    },
    {
      "entropy": 5.7548424243927006,
      "epoch": 5.475833458289599,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024465706722600326,
      "loss": 5.0488,
      "mean_token_accuracy": 0.23177414983510972,
      "num_tokens": 107130733.0,
      "step": 54780
    },
    {
      "entropy": 5.716497373580933,
      "epoch": 5.476333283350828,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024462170452594793,
      "loss": 5.0709,
      "mean_token_accuracy": 0.23280685842037202,
      "num_tokens": 107140084.0,
      "step": 54785
    },
    {
      "entropy": 5.73691725730896,
      "epoch": 5.476833108412055,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024458634259020266,
      "loss": 5.0324,
      "mean_token_accuracy": 0.23359175771474838,
      "num_tokens": 107148446.0,
      "step": 54790
    },
    {
      "entropy": 5.799814224243164,
      "epoch": 5.477332933473284,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002445509814196572,
      "loss": 5.2561,
      "mean_token_accuracy": 0.22049383223056793,
      "num_tokens": 107157556.0,
      "step": 54795
    },
    {
      "entropy": 5.840521764755249,
      "epoch": 5.477832758534513,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002445156210152012,
      "loss": 5.2578,
      "mean_token_accuracy": 0.2216998189687729,
      "num_tokens": 107166746.0,
      "step": 54800
    },
    {
      "entropy": 5.689421367645264,
      "epoch": 5.4783325835957415,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024448026137772436,
      "loss": 5.1307,
      "mean_token_accuracy": 0.23033791780471802,
      "num_tokens": 107176162.0,
      "step": 54805
    },
    {
      "entropy": 5.717631196975708,
      "epoch": 5.47883240865697,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002444449025081162,
      "loss": 5.0304,
      "mean_token_accuracy": 0.23749254792928695,
      "num_tokens": 107184931.0,
      "step": 54810
    },
    {
      "entropy": 5.645134687423706,
      "epoch": 5.479332233718199,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002444095444072665,
      "loss": 5.0041,
      "mean_token_accuracy": 0.24598631560802459,
      "num_tokens": 107194554.0,
      "step": 54815
    },
    {
      "entropy": 5.780370759963989,
      "epoch": 5.479832058779428,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002443741870760649,
      "loss": 5.1953,
      "mean_token_accuracy": 0.22417883425951005,
      "num_tokens": 107204347.0,
      "step": 54820
    },
    {
      "entropy": 5.80708417892456,
      "epoch": 5.480331883840655,
      "grad_norm": 1.25,
      "learning_rate": 0.0002443388305154007,
      "loss": 5.2118,
      "mean_token_accuracy": 0.2219604179263115,
      "num_tokens": 107214652.0,
      "step": 54825
    },
    {
      "entropy": 5.843624114990234,
      "epoch": 5.480831708901884,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024430347472616373,
      "loss": 5.2441,
      "mean_token_accuracy": 0.21486023664474488,
      "num_tokens": 107225189.0,
      "step": 54830
    },
    {
      "entropy": 5.717123699188233,
      "epoch": 5.481331533963113,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024426811970924335,
      "loss": 5.0979,
      "mean_token_accuracy": 0.23464725464582442,
      "num_tokens": 107234783.0,
      "step": 54835
    },
    {
      "entropy": 5.729621267318725,
      "epoch": 5.4818313590243415,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002442327654655292,
      "loss": 5.1643,
      "mean_token_accuracy": 0.23466860353946686,
      "num_tokens": 107244934.0,
      "step": 54840
    },
    {
      "entropy": 5.7612433433532715,
      "epoch": 5.48233118408557,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002441974119959108,
      "loss": 5.0993,
      "mean_token_accuracy": 0.2297697514295578,
      "num_tokens": 107253996.0,
      "step": 54845
    },
    {
      "entropy": 5.816912841796875,
      "epoch": 5.482831009146799,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024416205930127767,
      "loss": 5.1503,
      "mean_token_accuracy": 0.22840213924646377,
      "num_tokens": 107263853.0,
      "step": 54850
    },
    {
      "entropy": 5.82734694480896,
      "epoch": 5.483330834208028,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024412670738251902,
      "loss": 5.2121,
      "mean_token_accuracy": 0.2213237926363945,
      "num_tokens": 107273403.0,
      "step": 54855
    },
    {
      "entropy": 5.690635442733765,
      "epoch": 5.483830659269255,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024409135624052454,
      "loss": 5.0746,
      "mean_token_accuracy": 0.24194584041833878,
      "num_tokens": 107284132.0,
      "step": 54860
    },
    {
      "entropy": 5.79952392578125,
      "epoch": 5.484330484330484,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024405600587618358,
      "loss": 5.1585,
      "mean_token_accuracy": 0.22192201614379883,
      "num_tokens": 107294906.0,
      "step": 54865
    },
    {
      "entropy": 5.872737407684326,
      "epoch": 5.484830309391713,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024402065629038555,
      "loss": 5.253,
      "mean_token_accuracy": 0.21104557663202286,
      "num_tokens": 107305208.0,
      "step": 54870
    },
    {
      "entropy": 5.8670250415802006,
      "epoch": 5.4853301344529415,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024398530748401986,
      "loss": 5.2359,
      "mean_token_accuracy": 0.2152193918824196,
      "num_tokens": 107314521.0,
      "step": 54875
    },
    {
      "entropy": 5.707924509048462,
      "epoch": 5.48582995951417,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024394995945797583,
      "loss": 5.0416,
      "mean_token_accuracy": 0.23008397966623306,
      "num_tokens": 107323184.0,
      "step": 54880
    },
    {
      "entropy": 5.765757417678833,
      "epoch": 5.486329784575399,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002439146122131428,
      "loss": 5.1813,
      "mean_token_accuracy": 0.21799763292074203,
      "num_tokens": 107332786.0,
      "step": 54885
    },
    {
      "entropy": 5.841827249526977,
      "epoch": 5.486829609636628,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002438792657504102,
      "loss": 5.2338,
      "mean_token_accuracy": 0.22122581154108048,
      "num_tokens": 107341545.0,
      "step": 54890
    },
    {
      "entropy": 5.74621844291687,
      "epoch": 5.487329434697855,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024384392007066713,
      "loss": 5.1195,
      "mean_token_accuracy": 0.23066159784793855,
      "num_tokens": 107351504.0,
      "step": 54895
    },
    {
      "entropy": 5.737652778625488,
      "epoch": 5.487829259759084,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024380857517480314,
      "loss": 5.1192,
      "mean_token_accuracy": 0.22569798976182937,
      "num_tokens": 107361151.0,
      "step": 54900
    },
    {
      "entropy": 5.789899492263794,
      "epoch": 5.488329084820313,
      "grad_norm": 1.375,
      "learning_rate": 0.00024377323106370736,
      "loss": 5.1736,
      "mean_token_accuracy": 0.2378853052854538,
      "num_tokens": 107370209.0,
      "step": 54905
    },
    {
      "entropy": 5.77870626449585,
      "epoch": 5.4888289098815415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024373788773826905,
      "loss": 5.1595,
      "mean_token_accuracy": 0.2241970717906952,
      "num_tokens": 107379446.0,
      "step": 54910
    },
    {
      "entropy": 5.8956533908844,
      "epoch": 5.48932873494277,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002437025451993774,
      "loss": 5.2842,
      "mean_token_accuracy": 0.21840205043554306,
      "num_tokens": 107389072.0,
      "step": 54915
    },
    {
      "entropy": 5.864241743087769,
      "epoch": 5.489828560003999,
      "grad_norm": 1.25,
      "learning_rate": 0.00024366720344792177,
      "loss": 5.1849,
      "mean_token_accuracy": 0.2240106612443924,
      "num_tokens": 107399189.0,
      "step": 54920
    },
    {
      "entropy": 5.778935813903809,
      "epoch": 5.490328385065228,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002436318624847911,
      "loss": 5.1729,
      "mean_token_accuracy": 0.2238052800297737,
      "num_tokens": 107407999.0,
      "step": 54925
    },
    {
      "entropy": 5.704281568527222,
      "epoch": 5.490828210126455,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024359652231087476,
      "loss": 5.1632,
      "mean_token_accuracy": 0.23236638456583023,
      "num_tokens": 107418044.0,
      "step": 54930
    },
    {
      "entropy": 5.594397354125976,
      "epoch": 5.491328035187684,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002435611829270618,
      "loss": 4.9711,
      "mean_token_accuracy": 0.2357019767165184,
      "num_tokens": 107428437.0,
      "step": 54935
    },
    {
      "entropy": 5.8134605884552,
      "epoch": 5.491827860248913,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002435258443342414,
      "loss": 5.1903,
      "mean_token_accuracy": 0.22812223732471465,
      "num_tokens": 107437761.0,
      "step": 54940
    },
    {
      "entropy": 5.787441921234131,
      "epoch": 5.4923276853101415,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024349050653330273,
      "loss": 5.1936,
      "mean_token_accuracy": 0.2291065812110901,
      "num_tokens": 107446612.0,
      "step": 54945
    },
    {
      "entropy": 5.814890813827515,
      "epoch": 5.49282751037137,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024345516952513474,
      "loss": 5.1787,
      "mean_token_accuracy": 0.22310760766267776,
      "num_tokens": 107456485.0,
      "step": 54950
    },
    {
      "entropy": 5.809579372406006,
      "epoch": 5.493327335432599,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002434198333106265,
      "loss": 5.1504,
      "mean_token_accuracy": 0.22608141005039215,
      "num_tokens": 107465439.0,
      "step": 54955
    },
    {
      "entropy": 5.822672653198242,
      "epoch": 5.493827160493828,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024338449789066725,
      "loss": 5.2357,
      "mean_token_accuracy": 0.2171955019235611,
      "num_tokens": 107475010.0,
      "step": 54960
    },
    {
      "entropy": 5.714125061035157,
      "epoch": 5.494326985555055,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024334916326614582,
      "loss": 5.0507,
      "mean_token_accuracy": 0.24282897412776946,
      "num_tokens": 107485433.0,
      "step": 54965
    },
    {
      "entropy": 5.7587066173553465,
      "epoch": 5.494826810616284,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024331382943795134,
      "loss": 5.1279,
      "mean_token_accuracy": 0.2282642126083374,
      "num_tokens": 107495809.0,
      "step": 54970
    },
    {
      "entropy": 5.760294198989868,
      "epoch": 5.495326635677513,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00024327849640697275,
      "loss": 5.1369,
      "mean_token_accuracy": 0.23306455165147782,
      "num_tokens": 107506067.0,
      "step": 54975
    },
    {
      "entropy": 5.700497627258301,
      "epoch": 5.4958264607387415,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024324316417409904,
      "loss": 5.0639,
      "mean_token_accuracy": 0.22698360532522202,
      "num_tokens": 107514647.0,
      "step": 54980
    },
    {
      "entropy": 5.771798515319825,
      "epoch": 5.49632628579997,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024320783274021912,
      "loss": 5.233,
      "mean_token_accuracy": 0.21870225220918654,
      "num_tokens": 107524560.0,
      "step": 54985
    },
    {
      "entropy": 5.870864009857177,
      "epoch": 5.496826110861199,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000243172502106222,
      "loss": 5.2089,
      "mean_token_accuracy": 0.21983019709587098,
      "num_tokens": 107534174.0,
      "step": 54990
    },
    {
      "entropy": 5.731451559066772,
      "epoch": 5.497325935922428,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024313717227299647,
      "loss": 5.1297,
      "mean_token_accuracy": 0.23053516894578935,
      "num_tokens": 107543789.0,
      "step": 54995
    },
    {
      "entropy": 5.799831104278565,
      "epoch": 5.497825760983655,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024310184324143153,
      "loss": 5.146,
      "mean_token_accuracy": 0.22595046311616898,
      "num_tokens": 107552575.0,
      "step": 55000
    },
    {
      "entropy": 5.779407072067261,
      "epoch": 5.498325586044884,
      "grad_norm": 1.171875,
      "learning_rate": 0.000243066515012416,
      "loss": 5.121,
      "mean_token_accuracy": 0.22768126875162126,
      "num_tokens": 107562321.0,
      "step": 55005
    },
    {
      "entropy": 5.74169807434082,
      "epoch": 5.498825411106113,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024303118758683872,
      "loss": 5.1018,
      "mean_token_accuracy": 0.23163769245147706,
      "num_tokens": 107572942.0,
      "step": 55010
    },
    {
      "entropy": 5.714355707168579,
      "epoch": 5.4993252361673415,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00024299586096558856,
      "loss": 5.1127,
      "mean_token_accuracy": 0.23291805684566497,
      "num_tokens": 107582807.0,
      "step": 55015
    },
    {
      "entropy": 5.782852268218994,
      "epoch": 5.49982506122857,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024296053514955425,
      "loss": 5.1584,
      "mean_token_accuracy": 0.22753457725048065,
      "num_tokens": 107592173.0,
      "step": 55020
    },
    {
      "entropy": 5.78737587928772,
      "epoch": 5.500324886289799,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024292521013962465,
      "loss": 5.1279,
      "mean_token_accuracy": 0.23594476729631425,
      "num_tokens": 107602574.0,
      "step": 55025
    },
    {
      "entropy": 5.858513259887696,
      "epoch": 5.500824711351028,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024288988593668852,
      "loss": 5.2823,
      "mean_token_accuracy": 0.2186834007501602,
      "num_tokens": 107612308.0,
      "step": 55030
    },
    {
      "entropy": 5.7035534381866455,
      "epoch": 5.501324536412255,
      "grad_norm": 1.25,
      "learning_rate": 0.00024285456254163456,
      "loss": 5.0482,
      "mean_token_accuracy": 0.22878533899784087,
      "num_tokens": 107622555.0,
      "step": 55035
    },
    {
      "entropy": 5.753967142105102,
      "epoch": 5.501824361473484,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002428192399553516,
      "loss": 5.1478,
      "mean_token_accuracy": 0.22794820815324784,
      "num_tokens": 107632023.0,
      "step": 55040
    },
    {
      "entropy": 5.7503303527832035,
      "epoch": 5.502324186534713,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002427839181787283,
      "loss": 5.1503,
      "mean_token_accuracy": 0.2307886615395546,
      "num_tokens": 107642774.0,
      "step": 55045
    },
    {
      "entropy": 5.745674705505371,
      "epoch": 5.5028240115959415,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024274859721265323,
      "loss": 5.1413,
      "mean_token_accuracy": 0.23637868016958236,
      "num_tokens": 107653128.0,
      "step": 55050
    },
    {
      "entropy": 5.86475625038147,
      "epoch": 5.50332383665717,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002427132770580152,
      "loss": 5.2089,
      "mean_token_accuracy": 0.2219042018055916,
      "num_tokens": 107662610.0,
