swe-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407/checkpoint-10500/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.699254349627175,
  "eval_steps": 500,
  "global_step": 10500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.70417890548706,
      "epoch": 0.004142502071251036,
      "grad_norm": 18.0,
      "learning_rate": 2e-06,
      "loss": 14.324,
      "mean_token_accuracy": 9.699321235530078e-05,
      "num_tokens": 9149.0,
      "step": 5
    },
    {
      "entropy": 4.722008132934571,
      "epoch": 0.008285004142502071,
      "grad_norm": 19.5,
      "learning_rate": 4.5e-06,
      "loss": 14.1689,
      "mean_token_accuracy": 0.0,
      "num_tokens": 16800.0,
      "step": 10
    },
    {
      "entropy": 4.784619331359863,
      "epoch": 0.012427506213753107,
      "grad_norm": 23.875,
      "learning_rate": 7e-06,
      "loss": 13.9756,
      "mean_token_accuracy": 0.00012360939290374519,
      "num_tokens": 26617.0,
      "step": 15
    },
    {
      "entropy": 5.349014616012573,
      "epoch": 0.016570008285004142,
      "grad_norm": 30.125,
      "learning_rate": 9.5e-06,
      "loss": 13.1127,
      "mean_token_accuracy": 0.0,
      "num_tokens": 35855.0,
      "step": 20
    },
    {
      "entropy": 8.763514041900635,
      "epoch": 0.020712510356255178,
      "grad_norm": 6.3125,
      "learning_rate": 1.2e-05,
      "loss": 11.2931,
      "mean_token_accuracy": 0.0002482298179529607,
      "num_tokens": 45644.0,
      "step": 25
    },
    {
      "entropy": 10.648772144317627,
      "epoch": 0.024855012427506214,
      "grad_norm": 3.203125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.6872,
      "mean_token_accuracy": 0.014276380836963653,
      "num_tokens": 54641.0,
      "step": 30
    },
    {
      "entropy": 10.69749927520752,
      "epoch": 0.02899751449875725,
      "grad_norm": 3.109375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.4448,
      "mean_token_accuracy": 0.018782001547515394,
      "num_tokens": 63924.0,
      "step": 35
    },
    {
      "entropy": 10.672551250457763,
      "epoch": 0.033140016570008285,
      "grad_norm": 2.421875,
      "learning_rate": 1.95e-05,
      "loss": 10.1617,
      "mean_token_accuracy": 0.02377438023686409,
      "num_tokens": 73452.0,
      "step": 40
    },
    {
      "entropy": 10.593613147735596,
      "epoch": 0.037282518641259324,
      "grad_norm": 2.375,
      "learning_rate": 2.2e-05,
      "loss": 9.8966,
      "mean_token_accuracy": 0.039151473343372344,
      "num_tokens": 82731.0,
      "step": 45
    },
    {
      "entropy": 10.570558643341064,
      "epoch": 0.041425020712510356,
      "grad_norm": 2.1875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.7261,
      "mean_token_accuracy": 0.04260772801935673,
      "num_tokens": 92626.0,
      "step": 50
    },
    {
      "entropy": 10.598753547668457,
      "epoch": 0.045567522783761395,
      "grad_norm": 2.0,
      "learning_rate": 2.7e-05,
      "loss": 9.669,
      "mean_token_accuracy": 0.04115805197507143,
      "num_tokens": 101380.0,
      "step": 55
    },
    {
      "entropy": 10.600215435028076,
      "epoch": 0.04971002485501243,
      "grad_norm": 1.9453125,
      "learning_rate": 2.95e-05,
      "loss": 9.5621,
      "mean_token_accuracy": 0.041873040795326236,
      "num_tokens": 109895.0,
      "step": 60
    },
    {
      "entropy": 10.531742191314697,
      "epoch": 0.053852526926263466,
      "grad_norm": 2.015625,
      "learning_rate": 3.2e-05,
      "loss": 9.5289,
      "mean_token_accuracy": 0.04000609517097473,
      "num_tokens": 119203.0,
      "step": 65
    },
    {
      "entropy": 10.568301486968995,
      "epoch": 0.0579950289975145,
      "grad_norm": 2.3125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.3923,
      "mean_token_accuracy": 0.03999764695763588,
      "num_tokens": 127572.0,
      "step": 70
    },
    {
      "entropy": 10.564670372009278,
      "epoch": 0.06213753106876554,
      "grad_norm": 1.8671875,
      "learning_rate": 3.7e-05,
      "loss": 9.3471,
      "mean_token_accuracy": 0.042604190111160276,
      "num_tokens": 137161.0,
      "step": 75
    },
    {
      "entropy": 10.54567642211914,
      "epoch": 0.06628003314001657,
      "grad_norm": 1.921875,
      "learning_rate": 3.95e-05,
      "loss": 9.2877,
      "mean_token_accuracy": 0.04093840681016445,
      "num_tokens": 145991.0,
      "step": 80
    },
    {
      "entropy": 10.513956260681152,
      "epoch": 0.07042253521126761,
      "grad_norm": 1.8125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.1852,
      "mean_token_accuracy": 0.041304165683686736,
      "num_tokens": 154927.0,
      "step": 85
    },
    {
      "entropy": 10.496724605560303,
      "epoch": 0.07456503728251865,
      "grad_norm": 2.0,
      "learning_rate": 4.45e-05,
      "loss": 9.0189,
      "mean_token_accuracy": 0.046985261142253876,
      "num_tokens": 163234.0,
      "step": 90
    },
    {
      "entropy": 10.47362985610962,
      "epoch": 0.07870753935376967,
      "grad_norm": 1.890625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.9566,
      "mean_token_accuracy": 0.05403314344584942,
      "num_tokens": 171554.0,
      "step": 95
    },
    {
      "entropy": 10.420413589477539,
      "epoch": 0.08285004142502071,
      "grad_norm": 1.7890625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.8624,
      "mean_token_accuracy": 0.0495030127465725,
      "num_tokens": 180171.0,
      "step": 100
    },
    {
      "entropy": 10.291831016540527,
      "epoch": 0.08699254349627175,
      "grad_norm": 1.7890625,
      "learning_rate": 5.2e-05,
      "loss": 8.7136,
      "mean_token_accuracy": 0.05688973888754845,
      "num_tokens": 189216.0,
      "step": 105
    },
    {
      "entropy": 10.206434440612792,
      "epoch": 0.09113504556752279,
      "grad_norm": 1.5546875,
      "learning_rate": 5.45e-05,
      "loss": 8.735,
      "mean_token_accuracy": 0.05266922526061535,
      "num_tokens": 200022.0,
      "step": 110
    },
    {
      "entropy": 10.191181755065918,
      "epoch": 0.09527754763877382,
      "grad_norm": 1.65625,
      "learning_rate": 5.7e-05,
      "loss": 8.5465,
      "mean_token_accuracy": 0.05359771586954594,
      "num_tokens": 208686.0,
      "step": 115
    },
    {
      "entropy": 10.06379051208496,
      "epoch": 0.09942004971002485,
      "grad_norm": 1.7265625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.4093,
      "mean_token_accuracy": 0.054267140850424764,
      "num_tokens": 218266.0,
      "step": 120
    },
    {
      "entropy": 9.88507251739502,
      "epoch": 0.1035625517812759,
      "grad_norm": 1.4375,
      "learning_rate": 6.2e-05,
      "loss": 8.2985,
      "mean_token_accuracy": 0.04877174161374569,
      "num_tokens": 227658.0,
      "step": 125
    },
    {
      "entropy": 9.759512615203857,
      "epoch": 0.10770505385252693,
      "grad_norm": 1.21875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.0999,
      "mean_token_accuracy": 0.06026250720024109,
      "num_tokens": 236301.0,
      "step": 130
    },
    {
      "entropy": 9.516020011901855,
      "epoch": 0.11184755592377796,
      "grad_norm": 1.21875,
      "learning_rate": 6.7e-05,
      "loss": 8.1214,
      "mean_token_accuracy": 0.04890455044806004,
      "num_tokens": 246355.0,
      "step": 135
    },
    {
      "entropy": 9.335341739654542,
      "epoch": 0.115990057995029,
      "grad_norm": 1.09375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.9629,
      "mean_token_accuracy": 0.05138283483684063,
      "num_tokens": 256023.0,
      "step": 140
    },
    {
      "entropy": 9.04491310119629,
      "epoch": 0.12013256006628004,
      "grad_norm": 0.93359375,
      "learning_rate": 7.2e-05,
      "loss": 7.9599,
      "mean_token_accuracy": 0.046007866784930226,
      "num_tokens": 266893.0,
      "step": 145
    },
    {
      "entropy": 8.849730587005615,
      "epoch": 0.12427506213753108,
      "grad_norm": 0.90234375,
      "learning_rate": 7.45e-05,
      "loss": 7.7312,
      "mean_token_accuracy": 0.0518551766872406,
      "num_tokens": 276075.0,
      "step": 150
    },
    {
      "entropy": 8.576015663146972,
      "epoch": 0.12841756420878211,
      "grad_norm": 1.03125,
      "learning_rate": 7.7e-05,
      "loss": 7.7017,
      "mean_token_accuracy": 0.056488437578082085,
      "num_tokens": 285529.0,
      "step": 155
    },
    {
      "entropy": 8.416762542724609,
      "epoch": 0.13256006628003314,
      "grad_norm": 0.8046875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.686,
      "mean_token_accuracy": 0.05418187938630581,
      "num_tokens": 294403.0,
      "step": 160
    },
    {
      "entropy": 8.284829902648926,
      "epoch": 0.13670256835128416,
      "grad_norm": 1.03125,
      "learning_rate": 8.2e-05,
      "loss": 7.712,
      "mean_token_accuracy": 0.05654969438910484,
      "num_tokens": 303507.0,
      "step": 165
    },
    {
      "entropy": 8.158583068847657,
      "epoch": 0.14084507042253522,
      "grad_norm": 1.015625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.5667,
      "mean_token_accuracy": 0.0604190930724144,
      "num_tokens": 312393.0,
      "step": 170
    },
    {
      "entropy": 8.053034448623658,
      "epoch": 0.14498757249378624,
      "grad_norm": 1.015625,
      "learning_rate": 8.7e-05,
      "loss": 7.544,
      "mean_token_accuracy": 0.06070729456841946,
      "num_tokens": 321309.0,
      "step": 175
    },
    {
      "entropy": 8.106309843063354,
      "epoch": 0.1491300745650373,
      "grad_norm": 1.6015625,
      "learning_rate": 8.95e-05,
      "loss": 7.6075,
      "mean_token_accuracy": 0.05758780501782894,
      "num_tokens": 330217.0,
      "step": 180
    },
    {
      "entropy": 8.025533390045165,
      "epoch": 0.15327257663628832,
      "grad_norm": 0.94921875,
      "learning_rate": 9.2e-05,
      "loss": 7.5973,
      "mean_token_accuracy": 0.06240675076842308,
      "num_tokens": 339587.0,
      "step": 185
    },
    {
      "entropy": 7.989250421524048,
      "epoch": 0.15741507870753935,
      "grad_norm": 1.0703125,
      "learning_rate": 9.45e-05,
      "loss": 7.5142,
      "mean_token_accuracy": 0.06207436956465244,
      "num_tokens": 349367.0,
      "step": 190
    },
    {
      "entropy": 7.934307050704956,
      "epoch": 0.1615575807787904,
      "grad_norm": 0.9609375,
      "learning_rate": 9.7e-05,
      "loss": 7.4192,
      "mean_token_accuracy": 0.06271038055419922,
      "num_tokens": 358272.0,
      "step": 195
    },
    {
      "entropy": 7.99296875,
      "epoch": 0.16570008285004142,
      "grad_norm": 1.0703125,
      "learning_rate": 9.95e-05,
      "loss": 7.6202,
      "mean_token_accuracy": 0.06270747631788254,
      "num_tokens": 367709.0,
      "step": 200
    },
    {
      "entropy": 7.9447160243988035,
      "epoch": 0.16984258492129245,
      "grad_norm": 1.28125,
      "learning_rate": 0.000102,
      "loss": 7.4595,
      "mean_token_accuracy": 0.06487247794866562,
      "num_tokens": 376180.0,
      "step": 205
    },
    {
      "entropy": 7.915022087097168,
      "epoch": 0.1739850869925435,
      "grad_norm": 1.09375,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.574,
      "mean_token_accuracy": 0.06149500273168087,
      "num_tokens": 385636.0,
      "step": 210
    },
    {
      "entropy": 7.981282186508179,
      "epoch": 0.17812758906379453,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000107,
      "loss": 7.4416,
      "mean_token_accuracy": 0.06958327889442444,
      "num_tokens": 394486.0,
      "step": 215
    },
    {
      "entropy": 7.828207683563233,
      "epoch": 0.18227009113504558,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001095,
      "loss": 7.4059,
      "mean_token_accuracy": 0.05610249787569046,
      "num_tokens": 403765.0,
      "step": 220
    },
    {
      "entropy": 7.8190673828125,
      "epoch": 0.1864125932062966,
      "grad_norm": 1.390625,
      "learning_rate": 0.000112,
      "loss": 7.3815,
      "mean_token_accuracy": 0.06853485442698001,
      "num_tokens": 412988.0,
      "step": 225
    },
    {
      "entropy": 7.780495929718017,
      "epoch": 0.19055509527754763,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0001145,
      "loss": 7.3492,
      "mean_token_accuracy": 0.06311094798147679,
      "num_tokens": 422096.0,
      "step": 230
    },
    {
      "entropy": 7.860241508483886,
      "epoch": 0.19469759734879868,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.4279,
      "mean_token_accuracy": 0.0612465362995863,
      "num_tokens": 431333.0,
      "step": 235
    },
    {
      "entropy": 7.842223691940307,
      "epoch": 0.1988400994200497,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.4979,
      "mean_token_accuracy": 0.06898829713463783,
      "num_tokens": 440382.0,
      "step": 240
    },
    {
      "entropy": 7.790421056747436,
      "epoch": 0.20298260149130073,
      "grad_norm": 1.484375,
      "learning_rate": 0.000122,
      "loss": 7.3766,
      "mean_token_accuracy": 0.07021775655448437,
      "num_tokens": 448469.0,
      "step": 245
    },
    {
      "entropy": 7.793370962142944,
      "epoch": 0.2071251035625518,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001245,
      "loss": 7.4374,
      "mean_token_accuracy": 0.0661789108067751,
      "num_tokens": 457356.0,
      "step": 250
    },
    {
      "entropy": 7.799811267852784,
      "epoch": 0.2112676056338028,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000127,
      "loss": 7.4418,
      "mean_token_accuracy": 0.0664214115589857,
      "num_tokens": 465928.0,
      "step": 255
    },
    {
      "entropy": 7.788989782333374,
      "epoch": 0.21541010770505387,
      "grad_norm": 0.890625,
      "learning_rate": 0.0001295,
      "loss": 7.3311,
      "mean_token_accuracy": 0.06980733200907707,
      "num_tokens": 475269.0,
      "step": 260
    },
    {
      "entropy": 7.780997276306152,
      "epoch": 0.2195526097763049,
      "grad_norm": 1.640625,
      "learning_rate": 0.000132,
      "loss": 7.295,
      "mean_token_accuracy": 0.06686929501593113,
      "num_tokens": 485450.0,
      "step": 265
    },
    {
      "entropy": 7.6190821647644045,
      "epoch": 0.22369511184755592,
      "grad_norm": 1.109375,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.3345,
      "mean_token_accuracy": 0.07399698086082936,
      "num_tokens": 494029.0,
      "step": 270
    },
    {
      "entropy": 7.713169765472412,
      "epoch": 0.22783761391880697,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.294,
      "mean_token_accuracy": 0.07526539191603661,
      "num_tokens": 503253.0,
      "step": 275
    },
    {
      "entropy": 7.728218841552734,
      "epoch": 0.231980115990058,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001395,
      "loss": 7.2987,
      "mean_token_accuracy": 0.07004385031759738,
      "num_tokens": 512857.0,
      "step": 280
    },
    {
      "entropy": 7.657222604751587,
      "epoch": 0.23612261806130902,
      "grad_norm": 1.125,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.3416,
      "mean_token_accuracy": 0.07092855796217919,
      "num_tokens": 522319.0,
      "step": 285
    },
    {
      "entropy": 7.699072313308716,
      "epoch": 0.24026512013256007,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001445,
      "loss": 7.282,
      "mean_token_accuracy": 0.07054338790476322,
      "num_tokens": 531469.0,
      "step": 290
    },
    {
      "entropy": 7.673645305633545,
      "epoch": 0.2444076222038111,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000147,
      "loss": 7.3049,
      "mean_token_accuracy": 0.07169978618621826,
      "num_tokens": 540470.0,
      "step": 295
    },
    {
      "entropy": 7.67633023262024,
      "epoch": 0.24855012427506215,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001495,
      "loss": 7.2885,
      "mean_token_accuracy": 0.07255204916000366,
      "num_tokens": 550317.0,
      "step": 300
    },
    {
      "entropy": 7.703630876541138,
      "epoch": 0.2526926263463132,
      "grad_norm": 1.28125,
      "learning_rate": 0.000152,
      "loss": 7.2609,
      "mean_token_accuracy": 0.07455097585916519,
      "num_tokens": 559409.0,
      "step": 305
    },
    {
      "entropy": 7.531597137451172,
      "epoch": 0.25683512841756423,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.2618,
      "mean_token_accuracy": 0.0835557732731104,
      "num_tokens": 568726.0,
      "step": 310
    },
    {
      "entropy": 7.69538688659668,
      "epoch": 0.2609776304888152,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000157,
      "loss": 7.272,
      "mean_token_accuracy": 0.070323546230793,
      "num_tokens": 577926.0,
      "step": 315
    },
    {
      "entropy": 7.653127574920655,
      "epoch": 0.2651201325600663,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001595,
      "loss": 7.2668,
      "mean_token_accuracy": 0.0749148778617382,
      "num_tokens": 586695.0,
      "step": 320
    },
    {
      "entropy": 7.685172080993652,
      "epoch": 0.26926263463131733,
      "grad_norm": 1.34375,
      "learning_rate": 0.000162,
      "loss": 7.2778,
      "mean_token_accuracy": 0.072557582706213,
      "num_tokens": 596250.0,
      "step": 325
    },
    {
      "entropy": 7.5769373893737795,
      "epoch": 0.27340513670256833,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.297,
      "mean_token_accuracy": 0.07127823494374752,
      "num_tokens": 605761.0,
      "step": 330
    },
    {
      "entropy": 7.630343723297119,
      "epoch": 0.2775476387738194,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.2279,
      "mean_token_accuracy": 0.07685448080301285,
      "num_tokens": 614313.0,
      "step": 335
    },
    {
      "entropy": 7.6338982582092285,
      "epoch": 0.28169014084507044,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.2426,
      "mean_token_accuracy": 0.07610973864793777,
      "num_tokens": 624470.0,
      "step": 340
    },
    {
      "entropy": 7.5630439758300785,
      "epoch": 0.28583264291632143,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.1876,
      "mean_token_accuracy": 0.07631983868777752,
      "num_tokens": 633458.0,
      "step": 345
    },
    {
      "entropy": 7.555823183059692,
      "epoch": 0.2899751449875725,
      "grad_norm": 1.09375,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.1541,
      "mean_token_accuracy": 0.07695739194750786,
      "num_tokens": 643251.0,
      "step": 350
    },
    {
      "entropy": 7.54128098487854,
      "epoch": 0.29411764705882354,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000177,
      "loss": 7.2095,
      "mean_token_accuracy": 0.07986946441233159,
      "num_tokens": 652005.0,
      "step": 355
    },
    {
      "entropy": 7.541937828063965,
      "epoch": 0.2982601491300746,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001795,
      "loss": 7.1099,
      "mean_token_accuracy": 0.08143376260995865,
      "num_tokens": 661003.0,
      "step": 360
    },
    {
      "entropy": 7.468397426605224,
      "epoch": 0.3024026512013256,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000182,
      "loss": 7.0991,
      "mean_token_accuracy": 0.07771601304411888,
      "num_tokens": 670221.0,
      "step": 365
    },
    {
      "entropy": 7.572120714187622,
      "epoch": 0.30654515327257664,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0001845,
      "loss": 7.3705,
      "mean_token_accuracy": 0.06953227184712887,
      "num_tokens": 680756.0,
      "step": 370
    },
    {
      "entropy": 7.47379961013794,
      "epoch": 0.3106876553438277,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000187,
      "loss": 7.0678,
      "mean_token_accuracy": 0.08287248685956002,
      "num_tokens": 689641.0,
      "step": 375
    },
    {
      "entropy": 7.441448068618774,
      "epoch": 0.3148301574150787,
      "grad_norm": 1.71875,
      "learning_rate": 0.0001895,
      "loss": 7.0139,
      "mean_token_accuracy": 0.0792611964046955,
      "num_tokens": 698209.0,
      "step": 380
    },
    {
      "entropy": 7.345022249221802,
      "epoch": 0.31897265948632975,
      "grad_norm": 1.390625,
      "learning_rate": 0.000192,
      "loss": 7.0205,
      "mean_token_accuracy": 0.08480528593063355,
      "num_tokens": 706712.0,
      "step": 385
    },
    {
      "entropy": 7.502147245407104,
      "epoch": 0.3231151615575808,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001945,
      "loss": 7.2079,
      "mean_token_accuracy": 0.07785556949675083,
      "num_tokens": 716087.0,
      "step": 390
    },
    {
      "entropy": 7.44799690246582,
      "epoch": 0.3272576636288318,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.0887,
      "mean_token_accuracy": 0.07472220808267593,
      "num_tokens": 725913.0,
      "step": 395
    },
    {
      "entropy": 7.450920248031617,
      "epoch": 0.33140016570008285,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.1256,
      "mean_token_accuracy": 0.08386809080839157,
      "num_tokens": 735020.0,
      "step": 400
    },
    {
      "entropy": 7.34827938079834,
      "epoch": 0.3355426677713339,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000202,
      "loss": 7.0323,
      "mean_token_accuracy": 0.07652226313948632,
      "num_tokens": 744299.0,
      "step": 405
    },
    {
      "entropy": 7.45903434753418,
      "epoch": 0.3396851698425849,
      "grad_norm": 1.109375,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.0115,
      "mean_token_accuracy": 0.08094536513090134,
      "num_tokens": 753231.0,
      "step": 410
    },
    {
      "entropy": 7.382762813568116,
      "epoch": 0.34382767191383595,
      "grad_norm": 1.15625,
      "learning_rate": 0.000207,
      "loss": 7.069,
      "mean_token_accuracy": 0.0813817098736763,
      "num_tokens": 763043.0,
      "step": 415
    },
    {
      "entropy": 7.398455286026001,
      "epoch": 0.347970173985087,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002095,
      "loss": 7.1014,
      "mean_token_accuracy": 0.08094010204076767,
      "num_tokens": 771977.0,
      "step": 420
    },
    {
      "entropy": 7.3298886775970455,
      "epoch": 0.352112676056338,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000212,
      "loss": 6.9887,
      "mean_token_accuracy": 0.0867611140012741,
      "num_tokens": 780285.0,
      "step": 425
    },
    {
      "entropy": 7.321137142181397,
      "epoch": 0.35625517812758906,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002145,
      "loss": 7.0564,
      "mean_token_accuracy": 0.08066153451800347,
      "num_tokens": 789980.0,
      "step": 430
    },
    {
      "entropy": 7.415691232681274,
      "epoch": 0.3603976801988401,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.0081,
      "mean_token_accuracy": 0.0885832205414772,
      "num_tokens": 798700.0,
      "step": 435
    },
    {
      "entropy": 7.261818742752075,
      "epoch": 0.36454018227009116,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002195,
      "loss": 7.0726,
      "mean_token_accuracy": 0.07862451300024986,
      "num_tokens": 808758.0,
      "step": 440
    },
    {
      "entropy": 7.4208532810211185,
      "epoch": 0.36868268434134216,
      "grad_norm": 1.328125,
      "learning_rate": 0.000222,
      "loss": 7.0513,
      "mean_token_accuracy": 0.08192357867956161,
      "num_tokens": 817692.0,
      "step": 445
    },
    {
      "entropy": 7.318561601638794,
      "epoch": 0.3728251864125932,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002245,
      "loss": 7.0764,
      "mean_token_accuracy": 0.09159427806735039,
      "num_tokens": 827052.0,
      "step": 450
    },
    {
      "entropy": 7.326152086257935,
      "epoch": 0.37696768848384427,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.0672,
      "mean_token_accuracy": 0.08785221576690674,
      "num_tokens": 836442.0,
      "step": 455
    },
    {
      "entropy": 7.290367841720581,
      "epoch": 0.38111019055509526,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.9496,
      "mean_token_accuracy": 0.08379893153905868,
      "num_tokens": 846441.0,
      "step": 460
    },
    {
      "entropy": 7.274770307540893,
      "epoch": 0.3852526926263463,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.0006,
      "mean_token_accuracy": 0.08496225997805595,
      "num_tokens": 856108.0,
      "step": 465
    },
    {
      "entropy": 7.303304958343506,
      "epoch": 0.38939519469759737,
      "grad_norm": 1.125,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.9345,
      "mean_token_accuracy": 0.0826059028506279,
      "num_tokens": 865788.0,
      "step": 470
    },
    {
      "entropy": 7.17899341583252,
      "epoch": 0.39353769676884837,
      "grad_norm": 1.203125,
      "learning_rate": 0.000237,
      "loss": 6.9851,
      "mean_token_accuracy": 0.07972754761576653,
      "num_tokens": 874833.0,
      "step": 475
    },
    {
      "entropy": 7.176127529144287,
      "epoch": 0.3976801988400994,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002395,
      "loss": 6.9418,
      "mean_token_accuracy": 0.08143137246370316,
      "num_tokens": 883685.0,
      "step": 480
    },
    {
      "entropy": 7.265657234191894,
      "epoch": 0.40182270091135047,
      "grad_norm": 1.265625,
      "learning_rate": 0.000242,
      "loss": 6.9269,
      "mean_token_accuracy": 0.08215037062764168,
      "num_tokens": 893908.0,
      "step": 485
    },
    {
      "entropy": 7.271769285202026,
      "epoch": 0.40596520298260147,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002445,
      "loss": 7.0375,
      "mean_token_accuracy": 0.08106703236699105,
      "num_tokens": 903081.0,
      "step": 490
    },
    {
      "entropy": 7.133934879302979,
      "epoch": 0.4101077050538525,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000247,
      "loss": 6.8953,
      "mean_token_accuracy": 0.08535027876496315,
      "num_tokens": 912515.0,
      "step": 495
    },
    {
      "entropy": 7.248036670684814,
      "epoch": 0.4142502071251036,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002495,
      "loss": 6.9103,
      "mean_token_accuracy": 0.08826312869787216,
      "num_tokens": 921056.0,
      "step": 500
    },
    {
      "epoch": 0.4142502071251036,
      "eval_entropy": 7.152716042898866,
      "eval_loss": 7.043540954589844,
      "eval_mean_token_accuracy": 0.08588246503141897,
      "eval_num_tokens": 921056.0,
      "eval_runtime": 2.8013,
      "eval_samples_per_second": 1337.598,
      "eval_steps_per_second": 167.423,
      "step": 500
    },
    {
      "entropy": 7.137938928604126,
      "epoch": 0.4183927091963546,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000252,
      "loss": 6.8537,
      "mean_token_accuracy": 0.0894715815782547,
      "num_tokens": 931112.0,
      "step": 505
    },
    {
      "entropy": 7.109851598739624,
      "epoch": 0.4225352112676056,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002545,
      "loss": 6.9643,
      "mean_token_accuracy": 0.08381934091448784,
      "num_tokens": 941120.0,
      "step": 510
    },
    {
      "entropy": 7.161206293106079,
      "epoch": 0.4266777133388567,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000257,
      "loss": 6.9286,
      "mean_token_accuracy": 0.07928110212087632,
      "num_tokens": 950041.0,
      "step": 515
    },
    {
      "entropy": 7.187362718582153,
      "epoch": 0.43082021541010773,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002595,
      "loss": 6.9,
      "mean_token_accuracy": 0.08709124103188515,
      "num_tokens": 958608.0,
      "step": 520
    },
    {
      "entropy": 7.231058073043823,
      "epoch": 0.43496271748135873,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000262,
      "loss": 7.0099,
      "mean_token_accuracy": 0.08367124274373054,
      "num_tokens": 966901.0,
      "step": 525
    },
    {
      "entropy": 7.111437368392944,
      "epoch": 0.4391052195526098,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.8679,
      "mean_token_accuracy": 0.0846832774579525,
      "num_tokens": 976198.0,
      "step": 530
    },
    {
      "entropy": 7.2298952579498295,
      "epoch": 0.44324772162386084,
      "grad_norm": 1.125,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.9271,
      "mean_token_accuracy": 0.08150140345096588,
      "num_tokens": 985248.0,
      "step": 535
    },
    {
      "entropy": 7.10087308883667,
      "epoch": 0.44739022369511183,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.8356,
      "mean_token_accuracy": 0.08679510727524757,
      "num_tokens": 993289.0,
      "step": 540
    },
    {
      "entropy": 7.159850454330444,
      "epoch": 0.4515327257663629,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.8915,
      "mean_token_accuracy": 0.08651292622089386,
      "num_tokens": 1001784.0,
      "step": 545
    },
    {
      "entropy": 7.133292531967163,
      "epoch": 0.45567522783761394,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002745,
      "loss": 6.8731,
      "mean_token_accuracy": 0.09236607924103737,
      "num_tokens": 1010445.0,
      "step": 550
    },
    {
      "entropy": 7.113561677932739,
      "epoch": 0.45981772990886494,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000277,
      "loss": 6.9026,
      "mean_token_accuracy": 0.08968292474746704,
      "num_tokens": 1019084.0,
      "step": 555
    },
    {
      "entropy": 7.082435131072998,
      "epoch": 0.463960231980116,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002795,
      "loss": 6.8235,
      "mean_token_accuracy": 0.08769916668534279,
      "num_tokens": 1027263.0,
      "step": 560
    },
    {
      "entropy": 7.080044603347778,
      "epoch": 0.46810273405136704,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.9333,
      "mean_token_accuracy": 0.0789696842432022,
      "num_tokens": 1036275.0,
      "step": 565
    },
    {
      "entropy": 7.151401376724243,
      "epoch": 0.47224523612261804,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002845,
      "loss": 6.9681,
      "mean_token_accuracy": 0.08598844856023788,
      "num_tokens": 1045208.0,
      "step": 570
    },
    {
      "entropy": 7.106773471832275,
      "epoch": 0.4763877381938691,
      "grad_norm": 1.296875,
      "learning_rate": 0.000287,
      "loss": 6.8902,
      "mean_token_accuracy": 0.08505614772439003,
      "num_tokens": 1054703.0,
      "step": 575
    },
    {
      "entropy": 6.964013481140137,
      "epoch": 0.48053024026512015,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002895,
      "loss": 6.7593,
      "mean_token_accuracy": 0.08789120018482208,
      "num_tokens": 1063652.0,
      "step": 580
    },
    {
      "entropy": 6.994414854049682,
      "epoch": 0.48467274233637114,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000292,
      "loss": 6.853,
      "mean_token_accuracy": 0.08468487113714218,
      "num_tokens": 1072070.0,
      "step": 585
    },
    {
      "entropy": 7.174046993255615,
      "epoch": 0.4888152444076222,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002945,
      "loss": 6.8959,
      "mean_token_accuracy": 0.0872904784977436,
      "num_tokens": 1079992.0,
      "step": 590
    },
    {
      "entropy": 7.0769284725189205,
      "epoch": 0.49295774647887325,
      "grad_norm": 1.234375,
      "learning_rate": 0.000297,
      "loss": 6.8845,
      "mean_token_accuracy": 0.08729023486375809,
      "num_tokens": 1088862.0,
      "step": 595
    },
    {
      "entropy": 7.002944040298462,
      "epoch": 0.4971002485501243,
      "grad_norm": 1.25,
      "learning_rate": 0.0002995,
      "loss": 6.7078,
      "mean_token_accuracy": 0.09125389754772187,
      "num_tokens": 1097614.0,
      "step": 600
    },
    {
      "entropy": 6.979242372512817,
      "epoch": 0.5012427506213754,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000302,
      "loss": 6.8893,
      "mean_token_accuracy": 0.09072866439819335,
      "num_tokens": 1107347.0,
      "step": 605
    },
    {
      "entropy": 7.11456413269043,
      "epoch": 0.5053852526926264,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003045,
      "loss": 6.8977,
      "mean_token_accuracy": 0.0837775819003582,
      "num_tokens": 1116707.0,
      "step": 610
    },
    {
      "entropy": 7.103943634033203,
      "epoch": 0.5095277547638773,
      "grad_norm": 1.125,
      "learning_rate": 0.000307,
      "loss": 6.889,
      "mean_token_accuracy": 0.08799269050359726,
      "num_tokens": 1126356.0,
      "step": 615
    },
    {
      "entropy": 6.997752904891968,
      "epoch": 0.5136702568351285,
      "grad_norm": 1.125,
      "learning_rate": 0.0003095,
      "loss": 6.8317,
      "mean_token_accuracy": 0.08332241624593735,
      "num_tokens": 1136281.0,
      "step": 620
    },
    {
      "entropy": 6.969416952133178,
      "epoch": 0.5178127589063795,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000312,
      "loss": 6.7914,
      "mean_token_accuracy": 0.08854400739073753,
      "num_tokens": 1145466.0,
      "step": 625
    },
    {
      "entropy": 6.913108205795288,
      "epoch": 0.5219552609776305,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003145,
      "loss": 6.7835,
      "mean_token_accuracy": 0.09256977885961533,
      "num_tokens": 1154225.0,
      "step": 630
    },
    {
      "entropy": 6.981909942626953,
      "epoch": 0.5260977630488816,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000317,
      "loss": 6.8447,
      "mean_token_accuracy": 0.0854923166334629,
      "num_tokens": 1163872.0,
      "step": 635
    },
    {
      "entropy": 7.036574649810791,
      "epoch": 0.5302402651201326,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003195,
      "loss": 6.7584,
      "mean_token_accuracy": 0.09340214282274246,
      "num_tokens": 1172695.0,
      "step": 640
    },
    {
      "entropy": 6.875000143051148,
      "epoch": 0.5343827671913836,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000322,
      "loss": 6.6344,
      "mean_token_accuracy": 0.09556335210800171,
      "num_tokens": 1181291.0,
      "step": 645
    },
    {
      "entropy": 6.9969175338745115,
      "epoch": 0.5385252692626347,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.8093,
      "mean_token_accuracy": 0.08966382555663585,
      "num_tokens": 1189927.0,
      "step": 650
    },
    {
      "entropy": 6.955785942077637,
      "epoch": 0.5426677713338857,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.7423,
      "mean_token_accuracy": 0.09562051445245742,
      "num_tokens": 1197913.0,
      "step": 655
    },
    {
      "entropy": 6.968462419509888,
      "epoch": 0.5468102734051367,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.6775,
      "mean_token_accuracy": 0.09469703435897828,
      "num_tokens": 1207032.0,
      "step": 660
    },
    {
      "entropy": 6.832558679580688,
      "epoch": 0.5509527754763878,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.7769,
      "mean_token_accuracy": 0.09034623727202415,
      "num_tokens": 1215710.0,
      "step": 665
    },
    {
      "entropy": 6.937397909164429,
      "epoch": 0.5550952775476388,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.7588,
      "mean_token_accuracy": 0.08644548580050468,
      "num_tokens": 1225619.0,
      "step": 670
    },
    {
      "entropy": 6.87630820274353,
      "epoch": 0.5592377796188898,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000337,
      "loss": 6.6684,
      "mean_token_accuracy": 0.09475696682929993,
      "num_tokens": 1235000.0,
      "step": 675
    },
    {
      "entropy": 6.888253688812256,
      "epoch": 0.5633802816901409,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003395,
      "loss": 6.6846,
      "mean_token_accuracy": 0.09072612524032593,
      "num_tokens": 1243594.0,
      "step": 680
    },
    {
      "entropy": 6.984683322906494,
      "epoch": 0.5675227837613919,
      "grad_norm": 1.359375,
      "learning_rate": 0.000342,
      "loss": 6.8658,
      "mean_token_accuracy": 0.0922737754881382,
      "num_tokens": 1252279.0,
      "step": 685
    },
    {
      "entropy": 6.964951038360596,
      "epoch": 0.5716652858326429,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.7816,
      "mean_token_accuracy": 0.09199510142207146,
      "num_tokens": 1261314.0,
      "step": 690
    },
    {
      "entropy": 6.807141304016113,
      "epoch": 0.575807787903894,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000347,
      "loss": 6.6534,
      "mean_token_accuracy": 0.09733643308281899,
      "num_tokens": 1270854.0,
      "step": 695
    },
    {
      "entropy": 6.822793102264404,
      "epoch": 0.579950289975145,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003495,
      "loss": 6.6378,
      "mean_token_accuracy": 0.09607320502400399,
      "num_tokens": 1279183.0,
      "step": 700
    },
    {
      "entropy": 6.798097324371338,
      "epoch": 0.584092792046396,
      "grad_norm": 1.15625,
      "learning_rate": 0.000352,
      "loss": 6.7234,
      "mean_token_accuracy": 0.09325428456068038,
      "num_tokens": 1288347.0,
      "step": 705
    },
    {
      "entropy": 6.899397277832032,
      "epoch": 0.5882352941176471,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003545,
      "loss": 6.7217,
      "mean_token_accuracy": 0.09395038411021232,
      "num_tokens": 1297449.0,
      "step": 710
    },
    {
      "entropy": 6.90986647605896,
      "epoch": 0.5923777961888981,
      "grad_norm": 1.21875,
      "learning_rate": 0.000357,
      "loss": 6.6996,
      "mean_token_accuracy": 0.09421416223049164,
      "num_tokens": 1306232.0,
      "step": 715
    },
    {
      "entropy": 6.846466302871704,
      "epoch": 0.5965202982601492,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003595,
      "loss": 6.6972,
      "mean_token_accuracy": 0.09734119176864624,
      "num_tokens": 1315073.0,
      "step": 720
    },
    {
      "entropy": 6.890099143981933,
      "epoch": 0.6006628003314002,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000362,
      "loss": 6.7379,
      "mean_token_accuracy": 0.08920296430587768,
      "num_tokens": 1324629.0,
      "step": 725
    },
    {
      "entropy": 6.895062398910523,
      "epoch": 0.6048053024026512,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003645,
      "loss": 6.8505,
      "mean_token_accuracy": 0.08717696964740754,
      "num_tokens": 1334152.0,
      "step": 730
    },
    {
      "entropy": 6.952538633346558,
      "epoch": 0.6089478044739023,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000367,
      "loss": 6.827,
      "mean_token_accuracy": 0.08865289464592933,
      "num_tokens": 1343063.0,
      "step": 735
    },
    {
      "entropy": 6.843909358978271,
      "epoch": 0.6130903065451533,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003695,
      "loss": 6.7434,
      "mean_token_accuracy": 0.08947005942463875,
      "num_tokens": 1352632.0,
      "step": 740
    },
    {
      "entropy": 6.815534782409668,
      "epoch": 0.6172328086164043,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000372,
      "loss": 6.6749,
      "mean_token_accuracy": 0.09639113694429398,
      "num_tokens": 1361103.0,
      "step": 745
    },
    {
      "entropy": 6.879958820343018,
      "epoch": 0.6213753106876554,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003745,
      "loss": 6.6594,
      "mean_token_accuracy": 0.09863266348838806,
      "num_tokens": 1370658.0,
      "step": 750
    },
    {
      "entropy": 6.780733728408814,
      "epoch": 0.6255178127589064,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000377,
      "loss": 6.6808,
      "mean_token_accuracy": 0.09909494146704674,
      "num_tokens": 1379817.0,
      "step": 755
    },
    {
      "entropy": 6.838653945922852,
      "epoch": 0.6296603148301574,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003795,
      "loss": 6.6471,
      "mean_token_accuracy": 0.09778247401118279,
      "num_tokens": 1389329.0,
      "step": 760
    },
    {
      "entropy": 6.66096248626709,
      "epoch": 0.6338028169014085,
      "grad_norm": 1.0625,
      "learning_rate": 0.000382,
      "loss": 6.6749,
      "mean_token_accuracy": 0.09297099262475968,
      "num_tokens": 1399279.0,
      "step": 765
    },
    {
      "entropy": 6.80253233909607,
      "epoch": 0.6379453189726595,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003845,
      "loss": 6.603,
      "mean_token_accuracy": 0.09690207317471504,
      "num_tokens": 1408765.0,
      "step": 770
    },
    {
      "entropy": 6.849671459197998,
      "epoch": 0.6420878210439105,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.7456,
      "mean_token_accuracy": 0.09820655733346939,
      "num_tokens": 1418615.0,
      "step": 775
    },
    {
      "entropy": 6.82501745223999,
      "epoch": 0.6462303231151616,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.6678,
      "mean_token_accuracy": 0.0880895622074604,
      "num_tokens": 1427359.0,
      "step": 780
    },
    {
      "entropy": 6.867260122299195,
      "epoch": 0.6503728251864126,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.8025,
      "mean_token_accuracy": 0.09724401906132699,
      "num_tokens": 1435516.0,
      "step": 785
    },
    {
      "entropy": 6.804732322692871,
      "epoch": 0.6545153272576636,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.7582,
      "mean_token_accuracy": 0.09707285091280937,
      "num_tokens": 1444438.0,
      "step": 790
    },
    {
      "entropy": 6.864415264129638,
      "epoch": 0.6586578293289147,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.6723,
      "mean_token_accuracy": 0.10008604601025581,
      "num_tokens": 1453740.0,
      "step": 795
    },
    {
      "entropy": 6.7923424243927,
      "epoch": 0.6628003314001657,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003995,
      "loss": 6.6596,
      "mean_token_accuracy": 0.0992733120918274,
      "num_tokens": 1462083.0,
      "step": 800
    },
    {
      "entropy": 6.66512622833252,
      "epoch": 0.6669428334714167,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000402,
      "loss": 6.5895,
      "mean_token_accuracy": 0.10381095260381698,
      "num_tokens": 1471459.0,
      "step": 805
    },
    {
      "entropy": 6.769164514541626,
      "epoch": 0.6710853355426678,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004045,
      "loss": 6.5678,
      "mean_token_accuracy": 0.10211108550429344,
      "num_tokens": 1480516.0,
      "step": 810
    },
    {
      "entropy": 6.621196842193603,
      "epoch": 0.6752278376139188,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.4912,
      "mean_token_accuracy": 0.09601941630244255,
      "num_tokens": 1489268.0,
      "step": 815
    },
    {
      "entropy": 6.848507595062256,
      "epoch": 0.6793703396851698,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004095,
      "loss": 6.7021,
      "mean_token_accuracy": 0.09602859765291213,
      "num_tokens": 1498819.0,
      "step": 820
    },
    {
      "entropy": 6.716065549850464,
      "epoch": 0.6835128417564209,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000412,
      "loss": 6.6693,
      "mean_token_accuracy": 0.10195070579648018,
      "num_tokens": 1507852.0,
      "step": 825
    },
    {
      "entropy": 6.764216947555542,
      "epoch": 0.6876553438276719,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004145,
      "loss": 6.6844,
      "mean_token_accuracy": 0.1013753168284893,
      "num_tokens": 1517538.0,
      "step": 830
    },
    {
      "entropy": 6.660551071166992,
      "epoch": 0.6917978458989229,
      "grad_norm": 1.1875,
      "learning_rate": 0.000417,
      "loss": 6.567,
      "mean_token_accuracy": 0.10303043052554131,
      "num_tokens": 1525389.0,
      "step": 835
    },
    {
      "entropy": 6.7510017395019535,
      "epoch": 0.695940347970174,
      "grad_norm": 1.25,
      "learning_rate": 0.0004195,
      "loss": 6.5375,
      "mean_token_accuracy": 0.100385482609272,
      "num_tokens": 1534117.0,
      "step": 840
    },
    {
      "entropy": 6.690871620178223,
      "epoch": 0.700082850041425,
      "grad_norm": 1.171875,
      "learning_rate": 0.000422,
      "loss": 6.5814,
      "mean_token_accuracy": 0.10090366378426552,
      "num_tokens": 1542610.0,
      "step": 845
    },
    {
      "entropy": 6.683345317840576,
      "epoch": 0.704225352112676,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004245,
      "loss": 6.5864,
      "mean_token_accuracy": 0.10127236992120743,
      "num_tokens": 1551771.0,
      "step": 850
    },
    {
      "entropy": 6.748124027252198,
      "epoch": 0.7083678541839271,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000427,
      "loss": 6.7407,
      "mean_token_accuracy": 0.09586205929517747,
      "num_tokens": 1560231.0,
      "step": 855
    },
    {
      "entropy": 6.7252332210540775,
      "epoch": 0.7125103562551781,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004295,
      "loss": 6.6918,
      "mean_token_accuracy": 0.0975484773516655,
      "num_tokens": 1569578.0,
      "step": 860
    },
    {
      "entropy": 6.756966400146484,
      "epoch": 0.7166528583264291,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000432,
      "loss": 6.5691,
      "mean_token_accuracy": 0.10365280956029892,
      "num_tokens": 1579308.0,
      "step": 865
    },
    {
      "entropy": 6.701776933670044,
      "epoch": 0.7207953603976802,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004345,
      "loss": 6.7528,
      "mean_token_accuracy": 0.09729094281792641,
      "num_tokens": 1589506.0,
      "step": 870
    },
    {
      "entropy": 6.6439666748046875,
      "epoch": 0.7249378624689312,
      "grad_norm": 1.15625,
      "learning_rate": 0.000437,
      "loss": 6.543,
      "mean_token_accuracy": 0.10177418887615204,
      "num_tokens": 1598127.0,
      "step": 875
    },
    {
      "entropy": 6.685148239135742,
      "epoch": 0.7290803645401823,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004395,
      "loss": 6.5684,
      "mean_token_accuracy": 0.10523401126265526,
      "num_tokens": 1608073.0,
      "step": 880
    },
    {
      "entropy": 6.695839357376099,
      "epoch": 0.7332228666114333,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000442,
      "loss": 6.587,
      "mean_token_accuracy": 0.10461484640836716,
      "num_tokens": 1616272.0,
      "step": 885
    },
    {
      "entropy": 6.641492891311645,
      "epoch": 0.7373653686826843,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004445,
      "loss": 6.6058,
      "mean_token_accuracy": 0.09599140211939812,
      "num_tokens": 1625577.0,
      "step": 890
    },
    {
      "entropy": 6.810349798202514,
      "epoch": 0.7415078707539354,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000447,
      "loss": 6.6845,
      "mean_token_accuracy": 0.09567000940442086,
      "num_tokens": 1634902.0,
      "step": 895
    },
    {
      "entropy": 6.6667379379272464,
      "epoch": 0.7456503728251864,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.5977,
      "mean_token_accuracy": 0.1015617847442627,
      "num_tokens": 1644131.0,
      "step": 900
    },
    {
      "entropy": 6.560829305648804,
      "epoch": 0.7497928748964374,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.5118,
      "mean_token_accuracy": 0.11086556538939477,
      "num_tokens": 1652272.0,
      "step": 905
    },
    {
      "entropy": 6.772023248672485,
      "epoch": 0.7539353769676885,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.6107,
      "mean_token_accuracy": 0.10556575432419776,
      "num_tokens": 1660962.0,
      "step": 910
    },
    {
      "entropy": 6.561828851699829,
      "epoch": 0.7580778790389395,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.539,
      "mean_token_accuracy": 0.10557058155536651,
      "num_tokens": 1670570.0,
      "step": 915
    },
    {
      "entropy": 6.672806739807129,
      "epoch": 0.7622203811101905,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.5859,
      "mean_token_accuracy": 0.10312820002436637,
      "num_tokens": 1679674.0,
      "step": 920
    },
    {
      "entropy": 6.678398847579956,
      "epoch": 0.7663628831814416,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000462,
      "loss": 6.565,
      "mean_token_accuracy": 0.10954021513462067,
      "num_tokens": 1688849.0,
      "step": 925
    },
    {
      "entropy": 6.5227796077728275,
      "epoch": 0.7705053852526926,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004645,
      "loss": 6.4534,
      "mean_token_accuracy": 0.10681168809533119,
      "num_tokens": 1698565.0,
      "step": 930
    },
    {
      "entropy": 6.671809673309326,
      "epoch": 0.7746478873239436,
      "grad_norm": 1.09375,
      "learning_rate": 0.000467,
      "loss": 6.5485,
      "mean_token_accuracy": 0.11151607260107994,
      "num_tokens": 1708009.0,
      "step": 935
    },
    {
      "entropy": 6.625020933151245,
      "epoch": 0.7787903893951947,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004695,
      "loss": 6.5537,
      "mean_token_accuracy": 0.10801653489470482,
      "num_tokens": 1716497.0,
      "step": 940
    },
    {
      "entropy": 6.526713752746582,
      "epoch": 0.7829328914664457,
      "grad_norm": 1.140625,
      "learning_rate": 0.000472,
      "loss": 6.5447,
      "mean_token_accuracy": 0.11139797195792198,
      "num_tokens": 1725609.0,
      "step": 945
    },
    {
      "entropy": 6.60435209274292,
      "epoch": 0.7870753935376967,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004745,
      "loss": 6.4829,
      "mean_token_accuracy": 0.10714786946773529,
      "num_tokens": 1734165.0,
      "step": 950
    },
    {
      "entropy": 6.554976940155029,
      "epoch": 0.7912178956089478,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000477,
      "loss": 6.5828,
      "mean_token_accuracy": 0.10263153910636902,
      "num_tokens": 1744856.0,
      "step": 955
    },
    {
      "entropy": 6.635064172744751,
      "epoch": 0.7953603976801988,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004795,
      "loss": 6.5938,
      "mean_token_accuracy": 0.1016570694744587,
      "num_tokens": 1753386.0,
      "step": 960
    },
    {
      "entropy": 6.552286529541016,
      "epoch": 0.7995028997514498,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000482,
      "loss": 6.4448,
      "mean_token_accuracy": 0.1083413727581501,
      "num_tokens": 1762521.0,
      "step": 965
    },
    {
      "entropy": 6.489339590072632,
      "epoch": 0.8036454018227009,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004845,
      "loss": 6.5099,
      "mean_token_accuracy": 0.10716646909713745,
      "num_tokens": 1772403.0,
      "step": 970
    },
    {
      "entropy": 6.546666860580444,
      "epoch": 0.8077879038939519,
      "grad_norm": 1.125,
      "learning_rate": 0.000487,
      "loss": 6.3975,
      "mean_token_accuracy": 0.1070093922317028,
      "num_tokens": 1780924.0,
      "step": 975
    },
    {
      "entropy": 6.623412990570069,
      "epoch": 0.8119304059652029,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004895,
      "loss": 6.5443,
      "mean_token_accuracy": 0.10209290608763695,
      "num_tokens": 1789992.0,
      "step": 980
    },
    {
      "entropy": 6.60557599067688,
      "epoch": 0.816072908036454,
      "grad_norm": 1.1875,
      "learning_rate": 0.000492,
      "loss": 6.5226,
      "mean_token_accuracy": 0.11116698980331421,
      "num_tokens": 1799513.0,
      "step": 985
    },
    {
      "entropy": 6.434801244735718,
      "epoch": 0.820215410107705,
      "grad_norm": 1.25,
      "learning_rate": 0.0004945,
      "loss": 6.3764,
      "mean_token_accuracy": 0.10629670321941376,
      "num_tokens": 1807393.0,
      "step": 990
    },
    {
      "entropy": 6.577297115325928,
      "epoch": 0.824357912178956,
      "grad_norm": 1.203125,
      "learning_rate": 0.000497,
      "loss": 6.6109,
      "mean_token_accuracy": 0.10033297091722489,
      "num_tokens": 1817284.0,
      "step": 995
    },
    {
      "entropy": 6.540279865264893,
      "epoch": 0.8285004142502072,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004995,
      "loss": 6.4127,
      "mean_token_accuracy": 0.1117069512605667,
      "num_tokens": 1825591.0,
      "step": 1000
    },
    {
      "epoch": 0.8285004142502072,
      "eval_entropy": 6.394233569407514,
      "eval_loss": 6.601587295532227,
      "eval_mean_token_accuracy": 0.1064636675454279,
      "eval_num_tokens": 1825591.0,
      "eval_runtime": 2.7507,
      "eval_samples_per_second": 1362.188,
      "eval_steps_per_second": 170.501,
      "step": 1000
    },
    {
      "entropy": 6.598631191253662,
      "epoch": 0.8326429163214581,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999998547681232,
      "loss": 6.5313,
      "mean_token_accuracy": 0.10966545790433883,
      "num_tokens": 1834655.0,
      "step": 1005
    },
    {
      "entropy": 6.366255187988282,
      "epoch": 0.8367854183927091,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999992647639447,
      "loss": 6.3637,
      "mean_token_accuracy": 0.10939211472868919,
      "num_tokens": 1844187.0,
      "step": 1010
    },
    {
      "entropy": 6.511251926422119,
      "epoch": 0.8409279204639603,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999982209116618,
      "loss": 6.445,
      "mean_token_accuracy": 0.10647044554352761,
      "num_tokens": 1852768.0,
      "step": 1015
    },
    {
      "entropy": 6.5381862163543705,
      "epoch": 0.8450704225352113,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999967232133796,
      "loss": 6.4772,
      "mean_token_accuracy": 0.11856938228011131,
      "num_tokens": 1862095.0,
      "step": 1020
    },
    {
      "entropy": 6.492068910598755,
      "epoch": 0.8492129246064622,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999947716721196,
      "loss": 6.524,
      "mean_token_accuracy": 0.104226516187191,
      "num_tokens": 1871239.0,
      "step": 1025
    },
    {
      "entropy": 6.561237716674805,
      "epoch": 0.8533554266777134,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999923662918179,
      "loss": 6.5388,
      "mean_token_accuracy": 0.10700825974345207,
      "num_tokens": 1880065.0,
      "step": 1030
    },
    {
      "entropy": 6.622053861618042,
      "epoch": 0.8574979287489644,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999895070773266,
      "loss": 6.5681,
      "mean_token_accuracy": 0.10293134152889252,
      "num_tokens": 1889065.0,
      "step": 1035
    },
    {
      "entropy": 6.498682117462158,
      "epoch": 0.8616404308202155,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999861940344129,
      "loss": 6.5818,
      "mean_token_accuracy": 0.10599627569317818,
      "num_tokens": 1898238.0,
      "step": 1040
    },
    {
      "entropy": 6.620818948745727,
      "epoch": 0.8657829328914665,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999824271697597,
      "loss": 6.5041,
      "mean_token_accuracy": 0.11954014375805855,
      "num_tokens": 1906780.0,
      "step": 1045
    },
    {
      "entropy": 6.587788820266724,
      "epoch": 0.8699254349627175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499978206490965,
      "loss": 6.5855,
      "mean_token_accuracy": 0.10859053954482079,
      "num_tokens": 1916183.0,
      "step": 1050
    },
    {
      "entropy": 6.472416591644287,
      "epoch": 0.8740679370339686,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999735320065427,
      "loss": 6.4335,
      "mean_token_accuracy": 0.11546618118882179,
      "num_tokens": 1925382.0,
      "step": 1055
    },
    {
      "entropy": 6.469833660125732,
      "epoch": 0.8782104391052196,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999684037259216,
      "loss": 6.2955,
      "mean_token_accuracy": 0.12006475627422333,
      "num_tokens": 1933639.0,
      "step": 1060
    },
    {
      "entropy": 6.452498483657837,
      "epoch": 0.8823529411764706,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999628216594456,
      "loss": 6.5491,
      "mean_token_accuracy": 0.10926123335957527,
      "num_tokens": 1943044.0,
      "step": 1065
    },
    {
      "entropy": 6.504137325286865,
      "epoch": 0.8864954432477217,
      "grad_norm": 1.203125,
      "learning_rate": 0.000499956785818375,
      "loss": 6.4177,
      "mean_token_accuracy": 0.10931780487298966,
      "num_tokens": 1951777.0,
      "step": 1070
    },
    {
      "entropy": 6.467433595657349,
      "epoch": 0.8906379453189727,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999502962148843,
      "loss": 6.4799,
      "mean_token_accuracy": 0.10977004319429398,
      "num_tokens": 1961356.0,
      "step": 1075
    },
    {
      "entropy": 6.609413576126099,
      "epoch": 0.8947804473902237,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499943352862064,
      "loss": 6.529,
      "mean_token_accuracy": 0.10762318223714828,
      "num_tokens": 1970341.0,
      "step": 1080
    },
    {
      "entropy": 6.496694040298462,
      "epoch": 0.8989229494614748,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999359557739194,
      "loss": 6.383,
      "mean_token_accuracy": 0.11510946005582809,
      "num_tokens": 1978576.0,
      "step": 1085
    },
    {
      "entropy": 6.428346872329712,
      "epoch": 0.9030654515327258,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999281049653713,
      "loss": 6.4282,
      "mean_token_accuracy": 0.11499780714511872,
      "num_tokens": 1987229.0,
      "step": 1090
    },
    {
      "entropy": 6.449847650527954,
      "epoch": 0.9072079536039768,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999198004522557,
      "loss": 6.3618,
      "mean_token_accuracy": 0.11787119433283806,
      "num_tokens": 1996473.0,
      "step": 1095
    },
    {
      "entropy": 6.411348676681518,
      "epoch": 0.9113504556752279,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999110422513237,
      "loss": 6.3786,
      "mean_token_accuracy": 0.11625739708542823,
      "num_tokens": 2005551.0,
      "step": 1100
    },
    {
      "entropy": 6.447707653045654,
      "epoch": 0.9154929577464789,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999018303802414,
      "loss": 6.4902,
      "mean_token_accuracy": 0.11025101318955421,
      "num_tokens": 2015012.0,
      "step": 1105
    },
    {
      "entropy": 6.500209903717041,
      "epoch": 0.9196354598177299,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998921648575904,
      "loss": 6.4993,
      "mean_token_accuracy": 0.1120341457426548,
      "num_tokens": 2024435.0,
      "step": 1110
    },
    {
      "entropy": 6.45713210105896,
      "epoch": 0.923777961888981,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998820457028669,
      "loss": 6.3952,
      "mean_token_accuracy": 0.11182878091931343,
      "num_tokens": 2033991.0,
      "step": 1115
    },
    {
      "entropy": 6.520167350769043,
      "epoch": 0.927920463960232,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998714729364824,
      "loss": 6.4059,
      "mean_token_accuracy": 0.1159439042210579,
      "num_tokens": 2043117.0,
      "step": 1120
    },
    {
      "entropy": 6.4226549625396725,
      "epoch": 0.932062966031483,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998604465797634,
      "loss": 6.4284,
      "mean_token_accuracy": 0.11458204984664917,
      "num_tokens": 2052304.0,
      "step": 1125
    },
    {
      "entropy": 6.308629035949707,
      "epoch": 0.9362054681027341,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998489666549512,
      "loss": 6.354,
      "mean_token_accuracy": 0.1193151444196701,
      "num_tokens": 2061217.0,
      "step": 1130
    },
    {
      "entropy": 6.363573312759399,
      "epoch": 0.9403479701739851,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998370331852021,
      "loss": 6.3565,
      "mean_token_accuracy": 0.11765335276722907,
      "num_tokens": 2070796.0,
      "step": 1135
    },
    {
      "entropy": 6.606226682662964,
      "epoch": 0.9444904722452361,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998246461945872,
      "loss": 6.4943,
      "mean_token_accuracy": 0.11730922684073448,
      "num_tokens": 2080202.0,
      "step": 1140
    },
    {
      "entropy": 6.445854616165161,
      "epoch": 0.9486329743164872,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998118057080923,
      "loss": 6.4025,
      "mean_token_accuracy": 0.11324022263288498,
      "num_tokens": 2089788.0,
      "step": 1145
    },
    {
      "entropy": 6.401117420196533,
      "epoch": 0.9527754763877382,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997985117516183,
      "loss": 6.4367,
      "mean_token_accuracy": 0.11083348914980888,
      "num_tokens": 2098727.0,
      "step": 1150
    },
    {
      "entropy": 6.464313554763794,
      "epoch": 0.9569179784589892,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997847643519803,
      "loss": 6.322,
      "mean_token_accuracy": 0.11506827175617218,
      "num_tokens": 2107645.0,
      "step": 1155
    },
    {
      "entropy": 6.274763011932373,
      "epoch": 0.9610604805302403,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997705635369085,
      "loss": 6.3093,
      "mean_token_accuracy": 0.11011608317494392,
      "num_tokens": 2116613.0,
      "step": 1160
    },
    {
      "entropy": 6.368350839614868,
      "epoch": 0.9652029826014913,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997559093350474,
      "loss": 6.3013,
      "mean_token_accuracy": 0.11394222304224969,
      "num_tokens": 2125585.0,
      "step": 1165
    },
    {
      "entropy": 6.498920965194702,
      "epoch": 0.9693454846727423,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997408017759562,
      "loss": 6.4282,
      "mean_token_accuracy": 0.11357887014746666,
      "num_tokens": 2134795.0,
      "step": 1170
    },
    {
      "entropy": 6.396416473388672,
      "epoch": 0.9734879867439934,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997252408901084,
      "loss": 6.4224,
      "mean_token_accuracy": 0.11246276721358299,
      "num_tokens": 2143699.0,
      "step": 1175
    },
    {
      "entropy": 6.5009948253631595,
      "epoch": 0.9776304888152444,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997092267088922,
      "loss": 6.3638,
      "mean_token_accuracy": 0.1147896520793438,
      "num_tokens": 2152187.0,
      "step": 1180
    },
    {
      "entropy": 6.340474557876587,
      "epoch": 0.9817729908864954,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996927592646097,
      "loss": 6.4264,
      "mean_token_accuracy": 0.11860019415616989,
      "num_tokens": 2160941.0,
      "step": 1185
    },
    {
      "entropy": 6.451793146133423,
      "epoch": 0.9859154929577465,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996758385904776,
      "loss": 6.4028,
      "mean_token_accuracy": 0.11521425396203995,
      "num_tokens": 2168946.0,
      "step": 1190
    },
    {
      "entropy": 6.310120868682861,
      "epoch": 0.9900579950289975,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996584647206269,
      "loss": 6.315,
      "mean_token_accuracy": 0.11753729954361916,
      "num_tokens": 2177662.0,
      "step": 1195
    },
    {
      "entropy": 6.437413454055786,
      "epoch": 0.9942004971002486,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996406376901026,
      "loss": 6.4181,
      "mean_token_accuracy": 0.1255168005824089,
      "num_tokens": 2187461.0,
      "step": 1200
    },
    {
      "entropy": 6.393413114547729,
      "epoch": 0.9983429991714996,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996223575348635,
      "loss": 6.3,
      "mean_token_accuracy": 0.12105931863188743,
      "num_tokens": 2196878.0,
      "step": 1205
    },
    {
      "entropy": 6.42917308807373,
      "epoch": 1.0024855012427507,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996036242917828,
      "loss": 6.2979,
      "mean_token_accuracy": 0.11524101421236992,
      "num_tokens": 2206178.0,
      "step": 1210
    },
    {
      "entropy": 6.32910008430481,
      "epoch": 1.0066280033140016,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995844379986476,
      "loss": 6.1628,
      "mean_token_accuracy": 0.1222687229514122,
      "num_tokens": 2215330.0,
      "step": 1215
    },
    {
      "entropy": 6.457520198822022,
      "epoch": 1.0107705053852527,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995647986941587,
      "loss": 6.1389,
      "mean_token_accuracy": 0.12433998733758926,
      "num_tokens": 2224146.0,
      "step": 1220
    },
    {
      "entropy": 6.348219490051269,
      "epoch": 1.0149130074565038,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995447064179308,
      "loss": 6.1297,
      "mean_token_accuracy": 0.12844619452953338,
      "num_tokens": 2233646.0,
      "step": 1225
    },
    {
      "entropy": 6.102799320220948,
      "epoch": 1.0190555095277547,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499524161210492,
      "loss": 6.0188,
      "mean_token_accuracy": 0.12751995995640755,
      "num_tokens": 2243663.0,
      "step": 1230
    },
    {
      "entropy": 6.376331901550293,
      "epoch": 1.0231980115990058,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995031631132844,
      "loss": 6.1946,
      "mean_token_accuracy": 0.13004939407110214,
      "num_tokens": 2252564.0,
      "step": 1235
    },
    {
      "entropy": 6.368073844909668,
      "epoch": 1.027340513670257,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994817121686633,
      "loss": 6.1988,
      "mean_token_accuracy": 0.11914310157299042,
      "num_tokens": 2262504.0,
      "step": 1240
    },
    {
      "entropy": 6.29468092918396,
      "epoch": 1.0314830157415078,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004994598084198979,
      "loss": 6.0995,
      "mean_token_accuracy": 0.13067468032240867,
      "num_tokens": 2270553.0,
      "step": 1245
    },
    {
      "entropy": 6.198537969589234,
      "epoch": 1.035625517812759,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994374519111701,
      "loss": 6.0946,
      "mean_token_accuracy": 0.12411084324121475,
      "num_tokens": 2279319.0,
      "step": 1250
    },
    {
      "entropy": 6.336030292510986,
      "epoch": 1.03976801988401,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994146426875757,
      "loss": 6.2448,
      "mean_token_accuracy": 0.12142780497670173,
      "num_tokens": 2288749.0,
      "step": 1255
    },
    {
      "entropy": 6.316008758544922,
      "epoch": 1.043910521955261,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993913807951232,
      "loss": 6.1098,
      "mean_token_accuracy": 0.12393713593482972,
      "num_tokens": 2298272.0,
      "step": 1260
    },
    {
      "entropy": 6.254574918746949,
      "epoch": 1.048053024026512,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993676662807345,
      "loss": 6.1518,
      "mean_token_accuracy": 0.12555598765611647,
      "num_tokens": 2307230.0,
      "step": 1265
    },
    {
      "entropy": 6.343469524383545,
      "epoch": 1.0521955260977631,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993434991922443,
      "loss": 6.2458,
      "mean_token_accuracy": 0.11990855634212494,
      "num_tokens": 2316306.0,
      "step": 1270
    },
    {
      "entropy": 6.305857372283936,
      "epoch": 1.056338028169014,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004993188795784002,
      "loss": 6.1404,
      "mean_token_accuracy": 0.12261386215686798,
      "num_tokens": 2325672.0,
      "step": 1275
    },
    {
      "entropy": 6.25411434173584,
      "epoch": 1.0604805302402651,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992938074888628,
      "loss": 6.1622,
      "mean_token_accuracy": 0.12227201014757157,
      "num_tokens": 2335251.0,
      "step": 1280
    },
    {
      "entropy": 6.313401746749878,
      "epoch": 1.0646230323115162,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004992682829742052,
      "loss": 6.1248,
      "mean_token_accuracy": 0.1251654766499996,
      "num_tokens": 2344054.0,
      "step": 1285
    },
    {
      "entropy": 6.137462472915649,
      "epoch": 1.0687655343827671,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004992423060859132,
      "loss": 6.1157,
      "mean_token_accuracy": 0.127670818567276,
      "num_tokens": 2353493.0,
      "step": 1290
    },
    {
      "entropy": 6.367279481887818,
      "epoch": 1.0729080364540182,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992158768763849,
      "loss": 6.2362,
      "mean_token_accuracy": 0.12816474214196205,
      "num_tokens": 2362687.0,
      "step": 1295
    },
    {
      "entropy": 6.272933483123779,
      "epoch": 1.0770505385252693,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991889953989308,
      "loss": 6.208,
      "mean_token_accuracy": 0.12367280796170235,
      "num_tokens": 2372325.0,
      "step": 1300
    },
    {
      "entropy": 6.2988440990448,
      "epoch": 1.0811930405965202,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991616617077741,
      "loss": 6.2379,
      "mean_token_accuracy": 0.12832726687192916,
      "num_tokens": 2381987.0,
      "step": 1305
    },
    {
      "entropy": 6.4650335788726805,
      "epoch": 1.0853355426677713,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991338758580496,
      "loss": 6.125,
      "mean_token_accuracy": 0.12739104330539702,
      "num_tokens": 2390901.0,
      "step": 1310
    },
    {
      "entropy": 6.090523910522461,
      "epoch": 1.0894780447390224,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004991056379058045,
      "loss": 6.1601,
      "mean_token_accuracy": 0.12110325694084167,
      "num_tokens": 2400756.0,
      "step": 1315
    },
    {
      "entropy": 6.28847484588623,
      "epoch": 1.0936205468102733,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990769479079976,
      "loss": 6.004,
      "mean_token_accuracy": 0.12852884232997894,
      "num_tokens": 2409526.0,
      "step": 1320
    },
    {
      "entropy": 6.171352672576904,
      "epoch": 1.0977630488815244,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990478059225003,
      "loss": 6.194,
      "mean_token_accuracy": 0.12339239865541458,
      "num_tokens": 2419583.0,
      "step": 1325
    },
    {
      "entropy": 6.393887710571289,
      "epoch": 1.1019055509527755,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990182120080945,
      "loss": 6.1337,
      "mean_token_accuracy": 0.12826344668865203,
      "num_tokens": 2429439.0,
      "step": 1330
    },
    {
      "entropy": 6.1219724178314205,
      "epoch": 1.1060480530240264,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989881662244748,
      "loss": 6.1239,
      "mean_token_accuracy": 0.12123734056949616,
      "num_tokens": 2438467.0,
      "step": 1335
    },
    {
      "entropy": 6.308976459503174,
      "epoch": 1.1101905550952775,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004989576686322468,
      "loss": 6.0713,
      "mean_token_accuracy": 0.13452951610088348,
      "num_tokens": 2446619.0,
      "step": 1340
    },
    {
      "entropy": 6.185806179046631,
      "epoch": 1.1143330571665286,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004989267192929272,
      "loss": 6.1432,
      "mean_token_accuracy": 0.12741851955652236,
      "num_tokens": 2454998.0,
      "step": 1345
    },
    {
      "entropy": 6.214543724060059,
      "epoch": 1.1184755592377795,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004988953182689445,
      "loss": 6.117,
      "mean_token_accuracy": 0.12976484671235083,
      "num_tokens": 2463497.0,
      "step": 1350
    },
    {
      "entropy": 6.218276834487915,
      "epoch": 1.1226180613090306,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004988634656236378,
      "loss": 6.0278,
      "mean_token_accuracy": 0.1300082579255104,
      "num_tokens": 2471763.0,
      "step": 1355
    },
    {
      "entropy": 6.184001922607422,
      "epoch": 1.1267605633802817,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988311614212576,
      "loss": 6.0879,
      "mean_token_accuracy": 0.13324105516076087,
      "num_tokens": 2479949.0,
      "step": 1360
    },
    {
      "entropy": 6.281342220306397,
      "epoch": 1.1309030654515326,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987984057269648,
      "loss": 6.1428,
      "mean_token_accuracy": 0.1314686894416809,
      "num_tokens": 2489097.0,
      "step": 1365
    },
    {
      "entropy": 6.040010833740235,
      "epoch": 1.1350455675227837,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987651986068311,
      "loss": 5.9729,
      "mean_token_accuracy": 0.13274277672171592,
      "num_tokens": 2497625.0,
