tur-latn-100mb-after-ppt-shuff-dyck-100mb-ckpt500_seed10/checkpoint-21000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0991652921477484,
  "eval_steps": 3000,
  "global_step": 21000,
  "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
    }
  ],
  "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": 6.52103258112e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}