      "step": 55055
    },
    {
      "entropy": 5.751432943344116,
      "epoch": 5.503823661718399,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002426779577157029,
      "loss": 5.1023,
      "mean_token_accuracy": 0.23040810972452164,
      "num_tokens": 107672599.0,
      "step": 55060
    },
    {
      "entropy": 5.7423906326293945,
      "epoch": 5.504323486779628,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002426426391866048,
      "loss": 5.0492,
      "mean_token_accuracy": 0.24107517600059508,
      "num_tokens": 107683293.0,
      "step": 55065
    },
    {
      "entropy": 5.701886701583862,
      "epoch": 5.504823311840855,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024260732147160958,
      "loss": 5.1024,
      "mean_token_accuracy": 0.23002565354108812,
      "num_tokens": 107692999.0,
      "step": 55070
    },
    {
      "entropy": 5.76853928565979,
      "epoch": 5.505323136902084,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024257200457160577,
      "loss": 5.1878,
      "mean_token_accuracy": 0.23091915398836135,
      "num_tokens": 107701903.0,
      "step": 55075
    },
    {
      "entropy": 5.786064529418946,
      "epoch": 5.505822961963313,
      "grad_norm": 1.140625,
      "learning_rate": 0.00024253668848748196,
      "loss": 5.1768,
      "mean_token_accuracy": 0.2259308636188507,
      "num_tokens": 107712268.0,
      "step": 55080
    },
    {
      "entropy": 5.881660270690918,
      "epoch": 5.5063227870245415,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002425013732201268,
      "loss": 5.2397,
      "mean_token_accuracy": 0.21777946650981903,
      "num_tokens": 107722434.0,
      "step": 55085
    },
    {
      "entropy": 5.71565055847168,
      "epoch": 5.50682261208577,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024246605877042865,
      "loss": 5.0657,
      "mean_token_accuracy": 0.24216976165771484,
      "num_tokens": 107734200.0,
      "step": 55090
    },
    {
      "entropy": 5.731837272644043,
      "epoch": 5.507322437146999,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024243074513927615,
      "loss": 5.0795,
      "mean_token_accuracy": 0.23369455933570862,
      "num_tokens": 107743419.0,
      "step": 55095
    },
    {
      "entropy": 5.760212993621826,
      "epoch": 5.507822262208228,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024239543232755766,
      "loss": 5.1704,
      "mean_token_accuracy": 0.22336507439613343,
      "num_tokens": 107752652.0,
      "step": 55100
    },
    {
      "entropy": 5.794733285903931,
      "epoch": 5.508322087269455,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024236012033616177,
      "loss": 5.1105,
      "mean_token_accuracy": 0.23280623108148574,
      "num_tokens": 107762073.0,
      "step": 55105
    },
    {
      "entropy": 5.7783956050872805,
      "epoch": 5.508821912330684,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024232480916597682,
      "loss": 5.1011,
      "mean_token_accuracy": 0.23831462413072585,
      "num_tokens": 107771834.0,
      "step": 55110
    },
    {
      "entropy": 5.806090784072876,
      "epoch": 5.509321737391913,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002422894988178913,
      "loss": 5.2454,
      "mean_token_accuracy": 0.2160439521074295,
      "num_tokens": 107781276.0,
      "step": 55115
    },
    {
      "entropy": 5.769069242477417,
      "epoch": 5.5098215624531415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024225418929279348,
      "loss": 5.1449,
      "mean_token_accuracy": 0.22821730077266694,
      "num_tokens": 107790874.0,
      "step": 55120
    },
    {
      "entropy": 5.776853609085083,
      "epoch": 5.51032138751437,
      "grad_norm": 1.25,
      "learning_rate": 0.00024221888059157187,
      "loss": 5.2056,
      "mean_token_accuracy": 0.22336435616016387,
      "num_tokens": 107800622.0,
      "step": 55125
    },
    {
      "entropy": 5.760470628738403,
      "epoch": 5.510821212575599,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024218357271511492,
      "loss": 5.1427,
      "mean_token_accuracy": 0.2312928855419159,
      "num_tokens": 107810491.0,
      "step": 55130
    },
    {
      "entropy": 5.782981252670288,
      "epoch": 5.511321037636828,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024214826566431075,
      "loss": 5.066,
      "mean_token_accuracy": 0.23484028577804567,
      "num_tokens": 107819696.0,
      "step": 55135
    },
    {
      "entropy": 5.771374320983886,
      "epoch": 5.511820862698055,
      "grad_norm": 1.671875,
      "learning_rate": 0.00024211295944004768,
      "loss": 5.0592,
      "mean_token_accuracy": 0.23775051683187484,
      "num_tokens": 107829746.0,
      "step": 55140
    },
    {
      "entropy": 5.772552251815796,
      "epoch": 5.512320687759284,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024207765404321414,
      "loss": 5.1462,
      "mean_token_accuracy": 0.22624256163835527,
      "num_tokens": 107840759.0,
      "step": 55145
    },
    {
      "entropy": 5.74106273651123,
      "epoch": 5.512820512820513,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024204234947469838,
      "loss": 5.1258,
      "mean_token_accuracy": 0.23029852509498597,
      "num_tokens": 107849592.0,
      "step": 55150
    },
    {
      "entropy": 5.7406166076660154,
      "epoch": 5.5133203378817415,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024200704573538862,
      "loss": 5.1965,
      "mean_token_accuracy": 0.2252981722354889,
      "num_tokens": 107860508.0,
      "step": 55155
    },
    {
      "entropy": 5.842123556137085,
      "epoch": 5.51382016294297,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024197174282617308,
      "loss": 5.2339,
      "mean_token_accuracy": 0.21932739317417144,
      "num_tokens": 107869583.0,
      "step": 55160
    },
    {
      "entropy": 5.882249641418457,
      "epoch": 5.514319988004199,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024193644074793996,
      "loss": 5.2075,
      "mean_token_accuracy": 0.2162019804120064,
      "num_tokens": 107878116.0,
      "step": 55165
    },
    {
      "entropy": 5.7344237804412845,
      "epoch": 5.514819813065428,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024190113950157754,
      "loss": 5.0715,
      "mean_token_accuracy": 0.23743032813072204,
      "num_tokens": 107887616.0,
      "step": 55170
    },
    {
      "entropy": 5.877812719345092,
      "epoch": 5.515319638126655,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024186583908797395,
      "loss": 5.2743,
      "mean_token_accuracy": 0.21210546493530275,
      "num_tokens": 107897525.0,
      "step": 55175
    },
    {
      "entropy": 5.8410890102386475,
      "epoch": 5.515819463187884,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002418305395080172,
      "loss": 5.2435,
      "mean_token_accuracy": 0.21599075943231583,
      "num_tokens": 107909169.0,
      "step": 55180
    },
    {
      "entropy": 5.795981216430664,
      "epoch": 5.516319288249113,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024179524076259562,
      "loss": 5.1379,
      "mean_token_accuracy": 0.2261521950364113,
      "num_tokens": 107919364.0,
      "step": 55185
    },
    {
      "entropy": 5.791911697387695,
      "epoch": 5.5168191133103415,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002417599428525973,
      "loss": 5.2281,
      "mean_token_accuracy": 0.22411417663097383,
      "num_tokens": 107928404.0,
      "step": 55190
    },
    {
      "entropy": 5.707847690582275,
      "epoch": 5.51731893837157,
      "grad_norm": 1.25,
      "learning_rate": 0.00024172464577891018,
      "loss": 5.0594,
      "mean_token_accuracy": 0.24143005311489105,
      "num_tokens": 107937537.0,
      "step": 55195
    },
    {
      "entropy": 5.786009025573731,
      "epoch": 5.517818763432799,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024168934954242255,
      "loss": 5.2217,
      "mean_token_accuracy": 0.21854147166013718,
      "num_tokens": 107947822.0,
      "step": 55200
    },
    {
      "entropy": 5.733794784545898,
      "epoch": 5.518318588494028,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024165405414402214,
      "loss": 5.1813,
      "mean_token_accuracy": 0.22327460795640947,
      "num_tokens": 107956959.0,
      "step": 55205
    },
    {
      "entropy": 5.693645858764649,
      "epoch": 5.518818413555255,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024161875958459724,
      "loss": 5.0823,
      "mean_token_accuracy": 0.23209924548864364,
      "num_tokens": 107967607.0,
      "step": 55210
    },
    {
      "entropy": 5.793099069595337,
      "epoch": 5.519318238616484,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024158346586503572,
      "loss": 5.2128,
      "mean_token_accuracy": 0.22565873563289643,
      "num_tokens": 107977703.0,
      "step": 55215
    },
    {
      "entropy": 5.678117799758911,
      "epoch": 5.519818063677713,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024154817298622562,
      "loss": 5.0168,
      "mean_token_accuracy": 0.22642747759819032,
      "num_tokens": 107987232.0,
      "step": 55220
    },
    {
      "entropy": 5.775062704086304,
      "epoch": 5.5203178887389415,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024151288094905488,
      "loss": 5.0741,
      "mean_token_accuracy": 0.23351979106664658,
      "num_tokens": 107998407.0,
      "step": 55225
    },
    {
      "entropy": 5.773753213882446,
      "epoch": 5.52081771380017,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002414775897544114,
      "loss": 5.1423,
      "mean_token_accuracy": 0.22321850806474686,
      "num_tokens": 108007474.0,
      "step": 55230
    },
    {
      "entropy": 5.729519033432007,
      "epoch": 5.521317538861399,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024144229940318318,
      "loss": 5.0465,
      "mean_token_accuracy": 0.23444878160953522,
      "num_tokens": 108016408.0,
      "step": 55235
    },
    {
      "entropy": 5.744604015350342,
      "epoch": 5.521817363922628,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024140700989625804,
      "loss": 5.1929,
      "mean_token_accuracy": 0.22593776136636734,
      "num_tokens": 108026488.0,
      "step": 55240
    },
    {
      "entropy": 5.698582601547241,
      "epoch": 5.522317188983855,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002413717212345239,
      "loss": 5.0293,
      "mean_token_accuracy": 0.24201792180538179,
      "num_tokens": 108036452.0,
      "step": 55245
    },
    {
      "entropy": 5.80902328491211,
      "epoch": 5.522817014045084,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024133643341886857,
      "loss": 5.1859,
      "mean_token_accuracy": 0.22068763822317122,
      "num_tokens": 108047136.0,
      "step": 55250
    },
    {
      "entropy": 5.656038665771485,
      "epoch": 5.523316839106313,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024130114645017997,
      "loss": 5.0599,
      "mean_token_accuracy": 0.24203566759824752,
      "num_tokens": 108057297.0,
      "step": 55255
    },
    {
      "entropy": 5.731429576873779,
      "epoch": 5.5238166641675415,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024126586032934577,
      "loss": 5.0366,
      "mean_token_accuracy": 0.24135875105857849,
      "num_tokens": 108066923.0,
      "step": 55260
    },
    {
      "entropy": 5.827081346511841,
      "epoch": 5.52431648922877,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024123057505725392,
      "loss": 5.2024,
      "mean_token_accuracy": 0.22454309910535813,
      "num_tokens": 108075867.0,
      "step": 55265
    },
    {
      "entropy": 5.752426958084106,
      "epoch": 5.524816314289999,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024119529063479218,
      "loss": 5.0918,
      "mean_token_accuracy": 0.23638375252485275,
      "num_tokens": 108085095.0,
      "step": 55270
    },
    {
      "entropy": 5.778492593765259,
      "epoch": 5.525316139351228,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024116000706284814,
      "loss": 5.1643,
      "mean_token_accuracy": 0.23252712786197663,
      "num_tokens": 108094107.0,
      "step": 55275
    },
    {
      "entropy": 5.754909896850586,
      "epoch": 5.525815964412455,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024112472434230958,
      "loss": 5.1683,
      "mean_token_accuracy": 0.22252290099859237,
      "num_tokens": 108103645.0,
      "step": 55280
    },
    {
      "entropy": 5.721900367736817,
      "epoch": 5.526315789473684,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024108944247406433,
      "loss": 5.1054,
      "mean_token_accuracy": 0.22778012454509736,
      "num_tokens": 108113959.0,
      "step": 55285
    },
    {
      "entropy": 5.802838706970215,
      "epoch": 5.526815614534913,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024105416145899994,
      "loss": 5.1324,
      "mean_token_accuracy": 0.23515819907188415,
      "num_tokens": 108123593.0,
      "step": 55290
    },
    {
      "entropy": 5.718763399124145,
      "epoch": 5.5273154395961415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002410188812980041,
      "loss": 5.1228,
      "mean_token_accuracy": 0.22281937301158905,
      "num_tokens": 108133138.0,
      "step": 55295
    },
    {
      "entropy": 5.694383335113526,
      "epoch": 5.52781526465737,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024098360199196445,
      "loss": 5.0936,
      "mean_token_accuracy": 0.23312544375658034,
      "num_tokens": 108142798.0,
      "step": 55300
    },
    {
      "entropy": 5.80974817276001,
      "epoch": 5.528315089718599,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002409483235417686,
      "loss": 5.2503,
      "mean_token_accuracy": 0.22126837968826293,
      "num_tokens": 108153194.0,
      "step": 55305
    },
    {
      "entropy": 5.729055690765381,
      "epoch": 5.528814914779827,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002409130459483042,
      "loss": 5.0983,
      "mean_token_accuracy": 0.22856221497058868,
      "num_tokens": 108162803.0,
      "step": 55310
    },
    {
      "entropy": 5.732852458953857,
      "epoch": 5.529314739841055,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002408777692124588,
      "loss": 5.1605,
      "mean_token_accuracy": 0.23254209607839585,
      "num_tokens": 108172938.0,
      "step": 55315
    },
    {
      "entropy": 5.859977912902832,
      "epoch": 5.529814564902284,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00024084249333511992,
      "loss": 5.2158,
      "mean_token_accuracy": 0.2198292136192322,
      "num_tokens": 108182863.0,
      "step": 55320
    },
    {
      "entropy": 5.7008568286895756,
      "epoch": 5.530314389963513,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002408072183171751,
      "loss": 5.0471,
      "mean_token_accuracy": 0.2404790610074997,
      "num_tokens": 108192758.0,
      "step": 55325
    },
    {
      "entropy": 5.806861877441406,
      "epoch": 5.5308142150247415,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024077194415951197,
      "loss": 5.2524,
      "mean_token_accuracy": 0.21755337417125703,
      "num_tokens": 108202771.0,
      "step": 55330
    },
    {
      "entropy": 5.80497989654541,
      "epoch": 5.53131404008597,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024073667086301794,
      "loss": 5.2004,