      "step": 1370
    },
    {
      "entropy": 6.246344804763794,
      "epoch": 1.1391880695940348,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987315401278394,
      "loss": 6.216,
      "mean_token_accuracy": 0.12787129282951354,
      "num_tokens": 2507107.0,
      "step": 1375
    },
    {
      "entropy": 6.39310736656189,
      "epoch": 1.143330571665286,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004986974303578819,
      "loss": 6.2224,
      "mean_token_accuracy": 0.12255537360906602,
      "num_tokens": 2516629.0,
      "step": 1380
    },
    {
      "entropy": 6.220417118072509,
      "epoch": 1.1474730737365368,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986628693657621,
      "loss": 6.0929,
      "mean_token_accuracy": 0.13176359012722969,
      "num_tokens": 2525924.0,
      "step": 1385
    },
    {
      "entropy": 6.315376138687133,
      "epoch": 1.151615575807788,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004986278572211933,
      "loss": 6.2518,
      "mean_token_accuracy": 0.12930550798773766,
      "num_tokens": 2535586.0,
      "step": 1390
    },
    {
      "entropy": 6.2914097785949705,
      "epoch": 1.1557580778790388,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004985923939947987,
      "loss": 6.3282,
      "mean_token_accuracy": 0.12142003327608109,
      "num_tokens": 2546685.0,
      "step": 1395
    },
    {
      "entropy": 6.185162925720215,
      "epoch": 1.15990057995029,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004985564797581115,
      "loss": 6.0433,
      "mean_token_accuracy": 0.12628076002001762,
      "num_tokens": 2556873.0,
      "step": 1400
    },
    {
      "entropy": 6.250091075897217,
      "epoch": 1.164043082021541,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004985201145835747,
      "loss": 6.0778,
      "mean_token_accuracy": 0.13431011289358138,
      "num_tokens": 2565222.0,
      "step": 1405
    },
    {
      "entropy": 6.09890832901001,
      "epoch": 1.1681855840927922,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498483298544541,
      "loss": 6.0752,
      "mean_token_accuracy": 0.1278785362839699,
      "num_tokens": 2575467.0,
      "step": 1410
    },
    {
      "entropy": 6.252477359771729,
      "epoch": 1.172328086164043,
      "grad_norm": 1.15625,
      "learning_rate": 0.000498446031715272,
      "loss": 6.0932,
      "mean_token_accuracy": 0.1310752049088478,
      "num_tokens": 2583963.0,
      "step": 1415
    },
    {
      "entropy": 6.18706784248352,
      "epoch": 1.1764705882352942,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004984083141709391,
      "loss": 6.1308,
      "mean_token_accuracy": 0.1298580549657345,
      "num_tokens": 2593350.0,
      "step": 1420
    },
    {
      "entropy": 6.089730072021484,
      "epoch": 1.180613090306545,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498370145987623,
      "loss": 6.0316,
      "mean_token_accuracy": 0.13288710117340088,
      "num_tokens": 2602191.0,
      "step": 1425
    },
    {
      "entropy": 6.324441194534302,
      "epoch": 1.1847555923777962,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004983315272423127,
      "loss": 6.176,
      "mean_token_accuracy": 0.12697831466794013,
      "num_tokens": 2613590.0,
      "step": 1430
    },
    {
      "entropy": 6.06857738494873,
      "epoch": 1.1888980944490473,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004982924580129066,
      "loss": 5.9812,
      "mean_token_accuracy": 0.13065367564558983,
      "num_tokens": 2622072.0,
      "step": 1435
    },
    {
      "entropy": 6.248331546783447,
      "epoch": 1.1930405965202984,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004982529383782116,
      "loss": 6.1865,
      "mean_token_accuracy": 0.1266859605908394,
      "num_tokens": 2630904.0,
      "step": 1440
    },
    {
      "entropy": 6.229864883422851,
      "epoch": 1.1971830985915493,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982129684179432,
      "loss": 6.1479,
      "mean_token_accuracy": 0.12835109010338783,
      "num_tokens": 2639925.0,
      "step": 1445
    },
    {
      "entropy": 6.201044178009033,
      "epoch": 1.2013256006628004,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004981725482127251,
      "loss": 6.0555,
      "mean_token_accuracy": 0.13077252507209777,
      "num_tokens": 2649397.0,
      "step": 1450
    },
    {
      "entropy": 6.251532125473022,
      "epoch": 1.2054681027340513,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004981316778440892,
      "loss": 6.1979,
      "mean_token_accuracy": 0.1274854823946953,
      "num_tokens": 2658877.0,
      "step": 1455
    },
    {
      "entropy": 6.10891489982605,
      "epoch": 1.2096106048053024,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004980903573944756,
      "loss": 6.0963,
      "mean_token_accuracy": 0.1251230150461197,
      "num_tokens": 2668518.0,
      "step": 1460
    },
    {
      "entropy": 6.220654296875,
      "epoch": 1.2137531068765535,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004980485869472322,
      "loss": 6.2062,
      "mean_token_accuracy": 0.1263741858303547,
      "num_tokens": 2677459.0,
      "step": 1465
    },
    {
      "entropy": 6.302640914916992,
      "epoch": 1.2178956089478046,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004980063665866143,
      "loss": 6.1744,
      "mean_token_accuracy": 0.1222662165760994,
      "num_tokens": 2686668.0,
      "step": 1470
    },
    {
      "entropy": 6.034982395172119,
      "epoch": 1.2220381110190555,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004979636963977853,
      "loss": 5.9536,
      "mean_token_accuracy": 0.13268350437283516,
      "num_tokens": 2695545.0,
      "step": 1475
    },
    {
      "entropy": 6.225825643539428,
      "epoch": 1.2261806130903066,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004979205764668156,
      "loss": 6.1317,
      "mean_token_accuracy": 0.12943535372614862,
      "num_tokens": 2705514.0,
      "step": 1480
    },
    {
      "entropy": 6.211910057067871,
      "epoch": 1.2303231151615575,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004978770068806825,
      "loss": 6.0386,
      "mean_token_accuracy": 0.12719125896692277,
      "num_tokens": 2714373.0,
      "step": 1485
    },
    {
      "entropy": 6.15532283782959,
      "epoch": 1.2344656172328086,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497832987727271,
      "loss": 6.1529,
      "mean_token_accuracy": 0.12199215218424797,
      "num_tokens": 2723887.0,
      "step": 1490
    },
    {
      "entropy": 6.166454887390136,
      "epoch": 1.2386081193040597,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004977885190953724,
      "loss": 6.0362,
      "mean_token_accuracy": 0.12878855913877488,
      "num_tokens": 2733591.0,
      "step": 1495
    },
    {
      "entropy": 6.207261800765991,
      "epoch": 1.2427506213753108,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004977436010746846,
      "loss": 6.1471,
      "mean_token_accuracy": 0.13749015629291533,
      "num_tokens": 2741638.0,
      "step": 1500
    },
    {
      "epoch": 1.2427506213753108,
      "eval_entropy": 6.123490674155099,
      "eval_loss": 6.331637382507324,
      "eval_mean_token_accuracy": 0.12770517323730088,
      "eval_num_tokens": 2741638.0,
      "eval_runtime": 2.9708,
      "eval_samples_per_second": 1261.266,
      "eval_steps_per_second": 157.869,
      "step": 1500
    },
    {
      "entropy": 6.160883140563965,
      "epoch": 1.2468931234465617,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004976982337558124,
      "loss": 6.0387,
      "mean_token_accuracy": 0.133138719946146,
      "num_tokens": 2749711.0,
      "step": 1505
    },
    {
      "entropy": 6.207182788848877,
      "epoch": 1.2510356255178128,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004976524172302665,
      "loss": 6.2191,
      "mean_token_accuracy": 0.12747407108545303,
      "num_tokens": 2758121.0,
      "step": 1510
    },
    {
      "entropy": 6.156457328796387,
      "epoch": 1.2551781275890637,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004976061515904641,
      "loss": 6.2125,
      "mean_token_accuracy": 0.12646155208349227,
      "num_tokens": 2766764.0,
      "step": 1515
    },
    {
      "entropy": 6.284233474731446,
      "epoch": 1.2593206296603148,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004975594369297277,
      "loss": 6.1469,
      "mean_token_accuracy": 0.1318532146513462,
      "num_tokens": 2775561.0,
      "step": 1520
    },
    {
      "entropy": 6.1375274658203125,
      "epoch": 1.2634631317315659,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004975122733422861,
      "loss": 6.1051,
      "mean_token_accuracy": 0.1313116282224655,
      "num_tokens": 2784691.0,
      "step": 1525
    },
    {
      "entropy": 6.224356698989868,
      "epoch": 1.267605633802817,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004974646609232732,
      "loss": 6.0634,
      "mean_token_accuracy": 0.13405025228857995,
      "num_tokens": 2793893.0,
      "step": 1530
    },
    {
      "entropy": 6.1460289478302,
      "epoch": 1.2717481358740679,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004974165997687288,
      "loss": 6.1463,
      "mean_token_accuracy": 0.12966038882732392,
      "num_tokens": 2804479.0,
      "step": 1535
    },
    {
      "entropy": 6.288549613952637,
      "epoch": 1.275890637945319,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004973680899755972,
      "loss": 6.1883,
      "mean_token_accuracy": 0.12688734605908394,
      "num_tokens": 2813966.0,
      "step": 1540
    },
    {
      "entropy": 6.1371747016906735,
      "epoch": 1.2800331400165699,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000497319131641728,
      "loss": 5.998,
      "mean_token_accuracy": 0.13688533008098602,
      "num_tokens": 2822328.0,
      "step": 1545
    },
    {
      "entropy": 6.090888452529907,
      "epoch": 1.284175642087821,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004972697248658755,
      "loss": 5.9801,
      "mean_token_accuracy": 0.13497933447360994,
      "num_tokens": 2829845.0,
      "step": 1550
    },
    {
      "entropy": 6.0702019214630125,
      "epoch": 1.288318144159072,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004972198697476987,
      "loss": 5.9519,
      "mean_token_accuracy": 0.13229596316814424,
      "num_tokens": 2839040.0,
      "step": 1555
    },
    {
      "entropy": 6.079361152648926,
      "epoch": 1.2924606462303232,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004971695663877607,
      "loss": 6.043,
      "mean_token_accuracy": 0.13861832544207572,
      "num_tokens": 2848574.0,
      "step": 1560
    },
    {
      "entropy": 6.124926471710205,
      "epoch": 1.296603148301574,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004971188148875289,
      "loss": 6.0208,
      "mean_token_accuracy": 0.14211280643939972,
      "num_tokens": 2856812.0,
      "step": 1565
    },
    {
      "entropy": 6.180253314971924,
      "epoch": 1.3007456503728252,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004970676153493748,
      "loss": 6.1061,
      "mean_token_accuracy": 0.12911001816391945,
      "num_tokens": 2866001.0,
      "step": 1570
    },
    {
      "entropy": 6.168315410614014,
      "epoch": 1.3048881524440763,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004970159678765731,
      "loss": 6.1004,
      "mean_token_accuracy": 0.12929908260703088,
      "num_tokens": 2875181.0,
      "step": 1575
    },
    {
      "entropy": 6.167870616912841,
      "epoch": 1.3090306545153272,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004969638725733028,
      "loss": 6.0246,
      "mean_token_accuracy": 0.13357625305652618,
      "num_tokens": 2885103.0,
      "step": 1580
    },
    {
      "entropy": 6.009994029998779,
      "epoch": 1.3131731565865783,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004969113295446455,
      "loss": 5.9852,
      "mean_token_accuracy": 0.13527049124240875,
      "num_tokens": 2894055.0,
      "step": 1585
    },
    {
      "entropy": 6.2158726215362545,
      "epoch": 1.3173156586578294,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004968583388965863,
      "loss": 6.0608,
      "mean_token_accuracy": 0.13581639677286148,
      "num_tokens": 2903391.0,
      "step": 1590
    },
    {
      "entropy": 6.093079280853272,
      "epoch": 1.3214581607290803,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004968049007360132,
      "loss": 6.0605,
      "mean_token_accuracy": 0.13181073293089868,
      "num_tokens": 2911760.0,
      "step": 1595
    },
    {
      "entropy": 6.063480806350708,
      "epoch": 1.3256006628003314,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004967510151707169,
      "loss": 6.0138,
      "mean_token_accuracy": 0.13460284769535064,
      "num_tokens": 2920589.0,
      "step": 1600
    },
    {
      "entropy": 6.224989795684815,
      "epoch": 1.3297431648715825,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004966966823093903,
      "loss": 6.1592,
      "mean_token_accuracy": 0.12850213572382926,
      "num_tokens": 2929635.0,
      "step": 1605
    },
    {
      "entropy": 6.084767436981201,
      "epoch": 1.3338856669428334,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004966419022616287,
      "loss": 6.0021,
      "mean_token_accuracy": 0.13327318727970122,
      "num_tokens": 2938778.0,
      "step": 1610
    },
    {
      "entropy": 6.075594139099121,
      "epoch": 1.3380281690140845,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004965866751379296,
      "loss": 6.1079,
      "mean_token_accuracy": 0.1267368644475937,
      "num_tokens": 2947836.0,
      "step": 1615
    },
    {
      "entropy": 6.235728120803833,
      "epoch": 1.3421706710853356,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004965310010496922,
      "loss": 6.0749,
      "mean_token_accuracy": 0.12694141641259193,
      "num_tokens": 2957028.0,
      "step": 1620
    },
    {
      "entropy": 6.089836168289184,
      "epoch": 1.3463131731565865,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004964748801092168,
      "loss": 6.0329,
      "mean_token_accuracy": 0.12828338146209717,
      "num_tokens": 2966088.0,
      "step": 1625
    },
    {
      "entropy": 6.1745452880859375,
      "epoch": 1.3504556752278376,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004964183124297059,
      "loss": 6.1172,
      "mean_token_accuracy": 0.1309424474835396,
      "num_tokens": 2975271.0,
      "step": 1630
    },
    {
      "entropy": 6.151106929779052,
      "epoch": 1.3545981772990887,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004963612981252625,
      "loss": 6.1097,
      "mean_token_accuracy": 0.13138212487101555,
      "num_tokens": 2983799.0,
      "step": 1635
    },
    {
      "entropy": 6.1551590919494625,
      "epoch": 1.3587406793703396,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004963038373108908,
      "loss": 6.0853,
      "mean_token_accuracy": 0.13040598779916762,
      "num_tokens": 2992675.0,
      "step": 1640
    },
    {
      "entropy": 6.182181072235108,
      "epoch": 1.3628831814415907,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004962459301024954,
      "loss": 6.1637,
      "mean_token_accuracy": 0.12975483611226082,
      "num_tokens": 3002483.0,
      "step": 1645
    },
    {
      "entropy": 6.180190896987915,
      "epoch": 1.3670256835128418,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004961875766168816,
      "loss": 6.0541,
      "mean_token_accuracy": 0.13903512582182884,
      "num_tokens": 3010611.0,
      "step": 1650
    },
    {
      "entropy": 6.102008962631226,
      "epoch": 1.371168185584093,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004961287769717547,
      "loss": 6.0193,
      "mean_token_accuracy": 0.12874715775251389,
      "num_tokens": 3019928.0,
      "step": 1655
    },
    {
      "entropy": 6.066634035110473,
      "epoch": 1.3753106876553438,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004960695312857199,
      "loss": 6.0043,
      "mean_token_accuracy": 0.13317709937691688,
      "num_tokens": 3029003.0,
      "step": 1660
    },
    {
      "entropy": 6.0940875053405765,
      "epoch": 1.379453189726595,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004960098396782824,
      "loss": 6.0235,
      "mean_token_accuracy": 0.13606003373861314,
      "num_tokens": 3038049.0,
      "step": 1665
    },
    {
      "entropy": 6.055018329620362,
      "epoch": 1.3835956917978458,
      "grad_norm": 1.0,
      "learning_rate": 0.0004959497022698466,
      "loss": 6.0211,
      "mean_token_accuracy": 0.1442451775074005,
      "num_tokens": 3046674.0,
      "step": 1670
    },
    {
      "entropy": 6.138688325881958,
      "epoch": 1.387738193869097,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004958891191817163,
      "loss": 6.0594,
      "mean_token_accuracy": 0.1267915315926075,
      "num_tokens": 3056459.0,
      "step": 1675
    },
    {
      "entropy": 6.11493444442749,
      "epoch": 1.391880695940348,
      "grad_norm": 1.140625,
      "learning_rate": 0.000495828090536094,
      "loss": 6.0525,
      "mean_token_accuracy": 0.1292160801589489,
      "num_tokens": 3066294.0,
      "step": 1680
    },
    {
      "entropy": 6.1094911098480225,
      "epoch": 1.3960231980115991,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004957666164560815,
      "loss": 6.017,
      "mean_token_accuracy": 0.13060638532042504,
      "num_tokens": 3075420.0,
      "step": 1685
    },
    {
      "entropy": 6.1826433658599855,
      "epoch": 1.40016570008285,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004957046970656784,
      "loss": 6.1106,
      "mean_token_accuracy": 0.1357931189239025,
      "num_tokens": 3084373.0,
      "step": 1690
    },
    {
      "entropy": 6.090426683425903,
      "epoch": 1.4043082021541011,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004956423324897831,
      "loss": 6.0819,
      "mean_token_accuracy": 0.1346761591732502,
      "num_tokens": 3094212.0,
      "step": 1695
    },
    {
      "entropy": 6.164022970199585,
      "epoch": 1.408450704225352,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004955795228541917,
      "loss": 6.0205,
      "mean_token_accuracy": 0.14607309103012084,
      "num_tokens": 3102769.0,
      "step": 1700
    },
    {
      "entropy": 6.0860377788543705,
      "epoch": 1.4125932062966031,
      "grad_norm": 1.109375,
      "learning_rate": 0.000495516268285598,
      "loss": 6.0509,
      "mean_token_accuracy": 0.13397998213768006,
      "num_tokens": 3111465.0,
      "step": 1705
    },
    {
      "entropy": 6.0558473587036135,
      "epoch": 1.4167357083678542,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004954525689115935,
      "loss": 5.9885,
      "mean_token_accuracy": 0.13343949466943741,
      "num_tokens": 3121106.0,
      "step": 1710
    },
    {
      "entropy": 6.11056923866272,
      "epoch": 1.4208782104391053,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004953884248606669,
      "loss": 6.0095,
      "mean_token_accuracy": 0.14755313694477082,
      "num_tokens": 3130662.0,
      "step": 1715
    },
    {
      "entropy": 6.031003999710083,
      "epoch": 1.4250207125103562,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004953238362622034,
      "loss": 6.0571,
      "mean_token_accuracy": 0.1311265155673027,
      "num_tokens": 3139538.0,
      "step": 1720
    },
    {
      "entropy": 6.109453105926514,
      "epoch": 1.4291632145816073,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004952588032464855,
      "loss": 6.1066,
      "mean_token_accuracy": 0.1264252707362175,
      "num_tokens": 3148552.0,
      "step": 1725
    },
    {
      "entropy": 6.092384767532349,
      "epoch": 1.4333057166528582,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004951933259446919,
      "loss": 6.0087,
      "mean_token_accuracy": 0.13966907411813737,
      "num_tokens": 3156549.0,
      "step": 1730
    },
    {
      "entropy": 6.223468208312989,
      "epoch": 1.4374482187241093,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004951274044888975,
      "loss": 6.1692,
      "mean_token_accuracy": 0.12101665809750557,
      "num_tokens": 3166821.0,
      "step": 1735
    },
    {
      "entropy": 6.01346492767334,
      "epoch": 1.4415907207953604,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004950610390120731,
      "loss": 6.0518,
      "mean_token_accuracy": 0.1328336514532566,
      "num_tokens": 3174962.0,
      "step": 1740
    },
    {
      "entropy": 6.186932849884033,
      "epoch": 1.4457332228666115,
      "grad_norm": 1.046875,
      "learning_rate": 0.000494994229648085,
      "loss": 5.957,
      "mean_token_accuracy": 0.13281209170818328,
      "num_tokens": 3184027.0,
      "step": 1745
    },
    {
      "entropy": 6.017684364318848,
      "epoch": 1.4498757249378624,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004949269765316948,
      "loss": 6.0971,
      "mean_token_accuracy": 0.1264866605401039,
      "num_tokens": 3193186.0,
      "step": 1750
    },
    {
      "entropy": 6.227017116546631,
      "epoch": 1.4540182270091135,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004948592797985599,
      "loss": 6.1057,
      "mean_token_accuracy": 0.12643057182431222,
      "num_tokens": 3202676.0,
      "step": 1755
    },
    {
      "entropy": 6.240481233596801,
      "epoch": 1.4581607290803644,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004947911395852317,
      "loss": 6.2243,
      "mean_token_accuracy": 0.13314623907208442,
      "num_tokens": 3211950.0,
      "step": 1760
    },
    {
      "entropy": 6.1601158618927006,
      "epoch": 1.4623032311516155,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004947225560291565,
      "loss": 6.1576,
      "mean_token_accuracy": 0.12665070593357086,
      "num_tokens": 3221135.0,
      "step": 1765
    },
    {
      "entropy": 6.144661283493042,
      "epoch": 1.4664457332228666,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004946535292686746,
      "loss": 5.9717,
      "mean_token_accuracy": 0.13558048903942108,
      "num_tokens": 3230174.0,
      "step": 1770
    },
    {
      "entropy": 5.993796110153198,
      "epoch": 1.4705882352941178,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004945840594430209,
      "loss": 5.948,
      "mean_token_accuracy": 0.14206447452306747,
      "num_tokens": 3238898.0,
      "step": 1775
    },
    {
      "entropy": 6.077493000030517,
      "epoch": 1.4747307373653686,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004945141466923235,
      "loss": 5.988,
      "mean_token_accuracy": 0.14010899588465692,
      "num_tokens": 3247550.0,
      "step": 1780
    },
    {
      "entropy": 6.233349943161011,
      "epoch": 1.4788732394366197,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004944437911576041,
      "loss": 6.1399,
      "mean_token_accuracy": 0.12699589803814887,
      "num_tokens": 3256810.0,
      "step": 1785
    },
    {
      "entropy": 5.942909288406372,
      "epoch": 1.4830157415078706,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004943729929807772,
      "loss": 5.9276,
      "mean_token_accuracy": 0.13631228655576705,
      "num_tokens": 3265886.0,
      "step": 1790
    },
    {
      "entropy": 6.0962855339050295,
      "epoch": 1.4871582435791217,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004943017523046507,
      "loss": 6.0033,
      "mean_token_accuracy": 0.13771791383624077,
      "num_tokens": 3274522.0,
      "step": 1795
    },
    {
      "entropy": 6.160940313339234,
      "epoch": 1.4913007456503728,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000494230069272925,
      "loss": 6.1139,
      "mean_token_accuracy": 0.1388836421072483,
      "num_tokens": 3283388.0,
      "step": 1800
    },
    {
      "entropy": 6.049196195602417,
      "epoch": 1.495443247721624,
      "grad_norm": 1.125,
      "learning_rate": 0.0004941579440301923,
      "loss": 5.9919,
      "mean_token_accuracy": 0.13140289038419722,
      "num_tokens": 3291887.0,
      "step": 1805
    },
    {
      "entropy": 6.019881868362427,
      "epoch": 1.4995857497928748,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004940853767219372,
      "loss": 6.0475,
      "mean_token_accuracy": 0.1365261606872082,
      "num_tokens": 3300674.0,
      "step": 1810
    },
    {
      "entropy": 6.014644193649292,
      "epoch": 1.503728251864126,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004940123674945359,
      "loss": 5.8999,
      "mean_token_accuracy": 0.13877790048718452,
      "num_tokens": 3309745.0,
      "step": 1815
    },
    {
      "entropy": 6.119295310974121,
      "epoch": 1.5078707539353768,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004939389164952561,
      "loss": 6.0591,
      "mean_token_accuracy": 0.13487198427319527,
      "num_tokens": 3318888.0,
      "step": 1820
    },
    {
      "entropy": 5.966858196258545,
      "epoch": 1.512013256006628,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004938650238722564,
      "loss": 5.9897,
      "mean_token_accuracy": 0.13288637325167657,
      "num_tokens": 3328274.0,
      "step": 1825
    },
    {
      "entropy": 6.136635732650757,
      "epoch": 1.516155758077879,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004937906897745861,
      "loss": 6.0095,
      "mean_token_accuracy": 0.14079945385456086,
      "num_tokens": 3337579.0,
      "step": 1830
    },
    {
      "entropy": 6.079625654220581,
      "epoch": 1.5202982601491302,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004937159143521856,
      "loss": 5.9861,
      "mean_token_accuracy": 0.13965884745121002,
      "num_tokens": 3346262.0,
      "step": 1835
    },
    {
      "entropy": 6.045686912536621,
      "epoch": 1.5244407622203813,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004936406977558847,
      "loss": 6.0836,
      "mean_token_accuracy": 0.12937265560030936,
      "num_tokens": 3354833.0,
      "step": 1840
    },
    {
      "entropy": 6.17369589805603,
      "epoch": 1.5285832642916322,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004935650401374039,
      "loss": 6.1543,
      "mean_token_accuracy": 0.14100572615861892,
      "num_tokens": 3364314.0,
      "step": 1845
    },
    {
      "entropy": 6.07831563949585,
      "epoch": 1.532725766362883,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004934889416493524,
      "loss": 5.9693,
      "mean_token_accuracy": 0.13386041671037674,
      "num_tokens": 3373648.0,
      "step": 1850
    },
    {
      "entropy": 5.98188886642456,
      "epoch": 1.5368682684341342,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004934124024452295,
      "loss": 5.8443,
      "mean_token_accuracy": 0.14330715388059617,
      "num_tokens": 3382919.0,
      "step": 1855
    },
    {
      "entropy": 6.019197273254394,
      "epoch": 1.5410107705053853,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004933354226794232,
      "loss": 5.9833,
      "mean_token_accuracy": 0.1422867774963379,
      "num_tokens": 3391814.0,
      "step": 1860
    },
    {
      "entropy": 6.095344734191895,
      "epoch": 1.5451532725766364,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004932580025072098,
      "loss": 6.0449,
      "mean_token_accuracy": 0.13532943651080132,
      "num_tokens": 3400392.0,
      "step": 1865
    },
    {
      "entropy": 6.067120552062988,
      "epoch": 1.5492957746478875,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004931801420847545,
      "loss": 6.0924,
      "mean_token_accuracy": 0.12777185067534447,
      "num_tokens": 3410430.0,
      "step": 1870
    },
    {
      "entropy": 6.015319013595581,
      "epoch": 1.5534382767191384,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004931018415691102,
      "loss": 5.9287,
      "mean_token_accuracy": 0.13285826817154883,
      "num_tokens": 3420654.0,
      "step": 1875
    },
    {
      "entropy": 6.0661335468292235,
      "epoch": 1.5575807787903893,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004930231011182176,
      "loss": 6.0527,
      "mean_token_accuracy": 0.13392168357968331,
      "num_tokens": 3430514.0,
      "step": 1880
    },
    {
      "entropy": 6.158822917938233,
      "epoch": 1.5617232808616404,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004929439208909048,
      "loss": 6.0031,
      "mean_token_accuracy": 0.13457945808768274,
      "num_tokens": 3439551.0,
      "step": 1885
    },
    {
      "entropy": 5.95737943649292,
      "epoch": 1.5658657829328915,
      "grad_norm": 1.125,
      "learning_rate": 0.0004928643010468869,
      "loss": 5.9409,
      "mean_token_accuracy": 0.13493541032075881,
      "num_tokens": 3448160.0,
      "step": 1890
    },
    {
      "entropy": 6.1136833190917965,
      "epoch": 1.5700082850041426,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004927842417467659,
      "loss": 6.0155,
      "mean_token_accuracy": 0.14148347079753876,
      "num_tokens": 3457296.0,
      "step": 1895
    },
    {
      "entropy": 6.018945932388306,
      "epoch": 1.5741507870753937,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004927037431520302,
      "loss": 5.9566,
      "mean_token_accuracy": 0.14098128974437713,
      "num_tokens": 3466108.0,
      "step": 1900
    },
    {
      "entropy": 6.053025722503662,
      "epoch": 1.5782932891466446,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004926228054250541,
      "loss": 6.0408,
      "mean_token_accuracy": 0.1393430083990097,
      "num_tokens": 3475336.0,
      "step": 1905
    },
    {
      "entropy": 6.099352216720581,
      "epoch": 1.5824357912178955,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004925414287290979,
      "loss": 6.0325,
      "mean_token_accuracy": 0.1402113474905491,
      "num_tokens": 3483963.0,
      "step": 1910
    },
    {
      "entropy": 6.013989686965942,
      "epoch": 1.5865782932891466,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004924596132283075,
      "loss": 5.9376,
      "mean_token_accuracy": 0.14408371299505235,
      "num_tokens": 3492947.0,
      "step": 1915
    },
    {
      "entropy": 5.973308038711548,
      "epoch": 1.5907207953603977,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004923773590877135,
      "loss": 6.029,
      "mean_token_accuracy": 0.13019290938973427,
      "num_tokens": 3501877.0,
      "step": 1920
    },
    {
      "entropy": 6.14404263496399,
      "epoch": 1.5948632974316488,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004922946664732314,
      "loss": 6.074,
      "mean_token_accuracy": 0.12927663773298265,
      "num_tokens": 3510610.0,
      "step": 1925
    },
    {
      "entropy": 6.131452226638794,
      "epoch": 1.5990057995029,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004922115355516615,
      "loss": 6.0328,
      "mean_token_accuracy": 0.13440151512622833,
      "num_tokens": 3519829.0,
      "step": 1930
    },
    {
      "entropy": 6.053449392318726,
      "epoch": 1.6031483015741508,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004921279664906877,
      "loss": 6.0203,
      "mean_token_accuracy": 0.1330590568482876,
      "num_tokens": 3529408.0,
      "step": 1935
    },
    {
      "entropy": 5.981090784072876,
      "epoch": 1.6072908036454017,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004920439594588781,
      "loss": 5.9579,
      "mean_token_accuracy": 0.14276544600725175,
      "num_tokens": 3538068.0,
      "step": 1940
    },
    {
      "entropy": 6.187561845779419,
      "epoch": 1.6114333057166528,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000491959514625684,
      "loss": 6.0732,
      "mean_token_accuracy": 0.1400441735982895,
      "num_tokens": 3547302.0,
      "step": 1945
    },
    {
      "entropy": 6.01104588508606,
      "epoch": 1.6155758077879039,
      "grad_norm": 1.140625,
      "learning_rate": 0.00049187463216144,
      "loss": 5.9397,
      "mean_token_accuracy": 0.14568847864866258,
      "num_tokens": 3556087.0,
      "step": 1950
    },
    {
      "entropy": 6.153070068359375,
      "epoch": 1.619718309859155,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000491789312237363,
      "loss": 6.0312,
      "mean_token_accuracy": 0.13621370121836662,
      "num_tokens": 3566235.0,
      "step": 1955
    },
    {
      "entropy": 5.994067668914795,
      "epoch": 1.623860811930406,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004917035550255528,
      "loss": 5.9618,
      "mean_token_accuracy": 0.1392713963985443,
      "num_tokens": 3574762.0,
      "step": 1960
    },
    {
      "entropy": 5.9850420475006105,
      "epoch": 1.628003314001657,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000491617360698991,
      "loss": 5.9424,
      "mean_token_accuracy": 0.13408162221312522,
      "num_tokens": 3583359.0,
      "step": 1965
    },
    {
      "entropy": 5.994701671600342,
      "epoch": 1.6321458160729079,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004915307294315412,
      "loss": 5.9009,
      "mean_token_accuracy": 0.13760913535952568,
      "num_tokens": 3591436.0,
      "step": 1970
    },
    {
      "entropy": 5.980083847045899,
      "epoch": 1.636288318144159,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000491443661397948,
      "loss": 5.9868,
      "mean_token_accuracy": 0.1349792368710041,
      "num_tokens": 3601054.0,
      "step": 1975
    },
    {
      "entropy": 6.052795314788819,
      "epoch": 1.64043082021541,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004913561567738368,
      "loss": 5.9674,
      "mean_token_accuracy": 0.13878381252288818,
      "num_tokens": 3610641.0,
      "step": 1980
    },
    {
      "entropy": 5.99411187171936,
      "epoch": 1.6445733222866612,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004912682157357146,
      "loss": 5.9041,
      "mean_token_accuracy": 0.14029043316841125,
      "num_tokens": 3619351.0,
      "step": 1985
    },
    {
      "entropy": 5.911650848388672,
      "epoch": 1.6487158243579123,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004911798384609675,
      "loss": 5.8879,
      "mean_token_accuracy": 0.14449498876929284,
      "num_tokens": 3627341.0,
      "step": 1990
    },
    {
      "entropy": 6.012162733078003,
      "epoch": 1.6528583264291632,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004910910251278626,
      "loss": 5.9452,
      "mean_token_accuracy": 0.14296462759375572,
      "num_tokens": 3637132.0,
      "step": 1995
    },
    {
      "entropy": 6.013031959533691,
      "epoch": 1.6570008285004143,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004910017759155458,
      "loss": 6.0021,
      "mean_token_accuracy": 0.13557344153523446,
      "num_tokens": 3646196.0,
      "step": 2000
    },
    {
      "epoch": 1.6570008285004143,
      "eval_entropy": 5.924376028178852,
      "eval_loss": 6.150927543640137,
      "eval_mean_token_accuracy": 0.13550341723443093,
      "eval_num_tokens": 3646196.0,
      "eval_runtime": 2.7516,
      "eval_samples_per_second": 1361.773,
      "eval_steps_per_second": 170.449,
      "step": 2000
    },
    {
      "entropy": 6.026213884353638,
      "epoch": 1.6611433305716652,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004909120910040427,
      "loss": 5.9388,
      "mean_token_accuracy": 0.14352744221687316,
      "num_tokens": 3654658.0,
      "step": 2005
    },
    {
      "entropy": 6.142762756347656,
      "epoch": 1.6652858326429163,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004908219705742574,
      "loss": 6.2148,
      "mean_token_accuracy": 0.13257693275809287,
      "num_tokens": 3664509.0,
      "step": 2010
    },
    {
      "entropy": 6.105361080169677,
      "epoch": 1.6694283347141674,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004907314148079727,
      "loss": 5.917,
      "mean_token_accuracy": 0.14199795499444007,
      "num_tokens": 3672983.0,
      "step": 2015
    },
    {
      "entropy": 5.827013301849365,
      "epoch": 1.6735708367854185,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004906404238878496,
      "loss": 5.8004,
      "mean_token_accuracy": 0.14207299277186394,
      "num_tokens": 3681196.0,
      "step": 2020
    },
    {
      "entropy": 6.070524263381958,
      "epoch": 1.6777133388566694,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004905489979974265,
      "loss": 5.9619,
      "mean_token_accuracy": 0.13726090565323829,
      "num_tokens": 3690077.0,
      "step": 2025
    },
    {
      "entropy": 6.060069894790649,
      "epoch": 1.6818558409279205,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004904571373211197,
      "loss": 6.0231,
      "mean_token_accuracy": 0.13792718350887298,
      "num_tokens": 3700121.0,
      "step": 2030
    },
    {
      "entropy": 5.907645797729492,
      "epoch": 1.6859983429991714,
      "grad_norm": 1.046875,
      "learning_rate": 0.000490364842044222,
      "loss": 5.8567,
      "mean_token_accuracy": 0.14549633860588074,
      "num_tokens": 3709618.0,
      "step": 2035
    },
    {
      "entropy": 6.038069868087769,
      "epoch": 1.6901408450704225,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004902721123529031,
      "loss": 6.0487,
      "mean_token_accuracy": 0.12704455628991126,
      "num_tokens": 3719570.0,
      "step": 2040
    },
    {
      "entropy": 6.154643487930298,
      "epoch": 1.6942833471416736,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000490178948434209,
      "loss": 5.9776,
      "mean_token_accuracy": 0.14316998422145844,
      "num_tokens": 3729354.0,
      "step": 2045
    },
    {
      "entropy": 5.893994522094727,
      "epoch": 1.6984258492129247,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004900853504760614,
      "loss": 5.8869,
      "mean_token_accuracy": 0.14341987296938896,
      "num_tokens": 3738453.0,
      "step": 2050
    },
    {
      "entropy": 5.962806272506714,
      "epoch": 1.7025683512841756,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004899913186672577,
      "loss": 5.8651,
      "mean_token_accuracy": 0.14164601415395736,
      "num_tokens": 3747815.0,
      "step": 2055
    },
    {
      "entropy": 5.914563226699829,
      "epoch": 1.7067108533554267,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004898968531974702,
      "loss": 5.9676,
      "mean_token_accuracy": 0.14769200831651688,
      "num_tokens": 3757298.0,
      "step": 2060
    },
    {
      "entropy": 6.056203603744507,
      "epoch": 1.7108533554266776,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004898019542572462,
      "loss": 5.9602,
      "mean_token_accuracy": 0.13963632434606552,
      "num_tokens": 3765939.0,
      "step": 2065
    },
    {
      "entropy": 6.100585651397705,
      "epoch": 1.7149958574979287,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004897066220380073,
      "loss": 5.995,
      "mean_token_accuracy": 0.1360404372215271,
      "num_tokens": 3775160.0,
      "step": 2070
    },
    {
      "entropy": 5.958986473083496,
      "epoch": 1.7191383595691798,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004896108567320489,
      "loss": 5.8721,
      "mean_token_accuracy": 0.13895440623164176,
      "num_tokens": 3785181.0,
      "step": 2075
    },
    {
      "entropy": 6.122873830795288,
      "epoch": 1.723280861640431,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048951465853254,
      "loss": 5.9952,
      "mean_token_accuracy": 0.13867081850767135,
      "num_tokens": 3794531.0,
      "step": 2080
    },
    {
      "entropy": 5.962896156311035,
      "epoch": 1.7274233637116818,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004894180276335231,
      "loss": 5.9768,
      "mean_token_accuracy": 0.1366507463157177,
      "num_tokens": 3804602.0,
      "step": 2085
    },
    {
      "entropy": 6.0686544418334964,
      "epoch": 1.731565865782933,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004893209642299132,
      "loss": 6.0243,
      "mean_token_accuracy": 0.13472760021686553,
      "num_tokens": 3813394.0,
      "step": 2090
    },
    {
      "entropy": 6.06077013015747,
      "epoch": 1.7357083678541838,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004892234685174976,
      "loss": 5.9941,
      "mean_token_accuracy": 0.13705680817365645,
      "num_tokens": 3821604.0,
      "step": 2095
    },
    {
      "entropy": 6.069777965545654,
      "epoch": 1.739850869925435,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004891255406929359,
      "loss": 5.995,
      "mean_token_accuracy": 0.13641481548547746,
      "num_tokens": 3831977.0,
      "step": 2100
    },
    {
      "entropy": 6.0059997081756595,
      "epoch": 1.743993371996686,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004890271809537595,
      "loss": 5.9453,
      "mean_token_accuracy": 0.14286706522107123,
      "num_tokens": 3840684.0,
      "step": 2105
    },
    {
      "entropy": 5.989671230316162,
      "epoch": 1.7481358740679371,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004889283894983704,
      "loss": 5.8682,
      "mean_token_accuracy": 0.13856151327490807,
      "num_tokens": 3851147.0,
      "step": 2110
    },
    {
      "entropy": 6.041515731811524,
      "epoch": 1.752278376139188,
      "grad_norm": 1.09375,
      "learning_rate": 0.000488829166526042,
      "loss": 6.0099,
      "mean_token_accuracy": 0.1394909791648388,
      "num_tokens": 3859881.0,
      "step": 2115
    },
    {
      "entropy": 6.0923104763031,
      "epoch": 1.7564208782104391,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004887295122369176,
      "loss": 6.0555,
      "mean_token_accuracy": 0.13175006359815597,
      "num_tokens": 3870151.0,
      "step": 2120
    },
    {
      "entropy": 5.974537181854248,
      "epoch": 1.76056338028169,
      "grad_norm": 1.078125,
      "learning_rate": 0.000488629426832011,
      "loss": 5.9759,
      "mean_token_accuracy": 0.13857906460762023,
      "num_tokens": 3878398.0,
      "step": 2125
    },
    {
      "entropy": 6.02483057975769,
      "epoch": 1.7647058823529411,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004885289105132054,
      "loss": 5.9227,
      "mean_token_accuracy": 0.14162865728139878,
      "num_tokens": 3888007.0,
      "step": 2130
    },
    {
      "entropy": 5.9927177906036375,
      "epoch": 1.7688483844241922,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004884279634832532,
      "loss": 5.9141,
      "mean_token_accuracy": 0.14104892462491989,
      "num_tokens": 3896945.0,
      "step": 2135
    },
    {
      "entropy": 6.011741352081299,
      "epoch": 1.7729908864954433,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004883265859457754,
      "loss": 5.9408,
      "mean_token_accuracy": 0.14671368449926375,
      "num_tokens": 3906049.0,
      "step": 2140
    },
    {
      "entropy": 5.927330160140992,
      "epoch": 1.7771333885666944,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048822477810526186,
      "loss": 5.8978,
      "mean_token_accuracy": 0.137970220297575,
      "num_tokens": 3915050.0,
      "step": 2145
    },
    {
      "entropy": 5.98696722984314,
      "epoch": 1.7812758906379453,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048812254016706994,
      "loss": 5.9854,
      "mean_token_accuracy": 0.1412371039390564,
      "num_tokens": 3924031.0,
      "step": 2150
    },
    {
      "entropy": 6.043074274063111,
      "epoch": 1.7854183927091962,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004880198723374247,
      "loss": 6.0205,
      "mean_token_accuracy": 0.14472691193223,
      "num_tokens": 3933756.0,
      "step": 2155
    },
    {
      "entropy": 6.06718225479126,
      "epoch": 1.7895608947804473,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048791677482341845,
      "loss": 6.0066,
      "mean_token_accuracy": 0.13294869661331177,
      "num_tokens": 3943129.0,
      "step": 2160
    },
    {
      "entropy": 5.958523988723755,
      "epoch": 1.7937033968516984,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048781324783301016,
      "loss": 5.8428,
      "mean_token_accuracy": 0.1384035088121891,
      "num_tokens": 3951494.0,
      "step": 2165
    },
    {
      "entropy": 5.979408073425293,
      "epoch": 1.7978458989229495,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048770929157502505,
      "loss": 6.014,
      "mean_token_accuracy": 0.14348282739520074,
      "num_tokens": 3960588.0,
      "step": 2170
    },
    {
      "entropy": 6.067771530151367,
      "epoch": 1.8019884009942007,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004876049062591543,
      "loss": 5.9762,
      "mean_token_accuracy": 0.13583431839942933,
      "num_tokens": 3970135.0,
      "step": 2175
    },
    {
      "entropy": 6.0059386730194095,
      "epoch": 1.8061309030654515,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048750009209595443,
      "loss": 5.9479,
      "mean_token_accuracy": 0.13654519543051719,
      "num_tokens": 3979569.0,
      "step": 2180
    },
    {
      "entropy": 5.87956805229187,
      "epoch": 1.8102734051367024,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048739484929684714,
      "loss": 5.91,
      "mean_token_accuracy": 0.1421449676156044,
      "num_tokens": 3987885.0,
      "step": 2185
    },
    {
      "entropy": 6.040490770339966,
      "epoch": 1.8144159072079535,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048728917807411856,
      "loss": 6.0586,
      "mean_token_accuracy": 0.13321116119623183,
      "num_tokens": 3997726.0,
      "step": 2190
    },
    {
      "entropy": 6.041323566436768,
      "epoch": 1.8185584092792046,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004871830786409191,
      "loss": 5.887,
      "mean_token_accuracy": 0.14867799878120422,
      "num_tokens": 4006437.0,
      "step": 2195
    },
    {
      "entropy": 5.968663835525513,
      "epoch": 1.8227009113504558,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00048707655121126296,
      "loss": 5.8754,
      "mean_token_accuracy": 0.14638606756925582,
      "num_tokens": 4015326.0,
      "step": 2200
    },
    {
      "entropy": 5.972012090682983,
      "epoch": 1.8268434134217069,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048696959600002767,
      "loss": 5.9139,
      "mean_token_accuracy": 0.14176759198307992,
      "num_tokens": 4024294.0,
      "step": 2205
    },
    {
      "entropy": 5.954474020004272,
      "epoch": 1.8309859154929577,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048686221322295347,
      "loss": 5.9376,
      "mean_token_accuracy": 0.14578078538179398,
      "num_tokens": 4033341.0,
      "step": 2210
    },
    {
      "entropy": 5.941730737686157,
      "epoch": 1.8351284175642086,
      "grad_norm": 1.0,
      "learning_rate": 0.00048675440309664314,
      "loss": 5.8816,
      "mean_token_accuracy": 0.14992816746234894,
      "num_tokens": 4042446.0,
      "step": 2215
    },
    {
      "entropy": 5.9512083530426025,
      "epoch": 1.8392709196354597,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048664616583856166,
      "loss": 5.9209,
      "mean_token_accuracy": 0.13979735597968102,
      "num_tokens": 4051849.0,
      "step": 2220
    },
    {
      "entropy": 6.053277587890625,
      "epoch": 1.8434134217067109,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048653750166703516,
      "loss": 6.0092,
      "mean_token_accuracy": 0.14033474326133727,
      "num_tokens": 4061075.0,
      "step": 2225
    },
    {
      "entropy": 6.005611562728882,
      "epoch": 1.847555923777962,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004864284108012513,
      "loss": 5.9982,
      "mean_token_accuracy": 0.12854147851467132,
      "num_tokens": 4071022.0,
      "step": 2230
    },
    {
      "entropy": 5.967685556411743,
      "epoch": 1.851698425849213,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004863188934612583,
      "loss": 5.9007,
      "mean_token_accuracy": 0.14465969204902648,
      "num_tokens": 4080728.0,
      "step": 2235
    },
    {
      "entropy": 5.987842464447022,
      "epoch": 1.855840927920464,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004862089498679645,
      "loss": 5.8568,
      "mean_token_accuracy": 0.14544849768280982,
      "num_tokens": 4088712.0,
      "step": 2240
    },
    {
      "entropy": 5.957550859451294,
      "epoch": 1.8599834299917148,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004860985802431383,
      "loss": 5.8859,
      "mean_token_accuracy": 0.1428893633186817,
      "num_tokens": 4097488.0,
      "step": 2245
    },
    {
      "entropy": 5.878279829025269,
      "epoch": 1.864125932062966,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004859877848094071,
      "loss": 5.8605,
      "mean_token_accuracy": 0.14168896824121474,
      "num_tokens": 4106485.0,
      "step": 2250
    },
    {
      "entropy": 6.049147462844848,
      "epoch": 1.868268434134217,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004858765637902575,
      "loss": 5.8386,
      "mean_token_accuracy": 0.14742093831300734,
      "num_tokens": 4115806.0,
      "step": 2255
    },
    {
      "entropy": 5.996465444564819,
      "epoch": 1.8724109362054682,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004857649174100344,
      "loss": 5.9829,
      "mean_token_accuracy": 0.1403212234377861,
      "num_tokens": 4124636.0,
      "step": 2260
    },
    {
      "entropy": 5.9591474533081055,
      "epoch": 1.8765534382767193,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004856528458939409,
      "loss": 5.9603,
      "mean_token_accuracy": 0.13828378096222876,
      "num_tokens": 4133876.0,
      "step": 2265
    },
    {
      "entropy": 5.940617370605469,
      "epoch": 1.8806959403479702,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004855403494680373,
      "loss": 5.8493,
      "mean_token_accuracy": 0.14110578671097757,
      "num_tokens": 4143555.0,
      "step": 2270
    },
    {
      "entropy": 5.966775751113891,
      "epoch": 1.884838442419221,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048542742835924133,
      "loss": 5.8261,
      "mean_token_accuracy": 0.14103717133402824,
      "num_tokens": 4153007.0,
      "step": 2275
    },
    {
      "entropy": 5.793911170959473,
      "epoch": 1.8889809444904722,
      "grad_norm": 1.0,
      "learning_rate": 0.00048531408279532705,
      "loss": 5.8588,
      "mean_token_accuracy": 0.13940922021865845,
      "num_tokens": 4162107.0,
      "step": 2280
    },
    {
      "entropy": 5.9467339515686035,
      "epoch": 1.8931234465617233,
      "grad_norm": 1.0,
      "learning_rate": 0.00048520031300492507,
      "loss": 5.8562,
      "mean_token_accuracy": 0.1414430856704712,
      "num_tokens": 4170920.0,
      "step": 2285
    },
    {
      "entropy": 5.99835844039917,
      "epoch": 1.8972659486329744,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048508611921752124,
      "loss": 5.8883,
      "mean_token_accuracy": 0.14315918162465097,
      "num_tokens": 4180226.0,
      "step": 2290
    },
    {
      "entropy": 5.902116441726685,
      "epoch": 1.9014084507042255,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048497150166345703,
      "loss": 5.8181,
      "mean_token_accuracy": 0.1446967154741287,
      "num_tokens": 4189283.0,
      "step": 2295
    },
    {
      "entropy": 5.875041151046753,
      "epoch": 1.9055509527754764,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004848564605739287,
      "loss": 5.8647,
      "mean_token_accuracy": 0.14096341878175736,
      "num_tokens": 4199017.0,
      "step": 2300
    },
    {
      "entropy": 6.04994044303894,
      "epoch": 1.9096934548467275,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048474099618098644,
      "loss": 5.9904,
      "mean_token_accuracy": 0.13669014796614648,
      "num_tokens": 4207756.0,
      "step": 2305
    },
    {
      "entropy": 5.923856973648071,
      "epoch": 1.9138359569179784,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004846251087175348,
      "loss": 5.9005,
      "mean_token_accuracy": 0.1407478854060173,
      "num_tokens": 4216721.0,
      "step": 2310
    },
    {
      "entropy": 5.897194051742554,
      "epoch": 1.9179784589892295,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004845087984173312,
      "loss": 5.8789,
      "mean_token_accuracy": 0.14667138159275056,
      "num_tokens": 4225596.0,
      "step": 2315
    },
    {
      "entropy": 5.997752904891968,
      "epoch": 1.9221209610604806,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048439206551498644,
      "loss": 5.9253,
      "mean_token_accuracy": 0.14398268461227418,
      "num_tokens": 4234500.0,
      "step": 2320
    },
    {
      "entropy": 5.8700896263122555,
      "epoch": 1.9262634631317317,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004842749102459635,
      "loss": 5.8781,
      "mean_token_accuracy": 0.15120936185121536,
      "num_tokens": 4243240.0,
      "step": 2325
    },
    {
      "entropy": 5.906232309341431,
      "epoch": 1.9304059652029826,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048415733284657727,
      "loss": 5.8516,
      "mean_token_accuracy": 0.14731550291180612,
      "num_tokens": 4251892.0,
      "step": 2330
    },
    {
      "entropy": 5.97393217086792,
      "epoch": 1.9345484672742337,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004840393335539943,
      "loss": 5.8253,
      "mean_token_accuracy": 0.14521327316761018,
      "num_tokens": 4261447.0,
      "step": 2335
    },
    {
      "entropy": 5.948915719985962,
      "epoch": 1.9386909693454846,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048392091260623195,
      "loss": 5.9664,
      "mean_token_accuracy": 0.1478654459118843,
      "num_tokens": 4270120.0,
      "step": 2340
    },
    {
      "entropy": 5.98240385055542,
      "epoch": 1.9428334714167357,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048380207024215836,
      "loss": 5.8421,
      "mean_token_accuracy": 0.14822718501091003,
      "num_tokens": 4278300.0,
      "step": 2345
    },
    {
      "entropy": 5.955031204223633,
      "epoch": 1.9469759734879868,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048368280670149145,
      "loss": 5.8912,
      "mean_token_accuracy": 0.14697090536355972,
      "num_tokens": 4286454.0,
      "step": 2350
    },
    {
      "entropy": 5.854554843902588,
      "epoch": 1.951118475559238,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004835631222247989,
      "loss": 5.7636,
      "mean_token_accuracy": 0.14784407168626784,
      "num_tokens": 4294752.0,
      "step": 2355
    },
    {
      "entropy": 5.909240818023681,
      "epoch": 1.9552609776304888,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048344301705349735,
      "loss": 5.8298,
      "mean_token_accuracy": 0.14864546656608582,
      "num_tokens": 4303195.0,
      "step": 2360
    },
    {
      "entropy": 5.995252275466919,
      "epoch": 1.95940347970174,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048332249142985195,
      "loss": 5.9603,
      "mean_token_accuracy": 0.13669342547655106,
      "num_tokens": 4311961.0,
      "step": 2365
    },
    {
      "entropy": 5.914461421966553,
      "epoch": 1.9635459817729908,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004832015455969763,
      "loss": 5.8152,
      "mean_token_accuracy": 0.13769099935889245,
      "num_tokens": 4322267.0,
      "step": 2370
    },
    {
      "entropy": 5.856181859970093,
      "epoch": 1.9676884838442419,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048308017979883114,
      "loss": 5.939,
      "mean_token_accuracy": 0.14319972842931747,
      "num_tokens": 4330584.0,
      "step": 2375
    },
    {
      "entropy": 5.907060337066651,
      "epoch": 1.971830985915493,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048295839428022454,
      "loss": 5.7562,
      "mean_token_accuracy": 0.14656164199113847,
      "num_tokens": 4339215.0,
      "step": 2380
    },
    {
      "entropy": 5.901394891738891,
      "epoch": 1.975973487986744,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048283618928681127,
      "loss": 5.7878,
      "mean_token_accuracy": 0.1504148483276367,
      "num_tokens": 4347711.0,
      "step": 2385
    },
    {
      "entropy": 5.778523015975952,
      "epoch": 1.980115990057995,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004827135650650923,
      "loss": 5.7814,
      "mean_token_accuracy": 0.1501213528215885,
      "num_tokens": 4356982.0,
      "step": 2390
    },
    {
      "entropy": 5.905798053741455,
      "epoch": 1.984258492129246,
      "grad_norm": 1.125,
      "learning_rate": 0.0004825905218624139,
      "loss": 5.8019,
      "mean_token_accuracy": 0.148068468272686,
      "num_tokens": 4365378.0,
      "step": 2395
    },
    {
      "entropy": 6.008777904510498,
      "epoch": 1.988400994200497,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048246705992696756,
      "loss": 5.8635,
      "mean_token_accuracy": 0.1390143997967243,
      "num_tokens": 4374007.0,
      "step": 2400
    },
    {
      "entropy": 6.05829381942749,
      "epoch": 1.992543496271748,
      "grad_norm": 1.0,
      "learning_rate": 0.00048234317950778975,
      "loss": 5.9666,
      "mean_token_accuracy": 0.13901185542345046,
      "num_tokens": 4383909.0,
      "step": 2405
    },
    {
      "entropy": 6.023723077774048,
      "epoch": 1.9966859983429992,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004822188808547607,
      "loss": 6.0501,
      "mean_token_accuracy": 0.13643686026334761,
      "num_tokens": 4393145.0,
      "step": 2410
    },
    {
      "entropy": 5.890930223464966,
      "epoch": 2.0008285004142503,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004820941642186043,
      "loss": 5.7449,
      "mean_token_accuracy": 0.14210715517401695,
      "num_tokens": 4402024.0,
      "step": 2415
    },
    {
      "entropy": 5.920894384384155,
      "epoch": 2.0049710024855014,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004819690298508879,
      "loss": 5.5832,
      "mean_token_accuracy": 0.15544317960739135,
      "num_tokens": 4411377.0,
      "step": 2420
    },
    {
      "entropy": 5.783400630950927,
      "epoch": 2.009113504556752,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048184347800402083,
      "loss": 5.4847,
      "mean_token_accuracy": 0.1676163673400879,
      "num_tokens": 4419781.0,
      "step": 2425
    },
    {
      "entropy": 5.719689798355103,
      "epoch": 2.013256006628003,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004817175089312551,
      "loss": 5.4593,
      "mean_token_accuracy": 0.1607228234410286,
      "num_tokens": 4428427.0,
      "step": 2430
    },
    {
      "entropy": 5.824680519104004,
      "epoch": 2.0173985086992543,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048159112288668403,
      "loss": 5.4979,
      "mean_token_accuracy": 0.15756488591432571,
      "num_tokens": 4438156.0,
      "step": 2435
    },
    {
      "entropy": 5.819461822509766,
      "epoch": 2.0215410107705054,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048146432012524215,
      "loss": 5.532,
      "mean_token_accuracy": 0.16391444951295853,
      "num_tokens": 4446342.0,
      "step": 2440
    },
    {
      "entropy": 5.761379432678223,
      "epoch": 2.0256835128417565,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048133710090270446,
      "loss": 5.5658,
      "mean_token_accuracy": 0.15113780349493028,
      "num_tokens": 4455582.0,
      "step": 2445
    },
    {
      "entropy": 5.888222980499267,
      "epoch": 2.0298260149130076,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048120946547568596,
      "loss": 5.5826,
      "mean_token_accuracy": 0.15940120220184326,
      "num_tokens": 4464518.0,
      "step": 2450
    },
    {
      "entropy": 5.927597761154175,
      "epoch": 2.0339685169842583,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048108141410164133,
      "loss": 5.7022,
      "mean_token_accuracy": 0.1438803918659687,
      "num_tokens": 4475031.0,
      "step": 2455
    },
    {
      "entropy": 5.8743188858032225,
      "epoch": 2.0381110190555094,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004809529470388642,
      "loss": 5.5708,
      "mean_token_accuracy": 0.1615961655974388,
      "num_tokens": 4484609.0,
      "step": 2460
    },
    {
      "entropy": 5.786350297927856,
      "epoch": 2.0422535211267605,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004808240645464866,
      "loss": 5.6056,
      "mean_token_accuracy": 0.15688657909631729,
      "num_tokens": 4494238.0,
      "step": 2465
    },
    {
      "entropy": 5.79974250793457,
      "epoch": 2.0463960231980116,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004806947668844787,
      "loss": 5.4662,
      "mean_token_accuracy": 0.16647584587335587,
      "num_tokens": 4503258.0,
      "step": 2470
    },
    {
      "entropy": 5.838503170013428,
      "epoch": 2.0505385252692627,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048056505431364794,
      "loss": 5.5489,
      "mean_token_accuracy": 0.15769330710172652,
      "num_tokens": 4512071.0,
      "step": 2475
    },
    {
      "entropy": 5.770729064941406,
      "epoch": 2.054681027340514,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048043492709563877,
      "loss": 5.3786,
      "mean_token_accuracy": 0.17338205873966217,
      "num_tokens": 4520219.0,
      "step": 2480
    },
    {
      "entropy": 5.736357259750366,
      "epoch": 2.0588235294117645,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048030438549293193,
      "loss": 5.6183,
      "mean_token_accuracy": 0.15568529814481735,
      "num_tokens": 4529386.0,
      "step": 2485
    },
    {
      "entropy": 5.889974308013916,
      "epoch": 2.0629660314830156,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048017342976884424,
      "loss": 5.6105,
      "mean_token_accuracy": 0.14977954104542732,
      "num_tokens": 4538860.0,
      "step": 2490
    },
    {
      "entropy": 5.706206464767456,
      "epoch": 2.0671085335542667,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048004206018752773,
      "loss": 5.5646,
      "mean_token_accuracy": 0.15797781199216843,
      "num_tokens": 4547842.0,
      "step": 2495
    },
    {
      "entropy": 5.816332674026489,
      "epoch": 2.071251035625518,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047991027701396895,
      "loss": 5.5431,
      "mean_token_accuracy": 0.15634830445051193,
      "num_tokens": 4556238.0,
      "step": 2500
    },
    {
      "epoch": 2.071251035625518,
      "eval_entropy": 5.749874321382437,
      "eval_loss": 6.056258201599121,
      "eval_mean_token_accuracy": 0.14146501896605054,
      "eval_num_tokens": 4556238.0,
      "eval_runtime": 2.7516,
      "eval_samples_per_second": 1361.734,
      "eval_steps_per_second": 170.444,
      "step": 2500
    },
    {
      "entropy": 5.786322116851807,
      "epoch": 2.075393537696769,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004797780805139892,
      "loss": 5.5554,
      "mean_token_accuracy": 0.16125964000821114,
      "num_tokens": 4565684.0,
      "step": 2505
    },
    {
      "entropy": 5.756158685684204,
      "epoch": 2.07953603976802,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004796454709542432,
      "loss": 5.5027,
      "mean_token_accuracy": 0.1608697935938835,
      "num_tokens": 4575001.0,
      "step": 2510
    },
    {
      "entropy": 5.776489973068237,
      "epoch": 2.083678541839271,
      "grad_norm": 1.25,
      "learning_rate": 0.0004795124486022189,
      "loss": 5.6243,
      "mean_token_accuracy": 0.15420307517051696,
      "num_tokens": 4584128.0,
      "step": 2515
    },
    {
      "entropy": 5.726174306869507,
      "epoch": 2.087821043910522,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047937901372623694,
      "loss": 5.4739,
      "mean_token_accuracy": 0.16476461738348008,
      "num_tokens": 4593011.0,
      "step": 2520
    },
    {
      "entropy": 5.785531711578369,
      "epoch": 2.091963545981773,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047924516659545005,
      "loss": 5.5068,
      "mean_token_accuracy": 0.1623775690793991,
      "num_tokens": 4602676.0,
      "step": 2525
    },
    {
      "entropy": 5.84798231124878,
      "epoch": 2.096106048053024,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004791109074798425,
      "loss": 5.5896,
      "mean_token_accuracy": 0.15097921043634416,
      "num_tokens": 4611915.0,
      "step": 2530
    },
    {
      "entropy": 5.7544732093811035,
      "epoch": 2.100248550124275,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004789762366502296,
      "loss": 5.5609,
      "mean_token_accuracy": 0.15663281530141832,
      "num_tokens": 4621396.0,
      "step": 2535
    },
    {
      "entropy": 5.795808458328247,
      "epoch": 2.1043910521955262,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004788411543782571,
      "loss": 5.5689,
      "mean_token_accuracy": 0.15645280182361604,
      "num_tokens": 4630533.0,
      "step": 2540
    },
    {
      "entropy": 5.74350414276123,
      "epoch": 2.108533554266777,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004787056609364008,
      "loss": 5.4951,
      "mean_token_accuracy": 0.1582939773797989,
      "num_tokens": 4639023.0,
      "step": 2545
    },
    {
      "entropy": 5.719309282302857,
      "epoch": 2.112676056338028,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004785697565979657,
      "loss": 5.5339,
      "mean_token_accuracy": 0.16151303052902222,
      "num_tokens": 4647975.0,
      "step": 2550
    },
    {
      "entropy": 5.737587833404541,
      "epoch": 2.116818558409279,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004784334416370857,
      "loss": 5.5491,
      "mean_token_accuracy": 0.15539203137159346,
      "num_tokens": 4657555.0,
      "step": 2555
    },
    {
      "entropy": 5.7617528438568115,
      "epoch": 2.1209610604805302,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004782967163287231,
      "loss": 5.4565,
      "mean_token_accuracy": 0.15736441761255265,
      "num_tokens": 4667228.0,
      "step": 2560
    },
    {
      "entropy": 5.704279279708862,
      "epoch": 2.1251035625517813,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047815958094866756,
      "loss": 5.5323,
      "mean_token_accuracy": 0.15277542173862457,
      "num_tokens": 4676723.0,
      "step": 2565
    },
    {
      "entropy": 5.795382833480835,
      "epoch": 2.1292460646230325,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004780220357735363,
      "loss": 5.5207,
      "mean_token_accuracy": 0.1547121912240982,
      "num_tokens": 4686241.0,
      "step": 2570
    },
    {
      "entropy": 5.7523492813110355,
      "epoch": 2.1333885666942836,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047788408108077294,
      "loss": 5.5718,
      "mean_token_accuracy": 0.15764677077531813,
      "num_tokens": 4695195.0,
      "step": 2575
    },
    {
      "entropy": 5.813769340515137,
      "epoch": 2.1375310687655342,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004777457171486471,
      "loss": 5.63,
      "mean_token_accuracy": 0.15644090324640275,
      "num_tokens": 4704716.0,
      "step": 2580
    },
    {
      "entropy": 5.719523763656616,
      "epoch": 2.1416735708367853,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004776069442562541,
      "loss": 5.5895,
      "mean_token_accuracy": 0.15595096796751023,
      "num_tokens": 4713720.0,
      "step": 2585
    },
    {
      "entropy": 5.820378255844116,
      "epoch": 2.1458160729080364,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004774677626835138,
      "loss": 5.5662,
      "mean_token_accuracy": 0.15348037481307983,
      "num_tokens": 4722428.0,
      "step": 2590
    },
    {
      "entropy": 5.800876331329346,
      "epoch": 2.1499585749792876,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047732817271117084,
      "loss": 5.5822,
      "mean_token_accuracy": 0.15440100207924842,
      "num_tokens": 4732395.0,
      "step": 2595
    },
    {
      "entropy": 5.759647798538208,
      "epoch": 2.1541010770505387,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004771881746207934,
      "loss": 5.5637,
      "mean_token_accuracy": 0.15513806194067,
      "num_tokens": 4741071.0,
      "step": 2600
    },
    {
      "entropy": 5.729876518249512,
      "epoch": 2.1582435791217898,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047704776869477285,
      "loss": 5.5579,
      "mean_token_accuracy": 0.15268941819667817,
      "num_tokens": 4749876.0,
      "step": 2605
    },
    {
      "entropy": 5.7958738803863525,
      "epoch": 2.1623860811930404,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047690695521632344,
      "loss": 5.5113,
      "mean_token_accuracy": 0.15810582637786866,
      "num_tokens": 4758721.0,
      "step": 2610
    },
    {
      "entropy": 5.697278213500977,
      "epoch": 2.1665285832642915,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047676573446948114,
      "loss": 5.5424,
      "mean_token_accuracy": 0.16137579083442688,
      "num_tokens": 4767162.0,
      "step": 2615
    },
    {
      "entropy": 5.794365882873535,
      "epoch": 2.1706710853355426,
      "grad_norm": 1.125,
      "learning_rate": 0.00047662410673910376,
      "loss": 5.5584,
      "mean_token_accuracy": 0.16126850992441177,
      "num_tokens": 4776397.0,
      "step": 2620
    },
    {
      "entropy": 5.6881626605987545,
      "epoch": 2.1748135874067938,
      "grad_norm": 1.25,
      "learning_rate": 0.00047648207231087,
      "loss": 5.507,
      "mean_token_accuracy": 0.1613411232829094,
      "num_tokens": 4784822.0,
      "step": 2625
    },
    {
      "entropy": 5.762795734405517,
      "epoch": 2.178956089478045,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004763396314712786,
      "loss": 5.573,
      "mean_token_accuracy": 0.16014576703310013,
      "num_tokens": 4794272.0,
      "step": 2630
    },
    {
      "entropy": 5.7245035648345945,
      "epoch": 2.183098591549296,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004761967845076484,
      "loss": 5.5137,
      "mean_token_accuracy": 0.15319861993193626,
      "num_tokens": 4804314.0,
      "step": 2635
    },
    {
      "entropy": 5.677399730682373,
      "epoch": 2.1872410936205466,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004760535317081174,
      "loss": 5.519,
      "mean_token_accuracy": 0.15785249769687654,
      "num_tokens": 4812685.0,
      "step": 2640
    },
    {
      "entropy": 5.780713081359863,
      "epoch": 2.1913835956917977,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047590987336164207,
      "loss": 5.59,
      "mean_token_accuracy": 0.15907835513353347,
      "num_tokens": 4821761.0,
      "step": 2645
    },
    {
      "entropy": 5.792177724838257,
      "epoch": 2.195526097763049,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047576580975799705,
      "loss": 5.5878,
      "mean_token_accuracy": 0.1585995614528656,
      "num_tokens": 4830515.0,
      "step": 2650
    },
    {
      "entropy": 5.790842866897583,
      "epoch": 2.1996685998343,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004756213411877743,
      "loss": 5.5382,
      "mean_token_accuracy": 0.15319958925247193,
      "num_tokens": 4840741.0,
      "step": 2655
    },
    {
      "entropy": 5.652156352996826,
      "epoch": 2.203811101905551,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047547646794238277,
      "loss": 5.4936,
      "mean_token_accuracy": 0.16105724722146988,
      "num_tokens": 4849571.0,
      "step": 2660
    },
    {
      "entropy": 5.680575323104859,
      "epoch": 2.207953603976802,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047533119031404757,
      "loss": 5.4472,
      "mean_token_accuracy": 0.16898063719272613,
      "num_tokens": 4857194.0,
      "step": 2665
    },
    {
      "entropy": 5.678529882431031,
      "epoch": 2.212096106048053,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047518550859580966,
      "loss": 5.539,
      "mean_token_accuracy": 0.15916522443294526,
      "num_tokens": 4865894.0,
      "step": 2670
    },
    {
      "entropy": 5.780280065536499,
      "epoch": 2.216238608119304,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004750394230815249,
      "loss": 5.5643,
      "mean_token_accuracy": 0.15810834765434265,
      "num_tokens": 4875131.0,
      "step": 2675
    },
    {
      "entropy": 5.654790019989013,
      "epoch": 2.220381110190555,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047489293406586375,
      "loss": 5.4224,
      "mean_token_accuracy": 0.1665143996477127,
      "num_tokens": 4884098.0,
      "step": 2680
    },
    {
      "entropy": 5.683253002166748,
      "epoch": 2.224523612261806,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004747460418443106,
      "loss": 5.4703,
      "mean_token_accuracy": 0.15905565470457078,
      "num_tokens": 4892963.0,
      "step": 2685
    },
    {
      "entropy": 5.701637554168701,
      "epoch": 2.2286661143330573,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047459874671316305,
      "loss": 5.5293,
      "mean_token_accuracy": 0.15367296785116197,
      "num_tokens": 4902097.0,
      "step": 2690
    },
    {
      "entropy": 5.647844886779785,
      "epoch": 2.2328086164043084,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047445104896953153,
      "loss": 5.4472,
      "mean_token_accuracy": 0.16798048615455627,
      "num_tokens": 4910469.0,
      "step": 2695
    },
    {
      "entropy": 5.876229000091553,
      "epoch": 2.236951118475559,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047430294891133846,
      "loss": 5.7309,
      "mean_token_accuracy": 0.1496710143983364,
      "num_tokens": 4921024.0,
      "step": 2700
    },
    {
      "entropy": 5.72487506866455,
      "epoch": 2.24109362054681,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004741544468373179,
      "loss": 5.5663,
      "mean_token_accuracy": 0.15733755230903626,
      "num_tokens": 4930297.0,
      "step": 2705
    },
    {
      "entropy": 5.729690361022949,
      "epoch": 2.2452361226180613,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047400554304701474,
      "loss": 5.53,
      "mean_token_accuracy": 0.15848677456378937,
      "num_tokens": 4938971.0,
      "step": 2710
    },
    {
      "entropy": 5.639805555343628,
      "epoch": 2.2493786246893124,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047385623784078427,
      "loss": 5.5279,
      "mean_token_accuracy": 0.15552860796451567,
      "num_tokens": 4948531.0,
      "step": 2715
    },
    {
      "entropy": 5.787559366226196,
      "epoch": 2.2535211267605635,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004737065315197913,
      "loss": 5.6031,
      "mean_token_accuracy": 0.16300928741693496,
      "num_tokens": 4957430.0,
      "step": 2720
    },
    {
      "entropy": 5.746896505355835,
      "epoch": 2.2576636288318146,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047355642438600987,
      "loss": 5.5968,
      "mean_token_accuracy": 0.15746028870344161,
      "num_tokens": 4966681.0,
      "step": 2725
    },
    {
      "entropy": 5.689626312255859,
      "epoch": 2.2618061309030653,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004734059167422225,
      "loss": 5.4557,
      "mean_token_accuracy": 0.1679423928260803,
      "num_tokens": 4975327.0,
      "step": 2730
    },
    {
      "entropy": 5.666088104248047,
      "epoch": 2.2659486329743164,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047325500889201956,
      "loss": 5.5233,
      "mean_token_accuracy": 0.16318378522992133,
      "num_tokens": 4983741.0,
      "step": 2735
    },
    {
      "entropy": 5.736676263809204,
      "epoch": 2.2700911350455675,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004731037011397986,
      "loss": 5.6099,
      "mean_token_accuracy": 0.15960921347141266,
      "num_tokens": 4993171.0,
      "step": 2740
    },
    {
      "entropy": 5.83419246673584,
      "epoch": 2.2742336371168186,
      "grad_norm": 1.125,
      "learning_rate": 0.00047295199379076403,
      "loss": 5.639,
      "mean_token_accuracy": 0.151589272916317,
      "num_tokens": 5002810.0,
      "step": 2745
    },
    {
      "entropy": 5.685939884185791,
      "epoch": 2.2783761391880697,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000472799887150926,
      "loss": 5.5288,
      "mean_token_accuracy": 0.15614852607250213,
      "num_tokens": 5011875.0,
      "step": 2750
    },
    {
      "entropy": 5.7269433498382565,
      "epoch": 2.282518641259321,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004726473815271003,
      "loss": 5.5689,
      "mean_token_accuracy": 0.16298677772283554,
      "num_tokens": 5020780.0,
      "step": 2755
    },
    {
      "entropy": 5.73256287574768,
      "epoch": 2.286661143330572,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047249447722690747,
      "loss": 5.4877,
      "mean_token_accuracy": 0.158497454226017,
      "num_tokens": 5029896.0,
      "step": 2760
    },
    {
      "entropy": 5.687757587432861,
      "epoch": 2.2908036454018226,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047234117455877215,
      "loss": 5.5288,
      "mean_token_accuracy": 0.15659505352377892,
      "num_tokens": 5038956.0,
      "step": 2765
    },
    {
      "entropy": 5.689195823669434,
      "epoch": 2.2949461474730737,
      "grad_norm": 1.125,
      "learning_rate": 0.0004721874738319225,
      "loss": 5.4933,
      "mean_token_accuracy": 0.16915134489536285,
      "num_tokens": 5047005.0,
      "step": 2770
    },
    {
      "entropy": 5.674070739746094,
      "epoch": 2.299088649544325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004720333753563898,
      "loss": 5.5645,
      "mean_token_accuracy": 0.15010802298784257,
      "num_tokens": 5056475.0,
      "step": 2775
    },
    {
      "entropy": 5.762642955780029,
      "epoch": 2.303231151615576,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047187887944300747,
      "loss": 5.5326,
      "mean_token_accuracy": 0.16023070067167283,
      "num_tokens": 5064978.0,
      "step": 2780
    },
    {
      "entropy": 5.707161235809326,
      "epoch": 2.307373653686827,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004717239864034107,
      "loss": 5.6213,
      "mean_token_accuracy": 0.15539530366659166,
      "num_tokens": 5074235.0,
      "step": 2785
    },
    {
      "entropy": 5.734616947174072,
      "epoch": 2.3115161557580777,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047156869655003554,
      "loss": 5.4672,
      "mean_token_accuracy": 0.16697436273097993,
      "num_tokens": 5082386.0,
      "step": 2790
    },
    {
      "entropy": 5.639080142974853,
      "epoch": 2.315658657829329,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047141301019611875,
      "loss": 5.4486,
      "mean_token_accuracy": 0.1636665090918541,
      "num_tokens": 5090634.0,
      "step": 2795
    },
    {
      "entropy": 5.8338916301727295,
      "epoch": 2.31980115990058,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004712569276556967,
      "loss": 5.6254,
      "mean_token_accuracy": 0.15228739976882935,
      "num_tokens": 5099003.0,
      "step": 2800
    },
    {
      "entropy": 5.685038995742798,
      "epoch": 2.323943661971831,
      "grad_norm": 1.25,
      "learning_rate": 0.00047110044924360484,
      "loss": 5.7285,
      "mean_token_accuracy": 0.1518704243004322,
      "num_tokens": 5108631.0,
      "step": 2805
    },
    {
      "entropy": 5.7691676139831545,
      "epoch": 2.328086164043082,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047094357527547725,
      "loss": 5.492,
      "mean_token_accuracy": 0.15908450484275818,
      "num_tokens": 5117309.0,
      "step": 2810
    },
    {
      "entropy": 5.748100280761719,
      "epoch": 2.332228666114333,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004707863060677459,
      "loss": 5.5936,
      "mean_token_accuracy": 0.1596307374536991,
      "num_tokens": 5126305.0,
      "step": 2815
    },
    {
      "entropy": 5.909375190734863,
      "epoch": 2.3363711681855843,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047062864193764,
      "loss": 5.6575,
      "mean_token_accuracy": 0.15526335835456848,
      "num_tokens": 5135338.0,
      "step": 2820
    },
    {
      "entropy": 5.662262630462647,
      "epoch": 2.340513670256835,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047047058320318525,
      "loss": 5.5526,
      "mean_token_accuracy": 0.16301657408475875,
      "num_tokens": 5144518.0,
      "step": 2825
    },
    {
      "entropy": 5.736520290374756,
      "epoch": 2.344656172328086,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004703121301832034,
      "loss": 5.6585,
      "mean_token_accuracy": 0.15638706237077712,
      "num_tokens": 5154102.0,
      "step": 2830
    },
    {
      "entropy": 5.77762246131897,
      "epoch": 2.348798674399337,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047015328319731153,
      "loss": 5.5953,
      "mean_token_accuracy": 0.15536246746778487,
      "num_tokens": 5163719.0,
      "step": 2835
    },
    {
      "entropy": 5.71826581954956,
      "epoch": 2.3529411764705883,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004699940425659215,
      "loss": 5.5533,
      "mean_token_accuracy": 0.1597944587469101,
      "num_tokens": 5173192.0,
      "step": 2840
    },
    {
      "entropy": 5.660325384140014,
      "epoch": 2.3570836785418394,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046983440861023877,
      "loss": 5.5111,
      "mean_token_accuracy": 0.16499586999416352,
      "num_tokens": 5182438.0,
      "step": 2845
    },
    {
      "entropy": 5.761060523986816,
      "epoch": 2.36122618061309,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046967438165226275,
      "loss": 5.6253,
      "mean_token_accuracy": 0.15501290336251258,
      "num_tokens": 5191595.0,
      "step": 2850
    },
    {
      "entropy": 5.683302545547486,
      "epoch": 2.365368682684341,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004695139620147851,
      "loss": 5.5101,
      "mean_token_accuracy": 0.16470725685358048,
      "num_tokens": 5200174.0,
      "step": 2855
    },
    {
      "entropy": 5.7557755470275875,
      "epoch": 2.3695111847555923,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046935315002138994,
      "loss": 5.6179,
      "mean_token_accuracy": 0.15405628085136414,
      "num_tokens": 5210454.0,
      "step": 2860
    },
    {
      "entropy": 5.797947883605957,
      "epoch": 2.3736536868268434,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004691919459964525,
      "loss": 5.5461,
      "mean_token_accuracy": 0.15969752371311188,
      "num_tokens": 5219243.0,
      "step": 2865
    },
    {
      "entropy": 5.599535417556763,
      "epoch": 2.3777961888980945,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046903035026513894,
      "loss": 5.4581,
      "mean_token_accuracy": 0.16127343028783797,
      "num_tokens": 5228656.0,
      "step": 2870
    },
    {
      "entropy": 5.649978446960449,
      "epoch": 2.3819386909693456,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004688683631534056,
      "loss": 5.5285,
      "mean_token_accuracy": 0.15923455357551575,
      "num_tokens": 5237807.0,
      "step": 2875
    },
    {
      "entropy": 5.794434547424316,
      "epoch": 2.3860811930405967,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004687059849879981,
      "loss": 5.629,
      "mean_token_accuracy": 0.15792540162801744,
      "num_tokens": 5247432.0,
      "step": 2880
    },
    {
      "entropy": 5.662710428237915,
      "epoch": 2.3902236951118474,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046854321609645093,
      "loss": 5.4666,
      "mean_token_accuracy": 0.15640246868133545,
      "num_tokens": 5256680.0,
      "step": 2885
    },
    {
      "entropy": 5.708359909057617,
      "epoch": 2.3943661971830985,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046838005680708677,
      "loss": 5.5622,
      "mean_token_accuracy": 0.14960763677954675,
      "num_tokens": 5266793.0,
      "step": 2890
    },
    {
      "entropy": 5.644489765167236,
      "epoch": 2.3985086992543496,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046821650744901573,
      "loss": 5.4844,
      "mean_token_accuracy": 0.16081234514713288,
      "num_tokens": 5275028.0,
      "step": 2895
    },
    {
      "entropy": 5.687662172317505,
      "epoch": 2.4026512013256007,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004680525683521347,
      "loss": 5.5543,
      "mean_token_accuracy": 0.15716781169176103,
      "num_tokens": 5284162.0,
      "step": 2900
    },
    {
      "entropy": 5.802298831939697,
      "epoch": 2.406793703396852,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004678882398471269,
      "loss": 5.6351,
      "mean_token_accuracy": 0.14897942692041397,
      "num_tokens": 5293263.0,
      "step": 2905
    },
    {
      "entropy": 5.7214367389678955,
      "epoch": 2.4109362054681025,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046772352226546066,
      "loss": 5.5221,
      "mean_token_accuracy": 0.16579788774251938,
      "num_tokens": 5301373.0,
      "step": 2910
    },
    {
      "entropy": 5.654629802703857,
      "epoch": 2.4150787075393536,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004675584159393895,
      "loss": 5.5472,
      "mean_token_accuracy": 0.15706405341625213,
      "num_tokens": 5311009.0,
      "step": 2915
    },
    {
      "entropy": 5.7292389392852785,
      "epoch": 2.4192212096106047,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046739292120195083,
      "loss": 5.553,
      "mean_token_accuracy": 0.15957032442092894,
      "num_tokens": 5320341.0,
      "step": 2920
    },
    {
      "entropy": 5.703303718566895,
      "epoch": 2.423363711681856,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004672270383869657,
      "loss": 5.525,
      "mean_token_accuracy": 0.16170006692409516,
      "num_tokens": 5329250.0,
      "step": 2925
    },
    {
      "entropy": 5.660790014266968,
      "epoch": 2.427506213753107,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004670607678290379,
      "loss": 5.486,
      "mean_token_accuracy": 0.16461261510848998,
      "num_tokens": 5336515.0,
      "step": 2930
    },
    {
      "entropy": 5.666818332672119,
      "epoch": 2.431648715824358,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004668941098635532,
      "loss": 5.5664,
      "mean_token_accuracy": 0.16323853135108948,
      "num_tokens": 5346043.0,
      "step": 2935
    },
    {
      "entropy": 5.721542930603027,
      "epoch": 2.435791217895609,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046672706482667897,
      "loss": 5.5287,
      "mean_token_accuracy": 0.1586231455206871,
      "num_tokens": 5355124.0,
      "step": 2940
    },
    {
      "entropy": 5.67028284072876,
      "epoch": 2.43993371996686,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004665596330553633,
      "loss": 5.5369,
      "mean_token_accuracy": 0.15888493806123732,
      "num_tokens": 5364556.0,
      "step": 2945
    },
    {
      "entropy": 5.712566614151001,
      "epoch": 2.444076222038111,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046639181488733445,
      "loss": 5.5755,
      "mean_token_accuracy": 0.15751136243343353,
      "num_tokens": 5373778.0,
      "step": 2950
    },
    {
      "entropy": 5.770811271667481,
      "epoch": 2.448218724109362,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004662236106610999,
      "loss": 5.5566,
      "mean_token_accuracy": 0.1569676160812378,
      "num_tokens": 5382633.0,
      "step": 2955
    },
    {
      "entropy": 5.576747179031372,
      "epoch": 2.452361226180613,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046605502071594604,
      "loss": 5.4624,
      "mean_token_accuracy": 0.16361840963363647,
      "num_tokens": 5391442.0,
      "step": 2960
    },
    {
      "entropy": 5.590614557266235,
      "epoch": 2.4565037282518642,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004658860453919371,
      "loss": 5.4619,
      "mean_token_accuracy": 0.16674424558877946,
      "num_tokens": 5400504.0,
      "step": 2965
    },
    {
      "entropy": 5.691436147689819,
      "epoch": 2.460646230323115,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004657166850299149,
      "loss": 5.4869,
      "mean_token_accuracy": 0.16380043774843217,
      "num_tokens": 5408939.0,
      "step": 2970
    },
    {
      "entropy": 5.66795802116394,
      "epoch": 2.464788732394366,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004655469399714978,
      "loss": 5.5004,
      "mean_token_accuracy": 0.16163172274827958,
      "num_tokens": 5418934.0,
      "step": 2975
    },
    {
      "entropy": 5.624106788635254,
      "epoch": 2.468931234465617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004653768105590799,
      "loss": 5.5379,
      "mean_token_accuracy": 0.16006896495819092,
      "num_tokens": 5427894.0,
      "step": 2980
    },
    {
      "entropy": 5.696897411346436,
      "epoch": 2.4730737365368682,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046520629713583115,
      "loss": 5.5192,
      "mean_token_accuracy": 0.16015419960021973,
      "num_tokens": 5436830.0,
      "step": 2985
    },
    {
      "entropy": 5.637984657287598,
      "epoch": 2.4772162386081193,
      "grad_norm": 1.125,
      "learning_rate": 0.00046503540004569555,
      "loss": 5.5262,
      "mean_token_accuracy": 0.16234136521816253,
      "num_tokens": 5445504.0,
      "step": 2990
    },
    {
      "entropy": 5.732030582427979,
      "epoch": 2.4813587406793705,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046486411963339125,
      "loss": 5.5659,
      "mean_token_accuracy": 0.15476318076252937,
      "num_tokens": 5455378.0,
      "step": 2995
    },
    {
      "entropy": 5.726173019409179,
      "epoch": 2.4855012427506216,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046469245624440955,
      "loss": 5.6408,
      "mean_token_accuracy": 0.1584235519170761,
      "num_tokens": 5465334.0,
      "step": 3000
    },
    {
      "epoch": 2.4855012427506216,
      "eval_entropy": 5.639674754030923,
      "eval_loss": 5.9833292961120605,
      "eval_mean_token_accuracy": 0.1469943482738568,
      "eval_num_tokens": 5465334.0,
      "eval_runtime": 2.967,
      "eval_samples_per_second": 1262.891,
      "eval_steps_per_second": 158.072,
      "step": 3000
    },
    {
      "entropy": 5.791637706756592,
      "epoch": 2.4896437448218722,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046452041022501423,
      "loss": 5.6258,
      "mean_token_accuracy": 0.15460191220045089,
      "num_tokens": 5475701.0,
      "step": 3005
    },
    {
      "entropy": 5.727213287353516,
      "epoch": 2.4937862468931233,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004643479819222411,
      "loss": 5.5093,
      "mean_token_accuracy": 0.16252206414937972,
      "num_tokens": 5485244.0,
      "step": 3010
    },
    {
      "entropy": 5.609255790710449,
      "epoch": 2.4979287489643744,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046417517168389664,
      "loss": 5.4792,
      "mean_token_accuracy": 0.16647332459688186,
      "num_tokens": 5493587.0,
      "step": 3015
    },
    {
      "entropy": 5.652401065826416,
      "epoch": 2.5020712510356256,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004640019798585581,
      "loss": 5.6175,
      "mean_token_accuracy": 0.1556309536099434,
      "num_tokens": 5503117.0,
      "step": 3020
    },
    {
      "entropy": 5.7778074741363525,
      "epoch": 2.5062137531068767,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046382840679557233,
      "loss": 5.5637,
      "mean_token_accuracy": 0.16106179654598235,
      "num_tokens": 5512742.0,
      "step": 3025
    },
    {
      "entropy": 5.636971282958984,
      "epoch": 2.5103562551781273,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046365445284505516,
      "loss": 5.5246,
      "mean_token_accuracy": 0.15782141983509063,
      "num_tokens": 5521862.0,
      "step": 3030
    },
    {
      "entropy": 5.651063251495361,
      "epoch": 2.5144987572493784,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004634801183578907,
      "loss": 5.5415,
      "mean_token_accuracy": 0.1628969594836235,
      "num_tokens": 5531244.0,
      "step": 3035
    },
    {
      "entropy": 5.7396568775177,
      "epoch": 2.5186412593206295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046330540368573063,
      "loss": 5.5713,
      "mean_token_accuracy": 0.15206096023321153,
      "num_tokens": 5540786.0,
      "step": 3040
    },
    {
      "entropy": 5.720060348510742,
      "epoch": 2.5227837613918807,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046313030918099365,
      "loss": 5.5721,
      "mean_token_accuracy": 0.1614146649837494,
      "num_tokens": 5549961.0,
      "step": 3045
    },
    {
      "entropy": 5.705565547943115,
      "epoch": 2.5269262634631318,
      "grad_norm": 1.125,
      "learning_rate": 0.00046295483519686437,
      "loss": 5.5061,
      "mean_token_accuracy": 0.16351179033517838,
      "num_tokens": 5558941.0,
      "step": 3050
    },
    {
      "entropy": 5.665171957015991,
      "epoch": 2.531068765534383,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046277898208729306,
      "loss": 5.5118,
      "mean_token_accuracy": 0.16655909419059753,
      "num_tokens": 5567630.0,
      "step": 3055
    },
    {
      "entropy": 5.6335536479949955,
      "epoch": 2.535211267605634,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004626027502069947,
      "loss": 5.4747,
      "mean_token_accuracy": 0.16044125258922576,
      "num_tokens": 5576430.0,
      "step": 3060
    },
    {
      "entropy": 5.707994222640991,
      "epoch": 2.539353769676885,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046242613991144814,
      "loss": 5.5872,
      "mean_token_accuracy": 0.16125080436468125,
      "num_tokens": 5584785.0,
      "step": 3065
    },
    {
      "entropy": 5.570116853713989,
      "epoch": 2.5434962717481358,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046224915155689586,
      "loss": 5.5074,
      "mean_token_accuracy": 0.1572348579764366,
      "num_tokens": 5594583.0,
      "step": 3070
    },
    {
      "entropy": 5.726114130020141,
      "epoch": 2.547638773819387,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046207178550034244,
      "loss": 5.4746,
      "mean_token_accuracy": 0.162801356613636,
      "num_tokens": 5603498.0,
      "step": 3075
    },
    {
      "entropy": 5.751227521896363,
      "epoch": 2.551781275890638,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004618940420995547,
      "loss": 5.6074,
      "mean_token_accuracy": 0.15416808128356935,
      "num_tokens": 5613101.0,
      "step": 3080
    },
    {
      "entropy": 5.66825041770935,
      "epoch": 2.555923777961889,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004617159217130606,
      "loss": 5.585,
      "mean_token_accuracy": 0.15699986293911933,
      "num_tokens": 5621821.0,
      "step": 3085
    },
    {
      "entropy": 5.771618795394898,
      "epoch": 2.5600662800331397,
      "grad_norm": 1.125,
      "learning_rate": 0.0004615374247001483,
      "loss": 5.5981,
      "mean_token_accuracy": 0.15983579754829408,
      "num_tokens": 5631006.0,
      "step": 3090
    },
    {
      "entropy": 5.717100048065186,
      "epoch": 2.564208782104391,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004613585514208659,
      "loss": 5.5649,
      "mean_token_accuracy": 0.15318279713392258,
      "num_tokens": 5640159.0,
      "step": 3095
    },
    {
      "entropy": 5.654997730255127,
      "epoch": 2.568351284175642,
      "grad_norm": 1.125,
      "learning_rate": 0.00046117930223602016,
      "loss": 5.4329,
      "mean_token_accuracy": 0.16797446459531784,
      "num_tokens": 5648369.0,
      "step": 3100
    },
    {
      "entropy": 5.580140495300293,
      "epoch": 2.572493786246893,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004609996775071765,
      "loss": 5.4869,
      "mean_token_accuracy": 0.15920215398073195,
      "num_tokens": 5657669.0,
      "step": 3105
    },
    {
      "entropy": 5.672556781768799,
      "epoch": 2.576636288318144,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004608196775966575,
      "loss": 5.5745,
      "mean_token_accuracy": 0.16189952045679093,
      "num_tokens": 5666601.0,
      "step": 3110
    },
    {
      "entropy": 5.708433198928833,
      "epoch": 2.5807787903893953,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046063930286754267,
      "loss": 5.4922,
      "mean_token_accuracy": 0.16438818722963333,
      "num_tokens": 5675243.0,
      "step": 3115
    },
    {
      "entropy": 5.607330846786499,
      "epoch": 2.5849212924606464,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046045855368366756,
      "loss": 5.5122,
      "mean_token_accuracy": 0.15873268097639084,
      "num_tokens": 5683646.0,
      "step": 3120
    },
    {
      "entropy": 5.735720300674439,
      "epoch": 2.5890637945318975,
      "grad_norm": 1.265625,
      "learning_rate": 0.000460277430409623,
      "loss": 5.5464,
      "mean_token_accuracy": 0.1679003655910492,
      "num_tokens": 5693069.0,
      "step": 3125
    },
    {
      "entropy": 5.556051111221313,
      "epoch": 2.593206296603148,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046009593341075444,
      "loss": 5.4092,
      "mean_token_accuracy": 0.16623837649822235,
      "num_tokens": 5701527.0,
      "step": 3130
    },
    {
      "entropy": 5.590375089645386,
      "epoch": 2.5973487986743993,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004599140630531612,
      "loss": 5.5604,
      "mean_token_accuracy": 0.16305261999368667,
      "num_tokens": 5710379.0,
      "step": 3135
    },
    {
      "entropy": 5.747953128814697,
      "epoch": 2.6014913007456504,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004597318197036956,
      "loss": 5.48,
      "mean_token_accuracy": 0.1645922839641571,
      "num_tokens": 5719217.0,
      "step": 3140
    },
    {
      "entropy": 5.658558940887451,
      "epoch": 2.6056338028169015,
      "grad_norm": 1.25,
      "learning_rate": 0.0004595492037299624,
      "loss": 5.5902,
      "mean_token_accuracy": 0.15660670399665833,
      "num_tokens": 5728331.0,
      "step": 3145
    },
    {
      "entropy": 5.670481443405151,
      "epoch": 2.6097763048881526,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045936621550031806,
      "loss": 5.5444,
      "mean_token_accuracy": 0.16000198125839232,
      "num_tokens": 5737696.0,
      "step": 3150
    },
    {
      "entropy": 5.761334562301636,
      "epoch": 2.6139188069594033,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045918285538386973,
      "loss": 5.616,
      "mean_token_accuracy": 0.16404765099287033,
      "num_tokens": 5746191.0,
      "step": 3155
    },
    {
      "entropy": 5.724868297576904,
      "epoch": 2.6180613090306544,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045899912375047485,
      "loss": 5.6065,
      "mean_token_accuracy": 0.16577061861753464,
      "num_tokens": 5754824.0,
      "step": 3160
    },
    {
      "entropy": 5.630853986740112,
      "epoch": 2.6222038111019055,
      "grad_norm": 1.125,
      "learning_rate": 0.00045881502097074023,
      "loss": 5.5528,
      "mean_token_accuracy": 0.1559559613466263,
      "num_tokens": 5764863.0,
      "step": 3165
    },
    {
      "entropy": 5.6735029220581055,
      "epoch": 2.6263463131731566,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004586305474160212,
      "loss": 5.5744,
      "mean_token_accuracy": 0.15252353399991989,
      "num_tokens": 5774273.0,
      "step": 3170
    },
    {
      "entropy": 5.654740190505981,
      "epoch": 2.6304888152444077,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045844570345842116,
      "loss": 5.5444,
      "mean_token_accuracy": 0.1586697816848755,
      "num_tokens": 5784085.0,
      "step": 3175
    },
    {
      "entropy": 5.673130083084106,
      "epoch": 2.634631317315659,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004582604894707906,
      "loss": 5.5743,
      "mean_token_accuracy": 0.1666999340057373,
      "num_tokens": 5793921.0,
      "step": 3180
    },
    {
      "entropy": 5.719259262084961,
      "epoch": 2.63877381938691,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045807490582672634,
      "loss": 5.5366,
      "mean_token_accuracy": 0.1527248203754425,
      "num_tokens": 5803902.0,
      "step": 3185
    },
    {
      "entropy": 5.607631492614746,
      "epoch": 2.6429163214581606,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045788895290057083,
      "loss": 5.4507,
      "mean_token_accuracy": 0.16747543811798096,
      "num_tokens": 5812089.0,
      "step": 3190
    },
    {
      "entropy": 5.708329534530639,
      "epoch": 2.6470588235294117,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045770263106741137,
      "loss": 5.5981,
      "mean_token_accuracy": 0.15919173806905745,
      "num_tokens": 5821072.0,
      "step": 3195
    },
    {
      "entropy": 5.637457323074341,
      "epoch": 2.651201325600663,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004575159407030797,
      "loss": 5.4373,
      "mean_token_accuracy": 0.16173361092805863,
      "num_tokens": 5829725.0,
      "step": 3200
    },
    {
      "entropy": 5.542225885391235,
      "epoch": 2.655343827671914,
      "grad_norm": 1.125,
      "learning_rate": 0.0004573288821841505,
      "loss": 5.5196,
      "mean_token_accuracy": 0.16196130812168122,
      "num_tokens": 5838153.0,
      "step": 3205
    },
    {
      "entropy": 5.6365491390228275,
      "epoch": 2.659486329743165,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004571414558879413,
      "loss": 5.4599,
      "mean_token_accuracy": 0.15795936137437822,
      "num_tokens": 5847270.0,
      "step": 3210
    },
    {
      "entropy": 5.676420450210571,
      "epoch": 2.6636288318144157,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004569536621925114,
      "loss": 5.5895,
      "mean_token_accuracy": 0.15637888610363007,
      "num_tokens": 5855954.0,
      "step": 3215
    },
    {
      "entropy": 5.689839649200439,
      "epoch": 2.667771333885667,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004567655014766612,
      "loss": 5.4853,
      "mean_token_accuracy": 0.1627836525440216,
      "num_tokens": 5864435.0,
      "step": 3220
    },
    {
      "entropy": 5.640894746780395,
      "epoch": 2.671913835956918,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004565769741199314,
      "loss": 5.5571,
      "mean_token_accuracy": 0.15790077894926072,
      "num_tokens": 5873964.0,
      "step": 3225
    },
    {
      "entropy": 5.725470733642578,
      "epoch": 2.676056338028169,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045638808050260237,
      "loss": 5.5074,
      "mean_token_accuracy": 0.16316789835691453,
      "num_tokens": 5882738.0,
      "step": 3230
    },
    {
      "entropy": 5.639863920211792,
      "epoch": 2.68019884009942,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045619882100569317,
      "loss": 5.6406,
      "mean_token_accuracy": 0.15904040038585662,
      "num_tokens": 5891749.0,
      "step": 3235
    },
    {
      "entropy": 5.716983032226563,
      "epoch": 2.684341342170671,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004560091960109608,
      "loss": 5.5776,
      "mean_token_accuracy": 0.15503172799944878,
      "num_tokens": 5900551.0,
      "step": 3240
    },
    {
      "entropy": 5.711254215240478,
      "epoch": 2.6884838442419223,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004558192059008996,
      "loss": 5.595,
      "mean_token_accuracy": 0.154455304145813,
      "num_tokens": 5909254.0,
      "step": 3245
    },
    {
      "entropy": 5.6522393226623535,
      "epoch": 2.692626346313173,
      "grad_norm": 1.125,
      "learning_rate": 0.0004556288510587406,
      "loss": 5.5395,
      "mean_token_accuracy": 0.15719025135040282,
      "num_tokens": 5918357.0,
      "step": 3250
    },
    {
      "entropy": 5.7096232891082765,
      "epoch": 2.696768848384424,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045543813186845004,
      "loss": 5.5289,
      "mean_token_accuracy": 0.15661482959985734,
      "num_tokens": 5927742.0,
      "step": 3255
    },
    {
      "entropy": 5.649437093734742,
      "epoch": 2.700911350455675,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004552470487147294,
      "loss": 5.5935,
      "mean_token_accuracy": 0.15934477895498275,
      "num_tokens": 5937025.0,
      "step": 3260
    },
    {
      "entropy": 5.805260419845581,
      "epoch": 2.7050538525269263,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045505560198301443,
      "loss": 5.6019,
      "mean_token_accuracy": 0.15056679993867875,
      "num_tokens": 5946176.0,
      "step": 3265
    },
    {
      "entropy": 5.671889066696167,
      "epoch": 2.7091963545981774,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045486379205947396,
      "loss": 5.5669,
      "mean_token_accuracy": 0.15695112347602844,
      "num_tokens": 5955550.0,
      "step": 3270
    },
    {
      "entropy": 5.621965122222901,
      "epoch": 2.713338856669428,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004546716193310098,
      "loss": 5.4955,
      "mean_token_accuracy": 0.15768464505672455,
      "num_tokens": 5964467.0,
      "step": 3275
    },
    {
      "entropy": 5.697195816040039,
      "epoch": 2.717481358740679,
      "grad_norm": 1.078125,
      "learning_rate": 0.000454479084185255,
      "loss": 5.5231,
      "mean_token_accuracy": 0.15613993704319,
      "num_tokens": 5973906.0,
      "step": 3280
    },
    {
      "entropy": 5.676262474060058,
      "epoch": 2.7216238608119303,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004542861870105743,
      "loss": 5.5458,
      "mean_token_accuracy": 0.16207694113254548,
      "num_tokens": 5983701.0,
      "step": 3285
    },
    {
      "entropy": 5.698145246505737,
      "epoch": 2.7257663628831814,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045409292819606224,
      "loss": 5.5323,
      "mean_token_accuracy": 0.1577382504940033,
      "num_tokens": 5993367.0,
      "step": 3290
    },
    {
      "entropy": 5.663068532943726,
      "epoch": 2.7299088649544325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045389930813154297,
      "loss": 5.4883,
      "mean_token_accuracy": 0.16210660040378572,
      "num_tokens": 6003752.0,
      "step": 3295
    },
    {
      "entropy": 5.766804790496826,
      "epoch": 2.7340513670256836,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004537053272075693,
      "loss": 5.5558,
      "mean_token_accuracy": 0.16685177832841874,
      "num_tokens": 6012486.0,
      "step": 3300
    },
    {
      "entropy": 5.5676192283630375,
      "epoch": 2.7381938690969347,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045351098581542204,
      "loss": 5.5081,
      "mean_token_accuracy": 0.16729753464460373,
      "num_tokens": 6020742.0,
      "step": 3305
    },
    {
      "entropy": 5.710718393325806,
      "epoch": 2.742336371168186,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004533162843471089,
      "loss": 5.5413,
      "mean_token_accuracy": 0.1567221537232399,
      "num_tokens": 6030248.0,
      "step": 3310
    },
    {
      "entropy": 5.695099592208862,
      "epoch": 2.7464788732394365,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004531212231953641,
      "loss": 5.5763,
      "mean_token_accuracy": 0.1570330873131752,
      "num_tokens": 6039293.0,
      "step": 3315
    },
    {
      "entropy": 5.642256498336792,
      "epoch": 2.7506213753106876,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004529258027536473,
      "loss": 5.4417,
      "mean_token_accuracy": 0.16368594020605087,
      "num_tokens": 6048081.0,
      "step": 3320
    },
    {
      "entropy": 5.669494771957398,
      "epoch": 2.7547638773819387,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004527300234161428,
      "loss": 5.5984,
      "mean_token_accuracy": 0.16303794980049133,
      "num_tokens": 6057549.0,
      "step": 3325
    },
    {
      "entropy": 5.64669942855835,
      "epoch": 2.75890637945319,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004525338855777591,
      "loss": 5.6012,
      "mean_token_accuracy": 0.15381733626127242,
      "num_tokens": 6067372.0,
      "step": 3330
    },
    {
      "entropy": 5.687849664688111,
      "epoch": 2.7630488815244405,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045233738963412746,
      "loss": 5.457,
      "mean_token_accuracy": 0.16840497851371766,
      "num_tokens": 6075532.0,
      "step": 3335
    },
    {
      "entropy": 5.5901408195495605,
      "epoch": 2.7671913835956916,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045214053598160173,
      "loss": 5.4745,
      "mean_token_accuracy": 0.16109947264194488,
      "num_tokens": 6084389.0,
      "step": 3340
    },
    {
      "entropy": 5.622766542434692,
      "epoch": 2.7713338856669427,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045194332501725725,
      "loss": 5.4957,
      "mean_token_accuracy": 0.16854244321584702,
      "num_tokens": 6093536.0,
      "step": 3345
    },
    {
      "entropy": 5.716071701049804,
      "epoch": 2.775476387738194,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045174575713889014,
      "loss": 5.5674,
      "mean_token_accuracy": 0.16249414160847664,
      "num_tokens": 6102677.0,
      "step": 3350
    },
    {
      "entropy": 5.686060762405395,
      "epoch": 2.779618889809445,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004515478327450164,
      "loss": 5.6364,
      "mean_token_accuracy": 0.15851696133613585,
      "num_tokens": 6112430.0,
      "step": 3355
    },
    {
      "entropy": 5.725155210494995,
      "epoch": 2.783761391880696,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045134955223487093,
      "loss": 5.5313,
      "mean_token_accuracy": 0.16037144213914872,
      "num_tokens": 6121408.0,
      "step": 3360
    },
    {
      "entropy": 5.6291951656341555,
      "epoch": 2.787903893951947,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045115091600840757,
      "loss": 5.5594,
      "mean_token_accuracy": 0.1527995854616165,
      "num_tokens": 6131339.0,
      "step": 3365
    },
    {
      "entropy": 5.695040321350097,
      "epoch": 2.7920463960231983,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045095192446629697,
      "loss": 5.5165,
      "mean_token_accuracy": 0.16150978207588196,
      "num_tokens": 6140590.0,
      "step": 3370
    },
    {
      "entropy": 5.7630918502807615,
      "epoch": 2.796188898094449,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000450752578009927,
      "loss": 5.6064,
      "mean_token_accuracy": 0.16191209107637405,
      "num_tokens": 6149958.0,
      "step": 3375
    },
    {
      "entropy": 5.5992176055908205,
      "epoch": 2.8003314001657,
      "grad_norm": 1.125,
      "learning_rate": 0.00045055287704140115,
      "loss": 5.4601,
      "mean_token_accuracy": 0.16068203300237655,
      "num_tokens": 6159411.0,
      "step": 3380
    },
    {
      "entropy": 5.603084850311279,
      "epoch": 2.804473902236951,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004503528219635381,
      "loss": 5.5476,
      "mean_token_accuracy": 0.15345613062381744,
      "num_tokens": 6169264.0,
      "step": 3385
    },
    {
      "entropy": 5.713048839569092,
      "epoch": 2.8086164043082023,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004501524131798709,
      "loss": 5.5272,
      "mean_token_accuracy": 0.15406270325183868,
      "num_tokens": 6178374.0,
      "step": 3390
    },
    {
      "entropy": 5.732380008697509,
      "epoch": 2.812758906379453,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000449951651094646,
      "loss": 5.5217,
      "mean_token_accuracy": 0.1573767378926277,
      "num_tokens": 6186918.0,
      "step": 3395
    },
    {
      "entropy": 5.575137805938721,
      "epoch": 2.816901408450704,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044975053611282217,
      "loss": 5.5749,
      "mean_token_accuracy": 0.1632239282131195,
      "num_tokens": 6196400.0,
      "step": 3400
    },
    {
      "entropy": 5.593220329284668,
      "epoch": 2.821043910521955,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004495490686400706,
      "loss": 5.4258,
      "mean_token_accuracy": 0.16297936737537383,
      "num_tokens": 6204849.0,
      "step": 3405
    },
    {
      "entropy": 5.606288814544678,
      "epoch": 2.8251864125932062,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044934724908277313,
      "loss": 5.5297,
      "mean_token_accuracy": 0.160382042825222,
      "num_tokens": 6214121.0,
      "step": 3410
    },
    {
      "entropy": 5.674342155456543,
      "epoch": 2.8293289146644574,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044914507784802184,
      "loss": 5.5168,
      "mean_token_accuracy": 0.16667052805423738,
      "num_tokens": 6223163.0,
      "step": 3415
    },
    {
      "entropy": 5.644378805160523,
      "epoch": 2.8334714167357085,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004489425553436182,
      "loss": 5.5595,
      "mean_token_accuracy": 0.15832413882017135,
      "num_tokens": 6231581.0,
      "step": 3420
    },
    {
      "entropy": 5.681396913528443,
      "epoch": 2.8376139188069596,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044873968197807213,
      "loss": 5.6319,
      "mean_token_accuracy": 0.15717055946588515,
      "num_tokens": 6240015.0,
      "step": 3425
    },
    {
      "entropy": 5.746704483032227,
      "epoch": 2.8417564208782107,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044853645816060156,
      "loss": 5.618,
      "mean_token_accuracy": 0.15753672495484353,
      "num_tokens": 6249083.0,
      "step": 3430
    },
    {
      "entropy": 5.720417594909668,
      "epoch": 2.8458989229494613,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044833288430113104,
      "loss": 5.5847,
      "mean_token_accuracy": 0.15910181850194932,
      "num_tokens": 6258798.0,
      "step": 3435
    },
    {
      "entropy": 5.616397666931152,
      "epoch": 2.8500414250207124,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044812896081029136,
      "loss": 5.4288,
      "mean_token_accuracy": 0.17185353338718415,
      "num_tokens": 6267819.0,
      "step": 3440
    },
    {
      "entropy": 5.6485264778137205,
      "epoch": 2.8541839270919636,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004479246880994183,
      "loss": 5.5208,
      "mean_token_accuracy": 0.16037193909287453,
      "num_tokens": 6276915.0,
      "step": 3445
    },
    {
      "entropy": 5.701795244216919,
      "epoch": 2.8583264291632147,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004477200665805525,
      "loss": 5.5497,
      "mean_token_accuracy": 0.16226860135793686,
      "num_tokens": 6285880.0,
      "step": 3450
    },
    {
      "entropy": 5.694223499298095,
      "epoch": 2.8624689312344658,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044751509666643777,
      "loss": 5.6782,
      "mean_token_accuracy": 0.15221977159380912,
      "num_tokens": 6295408.0,
      "step": 3455
    },
    {
      "entropy": 5.652013063430786,
      "epoch": 2.8666114333057164,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044730977877052084,
      "loss": 5.4072,
      "mean_token_accuracy": 0.16112574934959412,
      "num_tokens": 6304110.0,
      "step": 3460
    },
    {
      "entropy": 5.673219871520996,
      "epoch": 2.8707539353769675,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044710411330695045,
      "loss": 5.5651,
      "mean_token_accuracy": 0.15716703981161118,
      "num_tokens": 6312577.0,
      "step": 3465
    },
    {
      "entropy": 5.815633296966553,
      "epoch": 2.8748964374482187,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044689810069057615,
      "loss": 5.6407,
      "mean_token_accuracy": 0.15225453972816466,
      "num_tokens": 6321766.0,
      "step": 3470
    },
    {
      "entropy": 5.69540114402771,
      "epoch": 2.8790389395194698,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004466917413369479,
      "loss": 5.5972,
      "mean_token_accuracy": 0.15052569210529326,
      "num_tokens": 6330327.0,
      "step": 3475
    },
    {
      "entropy": 5.617451906204224,
      "epoch": 2.883181441590721,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004464850356623152,
      "loss": 5.5372,
      "mean_token_accuracy": 0.161219622194767,
      "num_tokens": 6339596.0,
      "step": 3480
    },
    {
      "entropy": 5.652078914642334,
      "epoch": 2.887323943661972,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044627798408362597,
      "loss": 5.4389,
      "mean_token_accuracy": 0.16472167521715164,
      "num_tokens": 6348080.0,
      "step": 3485
    },
    {
      "entropy": 5.672922134399414,
      "epoch": 2.891466445733223,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004460705870185258,
      "loss": 5.5601,
      "mean_token_accuracy": 0.15549688041210175,
      "num_tokens": 6357725.0,
      "step": 3490
    },
    {
      "entropy": 5.644254064559936,
      "epoch": 2.8956089478044738,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004458628448853572,
      "loss": 5.5009,
      "mean_token_accuracy": 0.16381119638681413,
      "num_tokens": 6366821.0,
      "step": 3495
    },
    {
      "entropy": 5.726661252975464,
      "epoch": 2.899751449875725,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004456547581031589,
      "loss": 5.579,
      "mean_token_accuracy": 0.15758128613233566,
      "num_tokens": 6375641.0,
      "step": 3500
    },
    {
      "epoch": 2.899751449875725,
      "eval_entropy": 5.521543442821706,
      "eval_loss": 5.915812969207764,
      "eval_mean_token_accuracy": 0.150046447971101,
      "eval_num_tokens": 6375641.0,
      "eval_runtime": 2.7561,
      "eval_samples_per_second": 1359.533,
      "eval_steps_per_second": 170.168,
      "step": 3500
    },
    {
      "entropy": 5.617849683761596,
      "epoch": 2.903893951946976,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004454463270916646,
      "loss": 5.4478,
      "mean_token_accuracy": 0.16418685615062714,
      "num_tokens": 6384596.0,
      "step": 3505
    },
    {
      "entropy": 5.794553995132446,
      "epoch": 2.908036454018227,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004452375522713023,
      "loss": 5.777,
      "mean_token_accuracy": 0.14264359697699547,
      "num_tokens": 6395048.0,
      "step": 3510
    },
    {
      "entropy": 5.6611549854278564,
      "epoch": 2.912178956089478,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004450284340631938,
      "loss": 5.6103,
      "mean_token_accuracy": 0.15344774425029756,
      "num_tokens": 6404828.0,
      "step": 3515
    },
    {
      "entropy": 5.663045120239258,
      "epoch": 2.916321458160729,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044481897288915334,
      "loss": 5.4779,
      "mean_token_accuracy": 0.15861677676439284,
      "num_tokens": 6414670.0,
      "step": 3520
    },
    {
      "entropy": 5.5174237251281735,
      "epoch": 2.92046396023198,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044460916917168697,
      "loss": 5.4254,
      "mean_token_accuracy": 0.16657451838254927,
      "num_tokens": 6423446.0,
      "step": 3525
    },
    {
      "entropy": 5.599793004989624,
      "epoch": 2.924606462303231,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004443990233339918,
      "loss": 5.4952,
      "mean_token_accuracy": 0.1578425571322441,
      "num_tokens": 6433412.0,
      "step": 3530
    },
    {
      "entropy": 5.707996892929077,
      "epoch": 2.928748964374482,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004441885357999548,
      "loss": 5.4906,
      "mean_token_accuracy": 0.1628717914223671,
      "num_tokens": 6443520.0,
      "step": 3535
    },
    {
      "entropy": 5.676200866699219,
      "epoch": 2.9328914664457333,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004439777069941524,
      "loss": 5.5963,
      "mean_token_accuracy": 0.15572925359010698,
      "num_tokens": 6452933.0,
      "step": 3540
    },
    {
      "entropy": 5.624213981628418,
      "epoch": 2.9370339685169844,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044376653734184935,
      "loss": 5.4056,
      "mean_token_accuracy": 0.16887385100126268,
      "num_tokens": 6462113.0,
      "step": 3545
    },
    {
      "entropy": 5.549249124526978,
      "epoch": 2.9411764705882355,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004435550272689979,
      "loss": 5.4574,
      "mean_token_accuracy": 0.16682465970516205,
      "num_tokens": 6471411.0,
      "step": 3550
    },
    {
      "entropy": 5.7074384689331055,
      "epoch": 2.945318972659486,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044334317720223706,
      "loss": 5.5627,
      "mean_token_accuracy": 0.15771554186940193,
      "num_tokens": 6480682.0,
      "step": 3555
    },
    {
      "entropy": 5.644071817398071,
      "epoch": 2.9494614747307373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044313098756889154,
      "loss": 5.6091,
      "mean_token_accuracy": 0.15549242049455642,
      "num_tokens": 6490614.0,
      "step": 3560
    },
    {
      "entropy": 5.680739021301269,
      "epoch": 2.9536039768019884,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004429184587969708,
      "loss": 5.4566,
      "mean_token_accuracy": 0.1625344857573509,
      "num_tokens": 6499079.0,
      "step": 3565
    },
    {
      "entropy": 5.67983226776123,
      "epoch": 2.9577464788732395,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004427055913151689,
      "loss": 5.573,
      "mean_token_accuracy": 0.14837478697299958,
      "num_tokens": 6508556.0,
      "step": 3570
    },
    {
      "entropy": 5.678392362594605,
      "epoch": 2.9618889809444906,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004424923855528626,
      "loss": 5.59,
      "mean_token_accuracy": 0.155683271586895,
      "num_tokens": 6517934.0,
      "step": 3575
    },
    {
      "entropy": 5.643442249298095,
      "epoch": 2.9660314830157413,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004422788419401114,
      "loss": 5.4576,
      "mean_token_accuracy": 0.16748119741678238,
      "num_tokens": 6526995.0,
      "step": 3580
    },
    {
      "entropy": 5.613794136047363,
      "epoch": 2.9701739850869924,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044206496090765595,
      "loss": 5.5288,
      "mean_token_accuracy": 0.15781812220811844,
      "num_tokens": 6535264.0,
      "step": 3585
    },
    {
      "entropy": 5.682992553710937,
      "epoch": 2.9743164871582435,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044185074288691776,
      "loss": 5.4229,
      "mean_token_accuracy": 0.17119230926036835,
      "num_tokens": 6544498.0,
      "step": 3590
    },
    {
      "entropy": 5.585231161117553,
      "epoch": 2.9784589892294946,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044163618830999794,
      "loss": 5.54,
      "mean_token_accuracy": 0.15998248010873795,
      "num_tokens": 6554761.0,
      "step": 3595
    },
    {
      "entropy": 5.627832126617432,
      "epoch": 2.9826014913007457,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044142129760967643,
      "loss": 5.596,
      "mean_token_accuracy": 0.1564616858959198,
      "num_tokens": 6564039.0,
      "step": 3600
    },
    {
      "entropy": 5.723354148864746,
      "epoch": 2.986743993371997,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004412060712194115,
      "loss": 5.5456,
      "mean_token_accuracy": 0.1615665450692177,
      "num_tokens": 6572559.0,
      "step": 3605
    },
    {
      "entropy": 5.61319408416748,
      "epoch": 2.990886495443248,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004409905095733381,
      "loss": 5.5347,
      "mean_token_accuracy": 0.158642216026783,
      "num_tokens": 6581621.0,
      "step": 3610
    },
    {
      "entropy": 5.608583641052246,
      "epoch": 2.995028997514499,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044077461310626764,
      "loss": 5.5114,
      "mean_token_accuracy": 0.1594192534685135,
      "num_tokens": 6590353.0,
      "step": 3615
    },
    {
      "entropy": 5.687379932403564,
      "epoch": 2.9991714995857497,
      "grad_norm": 1.21875,
      "learning_rate": 0.000440558382253687,
      "loss": 5.5038,
      "mean_token_accuracy": 0.15936833918094634,
      "num_tokens": 6598822.0,
      "step": 3620
    },
    {
      "entropy": 5.65473690032959,
      "epoch": 3.003314001657001,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004403418174517573,
      "loss": 5.2736,
      "mean_token_accuracy": 0.1684224545955658,
      "num_tokens": 6606728.0,
      "step": 3625
    },
    {
      "entropy": 5.608128547668457,
      "epoch": 3.007456503728252,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044012491913731353,
      "loss": 5.1922,
      "mean_token_accuracy": 0.171491277217865,
      "num_tokens": 6616226.0,
      "step": 3630
    },
    {
      "entropy": 5.528424024581909,
      "epoch": 3.011599005799503,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004399076877478632,
      "loss": 5.0393,
      "mean_token_accuracy": 0.18953252136707305,
      "num_tokens": 6624801.0,
      "step": 3635
    },
    {
      "entropy": 5.629195356369019,
      "epoch": 3.015741507870754,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004396901237215858,
      "loss": 5.2488,
      "mean_token_accuracy": 0.17343629151582718,
      "num_tokens": 6633043.0,
      "step": 3640
    },
    {
      "entropy": 5.576674175262451,
      "epoch": 3.019884009942005,
      "grad_norm": 1.125,
      "learning_rate": 0.00043947222749733173,
      "loss": 5.1919,
      "mean_token_accuracy": 0.17088026255369188,
      "num_tokens": 6641197.0,
      "step": 3645
    },
    {
      "entropy": 5.592171764373779,
      "epoch": 3.024026512013256,
      "grad_norm": 1.125,
      "learning_rate": 0.00043925399951462153,
      "loss": 5.2255,
      "mean_token_accuracy": 0.17408303022384644,
      "num_tokens": 6650411.0,
      "step": 3650
    },
    {
      "entropy": 5.478409099578857,
      "epoch": 3.028169014084507,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004390354402136449,
      "loss": 5.1133,
      "mean_token_accuracy": 0.1816069394350052,
      "num_tokens": 6659405.0,
      "step": 3655
    },
    {
      "entropy": 5.61852650642395,
      "epoch": 3.032311516155758,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004388165500352597,
      "loss": 5.1873,
      "mean_token_accuracy": 0.18107183575630187,
      "num_tokens": 6667701.0,
      "step": 3660
    },
    {
      "entropy": 5.6003812789917,
      "epoch": 3.036454018227009,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004385973294209915,
      "loss": 5.2141,
      "mean_token_accuracy": 0.1739385098218918,
      "num_tokens": 6676801.0,
      "step": 3665
    },
    {
      "entropy": 5.48479471206665,
      "epoch": 3.0405965202982603,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004383777788130321,
      "loss": 5.0888,
      "mean_token_accuracy": 0.1855471447110176,
      "num_tokens": 6686005.0,
      "step": 3670
    },
    {
      "entropy": 5.496511220932007,
      "epoch": 3.044739022369511,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004381578986542393,
      "loss": 5.1157,
      "mean_token_accuracy": 0.17760300189256667,
      "num_tokens": 6694884.0,
      "step": 3675
    },
    {
      "entropy": 5.483774185180664,
      "epoch": 3.048881524440762,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004379376893881352,
      "loss": 5.2303,
      "mean_token_accuracy": 0.18040781468153,
      "num_tokens": 6703476.0,
      "step": 3680
    },
    {
      "entropy": 5.610590934753418,
      "epoch": 3.053024026512013,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004377171514589059,
      "loss": 5.2702,
      "mean_token_accuracy": 0.17431289851665496,
      "num_tokens": 6712576.0,
      "step": 3685
    },
    {
      "entropy": 5.457471132278442,
      "epoch": 3.0571665285832643,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004374962853114007,
      "loss": 5.0868,
      "mean_token_accuracy": 0.17861308753490449,
      "num_tokens": 6721532.0,
      "step": 3690
    },
    {
      "entropy": 5.460675001144409,
      "epoch": 3.0613090306545154,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004372750913911306,
      "loss": 5.1198,
      "mean_token_accuracy": 0.1840505540370941,
      "num_tokens": 6730055.0,
      "step": 3695
    },
    {
      "entropy": 5.507229948043824,
      "epoch": 3.0654515327257665,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004370535701442679,
      "loss": 5.1823,
      "mean_token_accuracy": 0.17917860001325608,
      "num_tokens": 6739310.0,
      "step": 3700
    },
    {
      "entropy": 5.547648572921753,
      "epoch": 3.069594034797017,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043683172201764537,
      "loss": 5.2717,
      "mean_token_accuracy": 0.1702258437871933,
      "num_tokens": 6748976.0,
      "step": 3705
    },
    {
      "entropy": 5.529606866836548,
      "epoch": 3.0737365368682683,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004366095474587547,
      "loss": 5.1182,
      "mean_token_accuracy": 0.17836759090423585,
      "num_tokens": 6758043.0,
      "step": 3710
    },
    {
      "entropy": 5.521498918533325,
      "epoch": 3.0778790389395194,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043638704691574646,
      "loss": 5.1912,
      "mean_token_accuracy": 0.17908665388822556,
      "num_tokens": 6767554.0,
      "step": 3715
    },
    {
      "entropy": 5.60535740852356,
      "epoch": 3.0820215410107705,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004361642208374283,
      "loss": 5.2083,
      "mean_token_accuracy": 0.17299294024705886,
      "num_tokens": 6776161.0,
      "step": 3720
    },
    {
      "entropy": 5.466820573806762,
      "epoch": 3.0861640430820216,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043594106967326495,
      "loss": 5.1217,
      "mean_token_accuracy": 0.17925718873739244,
      "num_tokens": 6785371.0,
      "step": 3725
    },
    {
      "entropy": 5.5335582256317135,
      "epoch": 3.0903065451532727,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043571759387337665,
      "loss": 5.1634,
      "mean_token_accuracy": 0.1835888460278511,
      "num_tokens": 6795242.0,
      "step": 3730
    },
    {
      "entropy": 5.547745561599731,
      "epoch": 3.0944490472245234,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004354937938885385,
      "loss": 5.249,
      "mean_token_accuracy": 0.1728634551167488,
      "num_tokens": 6805423.0,
      "step": 3735
    },
    {
      "entropy": 5.529517984390258,
      "epoch": 3.0985915492957745,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004352696701701796,
      "loss": 5.1689,
      "mean_token_accuracy": 0.17931755483150483,
      "num_tokens": 6814377.0,
      "step": 3740
    },
    {
      "entropy": 5.491488599777222,
      "epoch": 3.1027340513670256,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004350452231703819,
      "loss": 5.2307,
      "mean_token_accuracy": 0.17575548440217972,
      "num_tokens": 6824025.0,
      "step": 3745
    },
    {
      "entropy": 5.503986501693726,
      "epoch": 3.1068765534382767,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043482045334187954,
      "loss": 5.1449,
      "mean_token_accuracy": 0.18518491089344025,
      "num_tokens": 6833071.0,
      "step": 3750
    },
    {
      "entropy": 5.433632755279541,
      "epoch": 3.111019055509528,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043459536113805785,
      "loss": 5.1352,
      "mean_token_accuracy": 0.1775007888674736,
      "num_tokens": 6841983.0,
      "step": 3755
    },
    {
      "entropy": 5.491881084442139,
      "epoch": 3.115161557580779,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043436994701295237,
      "loss": 5.1331,
      "mean_token_accuracy": 0.18157943040132524,
      "num_tokens": 6851436.0,
      "step": 3760
    },
    {
      "entropy": 5.493215036392212,
      "epoch": 3.1193040596520296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043414421142124804,
      "loss": 5.1783,
      "mean_token_accuracy": 0.1788194492459297,
      "num_tokens": 6861810.0,
      "step": 3765
    },
    {
      "entropy": 5.526950836181641,
      "epoch": 3.1234465617232807,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043391815481827814,
      "loss": 5.0796,
      "mean_token_accuracy": 0.18713754117488862,
      "num_tokens": 6870418.0,
      "step": 3770
    },
    {
      "entropy": 5.399437761306762,
      "epoch": 3.127589063794532,
      "grad_norm": 1.125,
      "learning_rate": 0.0004336917776600237,
      "loss": 5.0979,
      "mean_token_accuracy": 0.18566158562898635,
      "num_tokens": 6879419.0,
      "step": 3775
    },
    {
      "entropy": 5.424571800231933,
      "epoch": 3.131731565865783,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004334650804031122,
      "loss": 5.2139,
      "mean_token_accuracy": 0.17417935729026796,
      "num_tokens": 6888768.0,
      "step": 3780
    },
    {
      "entropy": 5.586405467987061,
      "epoch": 3.135874067937034,
      "grad_norm": 1.125,
      "learning_rate": 0.0004332380635048165,
      "loss": 5.1836,
      "mean_token_accuracy": 0.17237520962953568,
      "num_tokens": 6897133.0,
      "step": 3785
    },
    {
      "entropy": 5.5270240783691404,
      "epoch": 3.140016570008285,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043301072742305474,
      "loss": 5.1976,
      "mean_token_accuracy": 0.17571339756250381,
      "num_tokens": 6904976.0,
      "step": 3790
    },
    {
      "entropy": 5.558376359939575,
      "epoch": 3.1441590720795363,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004327830726163886,
      "loss": 5.196,
      "mean_token_accuracy": 0.1767548218369484,
      "num_tokens": 6915138.0,
      "step": 3795
    },
    {
      "entropy": 5.543389368057251,
      "epoch": 3.148301574150787,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004325550995440226,
      "loss": 5.1739,
      "mean_token_accuracy": 0.17608616650104522,
      "num_tokens": 6923962.0,
      "step": 3800
    },
    {
      "entropy": 5.395559072494507,
      "epoch": 3.152444076222038,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043232680866580356,
      "loss": 5.1517,
      "mean_token_accuracy": 0.17694357335567473,
      "num_tokens": 6932102.0,
      "step": 3805
    },
    {
      "entropy": 5.490355110168457,
      "epoch": 3.156586578293289,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043209820044221893,
      "loss": 5.1846,
      "mean_token_accuracy": 0.18131535947322847,
      "num_tokens": 6940354.0,
      "step": 3810
    },
    {
      "entropy": 5.519707345962525,
      "epoch": 3.1607290803645403,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043186927533439654,
      "loss": 5.1749,
      "mean_token_accuracy": 0.17507922053337097,
      "num_tokens": 6949779.0,
      "step": 3815
    },
    {
      "entropy": 5.551403045654297,
      "epoch": 3.1648715824357914,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043164003380410335,
      "loss": 5.2449,
      "mean_token_accuracy": 0.17628766000270843,
      "num_tokens": 6958760.0,
      "step": 3820
    },
    {
      "entropy": 5.479237985610962,
      "epoch": 3.169014084507042,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004314104763137445,
      "loss": 5.1042,
      "mean_token_accuracy": 0.18383505642414094,
      "num_tokens": 6967535.0,
      "step": 3825
    },
    {
      "entropy": 5.487510824203492,
      "epoch": 3.173156586578293,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043118060332636264,
      "loss": 5.2234,
      "mean_token_accuracy": 0.17976367324590684,
      "num_tokens": 6976211.0,
      "step": 3830
    },
    {
      "entropy": 5.569280815124512,
      "epoch": 3.1772990886495442,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043095041530563665,
      "loss": 5.2344,
      "mean_token_accuracy": 0.176742222905159,
      "num_tokens": 6986001.0,
      "step": 3835
    },
    {
      "entropy": 5.433036470413208,
      "epoch": 3.1814415907207954,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043071991271588104,
      "loss": 5.2003,
      "mean_token_accuracy": 0.18441623449325562,
      "num_tokens": 6994568.0,
      "step": 3840
    },
    {
      "entropy": 5.5656046867370605,
      "epoch": 3.1855840927920465,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004304890960220446,
      "loss": 5.322,
      "mean_token_accuracy": 0.16879098266363143,
      "num_tokens": 7004744.0,
      "step": 3845
    },
    {
      "entropy": 5.5625487804412845,
      "epoch": 3.1897265948632976,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043025796568970975,
      "loss": 5.3119,
      "mean_token_accuracy": 0.167134191095829,
      "num_tokens": 7014315.0,
      "step": 3850
    },
    {
      "entropy": 5.508922958374024,
      "epoch": 3.1938690969345487,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043002652218509184,
      "loss": 5.1492,
      "mean_token_accuracy": 0.18394915163516998,
      "num_tokens": 7022258.0,
      "step": 3855
    },
    {
      "entropy": 5.534724378585816,
      "epoch": 3.1980115990057993,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004297947659750376,
      "loss": 5.259,
      "mean_token_accuracy": 0.17378944158554077,
      "num_tokens": 7031732.0,
      "step": 3860
    },
    {
      "entropy": 5.489095020294189,
      "epoch": 3.2021541010770505,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004295626975270246,
      "loss": 5.193,
      "mean_token_accuracy": 0.17550500780344008,
      "num_tokens": 7040478.0,
      "step": 3865
    },
    {
      "entropy": 5.512714910507202,
      "epoch": 3.2062966031483016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042933031730916024,
      "loss": 5.1635,
      "mean_token_accuracy": 0.1809170812368393,
      "num_tokens": 7049360.0,
      "step": 3870
    },
    {
      "entropy": 5.5673205852508545,
      "epoch": 3.2104391052195527,
      "grad_norm": 1.1875,
      "learning_rate": 0.000429097625790181,
      "loss": 5.2467,
      "mean_token_accuracy": 0.17765278667211531,
      "num_tokens": 7059271.0,
      "step": 3875
    },
    {
      "entropy": 5.450319290161133,
      "epoch": 3.2145816072908038,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004288646234394508,
      "loss": 5.2269,
      "mean_token_accuracy": 0.17899532318115235,
      "num_tokens": 7069038.0,
      "step": 3880
    },
    {
      "entropy": 5.491114950180053,
      "epoch": 3.218724109362055,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042863131072696115,
      "loss": 5.1796,
      "mean_token_accuracy": 0.1805991604924202,
      "num_tokens": 7077971.0,
      "step": 3885
    },
    {
      "entropy": 5.488774967193604,
      "epoch": 3.2228666114333056,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042839768812332905,
      "loss": 5.1946,
      "mean_token_accuracy": 0.16922565698623657,
      "num_tokens": 7086821.0,
      "step": 3890
    },
    {
      "entropy": 5.46287989616394,
      "epoch": 3.2270091135045567,
      "grad_norm": 1.15625,
      "learning_rate": 0.000428163756099797,
      "loss": 5.2217,
      "mean_token_accuracy": 0.17228440195322037,
      "num_tokens": 7095407.0,
      "step": 3895
    },
    {
      "entropy": 5.468328332901001,
      "epoch": 3.2311516155758078,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004279295151282314,
      "loss": 5.1369,
      "mean_token_accuracy": 0.18177154362201692,
      "num_tokens": 7103964.0,
      "step": 3900
    },
    {
      "entropy": 5.498992919921875,
      "epoch": 3.235294117647059,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004276949656811217,
      "loss": 5.2569,
      "mean_token_accuracy": 0.1760358139872551,
      "num_tokens": 7113571.0,
      "step": 3905
    },
    {
      "entropy": 5.572669744491577,
      "epoch": 3.23943661971831,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042746010823158006,
      "loss": 5.2735,
      "mean_token_accuracy": 0.17531929761171341,
      "num_tokens": 7122844.0,
      "step": 3910
    },
    {
      "entropy": 5.450087594985962,
      "epoch": 3.243579121789561,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042722494325333945,
      "loss": 5.172,
      "mean_token_accuracy": 0.17707158029079437,
      "num_tokens": 7132123.0,
      "step": 3915
    },
    {
      "entropy": 5.493407535552978,
      "epoch": 3.2477216238608118,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004269894712207533,
      "loss": 5.208,
      "mean_token_accuracy": 0.17930082976818085,
      "num_tokens": 7141068.0,
      "step": 3920
    },
    {
      "entropy": 5.45019793510437,
      "epoch": 3.251864125932063,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004267536926087947,
      "loss": 5.2217,
      "mean_token_accuracy": 0.17073139399290085,
      "num_tokens": 7150829.0,
      "step": 3925
    },
    {
      "entropy": 5.46892352104187,
      "epoch": 3.256006628003314,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004265176078930545,
      "loss": 5.1771,
      "mean_token_accuracy": 0.18690198361873628,
      "num_tokens": 7160814.0,
      "step": 3930
    },
    {
      "entropy": 5.543881416320801,
      "epoch": 3.260149130074565,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042628121754974166,
      "loss": 5.2803,
      "mean_token_accuracy": 0.1723069131374359,
      "num_tokens": 7170085.0,
      "step": 3935
    },
    {
      "entropy": 5.560903167724609,
      "epoch": 3.264291632145816,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004260445220556811,
      "loss": 5.2425,
      "mean_token_accuracy": 0.17817744761705398,
      "num_tokens": 7179366.0,
      "step": 3940
    },
    {
      "entropy": 5.578168535232544,
      "epoch": 3.2684341342170673,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042580752188831376,
      "loss": 5.304,
      "mean_token_accuracy": 0.17222840785980226,
      "num_tokens": 7188342.0,
      "step": 3945
    },
    {
      "entropy": 5.578952741622925,
      "epoch": 3.272576636288318,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004255702175256946,
      "loss": 5.3315,
      "mean_token_accuracy": 0.1791801705956459,
      "num_tokens": 7197466.0,
      "step": 3950
    },
    {
      "entropy": 5.46775631904602,
      "epoch": 3.276719138359569,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042533260944649273,
      "loss": 5.1905,
      "mean_token_accuracy": 0.17830806821584702,
      "num_tokens": 7205462.0,
      "step": 3955
    },
    {
      "entropy": 5.4653631210327145,
      "epoch": 3.28086164043082,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004250946981299893,
      "loss": 5.2049,
      "mean_token_accuracy": 0.17918088734149934,
      "num_tokens": 7215115.0,
      "step": 3960
    },
    {
      "entropy": 5.471656036376953,
      "epoch": 3.2850041425020713,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042485648405607776,
      "loss": 5.1558,
      "mean_token_accuracy": 0.1752260610461235,
      "num_tokens": 7224909.0,
      "step": 3965
    },
    {
      "entropy": 5.363120985031128,
      "epoch": 3.2891466445733224,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004246179677052617,
      "loss": 5.082,
      "mean_token_accuracy": 0.19152185171842576,
      "num_tokens": 7232946.0,
      "step": 3970
    },
    {
      "entropy": 5.45828218460083,
      "epoch": 3.2932891466445735,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004243791495586547,
      "loss": 5.1925,
      "mean_token_accuracy": 0.18146124333143235,
      "num_tokens": 7241104.0,
      "step": 3975
    },
    {
      "entropy": 5.594405555725098,
      "epoch": 3.297431648715824,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004241400300979791,
      "loss": 5.2484,
      "mean_token_accuracy": 0.17478115111589432,
      "num_tokens": 7250749.0,
      "step": 3980
    },
    {
      "entropy": 5.433809566497803,
      "epoch": 3.3015741507870753,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004239006098055651,
      "loss": 5.1399,
      "mean_token_accuracy": 0.1764796867966652,
      "num_tokens": 7258687.0,
      "step": 3985
    },
    {
      "entropy": 5.400339555740357,