      "mean_token_accuracy": 0.22740611881017686,
      "num_tokens": 108212626.0,
      "step": 55335
    },
    {
      "entropy": 5.810675764083863,
      "epoch": 5.531813865147199,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024070139842858034,
      "loss": 5.1222,
      "mean_token_accuracy": 0.24163019210100173,
      "num_tokens": 108222583.0,
      "step": 55340
    },
    {
      "entropy": 5.815518140792847,
      "epoch": 5.532313690208427,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024066612685708671,
      "loss": 5.2225,
      "mean_token_accuracy": 0.22211571782827377,
      "num_tokens": 108231953.0,
      "step": 55345
    },
    {
      "entropy": 5.883425903320313,
      "epoch": 5.532813515269655,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024063085614942458,
      "loss": 5.2654,
      "mean_token_accuracy": 0.22027381658554077,
      "num_tokens": 108241471.0,
      "step": 55350
    },
    {
      "entropy": 5.738061332702637,
      "epoch": 5.533313340330884,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002405955863064812,
      "loss": 5.0961,
      "mean_token_accuracy": 0.22957989424467087,
      "num_tokens": 108249654.0,
      "step": 55355
    },
    {
      "entropy": 5.750920534133911,
      "epoch": 5.533813165392113,
      "grad_norm": 1.25,
      "learning_rate": 0.00024056031732914407,
      "loss": 5.1719,
      "mean_token_accuracy": 0.2243879735469818,
      "num_tokens": 108258674.0,
      "step": 55360
    },
    {
      "entropy": 5.7290411472320555,
      "epoch": 5.5343129904533415,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024052504921830042,
      "loss": 5.1443,
      "mean_token_accuracy": 0.23182765692472457,
      "num_tokens": 108267576.0,
      "step": 55365
    },
    {
      "entropy": 5.725677537918091,
      "epoch": 5.53481281551457,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024048978197483773,
      "loss": 5.1342,
      "mean_token_accuracy": 0.22578500509262084,
      "num_tokens": 108276811.0,
      "step": 55370
    },
    {
      "entropy": 5.760149240493774,
      "epoch": 5.535312640575798,
      "grad_norm": 1.25,
      "learning_rate": 0.00024045451559964322,
      "loss": 5.1693,
      "mean_token_accuracy": 0.22341281324625015,
      "num_tokens": 108286489.0,
      "step": 55375
    },
    {
      "entropy": 5.803043413162231,
      "epoch": 5.535812465637027,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002404192500936042,
      "loss": 5.1437,
      "mean_token_accuracy": 0.22856416255235673,
      "num_tokens": 108296115.0,
      "step": 55380
    },
    {
      "entropy": 5.679873132705689,
      "epoch": 5.536312290698255,
      "grad_norm": 1.25,
      "learning_rate": 0.00024038398545760798,
      "loss": 5.0483,
      "mean_token_accuracy": 0.23392025232315064,
      "num_tokens": 108305437.0,
      "step": 55385
    },
    {
      "entropy": 5.672796630859375,
      "epoch": 5.536812115759484,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024034872169254173,
      "loss": 5.043,
      "mean_token_accuracy": 0.23860633820295335,
      "num_tokens": 108315545.0,
      "step": 55390
    },
    {
      "entropy": 5.761216878890991,
      "epoch": 5.537311940820713,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024031345879929278,
      "loss": 5.1067,
      "mean_token_accuracy": 0.2258922889828682,
      "num_tokens": 108325039.0,
      "step": 55395
    },
    {
      "entropy": 5.794339609146118,
      "epoch": 5.5378117658819415,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024027819677874824,
      "loss": 5.1878,
      "mean_token_accuracy": 0.22268156409263612,
      "num_tokens": 108334638.0,
      "step": 55400
    },
    {
      "entropy": 5.8042261600494385,
      "epoch": 5.53831159094317,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002402429356317954,
      "loss": 5.1977,
      "mean_token_accuracy": 0.22923283129930497,
      "num_tokens": 108343750.0,
      "step": 55405
    },
    {
      "entropy": 5.746529245376587,
      "epoch": 5.538811416004398,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024020767535932132,
      "loss": 5.081,
      "mean_token_accuracy": 0.23407482206821442,
      "num_tokens": 108353867.0,
      "step": 55410
    },
    {
      "entropy": 5.727075004577637,
      "epoch": 5.539311241065627,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024017241596221316,
      "loss": 5.0926,
      "mean_token_accuracy": 0.23232416361570357,
      "num_tokens": 108363465.0,
      "step": 55415
    },
    {
      "entropy": 5.846561765670776,
      "epoch": 5.539811066126855,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024013715744135807,
      "loss": 5.3325,
      "mean_token_accuracy": 0.2145149976015091,
      "num_tokens": 108372576.0,
      "step": 55420
    },
    {
      "entropy": 5.741961812973022,
      "epoch": 5.540310891188084,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024010189979764308,
      "loss": 5.1074,
      "mean_token_accuracy": 0.23033884763717652,
      "num_tokens": 108382127.0,
      "step": 55425
    },
    {
      "entropy": 5.825970554351807,
      "epoch": 5.540810716249313,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002400666430319553,
      "loss": 5.1666,
      "mean_token_accuracy": 0.22293677777051926,
      "num_tokens": 108391728.0,
      "step": 55430
    },
    {
      "entropy": 5.7499401569366455,
      "epoch": 5.5413105413105415,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024003138714518186,
      "loss": 5.1237,
      "mean_token_accuracy": 0.23200342208147048,
      "num_tokens": 108400990.0,
      "step": 55435
    },
    {
      "entropy": 5.765198659896851,
      "epoch": 5.54181036637177,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002399961321382097,
      "loss": 5.1674,
      "mean_token_accuracy": 0.2318272426724434,
      "num_tokens": 108411602.0,
      "step": 55440
    },
    {
      "entropy": 5.670980596542359,
      "epoch": 5.542310191432998,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023996087801192583,
      "loss": 5.0931,
      "mean_token_accuracy": 0.22516852617263794,
      "num_tokens": 108421976.0,
      "step": 55445
    },
    {
      "entropy": 5.87999382019043,
      "epoch": 5.542810016494227,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002399256247672173,
      "loss": 5.3356,
      "mean_token_accuracy": 0.21564442217350005,
      "num_tokens": 108431231.0,
      "step": 55450
    },
    {
      "entropy": 5.759003973007202,
      "epoch": 5.543309841555455,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023989037240497096,
      "loss": 5.0456,
      "mean_token_accuracy": 0.2309943690896034,
      "num_tokens": 108442177.0,
      "step": 55455
    },
    {
      "entropy": 5.757803583145142,
      "epoch": 5.543809666616684,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023985512092607387,
      "loss": 5.1343,
      "mean_token_accuracy": 0.22629011869430543,
      "num_tokens": 108451324.0,
      "step": 55460
    },
    {
      "entropy": 5.8716919898986815,
      "epoch": 5.544309491677913,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023981987033141284,
      "loss": 5.1814,
      "mean_token_accuracy": 0.22272633910179138,
      "num_tokens": 108461871.0,
      "step": 55465
    },
    {
      "entropy": 5.760394048690796,
      "epoch": 5.5448093167391415,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023978462062187485,
      "loss": 5.127,
      "mean_token_accuracy": 0.22807687073945998,
      "num_tokens": 108472262.0,
      "step": 55470
    },
    {
      "entropy": 5.833635044097901,
      "epoch": 5.54530914180037,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023974937179834681,
      "loss": 5.2406,
      "mean_token_accuracy": 0.21894096434116364,
      "num_tokens": 108482521.0,
      "step": 55475
    },
    {
      "entropy": 5.739714479446411,
      "epoch": 5.545808966861598,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023971412386171548,
      "loss": 5.1953,
      "mean_token_accuracy": 0.23004181534051896,
      "num_tokens": 108491631.0,
      "step": 55480
    },
    {
      "entropy": 5.7313450336456295,
      "epoch": 5.546308791922827,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023967887681286765,
      "loss": 5.0909,
      "mean_token_accuracy": 0.22239395529031752,
      "num_tokens": 108500720.0,
      "step": 55485
    },
    {
      "entropy": 5.7265801429748535,
      "epoch": 5.546808616984055,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023964363065269017,
      "loss": 5.0364,
      "mean_token_accuracy": 0.2403646484017372,
      "num_tokens": 108510387.0,
      "step": 55490
    },
    {
      "entropy": 5.790591144561768,
      "epoch": 5.547308442045284,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002396083853820699,
      "loss": 5.1099,
      "mean_token_accuracy": 0.2316739872097969,
      "num_tokens": 108520681.0,
      "step": 55495
    },
    {
      "entropy": 5.8588520050048825,
      "epoch": 5.547808267106513,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023957314100189354,
      "loss": 5.3296,
      "mean_token_accuracy": 0.22041382640600204,
      "num_tokens": 108531531.0,
      "step": 55500
    },
    {
      "entropy": 5.818482351303101,
      "epoch": 5.5483080921677415,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023953789751304783,
      "loss": 5.1813,
      "mean_token_accuracy": 0.2239350289106369,
      "num_tokens": 108540698.0,
      "step": 55505
    },
    {
      "entropy": 5.777743005752564,
      "epoch": 5.54880791722897,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023950265491641954,
      "loss": 5.1879,
      "mean_token_accuracy": 0.21418870985507965,
      "num_tokens": 108550601.0,
      "step": 55510
    },
    {
      "entropy": 5.855816507339478,
      "epoch": 5.549307742290198,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023946741321289523,
      "loss": 5.2391,
      "mean_token_accuracy": 0.21818089485168457,
      "num_tokens": 108560177.0,
      "step": 55515
    },
    {
      "entropy": 5.787144136428833,
      "epoch": 5.549807567351427,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023943217240336175,
      "loss": 5.2114,
      "mean_token_accuracy": 0.22115907371044158,
      "num_tokens": 108570259.0,
      "step": 55520
    },
    {
      "entropy": 5.681055402755737,
      "epoch": 5.550307392412655,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023939693248870554,
      "loss": 5.0637,
      "mean_token_accuracy": 0.24087909311056138,
      "num_tokens": 108581146.0,
      "step": 55525
    },
    {
      "entropy": 5.77838978767395,
      "epoch": 5.550807217473884,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002393616934698134,
      "loss": 5.2045,
      "mean_token_accuracy": 0.22362290918827057,
      "num_tokens": 108593942.0,
      "step": 55530
    },
    {
      "entropy": 5.7694775581359865,
      "epoch": 5.551307042535113,
      "grad_norm": 1.09375,
      "learning_rate": 0.00023932645534757192,
      "loss": 5.1663,
      "mean_token_accuracy": 0.22902638763189315,
      "num_tokens": 108602837.0,
      "step": 55535
    },
    {
      "entropy": 5.673811578750611,
      "epoch": 5.5518068675963415,
      "grad_norm": 1.046875,
      "learning_rate": 0.00023929121812286765,
      "loss": 5.0787,
      "mean_token_accuracy": 0.240499347448349,
      "num_tokens": 108614445.0,
      "step": 55540
    },
    {
      "entropy": 5.789804840087891,
      "epoch": 5.55230669265757,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023925598179658727,
      "loss": 5.1176,
      "mean_token_accuracy": 0.2306716650724411,
      "num_tokens": 108623873.0,
      "step": 55545
    },
    {
      "entropy": 5.749519348144531,
      "epoch": 5.552806517718798,
      "grad_norm": 1.375,
      "learning_rate": 0.00023922074636961694,
      "loss": 5.0806,
      "mean_token_accuracy": 0.23667407482862474,
      "num_tokens": 108633189.0,
      "step": 55550
    },
    {
      "entropy": 5.7712243556976315,
      "epoch": 5.553306342780027,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023918551184284355,
      "loss": 5.1194,
      "mean_token_accuracy": 0.2248160034418106,
      "num_tokens": 108642641.0,
      "step": 55555
    },
    {
      "entropy": 5.828954744338989,
      "epoch": 5.553806167841255,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002391502782171534,
      "loss": 5.2627,
      "mean_token_accuracy": 0.21560937017202378,
      "num_tokens": 108653308.0,
      "step": 55560
    },
    {
      "entropy": 5.776281499862671,
      "epoch": 5.554305992902484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002391150454934331,
      "loss": 5.1036,
      "mean_token_accuracy": 0.22459865510463714,
      "num_tokens": 108663362.0,
      "step": 55565
    },
    {
      "entropy": 5.640609216690064,
      "epoch": 5.554805817963713,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023907981367256897,
      "loss": 5.0462,
      "mean_token_accuracy": 0.24448426067829132,
      "num_tokens": 108673447.0,
      "step": 55570
    },
    {
      "entropy": 5.737593793869019,
      "epoch": 5.5553056430249415,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002390445827554475,
      "loss": 5.1418,
      "mean_token_accuracy": 0.2249154955148697,
      "num_tokens": 108682344.0,
      "step": 55575
    },
    {
      "entropy": 5.755420780181884,
      "epoch": 5.55580546808617,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023900935274295509,
      "loss": 5.1488,
      "mean_token_accuracy": 0.23041532337665557,
      "num_tokens": 108692091.0,
      "step": 55580
    },
    {
      "entropy": 5.851890802383423,
      "epoch": 5.556305293147398,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002389741236359781,
      "loss": 5.2671,
      "mean_token_accuracy": 0.22228560000658035,
      "num_tokens": 108702198.0,
      "step": 55585
    },
    {
      "entropy": 5.725716590881348,
      "epoch": 5.556805118208627,
      "grad_norm": 1.25,
      "learning_rate": 0.00023893889543540286,
      "loss": 5.082,
      "mean_token_accuracy": 0.23493741899728776,
      "num_tokens": 108711274.0,
      "step": 55590
    },
    {
      "entropy": 5.740524578094482,
      "epoch": 5.557304943269855,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023890366814211572,
      "loss": 5.1431,
      "mean_token_accuracy": 0.2264164999127388,
      "num_tokens": 108720303.0,
      "step": 55595
    },
    {
      "entropy": 5.80184531211853,
      "epoch": 5.557804768331084,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023886844175700307,
      "loss": 5.1692,
      "mean_token_accuracy": 0.22560289949178697,
      "num_tokens": 108729911.0,
      "step": 55600
    },
    {
      "entropy": 5.744702339172363,
      "epoch": 5.558304593392313,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023883321628095106,
      "loss": 5.127,
      "mean_token_accuracy": 0.228611022233963,
      "num_tokens": 108739623.0,
      "step": 55605
    },
    {
      "entropy": 5.804797697067261,
      "epoch": 5.5588044184535415,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023879799171484613,
      "loss": 5.2493,
      "mean_token_accuracy": 0.22592300176620483,
      "num_tokens": 108749823.0,
      "step": 55610
    },
    {
      "entropy": 5.805840635299683,
      "epoch": 5.55930424351477,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002387627680595743,