      "epoch": 3.3057166528583264,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004236608891643494,
      "loss": 5.179,
      "mean_token_accuracy": 0.17788545340299605,
      "num_tokens": 7267316.0,
      "step": 3990
    },
    {
      "entropy": 5.495499086380005,
      "epoch": 3.3098591549295775,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004234208686578748,
      "loss": 5.1947,
      "mean_token_accuracy": 0.17531768828630448,
      "num_tokens": 7276734.0,
      "step": 3995
    },
    {
      "entropy": 5.459540557861328,
      "epoch": 3.3140016570008286,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042318054877028876,
      "loss": 5.1834,
      "mean_token_accuracy": 0.18060380220413208,
      "num_tokens": 7285916.0,
      "step": 4000
    },
    {
      "epoch": 3.3140016570008286,
      "eval_entropy": 5.368396492655089,
      "eval_loss": 5.911151885986328,
      "eval_mean_token_accuracy": 0.1531766704055292,
      "eval_num_tokens": 7285916.0,
      "eval_runtime": 2.7544,
      "eval_samples_per_second": 1360.35,
      "eval_steps_per_second": 170.271,
      "step": 4000
    },
    {
      "entropy": 5.485665893554687,
      "epoch": 3.3181441590720797,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042293992998634314,
      "loss": 5.1885,
      "mean_token_accuracy": 0.17930345088243485,
      "num_tokens": 7294184.0,
      "step": 4005
    },
    {
      "entropy": 5.432258176803589,
      "epoch": 3.3222866611433304,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004226990127913918,
      "loss": 5.2016,
      "mean_token_accuracy": 0.178544519841671,
      "num_tokens": 7302647.0,
      "step": 4010
    },
    {
      "entropy": 5.458320236206054,
      "epoch": 3.3264291632145815,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004224577976713915,
      "loss": 5.2821,
      "mean_token_accuracy": 0.16796015501022338,
      "num_tokens": 7311455.0,
      "step": 4015
    },
    {
      "entropy": 5.627921485900879,
      "epoch": 3.3305716652858326,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042221628511289927,
      "loss": 5.3332,
      "mean_token_accuracy": 0.163201043009758,
      "num_tokens": 7320978.0,
      "step": 4020
    },
    {
      "entropy": 5.610194492340088,
      "epoch": 3.3347141673570837,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004219744756030723,
      "loss": 5.3411,
      "mean_token_accuracy": 0.1671421632170677,
      "num_tokens": 7330202.0,
      "step": 4025
    },
    {
      "entropy": 5.405560779571533,
      "epoch": 3.338856669428335,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042173236962966706,
      "loss": 5.1807,
      "mean_token_accuracy": 0.18122173845767975,
      "num_tokens": 7339121.0,
      "step": 4030
    },
    {
      "entropy": 5.506384181976318,
      "epoch": 3.342999171499586,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004214899676810375,
      "loss": 5.2531,
      "mean_token_accuracy": 0.1713682681322098,
      "num_tokens": 7348631.0,
      "step": 4035
    },
    {
      "entropy": 5.48051905632019,
      "epoch": 3.347141673570837,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004212472702461348,
      "loss": 5.2107,
      "mean_token_accuracy": 0.18103217482566833,
      "num_tokens": 7358354.0,
      "step": 4040
    },
    {
      "entropy": 5.467200136184692,
      "epoch": 3.3512841756420877,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004210042778145064,
      "loss": 5.2101,
      "mean_token_accuracy": 0.17799277156591414,
      "num_tokens": 7367031.0,
      "step": 4045
    },
    {
      "entropy": 5.4672215461730955,
      "epoch": 3.355426677713339,
      "grad_norm": 1.40625,
      "learning_rate": 0.00042076099087629443,
      "loss": 5.2203,
      "mean_token_accuracy": 0.1767668604850769,
      "num_tokens": 7375996.0,
      "step": 4050
    },
    {
      "entropy": 5.548968887329101,
      "epoch": 3.35956917978459,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042051740992223534,
      "loss": 5.279,
      "mean_token_accuracy": 0.16718391925096512,
      "num_tokens": 7385355.0,
      "step": 4055
    },
    {
      "entropy": 5.41011872291565,
      "epoch": 3.363711681855841,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004202735354436586,
      "loss": 5.108,
      "mean_token_accuracy": 0.18245522677898407,
      "num_tokens": 7394384.0,
      "step": 4060
    },
    {
      "entropy": 5.483056640625,
      "epoch": 3.367854183927092,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042002936793248546,
      "loss": 5.2523,
      "mean_token_accuracy": 0.18001522570848466,
      "num_tokens": 7403141.0,
      "step": 4065
    },
    {
      "entropy": 5.477651453018188,
      "epoch": 3.371996685998343,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004197849078812286,
      "loss": 5.2182,
      "mean_token_accuracy": 0.1779205918312073,
      "num_tokens": 7413331.0,
      "step": 4070
    },
    {
      "entropy": 5.5872838497161865,
      "epoch": 3.376139188069594,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004195401557829907,
      "loss": 5.2573,
      "mean_token_accuracy": 0.17058736681938172,
      "num_tokens": 7422362.0,
      "step": 4075
    },
    {
      "entropy": 5.449357652664185,
      "epoch": 3.380281690140845,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041929511213146343,
      "loss": 5.1764,
      "mean_token_accuracy": 0.17998363822698593,
      "num_tokens": 7431350.0,
      "step": 4080
    },
    {
      "entropy": 5.4769665718078615,
      "epoch": 3.384424192212096,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004190497774209266,
      "loss": 5.1622,
      "mean_token_accuracy": 0.18225474208593367,
      "num_tokens": 7440382.0,
      "step": 4085
    },
    {
      "entropy": 5.5300483226776125,
      "epoch": 3.3885666942833472,
      "grad_norm": 1.125,
      "learning_rate": 0.00041880415214624715,
      "loss": 5.2579,
      "mean_token_accuracy": 0.17903644889593123,
      "num_tokens": 7449357.0,
      "step": 4090
    },
    {
      "entropy": 5.504210472106934,
      "epoch": 3.3927091963545983,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004185582368028781,
      "loss": 5.1716,
      "mean_token_accuracy": 0.1800335317850113,
      "num_tokens": 7457544.0,
      "step": 4095
    },
    {
      "entropy": 5.432285499572754,
      "epoch": 3.3968516984258494,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004183120318868577,
      "loss": 5.1854,
      "mean_token_accuracy": 0.19055123776197433,
      "num_tokens": 7465915.0,
      "step": 4100
    },
    {
      "entropy": 5.4560328960418705,
      "epoch": 3.4009942004971,
      "grad_norm": 1.375,
      "learning_rate": 0.00041806553789480787,
      "loss": 5.1639,
      "mean_token_accuracy": 0.1798722505569458,
      "num_tokens": 7474551.0,
      "step": 4105
    },
    {
      "entropy": 5.477843856811523,
      "epoch": 3.405136702568351,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041781875532393423,
      "loss": 5.2476,
      "mean_token_accuracy": 0.17774663716554642,
      "num_tokens": 7483271.0,
      "step": 4110
    },
    {
      "entropy": 5.551624536514282,
      "epoch": 3.4092792046396023,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041757168467202404,
      "loss": 5.2635,
      "mean_token_accuracy": 0.16868621557950975,
      "num_tokens": 7493179.0,
      "step": 4115
    },
    {
      "entropy": 5.494732904434204,
      "epoch": 3.4134217067108534,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041732432643744576,
      "loss": 5.2252,
      "mean_token_accuracy": 0.17510553747415541,
      "num_tokens": 7501858.0,
      "step": 4120
    },
    {
      "entropy": 5.509705066680908,
      "epoch": 3.4175642087821045,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004170766811191481,
      "loss": 5.3021,
      "mean_token_accuracy": 0.17410938292741776,
      "num_tokens": 7511817.0,
      "step": 4125
    },
    {
      "entropy": 5.490671014785766,
      "epoch": 3.421706710853355,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004168287492166587,
      "loss": 5.2743,
      "mean_token_accuracy": 0.17766620516777037,
      "num_tokens": 7521655.0,
      "step": 4130
    },
    {
      "entropy": 5.497408199310303,
      "epoch": 3.4258492129246063,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041658053123008325,
      "loss": 5.252,
      "mean_token_accuracy": 0.17472622394561768,
      "num_tokens": 7531321.0,
      "step": 4135
    },
    {
      "entropy": 5.477228498458862,
      "epoch": 3.4299917149958574,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004163320276601046,
      "loss": 5.2403,
      "mean_token_accuracy": 0.1746629521250725,
      "num_tokens": 7540613.0,
      "step": 4140
    },
    {
      "entropy": 5.487975597381592,
      "epoch": 3.4341342170671085,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004160832390079816,
      "loss": 5.2993,
      "mean_token_accuracy": 0.1757570117712021,
      "num_tokens": 7549598.0,
      "step": 4145
    },
    {
      "entropy": 5.475250387191773,
      "epoch": 3.4382767191383596,
      "grad_norm": 1.125,
      "learning_rate": 0.00041583416577554815,
      "loss": 5.1797,
      "mean_token_accuracy": 0.1805828258395195,
      "num_tokens": 7558738.0,
      "step": 4150
    },
    {
      "entropy": 5.486615133285523,
      "epoch": 3.4424192212096107,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041558480846521213,
      "loss": 5.2191,
      "mean_token_accuracy": 0.1765519991517067,
      "num_tokens": 7566873.0,
      "step": 4155
    },
    {
      "entropy": 5.435937547683716,
      "epoch": 3.446561723280862,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041533516757995455,
      "loss": 5.1981,
      "mean_token_accuracy": 0.1783042773604393,
      "num_tokens": 7576645.0,
      "step": 4160
    },
    {
      "entropy": 5.4572593688964846,
      "epoch": 3.4507042253521125,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004150852436233283,
      "loss": 5.1985,
      "mean_token_accuracy": 0.17680279910564423,
      "num_tokens": 7584854.0,
      "step": 4165
    },
    {
      "entropy": 5.5248284339904785,
      "epoch": 3.4548467274233636,
      "grad_norm": 1.125,
      "learning_rate": 0.0004148350370994574,
      "loss": 5.3295,
      "mean_token_accuracy": 0.1705993190407753,
      "num_tokens": 7594022.0,
      "step": 4170
    },
    {
      "entropy": 5.466347408294678,
      "epoch": 3.4589892294946147,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004145845485130357,
      "loss": 5.1956,
      "mean_token_accuracy": 0.17772535532712935,
      "num_tokens": 7603259.0,
      "step": 4175
    },
    {
      "entropy": 5.420287275314331,
      "epoch": 3.463131731565866,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041433377836932594,
      "loss": 5.3028,
      "mean_token_accuracy": 0.1668353110551834,
      "num_tokens": 7613451.0,
      "step": 4180
    },
    {
      "entropy": 5.579280090332031,
      "epoch": 3.467274233637117,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041408272717415913,
      "loss": 5.2697,
      "mean_token_accuracy": 0.17549486607313156,
      "num_tokens": 7622218.0,
      "step": 4185
    },
    {
      "entropy": 5.3916911602020265,
      "epoch": 3.4714167357083676,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004138313954339327,
      "loss": 5.1244,
      "mean_token_accuracy": 0.17975206077098846,
      "num_tokens": 7630508.0,
      "step": 4190
    },
    {
      "entropy": 5.489048099517822,
      "epoch": 3.4755592377796187,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004135797836556106,
      "loss": 5.2042,
      "mean_token_accuracy": 0.18022646605968476,
      "num_tokens": 7640141.0,
      "step": 4195
    },
    {
      "entropy": 5.443240308761597,
      "epoch": 3.47970173985087,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041332789234672094,
      "loss": 5.2273,
      "mean_token_accuracy": 0.17134273797273636,
      "num_tokens": 7648764.0,
      "step": 4200
    },
    {
      "entropy": 5.443126392364502,
      "epoch": 3.483844241922121,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004130757220153561,
      "loss": 5.2169,
      "mean_token_accuracy": 0.18099185526371003,
      "num_tokens": 7658642.0,
      "step": 4205
    },
    {
      "entropy": 5.580880975723266,
      "epoch": 3.487986743993372,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004128232731701712,
      "loss": 5.3942,
      "mean_token_accuracy": 0.16530214846134186,
      "num_tokens": 7667524.0,
      "step": 4210
    },
    {
      "entropy": 5.503347635269165,
      "epoch": 3.492129246064623,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004125705463203833,
      "loss": 5.2072,
      "mean_token_accuracy": 0.18652101904153823,
      "num_tokens": 7675800.0,
      "step": 4215
    },
    {
      "entropy": 5.500323915481568,
      "epoch": 3.4962717481358743,
      "grad_norm": 1.25,
      "learning_rate": 0.00041231754197576986,
      "loss": 5.2621,
      "mean_token_accuracy": 0.17719916105270386,
      "num_tokens": 7683742.0,
      "step": 4220
    },
    {
      "entropy": 5.469090843200684,
      "epoch": 3.5004142502071254,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004120642606466685,
      "loss": 5.2636,
      "mean_token_accuracy": 0.17630812972784043,
      "num_tokens": 7692379.0,
      "step": 4225
    },
    {
      "entropy": 5.52761492729187,
      "epoch": 3.504556752278376,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004118107028439751,
      "loss": 5.254,
      "mean_token_accuracy": 0.17439160943031312,
      "num_tokens": 7702156.0,
      "step": 4230
    },
    {
      "entropy": 5.470154428482056,
      "epoch": 3.508699254349627,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041155686907914365,
      "loss": 5.2302,
      "mean_token_accuracy": 0.1728260651230812,
      "num_tokens": 7712135.0,
      "step": 4235
    },
    {
      "entropy": 5.464836168289184,
      "epoch": 3.5128417564208783,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004113027598641845,
      "loss": 5.1959,
      "mean_token_accuracy": 0.1753637671470642,
      "num_tokens": 7721347.0,
      "step": 4240
    },
    {
      "entropy": 5.445949077606201,
      "epoch": 3.5169842584921294,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004110483757116637,
      "loss": 5.2288,
      "mean_token_accuracy": 0.17813129127025604,
      "num_tokens": 7731351.0,
      "step": 4245
    },
    {
      "entropy": 5.498573207855225,
      "epoch": 3.52112676056338,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041079371713470195,
      "loss": 5.1307,
      "mean_token_accuracy": 0.1867936447262764,
      "num_tokens": 7740413.0,
      "step": 4250
    },
    {
      "entropy": 5.4525069236755375,
      "epoch": 3.525269262634631,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041053878464697343,
      "loss": 5.2345,
      "mean_token_accuracy": 0.17967701256275176,
      "num_tokens": 7749319.0,
      "step": 4255
    },
    {
      "entropy": 5.54470477104187,
      "epoch": 3.5294117647058822,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004102835787627047,
      "loss": 5.2767,
      "mean_token_accuracy": 0.17431822121143342,
      "num_tokens": 7757936.0,
      "step": 4260
    },
    {
      "entropy": 5.5463841438293455,
      "epoch": 3.5335542667771334,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004100280999966741,
      "loss": 5.3373,
      "mean_token_accuracy": 0.17934031113982202,
      "num_tokens": 7767349.0,
      "step": 4265
    },
    {
      "entropy": 5.427485036849975,
      "epoch": 3.5376967688483845,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040977234886421007,
      "loss": 5.1287,
      "mean_token_accuracy": 0.17617111504077912,
      "num_tokens": 7776448.0,
      "step": 4270
    },
    {
      "entropy": 5.451108407974243,
      "epoch": 3.5418392709196356,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040951632588119064,
      "loss": 5.2264,
      "mean_token_accuracy": 0.17633378505706787,
      "num_tokens": 7785768.0,
      "step": 4275
    },
    {
      "entropy": 5.602209663391113,
      "epoch": 3.5459817729908867,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004092600315640422,
      "loss": 5.3227,
      "mean_token_accuracy": 0.16932764798402786,
      "num_tokens": 7794803.0,
      "step": 4280
    },
    {
      "entropy": 5.463848304748535,
      "epoch": 3.550124275062138,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004090034664297382,
      "loss": 5.1837,
      "mean_token_accuracy": 0.17971521317958833,
      "num_tokens": 7803533.0,
      "step": 4285
    },
    {
      "entropy": 5.482815361022949,
      "epoch": 3.5542667771333885,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040874663099579883,
      "loss": 5.2329,
      "mean_token_accuracy": 0.1812317743897438,
      "num_tokens": 7812854.0,
      "step": 4290
    },
    {
      "entropy": 5.554046106338501,
      "epoch": 3.5584092792046396,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040848952578028915,
      "loss": 5.204,
      "mean_token_accuracy": 0.1752168819308281,
      "num_tokens": 7821683.0,
      "step": 4295
    },
    {
      "entropy": 5.42637448310852,
      "epoch": 3.5625517812758907,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004082321513018183,
      "loss": 5.2084,
      "mean_token_accuracy": 0.18373320996761322,
      "num_tokens": 7830967.0,
      "step": 4300
    },
    {
      "entropy": 5.483717727661133,
      "epoch": 3.566694283347142,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040797450807953897,
      "loss": 5.2312,
      "mean_token_accuracy": 0.17519243210554122,
      "num_tokens": 7841179.0,
      "step": 4305
    },
    {
      "entropy": 5.526522779464722,
      "epoch": 3.5708367854183924,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004077165966331454,
      "loss": 5.1941,
      "mean_token_accuracy": 0.17861186861991882,
      "num_tokens": 7849841.0,
      "step": 4310
    },
    {
      "entropy": 5.374811792373658,
      "epoch": 3.5749792874896436,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004074584174828735,
      "loss": 5.2229,
      "mean_token_accuracy": 0.17876483350992203,
      "num_tokens": 7859390.0,
      "step": 4315
    },
    {
      "entropy": 5.537761068344116,
      "epoch": 3.5791217895608947,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040719997114949856,
      "loss": 5.2098,
      "mean_token_accuracy": 0.17760876566171646,
      "num_tokens": 7868848.0,
      "step": 4320
    },
    {
      "entropy": 5.485729694366455,
      "epoch": 3.5832642916321458,
      "grad_norm": 1.25,
      "learning_rate": 0.00040694125815433533,
      "loss": 5.2625,
      "mean_token_accuracy": 0.1791510909795761,
      "num_tokens": 7877521.0,
      "step": 4325
    },
    {
      "entropy": 5.47872109413147,
      "epoch": 3.587406793703397,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004066822790192359,
      "loss": 5.2386,
      "mean_token_accuracy": 0.17770601958036422,
      "num_tokens": 7886355.0,
      "step": 4330
    },
    {
      "entropy": 5.513595104217529,
      "epoch": 3.591549295774648,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040642303426658967,
      "loss": 5.2643,
      "mean_token_accuracy": 0.17917318642139435,
      "num_tokens": 7895028.0,
      "step": 4335
    },
    {
      "entropy": 5.416304969787598,
      "epoch": 3.595691797845899,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040616352441932163,
      "loss": 5.157,
      "mean_token_accuracy": 0.1787642389535904,
      "num_tokens": 7904192.0,
      "step": 4340
    },
    {
      "entropy": 5.409717130661011,
      "epoch": 3.59983429991715,
      "grad_norm": 1.25,
      "learning_rate": 0.00040590375000089146,
      "loss": 5.2402,
      "mean_token_accuracy": 0.1671016678214073,
      "num_tokens": 7912328.0,
      "step": 4345
    },
    {
      "entropy": 5.533446025848389,
      "epoch": 3.603976801988401,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00040564371153529256,
      "loss": 5.2667,
      "mean_token_accuracy": 0.1743027612566948,
      "num_tokens": 7922594.0,
      "step": 4350
    },
    {
      "entropy": 5.476529502868653,
      "epoch": 3.608119304059652,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004053834095470509,
      "loss": 5.2288,
      "mean_token_accuracy": 0.16638234853744507,
      "num_tokens": 7932345.0,
      "step": 4355
    },
    {
      "entropy": 5.532585000991821,
      "epoch": 3.612261806130903,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040512284456122407,
      "loss": 5.2353,
      "mean_token_accuracy": 0.17749128341674805,
      "num_tokens": 7942800.0,
      "step": 4360
    },
    {
      "entropy": 5.442448711395263,
      "epoch": 3.616404308202154,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040486201710340016,
      "loss": 5.1496,
      "mean_token_accuracy": 0.18001354336738587,
      "num_tokens": 7951094.0,
      "step": 4365
    },
    {
      "entropy": 5.441343784332275,
      "epoch": 3.6205468102734053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040460092769969654,
      "loss": 5.2174,
      "mean_token_accuracy": 0.1730321764945984,
      "num_tokens": 7960652.0,
      "step": 4370
    },
    {
      "entropy": 5.571236658096313,
      "epoch": 3.624689312344656,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004043395768767592,
      "loss": 5.3051,
      "mean_token_accuracy": 0.1689101904630661,
      "num_tokens": 7969350.0,
      "step": 4375
    },
    {
      "entropy": 5.4977868556976315,
      "epoch": 3.628831814415907,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004040779651617612,
      "loss": 5.1953,
      "mean_token_accuracy": 0.17809638679027556,
      "num_tokens": 7978969.0,
      "step": 4380
    },
    {
      "entropy": 5.427513170242309,
      "epoch": 3.632974316487158,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040381609308240207,
      "loss": 5.2443,
      "mean_token_accuracy": 0.17994024604558945,
      "num_tokens": 7988531.0,
      "step": 4385
    },
    {
      "entropy": 5.485544919967651,
      "epoch": 3.6371168185584093,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040355396116690634,
      "loss": 5.2064,
      "mean_token_accuracy": 0.17305364310741425,
      "num_tokens": 7996576.0,
      "step": 4390
    },
    {
      "entropy": 5.475402927398681,
      "epoch": 3.6412593206296604,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004032915699440227,
      "loss": 5.2521,
      "mean_token_accuracy": 0.17335378378629684,
      "num_tokens": 8006279.0,
      "step": 4395
    },
    {
      "entropy": 5.493504381179809,
      "epoch": 3.6454018227009115,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000403028919943023,
      "loss": 5.2271,
      "mean_token_accuracy": 0.17998699843883514,
      "num_tokens": 8015476.0,
      "step": 4400
    },
    {
      "entropy": 5.4873686790466305,
      "epoch": 3.6495443247721626,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004027660116937009,
      "loss": 5.2477,
      "mean_token_accuracy": 0.17394260615110396,
      "num_tokens": 8025450.0,
      "step": 4405
    },
    {
      "entropy": 5.42911205291748,
      "epoch": 3.6536868268434133,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004025028457263712,
      "loss": 5.2456,
      "mean_token_accuracy": 0.18111443370580674,
      "num_tokens": 8034850.0,
      "step": 4410
    },
    {
      "entropy": 5.431485271453857,
      "epoch": 3.6578293289146644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004022394225718684,
      "loss": 5.1864,
      "mean_token_accuracy": 0.17571805864572526,
      "num_tokens": 8044526.0,
      "step": 4415
    },
    {
      "entropy": 5.437715530395508,
      "epoch": 3.6619718309859155,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004019757427615456,
      "loss": 5.141,
      "mean_token_accuracy": 0.17732379585504532,
      "num_tokens": 8054033.0,
      "step": 4420
    },
    {
      "entropy": 5.471504497528076,
      "epoch": 3.6661143330571666,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004017118068272739,
      "loss": 5.2639,
      "mean_token_accuracy": 0.1761350080370903,
      "num_tokens": 8063535.0,
      "step": 4425
    },
    {
      "entropy": 5.546802663803101,
      "epoch": 3.6702568351284177,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040144761530144096,
      "loss": 5.2413,
      "mean_token_accuracy": 0.1861344113945961,
      "num_tokens": 8073079.0,
      "step": 4430
    },
    {
      "entropy": 5.373814058303833,
      "epoch": 3.6743993371996684,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040118316871695,
      "loss": 5.1569,
      "mean_token_accuracy": 0.183153635263443,
      "num_tokens": 8082925.0,
      "step": 4435
    },
    {
      "entropy": 5.454323959350586,
      "epoch": 3.6785418392709195,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040091846760721857,
      "loss": 5.2731,
      "mean_token_accuracy": 0.17695521861314772,
      "num_tokens": 8092038.0,
      "step": 4440
    },
    {
      "entropy": 5.486633825302124,
      "epoch": 3.6826843413421706,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040065351250617774,
      "loss": 5.2104,
      "mean_token_accuracy": 0.17182173877954482,
      "num_tokens": 8101581.0,
      "step": 4445
    },
    {
      "entropy": 5.4517871856689455,
      "epoch": 3.6868268434134217,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040038830394827107,
      "loss": 5.2999,
      "mean_token_accuracy": 0.16965181678533553,
      "num_tokens": 8111392.0,
      "step": 4450
    },
    {
      "entropy": 5.539649868011475,
      "epoch": 3.690969345484673,
      "grad_norm": 1.203125,
      "learning_rate": 0.000400122842468453,
      "loss": 5.3106,
      "mean_token_accuracy": 0.16496230214834212,
      "num_tokens": 8120941.0,
      "step": 4455
    },
    {
      "entropy": 5.474268579483033,
      "epoch": 3.695111847555924,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003998571286021884,
      "loss": 5.2245,
      "mean_token_accuracy": 0.1711190015077591,
      "num_tokens": 8129983.0,
      "step": 4460
    },
    {
      "entropy": 5.5328563213348385,
      "epoch": 3.699254349627175,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003995911628854512,
      "loss": 5.3217,
      "mean_token_accuracy": 0.17576952427625656,
      "num_tokens": 8139428.0,
      "step": 4465
    },
    {
      "entropy": 5.537790441513062,
      "epoch": 3.7033968516984257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003993249458547234,
      "loss": 5.2564,
      "mean_token_accuracy": 0.1714064970612526,
      "num_tokens": 8148520.0,
      "step": 4470
    },
    {
      "entropy": 5.400145816802978,
      "epoch": 3.707539353769677,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039905847804699363,
      "loss": 5.0781,
      "mean_token_accuracy": 0.19130491316318513,
      "num_tokens": 8156908.0,
      "step": 4475
    },
    {
      "entropy": 5.433554601669312,
      "epoch": 3.711681855840928,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039879175999975677,
      "loss": 5.2379,
      "mean_token_accuracy": 0.17966145277023315,
      "num_tokens": 8165677.0,
      "step": 4480
    },
    {
      "entropy": 5.4159424781799315,
      "epoch": 3.715824357912179,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039852479225101217,
      "loss": 5.2058,
      "mean_token_accuracy": 0.1717654749751091,
      "num_tokens": 8174786.0,
      "step": 4485
    },
    {
      "entropy": 5.529488563537598,
      "epoch": 3.71996685998343,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039825757533926305,
      "loss": 5.2636,
      "mean_token_accuracy": 0.16528319716453552,
      "num_tokens": 8184150.0,
      "step": 4490
    },
    {
      "entropy": 5.472427845001221,
      "epoch": 3.724109362054681,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039799010980351494,
      "loss": 5.1701,
      "mean_token_accuracy": 0.18973219394683838,
      "num_tokens": 8192257.0,
      "step": 4495
    },
    {
      "entropy": 5.4878106117248535,
      "epoch": 3.728251864125932,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039772239618327535,
      "loss": 5.2456,
      "mean_token_accuracy": 0.17661782205104828,
      "num_tokens": 8201041.0,
      "step": 4500
    },
    {
      "epoch": 3.728251864125932,
      "eval_entropy": 5.359789646002275,
      "eval_loss": 5.869030475616455,
      "eval_mean_token_accuracy": 0.15561373882901186,
      "eval_num_tokens": 8201041.0,
      "eval_runtime": 2.9451,
      "eval_samples_per_second": 1272.304,
      "eval_steps_per_second": 159.25,
      "step": 4500
    },
    {
      "entropy": 5.483052015304565,
      "epoch": 3.732394366197183,
      "grad_norm": 1.25,
      "learning_rate": 0.00039745443501855156,
      "loss": 5.2076,
      "mean_token_accuracy": 0.1807943269610405,
      "num_tokens": 8209881.0,
      "step": 4505
    },
    {
      "entropy": 5.432073307037354,
      "epoch": 3.736536868268434,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039718622684985075,
      "loss": 5.233,
      "mean_token_accuracy": 0.1856631428003311,
      "num_tokens": 8218504.0,
      "step": 4510
    },
    {
      "entropy": 5.461060571670532,
      "epoch": 3.7406793703396852,
      "grad_norm": 1.234375,
      "learning_rate": 0.000396917772218178,
      "loss": 5.0646,
      "mean_token_accuracy": 0.19228380024433137,
      "num_tokens": 8226785.0,
      "step": 4515
    },
    {
      "entropy": 5.513548183441162,
      "epoch": 3.7448218724109363,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003966490716650357,
      "loss": 5.2095,
      "mean_token_accuracy": 0.18389481157064438,
      "num_tokens": 8236092.0,
      "step": 4520
    },
    {
      "entropy": 5.3846063137054445,
      "epoch": 3.7489643744821874,
      "grad_norm": 1.25,
      "learning_rate": 0.00039638012573242225,
      "loss": 5.2105,
      "mean_token_accuracy": 0.18304691314697266,
      "num_tokens": 8245297.0,
      "step": 4525
    },
    {
      "entropy": 5.519845771789551,
      "epoch": 3.7531068765534386,
      "grad_norm": 1.125,
      "learning_rate": 0.0003961109349628309,
      "loss": 5.2888,
      "mean_token_accuracy": 0.16889165788888932,
      "num_tokens": 8254772.0,
      "step": 4530
    },
    {
      "entropy": 5.504732084274292,
      "epoch": 3.757249378624689,
      "grad_norm": 1.265625,
      "learning_rate": 0.000395841499899249,
      "loss": 5.3241,
      "mean_token_accuracy": 0.17156866043806077,
      "num_tokens": 8264310.0,
      "step": 4535
    },
    {
      "entropy": 5.494646978378296,
      "epoch": 3.7613918806959403,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039557182108515624,
      "loss": 5.2377,
      "mean_token_accuracy": 0.1806105524301529,
      "num_tokens": 8274184.0,
      "step": 4540
    },
    {
      "entropy": 5.457763242721557,
      "epoch": 3.7655343827671914,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039530189906452457,
      "loss": 5.1825,
      "mean_token_accuracy": 0.18444707840681077,
      "num_tokens": 8283151.0,
      "step": 4545
    },
    {
      "entropy": 5.503646993637085,
      "epoch": 3.7696768848384425,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039503173438181605,
      "loss": 5.2488,
      "mean_token_accuracy": 0.17376102954149247,
      "num_tokens": 8292574.0,
      "step": 4550
    },
    {
      "entropy": 5.419807815551758,
      "epoch": 3.773819386909693,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039476132758198236,
      "loss": 5.1789,
      "mean_token_accuracy": 0.17953350245952607,
      "num_tokens": 8301281.0,
      "step": 4555
    },
    {
      "entropy": 5.42378077507019,
      "epoch": 3.7779618889809443,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039449067921046353,
      "loss": 5.2304,
      "mean_token_accuracy": 0.18252263218164444,
      "num_tokens": 8310201.0,
      "step": 4560
    },
    {
      "entropy": 5.5194062232971195,
      "epoch": 3.7821043910521954,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039421978981318703,
      "loss": 5.2667,
      "mean_token_accuracy": 0.17719817608594896,
      "num_tokens": 8320369.0,
      "step": 4565
    },
    {
      "entropy": 5.531326961517334,
      "epoch": 3.7862468931234465,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039394865993656625,
      "loss": 5.252,
      "mean_token_accuracy": 0.17671697437763215,
      "num_tokens": 8329519.0,
      "step": 4570
    },
    {
      "entropy": 5.40211820602417,
      "epoch": 3.7903893951946976,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003936772901274999,
      "loss": 5.2587,
      "mean_token_accuracy": 0.174149851500988,
      "num_tokens": 8338282.0,
      "step": 4575
    },
    {
      "entropy": 5.475307178497315,
      "epoch": 3.7945318972659487,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003934056809333705,
      "loss": 5.2536,
      "mean_token_accuracy": 0.17319595962762832,
      "num_tokens": 8346861.0,
      "step": 4580
    },
    {
      "entropy": 5.501742172241211,
      "epoch": 3.7986743993372,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039313383290204333,
      "loss": 5.2873,
      "mean_token_accuracy": 0.17214463353157045,
      "num_tokens": 8356078.0,
      "step": 4585
    },
    {
      "entropy": 5.4629617691040036,
      "epoch": 3.802816901408451,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003928617465818658,
      "loss": 5.2556,
      "mean_token_accuracy": 0.16976846903562545,
      "num_tokens": 8365487.0,
      "step": 4590
    },
    {
      "entropy": 5.553570556640625,
      "epoch": 3.8069594034797016,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039258942252166554,
      "loss": 5.2291,
      "mean_token_accuracy": 0.1798811972141266,
      "num_tokens": 8374622.0,
      "step": 4595
    },
    {
      "entropy": 5.455771589279175,
      "epoch": 3.8111019055509527,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003923168612707502,
      "loss": 5.1697,
      "mean_token_accuracy": 0.18383508920669556,
      "num_tokens": 8383312.0,
      "step": 4600
    },
    {
      "entropy": 5.4079009056091305,
      "epoch": 3.815244407622204,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003920440633789053,
      "loss": 5.1999,
      "mean_token_accuracy": 0.17800944447517394,
      "num_tokens": 8391987.0,
      "step": 4605
    },
    {
      "entropy": 5.477382802963257,
      "epoch": 3.819386909693455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039177102939639424,
      "loss": 5.1847,
      "mean_token_accuracy": 0.1740972265601158,
      "num_tokens": 8401263.0,
      "step": 4610
    },
    {
      "entropy": 5.497080659866333,
      "epoch": 3.8235294117647056,
      "grad_norm": 1.25,
      "learning_rate": 0.0003914977598739563,
      "loss": 5.2396,
      "mean_token_accuracy": 0.18734091222286225,
      "num_tokens": 8410110.0,
      "step": 4615
    },
    {
      "entropy": 5.532918787002563,
      "epoch": 3.8276719138359567,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003912242553628059,
      "loss": 5.3085,
      "mean_token_accuracy": 0.17287394851446153,
      "num_tokens": 8419321.0,
      "step": 4620
    },
    {
      "entropy": 5.5356667041778564,
      "epoch": 3.831814415907208,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003909505164146318,
      "loss": 5.2837,
      "mean_token_accuracy": 0.1744095876812935,
      "num_tokens": 8428376.0,
      "step": 4625
    },
    {
      "entropy": 5.479020452499389,
      "epoch": 3.835956917978459,
      "grad_norm": 1.125,
      "learning_rate": 0.000390676543581595,
      "loss": 5.2062,
      "mean_token_accuracy": 0.17393171042203903,
      "num_tokens": 8437263.0,
      "step": 4630
    },
    {
      "entropy": 5.409183502197266,
      "epoch": 3.84009942004971,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003904023374163289,
      "loss": 5.2173,
      "mean_token_accuracy": 0.18548498600721358,
      "num_tokens": 8445889.0,
      "step": 4635
    },
    {
      "entropy": 5.507781362533569,
      "epoch": 3.844241922120961,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003901278984719372,
      "loss": 5.4051,
      "mean_token_accuracy": 0.1659029632806778,
      "num_tokens": 8456086.0,
      "step": 4640
    },
    {
      "entropy": 5.501009941101074,
      "epoch": 3.8483844241922123,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038985322730199326,
      "loss": 5.2276,
      "mean_token_accuracy": 0.18063105046749114,
      "num_tokens": 8465150.0,
      "step": 4645
    },
    {
      "entropy": 5.488021993637085,
      "epoch": 3.8525269262634634,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003895783244605389,
      "loss": 5.2938,
      "mean_token_accuracy": 0.17158442884683608,
      "num_tokens": 8474172.0,
      "step": 4650
    },
    {
      "entropy": 5.424083518981933,
      "epoch": 3.856669428334714,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038930319050208315,
      "loss": 5.1718,
      "mean_token_accuracy": 0.1856976643204689,
      "num_tokens": 8483636.0,
      "step": 4655
    },
    {
      "entropy": 5.445523118972778,
      "epoch": 3.860811930405965,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038902782598160127,
      "loss": 5.133,
      "mean_token_accuracy": 0.18440633565187453,
      "num_tokens": 8491560.0,
      "step": 4660
    },
    {
      "entropy": 5.530062532424926,
      "epoch": 3.8649544324772163,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038875223145453346,
      "loss": 5.4289,
      "mean_token_accuracy": 0.18210761100053788,
      "num_tokens": 8501685.0,
      "step": 4665
    },
    {
      "entropy": 5.54461259841919,
      "epoch": 3.8690969345484674,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003884764074767842,
      "loss": 5.2639,
      "mean_token_accuracy": 0.17283273488283157,
      "num_tokens": 8510580.0,
      "step": 4670
    },
    {
      "entropy": 5.497996234893799,
      "epoch": 3.873239436619718,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003882003546047205,
      "loss": 5.2241,
      "mean_token_accuracy": 0.1733752891421318,
      "num_tokens": 8519595.0,
      "step": 4675
    },
    {
      "entropy": 5.499145746231079,
      "epoch": 3.877381938690969,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038792407339517123,
      "loss": 5.3103,
      "mean_token_accuracy": 0.17347073405981064,
      "num_tokens": 8529118.0,
      "step": 4680
    },
    {
      "entropy": 5.55419545173645,
      "epoch": 3.8815244407622203,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038764756440542574,
      "loss": 5.2642,
      "mean_token_accuracy": 0.1767756909132004,
      "num_tokens": 8539322.0,
      "step": 4685
    },
    {
      "entropy": 5.460342693328857,
      "epoch": 3.8856669428334714,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038737082819323285,
      "loss": 5.1742,
      "mean_token_accuracy": 0.174711674451828,
      "num_tokens": 8548707.0,
      "step": 4690
    },
    {
      "entropy": 5.474473476409912,
      "epoch": 3.8898094449047225,
      "grad_norm": 1.25,
      "learning_rate": 0.00038709386531679974,
      "loss": 5.2806,
      "mean_token_accuracy": 0.1729741796851158,
      "num_tokens": 8557500.0,
      "step": 4695
    },
    {
      "entropy": 5.450714111328125,
      "epoch": 3.8939519469759736,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038681667633479086,
      "loss": 5.2857,
      "mean_token_accuracy": 0.17091688215732576,
      "num_tokens": 8567465.0,
      "step": 4700
    },
    {
      "entropy": 5.398056554794311,
      "epoch": 3.8980944490472247,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038653926180632665,
      "loss": 5.139,
      "mean_token_accuracy": 0.18227409720420837,
      "num_tokens": 8576680.0,
      "step": 4705
    },
    {
      "entropy": 5.541038942337036,
      "epoch": 3.902236951118476,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038626162229098257,
      "loss": 5.284,
      "mean_token_accuracy": 0.16534818857908248,
      "num_tokens": 8585539.0,
      "step": 4710
    },
    {
      "entropy": 5.461390161514283,
      "epoch": 3.9063794531897265,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003859837583487877,
      "loss": 5.1118,
      "mean_token_accuracy": 0.18798414319753648,
      "num_tokens": 8593598.0,
      "step": 4715
    },
    {
      "entropy": 5.425908422470092,
      "epoch": 3.9105219552609776,
      "grad_norm": 1.25,
      "learning_rate": 0.0003857056705402241,
      "loss": 5.2572,
      "mean_token_accuracy": 0.17270071059465408,
      "num_tokens": 8603440.0,
      "step": 4720
    },
    {
      "entropy": 5.402763938903808,
      "epoch": 3.9146644573322287,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003854273594262253,
      "loss": 5.0748,
      "mean_token_accuracy": 0.18881514370441438,
      "num_tokens": 8612490.0,
      "step": 4725
    },
    {
      "entropy": 5.382251453399658,
      "epoch": 3.91880695940348,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003851488255681752,
      "loss": 5.1738,
      "mean_token_accuracy": 0.18037541210651398,
      "num_tokens": 8621994.0,
      "step": 4730
    },
    {
      "entropy": 5.483442497253418,
      "epoch": 3.922949461474731,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000384870069527907,
      "loss": 5.2547,
      "mean_token_accuracy": 0.17454733550548554,
      "num_tokens": 8631185.0,
      "step": 4735
    },
    {
      "entropy": 5.476728343963623,
      "epoch": 3.9270919635459816,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038459109186770213,
      "loss": 5.1373,
      "mean_token_accuracy": 0.18553917557001115,
      "num_tokens": 8639018.0,
      "step": 4740
    },
    {
      "entropy": 5.389010620117188,
      "epoch": 3.9312344656172327,
      "grad_norm": 1.125,
      "learning_rate": 0.00038431189315028905,
      "loss": 5.136,
      "mean_token_accuracy": 0.1870834484696388,
      "num_tokens": 8647672.0,
      "step": 4745
    },
    {
      "entropy": 5.4637034893035885,
      "epoch": 3.9353769676884838,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003840324739388419,
      "loss": 5.277,
      "mean_token_accuracy": 0.16738566309213637,
      "num_tokens": 8656638.0,
      "step": 4750
    },
    {
      "entropy": 5.4970618724823,
      "epoch": 3.939519469759735,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038375283479698,
      "loss": 5.2259,
      "mean_token_accuracy": 0.1828098863363266,
      "num_tokens": 8666167.0,
      "step": 4755
    },
    {
      "entropy": 5.554366827011108,
      "epoch": 3.943661971830986,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003834729762887659,
      "loss": 5.3116,
      "mean_token_accuracy": 0.16944513618946075,
      "num_tokens": 8676435.0,
      "step": 4760
    },
    {
      "entropy": 5.443045234680175,
      "epoch": 3.947804473902237,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003831928989787049,
      "loss": 5.2106,
      "mean_token_accuracy": 0.17927439361810685,
      "num_tokens": 8686011.0,
      "step": 4765
    },
    {
      "entropy": 5.438700103759766,
      "epoch": 3.951946975973488,
      "grad_norm": 1.125,
      "learning_rate": 0.0003829126034317435,
      "loss": 5.216,
      "mean_token_accuracy": 0.17709566801786422,
      "num_tokens": 8695063.0,
      "step": 4770
    },
    {
      "entropy": 5.548184633255005,
      "epoch": 3.956089478044739,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003826320902132684,
      "loss": 5.3238,
      "mean_token_accuracy": 0.16755364537239076,
      "num_tokens": 8704585.0,
      "step": 4775
    },
    {
      "entropy": 5.488041496276855,
      "epoch": 3.96023198011599,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038235135988910545,
      "loss": 5.2377,
      "mean_token_accuracy": 0.17685187458992005,
      "num_tokens": 8714097.0,
      "step": 4780
    },
    {
      "entropy": 5.452501726150513,
      "epoch": 3.964374482187241,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038207041302551844,
      "loss": 5.155,
      "mean_token_accuracy": 0.18494507074356079,
      "num_tokens": 8722890.0,
      "step": 4785
    },
    {
      "entropy": 5.4004967212677,
      "epoch": 3.968516984258492,
      "grad_norm": 1.25,
      "learning_rate": 0.0003817892501892078,
      "loss": 5.1608,
      "mean_token_accuracy": 0.17960449010133744,
      "num_tokens": 8732046.0,
      "step": 4790
    },
    {
      "entropy": 5.4106584072113035,
      "epoch": 3.9726594863297433,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038150787194730977,
      "loss": 5.2156,
      "mean_token_accuracy": 0.18292873501777648,
      "num_tokens": 8741750.0,
      "step": 4795
    },
    {
      "entropy": 5.450367879867554,
      "epoch": 3.976801988400994,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000381226278867395,
      "loss": 5.1913,
      "mean_token_accuracy": 0.17569466829299926,
      "num_tokens": 8750141.0,
      "step": 4800
    },
    {
      "entropy": 5.394807243347168,
      "epoch": 3.980944490472245,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038094447151746746,
      "loss": 5.1186,
      "mean_token_accuracy": 0.18763228952884675,
      "num_tokens": 8758567.0,
      "step": 4805
    },
    {
      "entropy": 5.4131927490234375,
      "epoch": 3.985086992543496,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003806624504659635,
      "loss": 5.213,
      "mean_token_accuracy": 0.1824996739625931,
      "num_tokens": 8767829.0,
      "step": 4810
    },
    {
      "entropy": 5.423103475570679,
      "epoch": 3.9892294946147473,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003803802162817503,
      "loss": 5.1052,
      "mean_token_accuracy": 0.1871612250804901,
      "num_tokens": 8776911.0,
      "step": 4815
    },
    {
      "entropy": 5.424696207046509,
      "epoch": 3.9933719966859984,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000380097769534125,
      "loss": 5.2108,
      "mean_token_accuracy": 0.18005358427762985,
      "num_tokens": 8786117.0,
      "step": 4820
    },
    {
      "entropy": 5.396023321151733,
      "epoch": 3.9975144987572495,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003798151107928136,
      "loss": 5.1608,
      "mean_token_accuracy": 0.18598888963460922,
      "num_tokens": 8795647.0,
      "step": 4825
    },
    {
      "entropy": 5.376370286941528,
      "epoch": 4.001657000828501,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037953224062796967,
      "loss": 5.1217,
      "mean_token_accuracy": 0.18068824261426925,
      "num_tokens": 8803480.0,
      "step": 4830
    },
    {
      "entropy": 5.468898296356201,
      "epoch": 4.005799502899752,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003792491596101733,
      "loss": 4.9411,
      "mean_token_accuracy": 0.19555393159389495,
      "num_tokens": 8812825.0,
      "step": 4835
    },
    {
      "entropy": 5.464945650100708,
      "epoch": 4.009942004971003,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003789658683104298,
      "loss": 5.0122,
      "mean_token_accuracy": 0.19209524542093276,
      "num_tokens": 8821911.0,
      "step": 4840
    },
    {
      "entropy": 5.468106603622436,
      "epoch": 4.014084507042254,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037868236730016866,
      "loss": 4.9396,
      "mean_token_accuracy": 0.1880512699484825,
      "num_tokens": 8830691.0,
      "step": 4845
    },
    {
      "entropy": 5.433845806121826,
      "epoch": 4.018227009113504,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037839865715124234,
      "loss": 4.9303,
      "mean_token_accuracy": 0.18797015398740768,
      "num_tokens": 8840410.0,
      "step": 4850
    },
    {
      "entropy": 5.357118368148804,
      "epoch": 4.022369511184755,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037811473843592534,
      "loss": 4.91,
      "mean_token_accuracy": 0.1958596184849739,
      "num_tokens": 8849662.0,
      "step": 4855
    },
    {
      "entropy": 5.440473842620849,
      "epoch": 4.026512013256006,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037783061172691273,
      "loss": 4.8919,
      "mean_token_accuracy": 0.19791371375322342,
      "num_tokens": 8858666.0,
      "step": 4860
    },
    {
      "entropy": 5.476719617843628,
      "epoch": 4.0306545153272575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003775462775973191,
      "loss": 4.9438,
      "mean_token_accuracy": 0.19154616743326186,
      "num_tokens": 8867885.0,
      "step": 4865
    },
    {
      "entropy": 5.392060422897339,
      "epoch": 4.034797017398509,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003772617366206773,
      "loss": 4.923,
      "mean_token_accuracy": 0.19716784507036209,
      "num_tokens": 8877458.0,
      "step": 4870
    },
    {
      "entropy": 5.341256952285766,
      "epoch": 4.03893951946976,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037697698937093783,
      "loss": 4.926,
      "mean_token_accuracy": 0.18977416455745696,
      "num_tokens": 8886995.0,
      "step": 4875
    },
    {
      "entropy": 5.460255432128906,
      "epoch": 4.043082021541011,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037669203642246685,
      "loss": 5.0154,
      "mean_token_accuracy": 0.19042324274778366,
      "num_tokens": 8895224.0,
      "step": 4880
    },
    {
      "entropy": 5.4328697681427,
      "epoch": 4.047224523612262,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037640687835004567,
      "loss": 4.9198,
      "mean_token_accuracy": 0.1944139927625656,
      "num_tokens": 8905216.0,
      "step": 4885
    },
    {
      "entropy": 5.429710817337036,
      "epoch": 4.051367025683513,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003761215157288692,
      "loss": 4.9663,
      "mean_token_accuracy": 0.19634376615285873,
      "num_tokens": 8915493.0,
      "step": 4890
    },
    {
      "entropy": 5.3624241828918455,
      "epoch": 4.055509527754764,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037583594913454514,
      "loss": 4.9252,
      "mean_token_accuracy": 0.1953189492225647,
      "num_tokens": 8924740.0,
      "step": 4895
    },
    {
      "entropy": 5.432469892501831,
      "epoch": 4.059652029826015,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003755501791430925,
      "loss": 4.9532,
      "mean_token_accuracy": 0.19013466387987138,
      "num_tokens": 8934726.0,
      "step": 4900
    },
    {
      "entropy": 5.440652370452881,
      "epoch": 4.063794531897266,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037526420633094044,
      "loss": 4.9609,
      "mean_token_accuracy": 0.1885475307703018,
      "num_tokens": 8944301.0,
      "step": 4905
    },
    {
      "entropy": 5.4080273628234865,
      "epoch": 4.067937033968517,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037497803127492757,
      "loss": 4.9874,
      "mean_token_accuracy": 0.19823203086853028,
      "num_tokens": 8953967.0,
      "step": 4910
    },
    {
      "entropy": 5.361791038513184,
      "epoch": 4.072079536039768,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003746916545523001,
      "loss": 4.888,
      "mean_token_accuracy": 0.19481140673160552,
      "num_tokens": 8962983.0,
      "step": 4915
    },
    {
      "entropy": 5.429137754440307,
      "epoch": 4.076222038111019,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003744050767407113,
      "loss": 5.0031,
      "mean_token_accuracy": 0.1879745066165924,
      "num_tokens": 8972019.0,
      "step": 4920
    },
    {
      "entropy": 5.387852048873901,
      "epoch": 4.08036454018227,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003741182984182198,
      "loss": 4.906,
      "mean_token_accuracy": 0.2037443846464157,
      "num_tokens": 8981751.0,
      "step": 4925
    },
    {
      "entropy": 5.394034767150879,
      "epoch": 4.084507042253521,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003738313201632889,
      "loss": 4.9007,
      "mean_token_accuracy": 0.19996586591005325,
      "num_tokens": 8990330.0,
      "step": 4930
    },
    {
      "entropy": 5.334316301345825,
      "epoch": 4.088649544324772,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003735441425547851,
      "loss": 4.9155,
      "mean_token_accuracy": 0.19760458320379257,
      "num_tokens": 8998366.0,
      "step": 4935
    },
    {
      "entropy": 5.31761622428894,
      "epoch": 4.092792046396023,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037325676617197697,
      "loss": 4.8439,
      "mean_token_accuracy": 0.2019736334681511,
      "num_tokens": 9006881.0,
      "step": 4940
    },
    {
      "entropy": 5.337348175048828,
      "epoch": 4.096934548467274,
      "grad_norm": 1.375,
      "learning_rate": 0.0003729691915945342,
      "loss": 4.8969,
      "mean_token_accuracy": 0.19937277734279632,
      "num_tokens": 9015415.0,
      "step": 4945
    },
    {
      "entropy": 5.404147434234619,
      "epoch": 4.1010770505385254,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000372681419402526,
      "loss": 4.9417,
      "mean_token_accuracy": 0.19824467450380326,
      "num_tokens": 9024578.0,
      "step": 4950
    },
    {
      "entropy": 5.3719419002532955,
      "epoch": 4.105219552609777,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003723934501764203,
      "loss": 4.9652,
      "mean_token_accuracy": 0.1880519673228264,
      "num_tokens": 9032637.0,
      "step": 4955
    },
    {
      "entropy": 5.3956928730010985,
      "epoch": 4.109362054681028,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003721052844970826,
      "loss": 4.8979,
      "mean_token_accuracy": 0.19484544545412064,
      "num_tokens": 9041513.0,
      "step": 4960
    },
    {
      "entropy": 5.380131006240845,
      "epoch": 4.113504556752279,
      "grad_norm": 1.25,
      "learning_rate": 0.00037181692294577457,
      "loss": 4.91,
      "mean_token_accuracy": 0.19260294139385223,
      "num_tokens": 9051024.0,
      "step": 4965
    },
    {
      "entropy": 5.3037251949310305,
      "epoch": 4.117647058823529,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037152836610415284,
      "loss": 4.8878,
      "mean_token_accuracy": 0.19715504497289657,
      "num_tokens": 9059494.0,
      "step": 4970
    },
    {
      "entropy": 5.3629344463348385,
      "epoch": 4.12178956089478,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037123961455426836,
      "loss": 4.8076,
      "mean_token_accuracy": 0.20269779115915298,
      "num_tokens": 9068166.0,
      "step": 4975
    },
    {
      "entropy": 5.36611852645874,
      "epoch": 4.125932062966031,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037095066887856434,
      "loss": 4.9139,
      "mean_token_accuracy": 0.19204219728708266,
      "num_tokens": 9077283.0,
      "step": 4980
    },
    {
      "entropy": 5.3835862159729,
      "epoch": 4.130074565037282,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003706615296598758,
      "loss": 4.928,
      "mean_token_accuracy": 0.19695132374763488,
      "num_tokens": 9086605.0,
      "step": 4985
    },
    {
      "entropy": 5.329501295089722,
      "epoch": 4.134217067108533,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003703721974814282,
      "loss": 4.9011,
      "mean_token_accuracy": 0.19619314521551132,
      "num_tokens": 9095567.0,
      "step": 4990
    },
    {
      "entropy": 5.312191390991211,
      "epoch": 4.1383595691797845,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037008267292683615,
      "loss": 4.8941,
      "mean_token_accuracy": 0.18934808075428008,
      "num_tokens": 9104457.0,
      "step": 4995
    },
    {
      "entropy": 5.389340496063232,
      "epoch": 4.142502071251036,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036979295658010227,
      "loss": 4.9553,
      "mean_token_accuracy": 0.19935455322265624,
      "num_tokens": 9113945.0,
      "step": 5000
    },
    {
      "epoch": 4.142502071251036,
      "eval_entropy": 5.237965325556838,
      "eval_loss": 5.855635166168213,
      "eval_mean_token_accuracy": 0.1573141107458804,
      "eval_num_tokens": 9113945.0,
      "eval_runtime": 2.7543,
      "eval_samples_per_second": 1360.434,
      "eval_steps_per_second": 170.281,
      "step": 5000
    },
    {
      "entropy": 5.397732496261597,
      "epoch": 4.146644573322287,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003695030490256161,
      "loss": 5.0267,
      "mean_token_accuracy": 0.19660246819257737,
      "num_tokens": 9122996.0,
      "step": 5005
    },
    {
      "entropy": 5.353362512588501,
      "epoch": 4.150787075393538,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000369212950848153,
      "loss": 4.8879,
      "mean_token_accuracy": 0.19993864893913268,
      "num_tokens": 9132079.0,
      "step": 5010
    },
    {
      "entropy": 5.396353435516358,
      "epoch": 4.154929577464789,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003689226626328725,
      "loss": 4.9975,
      "mean_token_accuracy": 0.1871913939714432,
      "num_tokens": 9141335.0,
      "step": 5015
    },
    {
      "entropy": 5.424786615371704,
      "epoch": 4.15907207953604,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036863218496531767,
      "loss": 5.0365,
      "mean_token_accuracy": 0.1909020200371742,
      "num_tokens": 9150049.0,
      "step": 5020
    },
    {
      "entropy": 5.413427305221558,
      "epoch": 4.163214581607291,
      "grad_norm": 1.25,
      "learning_rate": 0.0003683415184314139,
      "loss": 4.9105,
      "mean_token_accuracy": 0.1960063174366951,
      "num_tokens": 9159245.0,
      "step": 5025
    },
    {
      "entropy": 5.322030687332154,
      "epoch": 4.167357083678542,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036805066361746716,
      "loss": 4.9247,
      "mean_token_accuracy": 0.19489045292139054,
      "num_tokens": 9167892.0,
      "step": 5030
    },
    {
      "entropy": 5.281346797943115,
      "epoch": 4.1714995857497925,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036775962111016346,
      "loss": 4.9147,
      "mean_token_accuracy": 0.19946275949478148,
      "num_tokens": 9176067.0,
      "step": 5035
    },
    {
      "entropy": 5.337658357620239,
      "epoch": 4.175642087821044,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036746839149656734,
      "loss": 4.914,
      "mean_token_accuracy": 0.19304958283901213,
      "num_tokens": 9185318.0,
      "step": 5040
    },
    {
      "entropy": 5.347688484191894,
      "epoch": 4.179784589892295,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003671769753641207,
      "loss": 4.9583,
      "mean_token_accuracy": 0.1974531352519989,
      "num_tokens": 9194885.0,
      "step": 5045
    },
    {
      "entropy": 5.391808223724365,
      "epoch": 4.183927091963546,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036688537330064187,
      "loss": 4.9956,
      "mean_token_accuracy": 0.19615231305360795,
      "num_tokens": 9203906.0,
      "step": 5050
    },
    {
      "entropy": 5.32472858428955,
      "epoch": 4.188069594034797,
      "grad_norm": 1.265625,
      "learning_rate": 0.000366593585894324,
      "loss": 4.8371,
      "mean_token_accuracy": 0.20416567325592042,
      "num_tokens": 9213351.0,
      "step": 5055
    },
    {
      "entropy": 5.33026704788208,
      "epoch": 4.192212096106048,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003663016137337341,
      "loss": 4.9502,
      "mean_token_accuracy": 0.19210560172796248,
      "num_tokens": 9222191.0,
      "step": 5060
    },
    {
      "entropy": 5.368585538864136,
      "epoch": 4.196354598177299,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036600945740781197,
      "loss": 4.9675,
      "mean_token_accuracy": 0.19075418412685394,
      "num_tokens": 9231517.0,
      "step": 5065
    },
    {
      "entropy": 5.418367576599121,
      "epoch": 4.20049710024855,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036571711750586896,
      "loss": 4.999,
      "mean_token_accuracy": 0.20253635793924332,
      "num_tokens": 9241208.0,
      "step": 5070
    },
    {
      "entropy": 5.367947006225586,
      "epoch": 4.204639602319801,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003654245946175864,
      "loss": 4.9192,
      "mean_token_accuracy": 0.19680061638355256,
      "num_tokens": 9250237.0,
      "step": 5075
    },
    {
      "entropy": 5.3685387134552,
      "epoch": 4.2087821043910525,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036513188933301504,
      "loss": 5.031,
      "mean_token_accuracy": 0.18635352551937104,
      "num_tokens": 9260273.0,
      "step": 5080
    },
    {
      "entropy": 5.409731245040893,
      "epoch": 4.212924606462304,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003648390022425734,
      "loss": 4.9164,
      "mean_token_accuracy": 0.1915673181414604,
      "num_tokens": 9268878.0,
      "step": 5085
    },
    {
      "entropy": 5.328348445892334,
      "epoch": 4.217067108533554,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003645459339370468,
      "loss": 4.929,
      "mean_token_accuracy": 0.20126969665288924,
      "num_tokens": 9276324.0,
      "step": 5090
    },
    {
      "entropy": 5.336078357696533,
      "epoch": 4.221209610604805,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036425268500758594,
      "loss": 4.9996,
      "mean_token_accuracy": 0.19514807164669037,
      "num_tokens": 9285052.0,
      "step": 5095
    },
    {
      "entropy": 5.390273189544677,
      "epoch": 4.225352112676056,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00036395925604570614,
      "loss": 4.8837,
      "mean_token_accuracy": 0.20124629735946656,
      "num_tokens": 9293324.0,
      "step": 5100
    },
    {
      "entropy": 5.399580717086792,
      "epoch": 4.229494614747307,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003636656476432854,
      "loss": 4.9437,
      "mean_token_accuracy": 0.19105997085571289,
      "num_tokens": 9302131.0,
      "step": 5105
    },
    {
      "entropy": 5.278874444961548,
      "epoch": 4.233637116818558,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003633718603925642,
      "loss": 4.8534,
      "mean_token_accuracy": 0.19440596252679826,
      "num_tokens": 9311128.0,
      "step": 5110
    },
    {
      "entropy": 5.424720811843872,
      "epoch": 4.237779618889809,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003630778948861434,
      "loss": 4.9239,
      "mean_token_accuracy": 0.1933456376194954,
      "num_tokens": 9319849.0,
      "step": 5115
    },
    {
      "entropy": 5.307425880432129,
      "epoch": 4.2419221209610605,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003627837517169836,
      "loss": 4.9442,
      "mean_token_accuracy": 0.19091933965682983,
      "num_tokens": 9328355.0,
      "step": 5120
    },
    {
      "entropy": 5.376434850692749,
      "epoch": 4.246064623032312,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003624894314784037,
      "loss": 4.9141,
      "mean_token_accuracy": 0.1940031588077545,
      "num_tokens": 9337562.0,
      "step": 5125
    },
    {
      "entropy": 5.3480544090271,
      "epoch": 4.250207125103563,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003621949347640797,
      "loss": 4.9244,
      "mean_token_accuracy": 0.19829927235841752,
      "num_tokens": 9347851.0,
      "step": 5130
    },
    {
      "entropy": 5.402528762817383,
      "epoch": 4.254349627174814,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003619002621680439,
      "loss": 4.9332,
      "mean_token_accuracy": 0.19501807540655136,
      "num_tokens": 9356524.0,
      "step": 5135
    },
    {
      "entropy": 5.439273500442505,
      "epoch": 4.258492129246065,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003616054142846829,
      "loss": 5.0313,
      "mean_token_accuracy": 0.18848602771759032,
      "num_tokens": 9365521.0,
      "step": 5140
    },
    {
      "entropy": 5.269341659545899,
      "epoch": 4.262634631317316,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003613103917087371,
      "loss": 4.9373,
      "mean_token_accuracy": 0.1939264342188835,
      "num_tokens": 9374088.0,
      "step": 5145
    },
    {
      "entropy": 5.370335865020752,
      "epoch": 4.266777133388567,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003610151950352994,
      "loss": 4.9956,
      "mean_token_accuracy": 0.18929741978645326,
      "num_tokens": 9382683.0,
      "step": 5150
    },
    {
      "entropy": 5.333281707763672,
      "epoch": 4.270919635459817,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003607198248598137,
      "loss": 4.8809,
      "mean_token_accuracy": 0.20325784683227538,
      "num_tokens": 9391790.0,
      "step": 5155
    },
    {
      "entropy": 5.396844434738159,
      "epoch": 4.2750621375310685,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003604242817780739,
      "loss": 5.0043,
      "mean_token_accuracy": 0.19161476641893388,
      "num_tokens": 9400234.0,
      "step": 5160
    },
    {
      "entropy": 5.391296863555908,
      "epoch": 4.27920463960232,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036012856638622273,
      "loss": 4.9016,
      "mean_token_accuracy": 0.1907774806022644,
      "num_tokens": 9409513.0,
      "step": 5165
    },
    {
      "entropy": 5.337613248825074,
      "epoch": 4.283347141673571,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003598326792807504,
      "loss": 4.973,
      "mean_token_accuracy": 0.18558882176876068,
      "num_tokens": 9418608.0,
      "step": 5170
    },
    {
      "entropy": 5.444794988632202,
      "epoch": 4.287489643744822,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035953662105849357,
      "loss": 5.0223,
      "mean_token_accuracy": 0.19059320241212846,
      "num_tokens": 9428705.0,
      "step": 5175
    },
    {
      "entropy": 5.338778829574585,
      "epoch": 4.291632145816073,
      "grad_norm": 1.171875,
      "learning_rate": 0.000359240392316634,
      "loss": 4.9161,
      "mean_token_accuracy": 0.19613841623067857,
      "num_tokens": 9438149.0,
      "step": 5180
    },
    {
      "entropy": 5.3398984432220455,
      "epoch": 4.295774647887324,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003589439936526974,
      "loss": 4.8841,
      "mean_token_accuracy": 0.20296368300914763,
      "num_tokens": 9447566.0,
      "step": 5185
    },
    {
      "entropy": 5.279547691345215,
      "epoch": 4.299917149958575,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003586474256645525,
      "loss": 4.9566,
      "mean_token_accuracy": 0.19683022946119308,
      "num_tokens": 9458156.0,
      "step": 5190
    },
    {
      "entropy": 5.435497283935547,
      "epoch": 4.304059652029826,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035835068895040905,
      "loss": 5.01,
      "mean_token_accuracy": 0.19173871725797653,
      "num_tokens": 9467388.0,
      "step": 5195
    },
    {
      "entropy": 5.402412366867066,
      "epoch": 4.308202154101077,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035805378410881753,
      "loss": 4.9593,
      "mean_token_accuracy": 0.19687935560941697,
      "num_tokens": 9477515.0,
      "step": 5200
    },
    {
      "entropy": 5.358197641372681,
      "epoch": 4.312344656172328,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003577567117386675,
      "loss": 4.9786,
      "mean_token_accuracy": 0.19595407396554948,
      "num_tokens": 9485851.0,
      "step": 5205
    },
    {
      "entropy": 5.402158927917481,
      "epoch": 4.3164871582435795,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003574594724391863,
      "loss": 4.9802,
      "mean_token_accuracy": 0.19147001206874847,
      "num_tokens": 9495523.0,
      "step": 5210
    },
    {
      "entropy": 5.380666208267212,
      "epoch": 4.32062966031483,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003571620668099382,
      "loss": 4.9261,
      "mean_token_accuracy": 0.1964591324329376,
      "num_tokens": 9504195.0,
      "step": 5215
    },
    {
      "entropy": 5.35273551940918,
      "epoch": 4.324772162386081,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003568644954508227,
      "loss": 4.9544,
      "mean_token_accuracy": 0.19556818753480912,
      "num_tokens": 9513313.0,
      "step": 5220
    },
    {
      "entropy": 5.370005512237549,
      "epoch": 4.328914664457332,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003565667589620739,
      "loss": 5.0236,
      "mean_token_accuracy": 0.18786981403827668,
      "num_tokens": 9522387.0,
      "step": 5225
    },
    {
      "entropy": 5.367159080505371,
      "epoch": 4.333057166528583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003562688579442588,
      "loss": 4.9667,
      "mean_token_accuracy": 0.1943439081311226,
      "num_tokens": 9531522.0,
      "step": 5230
    },
    {
      "entropy": 5.376822233200073,
      "epoch": 4.337199668599834,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035597079299827625,
      "loss": 4.9793,
      "mean_token_accuracy": 0.18973538130521775,
      "num_tokens": 9540248.0,
      "step": 5235
    },
    {
      "entropy": 5.339836883544922,
      "epoch": 4.341342170671085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035567256472535585,
      "loss": 4.9336,
      "mean_token_accuracy": 0.19714388251304626,
      "num_tokens": 9549850.0,
      "step": 5240
    },
    {
      "entropy": 5.300297641754151,
      "epoch": 4.345484672742336,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003553741737270566,
      "loss": 4.9106,
      "mean_token_accuracy": 0.19903547316789627,
      "num_tokens": 9558929.0,
      "step": 5245
    },
    {
      "entropy": 5.291800212860108,
      "epoch": 4.3496271748135875,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035507562060526574,
      "loss": 4.9226,
      "mean_token_accuracy": 0.19506746232509614,
      "num_tokens": 9567627.0,
      "step": 5250
    },
    {
      "entropy": 5.43364782333374,
      "epoch": 4.353769676884839,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035477690596219763,
      "loss": 5.0623,
      "mean_token_accuracy": 0.18306090235710143,
      "num_tokens": 9576223.0,
      "step": 5255
    },
    {
      "entropy": 5.403663158416748,
      "epoch": 4.35791217895609,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035447803040039225,
      "loss": 5.018,
      "mean_token_accuracy": 0.1906593084335327,
      "num_tokens": 9585510.0,
      "step": 5260
    },
    {
      "entropy": 5.344839954376221,
      "epoch": 4.362054681027341,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003541789945227143,
      "loss": 4.9573,
      "mean_token_accuracy": 0.19684511572122573,
      "num_tokens": 9594843.0,
      "step": 5265
    },
    {
      "entropy": 5.388115930557251,
      "epoch": 4.366197183098592,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035387979893235184,
      "loss": 4.9438,
      "mean_token_accuracy": 0.19876011610031127,
      "num_tokens": 9604142.0,
      "step": 5270
    },
    {
      "entropy": 5.337083435058593,
      "epoch": 4.370339685169842,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003535804442328151,
      "loss": 5.0047,
      "mean_token_accuracy": 0.19857123345136643,
      "num_tokens": 9613669.0,
      "step": 5275
    },
    {
      "entropy": 5.431122589111328,
      "epoch": 4.374482187241093,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035328093102793506,
      "loss": 4.9803,
      "mean_token_accuracy": 0.192887544631958,
      "num_tokens": 9623502.0,
      "step": 5280
    },
    {
      "entropy": 5.4070333480834964,
      "epoch": 4.378624689312344,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035298125992186283,
      "loss": 4.9683,
      "mean_token_accuracy": 0.19389191567897796,
      "num_tokens": 9632041.0,
      "step": 5285
    },
    {
      "entropy": 5.3253696918487545,
      "epoch": 4.3827671913835955,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035268143151906764,
      "loss": 4.9906,
      "mean_token_accuracy": 0.19152809381484986,
      "num_tokens": 9641442.0,
      "step": 5290
    },
    {
      "entropy": 5.318687772750854,
      "epoch": 4.386909693454847,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035238144642433616,
      "loss": 4.9602,
      "mean_token_accuracy": 0.18918924331665038,
      "num_tokens": 9650857.0,
      "step": 5295
    },
    {
      "entropy": 5.494874811172485,
      "epoch": 4.391052195526098,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000352081305242771,
      "loss": 4.9958,
      "mean_token_accuracy": 0.198708176612854,
      "num_tokens": 9659173.0,
      "step": 5300
    },
    {
      "entropy": 5.333396673202515,
      "epoch": 4.395194697597349,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035178100857978994,
      "loss": 4.9176,
      "mean_token_accuracy": 0.19724688678979874,
      "num_tokens": 9668114.0,
      "step": 5305
    },
    {
      "entropy": 5.290274143218994,
      "epoch": 4.3993371996686,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035148055704112405,
      "loss": 4.9951,
      "mean_token_accuracy": 0.19266073405742645,
      "num_tokens": 9676628.0,
      "step": 5310
    },
    {
      "entropy": 5.300523328781128,
      "epoch": 4.403479701739851,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003511799512328168,
      "loss": 4.897,
      "mean_token_accuracy": 0.20140472203493118,
      "num_tokens": 9685801.0,
      "step": 5315
    },
    {
      "entropy": 5.341515064239502,
      "epoch": 4.407622203811102,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003508791917612232,
      "loss": 4.9751,
      "mean_token_accuracy": 0.1933399498462677,
      "num_tokens": 9694237.0,
      "step": 5320
    },
    {
      "entropy": 5.40983738899231,
      "epoch": 4.411764705882353,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035057827923300785,
      "loss": 4.8989,
      "mean_token_accuracy": 0.20040225237607956,
      "num_tokens": 9701904.0,
      "step": 5325
    },
    {
      "entropy": 5.3061418533325195,
      "epoch": 4.415907207953604,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035027721425514414,
      "loss": 4.8787,
      "mean_token_accuracy": 0.19985245913267136,
      "num_tokens": 9710784.0,
      "step": 5330
    },
    {
      "entropy": 5.287195825576783,
      "epoch": 4.4200497100248555,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003499759974349131,
      "loss": 4.9982,
      "mean_token_accuracy": 0.1949624612927437,
      "num_tokens": 9720452.0,
      "step": 5335
    },
    {
      "entropy": 5.354932880401611,
      "epoch": 4.424192212096106,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034967462937990194,
      "loss": 4.9701,
      "mean_token_accuracy": 0.18557771146297455,
      "num_tokens": 9729122.0,
      "step": 5340
    },
    {
      "entropy": 5.4259108543396,
      "epoch": 4.428334714167357,
      "grad_norm": 1.234375,
      "learning_rate": 0.000349373110698003,
      "loss": 5.02,
      "mean_token_accuracy": 0.1858140230178833,
      "num_tokens": 9738756.0,
      "step": 5345
    },
    {
      "entropy": 5.407877731323242,
      "epoch": 4.432477216238608,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003490714419974124,
      "loss": 4.9963,
      "mean_token_accuracy": 0.19534752666950225,
      "num_tokens": 9748547.0,
      "step": 5350
    },
    {
      "entropy": 5.372826719284058,
      "epoch": 4.436619718309859,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034876962388662885,
      "loss": 5.0508,
      "mean_token_accuracy": 0.18620429635047914,
      "num_tokens": 9758421.0,
      "step": 5355
    },
    {
      "entropy": 5.285065364837647,
      "epoch": 4.44076222038111,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003484676569744525,
      "loss": 4.8726,
      "mean_token_accuracy": 0.1962450250983238,
      "num_tokens": 9766855.0,
      "step": 5360
    },
    {
      "entropy": 5.314684295654297,
      "epoch": 4.444904722452361,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003481655418699836,
      "loss": 4.9712,
      "mean_token_accuracy": 0.1923555999994278,
      "num_tokens": 9776157.0,
      "step": 5365
    },
    {
      "entropy": 5.315577602386474,
      "epoch": 4.449047224523612,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034786327918262126,
      "loss": 4.8627,
      "mean_token_accuracy": 0.20385012179613113,
      "num_tokens": 9784386.0,
      "step": 5370
    },
    {
      "entropy": 5.351845407485962,
      "epoch": 4.4531897265948635,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034756086952206255,
      "loss": 4.9088,
      "mean_token_accuracy": 0.20008092224597931,
      "num_tokens": 9793838.0,
      "step": 5375
    },
    {
      "entropy": 5.381402921676636,
      "epoch": 4.457332228666115,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003472583134983005,
      "loss": 5.0079,
      "mean_token_accuracy": 0.19048677235841752,
      "num_tokens": 9802595.0,
      "step": 5380
    },
    {
      "entropy": 5.415020751953125,
      "epoch": 4.461474730737366,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034695561172162396,
      "loss": 4.9737,
      "mean_token_accuracy": 0.1940239518880844,
      "num_tokens": 9811309.0,
      "step": 5385
    },
    {
      "entropy": 5.285571432113647,
      "epoch": 4.465617232808617,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034665276480261526,
      "loss": 4.9379,
      "mean_token_accuracy": 0.19441236257553102,
      "num_tokens": 9820327.0,
      "step": 5390
    },
    {
      "entropy": 5.387330102920532,
      "epoch": 4.469759734879867,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003463497733521499,
      "loss": 4.9973,
      "mean_token_accuracy": 0.18807628452777864,
      "num_tokens": 9829588.0,
      "step": 5395
    },
    {
      "entropy": 5.450295066833496,
      "epoch": 4.473902236951118,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003460466379813947,
      "loss": 4.9936,
      "mean_token_accuracy": 0.18826743662357331,
      "num_tokens": 9838548.0,
      "step": 5400
    },
    {
      "entropy": 5.344222640991211,
      "epoch": 4.478044739022369,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003457433593018068,
      "loss": 5.0055,
      "mean_token_accuracy": 0.19389185756444932,
      "num_tokens": 9847925.0,
      "step": 5405
    },
    {
      "entropy": 5.375957345962524,
      "epoch": 4.48218724109362,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034543993792513243,
      "loss": 5.0326,
      "mean_token_accuracy": 0.18545857816934586,
      "num_tokens": 9856839.0,
      "step": 5410
    },
    {
      "entropy": 5.400127506256103,
      "epoch": 4.486329743164871,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003451363744634057,
      "loss": 4.9855,
      "mean_token_accuracy": 0.19444191157817842,
      "num_tokens": 9866519.0,
      "step": 5415
    },
    {
      "entropy": 5.405975961685181,
      "epoch": 4.4904722452361225,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034483266952894725,
      "loss": 5.0652,
      "mean_token_accuracy": 0.19181954264640808,
      "num_tokens": 9876866.0,
      "step": 5420
    },
    {
      "entropy": 5.462331819534302,
      "epoch": 4.494614747307374,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003445288237343632,
      "loss": 4.9846,
      "mean_token_accuracy": 0.1948983520269394,
      "num_tokens": 9884639.0,
      "step": 5425
    },
    {
      "entropy": 5.33106427192688,
      "epoch": 4.498757249378625,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003442248376925436,
      "loss": 4.9777,
      "mean_token_accuracy": 0.1957646831870079,
      "num_tokens": 9893819.0,
      "step": 5430
    },
    {
      "entropy": 5.2371148586273195,
      "epoch": 4.502899751449876,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034392071201666167,
      "loss": 4.9265,
      "mean_token_accuracy": 0.19253770560026168,
      "num_tokens": 9902419.0,
      "step": 5435
    },
    {
      "entropy": 5.392510509490966,
      "epoch": 4.507042253521127,
      "grad_norm": 1.109375,
      "learning_rate": 0.000343616447320172,
      "loss": 4.961,
      "mean_token_accuracy": 0.19620949327945708,
      "num_tokens": 9912434.0,
      "step": 5440
    },
    {
      "entropy": 5.389869260787964,
      "epoch": 4.511184755592378,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003433120442168097,
      "loss": 4.9099,
      "mean_token_accuracy": 0.20106928199529647,
      "num_tokens": 9921800.0,
      "step": 5445
    },
    {
      "entropy": 5.3106873512268065,
      "epoch": 4.515327257663629,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034300750332058924,
      "loss": 4.9893,
      "mean_token_accuracy": 0.1911606326699257,
      "num_tokens": 9930863.0,
      "step": 5450
    },
    {
      "entropy": 5.3417304992675785,
      "epoch": 4.51946975973488,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003427028252458028,
      "loss": 4.9967,
      "mean_token_accuracy": 0.19518981128931046,
      "num_tokens": 9940524.0,
      "step": 5455
    },
    {
      "entropy": 5.434265804290772,
      "epoch": 4.5236122618061305,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003423980106070194,
      "loss": 4.991,
      "mean_token_accuracy": 0.19182531535625458,
      "num_tokens": 9948949.0,
      "step": 5460
    },
    {
      "entropy": 5.366151189804077,
      "epoch": 4.527754763877382,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003420930600190834,
      "loss": 5.0001,
      "mean_token_accuracy": 0.19315124303102493,
      "num_tokens": 9958154.0,
      "step": 5465
    },
    {
      "entropy": 5.325886583328247,
      "epoch": 4.531897265948633,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003417879740971135,
      "loss": 4.9023,
      "mean_token_accuracy": 0.20153665095567702,
      "num_tokens": 9967023.0,
      "step": 5470
    },
    {
      "entropy": 5.410694360733032,
      "epoch": 4.536039768019884,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003414827534565014,
      "loss": 5.0019,
      "mean_token_accuracy": 0.19251520484685897,
      "num_tokens": 9976596.0,
      "step": 5475
    },
    {
      "entropy": 5.373929882049561,
      "epoch": 4.540182270091135,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034117739871291036,
      "loss": 4.9024,
      "mean_token_accuracy": 0.19974904358386994,
      "num_tokens": 9985330.0,
      "step": 5480
    },
    {
      "entropy": 5.342440366744995,
      "epoch": 4.544324772162386,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034087191048227445,
      "loss": 5.0139,
      "mean_token_accuracy": 0.19303976744413376,
      "num_tokens": 9996053.0,
      "step": 5485
    },
    {
      "entropy": 5.326486015319825,
      "epoch": 4.548467274233637,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003405662893807967,
      "loss": 4.9382,
      "mean_token_accuracy": 0.19153032153844834,
      "num_tokens": 10005517.0,
      "step": 5490
    },
    {
      "entropy": 5.385126543045044,
      "epoch": 4.552609776304888,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003402605360249482,
      "loss": 5.053,
      "mean_token_accuracy": 0.1852627784013748,
      "num_tokens": 10014354.0,
      "step": 5495
    },
    {
      "entropy": 5.326625633239746,
      "epoch": 4.556752278376139,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000339954651031467,
      "loss": 4.9381,
      "mean_token_accuracy": 0.19841257482767105,
      "num_tokens": 10023194.0,
      "step": 5500
    },
    {
      "epoch": 4.556752278376139,
      "eval_entropy": 5.222419834340305,
      "eval_loss": 5.8305840492248535,
      "eval_mean_token_accuracy": 0.15926696668301565,
      "eval_num_tokens": 10023194.0,
      "eval_runtime": 2.9539,
      "eval_samples_per_second": 1268.479,
      "eval_steps_per_second": 158.771,
      "step": 5500
    },
    {
      "entropy": 5.286475038528442,
      "epoch": 4.5608947804473905,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033964863501735647,
      "loss": 4.9338,
      "mean_token_accuracy": 0.1966310992836952,
      "num_tokens": 10032490.0,
      "step": 5505
    },
    {
      "entropy": 5.380674552917481,
      "epoch": 4.565037282518642,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033934248859988453,
      "loss": 4.9362,
      "mean_token_accuracy": 0.19466858208179474,
      "num_tokens": 10041613.0,
      "step": 5510
    },
    {
      "entropy": 5.289149475097656,
      "epoch": 4.569179784589892,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003390362123965817,
      "loss": 4.8937,
      "mean_token_accuracy": 0.19557223618030548,
      "num_tokens": 10050500.0,
      "step": 5515
    },
    {
      "entropy": 5.351538467407226,
      "epoch": 4.573322286661144,
      "grad_norm": 1.71875,
      "learning_rate": 0.0003387298070252407,
      "loss": 4.9404,
      "mean_token_accuracy": 0.19423381239175797,
      "num_tokens": 10059650.0,
      "step": 5520
    },
    {
      "entropy": 5.346185398101807,
      "epoch": 4.577464788732394,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003384232731039146,
      "loss": 4.9976,
      "mean_token_accuracy": 0.19463732093572617,
      "num_tokens": 10067882.0,
      "step": 5525
    },
    {
      "entropy": 5.283247089385986,
      "epoch": 4.581607290803645,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033811661125091584,
      "loss": 4.9269,
      "mean_token_accuracy": 0.19635584950447083,
      "num_tokens": 10077147.0,
      "step": 5530
    },
    {
      "entropy": 5.384209012985229,
      "epoch": 4.585749792874896,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033780982208481494,
      "loss": 4.9846,
      "mean_token_accuracy": 0.19521064311265945,
      "num_tokens": 10086149.0,
      "step": 5535
    },
    {
      "entropy": 5.307389402389527,
      "epoch": 4.589892294946147,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003375029062244392,
      "loss": 4.9092,
      "mean_token_accuracy": 0.20299234986305237,
      "num_tokens": 10094997.0,
      "step": 5540
    },
    {
      "entropy": 5.4333690166473385,
      "epoch": 4.5940347970173985,
      "grad_norm": 1.25,
      "learning_rate": 0.00033719586428887125,
      "loss": 4.9704,
      "mean_token_accuracy": 0.19187625497579575,
      "num_tokens": 10103585.0,
      "step": 5545
    },
    {
      "entropy": 5.3080726146698,
      "epoch": 4.59817729908865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033688869689744836,
      "loss": 4.8769,
      "mean_token_accuracy": 0.19785331487655639,
      "num_tokens": 10113877.0,
      "step": 5550
    },
    {
      "entropy": 5.329131698608398,
      "epoch": 4.602319801159901,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003365814046697608,
      "loss": 4.9829,
      "mean_token_accuracy": 0.18659879863262177,
      "num_tokens": 10123127.0,
      "step": 5555
    },
    {
      "entropy": 5.4331317901611325,
      "epoch": 4.606462303231152,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003362739882256504,
      "loss": 5.0533,
      "mean_token_accuracy": 0.1873202443122864,
      "num_tokens": 10132301.0,
      "step": 5560
    },
    {
      "entropy": 5.376758003234864,
      "epoch": 4.610604805302403,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003359664481852099,
      "loss": 5.0392,
      "mean_token_accuracy": 0.1848893031477928,
      "num_tokens": 10140985.0,
      "step": 5565
    },
    {
      "entropy": 5.3239668846130375,
      "epoch": 4.614747307373654,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033565878516878103,
      "loss": 4.9471,
      "mean_token_accuracy": 0.18900113552808762,
      "num_tokens": 10149668.0,
      "step": 5570
    },
    {
      "entropy": 5.349096965789795,
      "epoch": 4.618889809444905,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003353509997969537,
      "loss": 5.0037,
      "mean_token_accuracy": 0.19154171645641327,
      "num_tokens": 10158584.0,
      "step": 5575
    },
    {
      "entropy": 5.346112823486328,
      "epoch": 4.623032311516155,
      "grad_norm": 1.25,
      "learning_rate": 0.00033504309269056484,
      "loss": 4.9647,
      "mean_token_accuracy": 0.19203616976737975,
      "num_tokens": 10167657.0,
      "step": 5580
    },
    {
      "entropy": 5.267374849319458,
      "epoch": 4.6271748135874065,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003347350644706965,
      "loss": 4.8626,
      "mean_token_accuracy": 0.19540804028511047,
      "num_tokens": 10176657.0,
      "step": 5585
    },
    {
      "entropy": 5.372695875167847,
      "epoch": 4.631317315658658,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003344269157586754,
      "loss": 4.9731,
      "mean_token_accuracy": 0.2009139209985733,
      "num_tokens": 10185132.0,
      "step": 5590
    },
    {
      "entropy": 5.330093955993652,
      "epoch": 4.635459817729909,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033411864717607114,
      "loss": 4.8689,
      "mean_token_accuracy": 0.1938499853014946,
      "num_tokens": 10193428.0,
      "step": 5595
    },
    {
      "entropy": 5.328604459762573,
      "epoch": 4.63960231980116,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003338102593446951,
      "loss": 5.0322,
      "mean_token_accuracy": 0.18878376632928848,
      "num_tokens": 10202549.0,
      "step": 5600
    },
    {
      "entropy": 5.322787618637085,
      "epoch": 4.643744821872411,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033350175288659925,
      "loss": 4.8983,
      "mean_token_accuracy": 0.20287970155477525,
      "num_tokens": 10211161.0,
      "step": 5605
    },
    {
      "entropy": 5.347150087356567,
      "epoch": 4.647887323943662,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033319312842407483,
      "loss": 4.9845,
      "mean_token_accuracy": 0.18790417164564133,
      "num_tokens": 10220988.0,
      "step": 5610
    },
    {
      "entropy": 5.321266555786133,
      "epoch": 4.652029826014913,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003328843865796511,
      "loss": 4.8945,
      "mean_token_accuracy": 0.2011766642332077,
      "num_tokens": 10229383.0,
      "step": 5615
    },
    {
      "entropy": 5.443414688110352,
      "epoch": 4.656172328086164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033257552797609426,
      "loss": 4.9718,
      "mean_token_accuracy": 0.1887246623635292,
      "num_tokens": 10238622.0,
      "step": 5620
    },
    {
      "entropy": 5.507680892944336,
      "epoch": 4.660314830157415,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033226655323640553,
      "loss": 5.0702,
      "mean_token_accuracy": 0.19291428625583648,
      "num_tokens": 10248201.0,
      "step": 5625
    },
    {
      "entropy": 5.415904998779297,
      "epoch": 4.664457332228666,
      "grad_norm": 1.234375,
      "learning_rate": 0.000331957462983821,
      "loss": 5.0824,
      "mean_token_accuracy": 0.18392113000154495,
      "num_tokens": 10258239.0,
      "step": 5630
    },
    {
      "entropy": 5.355754089355469,
      "epoch": 4.668599834299917,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003316482578418094,
      "loss": 4.974,
      "mean_token_accuracy": 0.19172132462263108,
      "num_tokens": 10266915.0,
      "step": 5635
    },
    {
      "entropy": 5.390386533737183,
      "epoch": 4.672742336371169,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033133893843407126,
      "loss": 5.0978,
      "mean_token_accuracy": 0.18882705122232438,
      "num_tokens": 10276826.0,
      "step": 5640
    },
    {
      "entropy": 5.372137832641601,
      "epoch": 4.676884838442419,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033102950538453776,
      "loss": 5.0116,
      "mean_token_accuracy": 0.1879737451672554,
      "num_tokens": 10286041.0,
      "step": 5645
    },
    {
      "entropy": 5.400428771972656,
      "epoch": 4.68102734051367,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033071995931736897,
      "loss": 4.999,
      "mean_token_accuracy": 0.19514506608247756,
      "num_tokens": 10295258.0,
      "step": 5650
    },
    {
      "entropy": 5.356552267074585,
      "epoch": 4.685169842584921,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033041030085695323,
      "loss": 4.954,
      "mean_token_accuracy": 0.19490123093128203,
      "num_tokens": 10304726.0,
      "step": 5655
    },
    {
      "entropy": 5.338892889022827,
      "epoch": 4.689312344656172,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003301005306279055,
      "loss": 4.8983,
      "mean_token_accuracy": 0.20012467503547668,
      "num_tokens": 10312912.0,
      "step": 5660
    },
    {
      "entropy": 5.327005672454834,
      "epoch": 4.693454846727423,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032979064925506627,
      "loss": 4.9813,
      "mean_token_accuracy": 0.19570244252681732,
      "num_tokens": 10322046.0,
      "step": 5665
    },
    {
      "entropy": 5.324332284927368,
      "epoch": 4.697597348798674,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003294806573634999,
      "loss": 4.9375,
      "mean_token_accuracy": 0.2052298977971077,
      "num_tokens": 10331375.0,
      "step": 5670
    },
    {
      "entropy": 5.3730730533599855,
      "epoch": 4.7017398508699255,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003291705555784942,
      "loss": 5.0604,
      "mean_token_accuracy": 0.18693147450685502,
      "num_tokens": 10341465.0,
      "step": 5675
    },
    {
      "entropy": 5.467976140975952,
      "epoch": 4.705882352941177,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032886034452555814,
      "loss": 5.0767,
      "mean_token_accuracy": 0.18940733522176742,
      "num_tokens": 10350137.0,
      "step": 5680
    },
    {
      "entropy": 5.286528635025024,
      "epoch": 4.710024855012428,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003285500248304214,
      "loss": 4.9544,
      "mean_token_accuracy": 0.19064881652593613,
      "num_tokens": 10358666.0,
      "step": 5685
    },
    {
      "entropy": 5.301413583755493,
      "epoch": 4.714167357083679,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003282395971190328,
      "loss": 4.8422,
      "mean_token_accuracy": 0.20117190331220627,
      "num_tokens": 10366910.0,
      "step": 5690
    },
    {
      "entropy": 5.368517351150513,
      "epoch": 4.71830985915493,
      "grad_norm": 1.25,
      "learning_rate": 0.00032792906201755876,
      "loss": 5.0104,
      "mean_token_accuracy": 0.18896820694208144,
      "num_tokens": 10376435.0,
      "step": 5695
    },
    {
      "entropy": 5.301347875595093,
      "epoch": 4.72245236122618,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032761842015238256,
      "loss": 4.9189,
      "mean_token_accuracy": 0.20234665870666504,
      "num_tokens": 10384670.0,
      "step": 5700
    },
    {
      "entropy": 5.298396492004395,
      "epoch": 4.726594863297431,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003273076721501029,
      "loss": 4.901,
      "mean_token_accuracy": 0.19903917312622071,
      "num_tokens": 10393731.0,
      "step": 5705
    },
    {
      "entropy": 5.384547901153565,
      "epoch": 4.730737365368682,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003269968186375325,
      "loss": 5.002,
      "mean_token_accuracy": 0.19888944923877716,
      "num_tokens": 10402369.0,
      "step": 5710
    },
    {
      "entropy": 5.388984727859497,
      "epoch": 4.7348798674399335,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003266858602416967,
      "loss": 4.9758,
      "mean_token_accuracy": 0.1922182857990265,
      "num_tokens": 10411566.0,
      "step": 5715
    },
    {
      "entropy": 5.378957891464234,
      "epoch": 4.739022369511185,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032637479758983255,
      "loss": 5.035,
      "mean_token_accuracy": 0.18553909808397293,
      "num_tokens": 10420086.0,
      "step": 5720
    },
    {
      "entropy": 5.398101568222046,
      "epoch": 4.743164871582436,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032606363130938766,
      "loss": 5.0889,
      "mean_token_accuracy": 0.1900040939450264,
      "num_tokens": 10429746.0,
      "step": 5725
    },
    {
      "entropy": 5.369560623168946,
      "epoch": 4.747307373653687,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003257523620280181,
      "loss": 4.9626,
      "mean_token_accuracy": 0.19613697081804277,
      "num_tokens": 10438536.0,
      "step": 5730
    },
    {
      "entropy": 5.331012964248657,
      "epoch": 4.751449875724938,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032544099037358834,
      "loss": 4.9618,
      "mean_token_accuracy": 0.1978169471025467,
      "num_tokens": 10447471.0,
      "step": 5735
    },
    {
      "entropy": 5.372948026657104,
      "epoch": 4.755592377796189,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003251295169741688,
      "loss": 4.9748,
      "mean_token_accuracy": 0.20283176749944687,
      "num_tokens": 10457179.0,
      "step": 5740
    },
    {
      "entropy": 5.341614770889282,
      "epoch": 4.75973487986744,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032481794245803536,
      "loss": 5.0326,
      "mean_token_accuracy": 0.18564868420362474,
      "num_tokens": 10466291.0,
      "step": 5745
    },
    {
      "entropy": 5.323867321014404,
      "epoch": 4.763877381938691,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032450626745366806,
      "loss": 4.919,
      "mean_token_accuracy": 0.19624859541654588,
      "num_tokens": 10475452.0,
      "step": 5750
    },
    {
      "entropy": 5.311866760253906,
      "epoch": 4.768019884009942,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032419449258974924,
      "loss": 4.8934,
      "mean_token_accuracy": 0.20149471908807753,
      "num_tokens": 10484381.0,
      "step": 5755
    },
    {
      "entropy": 5.341735172271728,
      "epoch": 4.7721623860811935,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000323882618495163,
      "loss": 5.0851,
      "mean_token_accuracy": 0.1856486454606056,
      "num_tokens": 10495132.0,
      "step": 5760
    },
    {
      "entropy": 5.383388042449951,
      "epoch": 4.776304888152444,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003235706457989934,
      "loss": 5.023,
      "mean_token_accuracy": 0.18799879997968674,
      "num_tokens": 10504519.0,
      "step": 5765
    },
    {
      "entropy": 5.411660814285279,
      "epoch": 4.780447390223695,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032325857513052354,
      "loss": 4.9842,
      "mean_token_accuracy": 0.1945323258638382,
      "num_tokens": 10513575.0,
      "step": 5770
    },
    {
      "entropy": 5.416269826889038,
      "epoch": 4.784589892294946,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000322946407119234,
      "loss": 5.07,
      "mean_token_accuracy": 0.1892155572772026,
      "num_tokens": 10522148.0,
      "step": 5775
    },
    {
      "entropy": 5.3567101001739506,
      "epoch": 4.788732394366197,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032263414239480187,
      "loss": 4.989,
      "mean_token_accuracy": 0.1918441355228424,
      "num_tokens": 10531183.0,
      "step": 5780
    },
    {
      "entropy": 5.3522124767303465,
      "epoch": 4.792874896437448,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000322321781587099,
      "loss": 4.9343,
      "mean_token_accuracy": 0.1961425006389618,
      "num_tokens": 10540677.0,
      "step": 5785
    },
    {
      "entropy": 5.288934373855591,
      "epoch": 4.797017398508699,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003220093253261915,
      "loss": 4.9364,
      "mean_token_accuracy": 0.1887398898601532,
      "num_tokens": 10550009.0,
      "step": 5790
    },
    {
      "entropy": 5.345226097106933,
      "epoch": 4.80115990057995,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003216967742423377,
      "loss": 4.9516,
      "mean_token_accuracy": 0.19715719372034074,
      "num_tokens": 10559694.0,
      "step": 5795
    },
    {
      "entropy": 5.350609397888183,
      "epoch": 4.8053024026512015,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003213841289659872,
      "loss": 4.9777,
      "mean_token_accuracy": 0.1989009439945221,
      "num_tokens": 10567965.0,
      "step": 5800
    },
    {
      "entropy": 5.34626955986023,
      "epoch": 4.809444904722453,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003210713901277799,
      "loss": 4.963,
      "mean_token_accuracy": 0.19924664199352266,
      "num_tokens": 10577653.0,
      "step": 5805
    },
    {
      "entropy": 5.389357376098633,
      "epoch": 4.813587406793704,
      "grad_norm": 1.359375,
      "learning_rate": 0.000320758558358544,
      "loss": 4.9724,
      "mean_token_accuracy": 0.18880909979343413,
      "num_tokens": 10586903.0,
      "step": 5810
    },
    {
      "entropy": 5.329872035980225,
      "epoch": 4.817729908864955,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003204456342892955,
      "loss": 4.8922,
      "mean_token_accuracy": 0.19671850502490998,
      "num_tokens": 10595411.0,
      "step": 5815
    },
    {
      "entropy": 5.276799440383911,
      "epoch": 4.821872410936205,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032013261855123637,
      "loss": 4.9663,
      "mean_token_accuracy": 0.19169727563858033,
      "num_tokens": 10604147.0,
      "step": 5820
    },
    {
      "entropy": 5.370803022384644,
      "epoch": 4.826014913007457,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031981951177575366,
      "loss": 5.0421,
      "mean_token_accuracy": 0.19067293256521226,
      "num_tokens": 10614048.0,
      "step": 5825
    },
    {
      "entropy": 5.419658946990967,
      "epoch": 4.830157415078707,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003195063145944178,
      "loss": 5.0741,
      "mean_token_accuracy": 0.18566471636295317,
      "num_tokens": 10623399.0,
      "step": 5830
    },
    {
      "entropy": 5.3768425464630125,
      "epoch": 4.834299917149958,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003191930276389819,
      "loss": 4.91,
      "mean_token_accuracy": 0.19876255840063095,
      "num_tokens": 10632627.0,
      "step": 5835
    },
    {
      "entropy": 5.386160945892334,
      "epoch": 4.838442419221209,
      "grad_norm": 1.375,
      "learning_rate": 0.00031887965154137987,
      "loss": 4.9771,
      "mean_token_accuracy": 0.18994825184345246,
      "num_tokens": 10641155.0,
      "step": 5840
    },
    {
      "entropy": 5.340151739120484,
      "epoch": 4.8425849212924605,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003185661869337255,
      "loss": 5.0123,
      "mean_token_accuracy": 0.18744892179965972,
      "num_tokens": 10649661.0,
      "step": 5845
    },
    {
      "entropy": 5.295511531829834,
      "epoch": 4.846727423363712,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031825263444831137,
      "loss": 4.9226,
      "mean_token_accuracy": 0.19581829458475114,
      "num_tokens": 10658652.0,
      "step": 5850
    },
    {
      "entropy": 5.382693147659301,
      "epoch": 4.850869925434963,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003179389947176069,
      "loss": 5.0515,
      "mean_token_accuracy": 0.18901979625225068,
      "num_tokens": 10668703.0,
      "step": 5855
    },
    {
      "entropy": 5.36964898109436,
      "epoch": 4.855012427506214,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003176252683742578,
      "loss": 4.9579,
      "mean_token_accuracy": 0.19061000347137452,
      "num_tokens": 10677740.0,
      "step": 5860
    },
    {
      "entropy": 5.337471008300781,
      "epoch": 4.859154929577465,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003173114560510843,
      "loss": 4.9537,
      "mean_token_accuracy": 0.19722201973199843,
      "num_tokens": 10686429.0,
      "step": 5865
    },
    {
      "entropy": 5.351599311828613,
      "epoch": 4.863297431648716,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031699755838108027,
      "loss": 5.0432,
      "mean_token_accuracy": 0.19890908151865005,
      "num_tokens": 10696367.0,
      "step": 5870
    },
    {
      "entropy": 5.381594562530518,
      "epoch": 4.867439933719967,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003166835759974115,
      "loss": 4.9987,
      "mean_token_accuracy": 0.18672978281974792,
      "num_tokens": 10705712.0,
      "step": 5875
    },
    {
      "entropy": 5.394096040725708,
      "epoch": 4.871582435791218,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031636950953341486,
      "loss": 5.0241,
      "mean_token_accuracy": 0.18677301853895187,
      "num_tokens": 10715897.0,
      "step": 5880
    },
    {
      "entropy": 5.313100910186767,
      "epoch": 4.8757249378624685,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031605535962259666,
      "loss": 4.9354,
      "mean_token_accuracy": 0.18848999291658403,
      "num_tokens": 10724810.0,
      "step": 5885
    },
    {
      "entropy": 5.342027711868286,
      "epoch": 4.87986743993372,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003157411268986316,
      "loss": 4.9836,
      "mean_token_accuracy": 0.19233897030353547,
      "num_tokens": 10734661.0,
      "step": 5890
    },
    {
      "entropy": 5.353427410125732,
      "epoch": 4.884009942004971,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003154268119953615,
      "loss": 4.95,
      "mean_token_accuracy": 0.19514879137277602,
      "num_tokens": 10743642.0,
      "step": 5895
    },
    {
      "entropy": 5.3696380138397215,
      "epoch": 4.888152444076222,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003151124155467938,
      "loss": 4.9897,
      "mean_token_accuracy": 0.19674668163061143,
      "num_tokens": 10752812.0,
      "step": 5900
    },
    {
      "entropy": 5.333204650878907,
      "epoch": 4.892294946147473,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031479793818710056,
      "loss": 4.9884,
      "mean_token_accuracy": 0.18838852792978286,
      "num_tokens": 10762117.0,
      "step": 5905
    },
    {
      "entropy": 5.306610202789306,
      "epoch": 4.896437448218724,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031448338055061685,
      "loss": 4.9152,
      "mean_token_accuracy": 0.2044752523303032,
      "num_tokens": 10771515.0,
      "step": 5910
    },
    {
      "entropy": 5.339256954193115,
      "epoch": 4.900579950289975,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031416874327183994,
      "loss": 4.9615,
      "mean_token_accuracy": 0.1933135971426964,
      "num_tokens": 10780610.0,
      "step": 5915
    },
    {
      "entropy": 5.225224161148072,
      "epoch": 4.904722452361226,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031385402698542753,
      "loss": 4.918,
      "mean_token_accuracy": 0.19905627518892288,
      "num_tokens": 10789169.0,
      "step": 5920
    },
    {
      "entropy": 5.300740957260132,
      "epoch": 4.908864954432477,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031353923232619674,
      "loss": 4.9047,
      "mean_token_accuracy": 0.19498729556798935,
      "num_tokens": 10798636.0,
      "step": 5925
    },
    {
      "entropy": 5.357951354980469,
      "epoch": 4.9130074565037285,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003132243599291229,
      "loss": 4.991,
      "mean_token_accuracy": 0.19608666002750397,
      "num_tokens": 10808245.0,
      "step": 5930
    },
    {
      "entropy": 5.3833336353302,
      "epoch": 4.91714995857498,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031290941042933797,
      "loss": 4.9742,
      "mean_token_accuracy": 0.19343886524438858,
      "num_tokens": 10817164.0,
      "step": 5935
    },
    {
      "entropy": 5.308176517486572,
      "epoch": 4.92129246064623,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031259438446212953,
      "loss": 5.0158,
      "mean_token_accuracy": 0.18689876794815063,
      "num_tokens": 10826098.0,
      "step": 5940
    },
    {
      "entropy": 5.347735929489136,
      "epoch": 4.925434962717482,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031227928266293933,
      "loss": 5.0169,
      "mean_token_accuracy": 0.1899582639336586,
      "num_tokens": 10836427.0,
      "step": 5945
    },
    {
      "entropy": 5.395557880401611,
      "epoch": 4.929577464788732,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031196410566736215,
      "loss": 4.9738,
      "mean_token_accuracy": 0.19258994311094285,
      "num_tokens": 10845804.0,
      "step": 5950
    },
    {
      "entropy": 5.3557103157043455,
      "epoch": 4.933719966859983,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031164885411114453,
      "loss": 4.9527,
      "mean_token_accuracy": 0.1957266166806221,
      "num_tokens": 10855951.0,
      "step": 5955
    },
    {
      "entropy": 5.284099769592285,
      "epoch": 4.937862468931234,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031133352863018315,
      "loss": 5.0743,
      "mean_token_accuracy": 0.18917616903781892,
      "num_tokens": 10865528.0,
      "step": 5960
    },
    {
      "entropy": 5.398881435394287,
      "epoch": 4.942004971002485,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031101812986052406,
      "loss": 4.9951,
      "mean_token_accuracy": 0.1987755611538887,
      "num_tokens": 10874365.0,
      "step": 5965
    },
    {
      "entropy": 5.314361667633056,
      "epoch": 4.9461474730737365,
      "grad_norm": 1.25,
      "learning_rate": 0.00031070265843836107,
      "loss": 4.9216,
      "mean_token_accuracy": 0.19912609159946443,
      "num_tokens": 10883034.0,
      "step": 5970
    },
    {
      "entropy": 5.342735147476196,
      "epoch": 4.950289975144988,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003103871150000345,
      "loss": 4.9958,
      "mean_token_accuracy": 0.19518498629331588,
      "num_tokens": 10892307.0,
      "step": 5975
    },
    {
      "entropy": 5.3277805805206295,
      "epoch": 4.954432477216239,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031007150018202983,
      "loss": 5.0392,
      "mean_token_accuracy": 0.18803122043609619,
      "num_tokens": 10902074.0,
      "step": 5980
    },
    {
      "entropy": 5.365257978439331,
      "epoch": 4.95857497928749,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030975581462097673,
      "loss": 4.9232,
      "mean_token_accuracy": 0.1970903068780899,
      "num_tokens": 10911205.0,
      "step": 5985
    },
    {
      "entropy": 5.298571586608887,
      "epoch": 4.962717481358741,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030944005895364754,
      "loss": 4.9334,
      "mean_token_accuracy": 0.19873082935810088,
      "num_tokens": 10920371.0,
      "step": 5990
    },
    {
      "entropy": 5.278933000564575,
      "epoch": 4.966859983429992,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030912423381695584,
      "loss": 4.9293,
      "mean_token_accuracy": 0.19943152219057084,
      "num_tokens": 10928741.0,
      "step": 5995
    },
    {
      "entropy": 5.31387186050415,
      "epoch": 4.971002485501243,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030880833984795554,
      "loss": 4.9252,
      "mean_token_accuracy": 0.1994773268699646,
      "num_tokens": 10938074.0,
      "step": 6000
    },
    {
      "epoch": 4.971002485501243,
      "eval_entropy": 5.249049451051237,
      "eval_loss": 5.795785903930664,
      "eval_mean_token_accuracy": 0.162466109640944,
      "eval_num_tokens": 10938074.0,
      "eval_runtime": 2.762,
      "eval_samples_per_second": 1356.61,
      "eval_steps_per_second": 169.803,
      "step": 6000
    },
    {
      "entropy": 5.319996023178101,
      "epoch": 4.975144987572493,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003084923776838392,
      "loss": 4.9573,
      "mean_token_accuracy": 0.19452079087495805,
      "num_tokens": 10946756.0,
      "step": 6005
    },
    {
      "entropy": 5.295461893081665,
      "epoch": 4.9792874896437445,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030817634796193705,
      "loss": 4.9213,
      "mean_token_accuracy": 0.19939590990543365,
      "num_tokens": 10956313.0,
      "step": 6010
    },
    {
      "entropy": 5.413240003585815,
      "epoch": 4.983429991714996,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030786025131971555,
      "loss": 4.9929,
      "mean_token_accuracy": 0.1928446188569069,
      "num_tokens": 10964333.0,
      "step": 6015
    },
    {
      "entropy": 5.326666784286499,
      "epoch": 4.987572493786247,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003075440883947764,
      "loss": 4.9187,
      "mean_token_accuracy": 0.194312447309494,
      "num_tokens": 10974110.0,
      "step": 6020
    },
    {
      "entropy": 5.270960998535156,
      "epoch": 4.991714995857498,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003072278598248544,
      "loss": 4.9636,
      "mean_token_accuracy": 0.19501755386590958,
      "num_tokens": 10983186.0,
      "step": 6025
    },
    {
      "entropy": 5.332121467590332,
      "epoch": 4.995857497928749,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030691156624781734,
      "loss": 4.9471,
      "mean_token_accuracy": 0.20051537007093428,
      "num_tokens": 10992244.0,
      "step": 6030
    },
    {
      "entropy": 5.365746736526489,
      "epoch": 5.0,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000306595208301664,
      "loss": 4.9322,
      "mean_token_accuracy": 0.20733886510133742,
      "num_tokens": 11000005.0,
      "step": 6035
    },
    {
      "entropy": 5.359089994430542,
      "epoch": 5.004142502071251,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030627878662452273,
      "loss": 4.7695,
      "mean_token_accuracy": 0.2050136774778366,
      "num_tokens": 11009346.0,
      "step": 6040
    },
    {
      "entropy": 5.370102453231811,
      "epoch": 5.008285004142502,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003059623018546508,
      "loss": 4.7044,
      "mean_token_accuracy": 0.21563132405281066,
      "num_tokens": 11017844.0,
      "step": 6045
    },
    {
      "entropy": 5.33039960861206,
      "epoch": 5.012427506213753,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003056457546304326,
      "loss": 4.7333,
      "mean_token_accuracy": 0.21467170417308806,
      "num_tokens": 11026840.0,
      "step": 6050
    },
    {
      "entropy": 5.267746019363403,
      "epoch": 5.016570008285004,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003053291455903782,
      "loss": 4.72,
      "mean_token_accuracy": 0.2128629431128502,
      "num_tokens": 11034906.0,
      "step": 6055
    },
    {
      "entropy": 5.366693639755249,
      "epoch": 5.0207125103562555,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030501247537312293,
      "loss": 4.7654,
      "mean_token_accuracy": 0.21446437686681746,
      "num_tokens": 11044403.0,
      "step": 6060
    },
    {
      "entropy": 5.383611392974854,
      "epoch": 5.024855012427507,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003046957446174251,
      "loss": 4.719,
      "mean_token_accuracy": 0.21386451721191407,
      "num_tokens": 11052685.0,
      "step": 6065
    },
    {
      "entropy": 5.276443910598755,
      "epoch": 5.028997514498757,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030437895396216524,
      "loss": 4.5862,
      "mean_token_accuracy": 0.22429430484771729,
      "num_tokens": 11060830.0,
      "step": 6070
    },
    {
      "entropy": 5.327369451522827,
      "epoch": 5.033140016570008,
      "grad_norm": 1.25,
      "learning_rate": 0.00030406210404634475,
      "loss": 4.7504,
      "mean_token_accuracy": 0.21197423040866853,
      "num_tokens": 11070220.0,
      "step": 6075
    },
    {
      "entropy": 5.294449806213379,
      "epoch": 5.037282518641259,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003037451955090846,
      "loss": 4.6864,
      "mean_token_accuracy": 0.20806053876876832,
      "num_tokens": 11079171.0,
      "step": 6080
    },
    {
      "entropy": 5.31803297996521,
      "epoch": 5.04142502071251,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030342822898962377,
      "loss": 4.7565,
      "mean_token_accuracy": 0.20876048058271407,
      "num_tokens": 11089205.0,
      "step": 6085
    },
    {
      "entropy": 5.311614274978638,
      "epoch": 5.045567522783761,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003031112051273185,
      "loss": 4.7391,
      "mean_token_accuracy": 0.20910681486129762,
      "num_tokens": 11098339.0,
      "step": 6090
    },
    {
      "entropy": 5.328084325790405,
      "epoch": 5.049710024855012,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003027941245616405,
      "loss": 4.6407,
      "mean_token_accuracy": 0.22161855399608613,
      "num_tokens": 11107218.0,
      "step": 6095
    },
    {
      "entropy": 5.336871147155762,
      "epoch": 5.0538525269262635,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003024769879321761,
      "loss": 4.7647,
      "mean_token_accuracy": 0.2059190019965172,
      "num_tokens": 11116457.0,
      "step": 6100
    },
    {
      "entropy": 5.285742998123169,
      "epoch": 5.057995028997515,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003021597958786243,
      "loss": 4.7395,
      "mean_token_accuracy": 0.20309974104166031,
      "num_tokens": 11124834.0,
      "step": 6105
    },
    {
      "entropy": 5.300694370269776,
      "epoch": 5.062137531068766,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003018425490407962,
      "loss": 4.764,
      "mean_token_accuracy": 0.21726671755313873,
      "num_tokens": 11133920.0,
      "step": 6110
    },
    {
      "entropy": 5.400709104537964,
      "epoch": 5.066280033140017,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030152524805861344,
      "loss": 4.7608,
      "mean_token_accuracy": 0.21305971145629882,
      "num_tokens": 11143573.0,
      "step": 6115
    },
    {
      "entropy": 5.374997186660766,
      "epoch": 5.070422535211268,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030120789357210667,
      "loss": 4.7853,
      "mean_token_accuracy": 0.21126463860273362,
      "num_tokens": 11154255.0,
      "step": 6120
    },
    {
      "entropy": 5.311231470108032,
      "epoch": 5.074565037282519,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030089048622141453,
      "loss": 4.7193,
      "mean_token_accuracy": 0.2135860413312912,
      "num_tokens": 11163306.0,
      "step": 6125
    },
    {
      "entropy": 5.351924800872803,
      "epoch": 5.078707539353769,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030057302664678243,
      "loss": 4.7496,
      "mean_token_accuracy": 0.20993416160345077,
      "num_tokens": 11172763.0,
      "step": 6130
    },
    {
      "entropy": 5.342840385437012,
      "epoch": 5.08285004142502,
      "grad_norm": 1.40625,
      "learning_rate": 0.00030025551548856105,
      "loss": 4.737,
      "mean_token_accuracy": 0.21159224957227707,
      "num_tokens": 11182196.0,
      "step": 6135
    },
    {
      "entropy": 5.273936557769775,
      "epoch": 5.0869925434962715,
      "grad_norm": 1.25,
      "learning_rate": 0.0002999379533872051,
      "loss": 4.7876,
      "mean_token_accuracy": 0.19894420504570007,
      "num_tokens": 11192462.0,
      "step": 6140
    },
    {
      "entropy": 5.267487478256226,
      "epoch": 5.091135045567523,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002996203409832719,
      "loss": 4.7256,
      "mean_token_accuracy": 0.21519733220338821,
      "num_tokens": 11201805.0,
      "step": 6145
    },
    {
      "entropy": 5.3185279846191404,
      "epoch": 5.095277547638774,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002993026789174205,
      "loss": 4.7327,
      "mean_token_accuracy": 0.2225225269794464,
      "num_tokens": 11210114.0,
      "step": 6150
    },
    {
      "entropy": 5.329169511795044,
      "epoch": 5.099420049710025,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002989849678304099,
      "loss": 4.8153,
      "mean_token_accuracy": 0.2070423409342766,
      "num_tokens": 11219725.0,
      "step": 6155
    },
    {
      "entropy": 5.322601699829102,
      "epoch": 5.103562551781276,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029866720836309817,
      "loss": 4.7291,
      "mean_token_accuracy": 0.20557260662317275,
      "num_tokens": 11228583.0,
      "step": 6160
    },
    {
      "entropy": 5.33705506324768,
      "epoch": 5.107705053852527,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029834940115644086,
      "loss": 4.7298,
      "mean_token_accuracy": 0.2115030273795128,
      "num_tokens": 11237787.0,
      "step": 6165
    },
    {
      "entropy": 5.286470556259156,
      "epoch": 5.111847555923778,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002980315468514898,
      "loss": 4.7583,
      "mean_token_accuracy": 0.20592353045940398,
      "num_tokens": 11247136.0,
      "step": 6170
    },
    {
      "entropy": 5.239829063415527,
      "epoch": 5.115990057995029,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029771364608939196,
      "loss": 4.6924,
      "mean_token_accuracy": 0.20855813026428222,
      "num_tokens": 11256931.0,
      "step": 6175
    },
    {
      "entropy": 5.300743341445923,
      "epoch": 5.12013256006628,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002973956995113878,
      "loss": 4.6672,
      "mean_token_accuracy": 0.21277486830949782,
      "num_tokens": 11266457.0,
      "step": 6180
    },
    {
      "entropy": 5.272923183441162,
      "epoch": 5.1242750621375315,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002970777077588103,
      "loss": 4.643,
      "mean_token_accuracy": 0.21718240678310394,
      "num_tokens": 11274239.0,
      "step": 6185
    },
    {
      "entropy": 5.286969995498657,
      "epoch": 5.128417564208782,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002967596714730837,
      "loss": 4.7572,
      "mean_token_accuracy": 0.20912032276391984,
      "num_tokens": 11283239.0,
      "step": 6190
    },
    {
      "entropy": 5.284596729278564,
      "epoch": 5.132560066280033,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029644159129572185,
      "loss": 4.7243,
      "mean_token_accuracy": 0.20871085673570633,
      "num_tokens": 11291662.0,
      "step": 6195
    },
    {
      "entropy": 5.2875102996826175,
      "epoch": 5.136702568351284,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002961234678683272,
      "loss": 4.7338,
      "mean_token_accuracy": 0.21518487930297853,
      "num_tokens": 11300165.0,
      "step": 6200
    },
    {
      "entropy": 5.307006978988648,
      "epoch": 5.140845070422535,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002958053018325896,
      "loss": 4.7188,
      "mean_token_accuracy": 0.21759819090366364,
      "num_tokens": 11309729.0,
      "step": 6205
    },
    {
      "entropy": 5.370016050338745,
      "epoch": 5.144987572493786,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002954870938302846,
      "loss": 4.7562,
      "mean_token_accuracy": 0.20499182790517806,
      "num_tokens": 11318674.0,
      "step": 6210
    },
    {
      "entropy": 5.318466091156006,
      "epoch": 5.149130074565037,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002951688445032726,
      "loss": 4.7239,
      "mean_token_accuracy": 0.21144993752241134,
      "num_tokens": 11327583.0,
      "step": 6215
    },
    {
      "entropy": 5.306168127059936,
      "epoch": 5.153272576636288,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002948505544934971,
      "loss": 4.7492,
      "mean_token_accuracy": 0.2113962024450302,
      "num_tokens": 11338132.0,
      "step": 6220
    },
    {
      "entropy": 5.311028718948364,
      "epoch": 5.1574150787075395,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000294532224442984,
      "loss": 4.7927,
      "mean_token_accuracy": 0.20897248685359954,
      "num_tokens": 11347210.0,
      "step": 6225
    },
    {
      "entropy": 5.258694267272949,
      "epoch": 5.161557580778791,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029421385499383976,
      "loss": 4.6687,
      "mean_token_accuracy": 0.21956470906734465,
      "num_tokens": 11355479.0,
      "step": 6230
    },
    {
      "entropy": 5.31318998336792,
      "epoch": 5.165700082850042,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029389544678825035,
      "loss": 4.8328,
      "mean_token_accuracy": 0.1991332709789276,
      "num_tokens": 11364518.0,
      "step": 6235
    },
    {
      "entropy": 5.269758129119873,
      "epoch": 5.169842584921293,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029357700046847985,
      "loss": 4.7363,
      "mean_token_accuracy": 0.20779081732034682,
      "num_tokens": 11373589.0,
      "step": 6240
    },
    {
      "entropy": 5.3930335521698,
      "epoch": 5.173985086992544,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002932585166768694,
      "loss": 4.778,
      "mean_token_accuracy": 0.21061132550239564,
      "num_tokens": 11383209.0,
      "step": 6245
    },
    {
      "entropy": 5.310416650772095,
      "epoch": 5.178127589063794,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029293999605583545,
      "loss": 4.7087,
      "mean_token_accuracy": 0.21529638320207595,
      "num_tokens": 11392012.0,
      "step": 6250
    },
    {
      "entropy": 5.298952531814575,
      "epoch": 5.182270091135045,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029262143924786905,
      "loss": 4.7717,
      "mean_token_accuracy": 0.20748427361249924,
      "num_tokens": 11400489.0,
      "step": 6255
    },
    {
      "entropy": 5.269466066360474,
      "epoch": 5.186412593206296,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029230284689553404,
      "loss": 4.6912,
      "mean_token_accuracy": 0.21118670403957368,
      "num_tokens": 11409337.0,
      "step": 6260
    },
    {
      "entropy": 5.3224327087402346,
      "epoch": 5.190555095277547,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029198421964146605,
      "loss": 4.7322,
      "mean_token_accuracy": 0.21558451503515244,
      "num_tokens": 11418650.0,
      "step": 6265
    },
    {
      "entropy": 5.297787761688232,
      "epoch": 5.1946975973487985,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002916655581283709,
      "loss": 4.7468,
      "mean_token_accuracy": 0.20413282811641692,
      "num_tokens": 11427516.0,
      "step": 6270
    },
    {
      "entropy": 5.255175161361694,
      "epoch": 5.19884009942005,
      "grad_norm": 1.484375,
      "learning_rate": 0.00029134686299902397,
      "loss": 4.7413,
      "mean_token_accuracy": 0.2071158468723297,
      "num_tokens": 11435676.0,
      "step": 6275
    },
    {
      "entropy": 5.336490201950073,
      "epoch": 5.202982601491301,
      "grad_norm": 1.25,
      "learning_rate": 0.0002910281348962679,
      "loss": 4.7863,
      "mean_token_accuracy": 0.19594063609838486,
      "num_tokens": 11444520.0,
      "step": 6280
    },
    {
      "entropy": 5.3500142097473145,
      "epoch": 5.207125103562552,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029070937446301225,
      "loss": 4.7512,
      "mean_token_accuracy": 0.2061683714389801,
      "num_tokens": 11453759.0,
      "step": 6285
    },
    {
      "entropy": 5.378877973556518,
      "epoch": 5.211267605633803,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002903905823422317,
      "loss": 4.8813,
      "mean_token_accuracy": 0.19547753483057023,
      "num_tokens": 11463409.0,
      "step": 6290
    },
    {
      "entropy": 5.27807126045227,
      "epoch": 5.215410107705054,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002900717591769646,
      "loss": 4.7931,
      "mean_token_accuracy": 0.20699832290410997,
      "num_tokens": 11473440.0,
      "step": 6295
    },
    {
      "entropy": 5.271660184860229,
      "epoch": 5.219552609776305,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002897529056103122,
      "loss": 4.6289,
      "mean_token_accuracy": 0.2164784327149391,
      "num_tokens": 11482353.0,
      "step": 6300
    },
    {
      "entropy": 5.323471403121948,
      "epoch": 5.223695111847556,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002894340222854371,
      "loss": 4.7588,
      "mean_token_accuracy": 0.20732663720846176,
      "num_tokens": 11491331.0,
      "step": 6305
    },
    {
      "entropy": 5.23156099319458,
      "epoch": 5.2278376139188065,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002891151098455617,
      "loss": 4.6976,
      "mean_token_accuracy": 0.20581794679164886,
      "num_tokens": 11500553.0,
      "step": 6310
    },
    {
      "entropy": 5.197248125076294,
      "epoch": 5.231980115990058,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002887961689339672,
      "loss": 4.6446,
      "mean_token_accuracy": 0.21304180175065995,
      "num_tokens": 11509146.0,
      "step": 6315
    },
    {
      "entropy": 5.333087635040283,
      "epoch": 5.236122618061309,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002884772001939924,
      "loss": 4.7664,
      "mean_token_accuracy": 0.20489265024662018,
      "num_tokens": 11518172.0,
      "step": 6320
    },
    {
      "entropy": 5.309455251693725,
      "epoch": 5.24026512013256,
      "grad_norm": 1.296875,
      "learning_rate": 0.000288158204269032,
      "loss": 4.7733,
      "mean_token_accuracy": 0.20310726165771484,
      "num_tokens": 11527291.0,
      "step": 6325
    },
    {
      "entropy": 5.2990336418151855,
      "epoch": 5.244407622203811,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028783918180253567,
      "loss": 4.7745,
      "mean_token_accuracy": 0.20924709141254424,
      "num_tokens": 11536306.0,
      "step": 6330
    },
    {
      "entropy": 5.291861248016358,
      "epoch": 5.248550124275062,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002875201334380066,
      "loss": 4.7879,
      "mean_token_accuracy": 0.21372536867856978,
      "num_tokens": 11545879.0,
      "step": 6335
    },
    {
      "entropy": 5.2518861293792725,
      "epoch": 5.252692626346313,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002872010598190002,
      "loss": 4.7157,
      "mean_token_accuracy": 0.21597641110420226,
      "num_tokens": 11555067.0,
      "step": 6340
    },
    {
      "entropy": 5.24732780456543,
      "epoch": 5.256835128417564,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002868819615891229,
      "loss": 4.6881,
      "mean_token_accuracy": 0.21341271847486495,
      "num_tokens": 11563419.0,
      "step": 6345
    },
    {
      "entropy": 5.234815073013306,
      "epoch": 5.260977630488815,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002865628393920305,
      "loss": 4.7643,
      "mean_token_accuracy": 0.20835172683000563,
      "num_tokens": 11573018.0,
      "step": 6350
    },
    {
      "entropy": 5.213118267059326,
      "epoch": 5.2651201325600665,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002862436938714276,
      "loss": 4.697,
      "mean_token_accuracy": 0.20897249579429628,
      "num_tokens": 11581996.0,
      "step": 6355
    },
    {
      "entropy": 5.35305643081665,
      "epoch": 5.269262634631318,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002859245256710654,
      "loss": 4.7404,
      "mean_token_accuracy": 0.20999680757522582,
      "num_tokens": 11591706.0,
      "step": 6360
    },
    {
      "entropy": 5.342862844467163,
      "epoch": 5.273405136702569,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002856053354347411,
      "loss": 4.7787,
      "mean_token_accuracy": 0.21335438787937164,
      "num_tokens": 11600616.0,
      "step": 6365
    },
    {
      "entropy": 5.258344507217407,
      "epoch": 5.27754763877382,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002852861238062963,
      "loss": 4.6793,
      "mean_token_accuracy": 0.21405228972434998,
      "num_tokens": 11609650.0,
      "step": 6370
    },
    {
      "entropy": 5.259497785568238,
      "epoch": 5.28169014084507,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002849668914296157,
      "loss": 4.6831,
      "mean_token_accuracy": 0.21503762751817704,
      "num_tokens": 11618785.0,
      "step": 6375
    },
    {
      "entropy": 5.2954730033874515,
      "epoch": 5.285832642916321,
      "grad_norm": 1.1875,
      "learning_rate": 0.000284647638948626,
      "loss": 4.7581,
      "mean_token_accuracy": 0.20522686541080476,
      "num_tokens": 11628089.0,
      "step": 6380
    },
    {
      "entropy": 5.325137042999268,
      "epoch": 5.289975144987572,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002843283670072942,
      "loss": 4.8475,
      "mean_token_accuracy": 0.19651062339544295,
      "num_tokens": 11636954.0,
      "step": 6385
    },
    {
      "entropy": 5.289578056335449,
      "epoch": 5.294117647058823,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002840090762496268,
      "loss": 4.7935,
      "mean_token_accuracy": 0.21091917753219605,
      "num_tokens": 11646719.0,
      "step": 6390
    },
    {
      "entropy": 5.293153905868531,
      "epoch": 5.2982601491300745,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002836897673196681,
      "loss": 4.6893,
      "mean_token_accuracy": 0.20873350501060486,
      "num_tokens": 11655589.0,
      "step": 6395
    },
    {
      "entropy": 5.261083793640137,
      "epoch": 5.302402651201326,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028337044086149913,
      "loss": 4.7247,
      "mean_token_accuracy": 0.2081915482878685,
      "num_tokens": 11663703.0,
      "step": 6400
    },
    {
      "entropy": 5.304580163955689,
      "epoch": 5.306545153272577,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028305109751923616,
      "loss": 4.8185,
      "mean_token_accuracy": 0.2058854728937149,
      "num_tokens": 11672643.0,
      "step": 6405
    },
    {
      "entropy": 5.262790822982788,
      "epoch": 5.310687655343828,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028273173793702967,
      "loss": 4.6799,
      "mean_token_accuracy": 0.2053900897502899,
      "num_tokens": 11681449.0,
      "step": 6410
    },
    {
      "entropy": 5.297705030441284,
      "epoch": 5.314830157415079,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002824123627590629,
      "loss": 4.7525,
      "mean_token_accuracy": 0.20490695685148239,
      "num_tokens": 11690820.0,
      "step": 6415
    },
    {
      "entropy": 5.3078490734100345,
      "epoch": 5.31897265948633,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002820929726295504,
      "loss": 4.7686,
      "mean_token_accuracy": 0.21030567586421967,
      "num_tokens": 11699705.0,
      "step": 6420
    },
    {
      "entropy": 5.234976291656494,
      "epoch": 5.323115161557581,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002817735681927371,
      "loss": 4.7274,
      "mean_token_accuracy": 0.21462335735559462,
      "num_tokens": 11709417.0,
      "step": 6425
    },
    {
      "entropy": 5.313077116012574,
      "epoch": 5.327257663628831,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028145415009289644,
      "loss": 4.778,
      "mean_token_accuracy": 0.20744761675596238,
      "num_tokens": 11718661.0,
      "step": 6430
    },
    {
      "entropy": 5.279898405075073,
      "epoch": 5.3314001657000825,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002811347189743297,
      "loss": 4.7452,
      "mean_token_accuracy": 0.21239444017410278,
      "num_tokens": 11727568.0,
      "step": 6435
    },
    {
      "entropy": 5.248041486740112,
      "epoch": 5.335542667771334,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028081527548136446,
      "loss": 4.6759,
      "mean_token_accuracy": 0.2261684238910675,
      "num_tokens": 11736615.0,
      "step": 6440
    },
    {
      "entropy": 5.341589784622192,
      "epoch": 5.339685169842585,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002804958202583531,
      "loss": 4.7942,
      "mean_token_accuracy": 0.20758422166109086,
      "num_tokens": 11747458.0,
      "step": 6445
    },
    {
      "entropy": 5.308155250549317,
      "epoch": 5.343827671913836,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002801763539496718,
      "loss": 4.7912,
      "mean_token_accuracy": 0.21049614995718002,
      "num_tokens": 11756279.0,
      "step": 6450
    },
    {
      "entropy": 5.268953704833985,
      "epoch": 5.347970173985087,
      "grad_norm": 1.375,
      "learning_rate": 0.0002798568771997188,
      "loss": 4.7801,
      "mean_token_accuracy": 0.21439336389303207,
      "num_tokens": 11765022.0,
      "step": 6455
    },
    {
      "entropy": 5.214200973510742,
      "epoch": 5.352112676056338,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002795373906529137,
      "loss": 4.7259,
      "mean_token_accuracy": 0.20943275839090347,
      "num_tokens": 11774704.0,
      "step": 6460
    },
    {
      "entropy": 5.274822139739991,
      "epoch": 5.356255178127589,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002792178949536959,
      "loss": 4.6684,
      "mean_token_accuracy": 0.21818412989377975,
      "num_tokens": 11782309.0,
      "step": 6465
    },
    {
      "entropy": 5.255411052703858,
      "epoch": 5.36039768019884,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000278898390746523,
      "loss": 4.7146,
      "mean_token_accuracy": 0.2068032741546631,
      "num_tokens": 11792595.0,
      "step": 6470
    },
    {
      "entropy": 5.253266859054565,
      "epoch": 5.364540182270091,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027857887867587005,
      "loss": 4.7316,
      "mean_token_accuracy": 0.21163432002067567,
      "num_tokens": 11801536.0,
      "step": 6475
    },
    {
      "entropy": 5.31667275428772,
      "epoch": 5.368682684341342,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002782593593862276,
      "loss": 4.8114,
      "mean_token_accuracy": 0.19942343682050706,
      "num_tokens": 11810222.0,
      "step": 6480
    },
    {
      "entropy": 5.281545734405517,
      "epoch": 5.3728251864125935,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002779398335221011,
      "loss": 4.7882,
      "mean_token_accuracy": 0.20965925753116607,
      "num_tokens": 11819310.0,
      "step": 6485
    },
    {
      "entropy": 5.268137216567993,
      "epoch": 5.376967688483845,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002776203017280092,
      "loss": 4.7596,
      "mean_token_accuracy": 0.203800031542778,
      "num_tokens": 11827916.0,
      "step": 6490
    },
    {
      "entropy": 5.2395689487457275,
      "epoch": 5.381110190555095,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027730076464848225,
      "loss": 4.7627,
      "mean_token_accuracy": 0.21074754148721694,
      "num_tokens": 11836770.0,
      "step": 6495
    },
    {
      "entropy": 5.271307849884034,
      "epoch": 5.385252692626346,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027698122292806174,
      "loss": 4.7057,
      "mean_token_accuracy": 0.21504159569740294,
      "num_tokens": 11845917.0,
      "step": 6500
    },
    {
      "epoch": 5.385252692626346,
      "eval_entropy": 5.2211087598983665,
      "eval_loss": 5.805326461791992,
      "eval_mean_token_accuracy": 0.16286381451623527,
      "eval_num_tokens": 11845917.0,
      "eval_runtime": 2.7555,
      "eval_samples_per_second": 1359.835,
      "eval_steps_per_second": 170.206,
      "step": 6500
    },
    {
      "entropy": 5.27996506690979,
      "epoch": 5.389395194697597,
      "grad_norm": 1.296875,
      "learning_rate": 0.000276661677211298,
      "loss": 4.7218,
      "mean_token_accuracy": 0.21067458242177964,
      "num_tokens": 11855118.0,
      "step": 6505
    },
    {
      "entropy": 5.290355539321899,
      "epoch": 5.393537696768848,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002763421281427497,
      "loss": 4.6853,
      "mean_token_accuracy": 0.2137836202979088,
      "num_tokens": 11863793.0,
      "step": 6510
    },
    {
      "entropy": 5.385941362380981,
      "epoch": 5.397680198840099,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002760225763669823,
      "loss": 4.8313,
      "mean_token_accuracy": 0.2005970597267151,
      "num_tokens": 11873194.0,
      "step": 6515
    },
    {
      "entropy": 5.298235607147217,
      "epoch": 5.40182270091135,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027570302252856666,
      "loss": 4.7678,
      "mean_token_accuracy": 0.20786977410316468,
      "num_tokens": 11882566.0,
      "step": 6520
    },
    {
      "entropy": 5.248862600326538,
      "epoch": 5.4059652029826015,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027538346727207767,
      "loss": 4.7251,
      "mean_token_accuracy": 0.2184912756085396,
      "num_tokens": 11892267.0,
      "step": 6525
    },
    {
      "entropy": 5.307270288467407,
      "epoch": 5.410107705053853,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002750639112420934,
      "loss": 4.7704,
      "mean_token_accuracy": 0.20652723163366318,
      "num_tokens": 11901440.0,
      "step": 6530
    },
    {
      "entropy": 5.306009674072266,
      "epoch": 5.414250207125104,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027474435508319304,
      "loss": 4.8379,
      "mean_token_accuracy": 0.2083204060792923,
      "num_tokens": 11911259.0,
      "step": 6535
    },
    {
      "entropy": 5.290991640090942,
      "epoch": 5.418392709196355,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027442479943995645,
      "loss": 4.8012,
      "mean_token_accuracy": 0.20378943383693696,
      "num_tokens": 11920467.0,
      "step": 6540
    },
    {
      "entropy": 5.296770906448364,
      "epoch": 5.422535211267606,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002741052449569624,
      "loss": 4.7821,
      "mean_token_accuracy": 0.20670236200094222,
      "num_tokens": 11929971.0,
      "step": 6545
    },
    {
      "entropy": 5.34657564163208,
      "epoch": 5.426677713338857,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002737856922787869,
      "loss": 4.8034,
      "mean_token_accuracy": 0.2000308960676193,
      "num_tokens": 11938585.0,
      "step": 6550
    },
    {
      "entropy": 5.286465501785278,
      "epoch": 5.430820215410108,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027346614205000277,
      "loss": 4.7816,
      "mean_token_accuracy": 0.2058384746313095,
      "num_tokens": 11948264.0,
      "step": 6555
    },
    {
      "entropy": 5.299284839630127,
      "epoch": 5.434962717481358,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027314659491517777,
      "loss": 4.782,
      "mean_token_accuracy": 0.20998325943946838,
      "num_tokens": 11958459.0,
      "step": 6560
    },
    {
      "entropy": 5.328062868118286,
      "epoch": 5.4391052195526095,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002728270515188733,
      "loss": 4.7653,
      "mean_token_accuracy": 0.21042998731136323,
      "num_tokens": 11967994.0,
      "step": 6565
    },
    {
      "entropy": 5.375079822540283,
      "epoch": 5.443247721623861,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002725075125056435,
      "loss": 4.8042,
      "mean_token_accuracy": 0.2106100708246231,
      "num_tokens": 11976564.0,
      "step": 6570
    },
    {
      "entropy": 5.281209468841553,
      "epoch": 5.447390223695112,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002721879785200333,
      "loss": 4.7561,
      "mean_token_accuracy": 0.21402042210102082,
      "num_tokens": 11985682.0,
      "step": 6575
    },
    {
      "entropy": 5.310739374160766,
      "epoch": 5.451532725766363,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002718684502065777,
      "loss": 4.8023,
      "mean_token_accuracy": 0.20315634459257126,
      "num_tokens": 11995098.0,
      "step": 6580
    },
    {
      "entropy": 5.31499719619751,
      "epoch": 5.455675227837614,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027154892820980024,
      "loss": 4.8024,
      "mean_token_accuracy": 0.20829492509365083,
      "num_tokens": 12004538.0,
      "step": 6585
    },
    {
      "entropy": 5.283276653289795,
      "epoch": 5.459817729908865,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027122941317421175,
      "loss": 4.7472,
      "mean_token_accuracy": 0.20832152515649796,
      "num_tokens": 12012568.0,
      "step": 6590
    },
    {
      "entropy": 5.285381889343261,
      "epoch": 5.463960231980116,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000270909905744309,
      "loss": 4.762,
      "mean_token_accuracy": 0.20783878415822982,
      "num_tokens": 12021934.0,
      "step": 6595
    },
    {
      "entropy": 5.311660814285278,
      "epoch": 5.468102734051367,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027059040656457316,