      "loss": 5.2272,
      "mean_token_accuracy": 0.22620415985584258,
      "num_tokens": 108759797.0,
      "step": 55615
    },
    {
      "entropy": 5.841667652130127,
      "epoch": 5.559804068575998,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023872754531602199,
      "loss": 5.2904,
      "mean_token_accuracy": 0.21252412647008895,
      "num_tokens": 108769656.0,
      "step": 55620
    },
    {
      "entropy": 5.829286670684814,
      "epoch": 5.560303893637227,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023869232348507524,
      "loss": 5.1888,
      "mean_token_accuracy": 0.22124103158712388,
      "num_tokens": 108778508.0,
      "step": 55625
    },
    {
      "entropy": 5.693196439743042,
      "epoch": 5.560803718698455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023865710256762036,
      "loss": 5.1358,
      "mean_token_accuracy": 0.23426292836666107,
      "num_tokens": 108789770.0,
      "step": 55630
    },
    {
      "entropy": 5.794801378250122,
      "epoch": 5.561303543759684,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023862188256454332,
      "loss": 5.1189,
      "mean_token_accuracy": 0.22237320989370346,
      "num_tokens": 108799523.0,
      "step": 55635
    },
    {
      "entropy": 5.788484144210815,
      "epoch": 5.561803368820913,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023858666347673042,
      "loss": 5.1686,
      "mean_token_accuracy": 0.2249519109725952,
      "num_tokens": 108809151.0,
      "step": 55640
    },
    {
      "entropy": 5.758563280105591,
      "epoch": 5.5623031938821414,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023855144530506768,
      "loss": 5.1293,
      "mean_token_accuracy": 0.24015165865421295,
      "num_tokens": 108818524.0,
      "step": 55645
    },
    {
      "entropy": 5.758776521682739,
      "epoch": 5.56280301894337,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023851622805044117,
      "loss": 5.1593,
      "mean_token_accuracy": 0.22685603499412538,
      "num_tokens": 108828616.0,
      "step": 55650
    },
    {
      "entropy": 5.820524835586548,
      "epoch": 5.563302844004598,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000238481011713737,
      "loss": 5.2532,
      "mean_token_accuracy": 0.21884117573499678,
      "num_tokens": 108839143.0,
      "step": 55655
    },
    {
      "entropy": 5.821717309951782,
      "epoch": 5.563802669065827,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023844579629584117,
      "loss": 5.1749,
      "mean_token_accuracy": 0.223552705347538,
      "num_tokens": 108848852.0,
      "step": 55660
    },
    {
      "entropy": 5.730998945236206,
      "epoch": 5.564302494127055,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023841058179763976,
      "loss": 5.0538,
      "mean_token_accuracy": 0.23576657623052596,
      "num_tokens": 108859599.0,
      "step": 55665
    },
    {
      "entropy": 5.826196146011353,
      "epoch": 5.564802319188284,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023837536822001858,
      "loss": 5.2042,
      "mean_token_accuracy": 0.22516683787107467,
      "num_tokens": 108870338.0,
      "step": 55670
    },
    {
      "entropy": 5.809954643249512,
      "epoch": 5.565302144249513,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002383401555638638,
      "loss": 5.1832,
      "mean_token_accuracy": 0.23378636837005615,
      "num_tokens": 108880611.0,
      "step": 55675
    },
    {
      "entropy": 5.797666072845459,
      "epoch": 5.5658019693107414,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023830494383006123,
      "loss": 5.181,
      "mean_token_accuracy": 0.22433158606290818,
      "num_tokens": 108890316.0,
      "step": 55680
    },
    {
      "entropy": 5.78524694442749,
      "epoch": 5.56630179437197,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023826973301949695,
      "loss": 5.1721,
      "mean_token_accuracy": 0.21767204999923706,
      "num_tokens": 108899753.0,
      "step": 55685
    },
    {
      "entropy": 5.797933530807495,
      "epoch": 5.566801619433198,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023823452313305666,
      "loss": 5.2106,
      "mean_token_accuracy": 0.2232616886496544,
      "num_tokens": 108910732.0,
      "step": 55690
    },
    {
      "entropy": 5.888829898834229,
      "epoch": 5.567301444494427,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023819931417162627,
      "loss": 5.3196,
      "mean_token_accuracy": 0.21560386270284654,
      "num_tokens": 108920213.0,
      "step": 55695
    },
    {
      "entropy": 5.87577714920044,
      "epoch": 5.567801269555655,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002381641061360917,
      "loss": 5.1824,
      "mean_token_accuracy": 0.2258385732769966,
      "num_tokens": 108929538.0,
      "step": 55700
    },
    {
      "entropy": 5.777638673782349,
      "epoch": 5.568301094616884,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002381288990273387,
      "loss": 5.1407,
      "mean_token_accuracy": 0.22355872243642808,
      "num_tokens": 108939625.0,
      "step": 55705
    },
    {
      "entropy": 5.804617309570313,
      "epoch": 5.568800919678113,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002380936928462531,
      "loss": 5.1332,
      "mean_token_accuracy": 0.2255225211381912,
      "num_tokens": 108949145.0,
      "step": 55710
    },
    {
      "entropy": 5.761106824874878,
      "epoch": 5.5693007447393414,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023805848759372072,
      "loss": 5.1803,
      "mean_token_accuracy": 0.22926532924175264,
      "num_tokens": 108958469.0,
      "step": 55715
    },
    {
      "entropy": 5.736740922927856,
      "epoch": 5.56980056980057,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023802328327062727,
      "loss": 5.0833,
      "mean_token_accuracy": 0.23059680312871933,
      "num_tokens": 108966980.0,
      "step": 55720
    },
    {
      "entropy": 5.784594392776489,
      "epoch": 5.570300394861798,
      "grad_norm": 1.25,
      "learning_rate": 0.00023798807987785854,
      "loss": 5.1916,
      "mean_token_accuracy": 0.22571785300970076,
      "num_tokens": 108975984.0,
      "step": 55725
    },
    {
      "entropy": 5.770972013473511,
      "epoch": 5.570800219923027,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023795287741630017,
      "loss": 5.176,
      "mean_token_accuracy": 0.2216135859489441,
      "num_tokens": 108985794.0,
      "step": 55730
    },
    {
      "entropy": 5.74450798034668,
      "epoch": 5.571300044984255,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023791767588683788,
      "loss": 5.1307,
      "mean_token_accuracy": 0.21876550167798997,
      "num_tokens": 108994565.0,
      "step": 55735
    },
    {
      "entropy": 5.7138231754302975,
      "epoch": 5.571799870045484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023788247529035733,
      "loss": 5.0956,
      "mean_token_accuracy": 0.22918130606412887,
      "num_tokens": 109005093.0,
      "step": 55740
    },
    {
      "entropy": 5.80222454071045,
      "epoch": 5.572299695106713,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023784727562774411,
      "loss": 5.1481,
      "mean_token_accuracy": 0.2225553035736084,
      "num_tokens": 109013980.0,
      "step": 55745
    },
    {
      "entropy": 5.827111768722534,
      "epoch": 5.5727995201679414,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023781207689988387,
      "loss": 5.2188,
      "mean_token_accuracy": 0.21825421154499053,
      "num_tokens": 109023518.0,
      "step": 55750
    },
    {
      "entropy": 5.676475238800049,
      "epoch": 5.57329934522917,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023777687910766233,
      "loss": 5.041,
      "mean_token_accuracy": 0.24444643557071685,
      "num_tokens": 109032170.0,
      "step": 55755
    },
    {
      "entropy": 5.687872314453125,
      "epoch": 5.573799170290398,
      "grad_norm": 1.25,
      "learning_rate": 0.00023774168225196486,
      "loss": 5.0853,
      "mean_token_accuracy": 0.2299981966614723,
      "num_tokens": 109041141.0,
      "step": 55760
    },
    {
      "entropy": 5.7140185832977295,
      "epoch": 5.574298995351627,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002377064863336771,
      "loss": 5.0977,
      "mean_token_accuracy": 0.23374864757061004,
      "num_tokens": 109050788.0,
      "step": 55765
    },
    {
      "entropy": 5.7712925434112545,
      "epoch": 5.574798820412855,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023767129135368443,
      "loss": 5.1771,
      "mean_token_accuracy": 0.22898520529270172,
      "num_tokens": 109060584.0,
      "step": 55770
    },
    {
      "entropy": 5.826577281951904,
      "epoch": 5.575298645474084,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023763609731287255,
      "loss": 5.2319,
      "mean_token_accuracy": 0.21848107725381852,
      "num_tokens": 109068932.0,
      "step": 55775
    },
    {
      "entropy": 5.794831275939941,
      "epoch": 5.575798470535313,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023760090421212683,
      "loss": 5.2203,
      "mean_token_accuracy": 0.22280485779047013,
      "num_tokens": 109080223.0,
      "step": 55780
    },
    {
      "entropy": 5.9296417236328125,
      "epoch": 5.576298295596541,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002375657120523328,
      "loss": 5.328,
      "mean_token_accuracy": 0.21082387566566468,
      "num_tokens": 109090936.0,
      "step": 55785
    },
    {
      "entropy": 5.859537363052368,
      "epoch": 5.57679812065777,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023753052083437577,
      "loss": 5.201,
      "mean_token_accuracy": 0.22235283106565476,
      "num_tokens": 109099840.0,
      "step": 55790
    },
    {
      "entropy": 5.757027816772461,
      "epoch": 5.577297945718998,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023749533055914118,
      "loss": 5.1585,
      "mean_token_accuracy": 0.22159343957901,
      "num_tokens": 109109663.0,
      "step": 55795
    },
    {
      "entropy": 5.752348899841309,
      "epoch": 5.577797770780227,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002374601412275144,
      "loss": 5.137,
      "mean_token_accuracy": 0.22960332334041594,
      "num_tokens": 109119196.0,
      "step": 55800
    },
    {
      "entropy": 5.779278087615967,
      "epoch": 5.578297595841455,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023742495284038084,
      "loss": 5.17,
      "mean_token_accuracy": 0.2221122846007347,
      "num_tokens": 109128940.0,
      "step": 55805
    },
    {
      "entropy": 5.824253559112549,
      "epoch": 5.578797420902684,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023738976539862579,
      "loss": 5.2018,
      "mean_token_accuracy": 0.21868612766265869,
      "num_tokens": 109139614.0,
      "step": 55810
    },
    {
      "entropy": 5.822767210006714,
      "epoch": 5.579297245963913,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023735457890313455,
      "loss": 5.188,
      "mean_token_accuracy": 0.22080190479755402,
      "num_tokens": 109148624.0,
      "step": 55815
    },
    {
      "entropy": 5.640528535842895,
      "epoch": 5.579797071025141,
      "grad_norm": 1.125,
      "learning_rate": 0.00023731939335479247,
      "loss": 5.008,
      "mean_token_accuracy": 0.2503345459699631,
      "num_tokens": 109159494.0,
      "step": 55820
    },
    {
      "entropy": 5.757949352264404,
      "epoch": 5.58029689608637,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023728420875448482,
      "loss": 5.1073,
      "mean_token_accuracy": 0.2329590931534767,
      "num_tokens": 109170255.0,
      "step": 55825
    },
    {
      "entropy": 5.8042311668396,
      "epoch": 5.580796721147598,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023724902510309672,
      "loss": 5.1811,
      "mean_token_accuracy": 0.21633170992136003,
      "num_tokens": 109180341.0,
      "step": 55830
    },
    {
      "entropy": 5.819483137130737,
      "epoch": 5.581296546208827,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023721384240151333,
      "loss": 5.1911,
      "mean_token_accuracy": 0.21709214746952057,
      "num_tokens": 109189535.0,
      "step": 55835
    },
    {
      "entropy": 5.724277019500732,
      "epoch": 5.581796371270055,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00023717866065061998,
      "loss": 5.0628,
      "mean_token_accuracy": 0.2342119410634041,
      "num_tokens": 109199815.0,
      "step": 55840
    },
    {
      "entropy": 5.789330768585205,
      "epoch": 5.582296196331284,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023714347985130186,
      "loss": 5.1838,
      "mean_token_accuracy": 0.22644447088241576,
      "num_tokens": 109209227.0,
      "step": 55845
    },
    {
      "entropy": 5.745349550247193,
      "epoch": 5.582796021392513,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000237108300004444,
      "loss": 5.1233,
      "mean_token_accuracy": 0.22983393222093582,
      "num_tokens": 109219389.0,
      "step": 55850
    },
    {
      "entropy": 5.711888647079467,
      "epoch": 5.583295846453741,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002370731211109316,
      "loss": 5.1281,
      "mean_token_accuracy": 0.23427960723638536,
      "num_tokens": 109229483.0,
      "step": 55855
    },
    {
      "entropy": 5.814821386337281,
      "epoch": 5.58379567151497,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023703794317164967,
      "loss": 5.2241,
      "mean_token_accuracy": 0.22934310883283615,
      "num_tokens": 109238638.0,
      "step": 55860
    },
    {
      "entropy": 5.789039516448975,
      "epoch": 5.584295496576198,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023700276618748335,
      "loss": 5.1865,
      "mean_token_accuracy": 0.21983690410852433,
      "num_tokens": 109248628.0,
      "step": 55865
    },
    {
      "entropy": 5.796166467666626,
      "epoch": 5.584795321637427,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023696759015931756,
      "loss": 5.2683,
      "mean_token_accuracy": 0.21941974610090256,
      "num_tokens": 109259059.0,
      "step": 55870
    },
    {
      "entropy": 5.798108243942261,
      "epoch": 5.585295146698655,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023693241508803754,
      "loss": 5.1335,
      "mean_token_accuracy": 0.2295671671628952,
      "num_tokens": 109268609.0,
      "step": 55875
    },
    {
      "entropy": 5.823902893066406,
      "epoch": 5.585794971759884,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023689724097452808,
      "loss": 5.1786,
      "mean_token_accuracy": 0.22283689677715302,
      "num_tokens": 109278677.0,
      "step": 55880
    },
    {
      "entropy": 5.692731809616089,
      "epoch": 5.586294796821113,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023686206781967428,
      "loss": 5.0765,
      "mean_token_accuracy": 0.23523708581924438,
      "num_tokens": 109288675.0,
      "step": 55885
    },
    {
      "entropy": 5.842178010940552,
      "epoch": 5.586794621882341,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023682689562436104,
      "loss": 5.2171,
      "mean_token_accuracy": 0.21825551092624665,
      "num_tokens": 109298715.0,
      "step": 55890
    },
    {
      "entropy": 5.802306509017944,
      "epoch": 5.58729444694357,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023679172438947332,