      "loss": 4.7971,
      "mean_token_accuracy": 0.2041061446070671,
      "num_tokens": 12031158.0,
      "step": 6600
    },
    {
      "entropy": 5.3546699523925785,
      "epoch": 5.472245236122618,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002702709162794692,
      "loss": 4.7888,
      "mean_token_accuracy": 0.20316218733787536,
      "num_tokens": 12040392.0,
      "step": 6605
    },
    {
      "entropy": 5.243852329254151,
      "epoch": 5.4763877381938695,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026995143553344394,
      "loss": 4.6448,
      "mean_token_accuracy": 0.2206346407532692,
      "num_tokens": 12048412.0,
      "step": 6610
    },
    {
      "entropy": 5.254961729049683,
      "epoch": 5.48053024026512,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026963196497092483,
      "loss": 4.7667,
      "mean_token_accuracy": 0.20953905433416367,
      "num_tokens": 12057681.0,
      "step": 6615
    },
    {
      "entropy": 5.274502468109131,
      "epoch": 5.484672742336371,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00026931250523631905,
      "loss": 4.7542,
      "mean_token_accuracy": 0.20219781249761581,
      "num_tokens": 12066243.0,
      "step": 6620
    },
    {
      "entropy": 5.231963729858398,
      "epoch": 5.488815244407622,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026899305697401175,
      "loss": 4.7658,
      "mean_token_accuracy": 0.20898472964763642,
      "num_tokens": 12075312.0,
      "step": 6625
    },
    {
      "entropy": 5.279199838638306,
      "epoch": 5.492957746478873,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002686736208283649,
      "loss": 4.7728,
      "mean_token_accuracy": 0.2083324670791626,
      "num_tokens": 12083990.0,
      "step": 6630
    },
    {
      "entropy": 5.406312322616577,
      "epoch": 5.497100248550124,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026835419744371627,
      "loss": 4.8616,
      "mean_token_accuracy": 0.2029019698500633,
      "num_tokens": 12093301.0,
      "step": 6635
    },
    {
      "entropy": 5.235708093643188,
      "epoch": 5.501242750621375,
      "grad_norm": 1.375,
      "learning_rate": 0.0002680347874643775,
      "loss": 4.7055,
      "mean_token_accuracy": 0.21459992229938507,
      "num_tokens": 12102054.0,
      "step": 6640
    },
    {
      "entropy": 5.264314031600952,
      "epoch": 5.505385252692626,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002677153915346336,
      "loss": 4.7831,
      "mean_token_accuracy": 0.2037147119641304,
      "num_tokens": 12111432.0,
      "step": 6645
    },
    {
      "entropy": 5.2460126876831055,
      "epoch": 5.5095277547638775,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002673960102987412,
      "loss": 4.7019,
      "mean_token_accuracy": 0.2157286986708641,
      "num_tokens": 12120424.0,
      "step": 6650
    },
    {
      "entropy": 5.27989239692688,
      "epoch": 5.513670256835129,
      "grad_norm": 1.375,
      "learning_rate": 0.00026707664440092676,
      "loss": 4.8491,
      "mean_token_accuracy": 0.21006827503442765,
      "num_tokens": 12130210.0,
      "step": 6655
    },
    {
      "entropy": 5.415329027175903,
      "epoch": 5.51781275890638,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026675729448538655,
      "loss": 4.8268,
      "mean_token_accuracy": 0.21183697134256363,
      "num_tokens": 12139366.0,
      "step": 6660
    },
    {
      "entropy": 5.291393804550171,
      "epoch": 5.521955260977631,
      "grad_norm": 1.25,
      "learning_rate": 0.00026643796119628404,
      "loss": 4.6535,
      "mean_token_accuracy": 0.21639101058244706,
      "num_tokens": 12148385.0,
      "step": 6665
    },
    {
      "entropy": 5.328906440734864,
      "epoch": 5.526097763048882,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002661186451777495,
      "loss": 4.7895,
      "mean_token_accuracy": 0.21066424548625945,
      "num_tokens": 12156640.0,
      "step": 6670
    },
    {
      "entropy": 5.293066596984863,
      "epoch": 5.530240265120133,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002657993470738781,
      "loss": 4.8533,
      "mean_token_accuracy": 0.19991559982299806,
      "num_tokens": 12166532.0,
      "step": 6675
    },
    {
      "entropy": 5.226839828491211,
      "epoch": 5.534382767191383,
      "grad_norm": 1.25,
      "learning_rate": 0.00026548006752872915,
      "loss": 4.6878,
      "mean_token_accuracy": 0.2191959947347641,
      "num_tokens": 12175239.0,
      "step": 6680
    },
    {
      "entropy": 5.235377931594849,
      "epoch": 5.538525269262634,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002651608071863244,
      "loss": 4.7467,
      "mean_token_accuracy": 0.21195551753044128,
      "num_tokens": 12184588.0,
      "step": 6685
    },
    {
      "entropy": 5.350800323486328,
      "epoch": 5.542667771333885,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026484156669064665,
      "loss": 4.8113,
      "mean_token_accuracy": 0.20172822177410127,
      "num_tokens": 12194140.0,
      "step": 6690
    },
    {
      "entropy": 5.323469638824463,
      "epoch": 5.5468102734051365,
      "grad_norm": 1.390625,
      "learning_rate": 0.000264522346685639,
      "loss": 4.826,
      "mean_token_accuracy": 0.20553469508886338,
      "num_tokens": 12203504.0,
      "step": 6695
    },
    {
      "entropy": 5.323514175415039,
      "epoch": 5.550952775476388,
      "grad_norm": 1.25,
      "learning_rate": 0.0002642031478152031,
      "loss": 4.7729,
      "mean_token_accuracy": 0.21506688892841339,
      "num_tokens": 12212594.0,
      "step": 6700
    },
    {
      "entropy": 5.323617172241211,
      "epoch": 5.555095277547639,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026388397072319786,
      "loss": 4.8203,
      "mean_token_accuracy": 0.20215453058481217,
      "num_tokens": 12222101.0,
      "step": 6705
    },
    {
      "entropy": 5.289176988601684,
      "epoch": 5.55923777961889,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002635648160534385,
      "loss": 4.7928,
      "mean_token_accuracy": 0.20502501279115676,
      "num_tokens": 12232118.0,
      "step": 6710
    },
    {
      "entropy": 5.367046308517456,
      "epoch": 5.563380281690141,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026324568444969473,
      "loss": 4.7374,
      "mean_token_accuracy": 0.20731871724128723,
      "num_tokens": 12241275.0,
      "step": 6715
    },
    {
      "entropy": 5.213319921493531,
      "epoch": 5.567522783761392,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026292657655569,
      "loss": 4.7435,
      "mean_token_accuracy": 0.21404872387647628,
      "num_tokens": 12251125.0,
      "step": 6720
    },
    {
      "entropy": 5.2403851509094235,
      "epoch": 5.571665285832643,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026260749301509954,
      "loss": 4.7255,
      "mean_token_accuracy": 0.21383213102817536,
      "num_tokens": 12260613.0,
      "step": 6725
    },
    {
      "entropy": 5.374528121948242,
      "epoch": 5.575807787903894,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026228843447155005,
      "loss": 4.7293,
      "mean_token_accuracy": 0.2178867921233177,
      "num_tokens": 12269878.0,
      "step": 6730
    },
    {
      "entropy": 5.263756942749024,
      "epoch": 5.5799502899751445,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026196940156861736,
      "loss": 4.7927,
      "mean_token_accuracy": 0.20681553930044175,
      "num_tokens": 12278616.0,
      "step": 6735
    },
    {
      "entropy": 5.198074913024902,
      "epoch": 5.584092792046396,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026165039494982555,
      "loss": 4.7071,
      "mean_token_accuracy": 0.21244536191225052,
      "num_tokens": 12287104.0,
      "step": 6740
    },
    {
      "entropy": 5.224111557006836,
      "epoch": 5.588235294117647,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026133141525864605,
      "loss": 4.736,
      "mean_token_accuracy": 0.21215516626834868,
      "num_tokens": 12296503.0,
      "step": 6745
    },
    {
      "entropy": 5.283900499343872,
      "epoch": 5.592377796188898,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026101246313849557,
      "loss": 4.6961,
      "mean_token_accuracy": 0.21740568727254866,
      "num_tokens": 12306298.0,
      "step": 6750
    },
    {
      "entropy": 5.278920459747314,
      "epoch": 5.596520298260149,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002606935392327356,
      "loss": 4.7618,
      "mean_token_accuracy": 0.20829397290945054,
      "num_tokens": 12314826.0,
      "step": 6755
    },
    {
      "entropy": 5.257278776168823,
      "epoch": 5.6006628003314,
      "grad_norm": 1.25,
      "learning_rate": 0.00026037464418467035,
      "loss": 4.7013,
      "mean_token_accuracy": 0.2183123216032982,
      "num_tokens": 12323776.0,
      "step": 6760
    },
    {
      "entropy": 5.317180252075195,
      "epoch": 5.604805302402651,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002600557786375459,
      "loss": 4.7962,
      "mean_token_accuracy": 0.2113049641251564,
      "num_tokens": 12332374.0,
      "step": 6765
    },
    {
      "entropy": 5.237828922271729,
      "epoch": 5.608947804473902,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025973694323454916,
      "loss": 4.6588,
      "mean_token_accuracy": 0.21989188045263292,
      "num_tokens": 12341007.0,
      "step": 6770
    },
    {
      "entropy": 5.204132032394409,
      "epoch": 5.613090306545153,
      "grad_norm": 1.25,
      "learning_rate": 0.00025941813861880575,
      "loss": 4.7055,
      "mean_token_accuracy": 0.21825818568468094,
      "num_tokens": 12350330.0,
      "step": 6775
    },
    {
      "entropy": 5.299660015106201,
      "epoch": 5.6172328086164045,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025909936543337957,
      "loss": 4.747,
      "mean_token_accuracy": 0.20508041381835937,
      "num_tokens": 12358592.0,
      "step": 6780
    },
    {
      "entropy": 5.321096611022949,
      "epoch": 5.621375310687656,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002587806243212709,
      "loss": 4.7672,
      "mean_token_accuracy": 0.20989640206098556,
      "num_tokens": 12368195.0,
      "step": 6785
    },
    {
      "entropy": 5.260707855224609,
      "epoch": 5.625517812758907,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002584619159254154,
      "loss": 4.7008,
      "mean_token_accuracy": 0.21232942789793013,
      "num_tokens": 12377206.0,
      "step": 6790
    },
    {
      "entropy": 5.212645578384399,
      "epoch": 5.629660314830158,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025814324088868276,
      "loss": 4.7391,
      "mean_token_accuracy": 0.21497804969549178,
      "num_tokens": 12386595.0,
      "step": 6795
    },
    {
      "entropy": 5.1747864246368405,
      "epoch": 5.633802816901408,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025782459985387526,
      "loss": 4.6443,
      "mean_token_accuracy": 0.22247450202703475,
      "num_tokens": 12395923.0,
      "step": 6800
    },
    {
      "entropy": 5.2867666721344,
      "epoch": 5.637945318972659,
      "grad_norm": 1.375,
      "learning_rate": 0.0002575059934637268,
      "loss": 4.7958,
      "mean_token_accuracy": 0.2060307815670967,
      "num_tokens": 12405094.0,
      "step": 6805
    },
    {
      "entropy": 5.246973276138306,
      "epoch": 5.64208782104391,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002571874223609012,
      "loss": 4.7979,
      "mean_token_accuracy": 0.20800775438547134,
      "num_tokens": 12414834.0,
      "step": 6810
    },
    {
      "entropy": 5.329070091247559,
      "epoch": 5.646230323115161,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025686888718799136,
      "loss": 4.7416,
      "mean_token_accuracy": 0.20752692222595215,
      "num_tokens": 12423485.0,
      "step": 6815
    },
    {
      "entropy": 5.40279450416565,
      "epoch": 5.6503728251864125,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002565503885875173,
      "loss": 4.789,
      "mean_token_accuracy": 0.21221252679824829,
      "num_tokens": 12432149.0,
      "step": 6820
    },
    {
      "entropy": 5.262519645690918,
      "epoch": 5.654515327257664,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025623192720192564,
      "loss": 4.7558,
      "mean_token_accuracy": 0.21276623159646987,
      "num_tokens": 12441272.0,
      "step": 6825
    },
    {
      "entropy": 5.2058868408203125,
      "epoch": 5.658657829328915,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002559135036735877,
      "loss": 4.7395,
      "mean_token_accuracy": 0.20536566525697708,
      "num_tokens": 12449318.0,
      "step": 6830
    },
    {
      "entropy": 5.2281341552734375,
      "epoch": 5.662800331400166,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025559511864479864,
      "loss": 4.6851,
      "mean_token_accuracy": 0.21890474557876588,
      "num_tokens": 12458473.0,
      "step": 6835
    },
    {
      "entropy": 5.3270261764526365,
      "epoch": 5.666942833471417,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002552767727577758,
      "loss": 4.8205,
      "mean_token_accuracy": 0.21025011092424392,
      "num_tokens": 12467703.0,
      "step": 6840
    },
    {
      "entropy": 5.258397769927979,
      "epoch": 5.671085335542668,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025495846665465774,
      "loss": 4.675,
      "mean_token_accuracy": 0.21747020184993743,
      "num_tokens": 12476166.0,
      "step": 6845
    },
    {
      "entropy": 5.24764084815979,
      "epoch": 5.675227837613919,
      "grad_norm": 1.375,
      "learning_rate": 0.0002546402009775024,
      "loss": 4.7398,
      "mean_token_accuracy": 0.20348911136388778,
      "num_tokens": 12484880.0,
      "step": 6850
    },
    {
      "entropy": 5.317517852783203,
      "epoch": 5.679370339685169,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025432197636828663,
      "loss": 4.8191,
      "mean_token_accuracy": 0.20637772530317305,
      "num_tokens": 12494208.0,
      "step": 6855
    },
    {
      "entropy": 5.323177909851074,
      "epoch": 5.683512841756421,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002540037934689042,
      "loss": 4.7891,
      "mean_token_accuracy": 0.20799186676740647,
      "num_tokens": 12503647.0,
      "step": 6860
    },
    {
      "entropy": 5.3182206630706785,
      "epoch": 5.687655343827672,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025368565292116475,
      "loss": 4.8107,
      "mean_token_accuracy": 0.20570710450410842,
      "num_tokens": 12512869.0,
      "step": 6865
    },
    {
      "entropy": 5.284727287292481,
      "epoch": 5.691797845898923,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002533675553667927,
      "loss": 4.8668,
      "mean_token_accuracy": 0.19300749897956848,
      "num_tokens": 12523222.0,
      "step": 6870
    },
    {
      "entropy": 5.245349359512329,
      "epoch": 5.695940347970174,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025304950144742523,
      "loss": 4.7557,
      "mean_token_accuracy": 0.2118019163608551,
      "num_tokens": 12532343.0,
      "step": 6875
    },
    {
      "entropy": 5.290152835845947,
      "epoch": 5.700082850041425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002527314918046122,
      "loss": 4.7428,
      "mean_token_accuracy": 0.21624454557895662,
      "num_tokens": 12541508.0,
      "step": 6880
    },
    {
      "entropy": 5.261102771759033,
      "epoch": 5.704225352112676,
      "grad_norm": 1.25,
      "learning_rate": 0.0002524135270798136,
      "loss": 4.7716,
      "mean_token_accuracy": 0.2087491661310196,
      "num_tokens": 12550837.0,
      "step": 6885
    },
    {
      "entropy": 5.284794616699219,
      "epoch": 5.708367854183927,
      "grad_norm": 1.375,
      "learning_rate": 0.00025209560791439914,
      "loss": 4.7245,
      "mean_token_accuracy": 0.21842783987522124,
      "num_tokens": 12559859.0,
      "step": 6890
    },
    {
      "entropy": 5.260790061950684,
      "epoch": 5.712510356255178,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002517777349496465,
      "loss": 4.7821,
      "mean_token_accuracy": 0.210851489007473,
      "num_tokens": 12569137.0,
      "step": 6895
    },
    {
      "entropy": 5.3351655960083,
      "epoch": 5.716652858326429,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002514599088267401,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2035819560289383,
      "num_tokens": 12578111.0,
      "step": 6900
    },
    {
      "entropy": 5.30318021774292,
      "epoch": 5.72079536039768,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025114213018677004,
      "loss": 4.763,
      "mean_token_accuracy": 0.21490382850170137,
      "num_tokens": 12587604.0,
      "step": 6905
    },
    {
      "entropy": 5.3550975799560545,
      "epoch": 5.7249378624689315,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002508243996707305,
      "loss": 4.8096,
      "mean_token_accuracy": 0.20439844876527785,
      "num_tokens": 12596620.0,
      "step": 6910
    },
    {
      "entropy": 5.241341829299927,
      "epoch": 5.729080364540183,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025050671791951887,
      "loss": 4.7681,
      "mean_token_accuracy": 0.20889047533273697,
      "num_tokens": 12605452.0,
      "step": 6915
    },
    {
      "entropy": 5.215042495727539,
      "epoch": 5.733222866611433,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002501890855739337,
      "loss": 4.6706,
      "mean_token_accuracy": 0.21881663650274277,
      "num_tokens": 12614009.0,
      "step": 6920
    },
    {
      "entropy": 5.322356796264648,
      "epoch": 5.737365368682684,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024987150327467414,
      "loss": 4.7974,
      "mean_token_accuracy": 0.20980707705020904,
      "num_tokens": 12624135.0,
      "step": 6925
    },
    {
      "entropy": 5.1882896900177,
      "epoch": 5.741507870753935,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024955397166233855,
      "loss": 4.6483,
      "mean_token_accuracy": 0.22657137364149094,
      "num_tokens": 12632870.0,
      "step": 6930
    },
    {
      "entropy": 5.27833366394043,
      "epoch": 5.745650372825186,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024923649137742276,
      "loss": 4.8437,
      "mean_token_accuracy": 0.200515116751194,
      "num_tokens": 12643606.0,
      "step": 6935
    },
    {
      "entropy": 5.412599802017212,
      "epoch": 5.749792874896437,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002489190630603194,
      "loss": 4.8868,
      "mean_token_accuracy": 0.20398871451616288,
      "num_tokens": 12652926.0,
      "step": 6940
    },
    {
      "entropy": 5.342198324203491,
      "epoch": 5.753935376967688,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002486016873513159,
      "loss": 4.8111,
      "mean_token_accuracy": 0.20426811277866364,
      "num_tokens": 12661841.0,
      "step": 6945
    },
    {
      "entropy": 5.242041110992432,
      "epoch": 5.7580778790389395,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002482843648905937,
      "loss": 4.7458,
      "mean_token_accuracy": 0.20681970119476317,
      "num_tokens": 12670527.0,
      "step": 6950
    },
    {
      "entropy": 5.246947288513184,
      "epoch": 5.762220381110191,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002479670963182271,
      "loss": 4.7359,
      "mean_token_accuracy": 0.20286448150873185,
      "num_tokens": 12680346.0,
      "step": 6955
    },
    {
      "entropy": 5.329467821121216,
      "epoch": 5.766362883181442,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024764988227418114,
      "loss": 4.8331,
      "mean_token_accuracy": 0.20897539108991622,
      "num_tokens": 12691088.0,
      "step": 6960
    },
    {
      "entropy": 5.3361204147338865,
      "epoch": 5.770505385252693,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002473327233983116,
      "loss": 4.784,
      "mean_token_accuracy": 0.2111168012022972,
      "num_tokens": 12700296.0,
      "step": 6965
    },
    {
      "entropy": 5.274905633926392,
      "epoch": 5.774647887323944,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002470156203303622,
      "loss": 4.7758,
      "mean_token_accuracy": 0.20992494821548463,
      "num_tokens": 12709595.0,
      "step": 6970
    },
    {
      "entropy": 5.260739278793335,
      "epoch": 5.778790389395195,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002466985737099649,
      "loss": 4.8283,
      "mean_token_accuracy": 0.21001769304275514,
      "num_tokens": 12719052.0,
      "step": 6975
    },
    {
      "entropy": 5.242895555496216,
      "epoch": 5.782932891466446,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002463815841766371,
      "loss": 4.7222,
      "mean_token_accuracy": 0.21401812881231308,
      "num_tokens": 12727287.0,
      "step": 6980
    },
    {
      "entropy": 5.257803535461425,
      "epoch": 5.787075393537696,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002460646523697813,
      "loss": 4.7682,
      "mean_token_accuracy": 0.20726907700300218,
      "num_tokens": 12736511.0,
      "step": 6985
    },
    {
      "entropy": 5.357561779022217,
      "epoch": 5.7912178956089475,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024574777892868386,
      "loss": 4.7871,
      "mean_token_accuracy": 0.20512249171733857,
      "num_tokens": 12745597.0,
      "step": 6990
    },
    {
      "entropy": 5.2587878704071045,
      "epoch": 5.795360397680199,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024543096449251286,
      "loss": 4.6336,
      "mean_token_accuracy": 0.22447871565818786,
      "num_tokens": 12753492.0,
      "step": 6995
    },
    {
      "entropy": 5.272693634033203,
      "epoch": 5.79950289975145,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002451142097003177,
      "loss": 4.7795,
      "mean_token_accuracy": 0.2070946663618088,
      "num_tokens": 12761999.0,
      "step": 7000
    },
    {
      "epoch": 5.79950289975145,
      "eval_entropy": 5.147577573495633,
      "eval_loss": 5.778937816619873,
      "eval_mean_token_accuracy": 0.16521366807951857,
      "eval_num_tokens": 12761999.0,
      "eval_runtime": 2.7541,
      "eval_samples_per_second": 1360.519,
      "eval_steps_per_second": 170.292,
      "step": 7000
    },
    {
      "entropy": 5.260287237167359,
      "epoch": 5.803645401822701,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002447975151910276,
      "loss": 4.6991,
      "mean_token_accuracy": 0.2132972702383995,
      "num_tokens": 12770922.0,
      "step": 7005
    },
    {
      "entropy": 5.263891649246216,
      "epoch": 5.807787903893952,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024448088160344974,
      "loss": 4.7351,
      "mean_token_accuracy": 0.2070547252893448,
      "num_tokens": 12779539.0,
      "step": 7010
    },
    {
      "entropy": 5.276735162734985,
      "epoch": 5.811930405965203,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002441643095762688,
      "loss": 4.7366,
      "mean_token_accuracy": 0.21310780495405196,
      "num_tokens": 12788121.0,
      "step": 7015
    },
    {
      "entropy": 5.19331693649292,
      "epoch": 5.816072908036454,
      "grad_norm": 1.3125,
      "learning_rate": 0.000243847799748045,
      "loss": 4.625,
      "mean_token_accuracy": 0.2138730838894844,
      "num_tokens": 12797390.0,
      "step": 7020
    },
    {
      "entropy": 5.278093862533569,
      "epoch": 5.820215410107705,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024353135275721356,
      "loss": 4.8643,
      "mean_token_accuracy": 0.19812586903572083,
      "num_tokens": 12807876.0,
      "step": 7025
    },
    {
      "entropy": 5.222770547866821,
      "epoch": 5.824357912178956,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024321496924208235,
      "loss": 4.6698,
      "mean_token_accuracy": 0.2127540647983551,
      "num_tokens": 12816496.0,
      "step": 7030
    },
    {
      "entropy": 5.259526538848877,
      "epoch": 5.8285004142502075,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024289864984083166,
      "loss": 4.7574,
      "mean_token_accuracy": 0.21792919486761092,
      "num_tokens": 12825925.0,
      "step": 7035
    },
    {
      "entropy": 5.314914655685425,
      "epoch": 5.832642916321458,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002425823951915122,
      "loss": 4.7673,
      "mean_token_accuracy": 0.20973569005727768,
      "num_tokens": 12835479.0,
      "step": 7040
    },
    {
      "entropy": 5.204759502410889,
      "epoch": 5.836785418392709,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002422662059320442,
      "loss": 4.6815,
      "mean_token_accuracy": 0.215966072678566,
      "num_tokens": 12844024.0,
      "step": 7045
    },
    {
      "entropy": 5.23098201751709,
      "epoch": 5.84092792046396,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024195008270021595,
      "loss": 4.6855,
      "mean_token_accuracy": 0.20993933379650115,
      "num_tokens": 12852673.0,
      "step": 7050
    },
    {
      "entropy": 5.195737218856811,
      "epoch": 5.845070422535211,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024163402613368252,
      "loss": 4.7031,
      "mean_token_accuracy": 0.21670178920030594,
      "num_tokens": 12861191.0,
      "step": 7055
    },
    {
      "entropy": 5.288508749008178,
      "epoch": 5.849212924606462,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024131803686996457,
      "loss": 4.8364,
      "mean_token_accuracy": 0.2111321970820427,
      "num_tokens": 12870874.0,
      "step": 7060
    },
    {
      "entropy": 5.324412298202515,
      "epoch": 5.853355426677713,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024100211554644696,
      "loss": 4.7583,
      "mean_token_accuracy": 0.2117256596684456,
      "num_tokens": 12879896.0,
      "step": 7065
    },
    {
      "entropy": 5.279787683486939,
      "epoch": 5.857497928748964,
      "grad_norm": 1.375,
      "learning_rate": 0.00024068626280037753,
      "loss": 4.8001,
      "mean_token_accuracy": 0.20675662904977798,
      "num_tokens": 12888665.0,
      "step": 7070
    },
    {
      "entropy": 5.216130781173706,
      "epoch": 5.8616404308202155,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024037047926886578,
      "loss": 4.7429,
      "mean_token_accuracy": 0.20466429889202117,
      "num_tokens": 12897477.0,
      "step": 7075
    },
    {
      "entropy": 5.243073797225952,
      "epoch": 5.865782932891467,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024005476558888172,
      "loss": 4.7805,
      "mean_token_accuracy": 0.2017079696059227,
      "num_tokens": 12905934.0,
      "step": 7080
    },
    {
      "entropy": 5.28822956085205,
      "epoch": 5.869925434962718,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002397391223972541,
      "loss": 4.7859,
      "mean_token_accuracy": 0.20335040241479874,
      "num_tokens": 12915092.0,
      "step": 7085
    },
    {
      "entropy": 5.299696826934815,
      "epoch": 5.874067937033969,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023942355033066993,
      "loss": 4.866,
      "mean_token_accuracy": 0.20642060339450835,
      "num_tokens": 12924775.0,
      "step": 7090
    },
    {
      "entropy": 5.34201078414917,
      "epoch": 5.87821043910522,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023910805002567244,
      "loss": 4.7847,
      "mean_token_accuracy": 0.20166662335395813,
      "num_tokens": 12933677.0,
      "step": 7095
    },
    {
      "entropy": 5.293143224716187,
      "epoch": 5.882352941176471,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002387926221186603,
      "loss": 4.8081,
      "mean_token_accuracy": 0.2079706996679306,
      "num_tokens": 12942416.0,
      "step": 7100
    },
    {
      "entropy": 5.262480068206787,
      "epoch": 5.886495443247721,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023847726724588608,
      "loss": 4.785,
      "mean_token_accuracy": 0.21369751244783403,
      "num_tokens": 12952284.0,
      "step": 7105
    },
    {
      "entropy": 5.281225109100342,
      "epoch": 5.890637945318972,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002381619860434549,
      "loss": 4.8203,
      "mean_token_accuracy": 0.20545052736997604,
      "num_tokens": 12960694.0,
      "step": 7110
    },
    {
      "entropy": 5.235872554779053,
      "epoch": 5.894780447390223,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023784677914732356,
      "loss": 4.7394,
      "mean_token_accuracy": 0.21037458628416061,
      "num_tokens": 12969138.0,
      "step": 7115
    },
    {
      "entropy": 5.254066467285156,
      "epoch": 5.8989229494614746,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023753164719329868,
      "loss": 4.7607,
      "mean_token_accuracy": 0.20815321952104568,
      "num_tokens": 12978518.0,
      "step": 7120
    },
    {
      "entropy": 5.296635866165161,
      "epoch": 5.903065451532726,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023721659081703595,
      "loss": 4.7947,
      "mean_token_accuracy": 0.21491542011499404,
      "num_tokens": 12987506.0,
      "step": 7125
    },
    {
      "entropy": 5.276184558868408,
      "epoch": 5.907207953603977,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023690161065403838,
      "loss": 4.7789,
      "mean_token_accuracy": 0.19988380372524261,
      "num_tokens": 12996259.0,
      "step": 7130
    },
    {
      "entropy": 5.283065223693848,
      "epoch": 5.911350455675228,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002365867073396555,
      "loss": 4.7948,
      "mean_token_accuracy": 0.2045303091406822,
      "num_tokens": 13005430.0,
      "step": 7135
    },
    {
      "entropy": 5.252687072753906,
      "epoch": 5.915492957746479,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023627188150908174,
      "loss": 4.7581,
      "mean_token_accuracy": 0.21460192352533342,
      "num_tokens": 13014915.0,
      "step": 7140
    },
    {
      "entropy": 5.259900426864624,
      "epoch": 5.91963545981773,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023595713379735494,
      "loss": 4.7408,
      "mean_token_accuracy": 0.2052226707339287,
      "num_tokens": 13023471.0,
      "step": 7145
    },
    {
      "entropy": 5.302665424346924,
      "epoch": 5.923777961888981,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023564246483935602,
      "loss": 4.7136,
      "mean_token_accuracy": 0.20728942900896072,
      "num_tokens": 13033270.0,
      "step": 7150
    },
    {
      "entropy": 5.256060028076172,
      "epoch": 5.927920463960232,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002353278752698063,
      "loss": 4.6936,
      "mean_token_accuracy": 0.2116224244236946,
      "num_tokens": 13043300.0,
      "step": 7155
    },
    {
      "entropy": 5.246320581436157,
      "epoch": 5.9320629660314825,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002350133657232676,
      "loss": 4.7725,
      "mean_token_accuracy": 0.20378205180168152,
      "num_tokens": 13052152.0,
      "step": 7160
    },
    {
      "entropy": 5.305574131011963,
      "epoch": 5.9362054681027345,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002346989368341399,
      "loss": 4.8076,
      "mean_token_accuracy": 0.20854731053113937,
      "num_tokens": 13061704.0,
      "step": 7165
    },
    {
      "entropy": 5.284542512893677,
      "epoch": 5.940347970173985,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023438458923666057,
      "loss": 4.7279,
      "mean_token_accuracy": 0.2200355738401413,
      "num_tokens": 13071055.0,
      "step": 7170
    },
    {
      "entropy": 5.283852005004883,
      "epoch": 5.944490472245236,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002340703235649032,
      "loss": 4.8268,
      "mean_token_accuracy": 0.19662677496671677,
      "num_tokens": 13080023.0,
      "step": 7175
    },
    {
      "entropy": 5.302827405929565,
      "epoch": 5.948632974316487,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023375614045277593,
      "loss": 4.7712,
      "mean_token_accuracy": 0.20788494348526002,
      "num_tokens": 13088690.0,
      "step": 7180
    },
    {
      "entropy": 5.229621934890747,
      "epoch": 5.952775476387738,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023344204053402047,
      "loss": 4.741,
      "mean_token_accuracy": 0.21228492856025696,
      "num_tokens": 13097116.0,
      "step": 7185
    },
    {
      "entropy": 5.3526092052459715,
      "epoch": 5.956917978458989,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023312802444221066,
      "loss": 4.7924,
      "mean_token_accuracy": 0.20686232298612595,
      "num_tokens": 13106463.0,
      "step": 7190
    },
    {
      "entropy": 5.253214359283447,
      "epoch": 5.96106048053024,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002328140928107511,
      "loss": 4.7134,
      "mean_token_accuracy": 0.21423248201608658,
      "num_tokens": 13115769.0,
      "step": 7195
    },
    {
      "entropy": 5.277383041381836,
      "epoch": 5.965202982601491,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002325002462728766,
      "loss": 4.8214,
      "mean_token_accuracy": 0.2103443995118141,
      "num_tokens": 13124928.0,
      "step": 7200
    },
    {
      "entropy": 5.3181921482086185,
      "epoch": 5.9693454846727425,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023218648546164945,
      "loss": 4.7427,
      "mean_token_accuracy": 0.21340778917074205,
      "num_tokens": 13134106.0,
      "step": 7205
    },
    {
      "entropy": 5.448113822937012,
      "epoch": 5.973487986743994,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002318728110099599,
      "loss": 4.9364,
      "mean_token_accuracy": 0.20559266060590745,
      "num_tokens": 13143019.0,
      "step": 7210
    },
    {
      "entropy": 5.292605590820313,
      "epoch": 5.977630488815245,
      "grad_norm": 1.875,
      "learning_rate": 0.00023155922355052339,
      "loss": 4.6535,
      "mean_token_accuracy": 0.22634452879428862,
      "num_tokens": 13152675.0,
      "step": 7215
    },
    {
      "entropy": 5.260331392288208,
      "epoch": 5.981772990886496,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023124572371588,
      "loss": 4.7115,
      "mean_token_accuracy": 0.21955879777669907,
      "num_tokens": 13161094.0,
      "step": 7220
    },
    {
      "entropy": 5.227987623214721,
      "epoch": 5.985915492957746,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023093231213839344,
      "loss": 4.7869,
      "mean_token_accuracy": 0.20912853479385377,
      "num_tokens": 13170858.0,
      "step": 7225
    },
    {
      "entropy": 5.25358304977417,
      "epoch": 5.990057995028997,
      "grad_norm": 1.25,
      "learning_rate": 0.0002306189894502488,
      "loss": 4.8591,
      "mean_token_accuracy": 0.20006619840860368,
      "num_tokens": 13179668.0,
      "step": 7230
    },
    {
      "entropy": 5.306322860717773,
      "epoch": 5.994200497100248,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023030575628345246,
      "loss": 4.6917,
      "mean_token_accuracy": 0.2124774843454361,
      "num_tokens": 13188312.0,
      "step": 7235
    },
    {
      "entropy": 5.259900140762329,
      "epoch": 5.998342999171499,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022999261326982985,
      "loss": 4.7608,
      "mean_token_accuracy": 0.21093282997608184,
      "num_tokens": 13196989.0,
      "step": 7240
    },
    {
      "entropy": 5.266181182861328,
      "epoch": 6.0024855012427505,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022967956104102445,
      "loss": 4.6347,
      "mean_token_accuracy": 0.22529437392950058,
      "num_tokens": 13205830.0,
      "step": 7245
    },
    {
      "entropy": 5.3379847526550295,
      "epoch": 6.006628003314002,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022936660022849714,
      "loss": 4.5829,
      "mean_token_accuracy": 0.2291080743074417,
      "num_tokens": 13214711.0,
      "step": 7250
    },
    {
      "entropy": 5.2594372749328615,
      "epoch": 6.010770505385253,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022905373146352382,
      "loss": 4.6497,
      "mean_token_accuracy": 0.21611782610416413,
      "num_tokens": 13223736.0,
      "step": 7255
    },
    {
      "entropy": 5.279043388366699,
      "epoch": 6.014913007456504,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002287409553771952,
      "loss": 4.5512,
      "mean_token_accuracy": 0.22671631425619126,
      "num_tokens": 13232666.0,
      "step": 7260
    },
    {
      "entropy": 5.282508039474488,
      "epoch": 6.019055509527755,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022842827260041455,
      "loss": 4.5749,
      "mean_token_accuracy": 0.2172274261713028,
      "num_tokens": 13241770.0,
      "step": 7265
    },
    {
      "entropy": 5.257007074356079,
      "epoch": 6.023198011599006,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022811568376389745,
      "loss": 4.5213,
      "mean_token_accuracy": 0.2256094142794609,
      "num_tokens": 13250475.0,
      "step": 7270
    },
    {
      "entropy": 5.235157680511475,
      "epoch": 6.027340513670257,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022780318949816963,
      "loss": 4.5148,
      "mean_token_accuracy": 0.22576210647821426,
      "num_tokens": 13259587.0,
      "step": 7275
    },
    {
      "entropy": 5.24161467552185,
      "epoch": 6.031483015741508,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022749079043356623,
      "loss": 4.526,
      "mean_token_accuracy": 0.23990538716316223,
      "num_tokens": 13269681.0,
      "step": 7280
    },
    {
      "entropy": 5.24618501663208,
      "epoch": 6.0356255178127585,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022717848720023022,
      "loss": 4.5268,
      "mean_token_accuracy": 0.23160936087369918,
      "num_tokens": 13278186.0,
      "step": 7285
    },
    {
      "entropy": 5.230877685546875,
      "epoch": 6.03976801988401,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022686628042811142,
      "loss": 4.6364,
      "mean_token_accuracy": 0.2146809756755829,
      "num_tokens": 13287194.0,
      "step": 7290
    },
    {
      "entropy": 5.346965312957764,
      "epoch": 6.043910521955261,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000226554170746965,
      "loss": 4.6592,
      "mean_token_accuracy": 0.2186690464615822,
      "num_tokens": 13296774.0,
      "step": 7295
    },
    {
      "entropy": 5.336505460739136,
      "epoch": 6.048053024026512,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022624215878635023,
      "loss": 4.5649,
      "mean_token_accuracy": 0.22838946878910066,
      "num_tokens": 13306368.0,
      "step": 7300
    },
    {
      "entropy": 5.304216289520264,
      "epoch": 6.052195526097763,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022593024517562945,
      "loss": 4.5558,
      "mean_token_accuracy": 0.2314501956105232,
      "num_tokens": 13315393.0,
      "step": 7305
    },
    {
      "entropy": 5.279039192199707,
      "epoch": 6.056338028169014,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022561843054396636,
      "loss": 4.5367,
      "mean_token_accuracy": 0.2303796961903572,
      "num_tokens": 13324384.0,
      "step": 7310
    },
    {
      "entropy": 5.297043371200561,
      "epoch": 6.060480530240265,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022530671552032521,
      "loss": 4.5911,
      "mean_token_accuracy": 0.2289274513721466,
      "num_tokens": 13333191.0,
      "step": 7315
    },
    {
      "entropy": 5.239851713180542,
      "epoch": 6.064623032311516,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002249951007334693,
      "loss": 4.5829,
      "mean_token_accuracy": 0.22110097259283065,
      "num_tokens": 13342911.0,
      "step": 7320
    },
    {
      "entropy": 5.324529314041138,
      "epoch": 6.068765534382767,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022468358681195966,
      "loss": 4.5777,
      "mean_token_accuracy": 0.2222077503800392,
      "num_tokens": 13352464.0,
      "step": 7325
    },
    {
      "entropy": 5.271308374404907,
      "epoch": 6.072908036454018,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022437217438415393,
      "loss": 4.5855,
      "mean_token_accuracy": 0.22377583682537078,
      "num_tokens": 13361812.0,
      "step": 7330
    },
    {
      "entropy": 5.30517201423645,
      "epoch": 6.0770505385252696,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022406086407820503,
      "loss": 4.6385,
      "mean_token_accuracy": 0.22239355444908143,
      "num_tokens": 13372080.0,
      "step": 7335
    },
    {
      "entropy": 5.304878187179566,
      "epoch": 6.081193040596521,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022374965652205975,
      "loss": 4.6989,
      "mean_token_accuracy": 0.21319249719381334,
      "num_tokens": 13380897.0,
      "step": 7340
    },
    {
      "entropy": 5.345545053482056,
      "epoch": 6.085335542667772,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022343855234345796,
      "loss": 4.6819,
      "mean_token_accuracy": 0.21658573895692826,
      "num_tokens": 13389885.0,
      "step": 7345
    },
    {
      "entropy": 5.270920658111573,
      "epoch": 6.089478044739022,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022312755216993065,
      "loss": 4.6002,
      "mean_token_accuracy": 0.22368054687976838,
      "num_tokens": 13399734.0,
      "step": 7350
    },
    {
      "entropy": 5.220838260650635,
      "epoch": 6.093620546810273,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022281665662879903,
      "loss": 4.553,
      "mean_token_accuracy": 0.22898369729518891,
      "num_tokens": 13408641.0,
      "step": 7355
    },
    {
      "entropy": 5.246887588500977,
      "epoch": 6.097763048881524,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022250586634717362,
      "loss": 4.5968,
      "mean_token_accuracy": 0.22009286880493165,
      "num_tokens": 13418585.0,
      "step": 7360
    },
    {
      "entropy": 5.261246633529663,
      "epoch": 6.101905550952775,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022219518195195216,
      "loss": 4.5206,
      "mean_token_accuracy": 0.22703736275434494,
      "num_tokens": 13427897.0,
      "step": 7365
    },
    {
      "entropy": 5.145835208892822,
      "epoch": 6.106048053024026,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022188460406981925,
      "loss": 4.4938,
      "mean_token_accuracy": 0.22731132805347443,
      "num_tokens": 13437311.0,
      "step": 7370
    },
    {
      "entropy": 5.3025905132293705,
      "epoch": 6.1101905550952775,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022157413332724436,
      "loss": 4.5647,
      "mean_token_accuracy": 0.2319817513227463,
      "num_tokens": 13446359.0,
      "step": 7375
    },
    {
      "entropy": 5.309597969055176,
      "epoch": 6.114333057166529,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022126377035048072,
      "loss": 4.6076,
      "mean_token_accuracy": 0.224282668530941,
      "num_tokens": 13455302.0,
      "step": 7380
    },
    {
      "entropy": 5.340134811401367,
      "epoch": 6.11847555923778,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002209535157655646,
      "loss": 4.6821,
      "mean_token_accuracy": 0.2216806963086128,
      "num_tokens": 13466391.0,
      "step": 7385
    },
    {
      "entropy": 5.257795763015747,
      "epoch": 6.122618061309031,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022064337019831322,
      "loss": 4.5294,
      "mean_token_accuracy": 0.22612428516149521,
      "num_tokens": 13476130.0,
      "step": 7390
    },
    {
      "entropy": 5.179658079147339,
      "epoch": 6.126760563380282,
      "grad_norm": 1.6875,
      "learning_rate": 0.00022033333427432422,
      "loss": 4.5101,
      "mean_token_accuracy": 0.23560523688793183,
      "num_tokens": 13485105.0,
      "step": 7395
    },
    {
      "entropy": 5.251299571990967,
      "epoch": 6.130903065451533,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002200234086189738,
      "loss": 4.5861,
      "mean_token_accuracy": 0.21644460111856462,
      "num_tokens": 13494914.0,
      "step": 7400
    },
    {
      "entropy": 5.309022569656372,
      "epoch": 6.135045567522784,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021971359385741608,
      "loss": 4.6168,
      "mean_token_accuracy": 0.22761507630348204,
      "num_tokens": 13504293.0,
      "step": 7405
    },
    {
      "entropy": 5.378573846817017,
      "epoch": 6.139188069594034,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021940389061458115,
      "loss": 4.6942,
      "mean_token_accuracy": 0.2262974426150322,
      "num_tokens": 13513972.0,
      "step": 7410
    },
    {
      "entropy": 5.280871868133545,
      "epoch": 6.1433305716652855,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002190942995151743,
      "loss": 4.6214,
      "mean_token_accuracy": 0.2186795711517334,
      "num_tokens": 13522508.0,
      "step": 7415
    },
    {
      "entropy": 5.3220508098602295,
      "epoch": 6.147473073736537,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002187848211836746,
      "loss": 4.5227,
      "mean_token_accuracy": 0.2325556382536888,
      "num_tokens": 13531235.0,
      "step": 7420
    },
    {
      "entropy": 5.290721225738525,
      "epoch": 6.151615575807788,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002184754562443337,
      "loss": 4.552,
      "mean_token_accuracy": 0.2310631290078163,
      "num_tokens": 13541023.0,
      "step": 7425
    },
    {
      "entropy": 5.248855686187744,
      "epoch": 6.155758077879039,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021816620532117448,
      "loss": 4.5872,
      "mean_token_accuracy": 0.22816518545150757,
      "num_tokens": 13550622.0,
      "step": 7430
    },
    {
      "entropy": 5.257692337036133,
      "epoch": 6.15990057995029,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002178570690379899,
      "loss": 4.6074,
      "mean_token_accuracy": 0.21932708770036696,
      "num_tokens": 13559423.0,
      "step": 7435
    },
    {
      "entropy": 5.223868942260742,
      "epoch": 6.164043082021541,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021754804801834144,
      "loss": 4.5838,
      "mean_token_accuracy": 0.22118867188692093,
      "num_tokens": 13569306.0,
      "step": 7440
    },
    {
      "entropy": 5.22743239402771,
      "epoch": 6.168185584092792,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002172391428855585,
      "loss": 4.5444,
      "mean_token_accuracy": 0.22753443717956542,
      "num_tokens": 13578367.0,
      "step": 7445
    },
    {
      "entropy": 5.287085103988647,
      "epoch": 6.172328086164043,
      "grad_norm": 1.375,
      "learning_rate": 0.00021693035426273633,
      "loss": 4.5875,
      "mean_token_accuracy": 0.223337559401989,
      "num_tokens": 13586930.0,
      "step": 7450
    },
    {
      "entropy": 5.240332460403442,
      "epoch": 6.176470588235294,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021662168277273547,
      "loss": 4.5692,
      "mean_token_accuracy": 0.22664483785629272,
      "num_tokens": 13595891.0,
      "step": 7455
    },
    {
      "entropy": 5.32559986114502,
      "epoch": 6.1806130903065455,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002163131290381799,
      "loss": 4.6455,
      "mean_token_accuracy": 0.2130844384431839,
      "num_tokens": 13605786.0,
      "step": 7460
    },
    {
      "entropy": 5.268065881729126,
      "epoch": 6.184755592377797,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021600469368145624,
      "loss": 4.5417,
      "mean_token_accuracy": 0.2314801976084709,
      "num_tokens": 13614561.0,
      "step": 7465
    },
    {
      "entropy": 5.230492496490479,
      "epoch": 6.188898094449047,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002156963773247124,
      "loss": 4.5404,
      "mean_token_accuracy": 0.23531070798635484,
      "num_tokens": 13622853.0,
      "step": 7470
    },
    {
      "entropy": 5.208058404922485,
      "epoch": 6.193040596520298,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002153881805898561,
      "loss": 4.5268,
      "mean_token_accuracy": 0.22764941900968552,
      "num_tokens": 13631709.0,
      "step": 7475
    },
    {
      "entropy": 5.303780126571655,
      "epoch": 6.197183098591549,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002150801040985539,
      "loss": 4.6146,
      "mean_token_accuracy": 0.22441026866436004,
      "num_tokens": 13640854.0,
      "step": 7480
    },
    {
      "entropy": 5.210834121704101,
      "epoch": 6.2013256006628,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021477214847222958,
      "loss": 4.5465,
      "mean_token_accuracy": 0.22115834504365922,
      "num_tokens": 13649318.0,
      "step": 7485
    },
    {
      "entropy": 5.300684785842895,
      "epoch": 6.205468102734051,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021446431433206348,
      "loss": 4.6409,
      "mean_token_accuracy": 0.22462961375713347,
      "num_tokens": 13658738.0,
      "step": 7490
    },
    {
      "entropy": 5.152304601669312,
      "epoch": 6.209610604805302,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002141566022989907,
      "loss": 4.5493,
      "mean_token_accuracy": 0.22858153581619262,
      "num_tokens": 13667315.0,
      "step": 7495
    },
    {
      "entropy": 5.251782608032227,
      "epoch": 6.2137531068765535,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002138490129936998,
      "loss": 4.6604,
      "mean_token_accuracy": 0.21763822436332703,
      "num_tokens": 13676665.0,
      "step": 7500
    },
    {
      "epoch": 6.2137531068765535,
      "eval_entropy": 5.1646859366248155,
      "eval_loss": 5.785756587982178,
      "eval_mean_token_accuracy": 0.16465186759797748,
      "eval_num_tokens": 13676665.0,
      "eval_runtime": 2.7583,
      "eval_samples_per_second": 1358.427,
      "eval_steps_per_second": 170.03,
      "step": 7500
    },
    {
      "entropy": 5.27554235458374,
      "epoch": 6.217895608947805,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002135415470366323,
      "loss": 4.5764,
      "mean_token_accuracy": 0.2197849467396736,
      "num_tokens": 13685774.0,
      "step": 7505
    },
    {
      "entropy": 5.297122478485107,
      "epoch": 6.222038111019056,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021323420504798047,
      "loss": 4.6204,
      "mean_token_accuracy": 0.2204588070511818,
      "num_tokens": 13695700.0,
      "step": 7510
    },
    {
      "entropy": 5.313191175460815,
      "epoch": 6.226180613090307,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021292698764768675,
      "loss": 4.6384,
      "mean_token_accuracy": 0.22021855413913727,
      "num_tokens": 13704675.0,
      "step": 7515
    },
    {
      "entropy": 5.245783472061158,
      "epoch": 6.230323115161558,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021261989545544218,
      "loss": 4.6674,
      "mean_token_accuracy": 0.22652209997177125,
      "num_tokens": 13714562.0,
      "step": 7520
    },
    {
      "entropy": 5.311749649047852,
      "epoch": 6.234465617232809,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021231292909068534,
      "loss": 4.6169,
      "mean_token_accuracy": 0.22701555490493774,
      "num_tokens": 13723917.0,
      "step": 7525
    },
    {
      "entropy": 5.298685932159424,
      "epoch": 6.238608119304059,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021200608917260082,
      "loss": 4.5563,
      "mean_token_accuracy": 0.2285647138953209,
      "num_tokens": 13732482.0,
      "step": 7530
    },
    {
      "entropy": 5.306582593917847,
      "epoch": 6.24275062137531,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021169937632011838,
      "loss": 4.6429,
      "mean_token_accuracy": 0.22439165860414506,
      "num_tokens": 13742110.0,
      "step": 7535
    },
    {
      "entropy": 5.254005241394043,
      "epoch": 6.2468931234465614,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021139279115191123,
      "loss": 4.5787,
      "mean_token_accuracy": 0.2149302616715431,
      "num_tokens": 13750501.0,
      "step": 7540
    },
    {
      "entropy": 5.230757713317871,
      "epoch": 6.2510356255178126,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021108633428639534,
      "loss": 4.542,
      "mean_token_accuracy": 0.23032894283533095,
      "num_tokens": 13759786.0,
      "step": 7545
    },
    {
      "entropy": 5.274296474456787,
      "epoch": 6.255178127589064,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002107800063417275,
      "loss": 4.614,
      "mean_token_accuracy": 0.21465778052806855,
      "num_tokens": 13768476.0,
      "step": 7550
    },
    {
      "entropy": 5.288611221313476,
      "epoch": 6.259320629660315,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002104738079358048,
      "loss": 4.6444,
      "mean_token_accuracy": 0.20758169740438462,
      "num_tokens": 13778721.0,
      "step": 7555
    },
    {
      "entropy": 5.310076522827148,
      "epoch": 6.263463131731566,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002101677396862628,
      "loss": 4.6131,
      "mean_token_accuracy": 0.22510013729333878,
      "num_tokens": 13787108.0,
      "step": 7560
    },
    {
      "entropy": 5.224409532546997,
      "epoch": 6.267605633802817,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020986180221047463,
      "loss": 4.529,
      "mean_token_accuracy": 0.2325158953666687,
      "num_tokens": 13795021.0,
      "step": 7565
    },
    {
      "entropy": 5.214485740661621,
      "epoch": 6.271748135874068,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020955599612554966,
      "loss": 4.5873,
      "mean_token_accuracy": 0.22493835538625717,
      "num_tokens": 13804750.0,
      "step": 7570
    },
    {
      "entropy": 5.211126518249512,
      "epoch": 6.275890637945319,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020925032204833206,
      "loss": 4.5706,
      "mean_token_accuracy": 0.2246261343359947,
      "num_tokens": 13813910.0,
      "step": 7575
    },
    {
      "entropy": 5.337267208099365,
      "epoch": 6.28003314001657,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020894478059539995,
      "loss": 4.5359,
      "mean_token_accuracy": 0.22401334643363952,
      "num_tokens": 13822832.0,
      "step": 7580
    },
    {
      "entropy": 5.209841537475586,
      "epoch": 6.284175642087821,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020863937238306375,
      "loss": 4.5456,
      "mean_token_accuracy": 0.2343209609389305,
      "num_tokens": 13831951.0,
      "step": 7585
    },
    {
      "entropy": 5.173276138305664,
      "epoch": 6.2883181441590725,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020833409802736535,
      "loss": 4.5256,
      "mean_token_accuracy": 0.23267557322978974,
      "num_tokens": 13840422.0,
      "step": 7590
    },
    {
      "entropy": 5.287439966201783,
      "epoch": 6.292460646230323,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020802895814407636,
      "loss": 4.6094,
      "mean_token_accuracy": 0.2281896561384201,
      "num_tokens": 13850903.0,
      "step": 7595
    },
    {
      "entropy": 5.2715085506439205,
      "epoch": 6.296603148301574,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002077239533486972,
      "loss": 4.6792,
      "mean_token_accuracy": 0.2122771292924881,
      "num_tokens": 13859972.0,
      "step": 7600
    },
    {
      "entropy": 5.288576269149781,
      "epoch": 6.300745650372825,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020741908425645606,
      "loss": 4.6357,
      "mean_token_accuracy": 0.2200139805674553,
      "num_tokens": 13868980.0,
      "step": 7605
    },
    {
      "entropy": 5.261795234680176,
      "epoch": 6.304888152444076,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020711435148230707,
      "loss": 4.5812,
      "mean_token_accuracy": 0.22372327297925948,
      "num_tokens": 13877901.0,
      "step": 7610
    },
    {
      "entropy": 5.205504703521728,
      "epoch": 6.309030654515327,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020680975564092968,
      "loss": 4.5804,
      "mean_token_accuracy": 0.2176283985376358,
      "num_tokens": 13887081.0,
      "step": 7615
    },
    {
      "entropy": 5.293874454498291,
      "epoch": 6.313173156586578,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020650529734672686,
      "loss": 4.6802,
      "mean_token_accuracy": 0.22097607105970382,
      "num_tokens": 13896645.0,
      "step": 7620
    },
    {
      "entropy": 5.372445106506348,
      "epoch": 6.317315658657829,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020620097721382424,
      "loss": 4.6154,
      "mean_token_accuracy": 0.22537714540958403,
      "num_tokens": 13905975.0,
      "step": 7625
    },
    {
      "entropy": 5.31709041595459,
      "epoch": 6.3214581607290805,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002058967958560689,
      "loss": 4.5948,
      "mean_token_accuracy": 0.22527567744255067,
      "num_tokens": 13914605.0,
      "step": 7630
    },
    {
      "entropy": 5.19299464225769,
      "epoch": 6.325600662800332,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020559275388702775,
      "loss": 4.5124,
      "mean_token_accuracy": 0.23811782151460648,
      "num_tokens": 13924592.0,
      "step": 7635
    },
    {
      "entropy": 5.178159570693969,
      "epoch": 6.329743164871583,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020528885191998682,
      "loss": 4.5703,
      "mean_token_accuracy": 0.22409040480852127,
      "num_tokens": 13933513.0,
      "step": 7640
    },
    {
      "entropy": 5.153306102752685,
      "epoch": 6.333885666942834,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020498509056794952,
      "loss": 4.4654,
      "mean_token_accuracy": 0.23787264227867128,
      "num_tokens": 13942173.0,
      "step": 7645
    },
    {
      "entropy": 5.175963068008423,
      "epoch": 6.338028169014084,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020468147044363556,
      "loss": 4.4823,
      "mean_token_accuracy": 0.23301801979541778,
      "num_tokens": 13950205.0,
      "step": 7650
    },
    {
      "entropy": 5.228364753723144,
      "epoch": 6.342170671085335,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020437799215948017,
      "loss": 4.572,
      "mean_token_accuracy": 0.22810757607221605,
      "num_tokens": 13958956.0,
      "step": 7655
    },
    {
      "entropy": 5.239571762084961,
      "epoch": 6.346313173156586,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002040746563276319,
      "loss": 4.5771,
      "mean_token_accuracy": 0.22594334185123444,
      "num_tokens": 13967718.0,
      "step": 7660
    },
    {
      "entropy": 5.231808376312256,
      "epoch": 6.350455675227837,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020377146355995257,
      "loss": 4.6749,
      "mean_token_accuracy": 0.212095183134079,
      "num_tokens": 13976841.0,
      "step": 7665
    },
    {
      "entropy": 5.249170732498169,
      "epoch": 6.3545981772990885,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020346841446801495,
      "loss": 4.5391,
      "mean_token_accuracy": 0.23190926164388656,
      "num_tokens": 13984946.0,
      "step": 7670
    },
    {
      "entropy": 5.242449712753296,
      "epoch": 6.35874067937034,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020316550966310222,
      "loss": 4.6077,
      "mean_token_accuracy": 0.22002114057540895,
      "num_tokens": 13993521.0,
      "step": 7675
    },
    {
      "entropy": 5.212237119674683,
      "epoch": 6.362883181441591,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002028627497562065,
      "loss": 4.5656,
      "mean_token_accuracy": 0.2195043072104454,
      "num_tokens": 14002777.0,
      "step": 7680
    },
    {
      "entropy": 5.252022933959961,
      "epoch": 6.367025683512842,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020256013535802748,
      "loss": 4.6267,
      "mean_token_accuracy": 0.218884114921093,
      "num_tokens": 14012708.0,
      "step": 7685
    },
    {
      "entropy": 5.269517469406128,
      "epoch": 6.371168185584093,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002022576670789717,
      "loss": 4.6345,
      "mean_token_accuracy": 0.2195138677954674,
      "num_tokens": 14021088.0,
      "step": 7690
    },
    {
      "entropy": 5.245950841903687,
      "epoch": 6.375310687655344,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002019553455291505,
      "loss": 4.6405,
      "mean_token_accuracy": 0.2208177551627159,
      "num_tokens": 14031650.0,
      "step": 7695
    },
    {
      "entropy": 5.270649814605713,
      "epoch": 6.379453189726595,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002016531713183797,
      "loss": 4.6548,
      "mean_token_accuracy": 0.22094330042600632,
      "num_tokens": 14040741.0,
      "step": 7700
    },
    {
      "entropy": 5.246001815795898,
      "epoch": 6.383595691797846,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020135114505617752,
      "loss": 4.6229,
      "mean_token_accuracy": 0.2128319948911667,
      "num_tokens": 14049913.0,
      "step": 7705
    },
    {
      "entropy": 5.271371746063233,
      "epoch": 6.387738193869097,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020104926735176393,
      "loss": 4.594,
      "mean_token_accuracy": 0.2289719820022583,
      "num_tokens": 14059583.0,
      "step": 7710
    },
    {
      "entropy": 5.318186616897583,
      "epoch": 6.391880695940348,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020074753881405949,
      "loss": 4.5518,
      "mean_token_accuracy": 0.226786370575428,
      "num_tokens": 14067376.0,
      "step": 7715
    },
    {
      "entropy": 5.244753456115722,
      "epoch": 6.396023198011599,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002004459600516833,
      "loss": 4.6302,
      "mean_token_accuracy": 0.22130897492170334,
      "num_tokens": 14076415.0,
      "step": 7720
    },
    {
      "entropy": 5.263324642181397,
      "epoch": 6.40016570008285,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020014453167295298,
      "loss": 4.6138,
      "mean_token_accuracy": 0.21900541037321092,
      "num_tokens": 14085418.0,
      "step": 7725
    },
    {
      "entropy": 5.252635478973389,
      "epoch": 6.404308202154101,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019984325428588229,
      "loss": 4.5924,
      "mean_token_accuracy": 0.22715059965848922,
      "num_tokens": 14094454.0,
      "step": 7730
    },
    {
      "entropy": 5.240621948242188,
      "epoch": 6.408450704225352,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019954212849818077,
      "loss": 4.5292,
      "mean_token_accuracy": 0.23717418760061265,
      "num_tokens": 14103611.0,
      "step": 7735
    },
    {
      "entropy": 5.178173923492432,
      "epoch": 6.412593206296603,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019924115491725197,
      "loss": 4.4815,
      "mean_token_accuracy": 0.2312372922897339,
      "num_tokens": 14112145.0,
      "step": 7740
    },
    {
      "entropy": 5.236668729782105,
      "epoch": 6.416735708367854,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019894033415019253,
      "loss": 4.6137,
      "mean_token_accuracy": 0.21731938421726227,
      "num_tokens": 14121215.0,
      "step": 7745
    },
    {
      "entropy": 5.189931869506836,
      "epoch": 6.420878210439105,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019863966680379086,
      "loss": 4.5011,
      "mean_token_accuracy": 0.22405548691749572,
      "num_tokens": 14130106.0,
      "step": 7750
    },
    {
      "entropy": 5.183213758468628,
      "epoch": 6.4250207125103564,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001983391534845258,
      "loss": 4.5297,
      "mean_token_accuracy": 0.23860106468200684,
      "num_tokens": 14139071.0,
      "step": 7755
    },
    {
      "entropy": 5.213570594787598,
      "epoch": 6.4291632145816076,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019803879479856547,
      "loss": 4.5704,
      "mean_token_accuracy": 0.22312580049037933,
      "num_tokens": 14148018.0,
      "step": 7760
    },
    {
      "entropy": 5.198269939422607,
      "epoch": 6.433305716652859,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019773859135176626,
      "loss": 4.533,
      "mean_token_accuracy": 0.22430772781372071,
      "num_tokens": 14156369.0,
      "step": 7765
    },
    {
      "entropy": 5.217215204238892,
      "epoch": 6.43744821872411,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001974385437496713,
      "loss": 4.6167,
      "mean_token_accuracy": 0.2246893674135208,
      "num_tokens": 14165930.0,
      "step": 7770
    },
    {
      "entropy": 5.220076560974121,
      "epoch": 6.44159072079536,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019713865259750945,
      "loss": 4.5778,
      "mean_token_accuracy": 0.22422850430011748,
      "num_tokens": 14175557.0,
      "step": 7775
    },
    {
      "entropy": 5.290918493270874,
      "epoch": 6.445733222866611,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019683891850019391,
      "loss": 4.6409,
      "mean_token_accuracy": 0.21857522875070573,
      "num_tokens": 14185532.0,
      "step": 7780
    },
    {
      "entropy": 5.333722400665283,
      "epoch": 6.449875724937862,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019653934206232098,
      "loss": 4.7372,
      "mean_token_accuracy": 0.21307871043682097,
      "num_tokens": 14194307.0,
      "step": 7785
    },
    {
      "entropy": 5.297997760772705,
      "epoch": 6.454018227009113,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019623992388816935,
      "loss": 4.6676,
      "mean_token_accuracy": 0.22233052998781205,
      "num_tokens": 14203900.0,
      "step": 7790
    },
    {
      "entropy": 5.278000116348267,
      "epoch": 6.458160729080364,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001959406645816979,
      "loss": 4.5959,
      "mean_token_accuracy": 0.22168540805578232,
      "num_tokens": 14213232.0,
      "step": 7795
    },
    {
      "entropy": 5.233169317245483,
      "epoch": 6.4623032311516155,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019564156474654554,
      "loss": 4.542,
      "mean_token_accuracy": 0.22004218995571137,
      "num_tokens": 14221365.0,
      "step": 7800
    },
    {
      "entropy": 5.174768257141113,
      "epoch": 6.466445733222867,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019534262498602928,
      "loss": 4.5647,
      "mean_token_accuracy": 0.21907651722431182,
      "num_tokens": 14229878.0,
      "step": 7805
    },
    {
      "entropy": 5.2641740322113035,
      "epoch": 6.470588235294118,
      "grad_norm": 1.5,
      "learning_rate": 0.00019504384590314318,
      "loss": 4.6114,
      "mean_token_accuracy": 0.21852952986955643,
      "num_tokens": 14239132.0,
      "step": 7810
    },
    {
      "entropy": 5.308596181869507,
      "epoch": 6.474730737365369,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019474522810055745,
      "loss": 4.6157,
      "mean_token_accuracy": 0.22842988669872283,
      "num_tokens": 14247116.0,
      "step": 7815
    },
    {
      "entropy": 5.2250322818756105,
      "epoch": 6.47887323943662,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019444677218061673,
      "loss": 4.5981,
      "mean_token_accuracy": 0.22097215950489044,
      "num_tokens": 14255423.0,
      "step": 7820
    },
    {
      "entropy": 5.205509424209595,
      "epoch": 6.483015741507871,
      "grad_norm": 1.5,
      "learning_rate": 0.0001941484787453393,
      "loss": 4.5542,
      "mean_token_accuracy": 0.2294921875,
      "num_tokens": 14263836.0,
      "step": 7825
    },
    {
      "entropy": 5.199132061004638,
      "epoch": 6.487158243579122,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019385034839641558,
      "loss": 4.6596,
      "mean_token_accuracy": 0.2233857184648514,
      "num_tokens": 14273460.0,
      "step": 7830
    },
    {
      "entropy": 5.309156036376953,
      "epoch": 6.491300745650372,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001935523817352071,
      "loss": 4.6862,
      "mean_token_accuracy": 0.2182077720761299,
      "num_tokens": 14282925.0,
      "step": 7835
    },
    {
      "entropy": 5.278367805480957,
      "epoch": 6.4954432477216235,
      "grad_norm": 1.375,
      "learning_rate": 0.00019325457936274516,
      "loss": 4.5739,
      "mean_token_accuracy": 0.22955656200647354,
      "num_tokens": 14292059.0,
      "step": 7840
    },
    {
      "entropy": 5.25195198059082,
      "epoch": 6.499585749792875,
      "grad_norm": 1.5,
      "learning_rate": 0.00019295694187972955,
      "loss": 4.6196,
      "mean_token_accuracy": 0.22195079773664475,
      "num_tokens": 14301276.0,
      "step": 7845
    },
    {
      "entropy": 5.241566562652588,
      "epoch": 6.503728251864126,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019265946988652783,
      "loss": 4.6129,
      "mean_token_accuracy": 0.22002605795860292,
      "num_tokens": 14310259.0,
      "step": 7850
    },
    {
      "entropy": 5.249344348907471,
      "epoch": 6.507870753935377,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001923621639831733,
      "loss": 4.5115,
      "mean_token_accuracy": 0.2289004623889923,
      "num_tokens": 14318782.0,
      "step": 7855
    },
    {
      "entropy": 5.194301557540894,
      "epoch": 6.512013256006628,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019206502476936467,
      "loss": 4.4918,
      "mean_token_accuracy": 0.2372380942106247,
      "num_tokens": 14327571.0,
      "step": 7860
    },
    {
      "entropy": 5.149231624603272,
      "epoch": 6.516155758077879,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019176805284446414,
      "loss": 4.5138,
      "mean_token_accuracy": 0.23662362098693848,
      "num_tokens": 14336974.0,
      "step": 7865
    },
    {
      "entropy": 5.310570621490479,
      "epoch": 6.52029826014913,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019147124880749633,
      "loss": 4.6854,
      "mean_token_accuracy": 0.2186918169260025,
      "num_tokens": 14346367.0,
      "step": 7870
    },
    {
      "entropy": 5.280822849273681,
      "epoch": 6.524440762220381,
      "grad_norm": 1.375,
      "learning_rate": 0.00019117461325714768,
      "loss": 4.621,
      "mean_token_accuracy": 0.2232021763920784,
      "num_tokens": 14355641.0,
      "step": 7875
    },
    {
      "entropy": 5.255128335952759,
      "epoch": 6.528583264291632,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019087814679176436,
      "loss": 4.5581,
      "mean_token_accuracy": 0.22364246398210524,
      "num_tokens": 14364927.0,
      "step": 7880
    },
    {
      "entropy": 5.262372398376465,
      "epoch": 6.5327257663628835,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019058185000935175,
      "loss": 4.5182,
      "mean_token_accuracy": 0.2352157860994339,
      "num_tokens": 14373370.0,
      "step": 7885
    },
    {
      "entropy": 5.247505187988281,
      "epoch": 6.536868268434135,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019028572350757278,
      "loss": 4.5891,
      "mean_token_accuracy": 0.22501114308834075,
      "num_tokens": 14381820.0,
      "step": 7890
    },
    {
      "entropy": 5.2754909038543705,
      "epoch": 6.541010770505386,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018998976788374682,
      "loss": 4.5906,
      "mean_token_accuracy": 0.22035637646913528,
      "num_tokens": 14391158.0,
      "step": 7895
    },
    {
      "entropy": 5.244005107879639,
      "epoch": 6.545153272576636,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018969398373484898,
      "loss": 4.6212,
      "mean_token_accuracy": 0.22284531891345977,
      "num_tokens": 14399945.0,
      "step": 7900
    },
    {
      "entropy": 5.209062242507935,
      "epoch": 6.549295774647887,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018939837165750796,
      "loss": 4.5373,
      "mean_token_accuracy": 0.22396612912416458,
      "num_tokens": 14410275.0,
      "step": 7905
    },
    {
      "entropy": 5.287474346160889,
      "epoch": 6.553438276719138,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018910293224800584,
      "loss": 4.6253,
      "mean_token_accuracy": 0.2247585117816925,
      "num_tokens": 14420369.0,
      "step": 7910
    },
    {
      "entropy": 5.272275781631469,
      "epoch": 6.557580778790389,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018880766610227608,
      "loss": 4.6158,
      "mean_token_accuracy": 0.22317070811986922,
      "num_tokens": 14429078.0,
      "step": 7915
    },
    {
      "entropy": 5.198788452148437,
      "epoch": 6.56172328086164,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018851257381590264,
      "loss": 4.5659,
      "mean_token_accuracy": 0.2238755017518997,
      "num_tokens": 14438920.0,
      "step": 7920
    },
    {
      "entropy": 5.322684955596924,
      "epoch": 6.5658657829328915,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018821765598411917,
      "loss": 4.7707,
      "mean_token_accuracy": 0.20722351968288422,
      "num_tokens": 14449164.0,
      "step": 7925
    },
    {
      "entropy": 5.258245038986206,
      "epoch": 6.570008285004143,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018792291320180693,
      "loss": 4.6727,
      "mean_token_accuracy": 0.2097511887550354,
      "num_tokens": 14459418.0,
      "step": 7930
    },
    {
      "entropy": 5.303048753738404,
      "epoch": 6.574150787075394,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018762834606349455,
      "loss": 4.5937,
      "mean_token_accuracy": 0.2283338114619255,
      "num_tokens": 14468691.0,
      "step": 7935
    },
    {
      "entropy": 5.319290208816528,
      "epoch": 6.578293289146645,
      "grad_norm": 1.25,
      "learning_rate": 0.00018733395516335588,
      "loss": 4.6525,
      "mean_token_accuracy": 0.21864359378814696,
      "num_tokens": 14477936.0,
      "step": 7940
    },
    {
      "entropy": 5.276194095611572,
      "epoch": 6.582435791217896,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018703974109520976,
      "loss": 4.6318,
      "mean_token_accuracy": 0.2249257370829582,
      "num_tokens": 14487191.0,
      "step": 7945
    },
    {
      "entropy": 5.208021402359009,
      "epoch": 6.586578293289147,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018674570445251798,
      "loss": 4.5872,
      "mean_token_accuracy": 0.22356925159692764,
      "num_tokens": 14495731.0,
      "step": 7950
    },
    {
      "entropy": 5.175275182723999,
      "epoch": 6.590720795360397,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018645184582838465,
      "loss": 4.5618,
      "mean_token_accuracy": 0.22906572222709656,
      "num_tokens": 14505009.0,
      "step": 7955
    },
    {
      "entropy": 5.208952713012695,
      "epoch": 6.594863297431648,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018615816581555466,
      "loss": 4.6223,
      "mean_token_accuracy": 0.22183820903301238,
      "num_tokens": 14513843.0,
      "step": 7960
    },
    {
      "entropy": 5.204042577743531,
      "epoch": 6.5990057995028994,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001858646650064128,
      "loss": 4.6099,
      "mean_token_accuracy": 0.2285928636789322,
      "num_tokens": 14523405.0,
      "step": 7965
    },
    {
      "entropy": 5.288896274566651,
      "epoch": 6.603148301574151,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018557134399298215,
      "loss": 4.6753,
      "mean_token_accuracy": 0.22212832421064377,
      "num_tokens": 14532615.0,
      "step": 7970
    },
    {
      "entropy": 5.294111537933349,
      "epoch": 6.607290803645402,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001852782033669233,
      "loss": 4.5925,
      "mean_token_accuracy": 0.2272229015827179,
      "num_tokens": 14541297.0,
      "step": 7975
    },
    {
      "entropy": 5.213164806365967,
      "epoch": 6.611433305716653,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018498524371953295,
      "loss": 4.5626,
      "mean_token_accuracy": 0.2263799250125885,
      "num_tokens": 14549867.0,
      "step": 7980
    },
    {
      "entropy": 5.224558734893799,
      "epoch": 6.615575807787904,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018469246564174275,
      "loss": 4.6592,
      "mean_token_accuracy": 0.21291491091251374,
      "num_tokens": 14558708.0,
      "step": 7985
    },
    {
      "entropy": 5.24251537322998,
      "epoch": 6.619718309859155,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000184399869724118,
      "loss": 4.6346,
      "mean_token_accuracy": 0.21831366419792175,
      "num_tokens": 14567917.0,
      "step": 7990
    },
    {
      "entropy": 5.204243993759155,
      "epoch": 6.623860811930406,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018410745655685662,
      "loss": 4.5045,
      "mean_token_accuracy": 0.2368050381541252,
      "num_tokens": 14576417.0,
      "step": 7995
    },
    {
      "entropy": 5.229977369308472,
      "epoch": 6.628003314001657,
      "grad_norm": 1.5,
      "learning_rate": 0.00018381522672978806,
      "loss": 4.6018,
      "mean_token_accuracy": 0.22134475708007811,
      "num_tokens": 14585172.0,
      "step": 8000
    },
    {
      "epoch": 6.628003314001657,
      "eval_entropy": 5.170466900888536,
      "eval_loss": 5.771762847900391,
      "eval_mean_token_accuracy": 0.16584313252587307,
      "eval_num_tokens": 14585172.0,
      "eval_runtime": 2.7564,
      "eval_samples_per_second": 1359.395,
      "eval_steps_per_second": 170.151,
      "step": 8000
    },
    {
      "entropy": 5.283796501159668,
      "epoch": 6.632145816072908,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018352318083237167,
      "loss": 4.6133,
      "mean_token_accuracy": 0.21931763589382172,
      "num_tokens": 14593758.0,
      "step": 8005
    },
    {
      "entropy": 5.25171275138855,
      "epoch": 6.636288318144159,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018323131945369603,
      "loss": 4.6026,
      "mean_token_accuracy": 0.22325890362262726,
      "num_tokens": 14602558.0,
      "step": 8010
    },
    {
      "entropy": 5.286454439163208,
      "epoch": 6.6404308202154105,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018293964318247735,
      "loss": 4.5868,
      "mean_token_accuracy": 0.2302074611186981,
      "num_tokens": 14611463.0,
      "step": 8015
    },
    {
      "entropy": 5.298673295974732,
      "epoch": 6.644573322286661,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018264815260705863,
      "loss": 4.6359,
      "mean_token_accuracy": 0.22209562808275224,
      "num_tokens": 14619951.0,
      "step": 8020
    },
    {
      "entropy": 5.297376918792724,
      "epoch": 6.648715824357912,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018235684831540826,
      "loss": 4.6486,
      "mean_token_accuracy": 0.21894189715385437,
      "num_tokens": 14629332.0,
      "step": 8025
    },
    {
      "entropy": 5.236829566955566,
      "epoch": 6.652858326429163,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018206573089511862,
      "loss": 4.5896,
      "mean_token_accuracy": 0.22981232553720474,
      "num_tokens": 14639010.0,
      "step": 8030
    },
    {
      "entropy": 5.236099290847778,
      "epoch": 6.657000828500414,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018177480093340564,
      "loss": 4.6374,
      "mean_token_accuracy": 0.2158491298556328,
      "num_tokens": 14648507.0,
      "step": 8035
    },
    {
      "entropy": 5.27067232131958,
      "epoch": 6.661143330571665,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001814840590171065,
      "loss": 4.6256,
      "mean_token_accuracy": 0.22166715264320375,
      "num_tokens": 14657892.0,
      "step": 8040
    },
    {
      "entropy": 5.255868148803711,
      "epoch": 6.665285832642916,
      "grad_norm": 1.375,
      "learning_rate": 0.00018119350573267979,
      "loss": 4.5963,
      "mean_token_accuracy": 0.22356109023094178,
      "num_tokens": 14666460.0,
      "step": 8045
    },
    {
      "entropy": 5.226639270782471,
      "epoch": 6.669428334714167,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018090314166620302,
      "loss": 4.577,
      "mean_token_accuracy": 0.22053904086351395,
      "num_tokens": 14675046.0,
      "step": 8050
    },
    {
      "entropy": 5.208006191253662,
      "epoch": 6.6735708367854185,
      "grad_norm": 1.390625,
      "learning_rate": 0.00018061296740337225,
      "loss": 4.6145,
      "mean_token_accuracy": 0.21706026196479797,
      "num_tokens": 14685055.0,
      "step": 8055
    },
    {
      "entropy": 5.2562051773071286,
      "epoch": 6.67771333885667,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018032298352950077,
      "loss": 4.5895,
      "mean_token_accuracy": 0.22452397793531417,
      "num_tokens": 14695371.0,
      "step": 8060
    },
    {
      "entropy": 5.196543836593628,
      "epoch": 6.681855840927921,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018003319062951756,
      "loss": 4.5695,
      "mean_token_accuracy": 0.22941381335258484,
      "num_tokens": 14703671.0,
      "step": 8065
    },
    {
      "entropy": 5.226376628875732,
      "epoch": 6.685998342999172,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017974358928796682,
      "loss": 4.5936,
      "mean_token_accuracy": 0.23043138682842254,
      "num_tokens": 14712527.0,
      "step": 8070
    },
    {
      "entropy": 5.261101865768433,
      "epoch": 6.690140845070422,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017945418008900594,
      "loss": 4.6275,
      "mean_token_accuracy": 0.22156598865985871,
      "num_tokens": 14720711.0,
      "step": 8075
    },
    {
      "entropy": 5.291861820220947,
      "epoch": 6.694283347141674,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001791649636164049,
      "loss": 4.6582,
      "mean_token_accuracy": 0.21525126695632935,
      "num_tokens": 14729667.0,
      "step": 8080
    },
    {
      "entropy": 5.314059782028198,
      "epoch": 6.698425849212924,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000178875940453545,
      "loss": 4.5945,
      "mean_token_accuracy": 0.22837472409009935,
      "num_tokens": 14738614.0,
      "step": 8085
    },
    {
      "entropy": 5.294093370437622,
      "epoch": 6.702568351284175,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017858711118341742,
      "loss": 4.6149,
      "mean_token_accuracy": 0.22338873893022537,
      "num_tokens": 14747724.0,
      "step": 8090
    },
    {
      "entropy": 5.18295726776123,
      "epoch": 6.7067108533554265,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017829847638862255,
      "loss": 4.5611,
      "mean_token_accuracy": 0.22669446617364883,
      "num_tokens": 14757436.0,
      "step": 8095
    },
    {
      "entropy": 5.1606669425964355,
      "epoch": 6.710853355426678,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017801003665136823,
      "loss": 4.5674,
      "mean_token_accuracy": 0.22505855709314346,
      "num_tokens": 14765948.0,
      "step": 8100
    },
    {
      "entropy": 5.214610815048218,
      "epoch": 6.714995857497929,
      "grad_norm": 1.375,
      "learning_rate": 0.00017772179255346883,
      "loss": 4.5818,
      "mean_token_accuracy": 0.22175888866186141,
      "num_tokens": 14775397.0,
      "step": 8105
    },
    {
      "entropy": 5.253256893157959,
      "epoch": 6.71913835956918,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017743374467634433,
      "loss": 4.5613,
      "mean_token_accuracy": 0.22598690241575242,
      "num_tokens": 14783605.0,
      "step": 8110
    },
    {
      "entropy": 5.272920751571656,
      "epoch": 6.723280861640431,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017714589360101862,
      "loss": 4.6472,
      "mean_token_accuracy": 0.22028567790985107,
      "num_tokens": 14792210.0,
      "step": 8115
    },
    {
      "entropy": 5.183764934539795,
      "epoch": 6.727423363711682,
      "grad_norm": 1.3125,
      "learning_rate": 0.000176858239908119,
      "loss": 4.5437,
      "mean_token_accuracy": 0.22591645419597625,
      "num_tokens": 14801186.0,
      "step": 8120
    },
    {
      "entropy": 5.1869776248931885,
      "epoch": 6.731565865782933,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017657078417787414,
      "loss": 4.5084,
      "mean_token_accuracy": 0.22825928032398224,
      "num_tokens": 14810100.0,
      "step": 8125
    },
    {
      "entropy": 5.149010562896729,
      "epoch": 6.735708367854184,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017628352699011396,
      "loss": 4.4888,
      "mean_token_accuracy": 0.22637079060077667,
      "num_tokens": 14819335.0,
      "step": 8130
    },
    {
      "entropy": 5.2059454917907715,
      "epoch": 6.739850869925435,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001759964689242674,
      "loss": 4.6041,
      "mean_token_accuracy": 0.2253633037209511,
      "num_tokens": 14828686.0,
      "step": 8135
    },
    {
      "entropy": 5.2320653915405275,
      "epoch": 6.743993371996686,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001757096105593619,
      "loss": 4.6236,
      "mean_token_accuracy": 0.21879837960004805,
      "num_tokens": 14838312.0,
      "step": 8140
    },
    {
      "entropy": 5.163789844512939,
      "epoch": 6.748135874067937,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017542295247402234,
      "loss": 4.5531,
      "mean_token_accuracy": 0.22149926722049712,
      "num_tokens": 14846838.0,
      "step": 8145
    },
    {
      "entropy": 5.2038836002349855,
      "epoch": 6.752278376139188,
      "grad_norm": 1.25,
      "learning_rate": 0.00017513649524646919,
      "loss": 4.5897,
      "mean_token_accuracy": 0.21604187786579132,
      "num_tokens": 14857427.0,
      "step": 8150
    },
    {
      "entropy": 5.2381213188171385,
      "epoch": 6.756420878210439,
      "grad_norm": 1.375,
      "learning_rate": 0.0001748502394545181,
      "loss": 4.6175,
      "mean_token_accuracy": 0.2162717968225479,
      "num_tokens": 14866244.0,
      "step": 8155
    },
    {
      "entropy": 5.231541490554809,
      "epoch": 6.76056338028169,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001745641856755783,
      "loss": 4.604,
      "mean_token_accuracy": 0.22452767044305802,
      "num_tokens": 14875697.0,
      "step": 8160
    },
    {
      "entropy": 5.2111104965209964,
      "epoch": 6.764705882352941,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017427833448665136,
      "loss": 4.6399,
      "mean_token_accuracy": 0.22241395860910415,
      "num_tokens": 14884650.0,
      "step": 8165
    },
    {
      "entropy": 5.223736572265625,
      "epoch": 6.768848384424192,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017399268646433065,
      "loss": 4.576,
      "mean_token_accuracy": 0.22412953674793243,
      "num_tokens": 14894058.0,
      "step": 8170
    },
    {
      "entropy": 5.319792890548706,
      "epoch": 6.772990886495443,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017370724218479912,
      "loss": 4.6766,
      "mean_token_accuracy": 0.22078485786914825,
      "num_tokens": 14903219.0,
      "step": 8175
    },
    {
      "entropy": 5.262167882919312,
      "epoch": 6.7771333885666944,
      "grad_norm": 1.546875,
      "learning_rate": 0.00017342200222382934,
      "loss": 4.5863,
      "mean_token_accuracy": 0.22088341116905214,
      "num_tokens": 14913051.0,
      "step": 8180
    },
    {
      "entropy": 5.268294906616211,
      "epoch": 6.781275890637946,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017313696715678134,
      "loss": 4.6256,
      "mean_token_accuracy": 0.22337746620178223,
      "num_tokens": 14921777.0,
      "step": 8185
    },
    {
      "entropy": 5.3510373592376705,
      "epoch": 6.785418392709197,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000172852137558602,
      "loss": 4.7456,
      "mean_token_accuracy": 0.21156660765409468,
      "num_tokens": 14931558.0,
      "step": 8190
    },
    {
      "entropy": 5.284751319885254,
      "epoch": 6.789560894780448,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017256751400382384,
      "loss": 4.6518,
      "mean_token_accuracy": 0.22996071279048919,
      "num_tokens": 14940584.0,
      "step": 8195
    },
    {
      "entropy": 5.25944709777832,
      "epoch": 6.793703396851699,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017228309706656354,
      "loss": 4.5888,
      "mean_token_accuracy": 0.2282853052020073,
      "num_tokens": 14950830.0,
      "step": 8200
    },
    {
      "entropy": 5.285803890228271,
      "epoch": 6.797845898922949,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001719988873205213,
      "loss": 4.6033,
      "mean_token_accuracy": 0.223866605758667,
      "num_tokens": 14958990.0,
      "step": 8205
    },
    {
      "entropy": 5.251561021804809,
      "epoch": 6.8019884009942,
      "grad_norm": 1.375,
      "learning_rate": 0.00017171488533897916,
      "loss": 4.6243,
      "mean_token_accuracy": 0.22517223805189132,
      "num_tokens": 14967451.0,
      "step": 8210
    },
    {
      "entropy": 5.1832770824432375,
      "epoch": 6.806130903065451,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017143109169480005,
      "loss": 4.5623,
      "mean_token_accuracy": 0.22827257215976715,
      "num_tokens": 14976157.0,
      "step": 8215
    },
    {
      "entropy": 5.249496221542358,
      "epoch": 6.810273405136702,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017114750696042695,
      "loss": 4.6804,
      "mean_token_accuracy": 0.2116089418530464,
      "num_tokens": 14985591.0,
      "step": 8220
    },
    {
      "entropy": 5.271088886260986,
      "epoch": 6.8144159072079535,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017086413170788107,
      "loss": 4.5833,
      "mean_token_accuracy": 0.22722700536251067,
      "num_tokens": 14994513.0,
      "step": 8225
    },
    {
      "entropy": 5.245871353149414,
      "epoch": 6.818558409279205,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017058096650876143,
      "loss": 4.636,
      "mean_token_accuracy": 0.2236326038837433,
      "num_tokens": 15003252.0,
      "step": 8230
    },
    {
      "entropy": 5.27306432723999,
      "epoch": 6.822700911350456,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017029801193424313,
      "loss": 4.5747,
      "mean_token_accuracy": 0.21844292730093,
      "num_tokens": 15012342.0,
      "step": 8235
    },
    {
      "entropy": 5.2001677513122555,
      "epoch": 6.826843413421707,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017001526855507632,
      "loss": 4.547,
      "mean_token_accuracy": 0.221232670545578,
      "num_tokens": 15021336.0,
      "step": 8240
    },
    {
      "entropy": 5.2186760902404785,
      "epoch": 6.830985915492958,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016973273694158555,
      "loss": 4.5742,
      "mean_token_accuracy": 0.21934880167245865,
      "num_tokens": 15031016.0,
      "step": 8245
    },
    {
      "entropy": 5.28720235824585,
      "epoch": 6.835128417564209,
      "grad_norm": 1.484375,
      "learning_rate": 0.00016945041766366773,
      "loss": 4.6031,
      "mean_token_accuracy": 0.22269195318222046,
      "num_tokens": 15039181.0,
      "step": 8250
    },
    {
      "entropy": 5.259404563903809,
      "epoch": 6.83927091963546,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016916831129079187,
      "loss": 4.637,
      "mean_token_accuracy": 0.21551821529865264,
      "num_tokens": 15048563.0,
      "step": 8255
    },
    {
      "entropy": 5.262871503829956,
      "epoch": 6.84341342170671,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001688864183919973,
      "loss": 4.6002,
      "mean_token_accuracy": 0.22465775460004805,
      "num_tokens": 15056960.0,
      "step": 8260
    },
    {
      "entropy": 5.318361806869507,
      "epoch": 6.8475559237779615,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016860473953589266,
      "loss": 4.6458,
      "mean_token_accuracy": 0.2170533150434494,
      "num_tokens": 15065645.0,
      "step": 8265
    },
    {
      "entropy": 5.248215198516846,
      "epoch": 6.851698425849213,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016832327529065522,
      "loss": 4.5868,
      "mean_token_accuracy": 0.22645846605300904,
      "num_tokens": 15074544.0,
      "step": 8270
    },
    {
      "entropy": 5.274274206161499,
      "epoch": 6.855840927920464,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001680420262240289,
      "loss": 4.6135,
      "mean_token_accuracy": 0.2209447756409645,
      "num_tokens": 15083543.0,
      "step": 8275
    },
    {
      "entropy": 5.2044919490814205,
      "epoch": 6.859983429991715,
      "grad_norm": 1.453125,
      "learning_rate": 0.00016776099290332397,
      "loss": 4.6102,
      "mean_token_accuracy": 0.22180342972278594,
      "num_tokens": 15092291.0,
      "step": 8280
    },
    {
      "entropy": 5.312284231185913,
      "epoch": 6.864125932062966,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016748017589541525,
      "loss": 4.6575,
      "mean_token_accuracy": 0.21847364902496338,
      "num_tokens": 15100875.0,
      "step": 8285
    },
    {
      "entropy": 5.263402652740479,
      "epoch": 6.868268434134217,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001671995757667414,
      "loss": 4.633,
      "mean_token_accuracy": 0.22437605410814285,
      "num_tokens": 15111385.0,
      "step": 8290
    },
    {
      "entropy": 5.34099850654602,
      "epoch": 6.872410936205468,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016691919308330343,
      "loss": 4.6439,
      "mean_token_accuracy": 0.2297893062233925,
      "num_tokens": 15121153.0,
      "step": 8295
    },
    {
      "entropy": 5.261020040512085,
      "epoch": 6.876553438276719,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016663902841066404,
      "loss": 4.6279,
      "mean_token_accuracy": 0.2185002475976944,
      "num_tokens": 15130132.0,
      "step": 8300
    },
    {
      "entropy": 5.217970180511474,
      "epoch": 6.88069594034797,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016635908231394587,
      "loss": 4.6184,
      "mean_token_accuracy": 0.21961871534585953,
      "num_tokens": 15138878.0,
      "step": 8305
    },
    {
      "entropy": 5.2379903316497805,
      "epoch": 6.8848384424192215,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001660793553578307,
      "loss": 4.6795,
      "mean_token_accuracy": 0.21482877880334855,
      "num_tokens": 15148727.0,
      "step": 8310
    },
    {
      "entropy": 5.262538146972656,
      "epoch": 6.888980944490473,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016579984810655856,
      "loss": 4.6337,
      "mean_token_accuracy": 0.22103115022182465,
      "num_tokens": 15158107.0,
      "step": 8315
    },
    {
      "entropy": 5.269617080688477,
      "epoch": 6.893123446561724,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016552056112392594,
      "loss": 4.6093,
      "mean_token_accuracy": 0.2215348705649376,
      "num_tokens": 15168101.0,
      "step": 8320
    },
    {
      "entropy": 5.247559452056885,
      "epoch": 6.897265948632974,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001652414949732854,
      "loss": 4.5523,
      "mean_token_accuracy": 0.22715236395597457,
      "num_tokens": 15176376.0,
      "step": 8325
    },
    {
      "entropy": 5.233921813964844,
      "epoch": 6.901408450704225,
      "grad_norm": 1.375,
      "learning_rate": 0.00016496265021754365,
      "loss": 4.5871,
      "mean_token_accuracy": 0.22135200947523118,
      "num_tokens": 15184936.0,
      "step": 8330
    },
    {
      "entropy": 5.288531923294068,
      "epoch": 6.905550952775476,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016468402741916124,
      "loss": 4.6414,
      "mean_token_accuracy": 0.2231667757034302,
      "num_tokens": 15194846.0,
      "step": 8335
    },
    {
      "entropy": 5.2356529712677,
      "epoch": 6.909693454846727,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001644056271401506,
      "loss": 4.6247,
      "mean_token_accuracy": 0.22294401973485947,
      "num_tokens": 15204543.0,
      "step": 8340
    },
    {
      "entropy": 5.235690498352051,
      "epoch": 6.913835956917978,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016412744994207573,
      "loss": 4.5874,
      "mean_token_accuracy": 0.22003623247146606,
      "num_tokens": 15213459.0,
      "step": 8345
    },
    {
      "entropy": 5.248845911026001,
      "epoch": 6.9179784589892295,
      "grad_norm": 1.375,
      "learning_rate": 0.00016384949638605025,
      "loss": 4.689,
      "mean_token_accuracy": 0.20948727577924728,
      "num_tokens": 15223841.0,
      "step": 8350
    },
    {
      "entropy": 5.222267770767212,
      "epoch": 6.922120961060481,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016357176703273696,
      "loss": 4.6237,
      "mean_token_accuracy": 0.22164985090494155,
      "num_tokens": 15233594.0,
      "step": 8355
    },
    {
      "entropy": 5.226879024505616,
      "epoch": 6.926263463131732,
      "grad_norm": 1.375,
      "learning_rate": 0.0001632942624423463,
      "loss": 4.5502,
      "mean_token_accuracy": 0.23184365630149842,
      "num_tokens": 15242876.0,
      "step": 8360
    },
    {
      "entropy": 5.32651424407959,
      "epoch": 6.930405965202983,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016301698317463525,
      "loss": 4.6478,
      "mean_token_accuracy": 0.2237490728497505,
      "num_tokens": 15251581.0,
      "step": 8365
    },
    {
      "entropy": 5.283322143554687,
      "epoch": 6.934548467274234,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016273992978890658,
      "loss": 4.5114,
      "mean_token_accuracy": 0.2229705795645714,
      "num_tokens": 15260601.0,
      "step": 8370
    },
    {
      "entropy": 5.229434013366699,
      "epoch": 6.938690969345485,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001624631028440071,
      "loss": 4.5188,
      "mean_token_accuracy": 0.2311753362417221,
      "num_tokens": 15270123.0,
      "step": 8375
    },
    {
      "entropy": 5.300338697433472,
      "epoch": 6.942833471416735,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001621865028983271,
      "loss": 4.6608,
      "mean_token_accuracy": 0.2248455300927162,
      "num_tokens": 15278623.0,
      "step": 8380
    },
    {
      "entropy": 5.178774833679199,
      "epoch": 6.946975973487987,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001619101305097988,
      "loss": 4.5533,
      "mean_token_accuracy": 0.22850299924612044,
      "num_tokens": 15287789.0,
      "step": 8385
    },
    {
      "entropy": 5.093407869338989,
      "epoch": 6.9511184755592375,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00016163398623589563,
      "loss": 4.5289,
      "mean_token_accuracy": 0.23101640194654466,
      "num_tokens": 15296266.0,
      "step": 8390
    },
    {
      "entropy": 5.268917369842529,
      "epoch": 6.955260977630489,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016135807063363074,
      "loss": 4.6257,
      "mean_token_accuracy": 0.21622092723846437,
      "num_tokens": 15304785.0,
      "step": 8395
    },
    {
      "entropy": 5.181910562515259,
      "epoch": 6.95940347970174,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016108238425955584,
      "loss": 4.5133,
      "mean_token_accuracy": 0.22999098598957063,
      "num_tokens": 15313727.0,
      "step": 8400
    },
    {
      "entropy": 5.275093126296997,
      "epoch": 6.963545981772991,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001608069276697608,
      "loss": 4.7186,
      "mean_token_accuracy": 0.21603549271821976,
      "num_tokens": 15322911.0,
      "step": 8405
    },
    {
      "entropy": 5.30788402557373,
      "epoch": 6.967688483844242,
      "grad_norm": 1.375,
      "learning_rate": 0.0001605317014198714,
      "loss": 4.6083,
      "mean_token_accuracy": 0.22257586866617202,
      "num_tokens": 15330882.0,
      "step": 8410
    },
    {
      "entropy": 5.212507438659668,
      "epoch": 6.971830985915493,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016025670606504926,
      "loss": 4.5211,
      "mean_token_accuracy": 0.2204284578561783,
      "num_tokens": 15339915.0,
      "step": 8415
    },
    {
      "entropy": 5.217851781845093,
      "epoch": 6.975973487986744,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001599819421599899,
      "loss": 4.5829,
      "mean_token_accuracy": 0.2220682069659233,
      "num_tokens": 15348181.0,
      "step": 8420
    },
    {
      "entropy": 5.184613180160523,
      "epoch": 6.980115990057995,
      "grad_norm": 1.375,
      "learning_rate": 0.00015970741025892233,
      "loss": 4.5441,
      "mean_token_accuracy": 0.22897181510925294,
      "num_tokens": 15357128.0,
      "step": 8425
    },
    {
      "entropy": 5.257200241088867,
      "epoch": 6.984258492129246,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015943311091560721,
      "loss": 4.648,
      "mean_token_accuracy": 0.22485268414020537,
      "num_tokens": 15366878.0,
      "step": 8430
    },
    {
      "entropy": 5.329241704940796,
      "epoch": 6.988400994200497,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015915904468333635,
      "loss": 4.5985,
      "mean_token_accuracy": 0.2384787380695343,
      "num_tokens": 15376101.0,
      "step": 8435
    },
    {
      "entropy": 5.1715577125549315,
      "epoch": 6.9925434962717485,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00015888521211493135,
      "loss": 4.5334,
      "mean_token_accuracy": 0.220519782602787,
      "num_tokens": 15384942.0,
      "step": 8440
    },
    {
      "entropy": 5.182519721984863,
      "epoch": 6.996685998342999,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001586116137627423,
      "loss": 4.5697,
      "mean_token_accuracy": 0.22755053788423538,
      "num_tokens": 15394380.0,
      "step": 8445
    },
    {
      "entropy": 5.27120304107666,
      "epoch": 7.00082850041425,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015833825017864711,
      "loss": 4.499,
      "mean_token_accuracy": 0.23353623002767562,
      "num_tokens": 15402180.0,
      "step": 8450
    },
    {
      "entropy": 5.268194484710693,
      "epoch": 7.004971002485501,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015806512191405,
      "loss": 4.5391,
      "mean_token_accuracy": 0.23040981441736222,
      "num_tokens": 15411434.0,
      "step": 8455
    },
    {
      "entropy": 5.258645725250244,
      "epoch": 7.009113504556752,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001577922295198803,
      "loss": 4.4841,
      "mean_token_accuracy": 0.23560507148504256,
      "num_tokens": 15419905.0,
      "step": 8460
    },
    {
      "entropy": 5.223584079742432,
      "epoch": 7.013256006628003,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015751957354659206,
      "loss": 4.5132,
      "mean_token_accuracy": 0.22909362614154816,
      "num_tokens": 15429200.0,
      "step": 8465
    },
    {
      "entropy": 5.233704042434693,
      "epoch": 7.017398508699254,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015724715454416206,
      "loss": 4.4788,
      "mean_token_accuracy": 0.23763577193021773,
      "num_tokens": 15439328.0,
      "step": 8470
    },
    {
      "entropy": 5.257611846923828,
      "epoch": 7.021541010770505,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00015697497306208926,
      "loss": 4.4535,
      "mean_token_accuracy": 0.2374115392565727,
      "num_tokens": 15447739.0,
      "step": 8475
    },
    {
      "entropy": 5.276518440246582,
      "epoch": 7.0256835128417565,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001567030296493934,
      "loss": 4.3947,
      "mean_token_accuracy": 0.24155190736055374,
      "num_tokens": 15456598.0,
      "step": 8480
    },
    {
      "entropy": 5.272721004486084,
      "epoch": 7.029826014913008,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015643132485461395,
      "loss": 4.5496,
      "mean_token_accuracy": 0.2299117460846901,
      "num_tokens": 15466633.0,
      "step": 8485
    },
    {
      "entropy": 5.2553550720214846,
      "epoch": 7.033968516984259,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001561598592258094,
      "loss": 4.4583,
      "mean_token_accuracy": 0.24523903131484986,
      "num_tokens": 15476809.0,
      "step": 8490
    },
    {
      "entropy": 5.218363904953003,
      "epoch": 7.03811101905551,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00015588863331055541,
      "loss": 4.4565,
      "mean_token_accuracy": 0.2330383762717247,
      "num_tokens": 15484964.0,
      "step": 8495
    },
    {
      "entropy": 5.212063121795654,
      "epoch": 7.042253521126761,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001556176476559444,
      "loss": 4.454,
      "mean_token_accuracy": 0.2371162474155426,
      "num_tokens": 15493600.0,
      "step": 8500
    },
    {
      "epoch": 7.042253521126761,
      "eval_entropy": 5.144134729909998,
      "eval_loss": 5.7671732902526855,
      "eval_mean_token_accuracy": 0.16671444261188445,
      "eval_num_tokens": 15493600.0,
      "eval_runtime": 2.7478,
      "eval_samples_per_second": 1363.638,
      "eval_steps_per_second": 170.682,
      "step": 8500
    },
    {
      "entropy": 5.160946273803711,
      "epoch": 7.046396023198011,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015534690280858405,
      "loss": 4.4276,
      "mean_token_accuracy": 0.24355203807353973,
      "num_tokens": 15502614.0,
      "step": 8505
    },
    {
      "entropy": 5.328784036636352,
      "epoch": 7.050538525269262,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015507639931459615,
      "loss": 4.5512,
      "mean_token_accuracy": 0.22915739566087723,
      "num_tokens": 15511521.0,
      "step": 8510
    },
    {
      "entropy": 5.221444749832154,
      "epoch": 7.054681027340513,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000154806137719616,
      "loss": 4.4883,
      "mean_token_accuracy": 0.23399232029914857,
      "num_tokens": 15520220.0,
      "step": 8515
    },
    {
      "entropy": 5.216493368148804,
      "epoch": 7.0588235294117645,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001545361185687906,
      "loss": 4.4179,
      "mean_token_accuracy": 0.24791412502527238,
      "num_tokens": 15528924.0,
      "step": 8520
    },
    {
      "entropy": 5.180371522903442,
      "epoch": 7.062966031483016,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015426634240677824,
      "loss": 4.3999,
      "mean_token_accuracy": 0.23609755784273148,
      "num_tokens": 15539722.0,
      "step": 8525
    },
    {
      "entropy": 5.272413730621338,
      "epoch": 7.067108533554267,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015399680977774689,
      "loss": 4.4777,
      "mean_token_accuracy": 0.236843441426754,
      "num_tokens": 15548682.0,
      "step": 8530
    },
    {
      "entropy": 5.280804681777954,
      "epoch": 7.071251035625518,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015372752122537318,
      "loss": 4.4655,
      "mean_token_accuracy": 0.23580969572067262,
      "num_tokens": 15557508.0,
      "step": 8535
    },
    {
      "entropy": 5.23158540725708,
      "epoch": 7.075393537696769,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015345847729284168,
      "loss": 4.4983,
      "mean_token_accuracy": 0.24108577966690065,
      "num_tokens": 15566120.0,
      "step": 8540
    },
    {
      "entropy": 5.217205381393432,
      "epoch": 7.07953603976802,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015318967852284335,
      "loss": 4.4081,
      "mean_token_accuracy": 0.2468072861433029,
      "num_tokens": 15575166.0,
      "step": 8545
    },
    {
      "entropy": 5.20664610862732,
      "epoch": 7.083678541839271,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015292112545757476,
      "loss": 4.4513,
      "mean_token_accuracy": 0.22878976464271544,
      "num_tokens": 15585063.0,
      "step": 8550
    },
    {
      "entropy": 5.1814803123474125,
      "epoch": 7.087821043910522,
      "grad_norm": 1.5,
      "learning_rate": 0.00015265281863873663,
      "loss": 4.4606,
      "mean_token_accuracy": 0.23958255648612975,
      "num_tokens": 15594386.0,
      "step": 8555
    },
    {
      "entropy": 5.2313049793243405,
      "epoch": 7.091963545981773,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015238475860753338,
      "loss": 4.4954,
      "mean_token_accuracy": 0.23332029283046724,
      "num_tokens": 15603796.0,
      "step": 8560
    },
    {
      "entropy": 5.233847618103027,
      "epoch": 7.096106048053024,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015211694590467118,
      "loss": 4.4148,
      "mean_token_accuracy": 0.24896543323993683,
      "num_tokens": 15612752.0,
      "step": 8565
    },
    {
      "entropy": 5.224442100524902,
      "epoch": 7.100248550124275,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001518493810703575,
      "loss": 4.5041,
      "mean_token_accuracy": 0.23226284086704255,
      "num_tokens": 15621995.0,
      "step": 8570
    },
    {
      "entropy": 5.202006530761719,
      "epoch": 7.104391052195526,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015158206464430002,
      "loss": 4.476,
      "mean_token_accuracy": 0.2358950525522232,
      "num_tokens": 15630390.0,
      "step": 8575
    },
    {
      "entropy": 5.173835802078247,
      "epoch": 7.108533554266777,
      "grad_norm": 1.390625,
      "learning_rate": 0.000151314997165705,
      "loss": 4.4176,
      "mean_token_accuracy": 0.23947429656982422,
      "num_tokens": 15639723.0,
      "step": 8580
    },
    {
      "entropy": 5.201437520980835,
      "epoch": 7.112676056338028,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00015104817917327694,
      "loss": 4.3392,
      "mean_token_accuracy": 0.25074001848697663,
      "num_tokens": 15647755.0,
      "step": 8585
    },
    {
      "entropy": 5.218188905715943,
      "epoch": 7.116818558409279,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015078161120521676,
      "loss": 4.4926,
      "mean_token_accuracy": 0.241668264567852,
      "num_tokens": 15656949.0,
      "step": 8590
    },
    {
      "entropy": 5.180606842041016,
      "epoch": 7.12096106048053,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001505152937992211,
      "loss": 4.4276,
      "mean_token_accuracy": 0.2432338058948517,
      "num_tokens": 15666290.0,
      "step": 8595
    },
    {
      "entropy": 5.255083656311035,
      "epoch": 7.125103562551781,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015024922749248143,
      "loss": 4.4576,
      "mean_token_accuracy": 0.24036007076501847,
      "num_tokens": 15675728.0,
      "step": 8600
    },
    {
      "entropy": 5.205368947982788,
      "epoch": 7.1292460646230325,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014998341282168243,
      "loss": 4.4185,
      "mean_token_accuracy": 0.2462959334254265,
      "num_tokens": 15684519.0,
      "step": 8605
    },
    {
      "entropy": 5.216509628295898,
      "epoch": 7.133388566694284,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001497178503230015,
      "loss": 4.5147,
      "mean_token_accuracy": 0.230082406103611,
      "num_tokens": 15693813.0,
      "step": 8610
    },
    {
      "entropy": 5.244852066040039,
      "epoch": 7.137531068765535,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014945254053210712,
      "loss": 4.4876,
      "mean_token_accuracy": 0.2331735149025917,
      "num_tokens": 15702245.0,
      "step": 8615
    },
    {
      "entropy": 5.294452905654907,
      "epoch": 7.141673570836786,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014918748398415801,
      "loss": 4.4905,
      "mean_token_accuracy": 0.23399191945791245,
      "num_tokens": 15712246.0,
      "step": 8620
    },
    {
      "entropy": 5.239782047271729,
      "epoch": 7.145816072908037,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014892268121380244,
      "loss": 4.4502,
      "mean_token_accuracy": 0.2341538354754448,
      "num_tokens": 15720904.0,
      "step": 8625
    },
    {
      "entropy": 5.19270658493042,
      "epoch": 7.149958574979287,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014865813275517637,
      "loss": 4.4134,
      "mean_token_accuracy": 0.251400226354599,
      "num_tokens": 15730595.0,
      "step": 8630
    },
    {
      "entropy": 5.164038562774659,
      "epoch": 7.154101077050538,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014839383914190302,
      "loss": 4.391,
      "mean_token_accuracy": 0.24321847409009933,
      "num_tokens": 15739475.0,
      "step": 8635
    },
    {
      "entropy": 5.228014230728149,
      "epoch": 7.158243579121789,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001481298009070915,
      "loss": 4.4214,
      "mean_token_accuracy": 0.24496893882751464,
      "num_tokens": 15748672.0,
      "step": 8640
    },
    {
      "entropy": 5.247474336624146,
      "epoch": 7.16238608119304,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014786601858333566,
      "loss": 4.4569,
      "mean_token_accuracy": 0.2428860992193222,
      "num_tokens": 15757540.0,
      "step": 8645
    },
    {
      "entropy": 5.169031476974487,
      "epoch": 7.1665285832642915,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014760249270271341,
      "loss": 4.4624,
      "mean_token_accuracy": 0.23282100558280944,
      "num_tokens": 15767198.0,
      "step": 8650
    },
    {
      "entropy": 5.153016376495361,
      "epoch": 7.170671085335543,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014733922379678513,
      "loss": 4.4382,
      "mean_token_accuracy": 0.23907807618379592,
      "num_tokens": 15776818.0,
      "step": 8655
    },
    {
      "entropy": 5.30323224067688,
      "epoch": 7.174813587406794,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014707621239659312,
      "loss": 4.5796,
      "mean_token_accuracy": 0.22289534360170365,
      "num_tokens": 15786054.0,
      "step": 8660
    },
    {
      "entropy": 5.269830656051636,
      "epoch": 7.178956089478045,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001468134590326601,
      "loss": 4.5165,
      "mean_token_accuracy": 0.23531094938516617,
      "num_tokens": 15795367.0,
      "step": 8665
    },
    {
      "entropy": 5.288466930389404,
      "epoch": 7.183098591549296,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001465509642349881,
      "loss": 4.5155,
      "mean_token_accuracy": 0.2314830243587494,
      "num_tokens": 15805006.0,
      "step": 8670
    },
    {
      "entropy": 5.271142148971558,
      "epoch": 7.187241093620547,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00014628872853305808,
      "loss": 4.5543,
      "mean_token_accuracy": 0.23005914688110352,
      "num_tokens": 15813978.0,
      "step": 8675
    },
    {
      "entropy": 5.290530061721801,
      "epoch": 7.191383595691798,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014602675245582795,
      "loss": 4.5121,
      "mean_token_accuracy": 0.23368490040302276,
      "num_tokens": 15824156.0,
      "step": 8680
    },
    {
      "entropy": 5.286494398117066,
      "epoch": 7.195526097763049,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014576503653173219,
      "loss": 4.5721,
      "mean_token_accuracy": 0.2260148286819458,
      "num_tokens": 15833514.0,
      "step": 8685
    },
    {
      "entropy": 5.256912612915039,
      "epoch": 7.1996685998342995,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014550358128868036,
      "loss": 4.5304,
      "mean_token_accuracy": 0.23308124840259553,
      "num_tokens": 15842047.0,
      "step": 8690
    },
    {
      "entropy": 5.251299095153809,
      "epoch": 7.203811101905551,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014524238725405615,
      "loss": 4.5392,
      "mean_token_accuracy": 0.22702830731868745,
      "num_tokens": 15851916.0,
      "step": 8695
    },
    {
      "entropy": 5.24776177406311,
      "epoch": 7.207953603976802,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014498145495471664,
      "loss": 4.4608,
      "mean_token_accuracy": 0.23938020169734955,
      "num_tokens": 15860776.0,
      "step": 8700
    },
    {
      "entropy": 5.276491785049439,
      "epoch": 7.212096106048053,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014472078491699076,
      "loss": 4.5422,
      "mean_token_accuracy": 0.23557736426591874,
      "num_tokens": 15870272.0,
      "step": 8705
    },
    {
      "entropy": 5.245528697967529,
      "epoch": 7.216238608119304,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001444603776666784,
      "loss": 4.4457,
      "mean_token_accuracy": 0.24147832840681077,
      "num_tokens": 15879448.0,
      "step": 8710
    },
    {
      "entropy": 5.296770286560059,
      "epoch": 7.220381110190555,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014420023372904944,
      "loss": 4.5375,
      "mean_token_accuracy": 0.2281098335981369,
      "num_tokens": 15888806.0,
      "step": 8715
    },
    {
      "entropy": 5.216280794143676,
      "epoch": 7.224523612261806,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014394035362884274,
      "loss": 4.5189,
      "mean_token_accuracy": 0.23449675887823104,
      "num_tokens": 15898443.0,
      "step": 8720
    },
    {
      "entropy": 5.193191432952881,
      "epoch": 7.228666114333057,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014368073789026482,
      "loss": 4.4067,
      "mean_token_accuracy": 0.23813656866550445,
      "num_tokens": 15907759.0,
      "step": 8725
    },
    {
      "entropy": 5.237518215179444,
      "epoch": 7.232808616404308,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014342138703698882,
      "loss": 4.3953,
      "mean_token_accuracy": 0.2393280491232872,
      "num_tokens": 15916687.0,
      "step": 8730
    },
    {
      "entropy": 5.218048810958862,
      "epoch": 7.2369511184755595,
      "grad_norm": 1.453125,
      "learning_rate": 0.000143162301592154,
      "loss": 4.4798,
      "mean_token_accuracy": 0.2354348674416542,
      "num_tokens": 15925422.0,
      "step": 8735
    },
    {
      "entropy": 5.208735179901123,
      "epoch": 7.241093620546811,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014290348207836385,
      "loss": 4.469,
      "mean_token_accuracy": 0.23067065924406052,
      "num_tokens": 15934279.0,
      "step": 8740
    },
    {
      "entropy": 5.200824022293091,
      "epoch": 7.245236122618062,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001426449290176857,
      "loss": 4.4459,
      "mean_token_accuracy": 0.2411273807287216,
      "num_tokens": 15943565.0,
      "step": 8745
    },
    {
      "entropy": 5.171152496337891,
      "epoch": 7.249378624689312,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014238664293164934,
      "loss": 4.4387,
      "mean_token_accuracy": 0.2432919681072235,
      "num_tokens": 15952360.0,
      "step": 8750
    },
    {
      "entropy": 5.2569787979125975,
      "epoch": 7.253521126760563,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001421286243412459,
      "loss": 4.4644,
      "mean_token_accuracy": 0.23361512869596482,
      "num_tokens": 15960715.0,
      "step": 8755
    },
    {
      "entropy": 5.254800176620483,
      "epoch": 7.257663628831814,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014187087376692724,
      "loss": 4.4991,
      "mean_token_accuracy": 0.23598186820745468,
      "num_tokens": 15969543.0,
      "step": 8760
    },
    {
      "entropy": 5.139178991317749,
      "epoch": 7.261806130903065,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001416133917286043,
      "loss": 4.4371,
      "mean_token_accuracy": 0.2301303595304489,
      "num_tokens": 15978256.0,
      "step": 8765
    },
    {
      "entropy": 5.199715280532837,
      "epoch": 7.265948632974316,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001413561787456466,
      "loss": 4.4333,
      "mean_token_accuracy": 0.2337365508079529,
      "num_tokens": 15987646.0,
      "step": 8770
    },
    {
      "entropy": 5.227065134048462,
      "epoch": 7.2700911350455675,
      "grad_norm": 1.375,
      "learning_rate": 0.00014109923533688068,
      "loss": 4.4542,
      "mean_token_accuracy": 0.22979893684387206,
      "num_tokens": 15996767.0,
      "step": 8775
    },
    {
      "entropy": 5.201937913894653,
      "epoch": 7.274233637116819,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001408425620205897,
      "loss": 4.4668,
      "mean_token_accuracy": 0.23180714696645738,
      "num_tokens": 16007170.0,
      "step": 8780
    },
    {
      "entropy": 5.240734672546386,
      "epoch": 7.27837613918807,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001405861593145116,
      "loss": 4.5002,
      "mean_token_accuracy": 0.23389508724212646,
      "num_tokens": 16015354.0,
      "step": 8785
    },
    {
      "entropy": 5.272396230697632,
      "epoch": 7.282518641259321,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00014033002773583872,
      "loss": 4.4798,
      "mean_token_accuracy": 0.2297423854470253,
      "num_tokens": 16024600.0,
      "step": 8790
    },
    {
      "entropy": 5.288425016403198,
      "epoch": 7.286661143330572,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001400741678012164,
      "loss": 4.4897,
      "mean_token_accuracy": 0.23756824433803558,
      "num_tokens": 16033301.0,
      "step": 8795
    },
    {
      "entropy": 5.233310651779175,
      "epoch": 7.290803645401823,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013981858002674223,
      "loss": 4.514,
      "mean_token_accuracy": 0.22861506938934326,
      "num_tokens": 16043539.0,
      "step": 8800
    },
    {
      "entropy": 5.262283372879028,
      "epoch": 7.294946147473074,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001395632649279645,
      "loss": 4.5637,
      "mean_token_accuracy": 0.22659083157777787,
      "num_tokens": 16052861.0,
      "step": 8805
    },
    {
      "entropy": 5.248991298675537,
      "epoch": 7.299088649544325,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013930822301988177,
      "loss": 4.443,
      "mean_token_accuracy": 0.24139501601457597,
      "num_tokens": 16062454.0,
      "step": 8810
    },
    {
      "entropy": 5.189526510238648,
      "epoch": 7.3032311516155755,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013905345481694147,
      "loss": 4.4564,
      "mean_token_accuracy": 0.239634145796299,
      "num_tokens": 16071432.0,
      "step": 8815
    },
    {
      "entropy": 5.187956285476685,
      "epoch": 7.307373653686827,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001387989608330388,
      "loss": 4.4302,
      "mean_token_accuracy": 0.23870561569929122,
      "num_tokens": 16080411.0,
      "step": 8820
    },
    {
      "entropy": 5.241320371627808,
      "epoch": 7.311516155758078,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013854474158151602,
      "loss": 4.469,
      "mean_token_accuracy": 0.23667130917310714,
      "num_tokens": 16089396.0,
      "step": 8825
    },
    {
      "entropy": 5.252061271667481,
      "epoch": 7.315658657829329,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013829079757516106,
      "loss": 4.5584,
      "mean_token_accuracy": 0.22721774727106095,
      "num_tokens": 16099308.0,
      "step": 8830
    },
    {
      "entropy": 5.210462284088135,
      "epoch": 7.31980115990058,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013803712932620681,
      "loss": 4.4975,
      "mean_token_accuracy": 0.23123666197061538,
      "num_tokens": 16109336.0,
      "step": 8835
    },
    {
      "entropy": 5.27162618637085,
      "epoch": 7.323943661971831,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001377837373463298,
      "loss": 4.4839,
      "mean_token_accuracy": 0.23789117783308028,
      "num_tokens": 16118483.0,
      "step": 8840
    },
    {
      "entropy": 5.257375574111938,
      "epoch": 7.328086164043082,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001375306221466494,
      "loss": 4.4509,
      "mean_token_accuracy": 0.23670082092285155,
      "num_tokens": 16127302.0,
      "step": 8845
    },
    {
      "entropy": 5.224235773086548,
      "epoch": 7.332228666114333,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00013727778423772654,
      "loss": 4.5196,
      "mean_token_accuracy": 0.22835123986005784,
      "num_tokens": 16136361.0,
      "step": 8850
    },
    {
      "entropy": 5.301512670516968,
      "epoch": 7.336371168185584,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013702522412956304,
      "loss": 4.5757,
      "mean_token_accuracy": 0.2324121281504631,
      "num_tokens": 16146071.0,
      "step": 8855
    },
    {
      "entropy": 5.286683082580566,
      "epoch": 7.340513670256835,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013677294233160016,
      "loss": 4.5082,
      "mean_token_accuracy": 0.2335846468806267,
      "num_tokens": 16154606.0,
      "step": 8860
    },
    {
      "entropy": 5.188597393035889,
      "epoch": 7.3446561723280865,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013652093935271776,
      "loss": 4.469,
      "mean_token_accuracy": 0.24194480031728743,
      "num_tokens": 16163648.0,
      "step": 8865
    },
    {
      "entropy": 5.2929548740386965,
      "epoch": 7.348798674399337,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013626921570123357,
      "loss": 4.5272,
      "mean_token_accuracy": 0.22835434675216676,
      "num_tokens": 16172693.0,
      "step": 8870
    },
    {
      "entropy": 5.2380091667175295,
      "epoch": 7.352941176470588,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013601777188490156,
      "loss": 4.4437,
      "mean_token_accuracy": 0.23393087834119797,
      "num_tokens": 16181689.0,
      "step": 8875
    },
    {
      "entropy": 5.201257801055908,
      "epoch": 7.357083678541839,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001357666084109115,
      "loss": 4.5641,
      "mean_token_accuracy": 0.22886379808187485,
      "num_tokens": 16191254.0,
      "step": 8880
    },
    {
      "entropy": 5.259526205062866,
      "epoch": 7.36122618061309,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001355157257858875,
      "loss": 4.4818,
      "mean_token_accuracy": 0.2344774931669235,
      "num_tokens": 16199613.0,
      "step": 8885
    },
    {
      "entropy": 5.2244462966918945,
      "epoch": 7.365368682684341,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013526512451588718,
      "loss": 4.4569,
      "mean_token_accuracy": 0.23725605905056,
      "num_tokens": 16208345.0,
      "step": 8890
    },
    {
      "entropy": 5.250645637512207,
      "epoch": 7.369511184755592,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001350148051064008,
      "loss": 4.5205,
      "mean_token_accuracy": 0.23002041578292848,
      "num_tokens": 16217964.0,
      "step": 8895
    },
    {
      "entropy": 5.278724288940429,
      "epoch": 7.373653686826843,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013476476806234983,
      "loss": 4.565,
      "mean_token_accuracy": 0.22962198555469512,
      "num_tokens": 16227140.0,
      "step": 8900
    },
    {
      "entropy": 5.312851047515869,
      "epoch": 7.3777961888980945,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001345150138880865,
      "loss": 4.5029,
      "mean_token_accuracy": 0.23140528798103333,
      "num_tokens": 16237416.0,
      "step": 8905
    },
    {
      "entropy": 5.317034292221069,
      "epoch": 7.381938690969346,
      "grad_norm": 1.4375,
      "learning_rate": 0.00013426554308739214,
      "loss": 4.5531,
      "mean_token_accuracy": 0.22577800899744033,
      "num_tokens": 16246682.0,
      "step": 8910
    },
    {
      "entropy": 5.251986789703369,
      "epoch": 7.386081193040597,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013401635616347655,
      "loss": 4.5382,
      "mean_token_accuracy": 0.23102884888648986,
      "num_tokens": 16255450.0,
      "step": 8915
    },
    {
      "entropy": 5.131001996994018,
      "epoch": 7.390223695111848,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013376745361897712,
      "loss": 4.426,
      "mean_token_accuracy": 0.24305522590875625,
      "num_tokens": 16265014.0,
      "step": 8920
    },
    {
      "entropy": 5.271205043792724,
      "epoch": 7.394366197183099,
      "grad_norm": 1.5,
      "learning_rate": 0.00013351883595595743,
      "loss": 4.5095,
      "mean_token_accuracy": 0.2354437828063965,
      "num_tokens": 16273309.0,
      "step": 8925
    },
    {
      "entropy": 5.254529523849487,
      "epoch": 7.39850869925435,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001332705036759065,
      "loss": 4.4768,
      "mean_token_accuracy": 0.22674325853586197,
      "num_tokens": 16282129.0,
      "step": 8930
    },
    {
      "entropy": 5.20716290473938,
      "epoch": 7.4026512013256,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001330224572797377,
      "loss": 4.4394,
      "mean_token_accuracy": 0.23804527372121811,
      "num_tokens": 16291208.0,
      "step": 8935
    },
    {
      "entropy": 5.249595355987549,
      "epoch": 7.406793703396851,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013277469726778756,
      "loss": 4.4378,
      "mean_token_accuracy": 0.23947702646255492,
      "num_tokens": 16300111.0,
      "step": 8940
    },
    {
      "entropy": 5.250058317184449,
      "epoch": 7.4109362054681025,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001325272241398153,
      "loss": 4.4082,
      "mean_token_accuracy": 0.2453349322080612,
      "num_tokens": 16308653.0,
      "step": 8945
    },
    {
      "entropy": 5.2861817359924315,
      "epoch": 7.415078707539354,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013228003839500113,
      "loss": 4.5146,
      "mean_token_accuracy": 0.2294495865702629,
      "num_tokens": 16317986.0,
      "step": 8950
    },
    {
      "entropy": 5.239671325683593,
      "epoch": 7.419221209610605,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001320331405319458,
      "loss": 4.5021,
      "mean_token_accuracy": 0.23499785363674164,
      "num_tokens": 16326889.0,
      "step": 8955
    },
    {
      "entropy": 5.184326696395874,
      "epoch": 7.423363711681856,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013178653104866926,
      "loss": 4.5112,
      "mean_token_accuracy": 0.22870156764984131,
      "num_tokens": 16336121.0,
      "step": 8960
    },
    {
      "entropy": 5.236520719528198,
      "epoch": 7.427506213753107,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00013154021044260966,
      "loss": 4.5695,
      "mean_token_accuracy": 0.22427098006010054,
      "num_tokens": 16345216.0,
      "step": 8965
    },
    {
      "entropy": 5.153100252151489,
      "epoch": 7.431648715824358,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013129417921062275,
      "loss": 4.4096,
      "mean_token_accuracy": 0.24559928774833678,
      "num_tokens": 16353735.0,
      "step": 8970
    },
    {
      "entropy": 5.1912792205810545,
      "epoch": 7.435791217895609,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001310484378489803,
      "loss": 4.4268,
      "mean_token_accuracy": 0.24140621721744537,
      "num_tokens": 16362474.0,
      "step": 8975
    },
    {
      "entropy": 5.260259389877319,
      "epoch": 7.43993371996686,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013080298685336957,
      "loss": 4.5696,
      "mean_token_accuracy": 0.22756587266921996,
      "num_tokens": 16371948.0,
      "step": 8980
    },
    {
      "entropy": 5.208133554458618,
      "epoch": 7.444076222038111,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013055782671889205,
      "loss": 4.4525,
      "mean_token_accuracy": 0.2374194383621216,
      "num_tokens": 16380849.0,
      "step": 8985
    },
    {
      "entropy": 5.269089221954346,
      "epoch": 7.448218724109362,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001303129579400624,
      "loss": 4.5582,
      "mean_token_accuracy": 0.2278520569205284,
      "num_tokens": 16389495.0,
      "step": 8990
    },
    {
      "entropy": 5.237772464752197,
      "epoch": 7.452361226180613,
      "grad_norm": 1.5,
      "learning_rate": 0.0001300683810108078,
      "loss": 4.436,
      "mean_token_accuracy": 0.23841076344251633,
      "num_tokens": 16397744.0,
      "step": 8995
    },
    {
      "entropy": 5.221956491470337,
      "epoch": 7.456503728251864,
      "grad_norm": 1.375,
      "learning_rate": 0.0001298240964244666,
      "loss": 4.47,
      "mean_token_accuracy": 0.22928055822849275,
      "num_tokens": 16406785.0,
      "step": 9000
    },
    {
      "epoch": 7.456503728251864,
      "eval_entropy": 5.150845478846828,
      "eval_loss": 5.770321846008301,
      "eval_mean_token_accuracy": 0.16733991065576895,
      "eval_num_tokens": 16406785.0,
      "eval_runtime": 2.7565,
      "eval_samples_per_second": 1359.309,
      "eval_steps_per_second": 170.14,
      "step": 9000
    },
    {
      "entropy": 5.222792625427246,
      "epoch": 7.460646230323115,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001295801046737876,
      "loss": 4.4829,
      "mean_token_accuracy": 0.2424822673201561,
      "num_tokens": 16415768.0,
      "step": 9005
    },
    {
      "entropy": 5.179026937484741,
      "epoch": 7.464788732394366,
      "grad_norm": 1.59375,
      "learning_rate": 0.00012933640625092867,
      "loss": 4.4584,
      "mean_token_accuracy": 0.23162254691123962,
      "num_tokens": 16424722.0,
      "step": 9010
    },
    {
      "entropy": 5.19364447593689,
      "epoch": 7.468931234465617,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012909300164745636,
      "loss": 4.4487,
      "mean_token_accuracy": 0.23764137476682662,
      "num_tokens": 16433797.0,
      "step": 9015
    },
    {
      "entropy": 5.19287257194519,
      "epoch": 7.473073736536868,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012884989135434423,
      "loss": 4.3884,
      "mean_token_accuracy": 0.24451866745948792,
      "num_tokens": 16442537.0,
      "step": 9020
    },
    {
      "entropy": 5.255203294754028,
      "epoch": 7.477216238608119,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012860707586197218,
      "loss": 4.5171,
      "mean_token_accuracy": 0.22950493842363356,
      "num_tokens": 16451440.0,
      "step": 9025
    },
    {
      "entropy": 5.232228517532349,
      "epoch": 7.4813587406793705,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012836455566012586,
      "loss": 4.4732,
      "mean_token_accuracy": 0.23434990048408508,
      "num_tokens": 16459358.0,
      "step": 9030
    },
    {
      "entropy": 5.165243911743164,
      "epoch": 7.485501242750622,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012812233123799481,
      "loss": 4.374,
      "mean_token_accuracy": 0.2526957616209984,
      "num_tokens": 16467584.0,
      "step": 9035
    },
    {
      "entropy": 5.230511713027954,
      "epoch": 7.489643744821873,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012788040308417232,
      "loss": 4.4985,
      "mean_token_accuracy": 0.23179699778556823,
      "num_tokens": 16476674.0,
      "step": 9040
    },
    {
      "entropy": 5.191761779785156,
      "epoch": 7.493786246893124,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012763877168665383,
      "loss": 4.4058,
      "mean_token_accuracy": 0.2400926783680916,
      "num_tokens": 16485422.0,
      "step": 9045
    },
    {
      "entropy": 5.243845796585083,
      "epoch": 7.497928748964375,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012739743753283617,
      "loss": 4.5816,
      "mean_token_accuracy": 0.23172033131122588,
      "num_tokens": 16494024.0,
      "step": 9050
    },
    {
      "entropy": 5.245836687088013,
      "epoch": 7.502071251035625,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001271564011095169,
      "loss": 4.5502,
      "mean_token_accuracy": 0.22851696759462356,
      "num_tokens": 16503805.0,
      "step": 9055
    },
    {
      "entropy": 5.267697143554687,
      "epoch": 7.506213753106876,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001269156629028927,
      "loss": 4.4904,
      "mean_token_accuracy": 0.23017911314964296,
      "num_tokens": 16512559.0,
      "step": 9060
    },
    {
      "entropy": 5.261908674240113,
      "epoch": 7.510356255178127,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001266752233985589,
      "loss": 4.5267,
      "mean_token_accuracy": 0.23279943615198134,
      "num_tokens": 16521910.0,
      "step": 9065
    },
    {
      "entropy": 5.231717920303344,
      "epoch": 7.514498757249378,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012643508308150822,
      "loss": 4.4685,
      "mean_token_accuracy": 0.23291125595569612,
      "num_tokens": 16530802.0,
      "step": 9070
    },
    {
      "entropy": 5.175790023803711,
      "epoch": 7.5186412593206295,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001261952424361299,
      "loss": 4.4697,
      "mean_token_accuracy": 0.23915740251541137,
      "num_tokens": 16540519.0,
      "step": 9075
    },
    {
      "entropy": 5.179768180847168,
      "epoch": 7.522783761391881,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001259557019462088,
      "loss": 4.4624,
      "mean_token_accuracy": 0.2289350926876068,
      "num_tokens": 16549433.0,
      "step": 9080
    },
    {
      "entropy": 5.242737770080566,
      "epoch": 7.526926263463132,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012571646209492417,
      "loss": 4.4956,
      "mean_token_accuracy": 0.22793238312005998,
      "num_tokens": 16558128.0,
      "step": 9085
    },
    {
      "entropy": 5.226051378250122,
      "epoch": 7.531068765534383,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00012547752336484908,
      "loss": 4.4254,
      "mean_token_accuracy": 0.23739625066518782,
      "num_tokens": 16566636.0,
      "step": 9090
    },
    {
      "entropy": 5.249091672897339,
      "epoch": 7.535211267605634,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012523888623794892,
      "loss": 4.4516,
      "mean_token_accuracy": 0.23320165723562242,
      "num_tokens": 16575036.0,
      "step": 9095
    },
    {
      "entropy": 5.201520490646362,
      "epoch": 7.539353769676885,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001250005511955809,
      "loss": 4.4762,
      "mean_token_accuracy": 0.24034629613161088,
      "num_tokens": 16584146.0,
      "step": 9100
    },
    {
      "entropy": 5.257434415817261,
      "epoch": 7.543496271748136,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001247625187184929,
      "loss": 4.5323,
      "mean_token_accuracy": 0.23666058033704757,
      "num_tokens": 16593915.0,
      "step": 9105
    },
    {
      "entropy": 5.275080585479737,
      "epoch": 7.547638773819386,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001245247892868223,
      "loss": 4.4031,
      "mean_token_accuracy": 0.24238665997982026,
      "num_tokens": 16601490.0,
      "step": 9110
    },
    {
      "entropy": 5.221807432174683,
      "epoch": 7.551781275890638,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001242873633800955,
      "loss": 4.5061,
      "mean_token_accuracy": 0.22857069820165635,
      "num_tokens": 16610972.0,
      "step": 9115
    },
    {
      "entropy": 5.249258184432984,
      "epoch": 7.555923777961889,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012405024147722643,
      "loss": 4.4751,
      "mean_token_accuracy": 0.23071240037679672,
      "num_tokens": 16619388.0,
      "step": 9120
    },
    {
      "entropy": 5.131259632110596,
      "epoch": 7.56006628003314,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001238134240565158,
      "loss": 4.4362,
      "mean_token_accuracy": 0.24466467499732972,
      "num_tokens": 16628265.0,
      "step": 9125
    },
    {
      "entropy": 5.195134830474854,
      "epoch": 7.564208782104391,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001235769115956504,
      "loss": 4.3906,
      "mean_token_accuracy": 0.24812777638435363,
      "num_tokens": 16636809.0,
      "step": 9130
    },
    {
      "entropy": 5.277749156951904,
      "epoch": 7.568351284175642,
      "grad_norm": 1.546875,
      "learning_rate": 0.00012334070457170151,
      "loss": 4.5684,
      "mean_token_accuracy": 0.22674246579408647,
      "num_tokens": 16646084.0,
      "step": 9135
    },
    {
      "entropy": 5.308511877059937,
      "epoch": 7.572493786246893,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012310480346112472,
      "loss": 4.6059,
      "mean_token_accuracy": 0.22275809794664383,
      "num_tokens": 16656196.0,
      "step": 9140
    },
    {
      "entropy": 5.211741304397583,
      "epoch": 7.576636288318144,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012286920873975815,
      "loss": 4.5057,
      "mean_token_accuracy": 0.2300925672054291,
      "num_tokens": 16665770.0,
      "step": 9145
    },
    {
      "entropy": 5.211621952056885,
      "epoch": 7.580778790389395,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012263392088282226,
      "loss": 4.4462,
      "mean_token_accuracy": 0.23721508979797362,
      "num_tokens": 16674206.0,
      "step": 9150
    },
    {
      "entropy": 5.152486896514892,
      "epoch": 7.584921292460646,
      "grad_norm": 1.375,
      "learning_rate": 0.00012239894036491833,
      "loss": 4.4683,
      "mean_token_accuracy": 0.23260106295347213,
      "num_tokens": 16683977.0,
      "step": 9155
    },
    {
      "entropy": 5.175867319107056,
      "epoch": 7.5890637945318975,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012216426766002762,
      "loss": 4.4347,
      "mean_token_accuracy": 0.23842160552740096,
      "num_tokens": 16693117.0,
      "step": 9160
    },
    {
      "entropy": 5.248047780990601,
      "epoch": 7.593206296603149,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012192990324151079,
      "loss": 4.455,
      "mean_token_accuracy": 0.2401670128107071,
      "num_tokens": 16701631.0,
      "step": 9165
    },
    {
      "entropy": 5.263347434997558,
      "epoch": 7.5973487986744,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012169584758210635,
      "loss": 4.5414,
      "mean_token_accuracy": 0.23176029771566392,
      "num_tokens": 16710337.0,
      "step": 9170
    },
    {
      "entropy": 5.181923961639404,
      "epoch": 7.60149130074565,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012146210115393025,
      "loss": 4.4458,
      "mean_token_accuracy": 0.23728128969669343,
      "num_tokens": 16719046.0,
      "step": 9175
    },
    {
      "entropy": 5.2310608386993405,
      "epoch": 7.605633802816901,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00012122866442847455,
      "loss": 4.5318,
      "mean_token_accuracy": 0.22337554395198822,
      "num_tokens": 16727860.0,
      "step": 9180
    },
    {
      "entropy": 5.228456449508667,
      "epoch": 7.609776304888152,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012099553787660654,
      "loss": 4.5542,
      "mean_token_accuracy": 0.23585160374641417,
      "num_tokens": 16737839.0,
      "step": 9185
    },
    {
      "entropy": 5.181468772888183,
      "epoch": 7.613918806959403,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012076272196856808,
      "loss": 4.4547,
      "mean_token_accuracy": 0.22863250076770783,
      "num_tokens": 16746171.0,
      "step": 9190
    },
    {
      "entropy": 5.2023216724395756,
      "epoch": 7.618061309030654,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001205302171739742,
      "loss": 4.4386,
      "mean_token_accuracy": 0.23564775586128234,
      "num_tokens": 16755236.0,
      "step": 9195
    },
    {
      "entropy": 5.20301923751831,
      "epoch": 7.6222038111019055,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012029802396181256,
      "loss": 4.4617,
      "mean_token_accuracy": 0.23734355121850967,
      "num_tokens": 16764126.0,
      "step": 9200
    },
    {
      "entropy": 5.236660766601562,
      "epoch": 7.626346313173157,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012006614280044225,
      "loss": 4.5023,
      "mean_token_accuracy": 0.24664245396852494,
      "num_tokens": 16772861.0,
      "step": 9205
    },
    {
      "entropy": 5.207330226898193,
      "epoch": 7.630488815244408,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011983457415759274,
      "loss": 4.4976,
      "mean_token_accuracy": 0.23644336462020873,
      "num_tokens": 16782119.0,
      "step": 9210
    },
    {