      "loss": 5.1268,
      "mean_token_accuracy": 0.22922392785549164,
      "num_tokens": 109308566.0,
      "step": 55895
    },
    {
      "entropy": 5.760244417190552,
      "epoch": 5.587794272004798,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023675655411589593,
      "loss": 5.1219,
      "mean_token_accuracy": 0.22748637348413467,
      "num_tokens": 109317156.0,
      "step": 55900
    },
    {
      "entropy": 5.769362735748291,
      "epoch": 5.588294097066027,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023672138480451384,
      "loss": 5.106,
      "mean_token_accuracy": 0.23194352239370347,
      "num_tokens": 109327622.0,
      "step": 55905
    },
    {
      "entropy": 5.606499099731446,
      "epoch": 5.588793922127255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002366862164562118,
      "loss": 4.9928,
      "mean_token_accuracy": 0.25328775495290756,
      "num_tokens": 109337799.0,
      "step": 55910
    },
    {
      "entropy": 5.665244913101196,
      "epoch": 5.589293747188484,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023665104907187473,
      "loss": 5.1007,
      "mean_token_accuracy": 0.23945810794830322,
      "num_tokens": 109346571.0,
      "step": 55915
    },
    {
      "entropy": 5.709790468215942,
      "epoch": 5.589793572249713,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023661588265238737,
      "loss": 5.0197,
      "mean_token_accuracy": 0.24324010610580443,
      "num_tokens": 109356341.0,
      "step": 55920
    },
    {
      "entropy": 5.813706779479981,
      "epoch": 5.590293397310941,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023658071719863456,
      "loss": 5.2232,
      "mean_token_accuracy": 0.22254901230335236,
      "num_tokens": 109367336.0,
      "step": 55925
    },
    {
      "entropy": 5.771400880813599,
      "epoch": 5.59079322237217,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023654555271150103,
      "loss": 5.1367,
      "mean_token_accuracy": 0.22762976288795472,
      "num_tokens": 109376680.0,
      "step": 55930
    },
    {
      "entropy": 5.74250979423523,
      "epoch": 5.591293047433398,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002365103891918714,
      "loss": 5.1445,
      "mean_token_accuracy": 0.23118688613176347,
      "num_tokens": 109385650.0,
      "step": 55935
    },
    {
      "entropy": 5.734711456298828,
      "epoch": 5.591792872494627,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023647522664063054,
      "loss": 5.075,
      "mean_token_accuracy": 0.23066534250974655,
      "num_tokens": 109394821.0,
      "step": 55940
    },
    {
      "entropy": 5.815477657318115,
      "epoch": 5.592292697555855,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023644006505866305,
      "loss": 5.2062,
      "mean_token_accuracy": 0.2218490019440651,
      "num_tokens": 109404646.0,
      "step": 55945
    },
    {
      "entropy": 5.780820274353028,
      "epoch": 5.592792522617084,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023640490444685363,
      "loss": 5.1494,
      "mean_token_accuracy": 0.22855646461248397,
      "num_tokens": 109413942.0,
      "step": 55950
    },
    {
      "entropy": 5.797822332382202,
      "epoch": 5.593292347678313,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023636974480608687,
      "loss": 5.215,
      "mean_token_accuracy": 0.21982590854167938,
      "num_tokens": 109423405.0,
      "step": 55955
    },
    {
      "entropy": 5.734571743011474,
      "epoch": 5.593792172739541,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002363345861372474,
      "loss": 5.1877,
      "mean_token_accuracy": 0.22263875603675842,
      "num_tokens": 109433649.0,
      "step": 55960
    },
    {
      "entropy": 5.722483348846436,
      "epoch": 5.59429199780077,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023629942844121982,
      "loss": 5.1169,
      "mean_token_accuracy": 0.2337401807308197,
      "num_tokens": 109444254.0,
      "step": 55965
    },
    {
      "entropy": 5.806057405471802,
      "epoch": 5.594791822861998,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023626427171888865,
      "loss": 5.1359,
      "mean_token_accuracy": 0.2271878108382225,
      "num_tokens": 109454415.0,
      "step": 55970
    },
    {
      "entropy": 5.811261940002441,
      "epoch": 5.595291647923227,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002362291159711384,
      "loss": 5.2001,
      "mean_token_accuracy": 0.22651894241571427,
      "num_tokens": 109464987.0,
      "step": 55975
    },
    {
      "entropy": 5.583225107192993,
      "epoch": 5.595791472984455,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002361939611988536,
      "loss": 4.9012,
      "mean_token_accuracy": 0.24510067105293273,
      "num_tokens": 109475014.0,
      "step": 55980
    },
    {
      "entropy": 5.768626165390015,
      "epoch": 5.596291298045684,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023615880740291884,
      "loss": 5.0982,
      "mean_token_accuracy": 0.23285433501005173,
      "num_tokens": 109484809.0,
      "step": 55985
    },
    {
      "entropy": 5.848206663131714,
      "epoch": 5.596791123106913,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023612365458421843,
      "loss": 5.2585,
      "mean_token_accuracy": 0.21957770735025406,
      "num_tokens": 109494201.0,
      "step": 55990
    },
    {
      "entropy": 5.749364709854126,
      "epoch": 5.597290948168141,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002360885027436368,
      "loss": 5.1234,
      "mean_token_accuracy": 0.2306797280907631,
      "num_tokens": 109503380.0,
      "step": 55995
    },
    {
      "entropy": 5.69812970161438,
      "epoch": 5.59779077322937,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002360533518820585,
      "loss": 5.1666,
      "mean_token_accuracy": 0.22974963784217833,
      "num_tokens": 109512778.0,
      "step": 56000
    },
    {
      "entropy": 5.4931213617324826,
      "epoch": 5.598290598290598,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023601820200036782,
      "loss": 4.8484,
      "mean_token_accuracy": 0.26468845903873445,
      "num_tokens": 109522704.0,
      "step": 56005
    },
    {
      "entropy": 5.690971899032593,
      "epoch": 5.598790423351827,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023598305309944917,
      "loss": 5.0519,
      "mean_token_accuracy": 0.23947680741548538,
      "num_tokens": 109533032.0,
      "step": 56010
    },
    {
      "entropy": 5.7765336513519285,
      "epoch": 5.599290248413055,
      "grad_norm": 1.375,
      "learning_rate": 0.00023594790518018688,
      "loss": 5.1647,
      "mean_token_accuracy": 0.22449734807014465,
      "num_tokens": 109543007.0,
      "step": 56015
    },
    {
      "entropy": 5.7917078018188475,
      "epoch": 5.599790073474284,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002359127582434652,
      "loss": 5.1616,
      "mean_token_accuracy": 0.23281074166297913,
      "num_tokens": 109552632.0,
      "step": 56020
    },
    {
      "entropy": 5.7951336860656735,
      "epoch": 5.600289898535513,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023587761229016852,
      "loss": 5.1363,
      "mean_token_accuracy": 0.230520161986351,
      "num_tokens": 109562195.0,
      "step": 56025
    },
    {
      "entropy": 5.8157013893127445,
      "epoch": 5.600789723596741,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023584246732118104,
      "loss": 5.242,
      "mean_token_accuracy": 0.21995309591293336,
      "num_tokens": 109572364.0,
      "step": 56030
    },
    {
      "entropy": 5.7066374778747555,
      "epoch": 5.60128954865797,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023580732333738702,
      "loss": 5.0189,
      "mean_token_accuracy": 0.2312516987323761,
      "num_tokens": 109582254.0,
      "step": 56035
    },
    {
      "entropy": 5.723512458801269,
      "epoch": 5.601789373719198,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023577218033967064,
      "loss": 5.0866,
      "mean_token_accuracy": 0.2426062986254692,
      "num_tokens": 109591599.0,
      "step": 56040
    },
    {
      "entropy": 5.749347162246704,
      "epoch": 5.602289198780427,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023573703832891607,
      "loss": 5.1813,
      "mean_token_accuracy": 0.23132301419973372,
      "num_tokens": 109601219.0,
      "step": 56045
    },
    {
      "entropy": 5.76164984703064,
      "epoch": 5.602789023841655,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023570189730600754,
      "loss": 5.1678,
      "mean_token_accuracy": 0.22826303839683532,
      "num_tokens": 109611172.0,
      "step": 56050
    },
    {
      "entropy": 5.82293062210083,
      "epoch": 5.603288848902884,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023566675727182914,
      "loss": 5.0795,
      "mean_token_accuracy": 0.23018780648708342,
      "num_tokens": 109620100.0,
      "step": 56055
    },
    {
      "entropy": 5.810928010940552,
      "epoch": 5.603788673964113,
      "grad_norm": 1.21875,
      "learning_rate": 0.000235631618227265,
      "loss": 5.2144,
      "mean_token_accuracy": 0.22579940110445024,
      "num_tokens": 109628456.0,
      "step": 56060
    },
    {
      "entropy": 5.563017129898071,
      "epoch": 5.604288499025341,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023559648017319924,
      "loss": 4.9403,
      "mean_token_accuracy": 0.24788676500320433,
      "num_tokens": 109639413.0,
      "step": 56065
    },
    {
      "entropy": 5.7313783168792725,
      "epoch": 5.60478832408657,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023556134311051585,
      "loss": 5.1437,
      "mean_token_accuracy": 0.2334027424454689,
      "num_tokens": 109648720.0,
      "step": 56070
    },
    {
      "entropy": 5.767942142486572,
      "epoch": 5.605288149147798,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023552620704009902,
      "loss": 5.1637,
      "mean_token_accuracy": 0.22509979605674743,
      "num_tokens": 109658857.0,
      "step": 56075
    },
    {
      "entropy": 5.737338542938232,
      "epoch": 5.605787974209027,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023549107196283258,
      "loss": 5.1454,
      "mean_token_accuracy": 0.2319129928946495,
      "num_tokens": 109668330.0,
      "step": 56080
    },
    {
      "entropy": 5.708700227737427,
      "epoch": 5.606287799270255,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023545593787960067,
      "loss": 5.0729,
      "mean_token_accuracy": 0.24415443539619447,
      "num_tokens": 109677278.0,
      "step": 56085
    },
    {
      "entropy": 5.776048088073731,
      "epoch": 5.606787624331484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023542080479128713,
      "loss": 5.1739,
      "mean_token_accuracy": 0.22748820930719377,
      "num_tokens": 109687629.0,
      "step": 56090
    },
    {
      "entropy": 5.809965038299561,
      "epoch": 5.607287449392713,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023538567269877593,
      "loss": 5.2238,
      "mean_token_accuracy": 0.21620575785636903,
      "num_tokens": 109697187.0,
      "step": 56095
    },
    {
      "entropy": 5.78884744644165,
      "epoch": 5.607787274453941,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002353505416029511,
      "loss": 5.2391,
      "mean_token_accuracy": 0.22056455314159393,
      "num_tokens": 109707062.0,
      "step": 56100
    },
    {
      "entropy": 5.749438810348511,
      "epoch": 5.60828709951517,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002353154115046965,
      "loss": 5.0896,
      "mean_token_accuracy": 0.23181631565093994,
      "num_tokens": 109717587.0,
      "step": 56105
    },
    {
      "entropy": 5.800749588012695,
      "epoch": 5.608786924576398,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023528028240489578,
      "loss": 5.1999,
      "mean_token_accuracy": 0.2323201447725296,
      "num_tokens": 109728628.0,
      "step": 56110
    },
    {
      "entropy": 5.747981119155884,
      "epoch": 5.609286749637627,
      "grad_norm": 1.25,
      "learning_rate": 0.000235245154304433,
      "loss": 5.1047,
      "mean_token_accuracy": 0.23078362941741942,
      "num_tokens": 109738040.0,
      "step": 56115
    },
    {
      "entropy": 5.799719619750976,
      "epoch": 5.609786574698855,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023521002720419188,
      "loss": 5.188,
      "mean_token_accuracy": 0.2317604824900627,
      "num_tokens": 109748055.0,
      "step": 56120
    },
    {
      "entropy": 5.741240787506103,
      "epoch": 5.610286399760084,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023517490110505624,
      "loss": 5.0905,
      "mean_token_accuracy": 0.22768013775348664,
      "num_tokens": 109759391.0,
      "step": 56125
    },
    {
      "entropy": 5.812180852890014,
      "epoch": 5.610786224821313,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023513977600790985,
      "loss": 5.2322,
      "mean_token_accuracy": 0.22272959500551223,
      "num_tokens": 109770145.0,
      "step": 56130
    },
    {
      "entropy": 5.895798540115356,
      "epoch": 5.611286049882541,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023510465191363644,
      "loss": 5.2956,
      "mean_token_accuracy": 0.21123722940683365,
      "num_tokens": 109779210.0,
      "step": 56135
    },
    {
      "entropy": 5.640702962875366,
      "epoch": 5.61178587494377,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023506952882311978,
      "loss": 5.0018,
      "mean_token_accuracy": 0.24045850336551666,
      "num_tokens": 109789923.0,
      "step": 56140
    },
    {
      "entropy": 5.747842311859131,
      "epoch": 5.612285700004998,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002350344067372434,
      "loss": 5.1204,
      "mean_token_accuracy": 0.23279667496681214,
      "num_tokens": 109798993.0,
      "step": 56145
    },
    {
      "entropy": 5.831531286239624,
      "epoch": 5.612785525066227,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002349992856568911,
      "loss": 5.1563,
      "mean_token_accuracy": 0.22565697878599167,
      "num_tokens": 109808700.0,
      "step": 56150
    },
    {
      "entropy": 5.594302701950073,
      "epoch": 5.613285350127455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002349641655829464,
      "loss": 4.9606,
      "mean_token_accuracy": 0.2404107540845871,
      "num_tokens": 109818255.0,
      "step": 56155
    },
    {
      "entropy": 5.696468830108643,
      "epoch": 5.613785175188684,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023492904651629316,
      "loss": 5.148,
      "mean_token_accuracy": 0.23027893304824829,
      "num_tokens": 109827971.0,
      "step": 56160
    },
    {
      "entropy": 5.74213604927063,
      "epoch": 5.614285000249913,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023489392845781472,
      "loss": 5.163,
      "mean_token_accuracy": 0.2301846355199814,
      "num_tokens": 109836967.0,
      "step": 56165
    },
    {
      "entropy": 5.7576611042022705,
      "epoch": 5.614784825311141,
      "grad_norm": 1.25,
      "learning_rate": 0.00023485881140839472,
      "loss": 5.1535,
      "mean_token_accuracy": 0.22933966517448426,
      "num_tokens": 109847328.0,
      "step": 56170
    },
    {
      "entropy": 5.748749399185181,
      "epoch": 5.61528465037237,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002348236953689168,
      "loss": 5.0917,
      "mean_token_accuracy": 0.23434651792049407,
      "num_tokens": 109857493.0,
      "step": 56175
    },
    {
      "entropy": 5.755806016921997,
      "epoch": 5.615784475433598,
      "grad_norm": 1.25,
      "learning_rate": 0.00023478858034026417,
      "loss": 5.0637,
      "mean_token_accuracy": 0.2285470485687256,
      "num_tokens": 109867239.0,
      "step": 56180
    },
    {
      "entropy": 5.8132524490356445,
      "epoch": 5.616284300494827,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023475346632332062,
      "loss": 5.2113,
      "mean_token_accuracy": 0.2150292918086052,
      "num_tokens": 109876983.0,
      "step": 56185
    },
    {
      "entropy": 5.62470235824585,
      "epoch": 5.616784125556055,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023471835331896946,
      "loss": 5.0081,
      "mean_token_accuracy": 0.24719674736261368,
      "num_tokens": 109886005.0,
      "step": 56190
    },
    {
      "entropy": 5.8115684509277346,
      "epoch": 5.617283950617284,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023468324132809422,
      "loss": 5.2088,
      "mean_token_accuracy": 0.22648539692163466,
      "num_tokens": 109895151.0,
      "step": 56195
    },
    {
      "entropy": 5.796738815307617,
      "epoch": 5.617783775678513,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023464813035157823,
      "loss": 5.2072,
      "mean_token_accuracy": 0.21408575028181076,
      "num_tokens": 109904847.0,
      "step": 56200
    },
    {
      "entropy": 5.835349273681641,
      "epoch": 5.618283600739741,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023461302039030487,
      "loss": 5.2072,
      "mean_token_accuracy": 0.21901955902576448,
      "num_tokens": 109914055.0,
      "step": 56205
    },
    {
      "entropy": 5.813844919204712,
      "epoch": 5.61878342580097,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023457791144515755,
      "loss": 5.2184,
      "mean_token_accuracy": 0.21931398957967757,
      "num_tokens": 109923764.0,
      "step": 56210
    },
    {
      "entropy": 5.760696363449097,
      "epoch": 5.619283250862198,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023454280351701957,
      "loss": 5.1814,
      "mean_token_accuracy": 0.22721351236104964,
      "num_tokens": 109933757.0,
      "step": 56215
    },
    {
      "entropy": 5.806264877319336,
      "epoch": 5.619783075923427,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023450769660677417,
      "loss": 5.1993,
      "mean_token_accuracy": 0.2199242353439331,
      "num_tokens": 109942170.0,
      "step": 56220
    },
    {
      "entropy": 5.800165128707886,
      "epoch": 5.620282900984655,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023447259071530474,
      "loss": 5.1865,
      "mean_token_accuracy": 0.2245820179581642,
      "num_tokens": 109950697.0,
      "step": 56225
    },
    {
      "entropy": 5.747802114486694,
      "epoch": 5.620782726045884,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023443748584349455,
      "loss": 5.1359,
      "mean_token_accuracy": 0.2406544044613838,
      "num_tokens": 109960125.0,
      "step": 56230
    },
    {
      "entropy": 5.8818519115448,
      "epoch": 5.621282551107113,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002344023819922268,
      "loss": 5.313,
      "mean_token_accuracy": 0.21452655792236328,
      "num_tokens": 109969428.0,
      "step": 56235
    },
    {
      "entropy": 5.834206676483154,
      "epoch": 5.621782376168341,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002343672791623846,
      "loss": 5.1841,
      "mean_token_accuracy": 0.2250538855791092,
      "num_tokens": 109978586.0,
      "step": 56240
    },
    {
      "entropy": 5.779029273986817,
      "epoch": 5.62228220122957,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023433217735485128,
      "loss": 5.0614,
      "mean_token_accuracy": 0.23664415180683135,
      "num_tokens": 109987401.0,
      "step": 56245
    },
    {
      "entropy": 5.769055795669556,
      "epoch": 5.622782026290798,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023429707657050993,
      "loss": 5.1249,
      "mean_token_accuracy": 0.2234545335173607,
      "num_tokens": 109996817.0,
      "step": 56250
    },
    {
      "entropy": 5.749361324310303,
      "epoch": 5.623281851352027,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023426197681024357,
      "loss": 5.1039,
      "mean_token_accuracy": 0.23366525173187255,
      "num_tokens": 110005889.0,
      "step": 56255
    },
    {
      "entropy": 5.774554491043091,
      "epoch": 5.623781676413255,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002342268780749354,
      "loss": 5.1,
      "mean_token_accuracy": 0.22885786443948747,
      "num_tokens": 110015542.0,
      "step": 56260
    },
    {
      "entropy": 5.744083213806152,
      "epoch": 5.624281501474484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00023419178036546856,
      "loss": 5.1695,
      "mean_token_accuracy": 0.23187071233987808,
      "num_tokens": 110025333.0,
      "step": 56265
    },
    {
      "entropy": 5.744677925109864,
      "epoch": 5.624781326535713,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023415668368272598,
      "loss": 5.1136,
      "mean_token_accuracy": 0.2302869126200676,
      "num_tokens": 110033739.0,
      "step": 56270
    },
    {
      "entropy": 5.80875244140625,
      "epoch": 5.625281151596941,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002341215880275907,
      "loss": 5.1865,
      "mean_token_accuracy": 0.22479676306247712,
      "num_tokens": 110042871.0,
      "step": 56275
    },
    {
      "entropy": 5.813724517822266,
      "epoch": 5.62578097665817,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023408649340094575,
      "loss": 5.2123,
      "mean_token_accuracy": 0.22860124558210373,
      "num_tokens": 110053146.0,
      "step": 56280
    },
    {
      "entropy": 5.7978044033050535,
      "epoch": 5.626280801719398,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023405139980367407,
      "loss": 5.1237,
      "mean_token_accuracy": 0.22984054535627366,
      "num_tokens": 110062203.0,
      "step": 56285
    },
    {
      "entropy": 5.727661800384522,
      "epoch": 5.626780626780627,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023401630723665867,
      "loss": 5.1671,
      "mean_token_accuracy": 0.22430173307657242,
      "num_tokens": 110072490.0,
      "step": 56290
    },
    {
      "entropy": 5.676339864730835,
      "epoch": 5.627280451841855,
      "grad_norm": 1.25,
      "learning_rate": 0.00023398121570078245,
      "loss": 5.06,
      "mean_token_accuracy": 0.23685924857854843,
      "num_tokens": 110081712.0,
      "step": 56295
    },
    {
      "entropy": 5.8153472423553465,
      "epoch": 5.627780276903084,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023394612519692825,
      "loss": 5.2232,
      "mean_token_accuracy": 0.22350506633520126,
      "num_tokens": 110090908.0,
      "step": 56300
    },
    {
      "entropy": 5.765409374237061,
      "epoch": 5.628280101964313,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023391103572597904,
      "loss": 5.0803,
      "mean_token_accuracy": 0.23511358350515366,
      "num_tokens": 110099688.0,
      "step": 56305
    },
    {
      "entropy": 5.749918651580811,
      "epoch": 5.628779927025541,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023387594728881757,
      "loss": 5.1527,
      "mean_token_accuracy": 0.22451508939266204,
      "num_tokens": 110109117.0,
      "step": 56310
    },
    {
      "entropy": 5.791953754425049,
      "epoch": 5.62927975208677,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002338408598863267,
      "loss": 5.1381,
      "mean_token_accuracy": 0.22855116575956344,
      "num_tokens": 110119217.0,
      "step": 56315
    },
    {
      "entropy": 5.761813926696777,
      "epoch": 5.629779577147998,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002338057735193892,
      "loss": 5.1853,
      "mean_token_accuracy": 0.22625238597393035,
      "num_tokens": 110128408.0,
      "step": 56320
    },
    {
      "entropy": 5.804086780548095,
      "epoch": 5.630279402209227,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002337706881888878,
      "loss": 5.1977,
      "mean_token_accuracy": 0.22499352991580962,
      "num_tokens": 110136887.0,
      "step": 56325
    },
    {
      "entropy": 5.864183187484741,
      "epoch": 5.630779227270455,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002337356038957053,
      "loss": 5.2198,
      "mean_token_accuracy": 0.21995292007923126,
      "num_tokens": 110146547.0,
      "step": 56330
    },
    {
      "entropy": 5.787262725830078,
      "epoch": 5.631279052331684,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002337005206407244,
      "loss": 5.1323,
      "mean_token_accuracy": 0.22794411927461625,
      "num_tokens": 110155287.0,
      "step": 56335
    },
    {
      "entropy": 5.772494983673096,
      "epoch": 5.631778877392913,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023366543842482773,
      "loss": 5.1615,
      "mean_token_accuracy": 0.2270076483488083,
      "num_tokens": 110164611.0,
      "step": 56340
    },
    {
      "entropy": 5.8104105472564695,
      "epoch": 5.632278702454141,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023363035724889802,
      "loss": 5.2403,
      "mean_token_accuracy": 0.22734760493040085,
      "num_tokens": 110174748.0,
      "step": 56345
    },
    {
      "entropy": 5.757756185531616,
      "epoch": 5.63277852751537,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023359527711381783,
      "loss": 5.1061,
      "mean_token_accuracy": 0.22993528246879577,
      "num_tokens": 110184636.0,
      "step": 56350
    },
    {
      "entropy": 5.774715948104858,
      "epoch": 5.633278352576598,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002335601980204699,
      "loss": 5.2147,
      "mean_token_accuracy": 0.2312001809477806,
      "num_tokens": 110194500.0,
      "step": 56355
    },
    {
      "entropy": 5.853496742248535,
      "epoch": 5.633778177637827,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002335251199697367,
      "loss": 5.2728,
      "mean_token_accuracy": 0.2183099150657654,
      "num_tokens": 110202877.0,
      "step": 56360
    },
    {
      "entropy": 5.863200712203979,
      "epoch": 5.634278002699055,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023349004296250076,
      "loss": 5.2705,
      "mean_token_accuracy": 0.2128261595964432,
      "num_tokens": 110213663.0,
      "step": 56365
    },
    {
      "entropy": 5.809752607345581,
      "epoch": 5.634777827760284,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023345496699964468,
      "loss": 5.2449,
      "mean_token_accuracy": 0.21532247215509415,
      "num_tokens": 110222667.0,
      "step": 56370
    },
    {
      "entropy": 5.880796718597412,
      "epoch": 5.635277652821513,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023341989208205095,
      "loss": 5.1536,
      "mean_token_accuracy": 0.2268523558974266,
      "num_tokens": 110232510.0,
      "step": 56375
    },
    {
      "entropy": 5.764061069488525,
      "epoch": 5.635777477882741,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023338481821060203,
      "loss": 5.0885,
      "mean_token_accuracy": 0.2293630063533783,
      "num_tokens": 110243969.0,
      "step": 56380
    },
    {
      "entropy": 5.765289449691773,
      "epoch": 5.636277302943969,
      "grad_norm": 1.125,
      "learning_rate": 0.00023334974538618048,
      "loss": 5.1521,
      "mean_token_accuracy": 0.22531746029853822,
      "num_tokens": 110253831.0,
      "step": 56385
    },
    {
      "entropy": 5.807953882217407,
      "epoch": 5.636777128005198,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023331467360966852,
      "loss": 5.2058,
      "mean_token_accuracy": 0.22290192693471908,
      "num_tokens": 110263377.0,
      "step": 56390
    },
    {
      "entropy": 5.782846164703369,
      "epoch": 5.637276953066427,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023327960288194866,
      "loss": 5.1223,
      "mean_token_accuracy": 0.22445946633815766,
      "num_tokens": 110272729.0,
      "step": 56395
    },
    {
      "entropy": 5.755256414413452,
      "epoch": 5.637776778127655,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023324453320390314,
      "loss": 5.1494,
      "mean_token_accuracy": 0.23662446886301042,
      "num_tokens": 110282980.0,
      "step": 56400
    },
    {
      "entropy": 5.766554450988769,
      "epoch": 5.638276603188884,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023320946457641457,
      "loss": 5.1772,
      "mean_token_accuracy": 0.23008035272359847,
      "num_tokens": 110293333.0,
      "step": 56405
    },
    {
      "entropy": 5.721682214736939,
      "epoch": 5.638776428250113,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002331743970003651,
      "loss": 5.1279,
      "mean_token_accuracy": 0.22598986178636551,
      "num_tokens": 110303220.0,
      "step": 56410
    },
    {
      "entropy": 5.708129358291626,
      "epoch": 5.639276253311341,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023313933047663693,
      "loss": 5.1456,
      "mean_token_accuracy": 0.23241690546274185,
      "num_tokens": 110314833.0,
      "step": 56415
    },
    {
      "entropy": 5.811185264587403,
      "epoch": 5.639776078372569,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023310426500611255,
      "loss": 5.082,
      "mean_token_accuracy": 0.23683447241783143,
      "num_tokens": 110324888.0,
      "step": 56420
    },
    {
      "entropy": 5.837525272369385,
      "epoch": 5.640275903433798,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023306920058967408,
      "loss": 5.1452,
      "mean_token_accuracy": 0.23075545579195023,
      "num_tokens": 110335241.0,
      "step": 56425
    },
    {
      "entropy": 5.780348014831543,
      "epoch": 5.640775728495027,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023303413722820365,
      "loss": 5.1232,
      "mean_token_accuracy": 0.21966684609651566,
      "num_tokens": 110344732.0,
      "step": 56430
    },
    {
      "entropy": 5.720935535430908,
      "epoch": 5.641275553556255,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002329990749225836,
      "loss": 5.1731,
      "mean_token_accuracy": 0.2329125389456749,
      "num_tokens": 110354261.0,
      "step": 56435
    },
    {
      "entropy": 5.750528478622437,
      "epoch": 5.641775378617484,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023296401367369597,
      "loss": 5.1541,
      "mean_token_accuracy": 0.2286365270614624,
      "num_tokens": 110363268.0,
      "step": 56440
    },
    {
      "entropy": 5.722906160354614,
      "epoch": 5.642275203678713,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023292895348242293,
      "loss": 5.1138,
      "mean_token_accuracy": 0.2351672351360321,
      "num_tokens": 110373058.0,
      "step": 56445
    },
    {
      "entropy": 5.678517913818359,
      "epoch": 5.6427750287399405,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023289389434964664,
      "loss": 5.0702,
      "mean_token_accuracy": 0.2392196387052536,
      "num_tokens": 110384041.0,
      "step": 56450
    },
    {
      "entropy": 5.827167797088623,