      "entropy": 5.259128618240356,
      "epoch": 7.634631317315659,
      "grad_norm": 1.609375,
      "learning_rate": 0.00011960331850036352,
      "loss": 4.5288,
      "mean_token_accuracy": 0.22759294211864473,
      "num_tokens": 16790531.0,
      "step": 9215
    },
    {
      "entropy": 5.190987968444825,
      "epoch": 7.63877381938691,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011937237629522233,
      "loss": 4.4482,
      "mean_token_accuracy": 0.23928841948509216,
      "num_tokens": 16799516.0,
      "step": 9220
    },
    {
      "entropy": 5.233234834671021,
      "epoch": 7.642916321458161,
      "grad_norm": 1.546875,
      "learning_rate": 0.00011914174800800501,
      "loss": 4.4924,
      "mean_token_accuracy": 0.22671060413122177,
      "num_tokens": 16807830.0,
      "step": 9225
    },
    {
      "entropy": 5.257206058502197,
      "epoch": 7.647058823529412,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011891143410391387,
      "loss": 4.5441,
      "mean_token_accuracy": 0.22860875725746155,
      "num_tokens": 16816987.0,
      "step": 9230
    },
    {
      "entropy": 5.1971296787261965,
      "epoch": 7.651201325600663,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011868143504751732,
      "loss": 4.3915,
      "mean_token_accuracy": 0.23751573413610458,
      "num_tokens": 16825723.0,
      "step": 9235
    },
    {
      "entropy": 5.22703104019165,
      "epoch": 7.6553438276719135,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011845175130274848,
      "loss": 4.4707,
      "mean_token_accuracy": 0.24050054401159288,
      "num_tokens": 16835429.0,
      "step": 9240
    },
    {
      "entropy": 5.197973680496216,
      "epoch": 7.659486329743165,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011822238333290475,
      "loss": 4.4352,
      "mean_token_accuracy": 0.2480204850435257,
      "num_tokens": 16844693.0,
      "step": 9245
    },
    {
      "entropy": 5.204640674591064,
      "epoch": 7.663628831814416,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011799333160064617,
      "loss": 4.4947,
      "mean_token_accuracy": 0.23266191333532332,
      "num_tokens": 16853058.0,
      "step": 9250
    },
    {
      "entropy": 5.2966637134552,
      "epoch": 7.667771333885667,
      "grad_norm": 1.375,
      "learning_rate": 0.0001177645965679953,
      "loss": 4.5287,
      "mean_token_accuracy": 0.2271040141582489,
      "num_tokens": 16863017.0,
      "step": 9255
    },
    {
      "entropy": 5.1807972431182865,
      "epoch": 7.671913835956918,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011753617869633561,
      "loss": 4.3862,
      "mean_token_accuracy": 0.24129059463739394,
      "num_tokens": 16872054.0,
      "step": 9260
    },
    {
      "entropy": 5.279234790802002,
      "epoch": 7.676056338028169,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011730807844641085,
      "loss": 4.514,
      "mean_token_accuracy": 0.23627052307128907,
      "num_tokens": 16881554.0,
      "step": 9265
    },
    {
      "entropy": 5.198402547836304,
      "epoch": 7.68019884009942,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011708029627832431,
      "loss": 4.5037,
      "mean_token_accuracy": 0.2366371124982834,
      "num_tokens": 16890050.0,
      "step": 9270
    },
    {
      "entropy": 5.245404005050659,
      "epoch": 7.684341342170671,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011685283265153732,
      "loss": 4.4402,
      "mean_token_accuracy": 0.2328493371605873,
      "num_tokens": 16899298.0,
      "step": 9275
    },
    {
      "entropy": 5.274530220031738,
      "epoch": 7.688483844241922,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011662568802486905,
      "loss": 4.5442,
      "mean_token_accuracy": 0.22722618877887726,
      "num_tokens": 16908227.0,
      "step": 9280
    },
    {
      "entropy": 5.247538375854492,
      "epoch": 7.692626346313173,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011639886285649489,
      "loss": 4.5289,
      "mean_token_accuracy": 0.22495685517787933,
      "num_tokens": 16917708.0,
      "step": 9285
    },
    {
      "entropy": 5.16684193611145,
      "epoch": 7.6967688483844245,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011617235760394614,
      "loss": 4.372,
      "mean_token_accuracy": 0.24757057875394822,
      "num_tokens": 16926108.0,
      "step": 9290
    },
    {
      "entropy": 5.240006351470948,
      "epoch": 7.700911350455675,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001159461727241085,
      "loss": 4.4901,
      "mean_token_accuracy": 0.22910903990268708,
      "num_tokens": 16935821.0,
      "step": 9295
    },
    {
      "entropy": 5.22453088760376,
      "epoch": 7.705053852526927,
      "grad_norm": 1.5,
      "learning_rate": 0.0001157203086732217,
      "loss": 4.527,
      "mean_token_accuracy": 0.23414590507745742,
      "num_tokens": 16944899.0,
      "step": 9300
    },
    {
      "entropy": 5.225587940216064,
      "epoch": 7.709196354598177,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011549476590687811,
      "loss": 4.468,
      "mean_token_accuracy": 0.23945072144269944,
      "num_tokens": 16953745.0,
      "step": 9305
    },
    {
      "entropy": 5.300943613052368,
      "epoch": 7.713338856669428,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011526954488002227,
      "loss": 4.5719,
      "mean_token_accuracy": 0.22554791420698167,
      "num_tokens": 16962940.0,
      "step": 9310
    },
    {
      "entropy": 5.227784442901611,
      "epoch": 7.717481358740679,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011504464604694953,
      "loss": 4.4867,
      "mean_token_accuracy": 0.23215437233448027,
      "num_tokens": 16971322.0,
      "step": 9315
    },
    {
      "entropy": 5.254767274856567,
      "epoch": 7.72162386081193,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011482006986130533,
      "loss": 4.4245,
      "mean_token_accuracy": 0.24235508292913438,
      "num_tokens": 16980066.0,
      "step": 9320
    },
    {
      "entropy": 5.257081317901611,
      "epoch": 7.725766362883181,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011459581677608453,
      "loss": 4.5111,
      "mean_token_accuracy": 0.23566131442785263,
      "num_tokens": 16989202.0,
      "step": 9325
    },
    {
      "entropy": 5.192407178878784,
      "epoch": 7.7299088649544325,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011437188724362997,
      "loss": 4.5258,
      "mean_token_accuracy": 0.22844370156526567,
      "num_tokens": 16999267.0,
      "step": 9330
    },
    {
      "entropy": 5.163544845581055,
      "epoch": 7.734051367025684,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011414828171563214,
      "loss": 4.4724,
      "mean_token_accuracy": 0.2406766265630722,
      "num_tokens": 17008252.0,
      "step": 9335
    },
    {
      "entropy": 5.217708015441895,
      "epoch": 7.738193869096935,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011392500064312773,
      "loss": 4.469,
      "mean_token_accuracy": 0.23498304337263107,
      "num_tokens": 17017568.0,
      "step": 9340
    },
    {
      "entropy": 5.266183662414551,
      "epoch": 7.742336371168186,
      "grad_norm": 1.59375,
      "learning_rate": 0.000113702044476499,
      "loss": 4.4723,
      "mean_token_accuracy": 0.2342673808336258,
      "num_tokens": 17026206.0,
      "step": 9345
    },
    {
      "entropy": 5.217392492294311,
      "epoch": 7.746478873239437,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011347941366547303,
      "loss": 4.4547,
      "mean_token_accuracy": 0.2351702257990837,
      "num_tokens": 17035114.0,
      "step": 9350
    },
    {
      "entropy": 5.247145938873291,
      "epoch": 7.750621375310688,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001132571086591204,
      "loss": 4.5209,
      "mean_token_accuracy": 0.22884466052055358,
      "num_tokens": 17044459.0,
      "step": 9355
    },
    {
      "entropy": 5.253954124450684,
      "epoch": 7.754763877381938,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001130351299058547,
      "loss": 4.5402,
      "mean_token_accuracy": 0.22237540334463118,
      "num_tokens": 17053985.0,
      "step": 9360
    },
    {
      "entropy": 5.177068376541138,
      "epoch": 7.758906379453189,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011281347785343127,
      "loss": 4.3585,
      "mean_token_accuracy": 0.24387803971767424,
      "num_tokens": 17063136.0,
      "step": 9365
    },
    {
      "entropy": 5.221052789688111,
      "epoch": 7.7630488815244405,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011259215294894647,
      "loss": 4.5065,
      "mean_token_accuracy": 0.2302239328622818,
      "num_tokens": 17072423.0,
      "step": 9370
    },
    {
      "entropy": 5.219628095626831,
      "epoch": 7.767191383595692,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011237115563883693,
      "loss": 4.4576,
      "mean_token_accuracy": 0.2357018247246742,
      "num_tokens": 17080578.0,
      "step": 9375
    },
    {
      "entropy": 5.209923458099365,
      "epoch": 7.771333885666943,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011215048636887825,
      "loss": 4.4118,
      "mean_token_accuracy": 0.24526971578598022,
      "num_tokens": 17089995.0,
      "step": 9380
    },
    {
      "entropy": 5.149524831771851,
      "epoch": 7.775476387738194,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001119301455841846,
      "loss": 4.3985,
      "mean_token_accuracy": 0.23359975218772888,
      "num_tokens": 17099659.0,
      "step": 9385
    },
    {
      "entropy": 5.249895238876343,
      "epoch": 7.779618889809445,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011171013372920735,
      "loss": 4.5109,
      "mean_token_accuracy": 0.2335706502199173,
      "num_tokens": 17109972.0,
      "step": 9390
    },
    {
      "entropy": 5.286593961715698,
      "epoch": 7.783761391880696,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011149045124773441,
      "loss": 4.6067,
      "mean_token_accuracy": 0.22540475726127623,
      "num_tokens": 17120288.0,
      "step": 9395
    },
    {
      "entropy": 5.264656019210816,
      "epoch": 7.787903893951947,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00011127109858288945,
      "loss": 4.5056,
      "mean_token_accuracy": 0.24548871964216232,
      "num_tokens": 17130577.0,
      "step": 9400
    },
    {
      "entropy": 5.255543613433838,
      "epoch": 7.792046396023198,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011105207617713065,
      "loss": 4.5676,
      "mean_token_accuracy": 0.2278539225459099,
      "num_tokens": 17140782.0,
      "step": 9405
    },
    {
      "entropy": 5.183806562423706,
      "epoch": 7.796188898094449,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011083338447225027,
      "loss": 4.4729,
      "mean_token_accuracy": 0.23438529223203658,
      "num_tokens": 17150013.0,
      "step": 9410
    },
    {
      "entropy": 5.295999574661255,
      "epoch": 7.8003314001657,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011061502390937331,
      "loss": 4.5095,
      "mean_token_accuracy": 0.22928863763809204,
      "num_tokens": 17159349.0,
      "step": 9415
    },
    {
      "entropy": 5.261645221710205,
      "epoch": 7.804473902236952,
      "grad_norm": 1.5,
      "learning_rate": 0.0001103969949289568,
      "loss": 4.5491,
      "mean_token_accuracy": 0.23312466591596603,
      "num_tokens": 17168063.0,
      "step": 9420
    },
    {
      "entropy": 5.265921020507813,
      "epoch": 7.808616404308202,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011017929797078919,
      "loss": 4.5093,
      "mean_token_accuracy": 0.2266368091106415,
      "num_tokens": 17177087.0,
      "step": 9425
    },
    {
      "entropy": 5.238058090209961,
      "epoch": 7.812758906379453,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010996193347398886,
      "loss": 4.476,
      "mean_token_accuracy": 0.245611672103405,
      "num_tokens": 17186875.0,
      "step": 9430
    },
    {
      "entropy": 5.232633113861084,
      "epoch": 7.816901408450704,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010974490187700394,
      "loss": 4.538,
      "mean_token_accuracy": 0.22815486788749695,
      "num_tokens": 17195666.0,
      "step": 9435
    },
    {
      "entropy": 5.2528979778289795,
      "epoch": 7.821043910521955,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001095282036176107,
      "loss": 4.4768,
      "mean_token_accuracy": 0.23304449766874313,
      "num_tokens": 17205145.0,
      "step": 9440
    },
    {
      "entropy": 5.307254981994629,
      "epoch": 7.825186412593206,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010931183913291335,
      "loss": 4.5046,
      "mean_token_accuracy": 0.22929927557706833,
      "num_tokens": 17214106.0,
      "step": 9445
    },
    {
      "entropy": 5.19197301864624,
      "epoch": 7.829328914664457,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010909580885934265,
      "loss": 4.3982,
      "mean_token_accuracy": 0.24959655553102494,
      "num_tokens": 17223343.0,
      "step": 9450
    },
    {
      "entropy": 5.217424964904785,
      "epoch": 7.8334714167357085,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010888011323265517,
      "loss": 4.5325,
      "mean_token_accuracy": 0.22958433777093887,
      "num_tokens": 17233528.0,
      "step": 9455
    },
    {
      "entropy": 5.188608694076538,
      "epoch": 7.83761391880696,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010866475268793275,
      "loss": 4.4205,
      "mean_token_accuracy": 0.23154864311218262,
      "num_tokens": 17242026.0,
      "step": 9460
    },
    {
      "entropy": 5.163565731048584,
      "epoch": 7.841756420878211,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010844972765958096,
      "loss": 4.4507,
      "mean_token_accuracy": 0.24438020586967468,
      "num_tokens": 17251203.0,
      "step": 9465
    },
    {
      "entropy": 5.1432568550109865,
      "epoch": 7.845898922949462,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010823503858132894,
      "loss": 4.3773,
      "mean_token_accuracy": 0.24119524657726288,
      "num_tokens": 17260072.0,
      "step": 9470
    },
    {
      "entropy": 5.213344573974609,
      "epoch": 7.850041425020713,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010802068588622793,
      "loss": 4.5079,
      "mean_token_accuracy": 0.2356532707810402,
      "num_tokens": 17269015.0,
      "step": 9475
    },
    {
      "entropy": 5.1996417999267575,
      "epoch": 7.854183927091963,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010780667000665067,
      "loss": 4.4504,
      "mean_token_accuracy": 0.2433476373553276,
      "num_tokens": 17277721.0,
      "step": 9480
    },
    {
      "entropy": 5.223092889785766,
      "epoch": 7.858326429163214,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010759299137429073,
      "loss": 4.5235,
      "mean_token_accuracy": 0.2347540259361267,
      "num_tokens": 17286805.0,
      "step": 9485
    },
    {
      "entropy": 5.248261737823486,
      "epoch": 7.862468931234465,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010737965042016116,
      "loss": 4.5776,
      "mean_token_accuracy": 0.2246888503432274,
      "num_tokens": 17297577.0,
      "step": 9490
    },
    {
      "entropy": 5.249375915527343,
      "epoch": 7.866611433305716,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010716664757459404,
      "loss": 4.4837,
      "mean_token_accuracy": 0.23761717677116395,
      "num_tokens": 17306620.0,
      "step": 9495
    },
    {
      "entropy": 5.244761276245117,
      "epoch": 7.8707539353769675,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010695398326723938,
      "loss": 4.5249,
      "mean_token_accuracy": 0.2251071587204933,
      "num_tokens": 17314796.0,
      "step": 9500
    },
    {
      "epoch": 7.8707539353769675,
      "eval_entropy": 5.14112121972448,
      "eval_loss": 5.765201568603516,
      "eval_mean_token_accuracy": 0.16707931234956042,
      "eval_num_tokens": 17314796.0,
      "eval_runtime": 2.7492,
      "eval_samples_per_second": 1362.958,
      "eval_steps_per_second": 170.597,
      "step": 9500
    },
    {
      "entropy": 5.2463031768798825,
      "epoch": 7.874896437448219,
      "grad_norm": 1.609375,
      "learning_rate": 0.00010674165792706423,
      "loss": 4.5197,
      "mean_token_accuracy": 0.2386341780424118,
      "num_tokens": 17324048.0,
      "step": 9505
    },
    {
      "entropy": 5.237146329879761,
      "epoch": 7.87903893951947,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010652967198235214,
      "loss": 4.4868,
      "mean_token_accuracy": 0.23583484143018724,
      "num_tokens": 17333354.0,
      "step": 9510
    },
    {
      "entropy": 5.212380695343017,
      "epoch": 7.883181441590721,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001063180258607018,
      "loss": 4.4201,
      "mean_token_accuracy": 0.25275374203920364,
      "num_tokens": 17343026.0,
      "step": 9515
    },
    {
      "entropy": 5.161926698684693,
      "epoch": 7.887323943661972,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010610671998902671,
      "loss": 4.402,
      "mean_token_accuracy": 0.24219682812690735,
      "num_tokens": 17351161.0,
      "step": 9520
    },
    {
      "entropy": 5.2111059665679935,
      "epoch": 7.891466445733223,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010589575479355379,
      "loss": 4.42,
      "mean_token_accuracy": 0.23753047585487366,
      "num_tokens": 17360697.0,
      "step": 9525
    },
    {
      "entropy": 5.243257808685303,
      "epoch": 7.895608947804474,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010568513069982289,
      "loss": 4.5202,
      "mean_token_accuracy": 0.2338667094707489,
      "num_tokens": 17370303.0,
      "step": 9530
    },
    {
      "entropy": 5.2183421611785885,
      "epoch": 7.899751449875725,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00010547484813268594,
      "loss": 4.4817,
      "mean_token_accuracy": 0.23335253447294235,
      "num_tokens": 17379557.0,
      "step": 9535
    },
    {
      "entropy": 5.265551805496216,
      "epoch": 7.903893951946976,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010526490751630568,
      "loss": 4.5357,
      "mean_token_accuracy": 0.2362557277083397,
      "num_tokens": 17388206.0,
      "step": 9540
    },
    {
      "entropy": 5.273835611343384,
      "epoch": 7.908036454018227,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010505530927415544,
      "loss": 4.5787,
      "mean_token_accuracy": 0.23471909016370773,
      "num_tokens": 17398055.0,
      "step": 9545
    },
    {
      "entropy": 5.153480911254883,
      "epoch": 7.912178956089478,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010484605382901772,
      "loss": 4.3729,
      "mean_token_accuracy": 0.23927945792675018,
      "num_tokens": 17407214.0,
      "step": 9550
    },
    {
      "entropy": 5.246860647201538,
      "epoch": 7.916321458160729,
      "grad_norm": 1.5,
      "learning_rate": 0.00010463714160298354,
      "loss": 4.5115,
      "mean_token_accuracy": 0.23394761085510254,
      "num_tokens": 17416447.0,
      "step": 9555
    },
    {
      "entropy": 5.223754644393921,
      "epoch": 7.92046396023198,
      "grad_norm": 1.546875,
      "learning_rate": 0.00010442857301745184,
      "loss": 4.5218,
      "mean_token_accuracy": 0.2308438703417778,
      "num_tokens": 17424921.0,
      "step": 9560
    },
    {
      "entropy": 5.08242826461792,
      "epoch": 7.924606462303231,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001042203484931281,
      "loss": 4.3436,
      "mean_token_accuracy": 0.24109654277563095,
      "num_tokens": 17434425.0,
      "step": 9565
    },
    {
      "entropy": 5.196365737915039,
      "epoch": 7.928748964374482,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010401246845002407,
      "loss": 4.4766,
      "mean_token_accuracy": 0.234826922416687,
      "num_tokens": 17443879.0,
      "step": 9570
    },
    {
      "entropy": 5.211728239059449,
      "epoch": 7.932891466445733,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010380493330745641,
      "loss": 4.4595,
      "mean_token_accuracy": 0.241335928440094,
      "num_tokens": 17452276.0,
      "step": 9575
    },
    {
      "entropy": 5.223777341842651,
      "epoch": 7.937033968516984,
      "grad_norm": 1.5625,
      "learning_rate": 0.00010359774348404626,
      "loss": 4.4935,
      "mean_token_accuracy": 0.23449694067239762,
      "num_tokens": 17460706.0,
      "step": 9580
    },
    {
      "entropy": 5.2291717529296875,
      "epoch": 7.9411764705882355,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010339089939771814,
      "loss": 4.5232,
      "mean_token_accuracy": 0.23522074073553084,
      "num_tokens": 17469754.0,
      "step": 9585
    },
    {
      "entropy": 5.246913623809815,
      "epoch": 7.945318972659487,
      "grad_norm": 1.375,
      "learning_rate": 0.00010318440146569904,
      "loss": 4.5998,
      "mean_token_accuracy": 0.2202611029148102,
      "num_tokens": 17480057.0,
      "step": 9590
    },
    {
      "entropy": 5.193147659301758,
      "epoch": 7.949461474730738,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010297825010451806,
      "loss": 4.4757,
      "mean_token_accuracy": 0.2301161840558052,
      "num_tokens": 17490427.0,
      "step": 9595
    },
    {
      "entropy": 5.219490623474121,
      "epoch": 7.953603976801988,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001027724457300048,
      "loss": 4.4943,
      "mean_token_accuracy": 0.2302616700530052,
      "num_tokens": 17499105.0,
      "step": 9600
    },
    {
      "entropy": 5.252292966842651,
      "epoch": 7.957746478873239,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010256698875728938,
      "loss": 4.5353,
      "mean_token_accuracy": 0.23300056755542756,
      "num_tokens": 17508499.0,
      "step": 9605
    },
    {
      "entropy": 5.181434965133667,
      "epoch": 7.96188898094449,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010236187960080084,
      "loss": 4.435,
      "mean_token_accuracy": 0.24067785292863847,
      "num_tokens": 17516885.0,
      "step": 9610
    },
    {
      "entropy": 5.213464784622192,
      "epoch": 7.966031483015741,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00010215711867426678,
      "loss": 4.4018,
      "mean_token_accuracy": 0.2380411520600319,
      "num_tokens": 17524958.0,
      "step": 9615
    },
    {
      "entropy": 5.2348590850830075,
      "epoch": 7.970173985086992,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010195270639071243,
      "loss": 4.4817,
      "mean_token_accuracy": 0.2264284998178482,
      "num_tokens": 17534185.0,
      "step": 9620
    },
    {
      "entropy": 5.245913696289063,
      "epoch": 7.9743164871582435,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001017486431624596,
      "loss": 4.5063,
      "mean_token_accuracy": 0.2343599259853363,
      "num_tokens": 17542696.0,
      "step": 9625
    },
    {
      "entropy": 5.179908561706543,
      "epoch": 7.978458989229495,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010154492940112624,
      "loss": 4.3974,
      "mean_token_accuracy": 0.2521152153611183,
      "num_tokens": 17552751.0,
      "step": 9630
    },
    {
      "entropy": 5.275719881057739,
      "epoch": 7.982601491300746,
      "grad_norm": 1.28125,
      "learning_rate": 0.00010134156551762519,
      "loss": 4.5257,
      "mean_token_accuracy": 0.22792626172304153,
      "num_tokens": 17561973.0,
      "step": 9635
    },
    {
      "entropy": 5.2567394256591795,
      "epoch": 7.986743993371997,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010113855192216359,
      "loss": 4.5481,
      "mean_token_accuracy": 0.23013689368963242,
      "num_tokens": 17571238.0,
      "step": 9640
    },
    {
      "entropy": 5.238293600082398,
      "epoch": 7.990886495443248,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00010093588902424217,
      "loss": 4.4414,
      "mean_token_accuracy": 0.24236381351947783,
      "num_tokens": 17580189.0,
      "step": 9645
    },
    {
      "entropy": 5.219696187973023,
      "epoch": 7.995028997514499,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00010073357723265398,
      "loss": 4.529,
      "mean_token_accuracy": 0.22826115190982818,
      "num_tokens": 17590051.0,
      "step": 9650
    },
    {
      "entropy": 5.24488639831543,
      "epoch": 7.99917149958575,
      "grad_norm": 1.53125,
      "learning_rate": 0.00010053161695548418,
      "loss": 4.5126,
      "mean_token_accuracy": 0.23572204113006592,
      "num_tokens": 17599036.0,
      "step": 9655
    },
    {
      "entropy": 5.198426389694214,
      "epoch": 8.003314001657001,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010033000860010866,
      "loss": 4.3936,
      "mean_token_accuracy": 0.24143406003713608,
      "num_tokens": 17608433.0,
      "step": 9660
    },
    {
      "entropy": 5.227721166610718,
      "epoch": 8.007456503728251,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001001287525731934,
      "loss": 4.4047,
      "mean_token_accuracy": 0.24120991080999374,
      "num_tokens": 17617696.0,
      "step": 9665
    },
    {
      "entropy": 5.21623420715332,
      "epoch": 8.011599005799503,
      "grad_norm": 1.3984375,
      "learning_rate": 9.992784928069397e-05,
      "loss": 4.4525,
      "mean_token_accuracy": 0.2320122942328453,
      "num_tokens": 17626619.0,
      "step": 9670
    },
    {
      "entropy": 5.214959383010864,
      "epoch": 8.015741507870754,
      "grad_norm": 1.5078125,
      "learning_rate": 9.972729912785411e-05,
      "loss": 4.3274,
      "mean_token_accuracy": 0.2510974645614624,
      "num_tokens": 17635306.0,
      "step": 9675
    },
    {
      "entropy": 5.1959240436553955,
      "epoch": 8.019884009942006,
      "grad_norm": 1.6015625,
      "learning_rate": 9.95271025192055e-05,
      "loss": 4.3134,
      "mean_token_accuracy": 0.2534480854868889,
      "num_tokens": 17643526.0,
      "step": 9680
    },
    {
      "entropy": 5.268948078155518,
      "epoch": 8.024026512013256,
      "grad_norm": 1.4765625,
      "learning_rate": 9.93272598585665e-05,
      "loss": 4.4883,
      "mean_token_accuracy": 0.22948536276817322,
      "num_tokens": 17652474.0,
      "step": 9685
    },
    {
      "entropy": 5.17322416305542,
      "epoch": 8.028169014084508,
      "grad_norm": 1.3671875,
      "learning_rate": 9.912777154904171e-05,
      "loss": 4.4324,
      "mean_token_accuracy": 0.2421304926276207,
      "num_tokens": 17662025.0,
      "step": 9690
    },
    {
      "entropy": 5.1476397037506105,
      "epoch": 8.032311516155758,
      "grad_norm": 1.421875,
      "learning_rate": 9.89286379930207e-05,
      "loss": 4.3473,
      "mean_token_accuracy": 0.24823604971170427,
      "num_tokens": 17670908.0,
      "step": 9695
    },
    {
      "entropy": 5.143283033370972,
      "epoch": 8.036454018227008,
      "grad_norm": 1.3984375,
      "learning_rate": 9.872985959217779e-05,
      "loss": 4.3484,
      "mean_token_accuracy": 0.2501207798719406,
      "num_tokens": 17681242.0,
      "step": 9700
    },
    {
      "entropy": 5.1804948329925535,
      "epoch": 8.04059652029826,
      "grad_norm": 1.5234375,
      "learning_rate": 9.853143674747053e-05,
      "loss": 4.3046,
      "mean_token_accuracy": 0.25343292355537417,
      "num_tokens": 17689816.0,
      "step": 9705
    },
    {
      "entropy": 5.21025447845459,
      "epoch": 8.04473902236951,
      "grad_norm": 1.3984375,
      "learning_rate": 9.833336985913965e-05,
      "loss": 4.4057,
      "mean_token_accuracy": 0.24678899198770524,
      "num_tokens": 17699251.0,
      "step": 9710
    },
    {
      "entropy": 5.246069288253784,
      "epoch": 8.048881524440763,
      "grad_norm": 1.4609375,
      "learning_rate": 9.813565932670767e-05,
      "loss": 4.4405,
      "mean_token_accuracy": 0.24430164247751235,
      "num_tokens": 17708906.0,
      "step": 9715
    },
    {
      "entropy": 5.287776374816895,
      "epoch": 8.053024026512013,
      "grad_norm": 1.4921875,
      "learning_rate": 9.793830554897826e-05,
      "loss": 4.427,
      "mean_token_accuracy": 0.23462353646755219,
      "num_tokens": 17718071.0,
      "step": 9720
    },
    {
      "entropy": 5.18203387260437,
      "epoch": 8.057166528583265,
      "grad_norm": 1.4140625,
      "learning_rate": 9.774130892403566e-05,
      "loss": 4.3187,
      "mean_token_accuracy": 0.2543671905994415,
      "num_tokens": 17727690.0,
      "step": 9725
    },
    {
      "entropy": 5.186534881591797,
      "epoch": 8.061309030654515,
      "grad_norm": 1.4921875,
      "learning_rate": 9.754466984924348e-05,
      "loss": 4.3299,
      "mean_token_accuracy": 0.24183301031589508,
      "num_tokens": 17735934.0,
      "step": 9730
    },
    {
      "entropy": 5.200703716278076,
      "epoch": 8.065451532725767,
      "grad_norm": 1.4453125,
      "learning_rate": 9.734838872124432e-05,
      "loss": 4.3371,
      "mean_token_accuracy": 0.24433684200048447,
      "num_tokens": 17744323.0,
      "step": 9735
    },
    {
      "entropy": 5.13489556312561,
      "epoch": 8.069594034797017,
      "grad_norm": 1.3515625,
      "learning_rate": 9.715246593595859e-05,
      "loss": 4.325,
      "mean_token_accuracy": 0.2514267981052399,
      "num_tokens": 17753335.0,
      "step": 9740
    },
    {
      "entropy": 5.178722763061524,
      "epoch": 8.07373653686827,
      "grad_norm": 1.5,
      "learning_rate": 9.695690188858408e-05,
      "loss": 4.3817,
      "mean_token_accuracy": 0.24246701151132583,
      "num_tokens": 17761952.0,
      "step": 9745
    },
    {
      "entropy": 5.24194598197937,
      "epoch": 8.07787903893952,
      "grad_norm": 1.5390625,
      "learning_rate": 9.676169697359469e-05,
      "loss": 4.4415,
      "mean_token_accuracy": 0.24678991585969925,
      "num_tokens": 17771341.0,
      "step": 9750
    },
    {
      "entropy": 5.182179355621338,
      "epoch": 8.08202154101077,
      "grad_norm": 1.375,
      "learning_rate": 9.656685158474018e-05,
      "loss": 4.3511,
      "mean_token_accuracy": 0.2486436203122139,
      "num_tokens": 17780289.0,
      "step": 9755
    },
    {
      "entropy": 5.226418399810791,
      "epoch": 8.086164043082022,
      "grad_norm": 1.515625,
      "learning_rate": 9.63723661150449e-05,
      "loss": 4.4209,
      "mean_token_accuracy": 0.23876857608556748,
      "num_tokens": 17789317.0,
      "step": 9760
    },
    {
      "entropy": 5.293474245071411,
      "epoch": 8.090306545153272,
      "grad_norm": 1.53125,
      "learning_rate": 9.617824095680742e-05,
      "loss": 4.4883,
      "mean_token_accuracy": 0.23701299130916595,
      "num_tokens": 17797875.0,
      "step": 9765
    },
    {
      "entropy": 5.236095380783081,
      "epoch": 8.094449047224524,
      "grad_norm": 1.421875,
      "learning_rate": 9.598447650159931e-05,
      "loss": 4.4262,
      "mean_token_accuracy": 0.24168868958950043,
      "num_tokens": 17806603.0,
      "step": 9770
    },
    {
      "entropy": 5.318118476867676,
      "epoch": 8.098591549295774,
      "grad_norm": 1.5390625,
      "learning_rate": 9.579107314026459e-05,
      "loss": 4.4755,
      "mean_token_accuracy": 0.22686209678649902,
      "num_tokens": 17815747.0,
      "step": 9775
    },
    {
      "entropy": 5.208037662506103,
      "epoch": 8.102734051367026,
      "grad_norm": 1.4296875,
      "learning_rate": 9.559803126291911e-05,
      "loss": 4.3535,
      "mean_token_accuracy": 0.23870368748903276,
      "num_tokens": 17824568.0,
      "step": 9780
    },
    {
      "entropy": 5.260161352157593,
      "epoch": 8.106876553438276,
      "grad_norm": 1.40625,
      "learning_rate": 9.540535125894931e-05,
      "loss": 4.4381,
      "mean_token_accuracy": 0.24815064072608947,
      "num_tokens": 17834244.0,
      "step": 9785
    },
    {
      "entropy": 5.2032942295074465,
      "epoch": 8.111019055509528,
      "grad_norm": 1.453125,
      "learning_rate": 9.521303351701193e-05,
      "loss": 4.3399,
      "mean_token_accuracy": 0.24817773699760437,
      "num_tokens": 17843285.0,
      "step": 9790
    },
    {
      "entropy": 5.233155822753906,
      "epoch": 8.115161557580779,
      "grad_norm": 1.4296875,
      "learning_rate": 9.502107842503282e-05,
      "loss": 4.4295,
      "mean_token_accuracy": 0.2504579395055771,
      "num_tokens": 17852379.0,
      "step": 9795
    },
    {
      "entropy": 5.218298053741455,
      "epoch": 8.11930405965203,
      "grad_norm": 1.421875,
      "learning_rate": 9.48294863702063e-05,
      "loss": 4.3936,
      "mean_token_accuracy": 0.25013517439365385,
      "num_tokens": 17861948.0,
      "step": 9800
    },
    {
      "entropy": 5.22992057800293,
      "epoch": 8.12344656172328,
      "grad_norm": 1.4453125,
      "learning_rate": 9.463825773899465e-05,
      "loss": 4.3493,
      "mean_token_accuracy": 0.25128971934318545,
      "num_tokens": 17870445.0,
      "step": 9805
    },
    {
      "entropy": 5.237100839614868,
      "epoch": 8.127589063794533,
      "grad_norm": 1.4453125,
      "learning_rate": 9.444739291712673e-05,
      "loss": 4.4081,
      "mean_token_accuracy": 0.2527842879295349,
      "num_tokens": 17879380.0,
      "step": 9810
    },
    {
      "entropy": 5.2134534358978275,
      "epoch": 8.131731565865783,
      "grad_norm": 1.421875,
      "learning_rate": 9.425689228959791e-05,
      "loss": 4.4079,
      "mean_token_accuracy": 0.2444216340780258,
      "num_tokens": 17888631.0,
      "step": 9815
    },
    {
      "entropy": 5.195185136795044,
      "epoch": 8.135874067937033,
      "grad_norm": 1.3828125,
      "learning_rate": 9.406675624066868e-05,
      "loss": 4.3816,
      "mean_token_accuracy": 0.24284880310297013,
      "num_tokens": 17898541.0,
      "step": 9820
    },
    {
      "entropy": 5.272396850585937,
      "epoch": 8.140016570008285,
      "grad_norm": 1.5546875,
      "learning_rate": 9.38769851538642e-05,
      "loss": 4.3866,
      "mean_token_accuracy": 0.2464823067188263,
      "num_tokens": 17906664.0,
      "step": 9825
    },
    {
      "entropy": 5.201102447509766,
      "epoch": 8.144159072079535,
      "grad_norm": 1.484375,
      "learning_rate": 9.368757941197356e-05,
      "loss": 4.3965,
      "mean_token_accuracy": 0.24761822670698166,
      "num_tokens": 17916201.0,
      "step": 9830
    },
    {
      "entropy": 5.264723587036133,
      "epoch": 8.148301574150787,
      "grad_norm": 1.3203125,
      "learning_rate": 9.34985393970488e-05,
      "loss": 4.4115,
      "mean_token_accuracy": 0.2420223832130432,
      "num_tokens": 17925700.0,
      "step": 9835
    },
    {
      "entropy": 5.189842796325683,
      "epoch": 8.152444076222038,
      "grad_norm": 1.546875,
      "learning_rate": 9.330986549040432e-05,
      "loss": 4.3717,
      "mean_token_accuracy": 0.24570300728082656,
      "num_tokens": 17934293.0,
      "step": 9840
    },
    {
      "entropy": 5.179622411727905,
      "epoch": 8.15658657829329,
      "grad_norm": 1.375,
      "learning_rate": 9.312155807261602e-05,
      "loss": 4.3985,
      "mean_token_accuracy": 0.23710400462150574,
      "num_tokens": 17943541.0,
      "step": 9845
    },
    {
      "entropy": 5.253179264068604,
      "epoch": 8.16072908036454,
      "grad_norm": 1.3359375,
      "learning_rate": 9.293361752352045e-05,
      "loss": 4.3996,
      "mean_token_accuracy": 0.25183241069316864,
      "num_tokens": 17952616.0,
      "step": 9850
    },
    {
      "entropy": 5.236699962615967,
      "epoch": 8.164871582435792,
      "grad_norm": 1.3671875,
      "learning_rate": 9.274604422221433e-05,
      "loss": 4.356,
      "mean_token_accuracy": 0.24857900738716127,
      "num_tokens": 17961414.0,
      "step": 9855
    },
    {
      "entropy": 5.220491218566894,
      "epoch": 8.169014084507042,
      "grad_norm": 1.421875,
      "learning_rate": 9.255883854705344e-05,
      "loss": 4.4375,
      "mean_token_accuracy": 0.24180857241153716,
      "num_tokens": 17972077.0,
      "step": 9860
    },
    {
      "entropy": 5.184313631057739,
      "epoch": 8.173156586578294,
      "grad_norm": 1.46875,
      "learning_rate": 9.237200087565217e-05,
      "loss": 4.3655,
      "mean_token_accuracy": 0.24715562164783478,
      "num_tokens": 17980303.0,
      "step": 9865
    },
    {
      "entropy": 5.27153205871582,
      "epoch": 8.177299088649544,
      "grad_norm": 1.453125,
      "learning_rate": 9.218553158488238e-05,
      "loss": 4.4477,
      "mean_token_accuracy": 0.2402723342180252,
      "num_tokens": 17990410.0,
      "step": 9870
    },
    {
      "entropy": 5.205586528778076,
      "epoch": 8.181441590720794,
      "grad_norm": 1.4765625,
      "learning_rate": 9.199943105087319e-05,
      "loss": 4.3413,
      "mean_token_accuracy": 0.24648494720458985,
      "num_tokens": 17998931.0,
      "step": 9875
    },
    {
      "entropy": 5.221256732940674,
      "epoch": 8.185584092792046,
      "grad_norm": 1.453125,
      "learning_rate": 9.181369964900957e-05,
      "loss": 4.398,
      "mean_token_accuracy": 0.2453961968421936,
      "num_tokens": 18007853.0,
      "step": 9880
    },
    {
      "entropy": 5.213214921951294,
      "epoch": 8.189726594863297,
      "grad_norm": 1.484375,
      "learning_rate": 9.162833775393209e-05,
      "loss": 4.3602,
      "mean_token_accuracy": 0.24233943670988084,
      "num_tokens": 18016517.0,
      "step": 9885
    },
    {
      "entropy": 5.267703199386597,
      "epoch": 8.193869096934549,
      "grad_norm": 1.5078125,
      "learning_rate": 9.1443345739536e-05,
      "loss": 4.4452,
      "mean_token_accuracy": 0.24340564459562303,
      "num_tokens": 18025057.0,
      "step": 9890
    },
    {
      "entropy": 5.2295369625091555,
      "epoch": 8.198011599005799,
      "grad_norm": 1.4609375,
      "learning_rate": 9.12587239789703e-05,
      "loss": 4.4434,
      "mean_token_accuracy": 0.2375107914209366,
      "num_tokens": 18034371.0,
      "step": 9895
    },
    {
      "entropy": 5.196250915527344,
      "epoch": 8.202154101077051,
      "grad_norm": 1.421875,
      "learning_rate": 9.10744728446374e-05,
      "loss": 4.4147,
      "mean_token_accuracy": 0.24435777217149734,
      "num_tokens": 18043354.0,
      "step": 9900
    },
    {
      "entropy": 5.186095809936523,
      "epoch": 8.206296603148301,
      "grad_norm": 1.5,
      "learning_rate": 9.089059270819187e-05,
      "loss": 4.3615,
      "mean_token_accuracy": 0.24550262242555618,
      "num_tokens": 18051879.0,
      "step": 9905
    },
    {
      "entropy": 5.277436876296997,
      "epoch": 8.210439105219553,
      "grad_norm": 1.4140625,
      "learning_rate": 9.070708394054005e-05,
      "loss": 4.4187,
      "mean_token_accuracy": 0.2423293933272362,
      "num_tokens": 18062011.0,
      "step": 9910
    },
    {
      "entropy": 5.267582035064697,
      "epoch": 8.214581607290803,
      "grad_norm": 1.328125,
      "learning_rate": 9.052394691183921e-05,
      "loss": 4.5108,
      "mean_token_accuracy": 0.226079760491848,
      "num_tokens": 18071692.0,
      "step": 9915
    },
    {
      "entropy": 5.168189859390258,
      "epoch": 8.218724109362055,
      "grad_norm": 1.375,
      "learning_rate": 9.03411819914967e-05,
      "loss": 4.3456,
      "mean_token_accuracy": 0.24244830757379532,
      "num_tokens": 18080646.0,
      "step": 9920
    },
    {
      "entropy": 5.200239562988282,
      "epoch": 8.222866611433306,
      "grad_norm": 1.5078125,
      "learning_rate": 9.015878954816941e-05,
      "loss": 4.3671,
      "mean_token_accuracy": 0.24706092178821565,
      "num_tokens": 18088788.0,
      "step": 9925
    },
    {
      "entropy": 5.225918531417847,
      "epoch": 8.227009113504558,
      "grad_norm": 1.4140625,
      "learning_rate": 8.997676994976281e-05,
      "loss": 4.4078,
      "mean_token_accuracy": 0.24182597696781158,
      "num_tokens": 18098319.0,
      "step": 9930
    },
    {
      "entropy": 5.220188093185425,
      "epoch": 8.231151615575808,
      "grad_norm": 1.40625,
      "learning_rate": 8.979512356343023e-05,
      "loss": 4.4601,
      "mean_token_accuracy": 0.23988302946090698,
      "num_tokens": 18107740.0,
      "step": 9935
    },
    {
      "entropy": 5.236493062973023,
      "epoch": 8.235294117647058,
      "grad_norm": 1.46875,
      "learning_rate": 8.961385075557243e-05,
      "loss": 4.4468,
      "mean_token_accuracy": 0.23363285064697265,
      "num_tokens": 18117762.0,
      "step": 9940
    },
    {
      "entropy": 5.166470766067505,
      "epoch": 8.23943661971831,
      "grad_norm": 1.5625,
      "learning_rate": 8.943295189183632e-05,
      "loss": 4.3157,
      "mean_token_accuracy": 0.25396448373794556,
      "num_tokens": 18126212.0,
      "step": 9945
    },
    {
      "entropy": 5.272390270233155,
      "epoch": 8.24357912178956,
      "grad_norm": 1.5234375,
      "learning_rate": 8.925242733711484e-05,
      "loss": 4.4773,
      "mean_token_accuracy": 0.23481125980615616,
      "num_tokens": 18135622.0,
      "step": 9950
    },
    {
      "entropy": 5.23960394859314,
      "epoch": 8.247721623860812,
      "grad_norm": 1.5234375,
      "learning_rate": 8.90722774555456e-05,
      "loss": 4.4384,
      "mean_token_accuracy": 0.23727068603038787,
      "num_tokens": 18144768.0,
      "step": 9955
    },
    {
      "entropy": 5.195829963684082,
      "epoch": 8.251864125932062,
      "grad_norm": 1.4765625,
      "learning_rate": 8.889250261051063e-05,
      "loss": 4.4516,
      "mean_token_accuracy": 0.2447156623005867,
      "num_tokens": 18155107.0,
      "step": 9960
    },
    {
      "entropy": 5.217641305923462,
      "epoch": 8.256006628003314,
      "grad_norm": 1.328125,
      "learning_rate": 8.871310316463554e-05,
      "loss": 4.407,
      "mean_token_accuracy": 0.25019391179084777,
      "num_tokens": 18164064.0,
      "step": 9965
    },
    {
      "entropy": 5.198152494430542,
      "epoch": 8.260149130074565,
      "grad_norm": 1.5703125,
      "learning_rate": 8.853407947978845e-05,
      "loss": 4.4364,
      "mean_token_accuracy": 0.2350546672940254,
      "num_tokens": 18172714.0,
      "step": 9970
    },
    {
      "entropy": 5.216052627563476,
      "epoch": 8.264291632145817,
      "grad_norm": 1.328125,
      "learning_rate": 8.835543191707982e-05,
      "loss": 4.4267,
      "mean_token_accuracy": 0.24349797070026397,
      "num_tokens": 18182158.0,
      "step": 9975
    },
    {
      "entropy": 5.231154203414917,
      "epoch": 8.268434134217067,
      "grad_norm": 1.5625,
      "learning_rate": 8.81771608368613e-05,
      "loss": 4.3947,
      "mean_token_accuracy": 0.23894844949245453,
      "num_tokens": 18190605.0,
      "step": 9980
    },
    {
      "entropy": 5.247355031967163,
      "epoch": 8.272576636288319,
      "grad_norm": 1.46875,
      "learning_rate": 8.799926659872507e-05,
      "loss": 4.5748,
      "mean_token_accuracy": 0.2214917093515396,
      "num_tokens": 18199992.0,
      "step": 9985
    },
    {
      "entropy": 5.276301527023316,
      "epoch": 8.276719138359569,
      "grad_norm": 1.4609375,
      "learning_rate": 8.782174956150335e-05,
      "loss": 4.4308,
      "mean_token_accuracy": 0.2427278771996498,
      "num_tokens": 18208641.0,
      "step": 9990
    },
    {
      "entropy": 5.236592245101929,
      "epoch": 8.280861640430821,
      "grad_norm": 1.390625,
      "learning_rate": 8.764461008326723e-05,
      "loss": 4.4348,
      "mean_token_accuracy": 0.23547781854867936,
      "num_tokens": 18218355.0,
      "step": 9995
    },
    {
      "entropy": 5.219626712799072,
      "epoch": 8.285004142502071,
      "grad_norm": 1.4296875,
      "learning_rate": 8.746784852132661e-05,
      "loss": 4.4431,
      "mean_token_accuracy": 0.23401447087526323,
      "num_tokens": 18227217.0,
      "step": 10000
    },
    {
      "epoch": 8.285004142502071,
      "eval_entropy": 5.1498947468902,
      "eval_loss": 5.766971588134766,
      "eval_mean_token_accuracy": 0.16730784262611922,
      "eval_num_tokens": 18227217.0,
      "eval_runtime": 2.7493,
      "eval_samples_per_second": 1362.913,
      "eval_steps_per_second": 170.591,
      "step": 10000
    },
    {
      "entropy": 5.181580018997193,
      "epoch": 8.289146644573322,
      "grad_norm": 1.40625,
      "learning_rate": 8.729146523222877e-05,
      "loss": 4.3711,
      "mean_token_accuracy": 0.24438266903162004,
      "num_tokens": 18236410.0,
      "step": 10005
    },
    {
      "entropy": 5.25798020362854,
      "epoch": 8.293289146644574,
      "grad_norm": 1.5859375,
      "learning_rate": 8.711546057175804e-05,
      "loss": 4.4306,
      "mean_token_accuracy": 0.242798812687397,
      "num_tokens": 18244626.0,
      "step": 10010
    },
    {
      "entropy": 5.256560182571411,
      "epoch": 8.297431648715824,
      "grad_norm": 1.3828125,
      "learning_rate": 8.69398348949351e-05,
      "loss": 4.4429,
      "mean_token_accuracy": 0.2387913227081299,
      "num_tokens": 18253938.0,
      "step": 10015
    },
    {
      "entropy": 5.198227310180664,
      "epoch": 8.301574150787076,
      "grad_norm": 1.5234375,
      "learning_rate": 8.676458855601612e-05,
      "loss": 4.4351,
      "mean_token_accuracy": 0.2379624515771866,
      "num_tokens": 18262674.0,
      "step": 10020
    },
    {
      "entropy": 5.248046064376831,
      "epoch": 8.305716652858326,
      "grad_norm": 1.4453125,
      "learning_rate": 8.658972190849215e-05,
      "loss": 4.467,
      "mean_token_accuracy": 0.24207115918397903,
      "num_tokens": 18272558.0,
      "step": 10025
    },
    {
      "entropy": 5.2964802265167235,
      "epoch": 8.309859154929578,
      "grad_norm": 1.6171875,
      "learning_rate": 8.641523530508827e-05,
      "loss": 4.4912,
      "mean_token_accuracy": 0.24130599945783615,
      "num_tokens": 18282522.0,
      "step": 10030
    },
    {
      "entropy": 5.255693149566651,
      "epoch": 8.314001657000828,
      "grad_norm": 1.4375,
      "learning_rate": 8.624112909776314e-05,
      "loss": 4.4115,
      "mean_token_accuracy": 0.235798379778862,
      "num_tokens": 18290933.0,
      "step": 10035
    },
    {
      "entropy": 5.207491207122803,
      "epoch": 8.31814415907208,
      "grad_norm": 1.4296875,
      "learning_rate": 8.606740363770788e-05,
      "loss": 4.3971,
      "mean_token_accuracy": 0.2375014528632164,
      "num_tokens": 18300705.0,
      "step": 10040
    },
    {
      "entropy": 5.235866212844849,
      "epoch": 8.32228666114333,
      "grad_norm": 1.46875,
      "learning_rate": 8.589405927534571e-05,
      "loss": 4.3944,
      "mean_token_accuracy": 0.24056915044784546,
      "num_tokens": 18309889.0,
      "step": 10045
    },
    {
      "entropy": 5.247916793823242,
      "epoch": 8.326429163214582,
      "grad_norm": 1.46875,
      "learning_rate": 8.57210963603313e-05,
      "loss": 4.4581,
      "mean_token_accuracy": 0.23789738416671752,
      "num_tokens": 18319282.0,
      "step": 10050
    },
    {
      "entropy": 5.267377805709839,
      "epoch": 8.330571665285833,
      "grad_norm": 1.4921875,
      "learning_rate": 8.554851524154961e-05,
      "loss": 4.4805,
      "mean_token_accuracy": 0.24484300762414932,
      "num_tokens": 18328713.0,
      "step": 10055
    },
    {
      "entropy": 5.189864587783814,
      "epoch": 8.334714167357085,
      "grad_norm": 1.6328125,
      "learning_rate": 8.537631626711573e-05,
      "loss": 4.3413,
      "mean_token_accuracy": 0.2525541678071022,
      "num_tokens": 18337262.0,
      "step": 10060
    },
    {
      "entropy": 5.260660696029663,
      "epoch": 8.338856669428335,
      "grad_norm": 1.3359375,
      "learning_rate": 8.520449978437372e-05,
      "loss": 4.448,
      "mean_token_accuracy": 0.24227066338062286,
      "num_tokens": 18346699.0,
      "step": 10065
    },
    {
      "entropy": 5.20565242767334,
      "epoch": 8.342999171499585,
      "grad_norm": 1.4609375,
      "learning_rate": 8.50330661398962e-05,
      "loss": 4.3644,
      "mean_token_accuracy": 0.2394219756126404,
      "num_tokens": 18355466.0,
      "step": 10070
    },
    {
      "entropy": 5.193968391418457,
      "epoch": 8.347141673570837,
      "grad_norm": 1.46875,
      "learning_rate": 8.486201567948363e-05,
      "loss": 4.3675,
      "mean_token_accuracy": 0.24668571650981902,
      "num_tokens": 18363988.0,
      "step": 10075
    },
    {
      "entropy": 5.2645073413848875,
      "epoch": 8.351284175642087,
      "grad_norm": 1.296875,
      "learning_rate": 8.469134874816342e-05,
      "loss": 4.4934,
      "mean_token_accuracy": 0.2386019840836525,
      "num_tokens": 18373784.0,
      "step": 10080
    },
    {
      "entropy": 5.239567661285401,
      "epoch": 8.35542667771334,
      "grad_norm": 1.4765625,
      "learning_rate": 8.45210656901895e-05,
      "loss": 4.3679,
      "mean_token_accuracy": 0.23951668292284012,
      "num_tokens": 18382929.0,
      "step": 10085
    },
    {
      "entropy": 5.271424245834351,
      "epoch": 8.35956917978459,
      "grad_norm": 1.53125,
      "learning_rate": 8.435116684904136e-05,
      "loss": 4.4388,
      "mean_token_accuracy": 0.24105178266763688,
      "num_tokens": 18392399.0,
      "step": 10090
    },
    {
      "entropy": 5.268413257598877,
      "epoch": 8.363711681855841,
      "grad_norm": 1.3671875,
      "learning_rate": 8.418165256742346e-05,
      "loss": 4.5318,
      "mean_token_accuracy": 0.2282005712389946,
      "num_tokens": 18402255.0,
      "step": 10095
    },
    {
      "entropy": 5.221417999267578,
      "epoch": 8.367854183927092,
      "grad_norm": 1.328125,
      "learning_rate": 8.401252318726473e-05,
      "loss": 4.402,
      "mean_token_accuracy": 0.24990451484918594,
      "num_tokens": 18411677.0,
      "step": 10100
    },
    {
      "entropy": 5.1901134014129635,
      "epoch": 8.371996685998344,
      "grad_norm": 1.6015625,
      "learning_rate": 8.384377904971753e-05,
      "loss": 4.4151,
      "mean_token_accuracy": 0.2379810780286789,
      "num_tokens": 18420332.0,
      "step": 10105
    },
    {
      "entropy": 5.142832708358765,
      "epoch": 8.376139188069594,
      "grad_norm": 1.5625,
      "learning_rate": 8.367542049515733e-05,
      "loss": 4.3212,
      "mean_token_accuracy": 0.2413815438747406,
      "num_tokens": 18429542.0,
      "step": 10110
    },
    {
      "entropy": 5.23753342628479,
      "epoch": 8.380281690140846,
      "grad_norm": 1.46875,
      "learning_rate": 8.350744786318168e-05,
      "loss": 4.4211,
      "mean_token_accuracy": 0.25169758647680285,
      "num_tokens": 18439064.0,
      "step": 10115
    },
    {
      "entropy": 5.2034862518310545,
      "epoch": 8.384424192212096,
      "grad_norm": 1.4609375,
      "learning_rate": 8.33398614926097e-05,
      "loss": 4.4262,
      "mean_token_accuracy": 0.23721918910741807,
      "num_tokens": 18449001.0,
      "step": 10120
    },
    {
      "entropy": 5.22445592880249,
      "epoch": 8.388566694283346,
      "grad_norm": 1.328125,
      "learning_rate": 8.317266172148145e-05,
      "loss": 4.394,
      "mean_token_accuracy": 0.24567627161741257,
      "num_tokens": 18458105.0,
      "step": 10125
    },
    {
      "entropy": 5.239878511428833,
      "epoch": 8.392709196354598,
      "grad_norm": 1.484375,
      "learning_rate": 8.300584888705709e-05,
      "loss": 4.4433,
      "mean_token_accuracy": 0.23711068779230118,
      "num_tokens": 18466975.0,
      "step": 10130
    },
    {
      "entropy": 5.1691882610321045,
      "epoch": 8.396851698425849,
      "grad_norm": 1.390625,
      "learning_rate": 8.283942332581644e-05,
      "loss": 4.3307,
      "mean_token_accuracy": 0.25695629566907885,
      "num_tokens": 18475543.0,
      "step": 10135
    },
    {
      "entropy": 5.110739469528198,
      "epoch": 8.4009942004971,
      "grad_norm": 1.4140625,
      "learning_rate": 8.267338537345792e-05,
      "loss": 4.2864,
      "mean_token_accuracy": 0.2533396676182747,
      "num_tokens": 18484399.0,
      "step": 10140
    },
    {
      "entropy": 5.169474411010742,
      "epoch": 8.40513670256835,
      "grad_norm": 1.59375,
      "learning_rate": 8.250773536489833e-05,
      "loss": 4.4403,
      "mean_token_accuracy": 0.23278299421072007,
      "num_tokens": 18493168.0,
      "step": 10145
    },
    {
      "entropy": 5.14855318069458,
      "epoch": 8.409279204639603,
      "grad_norm": 1.34375,
      "learning_rate": 8.234247363427174e-05,
      "loss": 4.3975,
      "mean_token_accuracy": 0.2367030993103981,
      "num_tokens": 18502624.0,
      "step": 10150
    },
    {
      "entropy": 5.214443254470825,
      "epoch": 8.413421706710853,
      "grad_norm": 1.4453125,
      "learning_rate": 8.217760051492926e-05,
      "loss": 4.3817,
      "mean_token_accuracy": 0.2439698114991188,
      "num_tokens": 18510990.0,
      "step": 10155
    },
    {
      "entropy": 5.18201208114624,
      "epoch": 8.417564208782105,
      "grad_norm": 1.5390625,
      "learning_rate": 8.201311633943788e-05,
      "loss": 4.3685,
      "mean_token_accuracy": 0.24414715319871902,
      "num_tokens": 18519490.0,
      "step": 10160
    },
    {
      "entropy": 5.2158153533935545,
      "epoch": 8.421706710853355,
      "grad_norm": 1.4140625,
      "learning_rate": 8.184902143958013e-05,
      "loss": 4.4243,
      "mean_token_accuracy": 0.23719706237316132,
      "num_tokens": 18528827.0,
      "step": 10165
    },
    {
      "entropy": 5.168765115737915,
      "epoch": 8.425849212924607,
      "grad_norm": 1.546875,
      "learning_rate": 8.168531614635347e-05,
      "loss": 4.4106,
      "mean_token_accuracy": 0.24665599167346955,
      "num_tokens": 18537592.0,
      "step": 10170
    },
    {
      "entropy": 5.242663860321045,
      "epoch": 8.429991714995857,
      "grad_norm": 1.4609375,
      "learning_rate": 8.152200078996924e-05,
      "loss": 4.4967,
      "mean_token_accuracy": 0.24100762456655503,
      "num_tokens": 18547253.0,
      "step": 10175
    },
    {
      "entropy": 5.212944650650025,
      "epoch": 8.434134217067108,
      "grad_norm": 1.46875,
      "learning_rate": 8.135907569985248e-05,
      "loss": 4.4454,
      "mean_token_accuracy": 0.23465119898319245,
      "num_tokens": 18556658.0,
      "step": 10180
    },
    {
      "entropy": 5.2010722160339355,
      "epoch": 8.43827671913836,
      "grad_norm": 1.390625,
      "learning_rate": 8.119654120464077e-05,
      "loss": 4.4009,
      "mean_token_accuracy": 0.24089116603136063,
      "num_tokens": 18565594.0,
      "step": 10185
    },
    {
      "entropy": 5.216723346710205,
      "epoch": 8.44241922120961,
      "grad_norm": 1.546875,
      "learning_rate": 8.103439763218411e-05,
      "loss": 4.41,
      "mean_token_accuracy": 0.24066297113895416,
      "num_tokens": 18574033.0,
      "step": 10190
    },
    {
      "entropy": 5.19796347618103,
      "epoch": 8.446561723280862,
      "grad_norm": 1.484375,
      "learning_rate": 8.087264530954362e-05,
      "loss": 4.3997,
      "mean_token_accuracy": 0.2386510580778122,
      "num_tokens": 18583064.0,
      "step": 10195
    },
    {
      "entropy": 5.307903242111206,
      "epoch": 8.450704225352112,
      "grad_norm": 1.4453125,
      "learning_rate": 8.071128456299159e-05,
      "loss": 4.5014,
      "mean_token_accuracy": 0.23178473263978958,
      "num_tokens": 18591910.0,
      "step": 10200
    },
    {
      "entropy": 5.24851622581482,
      "epoch": 8.454846727423364,
      "grad_norm": 1.4765625,
      "learning_rate": 8.055031571801013e-05,
      "loss": 4.4145,
      "mean_token_accuracy": 0.2428817108273506,
      "num_tokens": 18600880.0,
      "step": 10205
    },
    {
      "entropy": 5.3274548053741455,
      "epoch": 8.458989229494614,
      "grad_norm": 1.640625,
      "learning_rate": 8.038973909929112e-05,
      "loss": 4.5911,
      "mean_token_accuracy": 0.22879723310470582,
      "num_tokens": 18610407.0,
      "step": 10210
    },
    {
      "entropy": 5.244914722442627,
      "epoch": 8.463131731565866,
      "grad_norm": 1.5,
      "learning_rate": 8.022955503073502e-05,
      "loss": 4.4515,
      "mean_token_accuracy": 0.2355645015835762,
      "num_tokens": 18620166.0,
      "step": 10215
    },
    {
      "entropy": 5.200718259811401,
      "epoch": 8.467274233637117,
      "grad_norm": 1.5390625,
      "learning_rate": 8.006976383545071e-05,
      "loss": 4.4013,
      "mean_token_accuracy": 0.23765162676572799,
      "num_tokens": 18628851.0,
      "step": 10220
    },
    {
      "entropy": 5.210610485076904,
      "epoch": 8.471416735708369,
      "grad_norm": 1.5703125,
      "learning_rate": 7.991036583575448e-05,
      "loss": 4.4211,
      "mean_token_accuracy": 0.24268756806850433,
      "num_tokens": 18637921.0,
      "step": 10225
    },
    {
      "entropy": 5.224740695953369,
      "epoch": 8.475559237779619,
      "grad_norm": 1.4609375,
      "learning_rate": 7.975136135316941e-05,
      "loss": 4.3577,
      "mean_token_accuracy": 0.24362326562404632,
      "num_tokens": 18647291.0,
      "step": 10230
    },
    {
      "entropy": 5.192912817001343,
      "epoch": 8.47970173985087,
      "grad_norm": 1.53125,
      "learning_rate": 7.959275070842502e-05,
      "loss": 4.3945,
      "mean_token_accuracy": 0.23919501453638076,
      "num_tokens": 18656252.0,
      "step": 10235
    },
    {
      "entropy": 5.161872673034668,
      "epoch": 8.483844241922121,
      "grad_norm": 1.3515625,
      "learning_rate": 7.943453422145626e-05,
      "loss": 4.4006,
      "mean_token_accuracy": 0.24831147193908693,
      "num_tokens": 18666199.0,
      "step": 10240
    },
    {
      "entropy": 5.2198844909667965,
      "epoch": 8.487986743993371,
      "grad_norm": 1.4609375,
      "learning_rate": 7.927671221140317e-05,
      "loss": 4.4452,
      "mean_token_accuracy": 0.24072585105895997,
      "num_tokens": 18675573.0,
      "step": 10245
    },
    {
      "entropy": 5.251893329620361,
      "epoch": 8.492129246064623,
      "grad_norm": 1.4609375,
      "learning_rate": 7.911928499660986e-05,
      "loss": 4.4227,
      "mean_token_accuracy": 0.24174360036849976,
      "num_tokens": 18684627.0,
      "step": 10250
    },
    {
      "entropy": 5.282138633728027,
      "epoch": 8.496271748135873,
      "grad_norm": 1.296875,
      "learning_rate": 7.896225289462426e-05,
      "loss": 4.4403,
      "mean_token_accuracy": 0.23444222658872604,
      "num_tokens": 18694287.0,
      "step": 10255
    },
    {
      "entropy": 5.195281553268432,
      "epoch": 8.500414250207125,
      "grad_norm": 1.6015625,
      "learning_rate": 7.880561622219735e-05,
      "loss": 4.4354,
      "mean_token_accuracy": 0.23889352083206178,
      "num_tokens": 18704014.0,
      "step": 10260
    },
    {
      "entropy": 5.249547338485717,
      "epoch": 8.504556752278376,
      "grad_norm": 1.4921875,
      "learning_rate": 7.864937529528232e-05,
      "loss": 4.4164,
      "mean_token_accuracy": 0.2446930304169655,
      "num_tokens": 18713030.0,
      "step": 10265
    },
    {
      "entropy": 5.260916042327881,
      "epoch": 8.508699254349628,
      "grad_norm": 1.5078125,
      "learning_rate": 7.849353042903428e-05,
      "loss": 4.4346,
      "mean_token_accuracy": 0.24542842358350753,
      "num_tokens": 18721911.0,
      "step": 10270
    },
    {
      "entropy": 5.174026393890381,
      "epoch": 8.512841756420878,
      "grad_norm": 1.4375,
      "learning_rate": 7.833808193780934e-05,
      "loss": 4.3815,
      "mean_token_accuracy": 0.2417333483695984,
      "num_tokens": 18730688.0,
      "step": 10275
    },
    {
      "entropy": 5.245500612258911,
      "epoch": 8.51698425849213,
      "grad_norm": 1.3671875,
      "learning_rate": 7.818303013516401e-05,
      "loss": 4.471,
      "mean_token_accuracy": 0.24101609736680984,
      "num_tokens": 18740214.0,
      "step": 10280
    },
    {
      "entropy": 5.202622604370117,
      "epoch": 8.52112676056338,
      "grad_norm": 1.4921875,
      "learning_rate": 7.802837533385482e-05,
      "loss": 4.4278,
      "mean_token_accuracy": 0.23473288416862487,
      "num_tokens": 18748837.0,
      "step": 10285
    },
    {
      "entropy": 5.159579133987426,
      "epoch": 8.525269262634632,
      "grad_norm": 1.6171875,
      "learning_rate": 7.787411784583733e-05,
      "loss": 4.3047,
      "mean_token_accuracy": 0.2515420287847519,
      "num_tokens": 18757482.0,
      "step": 10290
    },
    {
      "entropy": 5.1489521026611325,
      "epoch": 8.529411764705882,
      "grad_norm": 1.4375,
      "learning_rate": 7.772025798226582e-05,
      "loss": 4.4039,
      "mean_token_accuracy": 0.23894759714603425,
      "num_tokens": 18766326.0,
      "step": 10295
    },
    {
      "entropy": 5.334236145019531,
      "epoch": 8.533554266777134,
      "grad_norm": 1.6171875,
      "learning_rate": 7.756679605349241e-05,
      "loss": 4.5023,
      "mean_token_accuracy": 0.2361132800579071,
      "num_tokens": 18774287.0,
      "step": 10300
    },
    {
      "entropy": 5.202700662612915,
      "epoch": 8.537696768848384,
      "grad_norm": 1.234375,
      "learning_rate": 7.741373236906658e-05,
      "loss": 4.4648,
      "mean_token_accuracy": 0.24067246168851852,
      "num_tokens": 18785172.0,
      "step": 10305
    },
    {
      "entropy": 5.213409757614135,
      "epoch": 8.541839270919635,
      "grad_norm": 1.5078125,
      "learning_rate": 7.726106723773452e-05,
      "loss": 4.4701,
      "mean_token_accuracy": 0.2363409861922264,
      "num_tokens": 18794294.0,
      "step": 10310
    },
    {
      "entropy": 5.279458808898926,
      "epoch": 8.545981772990887,
      "grad_norm": 1.4765625,
      "learning_rate": 7.710880096743845e-05,
      "loss": 4.4568,
      "mean_token_accuracy": 0.23845926970243453,
      "num_tokens": 18803108.0,
      "step": 10315
    },
    {
      "entropy": 5.244077968597412,
      "epoch": 8.550124275062137,
      "grad_norm": 1.578125,
      "learning_rate": 7.695693386531612e-05,
      "loss": 4.4117,
      "mean_token_accuracy": 0.24705139845609664,
      "num_tokens": 18811215.0,
      "step": 10320
    },
    {
      "entropy": 5.221753454208374,
      "epoch": 8.554266777133389,
      "grad_norm": 1.4375,
      "learning_rate": 7.680546623770001e-05,
      "loss": 4.5008,
      "mean_token_accuracy": 0.23541390746831894,
      "num_tokens": 18820796.0,
      "step": 10325
    },
    {
      "entropy": 5.3375458240509035,
      "epoch": 8.55840927920464,
      "grad_norm": 1.453125,
      "learning_rate": 7.665439839011701e-05,
      "loss": 4.5435,
      "mean_token_accuracy": 0.22332278937101363,
      "num_tokens": 18830673.0,
      "step": 10330
    },
    {
      "entropy": 5.23153281211853,
      "epoch": 8.562551781275891,
      "grad_norm": 1.5078125,
      "learning_rate": 7.650373062728735e-05,
      "loss": 4.4414,
      "mean_token_accuracy": 0.24291413873434067,
      "num_tokens": 18839898.0,
      "step": 10335
    },
    {
      "entropy": 5.226158142089844,
      "epoch": 8.566694283347141,
      "grad_norm": 1.484375,
      "learning_rate": 7.635346325312437e-05,
      "loss": 4.3968,
      "mean_token_accuracy": 0.24004354774951936,
      "num_tokens": 18848380.0,
      "step": 10340
    },
    {
      "entropy": 5.212227916717529,
      "epoch": 8.570836785418393,
      "grad_norm": 1.53125,
      "learning_rate": 7.62035965707339e-05,
      "loss": 4.3686,
      "mean_token_accuracy": 0.24487358331680298,
      "num_tokens": 18857157.0,
      "step": 10345
    },
    {
      "entropy": 5.224075984954834,
      "epoch": 8.574979287489644,
      "grad_norm": 1.4609375,
      "learning_rate": 7.605413088241325e-05,
      "loss": 4.4296,
      "mean_token_accuracy": 0.24436374753713608,
      "num_tokens": 18866544.0,
      "step": 10350
    },
    {
      "entropy": 5.252039957046509,
      "epoch": 8.579121789560896,
      "grad_norm": 1.4140625,
      "learning_rate": 7.59050664896512e-05,
      "loss": 4.4748,
      "mean_token_accuracy": 0.23758187443017958,
      "num_tokens": 18876052.0,
      "step": 10355
    },
    {
      "entropy": 5.190032958984375,
      "epoch": 8.583264291632146,
      "grad_norm": 1.59375,
      "learning_rate": 7.575640369312684e-05,
      "loss": 4.3362,
      "mean_token_accuracy": 0.2482747182250023,
      "num_tokens": 18885141.0,
      "step": 10360
    },
    {
      "entropy": 5.208471202850342,
      "epoch": 8.587406793703398,
      "grad_norm": 1.5546875,
      "learning_rate": 7.560814279270928e-05,
      "loss": 4.3555,
      "mean_token_accuracy": 0.2501200929284096,
      "num_tokens": 18893108.0,
      "step": 10365
    },
    {
      "entropy": 5.22042293548584,
      "epoch": 8.591549295774648,
      "grad_norm": 1.5234375,
      "learning_rate": 7.5460284087457e-05,
      "loss": 4.409,
      "mean_token_accuracy": 0.24517705738544465,
      "num_tokens": 18901436.0,
      "step": 10370
    },
    {
      "entropy": 5.212440729141235,
      "epoch": 8.595691797845898,
      "grad_norm": 1.4453125,
      "learning_rate": 7.53128278756171e-05,
      "loss": 4.3879,
      "mean_token_accuracy": 0.25008017122745513,
      "num_tokens": 18910057.0,
      "step": 10375
    },
    {
      "entropy": 5.089436960220337,
      "epoch": 8.59983429991715,
      "grad_norm": 1.4296875,
      "learning_rate": 7.516577445462493e-05,
      "loss": 4.2756,
      "mean_token_accuracy": 0.2548051714897156,
      "num_tokens": 18918270.0,
      "step": 10380
    },
    {
      "entropy": 5.244073057174683,
      "epoch": 8.6039768019884,
      "grad_norm": 1.4609375,
      "learning_rate": 7.501912412110331e-05,
      "loss": 4.4713,
      "mean_token_accuracy": 0.23494393080472947,
      "num_tokens": 18928202.0,
      "step": 10385
    },
    {
      "entropy": 5.191400098800659,
      "epoch": 8.608119304059652,
      "grad_norm": 1.5546875,
      "learning_rate": 7.487287717086193e-05,
      "loss": 4.4384,
      "mean_token_accuracy": 0.24190011024475097,
      "num_tokens": 18937837.0,
      "step": 10390
    },
    {
      "entropy": 5.225221490859985,
      "epoch": 8.612261806130903,
      "grad_norm": 1.5078125,
      "learning_rate": 7.472703389889695e-05,
      "loss": 4.4046,
      "mean_token_accuracy": 0.24366991966962814,
      "num_tokens": 18946391.0,
      "step": 10395
    },
    {
      "entropy": 5.2270301342010494,
      "epoch": 8.616404308202155,
      "grad_norm": 1.4609375,
      "learning_rate": 7.458159459939007e-05,
      "loss": 4.4255,
      "mean_token_accuracy": 0.23782864958047867,
      "num_tokens": 18955269.0,
      "step": 10400
    },
    {
      "entropy": 5.198310232162475,
      "epoch": 8.620546810273405,
      "grad_norm": 1.453125,
      "learning_rate": 7.443655956570842e-05,
      "loss": 4.3521,
      "mean_token_accuracy": 0.2497437909245491,
      "num_tokens": 18964030.0,
      "step": 10405
    },
    {
      "entropy": 5.2013373374938965,
      "epoch": 8.624689312344657,
      "grad_norm": 1.421875,
      "learning_rate": 7.429192909040339e-05,
      "loss": 4.3724,
      "mean_token_accuracy": 0.24683944284915924,
      "num_tokens": 18973690.0,
      "step": 10410
    },
    {
      "entropy": 5.202197265625,
      "epoch": 8.628831814415907,
      "grad_norm": 1.578125,
      "learning_rate": 7.414770346521043e-05,
      "loss": 4.3663,
      "mean_token_accuracy": 0.2508851453661919,
      "num_tokens": 18981339.0,
      "step": 10415
    },
    {
      "entropy": 5.225391435623169,
      "epoch": 8.632974316487159,
      "grad_norm": 1.5703125,
      "learning_rate": 7.400388298104851e-05,
      "loss": 4.4009,
      "mean_token_accuracy": 0.23800765126943588,
      "num_tokens": 18990141.0,
      "step": 10420
    },
    {
      "entropy": 5.148018503189087,
      "epoch": 8.63711681855841,
      "grad_norm": 1.453125,
      "learning_rate": 7.386046792801911e-05,
      "loss": 4.3871,
      "mean_token_accuracy": 0.24623952209949493,
      "num_tokens": 18999217.0,
      "step": 10425
    },
    {
      "entropy": 5.212031412124634,
      "epoch": 8.64125932062966,
      "grad_norm": 1.5625,
      "learning_rate": 7.371745859540622e-05,
      "loss": 4.48,
      "mean_token_accuracy": 0.23905962109565734,
      "num_tokens": 19008554.0,
      "step": 10430
    },
    {
      "entropy": 5.171189689636231,
      "epoch": 8.645401822700912,
      "grad_norm": 1.4609375,
      "learning_rate": 7.357485527167522e-05,
      "loss": 4.3858,
      "mean_token_accuracy": 0.2464231014251709,
      "num_tokens": 19018159.0,
      "step": 10435
    },
    {
      "entropy": 5.20536789894104,
      "epoch": 8.649544324772162,
      "grad_norm": 1.625,
      "learning_rate": 7.343265824447254e-05,
      "loss": 4.3884,
      "mean_token_accuracy": 0.24610709100961686,
      "num_tokens": 19026985.0,
      "step": 10440
    },
    {
      "entropy": 5.231990766525269,
      "epoch": 8.653686826843414,
      "grad_norm": 1.3515625,
      "learning_rate": 7.329086780062522e-05,
      "loss": 4.4395,
      "mean_token_accuracy": 0.2351895123720169,
      "num_tokens": 19036745.0,
      "step": 10445
    },
    {
      "entropy": 5.198922967910766,
      "epoch": 8.657829328914664,
      "grad_norm": 1.46875,
      "learning_rate": 7.314948422613998e-05,
      "loss": 4.3908,
      "mean_token_accuracy": 0.24633614718914032,
      "num_tokens": 19045200.0,
      "step": 10450
    },
    {
      "entropy": 5.2414083003997805,
      "epoch": 8.661971830985916,
      "grad_norm": 1.6171875,
      "learning_rate": 7.30085078062031e-05,
      "loss": 4.3848,
      "mean_token_accuracy": 0.2518708869814873,
      "num_tokens": 19054102.0,
      "step": 10455
    },
    {
      "entropy": 5.244973087310791,
      "epoch": 8.666114333057166,
      "grad_norm": 1.453125,
      "learning_rate": 7.286793882517922e-05,
      "loss": 4.4706,
      "mean_token_accuracy": 0.24097106158733367,
      "num_tokens": 19062895.0,
      "step": 10460
    },
    {
      "entropy": 5.1470277309417725,
      "epoch": 8.670256835128418,
      "grad_norm": 1.5859375,
      "learning_rate": 7.272777756661152e-05,
      "loss": 4.3692,
      "mean_token_accuracy": 0.24908449947834016,
      "num_tokens": 19071523.0,
      "step": 10465
    },
    {
      "entropy": 5.185087251663208,
      "epoch": 8.674399337199668,
      "grad_norm": 1.484375,
      "learning_rate": 7.258802431322047e-05,
      "loss": 4.3948,
      "mean_token_accuracy": 0.24052246063947677,
      "num_tokens": 19080765.0,
      "step": 10470
    },
    {
      "entropy": 5.239206123352051,
      "epoch": 8.67854183927092,
      "grad_norm": 1.3671875,
      "learning_rate": 7.244867934690364e-05,
      "loss": 4.4266,
      "mean_token_accuracy": 0.2362300843000412,
      "num_tokens": 19089871.0,
      "step": 10475
    },
    {
      "entropy": 5.192302560806274,
      "epoch": 8.68268434134217,
      "grad_norm": 1.4140625,
      "learning_rate": 7.230974294873513e-05,
      "loss": 4.3649,
      "mean_token_accuracy": 0.241332483291626,
      "num_tokens": 19098928.0,
      "step": 10480
    },
    {
      "entropy": 5.248245811462402,
      "epoch": 8.68682684341342,
      "grad_norm": 1.4375,
      "learning_rate": 7.217121539896476e-05,
      "loss": 4.4706,
      "mean_token_accuracy": 0.2373034432530403,
      "num_tokens": 19107795.0,
      "step": 10485
    },
    {
      "entropy": 5.3205671310424805,
      "epoch": 8.690969345484673,
      "grad_norm": 1.5078125,
      "learning_rate": 7.203309697701784e-05,
      "loss": 4.4917,
      "mean_token_accuracy": 0.23045167624950408,
      "num_tokens": 19117060.0,
      "step": 10490
    },
    {
      "entropy": 5.2316793441772464,
      "epoch": 8.695111847555923,
      "grad_norm": 1.3984375,
      "learning_rate": 7.189538796149424e-05,
      "loss": 4.4418,
      "mean_token_accuracy": 0.2372008502483368,
      "num_tokens": 19126876.0,
      "step": 10495
    },
    {
      "entropy": 5.320941686630249,
      "epoch": 8.699254349627175,
      "grad_norm": 1.421875,
      "learning_rate": 7.175808863016806e-05,
      "loss": 4.4995,
      "mean_token_accuracy": 0.23131019175052642,
      "num_tokens": 19136888.0,
      "step": 10500
    },
    {
      "epoch": 8.699254349627175,
      "eval_entropy": 5.140484463177256,
      "eval_loss": 5.767979621887207,
      "eval_mean_token_accuracy": 0.1676151059996853,
      "eval_num_tokens": 19136888.0,
      "eval_runtime": 2.7771,
      "eval_samples_per_second": 1349.235,
      "eval_steps_per_second": 168.879,
      "step": 10500
    }
  ],
  "logging_steps": 5,
  "max_steps": 12060,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.8558289265408e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}