      "epoch": 5.643274853801169,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023285883627624922,
      "loss": 5.1458,
      "mean_token_accuracy": 0.22325925827026366,
      "num_tokens": 110393177.0,
      "step": 56455
    },
    {
      "entropy": 5.854349660873413,
      "epoch": 5.643774678862398,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023282377926311254,
      "loss": 5.2446,
      "mean_token_accuracy": 0.22182565778493882,
      "num_tokens": 110402096.0,
      "step": 56460
    },
    {
      "entropy": 5.745011854171753,
      "epoch": 5.644274503923627,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023278872331111868,
      "loss": 5.1432,
      "mean_token_accuracy": 0.22672748416662217,
      "num_tokens": 110412294.0,
      "step": 56465
    },
    {
      "entropy": 5.685494518280029,
      "epoch": 5.644774328984855,
      "grad_norm": 1.25,
      "learning_rate": 0.0002327536684211497,
      "loss": 5.0639,
      "mean_token_accuracy": 0.22658584713935853,
      "num_tokens": 110421824.0,
      "step": 56470
    },
    {
      "entropy": 5.7534912109375,
      "epoch": 5.645274154046084,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023271861459408754,
      "loss": 5.1277,
      "mean_token_accuracy": 0.23939052522182463,
      "num_tokens": 110431005.0,
      "step": 56475
    },
    {
      "entropy": 5.847652959823608,
      "epoch": 5.645773979107313,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023268356183081423,
      "loss": 5.1963,
      "mean_token_accuracy": 0.22397290170192719,
      "num_tokens": 110440630.0,
      "step": 56480
    },
    {
      "entropy": 5.769157361984253,
      "epoch": 5.6462738041685405,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023264851013221155,
      "loss": 5.1765,
      "mean_token_accuracy": 0.22225811779499055,
      "num_tokens": 110450200.0,
      "step": 56485
    },
    {
      "entropy": 5.7648584842681885,
      "epoch": 5.646773629229769,
      "grad_norm": 1.125,
      "learning_rate": 0.00023261345949916152,
      "loss": 5.1614,
      "mean_token_accuracy": 0.22321907132864,
      "num_tokens": 110461163.0,
      "step": 56490
    },
    {
      "entropy": 5.775086545944214,
      "epoch": 5.647273454290998,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023257840993254586,
      "loss": 5.2123,
      "mean_token_accuracy": 0.22257586270570756,
      "num_tokens": 110470644.0,
      "step": 56495
    },
    {
      "entropy": 5.64051513671875,
      "epoch": 5.647773279352227,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023254336143324657,
      "loss": 4.9522,
      "mean_token_accuracy": 0.24672135561704636,
      "num_tokens": 110479990.0,
      "step": 56500
    },
    {
      "entropy": 5.779343938827514,
      "epoch": 5.648273104413455,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023250831400214528,
      "loss": 5.218,
      "mean_token_accuracy": 0.22128663212060928,
      "num_tokens": 110490220.0,
      "step": 56505
    },
    {
      "entropy": 5.786227130889893,
      "epoch": 5.648772929474684,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023247326764012395,
      "loss": 5.2086,
      "mean_token_accuracy": 0.22279344499111176,
      "num_tokens": 110500227.0,
      "step": 56510
    },
    {
      "entropy": 5.786483240127564,
      "epoch": 5.649272754535913,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002324382223480642,
      "loss": 5.0573,
      "mean_token_accuracy": 0.23350415974855424,
      "num_tokens": 110510558.0,
      "step": 56515
    },
    {
      "entropy": 5.774311351776123,
      "epoch": 5.6497725795971405,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002324031781268479,
      "loss": 5.134,
      "mean_token_accuracy": 0.21979112327098846,
      "num_tokens": 110521531.0,
      "step": 56520
    },
    {
      "entropy": 5.815213298797607,
      "epoch": 5.650272404658369,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002323681349773567,
      "loss": 5.1878,
      "mean_token_accuracy": 0.23385708779096603,
      "num_tokens": 110532766.0,
      "step": 56525
    },
    {
      "entropy": 5.850406742095947,
      "epoch": 5.650772229719598,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002323330929004722,
      "loss": 5.2293,
      "mean_token_accuracy": 0.21736290454864501,
      "num_tokens": 110542792.0,
      "step": 56530
    },
    {
      "entropy": 5.827601909637451,
      "epoch": 5.651272054780827,
      "grad_norm": 1.25,
      "learning_rate": 0.00023229805189707616,
      "loss": 5.1951,
      "mean_token_accuracy": 0.2165027752518654,
      "num_tokens": 110553115.0,
      "step": 56535
    },
    {
      "entropy": 5.781102657318115,
      "epoch": 5.651771879842055,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023226301196805005,
      "loss": 5.174,
      "mean_token_accuracy": 0.22195959240198135,
      "num_tokens": 110561978.0,
      "step": 56540
    },
    {
      "entropy": 5.800445890426635,
      "epoch": 5.652271704903284,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023222797311427563,
      "loss": 5.1491,
      "mean_token_accuracy": 0.22088393121957778,
      "num_tokens": 110571773.0,
      "step": 56545
    },
    {
      "entropy": 5.874301242828369,
      "epoch": 5.652771529964513,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023219293533663439,
      "loss": 5.3017,
      "mean_token_accuracy": 0.21693543642759322,
      "num_tokens": 110581831.0,
      "step": 56550
    },
    {
      "entropy": 5.8563244342803955,
      "epoch": 5.6532713550257405,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023215789863600784,
      "loss": 5.1533,
      "mean_token_accuracy": 0.22290200591087342,
      "num_tokens": 110591053.0,
      "step": 56555
    },
    {
      "entropy": 5.779953098297119,
      "epoch": 5.653771180086969,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023212286301327757,
      "loss": 5.1931,
      "mean_token_accuracy": 0.22525351494550705,
      "num_tokens": 110601650.0,
      "step": 56560
    },
    {
      "entropy": 5.785735321044922,
      "epoch": 5.654271005148198,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023208782846932503,
      "loss": 5.2249,
      "mean_token_accuracy": 0.22044802755117415,
      "num_tokens": 110611663.0,
      "step": 56565
    },
    {
      "entropy": 5.812259197235107,
      "epoch": 5.654770830209427,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023205279500503162,
      "loss": 5.2007,
      "mean_token_accuracy": 0.22195306271314622,
      "num_tokens": 110622368.0,
      "step": 56570
    },
    {
      "entropy": 5.738581132888794,
      "epoch": 5.655270655270655,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023201776262127895,
      "loss": 5.1044,
      "mean_token_accuracy": 0.23256983309984208,
      "num_tokens": 110632446.0,
      "step": 56575
    },
    {
      "entropy": 5.7119073390960695,
      "epoch": 5.655770480331884,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002319827313189482,
      "loss": 5.0272,
      "mean_token_accuracy": 0.235806006193161,
      "num_tokens": 110642912.0,
      "step": 56580
    },
    {
      "entropy": 5.7092499256134035,
      "epoch": 5.656270305393113,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023194770109892095,
      "loss": 5.0603,
      "mean_token_accuracy": 0.239602954685688,
      "num_tokens": 110653192.0,
      "step": 56585
    },
    {
      "entropy": 5.773732900619507,
      "epoch": 5.6567701304543405,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023191267196207843,
      "loss": 5.1582,
      "mean_token_accuracy": 0.22364620715379716,
      "num_tokens": 110663299.0,
      "step": 56590
    },
    {
      "entropy": 5.723492622375488,
      "epoch": 5.657269955515569,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023187764390930206,
      "loss": 5.0044,
      "mean_token_accuracy": 0.2458738923072815,
      "num_tokens": 110672184.0,
      "step": 56595
    },
    {
      "entropy": 5.767896842956543,
      "epoch": 5.657769780576798,
      "grad_norm": 1.3125,
      "learning_rate": 0.000231842616941473,
      "loss": 5.182,
      "mean_token_accuracy": 0.22281378954648973,
      "num_tokens": 110681602.0,
      "step": 56600
    },
    {
      "entropy": 5.793287992477417,
      "epoch": 5.658269605638027,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023180759105947263,
      "loss": 5.1536,
      "mean_token_accuracy": 0.22033047974109649,
      "num_tokens": 110691101.0,
      "step": 56605
    },
    {
      "entropy": 5.775263833999634,
      "epoch": 5.658769430699255,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023177256626418207,
      "loss": 5.2066,
      "mean_token_accuracy": 0.22654726654291152,
      "num_tokens": 110700196.0,
      "step": 56610
    },
    {
      "entropy": 5.732609033584595,
      "epoch": 5.659269255760484,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002317375425564827,
      "loss": 5.108,
      "mean_token_accuracy": 0.23547713011503218,
      "num_tokens": 110710441.0,
      "step": 56615
    },
    {
      "entropy": 5.729812955856323,
      "epoch": 5.659769080821713,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002317025199372556,
      "loss": 5.0918,
      "mean_token_accuracy": 0.23469141721725464,
      "num_tokens": 110720835.0,
      "step": 56620
    },
    {
      "entropy": 5.753148794174194,
      "epoch": 5.6602689058829405,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023166749840738195,
      "loss": 5.1617,
      "mean_token_accuracy": 0.22357849031686783,
      "num_tokens": 110731343.0,
      "step": 56625
    },
    {
      "entropy": 5.835461711883545,
      "epoch": 5.660768730944169,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002316324779677429,
      "loss": 5.2305,
      "mean_token_accuracy": 0.22186302542686462,
      "num_tokens": 110741043.0,
      "step": 56630
    },
    {
      "entropy": 5.773494482040405,
      "epoch": 5.661268556005398,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023159745861921956,
      "loss": 5.19,
      "mean_token_accuracy": 0.22700108736753463,
      "num_tokens": 110750675.0,
      "step": 56635
    },
    {
      "entropy": 5.786199378967285,
      "epoch": 5.661768381066627,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023156244036269297,
      "loss": 5.2011,
      "mean_token_accuracy": 0.2161093071103096,
      "num_tokens": 110760839.0,
      "step": 56640
    },
    {
      "entropy": 5.769766473770142,
      "epoch": 5.662268206127855,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023152742319904423,
      "loss": 5.1376,
      "mean_token_accuracy": 0.22620550394058228,
      "num_tokens": 110770594.0,
      "step": 56645
    },
    {
      "entropy": 5.730558681488037,
      "epoch": 5.662768031189084,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002314924071291543,
      "loss": 5.0858,
      "mean_token_accuracy": 0.23665296137332917,
      "num_tokens": 110780059.0,
      "step": 56650
    },
    {
      "entropy": 5.815853071212769,
      "epoch": 5.663267856250313,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002314573921539042,
      "loss": 5.2329,
      "mean_token_accuracy": 0.21829502731561662,
      "num_tokens": 110789000.0,
      "step": 56655
    },
    {
      "entropy": 5.732690715789795,
      "epoch": 5.6637676813115405,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000231422378274175,
      "loss": 5.1376,
      "mean_token_accuracy": 0.22746024578809737,
      "num_tokens": 110797881.0,
      "step": 56660
    },
    {
      "entropy": 5.764768171310425,
      "epoch": 5.664267506372769,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002313873654908475,
      "loss": 5.2057,
      "mean_token_accuracy": 0.2255283921957016,
      "num_tokens": 110807645.0,
      "step": 56665
    },
    {
      "entropy": 5.652079343795776,
      "epoch": 5.664767331433998,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002313523538048027,
      "loss": 5.0306,
      "mean_token_accuracy": 0.2386926680803299,
      "num_tokens": 110816940.0,
      "step": 56670
    },
    {
      "entropy": 5.819542980194091,
      "epoch": 5.665267156495227,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002313173432169214,
      "loss": 5.1844,
      "mean_token_accuracy": 0.2220208078622818,
      "num_tokens": 110826507.0,
      "step": 56675
    },
    {
      "entropy": 5.789258146286011,
      "epoch": 5.665766981556455,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002312823337280845,
      "loss": 5.1022,
      "mean_token_accuracy": 0.23163911551237107,
      "num_tokens": 110835161.0,
      "step": 56680
    },
    {
      "entropy": 5.738721942901611,
      "epoch": 5.666266806617684,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023124732533917275,
      "loss": 5.054,
      "mean_token_accuracy": 0.2407779097557068,
      "num_tokens": 110844835.0,
      "step": 56685
    },
    {
      "entropy": 5.871937894821167,
      "epoch": 5.666766631678913,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023121231805106713,
      "loss": 5.3177,
      "mean_token_accuracy": 0.21547385603189467,
      "num_tokens": 110855327.0,
      "step": 56690
    },
    {
      "entropy": 5.738048505783081,
      "epoch": 5.6672664567401405,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002311773118646483,
      "loss": 5.0657,
      "mean_token_accuracy": 0.24296817183494568,
      "num_tokens": 110865018.0,
      "step": 56695
    },
    {
      "entropy": 5.81215271949768,
      "epoch": 5.667766281801369,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023114230678079707,
      "loss": 5.2596,
      "mean_token_accuracy": 0.2152879223227501,
      "num_tokens": 110874139.0,
      "step": 56700
    },
    {
      "entropy": 5.700995254516601,
      "epoch": 5.668266106862598,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023110730280039406,
      "loss": 5.0624,
      "mean_token_accuracy": 0.23371023684740067,
      "num_tokens": 110884874.0,
      "step": 56705
    },
    {
      "entropy": 5.794427061080933,
      "epoch": 5.668765931923827,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023107229992432005,
      "loss": 5.165,
      "mean_token_accuracy": 0.22078042328357697,
      "num_tokens": 110894089.0,
      "step": 56710
    },
    {
      "entropy": 5.8138669490814205,
      "epoch": 5.669265756985055,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023103729815345564,
      "loss": 5.1933,
      "mean_token_accuracy": 0.22645775526762008,
      "num_tokens": 110903113.0,
      "step": 56715
    },
    {
      "entropy": 5.872471523284912,
      "epoch": 5.669765582046284,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002310022974886815,
      "loss": 5.2587,
      "mean_token_accuracy": 0.2172664523124695,
      "num_tokens": 110913037.0,
      "step": 56720
    },
    {
      "entropy": 5.75441164970398,
      "epoch": 5.670265407107513,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023096729793087824,
      "loss": 5.1655,
      "mean_token_accuracy": 0.22719181030988694,
      "num_tokens": 110923546.0,
      "step": 56725
    },
    {
      "entropy": 5.789556884765625,
      "epoch": 5.6707652321687405,
      "grad_norm": 1.125,
      "learning_rate": 0.00023093229948092654,
      "loss": 5.1716,
      "mean_token_accuracy": 0.21951095014810562,
      "num_tokens": 110933532.0,
      "step": 56730
    },
    {
      "entropy": 5.775363492965698,
      "epoch": 5.671265057229969,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002308973021397068,
      "loss": 5.1316,
      "mean_token_accuracy": 0.23460191041231154,
      "num_tokens": 110943078.0,
      "step": 56735
    },
    {
      "entropy": 5.7419086456298825,
      "epoch": 5.671764882291198,
      "grad_norm": 1.203125,
      "learning_rate": 0.00023086230590809947,
      "loss": 5.1517,
      "mean_token_accuracy": 0.22247854471206666,
      "num_tokens": 110952519.0,
      "step": 56740
    },
    {
      "entropy": 5.701023721694947,
      "epoch": 5.672264707352427,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023082731078698528,
      "loss": 5.0588,
      "mean_token_accuracy": 0.24010335505008698,
      "num_tokens": 110963393.0,
      "step": 56745
    },
    {
      "entropy": 5.802209281921387,
      "epoch": 5.672764532413655,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002307923167772445,
      "loss": 5.2329,
      "mean_token_accuracy": 0.22059959471225737,
      "num_tokens": 110974429.0,
      "step": 56750
    },
    {
      "entropy": 5.8135683059692385,
      "epoch": 5.673264357474884,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023075732387975768,
      "loss": 5.2248,
      "mean_token_accuracy": 0.22507320940494538,
      "num_tokens": 110984844.0,
      "step": 56755
    },
    {
      "entropy": 5.882115364074707,
      "epoch": 5.673764182536113,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023072233209540523,
      "loss": 5.2478,
      "mean_token_accuracy": 0.22081060111522674,
      "num_tokens": 110993737.0,
      "step": 56760
    },
    {
      "entropy": 5.775233173370362,
      "epoch": 5.6742640075973405,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023068734142506748,
      "loss": 5.0544,
      "mean_token_accuracy": 0.22330053597688676,
      "num_tokens": 111003578.0,
      "step": 56765
    },
    {
      "entropy": 5.827782678604126,
      "epoch": 5.674763832658569,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023065235186962485,
      "loss": 5.2148,
      "mean_token_accuracy": 0.21858923733234406,
      "num_tokens": 111012588.0,
      "step": 56770
    },
    {
      "entropy": 5.784835767745972,
      "epoch": 5.675263657719798,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002306173634299576,
      "loss": 5.1654,
      "mean_token_accuracy": 0.2287621334195137,
      "num_tokens": 111021708.0,
      "step": 56775
    },
    {
      "entropy": 5.746915292739868,
      "epoch": 5.675763482781027,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002305823761069461,
      "loss": 5.1336,
      "mean_token_accuracy": 0.22708529680967332,
      "num_tokens": 111031845.0,
      "step": 56780
    },
    {
      "entropy": 5.835564708709716,
      "epoch": 5.676263307842255,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023054738990147057,
      "loss": 5.2979,
      "mean_token_accuracy": 0.22104129940271378,
      "num_tokens": 111041994.0,
      "step": 56785
    },
    {
      "entropy": 5.813997554779053,
      "epoch": 5.676763132903484,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023051240481441126,
      "loss": 5.2105,
      "mean_token_accuracy": 0.220838563144207,
      "num_tokens": 111050958.0,
      "step": 56790
    },
    {
      "entropy": 5.689022541046143,
      "epoch": 5.677262957964713,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023047742084664842,
      "loss": 5.0922,
      "mean_token_accuracy": 0.23935133963823318,
      "num_tokens": 111060967.0,
      "step": 56795
    },
    {
      "entropy": 5.827060985565185,
      "epoch": 5.6777627830259405,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023044243799906229,
      "loss": 5.1595,
      "mean_token_accuracy": 0.22860365808010102,
      "num_tokens": 111070121.0,
      "step": 56800
    },
    {
      "entropy": 5.825118207931519,
      "epoch": 5.678262608087169,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002304074562725329,
      "loss": 5.1762,
      "mean_token_accuracy": 0.22842006087303163,
      "num_tokens": 111078962.0,
      "step": 56805
    },
    {
      "entropy": 5.7704243659973145,
      "epoch": 5.678762433148398,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023037247566794035,
      "loss": 5.1494,
      "mean_token_accuracy": 0.23176665902137755,
      "num_tokens": 111089350.0,
      "step": 56810
    },
    {
      "entropy": 5.848315954208374,
      "epoch": 5.679262258209627,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002303374961861649,
      "loss": 5.2152,
      "mean_token_accuracy": 0.21993256658315657,
      "num_tokens": 111099970.0,
      "step": 56815
    },
    {
      "entropy": 5.762026691436768,
      "epoch": 5.679762083270855,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023030251782808653,
      "loss": 5.1432,
      "mean_token_accuracy": 0.22699429839849472,
      "num_tokens": 111110489.0,
      "step": 56820
    },
    {
      "entropy": 5.775433349609375,
      "epoch": 5.680261908332084,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023026754059458533,
      "loss": 5.1859,
      "mean_token_accuracy": 0.22680921405553817,
      "num_tokens": 111120347.0,
      "step": 56825
    },
    {
      "entropy": 5.7408246994018555,
      "epoch": 5.680761733393313,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023023256448654128,
      "loss": 5.1525,
      "mean_token_accuracy": 0.22343551069498063,
      "num_tokens": 111132029.0,
      "step": 56830
    },
    {
      "entropy": 5.854939699172974,
      "epoch": 5.6812615584545405,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023019758950483442,
      "loss": 5.195,
      "mean_token_accuracy": 0.21845096498727798,
      "num_tokens": 111141348.0,
      "step": 56835
    },
    {
      "entropy": 5.786709880828857,
      "epoch": 5.681761383515769,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023016261565034462,
      "loss": 5.1066,
      "mean_token_accuracy": 0.22745821326971055,
      "num_tokens": 111151359.0,
      "step": 56840
    },
    {
      "entropy": 5.68643741607666,
      "epoch": 5.682261208576998,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023012764292395195,
      "loss": 5.0933,
      "mean_token_accuracy": 0.2336590975522995,
      "num_tokens": 111160779.0,
      "step": 56845
    },
    {
      "entropy": 5.817161321640015,
      "epoch": 5.682761033638227,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023009267132653616,
      "loss": 5.1895,
      "mean_token_accuracy": 0.22340985238552094,
      "num_tokens": 111170429.0,
      "step": 56850
    },
    {
      "entropy": 5.79976863861084,
      "epoch": 5.683260858699455,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002300577008589772,
      "loss": 5.1839,
      "mean_token_accuracy": 0.21735592037439347,
      "num_tokens": 111179926.0,
      "step": 56855
    },
    {
      "entropy": 5.793803071975708,
      "epoch": 5.683760683760684,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023002273152215497,
      "loss": 5.1792,
      "mean_token_accuracy": 0.21889862567186355,
      "num_tokens": 111189641.0,
      "step": 56860
    },
    {
      "entropy": 5.757396173477173,
      "epoch": 5.684260508821913,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022998776331694927,
      "loss": 5.1245,
      "mean_token_accuracy": 0.2265615865588188,
      "num_tokens": 111200047.0,
      "step": 56865
    },
    {
      "entropy": 5.797581386566162,
      "epoch": 5.6847603338831405,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022995279624423988,
      "loss": 5.195,
      "mean_token_accuracy": 0.22285084873437883,
      "num_tokens": 111210426.0,
      "step": 56870
    },
    {
      "entropy": 5.737787532806396,
      "epoch": 5.685260158944369,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022991783030490648,
      "loss": 5.0762,
      "mean_token_accuracy": 0.23314360082149505,
      "num_tokens": 111220686.0,
      "step": 56875
    },
    {
      "entropy": 5.823957920074463,
      "epoch": 5.685759984005598,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002298828654998289,
      "loss": 5.2172,
      "mean_token_accuracy": 0.2275883972644806,
      "num_tokens": 111230829.0,
      "step": 56880
    },
    {
      "entropy": 5.7728245735168455,
      "epoch": 5.686259809066827,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002298479018298868,
      "loss": 5.1978,
      "mean_token_accuracy": 0.22092695087194442,
      "num_tokens": 111241450.0,
      "step": 56885
    },
    {
      "entropy": 5.767762231826782,
      "epoch": 5.686759634128055,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022981293929595986,
      "loss": 5.1601,
      "mean_token_accuracy": 0.22135827988386153,
      "num_tokens": 111251056.0,
      "step": 56890
    },
    {
      "entropy": 5.804527473449707,
      "epoch": 5.687259459189284,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022977797789892773,
      "loss": 5.179,
      "mean_token_accuracy": 0.2288795977830887,
      "num_tokens": 111260823.0,
      "step": 56895
    },
    {
      "entropy": 5.709134483337403,
      "epoch": 5.687759284250513,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022974301763967002,
      "loss": 5.017,
      "mean_token_accuracy": 0.23878639340400695,
      "num_tokens": 111271998.0,
      "step": 56900
    },
    {
      "entropy": 5.812494945526123,
      "epoch": 5.6882591093117405,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002297080585190664,
      "loss": 5.1945,
      "mean_token_accuracy": 0.22736174166202544,
      "num_tokens": 111281153.0,
      "step": 56905
    },
    {
      "entropy": 5.75702805519104,
      "epoch": 5.688758934372969,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022967310053799628,
      "loss": 5.1371,
      "mean_token_accuracy": 0.23373734802007676,
      "num_tokens": 111291055.0,
      "step": 56910
    },
    {
      "entropy": 5.726827144622803,
      "epoch": 5.689258759434198,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022963814369733937,
      "loss": 5.1031,
      "mean_token_accuracy": 0.23279448002576827,
      "num_tokens": 111300425.0,
      "step": 56915
    },
    {
      "entropy": 5.767966985702515,
      "epoch": 5.689758584495427,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002296031879979751,
      "loss": 5.1171,
      "mean_token_accuracy": 0.23333743661642076,
      "num_tokens": 111311810.0,
      "step": 56920
    },
    {
      "entropy": 5.741613817214966,
      "epoch": 5.690258409556655,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022956823344078293,
      "loss": 5.1027,
      "mean_token_accuracy": 0.230074642598629,
      "num_tokens": 111321637.0,
      "step": 56925
    },
    {
      "entropy": 5.787389755249023,
      "epoch": 5.690758234617884,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002295332800266423,
      "loss": 5.1639,
      "mean_token_accuracy": 0.22594217360019683,
      "num_tokens": 111330816.0,
      "step": 56930
    },
    {
      "entropy": 5.760404014587403,
      "epoch": 5.691258059679113,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022949832775643263,
      "loss": 5.1308,
      "mean_token_accuracy": 0.23826132863759994,
      "num_tokens": 111339871.0,
      "step": 56935
    },
    {
      "entropy": 5.839557695388794,
      "epoch": 5.6917578847403405,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002294633766310334,
      "loss": 5.2327,
      "mean_token_accuracy": 0.2245660826563835,
      "num_tokens": 111349045.0,
      "step": 56940
    },
    {
      "entropy": 5.803902673721313,
      "epoch": 5.692257709801569,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022942842665132386,
      "loss": 5.2233,
      "mean_token_accuracy": 0.22445495575666427,
      "num_tokens": 111358381.0,
      "step": 56945
    },
    {
      "entropy": 5.76609320640564,
      "epoch": 5.692757534862798,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022939347781818337,
      "loss": 5.2129,
      "mean_token_accuracy": 0.2265796422958374,
      "num_tokens": 111369098.0,
      "step": 56950
    },
    {
      "entropy": 5.812733459472656,
      "epoch": 5.693257359924027,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002293585301324912,
      "loss": 5.1667,
      "mean_token_accuracy": 0.2211843892931938,
      "num_tokens": 111380563.0,
      "step": 56955
    },
    {
      "entropy": 5.818071746826172,
      "epoch": 5.693757184985255,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022932358359512673,
      "loss": 5.1647,
      "mean_token_accuracy": 0.22788304537534715,
      "num_tokens": 111389831.0,
      "step": 56960
    },
    {
      "entropy": 5.804504537582398,
      "epoch": 5.694257010046484,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022928863820696912,
      "loss": 5.1477,
      "mean_token_accuracy": 0.22797502875328063,
      "num_tokens": 111399301.0,
      "step": 56965
    },
    {
      "entropy": 5.741389656066895,
      "epoch": 5.694756835107713,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002292536939688975,
      "loss": 5.1327,
      "mean_token_accuracy": 0.2416177436709404,
      "num_tokens": 111410590.0,
      "step": 56970
    },
    {
      "entropy": 5.773977327346802,
      "epoch": 5.6952566601689405,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002292187508817913,
      "loss": 5.1767,
      "mean_token_accuracy": 0.22800842374563218,
      "num_tokens": 111419715.0,
      "step": 56975
    },
    {
      "entropy": 5.74913067817688,
      "epoch": 5.695756485230169,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022918380894652952,
      "loss": 5.129,
      "mean_token_accuracy": 0.2367909461259842,
      "num_tokens": 111429435.0,
      "step": 56980
    },
    {
      "entropy": 5.7056783676147464,
      "epoch": 5.696256310291398,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022914886816399133,
      "loss": 5.0447,
      "mean_token_accuracy": 0.23927054703235626,
      "num_tokens": 111439490.0,
      "step": 56985
    },
    {
      "entropy": 5.832071352005005,
      "epoch": 5.696756135352627,
      "grad_norm": 1.25,
      "learning_rate": 0.0002291139285350558,
      "loss": 5.2107,
      "mean_token_accuracy": 0.22058423906564711,
      "num_tokens": 111449393.0,
      "step": 56990
    },
    {
      "entropy": 5.886922407150268,
      "epoch": 5.697255960413855,
      "grad_norm": 1.25,
      "learning_rate": 0.00022907899006060202,
      "loss": 5.2369,
      "mean_token_accuracy": 0.21760540902614595,
      "num_tokens": 111458465.0,
      "step": 56995
    },
    {
      "entropy": 5.769917058944702,
      "epoch": 5.697755785475084,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022904405274150902,
      "loss": 5.1094,
      "mean_token_accuracy": 0.22963700592517852,
      "num_tokens": 111468293.0,
      "step": 57000
    },
    {
      "epoch": 5.697755785475084,
      "eval_entropy": 5.527711975423984,
      "eval_loss": 5.342863082885742,
      "eval_mean_token_accuracy": 0.22514434524323945,
      "eval_num_tokens": 111468293.0,
      "eval_runtime": 23.8693,
      "eval_samples_per_second": 1300.622,
      "eval_steps_per_second": 162.593,
      "step": 57000
    }
  ],
  "logging_steps": 5,
  "max_steps": 100030,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.7679259763712e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}